Chris Bode: Hello and welcome to this Pharmaron Podcast, part of our DMPK Insights podcast series. Thanks for joining us. My name is Chris Bode, and I work on all types of DMPK studies as well as BCS classification studies performed by Pharmaron US Labs based in Exton, Pennsylvania, near Philadelphia.
Today, I’m happy to be speaking with Peter Turnbaugh. He’s a microbiologist and a professor of microbiology and immunology at the University of California, San Francisco. Welcome, Peter.
Peter Turnbaugh: Hi. Good to see you. Great to be with you.
Chris Bode: So, Peter is in the middle of so much of the groundbreaking research in a field that I find incredibly fascinating, the gut microbiome. And in the interest of full disclosure, I am absolutely not an expert on the gut microbiome, but it’s probably my favorite scientific topic. I follow it kind of as a hobby, always with the hope that I’ll get to work on it at some point. And although I think it’s been a really hot field for a decade or two, it seems like it’s still like it kind of flies under the radar still. I’m always trying to let people know about it, and I’ve organized two symposia on the gut microbiome under the banner of the Delaware Valley Drug Metabolism Discussion Group. In fact, that’s how I met Peter. He helped me organize the first one in 2016.
Before we really get started, I wanted to make sure I didn’t forget to let you know that Peter hosts his own blog called Science is Fun with an exclamation point. You can find it on Apple podcasts, maybe other places, too. And Peter, I’m so glad you told me about your podcast for a couple of reasons. One is this is my first podcast, so I took some comfort in knowing that at least you’ve done this before. And the ones I’ve listened to are really interesting and I’ve learned a lot.
So I know on your podcast, the episodes are very long.
Peter Turnbaugh: Oh yeah, maybe too long I have been told.
Chris Bode: I don’t know. They’re just packed with information and it’s really interesting. It seems like you, I don’t know if this is true of all of them. But the ones I’ve listened to, it seems like you lean heavily on your UCSF colleagues.
Peter Turnbaugh: Yeah, I mean, that’s not intentional. I’ve, you know, I try to reach out more globally. We’ve had some international guests. Fergus Shanahan was on, you know, I tend to do better with microbiome researchers that are in other places then once I get in a different field, I really, you know, they typically don’t reply to me unless they’re at UCSF.
Chris Bode: Yeah, no, it’s been great. I’ve learned a lot about things that I really never thought of before. Like, like the placenta, you know.
Peter Turnbaugh: Yeah, yeah, really fun.
Chris Bode: This is so fascinating. So for our listeners, there are so many fascinating things about the gut microbiome. And I’d like to share a couple of intriguing facts. First, when we’re born, we’re 100% human. But within a few years and for the rest of our lives, we’re more than 90% bacterial cells. There are 10 times as many bacterial cells in our bodies as human. And second, if the gut microbiome were an organ in terms of the total mass of bacteria, it’s as large as the liver. So those are just a couple of things that I found really fascinating and got me wanting to know more about it.
So Peter, can you talk about how you got into this field? Kind of your origin story, I guess?
Peter Turnbaugh: Yeah, definitely. I went to grad school at WashU. Really excited about genomics at the time; it was shortly after the human Genome project and there had already been some isolate bacterial genome sequenced. And so I was just really excited by the potential of being able to see all of the genes within an organism. And so, I didn’t really have a biological question in mind. It just kind of went in wanting to do something related to genomics and was also really fascinated by genetics and how genes work.
And fortunately ran into Jeffrey Gordon, who’s one of the really early pioneers in this area and he got me into it and that the timing couldn’t have been better. It was around the same time Ruth Lee joined his lab and you know, she brought all these culture independent methods from Norman Pace’s lab. And, and so we worked closely together on a lot of the early studies in the group.
Chris Bode: Yeah, I wondered if you had been in Jeffrey Gordon’s group. Obviously he’s one of the foundational people. And you mentioned Norm Pace. I was surprised when I read that his name kept coming up in the gut microbiome. I actually took, he was one of the lecturers in the biochemistry course I took at the University of Colorado Medical Center.
