 on how we understand microbiomes, because we're going through a period when technical advances are really beginning to overturn our ideas of microbial communities. And crucially, from my perspective as an evolutionary microbiologist, I'm interested in kind of what has agency in microbial communities? What are the entities which are evolving and how can we understand them? Okay, so the microscope was invented about 350 years ago. This is one of the microscopes like Lovendog used to reveal microscopic communities for the first time. And this is a quote about Lovendog here. So very few could follow Lovendog, mainly because of the technical state of the instrument and sheer lack of understanding in others, because microscopy takes the human consciousness into an entirely new world as with a newborn child. And I really read this through the lens of microbiology, because I feel like microbiology really does take us into this new and exciting and alien world. And one of the exciting things, of course, about living now is that we can get microscopes really easily. We can use these to look at the microscopic world and really bring this world to life. So what do we see if we get a microscope, we look down at what did Lovendog see when he looked at his microscope? Well, he saw a world that looked a bit like this or at least it represented like this. And it was represented like this because first of all, it's shocking. So this woman here is looking down at microscope, looking at this microbial world. And it's so shocked she's dropping her tea, perhaps also because she realized that these are the sorts of microbes which were living in the water used to make her tea. But of course, this is an illustration and it's been drawn with artistic license. And if you look at some of these microbes in this illustration, many of them look like animals. And indeed, this idea of thinking about the world, microscopic world as being a world full of little animals was really embodied by Lovendog's work. He called these microbes DHNs. I'd probably horribly mispronouncing that, sorry to any Dutch speakers, but the rough translation of this is little animals. And it was Oldenburg's translation which gave us the word animicule. So glass lenses were giving us a window into the microscopic world. And our macrocosm of animals and plants provided a sort of metaphorical lens to understand this alien microscopic world. And the consequence of this is that we've been really inclined to see microbiomes as collections of individuals. And we've been inclined to see them as collections of individuals which are kind of a bit like little animals, right? We name microbial species in the same way that we name animal species. And I kind of think this is quite interesting because this is an example of how our social, what we understand about the world is in sort of being imposed almost onto this other world. And it made me kind of wonder, it is really the way that the microbial world is, right? And the reason I was thinking about this is because the genome revolution is really overturned or at least is really beginning to challenge the idea of a microbial world consisting of these little individual species, right? So what would our conceptions of microbiology look like now if Love and Hook had invented the genome sequencer instead? We'd be going out and we'd be looking at these collections of genes. We'd perhaps be looking at pan genomes in the same way that James McEnany was talking to us about a couple of days ago and seeing pan genomes and seeing them as the sort of unit that we needed to understand rather than these microbial individuals. I think we have a different idea of agency, different idea of individuality would maybe use different metaphors to talk about the microbial world. So this is our kind of, I think the kind of culture clash I suppose or the kind of revolution we're seeing in the way of understanding microbial communities while undergoing at the moment. And this is because of course, we look inside our microbial communities, we see these, we call them microbial individuals but the important thing is that they are evolving quickly and they're evolving quickly by horizontal gene transfer. And this is really an understanding of the microbial world that was only really available to us for genome sequencing to understand the true extent of horizontal gene transfer. And it's really quite astonishing. So for example, bacillus is gaining and losing genes at four times the rate of mutation. Pseudomonas, it's four orders of magnitude higher than that. And this means that thousands of genes are being gained and lost in the time taken for just 1% amino acid divergence. So this fact that horizontal gene transfer is making this really important contribution to microbial evolution. And the fact that we're really starting to understand the microbiomes more and more by sequencing rather than just picking out strains and seeing what those individual microbes are mean that we really need to start thinking about mobile genetic elements, not as properties or traits of the microbes that carry them but as integral constituents of microbial communities. Okay, so yeah, no more technical elements. So this is my kind of thought interaction. I hope you're all okay with it. So I was really taken by the title for this conference series. I think it's a really nice title, this idea of a plasmids of vehicles for antimicrobial resistance spread. And of course, vehicle is another metaphor, right? We can sort of think about vehicles in lots of different ways or lots of different kinds of vehicles. This is another metaphor that we're taking from our world of trains and cars and we're sort of using it to try and understand this alien and essentially unknowable world of the microbiome. But I'm sort of using this as a little bit of way of framing my talk today. So I'm gonna talk about three different kinds of vehicles that we can sort of think about plasmids as being. We're gonna talk about delivery vans. I'm gonna talk about party bus and I'm gonna talk about taxi cabs. Okay, right. So I wanna understand plasmids and microbiomes. Plasmids, I mean, I don't even know why I bother with this slide because you all know this stuff already, right? They're