 Barba is the most promising scientist according to the Plasmid Conference in I don't know 2018, was it? Or 2017? I don't know. I didn't know I had that title, but it's good. That's a good title. Yeah, so I will just start sharing my experience a week. Yeah, no? Here we go. Here. Okay. So can you see the presentation, I guess? Yes. Okay, okay. So I'm going to talk today about, yeah, about two things, basically, the spread and evolution of Plasmid-mediated antibiotic resistance in hospitalized patients, in the gut microbiota of hospitalized patients. Okay, we have already heard quite a lot about how Plasmid-mediated resistance evolves. Briefly, there's transfer of conjugative Plasmids that distribute antimicrobial mechanisms across different bacteria, but acquiring a Plasmid may be really useful if there is antibiotic present, but it may be costly if there is no antibiotic because Plasmids tend to produce physiological alterations in the bacterial host that may lead to a fitness cost. And then this is not a static picture, but over time, Plasmid and bacteria can evolve and the cost can become compensated. The Plasmid copy number can be modulated. Different things can happen that fine-tune the relationship between bacteria and the Plasmids, right? So we are trying to understand the evolution of Plasmid-mediated resistance using carbapenemase-producing and terrobacteria. So carbapenemase-producing and terrobacteria are a group of bacteria that, such as epsilon pneumonia or E. coli that carry, are able to resist these carbapenem group of antibiotics that are last resort antibiotics that are usually used only in multi-drug resistant infections in hospitals. So these group of bacteria represent an important threat to patients, right? And what is important to us is that in general, these carbapenemases are the Betalactamase-type enzymes that degrade carbapenemase are usually included in conjugated Plasmids. So for the last six years, I guess we have been studying quite a lot specific carbapenemase, that is Oxa-48, which was described in the early 2000s in Turkey, but it's now distributed worldwide. And it's really important, I mean, it's really a good system for us because this carbapenemase is almost every single time encoded in an almost identical ink-L Plasmid, that is P-Oxa-48, and the P-Oxa-48 like Plasmids, but they're really similar. So this beauty here is the P-Oxa-48 Plasmid, which is conjugated, it conjugates quite a lot, all these conjugated genes, it carries different accessory genes, but well, it carries the carbapenemase gene right here. So an important thing about this Plasmid is that it associates with high-risk clones of mainly clefcyla pneumonia belonging to the sequence types 11, 15, 101, etc. So we wanted to understand or to study the evolution of these P-Oxa-48 medicated antibiotic resistance in hospitals, and to do that, I have been collaborating quite a lot with people, especially with Rafael Cantón, that is the head of the microbiology service in the Ramonica Hall Hospital in Madrid. So basically, Rafael, before I started working in the hospital, was performing a large project where they basically, they were sampling every single patient that was admitted to two different words, medical words, gastroenterology and uemology, and two different surgical words, neurosurgery and urology in the hospital. So every single patient that was admitted was sampled, looking for ESBLs and carbapenemase production in terrobacteria, just to find colonization in the gut microbiota of these patients. So over two years, more than 9,000 patients were sampled, and almost 3,000 swaps taken, and out of those samples, 250 P-Oxa-48 carbine terrobacteria were isolated from 105 patients. But the good thing about our noses is that once a patient was sampled positive, the patient will be sampled every single week after that until discharged from the hospital, and that's good because that will generate temporal series of P-Oxa-48 carbine terrobacteria that are really useful for us to study the evolution of the plasmid and in the gut microbiota of the patients. So I will talk about the spread and the evolution of the plasmid, then we'll start talking about how the plasmid spreads in the hospital, and this is work done or led by Ricardo León in the lab. So basically, we started studying the epidemiology of carbapenemase production in terrobacteria in the hospital. There are basically two layers of complexity that you may or you want to take into account. So basically, there's what we call the between-patient transfer and that basically is the transfer of the carbapenemase production in terrobacteria between different patients, and that can happen directly or indirectly through the hospital staff or surfaces and so on. But there is a second layer of complexity in the epidemiology of these bacteria in the hospital that is the within-patient transfer. Patients are usually frequently colonized by these carbapenemase producing in terrobacterian hospitals, and once the patients have been colonized, since the carbapenemases are encoded on conjugative plasmids, they can spread horizontally through conjugation, generating new carbapenemase producing in terrobacteria. So this is also an important part of the epidemiology of these bacteria. So okay, so we used the data from hernosis, and as I said, we found 105 patients that were colonized by P-oxa-40F producing in terrobacteria over the study period, and that's where I represent here. Every single row represents a patient. The color segment indicates the length of the stay from admission to discharge of the patient in the hospital in the different words. And if there's a black outline in the color, it means that from that particular patient, more than one different P-oxa-48 carrying clones, meaning from different species, were isolated, which is kind of an indicator of potential events within patient plasmid transfer. Okay, but that's just