Eating for Two Trillion: Dietary Intake Shapes Our Resident Gut Microbes

this program is presented by university of california television like what you learn visit our website or follow us on Facebook and Twitter to keep up with the latest UC TV programs you my name is Sunil Hollywood I'm a member of the Diabetes Center here at UCSF and I'm the enrichment core director for our nutrition obesity Research Center grant which runs in collaboration with Coast and the so grant and it's part of what we think is a really cutting edge very vanguard multidisciplinary funded effort pursuing obesity and diabetes from the basic science standpoint the clinical translational Sciences standpoint the the by behavioral and policy related standpoint simultaneously with all these arms working in lockstep and ideally in a very collaborative collegial manner and I think we've made a lot of progress over the last year in creating an infrastructure to do that kind of science and that kind of translational impact in our population locally and beyond that and so with that kind of multi disciplinary framework in mind I think having a microbiome session is really makes all the sense in the world because we're talking about one of the the most fundamental collaborations that there is in biology and that is the collaboration between us as a host and the teaming microbiota that occupied the spaces in and around our bodies and so with that in mind I'd like to introduce today's next speaker we are very lucky to have as one of our newer assistant professors at UCSF in the department's of microbiology and immunology peter turnbaugh and peter has done really groundbreaking work in understanding host microbiota interactions and has particularly focused since arriving here at UCSF on those interactions in terms of how we process foreign chemical substances be they drugs be they nutrient related substances or other substances such that what ultimately impacts the host is a function of the dual roles that both the microbiota and the host plate together as opposed to each individual component acting separately so we're really happy for Peter to speak to us today and look forward to his comments all right well yeah it's really really a pleasure to be here today it's been a really fascinating morning and afternoon and you know I think something that we've sort of touched on throughout all of these talks is the important interaction between the foods that we consume and the microbes that live in our gut and so I'm gonna focus on that subject for the next 40 minutes or so and so you know you've already gotten a really nice intro from Rob about the complex microbial communities that are in and on the human body and in particular the gut is home to the most numerous as well as diverse microbial community and as you heard in the in Rob's talk there's a tremendous amount of variation between individuals so much so that we can actually identify who you are based on your microbiome and so a lot of the research that's being done in the microbiome field is asking the very simple question of what the major factors are that drive these variations in microbial community structure and whether or not they're relevant to the host in terms of disease or other aspects of our physiology and so I want to return to the year of 1909 when Arthur Kendall wrote a really interesting paper in Jay BC and this was based on research that Arthur had done you know now over a century ago feeding chimpanzees different diets and trying to culture the bacteria that were in their stool I mean he had observed that diet seemed to have a very important role in shaping these microbes and so in particular he mentions you know as food passes through the elementary canal at different levels which be composed by various types of bacteria and that the predominating types of bacteria that take part in the decomposition will be determined largely by the nature of the diet um he also pointed out that you know this link between diets the intestinal microbiome and its end products had been very largely overlooked by 1909 and and I would argue that it remained very overlooked until about 10 years ago so Rob Rob covered a little bit of us and you've seen some of this data throughout the various talks the main tool that really revolutionized how people study microbial communities was based on this very simple basic biology observation which is that all microbes have the same gene or all bacteria anyway and you know so that allows you to develop molecular methods that allow you to profile microbial communities and so that's largely based on 16s the revit 16s ribosomal RNA gene and what you can then do is take a sample from the gut extract all the DNA in an unbiased way and then amplify that 16s ribosomal RNA fingerprint we then use sequencing methods to you know analyze large pools of samples at the same time this is originally done by a sort of traditional capillary Basinger sequencing and then 454 and more recently Illumina and that you know gives you these large data sets that represent the sort of spectrum of bacteria that are found in a given sample a lot of people in the field use chime that's the package that rods group developed there's also other software tools like mother.and and other methods but allow you to take these complex datasets and then group them in different ways and so traditionally people would use what's called an operational taxonomic unit and so that groups together sequences based on their similarity to each