 who is going to talk about dormancy, zlogro, nogro, et cetera. So thanks a lot, Jay, for being with us. OK. Someone let me know if my voice is projecting through there. OK. So hello. Welcome. Thanks for having me here. My first time in this part of Italy before. I'm on sabbatical right now, so it's given me an opportunity to explore some new places in Europe and meet some new people. So today I'm going to talk about, I actually have two lectures, one today and one tomorrow, that are sort of related. And so what I thought I would do today is kind of just lay the groundwork or fundamentals for what I think about this concept of dormancy, its role in the complexity and diversity of microbial systems. Before I do that, maybe I'll just give a little bit of context and try to describe the types of questions that our group is interested in. So we're interested in what generates and maintains biodiversity at different spatial and temporal scales. And we focus the most exclusively on microbial forms of life, bacteria, viruses, sometimes fungi and bees. And we're interested, I guess, from there, trying to understand to what degree microbial life forms follow laws and behaviors and patterns that have been documented for centuries in plant and animal systems. So are there general rules that we can apply to all domains of life? And that's a type of question that, due to methodological limitations, has been inaccessible by and large, just because of the way in which we study microbes, up until probably 15 years ago, the advent of high throughput sequencing, which allowed us, in terms of diversity, to sample at appropriate depths to characterize the diversity of very abundant microbial assemblages. So I guess it was maybe five or six years ago, a colleague, a postdoc of mine, we started to try to tackle this question by compiling as much biodiversity data that we could get our hands on. So there are lots of publicly available data sets for forests, inventories, plant data sets, insect data sets, where communities have been very well characterized in terms of the species and abundances that exist in those sites. And then we were able to start to pull together Amplicon data sets, 16S-Driven's normal RNA data sets through global efforts where people were trying to survey with very similar methods the diversity of microbial communities. Of course, there's different ways in which you're characterizing those communities. If you're an avian biologist or ornithologist, you're often listening or maybe looking through a pair of binoculars to characterize what organisms are in some actor environment that you can detect either through your eyes or your ears, whereas molecular techniques for extracting DNA and amplifying a single gene is going to be known. There's pros and cons with all these things, but when all is said and done, we were able to amass really large data sets. And to move forward with this question of whether or not microbes and macrobes exhibit similar patterns of biodiversity, we thought it important to consider one big difference, which is the abundance of individuals in a sample. So if you went out into a prairie or an old field and you wanted to survey plant communities, you might put down a one meter by one meter plot and you count all the individuals in that site. Maybe there'd be dozens of individuals or maybe a hundred. But in a gram of soil from that same plot, you'd probably have upwards of a billion individuals. So we know from kind of statistical sampling that the probability of encountering new things goes up if you sample more things. So we constructed these things called diversity abundance scaling laws for other data sets. And we found from this that microbes and macrobes did have similar scaling relationships across many orders of magnitude. So when we combined that with other types of biodiversity theory, we were able then to make predictions for how many microbial toxins there may be on a planetary scale. That number happens to be about 10 to the 12 or 1 trillion. So for those of you who are familiar with the census of life on Earth, if we think about plants and animals, there's somewhere on an order of 1 to 10 million estimated species. We're pretty confident with that number. So this means that when we include microbes, we're expanding the number of species that we share the planet with by six orders of magnitude. It's a big number. So we've been testing that with different types of data sets. We get the same answer. We've been thinking about the underlying assumptions about the species abundance distribution, which is really important for some of our projections. And we've also recently been thinking about whether or not there's simply been enough time on Earth to arrive at such a number. Thinking about things like the speciation rate, hard to measure, but we have some numbers, and extinction. How often do microorganisms like archaea and bacteria go extinct? How are they affected by mass extinction events that occurred in Earth's history? And if we took those things into consideration, could we even arrive at a number of 10 to the power of 12? Our estimates suggest that it is, and it kind of begs us questions moving forward about what the upper limits of diversity in life are on our planet. Those are some pretty cosmic questions, but those are the things that kind of ground and motivate our work as trying to understand that complexity and diversity in the microbial world. So when we start to ask these questions about the maintenance of biodiversity, the generous maintenance of biodiversity, we start to ask about ecological questions about how that diversity is maintained. Why are there so many species coexisting in a gram of soil and a liter of sea water or whatever in our guts? Why aren't there fewer numbers of species? So there's a lot of complexity. And there are multiple explanations and theories for how it is that we can pack species in like this, but many of those theories tend to focus on the differences among species. We can characterize those differences as being traits, which can be defined as the physiological, morphological, or behavioral characteristics of an individual that influences its performance or fitness in a given set of environmental conditions. I was told by Jacopo that there would be interruption. Should I be concerned at this point that there have been none? Okay, well, you know, so I have like 25 slides, which I know in certain audiences is a lot and other audiences is not so much. And so if we don't get through it, that's okay too. Yeah, maybe you're going together. It sounds like a lot of the backup of what you're doing is classical ecology and speciation theory, maybe. Will you show a definition of microbial species that sort of deals with horizontal gene transfer? Yeah, so I mean, those are really important. Maybe not even philosophical issues, but what is a microbial species or taxa? And we're using an operational definition because the data that was available to us was collected in that consistent manner. But it's important to recognize that that's a very conservative estimate of what a species is. So I think for this conversation, I'm going to move on into thinking about questions of traits. One of those traits will be dormancy. So I'm kind of just segwaying in from a broad perspective. But yeah, we could have a long discussion about what a microbial species is, how we define that in terms of rates of recombination and so on. Yeah, I conveniently try to work around that. I mean, I would be highly interested just to hear your overview of this. Yeah, I mean, I think for, you know, the same issue is true for... Here's a non-satisfying argument. We know there are a lot of cryptic species and non-microbial taxa. Things that we call a species will actually have probably multiple lineages within them, perhaps ones that don't satisfy either phylogenetic or biological definition of the species. And so for these types of large-scale data sets that I showed you, patterns, we're sort of course graining them at some level. And, you know, they may not satisfy every single definition of what a species is. But what I would say is that for the definition that we use, which was looking at similarity at a marker gene, there's all kinds