Divergent climatic, geographic, and taxonomic associations of human gut resistome burden and composition across 34 countries

Background Understanding how antimicrobial resistance (AMR) relates to gut microbial community structure is essential for interpreting the human gut resistome in its ecological context. Methodology We investigated the relationships between resistance burden, resistance-class composition, and microbial taxonomic profiles in 98 healthy young adults from 34 countries, recruited among first-year medical students. Country-level pooled shotgun metagenomics was integrated with individual-level 16S rRNA gene sequencing. Results At the country-pool level, clustering of resistance-class profiles identified three clusters (RCPC1-RCPC3). RCPC1 combined the highest AMR burden with greater relative abundance of taxa reported as potentially pathogenic, genotoxic, or aerotolerant. This cluster also showed a higher representation of putative short-chain fatty acid-producing taxa, indicating that elevated resistance burden can coexist with taxa associated with beneficial metabolic functions. At the individual level, clustering of 16S-based predicted metabolic pathways identified four clusters (PMC1-PMC4) sharing a broadly conserved functional background. Country-level analyses further distinguished factors associated with resistance abundance from those associated with resistance-class composition. Mean annual temperature showed an exploratory positive association with AMR abundance, corresponding to approximately 22% higher abundance per 10°C increase after adjustment for continental grouping and pool size. PM2.5 exposure showed a threshold-like pattern across WHO-defined tiers. Neither environmental variable was associated with resistance-class composition, whereas countries assigned to the Asian group showed enrichment of macrolide-lincosamide-streptogramin resistance. Conclusion Resistance burden and composition capture complementary features of the gut resistome and should be interpreted alongside microbial community structure. Higher resistance burden can coexist with putatively beneficial taxa, highlighting the limits of single-metric assessments and the potential value of integrated resistome-taxonomic profiling for One Health surveillance.