111313 Bio41

Hi, good morning everyone. So I'm sure that the thing that's on everyone's mind right now is the midterm that you just took. So you have to wait another hour to find out about that. But after class go across the road, and in front of Gilbert, your TAs will be there to hand out the midterms. And I know it's a very tough exam, harder than it has been in years past. Look through it, and think about what you have on there. And there are opportunities to think about your learning strategies that are available to you. OK, so that's going to come back after class. Problem sets, I already told you about. So these are just up here. And then we had a packet of lecture notes for my last sections that were printed out. If you have them, great. If you don't have them yet, there's some in the front and some somewhere in the back. You can also download those, if you want them as a PDF. So on Monday, we started linking back genetics and biochemistry. And specifically talking about how you can use genetics to figure out biochemical pathways. And I introduced this wonderful fungus called Neurospora, and how this was used to figure out basic metabolic pathways-- how do you make amino acids, how do you make vitamins, things like that. And that's kind of walking you back from the biochemical perspective into how this taught us about how genetics could be used in a lot of different ways. So I'm going to come back to that, come back to Neurospora and how that specific case was used by geneticists to figure out pathways, but then kind of back up and go more general, and kind of give you a feel for how geneticists are currently working out problems in metabolism, and development, human diseases, and other things. So we kind of ended near here where we were talking about the connections between genes and enzymes and pathways. And we all walked through in this kind of-- I gave you a diagram of a factory for cars, as kind of an analogy to say biochemical pathways work in a particular order. If you break the pathway in one place, it's predictable kinds of things accumulate and what kind of things can bypass that mistake. And so we looked at this kind of-- in this way of saying, if you have a mutation in a gene, it affects the ability-- that gene can no longer make an enzyme. Those enzymes are catalyzed in certain reactions. And so if you break the pathway here, by having a mutation here, we have very predictable outcomes of that. So something's going to accumulate before the block in the pathways. So the compound X is going to accumulate. And if we had a way of bypassing this, if we wanted to have something that lived in this case, we could feed it anything after the pathway and it still lives. So in the case of gene 2 here, we'd be accumulating compound X and compound Y or compound Z you could bypass. This is sort of a general principle, the way these pathways work. So now let's go and apply it to the specific things. I actually think it's kind of cool that they could figure out-- they have three mutations. So also remember, when you want to figure out a pathway through a genetics point of view, you accumulate a lot of mutations that failed to make the end product that you're interested in. And then you have to sort out what those mutations are doing. How many of those mutations are affecting the same gene? Are those mutations dominant or recessive? So you can interpret them as losing the function of the gene, or possibly elevating the function of the gene, and walking through. And so in the case of making argenine, first we had the set of mutations that-- the fungus couldn't grow unless you fed it argenine. And then I said, OK we're going to-- there were seven genes defined by that. We're going to look at three of those. And what was interesting about these three is that they all could grow if you fed them argenine, but two of them could also grow if you fed them these related compounds, so citrulline and ornithine. And actually, just based on this, you can actually figure out the way that these genes must work in a pathway. And so think about what can rescue them and what can accumulate. And so if something rescues it, it's acting-- the gene has to act before the thing that rescues it. So if we look at these data, so what we know is that we have this one mutation, argenine 1. And if you feed it argenine, it will live. But if you feed it citrulline, it will also live. And if you feed it-- live and grow-- ornithine it also can live. So all of these are able to bypass it. And so what you know from that is that that mutation has to be affecting a step earlier on, because all of these things can substitute and the thing will live. And so if we want to look at where these things are, we can just look at a chart like this and say OK, argenine has to be before any of these things I feed it. arg1. Now there's a question about whether arg1 goes here or here, these two arrows. Right now we can't tell unless we had another compound to test. So if we had a compound X to test, and we said if we feed arg1-- this mutant-- if we feed it X and it can't grow, do we put it before or after? After. So you can follow the same logic with these other ones. So arg2, we feed it citrulline, it can grow. Argenine, it will grow. But ornithine, it can't grow. So it has to be between these two. And then, the last one that's there is kind of obvious. Arg3, you can't feed it any of these things except for the argenine for it to live. So it has to be acting after those. So it's a really useful thing, because now they are able to put these enzymes in and correlate them with particular steps. If you remember from Monday, you had these compounds, citrulline and ornithine, and argenine. And they have particular carbons or nitrogens that are added to them. And so now you can say, well, I think maybe what arg2 is, is do the reaction that changes ornithine into citrulline. So this is some kind of cool pathway, figuring out that they did. But it turns out that this led-- opened the door for people thinking about genetics in different ways. And they said, well, in this case, we could figure out-- we had these three mutants, and we could figure out in what order they worked. So one of them catalyzed a reaction early in the pathway, one catalyzed a reaction later in the pathway. Maybe we could actually figure out the order in which genes work, using some other information. So let's see if we can generalize that. And again, coming back to this kind of forced analogy-- my apologies, I just really like these diagrams-- and think about what happens when you've got this kind of assembly line. We have a product that we need at the end. We have something that comes in. And on Monday, I told you the sad story of Carl here, who was so upset over a football game that he just left. And when he left, we didn't get an end product, because these steps all have to-- you can't