 Good morning. So I hope you still have some brain power left. That was an intense morning session. Very informative. So let me go with the first one. And I thought I will tackle first off the gaps and challenges that we're facing. And this is something that you may not know. I have no sound. Some sound. Okay, so this is a gap that I think all of you will recognize. State of NIH Fundings. It's pretty dry. I will say that's a big gap in challenges in my research. This one is another gap that I found. Don't know if you know what it represents. A bunch of bloodthirsty animals, also known as study section. I will be your proposal right there. Very, very daring. And if you do, we saw the first day there were a lot of animation on microbial composition 3D with Rob and Jacques. So I thought I will put a 3D animation of these gaps. It will go like that. So it will be you sending out your microbiome proposal. It's one of the... So big questions. Major question in the field and you think that it's going to be funded. You already have your team recruiting people. You smell the money. And all is good in your life. Then you got on e-comments and you saw the score. It's not very good. Actually, you have to think about your career because you put so much faith in this proposal that now you are down on your knees and thinking about plan B. But you're a scientist for God's sake. So you come back, call the PO and intense the team, change the aims and come back. And submit again and get the money because we need to move the field forward, right? So, keep the faith. Indeed. So I hope I'm not in trouble with the NIH staff yet. I'm a reviewer, right? So what I want to do here is to tackle a little bit generally the cancer and the microbiome interaction but specifically going after colorectal cancer. This is a cartoon that will be part of a review, invited review for Nature Cancer. But the idea is that different organ are linked to different form of cancer, microbiome driven. Sometimes it's like the liver microbial products. Sometimes it's an isolate in a different organ affecting a distal site like pancreas and oral pathogen. And other time will be like in the colon where you have a microbiome next to the organ that will be affected. And I will focus on that if you don't mind. Colorectal cancer is multifactorial diseases, disease as host genetics, diet and inflammation. And we learned for the past two days that inflammation is the powerful environmental factor shaping microbial constitution, composition, activities. So it's tantalizing to link this to cancer. Obviously there is a link between inflammation and cancer but could you bring in the microbiome and then make a sequential effect on inflammation, cancer, microbiome and cancer? And we know I will skip fast on that. I mean IBD cohorts have been screened for microbiome composition. We know there's cluster between healthy, Crohn's disease, ulcerative colitis, two different forms of IBD. Diversity is affected. This is an old stone age paper that was using I think TRFLP. New data, so many, many groups, different cohorts have all confirmed that there is a dysbiosis, a change in the microbial composition. We know a little bit in the resolution which microbes could be associated with disease, especially the proteobacteria, gamma proteobacteria and introbacteria ECs for Crohn's disease and a reduction of clostridia, butyrate producer and also TRG regulator. So we have dysbiosis that may be functionally linked to IBD. We're still to work the detail of that but it's there. And if you put this in motion in term of colorectal cancer, the colitis aspect of colorectal cancer was mostly thought as a host dysregulation. We have inflammatory mediators coming in from the patellium immune cells and it will hit somehow this cocktail of inflammatory mediator, will hit the patellium, cause dysplastic response, DNA damage, and progress to adenocarcinoma and many transcription factor in cytokine. Cytokines have been associated to that and causally found to be responsible for the disease. We don't have this figured out, but one thing that we know is this organ is hosting an abundant microbiome. So what is the microbiome doing? And for many years we thought that at least in the field of colitis, it was just a fuel to set more inflammatory response. So a trigger perpetrator of inflammation and that's how you get to the cancer. But the question is, is that all the microbiome could do? And I will focus on bacterial microbiome here. And people have done, in the past two years, a lot of survey using different compartments. I'm not going to name all the bacteria, but there is an abundance, an increased expansion of some group, a decrease of other in the luminal compartment. And if you look at tissue compartment, mucosal samples in early stage of cancer and late, there's also dysbiosis. It's very hard to pinpoint an association and in a cause to that. So it's mostly, and all the papers don't have time to highlight everyone, but there is an active group from different parts of the world looking at that, different cohorts. So it's clear we have dysbiosis. What is the cause? Effect relationship. That is difficult to do in patients. We need to introduce animal model. And the question we were asking is what happened to the microbiome during the onset of colitis? The progression is a microbiome entuned with the inflammatory response of the host and what does it do in terms of cancer development? We use a simple model. You heard about it, so germ-free mice. The reason why you have germ-free mice is because you avoid the