Watch Episode Here
Listen to Episode Here
Show Notes
This is a ~1 hour working meeting between Timothy Jackson (https://theconversation.com/profiles/timothy-n-w-jackson-115702) and myself, discussing our collaboration on how evolutionary and clinical toxinology connects to diverse intelligence and biomedical hacking.
CHAPTERS:
(00:00) Framing Chemical Ecology
(02:40) Galls, Venom, Symbiosis
(10:42) Microbiomes And Repurposed Molecules
(15:10) Psychoactives And Morphogenesis
(17:41) Psychedelics In Chemical Ecology
(24:12) Killing Versus Teaching Toxins
(26:31) Attractors And Drug Paradigms
(34:24) Agential Interventions And Escape
(42:35) Plasticity And Gene Expression
(49:14) Three Paper Project Outline
PRODUCED BY:
SOCIAL LINKS:
Podcast Website: https://thoughtforms-life.aipodcast.ing
YouTube: https://www.youtube.com/channel/UC3pVafx6EZqXVI2V_Efu2uw
Apple Podcasts: https://podcasts.apple.com/us/podcast/thoughtforms-life/id1805908099
Spotify: https://open.spotify.com/show/7JCmtoeH53neYyZeOZ6ym5
Twitter: https://x.com/drmichaellevin
Blog: https://thoughtforms.life
The Levin Lab: https://drmichaellevin.org
Transcript
This transcript is automatically generated; we strive for accuracy, but errors in wording or speaker identification may occur. Please verify key details when needed.
[00:00] Timothy Jackson: There's a massive amount to talk about. The basic outline is that we have these three papers. The first one is really focused on integrating a lot of what you're doing with morphoseuticals, electroceuticals, somatic psychiatry, multi-scale competence architecture, all this fantastic work that you've been doing with chemical ecology, broadly speaking. My real area of specialization is toxinology, the study of animal toxins and particularly venomous organisms. There are a lot of resonances between what you have been doing and chemical ecology more broadly, not just toxins. One of the things that I teach is a pharmacology subject. It's on the human relationship with drugs, but I take it all the way from chemical ecology, and then to the way that organisms find molecular tools in their environment, including in processes like toxin sequestration, but also various symbiotic processes. If we can talk about morphogenesis perhaps in a minute, there are all these fascinating examples of chemically mediated morphogenesis between organisms. The involvement of, say, symbiotic microbes in morphogenesis. That's common for marine invertebrates in the larval transition from a pelagic phase, when they're floating around on currents, to the benthic phase, where they become attached to a surface. It's common for some symbiotic bacteria to be involved in that transition. There is a lot of work attempting to identify the relevant molecules involved.
[01:55] Michael Levin: I think some low-hanging fruit there is a paper on that, particularly in the context of the synthetic morphology idea. That nature hacks itself and each other all the time in every direction, and we as engineers should be doing that too. I've been getting into plant galls and these amazing ways that life takes advantage of the competencies of the things around it. I'd love to hear you talk about this stuff because I don't know all of these examples. I think it would be amazing.
