The Peter Attia Drive - #283 ‒ Gut health & the microbiome： improving and maintaining the microbiome, probiotics, prebiotics, innovative treatments, and more

00:00:00.000 Hey, everyone. Welcome to the drive podcast. I'm your host, Peter Atiyah. This podcast,

00:00:16.580 my website, and my weekly newsletter all focus on the goal of translating the science of longevity

00:00:21.580 into something accessible for everyone. Our goal is to provide the best content in health and

00:00:26.780 wellness, and we've established a great team of analysts to make this happen. It is extremely

00:00:31.720 important to me to provide all of this content without relying on paid ads to do this. Our work

00:00:37.000 is made entirely possible by our members. And in return, we offer exclusive member only content

00:00:42.760 and benefits above and beyond what is available for free. If you want to take your knowledge of

00:00:47.980 this space to the next level, it's our goal to ensure members get back much more than the price

00:00:53.260 of the subscription. If you want to learn more about the benefits of our premium membership,

00:00:58.060 head over to peteratiyahmd.com forward slash subscribe. My guest this week is Colleen Cutcliffe.

00:01:06.940 Colleen received her bachelor's degree in biochemistry from Wellesley College and a PhD in biochemistry

00:01:11.920 and molecular biology from Johns Hopkins University. She then completed postdoctoral training at

00:01:16.900 Northwestern Children's Memorial Hospital and subsequently began working as a scientist in

00:01:21.340 the pharmaceutical industry. She's currently the CEO and co-founder of Pendulum Therapeutics,

00:01:26.280 a startup that is working to develop treatments for a variety of diseases by targeting the microbiome.

00:01:32.200 Although given that Pendulum has been around for about a decade, it seems a little odd to refer to

00:01:36.700 them still as a startup. I wanted to have Colleen on the podcast because quite frankly, she was the

00:01:43.200 first person I met and had deep discussions about the microbiome where I really felt like the person

00:01:49.420 knew what they were talking about. Now, I don't say that to be disparaging of the many other people

00:01:54.100 who have a lot to say about the microbiome, but my mind works in a particular way and I guess I've just

00:02:00.200 never had that connection with a person where when I ask questions, they seem to have answers that make

00:02:07.620 sense to me. So much of the work that I've seen around the microbiome has been interesting, but it's been

00:02:14.320 very difficult for me to understand how one could operationalize and make real causal effect from the

00:02:22.060 science that is being presented. And so over the course of many months, Colleen and I had a number

00:02:28.980 of discussions. I became quite interested in some of the products that her company sold and I even began

00:02:35.100 to see some of the benefits in my own blood work, something that I was incredibly skeptical of at the

00:02:41.080 outset. In fact, we realized after our first meeting that many people had been telling me about some of

00:02:47.940 these products over the previous four or five years and I had been generally quite dismissive without

00:02:53.620 looking more deeply at the data behind them. In this conversation, we really dive into all things

00:03:00.140 related to the microbiome. We talk about what it is and how it changes over time and how you can measure

00:03:05.100 it if that is something that is important and we'll talk about the importance or lack thereof. We talk about

00:03:10.540 probiotics versus prebiotics versus postbiotics and how much bacteria actually make it into various

00:03:16.540 products that one can buy. We speak about fecal transplants and the importance of fiber as well

00:03:22.020 as artificial sweeteners and antibiotics as it relates to the gut microbiome. We then speak specifically

00:03:28.100 about Ackermansia and Colleen's work at Pendulum around creating products that focus on not just

00:03:33.320 acromansia but creating substrate for it and other tools that will enable the gut to be fed in the best

00:03:41.760 possible way to improve metabolic health. This is a really interesting discussion on many levels and

00:03:48.040 again, I just want to stress how skeptical I have been of this topic for at least a decade. In fact,

00:03:54.200 there is probably no area of medicine that I have been more skeptical of than this one and of course,

00:04:00.520 with some of the high profile failures of companies in this space, I have somewhat smugly felt vindicated

00:04:07.100 in my skepticism. But I must say that I feel like that is changing and I think that the work that

00:04:13.680 Colleen and her team have shared with me along with work that has been shared by other people

00:04:18.180 have began to make me think that there may indeed be something to this gut biome. In other words,

00:04:24.320 it's not that I don't think that the gut biome matters, but the question is, can we change it

00:04:30.380 using tools other than our nutrition? And to me, that's one of the most important questions we dive

00:04:36.960 into in this discussion. So without further delay, please enjoy my conversation with Colleen Cutcliffe.

00:04:48.640 Hey Colleen, thank you so much for making the trip out to Austin to sit down. It's always more fun to do

00:04:53.480 these in person. Thanks for having me. We're going to take advantage of being in Austin.

00:04:57.680 Yeah, I know. This is a topic that we've received a lot of interest on and people have always wanted

00:05:03.780 to go deeper on this topic. And my reluctance to do so has been in part driven by a couple of things.

00:05:11.160 One is just my general lack of clarity around finding people that can speak about the topic with

00:05:17.400 some degree of rigor. And then secondly, it's just, it's a market of products that seems so

00:05:23.540 sketchy. And so when you and I met, God, it's probably been six months ago and connected,

00:05:29.580 we went for this walk and I was thinking, ah, this is like the most I've learned about this and

00:05:33.360 the rest is history. And so here we are. Maybe give folks a bit of a sense of your background

00:05:37.680 scientifically. So I think when we met, we figured out pretty quickly, we had both been at Hopkins

00:05:42.580 at almost about the same time that you, of course, were doing your PhD there. So what did you do your

00:05:47.780 PhD in again?

00:05:48.880 Yeah, my PhD was in biochemistry and molecular biology. So really thinking about enzymes and

00:05:54.120 pathways and how they all interact with each other. And then I did a postdoc at Northwestern.

00:05:59.880 We were looking for diagnostic markers for pediatric Wilms tumors. And then I moved out to the Bay Area

00:06:06.600 and I worked in pharma. We were developing drugs for Parkinson's disease. I sort of followed a pretty

00:06:10.900 traditional path of a scientist into pharmaceutical drug development. Then I did what everybody does

00:06:16.820 in Silicon Valley. I joined a startup company and this was a DNA sequencing instrument company.

00:06:21.360 And we went through rapid growth. We went public. And on the other side of that, I started this

00:06:25.680 company, Pendulum. And it was really premised in the fact that things like probiotics and yogurts

00:06:31.120 have been on the shelves for decades, but there actually hasn't been a new ingredient introduced in

00:06:35.580 over 50 years. And that's because microbiome science wasn't a real science until DNA sequencing

00:06:41.880 technology enabled us to really be able to survey the microbiome and create these metabolic maps and

00:06:47.080 start to approach it like a biochemical problem and a systems biology problem. And all of a sudden,

00:06:52.000 it seemed like, wow, there had been a lot of unlocks in DNA sequencing technologies that would allow this

00:06:57.220 entirely nascent field of science and medicine called the microbiome to produce real products.

00:07:02.800 And so here we are 10 years later in it. We'll definitely spend some time talking about

00:07:09.020 products and the evolution that you've had in that path. But let's talk more broadly just kind

00:07:14.500 of about this field. First of all, how do we define the microbiome? The microbiome is essentially all

00:07:20.200 of the microbes. So the bacteria, the viruses, fungi, yeast that reside in and on us. And they are on our

00:07:26.360 skin, they're in our nasal passages, they're in our lungs, and they're in our guts.

00:07:29.940 Got it. Now, our guts, which run mouth to anus, are outside of our body. So people don't think of

00:07:38.240 it always that way. But of course, they are. What allows the colonization of that? I mean,

00:07:42.620 is that something that is set at birth? Maybe taking a step back, when a child is in his mother's or her

00:07:48.600 mother's womb, there's amniotic fluid that's flowing through that spot. Is that a sterile area?

00:07:54.760 Well, interestingly, for a long time, as you know, we've all believed that was an entirely

00:07:59.380 sterile environment, and there were no microbes there at all. And some recent studies have started

00:08:04.140 to elucidate that there are some strains, but it's very minimal. When we think about the gut

00:08:10.080 microbiome of an adult versus somebody who's in the womb, I mean, it's incredibly much more diverse

00:08:15.840 once you become an adult. And in fact, really, the primary initial seeding of the microbiome is

00:08:21.180 through the delivery in the vaginal canal. And so we're going to get gross for a second,

00:08:25.920 but literally, as you're being delivered, you are consuming fecal matter that is in the vaginal

00:08:31.140 canal. And that's your first seeding of microbes. And infants have a very small diversity of microbes

00:08:37.240 that are really tied to mother's breast milk. And then as you start to eat foods, and as you start

00:08:43.460 to get exposure to other environments, then the diversity of your microbiome starts to really

00:08:47.040 grow and flourish. And then at some point on the aging process, the opposite starts happening.

00:08:52.240 You start to become less diverse in your microbiome as we age. So you start out almost

00:08:56.900 like a blank slate, you get a lot more diverse. And then as we age, you start to lose that diversity

00:09:01.440 and therefore some key functions in the microbiome.

00:09:04.460 When is peak diversity approximately? What decade of life?

00:09:07.320 Obviously, it varies from person to person. But if you can remember a time where you could eat or drink

00:09:10.980 whatever you wanted to, and you didn't have to worry about it, that would probably be the time.

00:09:14.620 Oh, so that's actually quite young.

00:09:15.980 Teenagers.

00:09:16.940 As a teenager, exactly. Yeah. When I could indeed eat a bowl of cereal for every meal

00:09:21.380 with no consequence. Yep.

00:09:23.060 By bowl, I mean a bowl the size of my head. So it's a box per bowl per meal. Okay, so we have a

00:09:28.140 relatively early peak in life for diversity. You hear all of these sort of bumper stickers,

00:09:33.240 slogans about the gut biome. Oh, it outnumbers us 10 to 1. Is that true?

00:09:37.540 I think those numbers have definitely come into question. I mean, they're nice to give people a

00:09:41.520 framework for the fact that you have a ton of microbes in you. And I think that's the

00:09:45.500 important part is that they are, whether they outnumber you 10 to 1 or 2 to 1, I think is

00:09:49.740 relatively probably not that important. But what is important is that they make up a huge portion

00:09:56.580 of your body, mass as well as functions. And so it's an important key part of your health.

00:10:01.820 Not even 1 to 1. Yeah, even 1 to 1 is huge.

00:10:04.380 So the idea is there are many cells that are not you between your mouth and your anus as there are

00:10:11.100 you. Exactly.

00:10:12.460 Now, obviously, just to have someone wrap their head around that, we're made up 70% of water.

00:10:18.640 So most of our mass is water, not the cells minus the water. Are these largely anhydrous cells? Like,

00:10:24.940 how do they weigh so little relative to the rest of us?

00:10:28.380 Oh, man, I don't know the water content of bacteria, but maybe I think about it a little

00:10:33.100 bit differently, which is more about the biochemist. Everything's going to come back to that. But it's

00:10:38.080 more about the biochemical functions. What's the output of each of these cells versus the output of

00:10:41.920 our cells? And I think when you look at it that way, these are real workhorses. So there's definitely

00:10:46.660 redundancy among bacterial cells, but each of them is having multiple functions and multiple outputs. And so

00:10:52.640 when you think about it at a cellular level, I would think more about what are the things being

00:10:56.980 produced by the cell? And bacteria tend to secrete a lot of things that they're producing, unlike the

00:11:01.300 cells in our body. And so there's a lot of function that's associated with the microbiome that's super

00:11:06.020 important. So I remember in my first and only biology course as a kid, because I didn't take bio

00:11:11.380 again until I decided to go to med school in ninth grade or whatever, we learned about prokaryotes and

00:11:17.060 eukaryotes. And I still remember to this day, the joke of our teacher, Mr. Jefferson, he said,

00:11:21.620 do you know what a eukaryote is? And everyone was like, no. And he goes, it's a portaging term.

00:11:26.840 You know, portaging when you carry canoes back and forth between rivers. Anyway, no,

00:11:31.200 it's a horrible joke, but I still remember it. So I'm going to have to cut this part out of the

00:11:34.940 podcast. It's so bad. Where are we going? Where I'm going with this is, can you explain to people

00:11:41.360 listening what the difference is between our cells and bacterial cells? Because there are some

00:11:46.020 fundamental differences between these things called prokaryotes and eukaryotes.

00:11:50.800 Yeah. I mean, I think that, first of all, every cell in our body needs the other cells and the

00:11:56.380 organs and the whole system in order to be able to survive and do their job. Whereas bacteria,

00:12:01.140 they don't need anybody else. And so every bacterial cell, every unit is its own living thing that can

00:12:06.640 replicate, perform functions, lose functions, be genetically modified and all of that. And so

00:12:12.240 it's sort of its own entity, its own living organism. Every cell is a living organism.

00:12:17.260 And then they divide really, really rapidly. Some of them as quickly as 10 to 15 minutes,

00:12:23.440 you're dividing. And so there's this other component, which is that some of these bacterial

00:12:28.140 strains, because they divide so quickly and because they're also under the pressures of evolutionary

00:12:32.320 processes is that they can evolve super quickly. And so we as humans, we have a long evolutionary

00:12:39.180 timeline because it comes in the form of you make a kid who makes a kid who makes a kid. Now,

00:12:43.940 imagine if that was happening every 10 minutes. You can evolve really, really rapidly. And so

00:12:48.260 that's part of the antibiotic crisis out there, which is to say that these things can become

00:12:53.380 resistant to antibiotics because of this division time.

00:12:56.400 You alluded to this already, but they have the capacity to secrete things significantly. We think of

00:13:02.440 bacteria and we hear it as a bad term. We think of bacteria is a bad thing. And there clearly are some

00:13:08.020 bad bacteria. But would I be oversimplifying if I said that most bacteria enjoy a kind of

00:13:16.100 complementary relationship with us as their host? Is that fair in terms of flora, such as the bacteria

00:13:21.560 on our skin, in our nasal passages, in our gut?

00:13:24.520 We've co-evolved with these microbes and these bacteria. And so generally speaking, when you're

00:13:28.980 co-evolving with something, there's some mutual benefit. I sort of cringe when people talk about good

00:13:33.580 bacteria and bad bacteria, although I do it as well, because it's the ecosystem and the context

00:13:38.500 of these bacteria that's actually more relevant. A good bacteria can become a bad bacteria in a

00:13:44.260 certain situation. And likewise, bad bacteria can become beneficial in a different context. And so

00:13:50.300 I think that it's important to know that they're all part of these different pathways and what they're

00:13:54.120 doing together. So for example, Clostridium difficile is something that I think people think is a terrible

00:13:59.920 pathogen and it's so bad for you. And oh my gosh, you better never get it. Almost all of us have

00:14:04.700 Clostridium difficile in our guts, but at the level that it's at and in the context of the ecosystem of

00:14:10.720 our strains, it's not having that kind of really nasty pathogenic impact. In my opinion, there aren't

00:14:17.300 really bad bugs.

00:14:19.060 People hear me rail about good cholesterol and bad cholesterol being meaningless terms. And of course,

00:14:23.380 cholesterol is simply cholesterol. It's where it ends up that can be good or bad. So that's

00:14:27.320 actually a great analogy. While we're on the topic of Clostridium difficile, which we'll come

00:14:30.960 back to in more detail, what is the prevalence of that as a function of total gut biome in a person

00:14:37.700 who's healthy and not otherwise in a pathologic state? Well, this is the convoluting part about

00:14:43.460 the microbiome, which is that if you sequence and do really deep sequencing and even biochemical

00:14:48.320 assays across a person's microbiome, and then you say, all right, now I want to do population studies.

00:14:53.420 What you find is that the difference between people is huge. And so the Human Microbiome

00:15:00.720 Project, in which they kind of looked across 10,000 plus people at all ages and different

00:15:05.660 demographics, really demonstrated that at the strain level, people are pretty different from

00:15:11.340 person to person. When you start to look at the functions, that's where you start to see some

00:15:14.980 redundancy. So it's hard to say for a particular strain, if someone gives you an actual number and

00:15:19.660 they don't give you a range, that's probably not correct.

00:15:23.760 Wow. Interesting. Tell me more about that project. So what were the observations it landed

00:15:29.180 at vis-a-vis how various factors, both modifiable and unmodifiable, either genetic or age and diet

00:15:37.340 being the most obvious modifiable factor, how did that impact the gut biome in these 10,000

00:15:41.320 people? Those differences magnify in the outcomes.

00:15:44.320 That initiative didn't have a longitudinal component to it or a perturbation of the system

00:15:51.420 and looking at before and after. So it really was just a-

00:15:54.240 A one-time observation.

00:15:55.280 One-time observation saying, okay, we just look across a population of people. And this

00:15:59.240 is super early on. So we didn't know anything about the microbiome. We barely knew how to

00:16:02.180 sequence the thing. And so, and even things like, well, what should the sample be? Should

00:16:05.660 it be a scoop of somebody's stool? Should it be the entirety of the stool? Should you do 16S

00:16:10.200 sequencing, which is just a gene that all microbes have? Or should you do the whole genome sequencing?

00:16:15.860 That's going to be a lot more expensive. This is a government-funded project. And so there

00:16:19.860 was a lot of unknowns at that time. So even just getting this information of, if I looked

00:16:24.360 across 10,000 people from skin to gut to vaginal microbiome, what does it look like? That was

00:16:32.220 a huge endeavor. And of course, coming out of that have been a ton of longitudinal studies

00:16:36.360 and studies where people have done actual interventions.

00:16:40.040 Going back to what you said earlier, it's not just bacteria, it's viruses, yeast, fungi.

00:16:44.740 What does the pie chart of that look like in terms of numbers? And then what does the pie

00:16:48.300 chart like that look like in terms of function? I'm guessing the bacteria are doing the majority

00:16:52.380 of the work. I do feel pretty comfortable saying viruses are bad. Is there an exception

00:16:56.580 to that rule? I mean, we certainly can harness viruses to do good things for us in terms of

00:17:01.240 recombinant DNA technology. But if there was not a single virus on this planet afflicting

00:17:06.960 humans, I think we'd be in a better place, right? I'm not aware of good viruses.

00:17:11.220 Well, this field is really nascent. I hesitate to even answer the question of what is the role

00:17:17.420 or the significance of all these different types of microbes. Interestingly, I think we also

00:17:23.500 don't really know what the role of viruses are in the microbiome. But one could imagine that

00:17:28.820 you have functions that are important and that you need to be healthy and that maybe these viruses

00:17:35.460 help accelerate movement of those particular genes from one bacteria to another. So I wouldn't say

00:17:42.460 the jury's totally out of my mind on that. Interesting. Yeah, that's a good point. Do we

00:17:46.580 know anything about how our microbiome compares in complexity to that of other animals?

00:17:53.160 A lot of microbiome studies are done in mouse models. And I mean, having been in pharma,

00:17:58.180 I feel like if you're a mouse and you've got cancer, you're made. We have so many different

00:18:02.340 cures for you. The translation of that into humans is not that great. And the microbiome is even worse

00:18:07.900 because animals eat different foods. Your diet is one of the biggest things that impacts your

00:18:12.500 microbiome. And so now you're saying, what is an animal's microbiome? How does it interact with

00:18:17.480 the host? And then what implications does that have for health? So even one more step removed. So

00:18:21.880 there are a lot of strains that exist from mouse to man. But I think, again, because it's context,

00:18:27.440 it's not totally clear that you can really use these animal models to predict what will happen

00:18:31.980 in a human. Has anybody looked at animals that are not living in captivity and not genetically

00:18:39.100 ridiculous, like the typical black six mouse and things like that? I mean, do we know what our pet

00:18:44.680 dogs look like or what animals in the wild look like? Do they more closely resemble us? Or again,

00:18:50.160 is it purely a function of what they're eating? A function of what they're eating is such a big

00:18:54.720 driving factor. So if you look at your pet dog, I mean, it really does depend on what you're feeding

00:18:58.240 them. Some people feed their dogs pretty much a grain-based diet. Some people are cooking chicken

00:19:02.680 for their dogs every night, you know. So when they look at pets, I think that's also hard because it

00:19:06.760 just ends up being parsed out by what you're feeding them. Just to get back to the reality of

00:19:11.520 these genetically modified mouse models, it's even more extreme in microbiome studies. So they literally

00:19:17.920 make, I call them like the bubble boy of mice. So they try to entirely deplete these mice of

00:19:23.620 microbiomes. So they're taking a mouse that has no microbiome and then they're infusing them with a

00:19:28.520 human microbiome and they're saying, okay, what happens now to the mouse? And so these are germ-free

00:19:33.320 mouse models that are really commonly used to understand the impact of a human microbiome on

00:19:39.100 the health of the animal. This is completely not what would ever happen in real life.

