Thaena

Your Gut Is On a Clock
·
Thaena Inc.

You don't have one biological clock. You have trillions. The one in your brain watches the sun. The ones in your gut watch your meals, and they listen to the chemistry your bacteria are producing, hour by hour. New research suggests that when you take a microbiome therapy may matter as much as what you take, and that the active ingredient might not be the bacteria at all.

Lit Review Friday · Learn Something with Thaena · Episode 23 · Published 2026 · Reading time: ~14 minutes

📝 In short
  • Does timing matter for a fecal microbiota transplant? Yes. Cao 2026 showed the same dose given twelve hours apart produced dramatically different outcomes in depressed mice. The early-window, fasted, rest-phase delivery restored mood-related behaviors and brain chemistry. The late-window did not.
  • Why does timing matter? Because the gut's molecular clock controls when the host is receptive. Bacterial metabolites set that clock. Light, food, and sleep are the three inputs that keep the system synchronized.
  • Do you need live bacteria for the benefit? Often, no. Autoclaved (Lin 2024) and pasteurized (Plovier 2017, Depommier 2019) preparations have retained therapeutic activity in human and animal studies. The active ingredient is frequently the metabolite, not the cell.
  • Does fecal microbiota transplantation work for depression in humans? Wang 2026 showed FMT plus escitalopram produced a 71.4% response rate vs. 35.0% for escitalopram alone (p=0.037) at two weeks in adults with depression already taking escitalopram. A 2025 meta-analysis of twelve trials in 681 patients found a substantial pooled effect (Zhang et al., Frontiers in Psychiatry).

🎧 Listen to the full episode: Your Gut Is on a Clock

Available on Spotify (Apple Podcasts, and wherever you listen coming soon)

The Two Clocks You Didn't Know You Were Running

For most of the past century, we have talked about the body's clock as if it were a single object. A pacemaker tucked behind the eyes, set by sunrise and sunset. That pacemaker is real. It is called the suprachiasmatic nucleus, a small cluster of neurons in the hypothalamus, and it does watch the sun.

What changed in the last decade is that we now know the master clock is just the conductor. It is not the orchestra.

Trillions of peripheral clocks live in your gut, liver, pancreas, immune cells. They are not watching the sun. They are watching food. When did you last eat. What did you eat. Whether the gut is empty enough to clear and rest. And on top of all of that, they are listening to the chemistry your gut bacteria are producing, hour by hour. The bacteria themselves oscillate. They migrate along the gut wall in waves. Different species peak at different times of day. They secrete bile acids, short-chain fatty acids, and tryptophan derivatives in rhythms. Those metabolites are how they tell your host cells what time it is.

A 2026 review in Frontiers in Medicine by Chen and colleagues lays this framework out under a name worth remembering: the circadian-microbiome-liver axis. A 2025 paper by Bautista and López-Cortés in Frontiers in Endocrinology uses a parallel term, chronobiome medicine. These are not Thaena hypotheses. They are a peer-reviewed framing of how the field now understands timing, microbiology, and host physiology as a single coupled system. Touitou and colleagues, writing on shift work and gut dysbiosis in Chronobiology International the same year, push the translation further: when the conductors disagree, the music falls apart, and the consequences for human health are measurable.

The Paper That Forced the Question

In April 2026, Cao and colleagues at Lanzhou University published a paper in Brain, Behavior, and Immunity asking a deceptively simple question: if you give a fecal microbiota transplant to depressed mice, does it matter what time of day you give it?

Same dose. Same donor. Same recipient genotype. Twelve hours apart. The answer was yes. Dramatically.

The setup

The team took male C57BL/6J mice, three to four weeks old, and put them through eight weeks of chronic unpredictable mild stress (CUMS), the standard rodent protocol for inducing depression-like behavior. By week eight the stressed mice showed reliable decreases in sucrose preference, increased immobility on tail-suspension and forced-swim tests, and reduced exploratory activity in the open field.

