A metagenomic analysis framework for characterization of antibiotic resistomes in river environment: Application to an urban river in Beijing.

River is considered generally as a natural reservoir of antibiotic resistance genes (ARGs) in environments. For the prevention and control of ARG risks, it is critical to comprehensively characterize the antibiotic resistomes and their associations in riverine systems. In this study, we proposed a metagenomic framework for identifying antibiotic resistomes in river sediments from multiple categories, including ARG potential, ARG hosts, pathogenicity potential, co-selection potential and gene transfer potential, and applied it to understand the presence, hosts, and co-occurrence of ARGs in the sediments of an urban river in Beijing. Results showed that a total of 203 ARG subtypes belonging to 21 ARG types were detected in the river sediments with an abundance range of 107.7-1004.1×/Gb, dominated by multidrug, macrolide-lincosamide-streptogramin, bacitracin, quinolone and sulfonamide resistance genes. Host-tracking analysis identified Dechloromonas, Pseudoxanthomonas, Arenimonas, Lysobacter and Pseudomonas as the major hosts of ARGs. A number of ARG-carrying contigs (ACCs) were annotated as fragments of pathogenic bacteria and carried multiple multidrug-ARGs. In addition, various biocide/metal resistance genes (B/MRGs) and mobile genetic elements (MGEs), including prophages, plasmids, integrons and transposons, were detected in the river sediments. More importantly, the co-occurrence analysis via ACCs showed a strong association of ARGs with B/MRGs and MGEs, indicating high potential of co-selection and active horizontal transmission for ARGs in the river environment, likely driven by the frequent impact of anthropogenic activities in that area.