legacy and a familial transmission of microbes that sometimes drive cage effect. Although we have a paper just showing that founder's effect is not the main cause of cage effect. But that's another ballgame. So you have wild-type mice, which is your control mice, or susceptible mice we use for colitis cancer, or AL-10 knockout, which is an immunosuppressive molecule, but no AL-10 knockout, this control or dysregulation of the immune response, and the phenotype is colitis. But they don't have colitis in the germ-free environment. If you move them into an SPF where they have environmental microbes coming in, then you have colitis. We use a compound that's called azoxymethane to initiate cancer, and then you survey at different time the microbiome in the stool, so two weeks will be a mild colitis, 12 weeks a strong colitis with high-grade dysplasia, and 20 weeks full colitis and invasive tumors. So you could sequence. And what we did here is to compare the microbiome of these mice, wild-type, AL-10 knockout. So this is two weeks cohort. Individual symbols is an individual mouse or the incorporated microbiome. And in the wild-type mice who will not develop colitis, after 12 and 20 weeks, the biome just assemble, and this is a college called Assembly of the Biome. There's no differences actually between the 12 and the 20 weeks. So they end up in a region of their microbiome is stable, and that's how they assemble. In the AL-10 knockout where you have here the colitis, and later on the colitis cancer, you could see that there is a clustering that is highly significant. So when cancer, when the colitis progress, it does shape the microbiome in the cancer, as well as another factor. If you do pie cross analysis on this community now that is in the colitis cancer, you can't read that, but the red symbols means that there are genes that are highly abundant. And most of them are in the type secretion system, one, two, three, five. And a secretion system is often associated with a more aggressive behavior from the microbiome. So you have a microbiome that is a little bit more prone to looking for trouble, if I may say. If you look at flagellin, which is another very important factor that microbes utilize to colonize, penetrate, and get close to the patellium. You need to be more mobile, and all these genes are highly abundant, higher in the colitis cancer. But if you go with the phylum level now, because you have generated a lot of data with the sequencing, where do you go from there? If you look here, this is a color code, so red means highly significant, and more abundant in the outer knockout than the wild type. You have two groups, two phylums that are highly, that change. So the proteobacteria and the vericromacrobia. So where do you go? We took a bias approach with the proteobacteria because we knew from the human sequencing that this group is also associated with a higher abundance in patient. If you go down to the family level and then compare over time, so two weeks, 12 and 20 weeks, this is the AL-10 and the wild type, you see always an increase, higher abundance, enterobacteriesies over time, although the niche diminish when you move to full colitis inflammation. That's an interesting observation, but yet you always have more of these microbial family in the animal that have colitis and cancer. This is Illumina sequencing. You can go down to species. This is with the RDP, it was a consensus 27, feeding the earciar and shigella. But we knew from the literature that at this point that people have identified E. coli, Adrenalin-invasive E. coli on IBD patient, on the mucosal, highly representative and also in colorectal cancer. So the E. coli was of great importance. This is the E. coli I'm talking about is not a he-hack-he-pick classical pathogenic microorganism. It's the one that you have in your bowel that somehow may acquire a more aggressive phenotype because of a change in the environment. And we did selected by us PCR analysis of 16 years looking at E. coli in these cohorts, and we found higher abundance of E. coli in the AL-10. We had the idea that the change with inflammation, the change of microbial composition, there is a group of bacteria that expand. There's many, many more bacteria, but we were interested in the E. coli because of the link with colorectal cancer in patient. So what do you do after that? Well, we took an E. coli that we had called NC-101 which is North Carolina 101 was found and isolated at North Carolina State University. It's non-pathogenic E. coli again. And then we'll go back to the germ-free AL-10 and then you mono-associate so you control the system. So what you do here, you introduce the E. coli NC-101 and an entrococcus ficalis are all human isolate that we knew also and use colitis in this system. And now if you change these microbial status in the mouse, what will happen to inflammation? Cancer. This is a astrological score of inflammation between the two mono-associated mice. E. ficalis, E. coli, very similar. This is tumor multiplicity and invasion and you could see that you have quite of a difference in the capacity to induce cancer. So the inflammation is quite similar but the cancer capacity is different. So there is again a link between the microbial constituent. Who do you have in there and how could they influence your cancer development? And we always thought that inflammation was the driving force and from there it seems that the microbes have something very important to say in influencing diseases. So what is that E. coli has that make it more susceptible or more aggressive in inducing cancer? And that goes back to looking and sequencing and having information in the genome of these bugs. We found that this bug has a PKS island which is a genotoxic island. 