Altered Gut Microbiota a Potential Indicator of Early Alzheimer Disease

Gautam Dantas, PhD


In this video, Gautam Dantas, PhD, discusses how the results of his study showed that altered gut bacteria may be an indicator of the earliest stages of Alzheimer disease.

Additional Resource:
Ferreiro AL, Choi J, Ryou J, et al. Gut microbiome composition may be an indicator of preclinical Alzheimer's disease. Sci Transl Med. 2023;15(700):eabo2984. doi:10.1126/scitranslmed.abo2984

For more Alzheimer disease content, visit the disease state hub.


I'm Gautam Dantas. I'm a professor of pathology and immunology here at Washington University School of Medicine in St. Louis. My lab largely studies microbes and microbial communities as they impact human health.

Consultant360: What was the impetus for this research? Why now?

Gautam Dantas, PhD: So the genesis of this study was actually a chance meeting with a colleague and friend of mine here, Dr. Beau Ances, who's a neurologist. And one of the things that his lab studies is various dementias and neurodegenerative disorders including Alzheimer's disease with particular expertise from the neuroimaging perspective. And we just happened to be hanging out actually outside one of our kids' soccer games, my lab studies the microbiome, and while we were chatting, he brought up this idea of the gut brain access that has been getting a little bit of traction. This concept that the bugs within our guts, through signals that they might make through small molecules communicate with our central nervous system. And he mentioned this really interesting concept in the Alzheimer's disease field, which is the idea that because it's a neurodegenerative disorder, it's something that progresses over periods of your lifetime, that there might be signatures of the disease that can be detected in the human well before the symptoms actually manifest.

This is purely on the human side. And he told me about this idea that WashU actually has been at the forefront of demonstrating of preclinical Alzheimer's disease or preclinical AD. And the concept is that AD is a disease that is diagnosed based usually on dementia, so on symptoms. And then people have gone and identified what correlates with those symptoms. And with people who have symptomatic AD, there are some pretty classical indicators now that you can detect in the cerebral spinal fluid and the blood and through imaging in the brain of these plaques and various proteins.

And so what has happened in the AD field is a demonstration that many of those features, many of those kind of molecular hallmarks in the individual that correlate with symptoms can actually be found in some individuals a good 10 or 15 years before symptoms come up. So this is in this preclinical AD stage people who in this case who fortunately been tracked for that period of time because perhaps a familial risk of AD, they've been shown that some of these markers start going up in their blood and in their CSF, a good 10 or 15 years before any symptoms come up.

And the reason that is really critical is you would imagine that with a progressive disorder, the earlier you catch it, the better a chance of doing something about it. So this is all preamble to say this is all stuff that's been done on the human side. And the question that we discussed, Beau and I, Dr. Ances and I, were much like you've got these human factors, could the gut microbiome, could these bugs in the gut be a similar kind of bellwether event? Could they also be correlated and hence could there be a much less invasive and much cheaper way of finding correlates with a much easier to work with analyze stool, right? Everyone poops. And so if you could analyze signatures in the gut if they existed, it'd be a much easier screening tool in the future rather than the important but much more invasive spinal tap and neuroimaging.

So that was the kind of basis of why we went about this. There was a clear of evidence based on the human side that you could track preclinical AD and you could tell early on what the risks were. And we were now looking for yet another correlate from the gut microbiome. And I'd say the one more thing that's important for this is people before us had clearly shown when they compared healthy individuals to symptomatic AD individuals, so the sort of two ends of the spectrum, to show that their gut microbiomes are different. So that's another important other people in the field. So there was already some evidence to suggest that the gut microbiome correlated with the people who are now clearly in one status versus the other. The question was, is that a single inflection point that it only changes at the very end or like preclinical AD from the host side? Is it something that's changing gradually so that you might be able to use it as a diagnostic?

C360: Were you surprised by the results of the study?

