RNA Self-assembly: Cooperation at the Origins of Life | Niles Lehman

well I'll try to make this as painless as I can so that you can head to the pool hall but I do want to thank Pierre for inviting me to give this talk the title of my talk is RNA self-assembly cooperation at the origins of life we saw this slide in the previous talk from Irene which is a depiction of the timeline of the history of the earth and we are focusing in these last waning hours of the of today's session on this this time frame that centers around 4.0 billion years ago and as Irene and Nigel before her mentioned this is this is our best guess for the time at which life originated life originated on the planet Earth and when we talk about the origin of life we have to try to define life in some way and so NASA's working definition of life is that it's a self-sustaining chemical system capable of Darwinian evolution and I underline the word self because that's sort of that the theme of my my talk today what do we mean by self and if we believe this definition we sort of have to come to grips with the fact that the origin of life is really the origin of self replication but again there's that term self so it's easy to figure out what self is in some kind of living systems such as my favorite organism here the southern hairy nosed wombat it's very clear what a self is and what a non-self is and it's probably reasonably clear in a bacterium okay and it may have may may or may not be clear in a bull Vox but you have a sense of what an organism might be but when we get down to the molecules when we get down the molecules that that led to life on the Earth from a lifeless planet that had simply chemicals on it the notion of self becomes a little bit fuzzy and I want to sort of challenge you to think about that and hopefully we'll make some inroads into that today now here's a molecule that a lot of people have been studying this is a an RNA molecule a ribosome a catalytic RNA that as rudimentary capability to make a partial copy of its self this is the so called RNA replicase ribozyme it was ultimately discovered selected for the in date bartels lab and this molecule is about 200 nucleotides long from from stem to stern and it can make a copy of about half of itself ish okay and it does this by template directed replication very much akin to some of the chemical reactions that Irene was just talking to us about okay now this this type of a replication is intrinsically to use this word against self selfish because it is making a copy of itself okay but this the theme of this conference is not so much about selfishness it's more about cooperation so I want to try to shift our thinking from selfishness to cooperation and I hope you don't throw too many bananas at me because the way I think about this transition or this this origin of selfishness and cooperation I got a lot of inspiration from one of my favorite figures that I've ever seen ever published and this was a figure that was published by Andy Ellington in 2001 this was a news and views that accompanied the description of this original replicates ribozyme and Andy had the insight to say well here's our replicates ribozyme alright and that ultimately led to the kind of biology we see today but that replicates ribozyme had to itself had had some sort of ancestry and that ancestry was rooted in simpler replication systems okay and so I sort of took the step immediately prior to the replication ribosome as something that I would like to try to emulate in the laboratory and he called it an assembly of ribosomes via tag sequences and as we go along today I think you will see that that the system I'll show you has a lot in common with that particular stage of luck that I'm really keen on this stage is that when I get up in the morning and I look in the mirror I look quite a bit like that so I figured that you know that was my calling I should study this stage of life but but I want to point out that if you look at these sorts of systems that predate in Andy's mind this replicates ribozyme they all have this this commonality that they're their collection of molecules a collection of molecules that aggregate li encode replication all right and so to me that seems very much like cooperation so I'm going to pitch this notion that cooperation was actually ancestral to selfish replication and if you think about if you think about it all the way down to the chemical level and you know at some sense I'm a chemist the cooperation can all go all the way back to maybe the origin of the first life like molecules here's a here's an adenine molecule okay obviously one of the critical components of our genetic system today and one a row showed in 1961 that you could take hydrogen cyanide and polymerize it and it makes adenine in a small yield but it still makes it and basically in Mike warped mind I'm taking I'm taking hydrogen five equivalents of hydrogen cyanide and they are cooperating to make a new molecule okay so this is this is what you might think of as chemical cooperation now when you think about self replication in the origin of life you have to think more in biological terms all right and to a chemist into a chemist the feature of molecules that makes them lifelike is this feature that we call Auto catalysis and auto catalysis is where the product of a reaction feeds back and helps to catalyze its own synthesis okay and so if we have this a plus B going to C and C feeds back to catalyze its synthesis from A to B that is an auto catalytic reaction okay and in some sense C is somewhat selfish in this regard right because it's catalyzing its own self replication so what I want to do is I want to say well when did this selfish type of auto catalysis