these quintessential mobile jet elements. And the key thing is that they're these Darwinian entities. This has been covered already. I'm sorry for putting this slide in. What I'm gonna say is that today on my talk, I'm gonna focus on three key ecological mechanisms which cast light on this kind of world of plasmids in microbial communities. And these ecological mechanisms, we'll be talking about molecular mechanisms as well. I'm saying this because ecological mechanisms are just as much important to understand as the molecular mechanism when we're trying to understand microbiology. It's important to understand ecological mechanisms as well. So first of all, I'm gonna talk about selection. So when environmental conditions favor plasmids, then I'm gonna talk about infectious transfer which is when the plasmid is in control. And finally, I'm gonna be talking about compensatory evolution. Okay, so this is a model system that I've done this work with. It's a set of large mercury-resistant plasmids which came from the exact same habitat as the bacterium that I work with Pseudomonas florescent SPW25. So when we take these plasmids, we put them in this bacterial host. This is a kind of a new pairing that we've not seen already in nature but it's one that is very realistic because we know that these organisms are sympatric. Okay, so selection. To understand how selection affects plasmid carriage, what we can do is we can take these plasmids, we can put them in a test strain and we can do what we call competition assays where we do a head-to-head competition between plasmid-free, plasmid-containing strains and measure the relative fitness. And what we can find when we do these kinds of experiments is we see that plasmids have a fitness cost, right? I would definitely not the first person to find this all throughout plasmid biology, we see plasmids have fitness costs. And one, the conception of plasmids is sort of like this adaptation module providing these key adaptive genes. And indeed, we can kind of see that with mercury-resistant plasmids because although they have a fitness cost when there's no mercury around, we can add mercury to the environment and then these plasmids become beneficial. So we can see these costs of plasmids becoming outweighed by selection. And so we might sort of think of plasmids through this light as being like a toolkit or like I say, like a delivery van, right? The bacteria are sitting there, it needs some new kind of trait. The plasmid is a vehicle that delivers it to those individuals. But I'm gonna try and convince you that this isn't the case. And the reason why is that we know that plasmid dynamics vary even when plasmids are beneficial. So this is an experiment done by Anastasia Kattara that I helped with a little bit but it's really Anastasia's experiment. What she did was she took the same mercury-resistant plasmid, put it into these five different species of pseudomonas and then basically asked, do these species maintain the plasmid? And what she found was when there's no selection for the plasmid, we see this really clear pattern where two of the species maintain the plasmid really well, right? So it's quite, you know, quite remarkable that they're actually maintaining the plasmid even though we know it's costly. And then these other three species lose it quite quickly, right? But this is when there's no selection for the plasmid. What happens when the plasmid is beneficial? Well, confusingly, it's kind of the same thing. So even when we've got selection for the plasmid in the short term, perhaps, it means the plasmid has maintained but in the long term, the plasmid gets lost. And this is because of transposition of the mercury-resistant genes from the plasmid onto the chromosome. So this tells us that selection isn't the whole story. We can't think of plasmids just as these delivery vans. Plasmids aren't just assisting because they're useful. Okay, so just as a more nuanced interaction. Okay, next we wanted to understand what happens in the context of a multi-species microbiome, so this was just for single species. What happens when we've got plasmid-favorable species occurring alongside unfavorable ones? So this is like 25,000 PCR reactions in one figure. So what I'm showing here is six independent replicas where we're looking at single species cultures of pseudomonas fluorescence, which if you recall it from the previous slide is one that likes to maintain plasmids and pseudomonas putida, which is one that tends not to maintain the plasmid. This is on the left panels, we're seeing what happens when there's no mercury around right panels with mercury, time on the x-axis, pattern of frequency on the y-axis. What we can see is again, pseudomonas fluorescence maintains this plasmid regardless of whether we've got selection or not. Pseudomonas putida gets rid of it regardless of whether we have selection or not. But when we co-culture these two species together in the context of a microbiome, we actually see the maintenance of this plasmid in pseudomonas putida, so in the unfavorable host. And we see the same pattern happening in the presence of mercury as well. So what this tells us is that infectious transfer can maintain a plasmid. So this is a plasmid being able to be maintained in a sort of source species or reservoir species, and then that's kind of spilling out into other members of the community. And this suggests the importance of understanding plasmids in their own right as these entities which can infect and move and Darwinian entities, they can be successful by infection. It doesn't necessarily need to be the case of a beneficial. And actually this property of infectious plasmid transfer has really interesting consequences for the non-plasmid genes. Because when we sequence clones from this experiment, we actually saw a lot of unexpected gene mobilization. Now this is a bit of a complicated figure and I'm really keen not to over go my time. So really what I want you to understand here is that the semicircle on the left shows the genomes from pseudomonas putida, the semicircles on the right show genomes from pseudomonas fluorescens. And these little triangles and the arrows between them show where genes have moved over the course of the