Abstract
The impact of two pesticides (S-metolachlor and propiconazole) and their respective main metabolites (ESA-metolachlor and 1,2,4-triazole) on bacterial denitrification in groundwater was studied. For this, the denitrification activity and the bacterial diversity of a microbial community sampled from a nitrate-contaminated groundwater were monitored during 20 days in lab experiments in the presence or absence of pesticides or metabolites at 2 or 10 μg/L. The kinetics of nitrate reduction along with nitrite and N(2)O production all suggested that S-metolachlor had no or only little impact, whereas its metabolite ESA-metolachlor inhibited denitrification by 65% at 10 μg/L. Propiconazole and 1,2,4-triazole also inhibited denitrification at both concentrations, but to a lesser extent (29-38%) than ESA-metolachlor. When inhibition occurred, pesticides affected the reduction of nitrate into nitrite step. However, no significant differences were detected on the abundance of nitrate reductase narG and napA genes, suggesting an impact of pesticides/metabolites at the protein level rather than on denitrifying bacteria abundance. 16S rRNA gene Illumina sequencing indicated no major modification of bacterial diversity in the presence or absence of pesticides/metabolites, except for ESA-metolachlor and propiconazole at 10 μg/L that tended to increase or decrease Shannon and InvSimpson indices, respectively. General growth parameters suggested no impact of pesticides, except for propiconazole at 10 μg/L that partially inhibited acetate uptake and induced a decrease in microbial biomass. In conclusion, pesticides and metabolites can have side effects at environmental concentrations on microbial denitrification in groundwater and may thus affect ecosystem services based on microbial activities.