eLife Episode 50: Inside Your Microbiome

How gut bacteria alter diabetes risk, and how the microbiome can manipulate your mood...
10 October 2018
Presented by Chris Smith
Production by Chris Smith.


This special edition of the eLife Podcast marks our 50th episode and we’ve decided to mark the milestone by focusing on a field that’s huge and tiny both at the same time: huge in terms of the rate at which the discipline’s growing and the impact it’s set to have our lives, and tiny because its subjects are microscopic. It's our microbiome, the community of micro-organisms that live on us and in us and outnumber our own human cells by maybe 50 fold: we’re literally passengers in our own bodies, and over the next 30 minutes we’ll hear how gut bacteria might alter your risk of diabetes, and how the microbiome can manipulate your mood...

In this episode


01:04 - Microbes on a diet

Soy-derived oils are toxic for beneficial bacteria in the small intestine but not in the gut.

Microbes on a diet
with Sara DiRienzi, Baylor College

Sara DiRienzi explains to Chris Smith what happens when mice are fed on a diet enriched with linoleic acid, an essential fatty acid that we eat an awful lot of these days…

Sara - So I got to thinking about what we do in our daily life and that's that we eat a lot of food, but what we eat from day to day frequently changes quite a bit. And that food that we consume, this is exactly what our microbes get to feed off of. So if we're changing everyday with they get to see, you would think that we would be changing what microbes live in our intestine every day. But that's not true. Over time, we actually see that our gut microbiome is reasonably stable and so the big broader question is how do we maintain a stable gut microbiome? Either the gut microbes themselves have mechanisms in order to deal with a rapidly changing diet or there's the host gut environment is somehow buffering that direct interface between the diet and the gut microbe, and changing how that interaction occurs.

Chris - Indeed, microbiologists often reassure people, even lonely people, that thanks to your microbiome you never dine alone! But the point is that this is really hard to test isn't it, because the microbiome is very diverse, diets are very diverse, people are very diverse? So how have you sought to control this and try and turn those enormous numbers of variables into something tractable?

Sara - Well, I also just want to bring up that this question is actually even more complicated when you're really talking about the diet that we eat. This is really a question for the small intestine and not a question of the large intestine. Less than 10 percent, sometimes even only 1 percent, ever really reaches the large intestine.

Chris - Okay. So just talk us through the experiment then and essentially what you did.

Sara - So if you follow with the very simple hypothesis that when we consume dietary fat it's broken down into its individual fatty acids, and those individual fatty acids are what interact with gut microbes. So what we can do is to simply take those individual fatty acids and directly put them on gut microbes that are found in the small intestine - these fatty acids. So, for instance, like linooleic acid directly inhibits gut microbes that are found in the small intestine.

Chris - That's rather strange isn't it because linoleic acid is one of the essential fatty acids? We can't make that in the body through our own metabolism, we're absolutely dependent on our diet as a source of that, so it seems rather counterintuitive that something were obliged to eat is going to poison our microbiome?

Sara - Exactly. Moreover linoleic acid is also the fatty acid that we consume the most of. It's highly abundant in vegetables and nuts so it's found in all of your vegetable oils. You know these are food you eat every single day.

Chris - And that being the case then if it's that toxic we shouldn't have any of these microbes left in our microbiome should we? We should have driven them all out of existence, so something's going on?

Sara - Exactly. So then we get to; well let's see what actually occurs in the small intestine. So to do this what we set up was a long term diet study. In this study we designed diets that were very carefully controlled and just differed in their percentage of fat versus carbohydrate. And the specific fat that we use in this diet was soybean oil rich in the linoleoic acid. So by putting put mice on these different diets we can see what happens to their gut microbiome. So when we do that, really kind of surprisingly, is you don't actually see very big changes at all, which is not what you would expect based off of what we see that happens in a test tube.

Chris - One of the perils of extrapolating to the in vivro situation; what goes on in vitro isn't it this? It suggests then that something must be endowing the microbiome when it's in vivo with more resiliency; that there's some protecting factor or something else is going on?

