Laboratory assessment of recovery of porcine circovirus 2 and porcine reproductive and respiratory syndrome virus using 2 commercial hollow-fiber ultrafilters

Abstract

Groundwater near swine farms is an uninvestigated reservoir for porcine reproductive and respiratory syndrome virus (PRRSV) and porcine circoviruses (PCVs). Enteric microorganisms are often collected from groundwater via dead-end ultrafiltration, but recovery of PRRSV and PCV with this method has not been assessed. We recovered PRRSV2 and PCV2 by dead-end ultrafiltration followed by polyethylene glycol (PEG) precipitation, nucleic acid extraction, and reverse-transcription quantitative real-time PCR. We also compared 2 commercial hemodialysis ultrafilters (Asahi Kasei Rexeed-25A, Nipro Elisio-25H) and compared PRRSV2 recovery in these filters to other waterborne microorganisms. On average, 8 ± 1% of PRRSV2 was recovered by dead-end ultrafiltration and PEG precipitation, compared to 25 ± 6% for adenovirus 41. Full-process recovery of bacteria in the same filters was 5–15%; Cryptosporidium parvum recovery was 42 ± 12%. PCV2 was detected in 4 of 12 replicate filters, but low stock concentrations precluded quantitative recovery estimates. Elisio-25H ultrafilters performed similarly to Rexeed-25A filters for all organisms tested and is an effective replacement for the Rexeed-25A, which is no longer available in the United States. Our recovery of PRRSV2 and PCV2 by dead-end ultrafiltration in the laboratory suggests that PRRSV2 detection limits are as low as 3–50 genomic copies/L in sample volumes of 100–1,500 L. Based on quantitative microbial risk assessment, these concentrations are relevant to PRRSV2 infection rates in the U.S. swine herd.

Keywords

adenovirus groundwater PCV2 PRRSV swine

Disease caused by the porcine reproductive and respiratory syndrome virus (PRRSV2; Arteriviridae, Betaarterivirus americense) is challenging to manage and causes significant expense to the pork industry with estimated costs of ~ $664 million per year in the United States.²⁵ Clinical signs include increased abortions and stillborn piglets in breeding females, slowed growth in growing pigs, pneumonia, and increased susceptibility to secondary infections. PRRSV can be spread via direct or indirect transmission, such as nose-to-nose contact, aerosols, and fomites.⁶¹ Similarly, porcine circovirus 2 (PCV2; Circoviridae, Circovirus porcine2) is ubiquitous in swine populations and causes disease of variable severity, with clinical signs including chronic wasting, pneumonia, skin lesions, and reproductive losses. PCV2 is transmitted by oronasal contact with body fluids and affected tissues.⁵¹ The route of introduction for any single farm outbreak is often difficult to determine despite advanced biosecurity.

Groundwater is a well-established pathway for virus transport and as a source of viral outbreaks.^18,29 Although human viruses have received the most attention in the scientific literature,²⁷ animal viruses are subject to the same fate and transport processes and have been detected in groundwater. For example, bovine adenovirus, bovine enterovirus, bovine polyomavirus, and bovine rotavirus A have been detected in groundwater contaminated by manure in the midwestern United States.^4,53,55 Non-enveloped viruses, such as PCV2, have been investigated and detected most frequently in groundwater, but nucleic acids from enveloped viruses, such as PRRSV2, have also been detected (e.g., from highly pathogenic avian influenza A virus during the 2015 midwestern U.S. outbreak³). Porcine viruses, including PCV2, have been detected in groundwater and water used for swine production in Brazil,^13,15 and PCV2 has been documented to leach through soil to tile drains following field application of swine slurry.³¹ Reports of porcine microbial source tracking (MST) markers in groundwater also indicate a potential pathway for dissemination of manure-borne pathogens,^{8,11,32,38,42,53,60} which could signal the possibility of PRRSV2 or PCV2 transmission among animals via groundwater.

Groundwater is a plausible transmission route for PRRSV and PCV2 to animals on swine farms. PRRSV and PCV2 can be shed in swine manure,^{35,43,47,48,58,59} and manure is spread to surrounding fields in the spring and fall. Spring and fall are periods of groundwater recharge,^41,49 a factor known to influence microbial contamination of groundwater,⁴ and these seasons also correspond to the timing of PRRSV outbreak cycles.¹⁰ Once in groundwater, cool, moist conditions and protection from sunlight allow viruses to survive for days to months,³⁹ and PRRSV has been documented to survive in well water 9 d post-inoculation,⁴⁴ suggesting that groundwater can be a reservoir. Many swine farms are supplied by groundwater from private wells, and water from these wells is not treated to remove pathogens prior to consumption by animals.³³ The potential for groundwater transmission of porcine viruses remains uninvestigated to date and presents a need for sampling and laboratory methods.

