Reading the L-Theanine Evidence: Part Four: Beyond the Brain

Abstract

The first three parts of this series covered what L-theanine has been tested for in healthy adults: attention, sleep, stress. This part covers what it has been shown to do everywhere else — in immunity, in the gut, and in developing neurons.

The mechanistic work is genuinely elegant, considerably more so than the human trials. It is also where the distance between a finding and a claim is largest, and where that distance is most often collapsed in marketing copy.

Three habits of reading do most of the work here: check what else was in the capsule, convert the animal dose, and read the disclosure statement against the affiliations printed above it.

1.Immunity: The Strongest Human Data in the Wrong Place

Shuna Chen, Bo Li and colleagues in the Department of Tea Science at Zhejiang University published a review of L-theanine and immunity in Molecules in 2023, and it opens by noting that the immunomodulatory effects have received less attention than the sedative and calming ones.

The clinical evidence they assemble is, in some respects, better than what exists for cognition. Randomised, double-blind, placebo-controlled trials found a significantly lower incidence of colds in supplemented participants than in placebo. Supplementation in healthcare workers over five months protected against influenza and was well tolerated. In older adults, supplementation before vaccination enhanced the immune response to the influenza vaccine. In gastric cancer patients undergoing distal gastrectomy, perioperative supplementation raised glutathione levels and improved several inflammatory markers.

These are real clinical endpoints — infections, vaccine responses, postoperative recovery — of a kind the cognition literature does not possess.

Now read the intervention. In the cold trials it was cystine and L-theanine. In the exercise studies, L-theanine and cystine. Before vaccination, L-theanine and L-cystine. In the surgical patients, L-theanine and cystine. In the chemotherapy work, L-theanine and cysteine.

Every one of the human immune findings comes from a combination, and the review's own mechanistic account explains why: the proposed pathway runs through glutathione synthesis, and glutathione is built from cysteine, glutamate and glycine. Cystine is the substrate. Theanine's contribution is understood as supporting that synthesis rather than performing it.

So the domain in which L-theanine has its most concrete human outcomes is also the domain in which it has never been tested alone. That is the same problem we met in Part Three with the magnesium-and-rhodiola trial, arriving from a different direction, and it is worth recognising as a general feature rather than a coincidence.

2.Depression and the Gut: An Elegant Mechanism at an Impossible Dose

Ying Peng, Daxiang Li, Shanshan Hu and colleagues at Anhui Agricultural University published the most mechanistically complete paper in this set in npj Science of Food, tracing how L-theanine acts on depressive-like behaviour in mice through the gut.

The model is chronic unpredictable mild stress, a standard rodent paradigm. The finding is a chain rather than a correlation: L-theanine reshaped the gut microbiota, increasing Lactobacillus and Roseburia; that restored production of short-chain fatty acids, particularly acetic, butyric and propionic acid; those in turn restored receptor function and barrier integrity in both the colon and the prefrontal cortex, with the tight-junction proteins ZO-1 and occludin recovering and the TLR9/NLRP3/caspase-1 inflammatory pathway downregulated. The depressive-like behaviour reversed.

This is careful work with a plausible causal structure, and the microbiome sequencing data are deposited publicly.

The dose is where it stops being about supplements. Three arms were tested — 100, 400 and 800 mg per kilogram — and the effects appeared at 400 and were most pronounced at 800.

The authors do the interspecies conversion themselves rather than leaving it to the reader, which is to their credit: 800 mg per kilogram in mice corresponds to a daily supplementation of 87.9 mg per kilogram for an adult human. For a 70-kilogram person that is roughly 6.2 grams a day.

The trials in Parts One to Three used 50 to 450 mg a day. The mechanism described here operates, as demonstrated, at something between fifteen and a hundred times a supplement dose — and the low arm, 100 mg per kilogram, corresponding to something closer to a large human dose, did not produce the effect.

Nothing about that invalidates the biology. It does mean that a sentence like theanine supports the gut–brain axis, if it is meant to describe this paper, is describing something that happens at six grams a day in mice under chronic stress.