22 genes, 54 key bits, a massive amount of multi-asimatic system and literature have found that in this pathogenic E. coli it's an extraintestinal E. coli sp15 that when they introduced that in the mouse as an ileal loop they found the any damage. So we thought since our E. coli has this PKS it may be responsible. So what you do, you do genetic deletion of the PKS island in the E. coli and you compare this island isogenic E. coli and go back to the animal model and ask the question do I have cancer in the system? Does it impact inflammation? What we found is that in term of inflammatory capacity at different time point which is 12, 14 and 18 weeks the inflammatory capacity of the wild type of mutant E. coli was very similar. So remove this island using massive inflammation now what happened to cancer you remove the PKS and you have you lose the ability or the strength of cancer promotion. That's the invasion multiplicity. So the E. coli was quite the PKS of E. coli was very important in driving the cancer susceptibility but not changing the inflammatory host response. This is an E. coli that was a murine isolate well is that gene has any relevance to human diseases so we went to a cohort I told you that E. coli was isolated from many many different population in UK and in US so we linked with a group in UK they had access to a bank of E. coli and we screened for the PKS island what we found is that this is clinical isolate now human we found that the prevalence of E. coli having this PKS was much more higher roughly 3 times higher in the population colorectal cancer and then the control population and that was confirmed by another group in France using a different cohort looking at different cytotoxic virulence factor but the PKS was also found to be 20% in the control roughly and 55% in the colorectal cancer similar to us so again it's an association the disease it doesn't mean that the PKS was causative of cancer but it shows that the E. coli has the capacity the E. coli that caused cancer in our system the gene is present in the human population so just to go back I said that the inflammation is a powerful force shaping the microbial composition but is it all it could do so what we're very interested to know is the inflammation if you have a group of bacteria that expand during inflammation if you take them and introduce them back in a mouse that is susceptible to cancer will it cause cancer by itself and what is the role of inflammation so what we did here again is germ free mice the Alten and also we use an Alten RAC2 these mice lack maturity and B cell so they can't go into chronic inflammation so you remove inflammation from that mouse basically and come back with the NC101 and what we found is that the inflammation the bugs induce inflammation in the Alten but in the Alten RAC2 there's no inflammation so now you remove the capacity to induce inflammation but you still have the E. coli with the PKS island will that induce cancer and it doesn't or it has a strong reduction so now I don't want to confuse you but it puts a lot of intricate relationship between inflammation, composition and activity of bacteria here we have no host inflammation and the bugs that have functional PKS can't induce the cancer so you need something by inflammation to get the bug going and we were interested in knowing what that is so we did RNA microbial RNA sequence on these cohorts see what is the relationship between inflammation and their transcriptomic response it's a little bit complicated but let's focus on this so the circles are the Alten alone so the Alten develop inflammation and cancer the mouse was mono-associated and they harvested stools from different time 2 weeks, 12 weeks and 20 weeks and performed the RNA seek so what you have is a clustering from a different time so you have the 2 weeks transcriptome of these different animals and then at 12 weeks and 20 weeks so the transcriptome change with the inflammatory response that will be your conclusion here with the rag now so you look at the transcriptome of the bugs that was in the Alten rag who doesn't develop inflammation and now it's moving very close to the inflammatory so the transcriptomic response of this E. coli was massively driven by adaptation environment adaptation and the host but we found a selected set of genes that are highly regulated by inflammation so adaptation of the bacteria and the host drive a lot of transcriptomic response inflammation will come and target select genes that likely and we're working on these genes I have no time to list use these genes but some of these genes likely will influence the capacity to induce cancer or the fitness of the bacteria so just to sum up what I've been telling you so far you have the microbiome that is inducing a host response in a susceptible host that will be inflammatory compartment response or a lot of inflammation this inflammation by itself doesn't just affect the host goes after the luminal microbial compartment and does change the ability or the composition of the microbiome and also likely the activities and some of these activities could be linked to genotoxin the PKS I've told you about and that is the perfect storm when you have an increased likelihood to develop colorectal cancer so since I'm in the translational section of the meeting I thought the gaps and challenges could be that now how do we move forward from