Jack Szostak (Harvard/HHMI) Part 1: The Origin of Cellular Life on Earth
By Micah Moen
My name is Jack Szostak, I’m a Professor of Genetics at Harvard Medical School, I’m an Investigator at Massachusetts General Hospital, where my labs are, and I’m also an Investigator of the Howard Hughes Medical Institute. In this lecture, what I’d like to tell you about is recent advances in work from my lab on the origin of cellular life on the early Earth. But before I get into those experiments, I’d like to step back from the origin of life per se, and talk a little bit about some insights from modern biology that bear on this question, in particular why the question has attracted so much interest and attention recently. So, this is one of the iconic images of hydrothermal deep sea vents. This is an environment characterized by very high temperature and pressure, and of course the surrounding area is just teeming with life. Here’s another example: an image from Norm Pace. You can see a layer of green cells growing inside the rock. These are photosynthetic cyanobacteria, and they’re living in the pores of the rock at very low pH. This is one of the famous hot springs in Yellowstone National Park. Again, a very high-temperature environment; again, full of life. And here’s yet another distinct kind of extreme environment, another very low pH environment. This is the Rio Tinto in Spain. Very acidic water, but again teeming with life: microbial, eukaryotic life. There are even more extreme examples of this kind of environment in acid mine drainage sites, where the water that’s flowing out is basically sulfuric acid at a pH close to zero. And again there is microbial life. So with all of these examples, what it’s telling us is just the remarkable extent which our planet has been colonized by life. And even environments that we would’ve considered incredibly hostile and extreme are apparently easily adapted to by life. And of course, this is a consequence of the power of Darwinian evolution, to lead to adaptations to diverse environments. So, if you put this together with recent observations from our astronomy colleagues, in terms of the discovery of extrasolar planets, it really puts into focus the question of whether there is life out there, apart from our planet. So this is an image of the Milky Way, of course. Up to a couple of years ago, astronomers had discovered on the order of 500 extrasolar planets, planets orbiting other stars. But more recently, as a result of the Kepler mission, a space telescope that is just pointed continuously at a very dense starfield, a large number of additional planets have been found, about 1200 candidates at the last count. And these are detected as the planets orbit around their star, and if they eclipse the star, if they transit in front of it, they block out some of the light, and you can detect that little dip in the intensity of the light. So this has given us a big enough sample to actually make extrapolations, and what I’ve heard from scientists associated with the Kepler mission is that those extrapolations suggest that there could be roughly on the order of 500 million, perhaps even a billion, Earth-like planets orbiting sun-like stars out there in our galaxy. And so, if you put that together with the fact that we know, on our planet, that at least microbial life can live in incredibly harsh and diverse environments, it’s pretty clear that there will environments out there on these other planets that could support life. So the question is, and the thing we all really want to know is: Is there life out there? Are we alone, or is the universe, is our galaxy, full of life? So this really comes down to the question you see here. Is it easy or hard for life to emerge from the chemistry of early planets? And, unfortunately, it’s going to be a long time before we can answer that question in the most satisfying way, by direct observation. Even to get indirect evidence from spectroscopy of planetary atmospheres may take 10, 20, 50 years, to look at Earth-like planets. So what can we do in the meantime to try to get some clues to answer this question? So, what we’ve been doing, and other people have been doing, is to go into the lab and do simple, chemical experiments and try to work out a complete, step-by-step, plausible pathway, all the way from simple chemistry to more complex chemistry to simple biology. And if we can actually show that there’s a continuous pathway with no super-hard steps along the way, then I think we can conclude that it’s likely that there is abundant life out there in our galaxy. On the other hand, it could be that our experiments show that there are some steps in that pathway that are extremely difficult, there are bottlenecks that might be very hard to overcome, in which case the emergence of life might actually be a very rare phenomenon. And in the extreme, we could be it. This could be the only place in our galaxy or even the universe where life has emerged. So we would like to try to get some insight into these questions by doing simple laboratory experiments. Now, there’s a related question, which is shown down here. If there is life out there, is