 Good morning Hopefully you had a chance to study some of the things yesterday that I went through possibly even read the first two chapters in the book and if you haven't got to start with that yet do it over the weekend because next week we're gonna Start doing even more fun stuff and then it's important that you know the basics before I'm Gonna start today's lecture. I'm gonna quickly recap what we talked about yesterday And then we're gonna spend some time on these study questions If there were lots of things we talked about but if I'm gonna itemized a few of the most important ones is I try to convince you How important proteins are and while proteins are not necessarily the only molecule? I like proteins because they capture the essence of what physics is about this strange boundary between things are too complicated to Deal with exactly with equations But they're still simple enough that we can understand them from physical principles And they do begin to capture the real complicated essence of biological processes I Jumped through the water and hydrogen bonding thing a little bit I'm gonna come back to that multiple times today and next week too So don't worry too much if you haven't understood all the details there and Then we spent quite some time talking about DNA and RNA in particular special molecules And we'll come back to that to the study questions in a few seconds Hopefully you all know what the central dogma is if not You will have a chance to learn it again in a few minutes Otherwise you should read up on it in the book and then I spent some time talking about the way we've been able to learn this in many ways science is challenging because The way we learn things historically is frequently not the same thing as we would do things today And of course to teach out to do things we should teach out how to do things today, right? But occasionally it helps to understand the state of science is frequently dependent on how we discovered things And that's why x-ray crystallography is still by far the most important method But I think cryo-EM in particular is race rapidly racing in importance But before we go on to the next stuff, let's spend some time focusing on these study questions And I've done depending on the dynamics in the group. We might end up doing this in slightly different ways. Let's Start with having you pick questions run Well, actually, no, let's start going through them in order But we can have anybody can decide to answer any questions And I would even the one thing you should not consider is whether you think your answer is right or not So just guess what you think it is because if if one of you happened to have the wrong answer I bet there are five more people got it wrong There are two options that either I realize that and then I spend some minutes explaining what the right answer is Or you keep quiet and let somebody else answer and then I won't have any idea that you misunderstood that concept It's entirely up to you, but I would prefer the first alternative So the first question is what are the time and length scales involved in biophysics, so what do they represent? Oh, sorry water. That should be a separate line time and length scales If we start with proteins, what type of scales are we talking about? Yeah, and that's the range of atoms and what time scales typically That's a bit of a tricky question. I kind of I kind of hinted about this But it's more complicated than you might think so what are the fastest time scales involved when we talk about proteins I would say nanoseconds if we're going to be really correct There are things like in photosynthesis and everything that it might be all the way down to femtore at those seconds But those are the exceptions and then it's electrodes so nanoseconds I think that's a great motivation, but what are the slowest time scales involved in proteins? I didn't tell you Normally in a cell I would say milliseconds or so, but normal is the keyword Protein some proteins when they insert in membranes they get 10 seconds And as we're going to come back to later in the course There are some processes like prions or infections that might have time frames of decades when they build up in your brain So can you just imagine that? time scales spanning from nanoseconds to 10 years or 100 years And then there are the I have these discussions too about population dynamics and everything We're not going to really get a talk so much about that so maybe we should leave that aside Why is water such a special molecule in biophysics? Yes Exactly and that has This too is a more complicated question that this is why water in general is special from all other solvents But all of physics and evolution has of course evolved to adapt the water I guess another way of phrasing this could be water is such an abundant abundant molecule on earth, right? So that if you want any type of process to survive on earth, it will have to adapt to water And then this has developed in symbiosis as we're going to see later proteins and everything we do they they always interact with water Your bodies are 70% water What is the typical size of a protein? Yeah, a few nanometers in terms of atoms. How many atoms? Yeah, I was a hundred is a very small protein and individual amino acid can easily be 15 atoms or so, right? So that a few thousand up to a million And then I I didn't explain what a hydrogen boner and acceptor was yesterday One thing I realized when I biked in this morning, I encourage you to watch the videos But from one point of view it's really stupid to watch the video from yesterday because you were likely present at that class You might want to look at YouTube my channel because you can find last year's videos and I present things I don't always say the same thing, right? So if you want a second opinion from me, sadly, you can actually what last year's video But what are hydrogen bond donors and acceptors? So we will we will come back to hydrogen bonds in detail But this has to do with this process, right? That if you start looking Inside a molecule when you have a few items such as oxygen and nitrogen they're exceptionally electronegative So they will tend to pull the electrons towards them And that means that the hydrogen then in particular end up with the deficit of electrons Which means that it will have a positive partial charge If you then put two such molecules next to each other You're effectively going to end up with one positive charge and one negative charge and then start to attract each other But this is really the way of describing this process who's attracting the electrons and who has a end up having a net partial charge But I will I think I will have 10 more slides about that the next few lectures I didn't tell you the energy of a hydrogen bond in water either, but maybe some of you looked it up So why do we use kilo calories? America's in particular love doing it. I occasionally use kilojones, but we talked about the molecule three minutes ago water How do you define a killer calorie? Yes, so these are really these are energy scales that are adapted roughly to the type of interactions We have at room temperature when it comes to molecules interacting Then again, we don't work with kilos of molecules, right? We work with individual molecules and then it works really well if you then say per mole We can use roughly the same numbers as you would use in the lab But when we say per mole then we can think about it on the individual molecule level and stuff And of course your money of your physicists you're more than welcome to use KT or You can well you can use Joel's skip the skip the per mole But then in everything you do you're gonna end up having carry this Multiplied by 10 tie up the race to the power of minus 25 and that gets very tedious and you're gonna keep forgetting And as sloppy as we are we should virtually every single energy I ever mentioned in this course is going to be per mole Do you think people are proper enough to remember that? We're gonna say a killer calorie and any type in a molecular level and anybody speaks about a killer calorie They really mean a killer calorie per mole So you're allowed to use any of these energy scales you want, but keep in mind which one you're using physicists loves KT, of course And I'll come back why KT is important later on What are the structural components of nucleic acids almost red amino acids? That's something I did instructed to read up on so what are the structural components of nucleic acids? Yes, so you pretty well, okay We even split the the sugar base into two parts right the part that is the base and then there are the parts that the sugar And then then this is connected to this chain of false fates There are a couple of names for these things the base or the sugar base is one the story is that Ribos and the oxyribos that had to do with the difference between DNA and RNA But in particular when you have these bases and then you're adding the sugar and then you're adding the false fates There are two names we use for these Nucleoside and nucleotide and which one is which? Yes, and That is important to keep in mind because suddenly you will see in books that people speak about nucleotides and it's important Remember, what is what? Name three pieces of information enabling DNA structure determination. I didn't I didn't Explain this in detail But this is an interesting question. I don't even remember exactly why I formulated the question that way, but it's an interesting question So if you're gonna determine the structure of DNA not the sequence, but the structure what is important or how do you do it? How would you do it? X-ray crystallography. Yeah, but that's a general technique. So if you to be able to determine this with x-ray crystallography What are the steps you would have to go through? hmm Exactly, the first thing is crystallography. You need to crystallize it. So you're gonna need to have provide some sort of crystal So first you're gonna need to purify the DNA, which is Actually not entirely trivial then you're gonna need to crystallize it You will need to create you will need to obtain this diffraction pattern, right? You will need to hope that your crystal is so pure that you get a good diffraction pattern And at some point you're gonna need to be able to build a model and make sure that your model agrees with the diffraction pattern I Gonna need to check what I what I also thought last year. I actually don't remember. This is not a good It's not a good very formulated in a good way Check my lecture last year and to see what what I was answering