Session 1: Origins of Life
Transcript of Part 1: The Origins of Life on Earth
00:00:00.00 My name is Jack Szostak, 00:00:01.25 I'm a Professor of Genetics at Harvard Medical School, 00:00:05.22 I'm an Investigator at Massachusetts General Hospital, where my labs are, 00:00:09.18 and I'm also an Investigator of the Howard Hughes Medical Institute. 00:00:14.17 In this lecture, what I'd like to tell you about is recent advances 00:00:21.06 in work from my lab on the origin of cellular life on the early Earth. 00:00:26.22 But before I get into those experiments, 00:00:29.08 I'd like to step back from the origin of life per se, 00:00:33.08 and talk a little bit about some insights from modern biology 00:00:39.00 that bear on this question, 00:00:41.01 in particular why the question has attracted so much interest and attention recently. 00:00:46.16 So, this is one of the iconic images of hydrothermal deep sea vents. 00:00:56.02 This is an environment characterized by very high temperature and pressure, 00:01:01.21 and of course the surrounding area is just teeming with life. 00:01:07.04 Here's another example: an image from Norm Pace. 00:01:12.04 You can see a layer of green cells growing inside the rock. 00:01:18.07 These are photosynthetic cyanobacteria, 00:01:21.29 and they're living in the pores of the rock at very low pH. 00:01:28.23 This is one of the famous hot springs in Yellowstone National Park. 00:01:34.26 Again, a very high-temperature environment; again, full of life. 00:01:43.29 And here's yet another distinct kind of extreme environment, 00:01:48.21 another very low pH environment. 00:01:50.14 This is the Rio Tinto in Spain. 00:01:53.27 Very acidic water, but again teeming with life: 00:01:59.00 microbial, eukaryotic life. 00:02:04.17 There are even more extreme examples of this kind of environment in acid mine drainage sites, 00:02:12.00 where the water that's flowing out is basically sulfuric acid at a pH close to zero. 00:02:18.24 And again there is microbial life. 00:02:21.11 So with all of these examples, 00:02:22.22 what it's telling us is just the remarkable extent which our planet 00:02:27.21 has been colonized by life. 00:02:31.17 And even environments that we would've considered incredibly hostile and extreme 00:02:38.20 are apparently easily adapted to by life. 00:02:43.10 And of course, this is a consequence of the power of Darwinian evolution, 00:02:48.10 to lead to adaptations to diverse environments. 00:02:53.16 So, if you put this together with recent observations 00:03:00.01 from our astronomy colleagues, in terms of the discovery of extrasolar planets, 00:03:06.02 it really puts into focus the question of whether there is life out there, 00:03:13.13 apart from our planet. 00:03:16.07 So this is an image of the Milky Way, of course. 00:03:22.20 Up to a couple of years ago, 00:03:24.21 astronomers had discovered on the order of 500 extrasolar planets, 00:03:29.04 planets orbiting other stars. 00:03:32.08 But more recently, as a result of the Kepler mission, 00:03:36.00 a space telescope that is just pointed continuously at a very dense starfield, 00:03:45.22 a large number of additional planets have been found, 00:03:48.24 about 1200 candidates at the last count. 00:03:52.28 And these are detected as the planets orbit around their star, 00:03:57.29 and if they eclipse the star, if they transit in front of it, 00:04:00.27 they block out some of the light, and you can detect that little dip in the intensity of the light. 00:04:06.23 So this has given us a big enough sample to actually make extrapolations, 00:04:12.02 and what I've heard from scientists associated with the Kepler mission 00:04:17.17 is that those extrapolations suggest that there could be roughly on the order of 00:04:23.06 500 million, perhaps even a billion, Earth-like planets 00:04:27.00 orbiting sun-like stars out there in our galaxy. 00:04:32.13 And so, if you put that together with the fact that we know, 00:04:37.03 on our planet, that at least microbial life can live in incredibly harsh and diverse environments, 00:04:44.28 it's pretty clear that there will environments out there on 00:04:49.12 these other planets that could support life. 00:04:52.03 So the question is, and the thing we all really want to know is: 00:04:56.01 Is there life out there? 00:04:58.01 Are we alone, or is the universe, is our galaxy, full of life? 00:05:03.15 So this really comes down to the question you see here. 00:05:07.25 Is it easy or hard for life to emerge from the chemistry of early planets? 00:05:14.13 And, unfortunately, it's going to be a long time before we can answer that question 00:05:18.28 in the most satisfying way, by direct observation. 00:05:23.16 Even to get indirect evidence from spectroscopy of planetary atmospheres 00:05:30.27 may take 10, 20, 50 years, to look at Earth-like planets. 00:05:38.28 So what can we do in the meantime to try to get some clues to answer this question? 00:05:46.27 So, what we've been doing, and other people have been doing, 00:05:49.29 is to go into the lab and do simple, chemical experiments 00:05:54.29 and try to work out a complete, step-by-step, plausible pathway, 00:06:02.17 all the way from simple chemistry to more complex chemistry to simple biology. 00:06:08.02 And if we can actually show that there's a continuous pathway 00:06:13.26 with no super-hard steps along the way, 00:06:17.20 then I think we can conclude that it's likely that there is abundant life 00:06:22.14 out there in our galaxy. 00:06:25.03 On the other hand, it could be that our experiments show 00:06:29.14 that there are some steps in that pathway that are extremely difficult, 00:06:35.01 there are bottlenecks that might be very hard to overcome, 00:06:38.29 in which case the emergence of life might actually be a very rare phenomenon. 00:06:44.04 And in the extreme, we could be it. 00:06:47.23 This could be the only place in our galaxy or even the universe where life has emerged. 00:06:53.28 So we would like to try to get some insight into these questions 00:06:58.11 by doing simple laboratory experiments. 00:07:00.20 Now, there's a related question, which is shown down here. 00:07:06.03 If there is life out there, 00:07:07.25 is it likely to be pretty similar to what we're familiar with on our planet? 