81. Microbiome 1: Introduction

In this episode, the ID:IOTS Podcast Microbiome Correspondent Dr Anastasia "Tash" Theodosiou, returns to talk about her favourite subject! 

This episode is part 1 of 2. Here we discuss: 
- What is the microbiome
- Viewing the microbiome as an organ system
- The microbiome's association with disease
- How we study the microbiome

Stay tuned for part 2 (in 1-2 weeks depending on editing), when we talk about the effect of medicine on the microbiome, microbiotoxicity, and what you can do to keep your patients' microbiomes in tip top shape! 

https://www.researchgate.net/profile/Anastasia-Theodosiou

https://x.com/doctoranastasia

Transcript

Callum: 0:00

Hello everyone. Welcome to the Idiots Podcast. That's infectious disease insight of free specialists. I'm Callum, that's Tash. That's going to be Ja and we're going to tell you everything you need to know about infectious disease. Hello,

Tash: 0:27

Hello. Thanks so much for having me back on the show.

Callum: 0:30

Always a pleasure to have you back and our first, second guest. For, we had first guest. What about a second? Second guest? Sorry, I can't resist a lot Rings reference. I'm delighted to be joined by dr. Anastasia. Theo. Who is a clinical microbiology and infectious diseases trainee based in Glasgow and is currently finishing her PhD on the early life microbiome. And you will, if you are a loyal listener, have heard her tones on her previous episodes on breastfeeding and the. Early life microbiome. So a warm welcome back Tash, and I know that you've got a lot of really interesting things to share with us, and I'm excited to learn more

Tash: 1:22

yeah, great. I'm really, really pleased to be here.

Callum: 1:25

So what are we here to talk about today? We've already talked about breastfeeding. We talked about the importance of, the microbiome in the early infant development about how breast milk is inherently designed to encourage the micro, the microbes that live in the infant gut and the importance of a sort of healthy start to life with exposure to the different organisms around the parents' microbiome. And if you haven't already, then check out episode number 56. I found it really interesting and I've actually listened to it. Since recording it because I thought there was a lot of really useful stuff in that, but given that was so well received and so interesting we were really eager to have Tash back to, to tell some more.'cause I think there are are other questions that remain to be answered, at least in my head. So Tash, have you some sort of basic initial introduction that you'd like to give us to the topic.

Tash: 2:23

I guess we focused a lot last time on the microbiome in early life, and particularly around breastfeeding, but then certainly in some of the chats you and I have had. Since then, we thought it'd be useful to look across the life course and think about how the microbiome is, is functionally important to our patients, and then specifically to us as infection doctors, looking at the microbiome as it relates to antibiotic prescribing. So I guess if we take a little step back just to cover some definitions again. So the microbiome, as you, as you may well know, is. The total community of, of microbes living on and inside a, a human host, usually defined as like a specific niche. Strictly speaking, the microbiota is all the living organisms, whereas the microbiome is like the stage on which that plays out. So, it's, it's the organisms plus the genetic material, the metabolites, and to some extent the host niche as well. And I. I always like throwing it out there. Can you tell me roughly how many bacteria you think each of us have?

Callum: 3:29

So I'm gonna say maybe roughly 30 trillion. Is that just a random guess Out the top of my head.

Tash: 3:37

K is currently reading off my show notes. That is quite cheeky, but yes. It's about 30 trillion, which happens to be the same number of cells as the same number of human cells as each of us has in the body. So it's roughly one to one. But I think taking it a step further. I would go as far as to say that it's not just, mind bogglingly large in its sort of scope, but functionally incredibly important so far that I would say that all of your loyal listeners should start thinking about it as an organ system in its own right. What do you think?