of reasons why you wouldn't expect there to be similar scaling relationships. Remember the binoculars or measuring trees in a forest? We identify species in so many different ways and so many different approaches. And yet, when we think about these patterns that emerge over borders of magnitude, we see that there are similar scenarios. And we actually statistically went in and tried to say, well, what if we defined a species of a microbe in a slightly different way in a more or less conservative fashion? And those findings are robust to those types of recalculations. So they're not sensitive to what appears to be a lot of different ways in which we classify what a species might be. That's for questions. Yeah, you want the questions. I said it just kept going. Can you maybe talk a little bit more about what a trait is for you? Sort of going from functional traits, binary traits. Because traits sometimes are stuff you measure externally and sometimes it's sort of more on a gene scale level that you look at it. What's your approach here? Yeah, so I mean, I think if you talk to other non-microbial biologists, they would say a trait is a phenotype, something that... And so how often do we as cell biologists or microbiologists actually measure phenotypes, whether it be metabolic or morphological? I would say that we're not really... That framework doesn't lend itself nicely to the way we study microbes, in part because we can't look at most microbes, we can't cultivate most microbes. We can infer things about their metabolic potential from looking at genomes and that's, I think, something that people have done and I think maybe that's okay. But if you applied that same definition to plants, you say, oh well, a plant has this gene. I don't think a lot of plant biologists would say that. So I'm not sure if that's where you're going, but I think we've had to... This framework is potentially useful. I think there are contexts where we want to know how organisms perform, microorganisms perform, and we have to come up with creative ways to figure out how they do that. They probably exist, but when you measure plants, you can go and we can measure... We can look at an individual, we can measure its photosynthetic capacity, we can measure its rates of respiration, all sorts of things, because they're studyable, they're independent. We can look at them and work with them in a non-destructive way. We can measure their fitness components, things like the number of flowers or seeds that are produced at the end of the growing season. Those are really hard things to measure at an individual level in microbial system. Of course, I think there are people here like, you know, I was just hanging out in Zurich and I know there are people here who work in these labs, you know, a lot of microfluidic devices where we can measure things like demographics in populations, but arguably in a Petri dish or in a device chamber, right? So, there is some kind of disconnect between if you want to study things in nature or in the conditions. I think that's what we wanted. Many biologists want to be able to study things that matter in an environment of some sort. So, yeah, I think there are challenges, but I don't know if that answers your question. I would say physiology, morphology, maybe behavior. If we can measure those things. Yeah, I think it answers most of it. But there's like in the trade-based community, there's also this these thought of like master trades. Okay. They always like to use size and I haven't quite understood size in microbes yet. It seems a little bit one-dimensional. Yeah. But what sort of the trades maybe you're really interested in? Are those like specific ones? I mean, if you say dormancy, that seems like a one-zero thing, right? Either you have it or you don't. Or is it a spectrum maybe also? A spectrum in terms of the metabolic status of an individual or whether or not an organism has the capacity to do it or not. The capacity but say another trade like swimming speed is a very continuous trade. That's not a I have it or not. Whether or not a cyanobacterium has the capacity to fix nitrogen or not. Of course then you could say if you do have the capacity to fix nitrogen at what rate do you do that? It probably varies among individuals. Yeah, I think the trade-based framework as far as I view it is something that accommodates both continuous and categorical characters. Thanks for opening up for questions. Now we're having fun. I just wanted to ask a little bit more about the abundance tax on richness plot that you showed. So that's really interesting and there's lots of ways that you might think about disentangling why that is but I just wanted to ask or just give an example of a place that I was at a hot spring in Japan recently and it's a sulfitic hot spring. It's really hot, 80 degrees and it's full of biomass. So I didn't do any cell counts but you can see these streamers in there. It's dense, it's super dense and so I think the abundance of cells is really high but when we've done molecular analyses the diversity is really low. So this is really different from a soil and I'm just thinking you know I'm really jet-legged I'm on a few high-spressos right now but I'm just thinking about the fluid flow regime of that situation in comparison to a soil and the nutrient and energy availability and how the cycling material and energy cycling in a soil is just so fundamentally different compared to this hot spring which is kind of like a chemostat it's sulfide and oxygen in and sulfate out and it's kind of chemostat. It's really, really steady. So I just wanted to kind of throw that out there and hear your idea about these funny outliers because I love the trait that you're able to capture with that plot and you're obviously able to capture something and that's cool but then I start to think about other things that I've seen and how can we understand that. Sure. So as I was putting some of these slides together that this basically is a compilation of some other studies that have done over the years one of the things that I decided to add at the last minute was this idea as it relates to dormancy this concept of residence time and so I'll show a slide where I have images of we've been thinking about chemostats and how flow rates and physical forces put constraints on both whether or not an organism can persist in an environment and I think that the example that you're using is a good one the contrasting soils versus a hydrothermal vent. Yeah. So maybe I'll show you some images and pictures and a few slides that will Yeah. Yeah. Time. Yeah. Okay. So I think I'll move on to the dormancy stuff. So Trace guides some of the ways and can explain the coexistence and diversity. There's other theories but trait based approaches are one of the groups of ideas to explain patterns of diversity and one trait that our group has been thinking about a lot for about 10 years now is dormancy and we take a really broad definition that may not satisfy everyone and we're careful about it so it's the ability of an individual to enter a reversible state of reduced metabolic activity so we can think about that for a second. It can maybe even write it on the board but so an individual it's an individual process can change. It can go between different metabolic states. It has to be reversible and so when we define it that way there are organisms like viruses that can engage in. We can argue whether or not those are forms of life but they can engage in things like latency herpes viruses do this there are bacteria and fungi that can form spores. Hopefully I'll show a little bit on some of the work that we've done on sporulation they're in the Mediterranean there are a lot of microbial single cell unicellular protists that can form cysts rotifers can engage in cryptobiosis worms can form dour stages insects, diapause amphibians can estimate fish can go quiescent birds can go into a torpor there are marsupials that can delay blastocyst formation and of course there are lots of mammals that can hibernate so there are a lot of examples there and what I'll say is that in all of those instances there are no known genes, pathways developmental programs that are conserved so it means that dormancy has evolved throughout