University Place

Origin of Life as a Chemical Ecological Problem

[Tom Zinnen, Outreach Specialist, Biotechnology Center, University of Wisconsin-Madison]

Welcome, everyone, to Wednesday Nite at the Lab. I’m Tom Zinnen, I work here at the U.W.-Madison Biotechnology Center. I also work for U.W.-Extension Cooperative Extension, and on behalf of those folks and our other co-organizers, Wisconsin Public Television, the Wisconsin Alumni Association, and the UW-Madison Science Alliance, thanks again for coming to “Wednesday Nite at the Lab.” We do this every Wednesday night, 50 times a year.

Tonight, it’s my pleasure to introduce to you David Baum. He’s a professor of botany, he was born in London, England, and he went to high school in Kingston upon Thames, and then he went to Oxford University where he majored in botany. Then he came to Washington University in St. Louis to get his PhD in evolutionary population, excuse me, evolutionary and population biology. He did a postdoc here at U.W.-Madison, and the went to a faculty position at Harvard where he got tenure and then he decided to come back here for the ice fishing.

[laughter]

And – No?

No. Not the ice fishing.

[laughter]

And came back to UW-Madison in 2001. Tonight, he’s going to talk with us about “The Origin of Life as a Chemical Ecological Problem: New Ideas and New Experiments.” Please join me in welcoming David Baum to “Wednesday Nite at the Lab.”

[applause]

[David Baum, Professor, Department of Botany, University of Wisconsin-Madison]

Wow, thank you, Tom. Thank you for inviting me to this.

So, the origin of life is usually framed as a problem of chemistry and physics, maybe a little bit of planetary science and geology. But what I’m going to try and argue today is that to really understand the problem, you have to also integrate ecology and evolutionary theory into your thinking. And when you do that, you’ll – we – I believe that you’ll start to see that there’s a whole class of experiments that ought to be done that haven’t historically been done which have the potential to really help us understand how life originated on this planet and where else in the universe we might find life.

So, what I’m going to do is I’ll kind of introduce you to my thinking, and at the end I’ll tell you about these experiments which we’re just starting, but they’re just – just – I’m very excited about them, which is why I’m so excited to be here and tell you about – about it.

[Title slide, The Origin of Life as a Chemical Ecological Problem – New Ideas and New Experiments, David Baum, Department of Botany and Wisconsin Institute for Discovery, U.W.-Madison]

So, the way that I want to get started is to start from sort of a more conventional view.

[slide titled, The conventional cellular perspective, with the bullet points that cells use energy and food to help them survive and reproduce and selection favors cells that are better at surviving and reproducing]

So, the conventional way of thinking life and how it evolved is from the perspective of cells. So, all life we know of is cellular. And cells are very clever little things that can take food from the environment and energy from the environment, grow and divide, and therefore reproduce.

And we understand very well that if any variant cell is better at surviving and growing and dividing, that variant –

[David Baum]

– will get enriched in the environment. It will – it will be preferentially passed down, this is natural selection. And that’s a very simple concept, and it explains how it is that once you have cells complexification can occur because natural selection doesn’t bias against complexity, and in some situations, it can favor complexity. So, we fully understand that once you have evolving cells, even of a very, very simple kind, they will complexify to give rise to the amazing organisms we see around us today.

The problem with this view is, how do you get started? Because if you take the view that you need a cell to evolve adaptively and adaptive – and adaptive evolution is the only way you can generate complexity, where does the first cell come from?

[slide titled, Were did the first cell come from?, with the bullet point that if being a cell is need to evolve by selection then the first cell must have arisen with the help of selection. Is this explanation plausible?]

You’re forced to the position that the first cell arose already able to function, already able to undergo adaptive evolution. And I would like to question whether that is really a plausible proposition.

[new slide titled, A cell has to do a lot, with a dotted circle under the title representing a cell]

So, think about what even the simplest cell needs to do. A cell is a membrane, and this membrane has to contain a certain set of key meta- metabolites, key chemicals.

[in the slide the word Metabolites appears within the dotted circle along with a curved arrow to show the metabolites interacting with one another]

And those chemicals interact with one another to take food –

[the slide animates on the word Food outside the dotted cell circle with a green arrow pointing into the dotted circle representing the cell taking on food]

– from the environment that has to be able to pass through this membrane and convert that food into waste –

[a blue arrow animates on the slide pointing from the inside of the dotted circle cell to outside the cell with the word Waste outside the circle, representing the cell getting rid of waste in the environment]

– and in the process, generate more of those metabolites –

[the dotted circle cell animates to become larger]

– so that the cell will grow. And all that has to happen under – without any prior – if – if we’re to – in order for evolution to start, this has to happen spontaneously –

[David Baum]

– for the first of these functional cells.