Peter Turnbaugh: Oh yeah, he’s had a huge impact. Yeah, that’s fascinating. Yeah, I mean the big, you know, as you probably know, the big revolution was this, you know, the use of PCR, you know, of specific genes like 16s to give you a profile of a microbial community. You know, before that, people were using, you know, really old methods that you’ve probably never heard of, you know, DGGE and RFLP and you know, other things to, you know, get kind of fingerprints, but you really didn’t get any information into what type of microbe it was. It was just all kind of abstract.
Chris Bode: Yeah, that’s really great. I mean, like, I’m not a microbiologist. I’m also, you know, the genome, I’m kind of on the outside of that too. But quickly, you appreciate how important it is to the genomics is and metagenomics I guess you often call it. Right, because it’s not really, it’s in the host, but it’s not the host that you’re sequencing.
Peter Turnbaugh: And I think like, you know, nowadays people at least in the field, don’t really appreciate how amazing these culture independent methods are, you know. You know, really around the time I started in grad school, a lot of people were still culturing and you know, it’s just so hard, you know, even if you are able to culture the microbes, the culture condition you choose really skews the community. And so you get really misleading answers, like you might think that E. coli is the number one bacteria in the gut.
And so being able to just like take any sample of interest and use these sequencing methods was really a big step forward.
Chris Bode: Are all the ones that are really important in terms of the gut microbiome, are they all anaerobic?
Peter Turnbaugh: Yeah, so I mean, that’s one of the big challenges. It’s been, you know, it’s been largely solved by the use of these, you know, we work mostly in these chambers that allow you to control the gas. And so, you know, but if you’re trying to, you know, definitely if you’re trying to culture on your benchtop, you’ll lose the vast majority of microbes that are in the gut.
Chris Bode: Yeah, I want to get back to that later. Just, you know, in terms of practical considerations for studying bacterial drug metabolism. And I know there’s so much more about the gut microbiome that doesn’t have anything to do with drug metabolism, but this is kind of, you know.
Peter Turnbaugh: Right.
Chris Bode: It’s a DMPK podcast series. And I have to say, initially I was skeptical whether gut bacteria could really be important in terms of drug metabolism, because I thought they were only in the large intestine. And, you know, most drug absorption takes place in the small intestine for most drugs. But clearly, gut bacteria are important for metabolism of some drugs. Does that mean that there are a lot of gut bacteria in parts of the small intestine?
Peter Turnbaugh: Yeah, that’s a great question. And, you know, this is a question we get all the time. You know, I think if you look at a textbook, kind of the classic chart that’ll show very low levels of colonization in the duodenum, the beginning of the small intestine. And then as you go throughout the small intestine, the density gets higher and higher until you reach a peak in the colon.
I think it’s important, or I guess what I’ve learned getting deeper and deeper into this area is that, you know, those textbook diagrams are a vast oversimplification of a much more complicated picture. And so, you know, in reality, there can be some individuals that have much higher levels in the small intestine. You might have heard of something called SIBO which is small intestinal bowel overgrowth. So you definitely see patients that can reach much higher levels. And then you can also reduce levels in the colon due to certain insults like antibiotics or, you know, diet, other factors.
And there’s also I think the big theme that we’ve learned in general is that the microbiome is incredibly variable between people. And so even though, you know, on average that kind of picture holds up, you know, it’s not true for everyone. I think really what we knew from the old literature, but, you know, I think people have been, you know, I think rightly really fixated on the massive levels of colonization in the colon, which reach like 10^11 cells per gram. So that, is obviously important given how abundant the cells are.
But that doesn’t mean that the stomach and small intestine is sterile. There’s always been microbes in the stomach and small intestine. And, you know, that dates back even to the early studies, and has led to some really big breakthroughs like the study of Helicobacter pylori.
Chris Bode: Yeah, yeah H. pylori in the stomach. So, would you say that the gut bacteria are important for drug metabolism, particularly for drugs that are not well absorbed, so the parent compound is still hanging around lower in the small intestine or is?
