Grab what you can—an evaluation of spatial replication to decrease heterogeneity in sediment eDNA metabarcoding

Abstract

Environmental DNA methods such as metabarcoding have been suggested as possible alternatives or complements to the current practice of morphology-based diversity assessment for characterizing benthic communities in marine sediment. However, the source volume used in sediment eDNA studies is several magnitudes lower than that used in morphological identification. Here, we used data from a North Sea benthic sampling station to investigate to what extent metabarcoding data is affected by sampling bias and spatial heterogeneity. Using three grab parallels, we sampled five separate sediment samples from each grab. We then made five DNA extraction replicates from each sediment sample. Each extract was amplified targeting both the 18S SSU rRNA V1–V2 region for total eukaryotic composition, and the cytochrome c oxidase subunit I (COI) gene for metazoans only. In both datasets, extract replicates from the same sediment sample were significantly more similar than different samples from the same grab. Further, samples from different grabs were less similar than those from the same grab for 18S. Interestingly, this was not true for COI metabarcoding, where the differences within the same grab were similar to the differences between grabs. We also investigated how much of the total identified richness could be covered by extract replicates, individual sediment samples and all sediment samples from a single grab, as well as the variability of Shannon diversity and, for COI, macrofaunal biotic indices indicating environmental status. These results were largely consistent with the beta diversity findings, and show that total eukaryotic diversity can be well represented using 18S metabarcoding with a manageable number of biological replicates. Based on these results, we strongly recommend the combination of different parts of the surface of single grabs for eDNA extraction as well as several grab replicates, or alternatively box cores or similar. This will dilute the effects of dominating species and increase the coverage of alpha diversity. COI-based metabarcoding consistency was found to be lower compared to 18S, but COI macrofauna-based indices were more consistent than direct COI alpha diversity measures.

Improved biodiversity detection using a large-volume environmental DNA sampler with in situ filtration and implications for marine eDNA sampling strategies

“…Sampling approaches and theory are understudied aspects of eDNA protocols (Dickie et al, 2018), and future work should evaluate the optimal sampling volume and strategy as a function of the environment and the biology of target taxa (Mächler et al, 2016). Studies in other environments have similarly demonstrated that increasing sample volumes can improve biodiversity detection (Bessey et al, 2020;Hestetun et al, 2021;Schabacker et al, 2020;Sepulveda et al, 2019). Because our eDNA sampler can efficiently pump a much larger volume than that which can be captured by a single Niskin bottle, it represents a better tool for collecting eDNA at deeper ocean depths (i.e., below ~100 m).…”

Through the eDNA looking glass: Responses of fjord benthic foraminiferal communities to contrasting environmental conditions

“…The underperformance by the 0.5‐g sample analyses may be resolvable with the collection of several (pseudo‐) replicate samples (i.e. > 2), or subsampling homogenized samples of larger volume (Hestetun, Lanzén, & Dahlgren, 2021; Hestetun, Lanzén, Skaar, & Dahlgren, 2021; van der Loos & Nijland, 2021). Currently, there is no consensus regarding the appropriate level of replication for assessment of total diversity, especially when considering effort/cost vs. benefit (e.g., review by Pawlowski et al, 2022).…”

Soil environmental DNA metabarcoding in low-biomass regions requires protocol optimization: a case study in Antarctica

“…Lear et al (2018) showed that the 18S gene is the most popular marker for targeting eukaryotes in metabarcoding research. 18S primers have been shown to outperform primers from other genes (von Ammon et al 2018, Sawaya et al 2019, Tytgat et al 2019, Hestetun et al 2021). The V9 subregion of the eukaryotic SSU (18S) rRNA gene - used in this research - has both highly conserved and highly variable sites, thus showing great taxonomic resolution potential (Amaral-Zettler et al 2009, Wu et al 2015).…”