New laboratory experiments show that a bacterium that lives on grape plants can break down guaiacol—an unpleasant-tasting substance which ruins wines made from grapes exposed to wildfire smoke. Claudia Castro of the U.S. Department of Agriculture's Agricultural Research Service and colleagues present these findings in the open-access journal PLOS One on October 1, 2025.
Wildfires are occurring increasingly often in wine-producing regions of the U.S. West Coast. Wine grapes can absorb substances from wildfire smoke that ultimately make their way into the bottle, resulting in a smoky, ashy-tasting pour. Prior research has shown that one such substance, guaiacol, can be broken down by certain species of bacteria living in soil, raising the possibility that bacterial metabolic processes could be harnessed to help fix the taste of smoke-tainted wines.
However, few studies have explored whether the ability to degrade guaiacol is also possessed by bacteria found naturally on grapevines as part of the plants' microbiome. To explore that possibility, Castro and colleagues collected leaves from two varieties of grape plants, chardonnay and cabernet sauvignon, and tested them in a lab for the presence of guaiacol-eating bacteria.
They found that the leaves harbored two strains of the same bacterial species, Gordonia alkanivorans, that can break down guaiacol in the lab. Analysis of their genomes revealed specific genes involved in this metabolic process, including a gene called guaA. When this gene was experimentally deleted in G. alkanivorans cells, the bacterium was unable to break down guaiacol, confirming that guaA is required for guaiacol degradation.
The researchers also exposed living merlot plants to smoke produced by a culinary smoker—simulating exposure to wildfire smoke—and tested which bacterial species were present on the plants before and after. They found that the microbiome on both leaves and berries (grapes) changed significantly in the days following smoke exposure, with a notable increase of bacteria in the Bacilli class, which are known to survive in extreme environments.
In recent years, smoke-tainted wine has cost industry stakeholders millions of dollars. These new findings could help pave the way for the development of biotechnological strategies that translate natural bacterial degradation of unpalatable substances into flavor-rescuing tools.
Claudia Castro adds: "In this study, we identified two members of the grape microbiome capable of degrading guaiacol, a volatile phenol that contributes to smoke taint of wine. We also identified changes in the grape microbiome caused by simulated wildfire smoke with the goal of providing genetic tools to help combat smoke taint."
"Working in the vineyard along with our collaborator Dr. Tom Collins and his team to set up the wildfire smoke simulation is one of my best memories of this study. Probably second best to finding a guaiacol-degrading microbe that lives on the surface of grapes!"
In your coverage, please use this URL to provide access to the freely available article in PLOS One: http://plos.io/4650m0C
Citation: Castro C, Badillo J, Tumen-Velasquez M, Guss AM, Collins TS, Harmon F, et al. (2025) Bacteria isolated from the grape phyllosphere capable of degrading guaiacol, a main volatile phenol associated with smoke taint in wine. PLoS One 20(10): e0331854. https://doi.org/10.1371/journal.pone.0331854
Author countries: U.S.
Funding: We thank the California Association of Winegrape Growers (CAWG) and the USDA-ARS for their financial support (U.S. Department of Agriculture 2030-21210-003-000D). We thank the Washington State Grape and Wine Research program and the USDA ARS for their financial support in the Collins lab (U.S. Department of Agriculture NACA agreement 58-2072-0-033). Additionally, this research was supported in part by the U.S. Department of Energy (DOE), Office of Biological and Environmental Research (BER), as part of Secure Biosystems Design Science Focus Area "Persistence Control of Engineered Functions in Complex Soil Microbiomes". The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
