Long-lasting effect of mercury contamination on the soil microbiota and its co-selection of antibiotic resistance.

Antibiotic resistance genes (ARGs) in the environment are an exposure risk to humans and animals and is emerging as a global public health concern. In this study, mercury (Hg) driven co-selection of ARGs was investigated under controlled conditions in two Australian non-agricultural soils with differing pH. Soils were spiked with increasing concentrations of inorganic Hg and left to age for 5 years. Both soils contained ARGs conferring resistance to tetracycline (tetA, tetB), sulphonamides (sul1), trimethoprim (dfrA1) and the ARG indicator class 1 integron-integrase gene, intI1, as measured by qPCR. The last resort antibiotic vancomycin resistance gene, vanB and quinolone resistance gene, qnrS were not detected. Hg driven co-selection of several ARGs namely intI1, tetA and tetB were observed in the alkaline soil within the tested Hg concentrations. No co-selection of the experimental ARGs was observed in the neutral pH soil. 16S rRNA sequencing revealed proliferation of Proteobacteria and Bacteriodetes in Hg contaminated neutral and alkaline soils respectively. Multivariate analyses revealed a strong effect of Hg, soil pH and organic carbon content on the co-selection of ARGs in the experimental soils. Additionally, although aging caused a significant reduction in Hg content, agriculturally important bacterial phyla such as Nitrospirae did not regrow in the contaminated soils. The results suggest that mercury can drive co-selection of ARGs in contaminated non-agricultural soils over five years of aging which is linked to soil microbiota shift and metal chemistry in the soil.