Effect of chemical clarification of oral fluids on the detection of porcine reproductive and respiratory syndrome virus IgG

Abstract

Routine collection and testing of oral fluid (OF) samples facilitates porcine reproductive and respiratory syndrome virus (PRRSV) surveillance in commercial swine herds in a cost-effective, welfare-friendly fashion. However, OFs often contain environmental contaminants that may affect liquid handling and test performance. Traditional processing methods (e.g., filtration or centrifugation) are not compatible with high-throughput testing because of the burden of additional processing costs and time. OF “clarification” using chemical flocculants is an alternative approach not widely explored. Therefore, we evaluated the effect of chitosan-based clarification treatment on a commercial PRRSV OF ELISA. Serum and individual OFs were collected from vaccinated pigs (n = 17) at −7 to 42 d post-vaccination and subdivided into 4 aliquots. Each aliquot was clarified (treatment A, B, C), with the 4th aliquot serving as untreated control. All samples were tested by PRRSV OF ELISA immediately after treatment and then were held at 4°C to be re-tested at 2, 4, 6, and 14 d post-treatment. Quantitative and qualitative treatment effects were evaluated. A Kruskal–Wallis test found no significant difference in ELISA S/P responses among treatments by days post-treatment. No difference was detected in the proportion of positive PRRSV antibody samples among treatments (Cochran Q, p > 0.05). Treatment of swine OFs using chitosan-based formulations did not affect the performance of a commercial PRRSV OF ELISA. Chitosan (or other flocculants) could improve OF characteristics and could be adapted for use in the field or in high-throughput laboratories.

Introduction

Since its recognition in 1991,³¹ porcine reproductive and respiratory syndrome virus (PRRSV; order Nidovirales, family Arteriviridae, genus Porartevirus, species Porcine reproductive and respiratory syndrome virus 1 and 2) has continually challenged pork producers and swine veterinarians. The cost of PRRSV to U.S. producers has been estimated at >$660 million per year.¹⁰ Using prevalence estimates from 11 countries, the costs of PRRSV to European producers has been estimated at €1.5 billion per year (de Paz X, et al. Boehringer Ingelheim survey, 2014, https://www.pig333.com/articles/prrs-cost-for-the-european-swine-industry_10069/). Likewise, in Asia, PRRSV imposes heavy economic consequences as a result of clinical and subclinical infections plus the cost of prevention and control measures.^16,36

The only 2 options for the control of PRRSV are: 1) eliminataion of the virus from the herd, or 2) reduction of clinical signs through enhancement and stabilization of herd immunity via vaccination or intentional exposure.¹⁵ In both cases, implementation of biosecurity protocols sufficient to stop the introduction of extraneous viruses is mandatory.

Regardless of the control plan, routine sampling and testing is necessary to verify that the chosen strategy is functioning, because clinical signs are not reliable or timely indicators of PRRSV infection.^5,6 Depending on the circumstances, the survey objective will either be surveillance (detection of PRRSV) or monitoring (tracking the circulation of the virus in an endemically infected population).²¹

Routine surveillance or monitoring using serum samples is not practical because of the time and cost associated with collecting a statistically sufficient number of samples on a continuous basis. In contrast, oral fluid (OF)-based testing offers the possibility of collecting population infectious disease data easily, quickly, inexpensively, and in a welfare-friendly fashion.^27,33 Both PRRSV antibody- and nucleic acid–based assays have been adapted to OF specimens,^12,25 and studies have shown that field research based on OF specimens equals or exceeds PRRSV detection based on serum.^2,27

Although offering practical advantages for routine PRRSV detection, OF samples commonly contain insoluble particles from the environment (e.g., feed, feces, and inorganic material). These contaminants have not been shown to directly affect test performance, but in the laboratory, these contaminants may affect liquid handling characteristics (e.g., the precision of pipetting). The only options for removing particulates from OFs are prolonged high-speed centrifugation or filtration, but neither is practical in a high-throughput laboratory in terms of time and labor. A third option is clarification (i.e., the removal of particles suspended in a solution, using chemical flocculants). Flocculants function by attracting the opposite-charged surface of the particles, thereby allowing them to flocculate (aggregate) into larger elements that can be removed more easily.