do this without having the thing before it happening. So if he leaves, then the pathway grinds to a halt. What's before him accumulates, so you get more stuff ahead of him. But if you manage to ship something in from another factory, you could get to the endpoint if you applied something in here. So what if we have the same situation where Carl's left, but Carl convinces his friend Edie over here, in Headlights and Parking Lights, I guess, to also leave. So now this one's gone, and this one's gone. So now we have two disruptions. We're still not going to get any endpoint, but now let's think about what accumulates. So if Edie was the only one who left, what would accumulate would be the thing that's in front of him, right? But since both of these guys left, we're not actually going to see what we'd normally if just Edie left. In fact, we're going to see what happens when Carl leaves. We're going to see this one accumulating. And on the same token, if we try to provide something after the fact, we have to bypass both of these guys. So if you block the pathway in two different ways, we're going to see slightly different effects. And that's actually useful for genetics, because that sort of thing, of not really knowing-- so in the case of Edie walking out, we couldn't actually tell that Edie walked out based on what accumulates, because this was already blocked. And so that's basically the definition of epistasis, that because of a blockage or missing gene 1, you can't see what happens when you're missing gene 2. That's the definition. And this turns out to be really useful to figure out in what order certain things have to act. So this is the old genetics definition. So Gene A is epistatic to Gene B if the genotype, the alleles or the mutations at A affects whether or not we can see what happens due to the alleles at the other gene. So if we looked again at what's going to accumulate, so back to a biochemical pathway. And this is just what we had before. So we have these three genes working in this order, if we happen to know this already. And I said, OK, if you have a mutation in gene 1 and we block the pathway here, we know what accumulates, what's before it and what feeds it, or rescues it, or bypasses it. Those are all the things afterwards. What happens when you've got a double mutant? So if we block the pathway in two different places, what's going to accumulate? W. So what's going to feed this double mutant, and what's going to rescue it? Y or Z. So the ones that are after both of these. And if we did the same thing, knocking out this one and this one-- it's hard to erase what-- I can erase that. Look at that. So if we block these guys out, what's going to accumulate is what's before both of them. So X. And then we're only going to be able to feed with compound Z. So if you think about what the phenotypes are here, so when we look at this double mutant here, when we think about what's accumulating, this phenotype of the double mutant is the same as the phenotype of the gene that's acting first in the pathway. So basically, if we think about what's accumulating, the epistatic gene, the one that we see the phenotype of, the first gene in the pathway is the epistatic one. That's probably hard to read. That says epistatic, believe it or not. But if we look at it this way, we say what can we bypass with? Then the story is a little bit different. It's actually the last gene in the pathway. So I'm going to come back to that in a couple of different ways. So we can figure out what order things are, but you have to pay attention to whether we're asking whether something's accumulating, or whether we're bypassing it. And then, this is actually just walking through with the exact situation that we're in. So rather than a conceptual gene 1, gene 2, gene 3, in the case of the argenines, if we make a double mutant between arg1 and arg2, we think about what phenotype we get out of that. If we're asking what can fix this, what can we feed it to fix the problem-- so we have mutations in both of these. The only things that we can feed it to fix the problem are the things that are after both of them. So C and argenine. And this-- so you can go through this. So what's going to happen with arg1 and arg3? What's going to feed-- argenine. So the thing that's at the end. And the same thing here. So in general, when we're asking you a question about what kind of-- we've got two mutations and the phenotype that you're seeing, we're asking what bypasses both of them? You have to bypass both of them, so the last gene in the pathway is the one whose phenotype you actually see. So if we're in this kind of question, here the epistatic gene is always the one that acts last. Because we have to go around all of them. So the last one is the one that's important here. But then that switches around when we ask you what accumulates. Because what accumulates is what-- if you block the pathway, what accumulates is in front of the very first gene. So again, if you were going to do this double mutant, the same double mutant that we did before, and say, in this case we asked what's going to accumulate here? We need a compound to accumulate. What accumulates is what's before the very first one. So here the epistatic gene is first acting, or earlier acting. So once you work through a couple of problems, this is going to become really, really fast. But it's actually a really, really powerful thing. You put two mutants together, and you can figure out, within a pathway, which one is controlling the step that would be before the other step. And I'll show you some examples of how it's been used to figure out not only biochemical pathways, but signaling pathways. And for those that are interested in things like how do cells communicate to avoid overproliferation in cancer situations, all of that actually gets worked out by doing kind of tests like this, so figuring out which gene acts before another one in a pathway. So this is just what I summarized. If you've got two blocks at a pathway, you can figure out what order the genes are working in. And the really key things are figuring out whether or not this is something about-- whether we're asking you what bypasses this path, and what can you fix it with? And this, to tell you the truth, is mostly applicable to biochemical pathways and something that I'm going to call compound accumulates. That is, that something gets blocked and you can see it. This turns out to be very generally useful. So most things will fall into this category. And I'll give you some examples. Except you don't spell useful with two l's. So we can use it for both biochemical pathways but also signaling pathways. All right, so at the end of this project that Beadle and Tatum did over in Jordan Hall with Neurospora, their conclusion after all of this, doing these epistasis tests, was that they could figure out the connection between