And then, to make matters worse, this cell, this first dividing, growing cell, after it divides, has to be – has to give rise to daughter cells –

[return to the A cell has to do a lot slide now with two cells with metabolites taking in Food and expelling Waste]

– that are at least as good as the parent cell, because if they’re both worse than the parent cell at growing and dividing, then, over time, they will degrade into non-functionality.

So, I view this, and I’m not alone in this –

[David Baum]

– as a – a very challenging proposition. The idea that without any adaptive process a cell that is functional, that is able to evolve, could arise spontaneously could happen. I think that’s very improbable. So, we need alternative models.

So, the – the alternative model that I think you may have heard of is the idea that the first replicating entity –

[slide titled, An alternative, with the bullet point of the question of a single self-replicating R.N.A. molecule]

– was not a cell, but it was a single molecule. A single, very talented R.N.A. molecule. As many of you all know, R.N.A. molecules are kind of interesting molecules because they can template information, but they also can function as catalysts. And so, the idea is if you had a single R.N.A. molecule that could replicate itself, then it could give rise to a population of descendants in which natural selection happens.

That’s all very well, but it also poses a lot of problems, okay.

[under the bullet point about self-replicating R.N.A. an new question animates on – how could a sufficiently long R.N.A. form?]

So, the first problem is, where would this very long R.N.A. come from? In – in – in natural environments, R.N.A. is not stable. We have to work hard to maintain R.N.A. stably in the lab. And in a natural environment it would take a lot of energy to generate these R.N.A. molecules. But supposing that a lone R.N.A. molecule –

[an new question animates on the slide below the previous question – if a sufficiently long R.N.A. molecule does form, how could it use energy to drive replication?]

– was formed and happened to be able to replicate itself without any prior history of adaptive evolution, how would it run that? In order to be able to replicate itself, it needs an input of energy. For example, it would need activated building blocks –

[David Baum]

– nuclear triphosphates. Where would they come from, and how would it be able to harness those indefinitely?

Okay, supposing that problem were overcome, it would then have to be able to copy itself actually accurately enough that its descendants, its copies, have a high probability of being as good as it at replicating. And a lot of theoretical work going back into the ’70s has shown that this is actually very hard to imagine given what we know about how accurately an R.N.A. molecule can function as – as a catalyst.

And, lastly, even if we accept that all these things could happen –

[return to the An alternative slide – now with two additional questions underneath the single self-replicating R.N.A. molecule question – namely, how could they copy accurately enough? And how could they go on to build a cell to live in?]

– how would this R.N.A. molecule build a cell around it? That is not really at all clear.

So, the bottom line is that if we rule out the idea of the first thing –

[David Baum]

– that could evolve being a cell, I would also add we – we could also rule out the first thing that could evolve adaptively as being a single R.N.A. molecule. So, we need yet another approach. And to take – to see that other approach, I think we need to sort of take a step back and think about what life is.

So, how do we define life?

[slide titled, Back to Basics, with N.A.S.A.’s “working” definition of Life – Life is a self-sustained chemical system capable of undergoing Darwinian evolution]

There are many definitions out there, but the most popular definition right now is the so-called “working definition.” This was developed by a committee at N.A.S.A., and it defines life as a self-sustained chemical system capable of undergoing Darwinian evolution.

So, I’m going to sort of break this down. I’m going to first talk about the self-sustaining part, and then I’ll talk about the evolution part.

[new slide titled, A self-sustaining system, with the bullet points that it is – a localized set of chemical species that can use food and energy to promote production of more of the same chemical species, the growth of the whole set into adjacent space, and a chemical ecosystem]

So, what does it mean to be a self-sustaining system? So, what this asks us to do is instead of looking at a cell as a single entity that can grow and divide, think of all the chemicals that make up that cell. The set of chemicals that make up a cell are of many different kinds, but they have the property that they cooperate. They work together to use stuff in the environment, food and energy, to make more of the same set of chemicals in roughly the same proportions so that –

[David Baum]

– as the cell grows, all the chemical species in there all double, ultimately, until the cell can divide.

So, what’s happening then is we have an ecosystem in the cell, a bunch of different species that cooperate and interact with one another to generate more and more of the same, and therefore to grow. So, we can think of a cell, any living system, as a chemical ecosystem.

So, as I said, modern cells can be thought of as these ecosystems, but they’re not like ecosystems we see in nature because they have – cells have a membrane around them. They have a natural edge. If you go out and you find an ecological community, there’s not an edge around it unless we put it there as humans. So, but – we also – but despite that –

[slide titled, Autocatalytic chemical systems, with a microscopic photo of a cell and the bulleted definition as a network of chemicals that cooperate to make more of one another and noting that it applies to modern cells and asks the question if a bounding membrane is necessary?]

– you can imagine a functioning life-like chemical system that didn’t have a membrane. And to help you visualize this –

[a photo of a sand dune animates on the slide]

– think of primary succession happening on something like a sand dune.

