Abstract:
Grazing is expected to exert a substantial influence on antibiotic resistance genes (ARGs) in grassland ecosystems. However, the precise effects of grazing on the composition of ARGs in grassland soils remain unclear. This is especially the case for grassland soils subject to long-term grazing. Here, we investigated ARGs and bacterial community composition in soils subject to long-term historic grazing (13–39 years) and corresponding ungrazed samples. Using a combination of shotgun metagenomics, amplicon analyses and associated soil physicochemical data, we provide novel insights regarding the structure of ARGs in grassland soils. Interestingly, our analysis revealed that long-term historic grazing had no impacts on the composition of ARGs in grassland soils. An average of 378 ARGs, conferring resistance to 14 major categories of antibiotics (80%), were identified in both grazing and ungrazed sites. Actinobacteria, Proteobacteria and Acidobacteria were the most prevalent predicted hosts in these soils and were also shown to harbour genetic capacity for multiple-resistant ARGs. Our results suggested that positive effects of bacterial community composition on ARGs could potentially be controlled by affecting MGEs. Soil properties had direct effects on the composition of ARGs through affecting the frequency of horizontal gene transfer among bacteria. Twelve novel ARGs were found in S. grandis steppe grasslands, indicating that different vegetation types might induce shifts in soil ARGs. Collectively, these findings suggest that soil properties, plants and microorganisms play critical roles in shaping ARG patterns in grasslands. Together, these data establish a solid baseline for understanding environmental antibiotic resistance in grasslands.
Description:
SUPPLEMENTARY MATERIAL : Fig. S1 Processes and mechanisms underlying impacts of grazing on antibiotic resistance genes (ARGs). MGEs: Mobile genetic elements. Fig. S2 Soil pH, content of total nitrogen (TN), total phosphorus (TP), total organic carbon (TOC), NH4-N, NO3-N, plant richness, and bacterial richness in each sample site. Values are mean ± SE. Fig. S3 The content of metallic elements in each sample site. The red dashed lines represent the risk screening value. Values are mean ± SE. Different letters indicate significant differences (P ≤ 0.05) between sample sites. Fig. S4 Four significant differential antibiotic resistance genes (ARGs; i.e. Tetracycline (only tet34), Aminogly-cosides (only APH2.llla), MexK, and abeM) in the ungrazed and grazing groups. Values are mean ± SE. * means significant difference at P ≤ 0.05 level. Fig. S5 Antibiotic resistance gene types in the fresh sheep feces sampled in the grazing plots. XJ and NM refer to the samples collecting in Sinkiang and Inner Mongolia, respectively. Fig. S6 The bacterial community compositions and relative abundance at the phylum level in each sample sites. 39-, 38-, and 19-year ungrazed (UN-39, UN-38 and UN-19) and corresponding long-term grazing grasslands (GN-1, GN-3 and GN-2) in the Inner Mongolia. 33- and 13-year ungrazed (US-33 and US-13) and corresponding long-term grazing grassland (GS) in Sinkiang. Fig. S7 The relationships between the relative abundance of total antibiotic resistance gene (ARGs) and mobile genetic elements (MGEs) in integrons (a) or plasmids (b). Table S1 Information about the locations, dominant species and management pattern of the sample sites. Table S2 12 unique antibiotic resistance genes (ARGs) in the GN-3 in Inner Mongolia Plateau. Table S3 Results of mantel correlations between the antibiotic resistance genes (ARGs) richness and heavy metals.