00:07:16.12 Will life that evolved independently elsewhere have the same 00:07:23.25 fundamental kind of biochemistry? 00:07:25.24 Will it be cells that are living in water, 00:07:29.00 using if not RNA and DNA, some nucleic acid to mediate heredity? 00:07:35.20 Will they use protein-like molecules to carry out biochemical functions? 00:07:41.18 Or could there be forms of life that are actually much different, much more diverse, 00:07:49.14 maybe using completely different kinds of molecules to mediate heredity 00:07:52.25 and to mediate function? 00:07:55.04 Or even forms of life that live in very different environments, 00:08:00.06 for example, in solvents other than water. 00:08:03.23 Again, this is the kind of thing that we can address by going into the lab 00:08:09.17 and doing simple experiments, and trying to build structures, 00:08:12.26 and assess the possibility of having living systems in 00:08:18.09 different kinds of environments and with different molecular bases. 00:08:25.16 So, let's try to think, then, 00:08:30.21 about how we can deduce something about early forms of life. 00:08:37.04 After all, if we want to experimentally investigate the beginnings of life, 00:08:41.26 we have to have some idea, some kind of conceptual model, 00:08:45.22 of what very primitive forms of life looked like. 00:08:50.17 And this has been a very difficult thing for people to think about, 00:08:55.03 because we're so biased by our view and our understanding of modern life. 00:09:00.25 So if we look at modern cells, they're incredibly complicated: 00:09:05.16 Just a lot of moving parts, very elaborate structures, 00:09:09.14 such as you can see here in this elaborate structure in a eukaryotic cell, 00:09:15.15 all the machinery involved in cell division. 00:09:19.00 If you go deeper and look at the underlying biochemistry, 00:09:23.09 if anything, it's even more complicated. 00:09:26.17 And this is just a small section of the chart of central metabolism, 00:09:32.09 so there are hundreds or thousands of enzymes that catalyze 00:09:36.22 all of the metabolic reactions that are required for cells to grow and divide. 00:09:43.10 Even the general organizational structure of modern cells is very complicated, 00:09:51.10 in the sense that it's highly self-referential. 00:09:54.24 So every aspect of this process, 00:09:59.03 this central dogma (the transmission of information from DNA to RNA 00:10:03.10 to proteins and then down to building structures with function), 00:10:08.21 every part of that depends on all the other parts. 00:10:12.12 So for example, the replication of DNA requires DNA, 00:10:16.14 but it also requires RNA and proteins, the polymerases. 00:10:20.11 The transcription of RNA requires DNA, 00:10:25.11 which is where the information's stored, but it also requires many proteins, 00:10:29.21 and it also requires many other RNA molecules. 00:10:33.02 And similarly, the formation of proteins occurs on a remarkably complex machine, 00:10:38.24 the ribosome, which is itself composed of RNA and proteins. 00:10:44.02 So, for decades, it was very hard for people to think of any reasonable way 00:10:51.07 in which such an internally self-referential system could emerge 00:10:57.25 spontaneously from a chemical environment. 00:11:01.16 And the answer to that really came from thinking about RNA 00:11:08.27 and the different things that it can do. 00:11:10.21 So this simplification in thinking came from the realization that RNA can not just 00:11:18.04 carry information but can also catalyze chemical reactions. 00:11:23.15 And that realization led immediately to the hypothesis that, 00:11:29.01 in primitive cells, RNA might be able to catalyze its own replication, 00:11:34.05 also carry out biochemical functions for the primitive cell. 00:11:39.10 And so then all you really need to think about is a cell with RNA molecules 00:11:45.14 encapsulated within some kind of primitive cell membrane 00:11:48.25 that itself could be a self-replicating structure. 00:11:52.08 So, the history of this idea actually goes back to the 1960s, 00:11:58.24 and three very smart people, Leslie Orgel, Carl Woese, Francis Crick, 00:12:05.05 hypothesized in part on the basis on the complex folded structure of tRNA, 00:12:13.01 that an early stage of life might've evolved RNA as the 00:12:19.01 sole macromolecular basis of evolved machinery. 00:12:24.25 And so, this lets you think of simple cells emerging with just a single biopolymer, 00:12:31.19 RNA, and that later on, as evolution 00:12:38.03 developed more complex cellular structures, information storage 00:12:42.14 became specialized in DNA, 00:12:44.18 and most functional activities because specialized as the job of proteins. 00:12:51.06 Now, although these ideas were put forth in rather elementary form in the 60s, 00:12:56.14 of course nobody took them seriously at the time, 00:12:59.21 because there was absolutely no experimental evidence for the idea 00:13:04.08 that RNA could catalyze chemical reactions. 00:13:06.23 At the time, people had just started to get very detailed, 00:13:11.19 high-resolution information about how proteins catalyzed reactions, 00:13:16.13 and the idea that a molecule like RNA could do the same thing seemed ludicrous. 00:13:22.26 So it wasn't until almost 20 years later, 00:13:25.05 with the work of Tom Cech and Sid Altman, 00:13:29.10 and the experimental demonstration that RNA molecules could actually 00:13:34.02 very effectively catalyze at least certain types of chemical reactions, 00:13:38.14 that people took this whole idea of an RNA-based early stage of life seriously. 00:13:45.00 And so that hypothesis, the "RNA world hypothesis," 00:13:48.23 was really summarized by Walter Gilbert in an article in 1986, 00:13:57.18 and this has really become the foundation of a lot of thinking 00:14:02.12 about early stages in the emergence of life. 00:14:08.01 So, apart from the basic facts, 00:14:11.20 that RNA does and can catalyze chemical reactions, 00:14:15.04 is there any other evidence that early life might have been 00:14:19.17 based more exclusively on nucleic acids? 00:14:24.21 And in fact, there are several lines of circumstantial evidence. 00:14:28.24 So one of them is the structure of many cofactors. 00:14:33.07 So here you see acetyl-CoA, just one example. 