Among the many options, chitosan (deacetylate chitin) is an abundant, biodegradable, biocompatible, and non-toxic flocculant that has been used in a variety of biological applications,⁹ for example, in production of foods³ and beverages,^4,7,17 improved drug delivery systems,²⁰ adjuvantation of vaccines,^14,32 and clarification of cell culture media.²⁹ Therefore, we evaluated the effect of chitosan-based clarification of OFs on the performance of a commercial PRRSV antibody immunoglobulin G (IgG) ELISA (IDEXX PRRS OF Ab test, IDEXX Laboratories, Westbrook, ME).

Materials and methods

Experimental design

Our study was conducted with the approval of the Iowa State University Office for Responsible Research. In brief, OF and serum samples of known PRRSV antibody status were generated by vaccinating (Ingelvac PRRS MLV, Boehringer Ingelheim Vetmedica, Duluth, GA) pigs under experimental conditions and then collecting samples over a period of 50 d. Following each collection, OF samples were subdivided into 4 aliquots, each of which was subjected to 1 of 4 treatments (untreated control [NC], A, B, or C) and then tested for PRRSV antibody (days post-treatment, 0 DPT). Thereafter, the treated OF specimens were stored at 4°C and re-tested at 2, 4, 6, and 14 DPT. At the end of the study, the PRRSV OF ELISA sample-to-positive (S/P) results were analyzed for the effect of treatment, storage time, and storage time-by-treatment interactions.

Animal care

PRRSV-negative pigs (n = 17) were acquired from a commercial swine farm at 14 wk of age (~40–50 kg) and housed in biosafety level 2 (BSL-2) research facilities accredited by the Association for Assessment and Accreditation of Laboratory Animal Care (AAALAC). Negative PRRSV status was verified by ELISA testing of serum samples collected 14 and 7 d prior to the arrival of the animals. In addition, the final set of serum samples was pooled (⩽5 samples per pool) and tested by PRRSV reverse-transcription quantitative PCR (RT-qPCR) to verify the absence of acute PRRSV infection.

Upon arrival (13 d prior to vaccination), pigs were randomly assigned ID numbers by blindly pulling ear tags from a bag and then assigned to individual pens consecutively by ear tag number. Pens (1.5 × 1.8 m) were constructed of solid partitions and gates. All pens had gates on at least 2 sides to allow interaction between animals in neighboring pens. Each pen was equipped with a nipple drinker and a bracket to hold a rope during OF collection. Pigs were fed a commercial swine diet (Heartland CO-OP, Prairie City, IA) twice per day. Animals were closely observed throughout the study by researchers, animal caretakers, and institutional veterinary staff. Five animals were observed through 28 d post-vaccination (DPV) and then removed because of facility space limitations; the remainder (n = 12) were observed through 42 DPV.

Vaccination

All animals were vaccinated intramuscularly at 0 DPV with 2 mL of a PRRSV modified-live vaccine (Ingelvac PRRS MLV, Boehringer Ingelheim, Ingelheim am Rhein, Germany) using a single-use syringe and needle (PrecisionGlide, Becton Dickson, Franklin Lakes, NJ).

Serum and oral fluid sample collection

Pigs were allowed to acclimate for 5 d before initiation of sampling. OF and serum samples were collected from −7 to 42 DPV (Table 1).

Table 1.

Number of samples by day post-vaccination and qualitative porcine reproductive and respiratory syndrome virus (PRRSV) ELISA results.

Specimen	Day post-vaccination*
	−7	−5	−3	0	3	6	7	8	9	10	14	21	28	35	42
Serum (no. samples)	17	—	—	17	4	9	4	9	4	9	17	17	17	12	12
PRRSV ELISA† (no. positive)	0	—	—	0	0	0	1	0	2	5	17	17	17	12	12
Oral fluid (no. samples)	16	16	17	15	15	13	16	16	15	15	16	15	12	12	12
PRRSV ELISA‡ (no. positive)§	0	0	0	0	0	0	0	3	8	14	16	15	12	12	12

Dash (—) = not included.

^*Ingelvac PRRS MLV, Boehringer Ingelheim Vetmedica.

^†PRRS X3 Ab test, IDEXX Laboratories.

^‡PRRS OF Ab test, IDEXX Laboratories.