it likely to be pretty similar to what we’re familiar with on our planet? Will life that evolved independently elsewhere have the same fundamental kind of biochemistry? Will it be cells that are living in water, using if not RNA and DNA, some nucleic acid to mediate heredity? Will they use protein-like molecules to carry out biochemical functions? Or could there be forms of life that are actually much different, much more diverse, maybe using completely different kinds of molecules to mediate heredity and to mediate function? Or even forms of life that live in very different environments, for example, in solvents other than water. Again, this is the kind of thing that we can address by going into the lab and doing simple experiments, and trying to build structures, and assess the possibility of having living systems in different kinds of environments and with different molecular bases. So, let’s try to think, then, about how we can deduce something about early forms of life. After all, if we want to experimentally investigate the beginnings of life, we have to have some idea, some kind of conceptual model, of what very primitive forms of life looked like. And this has been a very difficult thing for people to think about, because we’re so biased by our view and our understanding of modern life. So if we look at modern cells, they’re incredibly complicated: Just a lot of moving parts, very elaborate structures, such as you can see here in this elaborate structure in a eukaryotic cell, all the machinery involved in cell division. If you go deeper and look at the underlying biochemistry, if anything, it’s even more complicated. And this is just a small section of the chart of central metabolism, so there are hundreds or thousands of enzymes that catalyze all of the metabolic reactions that are required for cells to grow and divide. Even the general organizational structure of modern cells is very complicated, in the sense that it’s highly self-referential. So every aspect of this process, this central dogma (the transmission of information from DNA to RNA to proteins and then down to building structures with function), every part of that depends on all the other parts. So for example, the replication of DNA requires DNA, but it also requires RNA and proteins, the polymerases. The transcription of RNA requires DNA, which is where the information’s stored, but it also requires many proteins, and it also requires many other RNA molecules. And similarly, the formation of proteins occurs on a remarkably complex machine, the ribosome, which is itself composed of RNA and proteins. So, for decades, it was very hard for people to think of any reasonable way in which such an internally self-referential system could emerge spontaneously from a chemical environment. And the answer to that really came from thinking about RNA and the different things that it can do. So this simplification in thinking came from the realization that RNA can not just carry information but can also catalyze chemical reactions. And that realization led immediately to the hypothesis that, in primitive cells, RNA might be able to catalyze its own replication, also carry out biochemical functions for the primitive cell. And so then all you really need to think about is a cell with RNA molecules encapsulated within some kind of primitive cell membrane that itself could be a self-replicating structure. So, the history of this idea actually goes back to the 1960s, and three very smart people, Leslie Orgel, Carl Woese, Francis Crick, hypothesized in part on the basis on the complex folded structure of tRNA, that an early stage of life might’ve evolved RNA as the sole macromolecular basis of evolved machinery. And so, this lets you think of simple cells emerging with just a single biopolymer, RNA, and that later on, as evolution developed more complex cellular structures, information storage became specialized in DNA, and most functional activities because specialized as the job of proteins. Now, although these ideas were put forth in rather elementary form in the 60s, of course nobody took them seriously at the time, because there was absolutely no experimental evidence for the idea that RNA could catalyze chemical reactions. At the time, people had just started to get very detailed, high-resolution information about how proteins catalyzed reactions, and the idea that a molecule like RNA could do the same thing seemed ludicrous. So it wasn’t until almost 20 years later, with the work of Tom Cech and Sid Altman, and the experimental demonstration that RNA molecules could actually very effectively catalyze at least certain types of chemical reactions, that people took this whole idea of an RNA-based early stage of life seriously. And so that hypothesis, the “RNA world hypothesis,” was really summarized by Walter Gilbert in an article in 1986, and this has really become the foundation of a lot of thinking about early stages in the emergence of life. So, apart from the basic facts, that RNA does and can catalyze chemical reactions, is there any other evidence that early life might have been based more exclusively on nucleic acids? And in fact, there