Dead-end ultrafiltration is a validated technique for capturing viruses and other enteric microorganisms at low concentrations from large volumes of groundwater (hundreds to >1,000 L).^9,14 Primary concentration with ultrafilters is based on size exclusion; water and small molecules pass through the filter whereas microbes and larger particles are retained. Microbes are then backflushed off the ultrafilter for further sample processing or analysis.⁹ Ultrafilters have been used to recover waterborne viruses, bacteria, and protozoa,^30,55 with primary recovery efficiencies of 40–100% depending on the target microbe and experimental conditions.^{16,23,24,45,46,52} However, their use for recovering PRRSV and PCV2 has not been reported, to our knowledge, and recovery efficiencies for these organisms are unknown, which hampers their application in the field because their methodologic limits are not well defined.

Our objectives were to 1) investigate recovery of PRRSV2 and PCV2 by dead-end ultrafiltration in the laboratory to assess the potential for field application, and 2) compare recovery of waterborne pathogens by 2 ultrafilter types, Elisio-25H (Nipro) and Rexeed-25A (Asahi Kasei Medical). The Rexeed-25A has been recommended for water sampling because its 20-nm pore size is small enough to capture common waterborne pathogens, and it has performed well compared to other filter types.^14,24,52 However, given the limited availability of the Rexeed-25A in the United States, the Elisio-25H is a potential replacement as it has a similar molecular weight cutoff of 30 kDa.³⁶ We conducted recovery experiments for PRRSV2, PCV2, and other waterborne microorganisms, including human adenovirus, Campylobacter jejuni, Salmonella enterica subsp. enterica serovar Typhimurium, and Cryptosporidium parvum, thereby enabling comparison of our results to previous work on these organisms. Finally, we contextualized recovery of PRRSV2 by estimating sample-level probabilities of detection and annual probabilities of infection for a range of possible groundwater concentrations, with the latter based on quantitative microbial risk assessment (QMRA).¹⁹

Materials and methods

Organisms and stocks

Organisms used in recovery experiments included PRRSV2, PCV2, adenovirus 41, Salmonella Typhimurium, C. jejuni, and C. parvum. Adenovirus 41, C. jejuni, and Salmonella were obtained from laboratory stocks stored at −80°C. C. parvum oocysts were obtained from a commercial supplier (Waterborne). PRRSV2 was obtained from a commercial modified live vaccine (MLV; Ingelvac PRRS MLV, Boehringer Ingelheim), and PCV2 was obtained from banked porcine serum samples collected from commercial farms in Iowa because this material was readily available to us from a separate research project.

Stock concentrations of adenovirus 41, Salmonella, C. parvum, and C. jejuni were quantified by extracting nucleic acids, eluting into AE buffer, and following reverse-transcription and/or quantitative real-time PCR (RT-qPCR, qPCR) procedures as described below. Genomic copies were determined for PRRSV2 by submitting 0.5 mL of a reconstituted vial of Ingelvac PRRSV MLV to the Iowa State University–Veterinary Diagnostic Laboratory (ISU-VDL; Ames, IA, USA) to be tested by qPCR. A second vial of Ingelvac PRRSV MLV within the same lot number was utilized as the stock solution. Genomic copies were calculated for PCV2 by splitting serum samples and submitting one set of serum samples to the ISU-VDL for qPCR. Stocks were diluted such that the total number spiked in recovery control experiments for each organism was 10⁵–10⁷ gene copies, except for the PCV2 stock derived from serum samples, which was ~10² gene copies (Table 1).

Table 1.

Gene copies enumerated from spike by organism.

Organism type	Organism	Enumeration of spike (gene copies)
Bacteria	Campylobacter jejuni	5.7 × 10⁶
	Salmonella enterica subsp. enterica serovar Typhimurium	1.3 × 10⁶
Protozoa	Cryptosporidium parvum	9.0 × 10⁶
Virus	Adenovirus 41	6.4 × 10⁵
	Porcine circovirus 2	2.2 × 10²
	Porcine reproductive and respiratory syndrome virus	2.6 × 10⁷

Experimental design and recovery calculations

Test filters and control samples were used in experiments to isolate primary recovery (i.e., filtration and backflushing) from downstream laboratory steps.² For test filters, target organisms were spiked into water samples, concentrated by dead-end ultrafiltration, and backflushed (Fig. 1). For control samples, target organisms were spiked directly into the backflush solution to bypass dead-end ultrafiltration; test filters and controls were then processed identically after this point (Fig. 1). Experiments were conducted for Rexeed-25A and Elisio-25H filters (n = 6 each) with 1 set of 6 control samples used for both, yielding 18 samples.

Figure 1.

Parallel quantification of microbial targets to evaluate primary and full-process recovery. Before quantification by qPCR, microbial stocks were spiked into water and processed beginning with primary concentration by dead-end ultrafiltration (C_test), spiked into the preanalytical matrix and processed beginning with secondary concentration (C_control), or extracted directly (C_spike). Comparison of C_test to C_control isolates recovery of the primary concentration step (ultrafiltration and backflushing); comparison of C_test to C_spike includes efficiency of the secondary concentration step.

