Indole-3-Propionic Acid: The Metabolite Only Bacteria Make

Your cells can build almost every molecule in the tryptophan family. There is one step they cannot take. A paper published this February shows what sits on the other side of it, and why your gut lining depends on something you cannot make yourself.

Lit Review Friday · Indole-3-Propionic Acid: The Metabolite Only Bacteria Make · Published 2026 · 28 minute read

---

What Does IPA Actually Do?

Butyrate arrives with a reputation. Most people reading this have met it before, in a supplement aisle or a podcast. Indole-3-propionic acid has no such reputation, so it is worth starting with why anyone is looking at it at all.

Here is where the research has gone in the last decade. Every item below is what a published study reported, in the model it used.

Nerve, heart, liver, kidney, brain, gut. That is an unusually wide spread for one small molecule, and almost none of it is in the popular conversation.

Two honest notes before we go further. Almost all of that work is in animals, and a list of places a molecule shows up is not the same as a list of things it does for you. What makes IPA worth an episode is not the length of that list. It is a single fact underneath it, which is that your own cells cannot make this molecule at all.

There is one more detail in the Cell paper worth carrying forward, because it recurs throughout this piece. IPA was produced by Lactobacillus johnsonii working together with Clostridium sporogenes. One organism supplies, another finishes. That pattern turns out to be the whole story.

The Border You Cannot Cross

Tryptophan is an amino acid. You cannot make it. It arrives in your food, mostly in protein, and from there it goes in several directions at once. Some becomes serotonin. Some enters the kynurenine pathway. And some reaches your colon, where the bacteria living there take it apart and rebuild it into a family of molecules called indoles.

For about fifteen years, the story has been simple enough to fit on a slide. You supply the raw material. They do the chemistry. The indoles that end up circulating in your blood are theirs.

That story is now more interesting, and this post is about the specific place where it gets interesting. Because there is a single chemical step in the middle of it that your own cells cannot perform, and the molecule sitting on the far side of that step turns out to be doing something your gut lining depends on.

One reaction. On one side, a molecule you and your bacteria can both produce. On the other, a molecule that only exists in you because something else made it.

Where This Line of Research Started

It is worth pausing on how we got here, because the tryptophan derivatives were one of the first places anyone showed, in mechanistic detail, that a molecule made by gut bacteria could speak directly to a human cell through a named receptor.

In June 2018, two papers ran back to back in the same issue of Cell Host & Microbe. One was a review by Agus, Planchais and Sokol laying out how the gut microbiota governs tryptophan metabolism. The other, by Bhattarai and colleagues from Purna Kashyap's group, did something more specific. They took tryptamine, a tryptophan derivative that gut bacteria produce and that shows up in abundance in human stool, and they showed it acts on the 5-HT4 receptor on the cells lining the colon, changing how those cells move fluid.

That is not a correlation between a bacterium and a symptom. That is a ligand, a receptor and a measurable function. It set the shape that the entire field has been filling in ever since: find the metabolite, find the receptor, find the output.

The reason the tryptophan derivatives are more interesting to me than the short chain fatty acids is a matter of range. Butyrate, which we covered in an earlier episode, is largely consumed where it is made. The cells lining your colon burn it as fuel, and what escapes them is mostly cleared by the liver. It is a local currency. The indoles behave differently. They circulate, they act at low concentrations, and they turn up in immune regulation, in metabolic signaling and in tissues a long way from the gut. They are less like fuel and more like information.

It is worth putting numbers on that, because the gap is larger than the metaphor suggests. Cummings and colleagues did the definitive measurement in 1987, sampling the colon and then portal, hepatic and peripheral blood in the same people. Inside the colon, short chain fatty acids run at roughly a hundred millimolar. By the portal vein, the blood draining the gut, the total is down to about 375 micromolar. After the liver takes its share, 148. Out at the periphery, 79 micromolar, and that figure is all the short chain fatty acids together. The same paper notes that butyrate is taken up preferentially by the colonic epithelium, so butyrate specifically falls hardest of all.

One honest caveat on that comparison. There is no clean consensus figure for human serum IPA in the way there is for short chain fatty acids, so the two numbers above are not a like-for-like measurement. What they establish is the shape of the difference, not a precise ratio.

