Antimicrobial resistance is a growing global health concern, with soils serving as an important reservoir of antibiotic resistance genes (ARGs). However, how short-term extreme climatic events affect these genes remains poorly understood.

A research team led by Prof. YU Yong from the Northeast Institute of Geography and Agroecology of the Chinese Academy of Sciences has revealed how soil ARGs respond to different climate extremes.

The study, published recently in PNAS, was conducted in collaboration with Prof. Matthias C. Rillig from Freie Universität Berlin and Prof. Franciska T. de Vries from the University of Amsterdam.

To examine these responses, the researchers analyzed metagenomic data from a controlled experiment using soils collected from 30 grassland sites across 10 European countries. The experiment simulated four types of climate extremes — drought, flooding, freeze-thaw cycles, and heatwaves — and analyzed samples immediately after each disturbance and again after four weeks of recovery.

The results showed that geographic differences accounted for most of the variation in soil ARG composition, while climatic disturbances produced smaller changes that varied among event types. Among the four types of extremes, heatwaves produced the clearest decline in the relative abundance of ARGs across several major resistance classes, while drought, flooding, and freeze-thaw cycles generally had weaker effects.

Functional gene analysis provided clues to the heatwave response. Genes associated with cellular structures and signal transduction increased in relative abundance, while those involved in amino acid and carbohydrate transport and metabolism declined. Together with the lower relative abundance of ARGs, these patterns are consistent with a possible trade-off between maintaining resistance traits and coping with acute heat stress.

Environmental factors associated with ARG responses also differed between the disturbance and recovery phases. Immediate responses were closely associated with soil pH, temperature, and carbon and nitrogen status. During recovery, responses were linked more closely to soil moisture, water-holding capacity, and seasonal variation in temperature and precipitation.

The researchers also used soil properties and climate data to build random forest models that predict changes in ARG composition. Spatial cross-validation supported the models' predictive ability within the environmental range covered by the study, although performance varied among disturbance types.

According to the researchers, the results show that both the type of climate extreme and local environmental conditions play important roles in shaping soil ARG responses. The findings could help researchers anticipate how climate extremes may alter soil reservoirs of antibiotic resistance and incorporate these changes into One Health research, which considers human, animal, and environmental health as interconnected.