Primary recovery efficiency was calculated as R_primary = C_test/C_control, in which C_test is the mean target organism concentration measured by qPCR in test filters and C_control is the mean target organism concentration measured by qPCR in control samples (Fig. 1). This procedure represents parallel quantification for positive controls as described in the Environmental Microbiology Minimum Information (EMMI) Guidelines,² in which the value R_primary characterizes the primary concentration step (filtration and backflushing). In addition, efficiency of the full analytical process was calculated as R_full = C_test/C_spike, in which C_spike is the target organism concentration quantified in spikes (Suppl. Section 2). Recovery efficiency was not determined for PCV2 because the spike concentration was too low to generate quantitative results.

Spiking procedure

Organisms were spiked into 10 L of dechlorinated tap water in sterile carboys and concentrated using an Elisio-25H hemodialyzer filter or a Rexeed-25A hemodialyzer filter with a peristaltic pump (4 L/min; Fig. 2). After sampling, the filter was capped, labeled, and refrigerated at 4°C until completion of all replicates. Filters were backflushed with a solution containing 500 mL of reverse osmosis water, 0.50% Tween 80 (v/v), 0.01% sodium polyphosphate (w/v), and 0.001% Antifoam C (v/v) into a 1-L polypropylene bottle. Beef extract (1%; Gibco Bacto) was added to the eluate and briefly mixed before storing at −20°C overnight.

Figure 2.

Filters with spiked test water and peristaltic pump.

After thawing, 0.2 M NaCl and 8% polyethylene glycol (PEG) were added, and samples were mixed on a magnetic stir plate at 4°C for ≥2 h, then allowed to rest at 4°C for 2 h. Samples were split into two 500-mL conical tubes (Corning) and centrifuged at 4,700 × g for 45 min at 4°C. Supernatant from each tube was poured off; remaining pellets were resuspended in 1× TE buffer, combined with their other half, aliquoted into 2-mL tubes, and frozen at −80°C. Final concentrated sample volumes (FCSVs) were 2.0–5.0 mL.

Control samples (n = 6) were prepared by spiking 500 mL of backflush solution with microbial stocks; 1% beef extract was added prior to storing at −20°C overnight. These samples were concentrated by PEG precipitation and processed as described above, producing FCSVs of 1.9–4.2 mL.

DNA extraction, reverse transcription, and qPCR

We extracted 280 µL of FCSV and eluted into 140 µL of AE buffer. Extraction was performed using AVL buffer, carrier RNA, and the QIAamp DNA mini kit (Qiagen).

Quantitative PCR was performed (LightCycler 480 II; Roche) using hydrolysis probes (Suppl. Table 1).^1,26,28,37 Reactions for all targets were performed in duplicate. For all targets except PRRSV2, reaction volumes were 20 µL, consisting of 10 µL of LightCycler 480 probes mastermix, 4 µL of primers and probe, and 6 µL of template DNA (or cDNA for RNA organisms). Reactions for these targets began at 95°C for 5 min followed by 45 cycles of 10 s at 95°C and 1 min at 60°C.

For PRRSV, a reagent kit (VetMAX PRRSV 3.0; Applied Biosystems, ThermoFisher) was used to perform one-step RT-qPCR. The 20-µL reaction volumes consisted of 10 µL of Master Mix PRRSV 3.0, 2 µL of Sequences PRRSV 3.0, and 8 µL of sample template. The PRRSV 3.0 kit’s “fast mode” was followed according to instructions, except the number of cycles was increased from 40 to 45 for consistency with other assays and to improve sensitivity for analysis of groundwater samples. Reactions started with 5 min at 50°C followed by 10 min at 95°C and then finished with 45 cycles of 95°C for 3 s and 60°C for 30 s.

Standard curves were prepared in AE buffer containing 0.02% (w/v %) bovine serum albumin. The Roche LightCycler 480 software calculated Cq values using the second derivative maximum method and regressed them against the decimal logarithm of target concentration. Standard curve efficiencies were 91.2–99.6% with R² of 0.957–0.992 (Suppl. Table 2). Standard curves could not be prepared for PRRSV2 because the positive control sequence in the VetMAX PRRSV 3.0 kit is proprietary, so PRRSV2 was quantified by the LightCycler software based on the known concentration of positive control in kits (~125 copies/µL) and an assumed efficiency of 100%.

Hepatitis G Ultramer RNA Oligo (Integrated DNA Technologies) and salmon testes DNA (MilliporeSigma) were used to check for reverse transcription and PCR inhibition for each sample, respectively (Suppl. Table 1).^21,50 No samples exhibited inhibition. Bovine herpesvirus and bovine respiratory syncytial virus were used as DNA extraction positive controls and reverse transcription positive controls, respectively (Suppl. Table 1).^5,57

Quantitative PCR data have been deposited online as a USGS data release.¹²

Data analysis

Data from recovery experiments were used to estimate probabilities of detection and perform QMRA for PRRSV concentrations in groundwater. Results of these analyses are useful for planning field experiments because they 1) characterize the method’s probability of detection for a range of plausible PRRSV concentrations in groundwater, and 2) contextualize these concentrations in terms relevant to swine health.

