We’ve talked about the very origin of life, but certain transitions along its subsequent history were incredibly important. Perhaps none more so than the transition from unicellular to multicellular organisms, which made possible an incredible diversity of organisms and structures. Will Ratcliff studies the physics that constrains multicellular structures, examines the minute changes in certain yeast cells that allows them to become multicellular, and does long-term evolution experiments in which multicellularity spontaneously evolves and grows. We can’t yet create life from non-life, but we can reproduce critical evolutionary steps in the lab.
Support Mindscape on Patreon.
William Ratcliff received his Ph.D. in Ecology, Evolution, and Behavior at the University of Minnesota. He is currently Associate Professor in the School of Biological Sciences at Georgia Tech. Among his awards are a Packard Fellowship and being named in Popular Science‘s “Brilliant 10” of 2016.
- Web site
- Google Scholar publications
- Packard Foundation bio
- “Experimental Evolution of Multicellularity,” Ratcliff et al. (2012)
- “De Novo Evolution of Macroscopic Multicellularity,” Bozdag et al. (2021)
[accordion clicktoclose=”true”][accordion-item tag=”p” state=closed title=”Click to Show Episode Transcript”]Click above to close.
0:00:00.0 Sean Carroll: Hello, everybody. Welcome to the Mindscape Podcast, I’m your host, Sean Carroll. Life is complicated, I don’t mean your life, my life, those are also complicated, but as we all know, biology is complicated, the history of biology is complicated. Here on Mindscape, I’m interested in very big questions like the origin of life, but we also have the evolution of life, that’s interesting all by itself. And so there’s a bunch of phase transitions that we like to talk about that happened over the course of the evolution of life. One such transition was the origin of eukaryotes. Eukaryotes are cells that have nuclei in them, and we think these days that that came about from two different kinds of cells getting together, joining together, sharing their DNA, distributing it in the cell in interesting ways, you have mitochondria and then you have a nuclear DNA, and it’s all a big part of the interesting kinds of life we have on earth. I shouldn’t say interesting, it’s all interesting, the complex kinds of life we have on earth. Eukaryotes compared to prokaryotes, which don’t have nuclei, have a lot more capacity, a lot more potential for being complex such as being multicellular, and once you go from a single-celled organism to a multi-cell organism, then the possibilities of complexity grow enormously.
0:01:20.7 SC: So today’s guest, Will Ratcliff, is a biologist who is one of the world’s experts on exactly this transition from unicellularity to multicellularity in the evolution of life, not just studying what happens there in the record of actual life on earth, but doing artificial evolution, or what some people call directed evolution, doing an experiment in the lab where he takes, he and his group, take yeast cells and they let them evolve, they give a little bit of selection pressure to grow bigger, and they say… They don’t poke in there, the DNA, they just let mutations happen and watch what happens to these initially unicellular yeast organisms. And what happens very quickly is they become multicellular. In some sense, which will hopefully try to make clear in the podcast, yeast and many other eukaryotes are ready to be multicellular, in some sense, a little nudge pushes them in the right direction, but there are surprises that come up along the way.
0:02:22.9 SC: There are different ways to be multicellular, and there’s a big difference between being multicellular in the sense of just having a lot of cells sticking together versus being truly differentiated with different organs and things like that. And even though the experiments have not been going on very long, there’s still thousands of generations that we have the ability to look at, and it’s taught us something about this crucially important transition, which might have implications for life elsewhere, the idea of life, the future of life, even now that we human beings have the ability to actually go in there and poke at the genome of different kinds of organisms and see what happens. Maybe there’s whole kinds of genetic information schemes that would lead to different kinds of organisms that evolution was not able to find. I don’t know, I’m just a physicist and podcast host, but Will is an expert on this, so let’s go.
[music]
0:03:35.7 SC: Will Ratcliff, welcome to the Mindscape Podcast.
0:03:37.7 Will Ratcliff: Thank you, I’m glad to be here.
0:03:39.4 SC: So multicellularity, a big thing in the history of life, we… And no planning on my own part, I don’t plan series of podcasts in any interesting way, but we have been talking a little bit of the origin of life, recently did Batra Qachar about what we’ve learned about old life from paleo-genomics and so forth, but that was all single-celled stuff, basically. So why don’t you put us into context? When did multicellularity arise? What do we know about its origins?
0:04:11.3 WR: Yeah, no, I’m happy to be on this path of understanding life’s origins and the evolution of complexity and in some sense, they’ve already done the hard work because once you have a cell, you’ve already sort of… You’ve done 95% of it. I actually agree with Nigel Goldenfeld in his recent interview where he said, you know, “Once you have cells, who cares about elephants?” To some extent, I agree with him, to the other extent, I’ve made my entire career on disagreeing with him. So multicellularity is when you have an organism that’s composed of multiple cells, and it opens up all kinds of opportunities for complexity, because those cells can take on different roles, you can get all sorts of novel ecologist arising and you can have organisms with all these specialized functional parts. When multicellularity arose is not really… There’s no one answer because multicellularity actually has evolved many times in different lineages on earth, so we actually have fossils…
0:05:04.9 SC: What does many mean, like five times or 500 times?
0:05:10.7 WR: In between the two, so it’s probably around 100, although we actually don’t have a really robust estimate.
0:05:18.2 SC: That’s a lot, though. Okay, so that’s…
0:05:20.6 WR: It’s a lot. Some lineages, like the green algae, it’s evolved more than 25 times, independently in just green algae alone.
0:05:27.0 SC: But isn’t it true that life on earth was unicellular for a really, really long time?
0:05:34.6 WR: Yeah, so the majority of life… Life started at pre-cellular and then once you have sort of it becoming cellular, that’s really sort of almost a sea change in how life evolves, information is transmitted vertically, and you have sort of cellular revolution. And that was probably on the order of 3.8 billion years ago or even earlier, that’s about the origin of LUCA, the last universal common ancestor, and the coalescence of all things can be traced back to that point, but there was probably some cellular life before that, but it’s very, very hard to study anything that’s that old. And…
0:06:07.9 WR: Yeah, so the majority of the diversity of life that we have out here today is also single cell, but different single cell lineages begin exploring with multicellularity at different time points. So cyanobacteria began forming chains of cells in which you’d have cells that are specialized for carbon fixation, for photosynthesis or nitrogen fixation by basically taking nitrogen from the air and turning it into ammonium, which can be used in building proteins and stuff and that happened around 2.1 billion years ago. And then you have… And different types of bacteria played around with multicellularity between two billion and one billion years ago, and they’ve all remained pretty small and simple, and then you have eukaryotes arise and eukaryotes are a special kind of single cell organism. They’re actually the result of a fusion between archaea and alphaproteobacteria, that alphaproteobacterium was at one point a Symbion and became domesticated as our mitochondrion to sort of provide energy for our cells, and these eukaryotic cells have really sort of run away with, I don’t know, forming different kinds of organisms that bacteria and archaea don’t really seem to be able to form.
0:07:18.3 SC: Sorry. What you just said about the origin of the eukaryotes, the cells with nuclei in them, you sounded pretty confident. I’m not sure from my own naive physicists reading, are we pretty confident that we really know how that started and why it started, or this is…
[overlapping conversation]
0:07:40.4 WR: We’re getting more and more confident.
0:07:40.5 SC: Yeah, okay. Good.
0:07:41.0 WR: So as we find these archaea from the lokiarchaeota and other Asgard related… It’s a great acronym right? I’m not sure what it stands for, but there’s this clade of archaea that have been found in sea floor settings in, I guess, the near Arctic, and these things have all sorts of eukaryote specific cell biology, things that we thought only eukaryotes have, and it turns out that they are archaea. And if you look at them on a phylogenetic tree, which is like a genealogical tree, but applied to vast timescales and not just your local family, then you can see that they actually nest as a sister group to eukaryotes. So we’re pretty sure that archaea… If you begin to disentangle and unpack how eukaryotes arose, you really see that we basically have… That eukaryotic cells have a mixture of archaeal stuff and bacterial stuff, like the lipid membrane that we have surrounding our cells is actually bacterial. It’s not archaeal, they use a different kind of chemistry in their lipids, which make the membrane of the cell. And we think that actually came over from the mitochondria.
0:08:51.5 SC: So the nucleus was the archaea, and then the big puffy part was the bacteria or vice versa?
0:08:58.9 WR: So actually, I think those details, we don’t really know. Those are still the kinds of things we’re working out, but I think a better way of thinking about it is that the cell as a whole, like the macros, like the larger symbion was the archaea and that sort of got smaller bacteria living within it, and those bacteria that co-opted to being a part of this new eukaryotic cell, they can’t divide on their own anymore. They still have a remnant little genome. So if you look inside a mitochondria, it has its own genome, which is one of those things that made people realize 40 years ago, oh my gosh, these things were once free living, separate cells.
0:09:33.8 SC: Right. So this is… I know probably a lot of people in the audience know this very well, but maybe there are some who don’t, and it’s so amazing, I really wanna just sit and dwell on it, that there’s two different sets of DNA in every one of our cells, there’s our DNA, and then there’s these little free riding, not free riding, but symbiotic mitochondrial DNA in every one of our cells, and so we sort of genetically passed down both from generation to generation.
0:10:00.8 WR: Yeah, and the DNA inside of mitochondria is totally stripped down. If you imagine that it was at one point a car, now it is just the steering wheel, or I don’t know what the better analogy is. But it’s a 20000 DNA basis when a free-living bacteria is typically on the order of a few million. So they’re tiny, in fact most of the DNA from the genome of the mitochondria has been exported and is now residing inside our chromosomes.
0:10:29.9 SC: Okay, alright. So we co-opted the useful parts of it, and there’s… Some of the useful parts, other useful parts are still hanging out in the mitochondria.
0:10:37.8 WR: Yeah. Certain elements of it, it seems like it’s kind of hard to move them into the… Through those membranes into the mitochondria and it’s better just to have them made locally, but other stuff you want top-down host control, and so you just move that stuff into the domain of the host.