this we know there's this biosis and bacterial translocation when you have early onset of cancer or inflammation so could you modulate this microbiome composition pre and probiotics we heard about that bacterial therapy maybe the stone age fecal transplant or more targeted introduction of bacteria antibiotics could be although it's certainly not a long-term therapy genetically modified microbiota that you could introduce enzyme to help you out-compete activity of another one diet and cancer is highly link correlated could you change diet to impact the microbiome and decrease your cancer microbiome population we have pro-effective effector mechanism from these bacteria I've shown you the PKS could you target these genotoxin and also could you target prevent inflammation prevent the change of microbiome for real now for the needs we need to move the omics down to the target mechanism you don't want to be the mouse that was just a movie it's a happy ending but a lot of time there's no happy ending you do so you need to write a grant that will have a link between the data that generated by a sequencer and get mechanism to target related applications in my case I need access to clinical isolate all these bugs that we or family or group of bacteria that we identify from these biosis and sequencing need to be put in the bank we need to look at them in terms of sequencing sequence these bacteria from different cohorts could we found unique to a population and start asking question about the function in terms of cancer development we need to get better in the microbiology and we have the high technology that we're using but the old technology chemostat turbidostat multistage chemostat to study behavior of these bacteria if you have a clinical isolate which factor could influence the expansion behavior of these bacteria any enzymatic activity that could be controlled by diet, introduction or changing the setup of the bioreactor system so we need that and this look a little bit funky colonization protocol for germ free I mean people say you just introduce the bacteria and you're good to go that's not true I mean when we introduce complex community they are not all of them there you may lose a lot of them because there's no sequential pioneer bacteria coming in and setting the stage for the next wave of bacteria and I give you an example we had a clinical isolate provided by us from Emma Allen Verco in Guelph University with the fusobacterium nucleotom that people found associated with colorectal cancer we took that bug introduced it to the Altenac out and that bug was not able to colonize and stay so it needs another bacteria to come in so these when you start looking at consortium because cancer is not caused by one unlikely by one microorganism it's going to be an association interaction we need to duplicate that in the animal not going to be easy if we don't know how they interact with each other you can't put them all in a mix and hope for the best and with that my god I'm at 6 seconds of the end this is my lab I would like to recognize all the people involved in this work UNC Chapel Hill the UNC no biotic facility head by Barfor Sardar Instrumental UNC Charlotte for bioinformatics Cornell and University of Liverpool for the clinical isolate and I thank you for your attention okay we have time for a few questions crickets I think one question over here so Christian one thing we were talking about actually a journal club with that your cool paper is is the effect of the of the pks bearing organisms is it just is it an incremental genotoxic effect because you're getting things like AOM which is certainly a big genotoxic hit and a lot of environmental things will do that so when one focuses on the microbiome here are you seeing that there's a unique role of the organisms or is it sort of like recreating the natural setting where the hits are going to be sub carcinogenic and so any additional hits adds to the threshold right I mean that's a great question we like to see the progress it's hard to monitor the pks activity if you were to know that it's inducing a sequence of mutation and then you could start sequencing the host insert progression of cancer in your cohort over time and see that you know the pks is getting this hit here and there and then over time it accumulates to this you could try to link it to the activity of the pks it does one massive mutation or multiple mutation and you need the pks activity for that long to see that it's a difficult question to answer at this point we've been doing sequencing on tumors we were not able to find the usual hit suspect you know the apc wrasse and all these susceptible spot that could be mutated we didn't find them so we're not sure what is the pks endpoint in term of mutation we know it does DNA damage in vitro and we saw it in the mouse in vivo but we're not quite sure how to follow the details of it over time but I suspect you may have an activation a setup of pks activity and after that something else is coming could be inflammation of the host preventing repaired mechanism enhancing epigenetic changes and then the train left the station and you go into more aggressive cancer what we say for sure is that the pks does induce more invasive tumors so what they do to the pitalium has a long term consequences on the aggressiveness of cancer any other questions if not we'll move on thank you Christian ok so the final speaker in the session is Julian Davies from the University of British Columbia in Canada and he's going to talk about harvesting the molecular wealth of the microbiome