The team then wiped the gut bacteria of those stressed mice with an antibiotic cocktail. Once the gut was clean, they delivered a fecal transplant from healthy donor mice via oral gavage at one of two times of day. One group received the transplant at zeitgeber time 4. The other received the same transplant twelve hours later, at zeitgeber time 16.

Zeitgeber time is clock time relative to lights-on in a standard 12-hour light/dark cycle. ZT0 is lights on. ZT12 is lights off. Mice are nocturnal. They sleep, rest, and barely eat during lights-on, then feed actively once the room goes dark. ZT4 is mid-rest, fasted-leaning, low feeding activity. ZT16 is the active feeding window. The early-day window is when the host gut is empty, quiet, and primed to receive a microbial input. The late-day window is when the gut is busy digesting.

The translation to humans is direct. The rest-phase, fasted, pre-sleep window is when the gut is set up to integrate something new. The active-feeding window is not.

What ZT4 Restored That ZT16 Did Not

The behavioral readouts came in first. The early-window mice came out of their depression-like state. Their forced-swim immobility dropped from roughly 100 seconds back down to about 55. Their open-field activity restored to control-group levels. Tail-suspension immobility fell from 135 seconds to roughly 92, statistically indistinguishable from healthy controls.

The late-window mice improved a little. They did not separate from the depressed group on most measures.

Same bugs. Same dose. Same animals. The clock decided whether it worked.

📊 The Pattern Across Every Endpoint (Cao et al., 2026)

Same bugs. Same dose. Twelve hours apart. Across every measure Cao tested, the early-window transplant restored the system. The late-window did not.

  • Behavior (sucrose preference, tail suspension, forced swim, open field): ZT4 mice came back to control levels. ZT16 mice barely separated from the stressed group.
  • Neurotrophic support (BDNF in prefrontal cortex and hippocampus): ZT4 fully restored. ZT16 did not move it.
  • Neuroinflammation (IL-1β, NLRP3, microglial activation): ZT4 shut it down. ZT16 left it running.
  • Clock genes themselves (Clock and Bmal1 in PFC and hippocampus): ZT4 normalized. ZT16 stayed dysregulated.
  • Serotonin pathway (tryptophan and 5-HT in mood-regulating regions): ZT4 restored. ZT16 did not.

CUMS protocol 8 weeks. FMT dose 1×10¹² bacteria per dose, 0.25 mL fecal suspension daily for one week post-antibiotic. Behavioral assays n=11, RT-qPCR n=4–6, ELISA n=4–6.

Following the chemistry back to the brain

The team went looking for why. They followed the chemistry from the gut, into the bloodstream, and out into specific brain regions.

In the stressed mice, a single bacterium had bloomed. Alistipes, a Bacteroidetes-class commensal, was significantly elevated. Alistipes carries an enzyme that chews through tryptophan, the dietary amino acid your body needs as the precursor for serotonin. When Alistipes overgrows, it eats the serotonin precursor before your brain can use it. The team cross-referenced this against 100 healthy individuals under 40 years of age and 100 age- and gender-matched MDD patients in the GMrepo human gut database. Alistipes was elevated in the human MDD cohort relative to healthy controls. Same pattern, different species.

When the ZT4 transplant landed in the empty, fasted, receptive gut, it normalized the Alistipes bloom. Tryptophan availability in the prefrontal cortex and hippocampus returned to control levels. Tissue-level serotonin came back. ERK phosphorylation, a key part of the BDNF survival pathway, restored. Clock and Bmal1, the central transcription factors of the molecular circadian machinery, normalized. The inflammatory machinery, NLRP3 and microglial activation, calmed down.

The ZT16 transplant moved some of these markers in the right direction, but not all of them, and not as far. The most striking pattern is in the BDNF results. ZT4 restored BDNF in both the prefrontal cortex and the hippocampus. ZT16 did not move it at all. BDNF is one of the most heavily studied molecules in modern depression research. The same dose of the same bacteria moved it dramatically in one window and not at all in the other. Timing is not a footnote here. Timing is the variable.