0:10:51.0 SC: And it was this… We love phase transitions here on the Mindscape podcast, so this leap from prokaryotic-ness to eukaryotic-ness was a big phase transition, and you’re saying that that enabled new kinds of multicellularity or made it easier?
0:11:08.3 WR: It certainly seems that way, although we don’t have a first principles, universally accepted explanation for why that’s true. So when you look at the tapestry of multicellular organisms, there are bacterial and archaeal multicellular organisms but they’re very, very simple, they’ve been around for a really long time, billions of years, and they don’t seem to have really done much beyond making small groups of cells. Within eukaryotes, though, you have the entire complex tapestry of life that you see when you look out your window, right, you see trees, you see things growing on those trees, you see insects buzzing around those, you see mushrooms growing off of logs, all of those things are eukaryotic multicellular organisms. And multicellularity, one of the great features of it, is it’s like the anti [0:11:51.0] ____ thing. And unlike all multicellular families being alike in the same way, they’re, in fact, all very, very different. Every multicellular lineage does multicellularity in its own specific way. Free living algae in ponds that form spherical colonies of multicellular cells that spin around with the flagella that keep them mobile are very different from plants which are essentially growing into the air and competing for light, which are very different from animals, which are very different from fungi.
0:12:18.4 WR: So in some sense, it’s a challenge to try to come up with universal rules for understanding how multicellular organisms evolve and become more complex. On the other hand we actually… We can still do that, but you have to keep in mind that every lineage has its own ways of becoming multicellular that depend on its specific ecology and very importantly, the cell biology of the ancestry.
0:12:40.0 SC: Is it just too simplistic to imagine that being eukaryotic provides a bit more flexibility when it comes to being multicellular in the sense that a good, robust multicellular organism will… Each cell has the same DNA, but it gets expressed in different ways, and being a eukaryote enables more flavors for that.
0:13:02.2 WR: Yeah, that’s actually… You kinda nailed the explanation that people have, which is that eukaryotic cells have a much larger number of types of gene regulation available to them, and it’s easier to get cellular differentiation in eucaryotes. They also tend to be larger cells with more complex forms of cellular morphology. The ancestors of eukaryotes often have, at least if we look at… The best work in this area has been done by people looking at the unicellular relatives of animals. So animals evolved roughly 800 million years ago, and if you go back to the sort of family tree, just one step above animals, there’s things which are common ancestry of animals and things which are state single-celled. And if you look at those things that a state single-celled, they often have these complex life cycles where they can exist as an amoeboid and crawl around, which looks a lot like one of your blood cells that’s going off and trying to kill pathogens. It can exist as a flagellated state where it’s swimming, which looks a lot like a sperm cell. It can exist in many different temporal types of cell differentiation, which you could imagine gets reworked from being expressed in time to expressed in space. And now you have all these sort of modules to play with for building a multicellular organism.
0:14:16.9 SC: Yeah, and it’s also kind of cool because in some sense, a little bit of complexity at the micro-level is enabling much greater complexity at the macro level.
0:14:26.3 WR: Absolutely. But I wanna caution that this is consistent with what we understand, but I don’t know if it’s… It’s not universal accepted yet. Because bacteria and archaea are extraordinarily complex and diverse, and it’s kind of hard to find something that eukaryotes can do that they can’t do. That make sense?
0:14:45.5 SC: Yeah, no, that makes perfect sense. So this transition to eukaryotes was about 800 million years ago?
0:14:53.3 WR: That was probably more on the order of about two billion years.
0:14:55.7 SC: Oh, two billion years ago, okay.
0:14:57.1 WR: Yeah, animals were roughly 800 million years ago.
0:15:00.2 SC: I see. Okay. So eukaryotes… So the story that I have in mind of it taking a long time for multicellularity to arise sort of gets divided into… It took a medium long time to be eukaryotic and then another medium long time to be multicellular, right?
0:15:18.3 WR: That’s right. Yeah, so in the eukaryotic space, eukaryotes arose around two billion years ago, and then roughly a billion years later, you start to see the first lineages of bonified eukaryotic multicellular organisms arising and still persisting to the modern day. So you see… And this really occurs in red and green algae, interestingly enough. So you get these small little algae that form branched, topologically structured like little trees, branched multicellular groups, which if you look at the fossils and look at modern day ones, oh, my gosh. They look basically the same.
0:15:52.7 SC: Yeah, and this whole story is just bizarre to a physicist, because we think of systems as being characterized by time scales. And if you don’t know any better, there’s a certain time scale that governs everything that happens. And presumably, the time scale for a micro-organism is lifetime or it’s time to split or something like that. And those numbers are much, much smaller than a billion years. So for something interesting to happen, but for it to take a billion years to happen is just a reminder that physics is not very good at estimating certain things.
0:16:33.1 WR: Right. And the time scales that things… I think it’s something that we keep learning in evolutionary biology, is that the potential for something to evolve in a certain way is often kind of unlimited over geological time scales. Things can evolve very, very, very quickly given suitable selection. And often when you see stasis, it’s simply because you don’t have any selection, any significant selection on the organism’s state. And so during that two billion years to roughly 540 million years before the present, you had an era of very, very low environmental oxygen. And so that actually, in some ways, changes the expectations of whether you expect to see multicellularity and what types of multicellularity you expect to see. And after… Yeah.
0:17:24.9 SC: Well, I guess that was gonna be my question. I’m a big fan of multicellular organisms myself, being one, but from the evolutionary point of view, what was the benefit to becoming multicellular? And I guess, you’ve already started saying something about oxygen.
0:17:41.9 WR: Well, yeah. There’s a lot of different benefits to multicellularity. This is, again, goes back to reverse Anna Karenina problem, where every lineage has its own reasons to become multicellular, so there’s no one reason to do it. In some cases, you have organisms that are forming groups in order to avoid being consumed by predators, ’cause predators are filter-feeding and they have a small mouth, and if something bonks against their face and can’t get in their mouth, then they’re not eating. So it’s good to form a group. In other cases, you have organisms that are cooperating metabolically, they’re excreting stuff into the environment, and if you can get a clump of cells that are all cooperating and excreting enzymes into the environment, you can degrade environmental stuff much better than if it’s a chaotic mix of different species and things that are maybe not producing the enzymes, etcetera. There are…
0:18:27.0 WR: There’s actually a whole list of… We’re working on a review paper on this right now. I could go on and on, but it doesn’t really matter. What matters is there’s lots of different reasons to become multicellular. Maybe one that’s really relevant to algae or photosynthetic organisms is that light is this very valuable resource, but it’s directional. So if you can over-grow your competitors, you get the food. And that’s why we have 300-foot tall trees. They’re not getting more sun by being 300 feet tall than something sitting on the ground, they’re just getting to it first.
0:18:55.0 SC: Well, they’re not being overgrown by somebody else when you’re the tallest one in the forest.
0:19:00.2 WR: Exactly. So it’s an arms race. Trees are the result of a 450-million-year-long arms race.
0:19:06.1 SC: But then I do want to let you go into more about the oxygen stuff, ’cause that clearly was relevant.
0:19:10.9 WR: Yes, yes. Okay. So oxygen is very important in the sense that we really see sort of… If you look back at the history of multicellularity, you see these sort of experiments in… Let me just talk about how oxygen happened on Earth first. The great oxidation event, which I think is a great term, very much named by geologists because it took several hundred million years for this event to occur, occurred roughly 2.1 billion years ago in that era, a couple hundred million years, you have cyanobacteria cranking out so much oxygen that we go from a world that basically is anaerobic to one that actually has oxygen in the atmosphere and in the oceans. But it’s at very, very low levels. Maybe 1% of modern levels, maybe lower. It’s at very low levels and it stays low for the next one and a half billion years. And then in the Phanerozoic era, which is about 540 million years before now, it starts to really rise, and that’s actually concordant with this explosion in multicellular diversity. You have the Cambrian explosion in animals happening right around that time, plants, land plants kind of invade not that much longer. And fungi, they’ve been around for a while. We actually don’t have a great understanding of the origin and diversification of fungi, but fungi seem to really also diversify around this time as well.
0:20:27.1 WR: And so we’ve long thought that oxygen was this kind of catalyst for multicellularity, that if you give a system oxygen, you’re gonna enable the evolution of larger, more complex multicellular organisms, and there’s a very good physical reason for this, which is that if cells depend on having oxygen and you form a group, then you become diffusion-limited and those internal calls don’t have access to very much oxygen and therefore, that might set a maximum size on an organism, because basically, if you can’t get oxygen to the interior, maybe those calls can’t survive at all, that sets a maximum size to the organism. And so until you can invent a circulatory system, you’re stuck with diffusion and therefore size is just directly a function of oxygen concentration. We thought this way for a long time. This isn’t something we have actually tested, though. So a post-doc in my group, Ozon Bozdoc, did this cool experiment where he took a yeast system… Maybe this is a good time to introduce it.
0:21:21.2 SC: Well, keep talking. We’ll go back to the yeast. Don’t worry.
0:21:23.9 WR: Alright. Good, good, good. In any case… Although, okay.
0:21:29.5 SC: You can’t explain it without…
0:21:31.6 WR: It’s hard for me to explain our change in thinking about oxygen without referencing the experiment, ’cause that was what helped us understand… Should we go back and just talk oxygen in general?
0:21:39.2 SC: Yeah. No, let’s just note that the point is that… When I talked to Betuil, we talked about the great oxygenation event two billion years ago, and what you’re emphasizing is that’s nothing, there was another growth of oxygen by a factor of 100 later on and there’s this… It’s not completely clear why that happens, and we’re gonna illuminate maybe why it happens, but talk about yeast. So let’s pause there and talk about yeast for a while. What’s a yeast and why do you care about it?
0:22:13.4 WR: So yeast are a single-celled fungus and I care about them because they’re a really wonderful model organism for laboratory experiments and…
0:22:21.1 SC: And we all… Listen, we’re in the middle of a pandemic and everyone is baking bread, so we’re all familiar with yeast. But so, this is a… So it’s a fungus which is… What kingdom is that in? That’s a plant. That’s a… No, sorry, not a plant. It’s a fungus, obviously, but it’s a eukaryote…
0:22:39.5 WR: It’s a eukaryote. Yeah.