there. I'll check myself Roughly how many bases are there in the human genome? Yeah, that's a bit generous. Actually, well Actually, you perfectly right six billion is perfectly right the way I formulated the question. Why is six billion perfectly right? Exactly, so normally when we say basic we typically need base pairs right because if I have an adenine The partner of the adenine is not free. That has to be a timing, right? Or ever unless there is an error in the DNA So when I say base we frequently talk about base pairs if you determine the structure Sorry the sequence. There's no point in determining the sequence of both strands if I have the sequence on one strand the other one is given how many proteins are there then in the Roughly in the human genome. Yeah, 20,000 so when in The mid 90s when I started studying this we thought there were roughly 40,000 and then 30,000 and then 25,000 So this actually kept shrinking for a few years when we found that there are more and more of these that were either Duplicated or didn't really code for proteins So we will come back to that later in bioinformatics that it's just because you have this genome for a bacterium It's super easy. We can find what parts code for proteins But in a human or in general all higher organisms eukaryotes vertebrates It's very difficult to find out what parts of DNA code for proteins and it's a relatively small part And we don't really know what the other part does So to go back to this DNA double helix. How is the double helix stabilized? Hydrogen bonds and roughly how many hydrogen bonds are there? Yeah, so three and two respectively And if you can actually you can even do the math here, right? That if one hydrogen bond is in the ballpark of 2k cal or something There's gonna be an insane number of hydrogen bonds. It has a very high stabilization energy. It's very hard to tear DNA apart Which is good because otherwise we would be dead What does complementarity mean in DNA exactly did it let's see I don't think I have a specific question about that Did you read the Watson and Crick paper yesterday? So this is there's this is one of the most beautiful formulations in science, right? So what did they say? Like second paragraph from the end so they present the entire structure They are the ones who first proposed this complementarity and they proposed that the complementarity would also explain Shahgaff's rule so Shahgaff's rule was this observation that a and t always seems to occur in the same concentration and G and C always seem to occur in the same concentration And this is an important part of how science works I Shahgaff only observed that this is the case. It must mean something because he didn't understand what it meant But it was Watson and Crick that used that to realize that you must have base pair complementarity And the cool thing that this paper it has not escaped our notice that this provides a mechanism for inheritance period Just one sentence but that explains the entire basis of our dear genes evolved and everything basic Yeah, you thought that we hadn't thought about it. We have thought about it And that's really the birth of everything we do today in DNA genomics and everything goes prior to this Well for a long time We wasn't even sure whether the DNA contained the genetic information and then eventually we got to the point where it does contain the genetic information But prior to the Watson and Crick paper We had no idea how one generation actually passes on the genetic material to the next generation or how sales split and that's a beautiful You only need one sentence Well, you needed one sentence for this in 1953 today. This would have been supplementary information with figures 14 to 29 But I think it's a simply that sentence makes it worth reading the paper if you haven't done it So how does RNA differ from this? Yes, that's stupid Why shouldn't RNA be stable? But it wouldn't hurt to be stable, right? There is nothing good from being degraded Like generations of biochemistry students would be really happy if they didn't have to put their RNA on ice all the time It's worse than that RNA is also very sensitive to enzymes and everything. They would be very happy if RNA was a bit more stable well partly but Remember DNA How frequently do we when DNA is what is the main role of DNA DNA's role is just to store the information, right? It's like taking it back up and putting it in the basement. We rarely touch the DNA So for DNA it makes lots of cells that it should be stable We will have to open it now and then it doesn't really matter if that is a bit slow But if you think about RNA, so you kind of answer the question, right? But For RNA it would be really bad for the cell if it was very expensive to work with RNA because when the cell is working with RNA We're doing something with it. We are reading it from DNA We're transporting it and then we're gonna read it two seconds later in the ribosome And if this required you to have a very expensive operation to open it up It would use a ton of energy to produce proteins in the cell So for RNA it's actually important that it's not that stable or in particular that is not a double strand So a long time ago People actually discovered that RNA breaks down that has been known forever and there's been lots of studies Doing this so there were in the 1970s. There were a few groups studying this and Then there were some postdocs that found out that Shouldn't you start doing the same type of studies at DNA? What if DNA is not stable? That's the most stupid idea in the world, right? Because everyone knows that DNA is stable. We tend to teach it in undergraduate classes But the point is DNA isn't stable. It's more stable than RNA But it's not perfectly stable even in DNA despite all those hydrogen bonds you will occasionally have errors and And you can even show that DNA itself is so unstable that just due to the random errors You get for radioactivity and everything an organiser would be able to live more than a few weeks maybe a month and This was then the results that realized and that then your bodies must have an entire machinery to repair DNA To make sure that whenever whenever an error happened It's unlikely to happen in both bases, right? And if the error happens in one base This is kind of like a raided a disk or something in a computer. You still have the backup copy You have the other base so we can save this error By repairing in the collapsed base and then we haven't lost information if we only had one strand in DNA We would permanently lose the information with this happened The quote this was my uncle Thomas. He got the Nobel Prize for this a few years ago We use this today in drugs Because in particular cancers they undergo through very rapid cell division and if something goes undergoes very rapid cell division They are much more dependent on this machinery So surprisingly if you kick out the machinery that normally tries to repair things That's gonna hurt all the cells, but it will hurt cancer cells more So then you will selectively kill the cancer cells while you're gonna hurt the normal cells But not as much as the cancer cells And it's another one except these is seemingly a small minor stupid idea from a student or something, right? They can lead to a revolution when we learn how to use it. How does RNA perform its biological functions? Hmm. What is RNA polymerase? So what's the difference between DNA polymerase and RNA polymerase? There are lots of aces. So first what are aces? I spoke about that yesterday. Yes So why do we keep coming up with these names Well, the problem by only there are 20,000 proteins. There are so many processes, right? So in the 5th. Yeah, this is protein XYCP 29 That has the role of opening up DNA and reducing the activation barrier to if you're gonna mention that 20 times in an article people would die But they say when it's it's something related to the polymerization Extending a chain and this an ace any biochemist will instantly know that is the enzyme that activates that process and makes it happens So ace are enzymes or catalysts if you're not the chemists So what's the difference between DNA and RNA polymerase? I'm gonna be much easier. It's imagine if I If I was hung over and I had a student I have to write this exam so you're not you're not wrong, but As a physicist you frequently you won't understand the details, right? You need to learn to think like biologists here So what is a an ace first? It's an enzyme that much. We know it's a polymerase. What does a polymerase do if you had to guess? It polymerizes something right just gonna it's gonna create the long chain of something Good you have two long chains. You have a DNA chain and an RNA chain. So which one do you think keeps building DNA chains? DNA polymerase so you do any wild guesses which one might be involved in general building RNA chains RNA polymerase so when do you build DNA straight chains and when do you build RNA chains? Yes, so what's and then you have to remember what is the difference between replication? Translation so this is far easier in physics. I don't remember what an up and down at the top and the bottom quark is anymore These things that you just have to look at the world and see that you're replicating something Do you have any other yes? So if you have to from DNA to RNA is the RNA Exactly, so if you're replicating something you shouldn't have up with a copy, right? And you never copy at the RNA is not an exact copy of the DNA So the replication is when you need to start from DNA and create two copies of the DNA and Even without knowing anything you could write down the entire process just by looking at the words and thinking about what they mean So the other process they mentioned that's the translation when we're translating from something to something else So all these names are far easier than you think don't try to learn them by heart. I never bothered doing that I need to tell you I don't even I don't even Know that a definition between replication and translation by heart I always spend a second thinking about which one is which useless knowledge to know by heart So that comes down to this central dogma. What was the central dogma? So that and what can you say anything about this flow is? Unidirectional in particular right so it only goes from DNA to RNA to protein and I'm not gonna in the interest of time I'm not gonna repeat everything, but a typical study a question. I might ask you to explain the central dogma Point out the differences processes. What are the molecules involved and what are the enzymes