Influence of DNA from non-viable sources on the riverine water and biofilm microbiome, resistome, mobilome, and resistance gene host assignments

Shotgun metagenomic studies have revealed the diversity, relative abundance, and hosts of antibiotic resistance genes (ARGs) across environmental matrices. There is motivation to combine this method with viability-based techniques to better define ARG hazard. The objectives of this study were to evaluate the performance of different methods for extracellular DNA (eDNA) and putative "non-viable" cell DNA separation to understand the influence on ARG-host assignments. Paired water and biofilm samples were collected along a land use gradient. To study putative "viable-cell" DNA, samples were treated with propidium monoazide (i.e., PMA-DNA). To study eDNA, intracellular and extracellular DNA were separated. qPCR revealed differences in total 16S rRNA gene copies in water for filter vs. centrifuge-concentrated samples, but otherwise there were no differences in gene copy concentrations between DNA fractions. Next, metagenomic sequencing was performed on PMA-DNA and total DNA extracts revealing significant differences between the two for bacterial community structure and ARG profiles. Putative viable taxa containing pathogenic ARG hosts were identified in biofilm and water. Removing PMA-bound DNA improved N50 and assembly mapping compared to total DNA extracts. This study demonstrates the impact of different sample preparation methods on informing the hazard potential associated with riverine ARGs in water and biofilm.