 I know that we have double lectures, but because there are groups online, let's try to be here before 9 just so we can get started right at 9 for the folks who are joining us remotely and I can get through the material and you can get on to your 11 a.m. lecture. So this is lecture number 4 and I think I had already mentioned that tomorrow I intend to have more of a laboratory format. So what I've tried to do is condense some of the highlights of what I was going to do in lectures 4 and 5 into a jam-pack lecture today and I will not try to condense 4 hours of material into 2. That would be crazy, but Mateo, you need to turn off your sound. Yeah, good. I will, however, try to give the highlights both of the ways in which we can take some of the principles that I've tried to explain these first three days and apply them into the, into an applied context. I won't do both of my applications, so I've cut short some of the material, but I will try to explain ways in which some of the principles of this notion of viral lysis and being so efficient can be challenged. And then in the second part of the talk I will introduce an entirely different modality in which viruses can interact with their microbial hosts. So everyone's settled in now, you've found your, you've found your things that you're going to take notes with, and we can get started. Okay, so what I will do is just give a brief review and get us back into the same frame of mind, whether you're online or you're here. I've already told you that viruses can infect organisms across the diversity of life, including microbes, right, you now know all about this stuff, and viruses infect things like amoeba and archaea and bacteria at the base of food webs. I wanted to just give a few examples of how efficiently these viruses can act, at least in laboratory settings. These are three different examples, viruses of a cocalithophore that you can actually see these blooms and sort of milky seas from space, and the cliffs of Dover are really built out of the shell remains of these ameliani Huxleyi. And what you can see in each one of these cases that alone in the laboratory, of course, the time differs, that there can be the growth of the host in the absence of the virus. And with the virus, these populations can crash. This third example actually tracks both the intracellular contents as well as the extracellular contents. And what you can see in that case is that the host DNA is being degraded and viral DNA is being produced inside cells and then being released outside cells. So my claim has been from early on that viruses can have these very large effects on populations, really driving down populations quite rapidly. And these are just three examples. And as a consequence, as I tried to explain earlier in the week, and when I told you the first time might have been harder to remember, but hopefully the second time it's a bit easier, not only are they, these viruses killing these hosts, but they're releasing debris back into the environment and that debris can be taken up by other microbes. So for example, heterotrophic microbes that require already organic carbon and pulling that in to make more biomass for themselves can then take advantage of that dissolved organic matter. So viruses have this ecosystem effect and can actually stimulate the production and activity of non-targeted cells. So it's bad for the host they're targeting, but another host may actually benefit from that opportunity. So in what I do in my group and in what I've been trying to build up in these lectures is this notion of how what happens at one scale cascades and impacts others. So I already described the mechanisms by which a virus can infect an individual host leading to the death of that individual cell and the lysis and release of new virus particles. If we turn that into population dynamics we recognize that the death of one cell doesn't necessarily mean the death of the population, we can get induced oscillations for the reasons that I described. In my first lecture and then we can look at very large scales recognizing that there's something on the order of 10 to the 7 viruses per milliliter and 10 to the 6 hosts and of course that number varies and the ratio is usually somewhere between 1 and 100. And so you can imagine that if these kind of processes are happening all the time there would be a tremendous impact if viruses are continually knocking down populations. So that was a very quick recap of things that I've already told you and that you probably didn't know in advance of these lectures. So now you know them and you're hearing them again. But I want to change directions a little bit today and keep in mind that these viruses that we think of as environmental viruses can also be used in other means. So instead of just waiting to see you or watching or observing or trying to measure and understand what's happening with virus micro interactions in natural systems in some cases we may be able to isolate that virus from the environment and then use it potentially even therapeutically. And the reason is that some of these bacteria that are out in the environment get into us and do nasty things. For example, pseudomonas originosa, which can be an opportunistic pathogen colonizing potentially the lungs of individuals and causing all sorts of harm. This is another one of Cidobacter baumani. And yet if we can take a virus, these bacteria phage, and find ones that are specific to this pathogen, this bacterial pathogen, it might be possible to actually use viruses therapeutically, intentionally giving viruses either alone or in combination with an antibiotic as a means to cure an infection. These are two examples. One individual on the left who had a Cidobacter baumani infection and was cured through a combination with experimental bacterial phage therapy. And the one on the right, the individual in the upper right, a dentist in New Haven, had a fistula colonized by pseudomonas originosa. The gentleman on the bottom right is Professor Paul Turner, who realized that there were some ways that I'll explain later in this lecture to use a combination of antibiotics and phage to treat this particular infection to cure this person's life. And in both cases, there are now extensive clinical trials based on the work on the left, including a new phage therapeutic center-based in UCSD and also one at Yale. So this is not just a theoretical concept, this is actively happening in terms of compassionate therapy, and I'll talk more about it today. The two folks on the left ended up writing a memoir about their experience, and they titled it The Perfect Predator. And the perfect predator they're referring to is nothing other than the phage that we've been learning about this week. But because this is a advanced graduate class, we have to be a little careful with our adjectives and our adverbs, I want to go through and use that example as a setting off point to recognize that there are some limitations. Phage are far from perfect predators, right? And I'm going to revisit some of what I've told you as a means to review it briefly and synthesize it, but also put in this context of why we've already learned that they're far from perfect predators. And then Jacopo is here today and he asked me, are you going to talk about phage therapy tomorrow, so I decided to combine some things. So I'm actually going to talk about phage therapy, but in light of the fact that they're eco-evolutionary constraints, that they're not perfect. If they were, sure, we would use them all the time and they would be totally effective, but of course we don't and they're not. And so I'm going to explain why that is and what it might take to improve phage therapy. And then at the end, and I have enough time to do these three things, I'm going to try and open up an entirely different way in which viruses interact with host, which doesn't necessarily even involve killing. So I will try to unpack this piece by piece. And this first part should function as a review, but getting us taking the material that seemed just foundational earlier in the week and I'll put it into more of a relevant context. Why phage are far from perfect predators? So in the first lecture of the week, I went through and built up this notion that we have a life cycle in which viruses interact with these bacterial hosts and injecting their genetic material into the host, taking over host machinery, leading to the self assembly of these viral capsids, the integration of viral DNA into these capsids in the time release outwards. And we imagined a context, this particular chemostat environment in which we have resources being put in continuously and everything going out. And together we ended up explaining and understanding why this sort of combination between lysis at one scale and resource inflow and everything outflow would lead to these counterclockwise dynamics. It's exactly the same as block of ulterior like dynamics. And I'm not going to go over and detail this because we already talked about that quite a lot. And of course I also told you that if we were to include more details about the microscopic mechanisms rather than assuming that the viral lysis was instantaneous but rather had some time or duration in which viruses resided inside the host, on the left it's an infected class meaning there's the generation of infected cells and then over an exponential decay period they pop out. And the right is in which they are infected and then precisely at a time later they are released. So the difference between these two is on the left it's saying we have this kind of resonance time and on the right it's this kind of resonance time tau and 1 over eta is tau would be the mean. So they have the same latent period but they have very different distributions. And these become extreme benchmarks between which you can make more realistic latent period distributions. But the point that I tried to make was that then when you add this kind of information you can get these limit cycles. And that was interesting from a dynamical systems perspective but now in light of the claim that phage are perfect predators I want you to think about this in a different way. I want you to think about it as a way in which if we had wanted to eliminate this bacteria even though phage can kill individual bacteria they're not doing a very good job at eliminating them at the population scale. That generically what we're finding is coexistence. So for all this talk of perfect predators they're not quite perfect even ecologically. They may be very efficient but precisely because the phage depend on the availability of host or replicate. Their own efficiency means that they self limit. Okay so that's one concept we have to keep in mind when we're thinking about moving these concepts into a therapeutic context. And as I pointed out you can see these kind of oscillations in the laboratory. If you mix things together and it's true or the host populations are dropping to low levels but then they pop