involved? And then it's not quite as easy as saying that it's just the information flow, right? You need to know what is DNA? What is RNA? What is the transcription? What is the translation? What is the replication and we're doing up with the protein? But again, you don't have to know it all by heart 90% of this you can just get if you remember the words You can place the right word in the right place I think and that immediately leads to function 14 can function induce or lead to change this in structure and bait was you Just said the answer to that should be affirming. No, right? That's actually good question actually I should modify this someone when I talk about structure We mostly think about proteins because the structure of DNA is kind of given, right? Actually, it's not given but when it goes to DNA we mostly think about DNA as a sequence Which is stupid because you're right. It is a structure in DNA. Actually, there's much more structure in DNA than you think But can the function of a protein change the way the protein looks? Can you go kind of close this loop in the central dogma? So based on the plots we had yesterday What they'll know they also would be no but this goes down to this natural selection, right? So the individual cell really can't do it a part of the diesel post-translational modifications and things but the way nature and Evolution can do this because if one generation or gene has a property that makes it Better better to adopt then that will likely lead to more offspring and over billions of years You will then gradually have the gene shipped in that direction, but that's a much more subtle feature Number 15, we're going to come back to in bioinformatics with how does nature know when to start to stop reading DNA? What is the beginning of the protein and what is the end of the road? You have three billion bases? That's a fairly large library. Yes, and in particular bacteria. This is super easy There's a specific sequence of three bases that you have a another one of these enzymes Recognizing and then you start binding there and then you start reading it. It's much more complicated in humans And that has to do with the fact that not all parts are real proteins, right? And the similar there's a codon that says we're going to stop Number 16 is a question that we love to put on exams. Why are some amino acids more common than others in nature? Yes, so what what did we call those three the three this the sequence of three bases that codes for an amino acid What did we call codon? the reason why I love this is that The more you know the easier it is to make a mistake here because when the second people have learned a bit of bioinformatics They start off but all membrane proteins need to be hydrophobic and you throw the kitchen sink of information on this But it's actually very simple. This is determined directly by DNA and probably you can get this with mathematical statistics Which is a beautiful example that is this is based on math and physics not life science We already covered replication translation and transcription of it So I won't go into that detail again But I based on what you're saying you might want to spend a all of you might want to have a second look on the day The structure of DNA and RNA. Oh, sorry The codons there are four different bases, right and three of them code for an amino acid And that means in the first base you have one out of four in the second one You have four out of four and then the third base or one out of four in all three So that's four multiplied by four multiplied by four. That's 64. You only have 20 amino acids so there is a redundancy in the genetic code that Some amino acids are coded for by more than one codon And that tends up nature natural selection has created this a way so that the most common amino acid the ones that you need Of a lot the normal building blocks those are coded for by many codons while the very special rare say Tryptophan the large ones that we don't want that many of they only have one codon coding for it So the second you produced an amino acid. There is no longer any memory what codon it came from So that if you take something simple like alanine or valine, right? If it's coded for by many codons Is there a difference between the codons there? Was that your question or It's a very good question. We don't know the answer to that That's not there have been for membrane proteins There have been some discussions that some of the hydrophobic amino acids whether This could be used by the entire ribosome factory to recognize What pieces of RNA should eventually be transported into the membrane? But this is very much in speculation and everything so today. We just know that Some amino acids are coded for by one codon. We don't know whether it's in a deeper biological meaning to it It's a great research topic for the future No, so that the relative abundance of amino acids in your genes are discovered exactly Is determined exactly by these codons so you can see and they are out of these three billion base pairs To first approximation all of them occurred roughly the same concentration So it's to first or at least in nature as a whole so it appears to be random It's not based on how stable the amino acids are but we're gonna come back that late now Now that we're gonna convert to proteins because when it comes to proteins Not all sequences of amino acid will fold into stable proteins and then the things that you mentioned is highly relevant But whether when it comes to producing amino acids, it doesn't appear to play in the room We spoke a little bit about the methods to determine protein structure But simply I probably spend more time than I should have done on that yesterday. I might skip that What is the typical resolution of a good protein structure? I hate somebody whispering something, but I don't Ongstrom. Yes This is why no matter all the standardization and everything we have the reason why everyone It's not just in Sweden, but everyone all over the world keeps using Ongstrom Ongstrom is a very nice unit of measure that you talk about proteins because everything this a bond is 1.5 angstrom or so right you don't have to use that many decimals and everything So it's a nice natural length scale for proteins Can you think of some limitations in the way we determine structures in particular the ones? I mentioned yesterday compared to the real cellular environment where they exist Crystallizing is one Certain issue there right so that's proteins in a real cell a real cell is not a crystal and you're gonna pack proteins in a very strange way That's certainly a problem that you end up with the radiation damage in all these methods So what temperature do you perform these experiments? Either XA or cryo EM the second one might give you a hint 100 Kelvin right liquid nitrogen because it's cheap And I'm not sure about you even when it's a really cold the outside It's most of our cells don't live at 100 Kelvin There is something else that a protein in the cell typically occurs in a very aqueous environment There's just one protein and then there's water all around it. This is somewhat related to the crystals But even in cryo EM you might not strictly have crystals If you only had a few proteins you would need to collect data forever So you have a very high concentration of proteins in these samples whether it's a crystal and you hardly have any water Or just overexpressed proteins on the grid and in some cases the proteins might start to interact with each other Because you have too high concentration So I'm not saying that's bad where these these protein methods are good But one should be aware that we're not seeing a cell We're seeing a model of something that we created to try to get the protein structure We spoke a little bit I spoke a little bit about various things. So yesterday I spoke about iron cells I tried to remember why I added this It's one of these horrible Cases of a young kid who died about a year those actually about a year ago He had this facial tumor and then they brought him to the US to have surgery and then he died on the operating table The reason I brought up this to remember those ligand gaited iron channels I spoke about it's an extra area We're doing a whole lot of research this happens now on that when you perform surgery that people have it Strayer rare genetic reaction to the anesthetic and then they will literally die on the operating table like one out of 10,000 or so and This is of course based on a slight mutation in the protein I think we have here. We have that protein So it's a some sort of strange mutation in this protein that caused the body to react way too much to the anesthetics or some other protein And that's partly the reason why it's so important to understand exactly how these things work on the molecular level Occasionally if you're really unlucky, it might be sufficient with one amino acid in these structures being incorrect and Things can go that horribly wrong What we're going to talk more about throughout the class, but in particular they want to talk about interactions is Let's see. I think I have a movie there. Yes is that we are going to get back to the sequence I'm not going to talk that much about genes. So for us the genes is mostly can be a DC and t that they then are translated in the ribosome and Transcribed into a protein so you have a sequence of amino acids a long chain but hundred amino acids or so or more and Then these 100 amino acids, they will magically fold up into a protein. We're going to spend some time on that magic later in the class Depending on what this structure looks like that would then say create a binding pocket or create a hole so that Ios can go through a protein and it's really everything here comes from the sequence in your genes But what really primarily creates a particular function such as an iron channel is the three-dimensional structure of a protein That doesn't mean that the sequence is important But this is really the central dogma But in the way, it's very much parallel to the central dogma right because you have the DNA here that leads to the Structure that is created in the ribosome which then ends up folding and actually having some sort of function and binding But let's start with those pesky small amino acids How many of you have well you've looked at amino acids in upper secondary school have any if you looked at amino acids the last few years Okay, one The good thing is that they're not as complicated as physics, but there are some things you need to learn a little bit by heart here So if you