the tree of life independently numerous times so when that happens this is an example of convergent evolution and usually when there's a convergent evolution it's sort of interesting to biologists because it means that independently life is trying out different things and has arrived at a common solution to some major challenge and in the case of dormancy what that challenge is is living in unpredictable noisy and fluctuating environments so that's where I was a premise that I'll leave you with so the answer I guess if you look into it people say no but I was just thinking about this recently it's like well why isn't it I don't know it fits that definition but I think I would need to talk to there's probably many journals on like sleep biology or something and I'm sure there's some experts who would argue with me that this is not but yeah so we go in these periodic cycles of metabolic activity what I'd want to know is what is special about sleep that distinguishes it from dormancy I guess that would be my response to the question and I don't know the answer to that but I've thought about it just recently sorry the time scale okay with respect to the maybe the our lifespan so maybe if that's true then we would need to define some kind of scale at which metabolism reduces and resumes with respect to some other important if you want to differentiate sleep for dormancy yes if you don't want then if you want to make a difference I've just been avoiding it because it's not terribly relevant I mean it's interesting philosophically but I don't have like I mean so then making some kind of arbitrary cutoff this doesn't really make yeah maybe there's something isn't there a notion of environment as well like I mean sleep is not a way to cope with sort of an environment that is unfavorable right something that we do even in favorable environments on the contrary of like hibernation where the food will get low and so the organism are just putting themselves in dormancy to sort of like cope with this lack of resources and it feels like an important distinction for me I'm talking about someone who knows nothing scientifically both on dormancy and sleep but I would guess that metabolic activity should be different in sleep versus hibernation in mammals for example because mammals can go on for months without eating in dormancy and I don't think even if you can force them to sleep for months they would be able to go for like months just sleeping I don't know I'm just I would guess that if we could measure and look into the actual metabolic activity in sleep versus like hibernation in some mammals there would be a difference I don't know I mean there are things like circadian rhythms that are involved I mean yeah so maybe there is something fundamentally different here any other question comment that not necessarily is triggered by the environment like maybe dormancy is a reaction to the environment I don't know and sleeping it might be if you believe it's more efficient at night where we cannot do a lot of stuff before electricity to just save energy to the day where it is more efficient to be active maybe that was the environmental signal just the cycle of the sun but maybe you know also describing how this the relationship between the dormants with the environment can help to define better like some stuff the only thing I'll add to that is that I'll explain this in a slide is that there are examples of what we refer to as responsive dormancy where there are environmental signals that regulate transitions but it doesn't have to be environmentally triggered I have another proposition so maybe the sleep and being awake only works as a whole you cannot always be awake so you need to see it as one cycle as one thing whereas maybe some animals do not have to hibernate their life can still work and be okay without this the cycling possibly yes sorry again someone who knows nothing about this I think this makes sense because I mean I think sleep is something that like let's say mammals need physiologically in order to stay alive but what if I take like one mammal that normally hibernates because it lives in an environment where we're doing winter for example there's very low supply of food and put it somewhere else where it's constant spring and food is always abundant would they still like be hibernating with the same frequency or I don't know again someone who doesn't know anything I think with this discussion the interesting thing about sleep I'm sitting here thinking well I don't know a lot about sleep but I think what it does is it's an example of the situation that challenges my definition and because I've been thinking about this even more recently maybe it's something I need to revisit for the time being I'm going to use this generic definition and I'm doing this to kind of make some broad sweeping yeah it's a framework and maybe there's some ways in which it needs to be modified and sleep may be one example where we've found a biological phenomena that seems to match the definition but it's not quite something that we would call dormancy I think that's what I'm taking away from this conversation which is good I haven't had this discussion in a while in this doubt so I'll try to describe maybe a little bit more conceptually what dormancy is because I think it's really fascinating there are people who dedicate their whole lives to studying the production of aphipia and crustaceans the biochemical, molecular ways and the regulation that's involved and that's really fascinating but what I find is when I talk with people who work in those very specific fields on a specific group of organisms is that different languages come up there's a lot of in terms of trying to find commonality and generalities because of some of the details and so what I want to try to do is just present a framework it's a real simple one that will hopefully I can share some ideas on how we think about dormancy and of course no one has to agree but we could start with the simplest type of population biology model that you can conceivably imagine and so there's a number of individuals that are represented by the dimensions or size of the state variable which is the number of active individuals so all of these individuals we can imagine are just turned on and maybe they're even growing at their maximum growth rate and the population size is going to be balanced by some net reproductive rate and a mortality term which maybe says some density independent rate of mortality okay so what do we need to do now to incorporate in the simplest possible way to represent some of the features or fundamentals of dormancy we need a new first we need a new state variable which I'll just define as the number of inactive individuals in a population okay and then there have to be a couple new arrows or transitions processes if you will that will determine the changes in those state variables over time and so it was already mentioned like so maybe there are environmental triggers to dormancy and that's true individuals will over evolutionary time will invest in cellular machinery and sensing capabilities to determine changes in their environment and that can serve as a cue an environmental trigger if you will for transitioning and sometimes like so things like changes in water temperature or photo period or the detection of metabolites in your environment that may be a signal that there is high density or that the quality of your environment is deteriorating all those could be perhaps sensed and used as ways to inform decisions about transitioning into and out of dormancy and resuscitation is just this process of waking up from dormancy so it can be responsive this is one category of class of ways in which we can think about those transitions occurring in an environmentally deterministic responsive manner but sometimes there are environments that change so rapidly so unpredictably there's so much noise that investing in the cellular machinery isn't worthwhile and so what happens in those instances is that you get the evolution of stochastic switching where some proportion of individuals in a way a bet hedging manner will transition randomly into these states and depending on the optimality of those decisions as defined by the