other into what we consider is sort of the equivalent of a bacterial species then from that sort of data you can you know graph the relative abundance of different groups of bacteria you can you know generate phylogenetic trees or you can compare the sequences to each other and you can perform various types of clustering methods like principal coordinate and analysis and so that will hopefully give you sort of a basic background to how people are approaching these complex microbial communities in addition to targeting these marker genes through what's called meta genomic sequencing we can now analyze the entire sort of genes that are found in a community or even through RNA sequencing which genes are transcribed okay so for the purpose of this talk I wanted to you know go way back to the beginning of my PhD at Washington University School of Medicine in one of the few perks of living in st. Louis is that we have this giant park right next to the campus and all of the museums and zoos within that park are free and so the first thing that I did is a rotation student in Jeffrey Gordon's lab together with Ruth Leigh and a technician in the lab Sabrina was to just walk through the park to the zoo they let us in for free and then they let us dig around for all the stool samples we could get from all the elephants and and you know an interesting thing that we learned is that despite the fact that you know it's literally lying on the ground we had to go through a very rigorous process to be able to take these samples from the elephants they were worried that we were going to somehow hurt the animal you know I was more worried for us and so this data set that Ruth really led was it was the first sort of overall view of the microbial communities associated with all sorts of different mammals and so in terms of diet we had you know a nice selection of herbivores carnivores and omnivores we actually had three different locations in addition to the st. Louis Zoo we were able to get samples from the San Diego Zoo and risk friend and collaborator was able to go to Africa and get us samples from wild animals and then we also sampled a really tremendous diversity of mammals and so everything from monotremes like a kid knows their platypus to humans and so this is sort of the high level clustering that was done together with Rob Knights lab to sort of group all these different samples from the 60 different mammalian species and then you know label them based on different things that we knew about the animals and so the major thing that really stuck out from this analysis was that could group herbivores in green carnivores in red and then the omnivores in blue even though these come from very distinct mammals and if you're interested humans you know our favorite mammal are right here in the middle of the omnivores at least based on the samples that were available at the time more recently we wondered you know if you looked at a wild animal population you know animals are not static in the food that they're consuming and so over time you might expect the dietary intake would change and that might alter the microbiome and so we collaborated with Amy Peterson and Sarah Knowles who are wild animal immunologists from Edinburgh I mean what they do is they they very painstakingly go into these woods in Liverpool and they track this wild wood mouse over time and what's really unique about this study site is that they they have you know these two different woods that are about 6,000 meters apart and then they've been gridded out into very regular 10 by 10 square foot patches and so that's been used for four decades now to study sort of the ecology of wild mice and so we have you know detailed information about you know where the mouse was living as well as you know other important information like what pathogens that was infected by how old it was and sex and so forth and the striking thing we found is that you know following the microbial community over the course of two years we found a regular shift over time where in in between sort of July and August we saw a shift in the microbial community structure this is just showing sort of one summary of that on the y-axis here and then these are you know just months throughout the two years and so you can see that the overall microbial community tends to change in a similar time if you just take the data between the two years and correlate them to each other they're very reproducible so similar sort of change seemed to happen in both and we could identify individual bacteria that seem to be more enriched and sort of the spring versus fall and so we saw more lactobacillus in the spring and in the fall we saw more more alle sites the true group within the bacteria Dedes phylum and more helical vector which is very interesting because helical vectors a common pathogen of wild mice and so this is something we didn't necessarily expect you know we went into this thinking that maybe there would be a lot of differences between the the fields and between the woods we didn't really see much evidence for spatial segregation and so we were really beating our heads against the wall as to you know what might explain the seasonal effect and maybe not surprising given the intro it turns out that in that month every year the seeds fall off the trees and so the mice are given access to a new diet and so obviously this is a correlation and what Sarah's doing now is doing control the intervention experiments in these wild animal populations