are several lines of circumstantial evidence. So one of them is the structure of many cofactors. So here you see acetyl-CoA, just one example. But the working part of the molecule is the thioester out here, and for no obvious reason, there’s a nucleotide at the other end. And really the only way to make sense of that is the nucleotide is a “handle,” either a relic of a primitive ribozyme or something that was easy for primitive ribozymes to grab hold of and thereby, using this cofactor, catalyze reactions in a thioester-mediated way. Now there are other examples. Here is vitamin B12, another very important catalyst. Its working part is this complex corrin ring, but down here you see, again, a nucleotide. What’s it doing there? It’s probably another relic of the RNA world, when all of this complicated biochemistry was being catalyzed by RNA enzymes. Yet another example is the very way that the substrates for DNA synthesis are made, and they’re not made de novo, as you might expect if DNA came first. They’re actually made from preexisting ribonucleotides, and so the transformation of ribonucleotides to deoxynucleotides is catalyzed by the enzyme ribonucleotide reductase. And this unusual synthetic pathway can be viewed as the relic of the fact that, early in time, metabolism and RNA synthesis used ribonucleotides, and only later when DNA was invented or evolved, was there was requirement to make deoxynucleotides, and so they’re from the closest available substrate. Finally, perhaps the most important and dramatic piece of evidence for the early role of RNA in primitive forms of life is the actual structure of the ribosome. And so this is a slide from Tom Steitz showing a view into the active site of the large subunit. So this is the peptidyl transferase center, and this little green structure in here is a transition state analogue that marks it at the place in this giant machine where the chemistry is happening. And what you can see is that it’s these gray squiggles, which are the RNA, that completely make up that active site. So all proteins are generated by an RNA machine, the RNA central region of the ribosome itself. So again, this only makes sense in terms of an early stage of biochemistry dominated by RNA functions, which then over time evolved the ability to make proteins, which are now so important in all modern biochemistry. So, if we want to understand the origin of life, what we need to think about is not simply how to make these incredibly complex modern cells, but we need to think about how to go from chemistry to very simple, RNA-based cellular structures. So, what would the process look like? What’s the broader picture? When did this all happen on the early Earth? So, what was the timeframe in which these events took place? This is a slide from a review by Gerald Joyce, and it summarizes the broad sweep of events that were important in the origin of life. So we actually need to think of everything from planet formation, the beginning of the Earth itself around 4.5 billion years ago; over time as the Earth cooled, water could condense, we have a stable hydrosphere, we have liquid water on the surface; following that, increasingly complicated organic chemistry going on, probably in many different environments on the early planet; and then somehow that led up to the synthesis of RNA or RNA-like molecules on the Earth, which could start to carry out biochemical functions inside primitive cells; and then eventually lead to the emergence of much more complicated cells that would be biochemically similar to modern life. Now, the first really good evidence we have about the appearance of modern microbial life is roughly 3.5 billion years ago, so there’s a billion-year interval between the formation and cooling of the planet and the first good evidence for life. And basically, we have very little hard evidence about where all of these important events that led up to life emerging from chemistry, when they actually happened. And that goes along with the fact that we have very little concrete evidence concerning the environments in which those transitions took place. So, this is one of the difficult aspects of studying this question. We can’t actually go back, we can’t know for sure what the early environments were really like, we’ll never know exactly what really happened. So what’s our goal in studying these questions? What we’re trying to do is really come up with a plausibly realistic sequence of events so that we understand all of the transitions throughout this whole pathway, and we’d like to understand a complete pathway, from planet formation through early chemistry, more complicated organic chemistry, up to the assembly of those building blocks into the first cells, the emergence of Darwinian evolution, and then the gradual complexification of early life leading up to what we see now. So, let’s look a little bit more closely at the chemical steps. So in broad outline, what we think happened is that you start off with very simple molecules such as shown up here. There’s still a lot of debate about the nature of the early atmosphere. Scientific opinions have gone back and forth in