So, when a sand dune is colonized by the first few plants, if you’ve got a few little straggly grasses here that have managed to take root, well, they form a nucleus around which other members of that species and other species can start to collect because the sand is now being stabilized. And this community which can nucleate at one point can grow outward. And the notion here is that this community can function and grow even without having a boundary around it. Could that be true for primitive chemical systems?

[a new bullet point animates on the slide S.L.I.M.E. – Surface-associated life-like interacting molecular ensemble, along with a chemical equation]

And the answer is that, in theory, it could. So, a lot of theoretical work over the last sort of especially decade has shown that if you have a mineral surface or a solid surface and a subset of chemicals become associated with that surface –

[David Baum]

– sort of like what happens when these plants first colonize a sand dune, they can nucleate, and if they cooperate with one another, they can make more of the same chemicals in their – in their – in their vicinity. So, they can spread and grow over the surface.

Now, the little acronym I like to use for these systems is Surface-associated Life-like Interacting Molecular Ensemble.

[laughter]

Because that makes it S.L.I.M.E., okay, S.L.I.M.E.

[return to the Autocatalytic chemical systems slide]

So, I’m going to use the word SLIME because it – it’s catchy. So, S.L.I.M.E.s are chemical ecosystems that are unbounded but can nonetheless grow and function like a cell in that they can use material in the environment to make more of the same chemicals and – and expand.

[new slide titled, Autocatalytic chemical systems can self-propagate, featuring a yellow rectangle representing the base of the system with various capital A, B, Cs and Ds on it to represent cells on a mineral surface. Outside the yellow surface are random lower case a,b,c and ds and the key that an uppercase C helps a lower case a become an uppercase A, an uppercase A helps a lowercase b become an uppercase B, an uppercase D helps a lowercase c become and uppercase C, and an uppercase B helps a lowercase d become and uppercase D. All uppercase letters can stick to the surface]

So, to help you visualize this, here’s sort of a cartoon example. Imagine we have this mineral surface in kind of gulls, and there are these four different kinds of chemicals, big A, big B, big C, and big D, all of which have the property of sticking to the surface, but they’re not present in the solution, in – in the ocean. But, very rarely, they can be formed by conversion from some precursor. The precursors being little a, little b, little c, and little d, which don’t stick to the surface.

Now, these chemicals, I call them keystone species –

[uses spotlight function of slide system to highlight the uppercase letters on the yellow rectangle]

– which is an ecological term, have – form a catalytic loop. So, for example, big A helps little b make big B. And big B helps little d make big D. And big D helps, I lost that. Big D makes little c make big C, which helps little a make big A. So, you have a little closed loop, which means that if this is sitting in – in an environment in which little a, little b, little c, and little d are flowing in the ocean –

[on the slide all the lowercase letters fly around the screen and eventually hit their uppercase catalysts which in turn add more uppercase letters to the yellow surface]

– then this system, once it nucleates on the surface, will grow and spread over the surface, which I’ve sort of shown here. And this growth and expansion will continue, in principle, until either the – one loses the monomers, the building blocks, the A, B, C, and D, or you – the surface is – is full – full up or destroyed.

So, this is now how I think of the very beginnings of life, is of these SLIMEs forming spontaneously on mineral surfaces. And where do I imagine it might happen?

[new slide titled, Where would such nucleation and growth be most likely, featuring a photo of a hydrothermal vent on the ocean floor and the answer of solid surfaces bathed in a liquid phase rich in food and energy perhaps hydrothermal vents]

Well, there’s many possibilities, but the most sort of obvious place to think about would be something like a hydrothermal vent. So, a hydrothermal vent is a place in the oceans where water is percolated down into the – into the – into the crust, into – into – into the rocks, and it gets forced back up to mix with the ocean after having been changed chemically quite a lot.

So, what’s interesting is that a hydrothermal vent, there are a lot of chemical reactions that convert carbon, in the form of maybe carbon dioxide or methane into some of the small building blocks of life –

[David Baum]

– simple organic molecules. There’s also a lot of chemical potential energy here because the vent fluids are in a very different energetic state, they’re in a very different redux state –

[return to the Where would such nucleation and growth be most likely slide with Professor Baum using the spotlight feature of the slide program to highlight the plumes coming out of the photo of the hydrothermal vent]

– to the ocean they mix with. And so, the energy here that – that, in fact, many living organisms, many living microbes, use to grow and live.

So, but they could also, this energy and these little food substances –

[David Baum]

– could in principle promote and allow SLIMEs to form and then grow on these minerals around hydrothermal vents.

So that’s how to think of a SLIME as a self-sustaining and growing system. But what about evolution? In order to be living by my criterion –

[slide titled, Are SLIMEs living, with the N.A.S.A. definition of life again written in italics and the words Darwinian evolution underlined and asking the question can SLIMEs evolve adaptively?]

