Inadequate Waste Management Drives the Spread of Antimicrobial Resistance across the Watershed of a Large African City

Rapid urbanization in the Global South consistently outpaces sanitation infrastructure, transforming urban rivers into AMR reservoirs. Yet watershed-scale quantification of resistomes in such settings remains scarce, limiting exposure risk assessment and intervention planning. Here we present the most detailed spatial baseline for waterborne AMR in sub-Saharan Africa, using the Akaki River (Addis Ababa, Ethiopia) as a model system for rapidly urbanizing megacity watersheds with inadequate sanitation. We sampled 38 sites across an extreme pollution gradient during the dry season, when low flow produces maximum pollutant concentrations and the highest resident exposure risk. Water quality, microbiome marker data, 309 antibiotic resistance genes (ARGs), and 57 mobile genetic elements (MGEs) were quantified by high-throughput quantitative PCR. ARG and MGE abundances in central urban reaches exceeded 1010 gene copies L-1, indistinguishable from untreated sewage. Critically, clinically important ARGs co-occurred widely with broad-host-range MGEs, including the IncP-1α plasmid backbone, IS26, ISEcp1, IS6100, and class 1 integron integrase genes, indicating that ARGs are primed for potential horizontal gene transfer in waters where hundreds of thousands of Addis residents bathe, wash, and irrigate food crops. Dissolved oxygen depletion and elevated NH4-N correlated significantly with resistome expansion (p < 0.01), serving as robust chemical proxies for diffuse, system-wide raw sewage inputs that carry the actual mechanistic AMR drivers (e.g., heavy metals, antibiotic residues, and resistant hosts). The 38-site spatial design and dry-season worst-case conditions together provide a replicable model for AMR exposure assessment in comparable Global South cities.