Dr Dantas: Well, so it's interesting. Everything is not surprising in retrospect. I would say we were actually quite surprised. You go in with these hypotheses, that's the reason you do any kind of science. But I think an important caveat to bring up here is differences in the microbiome, you have to always ask are they cause or are they effect, right? Are they cause or they just correlation? And what I mean by that in this case is you could imagine a situation where if you compared only healthy to completely symptomatic AD, that the reasons that their microbiomes are different have nothing to do with the etiology of the disease, the cause of the disease. In fact, they could be simply reflections of dietary changes that occur. Because once you get into dementia, everything about you changes. And this has been a critique of microbiome studies that have been done in the gut brain access side of things to say how much of this that you're ascribing to the microbiome is really just a reflection of the fact that people who get to these later stage disease, that their behaviors are completely different. They might be eating completely different things.

So that's the reason we were quite cautious of going in to say, yes, this exists and we believe the data that's out there, but if we want to really evoke the microbiome as something that is actually changing in concert with the disease. This is actually a great litmus test for that. So I would say we were still surprised. As you alluded to, we did provide that evidence and now this sets up the next stage of what we would like to do, which is drill into this a little bit more to still disentangle this idea of correlation versus potential causation. But yeah, long way to say I was at least I will say pleasantly surprised that we found a pretty strong signal and a signal that now we're going to follow up on.

C360: What gaps in our knowledge still remain after the results of your study and what studies would be needed to fill these gaps?

Dr Dantas: So a few sort of top things that we're doing and I'm motivated to do based on this study. One is, as I mentioned, so right now what we've established is that the microbiome is a correlate of these important host factors that can distinguish preclinical ad from healthy. So one of the first things we want to do is those are essentially two snapshots and they're snapshots of different individuals. So we group those individuals together. What we'd like to understand is when we have the luxury of following individuals over time, your best control is yourself at a different time point because you control everything else. So one of the first things we'd like to understand is whether this hypothesis trajectory of change is actually observed within individuals by doing longitudinal sampling. So we are longitudinally sampling all of those individuals who we published on and we have at least a five-year window over which we're following them, piggybacking on studies that were already ongoing, which are basically the basis of the preclinical ad findings.

They're already being tracked in their neurodegenerative markers and their neuroimaging. All of that is being tracked. We're adding on this microbiome conference. So that's the first thing that we'd like to do to validate and verify. What we found is in addition to the group-based differences, some fraction of those individuals will regrettably progress. So some of the healthy people in our cohort would become preclinical because they have a familiar risk for AD and some of our preclinical people will now become symptomatic. We would like to see whether their microbiomes then switch status, if you will, as a way to again corroborate the group-based differences. So that's one. The second is what we did in this particular case from a sort of technical standpoint is we took the stools of these individuals and we sequenced all the DNA in the stool and the process that's called metagenomic sequencing.

And so what this basically does is it gives you the genomes of all of the bacteria that are in the stool sample and that's how we could make these inferences about who they were and potentially what they were doing. But there are other analytical methods that you can and we are applying to this that gives you a slightly higher resolution of not only who's there but what they might be doing towards potentially connecting the bacterial presence to how they might interact with the host. So one of those is to look at what genes actually being expressed. So the DNA is kind of your blueprint, it's what could be there, but you don't know exactly what enzymes they're going up and what metabolites are being made. And so to analyze that just from those same stools, you can sequence now the RNA fraction in addition to the DNA fraction.

So that's what we're doing, we're in the process of doing that, we've gone back to all of those stools and we're sequencing the RNA because our hypothesis is that in addition to being correlates, the microbes might actually be causally linked to the development of AD. And by looking at what those bugs are actually making, we might be able to connect that over to the host. And then accordingly, the next step also is to look at the host themselves. And so we're also in tandem developing some methods, and so this is a little bit more nascent because it's a little harder to do, is from the stool itself we're also trying to look at host signatures, host expression patterns, so maybe inflammatory markers, things that might tell you how, this might actually be that link, maybe the bug is making something bad and that's turning on some inflammation in the gut, which then translocates over to the CNS.

All speculation at this point. But things that because we have those particular samples, we're going back to try to connect those particular dots. And I will say we're very open-minded at trying to be, say we're happy to disprove hypotheses as much as prove them. Because I think one of the nice things with this kind of biomarker work is this has value even if there's no causative link. So that is to say the microbiome could very well be just a readout that it is not doing, it's possible that it's doing nothing to advance the disease, but rather it's responding to the disease.