– by that definition, it also has to be able to evolve adaptively. Can they?

Well, I’m going to suggest, and there’s a lot of emerging theory to support this claim, that they – they can. They would in fact evolve.

[new slide titled, SLIME adaptive evolution, with a small grey cloud like illustration representing a SLIME and a note that as SLIMEs expand, they maintain their ability to grow – heritability]

So, the first thing to be aware of is that as a SLIME is spreading over the surface, it’s carrying with it the ability to continue growing. So, there is a kind of heritability So as it grows, the ability to grow –

[the slide animates the SLIME cloud larger and larger]

– is carried forth in the SLIME. So, that is a form of heritability. And if, for any reason, one part of the SLIME was different than the other parts and that part could grow more rapidly, we would expect that part to take over and to maybe cover more of the surface in the long run.

Okay, but how might variants arise? In order to have adaptive evolution –

[David Baum]

– you need heritability, which is to say a sort of – sort of maintenance of some kind of character through time, but you also need a way to generate new variants. So, what – what does that look like? We dont have a genetic system here. Well, again, the modeling suggests that we should understand this in terms of network evolution.

[slide titled, Mutations=new chemical species added, featuring a slide with an illustration of eight cells that are all intaking food and interacting with one another to create autocatalytic loops]

So, autocatalytic chemical systems of which SLIMEs are an example are composed of little autocatalytic loops. And, in principle, there can be multiple of these loops that can exist. Each of them is called a core, an autocatalytic core. So, in this little example from this paper, there is this one sort of greenish core, which is composed of five keystone species that form a loop. So that, if – if those species associated with a surface, it would be a SLIME that could grow over the surface provided it got those food inputs. But if once sometime by chance a little bit of – of A and B were produced in the same area, now that also forms a second autocatalytic core that can be added to the first one. And the two can actually coexist, depending on the – on the details of the model. And so, these SLIMEs can actually evolve not through genetic mutations but through the addition of these autocatalytic cores.

[David Baum]

And, as a result, they are, in principle, evolvable.

So, the take-home here is that at least when one does this, approaches this from a theoretical point of view, these SLIMEs, which sound like they ought to be able to arise kind of easily, would begin adaptively evolving right away. And specifically, they would, over time you would expect them to get better at colonizing new surfaces, at resisting disruption and competing with other SLIMEs. So, we would expect sort of some sort of early complexification well before you have cells.

Now, if you remember when I talked about R.N.A., one of the weaknesses was that there wasn’t a go – a simple way to go from a self-replicating R.N.A. to a cell because we know after all that cells did eventually evolve.

[slide titled, From SLIME to cells, with the bullet points that – at some point cells did evolve and asking the question can a SLIME-y model explain the origin of the first evolvable cell?]

So, we have to – it only – this approach, this kind of ecological way of thinking about chemistry only helps us if it provides a path to the cell. So, can my SLIME-y model explain the origin of the first evolvable cell?

[new slide titled, SLIMEs should evolve towards improved colonization ability, featuring an illustration of an ocean with minerals on the bottom and within those minerals are two separate sets of minerals that are capable of creating SLIMES. One set of minerals is represented as having a green SLIME and two minerals down another mineral is represented as having a yellow SLIME]

And the answer is that it does so very naturally. So, to kind of talk you through this, imagine the following scenario. We have an ocean with patches of a certain kind of mineral that is suitable for SLIMEs and two different SLIMEs evolve independently, one in green and one in yellow. Now, those – every so often, minerals are going to be – are going to become unavailable. Some say they’ll get eroded away. They’ll get – theyll covered by something. And also, every so often, a new virgin mineral patch will appear due to a disruption.

Now, if a certain SLIME –

[the slide animates, and it shows an arrow of some of the green SLIME interacting with the ocean and ending up on the same mineral outcropping of the yellow SLIME so that the two differing SLIMEs are now on the same mineral]

– happened to produce a little piece of itself and release it every so often into the water column, this – this is a little dispersal unit or a propagule that would get up into the water column. It would allow that particular kind of SLIME to spread and colonize all these mineral surfaces. And so, it would gain an advantage. So, over time, you would expect to see more of the green and fewer – fewer of the yellow SLIME because it can keep colonizing as new environment becomes available.

Furthermore, if the SLIME, if there was a – a mutation that allowed that propagule, which is just an inert blob of – of protoplasm, to actually be able to do some metabolism in the water column, to take in food and grow a little bit, just a little bit of growth, maybe you can see that, that would give it potentially a competitive advantage because it would be able to colonize more effectively on a new surface.