0:22:41.2 SC: Yes, eukaryotic fungus. Single-celled.
0:22:44.9 WR: Yeah. It’s a single-celled eukaryote and they actually… The funny thing about the yeast… Yeast is a generic term for any single-celled fungus. The yeast that we use is Saccharomyces cerevisiae. And so there’s many different yeasts, and they can actually be very, very diverse. There’s yeast from very different branches of the fungal tree of life. It actually seems to be a kind of a common strategy, and this is the only multicellular group that does this, where you’ll have things which formed filamentous multicellular fungi, they’ll actually evolve to form single-celled yeast. In fact, the yeast that we work on, Saccharomyces cerevisiae, which is the bakers’ yeast, it’s the most common yeast used by humanity, it’s sort of domesticated by humans for brewing beer and baking bread. It’s also, in many ways, one of the most well-studied and extensively used workhorses of molecular and cellular biology in the research community. And it’s very, very easy to make it do what you want. You can basically do crazy synthetic biology experiments very easily in yeast because you can just kind of stick genes in them and they’ll express them and it works.
0:23:51.5 SC: That’s clearly very useful. And also, I get the impression that even if yeasts are unicellular, they’re kind of living on the edge of wanting to be multicellular.
0:24:01.0 WR: I think that’s probably true. Certainly in our case, our yeast descended from a multicellular ancestor, but that was several hundred million years ago, and during that time…
0:24:09.8 SC: So what does that even mean? So it used to be…
0:24:11.5 WR: And during that time they’ve really lost much of their genome that we know multicellular fungi have. They seem to have jettisoned it all.
0:24:18.3 SC: Okay. So they devolved in some sense?
0:24:21.9 WR: Yes, yes. Yes. And this is very rare in this palliative multicellularity. You don’t see unicellular descendants of animals or plants running around nature, but fungi seem to do this all the time.
0:24:32.2 SC: I see. So the yeast you’re dealing with were formerly a multicellular organism and now they like to be unicellular.
0:24:40.1 WR: Yeah. They’ve been unicellular for a long time.
0:24:42.3 SC: And then you… So it’s not completely surprising that you can prod them in different ways and mutate them and get them back to being multicellular?
0:24:52.0 WR: Yeah. It may not be that surprising, though I don’t think they’re actually reactivating latent ancestral pathways of multicellularity. I think it’s actually just pretty easy for them to… In our case, we understand the cell biology of how they become multicellular in our lab very well, and you can just break the machinery that normally allows daughter cells to separate from a mother cell after they’re done dividing, and then you just get these tree-like multicellular groups forming.
0:25:19.9 SC: So you’re not actually bringing back to life the multicellular organism that it used to be, you’re creating a different kind of multicellular yeast.
0:25:27.0 WR: Exactly.
0:25:28.3 SC: And these are the snowflakes. Is that right?
0:25:30.3 WR: Yes.
0:25:31.1 SC: So tell us about the snowflakes.
0:25:32.9 WR: Okay. So about 10 years ago when I was actually a fifth-year grad student, sort of, I guess, avoiding writing up my thesis, I started a… I think you’ve known them. I know them in my lab now. Sorry, if they’re listening to this. I started a collaboration with Mike Travisano at the University of Minnesota, to basically… We had a coffee chat where we were like, “What’s the coolest thing we could do in the lab?” After we… Discussing origin of life has been too difficult for us, biologists, and we’re not chemists, so it’s… Yeah, we don’t know how to do that. Like, maybe we should play around with evolving multicellularity. And so over December, I started doing these experiments selecting on single-celled yeast to form groups. And we did a really simple selection experiment in the lab, and these selection experiments are kind of like selective breeding in the way that we have… We’ve domesticated dogs and we’ve made wheat and corn from wild relatives, it’s basically a selective breeding experiment, the fancy term now is either experimental evolution or directed evolution. And so, we did this really simple selection experiment where we took our yeast, and every day we dilute them 100 fold into fresh media. Those cells grow up, and then we put them through sort of a race to get to the bottom of the test tube.
0:26:47.1 WR: So we give them a few minutes sitting on the bench to get to the bottom of the test tube and then any of the ones that are at the bottom of the test get transferred to fresh media, everyone else is killed. And so, this really incentivizes them to form groups, because groups settle through liquid media faster than single cells.
0:27:00.3 SC: So sorry, just to be clear, the mutations in the DNA are completely natural, you’re not inducing them…
0:27:06.8 WR: That’s right. Yeah.
0:27:07.4 SC: But the selection pressure is what you are… That’s the directed, indirected evolution, right?
0:27:12.6 WR: Exactly, exactly.
0:27:13.1 SC: Okay.
0:27:13.2 WR: And we don’t really think necessarily that settling faster is the reason that things would become multicellular in nature, although there’s a recent work out of [0:27:22.1] ____ lab showing this for a relative… A unicellular relative of animals. But the way that we think of it is really just a very expedient way to select on size, and that’s what all these lineages that are experimenting with multicellularity are getting advantages from. There has to be an advantage to science, otherwise things stay single celled. Like, most of them, they stay single celled. Some of them have an advantage of forming groups, and those ones are selected to form groups, and so, we’re just doing that selection in a really efficient way. We can select on a million clusters in a few minutes of sitting on the bench and a 10-second pipette transfer.
0:27:56.9 SC: So you’re not… I mean, alternatively, in a synthetic biology sort of context, we might go in there and remove a bit of DNA, a gene, and replace it with something else, but you’re… In this particular case, you’re actually just letting a mutation happen.
0:28:12.6 WR: So in the very first experiments, we didn’t know what would happen, and we just let these mutations occur and within a few weeks, these populations had all formed these beautiful little groups that we call Snowflake yeast. We came up with that term actually, ’cause they settle much like snow settling through the liquid media, and actually have this cool fractal like growth form, where if you break off a small branch, it’s self similar to the morphology of the larger group. You can imagine that they’re growing as a tree, and if you break off a small branch of the tree, it looks a lot like the bigger overall tree growth. And also, we were in Minnesota in December, so there was a lot of snow around.
0:28:48.0 WR: But you know, actually, all of our experiments that we do now, we start them out with using the synthetic approach to basically delete a gene that was the one that was most often mutated in the natural experiments at the beginning of our… To start our experiments, we just remove that bit of DNA to make them all snowflakes in the same way, and then we can basically replay multiple parallel runs of our experiments, but they’re all starting out in exactly the same multicellular state. These simple ancestral snowflakes.
0:29:14.9 SC: But is it safe to say that it’s just one gene you have to remove to make a unicellular yeast into a snowflake? I mean, that…
0:29:21.3 WR: That’s right.
0:29:21.4 SC: That gives evidence that it’s right there on the edge of being ready to be multicellular.
0:29:25.6 WR: It’s super easy.
0:29:27.2 SC: Yeah.
0:29:27.4 WR: It’s super easy, but that’s probably true for most unicellular organisms.
0:29:29.8 SC: Okay. Oh, okay, okay.
0:29:31.6 WR: So, yeah, we’ve played around with algae, and doing similar experiments in Chlamydomonas, which is a model green alga, it’s also very easy to form groups. And you can do similar things with bacteria and get them to form chains. And basically, cell division has to be done correctly in order to get one cell to pop off a daughter cell and have them both separate. And there’s a lot of ways to kind of break that division process in a way that gives you a simple group of cells.
0:29:57.3 SC: So, yeah. So basically, it’s very similar to being unicellular, but you’re sticking together, literally sticking together.
[chuckle]
0:30:02.7 WR: Exactly. Like, I think of them as naïve. They are not multicellular organisms, they are a dumb clump of cells.
0:30:08.5 SC: Yeah. Right. Okay.
0:30:09.1 WR: And in fact, to become a multicellular organism is an evolutionary transformation.
0:30:12.9 SC: Okay. And do these at all… They must occasionally appear in nature, if that’s the case, like, you must get a mutation that creates a yeast snowflake?
0:30:24.8 WR: I mean, certainly they do.
0:30:25.4 SC: Okay.
[chuckle]
0:30:27.3 WR: This mutation rate is fairly high, but we don’t typically see them floating around on rotting fruit or in the sap of oak trees, which is where you typically find Saccharomyces in its natural habitat. You don’t really find snowflakes, and I think the reason for that is that’s ecologically costly for them in this environment. If you’re making a living as a yeast and basically being moved from one sugar source to another, dispersal is a premium virtue.
0:30:52.7 SC: Right.
0:30:53.1 WR: And if you and… Your next eight generations of offspring are all stuck together in one big group, and a fly doesn’t come down and pick you up on its foot and transfer you, you’re dead.
0:31:01.6 SC: So it’s every cell for itself, basically.
0:31:02.8 WR: So I think it’s reverse bet hedging.
[chuckle]
0:31:04.4 SC: Okay. And… Okay, good. Are we then ready to talk about the oxygen, the role of oxygen, why it was helpful?
0:31:11.4 WR: Sure, yeah.
0:31:12.0 SC: Yeah.
0:31:12.2 WR: Yep. So our lab specializes in using experimental evolution to study key steps in the evolution of multicellularity and multicellular complexity. And our main model system are these snowflake yeast. So, getting back to oxygen, looking at the tapestry of oxygen and the history of Earth, we know that it sort of went from zero to maybe 1% two billion years ago, stayed very low for the next billion and a half years, and then it smashed up to modern levels around 500 million years ago. And we’ve long thought that oxygen was this fuel for the multicellular fire. That because it diffuses more deeply into tissues, the size of a simple multicellular organism without a circulatory system is just dependent on the concentration of oxygen in the environment and therefore, you have this sort of monotonic dependence. As you add more oxygen, you get bigger organisms.
0:32:04.1 WR: But no one’s really ever tested this, and there’s a couple of reasons, but one of them is that we haven’t had a model system that we could evolve of multicellularity. Something which we can put through a thousand generations and more in the lab and actually explore it’s evolution.