Peter Turnbaugh: Yeah, that’s a great point. This is mostly work that we haven’t done with other labs in pharmacology. I think as we’ve gotten deeper and deeper into pharmacology, we have learned that, as you know better than I do, the ADME properties of drugs are a lot more complicated than I imagined. And some drugs actually stay within the GI tract very long. We’re doing a lot of experiments right now on Rifaximin, which is a drug considered to be basically not absorbed at all. And so the whole drug gets excreted in the stool absent a tiny fraction that gets absorbed.
And so that can definitely happen. And that gives you potentially much higher doses of drug than what’s delivered in the colon as the lumen concentrated due to loss of water. And then in addition to that, a lot of drugs are released back into the gut after they’re absorbed. And so that’s something we think a lot about is kind of the classical drugs that are subject to enterohepatic circulation but also more novel kind of suggestions in the literature that drugs might, more drugs might kind of go through that route than we typically think about.
Chris Bode: Well, I know that some, there are issues with cancer drugs, for example, Irinotecan is one that it is glucuronidated in the liver, it’s excreted into the bile, it gets into the gut, and then bacteria that will deglucuronidate it, basically that’s fuel for the bacteria. But then you’ve got the parent compound that’s really toxic. And I know Matt Redenbow had done some work on, which I thought was fascinating to address that not with antibiotic to kill the bacteria, but with a selective inhibitor of the bacterial enzyme, the beta glucuronidase, but are there a lot? I’m sure there are more examples like that.
Peter Turnbaugh: Yeah, yeah, definitely. You know, one related to that, we’re very interested in fluoropyrimidines, which are a common drug used for cancer. They’re not, you know, those are not drugs that we typically think of as going through glucuronidation and the same release back into the gut, but they do have a lot of GI toxicities.
And you know, we have a preprint out, if you’re interested, kind of showing that changing the microbiome can affect the intestinal levels of the drug due to an enzyme that we showed inactivates 5-fluorouracil, and we think that’s important for controlling the local toxicity of 5-FU. And interestingly, 5-FU and irinotecan are given together in a cocktail. You can potentially have bacterial effects on multiple components of these cocktails that are given to cancer patients.
Chris Bode: Oh yeah, okay, very interesting. So probably like with Irinotecan, some patients have the bacteria that… I don’t know, maybe everybody has the bacteria that would de-glucuronidate that if it appears, if it’s excreted into the gut. But probably like different people do it to different extents and maybe even the same patient at different points in their life would probably handle it differently. Is that, is that fair to say?
Peter Turnbaugh: Yeah, that’s a great point. And definitely like kind of the direction we want to go, you know, in kind of our long term view of this area is trying to take more of a pharmacogenomic or pharmacogenetic approach. I think that we’re still at the point in the field of just kind of identifying what the main players are. It’s really helpful in the case of you mentioned the GUS enzyme or beta glucuronidase, or in the case of 5-FU preTA which is a bacterial version of dihydropyrimidine dehydrogenase, just finding those enzymes we think is really important because what we and others have found is that they don’t typically kind of fall within a single group of bacteria.
And so if you just know the names of the microbes, it’s really hard to predict whether or not they will metabolize the drug. Whereas if you know the enzyme, that is a better signature. And so, and you know, this is a bigger issue than human genetics where, you know, we all basically have the same genes. We just have different versions of them. In bacteria, you know, you can have two closely related microbes where one doesn’t have the gene at all and the other one does. And so, it’s very hard to kind of infer what bacteria are doing just from their name.
Chris Bode: Yeah, okay. And then I guess, I mean, in order to do that, to search for a, I mean in different bacteria, is there a sequence that you can look for and know, okay, this is going to be part of the gene for this enzyme, or are they different enough across bacteria that you couldn’t even look for that?
Peter Turnbaugh: Yeah. Now once you kind of find a gene of interest in a single organism, then what we love to do is, you know, go mining through all the genomes that are out there. And, you know, that’s really interesting. We often find, you know, and I’ll give you, go back to GUS as an example that Matt’s lab has really nicely worked out, you know, that those genes are found in many different microbes. And just to add like another layer of complexity, what he’s really nicely shown is that they actually, even though they’re similar to each other, when you look at the gene sequence, the proteins have different folds.
