Lab Study Finds Dietary Polyphenols Block Bacterial Communication System In Salmonella
A study published October 3, 2026, in Gut Microbes used machine learning to screen over 7000 plant compounds and found dietary polyphenols can block the chemical signaling system Salmonella bacteria use to coordinate infection, cutting bacterial invasion of human gut cells by up to 50% in lab dishes.
Researchers built a graph neural network classifier, AI2IMpred, trained on 215 known quorum-sensing interference molecules and 469 inactive compounds, then screened plant compounds. Quorum sensing is the signaling system bacteria use to coordinate biofilms and attacks on host cells. The model scored over 7000 phytochemicals; polyphenols made up 53% of the top 100 hits. Testing moved to Salmonella Typhimurium LT2 in culture. Of 263 candidate polyphenols, 100 were tested directly, and 38 showed antibacterial activity below 10 millimolar. Four compounds—pterostilbene, 4-methylcatechol, pyrogallol and 3-methylcatechol—were confirmed by binding assays to attach directly to bacterial proteins LuxS and LsrB, used to make and detect the signaling molecule. In human colon cell cultures, these polyphenols cut Salmonella's adhesion by up to 33% and invasion by up to 50%, without killing cells. Paired with nitrofurantoin, they cut the drug's effective inhibitory concentration twofold; no boost appeared with sulfonamides. Similar effects occurred in E. coli and Staphylococcus aureus, while Pseudomonas aeruginosa, lacking matching receptors, showed little response.
The authors describe polyphenols physically occupying binding pockets on LuxS and LsrB that bacteria use to produce and sense AI-2, disrupting genes for biofilm formation, movement and tissue invasion. Separately, known chemistry holds that hydroxyl- and oxygen-containing groups, common in polyphenols, readily form hydrogen bonds, helping explain why this compound class dominated the screen.
Salmonella's attack on gut cells, with polyphenols vs without
Data Panel
- Who
- Salmonella Typhimurium LT2, with cross-testing in E. coli MG1655, Staphylococcus aureus and Pseudomonas aeruginosa; also human colon epithelial cells (NCM460) in culture; no animal or human subjects
- Design
- Laboratory study: machine-learning virtual screen of 7000+ phytochemicals, followed by in vitro antibacterial assays, binding assays (surface plasmon resonance and molecular docking), gene expression analysis and human cell culture adhesion/invasion testing
- Dose
- Polyphenols tested at sub-inhibitory concentrations determined by growth curve assays for each compound and strain; binding assays used concentrations from 7.8 to 1000 micromolar
- Primary result
- 38 of 100 tested polyphenols showed antibacterial activity below 10 millimolar; 4 compounds (pterostilbene, 4-methylcatechol, pyrogallol, 3-methylcatechol) confirmed by binding assay to target LuxS and LsrB proteins
- Secondary
- Polyphenol treatment reduced Salmonella adhesion to gut cells by up to 33% and invasion by up to 50%; combining polyphenols with nitrofurantoin produced a twofold reduction in the antibiotic's effective inhibitory concentration; no synergy seen with sulfonamide antibiotics
- Funding / conflicts
- Funding and conflicts not reported in the abstract.
All results come from bacteria and human colon cells grown in dishes, not animals or people eating these compounds. The study doesn't report how much polyphenol intake would reach the concentrations used experimentally, so dietary benefit remains unproven. Funding and conflicts not reported in the abstract.
Dr. Axe's Take
This is bacterial chemistry in a dish, not a human nutrition trial. Polyphenols, the same plant compounds in berries, green tea, dark chocolate and grapes, physically jammed the signaling receptors Salmonella uses to organize an attack on gut lining here. Pterostilbene, found in blueberries, and pyrogallol and catechols found in tea and many fruits, stood out in these bacterial cultures. I would not treat this as proof that eating more polyphenols prevents foodborne illness, since nobody tested people eating food. I'd still build polyphenol-rich foods into meals as broad support for gut health: berries at breakfast, green tea in the afternoon, colorful vegetables at dinner.
— Dr. Axe