[new slide that animates several propagules of the green SLIME coming off of the original mineral grouping and transferring to several other outcroppings of minerals on the ocean floor eventually overtaking and eliminating the yellow SLIME altogether]

The same principle, selection for enhanced colonization ability, would then favor a variant again that was able to divide. So now, in the water column, it can do metabolism, it can grow, it can divide, and it can – it can colonize multiple mineral patches. And once you have a propagule or a blob that can divide on its own –

[slide animates several propagules now floating freely in the ocean and being able to propagate independent of the minerals from which it was formerly attached]

– then we don’t need to go back to the surface, and we have basically an evolvable, functioning cell.

[slide animates on the statement that SLIMEs would eventually yield functional cells]

So, in summary from this perspective, what’s nice about the SLIME model is that it actually provides a very natural selective steppingstone to a cell with – with division being a relative, the ability to –

[David Baum]

– divide being a relatively late addition in that sequence.

So, this model is – represents a change in concept. And I – I think the way that I would most starkly put this would be that it really suggests that when we think about the origin of life, the traditional view puts it in – in rather stark terms.

[slide titled, Rethinking the origin of life, featuring a graph with Time on the x-axis and Life-like-ness on the y-axis and the graph features a short period of non-life towards the beginning of Time and then a large straight upwards jump to a period of Time that there is Life]

That we go from non-living systems to the spontaneous appearance of – of something that has to be quite complicated but can now evolve adaptively a cell, and then we have life. That life is a – is a gulf that at some point was crossed.

[the slide animates off the previous graph line]

The perspective I’m trying to get across –

[slide animates on a new graph line that is a progressive curve from barely-life to life as we know it in opposition to the previous no-life/life binary]

– is a much more gradual transition. From very simple chemical networks, chemical ecosystems that can colonize and nucleate on mineral surfaces and grow –

[the slide animates on a large red area onto the timeline from barley-life to a quarter of the way to life-as-we-know-it and labels it SLIME]

– but then can gradually complexify, transitioning from sort of being unbounded SLIMEs at some point to being cells –

[slide animates on a larger orange area to the right of the red SLIME area on the timeline about 2/3rds of the way to life-as-we-know-it and labels it Cells]

– and eventually to adding –

[slide finally animates a green areal to the right of the orange cell area on the timeline all the way to life-as-we-know-it and labels it +Genetics]

– this familiar machinery of living – of living cells, which is to say D.N.A. and R.N.A. and a genetic – genetic system. So, this transition is a much more gradual one in my way of thinking.

[the slide animates on an arrow going the length of the SLIME section of the graph and asks the question of how long the SLIME area is on the timeline]

Furthermore, it raises the possibility that at least the early stages, these initial transitions, might occur easily and quickly –

[David Baum]

– rather than requiring a long wait for a miraculous transition.

So, taking this view, we can now ask, could SLIMEs actually arise easily? Because I think you can see that if SLIMEs could arise easily, suddenly whole new research programs become possible.

So, I’m going to suggest –

[slide titled, Could new SLIME-y life arise easily, with a bulleted list of why nothing argues against it]

– that while we don’t know that SLIMEs could arise easily, there’s nothing we know that says they couldn’t. Not a great argument. But it – but it – but it works, okay?

[laughter]

[slide animates on the bullet point – Life on Earth arose quickly]

So, the first thing to bear in mind is there is no evidence of a period when life lacked – when Earth lacked life. So, the oldest rocks, that could contain direct tangible evidence of life, do contain evidence of life. Now, that doesn’t mean that there wasn’t a long period after the planet formed –

[David Baum]

– when there might not have been life because for the first few hundred million years the – the planet was potentially habitable, we have no rocks to give us a record. It could be that life of some kind originated, you know, three days after the planet had water. It could have taken a couple hundred million years. We don’t know. But it’s at least possible that it happened quickly.

A second consideration is we might – you might argue –

[return to the Could new SLIME-y life arise easily slide, now with a new bullet point – Cellular life is likely to be very good at outcompeting new life]

– “Well, if life arose easily, wouldn’t we have – be aware of many independent origins of life?” Whereas all the data clearly show that every life form we know about today traces back –

[David Baum]

– to the same origin. Wouldn’t, if – if life could originate easily, wouldn’t there be many origins? Actually, Darwin himself pondered this question, and he pointed out, genius that he was, that if – once you have very able complicated cellular life around, it’s going to make it very hard for any other life to – to actually become complicated enough for us to see it, because life will eat it. Life will eat its food. And – and so, complex life will preempt simple life becoming complicated enough to be seen.

And then the last thing to be aware of is actually we don’t know that there aren’t SLIMEs all over the place in the environment –

[return to the Could new SLIME-y life arise easily slide with the new bullet point that – We wouldnt know if there were SLIMEs on mineral in the ocean]

– because we don’t have a good way to detect them.

