Water volume, biological and PCR replicates influence the characterization of deep-sea pelagic fish communities.

Abstract

The pelagic deep sea is challenging to investigate due to logistical constraints regarding access and collection of samples, however environmental DNA (eDNA) can potentially revolutionize our understanding of this ecosystem. Although recent advancements are being made regarding eDNA technology and autonomous underwater vehicles, no investigation has been performed to assess the impact of different experimental designs using gear found on many research vessels (i.e., CTD mounted with Niskin bottles). Here, we investigated the effects of sampled water volume, biological and PCR replicates in characterizing deep-sea pelagic biodiversity at the level of species and exact sequence variants (ESVs, representing intraspecific variation). Samples were collected at 450m depth at night in the northern Gulf of Mexico using Niskin bottles, and we targeted the fish community using the MiFish primer (12S rRNA). Our results show that 1L is insufficient to characterize deep-sea pelagic fish communities. The 5L and 10L treatments detected similar community structure (i.e., the combination of number of species and relative occurrence) and numbers of species per biological replicate, but the 10L treatment detected a higher total number of species, more ESVs, and a different community structure when considering ESVs. We found that five biological replicates can detect up to 80% of the species detected in this study in the water collected in both 5L and 10L treatments. PCR replicates also had an important role in species and ESV detection, which implies increasing PCR replicates if water volume is limited. We suggest that future studies collect at least 5L, 5 or more field replicates, and 5-10 PCR replicates to adequately investigate deep-sea pelagic biodiversity using eDNA, considering resource limitations. Our study provides guidance for future eDNA studies and a potential route to expand eDNA studies at a global scale.