look at the the What all these amino acids have in common is that they have and a carbon atom in the middle Which is called an alpha carbon the reason why it's called alpha is just really it's we we enumerate as the first carbon And then you have an amena group with this NH3 when it's free and then a carboxyl group which is COO minus when it's free and Then we typically have a small hydrogen directly on the alpha carbon and then you have some sort of sidechain we By are here. We just means that there is a sort of variant group that could be almost anything actually can be almost Way more than you think it could be Now in your bodies, this is typically limited to one out of 20 chains And those are the so-called essential or alpha amino acids and they're pretty much the only ones we're going to be looking at in this class These this architecture creates one peculiar thing that you might have studied enough for secondary school that if you have four If you have a central atom bound to four different groups different is the key word here if you if you Well, it's hard to draw this in two dimensions But the alpha carbon here if you this is low care oriented as a tetrahedron, so you replace these four groups here If you then take the mirror image of that no matter how you rotate the molecule on the left You can never get it to be the molecule on the right And these is that you might be able to use your fingers or something at least if you have high flexibility To try to understand that otherwise I can bring a small actually I'll try to bring a small molecular building tool kit on Tuesday so that you can have a chance to play around with this yourself So the what we typically say and that you say that the chemical properties of these are the same while Sorry the physical The physical properties might differ for instance in the way they rotate light or so But the chemistry should be the same the density is the same the weight is the same the energy in the bonds is the same That is not strictly true though in biology Because if you look at the DNA The DNA or the entire DNA chain is Like a spiral staircase, right, but it always goes in the same direction You never have a mirror image of the DNA And the same thing with proteins a protein you can virtually never take the mirror image of a protein So the problem is that because all if this was just one molecule You could argue that is chemical properties are the same But when you have hundreds or thousands of these molecules and they start interacting And See one of you come up here and shake my hand come up here. Yes So we are symmetric, right the first approximation. Let's shake hands. That works great. Let's shake hands again Yeah, you see that the problem is that for me as an individual it doesn't matter But when both of us are shaking hands, we have to be compatible, right? Thanks, that's all I needed And the problem here if you now can have a gigantic protein that's going to be an enzyme that binds to the DNA Well, the physics would be the same if I took the mirror image of my protein But then I would also need the mirror image of the DNA, right? So that nature has spontaneously ended up with one-sidedness here. We have no idea why Whether it's evolution or something so that you call this D and L amino acids Exactly, I never remember what this right and left here because that has to do with the exact definition It doesn't really matter But you should know that all chiral amino acids in nature is L With one exception. Glycine doesn't have a side chain It just has a hydrogen and then it has a second hydrogen and because there are not four different groups It can't actually be rotated For what this also gives by a lot most biological molecules they can we yes just a second We can actually use things like light and everything to detect the present We can also detect how much alpha helical structure do you have in it by using cd spectroscopy and working with these physical properties. Yes So whether they're used for anything Yes, oh, I should know this I think are there There might be some very very very very very special organism that has it It's one of us when it comes biologists different of physics to every single ruling biology. There is an exception Nobody's gonna fault you for saying that they never they're never interesting, but yes, they could be there Second you can actually use this imagine that you would like to create a special drug a protein drug Every other these is where to get a protein drug is to inject it. Why do you have to inject protein drugs? What happens if you take a protein drug and swallow it? Yeah, and what is your stomach really good at? Destroying proteins that's pretty much the stomach's job. It takes these big chains and cleaves them into amino acids Yeah, so good luck with that drug. It's gonna be destroyed immediately But if you now create a small protein built from D amino acids How do you think your stomach cleaves proteins by having enzymes right what do those enzymes bind to and recognize? L amino acids So in theory you could use this to create a molecule that in principle has the same physical properties You would still make sure that it binds and does what it should But this is a way that you can actually get a molecule to pretty much escape all those enzymes and get out in the bloodstream You still have to actually get out in the bloodstream which is a second complication So that's many of these things that are not really used in nature are actually used in biotechnology And that's also a reason why you might I'll we have one minute So I'm gonna show this next slide and I'll give you a break out of all these 20 natural amino acids Right occasionally it can be interesting from a biotechnology point of view to use amino acids that are not among these 20 Your body would not use them normally But just now and then it can be used but sometimes actually useful to deliberately have something that is not compatible with biology All of these amino acids you need to know by heart and be able to draw the structure No, I'm just kidding. I can't do that Actually the scare the sad part is I the sad part is the sad part is that I actually could that doesn't mean that you should It's an occupational hazard stuff come with working with these too much. There are abbreviations from this Here's where you should be physicists physicists don't bother about learning things by heart if you can look them up You can look this up I don't I never expect you to know the difference between an R and a W or anything But I think you will gradually in the course pick up some of these abbreviations anyway because long term in this field It will help you but no In practice we don't we rarely use the full names So his is short for histidine argue short for arginine lies a short for lysine if you read the other paper from Fred Sanger yesterday That he actually wrote out the entire sequence of insulin and this three-letter abbreviations That's what everybody did in the 50s. My father is a retired professor of my medical microbiology. He only knows the three-letter code And then we have all these bioinformaticians coming along and determining hundreds of billions of sequences You can't waste three first you can't waste three letters per amino acids Second if you want to put these in tables and compare them It's really irritating when they have different length. So then people come up with this idea It's really you only have 20 amino acids. We have 27 letters in the alphabet. Let's give them one letter codes So that's when you have these very short codes. H is for histidine Well, we can't use a for arginine because a is already used for alanine, right? That's our arginine. Yeah, let's use our there They're a bit stupid glutamine. Well Gly that's already used but cute. I mean well almost cute right cute. I mean They're a bit fun actually that's and it's there is no deeper thought behind this that if we've already used the letter We try to pick something that sounds phonetically similar Yes Tons of things one do you remember those ligand gated iron channels? I saw one of the things that's used when you jump between cells is an amino acid called the GABA gamma Immune butyric acid so instead of a hydrogen it has a second CH3 group So that's used as a neurotransmitter. It is this an amino acid that's only used in isolation But you will never see it in a protein. You could create it artificially in a protein, but nature doesn't use it Why I don't know we don't have that biology. I've already stolen a minute too much of your time So let's go and have some coffee and then I'm gonna continue after the break So one thing while I remember it I realized I kind of skipped through transcription and translation very quickly and I might even better stop It's sloppy with words So just to make sure that we don't look with up with 25 of you getting that wrong on the exam What is the difference between transcription and translation? Yes, and it's that might be the hardest one that's hardest understand just by reading the words maybe Maybe an analogy could be a transcripts of your grades or something right you're writing, but there's still letters While translation that's translating it to something completely different. That is a protein sequence is that or If nothing else works, you might have to be biological and learn it by heart I'm gonna continue with amino acids There are a ton of ways to represent amino acids and the reason why we why you will frequently see this different Blocks or so is that rather than worrying about the exact confirm it the exact details and counting the number of carbons In particular for proteins. We try to classify them. So for instance, are they small and Do they have a small hydrophobic side chains or are they say electrically charged arginine histidine and lysine can be charged positively and as part of a glutamic acid can have negative charges Or are they amino acids that are polar but not charged side chains? So there are a bunch of different ways to classify the amino acids And that's going to turn out to be very common when we look at podium structures and evolution if you have an arginine positive charge The likelihood that that can be mutated into lysine, which is another positive charge. That's quite high The likelihood that you will have an arginine positive charge Mutate to say a tryptophan with there's a large bulky hydrophobic amino acid. That's very low So the compatibility here matters more than the individual amino acid and that's also how evolution works And you cannot have pretty much as many ways of dividing