[slide animates on the bulleted supposition – lets take it seriously and study SLIME in the lab]

So, the upshot of this perspective is we don’t know that SLIMEs can arise easily, we don’t know that these evolving systems can arise easily, but we also don’t know they can’t. And the only way to explore this as a scientist is to get in the lab and try and do some experiments. So, that is what we want to do.

[new slide titled, An empirical research program with the hypothesis that – life began in the form of SLIMEs that could grow and evolve on mineral surfaces. The experiment is to select for SLIMEs (if they are present) that have particular features and see if there is a response to that selection]

This is where I transition to telling you what we’re hoping and trying to do in my lab and in the lab of – of – of a number of collaborators around the country. So, the hypothesis that we want to test is that SLIMEs, which is to say chemical systems that can colonize a mineral surface can grow and potentially evolve, can arise quite easily.

So, how – but how could we detect them? So, as an evolutionary biologist, it seemed obvious to me that we should go right to the heart of the matter. If a SLIME – if what were looking – if our definition of life –

[David Baum]

– is something that can evolve, we should be looking for things that can evolve. So, what we should do in our experiments is impose artificial selection for some characteristic on a chemical system that might contain a SLIME and then see if there is a response to that selection. If there is a response to that selection, that will point to the existence of something that could propagate itself and evolve. So thats – those are the experiments we’re trying to do. So, let me talk you through because this just really repeats that –

[slide titled, Chemical ecosystem selection, with the bullet points that they – want to impose selection on chemical mixtures, and use a response to selection as evidence that an evolvable autocatalytic system (a SLIME) is present)

– that we want to impose selection on mixed chemicals and then use the response as evidence there is in fact a – a life-like chemical system present in that mixture.

So, how – how are we going to do this?

[new slide under the title of Chemical ecosystem selection, with a new bullet point to – make a diverse chemical soup (e.g., modeled after Miller-Urey experiments) and featuring an illustration of said experiment]

So, the first thing we need is some kind of chemical soup. We need a mixture of the kinds of small molecules that could potentially form these – the – the – the keystone species of an autocatalytic system. Now, what we’ve been doing to date is creating ourselves, buying from – from Fisher or one another company, individual chemicals and mixing them together to make our own sort of prebiotic soup. But what we can do and hope to do, thanks to a collaborator, in the future is actually take the soups, create it in Miller-Urey type experiments.

So, Miller-Urey, Miller – Stanley Miller and Harold Urey did these experiments in the ’50s where they sparked primordial atmospheres and they showed that you could get liquid soups that contained many of the building blocks of life. And so, in the future we’ll actually being using real Miller-Urey soups, but right now we’re making our own from scratch.

[the slide animates on the next bullet point – add mineral grains and a photo of grains of pyrrhotite]

The next thing you need is some minerals, and there’s lots of minerals to use which we’re using.

[David Baum]

The experiments we planned this summer we’ll use four different minerals, but one of them is a pyrrhotite, which is an iron sulfur mineral that was abundant on the early Earth, and some -some theories from the ’80s suggest it’s a very good candidate surface –

[return to the Chemical ecosystem selection slide]

– for the formation of – of one of these autocatalytic systems.

[the slide animates on the next bullet point – add a source of chemical energy either Redox disequilibrium or high energy phosphate bonds – and the photo of the mineral grains is now replaced by a photo of orthophosphate assay]

The next thing we want to do is add some kind of energy because in order for creative work to happen, in order for a system to be able to maintain itself, you have to be – there has to be an input of energy. And what we are using as energy either redux disequilibria, so, for example, we might add ammonium and nitrate. So, these are the ingredients of a fertilizer bomb. So, they’re chemically, there’s a lot of chemical energy, but it’s chemical energy that is not easy to activate. So, that would be one thing we could add. Another thing we could add would be high energy phosphate bonds. That’s the kind of energy that our cells use. So, we actually use some experiments add A.T.P., which is a source of phosphate of energy.

[slide animates on new bullet point to – select cultures which use the most energy]

And then we can – once we have all these pieces in place, we can select among a set of cultures or tubes, vessels, for those that use the most energy. So, let me kind of talk you through that because it may not be obvious how this works.

[new slide titled, Artificial chemical ecosystem selection, featuring an illustration of a plate with wells in it named Gen 1 Seed with a red arrow pointing to the right that is titled incubate]

So, we can take a plate with 96 little wells in it, and we can put minerals in every well and soup and – and our energy sources and we can incubate it for a period of time.

[slide animates on a new version of the plate showing those wells that created the most energy with arrows from those wells going into a test tube where they are mixed]

Then we can measure how much energy was used up, or conversely, how much energy is left in the – in the well. And we can identify by a calorimetric reaction which wells had used the most energy. We can then take the grains out of those wells, mix them together –

[slide animates on a new plate from the most energetic wells that have been seeded and named Gen 2 and then this plate is also incubated]

– and feed them into a new plate, along with virgin grains of the same kind, a new soup and new solutions.