[02:40] Timothy Jackson: Goals are another really cool example, and I've seen you talk about them. Goals are another incredibly widespread thing. It's not just that there are these cynipid wasps that inflict galls. There's a whole fascinating thing with those wasps, where you've got the so-called inequaline species that are essentially hyperparasites. They don't make galls themselves, but they steal galls or appropriate the galls of other species. There's that cool stuff going on. You've got aphids that make galls and beetles that make galls. I think there's estimated to be tens or maybe even over 100,000 species of insect involved in gall formation. There are fungi that induce galls. What initially looks to be exotic — one organism influencing the morphogenesis of another through some chemical mediator — turns out to be a very, very common process. That's another thing with venom: we're not necessarily talking about morphogenesis there, although there can be very subtle things going on, such as the influence of the immune system. Venom is not necessarily all about killing, although people differ in their definition of venom. Toxins are certainly not necessarily all about killing. We tend to think venom is this exotic thing; snakes, scorpions, and spiders come to mind when people think of venomous organisms. There are probably over 200,000, maybe more, venomous organisms in the world. Venom has evolved over 100 times independently. It's a particularly specialized form of chemical ecology, and that's ubiquitous. Organisms mediating their interactions with other organisms through chemical means is a general phenomenon. Then we get these specific examples of it. When I teach the human relationship with drugs to pharmacology students at the University of Melbourne, I try to demonstrate that pharmacology could be thought of as a subset of biology. They don't get any biology in the pharmacology stream, or an integration of pharmacology with biology, which is mind-boggling to me. I start from basic principles of chemical ecology: organisms mediating interactions with each other in cooperative ways sometimes, and in antagonistic ways when we're talking about toxins. One fascinating thing is how the valences of relationships can flip: commensals can become parasites, commensals can become mutualists, or parasites can become mutualists. A shift in the context of the relationship — a shift in the reference frame of the relationship — has a profound effect on its valence. I move from these examples to the chemical ecology of humans and say that pharmacology is humans doing things in refined and specialized ways, but it's really something organisms do in general, which is search out useful molecular tools in their environment and appropriate them.
[06:41] Timothy Jackson: Vivid examples that I use are toxin sequestration. One organism appropriating the defensive toxins of another organism is the most common way this happens for their own defense. I'm wearing this chemical ecology t-shirt and the molecule here is the backbone of a cardiac glycoside and it's just an interesting molecule because it's one of these things that's often involved or is involved in a number of different cool examples of toxin sequestration. Wanderer butterflies: the caterpillars feed on milkweed and they take these cardiac glycosides that the plant produces and they sequester them in their own tissues. The butterfly, which only feeds on nectar, retains these stored toxins and the butterfly is itself toxic. In completely different sets of interactions, we have poisonous toads, which has been a major issue in Australia since the introduction of the cane toad. Most frog-eating critters in Australia are toad-naive because we don't have native toads, so they are sensitive to these cardiac glycoside toxins that the toads make. Elsewhere in the world where there are specialist snakes that feed on toads, you get things like tiger keelbacks, which have specialized glands in their neck or neutral glands. They feed on the toads and they sequester the toad toxins into these specialized defensive glands. They have bright orange and black bands on their necks, aposematic coloration to warn of the toxicity that they've basically stolen from their environment. Some examples of the influence of chemical ecology and symbiosis on morphogenesis involve transitions from pelagic to benthic for marine invertebrates. That's observed in a long list of different taxa: sponges, choanoflagellates, ascidians, worms. It seems like another common thing. I think this is involved with barnacles, which are arthropods. In some cases, it seems like there's a direct appropriation in the sense that the bacteria are being preyed upon and something that the bacteria is producing is being appropriated as part of that process of predation. In choanoflagellates that have both single-celled and multicellular life stages, they can actually live out their lives and reproduce as single-celled organisms. There's a rosette-inducing factor that certain bacteria produce and the choanoflagellates feed on these bacteria. As a result of getting this RIF1, they end up forming rosette-like multicellular colonies. That's been studied as a potential model system for the origins of multicellularity in a symbiotic interaction. In a number of cases, there are sponges which get key molecules that they can't make themselves from bacteria that they're symbiotic with. They may digest them at some point in the process: initially they're living with them, and once they've got enough of whatever it is they need, they digest them. This has been a really important process in the origin of eukaryotic cells. We believe Lynn Margulis gave us that idea, which we understand to be true, that this has been an incredibly important process in the origin of complex life and eukaryotic cells. It's a fascinating example of influences on morphogenesis and on these other major evolutionary transitions.
[10:42] Michael Levin: Does anybody use these bacteria as a vector to manipulate? We had a story about some commensal bacteria in our planaria that if you mess with the relative proportions of the species, you get worms with two heads with different visual systems, you get all sorts of stuff. It seems like that might be a cool vector for implementing some of these changes.