---

What Zhang and Colleagues Did

The anchor paper for this episode is from Yanan Zhang and colleagues in Shu Jeffrey Zhu's group at Zhejiang University, published in Nature Communications in February 2026. The question they asked is narrow and good: microbial metabolites are known to matter for the gut lining, but how exactly does one of them direct repair?

They started in humans rather than in mice. Working across four case-control cohorts of people with inflammatory bowel disease, they profiled metabolites and found tryptophan metabolism broadly reduced in Crohn's disease compared with non-IBD controls, with IPA specifically depleted. They then confirmed the reduction in stool by high-resolution mass spectrometry.

From there they moved into mechanism, and the toolkit is unusually complete. Germ-free mice, to establish that the microbiome is doing the work. Colonization with Peptostreptococcus russellii, a known IPA producer, to establish that a specific organism can restore it. Conditional knockouts of candidate receptors in intestinal epithelial cells, to work out how the signal is received. A conditional knockout of the downstream enzyme, to establish that the enzyme is required rather than merely correlated. And finally human colonic organoids, to check that the whole thing is not a mouse curiosity.

Which Receptor Does IPA Actually Use?

This is the part that surprised me, and it is the reason the paper is worth an episode rather than a paragraph.

If you have read anything about indoles, you have read about the aryl hydrocarbon receptor. AhR is a sensor inside your cells that reads chemistry from the outside world and adjusts how tolerant the tissue is. Indoles are its classic ligands. The association is so well established that "indole" and "AhR" travel together almost automatically.

Zhang's team checked. They gave IPA to mice in which AhR had been deleted specifically from intestinal epithelial cells, and separately to mice lacking the pregnane X receptor, another reasonable candidate. In both, IPA still raised the target genes, at levels comparable to normal mice. The effect does not run through either receptor.

It runs through PPAR-alpha, a nuclear receptor best known for its role in fat metabolism and as the target of the fibrate drugs. IPA binds its ligand-binding domain directly, at two specific residues, and the team pinned this down several ways: a canonical PPAR response element in the promoter of the target gene, a dose-dependent reporter assay showing that IPA enhances binding for PPAR-alpha and not for the other two PPAR isotypes, and a thermal shift assay confirming the physical interaction. By chromatin immunoprecipitation, IPA increased PPAR-alpha binding to that response element roughly three and a half fold.

There is a real clash here worth naming rather than smoothing over. The 2018 review that mapped this field states that IPA and indole are decreased in mice with chemically induced colitis, and that oral IPA has protective properties in that model, in a discussion built around the aryl hydrocarbon receptor. So the prior literature attributes IPA's benefit in this exact disease model to a different receptor.

Both can be true, and the resolution is the interesting part. IPA is a documented AhR ligand and has been listed as one for years. What Zhang shows is narrower than "IPA does not use AhR." It is that this particular job, in this particular tissue, does not require it. A single molecule can carry more than one conversation, and which one happens depends on the cell you are asking.

How Do Metabolites Actually Reach a Receptor?

Before we follow this any further, it is worth slowing down on what "binds a receptor" actually means, because the phrase covers two completely different events, and this episode happens to contain one of each.

Surface receptors versus receptors inside the cell

Tryptamine, in the 2018 study, acts on 5-HT4. That is a G-protein coupled receptor sitting on the outside surface of the colonocyte. The molecule never enters the cell. It docks on the outside, the receptor changes shape, and a cascade fires inside within seconds. Fast, local, reversible.

PPAR-alpha is nothing like that, and neither is the aryl hydrocarbon receptor. These are ligand-activated transcription factors that live inside the cell. AhR is described in the literature precisely as a cytoplasmic ligand-activated transcription factor. For IPA to act on PPAR-alpha, it has to physically cross the cell membrane, find the receptor inside, bind it, and then that whole complex goes to the DNA and changes which genes are transcribed. Slow, hours rather than seconds, and the output is a change in what the cell is building.

One of these is a doorbell. The other is somebody walking in and editing the blueprint.

Why these molecules have to be small and greasy

A cell membrane is a fatty barrier. Charged or bulky molecules do not cross it without a dedicated transporter. The indoles are small, largely lipophilic, ring-shaped molecules, which is exactly why they can drift through a membrane and reach a receptor sitting in the cytoplasm.