these as you want You can make decide to classify them based on size whether they're smaller large polar or non-polar hydrophobic or hydrophilic Whether they're positively or negatively charged Aliphatic that is just linear carbon chains or aromatic with rings Your guess is as good as mine. We're gonna come back to this So you don't have to don't worry too much about knowing all these ways by heart But you need to be aware that there are different ways of classifying them. I Already spoke a bit about the way we polymerize these chains and in particular that this leads to polypeptides That are there are polymers, but in contrast to plastics They're not homo polymers, but hetero polymers that's different and this is going to be really really really cool Because this is the way proteins are completely different from plastic and I also spoke about these yellow hydrogen bonds and The way the way they form is actually it's fairly complicated You'd usually need enzymes to help you do this But it has to do with electrons resonating here so that you keep up a water and then instead you create this peptide bond So here too They are more stable in the form when the peptide bond has formed But to go from the separate amino acid to forming this there is a fairly high energy barrier between these that we need to get over and that's usually where you have an enzyme helping you and Well, what happens in the stomach is that we do the exact opposite process Then we start from here and then we use another time to cleave it to have separate amino acids instead These peptide bonds though and the peptide bond is completely planar And by plane I mean it's stiff it can you can't rotate around that bond Exactly why is because you have an electron resonance all over the bond But that's quantum chemistry that we're not going to care so much, but you can't rotate around that bond On the other hand the bond before each alpha-carbon and after each alpha-carbon their normal bonds You can literally rotate as much as you want between them And if we now start looking at an entire protein, sorry that was in the wrong order a long chain here How many you have an insane number of degrees of freedom here right in principle every atom has three degrees of freedom X Y Z So there's probably like could it be 150 200 degrees of freedom Imagine if physics that we worry about one electron. This is an insanely complicated many but many body system and A biologist wouldn't even call it a protein. It's just a small polypeptide So here to be able to understand these molecules We need to find simpler ways of describing this you can't use X Y Z You can't computers will use X Y Z for all atoms, but there is no way our brains would be able to understand this We try to describe this with X Y Z coordinates And then we have many things for instance that hydrogen we know that that hydrogen will not suddenly end up there, right? It's gonna stay bound to the to the carbon On the other hand things can rotate around that bond so that those three hydrogens can start spinning around like a small propeller They do how important do you think that is for the structure? Well in one way this one degree of freedom, but it's not a particularly important degree of freedom So we would like to find a way to basically ignore the degrees of freedom that don't matter so much for the protein But can we focus on the degrees of freedom that will have a large impact on the way this protein looks? And that is exactly where we have this bond if you take this bond there either before the alpha carbon or after the alpha carbon If you're rotating that bond this entire part of the chain following is gonna Swoosh around in space, right? So when you change the conformations around these bonds, it's gonna have a profound impact on the entire structure and Even if you have a much more complicated side chain here, let's say we had a Arginine or something pointing out there even if you rotate all bonds in that arginine It will only have a local effect on how the side chain is pointing But when we're changing the bonds along this main chain or the backbone, that's gonna change the entire structure of the protein So already now instead of thinking 200 degrees of freedom This might be 1 2 3 4 5 6 7 8 9 10 11 12 15 or so You might have in the ballpark of 15 amino acids and each amino acid has two bonds it can rotate about So we have at most 30 degrees of freedom that is important That's a remarkable simplification We need it gets complicated to keep talking about the bonds before the alpha carbon and the bond after the alpha carbon So we need to define this a way that for historical reasons they are called Phi and Psi bonds By far the easiest way Do you think I know this definition by heart? I don't but I'm good at drawing amino acids So I know that we start with a nitrogen And then you do a six-hack pattern then we need an alpha carbon and After the alpha carbon there is a normal carbon with this oxygen bond and that's the entire mean it Well, you can if you really care you can draw the R group we lazy I'm not gonna draw hydrogens, and then it's the next amino acid. So then it's nitrogen alpha carbon R group and They will just continue that way The second I've done that. I only need to know Phi bond is the bond before the alpha carbon so that is a Phi bond and The one after the alpha carbon is a Psi bond And at this point I can just say okay Phi so that must be defined by carbon nitrogen alpha carbon carbon And that's probably how I drew this slide to So here you can do it any way you want either learn it by heart or learn how to draw amino acids I would strongly suggest you learn to draw amino acids because that takes you two minutes And it's gonna save you a ton of information a ton of work later on in the class So the way we did effectively we have these amino acids moving as if there were small local planes here that are fairly rigid Because that the rigidity here corresponds to the peptide bond the peptide bond can't rotate Yes, sure This is the peptide bond Here I was sloppy because we really should have all those hydrogens too, but I'm too lazy to draw that So that it's a peptide well that those oxygen carbon nitrogen hydrogen in principle It's only the bond between those two, but all four atoms are actually involved in it. I think I Yes, I think I might have that on the next slide here, so you have names for these bonds Occasionally we call the peptide bond an omega bond That is going to be likely the first and last time in your life. You hear that expression Nobody cares about the poor omega bond because a bond that pretty much never ever changes It's not worth our attention Remember what I said about biology and exceptions It would have been great if this bond never ever changes if it was always trans then I could literally forget about it But unfortunately we are in biology and in proline is a bit of a special amino acid So in proline's it can occasionally be cis that it's rather than being trans you can have the bond going back So that's the alpha carbon. Ah, let's see that you would have the alpha carbon there and the hydrogen there But that's an exception. I don't think I'm not I don't really think I'm gonna ask you about omega bonds ever but if you ever see an omega is the peptide bond people think about and Then just to make this complete for a small chain like this is completely irrelevant with all the side tears are but if you have a longer side chain occasionally we want to talk about the side tears and how are they oriented and We like Greek letters. If we started with Greek letters, we might as well continue. So they're called Chi But we're not gonna somewhere there our imagination runs out Different amino acids have different lengths of the side chain. So rather than coming up with very creative names We just call the first one k1 chi 1 the second one is chi 2 third one is chi 3 So as you go further out in the side chain, you just increase the number and pretty much nobody cares about anything except Chi 1 because Chi 1 will Determine how the entire if you this is a long side chain Chi 1 will determine how the entire side chain is flipping While Chi 3 will only determine how a small CH3 group or something is rotating at the very end and if So just great. It's the large motions that matter and that's why we primarily care in that fine psi so this Already now we can start thinking a little bit more like physicists and start to think of simplifications that yes Oh, so pro means proline. So that's this man is me and abbreviations This is I did so proline is a bit of a special amino acid We're gonna come back to that later on the course But so for normally if you had to guess the peptide bond will always be in a trans confirmation This is that these keep swapping sites, right? And that's how I always draw them Occasionally for proline, it's roughly 50 50 for proline Sometimes this bond can have another orientation Don't worry about that for now. We're gonna come back to that when we talk about podium structure in particular It's just that I can't say that it's always trans because strictly It's not correct and without really having talked much about a biology and what the sequence of these proteins are now We can we can start to What I love in physics is this concept of Gedanken experiment that we can think about how many how many ways can a protein be placed in? So if you take these fine psi torsions and here we need to Remember that I spoke yesterday about the power of models if you're gonna model things you need to learn to do Drastic oversimplification because otherwise they become too complicated So instead of 200 degrees of freedom We need to have a system that's simple enough that I can think about it with even without ever writing something down So if I take these fine psi torsions and say like well to actually say that it's different They have to move substantially. Let's say 10 degrees or so That means that it's each of these torsions can have roughly 36 states around the entire bond 360 divided by 10 If say 10 if you want to it's the order of magnitudes that matter here And that means that each amino acid can have 36 multiplied by 36 different ways of placing the chain because there are two of them Right and if you then have 100 amino acids Well, we should take that number and multiply that with another hundred numbers like that That's going to be 36 raised to the power of 2 raised to the power of 100 and that's roughly 10 to the power of 308 That's an interesting number I picked there was the reason I picked this number and if you want