Callum: 4:08

I completely agree. I think that. For as long as I've been practicing and learning about the body and medicine, I've had this quite basic understanding of, human and bacteria. And, up until recently, I feel like bacteria, microorganisms in the body were viewed as bad. And then we, I talk about healthy bacteria, but we had quite a low understanding. And I guess it's, it comes into that, that weighing of the risk. We are like, oh, they've got an acute kidney injury. If the antibiotics are, oh, they've, got some sort of liver toxicity, but, actually reframing that and saying, well, if this is an organ, then, organs are important. So if this is. If we are to think about this as an organ, I think everyone would know what the kidney does. The liver may be a bit more confusing. But if we're thinking about this as a, as an organ, what function does it play in the body?

Tash: 5:01

So I think it's really helpful to think about it in terms of function. You've absolutely nailed it. So rather than think about it in, in terms of composition, I think about it in terms of function, because the reality is there's 30. Trillion different bacteria, and there's lots of different variation between individuals about what that composition would look like in terms of what species are there, what ratios are in, but actually when you look at their product, so the actual function of a healthy microbiome, it's pretty well conserved across individuals. Yeah, so lots of different ways to make a healthy microbiome. Lots of different compositions, but the end functions are very similar. And the end functions really have to do with the type of stuff that the microbiome is making or is influencing the body's production. And I call these sort of the bioactive molecules. And I think there's a few things to say here. So one is that these bioactive molecules that are made by bacteria can access. Receptors on and inside human cells. So they're directly interfacing with our metabolic pathways. Now if that doesn't sound scream evidence of a shared evolutionary history, I dunno what does but more than that, they are. Sort of acting across different organ systems so they're not act only acting in the site where most of them are made, which in the human is mainly the gut. Although not exclusively, and we can talk about other microbiota as well, but mainly in the gut. But because these bioactive molecules. Can diffuse outta the gut and into the bloodstream. They can act at a distance almost like a, a kind of an endocrine system would. And increasingly we're recognizing relationships between the gut microbes and the brain or gut microbes in the genital tract. So I think these relationships are really interesting. I can talk through, I mean, there's loads of different bioactive molecules. I'm gonna focus in on four. So we've got short chain fatty acids. Which are made by the bacteria when they ferment dietary fiber. Probably the most functionally important is butyrate, but there's many. And I think what's really important with that one is it's profoundly anti-inflammatory and critical to maintaining normal cellular home metabolic homeostasis. So you've got your short chain fatty acids. Then you've got several hormones not usually being made directly by the bacteria, but the bacteria is influencing the gut mucosa production of these hormones. So these are all linked to glucose homeostasis. So things like insulin ghrelin, which controls satiety. Then you've got neurotransmitters like dopamine, serotonin, and GABA are the main ones, but there is also more minor ones. And then finally vitamins. So we know, from vitamin deficiency states. That people who lack a diverse microbiome can end up being quite significantly vitamin deficient. So those would probably be my four main categories. So short chain fatty acids, hormones, neurotransmitters, and vitamins.

Callum: 7:41

So four main groups, short chain, fatty acids, hormones, neurotransmitters and vitamins, and a mix of either the bacteria producing them or being fermented. But all of these are, are linked in with the microbiome. So without the microbiome, these things wouldn't, wouldn't function, I guess. Is that, is that what you're saying?