That is not a footnote. It is the reason this family of metabolites can talk to the genome at all, and the reason they work at such low concentrations. You do not need much of something whose job is to flip a transcriptional switch rather than to fuel a cell. Which is the honest contrast with butyrate: butyrate is fuel, consumed in quantity where it is made, and messages are cheap to send.

Why the same molecule shows up with different receptors in different papers

Because tissues express different receptors. A colonocyte, a cardiac muscle cell, a macrophage and a bone marrow stem cell are all running different subsets of the same genome. Hand all four of them the same molecule and you get four different answers, because they are not listening with the same equipment.

This is why the IPA literature looks chaotic on first read. In intestinal epithelium it is reported as PPAR-alpha. In colonic macrophages, PPAR-gamma. In CD4 T cells, PPAR-beta/delta. In cardiac muscle, AhR leading to PPAR-alpha. In bone marrow stem cells, PPAR-gamma being suppressed rather than activated.

And the microbes are not only supplying the ligand

This is the part that changes the relationship from delivery into something closer to co-regulation. Microbial metabolites change receptor signaling itself, not just receptor occupancy. Byndloss and colleagues showed in Science that stripping out the butyrate-producing microbes reduced epithelial signaling through PPAR-gamma, and that the downstream consequence was a change in the oxygen and nitrate environment of the gut lumen.

AhR signaling likewise changes what else the cell expresses, and activated AhR induces the enzymes that metabolize the very molecules that activated it. That is a feedback loop, not a switch.

So a microbial metabolite can act on at least three layers at once: which ligand is present, which receptors that tissue expresses, and how strongly those receptors are signaling. The question is not really which receptor IPA uses. It is which receptor a given tissue is listening with, right now, and what else is changing that.

Which makes Zhang's result less surprising, and more interesting

The Byndloss work is worth holding alongside this paper, because the two describe parallel machinery in the same cell type. Butyrate acts on PPAR-gamma and drives beta-oxidation, which keeps the lumen low in oxygen. IPA acts on PPAR-alpha and drives ketogenesis, which wakes up the stem cells. Same receptor family, same tissue, two different metabolites, two different host programs. Neither metabolite is one your own cells can manufacture.

What Happens Downstream

The gene PPAR-alpha switches on is Hmgcs2, which encodes the rate-limiting enzyme of ketogenesis. Turning it up means the intestinal cells produce more beta-hydroxybutyrate, the main ketone body. And beta-hydroxybutyrate, in this system, stimulates the LGR5-positive stem cells that sit at the base of the intestinal crypts and continuously rebuild the lining.

Restoring either IPA or beta-hydroxybutyrate reduced inflammation and barrier defects in their colitis models. Deleting the enzyme broke the chain. And in human colonic organoids, IPA induced the same enzyme and raised beta-hydroxybutyrate by close to fourfold, which is the result that keeps this from being a mouse story.

There is a detail here I find genuinely lovely, and it took a second paper to see it. Beta-hydroxybutyrate is not only a fuel. In 2013, Shimazu and colleagues showed in Science that it is an endogenous inhibitor of class I histone deacetylases, which is to say it changes which genes your cells can read. It is one of your body's own epigenetic instruments.

So follow the whole line. A molecule that only bacteria can make reaches a receptor in your intestinal cells, turns up an enzyme in your own genome, and the product of that enzyme is a molecule that goes on to alter how your own DNA is read. The microbes are not merely handing you a chemical. They are upstream of your own regulatory machinery.

The Same Molecule Keeps Turning Up

What makes IPA worth a whole episode rather than a single finding is that the field keeps arriving at it from unrelated directions.

A companion paper from the same group, also in Nature Communications this year, follows the identical chemistry from a different angle. There, Blautia coccoides supplies indole-3-lactic acid, other commensals convert it onward to IPA, and the resulting beta-hydroxybutyrate activates a reserve population of stem cells marked by HOPX. The same border, the same downstream fuel, a different stem cell compartment.

Then it shows up outside the gut entirely. Chen and colleagues, in the Journal of Neuroinflammation, report that after ischemic stroke in mice the gut shifts toward high indole-3-lactic acid and low IPA. The precursor accumulates and the product does not get made. Supplying an organism that performs the conversion restored the balance and reduced neuroinflammation through effects on microglia. Read alongside the gut papers, the striking thing is not the brain result. It is that the same reaction, the same bottleneck, is what fails.