[11:13] Timothy Jackson: I'm not aware of any work like that. That's a really incredible example though. And was that something that was an accident initially, you weren't culturing those bacteria, or were you?
[11:22] Michael Levin: The bacteria are hard to culture. This was some work with Ben Wolf, a microbiologist in our department. It started by asking the question of what kind of microbiome lives on these guys. We identified a few major types. Then we said, okay, what happens if you over-represent some of them? There was one particular kind that made all these changes in the area that was housing them.
[11:46] Timothy Jackson: I'm not aware of any work of that kind, but we could delve into the literature and see if we can find anything similar. My understanding is that we don't necessarily know in a lot of cases what the actual functional molecules are involved. A lot of these have been difficult systems to study in the lab. Most of what we know is based on culturing, basic studies of the microbiome of these organisms. There is an identification of an obligate relationship, but not necessarily what molecules are involved. Similarly with the galls, there are lots of ideas about what's actually going on there. There's some really fascinating stuff: in some cases there's the induction of meristem tissue, the transition of leaf tissue into meristem tissue, which is stem cells — pluripotent tissue that can take on various different configurations. Similar things are going on when the plant is engaged in fruit formation, flower formation, or the formation of reproductive organs. In some cases, invertebrates are expressing things that look a lot like plant hormones, including plant reproductive hormones. In other cases it looks like the same process viewed at different time scales or different stages of evolution. I think it's a wasp; it's a hormone involved in ovum formation within the wasp lineage itself. It's repurposed — similar enough to plant hormones that these wasp hormones can induce a similar effect with a few small modifications. That's what we see with toxins, which are more my area of specialization: you get what I call an endophysiologically active molecule, something with a key regulatory function in the body of the organism that produces it, but it has an activity that can readily be repurposed as a toxin. A straightforward example is a coagulation factor. Certain snakes have directly recruited coagulation factors into their venom and modified them a little bit. Now they're hypercoagulation factors, and when they inject them into target organisms, it clots the blood and chews up the organism's endogenous coagulation factors at an increased rate. Something similar is going on with the galls in the sense that you've got something involved in the induction of specific forms of tissue or in morphogenetic processes in the producing organism, and then it's repurposed to do something very similar in the target organism.
[15:10] Michael Levin: Are you aware of any overlap between the examples that you gave, the kinds of developmental control, and compounds that are known to be psychoactive?
[15:30] Timothy Jackson: Not to my knowledge. It's a really interesting question. Maybe you'd be specifically thinking of things that are plastogenic? So neuroplastogens.
[15:41] Michael Levin: For example, in plastogens, a lot of our work has been around treating the development of morphogenesis more broadly as the behavior of a collective intelligence. Once you view it as behavior, you can start to use all the tools from neuroscience. We've done all kinds of things, but one thing I would really like to do — we don't have the reagents yet because a lot of these are scheduled substances. I want the collective intelligence to hallucinate. We've tried SSRIs, which are quite interesting serotonergic modulators. We've done some other drugs that aren't scheduled that are in that category. It's pretty cool. I've seen embryos that aren't quite sure what species they're supposed to be, and they make some things that belong to an entirely different species. But we haven't even gotten to the good stuff. We don't have anything.
[16:40] Timothy Jackson: Are you seeking approval to work with some of the more tightly controlled substances?
[16:45] Michael Levin: I haven't started the process yet. I need to figure out exactly what that entails and how much of that we're actually able to do. I'd love to hear you talk about this issue: life forms that are very split off from us a really long time ago — mushrooms and things like this — that happen to have these molecules that are a very nice fit into some receptors in a vertebrate brain, and they don't kill you, but they do make you. What's up with that? Why is it that these fungi have such subtle molecules that have an effect on an animal that didn't exist for the longest time?