Tash: 8:01

Difficult to say. I mean, we know that when you take away the microbiome so for example, things like a, a germ free mouse behaves in a, very abnormal way physiologically. So I don't know what effect that would have on a human to have absently, none of those. Certainly some of those are made excuse exclusively by the bacteria and many of them are account for significant component. So, we need our microbiome to have normal functional levels of those four classes of essential mediators. I guess we might be putting the cart before the horse. That's, those are the products. But if you're thinking about what they do, there's myriad different functions that those products are involved in. But I think if we had to boil it down to just two words, I would probably say metabolism and inflammation. I. And regulating those two things. So I can go through each of those in turn. So I guess metabolism, I think many of us know that at a luminal level, the microbiota are involved in digesting our food. So I guess that's probably the one that kind of intuitively makes sense. But actually I'm more interested in talking about the mucosal side of things and once you actually get into the cells, so many of these bioactive molecules we talked about, so especially short chain fatty acids. Bile acids and endocannabinoids will directly interact with extracellular and intracellular receptors and activate all sorts of downstream pathways, mainly via G-protein coupled receptors. And you basically that, they will be locking in to the host's normal glucose metabolism, normal satiety and maintaining kind of low oxidative and nitric oxide states in the cells. So that's the normal pathway. Then the other sort of arm of that function is inflammation. So in normal, functional and again, I'm focusing very much on the gut we know that those bioactive molecules and the bacteria themselves help maintain a, sort of mucosal barrier. So with healthy tight junctions, they produce antimicrobial peptides. The commensals provide a physical sort of competition against pathogens and opportunistic organisms. And they also because they interface directly with the receptors with downstream pathways, so via things like toll-like receptors. You actually get a kind of intracellular and then bloodstream shift towards different cell types. So you get a shift towards more tolerogenic cells like regulatory T cells with anti-inflammatory cytokines and a shift away from pro-inflammatory cellular responses and cytokines. And this is especially true when you look at early childhood. So to the point that we're thinking that the microbiome in early life is programming. The, the child's subsequent immune responses to to pathogens and allergens.

Callum: 10:40

So, it's quite complicated and I guess you've, you've simplified that down although I'm still, still finding a little bit complicated and so. I guess to summarize what I took there was that there's two main functions, the metabolism, and that is the normal state of the microbiome producing chemicals that directly interface with parts of or normal sort of human physiology, relating to glucose homeostasis, ct. And reducing oxidative sta and nitrous oxide production. And the immunity, inflammation normal function is around a barrier, the gut antimicrobial peptides. And then shifting how our immune system is set up. What cell types are there around about the gut? is that about

Tash: 11:29

Exactly, and, and encouraging a sort of generally quite tolerant state so that you're not generating immune pathology or autoimmune disease like, like allergies and that sort of thing. And similarly, when you deviate from normal, and again, we'll come back to later what we mean by normal and how we, how we know normal from abnormal, but as you deviate from a sort of normal, healthy, microbiome. That you start replacing these kind of anti-inflammatory commensals with more pro-inflammatory patho. And you start having a, an increase in some of the permeability of the mucosal lining. And there's arguments here. There's some have proposed that this kind of low grade inflammation is potentially the kind of, the effect pathway linking to some of the diseases that we're gonna talk about later on.

Callum: 12:18

I'm looking forward to hearing about those So that's the function and you hinted at what happens when this goes wrong. So what do we know about this? And its link with, I guess, being medics. We're interested in disease states, so how does the microbiome link to our understanding of disease?