00:19:43.760 So for folks listening, again, who might not think much about bacteria, my vague recollection

00:19:49.540 of microbiology is one line in the sand we draw to divide them is, are they aerobic? Are they

00:19:55.720 anaerobic? Are they gram-positive? Are they gram-negative? Which is simply a staining technique. I guess I

00:20:01.800 should explain that aerobic are bacteria that require oxygen for respiration. Anaerobic are animals that

00:20:09.120 generate ATP without oxygen. And then you have facultative of each where they prefer to do it

00:20:15.820 one way, but they can do it the other way. Any other divisions worth talking about as we explain the

00:20:21.420 types of bacteria? The aerobic versus anaerobic is definitely one of the most important things

00:20:26.420 you think about. You've got bacteria on your skin that definitely loves oxygen. And then what we call

00:20:30.800 the gut microbiome is actually a strictly anaerobic area of the distal colon. So there's literally no

00:20:36.680 oxygen. Those strains can't even grow in the presence of oxygen. One other thing might be

00:20:40.360 localization. So your GI tract and the gut microbiome, there's this so-called gut lining.

00:20:46.240 So the cells, the epithelial cells, where there's this mucin layer, that's an important part of having

00:20:51.780 a well-fortified gut lining. There are strains that live in that mucin layer. And so that's a different

00:20:57.840 type of a strain. They feed off of mucin versus many of the strains that sort of feed off of things like

00:21:04.400 the foods that you eat are prebiotics. How does the characteristic, let's just focus on the bacteria,

00:21:10.460 how does the characteristic of the bacteria change from mouth to anus? Obviously, you're going into

00:21:15.660 less and less oxygen as you go down. So presumably, you don't have pure anaerobes in the mouth. But I

00:21:21.780 know from the little bit I remember about working in an ER, whenever somebody received a bite, you would

00:21:27.160 think, ah, how bad can a bite be? We were really conditioned to remember that those are some of the

00:21:31.620 dirtiest wounds a human can have. No less dirty than a feces-soiled wound. So even at the proximate

00:21:41.180 end of that bacterial lining, these are really frightening bacteria. Can you tell me anything

00:21:47.700 about how the bacteria change as you progress along the length of that? By the way, that's a very long

00:21:52.560 tube. It's not just measuring here to here. The listener would have to understand how tortuous

00:21:57.160 the small intestine and the colon can be. The primary thing that changes is this anaerobic

00:22:01.660 part. So obviously, your mouth, there's a lot of oxygen exposure. And then, as we said, when you get

00:22:05.840 to the distal colon, there's no oxygen there. So all along that path from the mouth to the stomach,

00:22:10.060 I think, is where you have a reasonable amount of oxygen exposure. Once something gets through the

00:22:15.300 stomach, there's sort of this in-between the strict anaerobic and the aerobic area. And a lot of the

00:22:21.740 lactobacillus and bifidobacterium strains that are on the labels of probiotics out there today,

00:22:26.080 they kind of reside in that small intestine area. And then you get to the recesses of the gut

00:22:30.740 microbiome, which is where all the action happens. So after your stomach has broken down foods and

00:22:34.740 they make their way to the distal colon, that's where an incredible amount of metabolism is

00:22:39.060 happening. And there, there's no oxygen. But yeah, the mouth microbiome is a really interesting

00:22:44.100 opportunity because what's happening in the mouth, the two ends that you mentioned, mouth and anus,

00:22:48.140 what's happening in the mouth and what's coming out the other side, give you an indication of

00:22:51.800 what's happening in the middle. And I'll just tell you a funny story. Early on when I started

00:22:55.420 this company, I met a guy who was really interested in hyena mouth microbiomes. The reason is because

00:23:01.180 the hyena is one of these bizarre animals that can eat a carcass of an animal after it's been dead for

00:23:08.260 extended periods of time, you know, a week, two weeks. No other animal will go near an animal that's

00:23:12.440 been dead for that amount of time because of the bacterial overload. This is just going to kill you.

00:23:16.440 And so it's so-called rotten meat that these hyenas are able to eat. And he had this big question of

00:23:20.840 why are they able to do that? And it turns out if you look at the microbiome of a hyena,

00:23:25.240 they make an incredible amount of antibiotics. And so he had this whole theory that if you could

00:23:32.040 understand what antibiotics are being generated in that hyena's mouth microbiome, it might be a new

00:23:37.740 source of antibiotics for us. And he was so extreme in this belief that the hyena had such a clean mouth

00:23:43.180 that in the context of doing this job of trying to get microbiomes, he got bit by a hyena. He went to the

00:23:48.460 emergency room and they're getting ready to give him antibiotics. And he said, no, I don't want antibiotics.

00:23:52.600 It's going to decimate my microbiome. Furthermore, I know that because I got bit by a hyena, I'm not going to get

00:23:57.220 an infection. They have the cleanest mouths. And he had to sign all these waivers. And apparently he never got an

00:24:01.760 infection. But the mouth microbiome is a source of potentially bacteria, but maybe other sources of

00:24:09.040 new science.

00:24:11.040 What is the classification? What's the org chart of these bacteria?

00:24:16.040 Now we're going to get back to seventh grade biology where you have the phylum, the family,

00:24:20.220 the genus, the species, the strain. And so that's the organization of them. And so...

00:24:25.180 I don't even remember that anymore. Say it again.

00:24:27.620 It's the phylum, the family, genus, species, strain. And so what DNA sequencing has enabled us

00:24:35.600 to do is to really look at strains. And maybe one interesting thing is even within strain identity,

00:24:40.440 so we give names to these strains based on their genomic makeup, even that part is really evolving

00:24:45.360 in our understanding as well. So for example, you know, methylation is sort of a hot topic when it

00:24:49.340 comes to humans. Bacterial genomes are also methylated. It's typically used for silencing certain genes.

00:24:55.080 But, you know, you sort of have this question of, is it really just the genetic makeup or these

00:24:59.120 post-modifications also changing a strain from one thing to another? Should you define a strain

00:25:04.720 by its genomic makeup, which is the traditional way we define things? Or should you define it based

00:25:09.160 on what it's doing and its actions? And that becomes more complicated. So I think we're going

00:25:14.300 to see that definition maybe evolve as we learn more.

00:25:17.900 Do you mostly operate in a world where you're thinking of the strain then?

00:25:21.900 Yeah, we operate in a world where we're thinking about the strain and the function. So we, for

00:25:27.300 example, as we manufacture our strains, we do occasionally these sort of whole genome audits

00:25:32.240 because these strains do divide and replicate most of them basically every two hours. You want to make

00:25:38.580 sure that whatever genetic modifications are just happening naturally don't actually impact the

00:25:43.480 function of the strain. So we literally do these audits. We'll do a whole genome sequencing

00:25:46.780 of these batches. We'll do a full panel of biochemical assays to understand, are they still

00:25:52.200 generating small molecules at the rate that we'd expect growth curves? And so I think that's one of

00:25:57.160 the challenges to actually being in this area is that these guys evolve really quickly.

00:26:02.380 Yeah. I mean, it's hard for me to kind of wrap my head around that because even though I acknowledge

00:26:05.460 what you're saying with respect to the speed of their multiplication or replication cycle,

00:26:09.440 when I think of my former life when I was in a hospital, you know, everybody knows what MRSA is,

00:26:14.840 but does it mean that there's like a new MRSA every pick your favorite period of time where

00:26:21.160 even something every month, there's a new MRSA in the hospital kind of thing that fortunately always

00:26:27.400 still seems to be responsive to vancomycin, but at some point it won't be like, I guess there's VRE,

00:26:33.160 right? Anc-resistant enterococcus. So it's hard for me to wrap my head around the speed with which

00:26:37.600 these things are mutating. So that within the span of a year of your life, does that mean

00:26:44.540 that your gut biome is changing, not just as a result of you changing, but because they're evolving?

00:26:52.920 I still don't understand exactly who's optimizing for what. Let me reframe my question. What are they

00:26:58.480 optimizing for in their rapid evolution?

00:27:01.060 They're optimizing for their environment. And so again, if you take diet as the primary

00:27:07.780 thing that can modify your microbiome, it's the primary thing that can modify your microbiome

00:27:11.700 because that's their food. They're living in your gut. They're waiting for you to feed them.

00:27:15.520 The reason that MRSA and VRE are evolving is to escape the antibiotic.

00:27:21.620 Yes.

00:27:22.340 Which is against our best interest. But what you're saying is, at least if I'm hearing you correctly,

00:27:27.140 it could be that the evolution of our gut biome is in our best interest because, in theory,

00:27:32.440 it's evolving to its environment, which is us.

00:27:35.900 Exactly. It's evolving to survive what we're feeding them and the environment that we're

00:27:40.240 creating for them.

00:27:41.280 Interesting. Okay. So very, very fluid.

00:27:44.420 It is. It is. You can go on a trip to another country and eat the food there for a week and come

00:27:50.860 back and your microbiome looks totally different in just a short period of time. I mean, many of us

00:27:55.100 have experienced that. Like, you go travel, you get GI distress from foods because you're not used to

00:27:59.320 having them, and you can whole hog change your microbiome. That's actually what really drew me

00:28:04.600 to this field. Because when you're talking about human genomics and human systems, the way in which

00:28:10.520 you can impact them is limited because it's an already existing system that's relatively immutable

00:28:15.840 without some real serious external force. You have to introduce a chemical small molecule to change

00:28:20.800 a pathway that has five backups to it in your system. But the microbiome is incredibly mutable,

00:28:27.360 and we're doing it all the time. And so when you think about the ability to change it and to have

00:28:31.020 real health implications, that's where it's at. That's why it's an exciting field.

00:28:36.140 Got it. Makes sense. Let's talk about how one measures these things, both in the lab, so what you

00:28:43.520 would do, but also maybe, for lack of a better word, over the counter. I mean, what I could do.

00:28:50.360 So if we were interested in understanding your genes, we could do them in a number of ways.

00:28:57.120 We could do the gold standard, which was we could do a whole genome sequence. We could sequence

00:29:02.580 every nucleotide of every single coding and non-coding gene in your body. And 20 years ago,

00:29:11.800 that would have cost close to a billion dollars. Today, that's about $1,000. Still not the most

00:29:19.160 practical test in the world because it yields a whole bunch of information one doesn't know what

00:29:23.080 to do with. There's the gold standard. Conversely, we could go on a fishing expedition and you could

00:29:28.380 say, well, I'm really worried about my risk of cancer. And we could do a commercial test that looks

00:29:34.080 at a whole bunch of known polymorphism SNPs, and we could screen for a hundred of those things.

00:29:39.540 It's a much more targeted look, and we could get that information. Walk me through the menu of

00:29:46.080 options that you as a scientist would embark on to do this, and then what a consumer can do.

00:29:52.540 It's actually quite similar when it comes to understanding the microbiome. So you can do

00:29:56.860 shotgun sequencing, where you're getting the entire genomes of all the different microbes,

00:30:01.020 and then you need to be able to assign which genome goes to which microbe, and there's a fair

00:30:04.560 amount of redundancy. And so you use long read paired with short read sequencing to be able to get

00:30:09.540 tell folks what that is so they understand. Yeah, because it's a bit complicated.

00:30:13.080 It's kind of exactly how it sounds. Short read DNA sequencing gives you, so let's assume,

00:30:18.220 let's take an example. You've got a thousand bases of DNA that you want to sequence.

00:30:23.740 A bacteria has typically how many genes and how many base pairs?

00:30:27.320 Really is a huge range of sizes. If you're trying to measure a certain piece of DNA, you can use

00:30:36.000 short read sequencing technologies, which allow you to, with high accuracy, get these short pieces of

00:30:42.060 your sequence. So you get maybe 200 base pairs, and so you would have a few of these, and then you

00:30:46.760 paste them together based on kind of the overlapping. The overlaps, yep.

00:30:49.540 And each piece is pretty accurate in terms of the sequence. Then you can do long read sequencing,

00:30:54.300 which will allow you to get in one shot that entire thousand base pairs, but that tends to be

00:30:58.440 a slightly lower fidelity read, and so you might have some errors in there. And so that's why kind

00:31:03.180 of the best way to do it is to do both. And I would say the reason that stitching together is

00:31:07.920 problematic is because bacteria have redundant genes. And so some bacteria might have five copies

00:31:14.460 of a gene versus another bacteria that has 10 copies of a gene, and it turns out it matters. And so

00:31:19.500 I'm not going to remember the name of this bacterial strain, but there is a strain that's been

00:31:23.720 studied pretty well that metabolizes digoxin.

00:31:27.260 A drug that's used to treat arrhythmias.

00:31:29.160 Yes. And one of the reasons why that drug fell from being first line is because the efficacy sort

00:31:33.960 of had such a wide, broad range across people, so it's hard to know what's the right amount

00:31:38.180 prescribed to somebody.

00:31:39.420 Wait, does it treat arrhythmias or heart failure?

00:31:41.260 It's heart failure.

00:31:41.940 Yeah, yeah, yeah. That's how long I've been away from the game.

00:31:45.220 What they found was that there's a microbe that can metabolize digoxin,

00:31:49.860 and so people who needed higher doses to have efficacy had higher amounts of this strain.

00:31:54.440 But then in another double click of that, it wasn't just that strain. It was how many copies

00:31:58.620 of this particular gene that it had. So if it had over five, it could metabolize digoxin. Less

00:32:02.780 than five, it couldn't.

00:32:03.780 And that means that that bacteria exists high in the GI tract, presumably, because if that were just

00:32:09.680 something in your cecum or beyond, presumably it wouldn't have impacted it. I mean, it has to be

00:32:15.940 somewhat in proximity to the liver. It has such an impact on the bioavailability of that drug,

00:32:21.820 right?

00:32:22.540 Isn't digoxin an orally?

00:32:24.120 Yes.

00:32:24.520 I actually don't remember the location of this bacteria, but I remember this part about the

00:32:27.980 number of genes being important as to whether it could even metabolize a thing or not.

00:32:31.940 Because normally we think a lot about the different genes in the P450 system in the liver,

00:32:37.280 which is heavily responsible for how most drugs are metabolized. And that clearly explains a lot of

00:32:42.280 the variability in human metabolism of drugs. I never knew about this. This is even one step

00:32:47.480 beyond that. And obviously not dependent on our genome, but the genome of this host.

00:32:52.200 Yeah. And these are mostly gut microbes. So it really is, especially for these oral

00:32:55.760 drugs, there is some path that it takes where it's interacting with these microbes and they

00:33:00.380 are able to metabolize these various drugs. Getting back to your sequencing question,

00:33:05.900 the number of replicates matter. So if you imagine you have these short snippets,

00:33:09.500 it's sort of a guessing game as to, well, gee, is this an actual just replicate or was it only

00:33:14.560 there one time?

00:33:14.980 Was it the second one?

00:33:15.860 Exactly. And so that's why you need-

00:33:17.580 A long sequence.

00:33:18.560 ...becomes your template to put these short reads on.

00:33:20.800 That costs several thousand dollars per sample to do.

00:33:24.060 Even today?

00:33:24.920 Even today, because you're really trying to get a comprehensive. And then just imagine that,

00:33:28.920 and now it's every bacterial strain out there. And P.S. there's a lot of redundancy in

00:33:33.180 bacterial strains. For a large number of them, we don't even know, much like the human genome,

00:33:38.660 we don't even know, what are these real bacteria? What do they do? And so there's just still a lot

00:33:43.600 of uncatalogued genes. And so that's a real endeavor. The second way is to really look at

00:33:48.860 a qPCR-based where you're really looking at a specific strain and you're trying to understand

00:33:54.480 how much of that exists.

00:33:56.040 Tell folks why the PCR would work that way, like what you have to use and how you have primers and

00:34:00.340 why that's analogous to the, let's look at your cancer genes approach.

00:34:05.340 Yeah. So you might have a lot of bacterial strains in your ecosystem, but you don't know

00:34:12.140 how much they are relative to each other. So you can get a catalog of everything that's in there,

00:34:16.700 but you don't really know how much of each one is in there from these sequencing methodologies.

00:34:21.820 Although there are some really interesting tools that people are starting to develop to try to get

00:34:24.860 at that. But the more accurate way to get at the quantitation or how much of this strain is

00:34:29.400 actually in the microbiome is this quantitative PCR. Basically make primers that are specific to

00:34:34.700 that strain. And then you're using PCR in a quantitative way to understand, well, how much

00:34:40.140 of it is in there compared to say some other strain or compared to the entirety of all the

00:34:44.940 different bacteria in there. And now you have an idea of, is this constituting 1% of my microbiome,

00:34:50.680 10% of my microbiome? And so you get this quantitative piece. If you're low in specific microbes or

00:34:56.100 specific functions, this becomes the way that you would really look at that. But there's an added,

00:35:01.280 very important complexity that the microbiome has that the human genome doesn't have. And that's

00:35:06.140 around this replication and ever-changing thing. And so it's really not good enough, as we talked

00:35:11.560 about with the Human Microbiome Project, it's not good enough to get one snapshot in time.

00:35:16.100 You really want to understand what is your baseline microbiome. And then if you ever change your diet,

00:35:21.280 you travel, you go on antibiotics, there's all kinds of things that can change your microbiome in very

00:35:25.200 cute ways, it all of a sudden becomes a different microbiome. And so this longitudinal data really

00:35:29.980 matters. And then additionally, because every microbe is replicating at a different rate,

00:35:35.400 the constitution or the fraction in which one microbe might be in your gut today might be different

00:35:40.220 tomorrow. And so you need longitudinal data that gets you the quantitative piece plus who are the

00:35:45.040 players. And then the third part is because they're also mutating, you need to understand,

00:35:49.380 well, did the functions change? So if you really want to understand the microbiome, I mean,

00:35:53.060 we were spending $5,000 to $6,000 per sample to understand longitudinally what's happening,

00:36:00.180 what are the different functions, how are these things changing? So it's pretty hard to do.

00:36:06.080 And what are the ideal functional assays? Because I think it's important that we don't lose sight of

00:36:10.220 what actually matters. It's what clearly matters when we look at our cells. I think we're really

00:36:16.120 understanding that today in humans that function matters more than genome. The protein is more

00:36:22.660 important than the gene. What do we look at in the gut biome? Are you looking at secretory products?

00:36:28.900 How are you determining the health of the function versus just the genome, which obviously must be

00:36:35.980 correlated with it, but not exact, right?

00:36:37.880 We are super interested in carbohydrate metabolism. And so we look at the output of that as the short

00:36:44.500 chain fatty acids. So butyrate, propionate, acetate. And so you can take an individual strain,

00:36:49.960 feed it a substrate, depending on where in that biochemical pathway it is, its substrate might

00:36:54.200 be slightly different. And then look at how much of those small molecules are being produced on a gas

00:36:59.360 chromatographer. And so you're running a enzymatic reaction. If you think about the bacteria as the

00:37:05.140 enzyme, you're giving a substrate and you're basically doing your old fashioned Michaelis

00:37:10.140 Mentencarves, where you're giving it increasing amounts of substrate. And you're basically looking

00:37:13.680 at the enzymology of going from substrate to that short chain fatty acid.

00:37:18.180 So let's use that as an example. You eat a piece of bread, starch, polysaccharide. Digestion,

00:37:25.280 of course, begins in the mouth. Amylase starts to break that down. It continues further in the stomach.

00:37:29.260 By the time it gets to, God, the jejunum, I mean, it's mostly just glucose monomers, right?

00:37:35.800 I think that's still being figured out, actually. What is the state of affairs when things arrive? I

00:37:41.240 mean, it's actually really hard to survey what's happening in the gut microbiome because we don't

00:37:44.480 have good sampling methodologies. That's super interesting. So you're saying even if you dropped

00:37:48.880 a tube down somebody's throat and you just sampled the slurry at the distal end of the duodenum or at

00:37:54.780 the proximal end of the jejunum, I mean, I've seen what it looks like when you see when you operate on

00:37:58.500 somebody. We can't tell what the composition of matter is there. I'm not squarely in this,

00:38:05.280 but I haven't seen anything pop out that was really compelling. The pushback all of those

00:38:08.860 technologies get is the ease with which things can get contaminated. And so essentially you might be

00:38:15.100 able to get a sample, but then in the process of pulling that back out, it becomes contaminated

00:38:19.180 with the other things that are along the track. It could be done intraoperatively. Hopefully someone

00:38:23.040 is doing that, right? If you're in there operating otherwise. Now the problem is you wouldn't be

00:38:27.420 operating on somebody without a bowel prep. And so you've completely destroyed the system.

00:38:33.960 Also, you're not operating on somebody in an anaerobic chamber.

00:38:36.980 The second they get exposed to oxygen, it's sort of different. Okay. So explain how carbohydrate

00:38:42.680 metabolism produces byproducts and what those byproducts are. Because people who listen to

00:38:47.960 this podcast, people like me, when I think of carbohydrate metabolism, I don't think of any of the

00:38:53.260 things you just said. I think of glucose. I think of glucose 1-phosphate. I think of glucose 6-phosphate.