[17:41] Timothy Jackson: It's a really fascinating question. I think the broader question about the prevalence of psychoactive molecules in plants and fungi in particular is still a little bit mysterious on one level. In many cases, we just don't know what these molecules do in the plants' or fungi's normal ecology, or whether they have regulatory functions in the molecular economy of the organisms that produce them, or they are specialized, I say exochemicals, right? They have an exo, outside, special function that's directed outside. A lot of them might make sense. Not everybody likes this connection and it may trivialize them, but as toxins in the sense that they might serve to disorient or confuse. A toxin doesn't necessarily—it's just a chemical that's mediating an antagonistic interaction. It doesn't necessarily have to kill or even make sick. It just can be a deterrent. It is possible that some of these molecules are deterrents for certain kinds of potential predators. There are obvious examples, not necessarily psychoactive, but something like capsaicin in chilies, which is specifically a deterrent for mammals because they have these TRPV1 pain receptors in their mouths, whereas birds don't have them. So the hypothesis goes, birds are better dispersers for chili fruit and seeds. Mammals, on the other hand, may destroy whole plants and aren't good dispersers for chilies. So they specifically are a defensive toxin for mammals. Weird things happen when a certain mammal, a certain naked ape decides that they enjoy a little bit of pain and starts to cultivate chilies. You get this flipping of the valence of a chemical ecological interaction from something that's antagonistic. So a defensive toxin becomes now a target molecule that humans are cultivating. Now chilies are all over the planet. It turns out this positive relationship with a particular mammal has been very good for the evolutionary history of the chili. That's something we see with a number of plants and fungi that make psychoactive molecules as well.
[20:37] Timothy Jackson: Cannabis has a global distribution and some people claim that there are no Cannabis sativa plants in the world that are descended from genuinely wild stock. So it's all the product of a history of domestication. Poppies, even psilocybe-producing fungi, are found everywhere in the world. Some of that at least is humans having spread it either intentionally or accidentally. They might be secondary metabolites, which essentially just means we don't know what they do; they're not part of the core regulatory function, so they probably have some externally oriented function, maybe in deterrence, some form of signaling. So they are signaling molecules, perhaps, and a toxin is a kind of signaling molecule, and often toxins target signalling pathways. And you get neurotoxins of all sorts in all sorts of venomous and poisonous animals that target very specific receptors in vertebrates or invertebrates. And obviously you can understand in a kind of antagonistic co-evolutionary or indeed a mutualist situation why there would be this tight relationship between, say, toxin and target, or just any molecule and its target molecule; it's sort of selected interaction partner. But why we get these apparently contingent effects? Well, again, it may be that we are just the bycatch or we're caught in the crossfire of some interaction. Maybe there's been a kind of evolutionary arms race between fungi that produce molecules like psilocybin and certain invertebrates for a long time. Invertebrates have serotonin and very similar receptors to us. A lot of these core signaling pathways and other molecular pathways are incredibly conserved. They've been around for hundreds of millions of years. So maybe it's aiming to have one effect on a common predator of the fruiting bodies of the mushroom, something that maybe disrupts the spore distribution of mushrooms. Maybe it's mammals, maybe specifically they're trying to deter mammals who eat entire fruiting bodies and take all the spores with them so they don't get dispersed properly. That basic function of disorienting and confusing turns out to be something that under certain circumstances is actually insight-generating, because it's breaking your frames of reference, which is potentially terrifying. Suddenly all of your meaningful information is turned upside down. That's potentially alarming, but it could also be very useful. So that's pretty vague and hand-waving all that I'm saying there, but I don't think we have a really good understanding of what the selection pressures have really led to the shaping of these things beyond the fact that there are these incredibly conserved pathways across the tree of life.
[23:34] Michael Levin: That's really interesting. It always seemed to me, mushrooms, it's very easy for them to kill you through microtubule-disrupting agents. And those things are conserved, wide, extremely wide. So that's pretty trivial; if you wanted to kill the predator, easy enough. The question is how do you get these super subtle kinds of things, which you've spoken to? I think it's really, really interesting. We'll see. We'll see what we can get hold of in terms of experimental.
