We investigated the effects of abiotic conditions on DNA degradation in different types of DNA samples in a non-controlled environment. We hypothesized three scenarios to understand the degradation process of DNA samples. The first scenario uses as a model sample simulating the presence of a species with a high release of DNA concentration (samples with L. fortunei DNA extract, initial average concentration of 1.6 ng/µL). The second scenario uses as a model a species with an intermediate release of DNA (samples with L. fortunei eDNA, initial average concentration of 0.015 ng/µL). And the third scenario uses a species releasing a low concentration of DNA (Cordylophora sp. eDNA samples, initial average concentration of 0.000048 ng/µL).
Our results show a strong effect of DNA degradation in samples under the influence of environmental variables, with DNA degraded about 16 times faster in samples with DNA extracts than in samples with environmental DNA from L. fortunei. The abiotic factors that most accelerate DNA degradation are time plus depth for the L. fortunei DNA extract, time, and turbidity interacting with time for Cordylophora sp. eDNA samples e just time for eDNA samples of L. fortunei.
Many abiotic factors in water samples can affect the detection of eDNA (Stewart, 2019). In a non-controlled environment, as in our study, the abiotic factors are correlated and interact with each other. In our study, the depth and turbidity associated with time influenced the degradation of DNA molecules in the limnetic environment depending of species and experiment. As showed in the analysis of environmental data, depth, turbidity, and temperature vary together and our statistical analysis kept the parameter with higher explanatory power in the model, but all this variable may influenced the DNA degradation in a synergistic way (Barnes et al., 2014; Caza-Allard et al., 2021). Apart from depth and turbidity parameters, the water flow rate (Hauger et al., 2020; Song et al., 2017), temperature (Robson et al., 2016; Strickler et al., 2015), pH (Torti et al., 2015) and climatic changes such as heavy rainfall or severe drought (King et al., 2022) can also influence DNA degradation. Climate shapes certain characteristics of watercourses, such that heavy rainfall creates an eDNA dilution effect (Buxton et al., 2017; Harper et al., 2019) and increasing concentrations of suspended particles (Deiner et al., 2017; Stoeckle et al., 2017; Wang et al., 2021), while drought can create scenarios with a low probability of detection, due to low water levels and other associated factors (Deiner et al., 2016; Sales et al., 2019).
In the eDNA samples from experiment II, there was an increase in DNA concentration in the first 24 hours, in contrast to experiment I, with the DNA extract. This DNA peak can be explained by the degradation of tissues and cell organelles leading to an increased release of intracellular DNA into the environment (Mauvisseau et al., 2021). However, this dynamic is complex, as other factors, both abiotic (such as pH) and biotic (microbial and enzymatic activity), can even synergistically influence the process of DNA release and degradation (Caza-Allard et al., 2021).
Other studies have already shown that the degradation of eDNA is very rapid and can even take place within minutes in aquatic environments (Sassoubre et al., 2016; Yamanaka and Minamoto, 2016). This process is triggered by hydrolysis, oxidation and microbial activity, among others (Lindahl, 1993; Torti et al., 2015). This rapid degradation of eDNA in aquatic communities is a very interesting finding, as the short persistence of DNA allows monitoring of organisms present in the environment in virtually real time (Seymour et al., 2018; Stefanni et al., 2022).
The DNA degradation in our study were higher in the eDNA samples of L. fortunei than in the samples containing Cordylophora sp. Studies with species from very different taxonomic groups also showed different degradation rates, as in the study by Andruszkiewicz Allan et al. (2021), who reported lower concentration in samples from jellyfish, followed by fish and shrimp. These differences may be related to physiological conditions (age or life stage) and density (Pilliod et al., 2013) which ultimately influence the state of eDNA produced and released by the organisms and thus on the concentration (Caza-Allard et al., 2021). In the study by Sansom and Sassoubre (2017) it was described that most of the eDNA released by the bivalve Lampsilis siliquoidea is due to the excretion of faeces or tissue cells from the body cavity. The latter is not the case in the cnidarian Cordylophora, which has a body cavity with little fluid and a chitinous periderm (Slobodkin and Bossert, 2010), suggesting that the anatomy of the organism may favour or hinder the excretion of eDNA into the environment. Material containing eDNA from cnidarians can be collected by the presence of larvae in the sexual reproductive phase and menonts (dormant stage) in the resting phase (Agostinis, 2016).
It is also important to assess the biomass of the organisms studied. In another study conducted by our group (Bertão et al., 2021) at the José Richa HPP, a higher density of L. fortunei was found compared to cnidarians, which may indicate a positive correlation found in relation to the higher concentration and DNA of L. fortunei found in this study.We have studied the effects of exposure to light at depths of up to 10 m on samples stored in opaque and transparent containers, as there is no consensus on the effects of light on DNA degradation. One example is that some studies have found no effects of UV light on sample integrity (Andruszkiewicz et al., 2017; Machler et al., 2018; Merkes et al., 2014) in contrast to Pilliod et al. (2014), who reported greater DNA degradation in samples exposed to UV light. Such seemingly inconsistent results highlight the need for additional studies under different conditions to better investigate the effects of UV light on DNA integrity (Eichmiller et al., 2016; Pilliod et al., 2013; Strickler et al., 2015). Here we were able to show that the opaque and transparent bottles showed no significant difference (p > 0.05) and the luminosity had no effect on the degradation rate of the samples. In the eDNA experiment, the samples were placed on the cables near the safety signal buoys of the reservoir, which might have caused shading at the first experimental depths (0, 0.3 and 0.6 m). However, this bias does not seem to have had a significant effect on the decomposition rates, as they showed a similar pattern to the extract samples placed on the cables away from the signal buoys in experiment I.