environment and how noisy it is these can confer fitness advantages to populations in terms of their geometric mean fitness over time so that's how dormancy evolves in different environments at least that's the existing theory the other arrow that I want to talk about just briefly is the mortality term associated with dormant individuals it's generally assumed that the size of that arrow would be smaller than it would be for active individuals that's the benefit you've reduced mortality in environments that would otherwise lead you to die in practice there are organisms that can be in a dormant state for days or weeks but we see other examples especially in microbial systems where people revive bacteria from ancient materials that are dated to hundreds of millions of years so samples from amber halite crystals and permafrost samples and so at a cellular level because this is what this class I think is about what determines the duration in which an organism can spend what's the time limit and so it's not a cost free state of existence you still have to maintain homeostasis you have to battle entropy you have to maintain energized membranes you need to maintain pH there are damage that's going to accrue over time misfolded proteins or genetic damages to DNA that need to be repaired and if you're completely turned off and you don't have enough endogenous reserves to meet those energetic basal metabolic requirements then an individual is going to die and that mortality rate is much lower than what it would be if the cell were remained in an active state so the thing that we need to take away from this is that there's deterministic and stochastic transitioning some are based on the environment and the rate at which environment changes and I'm also going to talk a little bit about this idea of mortality and dormancy and how that affects longevity of microbial cells and the iteration of this came about somewhat recently and I don't know if we need to dwell on this too much but I just want to focus on the information content and the idea that dormancy can give rise to something a term that's used in ecology and evolution called the seed bank and the seed bank is simply the sum of all metabolically inactive individuals and the genetic phenotypic and functional diversity and in some cases these seed banks can be quite large and by considering so what is the information content? If you think about population it could be different alleles it could be different physiological traits we've already been discussing and the ability for those organisms to resuscitate serves as a sort of memory for the system that can affect the dynamics of ecological and evolutionary populations and communities and also migration which I might talk a little about so to summarize the way we think about seed banks is that this is an emergent phenomena that comes from the individual behavior the transitions of individuals between inactive and inactive states and the production or accumulation of those individuals in communities it operates across both temporal and spatial scales and we can think about the processes that regulate dormancy on scales of minutes hours to millennia and they also affect spatial processes as well and these operate across biological levels of organization we can think about this at cellular levels we can think about it at the ecosystem scale it creates structure this memory, the potential for feedback so I think about this quite a bit I don't expect everyone here to really care or incorporate this into their research but I would argue that if you were completely ignorant of this or ignored these properties I would argue are common to all types of life then there might be instances where we might not be able to predict or understand dynamics the stability of the system, the resiliency as we would if we were thinking about dormancy or at least aware that it existed and it could be imparting a signature on a system I've been thinking what are you supposed to do when you're on sabbatical I'm supposed to think a little bit and I just went down to Zurich and gave a talk in this new center on the origins and prevalence of life and I've been reading a book when I came over here by Stuart Kaufman who's a physicist who thinks about autocatalytic sets and chemical systems at the origins of life and I started thinking about whether or not dormancy, how far back dormancy might go it turns out it's very ancient and so what I've been doing is I've been reading and trying to find examples in the literature of how far back we can date dormancy in Earth's history and of course Earth's history is quite long and some of these examples are very short on that time scale but slime molds have been recovered from amber that have conservation of sporo-carps with other structures that hold dormant spores the rest of these structures have completely diverged over time so they don't look anything alike but there's conservation in these dormancy structures suggesting that there's evolutionary stasis associated with them there's this group of organisms that are pre-mammalian tetrapads called synapsids that existed in the early Triassic about 250 million years ago they were found on different continents some in Antarctica still when it was very cold and harsh and very seasonal environment and they compared them to synapsid fossils that were found in other places at the same time you could look at their tusks and look at the growth rings and authors of this paper suggested that these organisms were capable of engaging in a torpor so 250 million years ago due to the harsh conditions where they lived in Antarctica going back to 480 million years ago we're at a time period where we can start to think about the transition of life and water onto land and there are algal fossil remains called cryptospores that are thought to be important for that transition of life from water to land and we can go into the pre-cambrian as well so now we're at a time when there are only microorganisms achonetes are a specialized cell type of cyanobacteria cyanobacteria themselves are about 3.85 million years old at least based on microfossil evidence but these achonetes are specialized cells that can store carbon reserve and aid in the desiccation tolerance and thermal tolerance of these cells and they've been recovered from shales dating to 2.1 billion years but all these examples are ancient I think we can agree on that but the complexity in which organisms have evolved even at 2.1 billion years ago there's a lot of genes that are involved in achonete formation so if we want to invoke something about dormancy may have been important in early earth conditions and through probably inherently dynamic harsh and fluctuating are there easier ways at which we could achieve dormancy so I've been thinking a little bit about some simpler forms so somebody may be familiar with persister cells these are organisms that can tolerate antibiotics and other environmental conditions not by through genetic changes but through in some cases things like toxin antitoxin modules and maybe other ways and in the simplest example there are two genes so one gene produces a toxin and another gene produces a molecule that will either an RNA or protein that will bind that toxin and as long as those two things are working in concert to one another the cell is going to be active but if toxin accumulations increase for whatever reason the antitoxin doesn't then the cell is either going to turn off or die so now we've kind of come down to just even like a simple two gene like system where we can achieve we would all kind of recognize as being a form of dormancy or at least I would recognize as a form of dormancy there's other people in this room who think about things like stochastic gene expression and how that can regulate not only the expression of a protein but also the metabolism of a cell we can envision a cell where there's a lot of proteins and molecules some of them are low abundance and there's a probability of one molecule encountering another one whether or not that's going to create a catalytic reaction and so those probabilities can be low in some cases so you can come up with a distribution of metabolic activity within a population of E. coli a growing