become strict maybe there was something that preceded this selfish type of auto catalysis that's more like I'm showing in the bottom panel in the slide much more of a cooperative system where we have two chemical systems a plus B going to C and a-prime and b-prime going to C Prime and if these two pathways can interact such that C prime catalyzes the synthesis synthesis of c and c catalyzes the synthesis of c prime talk about a tongue-twister then we start to develop a certain amount of rudimentary chemical cooperativity that may have bearings on on life and so maybe we can even take this notion of two systems interacting to more than two alright this is a drawing that appeared in one of manfred eigen and peter Schuster's papers in the 1970s this is the notion of three systems interacting cooperating to establish their own replication okay and maybe we can go beyond three into a network of interacting pathways and this sort of cooperative network was envisioned by Stu Kaufman in the 1990s and he called this an autocatalytic set all right the notion of an autocatalytic said is that you don't have one you don't have to you don't have three but you have a whole collection of genotypes that are collaborating in some sort of coalition to suddenly crystallize a set of reactions that ensures the replication of the whole set all right so I ask the question can we create a cooperative network of this type in the laboratory using catalytic RNAs all right because I want to sort of emulate this this putative type of replication that precedes self-replication and I wouldn't be standing here if the answer to this question weren't yes and what I'm going to show you real quickly is hopefully real quickly that by using catalytic RNAs that use the process the property of recombination which I will stress over and over again we can piece together copies of themselves to make a cooperative network all right so cooperation and recombination are the themes of this talk so what is recombination what is recombination to a chemist well recombination to a chemist is where you take blocks of genetic information and you shuffle them around to make new versions of sequences okay so on the left here we have a depiction of what I would call recombination where we have two strands a green strand and a red strand and these two strands are going to come together well brawn is going to be broken another Braund is going to be made at the same time and we're going to form two new strands in this case a short red strand and a chimeric blue sorry green red strand okay and you can contrast this to the the other kind of replication that people are more familiar with this notion of polymerization where you have a template in this case the green strand and then we make a copy on it by adding single red components to it one at a time okay it's my premise that this type of mechanism this recombination had utility in the origins of life because it's chemically simple and we can make very large jumps in sequence space in a very short period of time the other nice thing about recombination is that nature does this all right nature's very happy to do recombination reactions in fact the very first catalytic RNA that was ever discovered is in fact a recombining RNA group one introns which were discovered by Tom Cech actually perform recombination reactions okay they are self splicing RNA so the black portion on this diagram is the the catalytic RNA and it splices itself out of a nascent transcript in two sequential transesterification reactions all right in the first in the first step we have a G nucleotide attacking at one spot all right catalyzing a bond breakage and in the second step we get the blue fragment attacking the red fragment and we get the exons ligated together in the intron liberated alright now if you think about this what we're really doing is we're recombining say a sequence that's a thousand nucleotides long with the sequence that's one nucleotide long and we're ending up with two sequences that are 500 nucleotides long say okay so this is and in fact a recombination reaction it's an it's a disproportionate in reaction alright so we took advantage of this disproportionate in reaction this recombination reaction and we also took advantage of the fact that enzymes by their very nature can catalyze the ford step of a reaction as pretty much equally well as the reverse of a reaction to make a generalized recombination system out of a group one intron alright and here's the system that I came up with I took a group one intron which is a catalytic RNA that in nature self splices out of a nascent transcript alright and that's this gray sequence right here and I designed it I sort of engineered it in a crude sort of way so that it would take two exogenous pieces of RNA say a B and C D although I'm not showing you D and through a series of transesterification reactions which are very easy they're they're chemically facile reactions can catalyze the recombination of a B and C D into C B and AD although I'm not showing the ad on this slide okay so it's just a series of reactions that involve transesterification alright and by doing this i thought i could design a ribozyme that could recombine other rnas okay so the ribosome that I settled on ultimately to do these kinds of reactions is the azo Arcis ribozyme okay this is a relatively short group on intron it's about 200 nucleotides long and it's found in the ISO Lucille tRNA of the purple bacterium azo arcus ii the primary and secondary structure is shown on the left Scot Strobel crystallized it at Yale a few years back and there's a tertiary structure on the top this is a very robust ribozyme it's active up to very high temperatures and I thought this would