Dr. Xiaoyu Shan (Caltech) argued that environmental stresses such as drought concentrate natural antibiotic compounds in soils and select for resistance. She presented lab incubations (paired wet vs. drought treatments) and a computational pipeline applied to public metagenomes showing increased abundances of both antibiotic biosynthesis genes and resistance genes under drought-like conditions across cropland, wetland, grassland and forest soils globally.
Shan highlighted a matching signal in clinical surveillance: when she compared hospital-level resistance rates to regional aridity indices, drier places tended to have higher proportions of resistant clinical isolates; this pattern remained after restricting to high-income countries to control for economic confounders. She cautioned that correlation is not causation and said further work is needed to establish specific soil-to-human transmission pathways and timescales. “We think this transfer event should be very recent before any mutations had time to accumulate,” she said when noting identical gene sequences found both in soil Streptomyces and in clinical isolates.
Shan recommended broader evaluation of environmental drivers (drought, warming) and multiomics/physiologic studies to identify mechanisms that could plausibly move resistance from soil to humans through farming, dust inhalation or food-chain contacts.