something that's very difficult to do and so in the meantime we've decided to take a more controlled approach in the lab where you can much more readily alter the diet of mice and in addition to that we use a technique that's called meta biotics you saw some pictures of this earlier this is a method that's have been around since the 1950s that allows us to maintain mice and a sterile isolator all of the food is autoclaved or radiated for sterilization all of the bedding is sterilized on the area that goes through a HEPA filter and so that allows you to keep mice from birth until the end of their life in completely sterile conditions and it's a really powerful tool that allows us to just ask what sort of differences or what phenotypes are dependent on language and so that typically the groups that we think about first just to find notify otics-rio origin it literally means known life so it just means an animal where we actually know what's in it and then germ-free animals have no microbes conventionalized animals our former leader free animals that we have introduced microbes to and then conventional or SPF mice are just the animals that have an uncontrolled microbial community that are colonized from birth and so a few years ago now the first experiment we did was to take germ-free mice and then colonize them at 12 weeks of age while they were being fed a sort of standard Mouse chow that was low in fat and high in fiber plant polysaccharide and then we placed them on two different diets so one was just maintaining them on the standard Moscow and then the other is sort of meant to mimic the Western diet and so it's incredibly high and simple sugars or oligosaccharides and in fat and in animals that induces diet induced obesity over the course of five two or four to eight weeks and so these animals were kept on the two different diets and then eight weeks later sacrificed at the time we weren't able to do sort of you know many different time points and so we had a relatively small number of samples but I think the profile was so clear that hopefully appreciate it so these are five different mice and the initial acute our mouse donor sample here and then these are the five mice that were placed on the high-fat Western diet and you can see that if we cluster their microbial communities they split perfectly by diet interestingly the the human that are the sorry the mouse that was the donor which was also fed the low fat diet groups with the later time points on the low fat diet okay and we could also sort of measure overall changes in the community and so we saw that you the two major groups of bacteria in the gut the Firmicutes and the bacteria Dedes were altered so the Western diet caused an expansion of firmicutes and a decreased abundance of accurate Dedes another thing that you've heard a lot about is the overall diversity of microbes in the gut and we saw that the diversity of bacteria and what significantly decreased in mice that the Western diet relative to the control thing so the next thing we wondered is you know how quick these changes occur and so you know using four five four sequencing we are now able to really expand the number of samples and time points that we could look at and to this apologize the screen is a little small but this is a PCO a graph where again the the dots that are close to each other represent time points from mice that have similar microbial communities and these mice were colonized and one day after the colonization they are all still fed the low-fat diet you can see the time points cluster together we then like seven days and 28 days later and they they sort of move a little bit from the initial colonization but they're still grouping nicely and now we're going to switch half of the mice to the Western diet and ask how soon we see an effect on the microbial community and we were really shocked that just one day after the dietary intervention we saw a dramatic effect and that increased a little bit a week later and then was maintained over the following two months and so the Western diet at least for school samples has a very dramatic effect within only a single day we also we then wondered you know whether or not this was just something special about the end of the gut you know stool was somehow different or if the best thing that I could also affect the microbes that are found in the small intestine in different parts of the large intestine and so what's shown here is just samples that are colored by their location in the gut and so the stool is in sort of orange and then the small intestinal samples are in green there's also stomach cecum which is the part of the intestine between the small and large intestine in mice and the colon and what you can see is that all the groups are all the sample is taken on the Western diet or on the right of this graph and all the samples taken on the low-fat diet are on the left so despite the fact that there's a lot of differences in the types of bacteria that can colonize differently patience in the gut diet seems to have a much more dramatic effect than where you are along the length of the gastrointestinal tract so a couple years ago we decided to then you know see what would happen if we did a similarly dramatic dietary intervention in humans and so we wanted to design you know two very distinct diets that we could minister to people and then monitor daily changes in the gut microbiome and so to start with we decided to choose sort of the two most extreme diets that