And that seems to matter for the types of compounds that they can de-glucuronidate. And so even though it’s the same kind of host modification of chemicals, then he thinks of these different GUS enzymes can accept different types of substrates. And so, one might be really good at, like the enzyme from E. coli is really good at handling the irinotecan metabolite, whereas others might better suited at acting on glucuronated compounds and the food that we consume.
And so, I guess the other possibility is that you can find enzymes that are genes that are similar, but then one of them may not even have the function at all. It’s a lot of kind of trying to figure out what the key parts of the gene are that have to be conserved to predict the function.
Chris Bode: Yeah, and I guess if they’re that different, then that also there’s also the possibility of developing really selective inhibitors, if you want to hit one but not another one.
Peter Turnbaugh: Yeah, yeah. Like, it’d be great if you could keep most of your GUS activity, but only remove the activity against the drug.
Chris Bode: So if you block that enzyme, is that going to kill the bacteria or will they find another food source? They won’t have to, like, clip a glucuronide off of a….
Peter Turnbaugh: Yeah, that’s a great question. And yeah, that’s something that I’m really interested in working on more. I think in general, we don’t have a great sense for why bacteria are metabolizing drugs. You know, in the example of GUS, the speculation is that they’re using it to scavenge carbon you know, there’s a lot of kind of classical microbiology experiments you could do to tease out how important that enzyme is, what the natural role of it is, you know, when the drug is absent. And a lot of that still hasn’t been done.
So, yeah, I think that’s definitely a really interesting area because it would be useful to know kind of why microbes are doing this, because then you could potentially manipulate it either to kind of drive more or less metabolism, but we don’t know a lot about kind of the natural function of these enzymes.
Chris Bode: Yeah, I mean, not even so much the metabolism, but just to kind of modify the composition of the microbiome so that people will be healthier in general.
Peter Turnbaugh: Right, right. Yeah, and even related to the inhibition, it really makes a difference if the enzyme is critical to the growth of the microbe, because then the inhibitor becomes antibiotic.
Chris Bode: Yeah, that’s what I was thinking.
Peter Turnbaugh: But, you know, the kind of hope is that you can find targets that aren’t that for the fitness of the microbe, and that would be the ideal drug target. But the jury is still out in terms of what the kind of normal importance of these genes is.
I can tell you, at least when I was training in grad school, the thinking was that bacteria were really stingy. And so we typically think that bacteria don’t carry around genes that they don’t need. I think in reality it’s a lot more complicated. There’s a lot of, they may be messier than we appreciate.
Chris Bode: There’s still a place for the classic microbiology studies. It’s not all genomics.
Peter Turnbaugh: Yea, I don’t know if you saw. There was just a commentary out. I think it was from Andreas Balmler at Davis, our colleague at UC and talking about how we need more microbiologists in the microbiome field, and I totally agree with that sentiment. I say it all the time to people. A lot of people are similar to me in that they got into this because they care about genomics. And so we have amazing genomics in the microbiome field. Really elegant and impressive work that a lot of different labs are doing, but I think we’ve failed to kind of attract a lot of the classical microbiology people. And so that’s kind of the niche that our lab has been trying to fill is to do, you know, do some of these, like, microbiology experiments either in mice or in vitro.
Chris Bode: When I first met you, I think it hadn’t been too long since you published work on digoxin and the metabolism by gut bacteria. Can you talk about that? Because digoxin is a drug that’s not metabolized by humans.
Peter Turnbaugh: Yeah, we got to digoxin as similar to many of the other drugs we work on. It’s largely by kind of digging through the literature. So then at the time I was setting up high-throughput screens, right after I started my lab, I was working in the Harvard Center for Systems Biology as part of this Bauer Fellows program. And the goal was to try to get a better appreciation for which drugs are metabolized. But, you know, we really lacked a lot of this is even like before the Redinbo paper, there weren’t a lot of positive controls that we could draw from.
And you know, digoxin, as you might know, is actually in the textbook, if you read Goodman and Gilman’s Bible of Pharmacology. They mentioned that bacteria metabolized digoxin, which I thought was so exciting.