And IPA is not the only tryptophan derivative that talks to the endocrine system. In 2014, Chimerel and colleagues showed that free indole acts on the enteroendocrine L cells that release GLP-1, and the result is beautifully strange. Over short exposures indole blocks potassium channels, widens the electrical spike those cells fire, and increases GLP-1 release. Over long exposures it slows ATP production and reduces it. The same molecule, opposite effects, depending on duration. That is worth holding onto whenever anyone tells you a metabolite is simply good or simply bad.

---

The Twist: Some of These Were Never Theirs

Here is where I have to correct something I have said on this show more than once.

In June 2026, AbuSalim and colleagues in Joshua Rabinowitz's lab at Princeton published a paper in Nature Metabolism that asks a question the field had skipped. Everyone had been busy working out which bacteria produce which indoles. Nobody had systematically checked whether the bacteria were the only ones producing them.

They used isotope tracing, which is the right tool because it is the only one that answers the question directly. A metabolite floating in your blood carries no record of who built it. But if you feed a labelled version of the raw material, with heavier carbon atoms built into the tryptophan, you can follow those atoms into whatever gets made from them. Now the molecule has a return address.

They approached it seven ways, including germ-free mice, short infusions that reach the animal's own cells but not the gut bacteria, longer infusions that reach both, a labelled protein only bacteria can digest, and human cell culture with no bacteria anywhere.

The result reorganizes the family. Several indoles that everyone had filed as bacterial products are made by mammalian cells directly, and their levels in germ-free mice were not reduced at all. Indole-3-lactic acid is one of them. Indole-3-acetic acid is another. Your own tissues build them, and the same pattern held in the human samples they examined.

Because the paper does not stop at the deflation. It names what still requires bacteria, and the list is specific. Bacterial metabolism is needed to free the aromatic ring entirely, producing free indole, phenol and p-cresol. And bacterial metabolism is needed to reduce the amino acid the whole way down to the propionate.

Which is to say: indole-3-lactic acid, the precursor, is partly yours. Indole-3-propionic acid, the molecule Zhang's team traced into your stem cell compartment, is not. In germ-free mice it was significantly depleted. It sits on the far side of the one step you cannot take.

One honest refinement, because the paper is careful about it and the distinction matters. IPA in germ-free animals was significantly reduced rather than completely absent, and the authors point out why: there was IPA in the food itself. The molecule that vanished entirely in germ-free mice was indoxyl sulfate. So the precise claim is that bacteria are required to build IPA, not that every molecule of it inside you was made by a microbe. Some of it you ate.

I find this more compelling than the version of the story I had before. The old claim was that your bacteria make the signaling molecules in your blood, and that claim was always too broad, which meant it was one careful experiment away from being knocked over. This is that experiment. What survives it is smaller and much harder to argue with.

Two Factories, Different Rooms

There is a pattern here that I keep coming back to, and it is not new to biology.

Serotonin is the obvious parallel. Most of the serotonin in your body is in your gut, made by enterochromaffin cells, and the microbiome shapes how much of it appears. Reigstad and colleagues showed that short chain fatty acids from bacteria drive that production. Meanwhile the serotonin in your brain is made by your own neurons, from the same amino acid, behind a barrier the gut supply does not cross. Same molecule. Two production lines. Different rooms, different jobs.

The indoles look like the same arrangement, and we simply had not noticed. Some are made in the lumen by microbes. Some are made in your tissues by your own enzymes. Which is a more ordinary and more interesting picture than either extreme: not an outsourced factory, and not a self-sufficient one, but two overlapping systems that evolved alongside each other and now share a chemistry.

And within that shared chemistry, a small number of reactions are still performed by only one party. Those are the interesting ones. Not because they are the most abundant, but because they are the places where the relationship is genuinely load-bearing.

Do Microbial Molecules Actually Reach Distant Tissue?

There is an obvious objection to everything above, and it deserves a direct answer. If host tissue turns out to be making many of these molecules on its own, does anything genuinely microbial ever travel beyond the gut at all?

It does, and the cleanest recent demonstration comes from a different molecular family. Bacterial bile acid amidates are bile acids that gut bacteria have conjugated to amino acids beyond the usual taurine and glycine. The bacteria are unambiguously the ones performing that chemistry, which makes these molecules a clean test case for whether microbial products distribute systemically.