a particular gene and a particular enzyme, and the way that you could make amino acids. And so they got the Nobel Prize for this. This was a really big deal. And their conclusion was that mutations affect a process by changing an enzyme that is related to what the gene encodes. And we'll actually see in the next couple of lectures what that information is in a gene, what that information is in the DNA that gets translated into a protein. So this was important for that reason. But actually, really, to me, it is important because it sets up this general tool for understanding pathways. And so to kind of put that into perspective, you've actually, even without knowing this before, you've actually seen problems like this. And we've actually worked through problems with epistasis that are actually illustrating this kind of thing. We just talked about them in slightly different ways. So you might remember this one, back when we were just giving you some results from crosses and telling you, oh, you see some ratio. How many genes are involved? Is there epistasis? And we did something where we crossed a black mouse and a white mouse, got black mice, and got this kind of ratio out. Well, it turns out, and we talked about a white phenotype, not making pigment at all, being epistatic. And we just kind of left that conceptually there. But now that you think about biochemical pathways, you can start to think about how that might actually work. So perhaps you have something where it's a pigment precursor. So something that you know from biochemistry now a little bit, about molecules that can hold certain groups, certain functional groups in them that can be precursors to pigments. And then, specifically, a brown pigment, and specifically a black pigment. And remember when we talked about this, these two genes, we had gene C that was enabling color to be made, and gene B, that was making black color. So the B gene is probably working here. But we need to have a gene working earlier that enables you to go from some sort of pigment precursor into a specific pigment. And that's going to be C. Another thing that you'll see in a genetics problem is not what's up there. So another thing that you're going to see a lot in your homework problems is something where we tell you, there's this crazy pathway to make spots on butterfly wings, or something else. And there's lots of intermediates. And you need to figure out in what order those different intermediates accumulate. And you can figure this out by looking at what-- by basically applying this idea of epistasis. What blocks your ability to see something else? And we're going to do that over and over and over again. So imagine you have two mutants. If you-- both of them are not wild type, but they're both different from one another. And if you put them together-- there's mutant A and Mutant B-- put them together and you see the phenotype of mutant A, then we're going to be able to figure out which one comes first and which one comes second. So we're going to have-- so this is a very typical way that you'll see one of these tables. You're getting used to seeing these now. A bunch of crosses, so these are green-eyed mutants versus green-eyed mutants. If you cross the two-- if you cross it to itself, you still get green eyes. That's what you'd expect. But if we look at these double mutants-- and we won't talk right now about how you get the double mutants, just assume you're looking at double mutants-- a double mutant between something for the green phenotype and the black phenotype is black. So one thing that's important to talk about is what kind of problem is this? Are we bypassing something, or are we accumulating something? And the trick to knowing is that the only time it's ever a bypassing problem is when we have, somewhere in the text, you have now fed this thing this compound. Everything else-- when we say, you see these phenotypes, those are all compound accumulated type of problems. And that means that the epistatic gene, the phenotype that we see is acting first. So if you have this, then green and black, and you see black. So which one's first? OK, try that again. OK, so we have two things. We know that the phenotype that we see in the double mutant is the first block in the pathway. So when you see something like this, where we cross green to black, and we see black, you can conclude from that. Right, so the mutations in the gene for black are epistatic to the mutations that cause the green. So the mutations in the gene that lead to a black phenotype are epistatic to the mutations in the gene, a different gene, that leads to a green phenotype. So which of those genes, the one that can mutate to a black phenotype, or the one that can mutate to a green phenotype, would go first? Black, yes. So that's the absolute precise way of saying it. I've been shortcutting that. So thank you for bringing that up. [INAUDIBLE] OK. Should we just set up the old thing? [INAUDIBLE] Yeah, we can use a white board. You guys are really, really going to love my handwriting. So we have this chart, Luckily I wore something bright, so I can-- OK. So we have a chart in which we've done a lot of crosses. And every time you see one phenotype showing up in front of the other one, that's the one that seems to be acting first. So when we crossed-- so we know that in a pathway-- Saved. Usually what happens is you make enough of a fool of yourself doing something, it'll get fixed. So we can draw a little pathway. And that we think that black is before green. And geneticists always use these arrows to say that this one is-- so this one is before this one. So now we're going to do that with some other things. So we look at green crossed to all these other ones. If we look at green crossed to everything else, green versus orange versus purple or yellow, we're always seeing green. So we can figure out the order between those. So what's first? You guys figured it out? Yeah? OK. So he knew what? [INAUDIBLE] So black and green, and then what? [INAUDIBLE] OK but you guys have been walking through. So you know it's black and green and you said that all these are afterwards, right? That orange and purple and yellow are after green, right? So is orange next? So how do you know that orange is next? So let's look at when we crossed orange to orange it's still orange, orange to purple, and orange to yellow. So yeah, orange is next. And the which one? Then yellow. So when we do purple by yellow, it's purple. So yeah. So basically you're just trying to figure out, when you're looking at all this crosses, which one are we seeing? And the nice thing about this is you can double check yourself. So here you'd say, OK, I think it goes in this order. So if I cross purple by green, it better be green. So you look here, and you say, purple, green by purple is green. So you can always go back on these kinds of charts, draw