Chris Bode: I’ve forgotten that I missed that.
Peter Turnbaugh: The school of pharmacy at UCSF teaches digoxin being metabolized by bacteria. So, you know, like, we’ve made it into the legit pharmacology literature. But, you know, this is all before my time. It was, you know, the drug started. A lot of these key papers came out in the 80s, kind of showing that bacteria are able to make the main metabolite dihydrodigoxin. They kind of hit a wall because they didn’t, weren’t able to really link differences in the microbes they could culture with the levels of the drug.
And so at the time they were doing this by plating out stool samples and counting colonies and they were measuring the levels of these microbes that at the time were called Eubacterium lenta, and we now call them Agrathella lenta. And there was really no correlation at all with drug metabolism. And so as an example of, we know this microbe can do this activity in vitro, but it doesn’t predict anything in a patient. And so that’s kind of where it stopped.
And were fortunate enough to kind of figure out some of the microbiology. So I think the main goal starting out was just to find the enzyme. And the challenge was that there’s no genetic tools or warrants at the time in E. lenta. And so were able to confirm that digoxin is metabolized, but you know, how then do you find the gene? And I can remember it at the, I did an experiment that I thought was really obvious. Most of the people I explained it to at the time thought it would fail, you know, but I think it’s a really useful approach that we and others have used a lot since then.
We basically just gave the drug to the bacteria and then looked at what genes turned on by RNA-seq. And you know, at the time I was doing a lot of RNA sequencing. So I just kind of like, you know, I assumed that if you added a substrate you’re going to see more of the, whatever the enzyme is. But you know, you can imagine that would fail for a lot of different reasons. Like maybe the gene is always expressed or you know, like not under the control of the substrate.
Chris Bode: Yeah, it wasn’t obvious to me that that would work.
Peter Turnbaugh: I think there were, you know, GUS maybe is an example of something that the bacteria is already expressing. I don’t know if I don’t think they need the substrate to turn on the gene. But yeah, for Digoxin it worked beautifully. It’s still one of my favorite RNA-seq experiments because there’s basically no gene. There’s only one group of genes. This two gene operon that came on and it was like 10,000 fold upregulated. These genes are basically unexpressed. They’re completely off in the absence of the drug. And then when you have the drug, they light up. And it was really specific.
We didn’t see any other genes affected by digoxin, you know, to any kind of clear level. And so that really allowed us to like zoom in on those two genes immediately. And then later on, working with a really nice talented graduate student in Emily Balskus’s lab, were able to purify one of the proteins and show that it’s actually the enzyme that acts on digoxin.
Chris Bode: Yeah, that’s fantastic. Not only did it work, but was, you probably couldn’t have hoped for a better result.
Peter Turnbaugh: Yeah, yeah, we’ve still been chasing that kind of clarity of that original experiment, you know. For other compounds, it’s been, you know, sometimes it works as advertised. But, oftentimes, you know, the big issue we run into is a lot of drugs are, have off target antimicrobial effects. And so if you are working in a drug like 5-fluorouracil or methotrexate. You know, these are drugs that target conserved pathways that bacteria also need. And so if you do RNA sequencing, we see all sorts of changes that are, you know, not related at all to the metabolism of the drug.
Chris Bode: Okay, gotcha. But you know, the nice thing about digoxin is it really does not have antimicrobial activity that we’ve seen.
Peter Turnbaugh: Okay, and like that’s an important one because it’s a narrow therapeutic index drug. So, you know, if you know, you give a dose to somebody that doesn’t have that bacteria, they might be overdosed and you give it to somebody who does. They might not get the benefit at all because it’s all chewed up by the, by the bacteria, right?
Chris Bode: Yeah, exactly. That was definitely one of the reasons were excited to work on that and also kind of what drove us into cancer. A lot of chemotherapy drugs have similar narrow therapeutic windows. That’s definitely a motivator for drugs where dosing really does matter.
