From Lake Victoria to the tap : antibiotic resistance and pathogenic contamination of Kisumu City water supply and wastewater network

Abstract

Waterborne diseases and antimicrobial resistance (AMR) pose mounting public health threats across sub-Saharan Africa, particularly in rapidly urbanising regions dependent on untreated or poorly treated surface waters. This study applied shotgun metagenomic sequencing to characterise microbial communities, virulence factors and antibiotic resistance genes (ARGs) in water samples collected from Lake Victoria, River Wigwa, Dunga Water Treatment Plant, Nyalenda Wastewater Stabilisation Ponds and the tap water outlet in post-treatment supply pipe in Kisumu city (Kenya). Bacterial taxa dominated all metagenomes, with 121 classes represented. Cyanobacteria, particularly Planktothrix, were highly abundant in lake and tap water, whereas wastewater and river samples exhibited greater taxonomic diversity. Major human pathogens, including Pseudomonas aeruginosa, Klebsiella pneumoniae, Escherichia coli, Acinetobacter baumannii and Bacillus cereus/anthracis, were detected in nearly all samples, with unexpectedly high prevalence in tap water. Viral indicators of faecal contamination (adenoviruses, enteroviruses and torque teno viruses) corroborated widespread wastewater influence. Functional gene profiling revealed a rich resistome comprising aminoglycoside-modifying enzymes, β-lactamases, vancomycin-resistance operons and disinfectant-resistance determinants. The highest ARG and virulence gene frequencies occurred in tap and treatment-plant water, suggesting that incomplete disinfection and biofilm persistence promote the proliferation and exchange of ARGs between environmental and pathogenic taxa. In contrast, Lake Victoria water exhibited lower ARG abundance, reflecting natural self-purification processes. These findings underscore the inadequate water treatment and open wastewater systems create ecological ‘hotspots’ for ARG selection and horizontal gene transfer. Metagenomic surveillance integrated into One Health frameworks can enhance risk forecasting and guide interventions to mitigate AMR emergence and dissemination in freshwater systems serving over 35 million people across the Lake Victoria basin.

Description

DATA AVAILABILITY STATEMENT : The data that support the findings of this study are openly available in NCBI at http://www.ncbi.nlm.nih.gov/bioproject/1345882, reference number PRJNA1345882. SUPPLEMENTARY MATERIAL FIGURE S1 : Geographical locations of sampling points: 1—tap water (sample D002); 3—Lake Victoria (sample D003); 4—river Wigwa (sample D004); 5—stabilisation pond (sample D005); 6—sedimentation tank at the water treatment plant (sample D006). The map was exported from the OpenStreetMap server (https://www.openstreetmap.org/copyright?utm_source=chatgpt.com) operating under the Open Data Commons Open Database Licence (ODbL). Photos used in this figure are from O. Reva's private collection. FIGURE S2 : (A) Principal Component Analysis (PCA) plot of sampled microbiomes: D001—tap water; D003—Lake Victoria; D004—river Wigwa; D005—stabilisation pond; D006—sedimentation tank at the water treatment plant. (B) Euclidian distances between sampled microbiomes. Both PCA plot and distance matrix were calculated based on taxonomic profiles of bacterial genera using the program Past 4.02. DNA read counts associated with different genera were normalised prior to analysis using the centred log-ratio (CLR) transformation. TABLE S1 : Bacterial genera identified in metagenomic water samples.

Keywords

Waterborne diseases, Antimicrobial resistance (AMR), Antibiotic resistance gene (ARG), Metagenomics, Water microbiome, Wastewater, Pathogen surveillance

Sustainable Development Goals

SDG-06: Clean water and sanitation

Citation

Reva, O.N., Sifuna, A., Orata, F. et al. 2026, 'From Lake Victoria to the tap: antibiotic resistance and pathogenic contamination of Kisumu City water supply and wastewater network', Tropical Medicine and International Health, doi : 10.1111/tmi.70105.