Tash: 12:39

So I think if we take a step back, what happened early on in, in the microbiome science, so if we go back 10, 15 years ago, people started looking at the microbiome in, in healthy people and comparing it to the microbiome in, in people with a particular disease state. And they started finding differences. And so those association type studies became very, very common, very popular and lots and lots and lots of of diseases were looked at in this way. So it's quite a long list. The kind of evidence linking those varies the both in terms of the quality and the kind of reproducibility. But I think. The strongest links are those diseases that are centered on metabolic or inflammatory dysregulation. So where the link between a microbiome signal and a disease state appears strongest in humans are things like obesity colorectal cancer, inflammatory bowel disease, diabetes cardiovascular disease. So all of these are kind of diseases of either. Metabolic or inflammatory dysregulation. Now that's how things started off. So they started off as association studies where people were just spotting some, differences in microbiome profiles in different cohorts. It's really difficult to quantify exactly what you mean by increased risk, because again, this comes back to how we define a normal microbiome or an abnormal microbiome. And traditionally this was defined. By the fact that it was seen in a disease state so that the D, the dis difference was seen in a disease state versus the healthy person. So if you have no causality data and no intervention data, you can't actually say that it's an abnormal microbiome for all you know, that's a compensatory microbiome that's trying desperately hard to fight against the disease phenotype. But we've thankfully moved on a bit from there. So I think when people criticize microbiome science or are skeptical of it because they say it's association data, they're offering operating with a literature that's five to 10 years old. Because more recently there's been a real push towards more mechanistic models, and those are both in terms of animal models, but also human interventional studies. Again, this comes back to what we were saying about metabolism, inflammation. So in, in a lot of the mechanistic work where they're really starting to look at what's going on at a cellular and both immune cell and transcriptome level, we're seeing that. There's actually increased inflammation in these disease states and that it therefore appears plausible that the microbiome could be having a causal role, as in you're changing to a more pro-inflammatory metabolically dysregulated microbiome. And then you're seeing a disease that is characterized by increased inflammation. Altered metabolism. So that's the way things have moved. Now, if you want a smoking gun, you have to be able to show that you can induce disease by first changing the microbiome. And that's where the animal models have come in really importantly. You can do it with germ-free mice, or you can do it with microbiome depleted mice. So ones that have had, say, really heavy duty antibiotics. And essentially what you're doing is you are taking a microbiome from a diseased mouse or a diseased human. So someone who has the disease in question that you're looking at and you transplant their microbiome, usually through a stool transplant. So a stool gavage in a mouse. And what you find is that you not only. Change the microbiome in that mouse to represent the disease state or to represent the disease donor. But you also induce the phenotype. So in terms of demonstrating causality, there's many models, many different animal models that look like that. So there'll be like a sort of. Airway hypersensitivity model to, to look at asthma. There'll be a hyperinsulinemia model to look at human diabetes. They'll look at where they induce cancers and then see what happens. But essentially there's quite a lot of these different models where they take the microbiome, transfer it in, and induce the phenotype. So that's step one. Step two is that you then do another FMT or another fecal microbiome transplant from a healthy. Mouse or human, and you then restore the healthy phenotype. So again, adding to this idea that there could be a causal relationship. And what's really interesting is that in some of the, not all, but in some of these animal models, the disease phenotype is heritable to the mouse's offspring. So, and again, that's. Not that's not likely happening on a genetic level, but rather that the pup is then weaned and housed with the with the mum and passes on the microbiome that way, which again is emphasizing this idea that early life microbiomes can program downstream immunity and downstream health phenotypes.

Callum: 17:10

Wow. So we have this genetic material as human beings but much more diversity of genetic material in the bacteria. So we say like, oh, there's no genetic link. It's not. Heritable. But then you're saying there that in these animal models that there was a heritable component when the microbiome was affected. So it's really interesting to, we we're often looking at diseases like, what is the gene, what is the genetic link? And, but maybe we've been looking in the wrong place, so I guess we've got animal models, but we're not, we're not zebrafish. So, what human studies have we got?

Tash: 17:48

So they fragment into a couple of things. So I think in terms of showing a, causal link between the microbiome and disease. Probably, I think the most robustly studied one would be Clostridium difficile. So I think none of us would scoff at there being a microbiome basis for c diff infection. So essentially it's the, the archetypal one. You come in, you disrupt the gut microbiome and take something that is, a harmless colonizer in about. 5% of people, adults, and potentially many more in children and you encourage it to move to a disease state where it overgrows and becomes very tricky to deal with. So I think from a sort of disease one, that's probably the one that I think is most studied in humans, But also there, I guess you then get into interventional studies is the other way that you can show causality. So not necessarily trying to damage the microbiome to show causality, but rather trying to rescue the microbiome in some way in an interventional study. Because if we can show clinical improvement through interventional studies, then again what you would have to show is. You do an intervention, the microbiome changes more towards that of a healthy person, and then the phenotype changes to more of that of a healthy person. And those are starting to improve. Again, there's plenty we could talk about here. There's things like fecal microbiome transplant, not just for c diff, but for other indications like ulcerative colitis or IBS are probably the two where it's been looked at the most. Some really exciting data coming in with malignancy. And, sensitizing patients before immunotherapy. And then there's a different subset of data. So looking at probiotics, which are there's quite a spectrum in there. So there's some of them are not the best of science with a lot of industry involvement all the way through to some very good science that is incredibly hypothesis driven, and then also meta analysis. So, although no. Probiotics have made it into guidelines. There are several meta-analyses showing benefit of particular strains in particular indications. So the ones to highlight would be things like antibiotic associated diarrhea, especially in children, and not just any strain, just like you wouldn't say that any antibiotic is effective or any chemotherapies effective. It obviously is gonna be an agent specific thing. So probably the ones that have the best evidence base would be things like croce and lacto species. So, yeah, so I think there's, that's probably how I would think about it. I'd think the fecal microbiome transplant and probiotics as being two of the kind of big umbrellas of interventional research. As we move forward though, there's another very exciting area, which would be the live biotherapeutic products. So these are imagine if you think of, fecal microbiome transplant, but made in a lab, into a. Highly characterized pharmaceutical product where every capsule or every rectal infusion is exactly the same. And we've actually just had the first two agents reach FDA approval. I don't think they're used outside of research yet in the uk, but they've just gained approval. So these are live consortia of bacteria. That are regulated as drugs as distinct from fecal microbiome transplant. So that's really exciting because at the moment, the first one was for, for c diff, but I think it's not gonna be long before these are being looked at in the context of things like ulcerative colitis, IBS, et cetera.

