Concentration of Virus Particles from Environmental Water and Wastewater Samples Using Skimmed Milk Flocculation and Ultrafiltration

1. Comparison of serial ultrafiltration and skimmed milk flocculation to concentrate viruses in wastewater samples

1. Sample preparation
   1. Collect 2 L of 24 h flow-proportional composite raw (influent) wastewater samples. Samples were collected from the three major wastewater treatment plants (WWTPs) in Winnipeg, Canada, during the summer and fall of 2020 (Table 1).
   2. Transport the samples to the laboratory in light-proof bottles in an icebox and process them within 24 h. Collect wastewater physico-chemical and biological characteristics data.
2. Virus concentration assays
   NOTE: A sequential ultrafiltration method and an organic flocculation method, namely skimmed milk flocculation (SMF), were applied to assess the total recovery efficiency of each method using Armored RNA as a test virus²³. Other researchers have also used Armored RNA as a control virus for assessing the recovery of concentration methods for enveloped viruses, such as the family Coronaviridae^22,24,25,26.
   1. For the addition of Armored RNA, set the wastewater test volumes for the ultrafiltration and SMF methods as 140 mL and 500 mL, respectively. Using a pipettor, spike 5 x 10⁴ copies of Armored RNA into six fresh wastewater samples collected on one date and six fresh wastewater samples on another date, for each ultrafiltration method; further, spike six stored (at 4 °C) wastewater samples collected on a third, later date and processed days later for SMF.
      NOTE: Each of the six sample sets collected on different dates was composed of two samples from each of the three WWTPs. In our study, the first set of fresh wastewater samples were collected on July 8^th, 2020, the second on September 30^th, 2020, and the wastewater samples to be stored for SMF on October 14^th, 2020.
   2. Stir for 30 min at 4 °C to inoculate and homogenize the Armored RNA in the samples using a magnetic stirrer.
   3. Remove all particles or cells larger than 0.2 μm and perform ultrafiltration.
      1. Filter the spiked wastewater samples (120 mL from each 140 mL spiked sample) through cheesecloth and low-protein binding, 0.45 µm and 0.2 µm, 47 mm membrane disc filters, respectively, to remove large particles, sediments, eukaryotes, and bacteria²⁷. Process the filtered samples using the ultrafiltration methods described below.
         NOTE: The wastewater samples collected on July 8^th and September 30^th were used for ultrafiltration methods with 3,000 x g and 7,500 x g, respectively.
      2. For sequential ultrafiltration at 3,000 x g (UF-3k x g), concentrate a total of 120 mL of the test sample collected on the first date at 3,000 x g for 30 min by loading 60 mL of the sample twice to approximately 5 mL using a brand A centrifugal device, 30-kDa. Then, concentrate down the 5 mL volume at 3,000 x g for 30 min using a brand B centrifugal device, 30-kDa, to 500-1,200 µL.
      3. For sequential ultrafiltration at 7,500 x g (UF-7.5k x g), change the centrifugation speed for the brand B centrifugal device from 3,000 x g to 7,500 x g to obtain smaller final volumes, as per the manufacturer's recommendations, and keep the centrifugation speed of the brand A centrifugal device the same.
         NOTE: The tests were run with the samples collected on September 30^th.
   4. For SMF, perform the steps outlined below.
   5. Spike the wastewater samples (500 mL each) to directly concentrate using the SMF protocol^16,18, without prefiltration with cheesecloth and low-protein binding membrane disc filters, as described below.
   6. Dissolve 0.5 g of skimmed milk powder in 50 mL of synthetic seawater to obtain a 1% (w/v) skimmed milk solution and carefully adjust the pH of the solution to 3.5 using 1 N HCl.
      NOTE: Acidification makes viruses aggregate and precipitate out of water due to the change in their isoelectric point¹⁶ and improves the solubility of milk powder²⁸.
   7. Add 5 mL of skimmed milk solution to 500 mL of raw wastewater samples to obtain a final concentration of skimmed milk of 0.01% (w/v).
   8. Stir the samples for 8 h at medium speed and allow the formed flocs to settle for another 8 h at room temperature.
   9. Remove the supernatants carefully using serological pipettes without disturbing the settled flocs. Transfer a final volume of 50 mL containing the flocs to centrifuge tubes and centrifuge at 8,000 x g for 30 min at 8 °C.
   10. Carefully scrap the pellets using a sterilized spatula, and resuspend the remaining pellets in the tubes in 250 µL of 0.2 M sodium phosphate buffer (pH 7.5) after the supernatant is removed. Transfer the scraped and resuspended pellets to the same 1.5 mL mini-centrifuge tubes.
3. Perform viral RNA extraction from viral concentrates of wastewater samples and recovery efficiency assays using an RNA extraction kit with a 25:24:1 ratio of phenol:chloroform:isoamyl alcohol and β-mercaptoethanol, according to the manufacturer's instructions, to improve the extraction efficiency. Finally, elute the RNA in 50 µL of elution buffer.
4. Real time-quantitative polymerase chain reaction (RT-qPCR) analysis
   1. Quantify Armored RNA using tenfold dilutions (7.8 x 10⁴ to 7.8 gene copies) of a synthetic single-stranded DNA or gBlock construct. See Table 2 for the primers and probe sets that detect and quantify the Armored RNA.
   2. Design primer and probe sets using the primer design tool²⁹ and target a 95 bp region (gBlock construct) within the Armored RNA genome.
   3. Obtain calibration curves for each RT-qPCR run. Include negative controls in each qPCR run. Run standards, samples, and non-template controls in triplicate.
   4. Prepare each 10 µL RT-qPCR mixture in the following order: 2.5 µL of 4x master mix, 400 nM of each primer, 200 nM probe, and 2.5 µL of template, as well as ultrapure DNAse/RNAse- free distilled water.
   5. Perform thermal cycling reactions at 50 °C for 5 min, followed by 45 cycles of 95 °C for 10 s and 60 °C for 30 s on a real-time PCR system. If the CT was <40, consider the sample Armored RNA positive.
   6. Conduct qPCR and RT-qPCR tests using bovine serum albumin with raw sewage samples to assess the possibility of inhibitors or contaminants such as humic acids²⁴. No significant differences were observed between samples with and without the enzyme.
5. Calculate the concentrations from the standard curve. Calculate the recovery efficiency percentage by dividing the recovered concentration by the spiked concentration.
6. Transform gene copy numbers using the log₁₀ function for analysis. Run a general linear model using a statistical analysis tool on the qPCR data. Apply Tukey's test to detect differences among the methods and treatments. Take a p-value of 0.05 to be a minimum significance level for the test.