make a good recombinase ribozyme now the critical feature of recombination is recognition it's sequence recognition it's information transfer by base pairing rules and the way that the azo arkose ribozyme carries out a specific recombination reaction by breaking us breaking ribozyme breaking other rnas at a specific site is by using this three nucleotide to three nucleotide recognition process where three nucleotides on the ribosome on five prime end of the ribosome which are called the IGS which stands for internal guide sequence base pair with a target sequence on its substrate okay bye-bye Watson Crick base pairing here and here and a GU wobble there okay so in in the wild type azo arkose ribozyme its i GS is GU G all right it's very kind of important to remember that GU g and g ug is like a heat-seeking missile it goes out and targets triplets that have its conjugates e au which I'll call the tag and anywhere it finds that tag it will carry out this sequence of splicing reactions okay and so I can engineer this specificity to my own devious needs all right so what we did is we went in the laboratory we made synthetic RNAs a b and c d we put in this target c au where we wanted them to be recombined we threw in the azo Arcis ribozyme and sure enough it can reap recombine pretty much any two substrates to make the recombinant products okay so that's that's all fine and dandy but these are just synthetic oligos they don't really have any use in studying the origins of life yeah so this this you in in these short little guys that yields like 90% okay it's very it's very very good it's transient because the reverse reaction can occur but you know for about an hour - it's very efficient so what about self-replication what we did is that we said well we don't care about RNAs recombining other RNAs other little scraps of RNA we care about rnase recombining themselves so we took the azo arkose ribozyme and broke it into two pieces okay a five-prime half in red a 3-prime half in blue okay and the idea was was we could take the left-hand half and the right-hand half throw in a little bit of the full-length ribozyme and watch it make itself from its own pieces okay and in fact this is the the gel image from that that experiment where the the 5-prime half is radio labeled and after a couple of hours it makes the full-length product okay in really actually high yields the beautiful thing about this though is that I just lied to you this is not the result of that experiment this is the result of the negative control we did for that experiment this is the result of the the experiment where we didn't add any in the full-length ribozyme to begin with okay we just put the two halves in the test tube went away had lunch and came back and somehow they self assembled into a full-length ribosome okay so then we did the next logical thing we broke it into more pieces we broke it into four pieces that we're gonna call W X Y & Z red yellow blue and green okay we're gonna throw these in a test tube with a little bit of salt no full-length ribozyme to be found we're gonna walk away maybe go skiing and come back and watch the production of W into W X W X Y and finally into W X Y Z okay if you leave out any one of the four fragments nothing happens okay now this is in fact W X Y Z we get about a 30% yield after eight hours we can actually cut that band out of the gel sequence it it's all W X Y Z okay we're actually worried we could get W Y X Z for example what we didn't ok so what's what's happening here is that this is a self-assembling RNA ok the two halves are assembling the whole alright and if you look at the four piece notion we have these four pieces W XY and Z and they're all coming together and making themselves okay the way this actually happens if you're wondering is the four pieces come together and form sort of a loose assemblage alright a loose cooperative if you will base pairing interactions this loose assemblage is not a full-blown ribozyme because it's not covalently contiguous from one end to the other but it has enough catalytic activity that will start stitching these molecules together through subsequent recombination reactions until you get the full-length product once you get one of these then you're off to the races because this thing can start stitching itself back together even more rapidly okay so it goes to this what we call a trans state first alright so a non-covalent state and if you think about remember I was talking about cooperation in terms of formation of adenine from hydrogen cyanide well if you think about cooperate as a chemist might you really have to think about bond strengths all right this thing is held together only through hydrogen bonds the four pieces are held together through hydrogen bonds they're not full covalent bonds they're about eight kilojoules per mole of stabilization per hydrogen bond okay but when we get to a covalent bond all right that has you know roughly roughly three times it's much more stabilization as a hydrogen bond all right so what we've done here is we've sort of taken the information present in these molecules transiently gone through a hydrogen bonding state and sort of immortalize that information in covalent bonds all right so here's our sort of Bond type gradient from loose bonds all the way to some sort of covalently contiguous thing all right now in that assembly reaction that I showed you from W XY and Z what we've actually done is we've created a little small little network of reactions that extensive but but it is kind of interesting because you you start with the four molecules separately and then through a series of reactions with a lot of feedback steps a lot of auto catalytic contributions we get to