[24:12] Timothy Jackson: I think in terms of killing versus teaching, let's say, that's been a persistent discussion in toxinology circles studying just defensive toxins in general, and even including defensive venoms. And it gets into discussions of mimicry amongst snake complexes, where you have snakes that are obvious mimics of each other and one has a potentially fatal but not particularly painful bite. The other one has a highly painful but not likely to be fatal bite to something like a human or a large mammalian predator. Some people argue which is the actual model and which is the mimic: where does more learning occur? In terms of what's the best deterrence, is it just killing the animal that interferes with you? There's going to be lineage-level learning if you just kill organisms that interact with the poisonous organism. But it might be an alternate pathway and we can't say one is better than the other; it might be equally as good to not kill but to induce some profound learning in that organism. You could even speculate, especially if you were targeting a social organism in some way, if you can teach an individual, they can go and teach all the others to better avoid those. There are so many things that can be going on and so many contingencies in why different organisms are ultimately developing different functions for their toxins that it's going to be hard to disentangle all of those things.
[25:59] Michael Levin: Interesting. I think we have a very interesting set of stories to tell people about this notion of broader, but also specifically developmental and morphogenetic hacking, and how it is that in nature these things get explored and picked up. I could talk about some of the implications for morphogenic engineering and things like that.
[26:31] Timothy Jackson: I think it would be cool for the paper to go through a bunch of different chemical ecology type interactions from these mutualisms to various forms of antagonism and then how these things can shift over. How organisms disrupt or manipulate each other's physiology or metabolism and reorient the attracting set, essentially. That's where it gets really fine and elegant. Where we're not just killing, we're not just disrupting the thing to spiral it out of control unto death, but actually installing new set points. So shifting the organism's biomolecular economy, the way it's managed, the attractors that various homeostatic flows are under the influence of, actually shifting them somewhere else. What makes that cool is that it immediately brings us into line with what we're intending to do with any molecular tool, a drug. We're basically saying that this system is under the influence of some pathological set of attractors. And this becomes very clear in a dynamical systems treatment of mental health disorders. It's under the influence of some aberrant set of attractors, i.e. stuck in a rut. We are trying, with a targeted pharmacological intervention, to shift it out of that rut. Then you have a comparison at that level between psychotropic medications, SSRIs, which you mentioned, and ADHD medication, where there's chronic medication and the underlying principle is that we have this pathological state and we're going to slowly and force the whole system into what we think is a healthy normal state. Or you've got disruptive pharmacology with psychedelic therapies — the idea is to disrupt that habitual pathological pattern and give the system an opportunity to reconstitute itself, to recanalyze itself in relation to some healthier set of attractive states. There are two fundamentally different paradigms about the way that we use drugs. One is it's not really micromanaging, to use the terms that you often use. It's not really micromanaging, but it assumes with psychotropic medication that we're going to be able to force this system, under the constant influence of the exogenous chemical of the drug, into a healthier state. Whereas with the disruptive mode, there's none of that micromanaging. It's giving it a leg up, giving it an opportunity to reconstitute itself. That's where the importance of psychotherapy and setting the rails in a different way becomes so important. Peter, one of the things where I think we can start with this first paper is...