Probabilities of detection were estimated as a function of sample volume using the formula P_detection = 1 – exp(–λ_recovered),⁵⁴ in which P_detection is the probability of detection, λ_recovered is the expected number of PRRSV genomic copies recovered in a single sample, and sampled PRRSV numbers are assumed to be Poisson-distributed.¹⁹ The quantity λ_recovered is estimated as the product of C_groundwater × V_effective × R_full, in which C_groundwater is the PRRSV concentration in genomic copies/L, V_effective is the effective sample volume analyzed at the qPCR step, and R_full is the full-process recovery efficiency for PRRSV (Suppl. Sections 1, 2). Effective sample volume incorporates all dilution, concentration, and sub-sampling steps during laboratory processing (Fig. 1).

QMRA was conducted to determine the relevance of PRRSV concentrations detectable by qPCR to swine health. The hazard was defined as a single-day oral exposure to PRRSV via untreated groundwater for an individual nursery pig (3–9-wk-old). Exposure was assessed over a range of hypothetical PRRSV concentrations of 10^–3–10⁷ genomic copies/L. Daily water consumption was assumed to be 1.1 L/d,⁷ and the dose-response model was a beta-Poisson with α = 3.01 × 10^–1 and β = 1.44 × 10⁴.^6,19 This model extrapolates data for oral ingestion of PRRSV by nursery pigs,²² in which dose was quantified as 50% tissue culture infectious dose (TCID₅₀), to the low doses possible in groundwater exposures.¹⁹ Because not all virus detections by qPCR are likely to represent infectious particles, genomic copies were converted to TCID₅₀ using a dose harmonization factor of 2 genomic copies/TCID₅₀. This value was calculated from a study quantifying PRRSV in semen samples using both methods of quantification (Suppl. Section 2)²⁰ and is a point estimate derived as the geometric mean of multiple values of 0.35–43 genomic copies/TCID₅₀ (Suppl. Table 3).

Risk was characterized as daily probability of infection (P_inf) and annual probability of infection (P_inf,annual). Daily probability of infection was calculated using the equation

P_inf = 1 – (1 + D/β)^–α in which α and β are beta-Poisson dose-response parameters¹⁹ and D is average daily dose in units of TCID₅₀ (PRRSV concentration in TCID₅₀/L × daily water consumption in L). Annual probability of infection was calculated assuming 365 d of exposure per year and the cumulative risk equation: P_inf,annual = 1 – (1 – P_inf)³⁶⁵.¹⁹

Results

Both the Rexeed-25A and Elisio-25H hemodialyzers recovered PRRSV2 and PCV2 from water. Primary recovery efficiency of PRRSV2 by the Rexeed and Elisio filters was 64% and 78%, respectively (Table 2). PCV2 was recovered in 4 of 12 filters, 2 from each filter type, indicating that recovery was possible but that concentrations were near qPCR detection limits.

Table 2.

Primary recovery efficiency by organism and filter type.

Organism type	Organism	Elisio-25H, % (n = 6)	Rexeed-25A, % (n = 6)
Bacteria	Campylobacter jejuni	84	76
	Salmonella enterica subsp. enterica serovar Typhimurium	39	33
Protozoa	Cryptosporidium parvum	137	98
Virus	Adenovirus 41	46	40
	Porcine reproductive and respiratory syndrome virus	78	64

Porcine circovirus 2 was recovered in 4 of 12 filters (2 Rexeed-25A and 2 Elisio-25H), but recovery efficiency could not be quantified because detected concentrations were close to the qPCR assay’s detection limit.

Primary recovery efficiencies were similar between the Rexeed and Elisio filters for all organisms considered (Table 2). Primary recovery for adenovirus was 46% and 40% for Rexeed and Elisio, respectively. Primary recoveries for bacteria (C. jejuni, Salmonella) were 33–84%; the highest values occurred for C. parvum (98% and 137% for Rexeed and Elisio, respectively).

Full-process recovery efficiencies, which include losses due to downstream processing steps, were lower than primary recovery efficiencies (Table 3). For example, 7–8% of PRRSV2 was recovered, compared to 22–25% for adenovirus.

Table 3.

Full-process recovery efficiency by organism and filter type.

		Full-process recovery
		Elisio-25H, % (n = 6)		Rexeed-25A, % (n = 6)
Organism type	Organism	x̄ ± SE	CV	x̄ ± SE	CV
Bacteria	Campylobacter jejuni	5 ± 1	0.43	5 ± 1	0.62
	Salmonella enterica subsp. enterica serovar Typhimurium	15 ± 4	0.73	13 ± 3	0.55
Protozoa	Cryptosporidium parvum	42 ± 12	0.71	30 ± 7	0.58
Virus	Adenovirus 41	25 ± 6	0.58	22 ± 4	0.51
	Porcine reproductive and respiratory syndrome virus	8 ± 1	0.34	7 ± 2	0.54

Probabilities of detection extrapolated from our laboratory results increased as a function of simulated PRRSV concentrations in groundwater and sample volume (Fig. 3). Concentrations detectable by qPCR with a high level of confidence (i.e., 90% probability of detection) were 3–50 genomic copies/L for simulated sample volumes of 100–1,500 L (Fig. 3). Lower concentrations could be detected with less confidence. For example, the probability of detecting 1 genomic copy/L was 5–50% for simulated sample volumes of 100–1,500 L (Fig. 3).