3.Developing Neurons, and Eighteen Mice

The third study, by Mai Ibrahim, Masahiro Nishio and colleagues at Mie University, asks whether L-theanine affects the migration of cerebellar granule cells — the neurons that make up more than half the brain's total and whose migration during development, if disrupted, is associated with later neurological disorder.

The in vitro half is the interesting half. Granule cell microexplants from newborn mice were exposed to a range of concentrations, and migration distance was measured at 24, 48 and 72 hours. One micromolar produced the longest migration; higher and lower concentrations did less. The authors note that they extended the tested range into the nanomolar band because previous work had only gone up to 10 micromolar.

A concentration optimum with reduced effect on either side is a meaningful observation about how a molecule behaves. It is also, precisely, an observation about cells in a dish.

The behavioural half is much thinner: 18 male mice divided into four groups — a control and three doses — meaning four or five animals per arm, assessed on a maze and a Morris water maze. The abstract's own dose reporting is internally inconsistent, giving a single oral dose of 10 mg/kg in the methods and then reporting effects at 5 mg/ml and 0.05 mg/ml, which are concentrations rather than doses.

The paper's conclusion — that L-theanine may serve as a potential therapeutic agent in supporting cerebellar development — is hedged appropriately in the original. It travels badly. Cerebellar development happens before and shortly after birth; nothing in this work concerns an adult brain, and nothing in it involves a human.

4.Two Disclosure Statements Worth Reading Twice

Both of the primary studies in this part declare no competing interests. Both print information above that declaration which a reader might weigh differently.

In the gut–brain paper, one co-author's affiliation is Fujian Chunlun Group Co., Ltd. In the cerebellar paper, two co-authors are affiliated to Taiyo Kagaku Co., Ltd., and the acknowledgments thank Taiyo Kagaku for experimental support with L-theanine powder and maze apparatus — the compound under test and the apparatus used to test it. The custom maze is described in the methods, and pictured in Figure 1, as built by the same company.

We are not alleging misconduct and there is no indication of any. Journals differ in what they require declared, and supplying material to an academic laboratory is normal practice. But the reader who wants to weigh a finding needs the relationship, and in both papers it appears in the affiliations rather than in the field designed to carry it.

There is a related pattern across this whole series that becomes visible only when the papers are read together. One industry scientist, Makoto Ozeki, appears as a co-author both on the cerebellar migration study here and on the 200 mg four-week human trial that anchors the sleep evidence in Part Two. A small literature with recurring participants is not a corrupt one. It is a literature in which apparent independent replication needs checking before it is counted.

5.Three Rules This Part Suggests

The first: check what else was in it. If a positive human result comes from a capsule containing cystine, magnesium, rhodiola or B vitamins, the trial tested the capsule. This has now appeared in three of the four domains in this series.

The second: convert the dose before believing the mechanism applies to you. Milligrams per kilogram in a mouse is not milligrams per kilogram in a person, and where authors do the conversion themselves the answer is often startling — here, six grams a day.

The third: read the affiliations, not only the disclosure line. The disclosure is where a conflict is supposed to be declared; the affiliation is where it is always visible.

None of these rules requires expertise in immunology or in cerebellar development. They require reading the parts of a paper that summaries omit, which is most of what separates a claim from the evidence behind it.

Editorial Comment

MindHeaven® makes no claim that L-theanine supports immunity, influences the gut microbiome, protects neurons, or affects brain development. No such claim is authorised in the European Union, and nothing in the literature above would support one if it were.

We are conscious that this part is where a company in our position would ordinarily reach. Mechanistic findings are the most quotable material in any supplement literature precisely because they sound like effects while remaining technically true — a tight junction restored, a microbial genus increased, a neuron migrating further. Written carefully enough, a reader draws the conclusion the writer never stated.

We use L-theanine on the attention evidence in Part One, at 250 mg with caffeine, and that is the whole of our case for it. The immune findings belong to a compound plus cystine, the gut findings to a dose we do not supply and would not, and the developmental findings to cells in culture and to four mice per group.

Part Five closes the series with the paper that asks the question directly — whether the science on this compound matches the way it is sold — and with our own answer about the two products of ours that contain it.