Peter Turnbaugh: You know, when we organized our 2000, I guess it was 16 symposium, one of the speakers was Julia Cui, who’s at the University of Washington. She had shown some data that it was convincing to me that it’s not just her, but she was talking about CYP3A the host CYP3A, that it’s induced by small molecule metabolites produced by gut bacteria. And to me, that was like, that’s completely consistent with the fact that the activity of that enzyme in different people varies a lot. And just one., and to me it makes sense that, oh, it’s induced to different extents in different people. But is that, do you know, I mean, you’re not really a drug metabolism researcher, I understand, but do you know if that is kind of accepted now? It’s accepted by me, but I don’t know if it’s accepted in general.
Chris Bode: Yeah, I don’t know in terms of like the DMPK field, or industry, but yeah, I can definitely tell you in the literature, Julie’s work and you know, even like going way back to, there’s a seminal paper that Laura Hooper did when she was in the Gordon lab where, you know, at the time they were using microarrays. But, you know, it’s one of the first papers where they looked at host gene expression in the intestine in mice that were sterile versus mice colonized with specific bacteria.
And, you know, she found a lot of genes relevant to drug absorption, including P-glycoprotein or even the ABCB1A gene was actually one of the most differentially expressed genes in response to colonization. And so, yeah, I think there’s a growing literature kind of showing that the expression level of these genes changes in response to colonization. But the question, I think is how does that relate to function? And that’s turned out to be a little bit more complicated.
We’ve done a little bit of work on CYP3A, but have mainly focused on P-glycoprotein. And the surprising thing we found is that even though in our hands and in prior literature, colonization can induce the expression of PGP at the RNA level, we actually have found a microbe which turns out to be the same microbe that metabolizes digoxin that inhibit the activity of the transporter. And so we think actually the upregulation and in response to that microbe is kind of a compensatory response because the protein is getting blocked. And so I think it’s really important to kind of do the next, you know, ask whether or not the protein levels are the same and then also, like, is the protein functioning better or worse?
But, yeah, I think the general idea that the differences on the microbiome can control how our body handles drugs is really, you know, that’s a really wide open area that I’m really excited to see more people working.
Peter Turnbaugh: And then like, the host produces bile acids and those may, some of those may be required, I think, by gut bacteria, some of them may be toxic to certain gut bacteria. So it works both ways. There’s a lot of communication in both directions.
Chris Bode: Yeah, Well, I think we like to think that we control the microbiome, but I guess I’m convinced it’s mostly the opposite. But yeah, you know, we definitely have an immune system and, we have other kind of levers we can pull to try to maintain some sort of order.
Peter Turnbaugh: Oh, man, we could talk for hours about the effect of the gut microbiome on the immune system and the gut brain axis. But we’ll try to stick to DMPK.
Chris Bode: The most important area of microbiome. There you go.
Peter Turnbaugh: So, you know, one class of drugs that is just really huge right now are the GLP1 drugs, semaglutide and drugs like that. Do those have, I mean, do we know if there’s any interaction between those and the gut microbiome if they affect the gut microbiome or vice versa.
Chris Bode: Yeah, it’s definitely an area I’m really interested in. We didn’t really get into this, but then the thing I actually worked on when I was in grad school was diet and obesity with, you know, and how it relates to the microbiome. And then when I kind of started my own lab, I switched into mainly thinking about pharmacology, you know, but we’ve continued to think about diet and metabolic disease throughout the years. And the frustrating thing has been there’s never been a drug for, you know, a drug that really worked well for obesity.
And so, you know, I’m really excited now that these drugs seem to be working and that allows us to, you know, kind of think about pharmacology and with respect to diet and obesity. There’s not a lot out there, and surprisingly to me, given that there’s been really massive human studies of these drugs. And so, you know, nowadays I would expect that most of these clinical trials would have some sampling to look at the microbiome, but that really hasn’t been published.
There’s been some smaller studies in mice that have found something I think somewhat surprising, which is that these drugs really do seem to change the microbiome, and that’s not really well understood why that would be the case. And then in addition to that, there is a decent literature that kind of predates these drugs, including really nice work from Frederick Backhed, another former Gordon lab member, looking at how the microbiome controls GLP release. That’s, I think, a really interesting and related area, how the baseline levels of GLP1 and other incretins are controlled by microbes and that might kind of set the baseline for how these drugs are acting on people.
