Metagenomics combined with comprehensive validation as a public health risk assessment tool for urban and agricultural run-off

Early detection of emerging and life-threatening pathogens circulating in complex environments is urgently required to combat infectious diseases. This study proposed a public health risk assessment workflow with three stages, pathogen screening, pathogen genotyping, and risk assessment. In stage one, pathogens were screened with metagenomic sequencing, microfluidic chip, and qPCR. In stage two, pathogens were isolated and genotyped with multi-locus sequence typing (MLST) or conventional PCR. Finally, virulence genes from metagenomic data were assessed for pathogenicity. Two regions (Donggang and Zhanjiang) with potential public health concerns were selected for evaluation, each of which comprised of one urban and one farming wastewater sampling location. Overall, metagenomic sequencing reflected the variation in the relative abundance of medically important bacteria. Over 90 bacterial pathogens were monitored in the metagenomic dataset, of which 56 species harbored virulence genes. In Donggang, a pathogenic Acinetobacter sp. reached high abundances in 2018 and 2020, whereas all pathogenic Vibrio spp. peaked in October 2019. In Zhanjiang, A. baumanni, and other Enterobacteriaceae species were abundantly present in 2019 and 2020, whereas Aeromonas and Vibrio spp. peaked in November-2017. Forty species were subsequently isolated and subtyped by MLST, half of which were prevalent genotypes in clinical data. Additionally, we identified the African Swine Fever Virus (ASFV) in water samples collected in 2017, ahead of the first reported ASFV outbreak in 2018 in China. RNA viruses like Hepatitis A virus (HAV) and Enterovirus 71 (EV71) were also detected, with concentrations peaking in April 2020 and April 2018, respectively. The dynamics of HAV and EV71 were consistent with local epidemic trends. Finally, based on the virulence gene profiles, our study identified the risk level in wastewater of two cities. This workflow illustrates the potential for an early warning of local epidemics, which helps to prioritize the preparedness for specific pathogens locally.