Sara - Yeah. So we also took this study a step further. We used a compound called propidium monoazide. And what it does is that it allows us to identify microbes that have been affected by the inhibitory effects of linoleic acid. So what we see is that those gut microbes that were inhibited in the test tube, they're not showing that inhibition in the gut. In fact, they're actually doing better than some other got microbes. So, in fact, what we're seeing is that some of these gut microbes are not directly seeing linoleic acid in the gut the way that they directly see them in a test tube.

Chris - So that would suggest then that something is getting in the way of the access of that chemical into those microbes or, the microbes themselves are doing something different when they're in situ in the gut where they're supposed to be?

Sara - Absolutely.

Chris - But you don't know what, or do you?

Sara - No. Well we have some hypotheses. These gut microbes that we're talking about, they are known to live deep within the mucus. And so it might be that these specific microbes we're looking at have gone and just hidden within this mucus layer and so they essentially have a protective shield.

Chris - And can you get at that?

Sara - Yes. So we have some ways to move forward, and one of our approaches that we can use is a technique called experimental evolution. And what this technique allows us to do is to look at the genetic changes that gut microbes acquire as they're routinely exposed to a stressor. So in this case our linoleic acid is the stressor. When we do this experiment in a test tube, we see that gut microbes change the way that they metabolise fatty acids and they change their membranes. So then we can ask are the gut microbes that have survived this exposure, do they have those same changes in fatty acid metabolism and in their memory? And the answer is no they don't. And so this, again, is highlighting that these got microbes are not directly interacting with linoleic acid.

Variety of chemotherapy drugs in vials and an IV bottle.

07:00 - Antibiotics accelerate diabetes

A dose of antibiotics early in life may increase the risk of type 1 diabetes.

Antibiotics accelerate diabetes
with Martin Blaser, University of New York

Antibiotics undoubtedly save lives, but could we be saddling some individuals with a lifelong increased risk of a range of immune conditions if we treat them at a young age with antibiotic drugs? New York University’s Martin Blaser thinks we are. Although they don’t know precisely how yet, as he explains to Chris Smith, his team has found that perturbing the microbiomes of infant mice prone to developing autoimmune diabetes dramatically accelerates the trajectory of their disease, perhaps mirroring what we’re also now seeing in human patients with type 1 diabetes…

Martin - This is a disease that has been accelerating in recent years. It's doubling in its incidence every 20 or 25 years. During the period of time, when this disease has been doubling, antibiotics have come on the scene. And so antibiotics are a possible factor in this disease and other diseases like obesity and asthma. And the way they may be playing a role is by affecting the human microbiome, the organisms that live in and on the human body. We know that antibiotics perturb the microbiome. Our hypothesis is that perturbing it early in life is what's causing this big increase.

Chris - So what was the experimental method here?

Martin - We did our work in mice. There's a kind of mouse called the NOD mouse that spontaneously develops type 1 diabetes. And we wanted to see could we accelerate the diabetes by giving the mice antibiotics early in life? And in fact we found it could in male mice. There were differences between males and females, which we are exploring, but almost everything I'm going to tell you has to do with male mice.

Chris - Right. So you have cohorts of mice. You either give them, I'm presuming, a control where they don't get any antibiotics or you give them a dose of broad spectrum antibiotics and then you follow them up and you ask do they or do they not have a higher or lower rate of diabetes subsequently?

Martin - Yes, you said it very nicely. We give the mice one course of a macrolide antibiotic very early in life and then we follow them to see if they will develop diabetes. And here's where we saw that even one course of antibiotics both accelerated the diabetes, it happened earlier, and actually more mice became diabetic.

Chris - And can you show that in line with using that microlide antibiotic you also do genuinely perturb the microbiome, and you perturb it in a consistent way in the treated animals

Martin - Yeah. So our first findings had to do with the effect of the antibiotic on the microbiome. And we found very general kinds of differences that were consistent across mice, and we were able to name specific organisms that were significantly perturbed and that were associated with the development of diabetes.