up again to high levels. And again if this is a pathogen context it might be good news for a few days but bad news for some other days. And if that keeps repeating that might not be such good news for the patient overall. So we see theoretically we expect there to be potentially coexistence and empirically we see coexistence. And as I pointed out that's just the ecological part of the story. But as you can see in this example if we can imagine now we we wanted to eliminate this particular E. coli from some sort of context we added the phage we didn't quite eliminate it we dropped down the density but it kept popping back up. But then eventually the host density goes back to almost the levels at which it were before if not the same level. That doesn't seem so perfect with respect now to therapy. Again it's very interesting ecologically and from an eco evolutionary perspective but therapeutically also it is problematic. We use this phage try to eliminate it it led to coexistence and then we now no longer can even use this therapeutic phage because all we've done is a selection experiment. And now we have a bad bacteria that is resistant to the drug this phage that we had wanted to use to treat. Okay so there's some fundamentally interesting things about these dynamics but also you can see why they have applied relevance. And I explained that was because there was evolution and we can't avoid evolution in thinking about any kind of therapeutic application. Okay so I'm going to sum up more or less the first two lectures in this one slide but again in this claim of perfect predators that viruses yes can kill individual cells but they coexist with host populations. So it doesn't necessarily mean that we should not conflate the death of a host cell with the death of a population. Moreover evolution of resistance amongst the bacteria can lead to the loss of top-down control. At first we had viruses somehow suppressing or controlling the density but because of evolution then we lose that ability and now somehow the bacterial density is back to where it was. So that kind of suggests to me that maybe they're not perfect but it tends to be that non-perfect things are a bit more interesting to study. And it provides some challenges that we have to overcome and also we have to be more realistic about claims what can phage do given these limitations. I don't know if anyone here likes pomegranate juice. Does anyone like pomegranate juice? In the United States it's very fashionable to drink pomegranate juice. It has all these incredible things and powers and abilities it'll cure all your problems. Of course it doesn't have that ability but we are always looking for these miracle cures the perfect thing the perfect drink and it kind of changes yearly. I think it was a chai it was one does anyone know a chai has that come here yet? People know a chai. So that was supposed to be the next thing in pomegranates it's always something. So phage I think has suffered a little bit from this notion it's going to be this perfect solution. I just want to point out we have to be realistic about some limitations and that's why I asked you about pomegranate juice. Okay so what I want to now do is introduce some new material get it we got into the mind frame of why I'm doing this and I want to talk about phage therapy in light of these eco-evolutionary constraints. Any questions before I move on to this section? Yes if you shout it out I'll repeat it or use the mic. Okay like right now this the way we are considering phages is as if they like do what they normally do like go into the bacteria and infect them and wait for them to lyse. Can they be better predators if because I have heard that they are also used like to deliver DNA and other applications can they become better predators in like these other uses not directly like to lyse them. Right so in the third part of the talk I'll talk about different ways that we can think about phage although I won't do as much in the therapeutic side some of the therapeutic focus I will focus on today are obligately lytic phage so using them as killers. However you can also think about phage as being an efficient way to target and maybe deliver some package so if you can imagine genetically engineering a phage to inject genetic material that includes some other information that you might want this bacteria to have to do something else to so yes and you can also skip the phage entirely and in fact isolate some of the enzymes that the phage makes and decide forget about the phage that's too complicated let me just deliver these lytic enzymes and apply them and it turns out that's interesting in so far as predominantly work on gram positives because you want the enzyme to interact directly with the cell wall some people thought also working with gram negatives but you can think of the things that phage make phage itself as a killer or even phage as a delivery mechanism but I'm gonna I can't do all of that so I'm gonna focus on obligately phage killing today okay other questions fine so let me go into some new material phage therapy in light of eco evolutionary constraints so in the US the CDC the Centers for Disease Control and Prevention tends to try and prioritize different kinds of multi drug resistant pathogens based on really two factors one how bad is it to have this infection and also how limited the potential therapeutics that we have left that are available and sometimes some of the consequences of those therapeutics and so here you have some of the things you probably have heard of MRSA multi-resistant staph aureus candida strep gonorrhea C diff etc all of these are examples in which there's certainly been the evolution of resistance and even multi drug resistance and in some cases they're even untreatable forms or the treatments are pretty bad too in terms of their toxicity especially the case with with gonorrhea and some others one of the original rationales for sort of getting excited interest in phage was soon after their discovery whether by tort or durel and it's just past a hundred years or so since the discovery of phage so this whole field is actually no hundred years seems like a long time still relatively new so it's about a hundred years or so since people have been realized there was such a thing as a phage and I don't think I said this yet so I'm allowed to say it once even the word bacteriophage the second part phage from phagos from the Greek meaning to devour these were supposed to be bacteria eaters but of course they're not bacteria eaters is everyone here some people did you ever hear what I said I just saw some eyes that look like maybe they didn't hear me they are parasites but at the time they just seem to be something that that the experimentals didn't couldn't see but seemed to be eating up these colonies eating up the the bacteria and Felix Derrel one of the original discoverers of phage a co-discoverers about the same time based Institute Pasteur thought that well what we could do is actually use phage as a treatment and give it to people in this case he hears a quote from 1926 after being assured that there are no harmful effects of sugar bacteriophage it was used to try to treat people who had bacterial dysentery caused by sugar now again these are different standards in terms of what people considered no harm and different levels of understanding of what was going on with phage and even trying to isolate did were you just giving phage what about all the other debris that comes with it and so there are a lot of other things going on and for various reasons including the discovery of broad spectrum antibiotics and often the poor characterization of phage it's not something that many people have ever been offered by their clinician in the other Georgia I usually live in Georgia in the southeast United States but in Georgia the former Soviet Republic there are there's a long history of using phage as a therapeutic even now which I'll tell you in a moment you can get over the counter phage you can go to a pharmacy and get phage and take it in but let me move to the present and again just to point out that there's this long history of using phage and nonetheless people have tried to revisit it for all sorts of reasons and they've revisited in part because of the rise of these multi-drug resistant infections and also because we have a better understanding of how phage work how to prep them and so on this is one example that I will use as the basis for this first half of the talk here you have a control mouse and here you have a phage treated mouse the both sets whether control or phage have been exposed to a large potentially lethal dose of pseudomonas aeruginosa that calls it causes an acute pneumonia inside the untreated mice and what you're seeing is that you a time course because these are bioluminescent bacteria that you can actually observe the proliferation from time 0 to 4 6 hours in the untreated case and so you're actually seeing the evolution and time course of disease and this is the change of bioluminescence and you can already begin to see the difference even after 6 hours where this is a 10 to 1 ratio of application of phage relative to bacteria and here on the on the right side you can see the outcomes where if you don't use phage all the mice then are compassionally sacrificed after two days right if you use one in 10 after five but if you use at least the same amount if not more than a hundred percent of the mice survive so you can begin to see that people have begun to systematically examine again always caution in mice whenever you hear a new biological discovery you have to ask the question is it in mice and rather not in humans but nonetheless it's clear that even in a in vivo context you can take this principle of using phage and somehow we get this outcome a few years ago these concepts which had been really in these kind of you know study began to move into large-scale trials and so there was a well publicized well-known European based trial to treat burn wound patients with phage and it came out to a lot of hype and in 2015 but by 2016 they had a lot of problems they had intended to recruit 220 individuals and the only recruited 15 part of that was because phage are so specific if I have a phage we talked about that yesterday in terms of diversity those were only phage that infected Vibrio and what happened often was that these individuals unfortunately who get burn wounds are colonized by multiple kinds of bacteria and so if you have a phage for one kind you have to exclude the possibility of trying to apply that if in fact it's truly a kind of multi-strain colonization which it often is so you have your phage for E. coli's phage for Pseudomonas and yet if we have a mix of bacteria we can't actually proceed the problem with treating with an individual type is often also that there's this emergence of resistance and a prior response to these kind of failures has been to use cocktails so I talked to you before about when you add a back a