00:14:39.03 But the working part of the molecule is the thioester out here, 00:14:43.27 and for no obvious reason, there's a nucleotide at the other end. 00:14:48.13 And really the only way to make sense of that is the nucleotide is a "handle," 00:14:53.11 either a relic of a primitive ribozyme 00:14:56.24 or something that was easy for primitive ribozymes to grab hold of and thereby, 00:15:04.10 using this cofactor, catalyze reactions in a thioester-mediated way. 00:15:11.14 Now there are other examples. 00:15:14.21 Here is vitamin B12, another very important catalyst. 00:15:20.24 Its working part is this complex corrin ring, 00:15:25.07 but down here you see, again, a nucleotide. 00:15:29.22 What's it doing there? 00:15:30.28 It's probably another relic of the RNA world, 00:15:34.02 when all of this complicated biochemistry was being catalyzed by RNA enzymes. 00:15:40.12 Yet another example is the very way that the substrates for 00:15:45.25 DNA synthesis are made, and they're not made de novo, 00:15:50.29 as you might expect if DNA came first. 00:15:53.27 They're actually made from preexisting ribonucleotides, 00:15:58.02 and so the transformation of ribonucleotides to deoxynucleotides 00:16:03.17 is catalyzed by the enzyme ribonucleotide reductase. 00:16:08.13 And this unusual synthetic pathway can be viewed as the relic of the fact that, 00:16:17.04 early in time, metabolism and RNA synthesis used ribonucleotides, 00:16:23.25 and only later when DNA was invented or evolved, 00:16:29.10 was there was requirement to make deoxynucleotides, 00:16:32.15 and so they're from the closest available substrate. 00:16:38.03 Finally, perhaps the most important and dramatic piece of evidence 00:16:44.14 for the early role of RNA in primitive forms of life 00:16:50.06 is the actual structure of the ribosome. 00:16:53.21 And so this is a slide from Tom Steitz showing a 00:16:57.27 view into the active site of the large subunit. 00:17:02.23 So this is the peptidyl transferase center, 00:17:05.06 and this little green structure in here is a transition state analogue 00:17:09.17 that marks it at the place in this giant machine where the chemistry is happening. 00:17:14.28 And what you can see is that it's these gray squiggles, 00:17:17.29 which are the RNA, that completely make up that active site. 00:17:23.23 So all proteins are generated by an RNA machine, 00:17:29.19 the RNA central region of the ribosome itself. 00:17:34.25 So again, this only makes sense in terms of an early stage of biochemistry 00:17:40.00 dominated by RNA functions, which then over time evolved the ability 00:17:45.15 to make proteins, which are now so important in all modern biochemistry. 00:17:52.05 So, if we want to understand the origin of life, 00:17:57.03 what we need to think about is not simply how to make 00:18:02.14 these incredibly complex modern cells, but we need to think about how to go from chemistry 00:18:07.23 to very simple, RNA-based cellular structures. 00:18:13.08 So, what would the process look like? 00:18:16.04 What's the broader picture? 00:18:17.17 When did this all happen on the early Earth? 00:18:20.17 So, what was the timeframe in which these events took place? 00:18:25.07 This is a slide from a review by Gerald Joyce, and it summarizes 00:18:30.02 the broad sweep of events that were important in the origin of life. 00:18:34.20 So we actually need to think of everything from planet formation, 00:18:38.14 the beginning of the Earth itself around 4.5 billion years ago; 00:18:44.07 over time as the Earth cooled, water could condense, 00:18:48.01 we have a stable hydrosphere, we have liquid water on the surface; 00:18:52.06 following that, increasingly complicated organic chemistry going on, 00:18:57.08 probably in many different environments on the early planet; 00:19:01.00 and then somehow that led up to the synthesis of RNA 00:19:06.09 or RNA-like molecules on the Earth, 00:19:09.13 which could start to carry out biochemical functions inside primitive cells; 00:19:14.24 and then eventually lead to the emergence of much more complicated cells 00:19:19.29 that would be biochemically similar to modern life. 00:19:23.14 Now, the first really good evidence we have about 00:19:27.04 the appearance of modern microbial life is roughly 3.5 billion years ago, 00:19:33.17 so there's a billion-year interval between the formation 00:19:38.17 and cooling of the planet and the first good evidence for life. 00:19:43.21 And basically, we have very little hard evidence about 00:19:48.00 where all of these important events that led up to life emerging from chemistry, 00:19:53.01 when they actually happened. 00:19:55.03 And that goes along with the fact that we have very little concrete evidence 00:19:59.25 concerning the environments in which those transitions took place. 00:20:04.27 So, this is one of the difficult aspects of studying this question. 00:20:09.12 We can't actually go back, 00:20:10.28 we can't know for sure what the early environments were really like, 00:20:16.13 we'll never know exactly what really happened. 00:20:19.15 So what's our goal in studying these questions? 00:20:23.00 What we're trying to do is really come up with a plausibly realistic 00:20:29.16 sequence of events so that we understand all of the transitions 00:20:35.00 throughout this whole pathway, 00:20:37.05 and we'd like to understand a complete pathway, 00:20:38.29 from planet formation through early chemistry, 00:20:41.28 more complicated organic chemistry, 00:20:44.05 up to the assembly of those building blocks into the first cells, 00:20:48.29 the emergence of Darwinian evolution, 00:20:51.04 and then the gradual complexification of early life leading up to what we see now. 00:20:58.10 So, let's look a little bit more closely at the chemical steps. 00:21:03.23 So in broad outline, what we think happened is that you 00:21:07.17 start off with very simple molecules such as shown up here. 00:21:14.02 There's still a lot of debate about the nature of the early atmosphere. 00:21:20.00 Scientific opinions have gone back and forth in terms of 00:21:23.09 the structure and how reducing that atmosphere was. 00:21:27.26 But it's also been recognized that there could be very important local variation, 00:21:32.22 so even if the atmosphere was globally fairly neutral 00:21:38.20 or perhaps mildly reducing or mildly oxidizing, 00:21:41.16 there could be local environments that were more reducing. 