2. Filtration and concentration of DNA and RNA viruses for water-based epidemiology

1. For environmental surface water collection, collect 10 L of environmental surface water samples and use 10 L of deionized water samples for background control. Store all the samples in a 4 °C room until processing within 24 h following sample collection.
2. Perform concentration of viruses from environmental samples with SMF using the steps described below.
   1. Perform capsule and vacuum filtration using high-capacity groundwater sampling capsules and membrane disc filters of 0.45 µm, 0.2 µm, and 0.1 µm sizes to minimize noise during metagenomic sequencing, from larger fractions such as micro-eukaryotes and bacteria to smaller ones such as viruses.
   2. Prepare a 1% weight-by-volume pre-flocculated skimmed milk solution in 1.32 L of synthetic seawater with 13.2 g of skimmed milk powder. Adjust the pH of this suspension to 3.5 using 1 N HCl.
3. Adjust the pH of the environmental surface water samples to 3.5 using 1 N HCl.
4. Transfer 100 mL of the pre-flocculated acidified skimmed milk solution to 10 L acidified (pH 3.5) environmental water samples (final skimmed milk concentration of 0.01% [w/v]).
5. Using a magnetic stirrer and a magnetic stir bar, stir the samples for 8 h at room temperature and allow the flocs to sediment by gravity for an additional 8 h. Without disturbing the flocs, carefully remove the supernatant using a vacuum pump.
6. Aliquot and balance the remaining flocs according to the weight in 50 mL centrifuge tubes and spin them down at 8,000 x g for 30 min at 4 °C.
7. Discard the supernatant and dissolve the resulting pellet (which theoretically contains viruses of interest) in 200 µL of 0.2 M sodium phosphate buffer.
8. Treat the dissolved pellet with DNase I and RNase A, following the manufacturer's instructions, to eliminate free DNA and RNA present that may be co-precipitated owing to the kit used for nucleic acid extraction. Inactivate the DNase I and RNase A following the manufacturer's instructions.
9. Extract total nucleic acids from the dissolved pellet using a DNA/RNA extraction kit, as per the manufacturer's instructions.
10. Quantify the total amount of extracted DNA and RNA from effluent samples using a fluorometer. Samples with concentrations >1 ng/µL are considered optimal for DNA quantification and sequencing.
11. Perform qPCR analysis to target enteric viruses of interest and high-throughput sequencing to identify the viral community structure.

3. Direct precipitation and concentration of microbial fractions for water-based epidemiology

1. Collect 2 L of water samples from protected and impacted waterways. If the sample contains a considerable amount of debris, use a cheesecloth to remove them.
   NOTE: Collecting a biological duplicate per sample is highly recommended.
2. Prepare a 1% (w/v) skimmed milk solution by dissolving 4.4 g of skimmed milk powder in 440 mL of synthetic seawater. Adjust the pH of this suspension to 3.5 with 1 N HCl.
3. Adjust the pH of the freshwater samples to 3.5 using 1 N HCl.
4. Add 20 mL of skimmed milk solution to the previously pH-adjusted 2 L freshwater samples (final skimmed milk concentration of 0.01% [w/v]). Stir the samples at room temperature for 8 h.
5. Allow the flocs to sediment by gravity for another 8 h. Carefully remove the supernatants with a peristaltic pump. Transfer the sedimented flocs to a 50 mL centrifuge tube and spin them down at 8,000 x g for 30 min.
6. Discard the supernatant and resuspend the pellets with 200 µL of 0.2 M sodium phosphate buffer. Store the samples at -20 °C or proceed to select ~ 0.5 g of the floc for microbial DNA extraction by using the nucleic acid isolation kit of choice, following the manufacturer's instructions.
7. Determine the DNA nucleic acid concentration and purity with a high sensitivity fluorometer. Samples with concentrations >1 ng/µL should be optimal for DNA quantification and sequencing.