we could think of one was all animal products so this is my favorite diet it's a my favorite food salami then then we gave a delicious array of barbeque eggs and cheeses and then we also designed a very disgusting diet called the sambisa which was grains legumes fruits and vegetables okay and so I just want to you know emphasize to you that we really caused a dramatic difference in dietary intake to what people were consuming at baseline and so we were able to get these people to take daily diet logs throughout the whole course of the experiment the diets themselves were only five days and we had nine people that went through both arms and and one person that dropped out after the first and then your replacement but what you can see here this is grams of fiber intake per thousand Killick Alan soom so during the the few days of baseline and they're consuming sort of a similar amount of fiber to what people in America would normally take in I mean that went up about two and a half times during the plant-based diet and then as soon as they're let go onto their watch out or allowed to eat their normal diet they returned to the standard level on average and the animal-based diet we did something sort of more extreme which was in addition to taking out fiber we actually removed all carbohydrates from the diet and so they were only consuming protein and fat as macronutrients so it's sort of an extreme version of the Atkins okay and so as you would expect on the first diet there's a little bit of a drop in fat intake on the animal-based diet we actually increased fat intake to 70% of their daily calories and protein was about 30% and so that's just as a high level there's also you know many subtle differences between these two diets that we think are interesting okay the next thing we just asked was you know based on sequencing the microbiome overtime from daily samples collected throughout these two diets you know how similar was each day for a given person to their baseline microbial community structure and sort of a measure of how much of an effect the diet had on these peoples and microbiome and so in terms of the plant-based diet we were sort of surprised that we actually didn't see much of a change in the overall structure of the community whereas for the animal-based diet we saw that within a couple days of the diet I mean actually one day of when a food coloring that we added to the food was detectable we saw a significant increase in this sort of beta diversity measurement and so this suggests that the animal-based diet is really dramatically shifting these people from where they started we next ask the question of which bacteria were higher on the animal-based diet or lower on the animal-based diet I mean this is a really complicated figure where on the x-axis we're looking at the the fold change on plant the plant diet relative to baseline and then full change on the animal-based diet versus baseline on the y-axis and the sort of interesting thing is that you can then dig in and look at which bacteria we're up on the animal-based diet and we saw a number of bacteria like the waffle of odds where the eye as well as other members of the bacteria bt Asylum that have been shown to be very resistant to bile acids and this was very interesting to us because it's known from prior work but during you know consumption of high levels of dietary fat the amount of bile that's produced by the host is increased and so that might provide a selection pressure or you know provided a way of reducing the growth of some bacteria and increasing the growth of others on the flip side when we looked at the bacteria that we're down on this diet we saw a number of interesting bacteria like rose Barea a new bacterium recto Rick Talley that are known to be very efficient integrating plant polysaccharides or fiber and so that you know may be the reason that they're no longer you know seeing the fiber that was in the baseline diet okay we could also measure the sort of downstream products of microbial metabolism using these samples and so a product that you've heard a lot about is short chain fatty acids and these come in many different flavors the two that we hear a lot about all the time is acetate and butyrate these are produced when microbes ferment carbohydrate and so as you as you might predict on the plant-based diet we see more acetate and more butyrate relative to the animal-based diet the other thing that was really neat is that we can actually measure measure I so Valerie nice of butyrate which are produced during the bacterial fermentation of protein and again you can see those go up on the animal-based diet and so that suggests the the microbiome is able to sort of shift its metabolism from carbohydrate to protein based on what we're consuming you can also sort of see signals of that sort of shift in metabolism based on sequencing the expressed genes from these samples for example this is a pathway for using glutamine and glutamate we saw that all the enzymes were breaking down these amino acids were expressed at higher levels on the animal-based diet and the enzymes required to make these amino acids were expressed at higher levels on the plant-based diet and the next thing that you might wonder is you know what happens if you go out and find somebody that's been a lifelong vegetarian and then ask them to eat meat this is a only based on one person but I think it's informative but take that as a caveat so on the plant-based diet you know one of the