out your pathway, and then just make sure it fits all of the data. So that's the easy way. That's when we already give you the double mutants. But remember, we're geneticists here. So now you have to make your own double mutants. So we're not actually going to make you walk through all of this, but we'll give you something like this of a chart at the end. And so this is, actually, if you think about it, I just said we have these double mutants and they appeared magically in that previous chart. But what would happen if I actually said, you need to set up the crosses like the way an actual geneticist would do it? And so what you'd have is you'd have, say for example, orange by purple. That was one thing in the cross before. So if we have this mutant and we have this mutant, and in the F1, say that wild type is red eyes, so what if we get that result? So we're going to know two things immediately about these mutations, orange and purple. Think about what we know immediately about those. They're recessive, absolutely right. So they're both recessive. That's really useful, because I've hinted before that recessive is loss of function. It's much easier to interpret that. So what else do we know? So how many genes are we talking about? Two. So you knew that kind of from the chart before. But when presented with something like this, we may give you, you cross these two things, you got this. And so what can you tell? You can tell these two things. Now that you can tell that they're both recessive and we've got two different genes here, then we can give them symbols, like we had before. So they're recessive, so orange we'll represent as little o's. And so for this purple one, it's going to be-- those are terrible symbols to use, but-- and purple will be two little p's. Two big p's. So our F1 looks like that. And when we cross these guys to each other, you're going to get a chart like this. So we're going to save you some time, by not making you continually write out charts like this. But OK. So then, we can start to figure out, just by the ratios that we get in this generation-- before, when you saw those nine to sevens, or nine to threes to fours, they didn't really mean very much. But now we can actually use that information to tell you which one goes first. So if we look at this chart, a bunch of these are going to be wild type, right? Because these are two recessive mutations. So any time we have a big allele of the o or the p, it's just going to be wild type. So we can ignore, basically, all of those. Let them-- get those out of the picture. Very conveniently, they're red. That was not planned. So we've gotten rid of four, seven, nine. So nine red, and we just need to look at what else we have in here. So we've got these other things, where some of them have two small p's for purple, and other ones have two small o's for orange. OK, so there's seven of these boxes, right? So three of them just have the two small p's. So those are purple. And three of them have just the small o's. And then we've got this last one, this box. This box just keeps coming up, right? This box is the important one to look at. And that one last box, what color is it? So what we'll give you is-- so if we give you an answer where we say, you did this cross, and the results where you got nine wild type, three purple, and four orange, then you're going to know what's in this box. And you're also going to know what came first. Because if this double mutant looks orange, that's going to be represented in this ratio of four. And it's also going to tell you that that comes first in the pathway. So even though you didn't know you were doing these before, every time you solved problems like that, you were actually kind of comparing what two genes were doing. And sometimes we talked about two genes that had nothing to do with one another. So a pea pod color and a pea pod shape are not dependent on one another. But whenever it came to coat colors or things like that, we were actually that-- the result in that little, that last box is actually very helpful for telling us which came first. So I kind of keep going on on this. But pathways is pretty much-- it's the fundamental thing that biochemists and geneticists are constantly trying to do. We're trying to understand the world by organizing it into nice little pathways. But we have slightly different approaches. So if you're a biochemist-- and he literally is standing in front of the biochemistry book, which I love in this photo. This is genetics. So if you're a biochemist, your approach is, how can I simplify this system so I can ask one very, very specific question, preferably in a test tube? So how can I reduce things down to one enzyme working on one reaction? So you want to simplify. It's nice if you have a in vitro reaction, so something that you can do in a test tube, with a small number of components. And a lot of this is focused on basic fundamental metabolism. That's really, really good for biochemistry. Things that all cells need to do in an organism. So fundamental reactions. This is not absolute, but these are things that these systems are really good at. Now, geneticists, on the other hand, are going to look at something and say, well, I'm not going to try to reduce this down to what I can do in a test tube. I'm going to look at this whole organism. So I'm going to look at this whole fruit fly, or Neurospora, and ask, what is something doing in the context of this living being? And so what geneticists are going to do is basically say, this is this wonderful, happy, functioning thing and I'm going to try to break it. So they're going to make mutants, and use techniques like epistasis to figure out what order those mutants are acting in. And other genetic tests. And this is really good when you want to focus on something that is-- maybe that not every cell needs to do. So if you want to figure out how a couple of neurons in your eye work, genetics can be very powerful. Because it's hard to biochemically purify those neurons. But if those neurons aren't there, you can't see. So there's a phenotype. And so these two things are kind of complementary in trying to figure out the world. One that reduces it down to a very simple and very clear answer, but that you need to get enough stuff to do that, versus geneticists, who say, I want to understand a function. I'm going to break this function. And this gene is somehow important for that function. And now we have techniques to figure out what gene-- how that gene relates to a particular enzyme or protein. And so I wanted to give this-- a little bit of perspective, because I know a lot of you are interested in medicine, and genetic diseases, and cancer. And one of the things that I was thinking about in thinking about the pathways that we work on, is that if you think about some of the really classical signaling pathways that are implicated