"Same bugs. Same dose. Same animals. The clock decided whether it worked." — From the episode

The Molecular Bridge: How a Bile Acid Sets a Cell's Clock

If the gut bacteria are talking to the host clock, what is the actual molecule passing the message? In March 2026, Powell and colleagues, working between the Devlin and Thaiss labs at Harvard and Penn, published the most direct answer the field has yet produced. The paper appeared in PNAS, and it isolates a single bacterially produced molecule that grabs the wheel of the host's circadian machinery.

Powell's team built a human colonic cell line carrying a luciferase reporter for hPer2, one of the core clock genes. They screened a focused library of 88 gut microbial metabolites against the reporter, looking for any compound that would shift the period of the cell's circadian oscillation. Out of all 88, one stood out. Lithocholic acid (LCA) lengthened the circadian period in a dose-responsive manner.

LCA is a secondary bile acid. The host liver releases primary bile acids in response to a meal. Gut bacteria modify those primary bile acids in the lower GI tract, producing a family of secondary bile acids. The host cannot make secondary bile acids without the microbiome. They are, by definition, microbial output the host cannot replicate. Powell measured LCA concentration in human cecal contents at 4 to 275 micromolar (avg ~160). The dose-response curve in the cell-line experiments fell squarely inside that physiological range. This is the chemistry your gut is making, at the levels your gut is making it, hitting the dial on your colon's molecular clock.

The mechanism is specific. LCA inhibits casein kinase 1 delta and epsilon, the master kinases that regulate the period of mammalian circadian clocks. It also stabilizes CRY2, a core clock repressor. Together, those two effects lengthen the circadian cycle of human colon cells. A bacterially-made molecule reaching into a host cell and turning the clock dial.

It is not the bacterium that does this. It is the molecule.

💡 An aside: bile acids, again

Andrea has been openly biased about bile acids for years. She predicted, on the Ep17 episode, that secondary bile acids would turn out to be the molecular hand-off between the microbiome and the central nervous system. Powell 2026 is the cleanest mechanistic evidence yet that the bias was earned, not lucky. Bile acids are released in response to feeding. Bacteria modify them. The modified bile acids tell host cells what time it is. The molecule by which feeding talks to your peripheral clocks turns out to be made by the bacteria you feed.

If the Molecule Does the Work, Does the Bacterium Need to Be Alive?

Once you accept that the metabolite is the messenger, an obvious next question follows. If the molecule is the active ingredient, do you actually need the live bacterium that made it?

The honest answer from the literature, accumulated across multiple labs, is that you very often do not.

📊 The Dead-Bacteria Evidence
  • Lin et al., 2024, Int J Mol Sci: The team took Parabacteroides goldsteinii RV-01, a beneficial gut commensal, and autoclaved it at 121 °C for 15 minutes. Autoclaving is unambiguous. After this protocol, the bacterium is 100 percent non-viable. Nothing alive remains. Tested against HCT116 human colon epithelial cells, the autoclaved preparation produced anti-inflammatory effects, including TLR2 receptor activation and NF-κB suppression. The cells could not tell whether the bacterium was alive. The signal was the chemistry on its surface and the molecules it had already produced.
  • Plovier et al., 2017, Nat Med: Pasteurized Akkermansia muciniphila, heat-treated to inactivate the cells, retained metabolic-protective activity in mouse models of obesity and insulin resistance.
  • Depommier et al., 2019, Nat Med: A randomized, double-blind, placebo-controlled trial enrolled 40 overweight and obese insulin-resistant volunteers, of whom 32 completed the three-month supplementation with daily oral pasteurized or live A. muciniphila versus placebo. The pasteurized strain was safe, well-tolerated, and improved insulin sensitivity (+28.6%, p=0.002), reduced insulinemia (−34.1%, p=0.006), and reduced plasma total cholesterol (−8.7%, p=0.02) compared with placebo. Pasteurized, human-tested, dead bacteria, real benefit.