really striking things that's been found in the microbiome when we compare individuals from around the world is that these two groups of bacteria found in the gut the Bacteroides and the prep atella tend to be mutually exclusive so if you have a lot of credit ola you usually don't have many Bacteroides and vice-versa and nobody really knows what the reason for that shift is but it's potentially linked to dietary intake and so on the plant-based diet we found that the subject 6 that was a lifelong vegetarian and also an immigrant was it was fairly stable so maintained incredibly high levels of private olives so that's 60 to 80 percent of the bacteria in that individual stool whereas all the other individuals in the study sort of stayed at high Bacteroides and didn't markedly increase the amount of prep atella whereas on the animal-based diet we saw something really interesting which is the vegetarian although they again during their baseline samples started - telleth we could see three days and then four days after the diet that they were starting to shift towards a more Bacteroides dominated microbial community and comfortingly to us after going back to the baseline diet they sort of snapped back and so we didn't permanently destroy their revitol okay the other really interesting thing about dietary intake is that it's not just about how diet changes the resident microbes in our gut we're actually consuming live microbes as well and this is most well studied in the case of bacteria but also the cake also true for fungi viruses and other microbes and so one of the things that we were able to do in the study was to administer three artisanal cheeses that were donated by a an anonymous cheese vendor and we been sequence the bacteria and fungi that were in each of these cheese's so in the top here is blue cheese we saw mostly Staphylococcus was the most dominant group and then you can see there's a lot of Canada in the blue cheese right and then that that's true for the so we were able to profile the rinds as well as the center of each of these cheese sites you can also see string cheese which was actually still had high levels of streptococcus thermophilus and fish and as well as the fermented meats prosciutto and salami and so that provided us with sort of molecular data that we could then look for these microbes in the stool samples that we collected from these patients over time and so yeah I think I took out that for time so we were able to detect them based on 16s but the maybe more exciting thing is that when we looked at the RNA that was in these samples we were able to see that many of these bacteria as well as fungi had markedly increased expression on the animal-based diet when when people were consuming the cheese and so this and and because of the fact that RNA does not survive very long in the gut it implies that these microbes are actually surviving transit throughout the gastrointestinal tract and so you know we don't really know whether or not they did anything interesting in there but they at least were able to survive and we were able to actually confirm that by taking stool samples from people on their baseline time points and during the animal-based diet and show that we could actually culture out live fungi that were from the cheese that they were consuming and so one of the potential consequences of these changes is linked to inflammatory disease this is not the main purpose of the study but something that I think was very intriguing and it's connected to a series of studies that's been done in Eugene Chang's group at University of Chicago I mean what gene and his graduate seem Suzanne Dakota showed is that a few feed diets that are high in fat in particular saturated fats you can drive an increased host production of I'll acids those vile acids contribute to the expansion of Buffalo woods where the eye of the bacteria I mentioned earlier and if you're a mouse that's genetically susceptible to inflammatory bowel disease the BLA flow will then cause ulcerative colitis like phenotype and so you know one of the things that was sort of disturbing is that the number one group of bacteria that we saw enriched on the animal-based diet was buffalo included buffalo buds were the guy in green there we were able to confirm that using another independent method called a quantitative PCR based on a gene that is unique to buffalo which again we see is elevated on the animal-based diet and we also saw that this gene was actively expressed based on RNA sequencing in addition to that if you just look at the dietary intake records for these people we found that the strongest predictor of Bulava abundance was the amount of dietary fat that was consumed the bile acids were also elevated and so this provides the first sort of support in humans for for the hypothesis from the Chang lab the diet and the host production of bile acids may alter the microbiome in a way that could contribute to disease obviously this is a very short-term intervention so we have no idea whether or not these people if left on this diet would develop inflammatory bowel disease but it's something that's very important we think to look into okay um so you might be wondering you know if diet is really so important you know is there any role for host genetics or the immune system or other components of our own body in shaping the microbiome and so Rachel Carmody one of the postdocs in the lab decided to test those using animal models where we can sort of systematically change