phage to a bacteria we often see the evolution of resistance so what you might think about doing is just trying more types more types of phage to try to prevent these resistant mutants from growing and you can do this in all sorts of ways you can engineer them maybe so that they have different tail characteristics or surface characteristics some of them might be blocked on the outside some of them might be blocked on the inside but if you use enough variations maybe you can get it to work and people have built these phage engineering platforms as I said people have tried to actually deliver cocktails and in Georgia the former Soviet Republic you can buy these of course I'd have some caution there some prior work at least from about 10 years ago looked at what was inside these over the counter cocktails and it wasn't necessarily what was on the label that's itself an interesting story so what I'm gonna tell you today is an approach that we've tried to take and I'll tell you about one other approach which is to try to think ecologically and really equal from an egg you evolutionary perspective leveraging the principles that taught early in the week and try to use that as a means to be take a principled approach to phage therapy both in vivo and hopefully ultimately in the clinic and in doing so I want to introduce the players one of which is this multi drug resistance pseudomonasuriginosa I've already showed you some of those images of the time course that causes this fatal acute pneumonia a phage which has been shown in the past to prevent fatal acute pneumonia pack P1 the difference will be in this section of my talk I will also talk more about the ecosystem or the context on the first two days those models had nothing else it was a flask with resources bacteria and phage when you add bacteria and or phage into a animal context there's also an immune system and that immune system is gonna play a role as well so I'm gonna talk a little bit about what happens when you change the status of the immune system and just to say this is a long-standing project and one of the reasons I'm based in France for the years to continue to work with folks at the Institute past or on some of these issues and all of the experimental work if I you say the word we at any point was done by other people not me formerly Dwayne Roach who's now based at San Diego State under who's working on the supervision of Lauren to Barb you and now Sophia Zabrowski and Matthew Digiode also did some of the experiments I'll talk about okay so now I'm gonna get into some details how do you think about taking a phage bacteria interaction and imagining what could happen in a in vivo context to do so and I will try to recapitulate here in the center of the board I'm going to anchor what's happening with this bacteria in this phage keeping in mind that we've already introduced some notion that bacterial load would replicate that phage would somehow inhibit bacteria leading to the lice's release of more phage the phage might somehow decay or die and if we take these ingredients and of course there's limitations here we're going to get something that we expect to oscillate we're gonna get some coexistence maybe the fixed point or spiraling into a fixed point a number of years ago Jim bull and Bruce Levin had said well when we do a phage therapy application what we need to think about is the role of immune system and the immune system is going to play a role because the presence of this pathogen is going to potentially stimulate the immune system leading it to be more active at the site of infection and likewise the stimulated immune system will have a negative feedback and inhibit the growth of bacteria okay so what you should see on this slide in the solid black lines is the state of the art before we began to consider what would happen in these contexts and so you can see here that I've taken a standard kind of logistic growth assumption lysis viral release and decay and then I've added something like immune killing so there's some way in which we have some epsilon parameter they're coming into contact and also there's some sort of alpha B over B plus KN parameter here these are the two new pieces right so there's some killing rate based on contact and there's some stimulation the problem with this original model and the reason we decided to get involved here is that if you don't have the two things which I'm about to tell you about let's try to think through what would happen in this case if you notice here what is the form if I don't look at that red part for a moment maybe I'll actually I'll use this opportunity to write it so you can see the evolution make sure I'm using the right things okay we have a anchor of this bacteriophage interaction but we've added this new immune piece with killing and stimulation so my question is what is the limit to how high the immune response can be if we think of these as some sort of relevant immune cells whether neutrophils or macrophages or what's the limit here is there a limit you just help me out and jump it's not supposed to be a complicated question jump in is there any limit here to how much this can go I'm seeing some nose the original versions of these models said that we have some essentially exponential growth rate alpha but this alpha this growth rate goes with there's no bacteria would be zero we just have this on stimulated system but once bacteria go beyond some critical level in the immune system recognizes it takes off at alpha the issue with this kind of model means that if we were not to include the phage then ultimately this immune system is always going to just keep growing and growing in because it doesn't ever go down it eventually clears the infection so some of these early models basically said that the immune system should always work ultimately which is obviously not true because we have infections so one of the things that we decided to consider is that we have some saturation that the immune system somehow saturates and can only get so intense at the local site of infection that was one piece and then the other piece that we viewed as important was to say that once bacteria get at a certain high level that they have many ways to defend against the actions of the immune system whether through making biofilms virulence quorum sensing and so just because the immune system is active once the bacteria gets high enough it doesn't necessarily mean that the immune system is going to clear it so this is an example of the kinds of things that people do in this space right trying to add biological mechanisms and turn them into a nonlinear dynamic model and see the consequences are there questions about this before I move on so I'm going to use this is a harness to do some things next ever understand the pieces we have a standard of simplified phage bacteria model I've added an immune component in which their immune system is stimulated and kills but then I've added these other two new pieces as well okay good the nice thing about these kind of setup using this model I can eliminate pieces I can ask what would the model do in the absence of the immune system well you already know that we've already talked about that we get these lockable terra like dynamics and maybe they would eventually decay away but we would not have this perfect predator in an ecological sense we'd have oscillations although obviously the densities might be reduced so in the absence of the immune component we get an outcome that we have a persistent although oscillatory infection if we remove the phage then if the bacteria starts to grow it starts to stimulate the immune system you can see we end up at a new steady state which I would term as an infection the immune system is stimulated but yet because of that negative feedback term that inhibition of immune killing and high bacterial densities we don't then eventually just have the immune system going up and up and up it saturates and the bacteria also saturate if I were not to include these two terms this immune system would keep going up and it would be so high eventually that killing that product of if I don't have this epsilon times I becomes bigger than R eventually and we're guaranteed to have it clear so you can see that the phage alone doesn't kill eliminate the bacteria the immune system alone doesn't eliminate the bacteria but when we combine them it can and let me unpack this a little bit what you're seeing here are the results of combining the different pieces together and what you can see at the start we have a non-stimulated immune system the phage is added so we start to see these oscillations at the high points when the phage is not doing such a good job of controlling the bacteria starts to increase excuse me the museum starts to increase and then again it starts to increase and then once the immune system is high enough if the phage can ever drop the bacteria to a lower level the immune system now is no longer inhibited by these density-dependent factors and can clear the bacteria and I should point out that in this model the phage are eliminated before the bacteria which suggests this perfect predator's job may not be to sterilize the animal from the pathogen but rather drop densities low enough so that the immune system can then eliminate the rest of the infection they want me to go over that again or was that clear enough seems to have been clear okay this was what we had thought might happen and just to give you a little sense of what actually happens in science versus what we necessarily read about in textbooks where everything is very formal and proper we had had this idea and I had gone maybe five or six years ago six years now to Liverpool to a virus of microbes meeting and if you get into the space you might want to go to a future virus of microbes meeting there's one in Portugal this summer in July I believe mid-July July 14th through 18th and had gone there with a poster explaining some of these early ideas and one of the first talks I had gone there in part because there was a phage therapy day Dwayne Roach was presenting some of the work that they were doing and they had been thinking about things very much along the same lines I was hoping to get an experimentalist excited potentially about looking about phage therapy outcomes in immuno modulated context it turns out they were already doing in fact had already done it and had found these interesting results so here we again now time again going this way phage therapy saline mock you can see that in the untreated mouse we get this increase in bioluminescence and otherwise even six hours after treatment you can see the effect percent survival 100% survive in the phage treatment it's the same pack P1 pseudomonas originosa case and 100% of the mice die if they're not treated and you can also look at the radiance and to measure the bioluminescence and you can see that the phage free case gets close to the levels of detection it stays there there's not a recursion so this was what was known