Note on terminology: autoclaved (Lin) and pasteurized (Plovier, Depommier) are different protocols. Autoclaving uses high-pressure steam at 121 °C for 15 minutes and is more aggressive. Pasteurization uses moderate heat. Both render bacteria non-viable; the cell-wall chemistry that survives is somewhat different. Both have shown therapeutic activity in their own right.

People love to focus on engraftment. Will the bugs stick. Will they take up residence. Will the new ecosystem hold. But the metabolites are what is interfacing with you. Bacteria do not have to colonize permanently to leave a chemical signature. And in some of these studies, the signature is the entire therapeutic effect.

Does FMT Work for Depression in Humans?

It is fair to ask, at this point, whether any of the rodent and cell-line work translates to actual humans with actual depression. Two papers, both recent, anchor the human side of the picture. Read together, they do not say that microbiome interventions replace conventional psychiatric care. They say something more useful, and more interesting.

Wang et al., 2026, Scientific Reports

A randomized controlled trial led by Wang and colleagues, published in Scientific Reports in 2026, enrolled 46 adults aged 18–65 with depression (diagnosed per ICD-10) who were already taking a single antidepressant medication, escitalopram. The trial randomized them into two groups of 23: half continued escitalopram alone, the other half received escitalopram plus a fecal microbiota transplant delivered as oral capsules. Five participants in the FMT group and three in the control group dropped out during the intervention period, leaving 14 evaluable in the FMT group and 20 in the control group at the two-week endpoint. The primary measure was treatment efficacy rate at two weeks, defined as a 50 percent or greater reduction on the 24-item Hamilton Depression Scale (HAMD-24).

At two weeks, the response rate in the escitalopram-plus-FMT group was 71.4 percent (10 of 14 evaluable). In the escitalopram-alone group, 35.0 percent (7 of 20 evaluable). The difference reached statistical significance at p=0.037. Adverse-event incidence did not differ significantly between groups (28.6% vs 30.0%, p=0.928), with reported events limited to mild and self-resolving nausea, vomiting, and nasopharyngeal discomfort.

Two weeks is fast. It is too fast for stable bacterial engraftment, by the way. Most colonization studies show that even successful FMT engraftment unfolds over weeks to months as the donor community establishes residence. A two-week separation between groups is much more consistent with a metabolite-mediated mechanism: the donor community delivered a bolus of finished bacterial chemistry, and the host responded to that chemistry. Whether the bacteria themselves stayed or washed out, the chemistry was already at work.

Zhang et al., 2025, Frontiers in Psychiatry

A 2025 meta-analysis of randomized trials by Zhang and colleagues pooled 12 randomized controlled trials covering 681 patients across multiple FMT-for-depression studies. The pooled standardized mean difference for FMT versus control on depressive symptoms was −1.21 (95% CI −1.87 to −0.55; p=0.0003), a substantial effect size by meta-analytic standards.

The framing matters. The escitalopram still mattered in Wang's trial. The patients in Zhang's pooled trials were not abandoning conventional care. The intervention was layered on top, and it worked roughly twice as well as the conventional arm alone in Wang's data. This is the adjunct frame, and it is the right frame. We are not in a moment where the microbiome has replaced psychiatry. We are in a moment where we are discovering that the medicine cabinet your bacteria are running has been quietly shaping every other intervention all along.

Adjunct, not alternative. That is the line.

What This Means: Three Threads, One Pattern

Pull the four papers together and a coherent picture emerges. Three threads.

Thread one: timing is a variable, not a footnote. Cao 2026 shows the same microbial intervention delivered at different points in the host's circadian cycle produces dramatically different outcomes. The early-window, fasted, rest-phase delivery worked. The active-feeding, late-window delivery did not. Any future microbiome therapy that ignores timing is leaving most of the effect on the table.

Thread two: the metabolites are the messengers. Powell 2026 isolates a single bacterially-made molecule, lithocholic acid, that directly modulates a core clock gene's period in human colonic cells via a known kinase mechanism at physiological concentrations. The bacterium is not what reaches into the host cell. The molecule is. And the molecule keeps working whether the bacterium that made it is alive or dead, as Lin 2024 demonstrated for autoclaved P. goldsteinii and Plovier 2017 and Depommier 2019 demonstrated for pasteurized Akkermansia.

