Gut microbial metabolites are increasingly recognized as signaling molecules capable of influencing host physiology. New experimental findings from international research team led by Professor Marc-Emmanuel Dumas at Imperial College London & CNRS, together with Prof. Patrice Cani (Imperial & University of Louvain, UCLouvain), Dr. Dominique Gauguier (Imperial & INSERM, Paris) and Prof. Peter Liu (Ottawa Heart Centre), show that trimethylamine (TMA) directly inhibits the inflammatory regulator IRAK4, providing a mechanistic link between microbial metabolism, inflammation and glycemic control.1
Type 2 diabetes is a major global public health challenge. Beyond well-established factors such as diet, lifestyle and genetics, growing evidence suggests that the gut microbiota may contribute to metabolic health through the molecules it produces.
A recent study published in Nature Metabolism sheds light on one such interaction. The researchers found that trimethylamine (TMA), a metabolite produced by gut bacteria, improved glycemic control and reduced metabolic inflammation in experimental models by directly inhibiting interleukin-1 receptor-associated kinase 4 (IRAK4), a central regulator of innate immune signaling.¹
TMA should not simply be considered a precursor of TMAO
Gut bacteria produce TMA from dietary compounds including choline, L-carnitine and betaine. Once absorbed, TMA is converted in the liver into trimethylamine N-oxide (TMAO), which has received considerable attention because elevated circulating levels have been associated with cardiovascular risk.2
Yet TMA and TMAO are closely related molecules without necessarily having the same biological effects.
“This is actually one of the most interesting findings of the study,” explains Prof. Amandine Everard (senior research from the FNRS at UCLouvain), one of the study’s authors. The findings suggest that TMA has biological activities distinct from those of TMAO and should not simply be viewed as “TMAO before conversion”.
Previous research provides some clues to this apparent paradox. Dietary choline intake has been associated with improved insulin sensitivity in some studies, while experimental reduction of the liver enzyme responsible for converting TMA into TMAO improved insulin resistance and reduced atherosclerotic burden in diabetic mice.
The new study adds a potential mechanism to these observations: TMA directly inhibited IRAK4 and improved immune tone and glycemic control in mice with diet-induced obesity. According to Prof. Amandine Everard, TMA and TMAO may therefore have contrasting roles that are both context- and mechanism-dependent.
“A precursor and its downstream metabolite can have very different biological effects. The relationship between TMA, TMAO, the microbiome and human health is likely more nuanced than previously thought.”
Identifying the host target of a microbial metabolite takes the findings beyond association
While microbial metabolites have long been linked to host physiology, identifying the specific molecular targets through which they act remains a major challenge in microbiome research.
The researchers combined different experimental approaches, including primary human hepatocytes, human peripheral blood monocytic cells and mouse models of diet-induced obesity, to investigate how the choline–TMA pathway affects glucose metabolism.
Choline supplementation improved glucose tolerance, insulin sensitivity and hepatic insulin signaling. When microbial TMA production was blocked, however, these metabolic benefits disappeared, suggesting that conversion of choline into TMA was required for the observed effects. Chronic TMA administration subsequently reproduced some of these effects, improving glycemic homeostasis and reducing inflammatory markers.¹
The researchers then identified IRAK4 as a direct molecular target of TMA.
IRAK4 is a key component of signaling downstream of Toll-like and interleukin-1 receptors. These pathways are important for immune defense, but their persistent activation can promote pro-inflammatory signaling through pathways including NF-κB and MAPK and contribute to the chronic low-grade inflammation associated with insulin resistance.3
By inhibiting IRAK4, TMA reduced inflammation associated with a high-fat diet and preserved insulin signaling in the experimental models.¹
For Prof. Amandine Everard, moving from an association between a microbial metabolite and metabolic health to identifying both a causal effect and the specific host protein with which the metabolite interacts is a central contribution of the study.
“Molecules produced by the gut microbiota can act in a pharmacology-like manner, directly targeting host proteins that control inflammation and metabolism.”
The findings therefore provide an example of how microbiome research can move beyond describing changes in microbial composition towards identifying the molecular mechanisms through which microbial activity affects the host.
The findings are not yet a reason to increase dietary TMA precursors
The involvement of dietary compounds such as choline and L-carnitine in TMA production raises an important question for nutrition professionals: should these nutrients be increased to promote TMA production?
For now, the evidence does not support this approach.
Prof. Amandine Everard emphasizes that the study was designed to provide mechanistic evidence rather than establish dietary recommendations. Most of the mechanistic experiments were conducted in cellular systems and animal models, which do not reproduce the full complexity of human physiology.
“Nutrition professionals should avoid concluding that people should deliberately increase their intake of specific nutrients, such as choline or carnitine, in order to boost TMA production.”
The metabolic consequences of these nutrients are also unlikely to depend solely on how much is consumed. Gut microbiota composition, host metabolism and the overall dietary pattern may all influence how dietary substrates are metabolized and their subsequent effects on health.4
Interindividual variation in TMA metabolism is another important consideration. In addition, the study did not assess long-term clinical outcomes such as type 2 diabetes incidence, cardiovascular events or mortality.
Large longitudinal human studies will therefore be needed to determine whether the TMA–IRAK4 pathway identified experimentally is relevant to human metabolic health and whether it could eventually be targeted safely through nutritional or pharmacological strategies.
For healthcare and nutrition professionals, the immediate relevance of the findings is thus not a new dietary recommendation, but a broader shift in how interactions between diet, gut microbes and human physiology are understood. Characterizing the gut microbiome may require looking beyond which microorganisms are present to consider the metabolites they produce, the host proteins those metabolites interact with, and the biological responses that follow.
The discovery that TMA can directly target IRAK4 illustrates how microbial metabolites may function as molecular mediators between the gut microbiota and host immune and metabolic pathways—and why understanding these mechanisms may be essential before microbiome research can be translated into clinical practice.
References
- Chilloux J, Brial F, Everard A, et al. Inhibition of IRAK4 by microbial trimethylamine blunts metabolic inflammation and ameliorates glycemic control. Nature Metabolism. 2025;7:2531–2547.
- Mirji G, Bhat SA, Shinde RS. Trimethylamine-N-oxide: the microbial cue in immune-mediated disorders. Trends Endocrinol Metab. 2026 Mar 11:S1043-2760(26)00043-3. doi: 10.1016/j.tem.2026.02.008.
- Lin Y, Zheng L, Xu Y, Wang X, Li J, Zheng L, Liang G, Chen L. Interleukin-1 Receptor-Associated Kinase 4 (IRAK4) Degraders for Treating Inflammatory Diseases: Advances and Prospects. J Med Chem. 2025;68(2):902–914. doi: 10.1021/acs.jmedchem.4c01322.
- Lin H, Zhong Z, Zhang C, Jin X, Qi X, Lian J. An inverse association of dietary choline with atherosclerotic cardiovascular disease among US adults: a cross-sectional NHANES analysis. BMC Public Health. 2024 May 31;24(1):1460. doi: 10.1186/s12889-024-18837-8.