00:38:58.900 I think of pyruvate. I think of acetyl-CoA. I think of the Krebs cycle. I think of lactate as a byproduct

00:39:07.540 of oxidative phosphorylation. So you're thinking of things in a different level because you're obviously

00:39:13.040 looking at a different host. So explain the metabolism on that side of the ledger.

00:39:18.020 So in this case, when we talk about carbohydrates, we're really talking about these fibers. And so

00:39:22.300 our microbiome is uniquely positioned to metabolize fibers, a wide variety of which we actually can't

00:39:26.960 even metabolize ourselves. And fiber is one of the primary prebiotics that feeds all of your strains.

00:39:34.640 And so you're right. When we think about carbohydrate metabolism from the perspective of the microbiome,

00:39:38.480 it's really thinking about these fibers. There are primary and secondary fermenters in the microbiome

00:39:43.360 that can metabolize these fibers into certain short-chain fatty acids, which then become precursors

00:39:48.560 for the ultimate short-chain fatty acid, which is butyrate. Butyrate is this incredibly important

00:39:54.360 short-chain fatty acid that's been studied in a wide variety of conditions. And so your microbiome,

00:40:00.120 one of the most important molecules it makes, is butyrate. And so butyrate has a role in GI health.

00:40:05.700 The colon cells are the only cells that use butyrate as their source of energy, as opposed to glucose,

00:40:10.000 which is what's used by every other cell. And so when you don't have enough butyrate,

00:40:13.040 that's been associated with things like colon cancer and not having good colon health. But

00:40:17.580 butyrate is also a small molecule that triggers G-protein-coupled receptors to then release GLP-1

00:40:24.840 from these L cells in the microbiome. So it also plays a role in sort of this gut metabolism axis.

00:40:30.660 Butyrate becomes this really important small molecule that the gut is producing based on the

00:40:35.740 foods that you're eating. And that's kind of where we've really honed in.

00:40:38.680 So if we think about the difference between soluble and insoluble fiber,

00:40:42.840 are you referring mostly here to insoluble fiber, because you alluded to the fact that it's fiber

00:40:47.420 we aren't able to digest and metabolize, that is the food stock for these bacteria?

00:40:52.180 Yeah, that's really where we're focused in on is these insoluble fibers.

00:40:55.500 Insoluble fiber, I guess, is most readily available in vegetables, correct?

00:41:00.140 Exactly. Vegetables and fruits, yeah.

00:41:02.920 I was asked recently by a patient, I drink this green drink every morning,

00:41:07.760 and they're like, why do you drink it? Do you think it's a substitute for having

00:41:11.800 vegetables? And I said, I really don't. I said, I think it's a substitute for having vegetables with

00:41:17.500 respect to the vitamins you get, and probably even a lot of the polyphenols. But it's clearly

00:41:25.580 not a substitute for the fiber. Just based on the practicality, like you can look at the ingredient

00:41:30.260 label, there isn't enough fiber in this to be a substitute. So at the end of the day, like there

00:41:35.780 doesn't appear to be, if you buy the argument, which I think we're going to discuss, that fiber

00:41:40.120 is essential for gut health, which by extension means essential for human health, you have to be

00:41:46.460 eating a ton of fiber, insoluble fiber.

00:41:49.040 Yeah, you do. Although, I mean, there have been a myriad of studies showing that getting fiber from

00:41:53.920 these external sources, these powders, now you have to get them in the volumes needed. They can be

00:41:58.800 pretty impactful for your microbiome as well.

00:42:01.100 What are typical recommendations of fiber? Like how many grams per day? 18, 30, something like

00:42:06.540 that?

00:42:07.180 I actually don't know what the current standard recommended dose is, but I think it's somewhere

00:42:12.200 between 20 and 30 grams.

00:42:14.260 Yeah, it's hard to imagine getting that in a supplement.

00:42:16.460 And I think actually your average American's consuming like one to two. We're way off base

00:42:21.420 on how much fiber we're consuming. But just to get to your question, though, there's also a

00:42:25.480 delivery component to this too. So taking the fiber where you mix it in a drink might be a

00:42:31.020 lot less impactful, or you might need a lot more of it, I should say, than if you were to take it in

00:42:35.400 an enteric-coated capsule that got through the stomach acid and then closer to the distal colon.

00:42:41.300 It's a delivery question. So you might be able to get away with a lot less.

00:42:44.920 Oh, interesting. Why is that? I mean, I thought that given that it's insoluble fiber,

00:42:49.260 it would be impervious to the gut and the environment all along the way. Is that not the case?

00:42:55.360 The measurements of these things along the way has been elusive.

00:42:58.600 As my daughter would say, sus.

00:43:00.500 It's sus. We include inulin in our pill, an incredibly small amount. It's, you know,

00:43:07.140 on the order of one to 200 milligrams. So it's definitely not at the dietary fiber level.

00:43:12.200 But the reason we include it is because when we did our preclinical studies, we saw that if we

00:43:16.740 didn't include it, the strains really didn't colonize as well. So the concept behind that is

00:43:21.520 that by having the inulin right there next to the strains, as soon as that capsule dissolves,

00:43:26.180 these things start to get hydrated and start to become alive in very close proximity as their

00:43:30.080 food. And that's why I think about that kind of delivery and that proximity question as being

00:43:34.880 something we haven't really solved for yet. Interesting. Okay. So you eat a piece of celery,

00:43:40.320 which is basically all insoluble fiber. You mechanically break it into smaller pieces,

00:43:45.060 but functionally it's still cellulose that's making its way down. At what point in the gut does it have

00:43:52.140 to wait till it's in the colon or does the metabolism and production of butyrate begin in

00:43:57.000 the small intestine? There's some bifidobacterium in the small intestine that can help start to break

00:44:01.620 that down. But I think the general thought is that the majority of that metabolism is happening

00:44:06.080 in the distal colon. Okay. And which strains of bacteria are most responsible for the metabolism

00:44:14.800 of insoluble fiber and the production of butyrate as a byproduct? It's a multi-step pathway. So there

00:44:21.540 are clostridial strains, which can do this breakdown. There are bifidobacterium strains that

00:44:26.600 can do this breakdown. And then there is the strain acromantia, which was discovered in the early 2000s.

00:44:32.600 And it also plays a really important role in generating these short chain fatty acids.

00:44:37.900 And are these strains, do they belong to the same species?

00:44:40.760 These are the species names. Okay. Strain name would have two parts to it. So if I said

00:44:46.020 acromantia mucinophila, that would be the name of the strain, but I've shortened it to acromantia,

00:44:50.440 which really does represent the species, the higher level. Got it. So E. coli is?

00:44:55.660 Strain name. Is actually a strain name. Yeah. Eshera, whatever it would be the species.

00:45:01.240 Got it. Clostridium is the species. Clostridium difficile is the strain. Yep.

00:45:05.540 Okay. Let's pivot for a moment and actually talk about C. diff because that's one of the most

00:45:10.080 compelling arguments for an intervention in treating human disease by manipulating the gut

00:45:21.380 bacteria. So tell folks what Clostridium difficile is as a bacteria. And you already alluded to this,

00:45:28.060 but how does it go from being a benign slash benevolent participant in our coexistence with

00:45:34.540 the universe to one that could kill us? So Clostridium difficile is a strain. It exists,

00:45:38.660 and many of us have it in low levels in our gut microbiome and we walk around perfectly healthy

00:45:42.840 and fine with it. When you take an antibiotic, that's essentially almost like a nuclear bomb to

00:45:47.940 your microbiome. It kills everything in there, but in some cases it doesn't kill everything. And so

00:45:54.520 there are these strains of Clostridium difficile that after you take an antibiotic, it's killed all

00:45:59.980 the different strains off, but it didn't kill your Clostridium difficile strain. And the problem with

00:46:05.640 killing off all the other strains is now all the competition is gone. So you imagine you have this

00:46:10.820 ecosystem of different microbes and now you've just gotten rid of all of them. So now you have

00:46:14.580 this strain that has no competition and it can start to propagate unchecked. And so it's when you

00:46:19.100 start to have these really high levels of this strain, Clostridium difficile, in your microbiome,

00:46:23.660 that's when it starts to make you really sick. And actually ultimately it's fatal.

00:46:26.660 And so the way in which we treat infections is through more antibiotics. And so when you have

00:46:33.900 this Clostridium difficile infection, which is a result of having taken an antibiotic, almost

00:46:38.000 ironically, the cure also is another antibiotic. And really what you're trying to do is to get an

00:46:42.360 antibiotic that can kill that strain and get it back to its low levels until your microbiome can

00:46:47.280 reconstitute itself through the food that you eat. The success rate of those antibiotics varies

00:46:53.200 from person to person, but overall it's something like 70% successful. And remember,

00:46:57.320 fatality is on the other side of this.

00:46:59.140 I was just about to say, make sure people understand what non-success means here.

00:47:03.140 Yes. Yes. 70% success is not great when the 30% means that you're going to die on the other side.

00:47:07.220 So one of the concepts was really to go in the reverse direction and say, okay, if the problem is

00:47:12.680 that now there's no competition and this guy can propagate unchecked, what if we just load up the

00:47:17.960 person's gut with all these different microbes, reestablish an ecosystem, and then that's the way to

00:47:23.180 temper this thing down. And that's called a fecal microbiome transplant. It's exactly what it sounds

00:47:27.880 like. You literally take feces from a person and transplant it into another person. That thing has

00:47:33.400 like a 99% success rate.

00:47:36.480 When was this first done? I feel like this was at least in the 90s, maybe earlier.

00:47:40.300 Well, it was definitely done as one-offs before people were doing real studies with it. And actually

00:47:44.520 a ton of them were done in Australia. It's like one of the places where the most of these have been

00:47:48.920 done. But it has an incredible success rate. It's kind of gross.

00:47:52.000 Yeah. What was it like back in the day and how is it done today? I assume today it's done through

00:47:56.580 capsules?

00:47:57.700 Well, I think people are still using the methodology of actually going up through the colon, but people

00:48:02.280 have tried to create capsules and tried to be able to make it so that you can just consume these

00:48:05.880 capsules with a drink and then it gets there. And those seem to be effective. They're not as effective

00:48:10.760 as the enema version. And because probably in the process of getting that stool into a freeze-dried

00:48:17.660 format or into a pill format, you're losing probably some of the diversity that you just

00:48:22.300 get when you do it the old-fashioned way.

00:48:24.600 And there are risks associated with this or not? Are they overblown? I mean, there's certainly been

00:48:28.240 hoopla about this at the FDA at one point. What was that about?

00:48:33.220 Fundamentally, it doesn't feel like a therapeutic to give people shit. I think there's this instinct

00:48:39.540 that, okay, this isn't safe. But there is a reality around it, which is that the source of that person

00:48:44.740 is really important, right? How do you know that that's a healthy, so-called healthy stool donor?

00:48:49.760 And so what if they've got some weird pathogen that now you put it into this person who's already

00:48:54.220 in a depleted state?

00:48:54.640 Who's already compromised, yeah.

00:48:56.000 And then furthermore, because everybody's microbiomes do have these different functions,

00:49:00.240 it's possible that you might cure them of the C. diff. Okay, great. And you might not have given

00:49:03.920 them a new pathogen. That's fine. But you might have changed their metabolism of foods in a way

00:49:08.660 that's not beneficial to them. And so, for example, there are some of these case studies of someone

00:49:14.200 getting a fecal microbiome transplant, and now all of a sudden they have obesity issues that they

00:49:17.480 never had before. What really is at the heart of it is that we're very early in understanding

00:49:22.060 microbiome science. And so I think the FDA, their job is to make sure that people are safe. And so

00:49:27.000 they have to decide, is it more safe to give people this cure that has this incredibly high success rate

00:49:33.380 when fatality is on the other side? Or is it more safe to say, hey, let's keep learning more about

00:49:38.660 this before we actually introduce it into clinical care? And they actually tried to shut the whole

00:49:43.940 thing down. They essentially said, FMTs are not approved by the FDA. They're not safe until we

00:49:49.140 learn more about the science. What year was that-ish?

00:49:52.200 This was probably, I don't know, maybe 20 years ago. Essentially, these people who had these

00:49:58.740 physicians, these patients, patients' families who had all experienced this benefit, and then people

00:50:02.820 who were in the queue to get it done, essentially with pitchforks in hand, went to DC and said, this is

00:50:07.900 crazy. Like, this thing has such a high cure rate. How could you possibly take it away

00:50:11.280 as an option for people? And so they put their tail between their legs and reversed it and let

00:50:15.980 people do them.

00:50:17.480 Let's talk about an example you just gave there, which I know is sort of theoretical,

00:50:20.480 but I'm sure there are many cases like it, where you take a lean individual who's hospitalized for

00:50:25.980 some reason. They're given an antibiotic. In the process, they develop C. diff colitis.

00:50:29.680 They get a fecal transplant and they recover. The fecal transplant came from somebody who was obese

00:50:37.160 and now they develop obesity. So the hypothesis here has to be pretty clear that the gut bacteria

00:50:46.460 of the obese individual is playing a causal role in their obesity. So I want to tackle that topic,

00:50:53.280 but first, why wouldn't the habits of the recipient immediately override the bacteria that they got

00:51:03.360 on the receiving end, given that their habits are more in line with being lean? In other words,

00:51:08.520 given the rapid evolution of these bacteria, why do they persist in their phenotype?

00:51:14.760 Well, first of all, these are case studies. These are not clinical trials. And so I think

00:51:18.940 you're dealing with the uncontrollable nature of humans. But I would say one of the interesting

00:51:24.300 things is that these people's habits did not go back to what they used to have. And one of the

00:51:29.160 things we're learning about the microbiome is through this gut-brain connection, your microbiome

00:51:33.360 can actually influence your food cravings. And we actually have some data supporting that as well.

00:51:38.600 And so if you think about this new microbiome as maybe it's not metabolizing food as efficiently,

00:51:44.340 it gives you a predilection towards obesity, but on top of that, it changes your food cravings

00:51:49.120 you've actually now got a double whammy against you that keeps you from being your old self. And so

00:51:54.820 I think the microbiome, this gut-brain connection, its ability to generate neurotransmitters,

00:52:00.280 its ability to change cravings, is still at early stages of understanding, but it can basically

00:52:05.220 change your behavior. So you're not really the old person you are.

00:52:08.820 Profound. Two thoughts. The first is, could you be in that situation forced to eat your old way?

00:52:14.700 You're going to go to a one-week bootcamp where you're literally force-fed everything you used

00:52:19.360 to eat. Would that be sufficient to return your gut biome to the way it was? Talk about how the gut

00:52:26.740 and the brain connect. So my vague recollection is serotonin plays a role here. You've already talked

00:52:32.400 about GLP-1. GIP typically is secreted higher in the gut, but that presumably plays a role as well.

00:52:39.060 Again, this is in the early stages and we're going to learn so much more over time, but

00:52:43.720 the gut is a massive producer of a bunch of things that we've traditionally thought of as

00:52:49.260 neurotransmitters. So yeah, serotonin, dopamine, GABA, these GLP-1, GIPs. And then I would also say

00:52:56.500 there are neurons in the gut. And so there's this relationship between the neurons in the gut and

00:53:01.560 the neurons in the brain. The neurotransmitters being generated by the gut can make their way straight

00:53:05.740 to the brain through the vagus nerve. And so there is potentially an opportunity, for example,

00:53:10.700 to tackle stress and anxiety through the gut microbiome where you have these gut microbes

00:53:15.820 that can produce large amounts of GABA, which has been shown to be able to reduce stress and anxiety.

00:53:20.780 And so is there an opportunity to go after the gut to reduce stress and anxiety? And again,

00:53:26.280 kind of getting back to this cravings thing, it's really interesting for anybody who's ever felt stress

00:53:30.400 or anxiety, you know, actually your food cravings oftentimes change and kind of are hand in hand with

00:53:35.400 that. So this part about the gut-brain connection that involves both cravings and these neurotransmitters

00:53:41.880 I think is really, really fascinating. How much of that do you think is regulated by things outside

00:53:47.620 of food, meaning the gut flora as it comes to changing function, which then changes our physiologic

00:53:53.640 state? I mean, we've talked about food. We should talk about it some more. Clearly, a very important

00:53:58.020 piece. You've already talked about antibiotics. What are other modifiable factors that change this?

00:54:03.220 Certainly the two biggest changes in your microbiome are the things we've talked about,

00:54:07.200 antibiotics and nutrition. But beyond that, what we know is that there are, and this is especially

00:54:12.680 with regards to diversity and loss of function of your microbiome. We know that as you age,

00:54:16.880 your microbiome starts to become depleted. We know that when you go through periods of intense stress,

00:54:22.140 your microbiome becomes depleted. We know that when your circadian rhythm gets disrupted,

00:54:26.660 so you travel and day becomes night, night becomes day, that changes your microbiome.

00:54:30.580 By the way, when you say depleted, tell me what that means in a technical sense.

00:54:35.420 It means that what happens is that you have loss of function. So you start to become depleted in

00:54:40.860 specific strains that have specific functions. And so, for example, this strain that I alluded to

00:54:45.700 earlier, acromantia, is one of the ones that is deeply studied. And low levels of acromantia have

00:54:50.820 been associated with a really wide variety of diseases ranging from obesity to immunological

00:54:56.420 disorders to stress and anxiety and GI issues. And so, basically, the depletion means you're

00:55:02.420 losing certain strains and certain functions. And then for us women, when we go through menopause,

00:55:06.320 there's like a massive change in the microbiome on the other side of that lovely experience.

00:55:10.860 I was going to ask you about the vaginal microbiome,

00:55:14.220 but right now, are you saying that with respect to gut as well?

00:55:17.740 I'm saying that with respect to gut. Yep.

00:55:19.680 Okay. What are the factors that influence the vaginal microbiome besides the most obvious,

00:55:23.700 which must be menopause, pregnancy, or anything that has a dramatic shift in endometrial lining

00:55:30.100 and changes? So, presumably, if you sample a woman's vaginal microbiome across the month of her

00:55:35.720 cycle, given that her estrogen and progesterone levels are fluctuating, you're going to see

00:55:39.580 differences. But I assume there's a quote-unquote healthy versus an unhealthy. How does that manifest

00:55:46.540 itself? And what does one do about that? Well, I'm not an expert in the vaginal microbiome,

00:55:52.080 but certainly these lactobacillus strains are some of the more important strains in the vaginal

00:55:57.160 microbiome. And the name of the game there is actually production of acid. You want an acidic

00:56:01.080 environment there. So, it's a protection environment. It's a protection environment.

00:56:03.540 The goal is to keep yeast in particular and other things away.

00:56:06.120 Exactly. Sperm, everything. So, there is a healthy vaginal microbiome, which is really associated with

00:56:12.200 the acidity level that these bacteria keep the vaginal microbiome at. It can be influenced actually

00:56:17.520 also by the food that you eat. I don't know. It could be that I don't know, but it's not totally

00:56:22.720 elucidated. But for example, you might know that one of the things that gets recommended for you today

00:56:26.540 is cranberry juice. And you're not inserting that into your vaginal canal. You're consuming it. And so,

00:56:31.540 there's definitely this connection between the food that you're eating and then the composition of

00:56:35.040 your vaginal microbiome. Pregnancy is also another thing that really changes it. And one of the

00:56:40.500 implications that we did look into, this was some work that we were doing with the Mayo Clinic early on

00:56:44.560 in our company, is the idea of preterm labor. So, preterm labor has been highly associated with

00:56:51.800 bacterial vaginosis. And that is essentially, it's a broad name because I think we don't fully

00:56:56.980 understand what it means, but it's an infection of the, or a disruption of the vaginal microbiome.

00:57:03.360 This infection triggers preterm labor. And so, one of the ideas here is that, well, if you could

00:57:09.820 prevent bacterial vaginosis, you might be able to reduce the incidence of preterm labor. And there's

00:57:15.900 other idea, which is that once you have a preemie, you're a lot more prone to have a preemie again.

00:57:21.040 And these women usually also have this bacterial vaginosis, which happens right before they deliver

00:57:25.340 the baby. So, maybe there's something about your vaginal microbiome that makes you more susceptible

00:57:29.420 to bacterial vaginosis, which then leads to this preterm labor. So, then there's, of course,

00:57:34.500 the UTIs and yeast infections and things like that. Any other systemic complications from a

00:57:40.780 negatively altered vaginal microbiome? I mean, I guess you could argue preterm labor is systemic.

00:57:46.280 Outside of the pregnancy case, if you just talk about a woman who's entered menopause,

00:57:50.300 because I'm guessing being on hormones versus not on hormones is going to have a profound impact on

00:57:54.240 that. Do we know if any of the problems associated with menopause beyond the obvious ones like vaginal

00:58:02.660 dryness and hot flashes have their roots in a change in that microbiome?