experiment. And what we can see are chromosomal genes in both of these species. So it's chromosomal genes on the top half of the diagram. Moving onto these plasmids, moving between species and then in some cases jumping off the plasmid on the other side when they've sort of reached their destination. And these are quite big regions of the genome. So for example, this transposon Tn6291 is about 22 kb. They've got a large number of what we call accessory genes in them as well. Basically what we can see here are these transposons with the potential to carry diverse functions being picked up and carried around by the plasmids. So this is my next vehicle metaphor. It's a bit like a bus. But then of course a bus doesn't quite work because in this metaphor, the plasmid's actually moving around, getting the genes that it wants and assembling them potentially and helping to spread antimicrobial resistance that way. One thing we have in Liverpool are these kind of party buses. And so this is where people get in there for the drive-out and pick up all of their friends and they have a great time all together. And perhaps this is another kind of vehicle metaphor for trying to understand plasmids as these entities, which they've got their agency and they're moving around picking up the genes that they need in order to increase their own fitness and endow that on their host as they're moving around. But again, I don't think this is the whole picture. I think we need another metaphor on top of this. And this comes from this phenomenon that the plasmids are costly, right? I showed you some of this data before. We've heard about this again today in some of the talks earlier. And there are lots of different mechanisms by which plasmids might be costly, right? Dave Valtras wrote about this really clearly almost 10 years ago now and an updated paper by Alvaro and Craig more recently focusing on plasmids. This phenomenon of like, what causes plasmid fitness costs? And we can have lots of hypotheses about how this might happen. The fact that there's more DNA, there's more gene expression, there are various toxic genes, there are various potential negative interactions. But what we do know about plasmid fitness costs is that they can be resolved, right? And again, we've seen a lot of this, right? Alvaro and Craig, Ben Kerr and Hannah Jort. We've seen this in E. coli as well and Pseudomonas and with big plasmids and with small plasmids and lots of different kinds of genes involved. And so these compensatory mutations can reduce plasmid fitness costs. And we see that happening in our system as well. So evolution experiments have shown three key targets which seem like they're sort of evolving in parallel. So what I mean is when we do an evolution experiment, plasmid-containing treatments tend to have these mutations, plasmid-free treatments don't. So the key targets which tend to be hit in our particular bacterium with a plasmid that we've worked on are the GAC-AS system, which is a global regulator of secondary metabolism. I'm not gonna focus on as much today. There's a great paper by Ellie, you can read if you wanna know more about that. We also have this gene of unknown structure and function which was repeatedly mutated in different plasmid-evolution experiments. We call it PFLU-4242. As you probably can guess by the name, we don't really know what this gene is doing yet. And then there's a gene which is present on the plasmid which also was repeatedly mutated in about 25% of our evolved populations. And this is a putative lambda repressage. So this was published in Plost Biology. You can read it more. I'm going through all this too quickly. I'm sorry, I'm trying to keep the time. So we can do experiments which can show that disruption to these genes ameliorates plasmid-fitness cost. So if we knock out GAC-AS or we knock out PFLU-4242, there's no plasmids around, we have no fitness cost. If there is a plasmid around in the world type background with PQBR-57, you've got 70% fitness cost. But then mutating either of these genes resolves this fitness cost. And this is even more extreme with PQBR-103. It's like a 50% fitness cost with the plasmid. And then we just delete one gene, right? We delete GAC-AS or we delete PFLU-4242 and now fitness cost is resolved. And so this also tells us that the same mutations can ameliorate different plasmids as well. We can also find the plasmid mutation ameliorating plasmids fitness costs. So this is for context, right? This is a 30, sorry, a 300KB plasmid, right? 300KB plasmid, a single base pair mutation. And so what we can see here is that with the ancestral plasmid, we again have the 70% fitness cost, a single base pair mutation effectively resolves that fitness cost. So compensated evolution is important and the fact that compensated evolution could happen so effectively shows us that the main costs of plasmid doesn't come from having more DNA because they're still maintaining these plasmids. They just don't have the same cost. Doesn't come from gene expression. And I've sort of skipped over a little bit with that. I've got RNA-seq data which shows us don't have time for it in this talk. And it also suggested it's not toxic products either because what we can see again is that conjugation rates remain the same in compensated strains and it's very specific genes which are getting affected. And so this leaves us with the fact that it's potential interactions either with other mobile genetic elements or with chromosomes in the cell. So to understand what was kind of going on here we did an RNA-seq experiment. So in this diagram, each column is a different gene which has been differentially expressed following acquisition of the wild type strain with either of these plasmids. So this is telling us what happens when these bacteria gain the plasmid. And we can see a large number of genes which are up-regulated following acquisition of both plasmids. And it turns out that most of these are involved or associated with the SOS response and with signal transduction. And if you go into them in a bit more detail we can see that almost all of them are associated with predicted chromosomal mobile genetic elements. Now I've done a bit more work on this. I don't have time to talk about it