an indicator. Okay, so what we did is we took the 250 isolates that were recovered from this collection, and they basically kind of look like this. So this is the number of isolates by the month of the study, and the different colors indicate the species to which the isolate belong. And as you can see, the most common species is Klebsil anemone, and that's not surprising, because as I said, Klebsil anemone is usually associated to this plasmid, but we also found quite a lot of E. coli and other enterobacteria species associated to the plasmid in the gut of these patients. Okay, so we took the 250 isolates and we sequenced their genomes, both by Illumina and also for a few of them using long-rig technology. And what we did is kind of basically use the approach that Alicia has just introduced of trying to use epidemiological models with the collaboration of Ben Cooper from the University of Oxford. So basically what we did is we feed all the genomic information from the 250 isolates and the epidemiological information from the 105 patients to try to reconstruct the potential routes or the potential events of between patient transfer of P. oxa-48 carrying clones. Well, basically this is kind of the same figure as before, but now what, like on top of the patients or the length of the state of the patients, there are some arrows that basically indicate these potential events of between patient plasmid transfer. The width of the arrow indicates the probability assigned by the model to the transfer event and the color of the arrows indicate the clone or the sequence type involved in that transfer event. Okay, I mean, this has already been published, so if I won't go into much detail, if you're interested, you can check it out. And basically what we found is that we found quite frequent transfer events in patients. ST-11 of the genomic money was involved in transfer events in all the words. Neurosurgery work was the one with a higher or the highest frequency of transfers and the highest probability also. And although most of the transfer events were driven by Klebsiella, we also found that an E. coli spreading around or transferring around patients with this class. Some patients may act as super spreaders. There are really interesting epidemiological data that we got out from this analysis. Okay, so the plasmid transfer, so the plasmid current bacteria can transfer between patients. Okay, but we were also or we were really interested in the transfer within the patients. Okay, so basically the results kind of suggest that since there was kind of a high frequency of patients that were colonized by two different species of P oxa-48 carrying bacteria, this may be an indication of within patient transfer. But there's also another possibility that is that this is just caused by independent colonization events, by independent bacteria carrying the plasmid. So to kind of tell these two possibilities apart, what we did is we went and analyzed the genetic sequence of the plasmid of P oxa-48 in all the isolates. So basically what we wanted to do is to try to find the specific genetic signatures in the plasmid that will help us confirm if those were cases of within patient plasmid transfer. Okay, so to do that, what we did is we took the sequence of the plasmid and we construct, we compare the core plasmid sequence, which is good because in most of the strains, the core is really well conserved in terms of the fragment of the plasmid. So 90% is conserved in most of the strains. So in these strains, we compare the sequence of this core and basically that is what I represent here. So the outer ring indicates the genus of the isolate, the second circle indicated the name of the isolate from which the plasmid is recovered. And then the remaining circles indicate the presence or absence of certain SNPs that we found in the core ring. The first thing that I have to highlight here is that 80% of the plasmids were completely identical in the core region. So you can really tell, I mean you can really study any transmission events there because they are exactly the same. But for some of them we found mutations and what we did is we tried to find cases where these traceable plasmid variants that carry specific SNPs in the core region were present in different bacterial clumps. And why is that? Because we thought that if the isolation of different bacterial species carrying the same traceable plasmid variant happened in the same patient, in a single patient, that is a very strong indication of within patient plasmid transfer. So we found four cases of these potential traceable plasmid transfer events in the guard of patients. And this is, well again, this has been recently published. I won't take much longer in this, but basically what we found is for every single case where a mutation could tell us if there is a within patient plasmid transfer we found these plasmid transfer events. In patients where for example this patient here carries a specific plasmid variant and in the six different isolates belonging to four different species that were recovered from the guard of this patient, the exact same plasmid variant was present. And that plasmid variant was not present in any other patient in the stat. So this basically means that the plasmid is being transferred between these clumps in the guard microbiota of the patient. Another really interesting thing that we learned is that this high conjugation in the guard of patients also led to long-term p-oxa-48 guard carries. Because take for example these patients here was colonized by an ST-11 of the Psella. A few weeks later that particular ST was gone, but the same plasmid variant was present in a different E. coli in the guard microbiota. And then the patient was discharged and admitted again more than a year later. And the same specific plasmid variant was present now in a third different clone and in a different E. coli clone. But given that this