0:32:17.2 WR: We haven’t had a model system where we can toggle the way that they use oxygen, ’cause this actually turns out to be very important. If you assume that the cells in your organism all have to use oxygen to grow, well, then you really limit that… You’ve put an assumption to the system that’s very much gonna dictate the outcome, which is if you say, “If you don’t have any oxygen, that cell dies,” well, then you really restrict them to being small and there’s not much oxygen. But it turns out that the ancestral state of eukaryotes were called mixotrophs, that is they could ferment without oxygen when they had fermentable carbon or they could switch over to using mitochondria to respire when there was often available, and that’s a pretty sensical strategy in a world where oxygen is very, very low. You don’t wanna be an obligate aerobe if there’s not much oxygen in the world.
0:33:04.6 SC: So just to, again, to sort of linger on this ’cause it’s kind of cool. So since there is oxygen, if there is oxygen in the world, respiration is a useful way to bring energy to your internal cellular workings, and the alternative that you’re saying is fermentation, which again…
0:33:23.2 WR: That’s right.
0:33:24.9 SC: A lot of alcohol-related terminology here, but what exactly is fermentation in this context?
0:33:31.3 WR: Fermentation is just a metabolic strategy in which you break apart carbon polymers and you get some energy out of that.
0:33:38.5 SC: Okay.
0:33:39.8 WR: But you don’t use oxygen to burn them.
0:33:42.0 SC: So before there was a [0:33:42.1] ____ of oxygen, you could get energy from fermentation, if there was enough oxygen, it’s more efficient to just do respiration.
0:33:49.9 WR: Yeah, so oxygen provides two huge benefits, one, you can get a lot more energy out of the resource up to 18 times more energy, so you can get a lot more out of it. And two, some kinds of carbon compounds cannot be fermented but they can be respired. So in that sense, it’s kind of a co-factor in letting you eat something that if you don’t have oxygen, you can’t eat.
0:34:12.7 SC: Okay, good.
0:34:13.7 WR: Which is actually true for getting back to alcohol. Yeast can eat alcohol, but they need oxygen to do it. So when you’re making your beer, people have the air lock on there and they’re trying not to let oxygen get into their fermenter, and that’s because they don’t want The yeast to eat the alcohol.
0:34:30.8 SC: Got it. Okay, good. So I’m sorry, I interrupted you talking about the usefulness of the oxygen to the yeast.
0:34:38.3 WR: Okay, yeah. So what we have is a question. How does the evolution of multicellular size depend on the presence of oxygen? We’ve had this long assumption that it’s monotonic, that the more oxygen give them, the bigger they can get, and we have now this yeast model system where we can evolve them in the lab and we can go through about five generations of evolution per day, so in a year you’re at over a thousand generations, you can really push their evolution fast, and we could select on size very easily. And then we can actually mess with yeast, yeast are a great model organism, and we can basically make them fermentation only, so we can basically make them anaerobic, and we do that by basically breaking their mitochondria. They have mitochondria, we can break them, so their mitochondria don’t work anymore, they can only ferment. Then we can actually… We can make them respire only, so they can only grow using oxygen, or we can let them do both, we can let them be… The native state of yeast actually is to be a mixotroph just like the ancestral state of eukaryotes. They can either ferment or aspire depending on what that cell wants to do.
0:35:41.8 WR: So we took all three of these different kinds of Snowflake yeasts, which were otherwise identical at the start of the experiment, and we evolved them with selection for larger size. And we didn’t get what we expected, we expected that the more oxygen they had… And I should say, we also did this with different environmental oxygen, so we gave them low environmental oxygen or high environmental oxygen, and we gave them high by basically just bubbling that stuff through their tubes. So there’s a lot going on here. But what happens is when we evolve them, what we find is that unlike the expectation that the more oxygen they have, the bigger they get, we find that when they have no oxygen, that is they’re an anaerobe, or they have a low amount of oxygen, it doesn’t matter if they’re a mixotroph or an obligate aerobe or they have high amounts of oxygen, we end up getting this Nike swoosh, it’s not a monotonic increase. They get really big when they have no oxygen, they evolve to be really big when they have lots of oxygen, and when oxygen is limiting, they actually stayed small, didn’t really get very much bigger at all, despite a thousand generations of us every day selecting for bigger and bigger size.
0:36:47.3 SC: But is it only the fermenters that get big when there’s no oxygen?
0:36:52.0 WR: Actually that’s how we gave them no oxygen. Rather than removing oxygen from the test tubes, we just remove their ability to use it. So the fermenters are essentially in the no oxygen world in our experiment. And those guys got big, the ones which could use oxygen, if they had a little bit of oxygen, they were actually much smaller from the fermenters, but if we pumped in tons of oxygen to their test tubes, then they became very large.
0:37:18.7 SC: So what does that imply for us about the evolutionary history? You were saying that as oxygen increases, first it gets worse for the yeast. If [0:37:29.3] ____ a little bit and think that getting bigger is good, it indicates a healthy kind of unicellular yeast, as oxygen comes into the atmosphere, it’s initially bad, is that a safe conclusion?
0:37:43.5 WR: Yes, first let me say referencing bigger is healthier, I would say the bigger is necessary for evolving complexity, It provides opportunities for cells to evolve, different roles and functions, and in fact, Andy Knowle has written very persuasively that size itself is a driver of increased complexity. Once you have a big organism, now you really have an incentive to evolve the circulatory system and mechanisms of cellular communication and resource transport within the organism. So science probably came first and complexity probably followed, which is one reason that we’re so interested in science.
0:38:18.0 SC: This is a provocative statement. There’s a chicken and egg problem. Is it that once the cells get big, they want or need to be complex, so they can differentiate, or is it that once the cells get big, they have the capacity to become complex and that’s intrinsically useful?
0:38:37.5 WR: I would say neither. I would say once the groups get large, there is an evolutionary incentive to evolve the kinds of behaviors that drive increased complexity. So they don’t necessarily want to, there is no internal desire in a yeast cluster or a clump of cells in nature, but there is an advantage associated with that kind of differentiation once you have size, because once you have size you have all of these diffusion limitations that make growth very difficult and also it provides advantages. You have the opportunity for an inside outside environment. You can localize food to being in the center of your organism and therefore it doesn’t leak out. There’s all sorts of reasons why once you have a big group, there are advantages to evolving, essentially, the kinds of interactions and integration amongst the cells in the organism that typically drive increased complexity.
0:39:29.0 SC: Okay. And so then getting back to the evolutionary history, it sounds like we’re on the verge of saying that the first Great Oxygenation Event didn’t really help because a little bit of oxygen slows you down, but once it went way up, then it became beneficial, from… This is what your yeast experiments are teaching us, then it became beneficial to become more multicellular?
0:39:47.6 WR: Exactly. Exactly. That’s exactly right. That at first, we probably would have been better off without the Great Oxidation Event. And in fact, geologists call that period of time between two billion years and one billion years ago, the “Boring billion”, because there’s not that much happening in the biosphere that’s at least apparent in the geological record.
0:40:07.2 WR: I’m sure there’s all sorts of really interesting single cell microbial biochemistry and evolution happening, but you don’t see that really in the fossil record. And so during that low oxygen boring billion, oxygen may in fact have acted as a constraint. And I’ll just throw one more thing in there. The reason why we think that we got this Nikey swoosh effect is not just artifacts and metabolism or some sort of short-term proximate explanation, but we actually built an evolutionary model to understand this effect better. And it turns out that if you build a very simple first principles model, then you can re-capitulate this behavior and it’s extraordinarily robust. And the intuition actually is that oxygen is a valuable resource, and so if you have no oxygen, well, there’s nothing to compete for, if you give a system on a little bit of oxygen, well now all of a sudden, those groups that can better utilize that oxygen have an advantage. And in fact, that provides an incentive to get smaller and have a larger proportion of the cells in that group be capable of utilizing that oxygen to grow faster, to divide more, they tend to win in these evolutionary simulations, and then as you…
0:41:17.4 WR: And despite their lower survival, because we’re always saying bigger provides better survival, despite their lower survival, they still win by getting smaller, but then you pump in enough oxygen into the system that they can both get big and have high survival and use the oxygen well, and there’s no longer a conflict, they don’t have to get small to use the oxygen. And that’s really what we think is driving this suppression of multicellular size from oxygen, that small amounts effectively saying like, “Here’s a valuable resource, but if you’re big, you don’t get to use it very well.” And so that’s a very strong selective incentive to get smaller or stay small. And you could remove that by removing oxygen entirely or pumping tons of it into the system.
0:41:55.6 SC: So one of the lessons that I got from talking to Betuil was these transitions that had a big effect on life, like oxygen, there’s a give and take. There’s sort of geological or even astrophysical things that happened that changed conditions on earth, that make the environment for the living organisms different, but then the living organisms feed back in quite an impressive way to what is going on on Earth. So can we say something about which is more important for this second big boost in oxygen? Was it that life made more oxygen, photosynthesis, or was it something more geochemical going on?
0:42:32.9 WR: I am not the expert on this. I would defer to somebody like Chris Ryan Hart who really studies this problem, but my understanding of it is kind of a layman’s understanding, is that it’s a combination of the two, that you had big changes in the Earth’s temperature, you had Snowball Earth occurring at about that time. You have big changes in the composition of the oceans being bacterial dominated to algae dominated, you have sort of these combination of biotic and global geochemical cycles occurring, which combined drove this transition to a far higher atmospheric oxygen concentrations.
0:43:06.2 SC: Okay, good, good. Alright, so here we are. How long ago again?
0:43:12.0 WR: We started these experiments in the very beginning, about 10 years ago.
0:43:14.6 SC: No, I mean, how many millions of years ago was this the… Was the…
0:43:17.6 WR: Oh, sorry, sorry. Was which one?
0:43:20.6 SC: Was the more oxygenation creating multicellular life?
0:43:25.6 WR: Yeah, about 500.
0:43:26.6 SC: 500 million years ago. Okay, so more oxygen and at least, if not multicellularity, then at least growth in the size of the unicellular organisms would be expected by then. And as you’ve made the case already, this is a good allower or it opens the flood gates to becoming multicellular in interesting ways. Do we know, did the actual yeast like the fungi you’re looking at become multicellular or is it other things that did?
