Fig 1.
Schematic diagram of the pathogen filtration for detection of environmental nucleic acids.
A) Different pathogens in water can be filtered with B) a Nalgene™ Single Use Analytical Filter Funnels (Thermo Fisher Scientific, USA) using different filters and pumps. C) Filters are then stored in lysis buffers until processing. The lysis buffer and the filter are then transferred into a NucleoSpin Filters Midi column (Macherey-Nagel, Germany) and centrifuged at room temperature for 2 min at 5,580 x g to separate the filter from the buffer containing the nucleic acids. D) An automated extraction system such as the MagNA Pure 96 (Roche, France) can be used to extract nucleic acids from the lysis buffer, and ultimately E) from the elution can be performed qPCRs for pathogen detection and quantification. Created with BioRender.com.
Fig 2.
Design of experiments 1, 2 and 3.
Experiment 1. Trial A: flow rates were tested with an MCE filter and 1.2 mL of MagNA Pure LC RNA lysis buffer. Trial B: lysis buffers were tested with an MCE filter and a flow rate setting of 200 mL/min (peristaltic pump). Trial C: filters were tested with a flow rate setting of 200 mL/min (peristaltic pump) and 1.2 mL of MagNA Pure LC RNA lysis buffer. Experiment 2. Trial D: four lysis buffers were tested with a “sandwich” filtration method with an MCE filter on the bottom and a GF filter on the top in a Nalgene™ Single Use Analytical Filter Funnels and a flow rate setting of 200 mL/min (peristaltic pump). Experiment 3. Trial E: two flow rates were tested with 3 mL of DNA/RNA Shield™ lysis buffer and the “sandwich” filtration method. Trial F: two lysis buffers were tested with a flow rate setting of 3.8 to 4.0 L/min and the “sandwich” filtration method. Trial G: filters were tested with 3 mL of DNA/RNA Shield™ and 3.8 to 4.0 L/min flow rate setting. Created with BioRender.com.
Table 1.
Primer and probe sequences used for qPCR assays.
Fig 3.
Experiment 1, trials A, B, C. Differences in DNA and RNA yield from qPCR and RT-qPCR analyses, respectively, for the four different fish pathogens ISAV, SGPV, PRV1 and Y. ruckeri in the freshwater bucket experiment. RNA or DNA yields are shown in terms of cycle quantification (Cq) values, where the y-axis is reversed from high to low numbers since RNA or DNA yield increases with decreased Cq-values. Violin plots show Cq median (solid line). Statistical tests were performed with the GraphPad Prism v9.3.1 (* = p ≤ 0.05; ** = p ≤ 0.01; **** = p ≤ 0.0001; n = 3). A) Trial A: test of four different filtration flow rates; 200, 400 and 1000 mL/min (peristaltic pump) and 3.8 to 4.0 L/min (vacuum pump), using MCE filter and 1.2 mL of MagNA Pure LC RNA lysis buffer. B) Trial B: test of two different lysis buffers (MagNA Pure LC RNA lysis buffer and NucliSENS® lysis buffer) in two different volumes (1.2 and 2.4 mL) using an MCE filter and 200 mL/min flow rate setting. C) Trial C: test of four different filters (MCE, Nalgene™, Sterivex-GP and Sterivex-HV) using 200 mL/min flow rate setting and 1.2 mL of MagNA Pure LC RNA lysis buffer.
Fig 4.
Differences in DNA and RNA yield from qPCR and RT-qPCR analyses, respectively, for the three different fish pathogenic agents ISAV, PRV1 and Y. ruckeri in the second freshwater bucket experiment. RNA or DNA yields are shown in terms of cycle quantification (Cq) values, where the y-axis is reversed from high to low numbers since RNA or DNA yield increases with decreased Cq-values. Trial D tested four different lysis buffers in 4 mL volume (NucliSENS®, MagNA Pure 96 Bacteria, MagNA Pure 96 External, and DNA/RNA Shield™) using 200 mL/min flow rate setting (peristaltic pump) and “sandwich” filtration. The two different filters in the “sandwich” filtration (GF and MCE filters) were analysed separately. Violin plots show Cq median (solid line). Statistical significance levels were calculated only between different lysis buffers within the same filter type with the GraphPad Prism v9.3.1 (* = p ≤ 0.05; ** = p ≤ 0.01; **** = p ≤ 0.0001; n = 3).
Fig 5.
Differences in DNA and RNA yield from qPCR and RT-qPCR analyses, respectively, for the four different fish pathogens ISAV, SAV3, PRV1 and Y. ruckeri in the seawater bucket experiment. RNA or DNA yields are shown in terms of cycle quantification (Cq) values, where the y-axis is reversed from high to low numbers since RNA or DNA yield increases with decreased Cq-values. Violin plots show Cq median (solid line). Statistical tests were performed with the GraphPad Prism v9.3.1 (* = p ≤ 0.05; ** = p ≤ 0.01; *** = p ≤ 0.001; **** = p ≤ 0.0001; n = 3). A) Trial E: Test of two different filtration flow rate settings, 200 mL/min (peristaltic pump) and 3.8 to 4.0 L/min (vacuum pump), using the “sandwich” filtration method and 3 mL of DNA/RNA Shield™ lysis buffer. B) Trial F: Test two different lysis buffers (3 mL of NucliSENS® and 3 mL of DNA/RNA Shield™) using the “sandwich” filtration method and the 3.8 to 4.0 L/min flow rate setting (vacuum pump). C) Trial G: Filter control test where the “sandwich” filtration method was compared to the Nalgene™ Single Use Analytical Filter Funnels. Trial G used 3 mL of DNA/RNA Shield™ with a flow rate setting of 3.8 to 4.0 L/min (vacuum pump).
Table 2.
Stock concentrations (1:10) to estimate the corresponding Cq values and copy numbers.
Values are presented as mean ± standard deviation from two technical replicates (n = 2).
Table 3.
Freshwater experiment showing Cq values and estimated copy numbers after the filtration of 1 L of spiked water by the GF and MCE filters.
3.8 to 4.0 L/min flow rate setting and 3 mL of DNA/RNA Shield™ were used. Results are shown as mean ± standard deviation from six replicates (n = 6).
Table 4.
Seawater experiment showing Cq values and estimated copy numbers after the filtration of 1 L of spiked water by the GF and MCE filters.
3.8 to 4.0 L/min flow rate setting and 3 mL of DNA/RNA Shield™ were used. Results are shown as mean ± standard deviation from six replicates (n = 6).
Table 5.
Estimated copy numbers related to concentration in 1L and the recovery rate (GF+MCE) after the filtration of 1 L of spiked water.
Results are shown as mean ± standard deviation from six replicates (n = 6).
Fig 6.
Estimated pathogen copy numbers after the filtration of 1 L of spiked water.
A) ISAV (initial concentration 1 × 107 TCID50/mL, final concentration of 400 TCID50/mL in 12.5 L), B) SAV3 (initial concentration 1 × 108 TCID50/mL, final concentration of 800 TCID50/mL in 12.5 L) and C) Y. ruckeri (OD600 = 2.241 Abs in 1 mL). Filtration with the “sandwich” method was performed for both freshwater and seawater using a flow rate of 3.8 to 4.0 L/min and 3 mL DNA/RNA Shield™ lysis buffer. Violin plots with median (solid line). Statistical tests were performed with the GraphPad Prism v9.3.1 (* = p ≤ 0.05; ** = p ≤ 0.01; *** = p ≤ 0.001; **** = p ≤ 0.0001; n = 6).