Peter Turnbaugh: Yeah, or maybe some people don’t need them, they just need a healthier gut microbiome.
Chris Bode: Yeah, could you get the effects without the drug? I think it’ll be interesting to see kind of how far the microbes can push it, you know, as opposed to these, you know, agonists that are incredibly potent.
Peter Turnbaugh: Okay, yeah, I’ll have to keep my eye on that. That’s interesting that it hasn’t really been published yet, but I’ll, I guess people must, somebody must be working on that.
Chris Bode: Yeah, maybe. If you know anyone at Novo or the other companies, I’m happy to talk to them. Yeah, I guess more generally, I’m always surprised when companies invest so much at this point. Microbiome sequencing is pretty cheap, and so it’s weird to me that you would go to the trouble of doing a large scale phase three trial without throwing in some microbiome sequencing, even on a subset of the cohort.
Peter Turnbaugh: Yeah, especially with these drugs. I know that because there are side effects, you know, GI side effects, for sure.
Chris Bode: Yeah, like, even at a minimum, you might want to look at the microbiome of the people that have GI symptoms. You know, maybe it’s hard to know. Maybe it is happening and we just haven’t heard about it, but don’t have any kind of insider knowledge.
Peter Turnbaugh: Right, or the company got answer they didn’t really want.
Chris Bode: Right, the answer they were hoping for.
Peter Turnbaugh: Yeah, that really changed the microbiome.
Chris Bode: Yeah, I mean, I related to that, I think that the hard part, I think for industry and you know, as well as for academics is like, we don’t. It’s hard to say what the consequence of a change in the microbiome is. And so, you know, it’s very easy to do the study to measure changes in the microbiome in response to a drug. But then, you know, the question you might get from, you know, the FDA or from other folks at the company is like, you know, is this change good or bad? And that’s something that’s actually quite difficult to answer. And so, you know, I think maybe part of the hesitancy is, you know, they don’t really know what they would do with the data if it were collected.
Peter Turnbaugh: Yeah, because it’s not so much about bacteria that are good or bad. It’s more about the balance of all of them.
Chris Bode: Right, yeah. And I think really more specifically what you care or what I would care about is like, you know, are those changes, you know, do they matter for the direct mechanism of action of the drug or do they matter for the side effects of the drug? That’s something that you really have to do more experiments to figure out. We don’t really have a way to just look at the list of microbes and tell you the answer.
Chris Bode: I wanted to ask you something that maybe some of our listeners are wondering about, but I’m definitely wondering about. Is it really feasible to directly study bacterial drug metabolism in vitro? You mentioned, I mean, we talked about the fact that the bacteria we’d be interested in are anaerobic and you said there are chambers where you can work with them. But we’ll take microsomes or liver microsomes out of the freezer. And do a drug metabolism, study metabolic stability and measure metabolites. Is it that simple with gut bacteria too?
Peter Turnbaugh: Yeah, I mean, I think the good news is that, you know, with some exceptions, most of the microbes can be cultured. You know, so then it’s just mainly logistics and like, do you have the right equipment, you know, do you have personnel that are comfortable growing anaerobes? But, you know, it’s not really rocket science and that, you know, it’s mainly, you know, using special chambers that keep them away from oxygen.
I think maybe the question related question that you’re alluding to is whether or not the rate or extent of metabolism in vitro, can be extrapolated to an in vivo PK experiment. You know, that is a wide open question, and you know, it’s something we really don’t have a great handle on. You know, and I definitely think what we’ve learned over the last few 10 to 15 years has kind of complicated how you would imagine going about that because in the simplest model is that the only thing happening is this direct metabolism.
But as we’ve been discussing, you know, the same microbe that metabolizes digoxin can also control the transporter P-glycoprotein that influences absorption of digoxin. And while the transporter effect seems to be conserved among all of E. lenta species, only certain strains are able to metabolize the drug. And so you kind of have to think about the specific strain of microbe that’s in the community and also these more indirect ways that it could be influencing the drug. And so I think that suggests that it’s not going to be a one to one where you won’t be able to say this microbe was active in vitro. I would expect exactly this amount of change in pk.
