2.2 Molecular laboratory processes
We cleaned up each sample (comprised of 20 pitfalls pooled) following a Flotation-Filtration-Stereoscope protocol (FFS) that allowed us to have a ‘clean’ extraction (see Figure S1b-g). To prevent differences of biomass from causing biases at the DNA extraction and PCR steps (Elbrecht & Leese, 2015), we divided the specimens by size into: small (e.g., sizeDrosophila melanogaster ), medium (e.g., size Apis mellifera ), or large specimens (e.g., adult grasshopper). We then divided each sample in two subsamples, one including the complete bodies from small arthropods, and the second with the thorax (head included) from medium-sized arthropods and two legs from large arthropods. Subsamples (n=84) were processed independently for DNA extraction and library construction, but using the same library barcode identifier (thus, the final number of libraries retrieved was 42, one library per sample). For DNA extractions, we treated each bulk-sample with an electric homogenizer (Qiagen TissueLyser II: two times for 1 min at 30 Hz and 30 s at 30 Hz) using platinum beads with Qiagen DNeasy Blood & Tissue Kit and BSA buffer 1x.
For metabarcoding library construction, we used the double dual tagging method with the XT Illumina Adapter Kit as in Arribas et al. (2016). We amplified a 418-base pair region from the 5’ end of mitochondrial COI gene (within the standard barcode region for metazoa) with the primers B_F 5’ CCIGAYATRGCITTYCCICG 3’ (Shokralla et al., 2015) and Fol-degen-R 5’ TANACYTCNGGRTGNCCRAARAAYCA 3’ (Yu et al., 2012) modified to include Illumina overhang adaptors for subsequent nested PCR. For each subsample (n=84), we performed and pooled together three independent PCR replicates for each arthropod-size subsample. We included a negative control reaction with no DNA template in all experiments. All information regarding PCR reagents and conditions is available in Material S1 (Supporting Materials S1). Each pool of PCR amplicons was cleaned with Agencourt AMPure XP beads (Beckman Coulter) to purify the COI amplicon away from remaining primers and primer dimers. Then, we used primary amplicons as template for a limited-cycle PCR amplification to add dual-indices barcodes (N7 index and S5 index) and the P5 and P7 Illumina sequencing adapters using the Nextera XT Index Kit from Illumina. We sent the 42 resulting metabarcoding libraries and a negative control to the Cornell Institute of Biotechnology, Cornell University, USA, for sequencing on a lane of Illumina MiSeq 2x300 bp (Figure S2a-f).