the mouse genome and see whether or not that is stronger or less strong than a dietary intervention okay and so we started using a panel of outbred mice that we obtained from slab raters at Jackson Lab so so these are animals were every Mouse has a very distinct onset of genes and in addition to that based on the way that these mites were derived we can really easily sort of reconstruct their entire genome and so we know every mutation throughout the genome of each animal ok but the important thing is that we took 60 of these mice and then we we started them again on this low-fat diet put them through sort of two successive pulses of the high-fat diet and so they initially got it just for one week they were returned to the low-fat diet for a week and then put on the high fat diet for two months and then we just asked you know across these 60 animals how reproducible were the changes that we see in the microbiome and this sort of surprising thing is that you know despite the fact that there's a lot of genetic diversity in these animals we saw that in all of the mice and they started with sort of one community structure and within you know just one or two days again on the high-fat diet their microbiome was altered they sort of switched back during the recovery period and then you know during the following two months and returned to the same sort of state and so again that suggests that you know diet really has a dominant role in shaping the microbiome in addition to that we compare different inbred mouse strains where they have very similar genomes within each group and we found similar effects they even were I think surprising thing is that if you remove the entire immune system you still see these effects and so for example we can knock out mighty 88 which is a key adapter protein that's required to sense bacteria I'm using toll-like receptors I mean these mighty 88 mice still respond to the high-fat diet although maybe interestingly they they sort of regressed a little bit so that might be you know the maintenance of this altered state might be dependent on the immune system in addition to that you can knock out all being t-cells by taking out rag1 and the rag1 mice actually respond very similar to Wild Thing and so the next experiment that a really hard-working undergrad in the lab Asus Lovano decided to do was just to toggle the diets of these mice back and forth and to see you know how robust are these changes I'm too over time and so he had two groups of mice one that was oscillating and so every three days we went between the low-fat and high-fat diet he had a second group just to sort of control for differences in the facility that might be sort of causing our microbes to change over time that were oscillating in the on the opposite phase and then a control group that stayed on the low fat diet until the very end and a high-fat fed group that was switched immediately and then maintained and so I guess for time I'll sort of skip through the host data but the main finding that we observed was that the microbiome changed incredibly rapidly and in a reproducible fashion and so you can see here this is the oscillating group in the solid line where every three days with it you know immediately the microbiome is altered and stays in that altered state until it's returned to the prior diet and then the reverse group is shown on the dial and the dotted line and then these are the control groups of the low-fat you know stays pretty constant and then the high-fat chips right away and sort of stays in that lower area and so the other sort of related question is that you know these two diets that we've been comparing in mice have been relatively distinct from each other and so what happens if you do a more subtle dietary intervention would that still affect the microbiome and so raychel blended these two diets together where we have anywhere between you know the the complete low-fat diet - the complete high-fat diet and then you know different steps that either you know 25% high-fat 50 percent 75 percent and then we again measured the microbial community and we saw that there was a dose-dependent effect of the amount of high-fat high-sugar diet both in terms of sort of overall structure of the community as well as the abundance of the major two groups of bacteria and got the Firmicutes in the backyard Edie's and then you know as was mentioned in one of the earlier talks we think of these effects that are mediated by these high-fat diets could again have important consequences for the host one of the things that's been studied by us and and others is the potential role in metabolic disease one of our earlier experiments was to compare again mice and low fat versus high fat diet that were used to colonize a panel of germ-free recipients and then we just asked the question of whether or not the mice that got the high fat diet associated microbiome gained more fat than the control group and what we found is that although the the mice that got the control donor microbiome still gained some body fat there's about twice as much increase in the mice that were given the high fat associated community and the same is true in mice that are colonized by human samples and then put through the same dietary intervention and this is really surprising because in our hands anyway the mice did not actually change their caloric intake at all and they were very similar at the start of the experiment in terms of their weighting and body fat okay I'll just briefly