Chris - Now if that's the relationship that is causing the effect you're seeing, one would speculate then that if you put back the missing bugs or you just recolonise these animals with the microbiome they had before you fiddled with the antibiotics, you should be able to defer the diabetes risk. Did you do that experiment?

Martin - We are doing that experiment now which is to add back normal microbiota and ask the question can we get them back to their baseline?

Chris - Type 1 diabestes being an autoimmune condition. You're seeing an acceleration of the trajectory of the condition. You might think that it was some kind of loss of the normal gating processes that's holding the immune system back. Do you think that's what's happening when you perturb the microbiome in this way?

Martin - Yes. So the next part of the study was to look at the effects of these changes of the microbiome on the host. We focused mostly on the intestine of the host and our main focus there was on gene expression, which genes are turned on, which genes are turned off. And our first observation was that when we look at mice as they get older we can see a pathway of maturation. Over time, some genes are turned on and some genes are turned off. Now what we've found is that the antibiotic perturb microbiota very much altered this normal maturation.

Chris - There are a number of other immune related conditions that appear to be increasing in incidence in the modern era in line with things like diabetes and antibiotic use. Do you think we've got a common mechanism here then that we're exposing infants to large doses of antibiotics at critical window periods in the microbiome maturation period, and that is having a knock on effect for the way their immune system matures and, therefore, things like allergy risk and so on?

Martin - Yes, we agree with that too. We didn't study that in this paper but it is consistent with the parallel increases in such diseases like asthma. And in fact, when we looked at the gene expression we found many pathways that are seen in many different immunological conditions. We saw perturbation in these pathways and we also found perturbation in metabolic pathways also. Pathways about how the host is processing energy compounds for example. So we think this is all tied together, both metabolism and immunity.

Chris - So are we medically 'robbing Peter to pay Paul?' Because we're we're looking at the short term at how we keep young individuals safe and make sure that doctors don't get sued. But longer term are we saddling that individual with immunological consequences because of our defensive practice?

Martin - Yes, so that's an extremely important question. First let me just say that antibiotics are wonderful drugs. They are miraculous drugs, they've saved countless lives. But doctors and other health practitioners are using antibiotics more and more and more for very negligible illnesses in which it's thought that well since antibiotics have no real cost, no biological cost, we may as well try them. The mantra is 'it may not help but it won't hurt.' But we're finding more and more evidence that it may hurt and that it may change the arc of development. That's very concerning because antibiotics are so widely used in childhood all over the world.

Chris - Is it too late then? If this has been done to an individual is there any way of resetting the system or has this ship sailed?

Martin - That's a very important question. We don't know the answer. That's why we're doing restoration experiments to see, can we restore, can we get them back to their baseline? It's possible that the ship has sailed and that we have to focus on the next generation. On the kids who haven't been born yet so that we can prevent this fate for them. We can prevent this problem from getting worse.


13:57 - Microbes and sociability

Mice raised in sterile environments are less social.

Microbes and sociability
with Roman Stilling, University College Cork

According to the saying, “You are what you eat”. And to endorse that,  it’s looking increasingly likely that your microbiome might even affect how you behave in social situations. Mice uncolonised by bacteria, reared in so-called “germ-free” environments, and born to mothers who were themselves microbe-free tend to socialise a lot less than mice born normally. And in the brains of these microbially-deficient mice, neurones in the amygdala, which coordinates fear and other emotional responses, express very different patterns of gene expression, suggesting that the microbes in the intestines can influence how the brain handles information. Chris Smith hears how from University College Cork's Roman Stilling...

Roman - We were looking at the reasons that germ-free mice, so mice that are devoid of any microbiome, have social behaviour alterations - they are socially impaired. What we find in a classical paradigm that is called the "three chamber social interaction test" where you have one chamber with another mouse that the test mouse can interact with, or on the other side of the arena there's another chamber that has just an object like a ball or an egg cup or something like that. And we know that we usually see, when we just test normal mice that they interact a lot with their partner and less with the unsocial object. But with the germ free animals they interact less with their social partners.

