EXTRACELLULAR DNA IN SOIL SYSTEMS: STABILIZATION/DEGRADATION DYNAMICS AND IMPACTS ON MICROBIAL COMMUNITY ANALYSIS

Abstract

Extracellular DNA (eDNA) is any single- or double-stranded DNA molecule not contained within a cell. This eDNA is ubiquitous in soils, where it arises from either cell lysis or DNA secretion, and where it may persist for many years. Interest in the issue of soil eDNA has recently been piqued in the fields of molecular biology and microbial ecology because eDNA has the potential to bias sequence-based estimates of microbial community composition, inflate measures of alpha and beta diversity, and interfere with the detection of community shifts over time. In the first chapter, we examine the impacts of soil moisture, soil temperature, agricultural management, and habitat type on the degradation/persistence dynamics of eDNA in soil microcosms, using a synthetic eDNA marker which was traceable with both sequence-specific qPCR and 16S rRNA community sequencing. We found that despite very rapid degradation within the first week, a small fraction (< 1%) of the eDNA standard remained detectable with qPCR throughout the experiments (39 - 77 days). This suggests that eDNA may be indefinitely stabilized within soil. We also found that degradation/stabilization dynamics differed across gradients of environmental conditions and soil characteristics, with initial degradation rate (within the first week) being positively correlated with soil moisture, temperature, and tillage intensity, but negatively correlated with soil organic matter content. Longer-term stabilization (> 39 days) of the eDNA standard was highest at low moisture, low temperature, and low tillage intensity, but was not significantly correlated with soil organic matter. Additionally, among agricultural, forest, and meadow soils we found that forest soils had the slowest initial degradation rate, and meadow soils had the most stabilization of the eDNA standard. The eDNA standard was detectable by qPCR at all time-points for all treatments, but within the first week became only inconsistently detectable with high-throughput sequencing, the eDNA standard having dropped below the limit of detection for 16S rRNA gene sequencing. The time to disappearance below the sequencing limit of detection was calculated as ranging from 0.9 - 19.4 days, depending on treatment conditions. Therefore, we conclude that the ability of stabilized soil eDNA to bias estimates of microbial community structure depends on the sensitivity of the detection method and the objectives of the experiment. In the second chapter, we place the first chapter into a wider context by reviewing the impacts of eDNA on estimates of microbial communities in soils, and discussing evidence that eDNA-driven bias is or is not a problem in community structure estimates. We discuss the factors that influence the degradation/stabilization dynamics of eDNA in soils, techniques to reduce eDNA-driven bias, potential opportunities to exploit eDNA as a powerful tool for microbial community characterization, and important directions for future eDNA-related research. Taken together, this work constitutes a contribution of new knowledge and insight to the field of soil microbial ecology. This research and analysis will improve the ability of researchers to accurately characterize microbial communities using culture-independent methods, while understanding the extent to which eDNA may introduce bias to those estimates.