mention one of the ongoing experiments in the lab which is to now ask the question of you know what happens if we consume the same diet but just change how the food is prepared and so you could imagine that this would be really important for the microbiome for a couple different reasons one is that cooking the food that we consume aids in its digestion and so for example this is a picture of plant cells where in red this is the starch that's found in these very tight lobules and then after applying heat you can now see that the starch is more evenly dispersed throughout the cell and so this makes these carbohydrates much more susceptible to cleavage by enzymes in our body in addition to that when we cook our food we've all sorts of interesting chemistry occurs and that includes you know taking away compounds that are in raw food as well as producing new compounds and so Rachel decided to test those by administering either diets that were all meat or all all plants in this case a sweet potato in either raw or cooked fashion and in addition to the sort of overall impact of the substrate she was able to find that on the raw huber diet there was a very dramatic effect of cooking where all the raw samples are on the left here and all the cooked samples are on the right and so something about this raw tuber diet was having a very important effect on the microbiome we then looked at which bacteria were altered and we found that on the raw tuber diet there was an increase in bacteria Dedes of the expense of the firmicutes and it's also shown here but maybe more interestingly we found that there was evidence for antimicrobial effects of these raw diets and so this is an assay where we're just measuring the abundance of damaged bacteria in the gut of mice fed either the raw tuber in orange a cook tuber green ampicillin as a sort of positive control antibiotic in red and then chow in gray and you can see that on these raw tuber diets there's a dramatic elevation in damaged microbial cells similar to what happens when you give them into bionics so in conclusion and we found that your gut microbes are really influenced very closely by what we eat and that's true when we sort of compare mammals across mammalian evolution it's true in lab mice and we think is also true in humans and finally you know early our ongoing work in the lab suggests that it's not just about what we consume but how we prepare prepare the food and so just to wrap up as you've sort of gotten a sense for over the course of the day there's many potential consequences of changes in the microbiome that are caused by a change in value and that can be very fundamental aspects of our physiology whether or not that's our weight or at a paucity or other aspects of our metabolism or potentially our predisposition to disease many diseases recently have been linked to the microbiome including metabolic syndrome malnutrition IBD and cancer and so a lot of the work moving forward will be trying to test whether or not these diet dependent changes in the microbiome are contributing to the increase in disease that we see in Western populations okay so with that I want to thank Corrine Murray she was a postdoc in the lab who did the wild math study on Lawrence David who led the human dietary intervention and then Rachel who worked on all the results in mice that I showed you so with that thanks so much and they look forward to you yes so that's a really good question so in mice you know there's typically a few types of fat that are added to these foods we have used a couple different flavors and some of our early versions we were using animal fat and so it's sort of a purified saturated fat from beef we've also done experiments with a milk derived saturated fat and other labs have used many other diets so gene you know the the experiment I told you about that was done in the Chang lab actually compared many different fat sources and found that it was only the milk derived saturated fat that caused this dramatic expansion in Buffalo and so they're currently trying to figure out you know what specific lipids were in that source that could actually be causing that effect in humans we haven't really you know broken it down to that control of the level you know we're feeding real foods so these were the types of fat that you would see in you know beef and pork products but we haven't you know done sort of controlled fat experiments yet in humans but that would be a great experiment to do yeah we have not studied that I think yeah that's particularly was for me and Roberta's lab Emory but it's definitely a really interesting area I think you know similar to the sort of the fat consumption story that I mentioned for Bella flow in Andrews research the emulsifier consumption depends on sort of host preed is this a predisposition to disease so if you consume a diet rich and emulsifiers and you're you know a healthy wild type Mouse you know you don't actually develop inflammatory bowel disease or metabolic syndrome but it's sort of the combination of the emulsifier intake and another hit on that side that can drive disease and and that you know may well be true in humans as well and is a really important area for future research yeah no I think that's a really important point that you know often what we're talking about when we show an effect of diet on poor diet or antibiotics or other perturbations on the microbiome including enteric infections and other things that might happen in the gastrointestinal tract are