00:21:46.14 That, together with the input of various forms of energy 00:21:51.21 (for example, from electric discharges, lightning, 00:21:56.06 high-energy ultraviolet radiation, ionizing radiation) 00:22:01.08 these are all forms of very energetic processes that can basically 00:22:06.15 rip these small starting molecules apart into atoms, 00:22:11.26 which can then recombine to generate high-energy intermediates 00:22:15.17 with multiple bonds, molecules like cyanide and acetylene, 00:22:20.19 formaldehyde and so on. 00:22:22.18 And these molecules can then start to interact with each other 00:22:26.20 and gradually build up more complex intermediates, 00:22:29.13 ultimately leading to the things we really care about: 00:22:33.10 the lipids that will make membranes and vesicles, 00:22:37.25 the nucleotides that will assemble into genetic molecules like RNA, 00:22:42.16 amino acids that can assemble into peptides, 00:22:45.29 which may also play roles in primitive cells. 00:22:49.10 And somehow, and this is the question that my lab has really been focused on, 00:22:53.12 somehow all of these molecules come together 00:22:56.17 and assemble into larger structures that look and act like cells 00:23:01.10 that can grow and divide. 00:23:04.04 So how could that possibly happen, 00:23:06.17 and what would such a primitive cell look like? 00:23:09.24 So here is a schematic version of the way that we're thinking 00:23:13.21 about a primitive cell, or "protocell." 00:23:17.11 So what we think are the important components of a primitive cell 00:23:22.14 are basically two things: 00:23:25.01 a cell membrane and inside, 00:23:28.10 some kind of genetic material, maybe RNA, maybe DNA, 00:23:32.05 maybe something simpler, something more stable, we're not really sure. 00:23:37.12 So the first question is how could you assemble such composite structures? 00:23:43.23 So we want to be have a membrane boundary 00:23:46.13 that can keep important molecules encapsulated within 00:23:51.06 and essentially provide a distinction between the cell itself 00:23:54.22 and the rest of the universe. 00:23:57.05 We need to understand how these two components self-assemble, 00:24:02.09 how they come together. 00:24:04.07 And it actually turns out that that part is all fairly straightforward. 00:24:09.26 Self-assembly processes are critical in thinking about all of the steps, 00:24:15.12 and there are multiple different ways in which these components 00:24:19.06 can be made and can come together. 00:24:22.08 A much harder question and more interesting is: 00:24:28.08 Once you have structures like this, 00:24:30.16 how can they grow and then divide without any of the 00:24:34.17 complicated biochemical machinery that's present in all of modern life? 00:24:40.25 So since we're talking about the origin of life, 00:24:43.00 then by definition we didn't have highly evolved biochemical machinery around. 00:24:48.21 So it's sometimes hard to think about these problems 00:24:51.17 because modern cells use so much biochemical machinery 00:24:56.11 to mediate the process of cell growth and cell division. 00:25:00.12 It's almost hard to think of how could that be driven 00:25:06.04 by simple chemical and physical processes. 00:25:10.07 But that's in essence what we need to figure out 00:25:13.00 in order to understand this process. 00:25:15.07 There's no machinery around, 00:25:16.18 so we have to identify the chemical and physical processes 00:25:20.14 that will drive growth and then mediate cell division. 00:25:25.11 So that applies not only to the membrane, 00:25:27.11 but also the genetic material, whether it's RNA or something else. 00:25:31.13 There have to be simple chemical processes 00:25:35.16 that will drive the copying of that information, 00:25:38.12 that will allow the strands to separate 00:25:41.06 so that another round of copying can take place, 00:25:44.05 and that will allow that replicated material to be distributed into daughter cells. 00:25:50.10 So if we can identify chemical and physical processes that do all of that, 00:25:56.25 we would have a situation where essentially the environment 00:26:00.03 is driving a cycle of growth and division 00:26:04.15 that brings us back to this stage, 00:26:07.05 and you can go around and around that cycle again and again, 00:26:12.05 and that would be just very similar to the way in which 00:26:16.22 modern cells grow and divide. 00:26:18.23 The information within would be propagated and transmitted 00:26:22.23 from generation to generation, 00:26:25.08 and the important thing in terms of the emergence of Darwinian evolution is that, 00:26:31.00 during that continuous process of replication, 00:26:35.07 of course mistakes would be made. 00:26:38.18 Over time, more and more of sequence space would be surveyed, 00:26:43.28 and eventually we think, some sequence would emerge that did something 00:26:48.03 useful for the cell as a whole. 00:26:50.14 As soon as that happened, that sequence, 00:26:54.00 by conveying an advantage to its own cell, 00:26:57.24 whether in terms of growth rate or the efficiency of cell division 00:27:03.00 or the efficiency of survival, 00:27:05.22 it would have an advantage and it would gradually over generations 00:27:08.28 take over the population. 00:27:11.09 And so that is really the essence of Darwinian evolution. 00:27:15.05 You have a change in the genetic structure of the population as a result of natural selection. 00:27:21.01 And that is precisely what we would like to see 00:27:24.03 emerge spontaneously in our laboratory experiments. 00:27:27.17 We want to start with a chemical system 00:27:30.18 and watch it transition into the emergence of real Darwinian 00:27:35.18 evolution at a very simple level. 00:27:39.20 So, let's step back again and think about how all of these 00:27:43.25 molecules would be made in the environment of a primitive planet. 00:27:49.02 And of course, the first breakthrough in this research program 00:27:54.03 was the famous Miller-Urey experiment, 00:27:57.23 in which a mixture of reducing gases was subjected to an electric spark discharge, 00:28:02.10 and the products were analyzed. 