0:44:00.0 WR: You mean around 500 million years ago?
0:44:00.6 SC: Yeah, 500 million years ago.
0:44:01.7 WR: Yeah, so that’s the time period in which I think there’s sort of a flowering, a renaissance of multicellularity around the world. Now to be clear, I think a lot of things had already been multicellular, but they were kind of largely small, sort of unimpressive, things you wouldn’t really notice. Animals probably evolved to form small multicellular groups 800 million years ago, but it was until about 550 million years ago that there’s just this huge explosion in animal diversity. Land plants didn’t really begin to colonize land and begin this arms race of getting larger and more complex until around 450 million years ago.
0:44:40.6 WR: And in terms of the seaweeds, for example, the seaweeds like the red algae and green algae, they had been around since over a billion years ago, but those were very small and you really see this explosion in seaweed diversity and size around the same time. And then the brown algae, the dominant kelps that you see if you go to the ocean and look at the sand that’s covered with these being brown kelp, those actually are kind of the most recent kid on… They’re the new kid on the multicellular block. They evolved roughly 200 million years ago, and we don’t really have that much evidence of what their actual ancestor looked like, despite the fact that that’s one of the most recent multicellular transitions.
0:45:18.7 SC: I am so glad I’m not a biologist, it’s just too complicated. I don’t know how you keep the different colors of algae in mind but good. This might be a slight digression, but maybe it’s exactly the right time to talk about the fact that not only do the sizes and stickiness of your cells and your yeast change, but the shapes change as well, and you’ve done interesting work on the physics behind packing cells together and statistical mechanics and things like that. So it’s always important to remember that as organisms evolve, it’s not just changing the genome and giving yourselves new capacities, there’s real physics constraints here that shape what is possible and what is not.
0:46:05.3 SC: That’s totally true, and I think that it’s true for everything, and it’s maybe more true than in almost any other biological system that I know of, for the early evolution of multicellularity in the sense that forming a group of cells requires these simple multicellular organisms to confront stresses then act over length scales that they have no history of dealing with. They’re used to dealing with stresses that act over a few microns, they’re not used to stresses that act over 50 microns, 100 microns, they’re not used to dealing with avoiding how cell packing, causing strain to accumulate in groups and causes them to fracture. So understanding how multicellular organisms evolve is not just a question of biology and understanding cellular differentiation and gene regulation and all of these things that biologists tend to think about, but it’s also a question of how multicellular bodies become these sort of bio-physically tough, robust structures.
0:47:00.2 SC: Mm-hmm.
0:47:00.5 WR: So in a sense, you can think of this through the lens of Darwinian materials.
0:47:05.6 SC: Okay.
0:47:06.2 WR: That a body is a Darwinian materiel. And it’s something which evolves and it started out pretty bad and gets good.
[laughter]
0:47:12.5 WR: And how do they do that, right?
0:47:14.0 SC: Right.
0:47:14.9 WR: We kind of understand how modern organisms make strong, tough bodies, but they probably do it very differently from the way that nascent organisms would have done it, because modern organisms are essentially built through a process of development, which can take physical strain into account, which can optimize cell packing in a way that reduces strain accumulation. It’s all coordinated. Early multicellular organisms, they don’t have this coordination, so how do they do it?
0:47:44.2 SC: Right.
0:47:45.3 WR: And so this is a long-time collaboration with Peter Younker, who’s a soft-matter bio-physicist at Georgia Tech, and we recently have a… We put a paper on their archive with Ray Goldstein, who’s at Cambridge, looking at the statistical mechanics of cell packing in multicellular organisms. Now, we focused really on just two model organisms to do our experiments in. Number one, snowflake yeast, which is a dumb clump of cells, at this point, we’re talking essentially the origin, the original snowflake yeast that hasn’t evolved yet for any appreciable amount of time in the lab, and also a green algae, Volvox, which you may have seen pictures of this in various textbooks or just online, they form this cool sphere of cells where the entire center, it’s like a marble of extracellular matrix. And the cells are actually located on the surface.
0:48:35.1 SC: Okay.
0:48:36.6 WR: So we thought about what are the fundamental properties of a simple group of cells, right? Multicellular organisms really differ, if noise arises during the current growth and development, that can be catastrophic to an organism. We sort of know how existing organisms do it, we don’t know how nascent ones do, so we looked at the distribution of cell-free volumes within these clusters, and the reasoning for that, the logic was, that despite the fact that multicellular organisms are all very different, they can have cells growing like a tree, or they can have cells on the surface of a sphere of extracellular jelly, they all share one thing in common, which is that cells take up space. And if you sort of go back to what we understand about the packing of materials like sand grains or something, it turns out that you get universal packing statistics for things where their particles take up space, and in fact, this arises from maximum entropy considerations and the distribution of cell-free volumes, or the amount of free space around each particle, is distributed with a K-Gamma distribution. And that’s, in fact, indicative of this maximum entropy driven cell packing.
0:49:44.4 SC: You’ll have to explain what a K-Gamma distribution is.
0:49:47.4 WR: I think you’re probably in a… I’m not a physicist, so you’re probably…
[laughter]
0:49:49.5 WR: In a better place to explain that than I am.
[laughter]
0:49:53.1 SC: I actually don’t know that specific one, but basically what you’re saying is that there’re some probability distribution over different ways you can arrange yourself, and it’s like some power law or something like that, and it’s predictable based on saying, given the microscopic constraints, let’s maximize our entropy.
0:50:09.6 WR: Exactly. If every possible configuration of cell packing were equally probable, you would get a distribution of cell packings that is very, in fact, quite likely, and not a very unlikely distribution based on just this very simple assumption where there’s no free parameter fitting. You just need to know a few things about your system and you can get a predicted distribution of cell packings based on this assumption.
0:50:33.4 SC: And what does that have to do with the… Well, sorry. I guess what I should say is, are the differences between different kinds of microscopic morphologies mostly down to the shape of the individual cells, or is it the stickiness of the cells with each other or something more complex?
0:50:50.0 WR: I think the biggest difference has to do with how these cells arrange themselves in the group. So there’s a sort of palliative different ways that multicellular groups can form.
0:51:00.3 SC: Okay.
0:51:00.6 WR: You can have trees arising where essentially daughter cells bud from a mother cell and remain attached. So you can imagine that you’re adding ping-pong balls onto existing ping-pong balls and you just continue percolating the growth of this group, right? Or you can have something where you have a sphere of cells and a sphere… A spherical group and the cells are arranged on the surface. Or you can have cells that are sticky, actually sticky, and come aggregate together into a big blob because they’re attractive to each other. Or you can have something where you have one cell that gets really big and then undergoes a series of repeated, smaller and smaller divisions until it forms a group of nested cells within an exterior cell wall. All of these things actually occur in nature as it routes to multicellularity. One of the cool things about multicellularity is that it’s very, very diverse. And when we looked at the Snowflake yeast, which is basically a brand new model organism that’s had no time to adapt to being multicellular, it’s not even really a multicellular organism, it’s a dumb clump of cells. When we look at Volvox, which has been around for 20 million years and have development on a very bonafide multicellular organism, they both fit this maximum entropy distribution really, really well.
0:52:09.4 WR: And then if we simulate the growth of different groups, according to those four different things that I’ve just described, we can grow them through different rules and look at them and then partition space, they all fit to this maximum entropy distribution very, very well. And this is cool, it’s cool for a couple of reasons. Number one, it tells us that there’s sort of a multicellular ground state, that if you’re gonna form a group of cells, the distribution of cellular packing is probably going to be predicted by this maximum entropy prediction, and you can probably get away from that to some degree with coordinated multicellular development, but it’s a starting point. And number two, it helps us understand, I think one of the larger conceptual challenges, and understanding how multicellularity arises, which is it helps us understand how groups of cells get heritable traits, even without multicellular developmental programs. Maybe I should unpack that a little bit.
0:53:07.5 SC: Yes, this is exactly where I wanted to go because… Let’s phrase it in the following way, as I understand the Snowflake yeast, they’re multicellular in the sense that you got a bunch of cells and they’re sticking together to make a big snowflake kind of shape.
0:53:24.7 WR: Yes.
0:53:25.5 SC: But it’s not the kind of multicellularity that we’re familiar with in the macroscopic world where there’s huge cell differentiation and there’s organs and things like that, even though the underlying DNA is the same, and so that’s… Even if you say, okay, I’ve made a transition to multicellularity, there’s this other big transition looming in front of us about how you have cell differentiation within what you would call an organism, and that raises the question of what even counts as a true multicellular organism.
0:53:55.5 WR: Right. And I’m gonna sign set the philosophical discussion here, unless you wanna have it, ’cause it can take about 10 minutes.
0:54:01.9 SC: We’ll come back to it, but yeah, do the biology discussion first.
0:54:06.3 WR: Okay, cool, cool, cool. Yes, so I think the way to think about multicellularity is that forming a group of cells gives you a multicellular group, but it’s not a multicellular organism, and in order for something to be an organism, you need to sort of have… There’s actually a philosophical literature on what counts as an organism, and it kinda boils down to the organism contains interacting components in which there’s essentially mutual adaptation and essentially functionality. So those elements are baked into what an organism is. And in order to… And essentially, the way I like to think about multicellularity is that in order to get all these complex features… Which is frankly why we care about multicellularity. If multicellularity was just little clumps of cells, we wouldn’t really… We’d be talking about biofilms, we wouldn’t be talking about multicellular organisms. So in order to get those features and make multicellularity so rich as a system to have the sort of power to change the planet’s biosphere, which it did, you need these dumb clumps of cells to evolve to become more complex and therefore, this actually becomes an evolutionary description, where these clumps of cells actually gain these increased features across generations via Darwinian evolution. And so that means that groups of cells actually need to essentially have the properties of a Darwinian entity.
0:55:31.0 WR: And the cool thing about a Darwinian entity is that it’s an algorithm, this applies to anything, living or non-living. You only need a few things to be true in order to have a population of things evolve via Darwinian evolution. So you need to have whatever entity it is to replicate, it needs to have variation amongst different entities, that variation needs to be at least somewhat heritable and their success, either their ability to reproduce or survive, their persistence in the population, their frequency in the population, needs to depend on those traits that are varying and inheritable.