00:58:07.760 I think that work is still pretty early on. So, I would hesitate to go too far there. And even

00:58:12.740 the gut microbiome too, there are microbes that can, I'm not an expert in menopause or these hormones,

00:58:19.000 but I believe that estrogen has a modification that gets made that is a signal to the body to

00:58:26.100 not put that estrogen back into circulation, but to, I guess you must urinate it out.

00:58:30.080 But there are microbes in the gut, their job is to remove that modification. And so,

00:58:35.100 the effect is that it could potentially increase your circulating estrogen. And so,

00:58:39.480 this is all super early stage. If you go try to look up studies around menopause in the microbiome,

00:58:44.920 it's extremely sparse. And so, I think there's, we just really don't know.

00:58:49.900 But there could be a day, presumably, when there are bacterial products that are used to address some

00:58:55.360 of these issues. It certainly would make sense. Going back now to the example you gave, how robust

00:59:01.020 is the evidence in humans? And maybe we can start by the evidence in mice, because it gets all the

00:59:05.340 headlines, where you can take an obese mouse and you can do a fecal transplant from a lean mouse into

00:59:11.960 an obese mouse and make that obese mouse a lean mouse. Is that a relatively robust, reproducible

00:59:16.960 finding?

00:59:17.260 Yeah. A lot of people have done that. Even to take it one step further, they can make these mice

00:59:22.740 obese or lean using a human microbiome transplant. So, they basically would take the microbiome from

00:59:28.520 an obese woman, put it into a lean mouse and see that lean mouse become obese. And so,

00:59:33.580 we talked about the overlap between mice and men. There is some overlap there, because you can

00:59:37.820 actually change the metabolism of these mice through these fecal transplants.

00:59:41.140 Why has that not taken off in humans? I mean, what are the impediments to that? Are those studies

00:59:46.680 being done?

00:59:47.960 People have done fecal microbiome transplants into people with type 2 diabetes, for example. I think

00:59:52.820 there's two big challenges. One is, of course, the safety challenge, which is, do you really feel

00:59:58.280 confident that that stool donor is a stool that you want to have in your body and move forward with?

01:00:03.100 But more profoundly, and maybe linked to that, is the ability to reproduce the effect. So,

01:00:09.960 even if it's always Joey who's providing the stool, I mean, Joey travels. Joey goes on antibiotics. Joey

01:00:15.720 changes his diet. You may not always be eating the same stool from that donor. And so, in order to

01:00:21.460 generate consistent results, we do in drug development, dose response curves. You're really

01:00:27.860 trying to understand in this small molecule, in this setting, how much of it do you need and what

01:00:31.960 effect will it have? You can't do that if your intervention is a fecal microbiome transplant,

01:00:37.840 because that thing is different from person to person and even from the same person from time

01:00:41.920 to time.

01:00:41.960 It's a pretty messy drug.

01:00:42.640 It's a messy drug. When we started this company, the idea was the fecal microbiome transplant is super

01:00:48.340 compelling data. That tells you that there's something in that microbiome that can change

01:00:52.340 your physiology. But the name of the game is, how do you figure out what are the components of that

01:00:57.480 microbiome that are conferring that benefit? And how do you make a reproducible manufacturing and

01:01:02.800 quality process so that you know, every time I deliver this pill to you, I know exactly what's

01:01:07.220 in it. And it's the same every single time. And so, that becomes a challenge. How do you take that

01:01:11.760 whole kitchen sink, the whole fecal microbiome, and say, actually, it's these four strains or these

01:01:17.420 two strains or these 50 strains? That's the tricky part.

01:01:20.780 So, let's talk about what those strains might be and what their function is. So, for folks listening,

01:01:26.080 what are the strain names that we have to know here? Lactobacillus, you've already talked about.

01:01:30.640 Acromansia, you've already talked about. Yeah, I think if you start to look at labels on the

01:01:35.280 supplements that are out there, you're going to see a lot of lactobacillus and bifidobacterium.

01:01:38.560 Those are the two primary species that are pretty much in every product. Those strains are relatively

01:01:43.600 easy to grow. It's one of the reasons why those are the strains that are in a lot of probiotics.

01:01:47.500 Why are they easy to grow? I thought anaerobes are hard to grow.

01:01:50.420 These fall into that in between. They're facultative. You can grow them in some amount of oxygen. In fact,

01:01:55.420 an acromansia is not. It's a strict anaerobe. So, in fact, when we first started wanting to

01:01:58.980 manufacture acromansia, we just tried to outsource it to all these probiotics manufacturers around

01:02:04.000 the world. And they kept sending us back dead stuff. And it's because in order to manufacture

01:02:08.560 a strict anaerobe, your entire manufacturing system end-to-end has to keep all oxygen out.

01:02:14.460 A single molecule of oxygen kills the whole batch. And so, there's a real manufacturing challenge to

01:02:19.920 some of these next-generation probiotic strains.

01:02:23.200 Before we talk about that, let's talk more about the bifido and lacto. Are they a bit of the

01:02:27.320 shiny light problem? The keys under the shiny light?

01:02:29.960 Yeah.

01:02:30.360 Yeah. So, are they basically the ones that are most ubiquitous because they're easy to grow? Or is

01:02:35.660 there a really good physiologic case for their utility?

01:02:39.460 They're easy to grow. And so, they're prevalent. That's why so much work has been done on them.

01:02:44.900 Because there was no microbiome science when these strains started making their way into the market

01:02:47.940 in the 70s. There was just, what can I culture? And that's what's culturable. Nobody thought about

01:02:52.940 strict anaerobes or how to keep oxygen out. That's a pain and that's expensive. So, these are the

01:02:56.760 things that have been around since the 70s. And furthermore, they got grandfathered in by the FDA

01:03:01.640 as safe because they'd been on the market for so long. You want to bring a new strain to market,

01:03:06.020 you have to go through the entire process of the FDA that says, hey, it's a new strain,

01:03:09.980 but let me demonstrate to you all the thousands of reasons why this thing is safe.

01:03:13.540 And just to be clear, from a regulatory standpoint, if a company wants to put a new strain

01:03:18.600 into a probiotic, they're going down a GRAS pathway, a generally regarded as safe pathway,

01:03:24.440 or an IND? If you want to be able to sell it directly to consumers,

01:03:28.460 it has to be GRAS. You have to go down the GRAS pathway.

01:03:30.800 Are there people that are actually looking at a pure pharma strategy where they're making

01:03:36.000 an IND and going down the drug pathway? Absolutely. There are pharmaceutical,

01:03:42.260 actually, they're all startups, but pharmaceutical drug companies that are using the microbiome

01:03:46.920 as the therapeutic and filing INDs and doing these studies. And in fact, this fecal microbiome

01:03:51.860 transplant in the form of a pill, that's a therapeutic. There are multiple companies

01:03:55.280 out there that are developing that as a drug therapy.

01:03:58.900 Interesting. I mean, the closest thing that I can think of to that would be immunotherapy

01:04:03.820 using TIL. You have immunotherapy in the form of a drug where you use something like KTRUTA,

01:04:09.840 CTLA-4 inhibitor, where you have a single molecule that is an actual drug that triggers the immune

01:04:16.000 system to do something. But then you have tumor-infiltrating lymphocytes where you go and harvest a

01:04:21.660 patient's tumor, grow those T cells out, multiply them, you know, a log or more, and then re-infuse

01:04:29.020 them, actually several logs, and then re-infuse them. But of course, then every patient has their

01:04:33.320 own designer drug. That's still a purely regulated FDA IND problem. That sounds a lot more like the

01:04:41.700 fecal microbiome transplant, which is every patient has a different drug. The difference there is it's

01:04:46.920 not coming from them. By the way, it begs the question, should every patient going to the

01:04:51.660 hospital have a banked stool sample so that they can be their own transplant? You could have an

01:04:56.420 autologous transplant and save a lot of the risk. Yeah, I think that would probably be a super smart

01:05:02.700 thing to do, bank your own stool. You could do it in your own freezer. Right, just leave a stool

01:05:08.280 sample in your fridge when you go to the hospital, and if you can make it home and you're fine, you can

01:05:12.160 pitch it. Yeah. Yeah, interesting. What drew your interest in this other species called

01:05:18.000 acromantia? I think that acromantia has become an incredibly interesting strain over the last

01:05:25.780 decade. But when we first became interested in it, it was really centered around trying to

01:05:30.460 understand the difference in people with metabolic syndrome versus people who don't have metabolic

01:05:34.660 syndrome. We did, and a variety of people have done, these twin studies. These twin studies are

01:05:38.680 really interesting because you take a twin pair and you're looking for discordant twins where one

01:05:44.560 twin is healthy, the other twin has diabetes or is obese. And you're looking at their microbiome and

01:05:50.000 you're basically saying, hey man, genetically these two people are the same. This is a nurture, not nature

01:05:53.520 problem. And so you start to look at their microbiomes and you start to see patterns. Around the world, if you

01:05:59.260 compile this data together, you start to see that the twin that has obesity or type 2 diabetes has a

01:06:05.380 different microbiome from the twin that's healthy. And one of the hallmarks that appears to be

01:06:09.560 true across different cultures and different dietary patterns is this depletion in this strain called

01:06:14.620 acromantia. In absolute numbers. In absolute numbers. And what's the relative difference between these

01:06:21.700 populations? That I don't know, but I do think that it is interesting because across these different

01:06:27.420 populations, acromantia is, we don't have to get into this, but the so-called healthy gut microbiome, which is yet to be

01:06:32.900 defined, across the world, acromantia comprises the microbiome on the order of somewhere between like

01:06:39.200 5 to 10% or 5 to 8% for a healthy individual, no matter where you're living. And so being depleted or

01:06:46.160 having nothing show up is a really stark difference. Oh, it's that profound a difference. The unhealthy

01:06:51.640 person has none of it. The healthy person is 5 to 10%. Yes. At 5 to 10%, does that make acromantia the

01:06:57.560 most predominant species in the gut? I think it is among the most. I actually don't know that there's

01:07:04.380 a current tallying of winners because there's such a difference from person to person in terms of the

01:07:09.700 composition of the microbiome. Now, you said something a second ago that's very interesting,

01:07:13.640 which is you said, regardless of where a person lived, if they're a very metabolically healthy

01:07:20.500 individual, I'm paraphrasing a little bit, they're about 5 to 10% acromantia. But presumably that also

01:07:25.880 means it's independent of diet because you can point to people who are on radically different

01:07:32.580 diets who are metabolically healthy. So how is that explained? You can have people who are on an

01:07:38.580 all plant diet who are metabolically healthy and presumably eating a hundred grams or more of fiber

01:07:46.400 a day, plenty of food, and you can have somebody on a carnivore diet. And I've seen people on these who

01:07:54.420 are very metabolically healthy. I think it's an impossible diet to adhere to for someone like me

01:07:58.440 who are not eating a gram of fiber, but by all measurements are metabolically healthy. And then

01:08:04.520 of course, everything in between. So given the sensitivity of this flora to diet, how do we make

01:08:11.420 that explanation? I'm trying to think about any studies that really got at the heart of this.

01:08:18.280 Really the study that you would want to do is you want to take one of these individuals,

01:08:21.460 measure their microbiome, and you'd want to put them on a series of different diets and just see

01:08:25.500 like, is this something that is not changeable? Like no matter what you eat or what diet you're on,

01:08:30.820 this particular strain stays high.

01:08:32.980 In which case you would argue that the microbiome is a readout of their metabolic health,

01:08:39.500 not the cause of their metabolic health.

01:08:42.220 Could be. I was going to go in a slightly different direction, which is to say that

01:08:45.680 perhaps the host themselves play a much larger role in the composition of your microbiome than we

01:08:51.500 really understand. So as you change your diet, you would expect your microbiome to change. But if

01:08:56.560 there's something about the host, that part never changes, that might be the influence. To be honest,

01:09:02.760 I just don't think we know.

01:09:04.380 Wow. I mean, that's interesting. And by the way, it's possible that these studies haven't been done

01:09:08.700 across a broad enough dietary spectrum. So even though there's geographic diversity, maybe there

01:09:15.840 isn't as much dietary diversity. And that would be interesting as well, I suppose. For example,

01:09:21.120 not that we could go back and do this, but if you go and look at the Inuit who prior to the

01:09:26.340 adoption of Western food, were eating a seemingly ridiculous diet and yet were quite metabolically

01:09:32.880 healthy. Yeah. It does make me wonder, if I go back and look at this too, I mean, there are tribes

01:09:38.840 in South America that have been relatively untainted by anything outside of where they are. And so I know

01:09:46.260 that there are definitely microbiome studies that are happening in some of those tribal groups. That

01:09:50.160 could be sort of the beginnings of this kind of a study.

01:09:53.140 Quickly, the difference between a probiotic and a prebiotic.

01:09:57.020 Oh boy. Yeah, the vocabulary lesson. I'll introduce another one that's become popularized,

01:10:01.180 which is the postbiotic. So the microbiome, when we talk about it, is all these bacterial and yeast

01:10:07.180 strains. That's the probiotic. So the probiotic is the living organism itself. Prebiotic is the food

01:10:13.760 that feeds those organisms. So we talked about fibers and inulin and polyphenols and things like

01:10:18.780 that. Those are prebiotics. So prebiotics are the food. The probiotic is the organism itself.

01:10:23.800 And then what these organisms produce or what they secrete is now being called the postbiotic.

01:10:28.280 And so you would call butyrate would be a postbiotic. Yeah. And then maybe one more term

01:10:33.760 people might start seeing is symbiotic. And all symbiotic refers to is you've mixed two or more

01:10:38.640 of these things together, the pre and the probiotic together, or the probiotic and the

01:10:41.940 postbiotic. So a symbiotic has multiple. When people are consuming yogurt, they're consuming a

01:10:48.840 prebiotic, presumably. It's the bacteria in it that they're trying to. Yeah. There are lactobacillus

01:10:53.660 bacteria that are in that yogurt that stay alive in the context of yogurt. And so that's really what

01:10:59.340 you're consuming. And what is the perceived, believed, or realized efficacy of consuming

01:11:07.000 massive amounts of lactobacillus and bifidobacter? I think the most well-documented and even reported

01:11:16.880 from consumers' impact of consuming lactobacillus and bifidobacterium probiotics in general and

01:11:22.660 yogurts is around GI symptoms. So things like gas and bloating and diarrhea and constipation,

01:11:29.880 a lot of people report. And there have been studies showing both sides of it. But basically,

01:11:34.560 there have definitely been studies showing and people reporting they have better GI when

01:11:38.220 they're consuming these probiotics or these yogurts with probiotics.

01:11:42.240 If you go and buy yogurt off the shelf, how much lactobacillus is naturally within that?

01:11:48.360 I think they add it in. So it varies. It's not something that naturally occurs in yogurt.

01:11:52.740 I think it is naturally occurring in yogurt, but all the things that you're buying off the shelf,

01:11:55.580 they're also supplementing with additional lactobacillus. The real question to ask yourself

01:11:59.480 when you're buying yogurt is how much sugar is in it? Yeah, for sure. If you're buying fruity-flavored

01:12:04.740 yogurts. By the way, I want to come back and talk about sugars in a second. But I want to go back to

01:12:08.640 kind of this probiotic. How much, what is the dose effect, right? So I know that if you look at a

01:12:14.380 bottle of pick your favorite probiotic, it usually uses something called CFUs, colony forming units.

01:12:22.060 Can you explain what those are? Yes. Some brilliant marketer decided that that was going to become

01:12:27.060 the metric, the name of the game for probiotics. So colony forming units, remember back to seventh

01:12:32.220 grade biology where you might've been given a Petri dish and you had to like swab your mouth or swab

01:12:37.640 your hand or put your finger on it and then you see what grows. That's basically that tool. So you

01:12:42.460 take your pill or whatever, your yogurt, and you basically spread it out on a Petri dish and then

01:12:47.840 you count how many colonies form and that gives you a number. So you'll say, well, gee, this pill has

01:12:53.560 10 to the 9th colony forming units in it. So some marketer decided that that's the most important

01:12:59.680 thing is the number of colony forming units and I have 10 times more colony forming units than

01:13:04.220 somebody else. Maybe that's interesting, but it's less relevant than the function of what's

01:13:08.460 happening in your pill. And moreover, it only gives you one piece of data about what's in that

01:13:13.740 pill. So when you do that, the only thing you get to see are what's able to form a colony. But

01:13:19.580 actually in almost every supplement, every probiotic out there, the majority of what's in that pill is

01:13:25.000 dead probiotic. And you don't see any of that when you use this technology. There's a different tool

01:13:30.840 that can be used called flow cytometry. And essentially what you do is you take your capsule,

01:13:36.220 you put it into this flow cytometer, and what it pops out, its readout is live cells, dead cells,

01:13:41.920 and kind of in between cells. It's based on a staining of the membrane. And so it tells you

01:13:46.180 which of these cells is viable, like the membrane is really intact, which of them have a compromised

01:13:50.640 membrane, and which ones are kind of somewhere in between. And now you know exactly what's in your

01:13:56.300 pill. Because even if you could have the same number of live cells, it turns out that those dead

01:14:00.220 cells and those in between cells, they actually have a role to play. They have a function in there.

01:14:04.200 So these so-called postbiotics, that's what those guys are. And so you don't really know what's in

01:14:08.840 your pill unless you're using flow cytometry versus colony form. And when you do flow cytometry,

01:14:13.460 what do you stain for? Which surface receptors or molecules are you staining for? I don't know the

01:14:19.320 answer to that. I think there's two that we stain for, but I actually don't remember.

01:14:22.200 Okay. And it varies by bacteria, I'm assuming. Yeah. Well, there's some of these common ones among

01:14:25.720 almost all bacteria. Okay. So lacto, it's basically able to use oxygen and not use oxygen. So obviously,

01:14:32.040 if you're going to produce it, you're going to produce it under the conditions of oxygen because

01:14:35.160 it's cheaper and easier. I hear things about some of these things need to be refrigerated,

01:14:38.840 some don't. What's the specificity of that? Yeah, it's really about stabilizing something

01:14:44.120 that's meant to be inside the body. And so I think that a lot of these lactobacillus and bifidobacterium

01:14:49.460 strains, because we've been manufacturing them for so long, we figured out how to get them to be

01:14:54.440 stable through this process. So just to maybe take a step back in what is the process, you take these

01:14:59.900 strains and you're basically growing them into these big vats. If you've ever been to a brewery or a

01:15:05.680 vineyard, you see these big vats. Sort of the same in these manufacturing plants. You grow these strains

01:15:09.880 in a culturable media. And then when they get to a certain density, optical density, you then harvest

01:15:16.300 them. So centrifugation is one of the most common ways. You basically spin down the cells, get rid of the

01:15:22.220 media. And now you have this kind of paste of cells. How slowly do you need to spin them to

01:15:27.080 prevent them from dying? Lactobacillus and bifidose, at least the ones that are in the market

01:15:31.300 today, they're pretty hardy. You can spin them awfully fast and be done with this whole process

01:15:34.780 pretty quickly. And there's other strains that are a lot more sensitive. So you have to do this

01:15:38.820 process much more slowly. And so I think that depends on the strain. But you essentially throw away the

01:15:44.060 media, you take this paste of cells, and then you freeze dry them. And so that gets them into a

01:15:49.960 powder form. And once they're in a powder form, they tend to be pretty stable. And then you could throw

01:15:54.780 them in a pill, and you're off to the races. What's the yield on that? How much loss do you have

01:16:01.780 in live bacteria? Assuming you start with 100% live in the paste, when you just go through the act of

01:16:09.240 freeze drying them, what's your yield? That is where everybody dies. I mean, basically, a lot of times

01:16:15.400 you're losing like 90% of your cells just through that freeze drying process. It's a pretty harsh

01:16:19.920 process. So what people do is you try to figure out what cryoprotectants you can add that are not

01:16:25.680 going to harm the bacteria on the other side of it, but that they're going to help them through this

01:16:30.260 cryo phase. How does freeze drying actually work? It's both pressure and temperature. There are these

01:16:36.660 instruments called lyophilizers, and a lot of them use these sort of flat pans. And so you would put

01:16:42.560 your paste, it's like baking. You'd put your paste in this flat pan, you slide into the lyophilizer,

01:16:47.380 and then over a period of time, it's basically using temperature and cold, the sublimation process

01:16:54.100 of removing the liquid from the cells directly into a gas. Again, we're going back to seventh

01:16:57.820 grade biology. And on the other side of that, you have this powder, and that powder may be stable

01:17:01.940 at room temperature. It may still need to be refrigerated. But what you're trying to figure out

01:17:06.420 is what are the additives that you can incorporate that help it get through this freeze drying process

01:17:10.580 and remain viable. But the things that have to stay refrigerated, even after that process,

01:17:14.520 they're still relatively unstable. So you still have to keep them in the refrigerator.

01:17:17.780 For us, for example, when we first started making acromantia, it absolutely had to be refrigerated.