Callum: 21:04

speaking from personal experience, when fecal microbiome transplants were first a thing, it was quite do it yourself. And maybe unregulated. And as things have moved on, we've now got better processes, better donor screening guidelines about how to do that. And I guess that came about because clinicians saw the huge benefit that this. This had on treating people with recurrent c diff, where, where the huge recurrence rates and what we were doing wasn't working. People started doing this. So, as this becomes more developed, it could be really exciting and also maybe more palatable to patients. But I guess what we're gonna do is come back to specifically about c diff and talking about how we can mitigate the harms of affecting the microbiome in the second part of this two-parter.

Tash: 21:56

Hmm.

Callum: 21:57

So I guess my next question is, you've mentioned some research and I, I wondered maybe if what we, one of the things we could do is signpost people to some. What you might suggest is some seminal papers or, you talked about there's, there's good quality and there's low, lower quality evidence. So if we could cherry pick a couple of those to share just as a type examples in the show notes. But how do we get the data? So, obviously we're talking about the research and the evidence that underpins this. How do you or how do other researchers study the microbiome?

Tash: 22:26

the microbiome or the microbiota is a sort of overall community of organisms. So you need to extract the genetic material from those organisms in order to be able to study it. So the way you do that is, is sequencing and there's kind of two. Divergent approaches to that. So you can either do what's called amplicon sequencing, which is where you focus in on a very specific locus or area of the bacterial DNA or you can do a, what's called metagenomics, which would be sequencing absolutely everything in a sample. And there's kind of pros and cons of each. And I can go through them. Most of that is done by 16 s sequencing. Not all. There's other amplicons of interest, but, but most of it is, is 16 s sequencing. And I'm sure everyone's heard this because we use 16 s as a test in diagnostics, but if we take a step back to what it actually is so it refers to the 16 S-R-R-N-A gene, and this is a pan prokaryotic gene. So it's present in absolutely every pro, which of course includes all bacteria. And what's great about this. Gene is that it's incredibly highly conserved. So most of the gene is identical in all bacteria, and then it has these hypervariable regions that are completely different in every bacteria. So by sequencing just a couple of these hypervariable regions, you can actually differentiate between most bacteria. Now. Now the caveat is that you, if you sequence just a couple of the hypervariable regions, you'll usually only get to a species or even a genus level. So it's not good at telling you what strain is in your sample, but it's very good at giving you a sort of. Broad, relatively cheap community profile because you can do this with high throughput, relatively cheap sequencing platforms like Lumina and get quite a lot of DNA back quite quickly. And we've already mentioned that you can know people talk about sending off 16 s in diagnostics. That's slightly different. So what's happening there is that you're usually sending off a. You know what should be a sterile site from someone who's had an infection, but it's culture negative. So maybe you've got, A CSF or a joint fluid and nothing grew, but you're convinced that there's an infection. So you send it off for 16 s and what they're doing there is they're usually, they sequence more of the genes. So either the whole gene or quite a few more regions.'cause you really wanna get that species level resolution. But there, the goal isn't to. Define a community, but rather to use it as a diagnostic tool. So that's sequencing at amplicon, whereas metagenomics is even though it's very often misused because people will use the word metagenomics for pretty much anything, but what it should strictly be used for is, sequencing all the genomes in a sample. And so that would be absolutely everything, human, DNA, fungal, viral, the works. And then after that you pick apart the bits that you need and that actually will let you get down to strain level resolution potentially. Tends to be more computationally and, expense upfront. And it's also more prone to bias once you start getting into really low biomass sites. But those are the two approaches.

