Strong variation in sedimental antibiotic resistomes among urban rivers, estuaries and coastal oceans: Evidence from a river-connected coastal water ecosystem in northern China.

Sediment is thought to be a vital reservoir to spread antibiotic resistance genes (ARGs) among various natural environments. However, the spatial distribution patterns of the sedimental antibiotic resistomes around the Bohai Bay region, a river-connected coastal water ecosystem, are still poorly understood. The present study conducted a comprehensive investigation of ARGs among urban rivers (UR), estuaries (ES) and Bohai Bay (BHB) by metagenomic sequencing. Overall, a total of 169 unique ARGs conferring resistance to 15 antimicrobial classes were detected across all sediment samples. The Kruskal-Wallis test showed that the diversity and abundance of ARGs in the UR were all significantly higher than those in the ES and BHB (p < 0.05 and p < 0.01), revealing the distance dilution of the sedimental resistomes from the river to the ocean. Multidrug resistance genes contained most of the ARG subtypes, whereas rifamycin resistance genes were the most abundant ARGs in this region. Our study demonstrated that most antimicrobial resistomes were highly accumulated in urban river sediments, whereas beta-lactamase resistance genes (mainly PNGM-1) dramatically increased away from the estuary to the open ocean. The relative abundance of mobile genetic elements (MGEs) also gradually decreased from rivers to the coastal ocean, whereas the difference in pathogenic bacteria was not significant in the three classifications. Among MGEs, plasmids were recognized as the most important carriers to support the horizontal gene transfer of ARGs within and between species. According to co-occurrence networks, pathogenic Proteobacteria, Actinobacteria, and Bacteroidetes were recognized as potential and important hosts of ARGs. Heavy metals, pH and moisture content were all recognized as the vital environmental factors influencing the distribution of ARGs in sediment samples. Overall, the present study may help to understand the distribution patterns of ARGs at a watershed scale, and help to make effective policies to control the emergence, spread and evolution of different ARG subtypes in different habitats.