you know really fundamentally changes in the abundance of bacteria that are found in the gut and so there's actually very little precedent in humans anyway and in mice that you can remove or add that add in new organisms during that sort of perturbation it tends to be a change in abundance and so you know one way we think about it is that a lot of the stability is sort of the microbes that you're colonized early in life those particular screens or species might be carried on you know through many different perturbations later in life but their abundances relative to each other it can change on the timescale of days or maybe hours and so you know it's an important question whether or not you know which is more important you know how similar are the strains that are new versus me at a functional level and you know should B be more or less worried about changes and abundance or changes in overall composition yeah so there's a there's a couple things that are important to point out one is that humans are hard to experiment on and when we do mass experiments the experiment is ideal in the sense that they're on one diet and then shifted to diet bu you know every animal eats diet B all day long at will or every time they feed anyway and you know it's a very defined shift whereas in humans typically when people do nutritional interventions and I apologize everybody you know there are many concessions that are made to what you can get a person to eat you know typically the dogma is like no one will sign up for a trial where they eat the same food all day long for multiple days that sounds horrible and we did that for our study and you know God people sign up but you know I'm sure there's a selection bias so that you know wouldn't be a generalizable strategy and so you know you try to induce incentives to people to sort of you know consume very boring diets or to do it in a more controlled setting than they would normally do at home but that's undoubtedly you know one of the major issues life it's very hard in humans to to link you know dietary intake in the and the microbiome or any other trait yeah yeah it's actually it's a not and one of the inspirations for our experiment was there's this paper that was published back in 1923 in JAMA there was an arctic explorer that you know had seen that what was being consumed and was impressed by the fact that despite the fact that people were eating these high-fat diets they were very lean you know obviously they're living you know in the Arctic so maybe that helps but they actually did a study in 1923 where the Explorer in his you know crazy clinician friend ate meat for multiple years without any other source of nutrient and that was actually one of the first sort of proofs that you don't need other sources of vitamins in your diet but interestingly in that paper and just to emphasize how long people have been working on the microbiome there's a there's a paragraph on the microbiome in that paper and there's actually a follow-up study just about the profiling and so you know that idea that these high-fat diets are altering the microbiome is very old yeah so there is data on sort of full-out calorically restricted people some of which from the Gordon lab and others looking at the microbiome and then there is also just because he's on my slide Reiner dumper it's on the bottom right here is a visiting scholar who's here at UCSF right now he's in the audience and when I when I was in graduate school Reiner and I collaborated together on a study where he was able to administer controlled diets to individuals that only vary so they were controlled in terms of macro nutrient intake but that had differences in overall caloric intake and he was able to show for the first time that you know those changes in terms of caloric intake can affect the microbial community in the cop so that's published work yeah I think that's a really interesting question and one that you know other labs are working on in the context of early development you know so whether or not exposure to these diets in very early ages can have long lasting consequences probably the sort of related set of studies that's most well-known is work that's been done in Marty blazars lab showing that low doses of antibiotics early in development can you know cause increased weight and mice or increased body fat sorry in mice at later time points even though it doesn't really cause any permanent change to the microbiome although you know the related question is whether or not there is any sort of memory of our dietary intake in the microbiome itself and I sort of skipped over it do do complexity but you know one of the things we did in the oscillating experiment was to collaborate with Georg Gerber who's a mathematical model ER and he was able to find individual species and and Adeem families that were altered in response to the number of successive exposures to the high-fat diet I mean so at least in that experiment you know those groups were different on the final high-fat diet are in verses you know the first or second or third and so that you know there even though the overall structure is very reproducible there might be you know some individual you know components of the microbiome that can sort of remember what we you you

Channel: University of California Television (UCTV)

Views: 36,222

Rating: 4.6644592 out of 5

Keywords: microbes, pharmacology, nutrition, microbiome

Id: u56zXy0UCN8

Length: 48min 50sec (2930 seconds)

Published: Thu Jan 28 2016