Chris - So what was the additional set of questions you were able to ask of these mice when you did this?

Roman - In a previous study, we compared the gene expression in a specific brain region that we know is involved in social behavior to control mice. And we found that in the amygdala, this specific brain region, there were several genes differentially regulated that are associated with neuronal activity. Based on this finding, we wanted to see what is actually happening in a dynamic situation, so when they actually encounter a social interaction, what is then happening to this amygdala?

Chris - So how did you measure what the genes are doing in that part of the brain at the time that the animals are having their social relationship compared to when they're not?

Roman - What we need to take into account is that the gene expression is always a bit delayed. So we do this task just the normal way, put the right into an areana and let them explore. And then we wait one hour afterwards, take out the brain, and then we look specifically for the brain region we were interested in. We extract the new made RNA, then we send this for RNA sequencing, which is a method to look at the whole transcript home, so all RNAs that are in the cells that we analyzed at once with the next generation sequencing technology.

Chris - So what do you find then?

Roman - In a normal mouse when iT experiences an novel situation specific genes are switched on and and off. What we find now in the germ-free elements is that we also see these pathways and we also find something completely novel and unexpected. Genes that are in themselves regulating gene expression by modifying the chemicals that are produced during gene expression - the RNAs. And these RNAs, when they are they are produced by the cell to make new proteins then they need to be cut in to the right message, right, and this is why they're called messenger RNAs. And this RNA processing pathway that was heavily upregulated in our study.

Chris - Interesting. So the way that the RNA is being handled in the cell is changing. How is that being affected?

Roman - We don't know exactly how this is happening. But we see that the genes that are coding for factors that do the cutting, these genes themselves are turned on and off to a much higher degree than in control animals.

Chris - And are these changes in gene expression profile and splicing profiles reflected in a structural or connectivity difference in the amygdala so if you look at a sort of microscopic level can you see any differences?

Roman - We have done this in another study and actually did find more sinuses and more complex dendritictry in the germ-free animals, which somehow fits our narrative.

Chris - So how do you think that the message is transmitted between the microbiome, which is broadly in the intestine, and the developing nervous system?

Roman - To be honest we don't really know. But I think what we need to focus on in the future is the immune system. We know that the immune cells of the brain, the microglia, they are also impaired in their function in germ-free animals. That I think is certainly a way that we need to look closer at to study the interaction of the bacteria in the gut with the function of the brain.

Bacteroides biacutis, one of many commensal anaerobic Bacteroides spp. in the gastrointestinal tract.

19:19 - Vitamin B12 bacterial thieves

Some bacteria are able to steal vitamin B12 from us as it passes through the small intestine.

Vitamin B12 bacterial thieves
with Aaron Wexler, Vanderbilt University

Vitamins are molecules that we can’t make ourselves and that we rely on our diets to supply. But we’re not alone, because many of the chemicals we need bacteria depend upon too; and this can sometimes bring us into conflict with our own microbes. Aaron Wexler explains to Chris Smith how this works for vitamin B12...

Aaron - We're fundamentally interested in answering the question what does it take to be a successful gut bacterium. And so we know from previous work in our lab - we used a model gut bacterium, part of a geneus called bacteroides that's almost universally present among human populations - one of the things it takes is the ability to take up the small molecule vitamin B12.

Chris - Now when you say a "good microbiome" member, is that good for the bacterial outcome or is that good for the host?

Aaron - So when I was saying a successful gut bacterium, I was speaking predominantly on the side of the bacterium. So they want to colonize the gut, and so if they are not doing their job right they'll go extinct from that environment

Chris - But, equally, if they're too promiscuous and too greedy there could be consequences for the host too, couldn't there?

Aaron - That's certainly the case. Most of these bacteria are localised to the large intestine and they seem to do just fine there. And we get some benefits from them in terms of digestion, in terms of small molecules like Vitamin K, and folate. However, when they find themselves in the wrong location in the intestine, such as the small intestine, then we start seeing problems in people.