01:17:23.100 I mean, within hours, it would die. Meaning once you freeze dry it, you sort of shut its metabolism

01:17:28.320 down completely. And then the minute it gets even a little bit warmer, it basically warms enough to

01:17:34.680 the point where it's no longer cryogenically preserved, but there's no substrate for it and it dies.

01:17:39.720 Is that the actual mechanism by which it dies?

01:17:42.240 That's how it dies. Exactly. Because there's no substrate for it. And so it can't handle the heat.

01:17:46.360 And so it dies. The most important part is you do want it to be able to get through this freeze

01:17:51.240 drying process. And of course, everybody wants shelf stable product. But the most important thing is

01:17:55.720 that when you ingest this strain, it actually makes its way to the gut microbiome, is able to

01:18:01.760 reconstitute and to perform its functions, secrete the proteins it's supposed to secrete, secrete the

01:18:06.900 small molecules it's supposed to secrete. And so when you do this process, every time you make a

01:18:10.960 mod where you're like, oh my gosh, we just improved the viability 4x. You just like go to the other

01:18:16.400 side and say, okay, now I got to put in a human and see if I got the same output. So it's a pretty

01:18:20.540 lengthy process.

01:18:22.160 By the way, do you think that that 90% loss is typical on the most common strains that are used

01:18:28.000 commercially?

01:18:29.040 No. And I mean, we even grow some of those strains and you can maintain higher viability. You're never going

01:18:34.060 to get 100% through, but I think it's more like on the order of between 50 and 80%. So you're

01:18:39.140 definitely above the halfway mark.

01:18:41.100 Okay. And then how many of those make it into the person? So in other words, once they're reheated,

01:18:47.040 how many of them die along the way?

01:18:49.340 A lot of it depends on the encapsulation that you're using. So if you have these enteric-coated

01:18:56.680 capsules, you can make it through the stomach acid, and then you can use these time-release capsules so

01:19:01.460 that it takes a certain amount of time and you're making some assumptions around GI transit. And then

01:19:06.460 there's kind of the least expensive version, which probably all falls apart in the stomach and very

01:19:11.400 little of it gets to the actual destination.

01:19:13.680 So it wouldn't make sense to be drinking a probiotic in a liquid then?

01:19:18.100 You could get a lot more efficacy if you took it in a pill format. Part of it too is we talked

01:19:22.800 about these lactobacillus that are in yogurts. Part of it is that if you just have to get just on the

01:19:26.500 other side of the stomach and the small intestine, that is a different thing than trying to get all

01:19:31.280 the way to the gut microbiome. And is that where it seems that lacto and bifido need to be seated

01:19:37.980 is just outside the stomach? I don't know that the answer to that is known, but you can actually find

01:19:43.100 them all along the tract and even into the distal colon. And what is their functional output?

01:19:49.160 Do they make butyrate? Lactobacillus tend to produce lactic acid. Lactate besides that, yeah.

01:19:53.800 Bifidobacterium, various of those strains can produce short-chain fatty acids. They need to

01:20:00.700 be paired with secondary fermenters in order to get to the final butyrate.

01:20:06.280 Now, lactate is also a great substrate fuel for enterocytes, correct?

01:20:12.240 Yes. And so I think maybe that's why these kind of exist all along the track.

01:20:16.400 So in that sense, is the primary function of the lactobacillus to make food for the gut?

01:20:25.000 I don't think we know exactly the answer to that, but definitely the products of those

01:20:29.880 lactobacillus strains, there are a bunch of other strains that are dependent on having

01:20:33.560 lactobacillus.

01:20:35.140 Is it also in that 5% to 10% prevalence?

01:20:37.720 I actually don't know the answer to that.

01:20:39.100 Okay. Lots of companies out there are selling lactobacillus probiotics. You should put them in

01:20:44.780 the fridge and you should hope that their manufacturing process is such that you're

01:20:49.620 getting 80% instead of 10% of what they claim.

01:20:53.220 Yeah. This is where I think our government could maybe have an important role, which is to

01:20:57.720 put some guidelines and requirements around labeling. I'll give you an even more crazy thing.

01:21:03.040 Even the CFU, let's say you think CFU is a perfectly fine metric. The CFU that gets put on

01:21:07.760 that label could be the CFU at the time of manufacture, or it could be the CFU, if I say this thing's got a

01:21:13.960 two-year shelf life stability, it could be the CFU at the two-year time mark. And you're really

01:21:18.840 depending on, is this company going to tell me the higher number, or are they going to try to

01:21:24.400 legitimately tell me what the thing is at the end of the shelf life?

01:21:27.140 And is CFU by definition always alive? Because presumably if they form a colony, they were alive.

01:21:33.520 Yeah. Yeah.

01:21:34.640 Are there any recommendations one would make about how many CFU you need of each of those two bacteria?

01:21:42.520 Well, I think people tend to kind of converge around this like, oh, you need billions.

01:21:47.160 I don't know. I mean, I guess maybe I would ask the question in a slightly different way,

01:21:50.280 which is what are you trying to do?

01:21:52.940 What problem are you trying to solve?

01:21:53.720 What problem are you trying to solve? Exactly. So if you've got a problem you're trying to solve,

01:21:56.560 I generally think most people should operate under, I just need the minimum viable product

01:22:00.380 to solve my problem. So for you, that might be 10 billion. For somebody else, it might be 10 to the

01:22:06.400 seventh. Unfortunately, I think because of marketing, a lot of people are taking things and they don't even

01:22:11.260 really know why they're taking them. Well, it gets difficult when you don't know what the problem

01:22:15.600 is that you're trying to fix. And I think that therein lies a big challenge, which is what am I

01:22:20.980 objectively fixing? Now, it could be something symptomatic, as you said. It could be, boy,

01:22:25.420 if I take a probiotic, my GI symptoms vanish. I'm not bloated. Great. So just titrate to the point

01:22:30.740 where that's true. So going back to antibiotics for a second, for most people, taking antibiotics is

01:22:36.940 almost always oral. But of course, if you're in the hospital, it's not uncommon to take an

01:22:41.040 intravenous antibiotic. So for example, if you're having surgery, it's not uncommon right before the

01:22:46.840 surgeon cuts your skin for them to administer an intravenous antibiotic, usually something that

01:22:52.200 covers for skin flora and that reduces the risk of surgical wound infection. Do intravenous antibiotics

01:22:58.980 also have the same obliterating effect on the gut biome that oral ones do?

01:23:03.340 I have not investigated that, but these clostridium difficile infections arise a lot of times when

01:23:10.020 people are getting antibiotics that they're taking prior to surgeries.

01:23:13.900 I'm sure now that I think about it, people do get C. diff colitis from IV antibiotics.

01:23:17.740 That's a good point. So what do we know about the ideal response to helping your gut out when you

01:23:27.280 take antibiotics? This past year, I had to do two courses of Augmentin, which is a pretty powerful

01:23:35.020 antibiotic. Kicking and screaming, I did it, but I had a pharyngeal abscess and that's a no messing

01:23:41.020 around type of infection. And so I'm on steroids and Augmentin trying to avoid surgery. Luckily I did,

01:23:50.000 but truthfully, I did nothing after the fact and I seem to feel okay. Am I just lucky or should I have

01:23:55.320 done something? We know that you have an ecosystem in your microbiome. You take an antibiotic that

01:24:01.840 pretty much kills everybody. And then over some amount of time, you get a new microbiome from the

01:24:06.720 food, the treating, the environment you're in. And for a lot of people, that new microbiome is the same

01:24:11.480 as your old microbiome. But for a lot of people, it's not. And one of the really interesting things

01:24:16.660 are studies that have been done in kids where they show that kids who are systematically on antibiotics

01:24:21.100 later on in life are more prone to things like obesity, type 2 diabetes, even things like ADHD,

01:24:27.860 celiac disease. And so there's something, I think, in those developing years where

01:24:32.420 post-antibiotic, you've got a different microbiome than pre-antibiotic. So that's just something to

01:24:37.800 consider on the other side of an antibiotic. If ever you were going to clean up your diet,

01:24:41.900 that's a good time to do it because you're starting with a new blank slate of your microbiome.

01:24:46.380 And I think what happens for a lot of people is post-antibiotic, they're actually starting to

01:24:50.840 feel better and feel well. And then they're craving foods that maybe they haven't been

01:24:54.540 able to eat. And so you maybe aren't giving yourself the best shot at a good microbiome on

01:24:58.100 the other side. I used to believe that taking a probiotic during an antibiotic course seemed

01:25:04.260 illogical because the antibiotic is just going to kill that probiotic. But there have been studies

01:25:08.400 coming out showing that, I think in particular, this study that's really compelling showed that if you

01:25:13.700 did a fecal microbiome transplant after taking the antibiotics or, you know, at the tail end of that,

01:25:19.720 or if you were taking probiotics during the time of taking antibiotics, that somehow on the other

01:25:25.260 side of that, you had a, what we'll call a healthy gut microbiome. So you're able to have some of these

01:25:31.040 functions that we've talked about, the development of short-chain fatty acids, you've got acromancy in

01:25:35.440 there. And so the idea here is that even though that antibiotic is killing these strains,

01:25:41.080 you might be doing some kind of a seeding at maybe very low undetectable levels that's allowing

01:25:46.960 you to have a healthier microbiome on the other side of the antibiotics. So I guess if I were going

01:25:53.320 to go on antibiotics, I would probably double down on the probiotics that I'm taking. But again,

01:25:58.840 I think on the other side of that, really coming in hard and strong with a high fiber diet would be

01:26:04.320 your best chance of reconstituting with really good microbes.

01:26:07.540 Maybe a naive question, which shows how little I understand microbiology.

01:26:12.300 Antibiotics seem very specific. When you take Augmentin, you're really targeting a certain set

01:26:18.040 of antibiotics. Oh, this really targets this gram-negative strain, and this really targets

01:26:23.480 these gram-positive bacteria. And that's why there are so many different types of antibiotics,

01:26:29.000 right? I mean, it's not just the class of drugs like cephalosporins. You have the first gen,

01:26:32.600 the second gen, the third gen. And at one point, I actually remembered what they all did,

01:26:36.360 which I don't anymore. But the point is, there was remarkable specificity. So why is it that if

01:26:42.720 you just take one antibiotic for your skin infection, which presumably would be like a

01:26:47.880 first-generation cephalosporin, it obliterates your gut microbiome, which by the way, has nothing

01:26:54.160 in common with the bacteria on your skin. But we just described that these are all highly anaerobic

01:27:00.640 bacteria versus these guys that are totally aerobic. Why does an antibiotic even mess with

01:27:06.760 the gut biome?

01:27:08.140 Yeah, I think antibiotics are a lot more broad spectrum than maybe people or the physicians

01:27:13.360 that are recommending them know. I mean, yes, there's the gram-negative and gram-positive,

01:27:17.660 but really the name of the game for people who are manufacturing and producing antibiotics is

01:27:22.500 it's actually more useful to create a broad-spectrum antibiotic because you want your antibiotic to be

01:27:28.640 the one, the go-to of choice from a doctor. And if it is one that can tackle most different kinds

01:27:35.480 of infection, you're going to become the more popular antibiotic. So, and this is actually one

01:27:39.240 of the arguments in the antibiotic world, which is that antibiotics over time have become

01:27:43.240 more and more broad spectrum. They kill more and more different things, whereas you really want to go

01:27:48.500 in this opposite direction, which is-

01:27:49.780 You want precision antibiotics.

01:27:50.680 A lot more precision around them. There's all these phage therapies and things where people are

01:27:54.920 trying to get really specific, but that's new. Most of the antibiotics out there, they decimate

01:27:59.380 your microbiome. They don't kill everything, which is why you can have these things like

01:28:02.580 these C. diff infections.

01:28:03.980 Do you have a sense of just quantitatively, if you took a gram of stool from me now and looked at how

01:28:08.340 many bacteria were in there, and then you took a gram of stool from me after I was on an antibiotic

01:28:12.860 for 10 days, like Augmentin, very broad, powerful antibiotic, what's the log fold reduction in

01:28:18.540 viable bacteria?

01:28:19.660 I actually don't know the answer to that. That's probably published in lots of places.

01:28:24.780 Yeah. Interesting. Okay. So what I'm hearing you say is the best thing you can do if you're on an

01:28:31.260 antibiotic is, quote unquote, take advantage of the fact that you're starting with a new team

01:28:36.420 and go out of your way to eat the best possible diet. And we're not really talking much about

01:28:43.200 specifics outside of fiber, are we? What I would do now knowing this, what you've said is,

01:28:48.080 I already love eating salad and vegetables and fruits. So I'd probably go on the all salad,

01:28:54.600 all fruit diet for like, I don't know, a few days. I mean, that might sound dumb, but

01:28:58.980 like I would go overboard on those things. Are there other things you would be mindful of either

01:29:04.940 eating or not eating during that recolonization week?

01:29:08.240 I think it's primarily, as you said, these high fiber foods, adding the fruits in is important too,

01:29:13.020 because those are a source of fiber, but also polyphenols. So we know that polyphenols are

01:29:17.800 beneficial prebiotics for the microbiome, these certain strains as well. For example,

01:29:22.620 we know that polyphenols consumption results in higher levels of acromantia. I mean, I guess,

01:29:28.400 just avoid the high fat, high sugar foods.

01:29:30.800 Let's be specific about each of those. What is it about dietary fat or sucrose,

01:29:37.760 fructose and glucose that would be problematic?

01:29:40.980 It's not about the detrimental effects of them. It's actually about what they don't contain.

01:29:44.160 I see. So don't eat McDonald's because it's low in fiber, not because it's high in sugar and fat.

01:29:50.260 Yeah, I guess you're right. The high sugar, high fat foods don't tend to be high in fiber. So I was

01:29:53.540 making an assumption there, but it's really about trying to optimize for high fiber, high polyphenol

01:29:57.980 foods. And so anything that doesn't have that is not feeding.

01:30:01.340 Okay. So you don't have to avoid eating your meat, even though it's not a source of fiber.

01:30:04.520 You tend to make sure you better be eating a lot of fiber.

01:30:06.700 Yeah.

01:30:07.080 And polyphenol. Okay. Let's talk about artificial sweeteners for a second.

01:30:10.080 Highly contentious topic. Yes.

01:30:13.660 Recently, along with a couple of my colleagues, we put out a very lengthy piece of content to our

01:30:19.640 premium audience, to the newsletter. It's a 15,000 word treatise on all things related to

01:30:26.200 non-nutritive sweeteners and non-sugar sweeteners that are themselves nutritive. So it's a pretty broad

01:30:31.300 piece. Without recasting the entire thing, it's really clear that there's something going on with

01:30:37.940 non-nutritive sweeteners beyond their caloric or non-caloric impact. In other words, we all

01:30:45.460 understand that sucrose has its four kilocalories per gram. It's broken down into one part glucose,

01:30:52.420 one part fructose. We understand the metabolism of those things, and we understand that if consumed

01:30:56.660 in excess, you have probably some harm beyond just the caloric side vis-a-vis the fructose molecule

01:31:03.680 and not the glucose molecule. But it's now also clear that under isocaloric conditions,

01:31:09.100 high quantities of non-nutritive sweeteners are not entirely benign. And so I guess I'd start with

01:31:15.720 the question of what do we know about how the gut addresses these, if for no other reason because of

01:31:24.360 the fact that these are very foreign molecules in the concentrations we consume them. I mean,

01:31:28.020 we consume glucose and fructose for our entire existence. We're just seeing it in a higher

01:31:33.420 concentration now, but probably not as much of a multiple in concentration as we see aspartame

01:31:40.200 or sucralose. First of all, I'm going to give all the caveats that you are clearly a far deeper expert

01:31:46.040 in all of this than I am. I haven't spent hardly any time thinking about it. I know nothing about the

01:31:50.300 impact on the gut, really. Just the observations clinically about what we see in terms of sugar cravings

01:31:56.520 and other repetitive behaviors and metabolic symptoms, which I'm asking, do you think part

01:32:02.640 of that is manifested through the gut? And there was that very famous paper in Nature a few years

01:32:07.180 ago, which suggested that in mice. Yeah. Well, first of all, there have been a lot of studies

01:32:12.940 done in mice, and we've already talked about the advantages and disadvantages to getting too fired up

01:32:17.680 about mouse studies, especially in the microbiome. But I think that the data that's out there is

01:32:23.380 conflicting around the impact of these non-nutritative sweeteners on the microbiome.

01:32:27.320 And maybe it's because of what you just pointed out, which is these things are not all created

01:32:30.740 equal. And so by lumping them together and doing these studies, that might be causing some of this

01:32:36.660 conflict. I think it's relatively early stage. A lot of studies have been done in animals.

01:32:40.900 There are definitely studies which show that they can have a detrimental effect to some of these

01:32:44.740 beneficial microbes. In my mind, the jury's still out because I think we don't understand the

01:32:48.600 complexity of these different sweeteners. We also don't understand, obviously, the complexity of a

01:32:52.160 microbiome and the adaptation of the microbiome to consumption of these sweeteners. And so,

01:32:57.560 yeah, evolutionarily, maybe these things haven't been around for very long, but again, because

01:33:01.340 how fast- Because of the rapid evolution.

01:33:02.840 You replicate your- I mean, there's bacteria that can metabolize small molecule drugs. They've

01:33:08.080 definitely never seen before. They've never seen, yeah.

01:33:10.260 So I think that is going to be the name of the game is to understand how does your microbiome

01:33:14.800 evolve to these and how does it help or hurt you? And how's that linked to the metabolic pathway?

01:33:19.700 My takeaway is all of that and then layered onto that something you said earlier, which is you can

01:33:26.320 take five people who are the same weight and give them the same dose of digoxin, and they're going to

01:33:31.800 have five different PKs. For those listening who don't understand the term PK, it refers to like the

01:33:37.600 concentration of the drug within their body. In other words, a product of their metabolism.

01:33:41.560 I would say the same is probably true for aspartame, sucralose, saccharin, all of the above,

01:33:47.180 which is I have seen so many cases of people are trying to lose weight, trying to improve their

01:33:54.100 metabolic health, drinking six diet Cokes a day. They're saying, look, I'm getting zero calories in

01:34:00.340 here and nothing will budge. Do me a favor, substitute soda water for the diet Coke for a month. Let's see

01:34:08.180 if it makes a difference and the world changes. I don't know what to make of that because it's

01:34:13.800 anecdotal and I don't have perfect control over the situation. So it's certainly possible that when

01:34:19.500 they started drinking all the Topo Chico and started ditching the diet Coke, that they were

01:34:24.260 also doing 10 other things that changed. So I really don't know, but I've seen it enough

01:34:28.760 in both directions where it works and where it doesn't work. I do wonder if there are individual

01:34:34.760 factors where in that individual for whom it becomes a productive change, there's a lack of

01:34:41.260 symbiosis between the evolution of their bacteria in the high aspartame environment versus the

01:34:47.960 adaptation of another person in the context. My wife drinks diet Coke. I say that like it's somehow

01:34:53.260 a battle, but she freaking loves diet Coke. She probably has like one every other day. I don't know

01:34:58.020 that that's a high or low dose, but it is. She's in great shape. She's as healthy as a horse.

01:35:02.060 If I drink it, I can't stand the taste of it truthfully, but sometimes I'm like so parched

01:35:05.960 and thirsty, I'll drink one. It doesn't seem to cause me any ills. But again, I've seen people

01:35:10.760 and there's clearly an association between its use and otherwise.

01:35:15.160 That would be a super interesting study I think to do where you would basically take a bunch of

01:35:19.500 individuals, get some baseline data around their microbiome, and then you would put them,

01:35:23.140 you would start them either on a bunch of diet Cokes or you would start them on a bunch of just

01:35:27.880 soda water and then measure their microbiome over time. And then I'd do a little washout,

01:35:31.940 period, or maybe just switch them right over, do a crossover and see what happens. Because

01:35:35.240 this is kind of my frustration with a lot of the microbiome studies that are done is

01:35:38.700 they treat it the way that we do drug studies, which is you just have cohorts and you're just

01:35:43.260 comparing them to each other. Really in the microbiome, the person matters. And so crossover

01:35:47.960 designs are going to give you way more information about what's doing what in that person. And then you

01:35:53.220 can start to draw themes about pathways. But that would be an interesting design to do because

01:35:58.260 it could be that at the onset, if there was something about these people who could

01:36:03.760 live off of Diet Coke, and let's assume people prefer Diet Coke over unflavored soda water,

01:36:09.320 if there was something about their starting microbiomes that enabled them to lose weight

01:36:13.780 in that way or be healthy in that way, that could be a solution. So like I'll tell you,

01:36:18.320 there was a very interesting study that was done where they took all these people,

01:36:20.960 they put them all on the same diet, and everybody has experienced this. You don't have to be a physician.