in human disease, or human development, and you look at what we call those pathways-- they get called things like the notch pathway, or the hedgehog pathway, or the toll pathway, or the wnt pathway. And if you see these names, so what-- these are gene names, right? And these genes, if you figured out what protein they associated with, or are associated with, there are things like receptor kinases. So if you're a biochemist, biochemists are very methodical. They name things, they're all -ases, right? You got polymerases, kinases, carboxylases. But we call them by this. And the reason that we do is historically, we worked out these-- turned out to be biochemical pathways, by using genetics. This is a mutation that has a notched wing. This is a mutation that turns a fly embryo into something with lots of bristles. This one apparently means "cool" in German. And this one, I think, is a really great example of something where actually biochemistry and genetics come together, because this is something called the wnt pathway. And wnt comes from two different things. One is wingless, so another mutation, so a geneticist coming at this question, and integration one, which is a biochemist coming at this question. And it turns out it's the same person. And this is someone who's here at Stanford. So this is Roel Nusse. He's-- I think he's in the developmental biology program over in the medical school. You can find him. Just Google him. And he's got something that's called the wnt home page. And what he does is try to figure out very, very important signals. So how do cells tell their neighbor cells, look, it's time to divide, or it's not time to divide, you should go do this instead. And he looks at this partly as a biochemist looking at cancer and human cells, but also partly as a geneticist, using a whole bunch of different organisms to figure out how the pathway works. So it's-- genetics and biochemistry can get along very well together. And then this is just something I was going to use as an example, but I might actually not, for time. This is something that I took from a journal that talks about cancer. This is one of the major cancer pathways. It's actually called the epidermal growth factor. And it's a pathway by which a little signal that's secreted sees a receptor, so something at the surface of the cell. And that receptor sends a signal biochemically by changing whether or not it's phosphorylated, changing its shape and affecting some of these things, that affect these things and eventually leads-- it's the growth factor, so it eventually leads the cells to proliferate. And even though this looks like a biochemical pathway, it's pretty much worked out by Mike Simon, who's also in the biology program, who used genetics to really figure out all the parts up here. And Meera Sundaram who is not here, but is someone that I knew from a long time ago, worked out some really cool things in here, where as a geneticist, she said, the biochemists all say that this thing works directly with this, so they're directly linked. The substrate of this activity is this protein. But she found a protein that genetically-- a gene that encoded a protein that genetically appeared to be in between those two. And that confused people for a very long time, because a biochemist said, but this one talks to this one. So we don't need another protein. But it turns out, the thing that she found-- it's called KSR 1-- is one of often-mutated things in human cancer that actually helps to act as a scaffold. So it holds proteins together, so that this thing works very efficiently. And without it, even though in a test tube this thing can affect this one, in a normal cell it can't do it without this one. So there's a lot of things where genetics really can help us understand biochemistry. So you saw this already, and pretty much what I've been talking about for the last lecture and a half has been this link 1. So we're using genetics to figure out biochemistry. But the other way that they're linked is that the information in DNA actually can be used to make proteins. So we're going to now switch gears a little bit, and talk about that process whereby DNA is made into RNA and RNA is made into protein. So before I go there, though, it's a lot of stuff that I just gave you in the last half hour. So take a second to look at your notes, and ask any questions. Yeah? [INAUDIBLE] So if we're just asking you about phenotypes, and there isn't something in the problem where we specifically told you, you have this assay where we're giving you these compounds to feed them, then assume it's something that's it's one of these accumulation problems. So basically phenotype equals what accumulates. Yeah. That's a good way of thinking about it. All right, so at the end of this story with Neurospora, there was this idea that one gene equals one enzyme or one protein. So there's some connection between this and this. But it's not a direct connection. And there's something in between. And so there's two steps-- something called transcription, that I'll talk about, and something called translation that Professor Simoni will talk about. And what's really key to this-- the same thing over again-- so transcription is bright green, because it's very important. Translation is in black because I'm not talking about it. So in transcription, what you're doing is copying part of the DNA, but not all of the DNA. And before, when we talked about DNA replication-- and I'm going to make some comparisons between replication and transcription. So you can have-- if you start with DNA, there's basically two things you can do with it. You can replicate it, so you make-- those are two very ugly DNA strands-- but you make two of them, in which case you want to copy all of the DNA, and both strands of it. Or, you can do transcription, where you're going to selectively copy some parts of it. And this kind of idea-- this is called the central dogma of molecular biology, this process by which you go from DNA to RNA to protein. And this was actually a guy in Bio 41. So four years ago-- and he graduated with honors. Did an honors thesis, and is in medical school now. He sat kind of in the middle in the back. And after I started talking about the central dogma, he comes down at the beginning of class and lifts up his shirt-- which I was like, oh, wait, what's going on. And he was really, really committed. So he has DNA and RNA up here being made into protein. So really, really impressive. He told me he was going to work on Bio 42 going down his arm, but I never saw him again. So if any of you are-- I'm no longer shocked when people lift up their shirts down here. So if you've got tattoos, that's fine. And I'd love to add to the other things. But we're going to talk about what's happening on this guy's torso. So this going from DNA to protein-- there was an idea that there has to be something