even going back many years before that in a healthy mouse you could add phage and it seemed to clear infections and there was no recursion reversion of the disease one fact was more interesting is that they did some sort of treatment to the mouse and I know there'll be some jargon coming up here I'm gonna try to focus on the concepts rather than the particular cell types or some of the terminology but just want you to see now it says saline saline phage so the saline by itself is the standard one we know that's gonna fail but now we have these two treatments it's some kind of different mouse in a certain way and in one of these different mice when we don't add phage it's not a good outcome but when we add phage it's still not a good outcome same phage same bacteria the mouse is different something about the immune system has changed and now phage therapy no longer works this is a strong suggestion that these are not perfect predators we can't just add them and hope they're gonna fix all the problems that somehow it's part of a system and we have to understand the system dynamics okay so I don't know if you feel excitement but I certainly felt the excitement when I heard this talk it was one of the first talks in the meeting and we began to engage and work together the problem we faced of course is that there is this failure let me back up one before I go to the slide there is this failure mode you see the circles phage you see this bioluminescence increasing and I'll unpack a little bit more it turns out when one looks at these animals that have died because of this infection in the phage free case and ask the question what was in the lungs what were they colonized by because these bacteria were all supposed to be susceptible it turns out that they were all resistant to the phage infection so we have to expect that when we're doing a phage therapy treatment just like in our initial Monday Tuesday discussion that we should think about dynamics not just as being a homogeneous pool of bacteria but rather that the bacteria may be comprised of two different kinds or classes a susceptible type I think I don't have to change anything for that diagram to still work but also a resistance type which can also replicate and now I'll draw this circle here to make the point that the phage are targeting a subset of the bacteria but the immune system is seeing both so the question becomes if phage can only knock down the susceptible and we know that resistance is possible because there can be sometimes when there's a replication event we generate a resistant I don't need to worry about the loss of resistance or the reversion the question is how could phage possibly control why don't these take off so why does this even work in the first place forget about the immune deficient case how does this even work anyone want to just speculate understand the issues we have a healthy mouse immune system is there but we know that these resistant mutants are possible it's not as if somehow we've engineered this perfect phage so that there's no resistance yes there's resistance we see that's the failure when the immune system is deficient in some way but when the immune system is active this is still happening anyone want to just try and take a gander at perhaps why this could still work if anyone online wants to take a gander never understand these are what the question is yep correct so the resistant are the question is what are the resistant what is our mean it means that these better resistant to phage infection but they still can be cleared by the immune system that was your question so we have two kinds of bacteria they're both visible and can be eliminated by the immune system but only one of which can be eliminated by phage and so if fundamentally after everything I've told you Monday and Tuesday I think I would be wondering why is phage therapy even possible we just select for resistance they should take off yes yeah for the same reason as why coexistence was possible being able to be resistant to to the page makes them weaker to the immune system so you think maybe that there's a okay I get your that is possible so it's it's I think you're suggesting the suggestion is perhaps resistance to phage has a pliotropic effect leading to more vulnerability to the immune system is that what you're saying okay that could be but we don't even need to appeal to even that level of another consequence so let me just unpack it a little bit and ask people to look more carefully here I know it's a little bit hard to see did you want to have a comment maybe what could happen is that since the phages are already competing against this once we have more immune system to use against the resistant ones so that is maybe but that is the essence of what I'm trying to explain so let's take a look here at these diagrams I know they're a little bit hard to see phage are the ones that are up here dashed lines top one the sensitive bacteria are the next one the immune system is in the middle and the resistant are in the bottom everyone can see that phage sensitive immune resistant the immune system here is in a healthy mouse we know eventually it's going to clear but as you can see the presence of sensitive bacteria leads to the generation of resistant bacteria they do better than the sensitive bacteria because they're not killed by the phage but because the sensitive bacteria around and reach high densities before the phage can start to do anything the immune system although you see just a small increase is stimulated and now we have a higher level of the immune system so that when the phage start to eliminate the sensitive bacteria the immune system is a high enough level so that even though these phage are not doing anything to the resistant bacteria the immune system that stimulated can control this small subpopulation of resistance so in some fundamental sense the fact that we added this piece if we didn't have this piece we would just select for resistance it would take off but with this piece the phage are drawing this down so that the sum of this is controllable by the immune system I think I'm seeing enough nods to believe that that went into anyone have a question about that before I move on okay so this I think is quite important because it suggests remember the phage are obligate interest other parasites they must need the the susceptible host or proliferate and that leads to coexistence but the immune system is not right it's being produced it's eliminating things but it's not as if then it can go just disappear because aren't any hosts around it does raise the question of what exactly are the immune partners and the immune system is complicated so what the group then tried to do was to do different kinds of ways to manipulate the immune system so I'm just going to give a big caution you should not try and remember all these different terms in the next three or four slides but I want to try to give you the concepts the first thing they did is something that's called a mind 88 negative or immune activation deficient host in other words there are neutrophils around which are with along with macrophage some of the first responders in part of the innate immune system so they're going to just go and attack these foreign things well before your adaptive immune system learns to recognize and react however sometimes the signals don't reach those components so this is a case where they don't reach the components and in this case you can see we have the original case where we don't do anything and we have the mind 88 saline the mind 88 phage and what you can see is that the phage treatment doesn't do much better in fact it's really not that different from the non-treated case and when we look at the deficient signal you can see in the saline case it takes off there's some sort of deficiency the mice are going to die in the phage tree case it looks like after 24 hours things are going well but then there's a uptick again in the bioluminescence indicating the proliferation of these bacteria and when they look they found that those were resistant and this is precisely what our sort of models say the models say that what's going to happen is this unstimulated immune system you see that there's density but it's flat so it's not increasing these small differences factors of two or three make a difference we have phager taking off driving down the sensitive bacteria but now the resistant bacteria begins to proliferate the phage doesn't see it doesn't eliminate it and the immune system is not being activated doesn't respond and then we get failure because of the swap around the same time scale because this is a signaling system that should speak to the innate effect or cells part of the innate immune system and they knew enough about the mouse immune system they wanted to first look at different pieces so they asked what happens if you try to do phage therapy in a mouse that had no innate lymphoid cells B cells T cells and even if you don't know much about immunology you know that that sounds like a lot of the parts of the immune system right so all the kind of adaptive parts as well as parts of the innate system in lymphoid system and it turns out you can eliminate large portions of the immune system and phage therapy would still work effectively so this suggests that they're drilling in a little bit closer to who the alliance is with and it turns out that it suggests it's probably with neutrophils and so when I showed you that anti-gr1 case that's basically a drawdown using an antibody to draw down and deplete these mice of neutrophils so now not only is it they're not signaling correctly they don't even have these neutrophils to respond in the neutropenic case where you've drawn down the neutrophils you see that you can add phage but you just get the proliferation because this resistance population is now unchecked it can't be killed by the phage it also can't be eliminated by the immune system and that's exactly what we see in some of these models that are all based on this idea which means that phage is not a perfect predator but rather there's some synergy or alliance which is required for effective therapy I want to just say two more things and I'm a wrap up this section and move on to the third which is that you might ask and wonder can this even be added in advance or and no one has asked me why there's no relationship here right if I think about viruses then of course the immune system sees and reacts to viruses it turns out that that was a concern and time scales here are short but nonetheless what in this experiment happened is they added phage prophylactically four days before they added the bacteria and there is a time over which the system sort of dilutes out the phage and we actually then measured that by adding phage and then going back and finding how many could be extracted from the lungs but what you can see is that adding phage four days in advance still leads to a hundred percent survival here you can see what the dynamics say