[the slide animates on a new plate showing the wells in this new generation that have the most energy that are again taken out and mixed]

And then we can incubate again and again select. Every generation we select whichever wells, whichever cultures have the strongest response, the – the strongest signal.

[the slide animates on that this process is repeated seeding and mixing over many generations until it becomes wells of assay]

And the idea is that if we do this for many generations, we will see that the mean value of the trait, the mean amount of energy used, will gradually change. This would indicate a response to selection, which is hard to imagine how you can get a response to selection unless there’s some heritable system that is being preferentially passed on during the selection process.

So, these experiments are just beginning, so I won’t give you any results.

[David Baum]

We – we – we’ve just got all the equipment in, and we have – I have four students working over the summer who are going to be cranking through a bunch of these experiments.

There’s a second approach that we’ve already tried which is a little simpler but maybe a little less direct. And this is a natural selection variant. So, in a natural selection variant, we exploit the fact that if you serially transfer from one vial to another vial a subset of the mineral grains in the first vial, if you keep doing that many, many times, you are inadvertently selecting for any SLIMEs that might be present that – that are good at getting from one grain onto another grain. You are – you are selecting for the ability of systems to survive and – and spread from grain to grain because every generation you only take let’s say 10% of the grains to move them to the next generation. So, if you can – if you have a system on those 10% that can spread onto the other 90%, then those will get enriched over time.

And they, because these are chemically active systems, you would expect them to change the composition of the soup. So, what we can do is measure various features of the soup.

[slide titled, A natural selection approach, with the bulleted list of – serial transfers automatically favor SLIMEs that are better at propagating themselves from grain-to-grain and adaptive evolution of SLIMEs will change the soup. Additionally, there is an illustration at the bottom of the slide showing the process that was described above of transferring SLIME to vials]

And so, the – the thinking is that let’s say we did have some of these, of SLIMEs. Lets say there’s a red SLIME and a yellow SLIME. And the red SLIME maybe isn’t quite as good as the yellow SLIME at getting from grain to grain. Every time we take a subset of the – of the grains, and over time we’ll eventually dilute out the red ones and we have only yellow ones left. And since the yellow ones are going to have different chemistry than the red ones, we would correspondingly expect to see just changes in the chemistry of the solution after the incubation.

[new slide featuring a series of photos of students doing the above-described experiments along with photos of the vials as they are being used in the experiments]

So, here are a number of just pictures. These are quite simple experiments, and I’ve been very – I’ve been blessed with a great team of students who’ve been working on this. Mainly undergraduates, such – I see one of them in the room. So – so, this is, you know, making solutions. This is – these are the vials we incubate them in. This is actually grinding up the minerals. This is working in anaerobic chamber to avoid the minerals becoming oxidized in air. These are – this is an incubation happening. So, these are relatively simple experiments, and we’ve done this –

[new slide titled, Look for a response to selection, featuring two graphs of the data from the experiments one titled, what we have seen so far (no response) with data points all over the graph, and a second graph titled, what we would like to see, showing a group of data along a straight line with some limited variations]

– one experiment we’ve done for a hundred generations. And what we hoped to see would be a trend over time. Something like this. This is actually the result of a – these experiments are very similar to experiments that are sometimes done on microbial communities, which is what this is from. But this is what we found, which is we have 10 replicate seed lineages. None of them showed a trend over time. So, this tells us that we haven’t got the right conditions, if indeed we can find the right conditions.

There’s another thing –

[new slide titled, can also look for heritability, featuring a bar graph and a point graph, the bar and point graphs show generation 48 and 49 in ten replicates looking for heritability]

– we have been looking at which is maybe a more sensitive indicator that we have something interesting, which is to see if our systems show heritability. So, what this graph shows is there are pairs, generation 48, 49 for experimental replicate one. 48 and 49 for experimental replicate two. Now, if these systems are heritable, then generation 49 will resemble the – the character of its ancestor in generation 48. Its ancestor being the vial from which –

[David Baum]

– we took the grains. And so, you might expect to see a positive correlation between the values in these two generations, and that would indicate that there is something in the vials that is carried from generation to generation and effects a chemistry. However, in this case, while there is a very slight net positive slope, it is not significant.

[return to the can also look for heritability slide with the two graphs]

So, as of right now these experiments haven’t yielded any evidence of a SLIME, but we have a lot more conditions we want to try. We have a lot of ideas for interesting soups, interesting energy sources, and interesting traits to measure.

[new slide titled, Future work, with the bulleted list of – new soup and mineral combinations and packed-bed flow reactor and featuring two photos, one of freshman Alex Plum holding up a vial in front of a bulletin board and another of an experiment being done]

So – so, these experiments are just getting underway. And we have some new approaches that we plan to take in the future. I’m just highlighting here a student who designed a very beautiful little device that we may, hoping we’re gonna be rolling out to start working with in the future which does kind of a natural selection type experiment. It’s called a packed-bed flow reactor.