to do this You can try to actually your calculators might work all calculators This is not going to work with or if you try to do this in a small Python or C program The computer is going to give you not a number result because this number is too large to represent in double precision floating point on a computer So even for a hundred mere residue the number of confirmations here are pretty much more than the number of atoms There are in the universe Forget about it. There is no way the protein will be able to explore all those confirmations and yet We know from x-ray crystallography and everything right that there is one state of a protein that is the stable one And this creates a fun either because somehow nature we know that this based on physics physics manages to find one out of 10 to the power of 300 But there is no way it can test all of those So how does this work? I will come back to that later on but that Cyrus Leventhal was the first to formulate this as it's obviously a paradox the stuff that I talked about proliness related to this that When these proteins move actually they can't have 36 different states So most bonds tend to be either in the transform so that if you look if you forget about the small atoms or the atoms That are all part of the chain the main chain here if you trace out the big chain Either they go in a way they they keep stretching out So that's a transform so that those two yellow purple atoms are as far as a far away from each other as possible Or you could have alpha the alpha carbon carbon nitrogen and then another alpha carbon here So you could go back on the same side. So that would be a cis confirmation while this is a trans confirmation In general trans is much better because you stretch it out and have more room But it would be pretty boring with chains that are just long and straight And as we're going to see next week, there are other problems with that So occasionally you can have them flip over and actually be on the same side, which is going to be important Oh, yes, we even have an example of here here have the cis state And here you can also see why the cis state is normally not particularly good Right because you're going to start seeing the side chains will usually start bumping into each other, which is bad But proline is a bit special because proline doesn't really have a side chain and it's also very different around this nitrogen So for proline, you can actually put the residue here in trans and it's not really flashing a lot at least So due to the way proline looks Formally pros in that proline is actually not if you're going to be really really picky about this in chemistry Proline is not an amino acid but an amino acid But I'm never gonna ask you I'm I will never say that is wrong to call proline and amino acid So proline is the one exception out of 20 that can occasionally be in trans for the peptide bond No, sorry insist for the peptide bond, but even then remember our small protein 15 residues Well, 50 miles by 30 degrees of freedom in principle. This is a 30 dimensional space. I Not sure about you, but I can't think in 30 dimensions. Well, so we need even simpler ways of trying to represent these And Ramachandran came up with a pretty smart way of doing this that if you only have They're not 30 completely separate variables, right? So you have many amino acids and for each amino acid you have a Phi and a Psi bond So it's kind of the same bonds, but for different amino acids in the chain So what if we take just do a two dimension a single two dimensional plot and we brought the Phi bond here and the Psi bond there and then for every amino acid let's put a Black square where you have that particular amino acids and then the next amino acid put a black square and then a black square or a red square So it turns out that in theory they're all separate, but the amino acids tend to cluster So there are lots of amino acids that have Phi Psi bonds in that region There are some that have well a lot of them that have it up here and a few of them that have here and pretty much No, I have their values here Why is that? Yes So they're the scale in the diagram here Yeah, that's just the the angle of those two bonds the Phi and this is Phi and that's Psi So the reason why these readers are almost forbidden. It's that they would correspond to sidechains bumping into each other Again, these are large molecules, right? Then if you try in general there will be some things where even the even the backbone will bump into each other and Remember that we spoke about I was generous as I said maybe each of these variables Phi and Psi can have 36 different values 36 to the power of 2 that's about a thousand forget about a thousand We're talking about three regions so that you can have you can be that blue region or the blue region up there Or maybe the region out there So now we've gone from one thousand to for each amino acid rather than having a thousand states for it You pretty much have three states for it Again, it's a drastic oversimplification, but if we want to understand it starts being pretty useful And these are called the Ramachandran diagrams what you have it there on the right They are slightly different for different amino acids the general amino acid up there on the left looks roughly the way I showed you Proline There are almost no regions where the polling can be and that has to do with the strange ring in the polling It simply doesn't have a whole lot of degree of freedom if you try to change the Phi bond in the polling You can't do that because that sits in the ring So the only thing the polling can do it can have slightly two slightly different confirmations for the Psi and that's it Glycine on the other hand probably has at least four big areas it can be in You remember how Glycine looks There is a correlation here Hmm or to make it even simpler Glycine doesn't really have any side chain And if the big problem that what if it was limited our freedom here was that the side chains were clashing right if an amino acid Doesn't have a side chain. It's going to be small and flexible So if you look at the general amino acids versus glycine Which one do you think occurs most frequently in areas where we need to have a small turn in the protein or a small loop or something Where it needs to be flexible Glycine so nature uses there is a very strong correlation between these physiochemical properties of amino acids and where nature ends up using them in proteins And that's that's reason I even bring gradually bringing this up Proline on the other hand is special proline can't really it's very difficult for polling to occur inside healers or so because it would break things up and Some of you might know a little bit about protein structure. I'm going to come back to protein structure But we've given these names so we call this region an alpha helix and we call this real in a beta sheet for now You don't need to know what there But friend of order would say something here that nature physics is symmetric, right? The laws of physics are symmetric So why do you have a particular helix here and a beta sheet? Why don't if physics is symmetric shouldn't this diagram be symmetric? What is it that causes this diagram not to be symmetric? Exactly right where we have a hand we have a built-in Handedness in the amino acids and that will keep propagating to all the structures on all levels So so speak of this in physics terms. You've had a spontaneous symmetry break and Eventually this is what's what's going to create these gigantic structures So the blue parts here is an alpha helix you have alpha helix is in this transmembrane part to and up here You have lots of beta sheets But we're going to come back to later what that implies So all right now I kind of hinted that nature will use this diversity And we can use different proteins in different parts partly to create different structure and occasionally also because you might want If you're going to bind a negatively charged molecule, it might help a lot to have a positively charged amino acid, right? So on the one hand we want to use the diversity to build the protein itself and On the second we might want to use this amino acids to make to give the protein a particular function to interact with something else and This were two very very famous Discoveries in the mid 1900s Christian Ampelsen came up with a almost a postulate that From all these hundreds of amino acids no matter how complicated this is he postulated and actually proved that Proteins appear to adopt the structure that corresponds to the global minimum in free energy We're going to come back and talk about what free energy here means next week It's a it's one of those profound results There's a bit of remember what I said it's important to understand how Science was discovered today. This is so obvious that it hardly even goes in the textbooks But it's obvious that proteins will have obtained the energy minimum This was so non-obvious that he got the Nobel Prize for the discovery and a particular being able to prove it And the danger there is that some of the greatest discoveries Because when they make it into the textbook they eventually become obvious it wasn't obvious a hundred years ago and Then we had Cyrus Leventhal on the other hand formulating this as I just mentioned that One way or another this is obviously true, but we can't do that by randomly searching everything We're gonna come back to that in the course and try to understand why We already spoke a little bit about the size of the proteins But this just to recap that that you can go from say 50 or 100 amino acids up to 20 30,000 amino acids There are some gigantic proteins here and the way we usually measure their weight you measure the weight in atomic units So it's called dolphins usually we're not going to talk that much about it But if you see the Dalton or kilodolton number that is really the molecular weight in the interest of time I'm not gonna skip the case study when I go through titan, but this is an example of how this works in the body It's a very small protein that is somewhat Flexible so that you can pull it and when you pull it it will actually it acts like a rubber band and it will contract again And this is when what the body uses to express this and then you have gigantic long chains with many many many of these Domains and then these domains are gradually built up into larger filaments and eventually these filaments is what makes up your Fibrils that does turn into muscle fibers So every time you're twisting a muscle you then have billions of