under LB where there's this long tail distribution of metabolic activity such that some cells are highly active and then there may be lots of cells in that population that not achieving their maximal growth rate so if that's true and if that's universal and unavoidable that may be 4.5 4 billion years ago dormancy would have been some property that would be inherent to life that maybe contributed to the persistence of life at least those are some of the arguments that I'm still working on and they're not obviously some of it's conjecture it's hard to make strong inference about things that happened at this time but we're working through those sorts of ideas right now so this leads to the question how easy is dormancy and this is a little toy example that maybe maybe you will like or you won't like but there are these cellular automaton models that were created in the 1970s by John Conway zero player game of life models right and so you can think you can look at this array here on the left and you can take a local environment and you can focus in on one cell and each cell can have one of two states it's either going to be on or it's going to be off live or dead and then this deterministic game is going to be updated each time step and you're going to get a new confirmation based on these simple rules it's a little bit shaded so I'll go through them if you focus on this individual in the center let's just focus on that this is currently a live cell if it has less than two individuals in the next time step it's going to turn white it's going to be dead we can imagine in a biological sense that this is due to underpopulation or maybe some a lee effect right so if you don't have enough individuals you can't reproduce if a live cell in the middle has two or three living neighbors which is represented here then in the next time step it will remain alive if the live cell here had three or more neighbors it would die overpopulation in a dead cell here if it had three neighbors which is not represented in this situation then it would become alive in the next step because conditions were good and populations were growing okay so those are the simple rules this is the vanilla game of life model so I was working with a student named pat wall who's an informatician and we found this paper it was published in a book in 2006 which kind of introduced this idea and it was could you create dormant life in this type of cellular automaton model and the first thing you would need to do now again is create a new kind of state so you have white is still dead, orange in this case is alive and brown is going to be something that reflects dormancy in some way a dormant individual and to represent that we've just changed the model the rules a little bit so a live cell with zero living neighbors is going to die a live cell with one living neighbor is going to go dormant a live cell with two to three living neighbors will be alive in the next time step a live cell with more than three living cells is going to die and a dormant cell with two to three living neighbors would resuscitate so we know that there are certain things like chemical cues that are secreted by active individuals that can wake up dormant cells so I'll just show this simulation over time and let you look at the behavior and dynamics you can see that without dormancy the population size the number of active cells declines pretty rapidly by a hundred time steps these objects over here look like nothing's going on but they're actually updating these are called still life features that are common in the game of life but with this additional state variable and this small modification of rules we can get the persistence of life over time so this is just one simulation with one set of starting conditions but we've done other simulations where we've explored lots of different starting conditions we've mapped out the whole configuration space actually and what it seems like is that there's a lot of ways in which a lot of starting conditions that lead to dormancy lead to extinction without dormancy encoded just a curiosity does dormancy also have these still life features or is it something that is lost once you introduce this new state we haven't gone through and visualized and I don't know if there's an easy way to kind of computationally document whether or not there are still life dynamics but you could probably easily go through a series of these depending on the size so to do this kind of even somewhat more quantitatively we have to reduce the size by a lot but that would be kind of interesting maybe there is some way to kind of identify whether or not there are certain features or objects or domains if you will that are more common with or without this representation of dormancy Jake, I have a question about the game of life so this is a universal durian machine depending on the initial conditions you can do whatever you want so I guess in this case the initial conditions this is basically equal to program the computer so according to the initial input is the result that you have so I guess this is just a special case this is just one example just to illustrate so this student I've been working with recently moved to New York City and is working on some other things I thought I would pitch this idea it's half formed we have lots more simulations to describe the behaviors of all the possible initial starting conditions but for a much smaller array of size of a population so my guess is that if you try that random you are going to have everything basically you mean random initial conditions with random initial conditions? yeah so depending on the size of the grid of course what I can say is that we have worked out a smaller grid for computational and practical purposes and I mean there are distributions of potential outputs depending on starting conditions but we do see really big differences for example in the fraction of living cells if we include this state in these rules yeah so there are differences yeah it seems to be fairly consistent sorry two more questions maybe along the same lines how many neighbors you need to switch off and on depends your fraction pretty much directly the number of rules the size of rules how many neighbors you need to live that nicely defines your fraction of living cells yeah so I think one of the things that is that a question no that is a question or am I missing one of the things we've been thinking about is what if you change these rules they're not totally arbitrary but the other thing that I've been more concerned about is whether or not these results could be affected by the fact that you have two states versus three states and that light brown it may not be a function of dormancy per se it may be a function of you having three things versus two things and that's been trickier for me to kind of wrestle with intellectually and philosophically in definition you have to have another state a metabolic state in these models but if you had some other rules that were sort of divorced from the concept of dormancy would you still see more persistence and stability just because you have three states as opposed to two states if that makes sense well if you don't have a state maybe you can have migration that would probably give you persistence of life but I guess that when you do cellular automata model you can basically add individuals by having an influx with migration and you probably can see the persistence of life as well without the third state but it's a guess that's an important take away I think at some point is that in some cases dispersal in time and dispersal in space are substitutable and can give can be interchangeable I think there are instances when it's been shown mathematically that that's not the case but ecologists have tend to think about these things as being orthogonal that dispersal in space movement of individuals across space is somehow different than dispersal in time yeah I bet you're right I bet you if you could create a model probably somebody has where you could have migration from another patch so the point here is is basically is building up this idea of how easy is dormancy how I mean with the examples that I've described and some of the ones that I'm going to move on