it says that we should have some exponential decay but if you add in a high enough dose you basically have this dynamical response right at the beginning and there's another piece of the story here which is that there wasn't any evidence of a cytokine response in the same line versus the phage treatment and I'm again this gets a little bit jargony red means alerting the immune system if you add LPS parts of the bacterial cell wall and other cell cell surface contents the immune system will know about it if you add saline you don't see much in phage you don't see much either so somehow the phage which is why we didn't add this part are not really telling the immune system to react negatively to them that itself is a very interesting question why is it and how is it that phage are not stimulating negative responses from the immune system nonetheless that's not one we're going to cover today so there's not a lot of priming of host immunity which explains why didn't add some sort of weird negative feedback loop or even clearance of phage by the immune system which obviously would work to undermine our efforts of phage therapy okay any questions about that I've kind of explained a big picture of what I'm calling immunophage therapy and I have one last little piece to explain and then I have about 45 minutes to talk about the other thing let me get to the end of the session maybe there'll be questions yeah yeah go ahead so I don't understand why you should expect something different from that given that you don't have bacteria in the lung I mean you don't have bacteria the phages can replicate with the in the lung so it was partially from the immune perspective if you're an immunologist or even of you might not expect it because you work on models but if I'm an empiricist and I also don't know whether or not in those few days maybe I'm going to elicit remember this is a very rapid clearance so if I have four days it could be that yes there's some clearance but maybe that gives enough time to the immune system to start to clear things out even faster or learn that the phager shouldn't be there that could undermine again if there was a negative if this had a stimulation and a negative feedback loop it's trying to eliminate some of this part also there's a practical part too which is there's some questions of can you use these as prophylactics right rather than just a response and there's a timing issue so you have to use it within a certain amount of time but you can use it in advance we don't know how far back you can use it does that help a little bit yeah okay I mean again talk more about okay so let me give one more example of a phage therapy concept in practice one of the first slides I showed you was that gentlemen the dentist from New Haven Paul Turner which have used and now I've gone into clinical settings using phage to treat people with pseudomonas aeruginosa lung infections and there's a very interesting part of this story so I'm going to just try to explain it in the next few minutes there is a phage omk1 that can infect pseudomonas aeruginosa what's an interesting thing about this particular phage is it gets in accesses pseudomonas aeruginosa through an antibiotic efflux pump who has heard of an antibiotic efflux pump I'm a physicist that I probably zero almost effectively a few of you though term should explain what it is if you are to add antibiotics they get inside cells some cells chuck them back out expel them through these efflux pumps so recognize them and send them back out and if they have these efflux pumps you can add your antibiotics but you're not necessarily going to kill those bacteria it turns out that this particular phage gets into the bacteria through an efflux pump Paul is a eco evolutionary experimentalist really an evolutionary biologist and thought for various reasons what if we were to apply antibiotics and phage at the same time now you can imagine a type of bacteria that has the efflux pump we've added both so the antibiotics are going in but they're going out but the phage are going in and killing the bacteria that have the efflux pumps perhaps there would be a resistance excuse me I a mutation and now they don't have the efflux pump now the phage can't get in but the antibiotics can't get out so their ideas that if we use both together we might be able to eliminate these pathogens because whether or not you keep the efflux pump or we'll get rid of the efflux pump the bacteria is still going to die does everyone understand that the sort of different case here was before I only had one therapeutic agent here I have two and one of them is going to hit either one of the two options either the phage resensitive antibiotic resistant or the phage resistant antibiotic sensitive this is what is actually being used now in multiple clinical trials this concept now it doesn't hit everything but we began to ask the question is it even that's kind of complicated but is even that simple because when you have an initial composition we could have things that are antibiotic sensitive and we could apply just antibiotics here's the let me orient you here the levels of antibiotic this is the minimum inhibitory concentration so if I use more than that I contend to kill the bacteria that are sensitive antibiotics less I usually don't the outcome of bacterial density is listen the colors yellow as high blue as low and this proportion is whether or not all of them start off as antibiotics resistant and here they're antibiotic sensitive and what you can see is that if I add antibiotics we end up getting a selection so that by the end everything becomes essentially resistant to antibiotics sensitive to phage so if we add just antibiotics alone we've already seen that that's no good if we add antibiotics and there is an active immune system it would take a very high proportion of these antibiotic sensitives and a very high application to eliminate it otherwise we don't see elimination if we add antibiotics and phage we unlike the some of the original supposition suspect that what's going on is that when we use a lot of antibiotics we select for phage sensitivity and there's coexistence and when we don't use enough antibiotics we end up selecting for antibiotics sensitivity the last one is the one that we think is going on practically that when you have this dual application what this fundamentally is doing is driving down densities and with the immune system allows to clear what's notable from a therapeutic side is that we might even be able to use less antibiotics even less than the inhibitory concentration which would be very good news because we have less likely chance of generating some antibiotic resistant further selecting for antibiotic resistance in our treatment okay I realized that was a lot to take in the thing that you should take away from it is that these kind of models can also be applied to combination therapy and that's what we're trying to work on the same idea being that we have these synergies but that these resistant properties may be relevant to something else like you asked about maybe there's a consequence with respect to the immune system here we explored a consequence with respect to antibiotic sensitivity so those two were coupled okay if you want to read more about it then I recommend there's a nice essay Burmeister at all showing that in fact in general when we think about effective therapies they're often being done in combination there's some synergy when we use these single acting treatments they're unlikely to work because you basically end up selecting for resistance and keep in mind again for anyone I'm gonna post all these slides I've been posting them all in slack hopefully you're aware of that I turn off my slack notifications so I don't know if you reacted to them but I have put it and I'll put them out there today so this means synergies with the immune system combination treatments in a third class these evolutionary water cocktails really trying to think hard about where evolution might go so that you end up in some ways prospectively thinking about evolution and then adding different kinds of phage at the beginning okay I did two parts and I have one more part this will get back hopefully to some more foundational stuff and it's gonna be fun we're gonna have a great time so any questions before I embark on this last section cuz I'm gonna go in a very different direction in this last part they've been battled down because the last couple days I've been going to the board more and you see more satisfied when I'm going on the board today I've given you a lot of biological material this next part will be a bit more conceptual probably easier to hear given your background but we will we will see it's okay I think that we'll get some questions in a bit I have about 40 minutes left this is the last lecture on phage remember tomorrow I'm going to be doing a more interactive laboratory I'll post the information about the laboratory this afternoon so that if you want to look at it in advance you can do so I'll spend a few minutes at the end talking about it but I've been talking for the past four days three and a half really about this lidic modality the modality behind which viruses act as predators although I just showed you imperfectly but I want to point out that is not the only modality but by which viruses can interact with the microbial host there are a number here is one host DNA somehow is there virus and Jackson's genetic material I've been focusing on this route this lidic route replication release repeat for the last 40 minutes here just to broaden have some fun and also in the spirit of these are not perfect predators they're parasites sometimes they don't even kill what they do instead is integrate into the host genome the infected cell divides the blue bit is the phage and it now is replicated the particle is an under selection the life cycle you might say is under selection and the information about how the life cycle can work is contained in the genome of the phage which now has made it to a new cell occasionally this blue piece can be induced come out of the host genome and begin to reactivate the lidic cycle and essentially go this way right so usually these are operating the separate parallel circles but we can sometimes make a figure eight kind of diagram and end up over there and when that happens that's called induction so we have lysis and lysogeny and some viruses can only do the lidic cycle they're obligately lytic and the phage that can do both we often call temperate and I'll write that word on the board so we have these obligately lytic phage virulent phage that always kill and then we have virulent phage which can only do lysis and we have temperate phage which can do lysis or lysogeny and I'm going to get into some of the modeling the way we think about this but I want to give you a little bit more of the biological background before doing so so this is a modality it's not like this is just recently been discovered but it's been