 So, before we start, I just remind you what Matteo said earlier today. If you're interested to visit ICGB, I mean, the institute, we're going to have a, you know, we're going to have a 35, 40-minute presentation where I can highlight how to apply for a grant, the fellowships program, our collaboration with the industrial sector. And then we'll have some, we'll have a tool where some labs will briefly explain what they are doing. So if you're interested to come, there's still some places available. Just outside here on the left on the board, sign your name please. Thank you. Okay, so we start with the last two talks of the day and I'm very happy to introduce Jonas Kremer from UCSD from San Diego. He'll talk to us about microbial ecology in the human large intestine on the water flow and the growth dynamics of primary fermenters. Okay. Thank you very much. Thank you for the introduction and thank you for the invitation and I hope to explain you at the end what I really mean by this title. And since I have 40 minutes, I thought let's introduce some of the basics of microbial fermentation in the human large intestine first. But before doing that, I first want to acknowledge the people I work with on this project, especially Teri Waar, Alex Reusman, Markus Arnoldini, and Igor Segota. They helped a lot with experiments and with discussion. Louder. To screen. Hello. Yeah. All right. Fine. Should I try this? Is it better? Okay. Okay. So let's start with some basics about bacteria in our gut and so I put here some numbers, different parts of the gut, how many bacteria do we have there and you see that most of the bacteria are actually in the large intestine and their bacterial densities are really high, 10 to the 10. So bacterial densities are really only large in the large intestine, but their densities are really large, 10 to the 10 cells per gram or even 10 to the 11 cells per gram. And to put that number in perspective, that means that up to 70% of fecal dry mass is bacterial mass. So we produce a lot of these bacteria every day and the question is how important are they for our health. And there was great advancement in this research field, mainly driven by the great advancement in sequencing techniques. So what that allowed is to really investigate which bacterial species do we actually have in these samples. And here's shown one very famous example from the human gut microbiome project and what you see here is every column here is one analyzed fecal sample of a healthy human and then in color you see the fraction of different bacterial phyla and you see that most of the bacteria detas and fermicutes make up most of the bacterial biomass in the gut. That's for healthy humans, but then people of course went on and studied the composition for all kinds of diseases and there are more and more claims now that microbiota composition is at least strongly associated with many diseases. I hope to come back to that. One of my goals is to explain that actually. But then we have to go beyond sequencing now to really understand something and in the context of this meeting we maybe don't have to stress that so much, but what we have to do is we have to better understand the microbial ecology, this ecosystem of bacteria cells in the large intestine and roughly we can think of it like three rough classes that are minding what is going on. So we have to understand what is going on on the single cell level, how is metabolism happening there, how are cells growing and surviving. We have to as we heard this morning understand more about the interaction of different cells by cross feeding, but then also really important is that we have to think about the environment these cells are living in and in the human gut that's not so easy because we have to deal with a very complex organ and a lot of secretion and absorption events. But nevertheless let's try to think about some generic principle which should be valid for this ecosystem gut. So one thing is because we lose quite a lot of bacteria but the whole system is kind of in a steady state, there's a permanent strong growth of bacteria going on all the time. So it's a very dynamic ecosystem where every cell is replaced by every 40 hours or so. So what is now the energy driving this growth, it's simply what we eat and we can think about the numbers now a bit more what we eat or sorry what the bacteria eat, not what we eat, what the bacteria eat are actually all the carbohydrates which are not absorbed by the small intestine and this includes in particular resistant starch forms and fibers. That's one important thing to notice, the other thing is that these gut bacteria are actually really good in digesting these complex carbohydrates and to take really energy out of it. So let's look at this a bit more closely. So the first thing to notice then is if most of the nutrients are entering the large intestine from the small intestine, nutrients are also mostly available there. So growth is mostly happening in the ascending column. And the other thing is we can think now about how is this growth really happening and one thing to notice is that when these bacteria take up starch and fibers and they are able to break them down to monosaharides, they cannot do respiration. If they would be able to do respiration, they would be able to take out a lot of energy of every monosaharide with an end product, product water and CO2. But actually because there's no oxygen, they have to do fermentation and that means there's a lot of production of these fermentation products. So in particular acetate, butyrate, propionate. And the trick here for the human body