0:56:05.3 WR: And if you have all of those things met, then your population will evolve, right? Essentially, if selection is acting on a tree and it’s heritable, then your population will change over time to have more of that trait. It’s true for single cells, it’s true for ants, is true for dogs, it’s true for chemistry, it’s true for computer programs. This is just a feature of the world, right? It’s a logical argument. And so in order for a multicellular group to have those properties, you need to have some way that that group reproduces and makes more groups. It needs to have variation in multicellular traits, that variation needs to be heritable and selection needs to act on those multicellular traits. And if those properties are true, then over generations, you can see those multicellular groups evolving, and in fact, over long periods of time, I think that’s how complexity arises.
0:56:51.7 SC: Okay, but… I think I’m on board with everything you just said, but it sort of… I guess that’s a high level description of what we want to happen. Is there a sort of nitty-gritty description of at what point you go from a snowflake where basically, if I understand the snowflakes, every cell is identical, it’s just except for their location in the Snowflake. So at some point, there has to be a way of saying, “No, you become this kind of cell and you over there become that kind of cell.” Is that very far away from where we are now in the experiments or…
0:57:31.9 WR: No, I think that’s already happening, actually.
0:57:32.5 SC: Oh, okay.
0:57:32.9 WR: Yeah. But it took thousands of generations of laboratory evolution to get to that point.
0:57:37.6 SC: [chuckle] Well then, maybe this is a good point to be a little bit more explicit about the Long-Term Evolution experiment.
0:57:47.4 WR: Sure, yeah. So about five years ago in my lab, Ozon Bozdoc, a staff scientist in my group, set up this cool experiment where we’d actually been trying for years to figure out how to do a long-term yeast evolution experiment for… Selecting on a multicellular trait. And we kept running into these problems where they basically would… They would evolve to get larger and then they plateau, and then they just sit there for half a year and nothing was changing and we were getting ready to throw up our hands and think, maybe, maybe this is not a Darwinian system, maybe it doesn’t have the properties that I just described. And then he actually started playing around with these oxygen experiments, which actually led to a key breakthrough, which is that it turns out that we have been doing all of our experiments in a low oxygen environment and that oxygen was constraining our system, so they just could… There was no advantage to getting larger, and so it was pegging it at a small size. So we re-did the system with anaerobic yeast, so yeast with no functional mitochondria, with aerobic yeast and with mixotrophic yeast. And we did this thing where we grew them every day, and we basically let them sit on the bench and he took the ones that got to the bottom and transfer them to fresh media.
0:58:57.1 WR: And we did that every day. We’ve been doing it now for over 1000 transfers, which is around 5000 generations. And we kind of escalated the length of… The strength of settling selection, so now we’re only letting them settle for 10 seconds. And so they actually just fall like rocks to the bottom of the tube. Otherwise, they’re killed.
0:59:17.7 WR: And over this time, we saw the evolution of snowflake yeast that are visible to the naked eye. So they’re macroscopic. They’re bigger than a fruit fly.
0:59:28.4 SC: [chuckle] Okay.
0:59:29.5 WR: And this was pretty surprising because it required… So number one, this already tells us, just connecting to what I was talking about, this already tells us that they are a Darwinian entity, that they have these properties in order to allow them to evolve because they’re getting tens of thousands of times larger than their ancestors, and we can actually freeze… We freeze our populations in a minus 80 deep freezer about every 50 generations, and so we can actually go back and pull out our living fossil record. So we know exactly how they’re changing, we can compete existing organisms with any of their ancestral time points at any time that we want, we can sequence them, we know what all the mutations there are, we know what many of them are doing, not all of them, but a large number of them. We can really understand the system, and we can understand the biophysics of how these guys evolve to get big. So there’s a lot of different routes that we could take in the discussion at this point. We could talk about… We just sort of talked a little bit about how physics informs the origin of heredity. So maybe it’s a good time to sort of touch back to that, and then we can talk about the physics of how they evolve these bigger much tougher bodies.
1:00:34.3 SC: Well, first, I just wanna mention it’s… We’re all familiar with selective breeding, right? We make different breeds of dogs and things like that. And am I right to say that the difference here is you’re not being that selective, like you have a single thing that you’re doing, you’re letting things fall to the ground, but you’re not fiddling in a kind of interventionist way with this yeast versus that yeast.
1:00:58.5 WR: That’s right.
1:00:58.6 SC: It’s basically you’ve put them in a harsh environment and watched them evolve according to their own lights, not because you’re pre-selecting some kind of traits.
1:01:07.0 WR: That’s a really, really good point. In fact, people in the field often make the distinction between artificial selection, where you would be looking for a specific trait in picking those individuals and they’d be constituting the next generation or sort of experimental evolution in the lab where you have a selective scheme that you’re imposing, but you are letting them figure out the traits and phenotypes and things that best optimize fitness in that environment, and you’re just letting them run. And it’s definitely the latter.
1:01:34.4 SC: Right.
1:01:34.6 WR: We are not picking winners.
1:01:36.3 SC: And we should give some credit to the inspiration from Richard Lenski’s Long-Term Evolution experiment.
1:01:42.4 WR: That’s entirely why we’re doing this.
[chuckle]
1:01:43.9 WR: I don’t think we would have even had the thought to do this without his work.
1:01:46.1 SC: So explain what that is. Tell the audience what that experiment was.
1:01:49.9 WR: Yeah. So Rich Lenski is, in many ways, the sort of founder of the field of experimental evolution, and he has this incredible long-term experiment in E. Coli. It’s been going on now for maybe almost 80,000 generations, I know it’s between 75,000 and 80,000. And so it’s by far the longest-running continuous evolution experiment, and it’s been going for 30 years. And so throughout this really unique data set, we’ve learned a lot of fundamental things about the way evolution works ’cause you just… You can’t look at any other system in nature and look at so many generations in this exact same environment… We’ve learned that, basically, adaptation never stops. It’s a power law. It doesn’t asymptote. We’ve learned cool things about how novel gene function arises. We’ve learned cool things about just basic microbial evolution. We’ve learned cool things about the origin of and maintenance of diversified microbial strategies, and because he’s done this in 12 populations, we’ve learned a lot about chance and necessity in the process of evolution.
1:02:56.9 SC: And so you, of course, in your version of the experiment… Number one, I should say kudos because this is clearly a very long-term commitment for a lab or for a person or whatever and you have to keep it up, right?
1:03:11.0 WR: Yes.
1:03:11.1 SC: You can’t just leave it.
1:03:12.4 WR: Seven days a week.
1:03:12.6 SC: Seven days a week, yeah.
1:03:14.2 WR: We just don’t take off holidays.
[chuckle]
1:03:16.2 SC: Someone’s gotta be in the lab. And the difference being that you have a window on to this multicellular transition, presumably.
1:03:26.3 WR: That’s right.
1:03:26.4 SC: Whereas Lenski is looking at just… He’s stuck with unicellular E. Coli that are evolving in their own ways.
1:03:32.3 WR: Yeah, that’s right. I think that’s one of the really exciting things about our experiment, is that we are doing… We are in year three of 30 or so and I do… I really… And I agree, it’s hard work. Really hats off to Ozon for being this completely consistent, utterly committed scientist who has come in 1000 days in a row to do these experiments. We really don’t like to pause the experiment if we can help it ’cause it throws off the dynamics of the way that the populations are evolving.
1:04:00.5 SC: Well, I was just gonna ask, is it even possible in principle to pause it?
1:04:03.6 WR: It is.
1:04:03.8 SC: Can you just put everything in the freezer and come back?
1:04:07.3 WR: Yes.
1:04:07.4 SC: Okay.
1:04:07.6 WR: Exactly. And we’ve had to do it several times. The incubator overheats and everything dies or you get contamination and you have to go back to a time point. Those things happen, but we try to minimize it ’cause actually, the population dynamics are really interesting and take days to re-establish and are really non-trivial, especially when you get large groups that have different probabilities of survival, stuff like that.
1:04:28.6 SC: And are you also doing, sort of, separate parallel versions of it to see differences in ways things can evolve?
1:04:35.8 WR: Yes. So we have basically 15 evolving populations, five of which can’t use oxygen. So five are replicates of a genotype that started out with broken mitochondria, anaerobics. Five of which are mixotrophs and five of which are obligate aerobes.
1:04:50.0 SC: And are they all in the same environments, or are they different environments?
1:04:53.3 WR: They’re all in the same environment.
1:04:55.2 SC: Okay.
1:04:55.4 WR: Yeah.
1:04:55.7 SC: And are you hopeful that someday the ones that can’t use oxygen will discover how to? Is that something to imagine?
1:05:02.8 WR: I don’t think so. In fact, if we sequence them, you look at their mitochondrial genome, and it’s all gone except for a few origins of replication, so they are like…
1:05:12.1 SC: They’ve given up on oxygen.
1:05:12.2 WR: They are… In order of magnitude, more mitochondrial genomes, but they are sveral others and magnitude’s smaller. [chuckle]
1:05:20.4 SC: Right. This is…
1:05:21.4 WR: So pretty much, their mitochondria are gone.
1:05:23.0 SC: It’s one of the issues with evolution, right?
1:05:25.3 WR: Yes.
1:05:25.4 SC: Is that it generally precedes by small steps. So if there’s a place you could imagine getting to that would be better for you, but the only route to get there is through a valley that is worse for you, evolution will, roughly speaking, never get you there.
1:05:42.7 WR: That’s right, that’s right. That’s why a horizontal gene transfer can be very powerful and that you can bring in things from totally different lineages, and actually that’s the reason that synthetic biology, I think, has potential to do really cool things that you wouldn’t expect to evolve, because you would have never been able to put these things together, either valley-crossing or just moving machinery from one disparate lineage into another.
1:06:06.6 SC: Finally, some intelligent design after it’s been thought about for hundreds of years.
1:06:09.9 WR: Exactly.