Figure 3.

Estimated probability of detection of porcine reproductive and respiratory syndrome virus (PRRSV) in groundwater by qPCR. Probabilities are illustrated as a function of concentration and sample volume (100; 500; 1,000; 1,500 L). Assumes 8% full-process recovery efficiency (Table 3) and typical laboratory processing parameters for PRRSV (Suppl. Section 1). Probabilities of detection are for a single sample and improve for repeated sampling of a given groundwater source.

Hypothetical probabilities of infection predicted by QMRA for nursery pigs increased as a function of simulated PRRSV concentration in groundwater (Fig. 4). Predicted annual probabilities of infection via groundwater were less than the U.S. average for all exposure routes (~30%)⁵⁶ for concentrations <~100 genomic copies/L (Fig. 4). Concentrations with high confidence of detection by qPCR (3–50 genomic copies/L, depending on sample volume) had daily probabilities of infection <1% and annual probabilities of infection of 1–20%.

Figure 4.

Hypothetical probabilities of porcine reproductive and respiratory syndrome virus infection based on quantitative microbial risk assessment. The reference line for the U.S. average (~30%)⁵⁶ reflects infection via all exposure routes. Groundwater concentrations of 10–100 genomic copies/L produce infection risk consistent with the U.S. average; these concentrations can be detected by qPCR with high probability (≥90%) using sample volumes ≥500 L (Fig. 3).

Discussion

We found that primary recovery efficiency for PRRSV2 was greater than for adenovirus 41 and within the range of recovery efficiencies for other organisms considered. Recovery of PCV2 was not quantitative because the spike was obtained from serum samples on commercial swine farms, which were found to be too low for reliable quantification in recovery experiments. We selected serum samples with the highest number of starting genomic copies for our study, which allowed us to demonstrate that detection of PCV2 by dead-end ultrafiltration is possible, but quantifying recovery requires additional experiments.

Primary recovery efficiency of all organisms was similar between the 2 filters tested, typically differing by ≤10% between filters. The similarity of the filters for viruses, bacteria, and a protozoan indicates that the Elisio-25H is a suitable substitute for the Rexeed-25A, which is used extensively and is well validated.^24,46,52 Recovery of diverse microorganisms by ultrafiltration has been reported for various waters,^{17,23,30,45,46,52} but recovery efficiencies determined for the groundwater-supplied tap water used for our experiments may not represent other water types given the potential effects of the matrix.¹⁶ Assessment of RT-qPCR inhibition that results from matrix effects is a crucial element of environmental microbiology quality control.²

Comparison to previous studies is complicated by the use of different organisms (pathogens vs. pathogen surrogates), enumeration techniques (e.g., culture methods vs. qPCR), and experimental factors that are investigated to optimize recovery, such as spike concentration or filtration rate.^45,46,52 We recovered C. jejuni and Salmonella Typhimurium with a primary efficiency of 33–84%, which is within the range of 2 to >100% reported for bacteria.^17,24,45,52 Likewise, we recovered C. parvum with ≥98% primary efficiency and viruses with 40–78% primary efficiency; previous studies have reported recovery of C. parvum and viruses of 47–100% and 16–100%, respectively.^{17,23,24,45,46,52} Thus, our recovery procedure produced results comparable to those from previous studies. In particular, recovery of PRRSV2 was within the range of recovery reported for other viruses.

Comparison of primary concentration efficiency to full-process recovery indicates that losses occurred during the processes downstream of primary concentration. Such losses are not unexpected because the laboratory procedures are intended to capture a broad range of organisms (bacteria, viruses, protozoa) without being tailored to any one organism. Given relatively high primary recovery efficiency for PRRSV2 using dead-end ultrafiltration, subsequent processing steps (secondary concentration and/or nucleic acid extraction) could be modified or optimized to improve recovery of the virus. Alternatively, apparent losses following primary concentration may be due to differences in quantification of the spike.

The qPCR detection limits for PRRSV were sufficiently low using our filtration methods to capture PRRSV concentrations relevant to disease transmission on commercial swine farms. Approximately 30% of pigs produced annually in the United States may be infected with PRRSV,⁵⁶ with infection rates varying from farm to farm. The infections included in this number reflect all possible exposure routes (e.g., via water, food, pig-to-pig contact), such that the annual probability of infection due solely to groundwater must be ≤30%. Our QMRA predicts infection rates in this range for groundwater exposures at ≤100 genomic copies/L. Concentrations of ≤100 genomic copies/L are typical of groundwater exposures,⁵⁵ and based on our study, a 500-L ultrafilter sample could detect 10–100 genomic copies/L of PRRSV2 with ≥90% probability of detection even when accounting for imperfect recovery. This volume is well within reason for a single dead-end ultrafilter sample.⁴ Furthermore, repeated sampling of a farm’s groundwater source throughout the year would increase the probability of detection at a given concentration.