is really that our gut can take up these short-shade fatty acids and our whole blood system has of course access to oxygen so we can take that as an energy source. So how much energy contribution this bacteria fermentation to our daily calorie uptake is really a very unprecise number, but roughly it's 10% of our daily energy intake. Or if you want to remember this better, it's roughly one beer per day. So now we can think of basic growth conditions. What is important to set growth rates in the large intestine? And I talked already about a few, another things to consider are actually in terms of bacterial growth. It's really nice, temperature stable, salt concentrations are not varying much. But then pH varies a lot and we also have a strong flow. And I will come back to pH but first want to talk about strong flow. So let's consider flow rates through the gut in a bit more detail. And for that let me show one classical picture from a medical textbook. So what you see here is the water intake or the water uptake and secretion into the gut at different parts of the body. So we roughly drink and eat a total volume of 1.5 to 2 liters of liquid every day. But then there's a lot more secretion going on. So then because of saliva, because of bile acids and so on, roughly 10 liters of liquid is going through our small intestine every day. So then there's a lot of water uptake happening already at the end of the small intestine but still a lot of fluid volume is entering the large intestine every day and roughly it's 2 liters per day. So what does that mean now? That simply means that the average flow velocity at the beginning of the large intestine is really large and bacteria somehow have to deal with this. So picture somehow bacteria who have to grow now in very strong flow conditions. And the simple question here is what is actually if we have such a permanent flow, what is actually preventing washout of bacteria? Why do we have a stable population of bacteria at the beginning of the large intestine? And you can think about different factors counteracting this washout. For example, cells might be able to swim or cells might be able to attach to the wall. But if you look what's happening, many of these strains are not swimming and also wall growth is not enough to sustain these high bacterial densities to observe. So what I'm convinced is that the most important factor counteracting such a washout is actually the active contractions of the intestinal wall. So the idea is simply that we have contractions of the wall and by these contractions some backflow is generated against the average flow and this backflow in combination with bacterial growth is enough to maintain a stable bacterial population at the beginning of the large intestine. So I will come later on to the situation in the human large intestine but let me briefly introduce a project we did first and so our idea was to test this backflow growth versus washout idea in a very controlled setup. So for that we constructed kind of our own mini gut where we have simply a channel and bacteria grow into the channel, we can set flow velocities but in addition we constructed these smaller chambers where we can apply pressure. If we apply pressure we can generate local wall deformations and the idea is really now to generate like contraction patterns and to observe what is happening with bacterial densities and then you can first investigate simple mixing. So don't put bacteria in first but simply fluorescently labeled beads and you can observe what is happening with these beads over time and you see how repeated contractions of the wall lead to a mixing dynamics of these beads. And then you can estimate that actually the mixing you can generate in such a setup can be explained by an effective diffusion process with a fairly large diffusion number. So now what is happening if we put in bacteria into this device? Well we can observe fluorescently labeled bacteria and now investigate different mixing conditions or different flow velocities. So in the upper panel you see the situation where mixing or the frequency of wall contraction is kept constant and in the lower one you see the flow velocity is kept constant and what you see is for example this case here that indeed you simply get a washout of bacteria. This axis shows you the average bacterial density in the channel so you get washout of bacteria over time if contractions are not large enough but you get stable bacterial densities if contractions are frequent enough. And we went on and analyzed that further with a simple mathematical model. I will skip the details here but it's really a diffusion convection model including bacterial growth and if we compare the mathematical model with what we actually observe in spatial temporal dynamics then it's a pretty good match. For example you see here in this case A if flow velocities are large you simply get a washout of bacteria so color code here is time starting with this blue color over time red color we simply get a washout but the situation is really different if you go to lower flow velocities and in between we can get these spatial patterns. And in this paper we actually analyzed further the consequence of cross feeding on or in such a device where we have this mixing and flow but for the rest of the talk what's important to realize is that actually contractions can lead to quite substantial mixing and we can describe this whole process as at least fairly well by a diffusion process. Because I want to go back now to the situation in the human gut and discuss with you some numbers because the question is now if we want