^§Qualitative results on 0 DPT were identical among all treatments (NC, A–C) with the exception that one 8-DPV oral fluid sample (reported as positive above) was negative for treatments B and C.

Serum samples (n = 187) were collected using a single-use vacutainer system (Corvac integrated serum separator tube, 12.5 mL, Covidien, Minneapolis, MN). No pig was bled on consecutive days; rather, subsets of pigs were rotated through the sampling schedule to obtain serum samples for 3, 6, 7, 8, 9, and 10 DPV (Table 1). Blood samples were centrifuged (1,000 × g, 15 min), then serum was aliquoted into 2-mL tubes (Cryos, Greiner Bio-One, Monroe, NC) and stored at −80°C.

OF samples (n = 221) were collected with 3-strand twisted 100% cotton rope (Web Rigging Supply, Lake Barrington, IL) hung from a metal bracket fixed to one side of each pen. Brackets were placed such that samples could be collected without entering the pen. During the acclimation period, pigs were given access to the rope for two 30-min periods daily. The sampling procedure has been described in detail elsewhere.^8,33 In brief, pigs were allowed to interact and chew the rope for 30 min, then the wet portion of the rope was cut, inserted into a plastic bag (Seal-Top bag, Elkay Plastics, Commerce, CA), and passed through a towel wringer (WC38K, Dyna-Jet Products, Overland Park, KS). The OF that accumulated in the bag was then decanted into a 50-mL polypropylene centrifuge tube (Falcon, Fisher Scientific, Pittsburgh, PA). To maximize the volume collected, OF samples were collected twice daily (0800 h and 1600 h). OF samples collected in the morning were placed on ice, pooled with the afternoon sample, and then the composite sample aliquoted into 5-mL tubes (Fisher Scientific) and stored at −80°C.

PRRSV serum RT-qPCR

Viral RNA was extracted from 140 µL of serum and eluted to 90 µL of elution buffer (QIAamp viral RNA mini kit, Qiagen, Hilden, Germany) following the manufacturer’s instructions. The eluted RNA, primers, and probe were mixed with commercial reagents (EZ-PRRSV MPX 4.0 real-time RT-PCR, Tetracore, Rockville, MD). North American PRRSV, European PRRSV, and internal controls were included in every reaction. RT-qPCR reactions were performed (T-COR 8 thermocycler, Tetracore): 48°C for 15 min, 95°C for 2 min, 95°C for 5 s, and 60°C for 40 s (45 cycles and collection data step). The results were analyzed using an automatic baseline selected by the T-COR 8 software. Quantification cycle (Cq) values ⩽40 were considered positive for PRRSV.

PRRSV serum ELISA

Serum samples were tested for PRRSV antibody using a commercial ELISA (PRRS X3 Ab test, IDEXX Laboratories) performed according to the manufacturer’s instructions. In brief, 5 µL of serum were diluted in 195 µL (1:40) of sample diluent in 96-well polystyrene plates (Nunc, Roskilde, Denmark) and then 100 µL of the mixture was transferred to an ELISA plate followed by 30-min incubation (19–22°C). Plates were then washed 5 times with 300 µL of 1× wash solution, 100 µL of conjugate was added to each well, and the plates were incubated for another 30 min. The washing cycle was repeated, then 100 µL of tetramethylbenzidine–hydrogen peroxide substrate (TMB) was added to each well and the plates incubated for 15 min to visualize the reaction. Thereafter, 100 µL of stop solution was added to each well, and the plate was read (650 nm; EMax Plus microplate reader, SoftMax Pro 7 software, Molecular Devices, Sunnyvale, CA). The antibody responses in serum samples were calculated as S/P ratios using equation 1 below, where OD is the optical density. Serum samples with S/P ratios ⩾0.40 were considered PRRSV serum antibody positive.

$$\mathrm{S}/\mathrm{P\; ratio}=\frac{\left(\mathrm{sample\; OD}-\mathrm{negative\; control\; mean\; OD}\right)}{\left(\begin{array}{l}\mathrm{positive\; control\; mean\; OD}-\\ \mathrm{negative\; control\; mean\; OD}\end{array}\right)}$$ (1)

