Abstract
Characterisation of freshwater benthic biodiversity using DNA metabarcoding may allow more cost-effective environmental assessments than the current morphological-based assessment methods. DNA metabarcoding methods where sorting or pre-sorting of samples are avoided altogether are especially interesting, since the time between sampling and taxonomic identification is reduced. Due to the presence of non-target material like plants and sediments in crude samples, DNA extraction protocols become important for maximising DNA recovery and sample replicability. We sampled freshwater invertebrates from six river and lake sites and extracted DNA from homogenised bulk samples in quadruplicate subsamples, using a published method and two commercially available kits: HotSHOT approach, Qiagen DNeasy Blood & Tissue Kit and Qiagen DNeasy PowerPlant Pro Kit. The performance of the selected extraction methods was evaluated by measuring DNA yield and applying DNA metabarcoding to see if the choice of DNA extraction method affects DNA yield and metazoan diversity results. The PowerPlant Kit extractions resulted in the highest DNA yield and a strong significant correlation between sample weight and DNA yield, while the DNA yields of the Blood & Tissue Kit and HotSHOT method did not correlate with the sample weights. Metazoan diversity measures were more repeatable in samples extracted with the PowerPlant Kit compared to those extracted with the HotSHOT method or the Blood & Tissue Kit. Subsampling using Blood & Tissue Kit and HotSHOT extraction failed to describe the same community in the lake samples. Our study exemplifies that the choice of DNA extraction protocol influences the DNA yield as well as the subsequent community analysis. Based on our results, low specimen abundance samples will likely provide more stable results if specimens are sorted prior to DNA extraction and DNA metabarcoding, but the repeatability of the DNA extraction and DNA metabarcoding results was close to ideal in high specimen abundance samples.
Highlights
Modern biomonitoring and water quality assessments of freshwater ecosystems rely on standardised sampling and identification of benthic macroinvertebrates (Bonada et al 2006)
The weights of subsamples were similar overall (Fig. 2a; two-way ANOVA, extraction factor F2,5 = 0.55, site factor F2,5 = 1.04, p > 0.05) in spite of a crossover interaction, which was due to HotSHOT extraction subsamples being significantly heavier in the 2 m samples and significantly lighter in the 0.25 m samples compared to the PowerPlant Kit subsamples (Tukey’s HSD on extraction at each site, p < 0.05)
Weights of Blood & Tissue Kit and HotSHOT extraction subsamples did not correlate with the DNA yield (Spearman’s ρ, p > 0.05)
Summary
Modern biomonitoring and water quality assessments of freshwater ecosystems rely on standardised sampling and identification of benthic macroinvertebrates (Bonada et al 2006). Subsampling is often implemented to expedite the sample handling and identifications rely on taxonomic expertise and knowledge (e.g. availability of taxonomic keys for a specific group and location); the same samples can produce different taxonomic lists depending on the identifier and the subsample examined (Haase et al 2010). In this situation, approaches based on DNA metabarcoding promise more efficient and standardised sample processing (Hajibabaei et al 2011, Yu et al 2012), especially since large or replicate samples from the same location will not require significant additional effort
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
