364 Metabolites Mapped for Their Power to Reshape T Cell Immune States
A high-throughput atlas reveals how gut and endogenous metabolites reprogram T cell transcription, pointing to new immunomodulatory targets for inflammatory disease.
Summary
Researchers systematically exposed human T cells to 364 metabolites — including gut microbiome-derived compounds — and used advanced RNA sequencing to map how each one shifts T cell gene expression. They identified distinct T cell states ranging from baseline to highly activated, and developed an AI framework called DRUG-seq-PerturbFormer to predict which metabolites drive these transitions. A subset of metabolites was found to robustly reprogram immune programs. In a mouse model of colitis, top candidate metabolites reduced disease severity, suppressed inflammatory gene programs, and partially restored immune balance. This work establishes a comprehensive roadmap of metabolite-immune crosstalk, revealing that the metabolic environment — shaped by diet, gut bacteria, and host physiology — is a powerful regulator of T cell behavior with direct implications for treating inflammatory and autoimmune diseases.
Detailed Summary
The immune system does not operate in isolation — it is continuously tuned by the metabolic environment surrounding it. Gut microbiome-derived and endogenous metabolites are known to influence immunity, but a systematic, genome-wide picture of how they shape human T cell states has been missing. This study fills that gap with unusual breadth and precision.
Researchers at Shanghai's Ruijin Hospital and affiliated institutions used high-throughput DRUG-seq — digital RNA sequencing with perturbation profiling — to expose primary human T cells to 364 distinct metabolites and capture genome-wide transcriptional responses. The resulting atlas mapped T cell states across a spectrum from baseline through metabolically primed to highly activated, revealing structured trajectories rather than random variation.
To extract predictive signal from this large dataset, the team developed DRUG-seq-PerturbFormer, a multi-task deep learning model capable of quantifying the magnitude and directionality of each metabolite's effect on gene expression. This framework identified a subset of metabolites with the strongest reprogramming capacity — compounds that reliably shift T cells toward or away from inflammatory activation.
In a dextran sulfate sodium-induced colitis mouse model, representative metabolites from this shortlist reduced disease severity, suppressed pro-inflammatory transcriptional programs, and partially restored immune homeostasis. These in vivo results validate the transcriptomic predictions and demonstrate translational potential.
For longevity-focused readers, the findings matter because chronic low-grade inflammation — driven partly by shifts in gut microbiome composition and metabolite production that accompany aging — is a central driver of aging-related disease. Identifying specific metabolites that can tune T cell activity offers a route toward dietary or microbiome-based interventions to dampen inflammaging. Limitations include reliance on the abstract alone, mouse model validation, and the need for human clinical trials to confirm therapeutic relevance.
Key Findings
- 364 gut and endogenous metabolites were screened against human T cells, revealing structured transcriptional state transitions.
- A deep learning model (DRUG-seq-PerturbFormer) accurately predicted which metabolites drive T cell activation or suppression.
- A defined subset of metabolites robustly reprogrammed T cell inflammatory gene programs.
- In a colitis mouse model, top candidate metabolites reduced disease severity and partially restored immune homeostasis.
- Findings nominate specific metabolites as immunomodulatory targets relevant to inflammatory and aging-related diseases.
Methodology
The study used DRUG-seq (high-throughput digital RNA sequencing) to profile primary human T cell transcriptional responses to 364 endogenous and microbiota-derived metabolites. A multi-task deep learning framework (DRUG-seq-PerturbFormer) was developed to quantify perturbation magnitude and directionality. In vivo validation used a dextran sulfate sodium-induced mouse colitis model.
Study Limitations
This summary is based on the abstract only, as the full paper was not accessible. The in vivo validation used a mouse colitis model, which may not fully translate to human inflammatory disease. The therapeutic impact of individual candidate metabolites in human clinical trials remains to be established.
Enjoyed this summary?
Get the latest longevity research delivered to your inbox every week.
Enter your email to subscribe: