Water metagenomes reflect physicochemical water quality throughout a model agricultural pond.

Agricultural ponds are essential irrigation resources, though may also serve as reservoirs for pathogens and antimicrobial resistance (AMR) genes. While monitoring microbiological water quality is critical for food safety, the influence of sampling factors (e.g., when and where to collect samples) in making risk assessments and potential applications for using environmental covariates as indicators remain unclear. Here, we explored the hypothesis that metagenomes of agricultural waters change with spatiotemporal shifts in physicochemical water quality, i.e., across water depths over time. Water samples and underlying sediments were collected at a model pond at the surface and within the water column (0, 1, 2 m depths) throughout one day (i.e., 9:00, 12:00, 15:00). All samples were processed for shotgun metagenomic sequencing analysis and enumeration of various water quality parameters (e.g., temperature, nutrient concentrations, turbidity, pH, culturable Escherichia coli). At the pond surface, Microcystis aeruginosa and members of Cyanobacteria, along with genes encoding pathways related to photosynthesis and nucleotide biosynthesis, were enriched throughout the day. In contrast, within the water column (1-2 m depths) and sediments, diverse members of Proteobacteria and Actinobacteria were more dominant, along with encoded pathways related to respiration and amino acid biosynthesis. Various aspects of water quality (i.e., chlorophyll dissolved organic matter, ammonia, E. coli concentrations) correlated with water metagenome diversity, albeit not with any specific AMR genes or virulence factors. Nevertheless, de novo assembly of sequenced reads uncovered 22 unique strains encoding several AMR, virulence, or stress response genetic elements, thus linking metagenome functional potential to key taxa. Overall, our findings highlight distinctions in agricultural pond water metagenomes at the surface and in the water column and demonstrate the potential for metagenomic surveillance in water quality monitoring to support food safety.