So maybe as AD develops through its own sort of host neurodegenerative and inflammatory process, those signals could be transmitting to the gut and changing that architecture, while that'll mean that we can't likely use the microbiome as a therapeutic intervention, in that case, you can still absolutely propose to use it as a biomarker diagnostic intervention. So that's where I think our next stage are to very, as I said, open-mindedly asked that question. If there's a causative link, good because then maybe we could come up with ideas of manipulating the microbiome. If we can disprove that hypothesis, not a problem, we can then focus on the microbiome as a sort of cheap alternative for clinical diagnostics.

C360: Using the gut microbiome as a screening tool would be less invasive for the patient and easier to use in clinical practice, correct?

Dr Dantas: That is our hope right now with anything from a basic science perspective, it takes time to translate and takes time to validate. But that is our motivation, at the minimum is that since we found this signal and once we can validate that these signals are replicated in other cohorts and over time we are very enthusiastic about this idea that this could become a rapid and much cheaper screening tool. Now, I will say that ultimately you'll still really confirm stuff, need to look at the host markers, but just as an example, a reasonable PET scan is going to run you $10,000 a scan ish, and there is a hundred percent that you want to do that if you actually want to identify what's going on. But you and I are not going to do that on any kind of regular basis. Our economy would sink if everyone starts doing that, but we'd be much more likely to be like, well once a year, much like you do all sorts of other annual screens, I'll do a stool screen.

And in fact, maybe that's something that just becomes routinely something that's done because there's so many other biomarkers in the stool, it tells you about IBD status perhaps, or infectious agents. And now we just say, oh, and in addition we'll run a scan through to see whether against this database of Alzheimer's disease risk, whether it's something that you should consider doing the other scans for. And so I think that's a huge value proposition that should not engender a large amount of additional testing. Because that's the other thing, we don't necessarily just want to come up with another random test that people have to do and scare you. It needs to be something that is warranted and it's probably something that'll be tailored towards people already have some familial risk.

So the way we see this going as a first stage is potentially be developed as a routine sort of sequencing based screen. But then at the same time as we're doing these mechanistic studies, there is this enticing possibility that it is actually causative, or at least it may not be the only etiological agent, it's unlikely to be, but as long as the microbiome is contributing anything towards the progression of the disease, now that's a modifiable thing. So if either the ability to go in and remove the bugs that are bad, if you will, the pathogenic bugs that might be contributing to AD or maybe vice versa, maybe we find that people who are healthy have enrichment for particular bugs that maybe keep some of the house cellular processes low, then you can imagine a probiotic treatment that helps you ward against. So again, very speculative at this stage, but it's something that's certainly on our radar of things that we're trying to develop.

C360: What were the overall take-home messages from the study?

Dr Dantas: Yeah, I mean, again, I would say it's something where take home message. I think we've covered the eighties specific stuff. I would just say sort of more generally, I think it's quite exciting that we're in this stage where the microbiome field is advancing the go from what was required upfront as a very descriptive field, just to identify the landscape of who is in the forest, if you will, to now being able to use those techniques to begin to correlate with healthy versus disease states, especially this progressive disease states. And with some amount of humility to ask that question, do we find correlates? And when we find correlates, what they might mean? And so I think as a take home, I think we're in the era of the microbiome as an important medical entity. So I remember, I think it was the early 2000s when declared sort of the decade of the brain.

Now it's a little bit silly because obviously it's the century of the brain, but I do think the microbiome is now at that stage where we can very seriously consider this important microbial organ and human organ as something that is certainly a diagnostic feature, something that can correlate with all sorts of different things, and we can cheaply look at it, but actually might be something that we could therapeutically modify. And I think a lot of us are surprised that it's not just the GI diseases, the sort of obvious case. So obviously if you have diarrhea or something that we associate clearly with the gut, then you change the microbiome and something might occur. But now that we think of this network of interactions in the human body, the concept that you might be able to do something about neurodegeneration related to the gut, I think is quite exciting. And I think we are looking at this, I don't know if it's the decade of the microbiome, but certainly in the stage where we're in the years where the microbiome is going to be a big therapeutic and health target.

© 2023 HMP Global. All Rights Reserved.
Any views and opinions expressed are those of the author(s) and/or participants and do not necessarily reflect the views, policy, or position of Consultant360 or HMP Global, their employees, and affiliates.