in between DNA. You can't-- even though the information in DNA might be there, you need some sort of intermediates going from that information to making a protein. And there was a pretty clear reason for why people thought that that must happen, and that just had to do with where these things are going on. And so if you're a eukaryotic cell, like one of our cells, with a nucleus, you have chromosomes in the nucleus. But you can also observe the protein-making factories-- the ribosomes. So the chromosomes are here. And these ribosomes are out here. So you have a little bit of a problem. If you're making protein out here, but the information to make it is in here, somehow you need to bridge the gap. They just weren't happening in the right place. So people were looking for something that could move between these two parts of the cell. And RNA was very attractive for that. So RNA is found everywhere. You can find RNA in the nucleus, you can find RNA outside the cell-- or not outside the cell, but in the cytoplasm of the cell. So that was good. But also really importantly about RNA, is that RNA could interpret the information from DNA, because structurally it's very similar. So it's a polymer of linked sugars and bases. Or sugar-phosphate links with the base coming off the end. And those bases look an awful lot like those of in DNA. So you've got a sugar-phosphate backbone. Remember DNA has this polarity of 5 prime to 3 prime. RNA has the same type of polarity. In this case, you're going to pair anti-parallels, so you're going to be going this direction. And the bases look very similar, with a slight exception-- that RNA doesn't have a T base. It has a U base instead. The other exception is that the RNA, the sugar that's the part of the backbone of RNA, has an extra OH group on it. So this was an attractive idea, that this was going to somehow read part of the DNA and then leave it, and move over to the ribosome and be translated into making protein. So we're going to figure out how this happens. So basically, how do you go from making this RNA from here and then what does that RNA do? And this is actually a really exciting field right now, because for a long time, what we thought was, you make RNA, and then all the RNA just gets translated into making proteins. And that's very important, but now we know that a lot of RNAs don't go down this pathway. They're called non-coding RNAs. And they do a lot of really fascinating things in the cell. And we're not really going to talk about those in Bio 41. You'll probably hear about those in Bio 104. And so we're going to focus mostly on how do you make this RNA to begin with? And there are three steps. Every time I put up this slide, I think, this is the stupidest slide in the world. You have to start, you have to go through the middle, then you have to end. And that's true. And yet, I still put the slide in every year. So I haven't totally figure that out. But then I realize there's a reason that I do this. So if you're a prokaryote-- I'm going to talk mostly about prokaryotic biology, so biology in bacteria, just because the principles are the same with eukaryotes, but eukaryotes are just more complicated. So prokaryotes and eukaryotes have to initiate. So they have to start. But what's interesting is they have to start some place, not just anywhere on the chromosome. They're going to copy a very specific part of the DNA. And they actually have to copy it, and then they have to stop. We never really talked about this with DNA replication. When we talked about DNA replication, you started someplace, and you got to the end of the chromosome, and then you were done. But this is going to be different. So in eukaryotes, you have some RNA and then you have to process it. So on the five prime end of it, you're going to give it a little cap. That helps it be translated. On the three prime end, there's a bunch of A's added to it. And then parts in the middle are actually taken out, in a process called splicing. So how does this work? So I'm going to talk in very similar ways to back when we talked about DNA replication, about the more biochemical approach, which is let's try to simplify this. Let's come up with a in vitro assay that lets us figure out what protein is responsible for this very important activity. So this important activity being, how do you take DNA, a double stranded template, your chromosome, put a couple things in there and get RNA to be read-- made from that? So you need an assay mix. So we're going to put in a test tube our template, so some double stranded DNA needs to go in there, similar to the chromosome. We need the building blocks. So the nucleotides are going to get linked together in order to make the RNA, which is exactly the same as the DNA, in terms of being linked together-- the differences being here that you're going to use this U version instead of a T. We want to be able to see that we've actually made RNA. So we're going to use some radioactive UTP. And we're going to use this one because it's more specific. It doesn't get shared. ATP is, as you know, is cellular energy. And if we labeled that, everything would get really kind of messy. So we're going to label this one. Some other things that are important for the reaction, but aren't so important for you to know, and we're going to put this in a tube and see if we can make RNA. Now when I talked about DNA replication a long time ago, one of the things I said was that when they tried to just put a template and then these raw materials, the nucleotides, it didn't really work. You needed a primer, a short sequence of RNA or DNA that was already made. Turns out for RNA polymerase, you actually don't need that primer. So you can actually just put a very simple reaction together-- this template, these activated precursors, and some buffers and things-- and then start purifying proteins. Now you know how to purify proteins. So lots and lots of columns and sizes and charges and all sorts of things. And you finally get out what you want. And they found an enzyme that was able to make RNA, given these raw materials. So this was called RNA polymerase. And I told you back when we had DNA polymerase. So DNA polymerase was found by Arthur Kornberg over in the medical school. RNA polymerase was found by Roger Kornberg. So, you know, it's a family thing. There's another brother, Tom, who doesn't have a Nobel Prize, poor guy. But he's a fascinating guy. He's at UCSF, and he's really fantastic. So this is Roger's enzyme. And this is another multi-subunit enzyme, so it's got a couple very important things that are always there, called the core enzymes. So two alpha subunits of beta and a beta prime. But this enzyme sometimes comes along with some other proteins. So these four subunits are always together. But depending on how you purify it, it can also come with something called sigma 70 and omega. And