[new slide titled, if it works, featuring the bullet point that it will be really exciting!]

So, the question we have to ask is –

[laughter]

– well, if it works, it would be really exciting, okay?

[David Baum]

Now, okay, I’m going to be a bit over the top.

[laughter]

[scene from the movie Frankenstein on the screen]

[Dr. Frankenstein]

It’s alive! It’s alive! It’s alive, it’s alive!

[the movie scene is paused, and a speech bubble is put over one of the men restraining Dr. Frankenstein that says – Calm down. Its just SLIME.]

[laughter]

[David Baum]

So – so, okay, yeah it is just SLIME. And so, I think we should – we should pause a second and say this would not be a case of creating life because if this works –

[return to the if it works slide now with the new bullet points – we can figure out the SLIMEs chemistry and see what it shares with cellular life and what it does not, we can start the same experiment many times and see how repeatable the outcome is, and we can map out conditions yielding SLIMEs and use it to clarify where in the Universe life might be]

– this indicates that these kinds of systems are formed all the time. It wouldn’t be creating life, but it would be detecting these very simple earliest forms of life, which I do think would be really exciting. Not in that kind of crazy sense, I hope. And I hope I won’t get carted away. But so, the first thing about this is that we get the opportunity to actually test the theory because the theory is great, but wouldn’t it be great to actually have SLIMEs, to have these self-propagating chemical systems and see if they can evolve and see how they evolve.

[David Baum]

We can also ask interesting questions about the extent to which the actual chemistry we find in these systems resembles the chemistry of life. Is life the way it is just because it had to be this way? That every time life evolved on a planet like ours it would have our kind of chemistry? Or could it be different every time? Could there be chance that determines exactly how these chemical systems work?

Along the same lines, we could ask how repeatable are these phenomena. And, interestingly, we could ask what conditions are capable of yielding SLIMEs or life-like systems and which conditions aren’t. And that could give us really nice clues as to where else on the planet, on the planet, where else in the solar system and the universe we might expect to find these systems being initiated.

So, this project is – is now funded by N.A.S.A. –

[slide titled, CESPOoL – Chemical Ecosystem Selection Paradigm for the Origin of Life, with an illustration of various trays of wells of generations stacked upon one another and the aim of the project being to deploy the methods of experimental evolution to test the hypothesis that evolvable life-like chemical systems associated with mineral surfaces arise easily]

– and we call it, the – the – the approach we call chemical ecosystem selection. So, it’s a Chemical Ecosystem Selection Paradigm for the Origin of Life. N.A.S.A. likes catchy acronyms, this is CESPOoL project.

[laughter]

And obviously we’re looking for SLIME in our CESPOoL. And so, this is a multi-institution effort –

[new slide titled, At least, featuring the bulleted list of the goals to develop easy high-throughput C.E.S. protocols and devices and share them widely, and the scientific community can search for conditions conductive to the spontaneous emergence of life]

– that aims to develop these chemical ecosystem selection methods, obviously look and hope that we can find evidence of these life-like systems. But even if we can’t, our hope is that we can create protocols and devices that we can distribute to many, many labs and begin kind of a more systematic search because the number of chemical soups, the number of minerals –

[David Baum]

– the number of environments that you could try is infinite. And in the same way that it would be nice that S.E.T.I. you know, uses many citizen scientists to search in data sets, we like the idea of being able to get many labs on board searching for a set of conditions that generate these systems even if we can’t find them right away. So that’s our image.

This is a consortium. So, we’re the lead institution –

[slide titled, N.A.S.A. CESPOoL Consortium, that lists the member institutes and researchers – University of Wisconsin-Madison, Boston University, Brandeis University, Portland State University, Rensselaer Polytechnic Institute, Santa Fe Institute, University of Minnesota-Twin Cities, and University of Utah]

– but we have seven other partner institutions, all of which are doing variants on these experiments or developing models and theory to back them up.

[new slide titled, Thanks!, with a list of researchers and funders for this research]

So, I am excited to see where this leads – where this leads, and I hope you are too. But, of course, we wouldn’t have got here without a lot of people doing a lot of useful things. So, in addition to my collaborators in the CESPOoL consortium, I have a lot of colleagues here who have been terribly important and useful for ideas and techniques and support in the lab. I have a wonderful lab. Mainly I want to particularly pick out Lena Vincent, who’s my graduate student who’s been doing a lot of the work kind of getting us geared up for these experiments. And a big team of undergrads. So, these are all undergrads and one special student in there. And then, collaborators and our funders, N.A.S.A., N.S.F., and thanks to my department and the Wisconsin Institute for Discovery. And, lastly, thanks to you for your attention. Im happy to take questions.

[applause]