Source-tracking of antibiotic resistance genes in the watershed using molecular profiling and geospatial analyses
Date
2009
Journal Title
Journal ISSN
Volume Title
Abstract
Antibiotic resistance genes (ARG) have been found in many environmental matrices, including soils, groundwater, surface water, and sediments. Agricultural feeding operations and wastewater treatment plants are potential sources of ARG in rivers, or are sources of antibiotics that may select ARG from native river bacteria. The aim of this research is to identify ARG profiles that can characterize potential sources of ARG as well as native river environments and then use this knowledge to determine the sources and mechanisms involved in the spread of ARG to river environments. Initially, three wastewater treatment plants, six animal feeding operation lagoons, three sites along a pristine region of the Cache la Poudre River (PR), and a wildlife fish hatchery and rearing unit were compared with respect to the distribution, levels, and phylogenetic diversity of their ARG profiles. The tet genes tet(H), tet(Q), tet(S), and tet(T) were found to indicate agricultural influence, while high detection frequencies of tet(C), tet(E), and tet(O) were more typical of WWTP profiles. Sul(I) was detected in 100% of samples from source environments, but just once in the pristine river environment. The ARG profile of the pristine PR was dominated by tet(M) and tet(W), demonstrating their presence in an environment does not indicate anthropogenic disturbance. The tet(W) clone libraries from Pristine PR, WWTPs, and AFO lagoons, are each unique, as determined by both restriction fragment length polymorphism (RFLP) and phylogenetic analysis. Secondly, samples from the PR and South Platte River (SPR) in Northern Colorado were characterized with respect to the distribution, levels, and diversity of their ARG profiles. On the basis of the ARG indicator variables derived in the study of source environments, most river samples were classified as WWTP influenced by discriminant analysis. The relationship between spatial explanatory variables and the ARG response variables was determined with classification and regression tree (CART) analysis. There was good agreement between the classification of river sites according to spatial variables and source indicator variables, demonstrating the effectiveness of these indicators in source-tracking ARG. According to multivariate linear regression, sul(I) was significantly correlated with the inverse-distance weighted (IDW) number of cattle upstream of each river point (R2 = 0.83, p <0.0003), whereas tet(W) was not correlated with any explanatory variable tested. Tet(W) was isolated from two river environments: site PR4, located in Weld County downstream of Fort Collins; and site SPR3, located downstream of the confluence of the PR with the SPR. When compared to an existing clone library of tet(W) genes from animal feeding operations and wastewater treatment plants, PR4 was significantly different from the animal feeding operations (p<0.05); the SPR confluence (SPR3) was not significantly different from either environment. The PR4 environment was most similar to that of wastewater treatment plants, while SPR3 showed equal similarity with both source environments. A link between ARG indicator variables and spatial indicators was established. Furthermore, it was demonstrated that the ARG profiles of river samples were more similar to WWTPs than AFO lagoons or the pristine river. Based on this work, transport of ARG from sources may be a reasonable mechanism for ARG proliferation in riverine environments.
Description
Rights Access
Subject
antibiotic resistance
source tracking
wastewater
environmental engineering
water resources management