plasmid variant has this specific traceable mutation, this means that this patient has been colonized over more than a year from with this particular plasmid. Which is important because now it means that these carbapenema is encoded in plasmids may be moving towards the community and colonizing other people outside the hospital. Okay, so that's the first part of the talk. But now I want to present some new data in which we are kind of studying not only the spread of the plasmid, but also the evolution of the plasmid in the guard microbiota of patients. So this is work led by Javier in the lab. And basically what Javier did is he kind of took the sequences of p-oxa-48 plasmid and did a more in depth analysis looking not only at the core region, but at the complete plasmid sequence and comparing all plasmids with each other. And he, with this approximation, he basically identified more plasmid variants up to 35 p-oxa-48 plasmid variants isolated from these different patients in the hospital. And using these variants we wanted to understand if we could somehow trace the evolution of the plasmid in the guard of patients. So the first thing that we did is to do that is we kind of selected 14 of these variants that had kind of representative mutations of the entire collection that I represent here with these letters here from A to M. And each one as you can see carries either a deletion, an insertion, a different type of mutation, SNPs, synonyms, or not synonymous mutations, intergenic mutations, etc. So what we thought is like, okay, if these mutations are associated with the evolution of antibiotic resistance in the hospital, one good aspect that they may have an impact in plasmid associated traits that may be relevant for the evolution of resistance. So what we did is we took the 14 plasmid variants and the first series of experiments we just said, okay, so let's put them in an isogenic bacterial background and let's see what they do. Let's look at the plasmid associated traits. So we took the 14 variants and put them in an E. coli strain and we measure for this collection. Well, we sequenced them all just to be sure that they're completely isogenic apart from the presence of the different plasmid variants. And basically what we do, we measure fitness using growth curves and competition experiments, antibiotic resistant levels, conjugation rate and plasmid copy number to see if we could find any differences among them. So the first thing that I present here is the fitness effects here. I'm presenting the relative area under the growth cure relative to the plasmid free strain. But we have also done competition experiments and these two basically correlate perfectly. So it's kind of the same data. So I present here also the plasmid free is one and then all the different plasmid carrying the one that is highlighted in gray here is this plasmid variant C. And that's because that's the most common plasmid variant in the hospital. 60% of the isolates carry this particular plasmid variant. And as you can see, the plasmid tend to be quite costly in J53 in this equalized strain. But as you can also see, there are quite a lot of differences. So the most common plasmid variant is costly and there are other variants that are at least as high cost as this one. But there are a couple of variants that are almost cost free in J53. One of these, the first one carries a small deletion that inactivates the antimicrobial resistant gene, the OXA48 gene. And you can see that here. So this is the earth apenem resistant level of the different plasmid variant carrying clones. And as you can see, so the plasmid confer resistant to earth apenem. But in this particular clone here, it does not because the gene is deleted. And that's associated with an increase in fitness in J53. The second variant that also seems to be ameliorating the cost. So also carries a deletion. And this deletion is the large one. And it's going to delete some of the conjugation associated genes. And as you can see, this is the conjugation rates for its different variant. And this one is not conjugated at all, because it doesn't carry the genes anymore. But as you can also see, there are some differences in conjugation rates in the remaining clones. And finally, when we look at plasmid copy number again, so the plasmid copy number was about between 2.5 and 4 for most of the variants. But there was one variant with an increased plasmid copy number, which presented a mutation upstream the replication initiation gene, which could be responsible for this. And that one produced kind of a large cost and an enlarge or a high resistant level. So the thing here, or the take home message, is like, okay, different variants produce different effects, and they may be associated with the evolution of resistance. And actually, the most striking effect that we kind of observe is that there's a quite an important trade off between fitness and resistance for P oxa 48 carrying a 53 clones. So basically, this is the resistant level measure in I see 90 of her top NM. And this is the relative fitness, well, or the relative area under the growth cure, measuring fitness. And as you can see, there is a trade off between bacteria carrying these different plasmid variant being presented in a higher level of resistance, but also higher fitness cost in the absence of antibiotics. And also back to where the antibiotic resistant level is decreased or abolished and present a high fitness, which kind of indicate that probably the expression of the resistant mechanism is driving the cost in day 53, at least, but can also be really important for the evolution of P oxa 48 mediated resistance in vivo. If this translates to translates or stands in the in vivo in the wild type clones. So that's what we're doing what we're doing now. So we're trying to see if what we have learned from this E coli host background applies for the wild type bacteria clones that carry