Overview of environmental DNA

eDNA is defined as a pool of DNA isolated from environmental samples including sediment, soil, water, ice, air, feces, or even the surfaces of leaves (Pawlowski, Apothéloz-Perret-Gentil, and Altermatt 2020; Taberlet et al. 2012). eDNA can derive from whole organisms (e.g. diatoms in water) or gametes, or come from various tissues, secretions like mucus, blood, feces, urine, saliva, and shed skin, scales or hair. Mentions of eDNA appeared for the first time in the literature in 1987 (extraction of DNA from sediments; Ogram, Sayler, and Barkay 1987) and were extended to macro-organisms by the end of the 2000s (see Taberlet et al. 2018 for a review on the timeline for eDNA studies). eDNA differs from bulk samples, such as insects collected from malaise traps, because collection is generally passive and focused on environmental material. It also differs from soil samples taken with the intent of characterizing total diversity in soil via metagenomics, as the field often focuses on eukaryotic species. Note that the definition of eDNA is still debated (see Pawlowski, Apothéloz-Perret-Gentil, and Altermatt, 2020 for a review). In this manual we refer to the general concept of eDNA, that is a total pool of DNA isolated from environmental samples independently of its structural state (intra/extra-cellular) (Taberlet et al. 2012; Pawlowski, Apothéloz-Perret-Gentil, and Altermatt 2020).

Multiple factors shape the state and detectability of eDNA in the environment (review in Rourke et al. 2022). Some organisms are simply less detectable because they shed less DNA (Allan et al. 2021). Biomass, age, activity, and stress all relate to DNA concentration in the environment (Stewart 2019; Thalinger et al. 2021a). The probability of detecting any taxon using eDNA can also vary according to the stage of its life cycle (e.g. increased probability during the breeding season because of higher organism activity and release of gametes; Buxton et al. 2017; Buxton, Groombridge, and Griffiths 2018). eDNA exists in several states in the environment: as intracellular or intra-organellar DNA, dissolved DNA in the water, and particle adsorbed DNA (Mauvisseau, Harper, et al. 2022). As cells break down in the water, they release their mitochondria and/or chloroplasts. As cells and organelles undergo lysis, they release dissolved DNA, which may then adsorb onto suspended particles in the water (Mauvisseau, Harper, et al. 2022). Because there are typically many organelles and corresponding copies of mitochondrial/chloroplast DNA per cell, those genes are often preferred targets for eDNA assays.

Once DNA is released into the environment, several biotic and abiotic factors can influence its persistence and detection probability (Fig. 2). eDNA can diffuse and thus be transported tens of kilometres by external means such as human activities (e.g. boats) or animals (e.g. leeches, predator), or by internal sources such as stream flow (Deiner and Altermatt 2014). Three main processes control eDNA movement in stream: transport, retention, and resuspension (Shogren et al. 2017). Shogren et al. (2017) quantify transport as the distance an average eDNA particle travels in streams with different hydrologic signatures, and define retention as DNA that is trapped in porous benthic substrate interstices because of surface-subsurface exchanges, and resuspension as release of DNA from the streambed resulting in eDNA detection after the source of eDNA has been removed. While transport of eDNA may be an issue for deducing location of a detected species or local species richness, Deiner et al. (2016) showed that eDNA data gleaned from sampling stream networks can provide insights on biodiversity distribution across hierarchical spatial scales, including both terrestrial and aquatic species. In soil, eDNA derives mainly from organisms living below the ground surface (e.g. insect larvae, plant roots) and from the soil microbiome (Taberlet et al. 2018; Kirse et al. 2021).

Degradation of eDNA is affected by multiple key biotic and abiotic factors – light exposure, temperature, oxygen, pH, microbial activity and extracellular enzymes can shorten eDNA lifespan from a couple of week to only a few days (Barnes et al. 2014; Oehm et al. 2011; Strickler, Fremier, and Goldberg 2015). This means that eDNA signals can provide a relatively contemporary ‘snapshot’ of occurrences of taxa present in a sampled environment. However, in some environments where DNA degradation does not occur as rapidly (e.g. cold and dry sediments), eDNA can be preserved for (potentially) hundreds or even thousands of years and has been used to assess past biodiversity (Willerslev et al. 2003; Thomsen and Willerslev 2015). Recommendations on how to adjust sampling strategies to account for factors expected to decrease detection probability have been detailed in the practical guide for DNAbased methods for biodiversity assessment (Table 2 in Bruce et al. 2021). Overall, environmental parameters affecting eDNA state conversion and degradation are not completely understood (Mauvisseau, Harper, et al. 2022). Current best practice involves reporting as many parameters as possible, and using analytical controls during water sampling and processing.

Fig. 2 Illustration of the processes (shedding, degradation, retention, resuspension, transportation) influencing eDNA detection in the environment.

Standards for eDNA workflows and reporting are in development or have recently emerged (See CSA W214:21 and CSA W219:23). Current best practice includes meticulous documentation of assay development, sampling schema, multiple controls, filtration protocols, contamination precautions, metadata collection, wet-lab processing, dry lab processing, detection criteria, and statistical analyses (Goldberg, Turner, et al. 2016). See Fig. 3 for a workflow chart and recommended reporting at each stage.