A global survey was performed with 122 aquatic metagenomic DNA datasets (92 lake water and 30 seawater) obtained from the Sequence Read Archive (SRA). Antibiotic resistance genes (ARGs) and metal resistance genes (MRGs) were derived from the dataset sequences via bioinformatic analysis. The relative abundances of ARGs and MRGs in lake samples were in the ranges ND (not detected)-1.34 × 100 and 1.22 × 10-3-1.98 × 10-1 copies per 16S rRNA, which were higher than those in seawater samples. Among ARGs, multidrug resistance genes and bacitracin resistance genes had high relative abundances in both lake and sea water samples. Multi-metal resistance genes, mercury resistance genes and copper resistance genes had the greatest relative abundance for MRGs. No significant difference was found between epilimnion and hypolimnion in abundance or the Shannon diversity index for ARGs and MRGs. Principal coordinates analysis and permutational multivariate analysis of variance (PERMANOVA) test showed that stratification and geography had significant influence on the composition of ARGs and MRGs in lakes (p < 0.05, PERMANOVA). Coastal seawater samples had significantly greater relative abundance and a higher Shannon index for both ARGs and MRGs than deep ocean and Antarctic seawater samples (p < 0.05, Kruskal-Wallis one-way ANOVA), suggesting that human activity may exert more selective pressure on ARGs and MRGs in coastal areas than those in deep ocean and Antarctic seawater.
Keywords: Bacitracin resistance genes; Mercury resistance genes; Metagenomic analysis; Principal coordinates analysis; Stratification.
Copyright © 2019 The Authors. Published by Elsevier Ltd.. All rights reserved.