01:36:25.260 You go on a diet, somebody else goes on a diet, one person loses way more weight than the other

01:36:28.680 person. So they were trying to understand, is there something about the starting microbiome

01:36:33.040 that changes the way people respond to, they just did a regular high fiber diet. And they found that

01:36:39.080 if you start out with higher levels of acromantia, not to keep going back to this strain, but it really

01:36:43.880 is this keystone strain for a reason. This study, what they showed was that if you had higher starting

01:36:48.840 levels of acromantia, that was associated with all the metrics of responding better to the diet in

01:36:55.480 terms of BMI, hip to waist ratio, A1C, weight, all of these things, those people did better. And it's

01:37:03.120 correlative. If they had higher starting amounts, they responded better to the diets. So I do think

01:37:08.760 there's something about the microbiome that can help you or hurt you as you go through these

01:37:12.700 nutrition changes. To me, a corollary of what you just said is if your microbiome is suboptimal,

01:37:20.280 which we can define in a number of ways, but let's use one very specific example. If you are

01:37:24.040 completely deficient in or woefully deficient in acromantia, it is harder for you to have a

01:37:30.320 favorable response to a healthy intervention and, or you may be more susceptible to the downside

01:37:36.520 of a less healthy intervention. So on the one hand, you may be more impacted negatively by something

01:37:42.960 like non-nutritive sweeteners and you may be less responsive to dietary corrections. Would you agree

01:37:49.040 with that statement? That's what that data points to. Yeah. That kind of a hypothesis.

01:37:52.620 Let's talk about acromantia then going back to it. It's the species. There are several strains of it,

01:37:57.760 or when you refer to it, are you always talking about the same strain?

01:38:00.760 Most of the work out there has been done on particular strain called acromantia mucinophila.

01:38:05.660 There's other acromantias out there, but when people say the word acromantia,

01:38:09.980 they're really referring to that strain. Okay. How is it grown? You've already alluded to the fact

01:38:15.420 that it's a hardcore anaerobe and you made some insane statement earlier, which is if even one

01:38:22.300 molecule of oxygen is brought in the presence of its culture, it's dead. Yeah. First of all,

01:38:28.880 can you explain to me, I don't remember enough of my microbiology to know why that's the case.

01:38:33.740 I understand why an obligate anaerobe survives without oxygen. I understand that biochemistry.

01:38:41.540 I don't remember the biochemistry of why oxygen is toxic. It must be a free radical thing or

01:38:45.540 something. Yeah. They're insanely sensitive to it. Yeah.

01:38:50.300 Amazing. Yeah. Which is crazy because there's so much oxygen in the air around us, but this is really

01:38:55.400 this compartmentalized body part that we have. But there are these strict anaerobes. You'll see this,

01:39:01.900 especially we talked a little bit about these microbiome therapeutic companies that are

01:39:06.020 developing drugs. Almost every single one of those companies has had to develop their own

01:39:11.020 manufacturing plant because all these next generation strains that we're talking about

01:39:14.820 that are really in the gut microbiome, in the distal colon, they have this same feature,

01:39:19.380 the same issue, which is that you have to create this end-to-end closed system. And so all of us who

01:39:24.320 are developing these sort of next generation strains and trying to do studies with them, we all ended up

01:39:28.700 having to build our own manufacturing plants in order to solve for this precise problem of the

01:39:33.360 oxygen. Okay. So were there people out there trying to grow acromantia without super, super strict

01:39:39.360 controls? Yeah, for sure. There's a company that is touting pasteurized acromantia, dead acromantia as

01:39:48.380 the product that you really want. It's hard to not only grow the strain, but also to maintain its

01:39:52.820 viability on the other side of that as you try to create a product that people can take.

01:39:56.360 There are lots of products out there that supposedly include acromantia, right?

01:40:01.340 No. We have a whole game of whack-a-mole happening at our company for what we're calling

01:40:05.200 fake-ar-man-sia. So these are people who will throw a supplement on Amazon and say,

01:40:10.160 this is acromantia, and you can test all these things. Most of them just have lactobacillus or

01:40:14.420 bifidobacterium in them. So they're not even trying to fake it.

01:40:17.260 False advertising. Just false advertising.

01:40:19.340 Yeah. So walk through the process. So to grow acromantia, you have the same vat,

01:40:24.780 but now what do you have to pump into it? Nitrogen? How do you get rid of oxygen? How do

01:40:29.320 you create a no oxygen environment?

01:40:31.860 You get rid of oxygen by pumping in a competitor, nitrogen. At some point, we should have you come

01:40:36.220 visit the manufacturing site. We'll get you all gowned up and take you on a tour. Essentially,

01:40:40.500 it starts with a freezer master stock of the strain. So we always go back to the freezer stock

01:40:47.020 because of this genetic component. So we have a master stock of the strain, which has been very

01:40:51.760 heavily characterized. And we don't want to just keep growing this thing and then go off of that

01:40:56.860 growth and go off of that growth.

01:40:58.620 I can understand why you would want to start with the master stock because of the profound genetic

01:41:01.860 drift. What do you now do? You grow this in a pure nitrogen environment?

01:41:06.520 You're actually pumping in multiple other gases. Nitrogen is one of the primary ones. And actually,

01:41:11.640 the mix of gases matters. So that's actually part of the secret sauce of growing the strain.

01:41:15.420 The mix of gases matter. And everything is anaerobic. So in the beginning,

01:41:19.320 when you just have the small stock and you're doing the smaller cultures, you're literally

01:41:23.000 operating in an anaerobic chamber. So you've got the gloves going into this chamber and you're

01:41:27.040 working in that fashion. And then there is a physical tube that goes from that chamber into

01:41:33.420 an entirely closed then bag, transports into this bag filled with media, and then you're growing it

01:41:38.580 at these larger scales. Everything is an entirely closed lock system. And you're pumping in these other

01:41:43.840 gases in order to continue to keep oxygen out. Because really, no matter how closed the system is,

01:41:48.180 what we found is that oxygen makes its way in. So one of the most expensive parts of this process

01:41:53.420 is the pumping in all these other gases. And if you come visit our site, you'll see we just have

01:41:58.400 walls and walls lined with these tanks of these other gases that are getting pumped in.

01:42:03.100 And then after that, again, everything through lyophilization has to remain anaerobic. And then

01:42:08.580 once it's freeze dried, now it's in a stable state. So I could leave agromancia powder out on this table

01:42:14.220 for weeks and it would be fine. So it's really just leading up to that freeze dried state.

01:42:20.340 Wow. So freeze dried agromancia, I guess that kind of makes sense because a capsule presumably

01:42:26.440 still allows for diffusion of oxygen.

01:42:28.660 Absolutely. And the freeze dried state, you basically put them into a dormant state. So

01:42:32.980 they're not metabolically active at that moment.

01:42:35.560 And it's so interesting. So you don't have to keep them in a freezer at home. A fridge is good enough,

01:42:40.300 even though they had to be cooled to a much lower temperature, presumably to be freeze dried.

01:42:45.520 They can be warmed to as much as a refrigerator, which is probably what, 38 degrees or something?

01:42:50.620 Yeah. Well, actually, agromancia, the single strain is room temperature stable. So you can

01:42:54.780 actually have those pills at room temperature.

01:42:57.280 And that still doesn't permit enough heating.

01:42:59.600 Yeah.

01:43:00.220 The body temperature is what's necessary to heat it.

01:43:03.120 Even then, when you consume agromancia, it's the hydration that's actually your big problem once

01:43:08.440 you're in a freeze dried state. You're basically trying to keep that guy dormant. And so once

01:43:12.800 it's freeze dried, and when you buy our bottle, you'll see we have desiccant packets in there.

01:43:17.620 We have desiccant line bottles when we do our clinical trials. It's all about keeping moisture

01:43:21.380 out. Now, once they're in a freeze dried state, now once water enters the game, that's what brings

01:43:26.900 them back to life. And then they die because now there's oxygen. So once they're in their capsule

01:43:30.900 form, some people tell me like, oh, I threw out that little packet. I'm like, don't throw that out

01:43:34.540 because that's what's going to help you keep these guys stable.

01:43:38.020 Wow. So what's that process take in terms of time?

01:43:41.380 The time varies for each of the different strains and obviously also the concentration

01:43:45.080 and the number of bottles that you're trying to make.

01:43:46.940 So the people who do this, obviously, this is a high touch industry. They were doing what

01:43:52.480 before this industry? This is something that's done where?

01:43:56.240 You mean before manufacturing probiotics?

01:43:59.100 Yeah.

01:44:00.120 Probiotic manufacturing has been happening around the world for a very long time.

01:44:03.140 Using this high intense anaerobic process?

01:44:06.520 Actually, we built this thing from scratch. So we had PhD scientists and microbiologists

01:44:11.560 who were taking these small scale methodologies and then trying to figure out how do you grow

01:44:17.560 them at larger scale. So in fact, our manufacturing plant is in San Francisco because that's where

01:44:23.000 all of our PhD microbiologists were living. And so we actually have a pretty young team of

01:44:29.460 people that are coming, almost some of them straight out of their postdocs.

01:44:33.080 How are you doing the flow cytometry on the anaerobe without introducing oxygen?

01:44:39.300 The flow cytometer is in an anaerobic chamber.

01:44:42.480 I mean, I used to do flow cytometry every day. They're huge.

01:44:45.740 Yeah. We have custom built anaerobic chambers for all this stuff.

01:44:49.700 Wow. How does the person go in? Are they just putting their arms in?

01:44:53.800 Yeah. You just have these gloves. You stick your arms in.

01:44:56.460 But to prepare for the facts, you're mixing the antibodies, you're putzing around,

01:45:00.680 you're rinsing, you do all that through these little glove chambers?

01:45:03.420 All of it is done through these gloves.

01:45:04.720 Oh, that's hell on earth.

01:45:05.580 And then there's a little box on the side that you open it up, you put your thing in there,

01:45:09.360 you close it, you have to de-oxygenate it.

01:45:11.880 It's like a submarine.

01:45:12.780 It's like a submarine, the double chamber. It is. I think people get pretty adept at using

01:45:17.500 these gloves. I'm terrible at it, but they can do really fine work with them.

01:45:20.660 Wow. Okay. Aside from the fact that Acromantia, from an epidemiologic standpoint,

01:45:26.540 always shows up, how long has your company been selling an Acromantia probiotic?

01:45:32.240 We've only been selling it for a couple of years. I mean, we really spent about a decade doing

01:45:36.740 dev work and R&D work and preclinical and clinical work. I wanted to make sure the thing did what we

01:45:42.300 thought it was going to do.

01:45:43.420 Are you comfortable saying how much money you spent on that R&D?

01:45:46.720 A lot. Our company has raised $150 million and we've only had products in market for about two

01:45:53.520 to three years. So it was a heavy lift to build a manufacturing plant, do all the R&D work,

01:45:59.720 funding preclinical and clinical work. We fund other PIs to do work with these strains. I mean,

01:46:05.160 I think if you want to build a product that has real efficacy behind it, there's a lot of blueprint

01:46:10.520 from pharmaceutical drug industry that we've adopted because the incredible rigor that they have

01:46:15.800 around developing products with efficacy.

01:46:18.420 Yeah. And your industry is not one that's known for rigor. We'd be putting it mildly.

01:46:23.580 Yeah. And this was a really pivotal moment for our company when we really decided to sell to

01:46:29.180 consumers. So we'd always had this idea that the microbiome is really interesting because it's

01:46:34.020 mutable and you could develop things that could help people. There is a lot of evidence suggesting

01:46:38.820 through these fecal microbiome transplants that you can actually help people improve health

01:46:42.120 through microbiome intervention. And that if you followed a drug development rigorous pathway,

01:46:46.960 you could actually identify the subset of that fecal microbiome transplant that could help people.

01:46:52.500 And if you could figure out how to manufacture them, you could really have a product with efficacy

01:46:56.000 here. Now, how do you bring that to market and how do you choose to commercialize that?

01:47:00.940 We always believed that because these are grass, these are naturally occurring strains,

01:47:05.420 that you had an opportunity to bring this directly to the public that would allow you to

01:47:10.000 really democratize the availability of it. So if we launched Pendulum Glucose Control as a drug,

01:47:15.900 we would only be able to sell it to doctors who would prescribe it to people who had type 2 diabetes.

01:47:21.480 Anybody who's aging should be thinking about how to help their body metabolize glucose better. And so

01:47:27.280 we really felt like there was this natural product, things that could deliver efficacy that was

01:47:33.280 measurable that would be important, and then enabling these products to be available to everybody as

01:47:37.400 opposed to only through a prescription. That meant that the consumer market was really the way to go.

01:47:42.520 Also, a lot of people learn information about health on their own. I mean, Dr. Google is the most

01:47:48.100 famous doctor on earth. And so we really wanted to bring it directly to consumers. The downside of

01:47:53.000 that, and we've talked about this too, is it's really hard in this particular space to elevate it.

01:47:59.620 I mean, everybody is a marketing genius that's selling a probiotic. There are people who make you feel

01:48:05.900 like what they're selling is super innovative, and you've never seen anything like this before. But

01:48:10.000 when you really look at the ingredients, it's literally the same thing that everybody else has

01:48:13.920 been selling. And because a consumer is not going to go read clinical trials or necessarily compare

01:48:18.560 your ingredients to ingredients on all the other labels or do all that legwork, it becomes a really

01:48:23.340 different game to play where you're trying to deliver something that provides meaningful health

01:48:27.740 solutions for people. But you're also having to play the marketing game of how do I convince people or use

01:48:33.420 proxies to help them understand this works? Because you're not selling to a bunch of doctors, you're

01:48:38.140 selling to regular people.

01:48:40.500 Well, even if you are selling to doctors, I mean, quite frankly, it's complicated. I mean, it's not like

01:48:44.360 doctors would necessarily understand this. I struggle to understand this. Let's explain what

01:48:48.920 acromantia does to control glucose. You've alluded to this a couple of times now, you've alluded to a

01:48:53.200 product called glucose control. Let's kind of go back to the science of this. There's a clear and obvious

01:48:58.900 correlation as to why acromantia is beneficial, but that doesn't explain mechanistically what it's

01:49:03.400 doing. Do we have an insight into why acromantia is something that lowers an individual's blood

01:49:10.860 glucose or response to a glucose load? Yeah. So we'll do a deep dive into acromantia and what we

01:49:18.080 know today, knowing that we're going to learn all kinds of other stuff. But at a very fundamental level,

01:49:22.860 this is primarily through the GLP-1 pathway. And just to get people excited about that,

01:49:27.180 if you're hearing this and you've heard of ozempic, you've heard of GLP-1. Ozempic is a

01:49:33.500 mimetic of GLP-1. Yes. What we know about acromantia, there's three key things that we know

01:49:40.460 that it does that kind of result in this. So the first thing is that it has a surface protein called

01:49:46.600 AMUK1100 that appears to be able to bind to TLR2 receptors that helps stimulate these L cells. So

01:49:56.200 maybe we'll take one step back, which is that you have these L cells in your gut microbiome at the

01:50:01.140 lining of your gut microbiome. And L cells are the cells that secrete GLP-1. So a lot of people

01:50:06.760 don't know this. Your microbiome is really the guy secreting GLP-1. So you have this beautiful system

01:50:12.040 where you eat food, your microbiome metabolizes it, and right there are these cells that respond to that

01:50:18.140 food by secreting GLP-1. So acromantia has this surface protein called AMUK1100. It also secretes a

01:50:25.380 protein called P9. And P9 binds to these ICAM2 receptors in the L cells, which are also known to

01:50:32.520 stimulate them to produce GLP-1. And then the third thing that acromantia does is it's able to

01:50:37.680 produce a short-chain fatty acid called propionate. And propionate is upstream of butyrate. And so

01:50:43.360 there are a bunch of strains which can take propionate, convert it into butyrate. Butyrate binds

01:50:48.100 to G-protein coupled receptors 42 and 44, also in these L cells, and stimulates GLP-1 secretion.

01:50:54.840 And so these are all ways in which acromantia can stimulate GLP-1 secretion. And by the way,

01:51:00.900 there are only two strains shown to be able to directly stimulate GLP-1. One is acromantia

01:51:05.740 mucinophila. The other is clostridium butyricum. And they actually act together because clostridium

01:51:10.860 butyricum, as its name indicates, is a butyrate producer. And then GLP-1 has multiple benefits,

01:51:16.880 one of which is that you know all this stuff way deeper than I do, but it helps your body know that

01:51:22.160 to secrete insulin, to help you metabolize the sugars that are in your blood after eating a meal.

01:51:26.660 But it also has this other benefit, which is what is becoming really popular now, which is that

01:51:31.360 induces satiety. It makes you feel full. And it does that in two ways that are not totally understood.

01:51:36.640 But one is that it slows your GI transit, which gives you a feeling of fullness. But the other is

01:51:42.320 that it has some kind of a neurotransmitter mechanism that allows you to feel like you

01:51:47.340 don't have cravings. You're full. It makes it a very powerful small molecule. And so what we're

01:51:52.400 doing that's pretty distinct from what these small molecule drugs is doing is the natural way. We're

01:51:57.320 giving you the upstream bacteria that's enabling your body to produce GLP-1. The result of that is that

01:52:03.400 your body will produce GLP-1. If you've eaten food, it'll go back down. It'll produce it again after

01:52:08.260 you've eaten food, but it will raise your levels of GLP-1. So you get these benefits of lowered blood

01:52:13.340 glucose, reduced food cravings, lowered A1C, and the body weight impact.

01:52:19.300 Okay. A lot I want to unpack there, starting with just a quick summary. So you eat food,

01:52:25.160 provided you have a lot of acromantia and you're feeding it the right food, a couple of things

01:52:29.180 happen. It sounds like one of them is not at all related to a production or a secretory product,

01:52:35.780 but rather just this surface receptor. I think you said it was the AMIC-1100 or something like

01:52:41.440 that. Yes. It has a very fancy name. Yeah. And that surface protein by itself just stimulates L

01:52:47.180 cells in the enterocyte. We know that that makes GLP-1. Yeah. And I think that pathway is probably

01:52:53.160 the least well-understood. Short chain fatty acid and the P9 protein have had, I think, a lot more

01:52:58.040 work done on that. So it secretes P9. It secretes propionate. Propionate gets converted to butyrate

01:53:04.160 by another bacteria. Which ones? There is a class of clostridial strains that will do that secondary

01:53:10.600 conversion. And clostridium butyricum is sort of the most well-studied of those. Got it. What

01:53:15.880 about difficile? Is he doing anything good for us there? I actually don't know. We haven't studied

01:53:20.000 whether, and maybe people are afraid to bring that into the lab. Yeah. Yeah. And what does P9 do?

01:53:25.180 Directly stimulates L cells? Yes. It binds to this ICAM-2 receptor on the L cells and stimulates them.

01:53:30.620 ICAM receptors are receptors that are usually involved in, God, if my memory serves me correctly,

01:53:36.860 like kind of an immune response, right? Like an inflammatory response?

01:53:39.960 We have not done studies around that, but for sure that is an area of heavy interest.

01:53:45.500 Yeah. By vague recollection, I mean, we're literally going back 25 years, is that when a person is,

01:53:51.260 for example, septic, you have all of these ICAM modules that lead to vasodilation, leaky capillaries,

01:53:59.440 secretion of monocytes out into the peripheral tissue as macrophages. But it sounds like this is

01:54:05.160 a distinct process here. And maybe I should say too, like, I don't know how much of this is tied to

01:54:10.040 that process, but not only does acromansia stimulate these L cells to produce GLP-1, but acromansia has

01:54:15.740 another role, which maybe is very much tied to this, which is that it helps regulate the mucin

01:54:20.700 layer. And so that mucin layer turns out to be super important for if it gets too thin or if it's

01:54:27.120 too thick, it's a real so-called leaky gut or GI issues that become a result of not having enough

01:54:32.980 acromansia is really that you have too thin of a mucin layer. And so you start to get the ability

01:54:37.760 for pathogens to infiltrate and then also obviously for these molecules to secrete out.

01:54:41.680 And then tell me what's the fate of the butyrate made of the propionate?

01:54:45.480 It binds to these G protein coupled receptors. And that's also what stimulates the L cells to then

01:54:51.900 go release GLP-1. Got it.

01:54:54.680 So sort of multiple, these small molecules that are the signaling molecules for these receptors

01:54:59.340 on the L cells. So all roads point to more acromansia, more GLP-1. What differs from someone

01:55:07.740 taking Ozempic is they just have a mega high dose of GLP-1 all the time. And what you're talking about

01:55:16.680 here is, as you said, kind of a waxing and waning dose of GLP-1 that is more physiologic because it

01:55:23.280 comes with your meal. That's right.

01:55:25.180 So one would not expect this type of intervention to produce the amount of weight loss you would see

01:55:31.380 with carpet bombing somebody with GLP-1.

01:55:33.700 Absolutely not. And one of the other big differences is that the GLP-1s are injected.