the plasmid in the patient. So we're looking for intra patient plasmid evolution. And to do that, what we are doing is looking for patients where we can find more with more than one p oxa 48 isolate has been recovered over time. Sorry. Also that the same clone is present in the gap of the patients over time. And we can look at that just looking at the genomic sequences, but also that we're this in the same clone, we find different plasmid variants over time. So meaning a clone that seemed to be colonizing the patient over time, the plasmid variant changes in that clone, a percent of different mutations that may be involved in the evolution of of plasmid mediated resistance. So we found three cases where we're three patients with these characteristics that I mean, I'm not going to detail because I don't have enough time. But basically, we have patients with these conditions. And we also for this patient, we have the antibiotic treatments over time in the hospital. So we can link this potential evolution with with the antibiotic pressure. And what is difficult to do this is the technical part, because when you work with wild type strains, it's not as working with with the lab strains, where it's really easy to genetically manipulate. Right. So what we're doing is we take all these wild type strains and first we remove the plasmid, the P oxa 48 plasmid from them. And we use a CRISPR-Cas based system that that the big car from the Pasteur Institute is helped us to develop. So we can cure the the plasmid from the clone. And once we have cured the plasmid, what we do is we sequence the genome just to make sure that no mutation, no significant mutations have accumulated during the process. And then what we what we can do is to put back the plasmid, the plasmid that that particular clone was carrying and the alternative plasmid that was present in the same clone at different time points. Okay. So then we can compare the effect of the plasmid in an isogenic wild type background, which is kind of the of the key part of the experiment. And then we measure resistance, plasmid cost, by growth curve and competitions and plasmid copy number and look at that. And basically that's what we're doing now. And we have really exciting results. They're not ready to be presented yet, but but we are almost there. But hopefully they will be they will be able to to have them soon. And basically, that's everything I wanted to to present today. And I want to thank all the people in the lab, the different people involved in the different parts of the project that I already mentioned during the talk, and well, the collaborators and funding bodies. And if you have any questions, I'll be happy to answer. Wow, Alvaro, this is awesome. And it gets every time you present it gets better. We're we're quite we're really excited about this part. We've been working with the system for a long time. And it is true that it's I mean, we are quite excited. Yes. So are there any questions? I don't see anything in the chat. Is there there is Cornelia with the hand? Cornelia? I don't know. Very nice to hear your story. And I have a question concerning, do you have a hypothesis from where in the first place these plasmids came from to the gut? Because I have an idea. I mean, we know, we kind of know that our hospital is colonized by by these enterobacteria, especially Klebsiella pneumonia is really good at surviving in the hospital, in the environment, right? So we know that the plasmid is kind of colonized by by P oxa 48 carrying Klebsiella. So what it seems to happen is that when patients are admitted to the hospital, because I mean, and we can see that with our data, because since patients are sample over time, they were sampled at admission and then another time. And then if they were positive, they they were sampled again and again. So a lot of a lot of the times that the first sample was negative. But then the next sample was positive, meaning that we think that most of the colonizations occurred in the hospital. So yeah, because you first said that they were sampling the people at admission, and those that were positive were followed over time. So that's why I thought, well, maybe this would be quite because many of these isolates that you listed that carry the plasmid are typical plant associated bacteria. I guess you haven't heard my talk, but I think I did. I did. But I think that I believe that there might be really an overlooked link from the bacteria associated to produce because they are arriving in the gut well protected from the vex layer. And at least in our collection, we have a few that carry oxa 48. That's why I thought that that's that's interesting. I should now I mean, Teresa from the from from the hospital is doing a huge effort in sampling, not only patients, but everything in the hospital, ICU's especially, the kind of the ecology of the sinks, especially on the surfaces, but the ecology of active microbial system plasmids there is kind of mind blowing. So so yeah, but what you are, I mean, it's interesting and I need maybe this would be mainly interesting for those patients who have the plasmid already at admission. And I have to double check because it might be that we have oxa 48 also in N, for instance. Okay, okay. Yeah, I have to double check this, but that's why I was very curious. Yeah, because in our in our case in our study, every single, I mean, it was a perfect correlation between oxa 40 almost perfect correlation between oxa 48 and the incal plasmid oxa 48. Yeah, but basically it's in literature, you always find it in incal, or I mean, in as if in E. coli as 38 it is inserted in the chromosome, but I never seen. But we just recently sequenced sending n plasmid and if I'm not mistaken, it has a oxa 48. So I double check it and I will contact you. Yeah, that's interesting. Please. Yeah, we're p oxa 48 aficionados. Yeah. Okay. Thanks for was really nice talk. Thank you, Connie.