00:28:04.12 And amazingly, in that mix of products were many of the amino acids, 00:28:11.19 which are major components of the proteins of modern cells. 00:28:16.28 So that was really a revelation. 00:28:19.29 It really took people by surprise that the building blocks of biological structures 00:28:28.05 could be generated in such an easy manner. 00:28:33.01 Now, in fact that result was so powerful 00:28:36.08 that it might have actually been a little bit distracting. 00:28:40.15 Probably the really important thing that's made 00:28:43.19 in this kind of experiment is not amino acids per se, 00:28:48.03 but high-energy intermediates like cyanide and acetylene. 00:28:53.06 Those are the kinds of molecules that can assemble in 00:28:57.25 subsequent steps into nucleotides, the building blocks of genetic materials. 00:29:07.16 Those molecules are thought to have been made in primitive environments, 00:29:14.11 so that was an electric discharge experiment, 00:29:16.08 which is very analogous to the kinds of lightning displays 00:29:21.12 that you get in volcanic scenarios. 00:29:24.12 So this is the lightning that's going on in the ash cloud 00:29:29.14 of a currently erupting volcano in southern Chile. 00:29:34.02 So since the early Earth was thought to be highly volcanically active, 00:29:38.15 this seems like a very reasonable scenario. 00:29:42.13 What about some of the other molecules that we need 00:29:44.22 to build our primitive early cell? 00:29:50.29 We need to have lipid-like molecules, amphiphilic molecules 00:29:55.29 that can self-assemble into membranes and generate compartments spontaneously. 00:30:00.16 So these are molecules that are amphiphilic: 00:30:03.07 They have one part that likes to be in water, 00:30:05.26 and another part that doesn't like to be in water. 00:30:09.11 And the way that those preferences are balanced 00:30:12.29 is by forming membranes in which the nonpolar parts are on the inside 00:30:17.29 and the polar parts of the molecule face out into the water. 00:30:21.16 So it turns out that it's actually, again, 00:30:24.19 very easy to make molecules like that in a variety of different scenarios. 00:30:30.02 In fact, Dave Deamer and his colleagues showed that you can 00:30:34.08 extract molecules from the Murchison meteorite 00:30:37.11 (it's one of these carbonaceous chondrite meteorites that's rich in organic materials), 00:30:41.23 you can extract molecules that will self-assemble into a vesicle, 00:30:46.10 as you can see here. 00:30:48.04 So they spontaneously make membrane sheets that close up into small vesicles. 00:30:53.19 Here's another example. 00:30:56.00 This is an experiment that was done to 00:31:00.11 mimic processes going on in interstellar molecular clouds, 00:31:05.03 where you have various gasses that have condensed 00:31:07.22 on the surface of silica particles. 00:31:10.11 They're subjected to irradiation by ultraviolet light and ionizing radiation. 00:31:16.03 So if you make ices like that in the laboratory, 00:31:19.18 subject them to ultraviolet radiation, 00:31:22.29 you get a lot of complicated chemistry going on, 00:31:25.01 and then in that vast mix of products, 00:31:28.28 you can extract molecules which again will form membranes 00:31:32.01 and self-assemble into these vesicle compartments. 00:31:37.10 Here is yet another scenario. 00:31:39.03 This is a hydrothermal synthesis done by Bob Hazen and Dave Deamer. 00:31:46.05 Again, in hydrothermal processing, 00:31:50.00 you can grow carbon chains with oxygenated groups 00:31:55.05 such as carboxylates at the end, 00:31:57.05 and these self-assemble into membranes and make many compartments, 00:32:02.03 as you can see in this beautiful image. 00:32:05.29 So, what would be an example of an early Earth 00:32:09.29 environment where something like this could take place? 00:32:14.03 There are a series of experiments from the Simoneit Lab that 00:32:19.14 suggest that hydrothermal synthesis could happen deep down 00:32:24.28 in regions with high temperature and high pressure, 00:32:29.25 on the surface of catalytic minerals such as transition metal sulfides or oxides, 00:32:35.27 and those reactions would basically turn hydrogen and carbon monoxide 00:32:40.27 into fatty acids and related compounds. 00:32:45.05 So the next slide here is a movie that was prepared by Janet Iwasa, 00:32:50.29 that illustrates this process. 00:32:52.15 So we're going deep into the Earth, 00:32:54.23 down through the water channels of a geyser, 00:32:59.00 and here we're looking at the surface of these catalytic transition 00:33:03.28 metal minerals, and you can see hydrogen and carbon monoxide molecules 00:33:08.24 bouncing around the surface, and the mineral is catalyzing 00:33:13.11 their assembly into chains, which eventually will be released and float up, 00:33:21.00 and they'll be caught up in the flow of water 00:33:23.02 and thereby brought to the surface, where you can imagine these fatty acids, 00:33:28.15 fatty alcohols, and related molecules being aerosolized 00:33:32.22 and concentrated in droplets and perhaps even 00:33:35.20 building up into large deposits on the land surface. 00:33:41.05 So it doesn't seem like the prebiotic assembly of molecules 00:33:46.07 that could spontaneously form membrane vesicles is all that difficult. 00:33:53.03 It's definitely an understudied area of prebiotic chemistry, it needs more work, 00:33:57.16 but it looks, I think, reasonably plausible. 00:34:01.13 So the most prebiotically likely molecules 00:34:04.15 would be things like capric acid that you see down here. 00:34:08.24 Short chain, saturated fatty acids. 00:34:12.15 So we do experiments in the lab with molecules like this, 00:34:16.28 but we also use longer chain, unsaturated molecules 00:34:22.07 like myristoleic acid and oleic acid, 00:34:25.05 as model systems because they're just generally easier to work with. 00:34:29.16 So what happens if you just take one of these fatty acids 00:34:34.20 and shake it up in water with some salt and buffer? 00:34:38.06 Is it hard to make membranes? No. 00:34:40.20 What you can see if that you just spontaneously make vesicles 00:34:47.27 in a huge variety of complex structures, a huge range of sizes, 00:34:53.25 all the way from 30 microns (this large vesicle) 00:34:57.19 to many, many smaller vesicles ranging down to 30 nanometers. 