to test this idea for the human large intestine can we get some good numbers and or how can we get these numbers and how can we describe this but the good message here is that there are really amazing physiological studies from the 50s and 60s about flow so we have the human physiology numbers quite well and we can measure what these bacteria are doing now. So briefly the numbers for the human gut so let's first talk about dimensions. So the large intestine is roughly 1.9 meters long and the diameter is roughly 2.2 centimeters so that is where growth, where bacteria growth is happening. So now what about water uptake? I showed you this graph before with the total water turnover along the gut and I told you that roughly 2 liters enters the large intestine every day but that is of course not all because we know pretty well that not 2 liters are exiting the large intestine every day so there is a lot of water uptake going on and this water uptake is particularly happening at the beginning of the large intestine and so as a consequence what we have in the large intestine is a gradient in flow rates. So while we have really large flow velocities at the beginning of the large intestine because of water uptake flow velocities are actually becoming quite low in the more distal parts of the large intestine and growth is now somewhere happening in this gradient but the good thing is we have estimations for these flow velocity profiles. So if you want to have a simpler picture, a picture more like a strong flow waterfall situation at the beginning of the large intestine and really a smoothly slow flowing river at the end of the large intestine. So then probably the most challenging part is to estimate mixing and it's not so much as known about the details of mixing in the large intestine, we know much more about the physiology of these muscles in the small intestine but there are a few studies where people have tried to quantify in the context of diarrhea what these muscles are actually doing for particles and in particular there is this nice study by Hammer and Sidney Phillips and what they did is with healthy humans they injected radiolabeled particles at the beginning of the large intestine and then they followed over time the distribution of these particles while they were going through the large intestine. So what I did now is I analyzed this distribution and how it was becoming wider and wider and that allowed us to estimate roughly this mixing, the strength of mixing in the large intestine and if you do that you get a mixing described by a diffusion with a diffusion constant 10 to the 6 micrometer square per second. So we have a number for that but now to really understand how these bacteria are growing we have to understand how bacteria are growing and it's much harder to get really good numbers there but we can try to measure that now under very controlled conditions in the lab and people before have chosen a few organisms so I showed you this phyla plot again what you can do now is you can pick a few represent hopefully representative species from these different phylas and you can analyze how are they growing and you grow them under very controlled conditions under anaerobic conditions and then you observe growth by behavior so here is shown bacterial density measured in OD over time and you see that you can extract perfectly defined growth rates from that and now you can ask how are these growth rates depending on different conditions. What you can also do is you can quantify what fermentation products these bacteria strains are actually producing so this is done with HPLC measurements and you see that as a consequence of fiber or resistant starch consumption you see that these different bacteria species produce different patterns of these fermentation products. So now we have a number for that and now let me talk more about pH because what's important to realize now is that these fermentation products are all acids so that means if there's a lot of growth locally in our large intestine the pH has to go down and so we can ask now well is this relevant for growth and we can ask that question simply by analyzing bacterial growth in the lab so how is growth rate the doubling time of these gut strains how is that changing with pH and so what you see is these remarkable patterns that growth rates are going down a lot if pH is falling and bacteria detests seem to have an advantage compared to firmicutes when pH rates are more neutral. So now is this relevant in the gut where are we in the gut already end of the 80s people measured with I guess wireless capsules you would call it today how the pH is changing so you can capsules going through the gut are tracking constantly the pH value and what you see is that indeed pH is falling from more neutral values at the end of the small intestine to values closer to 5.5 and then it's going up to more neutral values again. So in this plot I showed you the relevant range in the gut is exactly in this range where growth rates are falling a lot. So now if we now want to include everything to understand bacterial growth we have to think more also about the uptake of short chain fatty acids but there are also estimations of that and actually I don't have time to go into this in detail now but the system is quite clever in the sense that short chain fatty acid uptake is often coupled to bicarbonate excretion and that again is acting as a buffer in the gut but we have numbers for that and can put that in the model as well. So now the model has many components it looks quite complicated but we have actually good estimations for every