the final product and the reaction kinetics of this thing are go as you would expect it starts off slow it zips up and then it poops out okay so this is a small selfish again you got to be careful with that word autocatalytic network so now now let's go from selfishness to an even more tangible form of cooperation all right if you think about that ribozyme that I told you about and you break it into two pieces what I've just showed you not the four piece case but just the two piece case is where one molecule comes along attacks another molecule gets a transesterification reaction to occur pieces together a covalent version of the molecule and then thing feeds back in an autocatalytic loop alright it's a very selfish sort of thing what I want to do well you know there's there's turnover but I'm not gonna worry about that what I want to do is I want to consider what would happen if we make if we force this system to cooperate all right what if we made the IGS and it's tagged to be non matching in any particular pair in any particular ribozyme all right this thing we've missed batch the central you hear and the C to form a C of C you pair which doesn't have a good base pairing stabilization all right now this thing can't put itself together but it might be able to put it put together some other molecule alright so it can act basically Jewel and if it puts together something that is also missing mismatch then that thing can't put itself together and so you can imagine creating a scenario where things have to cooperate if they don't cooperate they will never get off the ground so what we did was we created on paper this cooperative network where we broke the ribozyme into two pieces in three different ways W and X Y Z W X & Y Z and W X Y & Z okay so far so good but then we were dastardly and we forced the IGS and the tags and these systems to be mismatched and they're mismatched in such a way that this thing could in theory put this together this thing in theory could put this together and this thing in theory could put that together but none of them could put themselves together all right so we're forcing these RNA molecules to cooperate chemical cooperation all right when we do this we plot the yield of full length molecules as a function of time we see that in fact cooperation is going on if we take any one of the systems by itself nothing happens if we take two of the systems you get this sort of linear growth because say this system can put together that system but when you throw all three systems in the test tube all six RNAs in a test tube at the same time you get a much more significant yield and in much more dramatic fashion okay the next thing we did was we said well alright we can create a cooperative Network but is that better than a selfish Network is cooperation better than selfishness because you know I showed you in the beginning that you could have these little selfish assembly reactions so what we did was we plotted the yield of full length molecules as a function of time for two different experiments one in red because red is bad right hot-hot burns baby right red is bad that's selfishness all right a selfish system like I showed you before okay versus green which is good all right go green where we had the cooperative network and so we we pitted them not together but separately and unfortunately I have bad news selfishness is better than cooperation okay which is not what I was hoping would happen but then we thought well let's throw all of them together in the test two at the same time and see what happens if you throw them all together in the test tube at the same time you get a reversal of fortune you actually see that when everything is forced to exist in the same reaction milieu and compete for the same limiting resources the co-operative system outcompetes the selfish system all right so this is this is a very exciting result to me and I thought well this is this is really keen because it can really show cooperation at the molecular level and at least some crude form thank you of the benefit of cooperation all right and with the help of Irene our previous speaker we were able to mathematically model this and we're to confirm this result sort of from a mathematical point of view same same curves then we did the next logical experiment well let's just randomize these IGS and the tags make a whole bunch of different molecules and throw them in a test tube and see what happens so we randomize the IDS's and the tags we created these sets of molecules okay these sets of molecules in theory could assemble a whole variety of molecules okay because now the middle nucleotide of the IGS and the middle nucleotide of the tag are randomized all right so we through we created these random pools we threw them in a test tube all right of about 10 to the 14 molecules and then we fished out molecules every hour or so and then we've subjected them to high-throughput sequences and we asked what kinds of WXYZ molecules exist all right what kind of cooperation can exist and the answer is a lot of cooperation can exist there's a lot of possible cooperating molecules these are genotypes each of these circles represents a different genotype the sizes represents the frequency of the genotype some of them are obligatory co-operators some can be selfish and put themselves together we see all pretty much everything you could possibly imagine could be fished out of the system okay so I just want to finish up by saying that these kinds of interactions made me think about you know basic population ecology definitions of cooperation all right and if you think back to some basic ecology remember Hamilton's rule which which depicts the relationship between benefit and cost in