known since the very origins of modern phage biology going back at least 80 or so years to Lewoff and others so here you have the schematic you have some sort of decision switch and then there's lysis on the one hand and integration on the other I just want to give you a picture of what this can look like to make a point that these tempered phage this is a case of phage lambda which is often used to study but there are other examples here you have infecting phage can everyone see them little green things and the reason that they're green is this has been engineered to express a green fluorescent protein which gets embedded on the capsule which doesn't interfere with infection you can see it up there however is also engineered to express a red fluorescence whenever the lysogenic pathway is initiated and so what I hope you can see is infections in some cases these cells are infected and then burst an outcome or green particles in other cases you can see lysogeny and now the cell divides and both daughter cells are red I just you kind of see a version of this in action everyone get that piece of the puzzle so now we have some basic bio in mind and one of them point out this is just a wonderful system and we could talk about it for a long time these outcomes are not just coin flips or that are fixed forever you might think of these as some little parameter that the virus just carries around with it I'm a point seven temperate virus meaning 70% of the time I do lysis and 30% lysogeny the other one is a 10% right what is actually happening is that the outcome depends on how many viruses you add here is the viral concentration they were able to look and you can see here the average viral concentration measured at the single cell level population level and again single cell level the point here is that the more viruses you add per host the more likely is it that the host is multiply infected multiple viruses of the same kind are inside that host okay does everyone understand that back it up one side so take it away you saw something but you probably even know what you saw so let me try to say it again you see these infecting pages it could be that one lands on the host maybe two or three or four or five so just to get you all thinking because it's quite critical here we have this multiplicity of infection one two three four five here's the percent of times that this ends up not leading to the death of that host cell ever understand this relationship who thinks the more phage that enter a cell less likely there's going to be lysis and who thinks more likely to be lysis right because it percent lysogen maybe I should do it the other way around would you like me to put lysogen or lysis on this y-axis I don't really mind anyone have a preference my sergeant is fine raise your hand if you think you get more viruses you think they'll be more lysis as an outcome so add more viruses it should kill more it's okay don't you have to look around it's as we're not it's as you can have your own opinion one two three four five six seven eight nine ten about ten twelve who here thinks the more viruses we add the less lysis we're going to get two three four five six seven so more virus more lysis more virus less lysis there's always a third option who thinks it won't really make that much of a difference or is confused fine fine confusion and that you know this is a minimum right that's a lower bound on confusion might be more of you I appreciate the folks who admitted they were confused okay let me ask people who said more virus more lysis why do you think that's the case feeling okay so at the beginning they have to survive so you think they would replicate fine if I for example said that you were exposed to a certain level of whether it's ours kov2 or some other virus which you think will be worse for you more or less probably more seems worse right the more viruses you have it seems like that would be a worse outcome and out of these two outcomes it seems like killing is the worst outcome for those who said more virus less lysis any thoughts why did you say that someone I haven't heard from I've heard from you today I've heard from you today I haven't heard from so many other people more virus less lysis someone who made that suggestion who's willing to no I've heard from you I want to hear from I'm interested but I want to hear from some other folks today more virus less lysis why do you say that someone who raised their hand some of you did it it's okay you can share it even if it's just saying you just think so anyone have a thought remember I only have a few minutes left in this class so go for it I don't know maybe I already have a lot of viruses so I just care of go through generations of bacteria without killing them I don't know so maybe these bacteria already have some other viruses inside of them or they've been evolutionary exposed what are you trying to say meaning that they're already viruses inside or there's no need to destroy bacteria I already have a lot of viruses I just care about going on with the generations so so now the you are anthropomorphizing yourself as a virus you're saying I already have a lot of virus on the you is the virus right you've taken the perspective of the virus okay fine okay so maybe the reason why this is happening is somehow they're able to understand that there's so many of us around why bother making more sort of a group selection argument right these viruses are nothing just even selfishly about themselves they're worried about how all the viruses are doing that suggestion is actually not too dissimilar to some of the rationale that's been in this field for a while but I'm going to try to unpack today but it's a bit I have to be cautious about that the viruses are not such thoughtful things about all the you know the feelings of the other viruses and also it sounds a bit like group selection they're looking out for the others now maybe I tend to find that Darwinian natural section argument could end up in a group selection argument but we should make the other argument in practice this is the evidence it's not always this way more virus more Lysogyny so it turns out that it looks like this more virus you had you actually have less litic outcomes which I think from a practical side from like if you imagine the molecular consequences inside a cell it seems odd that you would add more of a virus and have in some sense a better outcome you have a higher chance of surviving five viruses infecting you than one you might look at around another way of course you didn't quite survive you're now Lysogen you're some sort of Chimera part virus part bacteria nonetheless that's what folks find and you can see here they were even able to do these experiments with single phage and count the outcomes and I want to point out that it's a bit fuzzy there so there's still casticity and it doesn't go from zero to 100 but it gets pretty close to 100 the more virus you have it's almost certainly going to be the fact that you're going to end up with a Lysogen and there's still a chance that you still might generate a Lysogen even when you're a single infecting phage I want to also point out that even the coming out part the induction process so there are two decisions to make if you're a temperate phage one is do you go this way or that way another one is do you keep going around the Lysogenic loop or do you pop back out induce and make induced Lysis and it turns out if you do bad things to the host like add a bunch of UV let me just explain this concept if you add a bunch of UV which is not good for the host the host might die you actually can induce higher levels of phage coming out nearly all of them at very high levels and the interesting mechanism there I don't have time to get into it if you're interested you can look up mark potassium is book a genetic switch but in essence when there's DNA damage there are these DNA repair enzymes that go around to try to fix things phage have a way of producing a molecule that keeps them inside these repressor molecules so they're keeping these little repressor molecules which usually protein show up as dimers and you have this sort of Pac-Man like enzyme that's going around supposed to be looking for DNA damage to repair it but this has evolved and actually cleaves these repressors and so now the phage detects DNA damage by proxy and it no longer stays in and pops out yes you had a question maybe grown but I mean even when the virus continue the number of virus continue growing the no I mean the the process of lysis reach a limit and it started constant so that's what you're seeing here is a snapshot in time these are not dynamics the x-axis is not dynamics yeah the x-axis is if I start with a fixed concentration then I get these kind of outcomes to figure out the population dynamics which I'm going to do in a little bit I have to combine both ideas together just my question is if that's continue growing I really should reach a limit or something but doesn't kind of continue growing right so you have to think that imagine we start with a low level of viruses then most of the cells will be in this category when they're infected and as the dynamics proceed because you're not making a lot of lysogens I'm going to explain this in a moment you're basically making more virus particles it will tend to move it over here but then you will turn those infected cells into lysogens and maybe then you will have fewer viruses around right so there's a feedback loop there which I think you're getting it so that's how things that's how they are temperate but I'm now going to ask the question of why be temperate so people had built and I will it does this just a reference really the premise is temperate phase do better when few hosts are available and extras more south extra cell mortality rate or high let me try to explain that it's a bit of a group selection argument but it basically says imagine you're in this case over here five viruses are inside those hosts I think you've said well let's think about all the virus I have enough already if you if the viruses were to lice that cell if that cells infected by five things the likelihood that we have uninfected cells out there might be pretty low and so maybe you burst that cell and there are no other cells to infect and now the infection is over and that could be bad or maybe the life outside for a viral particles quite bad it gets cleared for whatever reason very rapidly those are some factors that might seem to dissuade a virus from killing instead enter lysogeny the problem was this is a late early 80s paper in spite of the intuitive appeal of this hypothesis were unable to obtain solutions consistent with the here so it's been hypothesis now it turns out that this hypothesis is also seen in natural systems if one looks these are marine systems looking at low productive times and high productive times in other words when there are a lot of bacteria around things are pretty good with aren't as many bacteria and you look