[30:27] Timothy Jackson: Mapping your multi-scale competence architecture against what I have been thinking about for a long time as a layered ecology of molecules. We've got chemical ecology and that describes interactions at one level — that's between organisms mediating their interactions with molecular names as we discussed. But you've also got all these other levels. In terms of ecology, in terms of context specificity, you've got everything from protein folding right at the bottom to protein-protein interactions, which take place in a certain ecology of the molecules themselves. Molecules need to encounter appropriate interaction partners to have activities at all and to have selectable functions. Then you get into cell-cell interactions. Those can be within the bodies of organisms. The ecology of the organism itself, or the internal ecology of the organism, the cellular ecology there. And then you've got whole-organism interaction. So you're at the traditional level of chemical ecology then, but you've already built up three or four levels before you get to that. The way cultures can set contexts for the usage of drugs has a huge impact on the effect that drugs have. Leary's set and setting, mindset, and cultural context have a big impact on what a drug ends up doing for somebody. It connects to Fabrizio Benedetti and drugs and words having the same mechanisms of action. The words or the narrative surrounding a particular drug have a very significant effect on what it does to people. That takes us into discussions of addiction, the relationship between stigmatization of a certain subject and the likelihood that people develop unhealthy relationships with it. If we're doing the multiple-paper outline that we originally had, we can nest in this multi-layered view, introduce that initially, then focus on chemical ecology in the first paper, and then discuss other levels and some of your interests in neuropharmacology and drug-assisted psychotherapy a little bit later on. The other thing we were going to discuss in the middle is the origins of novelty. There are similarities between the way an organism discovers and appropriates a molecular tool and actually changes — balances of relationships change, and actual functions change. There's a form of exaptation going on there in which a molecule which has a particular selective history for a particular function is appropriated for some other function in the life history of a completely different organism. Because of the connection with exaptation, we can connect to xenobots and anthrobots. There are similarities between that change of context being related to a change of activity and function with origins of novelty at the molecular level in general.
[34:24] Michael Levin: Fantastic. I love that outline. It just occurred to me as you were saying, and I wonder if we could get the bots to appropriate some toxins from something else. Would they take some of the stuff? Another concept that we've been playing with more recently is this idea of agential interventions. This notion that there has to be a kind of impedance match between the tool you're using and the complexity of what you're trying to change. So if you really want to hit these high-level kinds of systems, the decisions that the physiological networks make and so on, maybe what you want is something that's also a little bit complex and context sensitive. So we've been looking at anthrobots as treatments because, in the theory, they share a lot of priors with the body itself. They have the ability to do complex things that individual molecules may not be able to do. And they come with all these sensors and amplification machinery and all this other stuff. So that also, I think, is something that we can talk about in all of the different examples that you show, how much, basically, to look at it from the perspective of the different elements of the system, including the compounds, and asking, what does the world look like if you're the toxin? What are you able to do? How much action are you able to do by yourself?
[35:56] Timothy Jackson: When I think of this multi-leveled ecology, which I briefly outlined then, I am thinking about something we also talked about a couple of years ago. I had a chat with Maxwell Ramstead about it, and there was an e-mail exchange with Carl. At each of these levels, when we're speaking of ecology, we're speaking of context-specific interactions; there is potentially a free energy principle account of what's going on at each of those levels that can then be built up. That would be a potential separate project from the papers we're talking about, but it might be something we can gesture towards in those papers and then bring those guys in if they're still interested and do a more detailed account of it. One thing I wanted to have a quick chat about is this origins-of-novelty discussion, because we had it a couple of times via e-mail. In terms of the xenobots or anthrobots and your notion that they are essentially escaping from the constraints of the cellular networks that they're normally involved in when they're becoming a normal skin cell, they're liberated from those constraint functions. So they have the capacity to embody a phenotype which is never foreseen in the history of evolution. They're a really vivid example of that, but I think that's a generic process in the sense that we have these models in organismal evolution, we have escape and radiate, and in molecular evolution, we have escape from adaptive conflict. Escape and radiate is an organism in the environment that it has a long history of selection and is involved in many different interactions, and it's occupying a fairly defined niche space. Then when an environment changes, when its context changes, all of its contextual constraint changes. If the environment changes too much, or if its niche space is too narrow or too specialized, it might die if everything changes too dramatically. We do see these cases often; invasive species are a good model system. When you take a toad out of a context where it has evolved interactions with predators that have toxin resistance and you take it to a new place, all of the constraint functions have changed and there aren't as many predators that can eat them. They can radiate. That lineage suddenly has a reduction of constraint pressure on it and is able to diversify more rapidly.
