Skip to main content

Infectious Diseases

Microbiome Connections to Disease Get Stronger

A fair amount of what you read about the human microbiome is hype. There’s no way around it. It’s quite difficult to study this area in a meaningful, reproducible way, and even the best work in the area can only go so far, as things stand now. When differences in (say) gut flora are actually found and worked out, we generally don’t know what the chicken-and-egg relationship between that and human disease might be, or which particular bacteria (or ratio, or blend) is responsible, in either direction.

But just because an area is difficult, or because it has a lot of media noise in it, doesn’t mean that progress isn’t being made. There’s some new work, for example, that suggests that the gut microbiome might have a real connection to multiple sclerosis. That sounds like an exercise in headline-grabbing, but it looks more solid than that. This team found several bacterial types associated with MS, following up on numerous earlier studies, specifically noting that relapsing/remitting MS patients tend to have higher proportions of Acinetobacter (generally rare in gut flora) and Akkermansia, and lower proportions of Parabacteroides. The enhanced species, when put into other animals via fecal transplant, have very noticeable effects on T-cell differentiation and also exacerbate the pathology seen in the widely used EAE rodent model of the disease. The authors also note that one of the Acinetobacter species has already been shown in the literature to produce peptides that mimic sequences in myelin basic protein, which makes you wonder if MS is (at least partly) a misfiring immune response to gut bacteria in general.

That hypothesis has been kicking around for a while, but this sort of work really strengthens it. So does the second paper mentioned above, which looks at 34 pairs of twins, one of whom has MS while the other does not (an extraordinarily powerful and useful data set). This team also found increased Akkermansia in the gut flora of the affected twins. They transplanted gut material from the affected subjects versus unaffected controls into a mouse model (RR mice) that have a mutation causing them to spontaneously develop inflammatory demyelinating disease. Only some of the human-derived bacteria were able to colonize the mouse gut (as you’d expect), but even so, the mice with the transplants from the MS-diagnosed twins showed a significantly higher rate of disease onset.

When they looked more closely at the altered gut microbiota in these mice, they found lower levels of the Sutterella genus, compared to the transplants from human patients. (Interestingly, they don’t seem to have noticed as much difference with Acenitobacter). To be sure, other studies of gut flora in MS patients versus controls have also disagreed on the raised and lowered profiles across taxa, which is one of the things that makes this such a hard area to work in. (One explanation is that the concentration and localization of various species probably vary a great deal across the whole intestinal tract, in ways that are difficult to sample and account for, and stool samples are not necessarily a good proxy).

But overall, there really does seem to be something here, although it’s way too early to start talking about the therapeutic implications (for starters, we’d better be able to agree on just which bacteria are responsible and how they might be having their effects). These studies on bacterial transfer are really doing a lot, though, to address that chicken-and-egg problem mentioned above. If adding in the relevant bacterial population can bring on trouble in this way, that significantly raises the odds for the bacterial trouble being ahead of the CNS phenotype.

These papers dovetail nicely with yet another new paper, from a team at Rockefeller/Mt. Sinai/Sloan-Kettering. They’ve found that (1) commensal gut bacteria are enriched in genes for the synthesis of N-acyl amides, (2) that the compounds thus produced are, in many cases, micromolar ligands for various GPCRs that look quite similar to endogenous human ligands for these receptors and (3) that the GPCRs identified are, in turn, disproportionately localized in the gut as well. Taken together, it would appear that this is a signaling network that has evolved among the bacteria over time which modulates their own environment, and provides a direct mechanism by which changes in gut flora could affect the host organisms (that is, us). These lipid-like-signaling receptors (such as the endocannabinoid one, GPR119) are part of the same family, and can have a number of metabolic and inflammation-pathway effects.

Overall, it looks like the microbiota/disease connections are getting stronger, and getting some more detailed foundation under them. No doubt there are going to be a lot more twists and turns as these stories go on, and I would be very skeptical of anyone claiming (at this early stage) to have a great new therapeutic microbiome breakthrough for something as complex as MS. But the field in general is real stuff, generating some very interesting real results, and is worth keeping an eye on.


Note: All opinions, choices of topic, etc. are strictly my own – I don’t in any way speak for my employer

23 comments on “Microbiome Connections to Disease Get Stronger”

  1. Saumil says:

    There is a really nice paper in Cell linking Fusobacterium nucleatum to recurrent colorectal cancer.