for visibly infected cells or lysogens and these aren't a marine system what you see is when times aren't there's not that many resource not that many cells around you tend to see more lysogens which is sort of consistent there aren't that many cells left let's enter lysogeny when there are a lot of cells around you rarely see lysogen you see a lot of visibly infected cells and this is called in ecological context a seasonal time bomb because depending on the season in the case where other may cells available the phage get into the host and then when there seems to be a lot of productivity a lot of cells around so maybe if they were to release they get the cells by themselves they induce and then we see a lot of visibly infected cells so everyone get this idea okay and I'm just going to set up what I'm going to do in the last 25 minutes or so I'll probably skip some slides here is that a few years ago people made a hypothesis that was the opposite so I've been showing you that somehow if we have perhaps more cells around we had lysis they actually said the opposite that lysogeny is positively correlated with increase in host density and productivity and to do so they were trying to explain the following observation that if you look at the number of viruses and the number of microbes and this is a one-to-one line you see that the data tends to be a little bit shallower you might not think that that's that impressive a data fit and you would probably be right nonetheless this is the claim published in nature 2016 that this was less than a one-to-one fit so their hypothesis as well maybe it's less than one-to-one because at higher densities instead of killing these viruses are becoming lysogens producing less viruses and that would lead to this decrease even though there's an increase in total number of the relative ratio of virus the microbes going down as microbes go up ever understand a virus going up but the ratio is going down because it's now less it's above one-to-one at low and less than one-to-one at high and I'll skip all that other stuff this model did some of the things that I've just talked about here it says that microbes grow their lice they have some intrinsic mortality and viruses release they decay in this piggyback the winter model they said that the virus release takes the standard form that I've just been sharing with you and multiplies it by n over k even though they said that there was an increase in lysogeny actually the lytic productivity increases within so it's sort of backwards and there's not even any lysogens so we went and investigated this and tried to explain that in fact this pattern of the ratio between virus and microbes can go down in a generic predator prey model it's just sort of a generic feature of these models it's not something special about lysogeny and not only did we explain that but people then looked at the relationship between microbial cells on the x- axis and indicators inside the viromes for what are called indicators of lysogeny are the viruses that are around probably temperate viruses the claim was that as microbes increase we should see more lysogeny does anyone see a relationship between the y and the x-axis there would you say that's a strong relationship what about that one maybe that one maybe goes up and down but certainly doesn't go straight increase our claim you can see the p-values and the r-squared relationships even if there was a relationship it's explaining very little of the variation and we didn't find any evidence really for significant relationships so the takeaway is there in fact is an interesting idea that we see in data from the single cell level but in empirical systems there seems to be this absence of a correlation and even absence of evidence between lysogeny proxies and cell densities and now I'm gonna have to I think I'm gonna skip a few things no I'm gonna do this now so I want to finish up in the last 20 minutes I know there's been an intense lecture I'm doing a lot of things today what environmental conditions should favor lysogeny rather than lysis I'll try to ask that with a lesson that you may have heard an expression a bird in the hand is worth two in the bush this is not the same expression that is made in every culture has anyone ever heard such a no has anyone ever heard a culturally equivalent term like this what is it you can say it in Spanish if you'd like how many how many I have to be flying how many in the hand versus how many flying right so you have one in the hand and a hundred are flying so this idea is that you have something and because you've acquired it it's worth many times the value of this if the uncertainty of maybe acquiring it so I I've said this before but I find it interesting to know the equivalence value in different cultures it tends to be this kind of statement and I imagine depending on the uncertainty of how hard is to acquire resources different cultures have established different equivalences this is just two in the bush a hundred says ah I better take what I have because it's pretty darn hard to get that right so it's something about sometimes how hard it is to acquire resources keeping in mind that a virus needs to find a cell to communicate replicate it found one why kill it sure there's a lot of them out there but there better be a sufficient number of them going back to your point because if there aren't that many out there perhaps it's not worth killing this all you should stick with it even if you're not killing it with it you're still replicating along with it but what is it what is this equivalence in this virus micro world so to do that I want to go back to some stuff and hopefully this last piece will both be exciting and connect to things that I did earlier in the week let's go reason through a little bit the dynamics of virus and microbes but not the population level but at the individual level and to do so I'm going to anchor myself in these litic viruses here we have a bunch of cells phage comes into contact injects it's genetic material into the cell I'm going to call this the mother virus okay called the mother virus not a mother cell because then there's this license event and there are all these virus particles that come out and when I say the word virus I assume you think of a particle I assume that's what you think in your mind when I say the word virus a hundred viruses come out that's not the fitness of that particular virus most of those virus particles maybe they're defective maybe they absorb to some other material they never find a new host in this example only three daughter viruses are produced infected cells containing that viral genome I would argue that that at an individual level means that this mother virus produced three daughter viruses in her lifetime I know that's a weird thing for me to say at some level but I'm going to say now I'm going to look at an alternative world and maybe not blow your mind but I will try to get you to think differently here we have this virus comes in it's a temperate virus can do either in this particular case this mother virus has initiated Lysogeny the mother virus divides made a daughter virus her fitness is her own divides again number two divides a third time number three and then the mother virus dies I would argue that this mother virus also has a fitness of three but there's been no Lysogeny from the individual perspective these two mechanisms now I haven't talked about time yet and time is important have produced individually the same number of progeny using completely different mechanisms and I'm going to use this as a means to try to revisit this question of why be temperate not how there's all these molecular details and if you think it was painful listening to me talk about Facebook imagine me telling you all the molecular details of how Lysogeny works but really be painful I just gave you a taste but I want to try to explain the concepts here from a dynamical systems perspective two vastly different strategies can lead to the same I put in quotes fitness your individual fitness how does this depend on cell densities there's context so I gave you this example three and three but it must depend on what's going on in the environment okay to do this I'm going to show you the litic because I think we've grounded in that and then I'm going to extend it to the late and even a third time here we have our individual view here we have our population view susceptible cells infected cells lysis and I've written models like this enough that I think you can accept that we have all the ingredients we have cell replication we have infection death of cells infected cells as well as viral decay infection and then over some exponentially distributed period we get a lysis event releasing beta new viruses back into the environment okay now how would we figure out if there's invasion if I ask that on an exam then you would probably find the fixed point you could probably do it for just the infected subsystem by the way don't do the full thing find do a linearization find the eigenvalues see if any of them are positive and then see what the condition is for them to be positive a threshold criteria it turns out if you do all those moves and remember I talked about this earlier in the week you find that the eigenvalue one eigenvalue is positive if and only if this number is greater than one this number which is based on these parameters can be written equivalently as the birth size times the probably a viral absorption to host before decay probably lysis before washout I just flip that up there but I just want to keep that in mind if we think about this cycle here we have the mother virus here's the chance that it actually burst before it dies due to some other mechanism because washed out eaten by a grazer who knows what this is how many new virus particles are produced this is how many of those virus particles find a new daughter cell probability times the number times probabilities the average number of new infected cells caused by this horizontal route which is why I write our HR horizontal must be greater than one again if you were to do the sort of algebraic messy thing and I went over that a little bit you would get the same conclusion but intuitively you can think of this just as a threshold criteria of does one infected cell cause more than one infected cell on average strictly through a horizontal transmission route meaning leaving that cell and going horizontally to some other cell you should also note that there is an S star there that's key because the probability of finding a new host once I produce this virus depends on how many hosts are around so now we get to go back to that claim by steward and Levin that when cells aren't that available doesn't seem to favor lysis now we see it pop out immediately and also we can see the fact that when extra more cellular mortality rate is high M that also isn't going to favor this because that's going to drive down the ability to transmit horizontally any questions okay