Our QMRA was limited in scope to nursery pigs based on the relevance of this age group with respect to PRRSV infections. Although PRRSV is detrimental to other stages of swine production in which a QMRA may be warranted (especially the breeding herd), breeding herds make up a smaller percentage of the total U.S. swine herd,⁴⁰ have stricter biosecurity practices, and have a lower overall prevalence of PRRSV than do nursery or finisher pigs. QMRA was also limited by assumed inputs such as water consumption, the dose harmonization factor, and dose-response model. Water consumption varies based on age, stage of production and lactation status, facility and drinker type, temperature, and other factors,⁷ and few studies can distinguish true water consumption while accounting for the amount of water wasted.³⁴ Nonetheless, exposure to water via daily drinking is well-quantified compared to other potential means of exposure (e.g., if a pig inhales contaminated water or is bitten, cut, or scraped and contacts contaminated water). Thus, QMRA for exposure via drinking water is feasible based on current data, whereas other exposure routes cannot yet be considered. Likewise, we used the only available data for dose harmonization to convert genomic copies to TCID₅₀.²⁰ Future work could refine this value by comparing qPCR and culture-based abundance of wild-type PRRSV in environmental and/or manure samples. QMRA dose-response models account for the combined effects of potential inactivation in the stomach and/or inability to reach lymphoid tissue and cause infection.¹⁹

Active organized elimination efforts are established or are in developing stages for endemic swine disease and for preparation of the occurrence of foreign animal diseases entering the United States. The persistence of endemic diseases considering current stringent biosecurity suggests a pathway for unexplored disease transfer, and water could represent one of these pathways. Research efforts have been focused on testing animals for pathogens and pathogen transmission, feed, air, fomite, and insect transmission, leaving water as an unexplored mechanism. Previous literature demonstrates that groundwater can serve as a potential reservoir for livestock pathogens and exposes a need for a method for determining their frequency in the field. Our research findings demonstrate that veterinarians and researchers can utilize dead-end ultrafiltration to concentrate groundwater pathogens with either Rexeed-25A or Elisio-25H filters with comparable recovery efficiencies for many types of pathogens, including PRRSV2 at clinically relevant concentrations. Future research studies could concentrate on applying dead-end ultrafiltration in field settings through sampling wells supplying water to livestock farms, determining the frequency of recovery of target pathogens, and evaluating if the recovery is at concentrations that could infect livestock herds.

Supplemental Material

sj-pdf-1-vdi-10.1177_10406387251322506 – Supplemental material for Laboratory assessment of recovery of porcine circovirus 2 and porcine reproductive and respiratory syndrome virus using 2 commercial hollow-fiber ultrafilters

Supplemental material, sj-pdf-1-vdi-10.1177_10406387251322506 for Laboratory assessment of recovery of porcine circovirus 2 and porcine reproductive and respiratory syndrome virus using 2 commercial hollow-fiber ultrafilters by Aaron D. Firnstahl, Gabrielle E. Doughan, Sarah A. Opelt, Rachel M. Cook, Joseph A. Heffron, Karen M. Krueger, Mark A. Borchardt, Locke A. Karriker, Joel P. Stokdyk and Tucker R. Burch in Journal of Veterinary Diagnostic Investigation

Footnotes

Acknowledgements

Use of trade, firm, or product names is for descriptive purposes only and does not imply endorsement by the U.S. Government. USDA is an equal opportunity provider and employer.

Declaration of conflicting interests

The authors declare no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

Funding was provided by USDA–Agricultural Research Service Project 5090-12630-006-000D.

ORCID iDs

Aaron D. Firnstahl

Gabrielle E. Doughan

Joel P. Stokdyk

Tucker R. Burch

Supplemental material

Supplemental material for this article is available online. Underlying data are available as a USGS data release located at

References

Best

, et al Applicability of a rapid duplex real-time PCR assay for speciation of Campylobacter jejuni and Campylobacter coli directly from culture plates. FEMS Microbiol Lett 2003;229:237–241.

Borchardt

, et al The environmental microbiology minimum information (EMMI) guidelines: qPCR and dPCR quality and reporting for environmental microbiology. Environ Sci Technol 2021;55:10210–10223.

Borchardt

, et al Avian influenza virus RNA in groundwater wells supplying poultry farms affected by the 2015 influenza outbreak. Environ Sci Technol Lett 2017;4:268–272.

Borchardt

, et al Sources and risk factors for nitrate and microbial contamination of private household wells in the fractured dolomite aquifer of northeastern Wisconsin. Environ Health Perspect 2021;129:067004.

Boxus

, et al Real time RT-PCR for the detection and quantitation of bovine respiratory syncytial virus. J Virol Methods 2005;125:125–130.

Brookes

, et al Import risk assessment incorporating a dose–response model: introduction of highly pathogenic porcine reproductive and respiratory syndrome into Australia via illegally imported raw pork. Prev Vet Med 2014;113:565–579.