omega is probably to help the structure, help everything fold together. But sigma 70 has another really important role, and I'll tell you more about that. So if we talk about this whole thing, all of these subunits, that's called a holoenzyme. So if you see that on an exam or something, holoenzyme means including all of these different subunits. Core enzyme doesn't have the sigma 70. So that's the enzyme. What does it do? In terms of a biochemical reaction, it's basically the same as the DNA polymerase. So there's going to be a template here. And what it's going to do is allow a new nucleotide triphosphate. So this is the new thing coming in. This is RNA, so it's got an OH group here. It's got those three phosphates, it's going to bring this in and add it to a growing chain. Now the things that are different-- so this is different. This reaction is exactly the same as when you make DNA or RNA. And here it's written as a primer strand here, just in the way this is illustrated. It doesn't actually need one. So RNA polymerase can actually start-- link two nucleotides together, and it doesn't need to have a primer ahead of it. But this reaction-- if there is a primer ahead, and there's got a three prime OH group, this can bring in something new and form a bond here. So a very, very simple reaction. And this reaction releases these two phosphates. Those two phosphates break down, so you've got energy. So it's a energetically favorable process. So, pretty similar to DNA polymerase. They're both using DNA, but they're making something different from it. And I just mentioned this, that RNA polymerase doesn't require a primer. And there are a couple other things that are different. So DNA polymerase, if you remember, so DNA is one of those things that you want to copy once, you want to copy perfectly, you want to make sure there are no mistakes. And so a lot of that enzyme is designed to make sure it can go back and fix things, has these exonuclease activities, can back up and change things. And RNA polymerase really doesn't have any of those. So it used to be that I'd say that RNA polymerase doesn't-- and that's called proofreading, when you go back and check. So RNA polymerase-- it used to be thought that it doesn't proofread at all, it doesn't proofread well. And I have to say that, because Steve Block in the biology department pretty much proved that there is a little bit of proofreading activity in a very different mechanism. So now I'll just say it doesn't proofread very well, certainly not as good as DNA polymerase. And people think that part of the reason is, when you're making RNA, RNA is sort of a temporary messenger, something that's going to be used. You're going to make protein from that. You can tolerate a few mistakes, whereas if you make a mistake in DNA, that's inherited by the next generation and that's a disaster. So I always find it useful in my own head to think about two processes by comparing them. And so in terms of DNA replication and RNA transcription, we're seeing something using exactly the same template. So DNA, double stranded DNA, using exactly the same kind of mechanism, so taking nucleotide triphosphate, so NTPs, and adding them. Their base has to be complementary to something. They add them in a polarized way. So they add them three prime to five prime. But the products are a little bit different. So the template here is always double stranded DNA. If you remember back to DNA replication, there were a lot of enzymes. So there was DNA polymerase III. That was sort of the major enzyme. But there was also polymerase I. And then, remember, that's just the sort of the central part of it. If you wanted to unwind the DNA to get the polymerase in there, we had all those other things. So we had topoisomerases, helicases to open up the DNA, primases to get the thing started. What else do we have? Ligase. Probably forgotten something else. So single stranded binding proteins, et cetera. So a lot of enzymes. So you'll be pleased to know that for RNA transcription, you need RNA polymerase. That's it. And I'll kind of go through why you might only need one enzyme for that. So that's important. There's a whole entourage of things that are involved here. This is going to be very stripped down. Just that one enzyme that I showed you, with some of its subunits. So then you also have this problem of where do you start, and where do you finish? And so for copying DNA, the important thing is to copy all of it one time. So you start at the origin, the origin of replication. And you stop at the end of the chromosome. And you have to copy both strands. But when you're dealing with RNA transcription, what you're going to do is you're going to be starting the beginning of a gene within one chromosome. If you're a bacterium, you're going to have 6,000. 6,000 or so genes. If you're us, you've got 30,000 genes. So you're going to be starting and stopping at the beginning and the end of genes. And you're only going to be copying from one strand of that DNA. And so we're going to have to start thinking about those two strands differently. So one of those strands ends up being called the coding strand, and the other one the template strand. And you'll see in a moment which ones those are. So this immediately tells you a couple of things. So somehow, if this RNA polymerase is going to copy part of this DNA, it needs to be able to figure out where it needs to go to start. And so if you think about a chromosome, it's quite long. And in there, there'll be the genes interspersed. You'll have something-- so some part of that DNA is actually going to be copied and made into RNA. So let's represent that here. But there is-- remember, this DNA is all of those A's and T's and C's and G's. So even though this is the only part that's going to be copied, there's actually some information here that tells that RNA polymerase, hey, come over here and start and copy this strand of the DNA in this direction, and do it this many times. So we're going to use this other information. And this is something called a promoter. A promoter is something that's next to the part of the DNA that's actually going to be copied, and it recruits that RNA polymerase. So the role of this promoter is to attract the RNA polymerase, we need to figure out-- so it's going to go in one direction. And it's going to copy just on one strand going this way. It's going to ignore the other strand. And so we want to know, how do you identify these promoter sites? So how does RNA polymerase identify those promoter sites, and how do we as experimentalists figure out where those promoter sites are? And the answer is going to be that we're going to find out specific sequences of DNA. Specific-- two A's, and a T will tell this RNA polymerase that this is the place to start. And then allow that to initiate. And mRNA is a messenger RNA. So how do you figure out where RNA