these proteins undergoing this contraction or relaxing based Steered entirely by your nerve signals, but we're gonna be looking at a ton of proteins later in the class I won't spend too much time on that one There are three large classes of the proteins that we're gonna start studying next week first fibrous proteins Boring things like hair nails fingers that they're large and they're important structure components of your body And then we have these water soluble proteins that are called globular that are very much the worst horses And then that's the love story of my life which is membrane proteins and you might there might be a slight bias in the course I have to confess These are so beautiful because they're they're producing physical processes. They're transporting things They're open. They're the windows and doors of yourselves. They're so intimately collected to biology and that's why we love You can study Proteins actually fairly complicated that when you start studying them because you can study produce like a study any other compound like a salt or so So on the one hand a single domain like that is like any other it is a molecule, but it's a large molecule It's hard. It's almost like a sphere that it's not trivial to crystallize it But you can crystallize it But eventually when you build up these really really large assemblies, right then they become like skin or so and they can become soft There is a remarkable diversity not just on the amino acid level, but also how we build these assemblies and create different functionality in our bodies and Not only are they occasionally a bit softer But many of these proteins they change shape slightly when they work an iron channel for instance It has to open or close otherwise. We can't really control how many aisles flow into our nerves and Most of these proteins tend to be in either or state here They're either folded or not folded an iron channel is either open or closed sure when it's actually Closing it has to go under through a transition But they tend to be reasonably well-defined states and then they move between these states rather than moving all over the playing field all The time and why we don't really know yet, but we're gonna study that in the class together So most of these things seem to be almost like a light bulb that it either works or it doesn't work And we're gonna continue with this theme of trying to drastically simplify things Remember that I talked we talked about Ramos Shandong's right? If you look at the structure if we think like whether we're physicists or biologists It's nice to think in terms of recurring patterns So we had 20 amino acids and technically one molecule is not the same as another molecule But it's the same 20 amino acids and that's really this is a sequence But since that is where we're starting it kind of makes sense to at least call that the primary structure a primary sequence You can use either word there, but that is really because each of these amino acid has some internal structure It's just that it's pointless to draw it So that is what we're starting from and it's a single chain The next step has to do with this regular patterns that we saw in the Ramos Shandong diagram That appears to be two big regions where we spend a lot of time and that's what we call the secondary structure We're gonna be looking a bit more about secondary structure in a second. I Already showed you some proteins at some point You have many of these secondary structure elements building up even larger proteins and the neat thing here here We kept completely forget about atoms. I even forget about amino acids on this level So here I'm like oh, there's a helix there and a second helix there and a third helix there So I deliberately choose to take a step back and look at the larger scale If I look at an entire ribosome I can't even think about helices and a ribosome is an assembly of roughly 60 proteins So then I need six oh wait a second. What is the protein responsible for binding the RNA? What is the protein responsible for pushing things out? So then you might go all the way and start to looking at assemblies of multiple subunits So the subunits is what's called tertiary or third structure while this multiple subunits would be called a quaternary structure So the fourth order structure and we're gonna be moving up and down the scale So on the very lowest level we want the atomic detail and on the high level we focus on the principles But we need to start somewhere it would be strange to start at the highest level and assume that know the other So we went through the amino acids. So let's look a little bit at the secondary structure. So the secondary structure By far the most well-populated ones are the two on the left here that you have an alpha helix and a beta sheet And they look like this. So the alpha helix is literally a helix It's not a double helix like DNA, but just a single helix, but it's definitely wound up and then the beta sheets That's when we have the amino acid stretched out instead. So in the beta sheet you literally have Yes, like a drood them here. They the side chains are on alternating sides and it's at stretched out as possible So if you just look at them there, which one do you think is most stable? Why? So that's what which one do you think is most flexible you're right actually, but The beta sheet is much more flexible right because that there aren't really any hydrogen bonds restricting its motion While the alpha helix has a lot of hydrogen bonds. I haven't draw them here, but this is somehow it tighter Sometimes that's good. Sometimes that is good nature appears to use both of them So if we start by looking a little bit about the helix is there are in principle There are different ways of forming a helix, but in practice we virtually only see one So what really stabilizes the alpha helix is that once you place the amino acids here The oxygen on one amino acid that one Can hydrogen bond to the hydrogen a few amino acids later? And it's also so it's the same hydrogen axis and both of these sit in the peptide bond So that oxygen will form a hydrogen bond to the hydrogen for amino acids apart So By far the most common one is that you form a bond residue number I hydrogen bonds to rescue for four units away. So I to I plus four in Principle it is possible to twist it a bit harder that it's only three then it's a three ten helix Or you can let it lose and form a five or pi helix Forget about those the one you need to know is that a normal helix makes a hydrogen bond to a residue for amino acids part I can show you a couple. Oh, yes here. You see them top down So on the top left you have the alpha helix there You can almost you might almost be able to see that it's a fairly balanced It's a real it's a relaxed molecule all the sides are not all the backbone angles and everything there are nice stable minimal conditions So it's going to be a low energy low free energy state where the entire helix is quite happy If you form this three ten helix, that's kind of like taking a spring but twisting it really hard Sure, it is possible to physically put it that way But the entire molecule is going to be a bit strained in particular here. There's very little space here So the atoms are really going to be colliding here And if you do the opposite we take the spring and try to tear it apart a bit Then you get this pie helix and here you the point here You were essentially going to have vacuum on the inside here and nature reports vacuum too So occasionally we absolutely need those structures and then nature will form it But they're so rare that 99% or 95% of the helix is at least they're going to be alpha helix is and Now even if the helix is 20 residues here rather than thinking we no longer thinking about the atoms We forgot that a long time ago. Did you hear me talk about the amino acids here? No, did you hear me talk about the ramachandran triangles? No So now I can think about an entire building block of 20 amino acids as an alpha helix And that is now one brick that we can use to build larger scale structures So by approaching this hierarchically, that's the only way we will be able to understand the structures Although they all consist of atoms of course Then on the other hand we have this beta sheet The beta sheet is not going to be stable at all because it has absolutely no hydrogen bonds or anything to stabilize it So that sounds a bit stupid. Why on earth would the beta sheet ever occur then? It is more flexible, but if it was only a matter of flexibility We could just have a random orientation of the amino acids So if you have one of these and then I add a second copy a second sheet They can hide it and bond to each other right so if you pair put multiple of these together they can form structure So if you look at the beta sheets, they're a bit they're a bit difficult that You only have helices sheets and turn but the beta sheets are slightly in that they form Instead of forming a local high the helix is always local the helix forms a hydrogen bond to another rescue That is just for residues apart The beta sheet will form hydrogen bonds to something that could be hundred residues away So this is more of a global structure, but it can be stable And the reason we see both helices and sheets is that again these hydrogen bonds tend to be paired in them If you only had one when we only have one of these beta things I usually call it a single strand and if I only have one of them That would likely just become flexible So unless it forms these hydrogen bonds it's not going to stay that way and if it's neither helices or sheets We frequently call them turns or random coil. That's less defined structure. Yes Yeah, so that I'll come back to that in a second turn is actually random coil if there is no structure at all turn is relatively well defined That's like taking a small u-turn Because beta sheets are a bit more complicated And if you I mentioned that we're going to put them together, right? But are you going to make them all parallel or you're going to make them anti-parallel? Are you going to go up up up or up down up down up down up down? There are advantages to both it depends on where you're going The advantage is going up down up down is that in theory you could start here And then have a small loop and then you go down and then you have a small loop and then you go up So if you literally only want beta sheet, this is very efficient to make them anti-parallel Or you can make them you can have a helix that goes up And then you go away and you go out of the screen here Maybe have