PRRSV oral fluid ELISA

OF samples were tested for PRRSV antibody using a commercial ELISA (PRRS OF Ab test, IDEXX Laboratories) performed according to the manufacturer’s instructions. In brief, 100 µL of OF was diluted 1:1 with sample diluent in 96-well polystyrene plates (Nunc), and then 100 µL of mixture was transferred to the ELISA plate. After 2-h incubation (19–22°C), plates were washed 5 times using 350 µL of 1× wash solution, then 100 µL of conjugate was added to each well and the plates incubated for 30 min. The wash cycle was repeated, 100 µL of TMB were added to each well, and the plates were incubated for 15 min to visualize the reaction. Thereafter, 100 µL of stop solution was added to stop the reaction and the plates read (450 nm; ELISA plate reader, Biotek Instruments, Winooski, VT; xChek software, IDEXX Laboratories). Antibody responses in OF samples were calculated as S/P ratios using equation 1. OF samples with S/P ratios ⩾0.40 were considered positive.

Oral fluid clarification treatments

Three chemical clarification treatments (A–C) were prepared (Table 2). In brief, treatment A consisted of 100 ppm chitosan oligosaccharide lactate (523682, Sigma-Aldrich, St. Louis, MO), 0.1% polyethylene glycol sorbitan monolaurate (Tween 20, Sigma-Aldrich), 0.5% bovine serum albumin (BSA; Jackson ImmunoResearch, West Grove, PA), and 1 ppm xylene cyanol in phosphate-buffered saline (1×, pH 7.4). Treatment B was identical to treatment A, excluding the Tween 20. Treatment C was identical to treatment A, excluding the chitosan oligosaccharide lactate. Xylene cyanol, a blue dye, was added to all formulations (A–C) to allow for convenient visual differentiation of treated versus untreated samples. BSA (0.5%) was included in all formulations (A–C) to block nonspecific binding.^26,30 Chitosan was used at 100 ppm (0.01%) in formulations A and B, with treatments NC and C used for comparisons. Tween 20 (1.0%) was added to formulations A and C to further reduce nonspecific binding, with treatments NC and B used for comparisons. All components were mixed at room temperature until dissolved.

Table 2.

Treatments evaluated for their effect on swine oral fluid porcine reproductive and respiratory syndrome virus antibody detection.

Component	Treatment formulations
	Untreated control	A	B	C
1× phosphate-buffered saline (mL)	—	97.9	98.9	98.9
Bovine serum albumin* (g)	—	0.5	0.5	0.5
Stock solution 0.1% xylene cyanol† (µL)	—	100.0	100.0	100.0
Stock solution 1.0% chitosan‡ (mL)	—	1.0	1.0	—
Tween 20§ (mL)	—	1.0	—	1.0
Total components cost per sample
U.S. dollars	0.0	0.033	0.032	0.033
Euros	0.0	0.027	0.026	0.027

Dash (—) = not included.

^*001-000-162, Jackson ImmunoResearch, West Grove, PA.

^†X4126, Sigma-Aldrich, St. Louis, MO.

^‡523682, Sigma-Aldrich.

^§P1379, Sigma-Aldrich.

To avoid diluting OF samples prior to ELISA testing, 1 mL of each treatment was pipetted into 5-mL round-bottom polystyrene tubes (Falcon), held at −80°C for 24 h, and then lyophilized (FreeZone, Labconco, Kansas City, MO) for 15 h. After lyophilization, clarification treatment tubes were closed with polyethylene caps (Falcon), vacuum sealed in plastic bags, and stored at room temperature (19–22°C) until use.

Immediate effect of treatment

On 0 DPT, OF samples (n = 221; >4 mL) were thawed at room temperature (19–22°C) and vortexed for 5 s. OF (1 mL) was then added to each of the 3 clarification treatments tubes (A–C) and to an empty 5-mL round-bottom polystyrene tube (NC). Samples were vortexed to resuspend the chemical components (5 s), centrifuged at 1,200 × g for 3 min at 4°C, randomly ordered (www.random.org), and tested for PRRSV antibody (IDEXX Laboratories).

Temporal effect of treatment

After testing on 0 DPT, all samples (n = 884; A–C, NC) were held at 4°C in an environmental chamber (Caron, Marietta, OH) and re-tested on 2, 4, and 6 DPT. A subset (n = 352; i.e., OF samples collected on −7, 6, 7, 8, 9, 10, 14 and 42 DPV) was held and tested on 14 DPT. OF samples were neither vortexed nor centrifuged prior to testing on 2, 4, 6, and 14 DPT.

