Fate and removal of bacteria and antibiotic resistance genes in horizontal subsurface constructed wetlands: Effect of mixed vegetation and substrate type.

This study aimed to investigate the influence of cropping method and substrate type on the fate and the removal of bacterial and antibiotic resistance genes (ARGs) indicators from primary wastewater by constructed wetlands (CWs) during startup and maturation stages. Four small-scale CWs differing in their plantation pattern (monoculture vs. polyculture) and substrate type were constructed and operated under field conditions. While for bacteria, the greatest impact of the cropping method and substrate type on removal was during the startup stage rather than the maturation stage, for ARGs, such impact was significant at both stages. During startup, the removal efficiencies of heterotrophic bacteria, fecal coliforms, E. coli, 16S rRNA genes and lacZ increased with the operation time. At maturation, the removal efficiencies were constant and were within the range of 89.2-99.4%, 93.7-98.9%, 89-98.8%, 94.1-99.6% and 92.9-98.7%, respectively. The removal efficiencies of intl1, tetM, intl1, sul1, ermB and total ARGs were also increased with the operation time. However, they were ARG type and configuration-dependent; at maturation they ranged between 50.7%-89.4%, 85.9%-97%, 49.6%-92.9%, 58.2%-96.7% and 79.9-94.3%, respectively. The tuff-filled serially planted CW was also the only one capable of removing these genes at similar high efficiency. Metagenomic analysis showed that none of the ARGs was among the most common ARGs in water and biofilm samples; rather most ARGs belonged to bacterial efflux transporter superfamilies. Although ARGs were removed, they were still detected in substrate biofilm and their relative concentrations were increased in the effluents. While the removal of both bacteria and ARGs was higher during summer compared to winter, the season had no effect on the removal pattern of ARGs. Hence, combination of the serial plantation with substrate having high surface area is a potential strategy that can be used to improve the performance of CWs.