because I want to talk about what happens with compensation. So what happens with compensation? So we see when the bacteria gain these plasmids we've got this massive gene differential expression. If we get these mutations, the mutation by itself actually has no effect on gene expression at all. And this mutation in the context of the other plasmid brings down almost all of those differentially expressed genes. And the same is true for that single plasmid mutation. All of these differentially expressed genes which were causing mischief in the wild type strain have come back down to being non-significant here. Okay, but there are some genes which remain up-regulated and what are these? Turns out that these actually encode the transposates for that chromosomal transposon that we saw jumping between the two different species a few slides before. So what this is telling us is that this transposon on the chromosome knows when the plasmid is here even when the plasmid isn't causing all of this disruption. And this really makes sense from the perspective of the transposon because a transposon obviously can't move between cells by itself. It needs other mobile genetic elements like plasmids. And a plasmid arriving is really great opportunity for this mobile genetic element to start transposing, moving onto the plasmid and then giving it the ability to use that plasmid as a vehicle for horizontal gene transfer to move into new backgrounds. So this is the plasmid as a taxi cab and this signal that the mobile genetic element is getting that causes it to up-regulate its transposon is almost like the ping on the phone which tells it this is cab has arrived. Okay, so these are my summaries. Selection for pathway decoded trait is not enough. Infectious transfer is important. Compensation is important. And we really need to start thinking about mobile genetic elements in their own right not as just these traits or not as just like extra features in a microbiome or in a pendulum. They're actually entities with their own fitness interests and have all as such. Okay, so this is my research group as it stands. I'm in Liverpool. We've got a bunch of projects which are kind of ongoing and some really great students post-docs working on this at the moment will probably be recruiting more people soon. So if you're interested in any of this kind of stuff please get in touch. These are my kind of key collaborators in Manchester, Sheffield, York and Liverpool who've been involved in this work or involved in the associated discussions. Thanks everybody for listening. This is my kind of concluding slide. Thank you very much, James. Sorry, that was a little bit... I need for one more few questions. Well, I'm going to read out. Jan has got a question in the chat. Jamie, if you find that the same compensatory mutations very repeatedly and quite drastically ameliorate fitness cost, why do you think the plasmids maintain these genes in the first place? Why aren't they selected against? Yeah, that's a great question. And of course there are compensatory mutations that occur on the plasmids and compensatory mutations which occur on the chromosome. And it's going to be because there's come some kind of pleiotropic effect, right? The fact that we see these genes. So for example, it may be that the chromosomal genes are beneficial under certain kinds of environments and so breaking them has these negative effects in those certain environments. And it looks like it may be the same with the plasmids as well. So it's a bit long-winded explanation. I can go into it in a bit more detail that we've got a chance in the discussion. But essentially the plasmid mutation causes an increased fitness cost in the presence of other plasmids. That's what we suspect is going on. So compensation for this particular plasmid is good and accessible when it's by itself but when it's in a multi-plasmid context, it's got a problem. And so that could be why this gene exists in deep natural isolates. Goni has a question as well. Yeah. I really enjoyed your talk a lot, it was fantastic. I just want to learn from you that you have thoughts about how to determine fitness costs under the conditions, for instance, of the sugar beet rhizosphere because the fascinating thing of these plasmids is that they originate from the rhizosphere of sugar beets and they appeared at a certain time in the year and as a response to triggers which were likely the exudates. So do you have any thoughts of how to determine fitness costs under conditions where we do not have this massive growth? So I've done, I measure fitness in soil and we can still see fitness costs in soil. So we can take this bacteria, put them in soil. Again, we see the fitness costs. There are experiments ongoing as far as I'm aware to actually test it in the sugar beet rhizosphere as well. I think Mike might be doing some, might have some results on that, I don't want to give anything away. But yeah, that's definitely something we're interested in trying to understand. Yeah. But do you see growth in the rhizosphere? Because I think this is a tricky thing. I typically, when we introduced plasmid-free and plasmid-containing strain into the rhizosphere and followed this, we are not seeing growth. We are seeing this slight decline and we determined then the proportion of plasmid-carrying and plasmid-free populations. This was for us in NATO-Bakta. But I always got the criticism, you do not have growth, but this is a complicated point. I mean, microbes are growing in their natural habitats, right? So there is growth in those kinds of environments. I mean, I actually think that some of these plasmid-fitness costs in our case actually comes from cells being killed rather than a burden on replication. So we would probably see that happening. Again, I skipped over the details, but if you read the paper, we actually show that some of the genes which are up-regulated actually have a cost by themselves because they're these hole in genes which make holes in the cell membrane. And so it could be that, I mean, I wouldn't be surprised if we actually saw a fitness cost even without any growth because just from cells being killed by inappropriate activation of the SOS response. Thanks. Okay, we have to move ahead.