part. The structure is again as for this in vitro setup important is to include mixing and flow but then really all these growth terms and consumption of nutrients and now given that model we can ask what is it predicting house growth dynamics happening in the large intestine and I show you here a situation for a western diet that means that's for the amount of nutrients people estimated to reach the large intestine every day for a typical British diet and you see here the steady state after five days of simulations and so versus distance and what you see here is that nutrients are there at the beginning but then because of bacterial growth nutrient levels are falling confirming this idea that the growth zone is only at the beginning of the large intestine. You see that bacterial densities allow at the beginning because of flow but then because of bacterial growth bacterial densities are going up but importantly now there's a difference between bacteria detas and fermicutes and that's because of the difference in pH the growth rates are not necessarily the same and the model also takes care of short-chain phase the acid productions or all these fermentation products so because of bacterial growth fermentation products are going up and then because of uptake fermentation products are going down again and the pH value is changing accordingly from neutral to more acidic to more neutral values again so now we can ask we put in numbers we had we constructed this model are these observations in agreement with what has been observed and they are quite well so in particular this pH range is confirmed the short-chain fatty acid range with higher values at the beginning than at the end is confirmed and the bacterial densities are also confirmed so bacterial densities in the sequel when people measure that they're really much higher sorry much lower than than bacterial densities and feces okay so now now we have a model so now if we leave this model we can use this model now and ask can we test some some consequences if we change something here and can we maybe even explain this plot of variation from person to person in phyla composition so for that what I will show you in the rest of my talk is what is happening if we change in the model either the amount of nutrients flowing into the large intestine or the amount of water uptake so let's start with nutrient inflow so this is what you get if you change nutrient inflow and this for for typical Western diet in terms of here nutrient inflow is measured in terms of glucose equivalence for typical Western diet this value is 300 but what you see is that you that you get it strong change in phyla composition if you change if you go to higher nutrients and the explanation is really simple because the reason for this all it is is we have more nutrients coming in we have more fermentation pH values are dropping more but more acidic pH gives a growth advantage for fermicutes right so now we can ask is this in agreement with what has been observed and there's one famous study where people indeed observed a change in microbiota composition when putting people on a diet so they took a subset of people with a very high fraction of fermicutes and they observed that if they put people on a diet then bacteria or a detest fraction went up completely in agreement with what we are saying here okay so the other thing is what is happening if we change water uptake and for that let me introduce the Bristol stool scale which has been introduced in the 19th by a rough score how soft or hard to stool is and the nice thing for that is that everyone can measure this basically and more recently people have studied how stool consistency is related to microbiota composition and what has been found by a study of Filoni and all only last year is that actually the Bristol stool score in their correlative study was the most important factor explaining microbiota composition and if you analyze bacterial samples from healthy people and you classify them according to Bristol stool score then at least you see a strong correlation in the fraction of bacteria detest to fermicutes so why I'm talking about that the reason is that Bristol stool score is directly related to transit time and water uptake and so in our model if we change water uptake from low to high then what is happening is that because we have more water uptake concentrations are all getting higher so all these fermentation products also have a higher concentration in the gut so that means pH is actually going down more down if we have more water uptake and if we correlate then got transit time with water uptake and Bristol stool score we get exactly this result that low Bristol stool score high fraction of fermicutes so now we can put everything together and ask for healthy humans if we if we take the typical range observed in healthy humans in Bristol stool score which is huge and the typical range of nutrient amount entering the large intestine which variation in phyla composition do we get and we really get here a change from almost only bacteria detest to almost only fermicutes so what it suggests is that this pH feedback driven by flow and water uptake is really responsible for this huge variation in person to person microbiota composition and of course we hope now that people are taking up this idea and check this idea more directly in humans by really correlating pH measurements with microbiota composition and transit time so with that let me thank again the people involved in this project and let me show you the summary slide with the two projects I talked about and thank you for your attention