1:06:10.0 SC: Okay, so… Good, so what… You wanted to say, with that on the ground, on the table, something about the physics of what we’re learning from this experiment?
1:06:19.6 WR: Okay, so in order… So we’re selecting for larger size and again, over… I guess I should say our listeners can take a look at our pre-print to read about this work, it’s on the BioR archive, or if you just find my Twitter thread… Twitter feed, just Google my name, Will Ratcliff, I sort of walked through the paper in that Twitter thread, so you don’t have to read a scientific paper to see what we’re doing, but it actually has a bunch more cool visuals…
1:06:43.7 SC: I can link to both in the show notes, so people should check them out on preposterousuniverse.com/podcast, yeah.
1:06:50.5 WR: Oh, perfect. Excellent. Show notes, perfect. So over the course of 5000 generations of experimental evolution, the anaerobic yeast evolved to get 20,000 times bigger than their ancestor. The aerobic and mixed atrophic yeast, which can use oxygen, are actually pegged, they’re just a few times bigger than their ancestor. So we’ve selected them for over 5000 generations, they don’t do anything in terms of size which is kinda neat. I think it’s a nice follow-up to the oxygen work that I mentioned before, a nice long-term evolution experiment. So the anaerobes, they’ve gotten huge, and we actually know a lot about the physics of how multicellular groups grow to a certain size, how snowflake yeasts grow to a certain size before they break, and it turns out that fracture in this case is driven by cells dividing into the interior of the group, causing packing, strain accumulating from cell-cell packing, and when that strain accumulating from packing exceeds the strength of a cell-cell connection, you break a branch. And much like cutting a branch off a tree, when you break a branch of a snowflake, it floats away.
1:08:00.7 SC: And that’s actually cool because, connecting to the Darwinian algorithm argument I talked about before, this is a way that one group can make multiple groups. As it grows, it breaks up branches, those branches then grow back to approximately their parent size, they begin to break and you get this sort of life cycle. And there’s actually been a fair amount of work devoted to understanding how multicellular life cycles arise, and I think this physical packing causing strain mechanism is extremely simple, one, it’s robust, it gives a very nice life cycle, and it does something which is actually very important in the context of biology here, it introduces genetic bottlenecks.
1:08:31.2 WR: If you imagine that you have a branch of cells, or imagine that you have a ping-pong ball and you put a ping pong ball on the top of it, and each of those cells has a new ping pong ball on top of it and that all of them get another ping pong ball popped off, that bottom ping-pong ball is a sort of parent of everything downstream, right? It’s the progenitor cell, and it’s three generations back, but it’s the progenitor cell. So if you imagine that that’s actually just a branch in a bigger Snowflake, and you were to break that branch off, then that cell at the base of that branch is actually the genetic parent of everything downstream. So as mutations arise within these branches and you break the branches, you’re actually segregating genetic variation between groups. That actually turns out to be profoundly important for allowing mutations to arise, which change the properties of cells which have emergent properties that change the properties of groups, and selection can act on the groups themselves, but because all the cells in the group have the same genotype, they have the same mutation, now there’s a statistical correlation between the group level trait and a genetic trait in the cells. And so if you select on that emergent group trait, while they also… There’s an underlying genetic basis, and now you can actually drive continued Darwinian evolution.
1:09:50.4 SC: So this is, though, not what we see in cell differentiation and big organisms where the underlying DNA is all the same, but they’re just expressed differently, this is true genome differentiation between different parts of the Snowflake, then maybe a part can break off and really be a different kind of organism.
1:10:11.6 WR: I think the way that I would compare this to extant organisms is that this is similar to having mutations in a fertilized egg, and so… Or in a lineage of cells that leads up to a fertilized egg. And so the egg is now the bottleneck, that entire organism develops with the same genetic under basis. And so in any organism that has single cell propagules, you have a bottleneck proper. In our organism, you don’t have single cell propagules, but you have these branches that actually act kind of in the same way, but it basically… In fact, you can calculate the maximum waiting time for genetic variation to be partitioned between groups, and it’s not very long. It sort of sets up an expectation that sooner or later, any mutation that arises, will be partitioned off into its own group, and then you’ll never have that mixture between that group and another group again. So it’s a one-way direction towards breaking up mutations that arise within an organism and making it between organisms.
1:11:06.7 SC: And then is there a competition in the test tube between the different…
1:11:10.8 WR: Oh yes, it’s an arms race. Just like the trees getting taller is an arms race. In our case, we don’t transfer all of the clusters that survive in a given set of time, like 30 seconds of sinking, we only transfer the bottom of the group of cells that are racing to the bottom. And so there’s a continual arms race dynamic here where it’s not good enough just to get to the bottom in 10 seconds or 30 seconds, you’ve gotta beat everyone else trying to get to the bottom in 30 seconds.
1:11:35.8 SC: Have you ever read the short story by Theodore Sturgeon called Micro-cosmic God?
1:11:39.7 SC: No.
1:11:42.2 SC: Oh, you should.
[laughter]
1:11:43.9 SC: It’s from many decades ago, but it’s basically a genius inventor buys his own island and then creates some micro-organisms and subjects them to really, really harsh conditions that they evolve really quickly and develop intelligence, and then he makes them invent things for them, and he sells them to the rest of the world and becomes rich and…
1:12:05.8 WR: Oh my gosh, I love it.
1:12:06.2 SC: Complications ensue. I’m just saying… I’m just saying that if this academia thing doesn’t work out for you, there might be another way to earn a living.
1:12:15.8 WR: Might take a few more transfers. [chuckle]
1:12:17.2 SC: Well, yeah, because it does… It’s pretty rapid, your evolution, but it’s not…
1:12:22.6 WR: It’s very fast.
1:12:24.5 SC: That rapid, but it’s still generations… Several generations per day, not a million generations per day.
1:12:31.3 WR: Yeah, we’re getting five generations a day and it’s frustrating, ’cause it’s a log of the numerical number increase, so you have these hundred-fold expansion of population size a day and that’s… Six and three-quarters generations, and it’s not even really a hundred-fold. It’s like 60-fold.
1:12:48.1 SC: Well, but it goes…
1:12:48.4 WR: It’s not that many generations.
1:12:49.7 SC: It goes back to my point that physicists would be puzzled by how something that has a time scale and measured in minutes or hours can change over a time scale of a billion years in a profound way. But of course, the reason why, if you forced a physicist at gunpoint to say, “How could that happen?” Well, there must be… The billion-year time scale must be the small time scale times a big number, a big dimensionless number. And that big dimensionless number must have to do with the space of possibilities for these genomes to evolve, which is just huge. There’s a whole bunch of different mutations that could go on. So, how do you… I guess in part you’re dealing with that just by making the environment really harsh and therefore nudging them to evolve quickly, but is there some… How do you deal with the fact that even though you’re doing a lot of generations, it’s nothing compared to a billion years worth of evolution?
1:13:49.1 WR: It’s nothing. [chuckle]
1:13:51.1 WR: Yeah, so we can do some cool tricks to get around some of these things, which is we can use synthetic biology tools to bring in features that evolved over millions of years and see what they do in our system, which is really fun. So for example, Tony Burnetti, a postdoc in our group, who’s a synthetic biology wiz, has been basically engineering our yeast to express myoglobin, which is an oxygen binding molecule, and we actually took the sequence from a sperm whale, we optimized it to be expressed in yeast, stuck it into yeast and it works. They’re making sperm whale myoglobin. They’re pink. They’re pink yeast. And so, we basically think that these oxygen binding molecules, their whole thing is not like holding on oxygen as a battery, it’s to increase the diffusion rate of oxygen through a tissue. And so we can test the hypothesis that increasing diffusion rates of oxygen through a tissue would interact synergistically over evolutionary time to drive something to become bigger by removing constraints of size that are associated with oxygen, and sure enough, that’s what it appears to be doing. I mean, these papers are not out yet. This is work that’s ongoing, but we took myoglobin from a sperm whale, myohemerythrin from a peanut worm, which lives in deep sea sediments and has a very low oxygen environment, and then we actually took a…
1:15:16.0 WR: There’s a published paper from Joe Thornton’s Group where they use ancestral sequence reconstruction to identify the ancestor of myoglobin and hemoglobin. So we made that gene too and stuck it into our yeast.
1:15:28.1 SC: So it’s not just sitting around twiddling your thumbs waiting for evolution to happen in the test tubes, you’re sometimes taking an active role and poking at it and seeing what can happen.
1:15:40.2 WR: Absolutely, yeah. There’s actually a whole lot more crazy synthetic biology that Tony is up to. I can tell you if you’re interested.
1:15:48.3 SC: Yeah.
1:15:49.8 WR: But I kinda also wanna get back to the physics of this.
1:15:51.5 SC: I’m sorry, yeah.
1:15:52.4 WR: Because I feel like we never really went down that path.
1:15:53.6 SC: We got off that trail. But yes, let’s complete the physics discussion.
1:15:56.8 WR: Okay. So initially, what all of our yeasts that are evolving to become macroscopic do is they make more and more elongate cells. And we really understand how this affects size and by affecting the, essentially, the amount of free space around individual cells and the amount of jamming that occurs inside a cluster. If you imagine before that you were building a snowflake yeast by adding ping-pong balls onto the tree of ping-pong balls, you can see how that would get very dense very quickly. It would kind of jam and break apart. But now imagine you’re doing the same thing with hot dogs, you have… The same group has a lot more free space, it’s a lot more fluffy, and as a result, you don’t have as much jamming, and when they do finally jam, they’re at a much, much larger size. So all of our replicate populations are basically getting mutations which increase the length of cells, which increases group size, and this slowly… This increase in group size is fairly slow, over several hundred generations, we’re getting 10-fold increase in size, and then we see the surprising break from the linear march towards slightly increased size. We see things that are orders of magnitude bigger with just a little bit more cell length, and if we in fact, look at how these things are packing… Normally we expect the more elongate cells are fluffier and is less densely packed.
1:17:16.0 WR: But if you make… These things have slightly longer cells and all of a sudden they’re packing more and more densely within these clusters, which is not what we expect based on or physical first principles. And so it turns out what’s happening is we’re having this shift, which in fact, if you’re into phase transitions, I think this is something which probably is one.
