Abstract
Environmental DNA (eDNA) metabarcoding is a relatively new monitoring tool featuring in an increasing number of applications such as the facilitation of the accurate and cost effective detection of species in environmental samples. eDNA monitoring is likely to have a major impact on the ability of salmonid aquaculture industry producers and their regulators to detect the presence and abundance of pathogens and other biological threats in the surrounding environment. However, for eDNA metabarcoding to develop into a useful bio-monitoring tool it is necessary to (a) validate that sequence datasets derived from amplification of metabarcoding markers reflect the true species’ identity, (b) test the sensitivity under different abundance levels and environmental noise and (c) establish a low-cost sequencing method to enable the bulk processing of field samples. In this study, we employed an elaborate experimental design whereby different combinations of five biological agents were crossed at three abundance levels and exposed to sterile pre-filtered and unfiltered seawater, prior to coarse filtering and then eDNA ultrafiltration of the resultant material. We then benchmarked the low-cost, scalable, Ion Torrent sequencing method against the current gold-standard Illumina platform for eDNA surveys in aquaculture. Based on amplicon-seq of the 18S SSU rDNA v9 region, we were able to identify two parasites (Lepeophtheirus salmonis and Paramoeba perurans) to species level, whereas the microalgae species Prymnesium parvum, Pseudo-nitzschia seriata, and P. delicatissima could be assigned correctly only to the genus level. Illumina and Ion Torrent provided near identical results in terms of community composition in our samples, whereas Ion Torrent was more sensitive in detecting species richness when the medium was unfiltered seawater. Both methods were able to reflect the difference in relative abundance between treatments in 4 out of 5 species when samples were exposed to the unfiltered seawater, despite the significant amount of background noise from both bacteria and eukaryotes. Our findings indicate that eDNA metabarcoding offers significant potential in the monitoring of species harmful to aquaculture and for this purpose, the low-cost Ion Torrent sequencing is as accurate as Illumina in determining differences in their relative abundance between samples.
Highlights
The salmonid aquaculture industry is undergoing explosive growth globally
P. perurans is the causative agent of amoebic gill disease, which is a major source of commercial loss for aquaculture in Tasmania, and affects industries in both North and South America as well as in Europe (Young et al, 2008)
The Illumina MiSeq data consisted of fewer raw reads (6,115,810 paired reads) across all samples compared with the Ion Torrent output (9,350,400 single reads)
Summary
The salmonid aquaculture industry is undergoing explosive growth globally. The industry is beset by parasitic disease and is often the subject to mass mortalities of farmed fish due to toxin-producing Harmful Algal Blooms (HABs) (Smayda, 2006; Hinder et al, 2011). Economic losses associated with certain agents, as for example sea lice, accounts for up to £ 470 M/year for major producers like Norway (Liu and Bjelland, 2014). The presence and abundance thresholds of these potentially damaging organisms in the environment and around aquaculture sites must be closely monitored. Traditional microscopy methods for algal and copepod larval species identification are timeconsuming, demand expertise and are not always accurate when abundances are low or when cryptic species are involved. Parasite counts on the fish themselves are both time consuming and impose significant handling stress
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