Thread three: the human data is converging. Wang 2026 shows an FMT-plus-SSRI combination roughly doubles the two-week treatment-response rate compared with SSRI alone in adults with depression on stable escitalopram. The two-week timescale is itself a mechanistic clue: too fast for engraftment-driven effects, which strengthens the metabolite-as-messenger interpretation. Zhang 2025 pools 12 trials and finds a meaningful effect size for FMT on depressive symptoms across the field. Mouse work, molecular work, and human work are pointing the same direction.

If the metabolites are doing the work, you do not need the live bacteria. You need the chemistry of a healthy community, delivered when the body is set up to receive it.

The Postbiotic Argument, Carefully Stated

This is where the science meets Thaena's working hypothesis, and it is worth being precise about the claim and what supports it.

Chen and colleagues, in the 2026 review that names the circadian-microbiome-liver axis, write the postbiotic frame plainly. From the paper, verbatim: "Postbiotics, such as SCFAs, bacterially derived peptides, or membrane vesicles, bypass the need for bacterial colonization and may be used to directly influence metabolic or immune pathways." Chen lists postbiotic metabolites alongside time-restricted nutrition, chronobiotic probiotics, and circadian-timed pharmacotherapy as the four therapeutic strategies that target the axis. This is a peer-reviewed framing of postbiotic intervention, in a review co-edited under journal standards, written by authors with no commercial relationship to Thaena.

Thaena's specific bet is narrower than that. The bet is that a sterilized human-sourced postbiotic, derived from a healthy donor microbiome, captures something a single-strain probiotic cannot: the emergent chemistry of a functional community, already produced, no longer dependent on whether the bacteria can survive a hostile gut, a transit time, or a stomach acid bath. The metabolites are finished. They are stable. They do not need to engraft.

We believe this is a plausible mechanism. We are watching, in real time, the literature converge on the underlying logic. We are also clear about what we have not done. We have not run our own randomized, placebo-controlled trial on mood. We have not published a Thaena-branded human study in a peer-reviewed journal. The trial work needed to validate the specific clinical claims is part of what Thaena is working toward, not something we are claiming today. This is a thesis, not a proven clinical outcome.

A practical landing

If you read Cao 2026, Powell 2026, and Lin 2024 together, the operational implication is specific. The rest-phase, fasted, pre-sleep window is when the host gut is most receptive to a microbial input. The active ingredient is the chemistry, not the live cell. And the chemistry should be delivered when the gut is about to enter its longest fasted clearance window of the day.

For Andrea, that has translated into a personal experiment. ThaenaBiotic at night, on an empty stomach, before bed. Sleep mask on. Fasting through to morning. Three behaviors aligned to what the data is suggesting about timing. This is a usage suggestion, not a treatment claim. It is the kind of behavior change that a single person can run, observe, and refine. Thaena's product is not approved to treat, cure, prevent, or diagnose any disease. The night-time, fasted, pre-sleep usage pattern is a way to align the timing of a postbiotic with the host clock state the Cao paper points to as most receptive.

We have been hearing, unprompted, from people taking ThaenaBiotic that something feels different after a few weeks of consistent use. Sleep is mentioned. Reactivity is mentioned. Mood is mentioned. We are listening, gathering, building the evidence base. The mechanism by which it could be working is plausible and consistent with the literature laid out above. The clinical proof is not yet ours to claim.

The Honest Limitations

Every paper in this episode comes with caveats that any clinician reading the comments will catch, and they are worth naming clearly.