00:35:02.29 Many of these vesicles are composed of multiple sheets of membrane, 00:35:08.04 so stacks of membranes. 00:35:10.04 You can see some of these vesicles have smaller vesicles inside them. 00:35:14.17 So it's a very heterogeneous, complex mixture. 00:35:19.29 Now, the other thing that's really important about this is that these vesicles, 00:35:25.25 these membranes, have very, very different properties 00:35:29.02 from modern biological membranes. 00:35:31.26 Modern membranes are basically evolved to be good barriers, 00:35:37.09 so that cells can control the flow of all molecules in and out 00:35:42.04 using complicated protein machines. 00:35:47.26 For a primitive cell, you wouldn't want a situation like that... 00:35:51.03 that would be suicidal. 00:35:52.15 These molecules have to let stuff get across, 00:35:55.03 they have to have dynamic properties 00:35:56.26 that can let them grow and equilibrate. 00:36:00.16 So the next slide is actually a movie, again prepared by Janet Iwasa, 00:36:05.13 to illustrate the dynamic properties of these vesicles, 00:36:09.29 which are so different from modern membranes. 00:36:13.06 And so what you can see here is, first of all, 00:36:15.08 the motion on the surface, a lot of oscillations, diffusion. 00:36:20.27 In the membrane itself, these molecules, the individual molecules 00:36:25.05 are rapidly flip-flopping back and forth from inside to outside, 00:36:29.23 they're constantly entering the membrane, leaving the membrane, 00:36:34.16 so there's a lot of exchange reactions that are 00:36:37.29 going on on very rapid timescales, on the order of a second or less. 00:36:42.23 So they're very dynamic structures. 00:36:44.23 And these dynamic motions are also probably 00:36:49.09 very important in terms of permeability. 00:36:51.26 They allow the formation of transient defects in the membrane, 00:36:55.20 which let molecules get across spontaneously 00:36:58.21 without any complicated machinery. 00:37:02.19 There's another property of these vesicles which I find quite fascinating. 00:37:09.02 So as you saw in the illustration, the molecules that make up 00:37:12.28 any given vesicle come and go and therefore exchange between vesicles 00:37:19.07 on the timescale of roughly a second. 00:37:22.09 In this slide what you see are two populations of vesicles 00:37:25.16 that were labeled with phospholipid dyes, 00:37:28.18 so they're not exchanging between vesicles. 00:37:31.18 The picture here was taken after about a day, 00:37:35.18 and so you can see that they haven't all just fused and mixed up, 00:37:39.01 there are still red vesicles and green vesicles. 00:37:42.11 And yet we know from our other experiments that the molecules 00:37:45.27 that make up any one of these vesicles are changing 00:37:50.29 on a very rapid timescale, yet the structures themselves 00:37:54.23 maintain their identity on the timescale of weeks or months. 00:38:02.03 What about the nucleic acids then? 00:38:05.05 We've talked a lot about the building blocks of membranes, 00:38:07.27 the way they self-assemble, 00:38:09.12 and the properties of the membranes that they assemble into... 00:38:12.27 let's go back to the genetic materials and think about 00:38:15.21 what kinds of building blocks we need to assemble molecules like RNA. 00:38:22.06 Now, again, we have a difference between the molecules used in modern life... 00:38:28.12 so these of course are nucleoside triphosphates, 00:38:32.11 they're almost ideal substrates for a highly evolved cell 00:38:37.13 with very, very powerful catalysts. 00:38:41.15 These molecules are kinetically trapped in a high-energy state. 00:38:47.13 They don't spontaneously act very well at all, 00:38:51.29 so it takes a very sophisticated catalyst to 00:38:54.27 use molecules like this as a substrate. 00:38:57.26 They're also of course very polar, the triphosphate group is highly charged, 00:39:02.16 and that prevents these molecules from leaking out of the cell, 00:39:06.23 which would be a bad thing. 00:39:08.14 On the other hand, in a primitive cell, 00:39:11.19 if you imagine that substrates, food molecules, 00:39:15.05 are being made in chemical processes out in environment, 00:39:19.15 it needs to be possible for those molecules to get across the membrane 00:39:23.17 spontaneously and get into the interior of the cell. 00:39:27.18 So then we to think about different kinds of substrates, 00:39:31.19 molecules that are less polar so they can get into the cell, 00:39:36.25 and more chemically reactive, so that they can polymerize without the need 00:39:43.15 for very sophisticated, advanced, highly evolved catalysts. 00:39:47.24 And so molecules like this were first made by Leslie Orgel 00:39:52.20 and his students and colleagues 20-30 years ago, 00:39:57.13 and studied in quite a bit of detail as models for the early replication of RNA. 00:40:06.17 So, this brings us back to the question of 00:40:12.02 what was the first genetic material? 00:40:14.17 Was it RNA, in fact? 00:40:17.14 Or is RNA so complicated, 00:40:19.24 or its building blocks so hard to make, 00:40:23.03 that life more likely began with something simpler, 00:40:27.18 something easier to make, 00:40:29.12 maybe something more stable that could accumulate, 00:40:31.24 like DNA for example? 00:40:34.16 So this is an area of active debate and investigation, 00:40:39.10 we really don't know the answer to this question, 00:40:43.13 but lots of people are doing experiments and trying 00:40:45.22 to work out chemical pathways leading up to RNA, 00:40:49.09 for example, the Sutherland Lab in the UK has made a lot of progress in this area. 00:40:55.16 We're studying how these molecules could be assembled and replicated. 00:41:01.27 So one of the satisfying thinks about thinking about RNA 00:41:04.22 as the first genetic material, 00:41:06.10 is that we actually have two different chemical physical processes 00:41:12.29 that can lead to the polymerization of activated building block 00:41:17.01 into long RNA chains. 00:41:19.16 The first of these was discovered by Jim Ferris, working with Leslie Orgel, 00:41:26.27 and that was the discovery that a common clay mineral 00:41:30.20 known as montmorillonite can catalyze the assembly of 00:41:34.19 nucleotides into RNA chains. 