Brumm

Patterns of drinking water use in pork production facilities. Nebraska Swine Rep 2006;221:10–13. https://digitalcommons.unl.edu/coopext_swine/221/

Carrey

, et al Combining multi-isotopic and molecular source tracking methods to identify nitrate pollution sources in surface and groundwater. Water Res 2021;188:116537.

Cashdollar

Wymer

Methods for primary concentration of viruses from water samples: a review and meta-analysis of recent studies. J Appl Microbiol 2013;115:1–11.

10.

Corzo

. Developing the Swine Health Monitoring Project (SHMP) to build capacity and enable the Swine Health Information Center. Swine Health Information Center, 2022 Feb 11. SHIC 20-172. https://swinehealth.org/results/

11.

Demoliner

, et al Microbial source tracking in small farms: use of different methods for adenovirus detection. Water Air Soil Pollut 2021;232:63.

12.

Firnstahl

, et al Microbial concentrations derived from laboratory experiments with hemodialyzers performed in Marshfield, Wisconsin. U.S. Geological Survey data release. 2024. doi:10.5066/P1BZPMBO

13.

Fongaro

, et al Human and animal enteric virus in groundwater from deep wells, and recreational and network water. Environ Sci Pollut Res Int 2015;22:20060–20066.

14.

Gallardo

, et al Protocol for collection of water samples for detection of pathogens and biothreat agents. EPA/600/R-21/280. U.S. Environmental Protection Agency, 2022.

15.

Garcia

LAT

, et al Surveillance of human and swine adenovirus, human norovirus and swine circovirus in water samples in Santa Catarina, Brazil. J Water Health 2012;10:445–452.

16.

Gibson

, et al Measuring and mitigating inhibition during quantitative real time PCR analysis of viral nucleic acid extracts from large-volume environmental water samples. Water Res 2012;46:4281–4291.

17.

Gibson

Schwab

KJ.

Tangential-flow ultrafiltration with integrated inhibition detection for recovery of surrogates and human pathogens from large-volume source water and finished drinking water. Appl Environ Microbiol 2011;77:385–391.

18.

Goyal

, et al Viruses in groundwater beneath sewage irrigated cropland. Water Res 1984;18:299–302.

19.

Haas

, et al Quantitative Microbial Risk Assessment. 2nd ed. Wiley, 2014:427.

20.

Han

, et al Effect of the modified live porcine reproductive and respiratory syndrome virus (PRRSV) vaccine on European and North American PRRSV shedding in semen from infected boars. Clin Vaccine Immunol 2011;18:1600–1607.

21.

Haugland

, et al Comparison of Enterococcus measurements in freshwater at two recreational beaches by quantitative polymerase chain reaction and membrane filter culture analysis. Water Res 2005;39:559–568.

22.

Hermann

, et al Probability of porcine reproductive and respiratory syndrome (PRRS) virus infection as a function of exposure route and dose. Vet Microbiol 2005;110:7–16.

23.

Hill

, et al Multistate evaluation of an ultrafiltration-based procedure for simultaneous recovery of enteric microbes in 100-liter tap water samples. Appl Environ Microbiol 2007;73:4218–4225.

24.

Holowecky

, et al Evaluation of ultrafiltration cartridges for a water sampling apparatus. J Appl Microbiol 2009;106:738–747.

25.

Holtkamp

, et al Assessment of the economic impact of porcine reproductive and respiratory syndrome virus on United States pork producers. J Swine Health Prod 2013;21:72–84.

26.

Hoorfar

, et al Automated nuclease PCR assay for identification of Salmonella enterica. J Clin Microbiol 2000;38:3429–3435.

27.

Hynds

, et al Contamination of groundwater systems in the US and Canada by enteric pathogens, 1990–2013: a review and pooled-analysis. PLoS One 2014;9:e93301.

28.

Jothikumar

, et al Quantitative real-time PCR assays for detection of human adenoviruses and identification of serotypes 40 and 41. Appl Environ Microbiol 2005;71:3131–3136.

29.

Keswick

Gerba

CP.

Viruses in groundwater. Environ Sci Technol 1980;14:1290–1297.

30.

Knappett

PSK

, et al Efficacy of hollow-fiber ultrafiltration for microbial sampling in groundwater. Ground Water 2011;49:53–65.

31.

Krog

, et al Leaching of viruses and other microorganisms naturally occurring in pig slurry to tile drains on a well-structured loamy field in Denmark. Hydrogeol J 2017;25:1045–1062.

32.

Lamendella

, et al Evaluation of swine-specific PCR assays used for fecal source tracking and analysis of molecular diversity of swine-specific “bacteroidales” populations. Appl Environ Microbiol 2009;75:5787–5796.

33.

Little

, et al Water distribution systems in pig farm buildings: critical elements of design and management. Animals (Basel) 2021;11:3268.

34.

Little

, et al Water consumption and wastage behaviour in pigs: implications for antimicrobial administration and stewardship. Animal 2022;16:100586.

35.

Liu

, et al Epidemiology investigation of PRRSV discharged by faecal and genetic variation of ORF5. Transbound Emerg Dis 2021;68:2334–2344.