Callum: 25:30

And you've explained that very succinctly. I imagine that with all this huge amount of data, the big challenge comes. You have this big readout, and then how do you interpret that? How do you compare that between different people? I've seen some presentations by yourself and others about this, and I was amazed at how, how complicated it is to get answers. But as a brief introduction that's really helpful. So thanks. We've covered quite a few things in this sort of introduction to the microbiome. Tash has come on as our microbiome correspondence, I think as James titled you or so. Yeah. So maybe we could just wrap up and, and summarize what we've talked about briefly before we talk about the next part, well I'm gonna try and summarize it'cause that helps me learn. And if I get it wrong, then you can jump in or add something. So you talked about what the microbiome is, which is essentially the, the microorganisms that we live within us as humans. And that's very diverse and depends on the body site and is, much more genetically rich in a way than our own genome. And there's at least as many bacterial cells as our human cells. We talked about how that is. I. Important as essentially thinking of it as a human organ system. So the microbiome in itself, particularly in the gut, but also in other areas like the skin, upper spirits, tract, genitals have a really important role in the function of US as humans. And so that's by acting through bioactive molecules, either directly or indirectly. And they have four main groups, the short chain fatty acids, hormones, neurotransmitters, and vitamins. But overall we can think about the function of this organ as two most important things, the metabolism and inflammation or immunity. Why does this matter? So it's associated with the disease states, lots of association studies and has took us through a bit about how the quality of our data has moved on from these sort of more association studies. Where we looked at. Disease states and what the microbiome differed between normal and abnormal quote unquote towards studies now where we're looking at mechanistic models and animal models to try and determine is there a causation there. And we're now into the state of interventional studies and we heard a little bit about how you studied the microbiome, which is through either amplicon or metagenomics sequencing. That. Do you think that summarizes what we've talked about? Anything to, to add to that? Tash?

Tash: 28:01

put column.

Callum: 28:03

So hopefully that helps you to put away in your head what we've just heard about. Thanks very much for coming on and I'm sure lots of other people will be interested to, to learn as well, any closing statements for, for this part of our two-parter, or maybe we can tell them what the next part's gonna be on. How about that?

Tash: 28:19

Yeah. Well I think we've talked so much about, how there's this, potentially we should think about it as an organ system and that it seems to be associated with disease and I think as infection doctors. The big question is then, how are we contributing to this cycle of disease and deviation from normal? And I think that's what we're gonna focus completely on in the next session. So looking at what happens when we prescribe antibiotics, what impact that has on the microbiome, and how we should think about all of this exciting science in our prescribing decision making.

Callum: 28:51

And what would you call that if you were to give it a name?

Tash: 28:53

Well, if only we had a word. I mean, we have nephrotoxicity, we have hepatic toxicity. It turns out you can just make up a word and people will let you publish it. So if you check out the show notes in preparation for the next episode you can read all about microbio toxicity. Which is the term that my colleagues and I coined to to provide a framework for thinking about the bystander effects of antibiotics on a patient's microbiome.

Callum: 29:20

Thank you, Ash.

Tash: 29:21

Cool. Thank you.