DNA metabarcoding of storage ethanol and conventional morphometric identification of stream macroinvertebrates (New Brunswick, Canada)

Published: 21 November 2018| Version 1 | DOI: 10.17632/xj3g36f76p.1
Maitane Erdozain, Terri Porter, Dean Thompson, Mehrdad Hajibabaei, Karen Kidd, Dave Kreutzweiser, Paul Sibley


Stream macroinvertebrates were collected in 15 forest streams in northern New Brunswick, Canada. Within Black Brook (BB), we selected 12 low order streams and their respective catchment areas, which represented a gradient in forest harvesting intensity. The second location was in the unmanaged forest of Mount Carleton Provincial Park (MC) where three headwater streams were sampled. Invertebrates were collected by electroshocking (3 passes of 20 seconds separated by 10-second breaks) within a 100-cm long x 25-cm wide rectangular metal frame; then, five rocks within the rectangle were chosen and carefully inspected to capture attached macroinvertebrates which were added to the sample. This sampling procedure was repeated for 3 different stream riffles with one each at the upstream, middle and downstream sections of the 60-m sampling reach constituting subsamples. In the lab, aquatic insects were identified to genus, with the exception of Chironomidae and Simuliidae – which were identified to family, and classified according to their functional feeding group (FFG) using Merritt et al. (2008). Then, one piece of tissue (usually one leg, but the anterior or posterior end of the body in the case of Dipterans) was pulled from each individual and transferred into a single glass vial filled with 95% ethanol to form a pooled composite for each subsampling site and subsequently submitted for DNA metabarcoding analysis. Attached are the results of these conventional morphometric identifications (stream, replicate, order, family, genus, functional feeding group and number of individuals). DNA was isolated from the preservative ethanol and two fragments (BR5 and F230R) of the cytochrome C oxidase subunit 1 gene were amplified from each sample through a one-stage PCR. Bioinformatic methods involved processing sequence reads obtained from each subsample. Taxonomic assignments were performed using the stand-alone Ribosomal Database Classifier 2.12 with the CO1 Eukaryote v2 training set and they were only used if they met minimum bootstrap support cut-offs: genus bootstrap proportion (BP) >= 0.50, family BP >= 0.30, order BP >= 0.10.The final taxonomy table and FASTA file of exact sequence variants are included within the DNA metabarcoding folder. Using these two datasets, we compared stream macroinvertebrate community metrics based on conventional morphometrics vs. non-destructive DNA metabarcoding from storage ethanol to assess forest management impacts on headwater streams across a gradient of intensively managed forest catchments in eastern Canada. The two approaches demonstrated substantial congruence in the detection of taxa, but DNA metabarcoding from preservative ethanol identified significantly fewer genera (3.3 on average) and families (2.0) than conventional morphometrics. Further details on methods and results can be found in the associated article.



University of New Brunswick Saint John


Forestry, Ecological Indicators, Cytochrome C Oxidase, Ethanol, Stream, mtDNA, Macroinvertebrates