to with certain groups of bacteria are really complex but I'm just been curious about whether or not there are simple easy ways to achieve this definition of dormancy and how might we do this in the easiest possible way okay so let's talk a little bit about microbes because we haven't really been talking about them dormancy is not something that's probably new to most microbiologists there's been a lot of focus on the role of dormancy in human health and pathogens on the left I'm showing a picture here that's supposed to kind of depict Vibrio cholera which is in coastal waterways it can enter something called a viable but not cultural phase which is really dependent on temperature and basically limits the geographical range of where we find Vibrio cholera in North America I'm not sure about in Europe we have these deer ticks called Ixodes and they have a spirochete in them that in fact you get Lyme disease people from Illinois shaking their head mostly in northeastern United States for the past two decades but now expanding westward and you get this bull's eye rash or lesion on your arms and you go to the doctor and they immediately give you I think a moxicillin and that rash goes away and you don't have to suffer flu-like symptoms for a while but there are reports that ten years after having Lyme disease and no subsequent known exposure to these individuals will start to have symptoms and there's been speculation that maybe some of these spirochetes which are known to be slow growing microorganisms can kind of lay low below the host immune system and then re-emerge later the last example is that of Mycobacterium tuberculosis the causative agent of TB kills about 2 million people a year and one third of the world's population is just walking around with what are called infection infections so in all these examples, at least some of them we know a lot about the molecular mechanisms that determine transition and allow for persistence during dormant states but when we first started getting into this idea of this framework I would say not so much was known about the prevalence of dormancy or seed banks in the wild or in natural systems so the first thing was with a former post-doc Stuart Jones this is sort of my default before we start lifting a pipad or doing anything I just kind of want to see what's going on in the literature first and try to get as much data as I can in the easiest way, shortest way possible so we looked in the literature for people who had tried to come up with single cell approaches to quantifying the fraction of active and inactive cells you can do this in a couple different ways you know, in the 1990 late 1990s, early 2000s people were still doing a lot of fluorescent and situ hybridization using probes that would bind to ribosomes 16S ribosome RNA so you could look at the total number of cells and the number of cells that had a positive report based on the number of ribosomes which are used for protein translation so that's one way there are metabolic stains that people use for looking at electron or chain and ATP which can be quantified either microscopically or through flow cytometry so these are the results we gather from the literature what can you hope for, right? what's sort of interesting is that among the data that we collected there's not a whole lot of variation within a given ecosystem but there's a lot of variation across ecosystems so the human gut and wastewater treatment plants based on the data that we had available to us somewhere like 25 to 30% but in soils in soils we had foreshadowing to this problem we see that upwards of 90% of all those individuals that are important for what? retaining organic carbon large reservoir of organic carbon that we don't want to go in the atmosphere storage of carbon recycling of nutrients production and consumption of trace gases of the atmosphere fertility and production of food systems 90% of all those microorganisms at any given point in time appear to be metabolically inactive any thoughts on why we might see differences among ecosystems and the proportion of active cells? inactive cells also have ribosomes they just have a bit fewer ribosomes it becomes a detection issue like you're going to sure, so if you have a cell and maybe it has 3 ribosomes in it or 10 ribosomes in it if you do an assay you go from 100,000 to 10,000 that's sort of the numbers we're talking about so you don't think there are any cells that have fewer than 10,000 ribosomes in them? well they're probably small cells I'm just giving you numbers for E. coli I don't know how it is this is also going to vary by species this is definitely not E. coli nobody's going to vary by species so this is not a model system in the laboratory these are samples that were obtained from the natural environment I have a similar question kind of going in the same direction for different ecosystems would you expose or for these different niches would you also expect to have experimental biases in them during the measurements could you comment on that? we went to the literature and people do all of these methods approaches this is sort of what we would call maybe a pseudo meta analysis where you're going to the literature and just finding whatever data that exists meets some criteria that had to fit one of those methods where there were single cell assays that either used a metabolic stain or were using for us an in situ hybridization targeting ribosomes and ribosome activity and so every person probably does things in a slightly different way the incubation times, the conditions under which they're doing it how samples are processed so we would assume that probably all of these approaches were done in slightly different ways yet we see this general pattern and we didn't try to test any hypotheses I should say with this, this was just purely discovery and exploration we knew that people were using some at this point and then there had been no systematic work done across the ecosystem to try to evaluate the extent to which individual cell level the proportion of active cells in samples so we had no expectations going in I don't even think it's really that interesting of a figure other than it just says that there's variation yeah so if you would expect to talk a little bit about these measurements but if you maybe would actually introduce the bias just depending on which kind of system you're measuring in I don't know if detection varies etc so you might imagine that some of those techniques might be particularly hard in soils because cells are attached to minerals and organic material and there has to be an extraction process and then visualizing cells given the matrix that we do with wastewater treatment plant some of these techniques have been developed at least in environmental microbiology and oceans and freshwater ecosystems so I would tend to trust those approaches in those environments more so what I can say is the only number that seems to be sort of remarkable is soils and there are different ways to kind of get at that same information by extracting things like other biomolecules from soils using fumigation techniques etc and there is support before and after this paper suggesting that there's a lot of microbial biomass in soil samples that's not contributing to the activity that's measured at any given point in time I'll just make a wild random guess here why soils could be more dormant from my limited understanding is that because soil is a very porous structure you have these very small niches of nutrients and eventually they decline a lot faster whereas in an ocean freshwater or wastewater they tend to just get new nutrients all the time but porous structures have this inexplicable ability to just not get new nutrients ever that's my I don't know that would be my guess but my question is more what is inactive here is dormancy when I think of a micro you go into stationary is that already a dormancy because it sort of expands your lifetime you know sort of orders of magnitude sometimes above your like max metabolism save the like the E. coli that grows in 20 minutes but it can also live a week in that is it a dormant state or is it not so it has reduced metabolic activity right I think so yes and then when conditions change