⚠️ What this body of work cannot yet say
  • Cao 2026 is a mouse model. The CUMS protocol produces depression-like behavior, not human major depressive disorder. Mouse circadian biology is genuinely different from human circadian biology. Mice are nocturnal; humans are diurnal. The translation of ZT4 to "rest-phase, fasted, pre-sleep" is a defensible mapping but not a one-to-one equivalence.
  • Cao reports that serum tryptophan and 5-HT levels were not changed by FMT, even though tissue-level tryptophan and serotonin in the prefrontal cortex and hippocampus were restored. The authors attribute this to tight host regulation of circulating amino acid pools. It means a clinician looking at routine bloodwork would not see a serotonin shift even if the brain effect is real. Worth knowing.
  • Wang 2026 is small. 46 enrolled, 34 evaluable at two weeks (14 vs 20 due to uneven dropout), two-week endpoint. The 71.4 percent versus 35.0 percent treatment-response split is real and statistically significant, but a trial of this size is the kind that needs replication before the field treats it as settled.
  • Zhang 2025 has heterogeneity. A contemporaneous meta-analysis (Fu et al., 2026, Frontiers in Psychiatry) included seven studies in 235 subjects and did not find a significant overall treatment effect of FMT on depression, though their subgroup analyses suggested better antidepressant effect in populations with confirmed depression. The meta-analytic field is not yet in consensus on FMT for depression. Zhang's effect size is meaningful but not the only word.
  • Powell 2026 is in cell culture. Human colonic cells in a dish are not a human gut. The dose-response curve falls in the physiological range, which is reassuring, but the leap from a luciferase reporter to an in vivo clock-gene effect in living tissue requires further work.
  • Lin 2024 used colon epithelial cells, not the full immune environment of a living gut. The TLR2 activation and NF-κB suppression are clear, but the systemic anti-inflammatory translation requires animal and eventually human work.

None of these limitations invalidate the through-line. They constrain the language. Mouse models earn the strongest claims about mechanism, modest claims about cause-and-effect, and weaker claims about clinical translation. The convergence across the four papers is what makes the picture worth taking seriously, not any one of them in isolation.

Frequently Asked Questions

Should I take my probiotic or postbiotic at night?

The strongest current mechanistic evidence comes from a 2026 mouse study (Cao et al., Brain, Behavior, and Immunity) showing that microbial input during the host's resting, fasted phase produced dramatically better outcomes than during the active feeding phase. Translated to humans, this lines up with evening or pre-sleep delivery on a relatively empty stomach. This is a usage suggestion based on emerging mechanistic evidence, not a clinical guideline. Most product labels still recommend a morning or with-food schedule based on older assumptions about gastric pH and bacterial survival; that framing predates the timing-of-receptivity research.

Does the time of day affect how my gut bacteria work?

Yes. The trillions of bacteria in your gut oscillate on a 24-hour rhythm. Different species peak at different times of day. They migrate along the gut wall in waves and produce metabolites — bile acids, short-chain fatty acids, tryptophan derivatives — in time-dependent patterns. These metabolites in turn signal your host cells about what time it is. The system is bidirectional: feeding rhythms set bacterial behavior, and bacterial behavior reinforces host clocks. Disrupting the timing (shift work, jet lag, late-night eating, light at night) breaks the synchronization.

What is the difference between probiotics, postbiotics, and FMT?

Probiotics are live bacterial strains intended to colonize the gut. Postbiotics are the bioactive compounds bacteria produce, or the inactivated bacteria themselves and their cell-wall components, delivered without the live cells. Fecal microbiota transplantation (FMT) is the transfer of a complete microbial community from a healthy donor to a recipient. The emerging evidence base (Lin 2024, Plovier 2017, Depommier 2019) suggests that the metabolites — not the live cells — are often the active ingredient, which is the case for postbiotic-based interventions.

Can gut bacteria affect mood?

A growing body of human and animal research suggests yes. Gut bacteria produce neurotransmitter precursors (tryptophan for serotonin, glutamate for GABA), modulate the inflammation that affects brain function, and signal the brain through the vagus nerve and bloodstream. A 2025 meta-analysis of twelve randomized trials (Zhang et al., Frontiers in Psychiatry, n=681) found that fecal microbiota transplantation produced a statistically significant reduction in depressive symptoms across diverse trial populations. The mechanism appears to run through bacterial metabolites that influence host clock genes, neurotransmitter availability, and neuroinflammation.