Chris Bode: Yeah, not even, I wouldn’t be thinking, trying to scale it quantitatively, but more just like, you know, I have this new drug candidate, could it be metabolized by gut bacteria and you know, what metabolites might I look for? And I mean, can you?
Peter Turnbaugh: Yeah, I think to that, you know, that question is much more answerable. So then, you know, the main challenge there, I think is just the diversity of microbes, you know, as opposed to the liver where there’s a finite number of enzymes, there’s always another microbe to test. And so, yeah, I think we in general try to avoid saying that the microbiome can’t do something because it might just be we haven’t looked at the right microbe. But you can definitely, you know, you can within reason, screen 100 microbes or you know, a set of complex microbial communities and then see whether or not you find anything or not, and then if you find metabolites, then you know that those can be produced by microbes.
Chris Bode: So I feel like I’m looking at this so naively, but would it be useful to start with fecal samples? Fecal homogenates. Are those all dead bacteria? Are there enough live ones that you would get an idea of what might be happening in the gut?
Peter Turnbaugh: Yeah, no, it’s not naive. Like, yeah, this is, we call it, well there’s a lot of different words for this but we like to call it ex vivo incubations which you know, sounds cleaner. People do other stuff. So in our lab, the way that we do ex vivo incubations, you’re just diluting the cells into media and then incubating. And so that I think is a useful way to ask whether or not this complex community that you just pulled out of a human is doing something to a drug. The challenge is that as you culture, I think we talked about way at the beginning, the culturing process is going to change the community. And so we currently don’t have a way to kind of fix the community in place, maintain the same structure that it was in the patient.
And so that leads to all sorts of confusing results where you have expansions of different microbes that you don’t want in your sample. And so one solution for that, I think they call it fecalase. But you can, you know, you can just take a stool sample and lyse all the cells so that what’s left is just a big collection of enzymes and then test that for activity. And so, and then there’s different medias that people use and you know, there’s various options kind of within those two major categories.
Chris Bode: And like how rigorous do you have to be about excluding oxygen, from the initial sample and all the way through processing and shipping and storage?
Peter Turnbaugh: Yeah, exclude oxygen. It’s a good question. We definitely think it’s important to the, and yeah, this maybe won’t make you want to do this approach. I mean one of the nice things about these like DNA based profiling methods is that they tend to be less sensitive to that because the DNA is really stable. Yeah, right. And so, you know, if you’re doing sequencing, you don’t have to worry about that as much.
We, a while ago now had a paper where were developing methods of kind of measuring live, live and dead bacteria within a sample. And for those metrics, it was really important that the sample got straight into the anaerobic conditions. So even, you know, within hours, you can start to see the a lot of the cells dying, and, you know, that makes sense given that the most abundant members of the human gut microbiota are also very oxygen sensitive.
So, yeah, I mean, I think that’s probably one of the reasons why even if you’re culturing under anaerobic conditions on petri dishes, we often find, you know, the usual suspects like E. Coli and enterococcus and, you know, all of the microbes that can handle oxygen.
Chris Bode: Yeah, you unintentionally select for those.
Peter Turnbaugh: Just like in, you know, it’s really hard if you’re thinking about a clinical study because then you’re, you know, you’re typically sampling patients that are not even near the lab. And so as opposed to collecting a local sample or in mice, we have the advantage of being able to actually, you can even bring the organ into the anaerobic conditions.
Chris Bode: Okay, Peter, this has been a fascinating and insightful discussion, and I thank you so much for your time. Really appreciate it.
Peter Turnbaugh: Sure, yeah, no problem. It was really fun.
Chris Bode: So thank you all for listening to this episode of Pharmaron’s DMPK Insights podcast series. We would like to remind you that our DMPK webinar series is also available on demand, covering a variety of key questions related to DMPK science and drug discovery and development. And stay in touch for more episodes in the Pharmaron DMPK Insights podcast series. Thank you and goodbye for now.