1:17:33.8 SC: Sounds like it, yeah.
1:17:35.0 WR: We’re working on this to know if it really is. We see this shift from cells being basically, branches on a tree that if you break a single cell, the entire branch falls away to cells being entangled like vines, such that the cells wrap around adjacent cells. And now if you wanna move that one cell, then you’re moving its entangled components, and its entangled components, entangle components, and it turns out that that entanglement percolates throughout the entire cluster, and these groups become orders of magnitude more tough. In fact, if you look at the materials property of them, they go from being a 100 times weaker than gelatin as the ancestral snowflake yeast. If you break a single cell-cell bond, the material breaks apart, which is great for getting a lifecycle, but very bad for making a tough organism. You go from 100 times weaker than gelatin to snowflake yeast that are as tough and strong as wood.
1:18:26.9 SC: And it’s like a cord that is made from wrapping thread, various strands of thread. Yeah.
1:18:32.4 WR: Yeah.
1:18:32.4 WR: Exactly.
1:18:34.4 SC: And is that primarily… Or is this an answerable question? Is that primarily because of the configuration space and some physics, or is it because you’re selecting for bigness and toughness and the yeast discovered how to do that in a new way?
1:18:48.7 WR: I think it’s the latter, but I’d like to talk to Peter Yonker and my physics collaborator more to be sure he sees it the same way I do. But we are providing a massive fitness incentive to get big and out-compete your competitors if you get big. And these things that are entangled are 1000 times bigger than anything that came before them, they’re humongous. And in fact…
1:19:10.0 SC: Are you gonna increase the size of your test tubes at some point? Are they gonna become too big, your yeast?
1:19:16.4 WR: I don’t… I doubt it.
1:19:18.3 SC: Okay.
1:19:20.3 WR: Because… And there’s a reason baked into the way that we transfer them, we don’t actually let the entire test tube compete to settle at the end of every day, we sub-sample 10% of it and those yeast compete to be transferred. So they actually have a 90% chance of dying without ever even having a chance to compete for faster settling and that, I think, actually sets a sort of maximum size which they can grow before dilution just kinda gets the better of them, which… And actually, that’s very important ’cause if it’s something which we’re able to grow forever, it would actually be a very boring organism, it would probably kind of break our system. We want things that actually reproduce and die.
1:19:54.6 SC: Yeah. Yeah. So there’s some constraints.
1:19:54.7 WR: And we enforce that through this mechanism.
1:19:57.6 SC: Okay.
1:20:00.6 WR: But getting back to phase transitions, I think that, this is something we’re working on, but I think that this entanglement idea is probably an example of a phase transition, that as soon as something has the potential to entangle its neighbors, that percolates throughout the entire cluster and you go from something where you’re one fracture away from reproduction to having to break hundreds of bonds and there’s not very much middle ground.
1:20:20.6 SC: And it sounds like something that is actually taking advantage of the feature of multicellularity, and it’s not just getting bigger by adding more cells, it’s getting a good survival characteristic by doing something that a multicellular organism can do that a uni-cellular organism can’t.
1:20:39.9 WR: That’s right. And in fact, if you look… We spend a lot of time thinking about entanglement in biology, and it turns out it’s a relatively understudied research area, but if you look at the largest fungi that have ever existed, there’s something called prototaxities, afungus from the Devonian 400-ish million years ago that formed these 3-foot tall, 20… Sorry, 3-foot wide, 25-foot tall columns of fungal Mycelium. And if you look at… And they’re fossilized so well, that you can take thin sections of them and look at them, and you see… I mean they’re beautifully entangled if you’re basically going in two different directions. They’re wrapping around each other. It’s gorgeous. And actually, as Peter and myself and perhaps David, who as well, who’s at Georgia Tech and works on the mechanics of biology, we’re beginning to think more about how prevalent is entanglement actually in biological systems? For example, is rhino horn an example of a… A good example of an entangled material? It’s made from hair that’s kind of all meshed up and matted down in this horn, and those hairs are not just perfectly parallel. So how much is entanglement actually underlying a lot of the materials properties of multicellular organisms, but we don’t typically think of it that way because they have glue or they have bonds between cells or something like that.
1:22:00.9 SC: But it’s clearly a nice way that the possibility of cooperative behavior is changing the qualitative features of the organism, right?
1:22:12.2 WR: That’s right. So getting back to something you said earlier, the way we introduced talking about the multicellularity Long-Term Evolution experiment was, are we gonna see cellular differentiation? When are we gonna see cells performing different things? And we’re actually investigating this in the context of entanglement right now in the sense that our yeast cells are all genetically the same, but that does not mean that they’re physiologically the same. In fact, we see big differences, evolved differences that have evolved over our experiment in the behavior of young cells that are sort of weighted towards the surface and older cells that are on the interior of the cluster. It turns out that the older cells are much more likely to divide from the middle of the cell making a right-angled branch, which is something which drives entanglement.
1:23:00.0 WR: New cells, young cells first generation cells are much more likely to be popped off the tip of the cells. And the fact it looks like this is, to some extent, being coordinated by changes in the expression of chaperone proteins, which have a very broad role in the cell, but if we play with the expression of these chaperone proteins, we can actually change these phenotypes, and so basically what we see is that over cellular age, these things are growing outwards, so the internal cells are old, the external cells are young, we’re seeing the evolution of new age-dependent cellular behaviors, which may be in fact important for driving the entanglement phenotype and giving these groups really robust materials properties.
1:23:40.6 SC: Well, and also…
1:23:43.3 WR: So this should be an example.
1:23:44.4 SC: Yeah, is it too much to say like, this could be the hint of the beginning of true cellular differentiation and expression?
1:23:50.9 WR: That’s exactly how we are thinking of it, is that it’s sort of like, if you don’t have… This is one way to get this, there is a reliable distribution of… Like, age provides a reliable signal for where the cell is, if it’s young, it’s probably on the outside, if it’s old it’s probably on the inside, and we know that yeast actually have all these things that are expressed differentially in age already, so that’s actually a lever that’s easy for them to pull. And what we seem to be seeing is them pulling on this lever of changing the expression of this chaperone protein, which changes the distribution of budding, which then changes how entanglement proceeds.
1:24:27.6 SC: That’s pretty amazing.
[chuckle]
1:24:27.7 WR: That’s pretty cool.
1:24:30.8 SC: Yeah and the most amazing thing is that you’re not taking billions of years to see it happen. By just nudging it a little bit, you’re able to see it in the lifetime of a postdoc.
1:24:38.2 WR: Yeah, that’s right. And the other thing that’s really cool from my perspective is that a lot of this stuff is stuff that I would never have been able to predict before doing the experiment. A lot of experiments in biology are hypothesis-driven, where you already have a pretty good idea of the way the system works before you even propose the experiment, you set it all up to usually test some very low-dimensional linear effect and ignoring all the complications of the real world and, is this additive effect significant? Yes, it is. And those experiments are great, don’t get me wrong, but they’re also limiting in that you have to already kind of understand your system well enough to be able to propose that kind of experiment before you can do it. And in our case, one of the beautiful and fun things about experimental evolution is that you don’t actually necessarily have to know what you’re gonna get, and you can be really surprised by the twists and turns that your system goes on.
1:25:28.2 SC: Well, I think that is the perfect place to draw this to a close, ’cause you’ve given us an enormous amount to think about, and that’s what science is all about, being a little bit surprised about what you learn. And I’m hoping that your experiment goes on for at least 30 years. I think you’re being modest and it could be at least 300 years, you never know.
1:25:47.2 WR: That’d be awesome.
[chuckle]
1:25:49.7 SC: Alright.
1:25:49.8 WR: I will happily hand the reins over to someone else.
1:25:53.1 SC: Well, unless they solve aging by then, you could still be the boss for the next 300 years. I don’t know.
1:25:57.9 WR: Oh yeah. Well, that sounds horrible. No, no. In fact, maybe even a better thing would be to encourage other people to start their own forks experiment. Yeah.
1:26:04.6 SC: We can also do the… Oh that’s true. I was gonna say their own experiments, but of course we can share, right?
1:26:10.0 WR: Absolutely.
1:26:10.3 SC: You can pass these around and have it evolve differently in different labs. That sounds good to me. Alright, good. A project out there for some of the folks listening to the Mindscape Podcast. So Will Ratcliff, thanks so much for being on Mindscape.
1:26:22.4 WR: Thank you so much. I had a great time.
[music][/accordion-item][/accordion]
I wonder if reactive oxygen species like H202 can select for multicellular grouping as a means to avoid oxidative stress. Perhaps there are certain ecological situations this could happen–one thinks of the two big oxidation events and scenarios that might occur in areas with much photosynthesis. Maybe there are some particular conditions in which the effects of grouping on preventing diffusion of oxygen or its reactive species would be an advantage for grouping, perhaps for anaerobes, rather than more of a constraint (e.g., as envisioned for aerobic organism late in evolution as talked about @ 23:55).
Pingback: Sean Carroll's Mindscape Podcast: William Ratcliff on Multicellularity, Physics, and Evolution - 3 Quarks Daily
William, thanks for ruining it.. I’m never going to look at my beer and pizza the same way again! 🙂
It’s critical to keep in mind that the evolution of all organisms on Earth from single-celled life forms to plants, animals and human beings depend on geology, physics, chemistry and biology. That’s why in important to have experts available in each of these areas when preforming experiments like those done by William Ratcliff, which he seems to have provide for.
It’s also worth noting that while each of these disciplines are different, they each use the so-called scientific method when performing and evaluating experiments.
The closing comment in the introduction was “We can’t yet create life from non-life, but we can reproduce critical evolutionary steps in the lab.” If we actually did have the ability to create life from non-life, the question becomes – Should we? Sounds like an interesting topic for a future Mindscape episode.
Teaching grandkids at home. Boy did Biology change. Animals are a bag of atoms until the electricity flow. I wonder if somehow there is a sort of feedback loop that has to be tied into or something related to quantum mechanics. Or whatever is beyond. Fields nacelle to the One like the Enneads explain.