Molecular analyses
For all sample treatments, eDNA extraction was performed in a dedicated room using the NucleoSpin® Soil Mini Kit (Macherey-Nagel, Germany), after a preliminary step where 15 g of soil were mixed with 20 ml of phosphate buffer for 15 minutes as described in (Taberlet, Coissac, Pompanon, Brochmann, & Willerslev, 2012); and with a final elution in 150 μl. We also included one extraction negative control per treatment.
Environmental DNA of bacteria, fungi and eukaryotes was amplified using primers designed for markers Bact02 (Taberlet et al., 2018), Fung02 (Epp et al., 2012; Taberlet et al., 2018) and Euka02 (Guardiola et al., 2015), respectively. Bact02 and Fung02 relate to fragments of about 220-250 bp, while Euka02 generally relates to fragments <150 bp. The three markers are well suited for metabarcoding analyses, as all have a very low number of mismatches in the priming region across target organisms, and the relatively short length of amplified fragments allows their use with potentially degraded DNA (Taberlet et al., 2018). To allow bioinformatic discrimination of PCR replicates after sequencing, eight-nucleotide long tags were added on the 5’ end of both forward and reverse primers, so that each PCR replicate was represented by a unique combination of forward and reverse tags. Tags had at least five nucleotide differences among them (Coissac, 2012). Samples were randomized on a 96-well plate, along with the five extraction controls, eight bioinformatic blanks, six PCR negative controls and two PCR positive controls. PCR positive controls were included to check for potential cross-contaminations and to monitor amplification and sequencing performances. The positive control was a 1:10 dilution of the ZymoBIOMICS™ Microbial Community DNA Standard II (Zymo Research, USA) constituted of genomic DNA of eight bacterial and two fungal strains (i.e., Pseudomonas aeruginosa, Escherichia coli, Salmonella enterica, Lactobacillus fermentum, Enterococcus faecalis, Staphylococcus aureus, Listeria monocytogenes, Bacillus subtilis, Saccharomyces cerevisiae, Cryptococcus neoformans ) at known concentrations.
In order to avoid over-amplification of template DNA and to limit chimera formation, we determined the optimal number of amplification cycles and DNA extract dilution using qPCR. The qPCR assay was conducted on 48 randomly selected samples, using 1 μl of 1:1000 diluted SYBR® Green I nucleic acid gel stain (Invitrogen™, USA), with a real-time PCR thermal cycler set to standard mode. qPCR was performed for both 1:10 diluted and undiluted template eDNA.
For Bact02 and Fung02, PCR reactions were performed on 1:10 diluted template DNA, using 32 and 44 cycles respectively. For Euka02, we performed 34 cycles on undiluted DNA. All PCR reactions consisted of 10 μl of AmpliTaq Gold 360 Master Mix 2X (Applied Biosystems™, Foster City, CA, USA), 2 μl of primers mix at initial concentration of 5 μM of each primer, 0.16 μl of Bovine Serum Albumin (corresponding to 3.2 μg; Roche Diagnostic, Basel, Switzerland) and 2 μl of DNA extract, for a final volume of 20 μl. The PCR profiles had an initial step of 10 min at 95°C, followed by several cycles of a 30 s denaturation at 94°C, a 30 s annealing at 53°C (Bact02), 56°C (Fung02) or 45°C (Euka02), and a 90 s elongation for Bact02 and Fung02, or a 60 s elongation for Euka02 at 72°C, followed by a final elongation at 72°C for 7 minutes. The amplification was performed in 384-well plates, with four replicates for each sample. After amplification, PCR products of the same marker were pooled together in equal volumes and a 5-μl aliquot of the pooled amplicons was visualized by high-resolution capillary electrophoresis (QIAxcel Advanced System, QIAGEN, GERMANY) to verify the expected fragments length and to monitor primer dimers. Pooled amplicons were purified using the MinElute PCR Purification Kit (QIAGEN, GERMANY) following the manufacturer’s protocol. Six subsamples of the pool of amplicons were purified separately for each marker, and then combined again before being sent for library preparation and sequencing to Fasteris (SA, Geneva, Switzerland). One library was prepared per marker using the MetaFast protocol (Taberlet et al., 2018) and then sequenced using the MiSeq (Fung02 and Bact02) or HiSeq 2500 (Euka02) platforms (Illumina, San Diego, CA, USA) with a paired-end approach (2 × 250 bp for Fung02 and Bact02, and 2 × 150 bp for Euka02).