which means that we could plot this individual level fitness is RHOR as a function of the birth size and the environmental context I could do it for other variables but I didn't hear for birth size you can see I'll be on this side here you could have a hundred virus particles produced but still go extinct in other words never invade that population if the population were too slow or you could have a very seemingly inefficient virus but in a giant population of hosts and they could proliferate as a whole more susceptible cells around is favoring a lytic mode less is favoring less and obviously these traits obviously make a difference so the takeaway here is that ecological conditions with more susceptible health cells and viral traits with both more efficient infections and that don't have immediate mortality in other words their extra cellular variant survivability is high favor this late lytic antigenistic mode let me do the same thing now for latent viruses for latent viruses I have to write a new kind of model I have logistic growth I have infection cell death here I have a new variable called L for lysogens lysogens are cells with an integrated viral genome inside you can see here we make these lysogens and I've kind of combined them so that sometimes they immediately generate these lytic modes that I have a probably P and otherwise they're going to reproduce and make more lysogens if I want to just look at only the vertical mode I can set P to 0 Q to 1 and just focus on if I started with a single lysogen so I have this virus it integrates I start with a lysogen will the lysogen population proliferate well it will depend on this ratio and this ratio can think of 1 over D prime is the lifetime of that cell because as a cell death rate of D prime 1 over D prime is how long it lives so the time on which it lives times its division rate but the division rate also depends on cell context and also perhaps changes in the intrinsic growth rate because it's a lysogen not in fact not a susceptible cell so all this is is division rate time cell lifespan and again if you were to do the modeling parts and ask the same questions you would find the same answer this is a threshold criteria it says that these lysogens can take off and proliferate what's interesting though is it has exactly the opposite relationship to cell density the more cells around the less good it is to do lysogen because you're a rare lysogen and so if the cell density gets very high then you have to compete with all these other cells and if there is a benefit of lysogen you may be able to do this for a very long period if there's a little bit less a little less and obviously if there's a cost it may not be possible at all right if that phase comes with a cost to you as a cell and division you might not be able to invade but you can see there's a negative relationship for lysogen a positive relationship for lysis so this says ecological conditions with a reduced niche competition direct cell benefits or low variant survivorship all favor latent strategies I want to do one last thing and I had some other things I was gonna do but I'm gonna get rid of those I just want to point out that there's even other modes that are called chronic viruses that can release virus particles without killing the cell so they bud off sometimes with an increased death rate if you were to analyze this system and again it's that same fine-fixed points linear eyes calculate eigenvalues but it turns out that in this case there are two ways that you can get new progeny here we have the mother virus and in a chronic produced case you can produce virus particles along the way so here there's a division event but then some virus particles get out and infect horizontally so we have to add these two components together and we still can get the same level of productivity through this mixed modality and when we calculate the threshold criteria for invasion we get two components a vertical part which is how many new cells do I divide while I'm a chronically infected cell and a horizontal part which is a little bit different it says how long 1 over D prime how long am I infected the rate of virus production during my infectious period times the probability those viruses find a new cell it was quite intuitive and I prefer this mode because I think you can understand it and I don't have to do any of the algebra it has all the essence of the ideas if you then look at these kind of cases you see that when cell densities are low having a lytic mode is very difficult to sustain when cell densities are very high that can be better but at low densities being chronic or being a temperate phage has advantages because you found this host in a bloom the end is not high enough to go and kill it and find some other new hosts okay if you want to read more there's a paper that I will put on this afternoon it has all the details of how we do these calculations in fact tomorrow my hope is that you're going to do some of these calculations and build some of these models and maybe even as homework get into some of this okay good and the takeaway point is this is really a much larger range than anticipated in which non-lytic modes tend to be favored and now I'm going to try to pull this all together in this one slide and I don't know I have about ten more slides to go but I'm not give any of those slides and just wrap it up here with two or three conceptual things I started by asking in the second part what environmental conditions should favor lysogen over lysis and I've ended up using the basic reproductive number and this is a nice segue to my next week set of lectures on epidemics and epidemiology in each of these cases I've tried to calculate invasion through a threshold criteria the average number of new infected cells produced by a single infected cell in otherwise susceptible population and I haven't used virus particles because otherwise it would seem that lysogeny isn't a totally untenable path it postpones making virus particles to some indeterminate future but if we measure viral fitness in terms of life cycles starting in cells and ending in cells we can actually compare directly we can see the benefits accrue at low densities okay I'm going to skip a bunch of stuff including Feynman like diagrams doesn't even matter I can tell you about that some other time we've taken this diagrammatic approach but I wanted then go to a big picture question what is a virus when I say the word virus I'm going to assume how many of you just think of this or SARS-CoV-2 you think of a particle some of you might think of the genome right maybe you think of the genome that's the virus there's this other state in which you have litically infected cells or even a lysogen which of the genome is inside a cell what I'd like you to try to think about when we think about viruses that actually all of the above and in fact these are states as part of a life cycle and the life cycle is under selection and the genome contains the information that embeds this life cycle trying to figure out the environmental impacts and conditions by which choosing amongst these pathways and evolving different levels of choice amongst these pathways is something that I've been hoping to do much more this year I still hope to but it may take many years to come if you're interested in this topic in a broad sense the other paper that I'll post is a really biologically oriented review the virus evolution paper will be much more readable given your background it'll be more of a dynamical systems point of view nonetheless also post this which talks about these sort of rules of life for viruses microorganisms in a broad sense it just came out late this past year so it should still be relatively fresh okay wrapping up the dominant paradigm of rapid lysis and large phage impacts on mortality should be reexamined if we're going to do a lot of things first of all interesting fundamental questions it's relevant to phage therapy because we have to deal with the fact that these phage are not perfect predators and it certainly is relevant in environmental context in which there's a lot of these examples of phage not acting politically but in fact acting through this other temper path integrating rather than killing okay that is the end of my series of phage ecology lectures you made it through tomorrow I'm gonna ask you to do some work I've been doing a lot of work this week and I'm tired when you do something young strong you will do stuff so tomorrow I'm going to ask you and I'm gonna work with my town trying to figure out the room but each of you I understand has a computer you can work either with partners there will be a laboratory with both a Python and R I also have a MATLAB version but I think I'm just do Python and R versions which will allow you to do work if you are not that experienced encoding it has a long introduction period which actually will train you up to the point to build and simulate nonlinear dynamical systems so my view is that this is a course for you to learn things I don't know where you're starting from what I'm going to ask you to do is do something I don't know your level so if you're more beginner I'll ask you to do the exercises the exercise answers will be included don't look at them try to figure them out but I want them there because I want you to actually know the answers under the reference then at the second part of the laboratory will include more advanced questions recapitulating some of what I did here today and I haven't provided the answers I want you to try to see how far you can get next week I'll also do another laboratory and we may talk about what you did because it won't be enough time just tomorrow so it can be something that you can do along the way okay and I think there's some sort of grade for this course doing something would be helpful as everyone give me some kind of Python notebook or our markdown document something that I can see that you tried okay that's all for today last comments are just that if you know of people interested in this space I do have postdoc positions PhD positions if you want to learn more about viral nanostructure cross scales I'm going to continue to advertise this and when I leave next week and go back the week after I'm running a full week workshop and some of the talks will be online and if you happen to be based in Paris not this spring it's getting a little bit late or next spring I may have some internships available so just want to let people know about those and I think I'm out of time today so just want to ask can you raise your hand if you have a laptop great so bring your laptop tomorrow at 9 we're gonna do a lab you have Python is it Python okay yes yes it looks like does anyone want or should I just give the Python version well I'm just no Julia you want to if you want to translate into joy you can do it no I don't have Julia I don't have every because remember this yeah no I'll bring the Python right tomorrow see you all enjoy coffee break okay thanks