What is the circadian-microbiome axis?

The circadian-microbiome axis is the bidirectional system in which your gut bacteria and your body's cellular clocks influence each other on a 24-hour cycle. Your peripheral clocks (in liver, gut, immune cells) are entrained primarily by feeding times rather than light. Bacterial metabolites act as zeitgebers, time cues, that synchronize host cellular clocks. Disruption of the system (shift work, jet lag, late eating, light at night) breaks this synchronization and contributes to metabolic, immune, and mood dysregulation. The framework is reviewed in Chen et al., 2026 (Frontiers in Medicine) under the related term "circadian-microbiome-liver axis."

The Bottom Line

You are not one clock. You are an orchestra. The conductor in your brain reads the sun. The musicians in your gut, liver, and immune cells read your meals, your sleep, and the chemistry your bacteria are producing. When the conductor and the musicians disagree, the music falls apart.

Three things set this orchestra. Light. Food. Sleep. If those three are broken, no microbiome intervention is going to land cleanly, because the host's clock will be telling its cells the wrong story.

The bacteria themselves are timekeepers. Their metabolites are how they tell your cells what time it is. Bile acids, short-chain fatty acids, tryptophan derivatives, signaling molecules that pulse on schedule. Cao 2026 shows those signals only land when the host is set up to receive them. Powell 2026 shows the molecular machinery by which a single bile acid sets the dial. Lin 2024, Plovier 2017, and Depommier 2019 show the molecule keeps working without the live bacterium. Wang 2026 and Zhang 2025 show the human data is converging in the same direction.

The postbiotic frame is the practical one. You do not need new bacteria to live in you forever. You need the chemistry of a healthy community, delivered when your body is ready to receive it.

Summary: What This Means
  • You have two clocks at the system level: a master clock in the brain that reads light, and trillions of peripheral clocks in your gut, liver, and immune system that read food and microbial chemistry.
  • The circadian-microbiome-liver axis (Chen 2026) and chronobiome medicine (Bautista 2025) are peer-reviewed framings of how the field now understands this coupled system.
  • In Cao 2026, the same FMT dose given 12 hours apart produced dramatically different outcomes in depressed mice. The early-window, fasted-leaning, rest-phase delivery restored mood-related behaviors, BDNF, clock genes, and tryptophan/serotonin. The late-window did not.
  • The mechanism traces to specific bacterially-produced metabolites. Powell 2026 isolates lithocholic acid, a secondary bile acid, that lengthens the circadian period of human colonic cells by inhibiting CK1δ/ε and stabilizing CRY2.
  • Lin 2024 (autoclaved P. goldsteinii), Plovier 2017, and Depommier 2019 (pasteurized A. muciniphila) all show that dead bacteria retain therapeutic activity. The signal is the chemistry, not the live cell.
  • Human RCT data is converging. Wang 2026 shows 71.4 percent versus 35.0 percent treatment-response rate at two weeks for FMT-plus-escitalopram versus escitalopram alone (HAMD-24 reduction ≥50%; n=14 vs 20 evaluable). Zhang 2025 meta-analyzes 12 RCTs (n=681) with a pooled effect size of SMD = −1.21 (p=0.0003).
  • Adjunct, not alternative. The microbiome work is making conventional psychiatric care work better, not replacing it.
  • If the metabolites are the active ingredient, the postbiotic frame becomes practical: deliver the finished chemistry of a healthy community when the gut is set up to receive it. Thaena's bet is that a sterilized human-sourced postbiotic captures the emergent chemistry no single strain can. We believe this is a plausible mechanism. It is a thesis, not a proven clinical outcome.

Light, food, sleep, and chemistry. The orchestra. Tune it.

Stay curious. Take care of your microbes. The orchestra inside you is keeping time with you whether you know it or not.