Writing in Cell Systems in 2026, a group spanning the University of California San Diego and Penn State profiled 690 samples across 14 organs, digestive tract contents and biofluids in mice, sampled across the day. They found these bacterial molecules distributed widely through the body, oscillating on a schedule, with the tightest temporal synchronization in the ileal contents and progressively looser rhythmicity out in the periphery.

Worth stating plainly what that study did and did not do. It is a mouse study, it maps distribution rather than demonstrating a function at any of those sites, and it concerns bile acid conjugates rather than indoles. It establishes that microbial-only molecules travel and arrive on a schedule. It does not establish what they do once they get there.

Where This Leaves the Postbiotic Question

If a specific reaction is performed only by particular members of a community, then the practical question is what happens when those members are depleted or gone. Zhang's cohorts point at exactly this. People with Crohn's disease had less IPA, and the tryptophan pathway as a whole was turned down. The obvious responses are to put the organisms back, which is what fecal microbiota transplant attempts and what probiotics attempt, with the significant limitation that a strain arriving in an already disturbed gut has to survive and establish there before it can perform any chemistry at all.

With that said, here is the general reasoning that this whole line of work supports, and that keeps me interested in it.

The chemistry described in this post takes more than one organism. One species supplies the precursor. A different species performs the reduction. Neither one alone gets you to the end of the road. That is a decent argument for thinking at the level of a community rather than a strain, because the emergent chemistry only appears when the ecosystem is actually functioning.

ThaenaBiotic is a full-spectrum postbiotic derived from screened healthy human donors, carrying 10,000+ molecular signals and no live bacteria. The reason we built it that way is the reasoning above: capture what a working community produces, rather than betting on one organism establishing itself somewhere it may not take.

How far does that go? Not as far as I would like yet, and I would rather say so. The interventions in these papers gave purified compounds to animals, at controlled doses, and measured specific readouts. That is a different thing from a supplement, and I am not going to pretend the distance between them is small. The literature is consistent with our thesis. It is not a hypothesis a single trial has proven, and the human trial data needed to settle it is part of what we are working toward. ThaenaBiotic is one tool in your toolbox, not the toolbox.

---

The Honest Limitations

Nearly all of the mechanism is mouse. The receptor work, the knockouts, the stem cell readouts and the colitis models are all in mice. The human contribution is the patient cohorts showing IPA is lower in Crohn's disease, and the colonic organoid result. Those matter, and they are not the same as showing that raising IPA helps a person.

The doses were high, though better characterized than a single number suggests. The main intervention was 200 mg/kg by oral gavage, which is a large amount of purified compound. The paper also tested 20 mg/kg, which produced only modest effects, and a prolonged lower dose given in drinking water over two weeks, which worked. So there is a dose-response here rather than one heroic dose. What remains genuinely unknown is whether the IPA a normal gut community actually produces reaches a concentration that does any of this without supplementation.

Lower IPA in disease does not establish direction. Inflamed intestines are different environments in many ways at once. Reduced IPA could be a cause, a consequence, or both. The mouse work is what supplies the causal arrow, and it supplies it in mice.

The human evidence in the AbuSalim paper is thin, and the authors say so. Two small human cohorts, both consistent with the mouse pattern, and the authors write plainly that larger studies are merited before concluding host metabolism is sufficient to maintain these levels.

Several enzymes on the host side are still unidentified. AbuSalim's team named one enzyme performing the first step. The route to at least one other indole remains, in their words, an open question in mammals and in bacteria alike.

Tolerance signals are not universally good. Tryptophan metabolites acting through the aryl hydrocarbon receptor promote immune tolerance, which is protective in an inflamed gut and unhelpful inside a tumour, where the same axis suppresses anti-tumour immunity. The context decides the meaning. That is a recurring lesson in this series and it applies here.

---

Frequently Asked Questions

What is indole-3-propionic acid and where does it come from?

It is a molecule built from the amino acid tryptophan, which you get from protein in food. Specific gut bacteria perform the final chemical step that produces it. Human cells make several related tryptophan molecules, but not this one.

Do human cells make their own indoles?

Yes, more than the field expected. Isotope tracing work published in 2026 found that mammalian cells produce several indole and phenol molecules independently of any microbiome, and that their levels held steady even when the microbiome was disrupted. A smaller set still requires bacteria.

Why do gut bacteria matter if our own cells make similar molecules?