Chris - And do those problems include include things related to B12?

Aaron - Yes. So these bacteria are very good at taking up vitamin B12. One of the focuses of this paper was a particular protein called BTG that we identified, which is on the surface of these bacteria and has a very high affinity for vitamin B12. So when these bacteria with these proteins with high affinity for B12 find themselves in the small intestine and grow to large densities, then they start taking the B12 away from our proteins.

Chris - What, because that particular B12 binding protein is so good at doing it's job it's better than our own one, and so if you put the two head to head in a competition the bacteria win?

Aaron - That's certainly the case in vitro. So when we take this human protein called Intrinsic Factor that we used to take a vitamin B12, give it vitamin B 12. But then we incubate it with this bacterial protein , we find that in a very short period of time the bacterial protein has taken almost all of the B12 away from this human protein. So we can reason that a similar process might be occurring in humans and in the small intestine when these bacteria grow to too larger numbers.

Chris - It's like throwing a bulldog and a poodle a lamb chop isn't it and asking who's going to win? Why has the bacterium evolved to have this incredible scavenging power for B12 then.? Obviously it's a valuable molecule but this sounds like a case of overkill to die for?

Aaron - That's right. I think it relates more to competition with the bacteria over B12 versus competition with humans for B12. So I think stealing B12 from intrinsic factor, the human protein, I think that's accidental. And I think it's a consequence of competition for vitamin B12 between gut bacteria versus between humans and gut bacteria. So these bacteria are, like I said, they live predominantly in the large intestine and the densities of bacteria there are are very high. And so with high bacterial densities you get a high level of inter-bacterial competition for nutrients. The human intrinsic factor has also evolved with humans over the course of our evolutionary history and it's not used to competing with gut bacteria for vitamin B12. And so when these bacteria find themselves in high numbers in the small intestine where intrinsic factor is, it just so happens that these bacteria will win that competition pretty much every time.

Chris - So does this fill in that sort of clinical gap then, because we knew that classes of patients do end up with a deficiency in B12, and they do have a bacterial overgrowth and now this explains why these bacteria are able to render us B12 deficient?

Aaron - I think it's reasonable to say that this is an explanation for that. We don't know for sure that this is happening in these patients because we haven't done the experiments to prove that. However the in vitro data that we did collect suggests that this is a reasonable hypothesis.

Chris - Now how did you actually stumble on this in the first place?

Aaron - We were studying B12 uptake in these gut bacteria. And we found that a certain class of gut bacteria called Bacteroides contained the same four components involved in vitamin B12 uptake as E. coli does, and E. coli is the classic bacterium in which vitamin B12 update had been studied. So these bacteroides species in our gut contain the same four proteins, but they contain extra genes that are adjacent to these genes encoding these transporters in bacteroides, and these genes didn't have known functions. And one of these genes, which we called BTG, turned out to be the surface protein with high affinity for vitamin B12.

Chris - So what did you do, a little bit of conclusion jumping where you thought well, they're in the same region so they might be involved? So you then went and looked at them and found oh hey presto, one of them really is involved in this and this is the function?

Aaron - Yes. So we couldn't assume, just based on their genetic location, that they were involved in B12 transport. To get at that answer we took a molecular genetic approach first by deleting BTG, and when we did that we found that these bacteria struggled to grow in low concentrations of B. The second clue came from studies we did where we found the intracellular concentration of B12 was lower in bacteria that lacked this protein.

Chris - So what are you going to look at next?

Aaron - We have some clues that suggest that BTG isn't the end of the story. That there may be other components at play and we don't know what they're doing. Another thing would be to try to get at the clinical side which is to try to figure out if BTG really is the key piece that explains why a small intestinal overgrowth of bacteria leads to vitamin B12 deficiency in people.

Human fingerprint

25:27 - Microbiome pecking order

The order in which we acquire bacteria early in life affects the composition our gut microbiome as adults.