a helix or something and then we come back a while later And then we go up again and then we might go out and then we have a second helix and then we go up again So this is a bit more of a complicated structure, but both of them are perfectly stable and fine They will you will see that they have slightly different properties and they're hydrogen bonds here, but that's a bit more complicated So at this point you're likely thinking that why on earth beta sheets? They just look complicated They're not all as beautiful and simple as alpha helices and Nothing could be more wrong. These are beautiful. In many ways. These are simpler structures They're easier building blocks than the alpha helices Because by the time you put three or four of these together First you're gonna form you're literally gonna form a small sheet like a piece of paper or something Now when you have this sheet in particular for anti-parallel ones, you're gonna have the fun remember that when they draw this here you had This side's in is pointing up this site in is pointing down the next one is going to point up and the fourth one is going to point down So every second site in will appear on opposite sides, right? And if you have a very large sheet that's going to be true for each of these beta strands So we now create a structure where let's say I make all the odd amino acids hydrophobic So that they're don't like water. I like make all the even amino acids hydrophilic. So they like water You will now have a sheet that is hydrophilic on one side, but hydrophobic on the other, right? Are you with me? One such sheet is going to be fairly useful less But if you put take two such sheets together So we have a top-level sheet here and then a bottom-level sheet under it If you make both of these say hydrophilic on the outside It's going to be water soluble because it likes water on the outside, but we can make the inside hydrophobic So what could you ever use that for? Well, this is a small protein called fatty acid binding protein So your body needs transports fat all the time, but fat is not water soluble So this is a small protein though that is fat soluble on the inside So we can bind the fatty acids there and once the fatty acid is bound here This entire protein can then transport the fatty acid in water because it's hydrophilic on the outside So it's an example the reason I bring this up fairly early I think it's a beautiful example of this by by assembling a particular order of amino acids We can create biological function or in this case is chemical function. It's a very simple chemical function really But try to achieve that with a normal traditional molecule or physics Then it's a fairly complicated procedure to create something that is both water and fat soluble How do you think people discovered all these secondary structures who came up with the idea? Because it's pretty cool. I in one way they don't exist It's just a mind exercise. The only thing that exists are the atoms Even the concept of ML and amino acid you could argue that exists is just a molecule But these are just atoms with XYZ coordinates The only sense and with a beta sheet exists is that this is our Conceptual model to try to understand the function on a higher level, right? But the physics here is created by the individual atoms Then we choose to represent this on a higher level to understand it better because if I just if I just showed you a blob here With 600 atoms, I bet that you wouldn't see the point of it So how did I spoke yesterday about how we determined structures? x-ray, right? And nothing could be more wrong. This is really cool This was postulated theoretically based on how amino acid looked before the structures were available by Linus Pauling and and So yesterday I was kind of ridiculing Linus a bit about his DNA structure This is of course the reason why it's possibly the largest genius in life science of the 90s and 1900s They there's a series of eight papers in PNAS And I think I put most of them at least the review is available on canvas They came up with these suggestions for all this different helices and the beta sheets and everything in 51 which was roughly 10 years before we got the first protein structures and then when we got the structures They found that this is the way they'd actually do look internally So that had they done this afterwards? It was just a bit a neat model when you do this 10 years before you know the answer That's when that's why you're a Nobel Orient The review paper is well worth reading So if you look at this Ramos Shandon diagram The large part there is really the alpha helix, which is right-handed and the reason we don't have a left-handed helix has to do with the chirality the handedness of the individual amino acids and This way it gets complicated then individual amino acids. We call them left-handed, but the alpha helix is right-handed There is no pattern here. The alpha helix is right-handed because it goes that way. That's the way you have a normal screw, right? The amino acids handed this has nothing to do with the helix is handed this If you just count this it also turns out that each amino acids twist roughly 100 degrees in an alpha helix So there are exactly 3.6 amino acids or residues per turn I will use amino acid that residue interchangeably Residue is really any individual molecule in a long chain The DNA bases are residues in the DNA chain in a way But when you just say residue people usually will assume that you mean amino acid residues There are some other peculiar properties that if you look inside the alpha helix is all these peptide bonds They're quite they're not charged But the oxygen here likes to attract electrodes and the hydrogen there is fairly happy to give up its electrodes So what will happen in each peptide bond is you're gonna have more negative charge up towards the oxygen here And much more positive charge down towards the hydrogen Now the net charge here is still plus minus zero of that entire group But this creates a dipole in the cells that you have the positive and negative parts Systematically offset from each other and you can even calculate roughly how large that dipole is and it's a few device In general that wouldn't matter because in solution do they would point in random directions But what happens in the helix due to these hydrogen bonds? First you have one peptide bond there. It's pointing in that direction You have a second peptide bond there that points in that direction. Do you see all these dipoles will line up perfectly? So even in a small helix you can have 20 dipoles lining up perfectly So that the entire helix will actually have a fairly large dipole pointing in one direction And now the point here is not to go through dipoles in physics, but this is another example how nature uses this Those ion channels I showed you this was discovered some 20 years ago These it's called a KCSA and the first day that this is an ion channel that conducts K ions or potassium ions So this and I and it's used this one particular one is bacterial But they're used in your nervous system too and we want them to be selective so that they conduct potassium, but not sodium Is that a problem? I don't expect you to be chemistry undergraduate. So what's the difference between potassium and sodium? I'm going to draw these two ions for you. So I'm asking you to create a hole and Let's first Here is the potassium ion and here is the sodium ion So please create a hole let's let's through this But not this that's a bit complicated, right? Because the first approximation anything that let's do that will obviously let through that one too Nature never makes a mistake here. Actually it does one in roughly hundred billion ions or so It's insanely efficient and the way this is generated is that In this channel you have a bunch of helices lining up. They're literally almost as if it was an accelerator They're pointing inwards. So you have lots of helis dipoles pointing into almost a binding site in here So what happens when this ion comes in ions don't exist in isolation remember what I said yesterday water water water So any ion will have a layer of water around its binding water And they will let you particularly the positive ions are going to be conducting Attracting the negative the oxygen and this will literally be bound water to the ion And if you know your physics if you have a very small radius here The oxygens will be able to get closer to the positive charts and so that a small ion is going to bind its water harder So what nature has done here is that you have a very very narrow pore here So to actually to go through this pore We have to start by stripping the water from the ion And that's what these helices do with their dipoles They're creating a binding site where the water I will let go of the water and the cool thing that the large ion It's larger. So it's not binding its water as hard. So this one would be able to get rid of its water and go through this one on the other hand won't and Once you add in the water here that small ion is going to look something like that So together with the crystal water, it's actually sorry with the ion water. It's actually much larger The reason for bringing this up is again, you see how nature uses fairly simple physical principles to achieve its functionality This accuracy the accuracy here is far better than any computer. You can imagine These biological systems are like their orders of magnitude rarer that they make errors than the best computer in the world Which is also fascinating, right? Because they're very floppy and simple rules and you would imagine that it would go wrong all the time Which occasionally does but there are lots of error correction mechanisms Let's see. Oh, yes. This is an old movie. I got from Mike Levitt We're gonna we're gonna end I will bring I will bring this up tomorrow is that This is some of the first this I think it was the first computer graphics ever created And actually given that we're already and I will wait for this tomorrow But many of these people that you see when you see them We have you frequently see them when they got them when they're senior and they're all dressed in suit and ties So Cyrus level he was actually a computer geek sitting down in the computer lab all the time and writing all these programs They're gonna speak a little bit about that tomorrow how we started to visualize proteins and everything and in particular what Cyrus did here So there are like some 10 slides or so at the end that I will skip But that's perfectly fine and I'll talk about those tomorrow is