Because the overlap is partial. Certain chemical steps are performed only by microbes, so the molecules downstream of those steps depend entirely on which organisms are present. That is a narrower claim than the field used to make, and a sturdier one.

How do gut bacteria influence the gut lining?

In the 2026 Nature Communications study, a bacterial tryptophan metabolite switched on a host enzyme in intestinal cells, which raised production of a ketone body, which stimulated the stem cells that continually rebuild the lining. The signal came from outside, the machinery it activated was the host's own.

Do you need live bacteria to get the benefit of a microbial metabolite?

The receptors involved respond to molecular shape, not to whether a cell is alive. In these studies the compounds were administered directly and the effects still appeared, which is what makes the postbiotic question worth asking. Whether a supplement reproduces those results in people is not something these papers tested.

What does it mean for a molecule to bind a receptor?

It covers two different events. Some receptors sit on the outer surface of a cell, and the molecule docks there without ever going inside. Others sit within the cell and act on the DNA directly, which requires the molecule to cross the membrane first. The second kind changes what the cell builds rather than triggering an immediate response.

Why do different studies report different receptors for the same gut metabolite?

Because different tissues express different receptors. The same molecule reaching a gut lining cell, an immune cell and a bone marrow cell can produce three different outcomes, since each is equipped to listen differently. In the research on this particular metabolite, at least three related receptors have been reported across different tissues.

Does tryptophan in food change any of this?

Tryptophan is essential, meaning you cannot make it and must eat it, and every pathway described here begins with it. The Zhang study found that both dietary tryptophan and the presence of specific commensal organisms were needed to sustain the metabolite. The raw material alone was not enough.

---

The Bottom Line

I had the shape of this wrong for a long time, and the correction is an improvement. It was never true that your microbes manufacture every signaling molecule in your blood. What is true is more specific and more interesting: a functioning microbial community performs a small number of chemical operations that your own cells cannot, and the products of those operations are wired into machinery you would recognize as entirely your own.

That is not an outsourced factory. It is a system that grew up expecting company.