terms of the structure and how reducing that atmosphere was. But it’s also been recognized that there could be very important local variation, so even if the atmosphere was globally fairly neutral or perhaps mildly reducing or mildly oxidizing, there could be local environments that were more reducing. That, together with the input of various forms of energy (for example, from electric discharges, lightning, high-energy ultraviolet radiation, ionizing radiation) these are all forms of very energetic processes that can basically rip these small starting molecules apart into atoms, which can then recombine to generate high-energy intermediates with multiple bonds, molecules like cyanide and acetylene, formaldehyde and so on. And these molecules can then start to interact with each other and gradually build up more complex intermediates, ultimately leading to the things we really care about: the lipids that will make membranes and vesicles, the nucleotides that will assemble into genetic molecules like RNA, amino acids that can assemble into peptides, which may also play roles in primitive cells. And somehow, and this is the question that my lab has really been focused on, somehow all of these molecules come together and assemble into larger structures that look and act like cells that can grow and divide. So how could that possibly happen, and what would such a primitive cell look like? So here is a schematic version of the way that we’re thinking about a primitive cell, or “protocell.” So what we think are the important components of a primitive cell are basically two things: a cell membrane and inside, some kind of genetic material, maybe RNA, maybe DNA, maybe something simpler, something more stable, we’re not really sure. So the first question is how could you assemble such composite structures? So we want to be have a membrane boundary that can keep important molecules encapsulated within and essentially provide a distinction between the cell itself and the rest of the universe. We need to understand how these two components self-assemble, how they come together. And it actually turns out that that part is all fairly straightforward. Self-assembly processes are critical in thinking about all of the steps, and there are multiple different ways in which these components can be made and can come together. A much harder question and more interesting is: Once you have structures like this, how can they grow and then divide without any of the complicated biochemical machinery that’s present in all of modern life? So since we’re talking about the origin of life, then by definition we didn’t have highly evolved biochemical machinery around. So it’s sometimes hard to think about these problems because modern cells use so much biochemical machinery to mediate the process of cell growth and cell division. It’s almost hard to think of how could that be driven by simple chemical and physical processes. But that’s in essence what we need to figure out in order to understand this process. There’s no machinery around, so we have to identify the chemical and physical processes that will drive growth and then mediate cell division. So that applies not only to the membrane, but also the genetic material, whether it’s RNA or something else. There have to be simple chemical processes that will drive the copying of that information, that will allow the strands to separate so that another round of copying can take place, and that will allow that replicated material to be distributed into daughter cells. So if we can identify chemical and physical processes that do all of that, we would have a situation where essentially the environment is driving a cycle of growth and division that brings us back to this stage, and you can go around and around that cycle again and again, and that would be just very similar to the way in which modern cells grow and divide. The information within would be propagated and transmitted from generation to generation, and the important thing in terms of the emergence of Darwinian evolution is that, during that continuous process of replication, of course mistakes would be made. Over time, more and more of sequence space would be surveyed, and eventually we think, some sequence would emerge that did something useful for the cell as a whole. As soon as that happened, that sequence, by conveying an advantage to its own cell, whether in terms of growth rate or the efficiency of cell division or the efficiency of survival, it would have an advantage and it would gradually over generations take over the population. And so that is really the essence of Darwinian evolution. You have a change in the genetic structure of the population as a result of natural selection. And that is precisely what we would like to see emerge spontaneously in our laboratory experiments. We want to start with a chemical system and watch it transition into the emergence of real Darwinian evolution at a very simple level. So, let’s step back again and think about how all of these molecules would be made in the environment of a primitive planet. And of course, the first breakthrough in this research program was the famous