it can resume its activity yes I'd be okay with that I mean we can argue about definitions I don't know I mean maybe it's productive to do that but I mean I think it captures some aspect of a micro organisms ability to go some period of time under conditions that are not suitable for growth and reproduction and then when conditions change it can resume growth and personally at the scope and the level that I'm trying to talk I would say that that probably satisfies some definition of dormant we can create another category of thing if you want to call it that but I would argue that it should be it should have an equally good definition for that to make it distinct right so this is like a colleague of mine who's like it's like lumping things like when do we lump and when do we split maybe I'm maybe I'm lumping it too high of a level so for the forgetting these numbers operationally what's the difference between a dormant cell and a dead cell dormant cell is one that one can wake up in the in the lab with known with a known change of so in this case we don't know whether or not sorry the y-axis is labeled as inactive cells so we can look at cells using one of those analytical methods and we can somebody else scored that cell is either having with a given threshold and sensitivity it was either exhibited activity or did not exhibit activity so inactive can mean dead also in active in this case in active in this case could be dead and and and molecularly we have explored that distinction as well so we do some environmental metagenomics and we process samples where we can treat samples with things like well there there are kits called propitium iodine kits that can look for compromised membranes or you can treat samples with DNAs which will degrade all extracellular DNA that would otherwise be amplified so there's probably some contribution to dead cells and other studies that we've done it's it's relatively small like five to ten percent but that would still those those numbers would be included here since you have these changes and you have dormancy in soil does dormancy correlate with any other relevant trait that can add information to this picture so so yeah so the question is does dormancy correlate with other yeah so I mean this is sort of not a PC term used anymore but in the 1970s when people first started thinking about dormants in the oceans they refer to them as dwarf cells because they're small so so people were I'm not sure exactly what technique is looking looking into the microscope and people found that cells that are starved for for nutrients and energy or stress in other ways tend to on average get smaller so that would be a trait that would be correlated in some level in some ecosystems with but not when they are alive though are you saying the distinguishing between live and dead yeah because even in stationary phase in the day so they become smaller so I was wondering if there is any correlation with traits that you can measure when in principle they are not dormant yeah so the only trait I know for distinguishing like if you saw a cellular structure and you thought well that could either be dormant or dead the thing that I think a lot of people will in practice use are looking for signatures of a compromised membrane so we have a hole in a membrane then that's likely not to be repaired and that cell would be dead there might be other kind of in principle theoretical ways that you could know when a cell is alive versus dead based on for example ATP charge on a membrane or something like that but in an environmental sample like some of the ones we are talking about I think that might be dormancy is present in cells that are usually I don't know fast growers are making it up but just a trait that you can measure across pieces and you can say okay these you see dormancy in this branch of the phylogenetic tree what other traits do you see in that branch of the phylogenetic tree that are not that was but maybe we can discuss yeah I'm not aware but I do think that one of the things you said is true is that there's probably there's definitely different ways in which organisms can achieve this they do it in very different ways some are well characterized and in some for some lineages we don't know if if those organisms have active measures like so when I talk to some colleagues of mine who are mathematicians they talk about they use this language of like falling into dormancy it's like this passive process that I'm envisioning like a cell is just like falling backwards can no longer stand up it's so tired but that's not true for at least some microorganisms so I'm yeah so it's an interesting idea like whether or not there are other characteristics that might help guide an investigation like this it's an interesting plot and I wonder if you try to quantify it in terms of the because one attractive hypothesis is that the dormancy is influenced by the frequency of environmental fluctuations like nutrient supply disruptions and things like this so one may argues that guts and wastewater treatment plants are relatively stable environments guts you know of a live organism gets nutrients several times a day and on the other hand the soil is probably fairly unpredictable so kind of a constructive way to try to probe it is to look closer at the oceans and in the ocean the theory is that the more temperate oceans are less fluctuating and more temperate ones are more is there any systematic change in dormancy across latitude in the ocean I'm not aware of any systematic study but I think that's an interesting so could you in a comparative sense look for these traits in the activity of cells in more stable or less stable or fluctuating environments the one I didn't think about that a long time ago I was doing some work in soils and you know the question came as well what is it the environment that the cells are experiencing what is fluctuating for a cell at the scale of a micron versus the scale of a gyre or a body of water in the ocean but yeah I think you one could do that another hypothesis of course is phages so nobody mentioned phages yet so I will be the first one if you are dormant and not divided you are kind of invisible to phages could it be a strategy to hide from from a phage infestation we published some work on this recently and I think I'm going to talk about that tomorrow in the context of sport and bacteria not to give too much away but you have phenotypic switching of a cell and when it becomes a dormant cell in the case of an endospore it's not expressing receptors that allow phage particles to intact and so this is a form of physiological defense or tolerance against phage particles it it affects the co-evolution of those interacting populations there was one question from the back and then I suggest move forward because we are of course late well yeah partially Sergei already stole my question but nevertheless I was thinking that maybe it could be like inversely proportional to some kind of organic carbon influx because in God's waste water treatment plants you basically will have a steady influx of organic carbon in oceans in fresh water you will have so that will produce this organic carbon and in soil you basically just got root exudate so it's the least peak influx of organic carbon that I'm speculating basically the plant would photosynthesize and exudate organic carbon into the soil so it could be like that in this case microbial dormance is basically a consequence to the absence of organic carbon how would you comment that there's two ideas that came up that this figure with all of its caveats let us to kind of pursue one would be I think as you're suggesting is the productivity or energy inputs to a system should regulate the metabolic activity of cells and the other one we can think about soils and this was brought up over here as well is that there's something about the porosity or physical structure that influences the turnover and the input rate of new resources and the turnover of cellular biomass within a given environment and so I think this is the thing I hesitated to include but I was thinking about I could start with the premise that cellular life is at the mercy of currents and flows so this is an assumption and a chemostat would