  2. GutDecipher says:

    One of the greatest difficulties that plagues the utility of microbiome datasets is that genetic profiles can differ wildly from functional profiles, almost certainly due to horizontal gene transfer and the incredibly dynamism of most species. In that regard, “just what bacteria are responsible” is not necessarily the same as “which bacteria in the sample are responsible”, further muddying the whole picture.

    When functional data is multiplexed w/ phylogenetic data, we’ll see truly interesting results among diseases states, especially in autoimmune dysfunction (if anyone has an example of a paper where this is done for human disease samples, please let me know). MS could be the first one to fall etiologically speaking, but there are intriguing microbiome hypotheses on other diseases including ulcerative colitis, eczema, and even autism.

    I need to read the second paper you linked, but providing further putative mechanistic basis for the gut-brain axis is truly exciting. Please keep highlighting microbiome-related work!

    BTW love the URL slug for today, very matter-of-fact.

    1. Derek Lowe says:

      Yeah, that was a placeholder title, and I hit “publish” before I’d redone it (!)

    2. Tom Bevis says:

      What little I could understand of this article made me wonder if the same effect could be linked to diseases like Lupus or Crones. I’m glad that smarter people than me are looking into these things!

  3. cynical1 says:

    No, it really is headline grabbing and only that. Face palm.

    I think that if you had read as many papers using the EAE model as I have, you would simply have turned the page and rolled your eyes on this. You can induce, increase, decrease effects on the EAE model using anything under the sun. It’s a pretty crappy model. (Yes, I know, it’s all we have.) If you’re using a model that isn’t using a relevant autoantigen to induce a disease state in an animal that does not have an MS equivalent of its own, I’m going to suggest that you’re going believe whatever you want with the result. So yes, putting a fecal transplant into a trangenic species will effect outcomes in the EAE model. And who cares?

    The other thing is that MS patients typically do not have the same environment than healthy controls or their identical twin. They take drugs – lots of them, they are typically less mobile, diet differences, etc. Who cares about small but significant differences in the amount of gut flora in a small patient sampling? Do they see elevated antibody response in the CSF to these gut flora? What are all those oligoclonal bands in the CSF directed at in MS patients? Gut flora? Really? Are MS patients T-cells cross reactive to a myelin protein (not MBP) and whatever these gut flora are supposed to be producing that would induce an antibody response that crosses the blood brain barrier? Do levels of the gut flora correlate with disease activity in MS patients? Do interferons reduce the T-cell directed response to these gut flora or reduce their level? BTW, we already know that MBP isn’t the autoantigen in MS. If you think the beta amyloid hypothesis isn’t holding up in Alzheimer’s, then the MBP hypothesis in MS is dead and buried six feet under.

    I also do not see how those two papers dovetail into micromolar ligands for GPCRs. I wasn’t aware that most endogenous ligands for GPCRs that “look quite similar” to these N-acyl amides were usually in the micromolar range. What’s that got to do with the EAE model of MS and “small but significant” differences in gut flora in MS patients.

    If this research is so robust, why don’t they just give a fecal transplant from the identical twin to bring them into Food Babe harmony and record the results?

    I’m going to go have a yogurt because it will replenish my gut flora and make my headache and arthritis go away……..

    1. Derek Lowe says:

      EAE does indeed suck as an MS model. But if gut microbiota can affect it that much, there may well be immunological effects in humans that are more relevant. . .

  4. luysii says:

    Along these lines [ Nature vol. 549 pp.. 48 – 53 ’17 ] showed that gut bacteria produce N-acyl amides which bind to a G Protein Coupled Receptor (GPCR) called GPR119 as well. GPR119 is an endocannabinoid receptor. Anandamde is an endocannabinoid which is the N-acyl amide of arachidonic acid and ethanolamine.

    If plants can make compounds to affect human physiology. Why not gut bacteria?

    What is an endocannabinoid receptor doing in the gut — it is regulating metabolic hormones and glucose homeostasis. The gut N-Acyl amides bind GPR119 as tightly as our endogenous ligands

    1. Lane Simonian says:

      Glutathione–a critical antioxidant–is one of the most important “products” of certain gut bacteria.