Miller-Urey experiment, in which a mixture of reducing gases was subjected to an electric spark discharge, and the products were analyzed. And amazingly, in that mix of products were many of the amino acids, which are major components of the proteins of modern cells. So that was really a revelation. It really took people by surprise that the building blocks of biological structures could be generated in such an easy manner. Now, in fact that result was so powerful that it might have actually been a little bit distracting. Probably the really important thing that’s made in this kind of experiment is not amino acids per se, but high-energy intermediates like cyanide and acetylene. Those are the kinds of molecules that can assemble in subsequent steps into nucleotides, the building blocks of genetic materials. Those molecules are thought to have been made in primitive environments, so that was an electric discharge experiment, which is very analogous to the kinds of lightning displays that you get in volcanic scenarios. So this is the lightning that’s going on in the ash cloud of a currently erupting volcano in southern Chile. So since the early Earth was thought to be highly volcanically active, this seems like a very reasonable scenario. What about some of the other molecules that we need to build our primitive early cell? We need to have lipid-like molecules, amphiphilic molecules that can self-assemble into membranes and generate compartments spontaneously. So these are molecules that are amphiphilic: They have one part that likes to be in water, and another part that doesn’t like to be in water. And the way that those preferences are balanced is by forming membranes in which the nonpolar parts are on the inside and the polar parts of the molecule face out into the water. So it turns out that it’s actually, again, very easy to make molecules like that in a variety of different scenarios. In fact, Dave Deamer and his colleagues showed that you can extract molecules from the Murchison meteorite (it’s one of these carbonaceous chondrite meteorites that’s rich in organic materials), you can extract molecules that will self-assemble into a vesicle, as you can see here. So they spontaneously make membrane sheets that close up into small vesicles. Here’s another example. This is an experiment that was done to mimic processes going on in interstellar molecular clouds, where you have various gasses that have condensed on the surface of silica particles. They’re subjected to irradiation by ultraviolet light and ionizing radiation. So if you make ices like that in the laboratory, subject them to ultraviolet radiation, you get a lot of complicated chemistry going on, and then in that vast mix of products, you can extract molecules which again will form membranes and self-assemble into these vesicle compartments. Here is yet another scenario. This is a hydrothermal synthesis done by Bob Hazen and Dave Deamer. Again, in hydrothermal processing, you can grow carbon chains with oxygenated groups such as carboxylates at the end, and these self-assemble into membranes and make many compartments, as you can see in this beautiful image. So, what would be an example of an early Earth environment where something like this could take place? There are a series of experiments from the Simoneit Lab that suggest that hydrothermal synthesis could happen deep down in regions with high temperature and high pressure, on the surface of catalytic minerals such as transition metal sulfides or oxides, and those reactions would basically turn hydrogen and carbon monoxide into fatty acids and related compounds. So the next slide here is a movie that was prepared by Janet Iwasa, that illustrates this process. So we’re going deep into the Earth, down through the water channels of a geyser, and here we’re looking at the surface of these catalytic transition metal minerals, and you can see hydrogen and carbon monoxide molecules bouncing around the surface, and the mineral is catalyzing their assembly into chains, which eventually will be released and float up, and they’ll be caught up in the flow of water and thereby brought to the surface, where you can imagine these fatty acids, fatty alcohols, and related molecules being aerosolized and concentrated in droplets and perhaps even building up into large deposits on the land surface. So it doesn’t seem like the prebiotic assembly of molecules that could spontaneously form membrane vesicles is all that difficult. It’s definitely an understudied area of prebiotic chemistry, it needs more work, but it looks, I think, reasonably plausible. So the most prebiotically likely molecules would be things like capric acid that you see down here. Short chain, saturated fatty acids. So we do experiments in the lab with molecules like this, but we also use longer chain, unsaturated molecules like myristoleic acid and oleic acid, as model systems because they’re just generally easier to work with. So what happens if you just take one of these fatty acids and shake it up in water with some salt and buffer? Is it hard to make membranes? No. What you can see if that you just spontaneously make vesicles in a huge