Antibiotic resistance gene profile changes in cropland soil after manure application and rainfall

Authors

Morgan A. Meyers, USDA-ARS, University of Nebraska-LincolnFollow
Lisa M. Durso, USDA-ARS, University of Nebraska-LincolnFollow
John E. Gilley, University of Nebraska-LincolnFollow
Heidi Waldrip, USDA-ARSFollow
Lana Castleberry, USDA-ARSFollow
Amy Schmidt, University of Nebraska-LincolnFollow

Date of this Version

2020

Citation

J. Environ. Qual. 2020;49:754–761.

DOI: 10.1002/jeq2.20060

Comments

2020 Author(s)

Abstract

Land application of manure introduces gastrointestinal microbes into the environ- ment, including bacteria carrying antibiotic resistance genes (ARGs). Measuring soil ARGs is important for active stewardship efforts to minimize gene flow from agri- cultural production systems; however, the variety of sampling protocols and target genes makes it difficult to compare ARG results between studies. We used polymerase chain reaction (PCR) methods to characterize and/or quantify 27 ARG targets in soils from 20 replicate, long-term no-till plots, before and after swine manure application and simulated rainfall and runoff. All samples were negative for the 10 b-lactamase genes assayed. For tetracycline resistance, only source manure and post-application soil samples were positive. The mean number of macrolide, sulfonamide, and inte- grase genes increased in post-application soils when compared with source manure, but at plot level only, 1/20, 5/20, and 11/20 plots post-application showed an increase in erm(B), sulI, and intI1, respectively. Results confirmed the potential for tempo- rary blooms of ARGs after manure application, likely linked to soil moisture lev- els. Results highlight uneven distribution of ARG targets, even within the same soil type and at the farm plot level. This heterogeneity presents a challenge for separating effects of manure application from background ARG noise under field conditions and needs to be considered when designing studies to evaluate the impact of best manage- ment practices to reduce ARG or for surveillance. We propose expressing normalized quantitative PCR (qPCR) ARG values as the number of ARG targets per 100,000 16S ribosomal RNA genes for ease of interpretation and to align with incidence rate data