      1. steve says:

        I think it’s pretty obvious why endocannabinoid receptors would be found in the gut. They’re there in case you eat the funny brownies.

  5. GutDecipher says:

    Although now that I’m looking more into it and discussing with others, there seems to be a lot of suspicion of the transfer data (Fig 3 of the second paper). Any consternation that the ~0.001% relative abundance difference of Sutterella (3B) is truly what’s driving a nearly 40% difference in the rate of spontaneous EAE incidence (3A)?

    1. tangent says:

      Thank you.

      I want a new rule: you can’t cite any work by p-value alone, you must quote an actual metric of interest. Just like making a scientific claim with no source will get you a “citation needed” eyeroll, citing a paper and dropping the effect size should get an eyeroll. (And of course, a paper that doesn’t have an effect size should be cast into the pit during the review process.)

  6. Barry says:

    Acetinetobacter sounds like an interesting target for a recombinant antigen vaccine. You certainly wouldn’t want a traditional vaccine that might induce cross reactivity with myelin. But whom would you target? Even if acetinobacter is causal, the clinical trial would have to be vast to show impact on a relatively rare disease like MS.
    But once the auto-immune attack has been provoked, just eliminating the acetinobacter wouldn’t be expected to show any benefit–so you can’t wait to treat after diagnosis. There’s no prospect that anyone could make their R&D costs back on such a vaccine.

  7. anon electrochemist says:

    $5 says ignoring glycosylation will trash your chance of success in these endeavors.

  8. Barry says:

    And while we’re in the gut biome…what do we understand about parasitic worms lulling the host’s immune system? At least hookworms and flatworms (which are only very, very distantly related) have been shown to alleviate Crohn’s disease. Have we characterized how?

    1. GutDecipher says:

      Something something TRegs

      1. Barry says:

        yeah, that’s the level of “explanation” I’ve seen to date. But if they’re exporting a soluble signal, I want to see it. And its receptor is potentially an important drug target. And since we’ve seen this with two unrelated parasites, it’s likely they’ve converged on this stratagem, possibly with different chemistry.

        1. Philip says:

          Not from the parasites that have been shown to be beneficial in IBD, but it has been shown that parasites can produce and excreted/secreted product (as yet, unidentified) that can stimulate Treg generation through mimicking TGF-beta (using the TGF-beta receptor)

  9. Joe says:

    Dr Thomas Borody, Centre for Digestive Diseases, has reversed MS via an microbiome transplant. Restoring the health and diversity of the microbiome affected three MS patients. They were in wheelchairs, now they are walking. Remarkable.

    1. Anonymous Layperson says:

      Cites for this individuals papers and data sets? There are none? Didn’t think so. Without empirical evidence, claims like this are no better than the folks hawking apricot pits and crystals for cancer.

  10. DrOcto says:

    ”When they looked more closely at the altered gut microbiota in these mice, they found lower levels of the Sutterella genus, compared to the transplants from human patients. (Interestingly, they don’t seem to have noticed as much difference with Acenitobacter).”

    My first thought was that the offending species in terms of an undesired immuno-response was being transferred from Acenitobacter to another mouse friendly bacteria via plasmid transfer.

  11. LiqC says:

    Nomenclature question here, how did the “N-acylamide” end up with that name? They’re fatty acid amides. N-acylamide implies an imide, with two acyl groups on one nitrogen

  12. DLunsford says:

    It’s a great time to be a mouse veterinarian!

  13. Lane Simonian says:

    It is the weekend so instead of giving a link, I will try to summarize the following article.

    “Gut Microbiota, Nitric Oxide, and Microglia as Prerequisites for Neurodegenerative Disorders”

    Certain gut bacteria (via g protein coupled receptors) led to the secretion of the amyloid precursor proteins (via activation of protein kinase C) and to various reactive oxygen and nitration species (via activation of NMDA receptors). All of these can apparently make it to the brain through communications between the enteric and the central nervous systems.

    The authors conclude that various pharmacological and lifestyle interventions such as NOS inhibitors, potassium channel modulators, NMDA receptor antagonists, probiotics, diet, and exercise can be used to rectify the problems leading from aberrant nitric oxide signalling in Alzheimer’s disease.

Comments are closed.