01:55:39.500 So it's going right into the bloodstream, whereas it's a microbiome effect. And to your point,

01:55:43.820 you get the waxing and waning, more physiologically relevant, but it's not going to be the same as

01:55:49.000 just hammering a bunch of GLP-1s straight into the bloodstream all the time, all day, all night.

01:55:53.720 So now let's talk about the data. So what was the first study that demonstrated that

01:55:59.040 acromantia could play a role from an intervention perspective from impacting metabolism vis-a-vis

01:56:06.600 blood sugar?

01:56:08.280 Yeah. Some of the earliest studies are really done. I mean, first of all, it was discovered by

01:56:12.260 Dr. Lee Kaplan over at MGH, and he's a bariatric surgeon. And so his initial interest was what's

01:56:19.060 happening to the microbiome when we do this bariatric surgery and came to really be one of

01:56:23.340 the first people to really look at these microbes and discovered that acromantia appeared to be

01:56:28.220 associated with or inversely associated with obesity, and then started doing these in vitro

01:56:32.660 studies to figure out, like, what is it doing? The work really has to go credited back to him in the

01:56:37.140 early 2000s.

01:56:38.640 I want to point out one of the observations that came from his work, which is a really remarkable

01:56:42.660 observation, which is at least as far as the Roux-en-Y gastric bypass, which is a real

01:56:48.560 organization, a reorganization of the plumbing of a person's gut, that you took a person who was obese

01:56:56.000 with type 2 diabetes, and you do a Roux-en-Y gastric bypass on them. And within days of surgery,

01:57:06.220 their glycemic control improves even before they've lost weight. This is a very important point,

01:57:12.300 because nobody would find it that surprising if a person after a gastric bypass loses 100 pounds and

01:57:18.380 their diabetes goes away. That would be a, oh shucks, of course, moment. It's why does their

01:57:23.140 diabetes resolve in the days following surgery when the real weight loss hasn't started? And there

01:57:29.000 have been lots of suggestions to this. One could be the fasting. You're not eating anything a day

01:57:33.900 before surgery or a day or two after surgery. That's a pretty significant fast that depletes all of

01:57:39.840 the glycogen in the muscles and much of the liver glycogen. Is that enough to kickstart them? But it's hard to

01:57:47.060 ignore that you're also completely changing their gut biome by re-plumbing their GI system. So can you

01:57:55.360 say more? Because I'm not familiar with those data in terms of what they saw as far as pre- and post-surgical

01:58:02.520 gut biome.

01:58:03.820 I'm also not an expert in these surgeries. They have been a source. A lot of publications have sort of used

01:58:11.720 them to try to understand what are the key gut microbes out there. And I don't think it answers

01:58:16.760 this question of why do you see such an immediate response sometimes hours after the surgery. Just

01:58:22.700 thinking about the time length of these different pathways to have effect is most likely something

01:58:27.360 hormonal. But I don't think we have the answer to that.

01:58:29.920 It's hard to imagine it's not something related to GLP-1.

01:58:32.800 Yeah. So I don't think we have the answer to that specific question. But definitely that was where

01:58:39.180 Akramansu was born out of. They do these gastric bypass surgeries in, there've been a bunch of

01:58:44.000 animal models. This was some of the very early work really just trying to lay the foundation for

01:58:48.420 the fact that your microbiome could play a role in weight loss or weight gain. And so they would take

01:58:53.860 these microbiomes before and after the surgeries and then they would put them into these mouse models and

01:58:58.580 they would show that you could change the metabolism of the mouse. It's been really hard in the field

01:59:02.940 to take that data, that fecal microbiome data, where you feel like, oh, there's a glimmer of hope here

01:59:08.320 and to distill it down to what are the things that are actually doing that. That's the big challenge

01:59:13.360 of the field. We're not there. So when you say like, oh, what did they find? Well, they found that

01:59:17.080 the whole gamish could change things, but we don't really know what in there is doing that.

01:59:21.940 Yeah. One hypothesis is that Akramansia is doing part of that, at least.

01:59:26.320 Yes. And so the easiest way to test that would be to give people Akramansia.

01:59:30.600 Okay. So you guys did an experiment like that. Who did you do that in collaboration with?

01:59:35.520 I'll point to the clinical trial that's published in BMJ because I think that was probably

01:59:40.120 the most rigorous trial that we've done. And then I'll talk about the downside of running too many

01:59:45.120 trials and things that involve humans. But in that trial, we didn't just have Akramansia. We actually

01:59:49.960 had it in combination with four other strains. The idea here is, as I said, Akramansia can produce

01:59:55.200 propionate, but without the help of another strain, it can't produce butyrate. And so we had this idea

01:59:59.980 of let's include primary and secondary fermenters. Let's not make Akramansia the only primary

02:00:05.440 fermenter. Let's add some more in there. And really thinking about what is the team members that you'd

02:00:10.620 want in here that can metabolize fiber into butyrate, trying to optimize for butyrate production.

02:00:15.280 At the time that we did this study, nobody even knew what P9 was or AMUK1100. We just knew that

02:00:20.200 we thought it was only playing a role in the mucin regulation. We didn't even really know that it was

02:00:24.220 able to stimulate GLP-1 even at that time. So we really thought it was these butyrate producers

02:00:28.940 that are doing that. Which are the clostridial strains?

02:00:31.800 Yep. Which are these clostridial strains? And then also this bifidobacterium infantis strain,

02:00:35.780 which is also a primary fermenter. So we basically did a placebo-controlled double-blinded randomized

02:00:41.700 trial where we had three arms. One was placebo. One was this full formulation of all five team members.

02:00:47.240 And then the third one was a subset that did not include acromantia. And these were in people with

02:00:52.940 type 2 diabetes. Initially, we wanted it to be people who just had type 2 diabetes but weren't

02:00:58.320 on any medication. It turns out that's like almost impossible to find.

02:01:01.320 Yeah, it's a hard study to find.

02:01:02.300 Yes. So then we said, okay, we'll have it with people on metformin. And anyway, a lot of people

02:01:06.160 are on metformin. So if your thing is going to benefit people, it ought to work on top of metformin.

02:01:10.980 Recruiting into clinical trials is hard. It takes a long time. We were a startup company. We're like,

02:01:15.060 we're going to run out of money if we don't expand this out. So then we started including

02:01:18.800 people on sulfonylureas. So finally, we get all the people in trial. It's still a pilot trial,

02:01:23.380 76 people, you know, across these three different arms. And what we found is that compared to the

02:01:27.440 placebo group, the people who were on the full five-strain formulation over a 90-day period saw

02:01:33.300 their A1C go down by 0.6 and their blood glucose spikes go down by 34%.

02:01:38.960 Measured how?

02:01:40.240 That was done using a pretty old school oral glucose tolerance test. They literally came

02:01:44.760 into the clinic, they got their sugar, and then they just had blood drawn every 15 minutes for

02:01:48.340 two hours. And the reason we did that rather than a CGM was because we wanted to use all the gold

02:01:54.820 standard traditional methodologies. We don't want anybody to think that the data was weird.

02:02:00.160 So hemoglobin A1C came down by 0.3%, which is a pretty big drop.

02:02:04.320 0.6%.

02:02:06.040 Compared to placebo, yeah.

02:02:07.120 Compared to placebo in 90 days. And peak glucose level fell by about a third, you said?

02:02:13.420 Area under the curve. So the entirety area under the curve, yeah.

02:02:16.260 Okay. Fell by about a third?

02:02:17.580 Yep.

02:02:18.380 Got it. And that was placebo to five strains. Acromantia plus four strains to do the conversion

02:02:26.320 of propionate into butyrate.

02:02:28.640 We also have strains that do the redundant function of acromantia, which is they'll do

02:02:32.540 that primary fiber into the propionate or acetate.

02:02:36.640 I see. Okay. And then how did the group absent acromantia do?

02:02:41.820 They had some efficacy, but it wasn't like the five strain formulation. So my co-founder

02:02:48.560 and biostatistician would say that was not statistically significant and the data is published so anybody

02:02:54.340 can go look at it. I would say it looked like it was in between the placebo and the five strain

02:02:57.900 formulation.

02:02:58.780 But it could have been very underpowered and that's why you didn't see a difference.

02:03:01.480 It also could have been very underpowered, yeah.

02:03:03.240 So this was a, thinking about this now, not as a scientist, but as an entrepreneur, this

02:03:08.040 was a do or die moment.

02:03:10.100 There's going to be a binary outcome here. Either we were going to be a company on the

02:03:13.060 other side of this or we were probably going to close up shop because it was all in on this

02:03:16.840 trial.

02:03:17.140 What year was that experiment done?

02:03:19.900 So this was maybe 2018. We found it in 2012. All the stuff leading up to this

02:03:27.820 was we had to identify the strains. We had to figure out how to manufacture them.

02:03:30.520 So six years of basic science.

02:03:32.120 Six years of basic science leading up to this trial. And I was already starting to think

02:03:36.720 about, it's very rare that your first clinical trial out of the gates works. This is why we

02:03:42.000 did the three strain and the five strain.

02:03:43.860 Yeah. Why didn't you just do the placebo versus the five strain?

02:03:47.980 Because acromantia was so hard to grow.

02:03:50.220 Oh, so you were hoping to see that the four strain would be great and you wouldn't have

02:03:55.300 to grow acromantia.

02:03:56.340 Yeah. We were hoping we didn't need the diva in there. It turns out we did.

02:04:00.680 So what's interesting scientifically, but would have added too much overhead to the study,

02:04:05.460 and maybe it's irrelevant if the other four are not that expensive to grow, is with acromantia

02:04:09.680 alone, how would that have performed relative to the five strain pot?

02:04:13.940 Yeah, of course, if we had been able to do that subsequently. But I will say that lots

02:04:19.000 of people have been doing work with acromantia solo. And I put vagus odds down that by itself,

02:04:25.500 it's probably not as impactful as a formulation.

02:04:28.440 Because you don't have the assistance in converting propionate to butyrate.

02:04:32.520 Exactly.

02:04:33.280 And then, sorry, dumb question. I think you already addressed this and I've forgotten already.

02:04:38.100 Butyrate is stable. It's a short-change fatty acid. Why aren't we just all mainlining butyrate?

02:04:42.720 There have been a ton of preclinical and, of course, in vitro work showing the massive benefits

02:04:48.100 of butyrate across a bunch of different states, including metabolism. And yet none of them is

02:04:52.400 translated into humans. Even when we try to do like these enemas, these clinical trials,

02:04:57.540 they've never looked the way they did in the animal studies. The animal models are like,

02:05:00.740 blow your mind, blow you out of the water. And the human trials really don't show that.

02:05:04.880 And I think for this particular pathway that we're talking about here,

02:05:08.700 it's a localization problem. So earlier, I mentioned that butyrate is source of energy for

02:05:14.240 all these colonic cells. The issue is when you're delivering butyrate all along that track as this

02:05:19.600 butyrate is basically being absorbed by all these different cells. And where you need it to get to

02:05:23.660 for this pathway is to the gut lining where the G-protein coupled receptor is sitting,

02:05:28.960 bind to that G-protein coupled receptor.

02:05:30.540 So it's not that acromantia just makes butyrate. It's that it delivers it to the place where it

02:05:36.020 needs to be.

02:05:36.800 Exactly. Literally, the physical proximity is what's enabling this to work.

02:05:41.940 Wow. Let's go back to actually something you said a moment ago, which is this trial was done

02:05:45.780 in people with type 2 diabetes. Average hemoglobin A1c coming in was what? Roughly. These are kind of

02:05:51.680 early diabetics. They're in the 7s.

02:05:53.880 Yeah, they were in the 7s. Between 7.5 and 8.2, I think was the range for all the groups. I have to go

02:05:58.500 back and double check that. Is there any sense, and I know you didn't do this in the study because

02:06:02.720 you said it was only 90 days, but is there any plan to redo this to find out what happens if they

02:06:09.280 had stayed on for a year? In other words, could you take someone with type 2 diabetes and actually

02:06:13.880 knock 1.2% off their hemoglobin A1c, which for many people would now take them straight out of the

02:06:20.220 diabetic range? Yeah. We have multiple ongoing studies, and maybe this will get back to kind of

02:06:25.060 the personal story of the entrepreneurship, which is that on the heels of this study,

02:06:29.480 we were like, holy shit, we made something that works. Now we got to figure out how to commercialize

02:06:33.620 it, but before we do that, we should probably replicate the study, or maybe we should do them

02:06:37.400 in parallel. By the way, did you get any pushback from your investors on that?

02:06:41.100 Doing another study?

02:06:42.120 Yeah, because if you're wearing your scientific hat, that's the obvious thing to do because

02:06:46.200 scientists always want to replicate things and make sure and add another question. If you're an

02:06:50.880 investor, you might be like, are you crazy? We just got the answer. It's the best possible

02:06:55.840 answer. Don't ever ask another question. I mean, there's a reason virtually no supplement

02:07:01.660 companies will ever, ever, ever do a clinical trial. They don't want to know the answer.

02:07:07.220 It's true. It's true. Well, there's two things. First of all, we have been super fortunate in having

02:07:13.240 investors that believe in the underlying premise of our company. And the underlying premise of the

02:07:17.900 company is that the supplement space is riddled with tons of products that don't really do

02:07:23.940 anything. And therefore, it's very hard for anybody to own market share. It's actually a

02:07:27.780 highly fragmented space. Sometimes you get a player that comes in, they take market share because

02:07:31.460 they've done something fancy on the marketing side, but then somebody else comes in and takes

02:07:35.720 their market share. You just sort of see this game happening. Our underlying premise was if you

02:07:40.080 could deliver something that actually helped people in measurable ways, you could take that whole

02:07:44.200 market over because now you've created something that works. Well, in order to know if it works...

02:07:48.360 Yeah, you have to have a study and you have to make a claim. So you can now make a claim

02:07:53.260 with a product. That's a very big deal. Are there other probiotic companies out there that can make

02:07:58.660 claims? No, this claim around lowering A1C and blood glucose spikes...

02:08:03.340 You own that claim?

02:08:04.620 Yes. We are the only ones that make that. Well, somebody else could do a study and make that claim

02:08:07.660 too, but they haven't. But currently, nobody else can make it.

02:08:09.280 Nobody else does. The other reason I think our investors got behind it is, first of all,

02:08:13.020 they understand that we're trying to build products that can change lives. And we have

02:08:16.480 investors that have gotten behind game-changing, category-creating products in the world. So

02:08:23.580 Apple and these sorts of companies that maybe didn't have a predecessor to them. And so...

02:08:29.160 That's a very different risk tolerance than biotechnology. Biotech is the graveyard

02:08:34.940 of some great investors. It's true. I don't know what to say. They're behind this mission

02:08:40.380 that the thing has to work. And so therefore, you've got to have these studies that work.

02:08:43.680 And I will say though, I have learned by having products in market and enabling people to run

02:08:48.640 their own studies on themselves, especially in the context where nutrition and your ecosystem,

02:08:54.620 your microbiome are still pretty much a black box of the scientific community. We've been able to

02:08:58.640 garner so much data from our customers who are just sharing this stuff voluntarily with us

02:09:03.140 in order to be able to do better and better product development. So I am actually now a

02:09:08.160 believer that the clinical trial is important because it gives you a very controlled environment

02:09:12.600 to know whether there's a there there. But unless you can change behavior in the real world

02:09:17.940 and have people actually experience benefit that they can come back and tell you about,

02:09:22.300 you haven't made a product that's very interesting at all. And so I think that that part is important.

02:09:27.440 The other part to this story that's important on our strategy is that because we have this

02:09:32.700 clinical data and because a consumer may not be able to go read a paper, but a doctor can,

02:09:38.100 that we go through healthcare professionals, we go directly to them, but then we also go through

02:09:43.400 them to help create credibility or knowledge for consumers. And so if you want to convince a bunch

02:09:50.100 of doctors that the thing you have works, they're going to see more than a pilot. And they're also going to

02:09:55.300 want to see it run by somebody that's not yourself. And so we immediately launched these studies and

02:10:00.400 then COVID hit and every study one by one got brought to its knees because people with diabetes

02:10:06.300 are more prone to COVID complications that that came out a little bit later on, but people were

02:10:10.920 getting COVID and then you're like, well, I don't know what this COVID is doing to the microbiome.

02:10:14.560 And then people were totally not adherent to any protocols. They couldn't come into clinic.

02:10:18.720 So one by one, we lost a lot of money on trials that we were trying to do follow-ups on.

02:10:23.120 And then we just decided, my chief medical officer said, all right, we have to wait until everything

02:10:29.000 is cleared out before we even start another clinical trial because we're not going to lose

02:10:32.820 another dollar. And so literally just in this last year, 2023, is when we felt like there was enough

02:10:39.740 of a handle on this that we could actually start doing clinical trials, but we've done it in a

02:10:43.620 different way, which is now we've sort of built up some momentum. So it's all third-party PIs and

02:10:49.120 academic and clinical institutions that we're just giving free product to, to run these trials.

02:10:53.120 So we'll get that data. I do think it's important if you want to get the medical community behind

02:10:57.700 you. And I think that'll be a differentiator.

02:11:00.120 That means they have to apply for their own grants to fund the study. They're just having

02:11:05.080 a significant part of the cost taken out of it, which is the drug.

02:11:08.580 Yeah.

02:11:09.200 But that's good. So there's interest. There are PIs out there who are like, yeah, I would happily

02:11:13.580 go and seek the grants to do this study, especially given that I can mitigate the drug cost.

02:11:18.260 Yeah. And the government is putting out, you know, NIH grants. There's quite a few of them

02:11:22.180 centered around microbiome interventions.

02:11:24.220 So how many such trials are underway right now?

02:11:26.200 Right now we have about a dozen around the world that are happening. Not all of them are

02:11:30.820 for type 2 diabetes. We actually have several where investigators have their own hypothesis

02:11:34.940 about what these strains are doing. So for example, we sort of talked about the gut-brain axis.

02:11:40.320 We have an investigator that's been funded by the Milken Institute that is looking at the role

02:11:45.060 of acromantia in bipolar disorder. I mean, that's kind of amazing if you really think

02:11:49.360 about it. So you've got basically a dozen PIs around the world on the basis of one relatively

02:11:56.320 small pilot study that was very well done, published in a very reputable journal. And that alone has

02:12:01.860 generated that much interest that they're finally willing to study a commercial product.

02:12:06.200 Well, I don't think it was just that study. There's like over 3,000 publications on acromantia

02:12:11.400 that people around the world have been publishing on, you know, mostly academics.

02:12:15.440 But those studies weren't intervention studies.

02:12:17.480 No, a lot of them are correlative studies.

02:12:19.020 Right. But that's my point. What you did is quite unique in that you finally bridged a gap,

02:12:25.440 which was correlate, correlate, correlate. But until you can test an intervention under blinded,

02:12:34.340 randomized conditions, you can't know if there's an associative link. And you've at least suggested that.

02:12:40.260 Absolutely. And I think that you're right. The fact that we can make it and that it was able to

02:12:45.780 have this efficacy does give people some belief that, okay, well, if I want to go test an

02:12:49.840 intervention, these are the people that have made it.

02:12:52.080 Right. They could never invest in the CapEx to go and make it.

02:12:55.440 No, I mean, our manufacturing plant, it cost us $10 million to make that plant.

02:13:00.360 Yeah. Okay. So how many products do you sell today? Pendulum sells the product that was tested in

02:13:06.780 that trial. Is that the product called glucose control?

02:13:10.000 Yes. So Pendulum glucose control is marketed for people with type 2 diabetes, lowers A1c,

02:13:15.720 lowers blood glucose spikes, and it is the exact formulation that was in that clinical trial.

02:13:20.080 And do you still use that same quality control metric of actually doing flow cytometry on those

02:13:26.380 products?

02:13:27.360 Yeah.

02:13:28.220 I take glucose control. If I looked at that bottle, which I can't say I've looked at and

02:13:32.380 remember, it doesn't say anything about CFUs on the back.

02:13:34.980 It says AFU, active fraction unit.

02:13:37.480 And what does it report it in?

02:13:39.300 It's in the number of cells that appeared in that active fraction. So it's still going to look like,

02:13:45.860 I think it's like 10 to the 8th or 10 to the 9th for most of the strains.

02:13:48.980 Got it. Okay. What other products do you sell?

02:13:51.700 We released Pendulum glucose control. And I would say one of the hallmarks of our business that we've

02:13:56.580 been very proud of is the repeat business. So we get a lot of, once people try it, they really stick with the product

02:14:02.380 We actually got a lot of feedback from people. We were getting amazing reports of, oh, my A1C is

02:14:07.180 lowered, my blood glucose spikes have lowered. When we launched our first version of glucose control,

02:14:12.260 we also offered people free A1C testing every 90 days and a nutrition coach, which we had a team

02:14:18.720 of registered dietitians. So we were like, this is going to be a solution, right? We're going to give

02:14:22.860 you your nutrition coaching. We're going to give you the test that's going to tell you if it's working.