variety of complex structures, a huge range of sizes, all the way from 30 microns (this large vesicle) to many, many smaller vesicles ranging down to 30 nanometers. Many of these vesicles are composed of multiple sheets of membrane, so stacks of membranes. You can see some of these vesicles have smaller vesicles inside them. So it’s a very heterogeneous, complex mixture. Now, the other thing that’s really important about this is that these vesicles, these membranes, have very, very different properties from modern biological membranes. Modern membranes are basically evolved to be good barriers, so that cells can control the flow of all molecules in and out using complicated protein machines. For a primitive cell, you wouldn’t want a situation like that… that would be suicidal. These molecules have to let stuff get across, they have to have dynamic properties that can let them grow and equilibrate. So the next slide is actually a movie, again prepared by Janet Iwasa, to illustrate the dynamic properties of these vesicles, which are so different from modern membranes. And so what you can see here is, first of all, the motion on the surface, a lot of oscillations, diffusion. In the membrane itself, these molecules, the individual molecules are rapidly flip-flopping back and forth from inside to outside, they’re constantly entering the membrane, leaving the membrane, so there’s a lot of exchange reactions that are going on on very rapid timescales, on the order of a second or less. So they’re very dynamic structures. And these dynamic motions are also probably very important in terms of permeability. They allow the formation of transient defects in the membrane, which let molecules get across spontaneously without any complicated machinery. There’s another property of these vesicles which I find quite fascinating. So as you saw in the illustration, the molecules that make up any given vesicle come and go and therefore exchange between vesicles on the timescale of roughly a second. In this slide what you see are two populations of vesicles that were labeled with phospholipid dyes, so they’re not exchanging between vesicles. The picture here was taken after about a day, and so you can see that they haven’t all just fused and mixed up, there are still red vesicles and green vesicles. And yet we know from our other experiments that the molecules that make up any one of these vesicles are changing on a very rapid timescale, yet the structures themselves maintain their identity on the timescale of weeks or months. What about the nucleic acids then? We’ve talked a lot about the building blocks of membranes, the way they self-assemble, and the properties of the membranes that they assemble into… let’s go back to the genetic materials and think about what kinds of building blocks we need to assemble molecules like RNA. Now, again, we have a difference between the molecules used in modern life… so these of course are nucleoside triphosphates, they’re almost ideal substrates for a highly evolved cell with very, very powerful catalysts. These molecules are kinetically trapped in a high-energy state. They don’t spontaneously act very well at all, so it takes a very sophisticated catalyst to use molecules like this as a substrate. They’re also of course very polar, the triphosphate group is highly charged, and that prevents these molecules from leaking out of the cell, which would be a bad thing. On the other hand, in a primitive cell, if you imagine that substrates, food molecules, are being made in chemical processes out in environment, it needs to be possible for those molecules to get across the membrane spontaneously and get into the interior of the cell. So then we to think about different kinds of substrates, molecules that are less polar so they can get into the cell, and more chemically reactive, so that they can polymerize without the need for very sophisticated, advanced, highly evolved catalysts. And so molecules like this were first made by Leslie Orgel and his students and colleagues 20-30 years ago, and studied in quite a bit of detail as models for the early replication of RNA. So, this brings us back to the question of what was the first genetic material? Was it RNA, in fact? Or is RNA so complicated, or its building blocks so hard to make, that life more likely began with something simpler, something easier to make, maybe something more stable that could accumulate, like DNA for example? So this is an area of active debate and investigation, we really don’t know the answer to this question, but lots of people are doing experiments and trying to work out chemical pathways leading up to RNA, for example, the Sutherland Lab in the UK has made a lot of progress in this area. We’re studying how these molecules could be assembled and replicated. So one of the satisfying thinks about thinking about RNA as the first genetic material, is that we actually have two different chemical physical processes that can lead to the polymerization of activated building block into long RNA chains. The first of these was discovered by Jim Ferris, working with Leslie Orgel, and that was the discovery that a common clay