Four antibiotics[azithromycin (AZM), sulfamethoxazole (SMZ), ciprofloxacin (CIP), and tetracycline (TCY)], and the antibiotic resistance genes (ARGs)[sulfonamides (sul1 and sul2), tetracyclines (tetX and tetM), quinolones (qnrS and qnrD), macrolides (ermB), and 16S rDNA] were selected as target compounds. Artificial ecosystems were constructed with combinations of two emergent plants and Microcystis aeruginosa (Acorus calamus+Cordyceps, algae+Cordyceps, algae+Acorus calamus, and algae+Acorus calamus+Cordyceps) in an indoor-simulated river system. Throughout the artificial ecosystems, changes in antibiotic concentrations and other pollution indicators (i.e., COD, NH4+-N, TP, and TN) were monitored in different media (the aqueous phase, sediment phase, and in plants), and the distribution and removal of ARGs in aqueous and sediment phases were explored. Removal of the target compounds was calculated based on mass balance, and the correlation between ARG abundance and environmental factors in the aqueous and sediment phases was analyzed. The results showed that the constructed artificial ecosystem achieved removal rates of COD, NH4+-N, TP, and TN ranging from 60.2% to 74.8%, 63.4% to 77.4%, 64.0% to 73.2%, and 46.8% to 54.8%, respectively. The antibiotics in the aqueous phase were notably removed and the artificial ecosystem 'algae+Acorus calamus+Cordyceps' achieved the best removal efficiency for the four antibiotics. Removal rates of the antibiotics in the sediment phase were ranked in the order TCY>CIP>AZM>SMZ; the removal efficiency of TCY in the 'algae+Acorus calamus+Cordyceps' system reached up to 53.5%. The total removal rates of antibiotics obtained by the ecosystems were ranked in the following order:algae+Acorus calamus+Cordyceps > algae+Cordyceps > algae+Acorus calamus > Acorus calamus+Cordyceps. Removal of the four ARGs was very efficient and was higher in the aqueous phase than in the sediment phase. Correlations between the ARGs, the other pollution indicators, and the antibiotics were variable; tetX and environmental factors were correlated in the aqueous phase, while AZM and its corresponding ARGs were not significantly correlated in the sediment phase. The results showed that ARGs can be targeted under corresponding antibiotic pressure and other types of environmental pressure. In the study system, the concentrations of antibiotics did not directly affect the transmission of ARGs. Overall, this study shows that artificial ecosystems constructed with emergent plants and Microcystis aeruginosa can be effective at purifying water and reducing the environmental risks of antibiotics in urban rivers.
Keywords: Microcystis aeruginosa; antibiotic resistance genes; antibiotics; emergent plant; redundancy analysis; urban rivers.