36.

Maduell

, et al Efficacy and safety of the medium cut-off Elisio HX dialyzer. Blood Purif 2023;52:68–74.

37.

Mary

, et al Multicentric evaluation of a new real-time PCR assay for quantification of Cryptosporidium spp. and identification of Cryptosporidium parvum and Cryptosporidium hominis. J Clin Microbiol 2013;51:2556–2563.

38.

Miura

, et al Recovery of nucleic acids of enteric viruses and host-specific Bacteroidales from groundwater by using an adsorption-direct extraction method. Appl Environ Microbiol 2021;87:e00710–00721.

39.

Murphy

. Persistence of pathogens in sewage and other water types. In: Rose

, et al eds. Global Water Pathogen Project (Yates M, ed. Part 4: Management of Risk from Excreta and Wastewater). Michigan State University, UNESCO, 2017 May 10. https://www.waterpathogens.org/book/persistence-in-sewage

40.

National Agricultural Statistics Service, Agricultural Statistics Board, U.S. Department of Agriculture. Quarterly hogs and pigs. USDA, 2024 Mar 28. https://downloads.usda.library.cornell.edu/usda-esmis/files/rj430453j/q811n8305/s46573384/hgpg0324.pdf

41.

Neher

, et al Seasonal variations in export of antibiotic resistance genes and bacteria in runoff from an agricultural watershed in Iowa. Sci Total Environ 2020;738:140224.

42.

Ohad

, et al Microbial source tracking in adjacent karst springs. Appl Environ Microbiol 2015;81:5037–5047.

43.

Patterson

, et al Shedding and infection dynamics of porcine circovirus type 2 (PCV2) after experimental infection. Vet Microbiol 2011;149:91–98.

44.

Pirtle

Beran

GW.

Stability of porcine reproductive and respiratory syndrome virus in the presence of fomites commonly found on farms. J Am Vet Med Assoc 1996;208:390–392.

45.

Polaczyk

, et al Ultrafiltration-based techniques for rapid and simultaneous concentration of multiple microbe classes from 100-L tap water samples. J Microbiol Methods 2008;73:92–99.

46.

Rhodes

, et al Evaluation of hollow-fiber ultrafiltration primary concentration of pathogens and secondary concentration of viruses from water. J Virol Methods 2011;176:38–45.

47.

Rose

, et al Epidemiology and transmission of porcine circovirus type 2 (PCV2). Virus Res 2012;164:78–89.

48.

Rossow

, et al Experimental porcine reproductive and respiratory syndrome virus infection in one-, four-, and 10-week-old pigs. J Vet Diagn Invest 1994;6:3–12.

49.

Schilling

, et al Using baseflow to quantify diffuse groundwater recharge and drought at a regional scale. J Hydrol 2021;602:126765.

50.

Schlueter

, et al Reverse transcription-PCR detection of hepatitis G virus. J Clin Microbiol 1996;34:2660–2664.

51.

Segalés

, et al Circoviruses. In: Zimmerman

, et al, eds. Diseases of Swine. 11th ed. Wiley, 2019:473–487.

52.

Smith

Hill

VR.

Dead-end hollow-fiber ultrafiltration for recovery of diverse microbes from water. Appl Environ Microbiol 2009;75:5284–5289.

53.

Stokdyk

, et al Assessing private well contamination in Grant, Iowa, and Lafayette counties: the southwest Wisconsin groundwater and geology study. Open-file report 2023-02. University of Wisconsin–Madison Division of Extension, 2023 May 12. doi:10.54915/tbjg5785

54.

Stokdyk

, et al Determining the 95% limit of detection for waterborne pathogen analyses from primary concentration to qPCR. Water Res 2016;96:105–113.

55.

Stokdyk

, et al Viral, bacterial, and protozoan pathogens and fecal markers in wells supplying groundwater to public water systems in Minnesota, USA. Water Res 2020;178:115814.

56.

Trevisan

, et al Swine Disease Reporting System (SDRS) detection dashboards. Field Epidemiology, Iowa State University, 2024. https://fieldepi.org/SDRS/

57.

Wang

, et al Validation of a real-time PCR assay for the detection of bovine herpesvirus 1 in bovine semen. J Virol Methods 2007;144:103–108.

58.

Wills

, et al Porcine reproductive and respiratory syndrome virus: routes of excretion. Vet Microbiol 1997;57:69–81.

59.

Yoon

, et al Persistent and contact infection in nursery pigs experimentally infected with porcine reproductive and respiratory syndrome (PRRS) virus. Swine Health Prod 1993;1:5–8.

60.

Zhang

, et al Tracing fecal pollution sources in karst groundwater by Bacteroidales genetic biomarkers, bacterial indicators, and environmental variables. Sci Total Environ 2014;490:1082–1090.

61.

Zimmerman

, et al Porcine reproductive and respiratory syndrome viruses (porcine arteriviruses). In: Zimmerman

, et al, eds. Diseases of Swine. 11th ed. Wiley, 2019:685–708.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.96 MB