Microbiome pecking order
with Jens Walter, University of Alberta

Jens Walter explains to Chris Smith how the first types of bacteria to colonize the gut could shape our microbiome as adults...

Jens - One puzzling finding is that each individual has a unique and stable community of gut microbes that is almost as personal as a fingerprint. Studies have not tried to identify what causes this individuality and found that factors such as genetics, diet, environment or lifestyle, also you know health of its theological state only account for around 30 percent of the variation.  And even identical twins actually have different microbiomes. So this suggests that something unknown actually contributes to microbiome individuality. So what we tried in our study now is to determine the factors that contribute to microbiome assembly, and more specifically we tested the hypothesis that differences in the order by which we acquire microbiomes early in life can actually cause variation later in life, and the high level of individual variation that we are observing.

Chris - Why do you think that the order in which we pick up microbes in this way might make a difference at all?

Jens - This is very much based on the ecological theory that colonisation history, or more specifically priority effects, actually influence the outcome of community asembly. Priority effects mean, basically, that an early coloniser will often have an advantage because it is occupying the niche first, and it will also have an impact on the ecosystem by influencing colonisers that come later. Due to this, the order or the timing by which these members in a community arrive will then shape how the community looks in the end and how it functions.

Chris - I suppose it's a bit like if you walked into a room and there was a bunch of empty seats and there was no one sitting in any of them you'd choose any seat to sit in? But if you walked into the same room and there were two or three people sitting down already you might change where you sat and who you spoke to?

Jens - That's an excellent metaphor actually for what we are feeling. In inequality, we are actually referring to this as niche pre-emption. The one who comes first basically sits first.

Chris - Now how did you test this though? Because it's a nice theory, it's a nice idea, but how can you subject that to proper scientific scrutiny to find out if that's really what's going on?

Jens - So we have given this quite a bit of thought. So what we did is we used mice that don't contain any microbes, meaning they were germ free. And these mice were further genetically identical and they were housed exactly under the same condition using airtight plastic bubbles. So what we were basically trying to do is we controlled all of the experimental variables that we could think of. We then associated these mice with either entire microbial communities, or specific bacterial strain cocktails in combination with communities, and we introduced these entities in different succession. We could specifically test if the timing and order of early life colonisation influences how the bacterial community is composed.

Chris - What about the immune system though?  Because I put it to you that when I put a group of bacteria in, they might do something to the immune system which that, in turn, affects the receptivity of the location for whoever comes in next?

Jens - Now that is an excellent question and we are actually giving this quite a bit of thought. We think that early colonisers can somehow interact with the immune system and actually in enthuse tolerogenic immune responses that would then favour this early colonisation. To test for this we actually repeated our experiments in mice that do not have an immune system and therefore cannot actually at least develop immunological tolerance to these early colonisers. And when we did these experiments in these immuno deficient mice we still found a clear evidence for priority effects.

Chris - So what are the implications of this then? Because we've got lots of babies popping out all around the world every day, not all of them are being born via the normal route. Some are being born by say caesarean section. Some are being given big doses of antibiotics when they're born, usually for very good reason, but that could therefore have, based on what you're finding, very profound effects on the future microbiome structure for those individuals.

Jens - We can make two major conclusions here. First, as you said, we get an understanding about you know what actually shapes our microbiome. And since we showed that the early acquisition is important, clinical practices that influenced this early acquisition of the microbiome, such a caesarean sections or antibiotics, formula feeding will actually then influence how the microbiome looked later in life. So appreciation for these ecological events early in life might help us to develop different clinical practices to potentially prevent a barren assemblies of gut microbiomes. The other main thing that we learn is that if we want to modulate the gut microbiome more permenantly we probably have to do this early in life, because early colonisers have a higher success of actually establishing themself in the ecosystem. And not only this, these early colonisation effects might also allow us to specifically out of the entire community establish a more healthy microbiomes or prevent the establishment of microbiomes that might have a detrimental effect in the end.


Add a comment