[39:15] Timothy Jackson: That's a very brief and not very articulate way of talking about escape and radiate co-evolution, but it's a common mechanism that is pointed to in diversification of organisms. With escape from adaptive conflict, the basic idea there is that gene duplication, one of the theories for speaking about how gene duplication is involved in the origins of novel functions at the molecular level. Gene duplication results in redundancy, and you've got one of those genes holding down the fort, carrying on the ancestral function. It's still under the same selection pressure, the same constraints associated with a function that it's had a long evolutionary history with. But because you've now got redundancy, you've got another copy, or maybe many copies in some cases. We see these incredibly huge redundant arrays of genes sometimes, and this is really common with toxins. There are reasons why toxins are often recruited from these multi-gene families: there's a reduction of constraint on the function of each individual member. There's a transformation of the fitness landscape. Think something that was rugged for that individual gene becomes smooth for a network of, or an array of redundant genes. Where it gets really interesting is that when a gene product is directed outside the body of the organism that produces it, its constraints are completely different. For example, you go from not wanting to poison yourself by producing too much of this gene product to maybe producing large quantities of it in very particular tissues. There's a gene expression, there's a regulatory discussion to have here as well; it could be something that is positively selective because, for a toxin, sometimes more is just better. Whereas for something that has tightly constrained gene dosage requirements within the body of the organism that produces it, you can't just accumulate duplicates without suppressing the expression levels of each of them. Otherwise, you just end up getting far too much of the gene product. All of that is to say that it seems like a similar thing with your frog skin cells or your human lung cells for the anthropots. Not that one would ever have predicted that they'd have this particular response to escape from adaptive conflict, but it seems like the same thing. They had an ancestral function. They were tightly constrained by their context, their environment. Having been taken out of that context by you guys, they have suddenly been freed up to explore other competencies they had that were previously suppressed. That almost gets mapped onto an exaptation story. Change of context equals change of constraint equals change of phenotype ultimately.
[42:35] Michael Levin: I agree with that. I think all that makes a lot of sense. Two things. One is, in the standard escape and radiate story, it takes multiple generations to get out. And so, I think this is interesting because what it tells us is that there's actually a tremendous amount that can be done in one. It's with the tadpoles that we had where we put the eyes on the tail and they could see fine from those eyes, even though the eye was connected to the spinal cord, not to the brain. We didn't need multiple rounds of selection to get into this new sensory-motor system architecture, one generation, it's fine, they can see. So that plasticity, I'm going to assume that plasticity really potentiates this escape and radiate thing. If you were already capable of doing that out-of-the-box, then of course, put that together with rounds of generations.
[43:32] Timothy Jackson: Just to say that because it's a particularly vivid example of plasticity, whereas maybe with something like a toad, you're not seeing this morphological change, although there are changes in morphology in some of the Australian populations of toads, for example, but you're not necessarily seeing that immediately, but you are seeing a total change of behavior straight away. So instead of being more cryptic because you're worried that your environment is full of predators that might feed on you, suddenly there can be selection on boldness. Now you might be foraging at different times of the day. Juvenile toads in Australia can be diurnal. And that's really helpful for them because they're avoiding predation from adult toads, which are nocturnal, which now have become the most significant toxin-resistant predator in their environment. Whereas in an environment where there are tons of potentially diurnal predators that could feed on juvenile toads, they might be constrained in that way. So you would see, and we could tell similar stories at the molecular level in terms of, now you can suddenly bind and occlude a certain receptor, which, if you did too much of that within the body of the organism that produces it, would have this autotoxic effect. So I think that there are similarities with immediate shifts in at least behavioral phenotypes when you're talking about organisms. It may not seem as vivid, but I think it's still not necessarily something that takes dozens of generations to happen.