polymerase goes? So RNA polymerase is going to be cruising around. But it can recognize those promoter sites. So what you do is you allow an RNA polymerase to go on to a defined strand of DNA, and it will find where the promoter is. If you don't give it enough of the nucleotides, the things it's going to use to build another strand, then it has to just stay there. So it'll find where it should start, but it doesn't have the raw material to actually translate yet, or transcribe yet. So it'll sit there. And then you can do this nice little trick. You have the RNA polymerase sitting there, where it wants to copy something-- it can't do it yet-- sitting on the promoter. And this protein is basically protecting, wrapped around that DNA. So then you can add an enzyme that digests DNA, and it will digest everything outside of the stuff that's hidden, protected by this protein. So then you can do another trick to release this DNA from the enzyme afterwards, and then you can sequence what this DNA is to figure out what those bases are. So it's a nice little trick to figure out where the RNA polymerase is usually sitting, and that's the promoter region. So this actually takes a long time to do. So people did this to maybe 20 or 30 or 40 different regions, and found out where the RNA polymerase was sitting. And luckily, they started to see a pattern. So when they looked at the sequence of DNA that was always protected, they found that there were certain things in common. So these were actual sequences from different pieces of DNA that they gave it. And they saw there's a region-- so the transcription started here. When the RNA was made, this was the very first part of it. They would always see something that looked basically like TAT AAT. And that was a certain number of base pairs away. And so this is how people started to find out different sections that constitute a promoter. So this is sometimes called the TATA box, and sometimes called a minus 10 region. And it's a really defined region that now, we don't actually do the biochemical experiment anymore. It's much easier to get DNA sequence, and we look for things that have these particular bases. So how does this work? It turns out that RNA polymerase can wrap around and actually recognize those particular bases. So the RNA polymerase, it's a pretty big enzyme-- takes up a lot of room. And what it does is physically interact with things in these two places. And what's really useful about that is that it gives this thing a direction. So it's a particular size and shape. It has one end of it that's right sitting there at that plus end. That's where the active site is going to be, ready to do something. But it knows where that site is, because it kind of cozies up to the DNA, recognizing something over here that's not only a particular sequence, but a particular distance away from where it wants to start, and this other sequence. And so it orients in one direction. It's always going to be able to put its active site right here and not on the other direction, because it needs to recognize these sequences this far, 35 base pairs away, and these sequences, 10 base pairs away from here. So it's a nice little physical system that says, I'm going to take this enzyme to recognize the DNA, and it's going to be pointing in one direction. So the other thing that you may have noticed is when I said these all have sequences that look like TAT AAT, the truth is that none of them has a sequence that's exactly the same as this. So this is the idea of the consensus sequence. If you look at about 100 of them, this is what happens most of the time. These are the most predominant bases. But they're not exactly the same as this. Turns out, the closer you are to this exact sequence, the more attractive you are to RNA polymerase. So genes that have this at the beginning are very attractive, and actually attract lots of RNA polymerase and get made more-- the RNA gets made more often. If they're a little bit different from that, then it's less attractive. And if they're totally different, if they are full of G's here, then RNA polymerase can't recognize those. So it's a nice way of having in front of a gene, some information that tells you hmm, this is a gene that should be transcribed, made into RNA at a particular place, and at a particular amount. So then the question is, how does RNA polymerase actually recognize those things? So how does it know that it should be in this place? It's got to have some sites. And the protein that can actually recognize that minus 35 set of sequences in the minus 10-- and that turns out to be the sigma 70 subunit. So I said that this is something that sometimes comes off when you purify it. It isn't always associated with the core, so you can have it come off here. And what this allows RNA polymerase to do is actually pretty interesting. So RNA polymerase can work as an enzyme to keep adding new nucleotides to a growing chain. But it doesn't know where to start doing that, unless you have this-- that sigma 70 subunit. And so there's an idea that this thing is a part of this enzyme that allows it to see a certain sequence. But when it has-- when the RNA polymerase has the sigma subunit, it also moves really quickly. And this took me awhile to get my head around this. But let me see if I can walk you through this. So you've got DNA, and you've got a cell. Your RNA polymerase always wants to be scanning the DNA, but he wants to be kind of scanning through it pretty quickly. So if you have a version of the RNA polymerase that has a sigma 70 subunit with it, it's able to kind of glide along the DNA, scanning it, looking for the sequences that it likes. And it moves pretty quickly doing that. So when it finds those, then that sigma 70 subunit actually has the parts of the protein that can recognize the DNA. So it'll set it down in one particular place. So this, it lowers the general affinity of the RNA polymerase for DNA. So it will associate with DNA, but it can kind of glide along there. And it finds where it wants, and this will let it recognize that. And then the RNA polymerase basically kicks it out. So this helps it find where it should start. But then when it's time for the polymerase to do its polymerization action, to actually sit there and slowly add new nucleotides accurately, and do the enzymatic process, it wants to not be moving really quickly around the DNA. It wants to clamp on. So then it kicks out the sigma 70 subunit, and can kind of do its little reaction. So that's sort of the idea of why you'd have this sigma 70. So, initiation-- we're going to talk a lot about how you control that, because it's really important. This is going to go pretty quickly. And this is something new-- termination. We never really bothered to think about termination, but it actually turns out to be pretty important for RNA. So I think we will start this on Friday.