00:41:37.04 So this illustrates the structure of this clay, 00:41:39.25 it's a layered hydroxide mineral. 00:41:43.03 In between the layer, the aluminum silicate layers, 00:41:47.21 there's water, and in these inner layers, 00:41:50.27 organic molecules can accumulate, and when they're brought close together, 00:41:54.23 they can react each other and start to polymerize. 00:41:58.25 So here is some of the experimental data. 00:42:01.29 So over a period of days, you start off with small chains, 00:42:07.08 and then gradually they get longer and longer, up to lengths of roughly 40, 00:42:12.18 and in more recent experiments up to 50 or 60, nucleotides long. 00:42:17.27 So I wanted to illustrate that with this movie, 00:42:20.16 another one of Janet Iwasa's animations, 00:42:24.16 to show roughly how we think this works. 00:42:27.15 So these chemically activated building blocks like to stick to the 00:42:31.13 surface of the clay mineral, 00:42:33.20 and when they stick in such a way that they're lined up with each other, 00:42:38.05 they can react and assemble a chemically linked backbone, 00:42:43.16 as you see here. 00:42:47.28 Now, there is another process that can do that same thing, 00:42:51.05 which is very interesting because it's so counterintuitive. 00:42:54.05 It turns out if you take these same building blocks and just have them 00:42:57.29 in a dilute solution and put that on your bench, nothing happens. 00:43:03.10 But if you take that same solution and put it in the freezer 00:43:07.08 and then come back the next day, you'll find RNA chains. 00:43:11.20 Why is that? 00:43:13.09 It's because when water freezes and forms ice crystals, 00:43:16.26 that during the growth of the ice crystals, 00:43:19.12 other molecules (solutes) are excluded from the growing crystal, 00:43:23.23 and so they end up concentrated as much as a thousand fold 00:43:27.21 in between the grains of ice, 00:43:31.04 and so when they're so concentrated, again they can react and polymerize. 00:43:35.08 So having two different processes that can lead the assembly of 00:43:38.23 RNA chains is actually a very satisfying thing... 00:43:41.29 that's something we look for in this field, 00:43:43.29 if there's more than one way of solving a problem, 00:43:47.04 it makes the whole solution seem more robust. 00:43:51.15 Now, the hardest problem, perhaps, 00:43:54.07 is once you've got RNA chains like this, how can they be replicated? 00:43:59.16 So much of our early thinking was based on RNA catalysis, 00:44:04.19 and in fact the whole basis of the RNA world is the idea 00:44:07.16 that RNA can act as an enzyme that could catalyze its own replication. 00:44:13.18 And Dave Bartel, when he was a student in my lab many years ago, 00:44:19.17 actually evolved an RNA enzyme with a catalytic activity, 00:44:25.11 that can ligate together pieces of RNA. 00:44:29.04 And Dave subsequently evolved this ribozyme into an even more complex structure 00:44:35.07 that is really an RNA polymerase made out of RNA. 00:44:40.09 Now, that's a very impressive proof of principle, 00:44:43.26 but unfortunately, despite many advances over the years, 00:44:47.26 we're still far from having an RNA molecule that can 00:44:51.09 completely catalyze the copying of its own sequence. 00:44:55.28 So, what we've decided to do is to actually again step back 00:45:01.07 and try to look at the underlying chemistry 00:45:04.03 and see if there might be ways of adjusting or playing 00:45:10.05 with the chemistry of RNA polymerization that would simplify this problem. 00:45:17.04 Ideally, perhaps we will be able to find a complete chemical process 00:45:22.00 that could drive RNA replication. 00:45:25.15 Now, that's a very difficult task, 00:45:28.16 Leslie Orgel and his colleagues worked on that for many years, 00:45:32.17 got partway to a solution, 00:45:34.22 but were never able to have complete cycles of replication. 00:45:39.23 But we have decided to go back and look at some model systems 00:45:44.16 and see if we can get some clues as to how to approach that problem, 00:45:48.24 perhaps in some fresh ways. 00:45:51.13 So, just to illustrate what we're really after, 00:45:53.24 I'm going to show another of Janet Iwasa's movies, 00:45:57.16 and so what you see here is an RNA template, a single-stranded molecule, 00:46:02.00 floating in a solution full of activated monomers, 00:46:05.05 which then find their complementary bases, 00:46:07.26 so they use Watson-Crick base pairing to line up on the template, 00:46:11.06 and then they basically click together to build up a complementary strand, 00:46:16.16 generating a duplex product. 00:46:20.25 So we're after some kind of simple, 00:46:25.03 chemical system that would drive that process very efficiently. 00:46:30.18 So, if we could get to that point, 00:46:33.06 then we would be back to being able to assemble this kind of model system, 00:46:39.10 a model protocell, composed of a membrane compartment boundary 00:46:45.07 and replicating genetic material on the inside. 00:46:50.28 Now, when we're thinking of a complex composite system like this, 00:46:56.13 the question often arises as to, 00:46:58.06 well, why really bother with the membrane compartment? 00:47:01.10 Why not just let the RNA molecules replicate in solution? 00:47:05.17 And one way of thinking about that is that, 00:47:09.22 for Darwinian evolution to emerge, 00:47:12.27 molecules that are in some way better than their neighbors 00:47:15.25 have to have an advantage for themselves. 00:47:19.01 So if we think about RNA replicases floating around in solution, 00:47:24.02 so these would be RNA molecules that catalyze the replication 00:47:27.24 of another RNA molecule, 00:47:30.14 it doesn't really help if you have a mutation which is faster or more accurate, 00:47:37.23 if all it's doing is copying random, other RNAs 00:47:41.01 that it bumps into in solution. 00:47:44.01 It has to have an advantage for itself. 00:47:47.18 And the simplest way to imagine that happening 00:47:50.13 is to encapsulate these molecules within a vesicle, 00:47:54.20 so that they're always copying molecules that are related by descent. 00:48:01.03 Now, the self-assembly of these kinds of complex structures 00:48:06.22 is something that's actually quite simple. 