mineral known as montmorillonite can catalyze the assembly of nucleotides into RNA chains. So this illustrates the structure of this clay, it’s a layered hydroxide mineral. In between the layer, the aluminum silicate layers, there’s water, and in these inner layers, organic molecules can accumulate, and when they’re brought close together, they can react each other and start to polymerize. So here is some of the experimental data. So over a period of days, you start off with small chains, and then gradually they get longer and longer, up to lengths of roughly 40, and in more recent experiments up to 50 or 60, nucleotides long. So I wanted to illustrate that with this movie, another one of Janet Iwasa’s animations, to show roughly how we think this works. So these chemically activated building blocks like to stick to the surface of the clay mineral, and when they stick in such a way that they’re lined up with each other, they can react and assemble a chemically linked backbone, as you see here. Now, there is another process that can do that same thing, which is very interesting because it’s so counterintuitive. It turns out if you take these same building blocks and just have them in a dilute solution and put that on your bench, nothing happens. But if you take that same solution and put it in the freezer and then come back the next day, you’ll find RNA chains. Why is that? It’s because when water freezes and forms ice crystals, that during the growth of the ice crystals, other molecules (solutes) are excluded from the growing crystal, and so they end up concentrated as much as a thousand fold in between the grains of ice, and so when they’re so concentrated, again they can react and polymerize. So having two different processes that can lead the assembly of RNA chains is actually a very satisfying thing… that’s something we look for in this field, if there’s more than one way of solving a problem, it makes the whole solution seem more robust. Now, the hardest problem, perhaps, is once you’ve got RNA chains like this, how can they be replicated? So much of our early thinking was based on RNA catalysis, and in fact the whole basis of the RNA world is the idea that RNA can act as an enzyme that could catalyze its own replication. And Dave Bartel, when he was a student in my lab many years ago, actually evolved an RNA enzyme with a catalytic activity, that can ligate together pieces of RNA. And Dave subsequently evolved this ribozyme into an even more complex structure that is really an RNA polymerase made out of RNA. Now, that’s a very impressive proof of principle, but unfortunately, despite many advances over the years, we’re still far from having an RNA molecule that can completely catalyze the copying of its own sequence. So, what we’ve decided to do is to actually again step back and try to look at the underlying chemistry and see if there might be ways of adjusting or playing with the chemistry of RNA polymerization that would simplify this problem. Ideally, perhaps we will be able to find a complete chemical process that could drive RNA replication. Now, that’s a very difficult task, Leslie Orgel and his colleagues worked on that for many years, got partway to a solution, but were never able to have complete cycles of replication. But we have decided to go back and look at some model systems and see if we can get some clues as to how to approach that problem, perhaps in some fresh ways. So, just to illustrate what we’re really after, I’m going to show another of Janet Iwasa’s movies, and so what you see here is an RNA template, a single-stranded molecule, floating in a solution full of activated monomers, which then find their complementary bases, so they use Watson-Crick base pairing to line up on the template, and then they basically click together to build up a complementary strand, generating a duplex product. So we’re after some kind of simple, chemical system that would drive that process very efficiently. So, if we could get to that point, then we would be back to being able to assemble this kind of model system, a model protocell, composed of a membrane compartment boundary and replicating genetic material on the inside. Now, when we’re thinking of a complex composite system like this, the question often arises as to, well, why really bother with the membrane compartment? Why not just let the RNA molecules replicate in solution? And one way of thinking about that is that, for Darwinian evolution to emerge, molecules that are in some way better than their neighbors have to have an advantage for themselves. So if we think about RNA replicases floating around in solution, so these would be RNA molecules that catalyze the replication of another RNA molecule, it doesn’t really help if you have a mutation which is faster or more accurate, if all it’s doing is copying random, other RNAs that it bumps into in solution. It has to have an advantage for itself. And the simplest way to imagine that happening is to encapsulate these molecules within a vesicle, so that they’re always copying molecules that are related by descent. Now, the self-assembly of these kinds of complex structures is something