Growing interest and affordability of environmental DNA and RNA (eDNA and eRNA) approaches for biodiversity assessments and monitoring of complex ecosystems have led to the emergence of manifold protocols for nucleic acids (NAs) isolation and processing. Although there is no consensus on a standardized workflow, the common practice for water samples is to concentrate NAs via filtration using varying pore size membranes. Using the smallest pore is assumed to be most efficient for NAs capture from a wide range of material (including sub-cellular particles), however a trade-off must occur between detection of a meaningful molecular signal and cost/time effort when processing samples using fine pore membranes.

Comparative studies involving formal efficiency assessments are lacking, which restricts informed decision-making around an optimized sampling approach for applications such as biosurveillance (i.e., detection and monitoring of target taxa - nuisance organisms, endangered and indicator taxa or other species of economic or cultural importance). Here, we present an experimental study using an easily cultured microalgal species (Alexandrium pacificum) to test different filter membranes for capturing NAs in the context of cost/time effort and cell fractions encountered in nature (whole cells, partially lysed, and naked NAs).

The results showed no statistically significant difference between membrane types for capturing target eDNA signal from intact and partially lysed cell treatments. In terms of time effort and volume processed, higher efficiency ratings were obtained with the larger pore size (5 µm) cellulose membranes. Positively charged nylon demonstrated enhanced capture of naked NAs, and especially eRNA signal, across treatments.

Our findings support using coarse pore size filters for adequate capture of target NA signal (from both eDNA and eRNA) with less processing time. The framework presented here can provide a quick and robust feasibility check and comparative assessment of new and existing NA processing technologies, and allows sufficient control over multiple parameters, including physical-chemical water properties, temporal scales, and concentration and type of input material.

A two-tier experimental set-up was applied in this study. In the first experimental round, the performance of different filter membranes for capturing various fractions of eDNA/eRNA was assessed. We selected cellulose acetate (CA) membranes (repeatedly reported performing well in eDNA studies) of three pore sizes (5 μm, 1.2 μm, 0.45 μm) and a positively charged nylon (PCN) 1.2 μm pore size membrane (as a potential alternative to commonly used filter types). A crossed experimental design was applied with three experimental factors manipulated:

• eDNA/eRNA fractions, four levels: 1) intact (whole) cells of microalgal culture, 2) broken (sonicated) cells, 3) naked DNA, and 4) naked RNA.
• Seawater, two levels: 1) pre-filtered (0.35 μm) and 2) non-filtered A. pacificum free seawater (determined via light microscopy and ddPCR assay).
• Filter type, four levels: 1) 5 μm cellulose acetate (CA) membrane, 2) 1.2 μm CA, 3) 0.45 μm CA, 4) 1.2 μm positively charged nylon (PCN) membrane.

The membranes (three replicates for each filter type) with the captured material were immediately transferred to sterile centrifuge tubes and kept frozen at -80 °C until processed for dual DNA/RNA extraction.

The second experiment was focused on establishing input and output parameters for an efficiency model and was run with two concentrations of sonicated A. pacificum cells spiked into target-free (determined via light microscopy) seawater. The spiked water was filtered through the same four types of membrane filters as outlined above (six replicates each), while recording a) the time required to filter 250 mL of water sample and b) the volume filtered in 20 min (through continued filtering past the 250 mL mark in [a]). The filters from step (b) were immediately frozen at -80 °C until simultaneous DNA/RNA extraction.

RNA and DNA were extracted simultaneously from each filter using the Zymo Quick-DNA/RNA™ Miniprep Kit (Zymo Research, California USA). The filters were placed into individual ZR BashingBead

Lysis Tubes and 0.8 mL of lysis buffer from the kit was added to each tube which was adequate volume to cover the filter. These were then homogenized by bead beating (1,500 RPM, 2 min; 1600 MiniG Spex SamplePrep NJ, USA) such that material on all parts of the filter was exposed to the buffer and beads. After centrifugation (3,000 × g, 5 min, 20 °C). DNA and RNA were co-extracted following the manufacturer’s protocol.

Trace DNA in isolated RNA samples was eliminated by two sequential DNase (TURBO DNA-free^TM Kit, Thermo Fisher Scientific, MA, United States) treatments following Langlet et al. (2013). Treated RNA was diluted to 10 ng/µL equimolar concentrations and reverse transcribed into cDNA using the SuperScript^TM III reverse transcriptase (Thermo Fisher Scientific, MA, United States). All extracted products were stored frozen (−20 °C) for DNA and cDNA, and at −80 °C for RNA until analysis.

Alexandrium pacificum-specific copy numbers were quantified using ddPCR for all samples, including all negative extraction controls (DNA and RNA) on a QX200 Droplet Digital PCR SystemTM (Bio-Rad, CA, United States). For the ddPCR assays, A. pacificum specific primers (ACTA-416-F and ACTA-605-R); targeting 204bp from the large ribosomal subunit) and Taqman probe (ACTA-456-P) were used as described in Ruvindy et al. (2018). Each ddPCR reaction included 0.5, 1.0, and 0.05 µM of the forward primer, reverse primer and probe respectively, 1 × BioRad ddPCR Supermix for probes, 1 µL pre-diluted DNA (1:25 for the first experiment and 1:50 for the second experiment) or 1 µL undiluted cDNA, and sterile water in a total reaction volume of 21 µL. The BioRad QX200 droplet generator partitioned each reaction mixture into approximately 20,000 nanodroplets by combining 20 µL of the reaction mixture with 70 µL of BioRad droplet generation oil. After processing, this resulted in a total droplet volume of 40 µl, which was transferred to a PCR plate for amplification using the following cycling protocol: hold at 95 °C for 10 min, 40 cycles of 94 °C for 30 sec and 54 °C for 1 min, and a final enzyme deactivation step at 98 °C for 10 min. Each well of the plate was then individually analyzed on the QX200 instrument to establish the threshold value separating negative and positive droplets and perform absolute quantification of target DNA or cDNA. A positive control of extracted DNA from the A. pacificum CAWD234 culture and negative (sterile water) control was included on each plate. For the first experiment, the results were converted to copies per sample using the formula: number of copies per µL × 22 [the initial volume of the PCR reaction (µL)] × 50 [DNA elution volume] or 25 [RNA elution volume] (µL). Since various sample volumes were processed through filter membranes as part of the second experiment, the ddPCR results were standardized to number of copies per “cell” of input material (A. pacificum), applying the following formula. Total volume (mL) of material that passed through the filter in 20 min multiplied by the “cell” concentration (equivalent to 9.8 cells/ml or 91 cells/ml, see above for details) to calculate total number of “cells” on the filter. This number was divided by 50 µL (DNA elution volume) or 25 µL (RNA elution volume) to calculate number of “cells” per µL. Copies per µL from ddPCR results were divided by “cells” per µL to obtain copies per “cell”.