---

References

1. Zhang Y, Tu S, Shao X, et al. Microbiota-derived IPA protects against colitis by regulating intestinal HMGCS2-mediated ketogenesis to facilitate mucosal healing. Nature Communications. 2026;17(1). https://doi.org/10.1038/s41467-026-69341-z · FREE FULL TEXT
2. AbuSalim JE, Olszewski K, Youssef S, et al. Host metabolism can produce many indoles and phenols independently of the microbiome. Nature Metabolism. 2026. https://doi.org/10.1038/s42255-026-01550-8 · FREE FULL TEXT
3. Bhattarai Y, Williams BB, Battaglioli EJ, et al. Gut microbiota-produced tryptamine activates an epithelial G-protein-coupled receptor to increase colonic secretion. Cell Host & Microbe. 2018;23(6):775-785. https://doi.org/10.1016/j.chom.2018.05.004 · FREE FULL TEXT
4. Agus A, Planchais J, Sokol H. Gut microbiota regulation of tryptophan metabolism in health and disease. Cell Host & Microbe. 2018;23(6):716-724. https://doi.org/10.1016/j.chom.2018.05.003
5. Shimazu T, Hirschey MD, Newman J, et al. Suppression of oxidative stress by beta-hydroxybutyrate, an endogenous histone deacetylase inhibitor. Science. 2013;339(6116):211-214. https://doi.org/10.1126/science.1227166 · FREE FULL TEXT
6. Zhang Y, Meng J, Tu S, et al. A microbiota-IPA axis facilitates intestinal stem cell-mediated regeneration in colitis through a Hopx-associated program. Nature Communications. 2026. https://doi.org/10.1038/s41467-026-70062-6 · FREE FULL TEXT
7. Chen JM, Zhang C, Yu LL, et al. Microbiota-derived IPA mitigates post-stroke neuroinflammation by inhibiting TREM2-dependent pyroptosis. Journal of Neuroinflammation. 2026;23:47. https://doi.org/10.1186/s12974-025-03660-8 · FREE FULL TEXT
8. Chimerel C, Emery E, Summers DK, Keyser U, Gribble FM, Reimann F. Bacterial metabolite indole modulates incretin secretion from intestinal enteroendocrine L cells. Cell Reports. 2014;9(4):1202-1208. https://doi.org/10.1016/j.celrep.2014.10.032 · FREE FULL TEXT
9. Reigstad CS, Salmonson CE, Rainey JF, et al. Gut microbes promote colonic serotonin production through an effect of short-chain fatty acids on enterochromaffin cells. FASEB Journal. 2015;29(4):1395-1403. https://doi.org/10.1096/fj.14-259598 · FREE FULL TEXT
10. Zhang J, Zhou Y, Mei X, et al. Molecular mechanisms and therapeutic potential of tryptophan metabolism in gut-brain signaling transduction: a narrative review. Journal of Neuroinflammation. 2026;23:131. https://doi.org/10.1186/s12974-026-03763-w · FREE FULL TEXT
11. Byndloss MX, Olsan EE, Rivera-Chávez F, et al. Microbiota-activated PPAR-γ signaling inhibits dysbiotic Enterobacteriaceae expansion. Science. 2017;357(6351):570-575. https://doi.org/10.1126/science.aal3535 · FREE FULL TEXT
12. Rothhammer V, Quintana FJ. The aryl hydrocarbon receptor: an environmental sensor integrating immune responses in health and disease. Nature Reviews Immunology. 2019;19(3):184-197. https://doi.org/10.1038/s41577-019-0125-8
13. Gao J, Xu K, Liu H, et al. Impact of the gut microbiota on intestinal immunity mediated by tryptophan metabolism. Frontiers in Cellular and Infection Microbiology. 2018;8:13. https://doi.org/10.3389/fcimb.2018.00013 · FREE FULL TEXT
14. Liu JR, Miao H, Deng DQ, Vaziri ND, Li P, Zhao YY. Gut microbiota-derived tryptophan metabolism mediates renal fibrosis by aryl hydrocarbon receptor signaling activation. Cellular and Molecular Life Sciences. 2021;78(3):909-922. https://doi.org/10.1007/s00018-020-03645-1 · FREE FULL TEXT
15. Cummings JH, Pomare EW, Branch WJ, Naylor CP, Macfarlane GT. Short chain fatty acids in human large intestine, portal, hepatic and venous blood. Gut. 1987;28(10):1221-1227. https://doi.org/10.1136/gut.28.10.1221 · FREE FULL TEXT
16. Reilly ER, Charron-Lamoureux V, Mannochio-Russo H, et al. Systemic rhythmicity of host and bacterial bile acid amidates in the mouse. Cell Systems. 2026;17(4):101541. https://doi.org/10.1016/j.cels.2026.101541 · FREE FULL TEXT
17. Kim CS, Jung S, Hwang GS, Shin DM. Gut microbiota indole-3-propionic acid mediates neuroprotective effect of probiotic consumption in healthy elderly: a randomized, double-blind, placebo-controlled, multicenter trial and in vitro study. Clinical Nutrition. 2023;42(6):1025-1033. https://doi.org/10.1016/j.clnu.2023.04.001
18. Serger E, Luengo-Gutierrez L, Chadwick JS, et al. The gut metabolite indole-3 propionate promotes nerve regeneration and repair. Nature. 2022;607(7919):585-592. https://doi.org/10.1038/s41586-022-04884-x
19. Jia D, Wang Q, Qi Y, et al. Microbial metabolite enhances immunotherapy efficacy by modulating T cell stemness in pan-cancer. Cell. 2024;187(7):1651-1665. https://doi.org/10.1016/j.cell.2024.02.022
20. Zhou Y, et al. Indole-3-propionic acid protects against heart failure with preserved ejection fraction. Circulation Research. 2024;134(4):371-389. https://doi.org/10.1161/CIRCRESAHA.123.322381 · FREE FULL TEXT
21. Rothhammer V, Mascanfroni ID, Bunse L, et al. Type I interferons and microbial metabolites of tryptophan modulate astrocyte activity and central nervous system inflammation via the aryl hydrocarbon receptor. Nature Medicine. 2016;22(6):586-597. https://doi.org/10.1038/nm.4106 · FREE FULL TEXT
22. Zhao ZH, Xin FZ, Xue Y, et al. Indole-3-propionic acid inhibits gut dysbiosis and endotoxin leakage to attenuate steatohepatitis in rats. Experimental & Molecular Medicine. 2019;51:1-14. https://doi.org/10.1038/s12276-019-0304-5 · FREE FULL TEXT

This post accompanies the Lit Review Friday episode of Learn Something with Thaena.