that’s actually quite simple. So, at the lowest level, the formation of a membrane vesicle can just encapsulate whatever is there in the surrounding solution. However, it’s intriguing that there are ways of making the process more efficient, and one of the most interesting ways of doing that is to take advantage of that same clay mineral, montmorillonite, that we’ve already seen can catalyze the assembly of RNA strands. And so what you can see in this picture, which was generated by Shelly Fujikawa and Martin Hanczyc when they were in my lab about eight years ago… what you can see is that we have here a clay particle, which has RNA molecules bound to its surface, so the orange color is a dye-labeled RNA, and it turns out these clay particles can catalyze the assembly of membrane sheets from fatty acids. And what’s happened here is that this clay particle has catalyzed the assembly of this large surrounding vesicle as well as the many smaller vesicles encapsulated within. So what we now can see is that a single very common, abundant mineral can catalyze the assembly of a genetic material, it can catalyze the assembly of compartment boundaries (cell membranes), and it can help bring them together. So very intriguing as a way of simplifying the assembly of cell-like structures on the early Earth. Here’s another picture: clay particle inside a vesicle. Here the boundary is quite dramatically evident, so this is a stack of many layers of membrane bilayers. Here’s yet another example where the large outer vesicle is filled with hundreds of smaller vesicles, all assembled under the catalytic influence of this clay particle in the middle. So, assembling these things looks fairly simple. What about the process of growth and division? After all, that’s what we really need to generate cell-like structures that can propagate. And at this point, what I can say is that we’ve come up with a process that looks fairly robust. We can start with vesicles and food in the form of fatty acid micelles. They grow remarkably into filamentous structure, which can then divide very easily into daughter cells, and this generates a cycle that can go around and around indefinitely. And in the next part of this lecture, I’ll go into much more detail about the nature of this process and the mechanism by which this happens. But, putting this cycle together with our thinking about nucleic acid replication, we can actually start to imagine what a primitive cell cycle would have looked like. And so this is shown in this figure from a Scientific American article that I wrote with Alonso Ricardo from my lab, and it summarizes some of our ideas about the ways in which the early Earth environment might help to drive cell growth and division. So the idea is that the general environment should be rather cold, perhaps even an ice-covered pond, something you might find in an arctic or alpine environment. There are many examples on the modern Earth. The reason for wanting a cold environment in general is that the copying chemistry seems to go better at low temperatures. The low temperature helps the building blocks to bind to the template and facilitates the copying process. But then we know that eventually, once copying is complete, you have to get the strands apart so that you can undergo another round of copying. Simplest way for that to happen is to invoke high temperatures. And so what we like to think about are convection cells driven by geothermal energy; so essentially in a hot spring type of environment, you could have a pond that’s mostly cold, but every now and then, these particles would get caught up in a plume of hot water rising from a hot spring. They’d be transiently exposed to high temperatures that would result in strand separation. It also allows for a rapid influx of nutrients from the environment to feed growth and replication through the next round. And then that would generate a cycle in which the entire process of growth and replication and division is driven by fluctuations in the environment. This is driving us to talk to geologists and to search for analogues of this kind of environment on the modern Earth. Here is a beautiful image of an Antarctic lake in which you see stromatolites, these mounds here are microbial growths on the surface, and the reason that it’s liquid is of course there is heat rising up from below geothermally, so it’s not a perfect analogue of the scenario I described. We’d like to find environments like this where there are hot springs generating convection cells that could drive the whole cycle. So that would be very satisfying if we could identify such environments. So, what I’ve tried to show in this lecture is basically the context of the environment and the chemistry leading up to the assembly of primitive cells, in a way that’s plausible on the early Earth. And what we’ll head into in the next two parts are a more detailed look at the chemistry of membrane assembly, growth, and division; and the chemistry of nucleic acid replication. And all of this work is of course has been done through many very talented students and postdocs in the lab who you can see here on this slide. Thank you.