00:48:09.28 So, at the lowest level, 00:48:12.26 the formation of a membrane vesicle can just encapsulate 00:48:16.01 whatever is there in the surrounding solution. 00:48:19.17 However, it's intriguing that there are ways of making the process more efficient, 00:48:24.02 and one of the most interesting ways of doing that is 00:48:28.05 to take advantage of that same clay mineral, montmorillonite, 00:48:31.25 that we've already seen can catalyze the assembly of RNA strands. 00:48:36.24 And so what you can see in this picture, 00:48:39.19 which was generated by Shelly Fujikawa and Martin Hanczyc 00:48:44.11 when they were in my lab about eight years ago... 00:48:47.03 what you can see is that we have here a clay particle, 00:48:51.15 which has RNA molecules bound to its surface, 00:48:55.21 so the orange color is a dye-labeled RNA, 00:48:59.02 and it turns out these clay particles can catalyze the 00:49:01.28 assembly of membrane sheets from fatty acids. 00:49:08.15 And what's happened here is that this clay particle has catalyzed 00:49:11.15 the assembly of this large surrounding vesicle 00:49:15.29 as well as the many smaller vesicles encapsulated within. 00:49:20.03 So what we now can see is that a single very common, 00:49:25.11 abundant mineral can catalyze the assembly of a genetic material, 00:49:30.17 it can catalyze the assembly of compartment boundaries (cell membranes), 00:49:34.08 and it can help bring them together. 00:49:36.13 So very intriguing as a way of simplifying the assembly of 00:49:39.14 cell-like structures on the early Earth. 00:49:43.02 Here's another picture: clay particle inside a vesicle. 00:49:48.01 Here the boundary is quite dramatically evident, 00:49:52.07 so this is a stack of many layers of membrane bilayers. 00:49:56.26 Here's yet another example where the large outer vesicle 00:50:00.15 is filled with hundreds of smaller vesicles, all assembled under 00:50:04.08 the catalytic influence of this clay particle in the middle. 00:50:10.11 So, assembling these things looks fairly simple. 00:50:14.02 What about the process of growth and division? 00:50:16.21 After all, that's what we really need to generate 00:50:19.07 cell-like structures that can propagate. 00:50:22.18 And at this point, 00:50:25.01 what I can say is that we've come up with a process that looks fairly robust. 00:50:31.09 We can start with vesicles and food in the form of fatty acid micelles. 00:50:37.16 They grow remarkably into filamentous structure, 00:50:42.13 which can then divide very easily into daughter cells, 00:50:45.19 and this generates a cycle that can go around and around indefinitely. 00:50:51.08 And in the next part of this lecture, 00:50:53.26 I'll go into much more detail about the nature of this process 00:50:57.09 and the mechanism by which this happens. 00:51:01.02 But, putting this cycle together with our 00:51:06.17 thinking about nucleic acid replication, 00:51:09.13 we can actually start to imagine what a 00:51:12.13 primitive cell cycle would have looked like. 00:51:15.12 And so this is shown in this figure from a Scientific American article 00:51:20.04 that I wrote with Alonso Ricardo from my lab, 00:51:23.03 and it summarizes some of our ideas about the ways in which 00:51:28.28 the early Earth environment might help to drive cell growth and division. 00:51:35.22 So the idea is that the general environment should be rather cold, 00:51:40.15 perhaps even an ice-covered pond, 00:51:44.24 something you might find in an arctic or alpine environment. 00:51:48.29 There are many examples on the modern Earth. 00:51:52.16 The reason for wanting a cold environment in general is that the 00:51:56.08 copying chemistry seems to go better at low temperatures. 00:52:01.06 The low temperature helps the building blocks 00:52:03.10 to bind to the template and facilitates the copying process. 00:52:07.10 But then we know that eventually, once copying is complete, 00:52:10.28 you have to get the strands apart so that you can 00:52:13.09 undergo another round of copying. 00:52:15.21 Simplest way for that to happen is to invoke high temperatures. 00:52:19.23 And so what we like to think about are 00:52:21.21 convection cells driven by geothermal energy; 00:52:26.12 so essentially in a hot spring type of environment, 00:52:30.13 you could have a pond that's mostly cold, 00:52:32.20 but every now and then, these particles would get caught up 00:52:35.01 in a plume of hot water rising from a hot spring. 00:52:39.01 They'd be transiently exposed to high temperatures 00:52:41.24 that would result in strand separation. 00:52:45.02 It also allows for a rapid influx of nutrients from the environment 00:52:50.02 to feed growth and replication through the next round. 00:52:54.08 And then that would generate a cycle in which the 00:52:57.25 entire process of growth and replication and division 00:53:01.07 is driven by fluctuations in the environment. 00:53:04.20 This is driving us to talk to geologists and to search 00:53:09.19 for analogues of this kind of environment on the modern Earth. 00:53:13.13 Here is a beautiful image of an Antarctic lake 00:53:18.19 in which you see stromatolites, 00:53:20.15 these mounds here are microbial growths on the surface, 00:53:25.18 and the reason that it's liquid is of course there is heat rising up from 00:53:29.18 below geothermally, so it's not a perfect analogue 00:53:34.15 of the scenario I described. 00:53:36.05 We'd like to find environments like this where there are hot springs 00:53:39.12 generating convection cells that could drive the whole cycle. 00:53:43.05 So that would be very satisfying if we could identify such environments. 00:53:48.29 So, what I've tried to show in this lecture is basically 00:53:54.13 the context of the environment and the chemistry 00:53:57.10 leading up to the assembly of primitive cells, 00:54:00.25 in a way that's plausible on the early Earth. 00:54:03.17 And what we'll head into in the next two parts 00:54:06.18 are a more detailed look at the chemistry of membrane assembly, 00:54:11.11 growth, and division; 00:54:12.16 and the chemistry of nucleic acid replication. 00:54:15.25 And all of this work is of course has been done through many 00:54:20.17 very talented students and postdocs in the lab 00:54:25.03 who you can see here on this slide. 00:54:28.13 Thank you.