[45:10] Michael Levin: No, that's great. That's very interesting. The only other thing I was going to say is I didn't know this the last time we talked; this is new data. We're in the process of writing this up, so there'll be a pre-print on it in a month or so. We asked the question of what genes do xenobots express and anthrobots express that the normal cells don't express? What new genes are they turning on that normal embryos on the one hand and normal lung tissue on the other hand don't express? Hundreds and hundreds of new genes for their new lifestyle are basically being triggered. And again, no transgenes, we didn't put, we had no drugs, nothing, just a slightly new lifestyle. Even the medium is the same; there's literally very little in the environment that's different, but they are in a different configuration now and their behavior is different. And now they're dipping into the genome to pull out a whole other set of transcriptional programs — some very interesting stuff there.
[46:11] Timothy Jackson: That's incredible. You have that other example with the barium exposure of the planaria, where it's different, but similar in the sense that they very rapidly changed their gene expression. I think this is the way that you would frame it. You're revealing a plasticity which might be much more generic than we normally imagine it to be. There are going to be other examples, and we're talking about drugs and toxins. We can look at the way that an exogenous molecule dramatically and rapidly changes the gene expression of the target organism. You can talk about that; there'd be great examples from parasites. We can talk about zombie ants, but you can talk about parasitoid wasps that inject eggs and toxins or effector molecules that immediately change the immune system and patterns of gene expression in the organisms that they're injected into. Or you can talk about drinking alcohol and how that affects patterns of gene expression over a 12-hour period. They're absolutely not undermining the uniqueness of the xenobots and the anthrobots, but I think what they're pointing to is an incredibly vivid example of processes which actually turn out to be much more generic. People don't necessarily think about a hangover as associated with changes of gene expression in the brain, but it is. That's part of what's going on there. Why are you so sensitive to light when you wake up and can't get back to sleep? Not you, Mike, not us sensible, mature adults, but someone who has drunk too much, passes out, and then wakes up early in the morning feeling really wired and can't go to sleep. This is because you've had a glutamate suppression effect when you've drunk too much alcohol, and you've had a change in gene expression in certain glutamatergic pathways to compensate for that effect. Now the alcohol is gone, so glutamate suppression is over, but you've upregulated, say, glutamate receptors. Suddenly you've got an excess; your glutamatergic activity is peaking in an unhealthy way. That's because there's been this compensatory swing in your gene expression. Those sorts of things are another vivid example of something that's going on all the time: the way that things we put in our body, behaviors we engage in, and things we perceive are changing our gene expression, from very fine-grained levels to quite profound ones. They feel like incredible model systems for uncovering those underlying generic mechanisms.
[49:14] Michael Levin: I see it. Super cool. I think this content that we've outlined is amazing. I guess the question is, do we already have a specific outline of all this, or should we make one? I've been taking notes.
[49:32] Timothy Jackson: We have an outline, but it's an old one. I think it's certainly enough and suitable for us to work from. If you've got more notes, and we've got this recording we can refer to, any particular thoughts that have come up on the basis of this we should incorporate. I see it as the original outline: chemical ecology and your framework and how they are conceptually related. There are all of these fantastic resonances and layered ecologies that we can use to understand molecules. We need this layered perspective to understand even why a molecule looks the way it is in evolutionary terms. That's something we haven't really talked about, but we can introduce that and maybe the potential free energy principle connections in the initial article while showing the resonances between chemical ecology and the work you've been doing. The next article was going to be more focused on origins of novel functions. One thing we just talk about in passing in the first article is molecular tools and how molecular tools are discovered and derived. The second paper can be about origins of novelty more broadly, using molecular tools and things like toxin sequestration and change of context equaling change. These analogies — escape and radiate, escape from adaptive conflict — and things going on with xenobots and anthrobots could be the focus. The third paper was going to be about how these things might hit the road: integrating this somatic psychiatry perspective — how the interaction between chemical ecology and your work, defined as somatic psychiatry, feeds back into psychiatry. What are we learning about these mechanisms from recognizing resonances at different levels of description that could be useful for treatment, both as a perspective on mental health disorders and for how we actually treat them.
[52:13] Michael Levin: I like that a lot. Yeah.