an interaction between two different strategies all right and I'm thinking well maybe in this system we can have strategies because we have cooperation when we have selfishness clearly we have strategies that we can exploit so just to give you a flavor of what we're doing now with the system is that we're saying well we can we can take these IGS and IGS tag partners and play games with them all right we can we can imagine all kinds of various pairwise interactions okay and we can come up with a variety of games that we might play with these all right and so these are games and in a game theoretic sense all right we're doing this again with collaboration with with Irene and Martin Nowak all right so we might have a system where a makes b and b makes a but they can't make themselves a couple of selfish guys something that can make itself on something else and vice versa and we can play all sorts of games just to give you a brief hint of the kind of games that we're playing is that we were trying to do something where we can generate a payoff matrix of costs and benefits all right and so these are data where we took two different tags and competed them in a system where W and XYZ are coming together we came up with a payoff matrix here and we're trying to match it to a particular kind of game theoretic scenario all right with costs and benefits and we're just starting this type of work and we think there's a lot of possibility in applying game theory to molecules all right so what I've told you today is that first of all that group one introns can be engineered to be generalized RNA recombinases second that the azo Arcis ribozyme can put itself together basically through a series of recombination reactions we can get self crude sense of self replication in this system and we see a whole bunch of type of assembly type things like Auto catalysis two membered networks three member networks very complex networks okay so this is cooperation the molecular level we think it has relevance to the origins of life itself so I want to thank a lot of the graduate students I want to thank Irene for helping me out on these analyses and thank NASA for giving me money and thank you for your attention this is amazing first and I would claim though that your notion of selfless selfishness and cooperation is exactly the opposite of what would the social evolution is called these things the the self-replicating ones are the co-operators in time in terms of efficiency and fastness of like fitness and then the the cycle which you have to think about it as a cycle as an entity is actually the cheater because it it doesn't grow that well efficiently but it can actually steal Fitness from the cooperator which is the self-replicating so I would claim that there's just the notion of cooperation and cheating from the chemical perspective is completely the opposite of what the evolutionary biologists will say okay a lot of you what you said got lost in the in the in the haze but but I think your bottom line was that maybe it's possible think about this in a completely different or opposing opposite fashion and it may be I'll talk to you more about that because from a chemical point I'm thinking from a chemical point of view I see things in the way that I do but I'm open to new ideas absolutely so for the I eyesight inertia this talk was really cool kind of blew my mind a little bit so for the game theory stuff I guess what do you hope to learn about game theory maybe generally or specifically game theory when it comes to cooperating chemicals that you wouldn't learn my modeling or game theory with microbes okay what do I hope to learn and why can't I do it on the computer great questions what I hope to learn is what types of games are relevant to the origins of life what turps of games are relevant to the kinds of molecules that we actually think are involved in the origins of life and that sort of ties into the answer to the second part of your question why can't we do this on the computer there are biochemical realities that RNA molecules have that can't yet be modelled on the computer and they always surprise us and they show us things that we wouldn't necessarily predict it on the computer and the thing about RNA is that it probably has some kind of pre biological relevance it may not have been the polymer but it has the features of polymers and so I think we can we we we can do these sorts of things and sort of tie theory and empiricism together which is what I'm really trying to do yeah we got about five years so you have time to make it in Mexico that means we are going back to the RNA world yeah watch out it's coming so as long as I have the mic was there was there a rate efficiency trade-off between your co-operators and selfish reproducers was there a rate efficiency trade-off the answer is yes the selfish replicators get going faster because they don't require some sort of intermolecular interaction the co-operators have sort of a time lag because they require this network to be set up so they're a little bit slower and getting started but ultimately as you saw they can win you

Channel: aoflex

Views: 10,637

Rating: 4.7457628 out of 5

Keywords: abiogenesis, rna, origin of life, Niles Lehman, Portland State University, evolution, biology, microbiology, molecular biology, genome, genetics, prokaryotes, virus, transfection, bacteria, dna, grna, splicing, generic repair, knockout, lecture, seminar, cells, KITP, UCSB, UC Santa Barbara, natural selection, Darwin, cell biology, medicine, medical, stem cell, genome engineering, biotechnology, synthetic biology, coyne, dawkins, rna world, Self-assembly, entropy

Id: vrpADqF3VBo

Length: 35min 38sec (2138 seconds)

Published: Wed Mar 25 2015