References

  1. Cao P, Li Y, Zhang S, Li C, Sun Y, An B. Study on the efficacy and mechanism of fecal microbiota transplantation for depression based on circadian rhythm. Brain Behav Immun. 2026;131:106186. https://doi.org/10.1016/j.bbi.2025.106186
  2. Powell CE, McSween AM, Dohnalová L, Kim CH, Eisert RJ, Sun Z-YJ, Seo H-S, Marquardt V, Dhe-Paganon S, Thaiss CA, Devlin AS. Gut microbiome-produced bile acid metabolite lengthens circadian period in host intestinal cells. Proc Natl Acad Sci USA. 2026;123(11):e2506313123. https://doi.org/10.1073/pnas.2506313123
  3. Wang L, Zhang S, Liu Y, Li D, Tian G, Li X, Li Y. A study on the efficacy and safety of fecal microbiota transplantation as an adjunctive therapy for treating depressive episodes. Sci Rep. 2026;16:13417. https://doi.org/10.1038/s41598-026-41801-y
  4. Zhang X, Li Y, Guo Y, Sun J, Yang Y. Clinical efficacy of fecal microbiota transplantation in alleviating depressive symptoms: a meta-analysis of randomized trials. Front Psychiatry. 2025;16:1656969. https://doi.org/10.3389/fpsyt.2025.1656969
  5. Fu J, Zhang Y, Gao J, Lan M, Zhang Z, Liang R, Zhou H, Liu S, Zhou Z, Zhao J, Yu X, Liu Y, Han P, Chen X, Lin C, Guo Q. Efficacy effects of fecal microbiota transplantation on depressive symptoms: a meta-analysis based on randomized controlled trials. Front Psychiatry. 2026;17. https://doi.org/10.3389/fpsyt.2025.1629290
  6. Lin T-L, Chen W-J, Hung C-M, Wong Y-L, Lu C-C, Lai H-C. Characterization and safety evaluation of autoclaved gut commensal Parabacteroides goldsteinii RV-01. Int J Mol Sci. 2024;25(23):12660. https://doi.org/10.3390/ijms252312660
  7. Plovier H, Everard A, Druart C, et al. A purified membrane protein from Akkermansia muciniphila or the pasteurized bacterium improves metabolism in obese and diabetic mice. Nat Med. 2017;23(1):107–113. https://doi.org/10.1038/nm.4236
  8. Depommier C, Everard A, Druart C, et al. Supplementation with Akkermansia muciniphila in overweight and obese human volunteers: a proof-of-concept exploratory study. Nat Med. 2019;25(7):1096–1103. https://doi.org/10.1038/s41591-019-0495-2
  9. Chen Y-J, Yang B-W, Gu Z-C, Han J. Interplay between circadian rhythms, gut microbiota, and MASLD: from mechanistic foundations to therapeutic opportunities. Front Med. 2026;13:1767462. https://doi.org/10.3389/fmed.2026.1767462
  10. Bautista J, López-Cortés A. Chronobiome medicine: circadian regulation of host–microbiota crosstalk in systemic physiology. Front Endocrinol. 2025;16:1691172. https://doi.org/10.3389/fendo.2025.1691172
  11. Touitou Y, Perlemuter G, Touitou C. Shift work, gut dysbiosis, and circadian misalignment: the combined impact of nighttime light exposure, nutrients, and microbiota rhythmicity. Chronobiol Int. 2025;42(10):1275–1290. https://doi.org/10.1080/07420528.2025.2540039
  12. Liu L, Wang H, Chen X, Zhang Y, Zhang H, Xie P. Gut microbiota and its metabolites in depression: from pathogenesis to treatment. eBioMedicine. 2023;90:104527. https://doi.org/10.1016/j.ebiom.2023.104527

Additional background reading

The following papers informed the episode's NotebookLM-driven research process and are included for readers who want to follow the conversation further. They are not directly cited in the episode body.

  1. Meyyappan AC, et al. 2025.
  2. Aroca-Crevillén A, et al. 2026.
  3. Mahen E, et al. 2025.

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