02:14:25.900 How did you offer that for free? You didn't want to make money, I'm guessing.

02:14:29.260 We don't offer it now. Somebody even said, oh, my gosh, everyone should go buy this product right

02:14:34.340 now because these are non-scalable. So at some point, they're not going to offer these things.

02:14:38.580 And I laughed when I saw that. I was like, oh, you don't know. But then eventually I realized,

02:14:41.400 though, that's true. I think that it's still important for us to offer nutrition information.

02:14:47.220 It just doesn't necessarily have to come through a registered dietitian and a personalized

02:14:49.700 one-on-one coaching. And maybe AI will help us get there faster than we would otherwise.

02:14:54.500 And what other things did you hear from customers? So did you see in the real world

02:14:59.940 comparable reductions in hemoglobin A1C?

02:15:03.600 It was all higher. The reports we got back were higher A1C drops, higher blood glucose drops.

02:15:09.360 Do you have a sense of how much selection bias was going into that? Were there a good number

02:15:13.680 of people saying, what the hell, like it didn't get better?

02:15:16.280 No, actually. Moreover, I would say, you might say, well, gee, those people just didn't tell you,

02:15:20.880 but you can see it in the numbers. You can see it in the return purchase rate. And pendulum

02:15:24.860 glucose control is $165 a month. That's a real out-of-pocket expense.

02:15:29.580 I don't really know the space that well, but how expensive are the top-selling probiotics?

02:15:35.960 Normal probiotic is $20 a month. A premium high-end probiotic is $49 a month. So $165 is way out there.

02:15:43.340 I think you've explained why it has to be that expensive. I won't ask you what your margin is on

02:15:47.340 it, but I'm guessing it's not that high given your manufacturing process.

02:15:49.940 Well, what's even worse is that you're teasing me about offering this test and this nutrition

02:15:53.560 coaching and how did you do that? We were not only losing because of those, but we were actually

02:15:56.960 losing on every bottle we sold because it was costing us more to make it than the $165 we were

02:16:01.360 selling for. So it was all kind of a hot mess out of the gates. But I think what we believed was that

02:16:06.440 we knew that as we scaled, a lot of those costs would go down. So just like everything else,

02:16:10.720 if you're buying 10 bottles versus 10,000 bottles, the same bottle costs you a lot less.

02:16:16.080 Can you currently make a bottle for less than whatever that cost is?

02:16:19.240 We now do make it for less than what we're pricing at, but we also have to ship it cold.

02:16:23.720 So there's also that transport.

02:16:25.500 Yeah, that's an important point. It arrives, you get three bottles, which is a three-month supply.

02:16:30.180 It arrives in an ice pack, in a big box, just like if you order Repatha or something like a PCSK9

02:16:36.280 inhibitor, same thing. Shows up cold. You got to put it in the fridge right away. Can't travel with it.

02:16:41.580 I mean, there's some logistic challenges associated with these things.

02:16:44.480 There was definitely complaints. But I think what we wanted to do was to give the product

02:16:48.560 the best chance of success by giving people these tools to be able to help them measure. And then

02:16:52.920 also, a lot of people don't even know what a high-fiber diet is supposed to look like. And so

02:16:56.080 just giving those tools to people. So we got all this great feedback. Efficacy was higher than we'd

02:17:00.760 even seen in the trial. But we also got a lot of complaints. Hey man, how is this $165 a month?

02:17:05.840 Why isn't my insurance covering it? Why does it have to be cold? I can't travel with it.

02:17:09.740 What am I supposed to do in the time that I'm traveling? I don't have type 2 diabetes.

02:17:13.800 We kind of took all that and said, well, let's just run some marketing tests on a lower dose version.

02:17:19.800 That had huge uptake. And so we launched a product this year called Metabolic Daily. And it's

02:17:25.980 literally the same thing as glucose control, but just at a lower dose that allows us to get down to

02:17:29.880 this price point of $49 a month, which feels a lot more manageable for people when they're thinking

02:17:35.200 about, especially if you don't have type 2 diabetes, but you want to help your body metabolize

02:17:39.700 sugars and carbs better and things like that. And that product still has all five strains in it?

02:17:43.940 That product still has all five strains in it. Yeah.

02:17:45.660 So it needs to be refrigerated?

02:17:47.100 All of our products really should be refrigerated. That does maintain the viability for longer.

02:17:51.380 Those two products. And then we also released the single strains that are

02:17:54.640 components of these products. So we had a lot of people who came back to us and said,

02:17:59.480 hey, I'm buying your pendulum glucose control, but all I really want is acromantia. I just want that

02:18:03.600 one strain. We were sort of like, who the hell knows what acromantia is? This is just like some

02:18:07.920 really educated people coming to us. And it's not a real market. So we did a market test. We made a

02:18:13.880 thousand bottles of acromantia. We literally just called it acromantia. And we put on there like,

02:18:18.160 that's for gut health. We made no claims about it. People were coming to us for a wide variety of

02:18:22.340 reasons of why they wanted acromantia from metabolism issues to GI issues to neurological disorders.

02:18:28.740 So we were like, we'll just sell it as just acromantia. That's the value prop.

02:18:32.700 We made a thousand bottles. In less than 10 days, every bottle was gone.

02:18:36.800 And that's an expensive product because it's acromantia, right?

02:18:39.280 Acromantia is one of the hardest guys for us to make. And so that is a product that's one of

02:18:43.200 our best sellers is just pure acromantia. And actually a lot of practitioners use it because

02:18:47.200 let's say they ran a gut microbiome test, a person's low in acromantia. They just want to give

02:18:52.040 them the minimal viable product to boost this strain back up.

02:18:54.680 But we're really on the outskirts of knowledge here. We really don't know what it's doing by itself.

02:18:59.260 Well, we know that it's secreting some of these proteins.

02:19:01.520 But in terms of outcomes, that's more of a stretch, right? At this time.

02:19:05.840 I would say that what we've learned from having it in the market is that people are

02:19:10.520 getting a variety of different measurable impact from it. And one of the most interesting

02:19:16.760 things that we learned is that we had a bunch of people telling us, actually, the reason why

02:19:22.140 I stay on it is because I don't have any more sugar cravings. I went to the Christmas party and

02:19:27.580 didn't eat a single cookie. And I'm usually the guy sitting at the table eating all the cookies.

02:19:30.420 We started to see this theme of sugar cravings. And as we talked about earlier,

02:19:34.640 one of the hallmarks of GLP-1 is improvement in satiety. We just did a little pilot study

02:19:40.120 ourselves. We said, go on this product for 90 days and do this diagnostic test on cravings.

02:19:46.240 Was it a blinded study? Was there a placebo arm?

02:19:48.200 No. So this is a pilot. This is just a proof of concept. And so what we see is that people have a

02:19:55.160 significant reduction in these food cravings. And so what that's now being parlayed into is

02:20:00.680 funding a clinical trial to look at that. And so this is kind of where I think the consumer

02:20:05.140 space is a really interesting place to play because rather than going all in with millions

02:20:09.920 of dollars on a clinical trial based on a hypothesis in your head, you now can put products

02:20:15.060 out there to people, hear back what they're experiencing.

02:20:18.280 Yeah. Generate hypotheses from customer data.

02:20:20.480 Concept data. And then you feel like, okay, a priori, I have some sense this thing's going to

02:20:24.140 work. And you could even get really sophisticated and say, okay, well, maybe I'd like to know

02:20:28.020 what are people starting microbiomes or what are their diet or what are their lifestyles or their

02:20:32.060 demographics to try to understand who might this thing have the most efficacy for.

02:20:36.100 Now, you also haven't done the study, I'm guessing, where for those, look at your pilot study or your

02:20:42.840 76 person study. It would be very interesting if you go back in time, if you had the funding to

02:20:48.140 look at what the constitution of the gut biome was prior to the study, after the study,

02:20:52.600 and if that predicted response. In other words, is the greater the deficit of acromantia at the

02:20:59.740 outset of the study, a predictor of a greater response upon normalization? Because even though

02:21:04.440 the average hemoglobin A1c came down by 0.6, there must have been people for whom it came down by over

02:21:10.880 a percent and people for whom it only came down by 0.2. So do you have a sense of what that relationship

02:21:17.600 is? Yeah. Well, actually in that trial, we did, we got four stool samples from people during the

02:21:22.400 trial. So one baseline, I think it was like 30 days in after the 90 day mark. And then we did a

02:21:28.300 washout period. So you went for a month without taking anything. And then we got a stool sample.

02:21:32.560 And I'll tell you this, we racked our brains about how are we going to get people to provide

02:21:36.740 four stool samples? I mean, this is like a real ask. We wanted the whole sample. So we literally,

02:21:42.520 it was like these cool whip, that's what we gave to people. And they had to literally shit in a

02:21:47.080 bucket, put it in their own freezer until they could get it to the clinic and everything had

02:21:50.600 to stay frozen. And so we had a hundred percent compliance. We had people who dropped out of the

02:21:54.580 study who were still sending us their shit. I mean, it was amazing how much sample people were willing

02:22:00.560 to share. By the way, I remember my brother got back from like, I don't know, he was somewhere in

02:22:06.320 Africa or something. And he had some awful GI bug. And eventually like he had to collect a stool sample.

02:22:10.800 I was like looking at his fridge and there was this bag in there. And I was like,

02:22:14.200 what's that? And he's like, ah, it's just some shit. And I was like, no, seriously,

02:22:16.920 what's that? He goes, ah, it's just some shit. And I was like, dude, if you don't want to tell

02:22:19.640 me that's fine. But he's like, no, no, that's what it is. That's literally what it is. Yeah.

02:22:24.340 That's literally what it is. People are sending us shit. And the reason is, because as I said,

02:22:27.920 we were an early stage startup. I was like, the chances of this clinical trial isn't going to work.

02:22:31.740 I got to know we have to have this microbiome information so that in my head, I thought for sure,

02:22:36.840 if this thing doesn't work. I want to at least know it didn't work. Exactly. Did it get delivered

02:22:40.540 or not? And so we really wanted to see the presence and absence of our strains. But to

02:22:44.160 your point, we also really want to understand, could you do this predictive modeling of who

02:22:48.780 would be better or worse responders? And we couldn't. It turns out that it could be that

02:22:53.840 the N is too small. It could be that we don't know enough about these individuals. But the microbiome

02:22:58.200 alone, or even the levels of acromantia alone, are not enough of a predictor.

02:23:03.640 They're not enough of a predictor of response. And tell me this, did anybody who took acromantia

02:23:09.680 have a significant increase from pre-study to post-study during the 90 days? And then what

02:23:15.160 was the fall off from 90 days through the washout?

02:23:20.260 Every participant in the study showed an increase in all the strains. And so that was very rewarding

02:23:27.800 because we invest a lot in the encapsulation to get the thing delivered.

02:23:30.640 So that was a great proof of concept. We actually managed to deliver something that in theory

02:23:35.780 is undelivered. Exactly. Deliver the goods. And then the washout period was 30 days. So after the

02:23:41.320 90 days, there was a 30-day washout period. And most people lost the strains after that period.

02:23:47.840 But there were about 15 to 20% of participants who were able to maintain their acromantia levels even

02:23:53.900 after not taking the pills anymore. I was exactly about to ask you that question. Is there any chance

02:23:58.860 that those are the people who just consumed the best diets and ate the most fiber after?

02:24:03.000 So this is another decision that may be in retrospect. We didn't do diet logs. And the

02:24:07.380 reason we didn't do diet logs is because I'm told that people are liars. When you do diet logs in

02:24:12.440 clinical trials, what you get back is all the days somebody was good and then big gaps of missing

02:24:16.580 days where they ate something terrible or did something bad.

02:24:19.020 I can almost think of doing a diet log here where the only thing you're asking them to log is fiber.

02:24:23.200 And you give them a really clear sense of fiber. And it's just all we're doing is counting grams of

02:24:28.760 fiber per day. Take a picture of everything you eat that is one of these things on this laminated

02:24:33.980 card. Every time you eat a carrot or a celery or this or an orange, just take a freaking picture

02:24:38.860 of it and tell me how much of it you ate. And that's it. I don't care how many calories,

02:24:43.000 how many diet Cokes. I don't care about anything. Just eat all the cookies you want if you want.

02:24:47.780 I just want to know this. Yeah. I wish I'd met you when we were doing the trial design.

02:24:51.920 So we didn't do that. What we did was we asked people, what are you? Are you an omnivore? Are

02:24:55.680 you a vegetarian? We have all that. There doesn't appear to be a correlation with that. And then we

02:24:59.740 ask people, please don't change your diet. The whole point of this study is we don't want people to have

02:25:03.580 to undergo a behavioral change in order to see an improvement. It should be just the microbiome

02:25:08.140 intervention. But of course, we don't know if people change their diets. I think what we want to tackle

02:25:12.960 now is a much more direct thing, which is to say that you can deliver people meals and they don't

02:25:18.580 have to eat the meal every day. But basically, if you say part of the trial is you're going to be

02:25:22.960 on the product and then we're going to deliver you a, it's basically a high fiber meal. We want you to

02:25:28.020 have it for lunch three days in the week. And then you have your control group where they just get the

02:25:31.640 pills. And so I think by deliberately making sure that people are getting this added fiber, you can sort

02:25:36.780 of compare whether that fiber is helpful. It almost feels like a no brainer that people who will have

02:25:42.240 higher fiber food will do better, but you will measure their microbiomes and we'll see whether

02:25:46.720 the strains are actually even higher for them too. So the next two years really is an interesting time

02:25:52.360 for your company, but more importantly, I think just our scientific understanding of what's going on

02:25:57.140 on the basis of a product your company has figured out how to make. So if nothing else,

02:26:02.820 you guys have figured out how to make something from a manufacturing process that no one else was

02:26:08.280 going to figure out. There's no university that could ever have figured this out because they

02:26:11.180 wouldn't have been able to put the resources in it. No one was going to spend a hundred million

02:26:14.900 dollars to figure out how to make an obligate anaerobe when the maximum NIH grant is $480,000

02:26:22.260 a year. It's not going to happen. And now it enables a bunch of people to ask these questions

02:26:27.480 about what's the impact on depression? What's the impact on ADHD? What's the impact on obesity,

02:26:32.460 type two diabetes? It's super fascinating. In many ways, it feels like the bridge between the diet and the

02:26:37.840 drug. Cause we already have a really good sense of like, it's not like it's rocket science from an

02:26:43.360 efficacy standpoint to get somebody to lose weight by changing their diet. The effectiveness is the

02:26:47.860 problem. It's too hard to implement for most people. At the other end of the spectrum, we clearly know

02:26:53.520 now how to do it with drugs. And that's really changed. I think it really started in 2014 when

02:26:59.200 liraglutide came out, but clearly semaglutide was the game changer. And that's three years ago was when

02:27:05.860 we saw the pivot from semaglutide as just a diabetic drug to an obesity drug. And it's exciting

02:27:11.540 for me to just watch this. And frankly, even look at other questions that haven't been answered yet,

02:27:17.100 like the dose response is more better. Do we know anything about how much is too much? And what's

02:27:22.920 the maximal dose, the minimum effective dose and the median effective dose? Like there's all this other

02:27:28.860 stuff that I don't know, it'll be five years and we'll be sitting here and hopefully having a more

02:27:33.240 interesting discussion. Absolutely. I mean, I think the dosing is an interesting one too, because

02:27:37.560 the right question is, what is the colonization that's happening? Because it's not just about

02:27:42.740 what you're delivering, it's about what's colonizing and it's different from person to

02:27:45.700 person. So I think if we could crack that nut on the dose to colonization ratio and in what context

02:27:51.040 that's better, you start to see improvements. But absolutely, it is the gap between nutrition and then

02:27:56.940 these downstream small molecules. And the microbiome has been a black box. And now we've got some

02:28:01.380 tools here. We're just at the beginnings of it. But I think for us, the big name of the game is

02:28:06.340 how do we get these products into as many people's hands and as many investigators' hands as we can

02:28:11.080 to just create more and more data around what is your starting microbiome? What is your lifestyle?

02:28:18.180 How are these things tied together? And what are the health outcomes that can really come from a

02:28:21.800 different microbiome? Are there any other bacteria out there as specific strains or even species

02:28:28.420 that you think are going to be worth investigating based on the literature today? And we've talked a

02:28:35.940 lot about acromantia and all the reasons why you had so much data to stand on the shoulders of to go

02:28:41.020 down that rabbit hole. Are there others out there that you think offer promise and that maybe in five

02:28:45.460 years, there are many other strains out there that people are taking. So you might have in your fridge,

02:28:50.620 a bottle of acromantia strain two, strain three, strain four.

02:28:54.360 Absolutely. I assume you're not going to say what they are. I'm guessing that's somewhat

02:28:59.100 proprietary at this point. Absolutely. There are other, I think,

02:29:02.140 of these sort of key strains that I'll just kind of allude to the link between our microbiome and

02:29:08.100 our immune response is super interesting. There are other strains. We'll stay tuned. All right.

02:29:13.260 Thanks, Colleen. Thank you so much for having me.

02:29:16.660 Thank you for listening to this week's episode of The Drive. It's extremely important to me to provide

02:29:21.560 all of this content without relying on paid ads. To do this, our work is made entirely possible by

02:29:27.120 our members. And in return, we offer exclusive member-only content and benefits above and beyond

02:29:33.160 what is available for free. So if you want to take your knowledge of this space to the next level,

02:29:37.580 it's our goal to ensure members get back much more than the price of the subscription.

02:29:42.340 Premium membership includes several benefits. First, comprehensive podcast show notes that detail

02:29:48.460 every topic, paper, person, and thing that we discuss in each episode. And the word on the street

02:29:54.080 is nobody's show notes rival ours. Second, monthly Ask Me Anything or AMA episodes. These episodes are

02:30:02.160 comprised of detailed responses to subscriber questions typically focused on a single topic

02:30:07.020 and are designed to offer a great deal of clarity and detail on topics of special interest to our

02:30:12.400 members. You'll also get access to the show notes for these episodes, of course.

02:30:15.900 Third, delivery of our premium newsletter, which is put together by our dedicated team of research

02:30:22.040 analysts. This newsletter covers a wide range of topics related to longevity and provides much more

02:30:27.980 detail than our free weekly newsletter. Fourth, access to our private podcast feed that provides you

02:30:35.280 with access to every episode, including AMA's sans the spiel you're listening to now and in your regular

02:30:41.580 podcast feed. Fifth, the Qualies, an additional member-only podcast we put together that serves

02:30:48.300 as a highlight reel featuring the best excerpts from previous episodes of The Drive. This is a great

02:30:53.860 way to catch up on previous episodes without having to go back and listen to each one of them.

02:30:58.540 And finally, other benefits that are added along the way. If you want to learn more and access these

02:31:03.880 member-only benefits, you can head over to peteratiamd.com forward slash subscribe. You can

02:31:10.420 also find me on YouTube, Instagram, and Twitter, all with the handle peteratiamd. You can also leave

02:31:16.360 us a review on Apple Podcasts or whatever podcast player you use. This podcast is for general informational

02:31:23.140 purposes only and does not constitute the practice of medicine, nursing, or other professional healthcare

02:31:28.120 services, including the giving of medical advice. No doctor-patient relationship is formed. The use of

02:31:34.580 this information and the materials linked to this podcast is at the user's own risk. The content on

02:31:40.680 this podcast is not intended to be a substitute for professional medical advice, diagnosis, or treatment.

02:31:46.080 Users should not disregard or delay in obtaining medical advice from any medical condition they have,

02:31:51.440 and they should seek the assistance of their healthcare professionals for any such conditions.

02:31:55.840 Finally, I take all conflicts of interest very seriously. For all of my disclosures and the

02:32:01.440 companies I invest in or advise, please visit peteratiamd.com forward slash about where I keep an up-to-date

02:32:09.400 and active list of all disclosures.

02:32:25.840 Thanks for being here.

02:32:27.040 Thanks for having me on YouTube, too.

02:32:33.380 I'll see you next week.

02:32:42.280 Bye-bye.

02:32:42.480 Bye-bye.

02:32:43.720 Bye-bye.

02:32:45.620 Bye-bye.

02:32:46.360 Bye-bye.

02:32:49.620 Bye-bye.

02:32:50.240 Bye-bye.

02:32:50.780 Bye-bye.

02:32:52.320 Have a wonderful life.

02:32:53.220 Bye-bye.

02:32:54.800 Bye-bye.

The Peter Attia Drive - December 18, 2023

#283 ‒ Gut health & the microbiome： improving and maintaining the microbiome, probiotics, prebiotics, innovative treatments, and more ｜ Colleen Cutcliffe, Ph.D.

Episode Stats

Length

Words per Minute

Word Count

Sentence Count

Misogynist Sentences

Hate Speech Sentences

Summary

Transcript