be a perfect example and the theory for a chemostat has been around for at least 40 years I would say formally lots of different models for understanding as we adjust the dilution rate of a cell and there's a constant input and output then cells should achieve equilibrium and the dilution rate would be equal to its growth rate so that theory is pretty well worked out for single populations effectively there are some ecologists in the 80s started to develop competition theory based around chemostats for glow limiting resources at different dilution rates but to the best of my knowledge there hadn't been a systematic development of any kind of theory or framework for understanding how variation in flow rates would influence the abundance the productivity the diversity or the types of traits that you would expect to see under a given set of conditions in flow rates which are going to be proportional to things like migration and the rate at which you're encountering new resource particles in the environment which is the little simulation that we see two-dimensional simulation on the left so with a former post-doc Ken Losey and I we worked on building some individual based models socastic individual based models where we we seeded microbes or microbelike entities into these flowing environments where traits were drawn randomly from uniform distributions and we saw a whole bunch of different patterns of abundance, productivity and diversity that emerged from those simulations one of which is the percent of dormant cells increasing in reactors of different volume sizes coated by the different colors so right now I have a PhD student, Emmy Mueller who is testing some of those ideas in an array of chemostats where she's used bacterial communities from a lake and she introduces them into these reactors where filtered lake water is used as a medium and we just dial up the residence time in this case across six orders of magnitude and you can see that there are fundamental things like abundance changes there are changes in productivity at the aggregate level and also at the single cell level and patterns and changes in the composition and richness of communities as a function of residence time so it's the average amount of time that a molecule or a unit of water stays in the system so I didn't intend to go into this in really deep but I was trying to think where have we come with some of these ideas and what came out of that figure and one of them was thinking about how, yes productivity and energy inputs could matter but also there's something about the physical turnover which microorganisms live that might lead to so think about the human gut we have lots of bacteria in our gut is it a good decision for organisms if the residence time of an insect gut is on the order of like hours or minutes is it a good idea for that organism to engage in dormancy because it will be washed out of the system at some point so only active individuals which can replicate on the time scale of the residence time of the system in which it inhabits will be able to persist in that system otherwise they're going to succumb to wash out and at extremely high residence times the input rates of migrations and resources are going to be so low that dormancy may be the only way in which an organism can persist going back to our original model we talked about that area so how we should I okay maybe this is okay maybe this is okay place to stop them I don't know there's other little vignettes but we went from it's quarter after 10 now yeah okay I can pick up so I'm going to talk tomorrow and address this issue that Sergey asked which is maybe I'll just use a couple minutes to give you a preview for tomorrow if that makes sense I'll skip ahead I think maybe Will Shoemaker who's already been here in the course he's not here today but the duration of time how long can microbes stay dormant because it's put some limits on the extent of dormancy and the size of the seed bank we're thinking a lot about endospore forming bacteria as a way a model system for studying dormancy it opens up a lot of opportunities for manipulation it's an ancient complex trait it's really time consuming to engage in and we've been thinking about what is the cost of making a spore a conundrum that this cell should do what it does because there's this idea of falling into dormancy which might not take very much input you're starving you don't have the essential resources for maintaining your cell or reproducing falling into dormancy seems like that would be a great option but some bacteria like endospore forming bacillus and clostridia which are abundant and cosmopolitan distributed seem to have major chunks of their genome on the order of five to ten percent five hundred genes that are needed to make a productive endospore and it takes a long time to do it so we've been thinking about well what does it actually cost to make an endospore and we've been using genomic transcriptomic and proteomic data to get at estimates of the replication, transcriptional and translational cost associated with all those five hundred genes during the developmental production of an endospore and it seems like it's quite expensive if we relativize this to the total cost of what it means for cell division the sum of in germination costs are somewhere around 35 to 40 percent of the total cell budget more than essential proteins more than a whole bunch of other processes that are involved in the stress response of bacillus and potency, motility and basal metabolism so it seems to be really cost very costly there's risk involved and how cells commit to this and what it leads me to ask I'll leave it on this note is are there other other other processes I'll leave you with this question something that seems to be expensive something that seems to be risky yet common and been around for a long time how is it maintained and in fact it can be lost we've done this in our experiments with Will other people have done this if you passage a spore forming bacteria under good conditions for weeks mutations will hit one of those genes, one of those targets in the genome that's required for making an endospore so a trait that's been around for a good chunk of three billion years on earth can be lost in a month and if you look throughout the of the formicities or now the basalota you can see that this trait has been lost repeatedly through time but there are two major lineages the clostridiales and the bacillus that seem to be able to retain this trait what is it that is it just harsh environmental conditions or are there other things that are going on in nature, in the world that could potentially reinforce the maintenance of this particular ancient trait one we've already heard proposed is phage it could also be predation so if there's any kind of size selectivity of a predator or parasite that would potentially add a little bit of extra bonus to dormancy and then the other thing that I won't get to but I want to make a point of is that the movement of individuals in space dispersal migration if you will can be assisted and can be positively correlated with dormancy. Individuals that are moving from one environment to the next there's risk associated with that for passively moving individuals there's risk of dying in transit and there's a risk of landing in a new patch that's not optimal for growth and reproduction when you arrive and if you have the ability to be dormant in transit or when you arrive this would provide an opportunity to wait until conditions can change where you can subsequently invade and colonize a new patch so these are some of the the two ways in which I'm thinking that are non-traditional at least in mainstream ecology where we may find that dormancy can be reinforced when otherwise it might appear at the surface that this trait could be costly and maybe not adaptive so I'll leave it on that and then tomorrow we'll talk a little bit more detail about some phage, bacteria, interactions and co-evolution great so gravity so we had so many questions, yeah, still works surprisingly, I suggest that if there is one pressing question Jake can answer it and in the meanwhile we transition to the next presentation no pressing question okay, very good so let's thank