Data analysis

The effect of clarification treatments on the PRRSV OF ELISA S/P ratios and antibody status (positive or negative) was analyzed (MedCalc 17.6, MedCalc Software, Ostend, Belgium; SAS 9.4, SAS Institute, Cary, NC). Initially, the effect of OF treatment was analyzed separately for OF samples collected prior to the expected appearance of detectable antibody (⩽7 DPV, “early” samples) and OFs collected after the expected appearance of antibody (⩾14 DPV, “late” samples). The assumption of normality for the “early” and “late” datasets was rejected (Shapiro–Wilk test), and transformation of the data did not achieve normality. Therefore, a nonparametric approach (Kruskal–Wallis test) was used to evaluate the effect of treatment on S/P values for each dataset within each day post-treatment (0, 2, 4, 6, and 14). Thereafter, the complete dataset was analyzed for the effect of treatment (NC, A–C) and storage time (0, 2, 4, 6, 14 DPT) on S/P ratios using a repeated-measure multiple-comparison test with Tukey adjustment (Proc GLIMMIX). The Cochran Q test was used to evaluate differences in the proportion of antibody-positive results (S/P ⩾ 0.40) among treatments within days post-treatment and within treatments among days post-treatment.

Results

Pigs were determined to be naive for PRRSV infection at the time of arrival on the basis of negative results using ELISA and RT-qPCR on serum samples collected prior to arrival. In addition, all serum and OF samples collected prior to 7 DPV tested negative by ELISA. No clinical signs or adverse health events were observed over the course of the observation period.

A total of 187 serum samples were collected over the course of the study and tested by PRRSV ELISA (Table 1). Serum antibody ontogeny (ELISA S/P response) is shown in Figure 1.

Figure 1.

Mean porcine reproductive and respiratory syndrome virus (PRRSV) ELISA sample-to-positive (S/P) results from serum samples (PRRS X3 Ab test, IDEXX Laboratories, Westbrook, ME) and oral fluid samples (treatments A–C, untreated control; PRRS OF Ab test, IDEXX Laboratories) by day post-vaccination (DPV). Oral fluid samples were tested immediately following treatment (see Table 1).

A total of 221 OF samples of volume sufficient to make 4 aliquots (each of 1 mL) were collected over the course of the experiment. The lyophilized treatments (A–C) were readily rehydrated by adding 1 mL of OF and vortexing briefly. Untreated samples were readily differentiated from treated samples by the light blue color conferred by the xylene cyanol present in formulations A–C.

PRRSV OF ELISA test results for all OF samples on 0 DPT are given in Figure 1 and Table 1. Among 0, 2, 4, 6, and 14 DPT, no difference was detected among treatments in the proportion of antibody samples positive by PRRSV (Cochran Q test, p > 0.05; Table 1).

The effect of OF treatment on ELISA S/P responses was analyzed separately for “early” (⩽7 DPV) and “late” (⩾14 DPV) samples for 0, 2, 4, 6, and 14 DPT (Table 3). At 0, 2, 4, and 6 DPT, a statistically significant difference in the S/P response for “early” samples was detected between treatments A or C versus NC or B (Kruskal–Wallis test, p < 0.0001). For “late” samples, no differences were detected between treatments (NC, A–C).

Table 3.

Quantitative effect of treatment on porcine reproductive and respiratory syndrome virus (PRRSV) oral fluid ELISA^* sample-to-positive (S/P) mean ratios by day post-treatment (DPT).

DPT	n	Untreated control	Treatment A	Treatment B	Treatment C
Early sample†
0	108	0.06‡ (0.05, 0.08)	0.02§ (0.01, 0.04)	0.05‡ (0.04, 0.07)	0.03§ (0.01, 0.05)
2	108	0.05‡ (0.03, 0.06)	0.00§ (−0.01, 0.01)	0.04‡ (0.02, 0.05)	0.00§ (−0.01, 0.01)
4	108	0.06‡ (0.05, 0.08)	0.01§ (−0.01, 0.02)	0.09‡ (−0.01, 0.18)	0.01§ (−0.01, 0.02)
6	108	0.04‡ (0.03, 0.06)	0.00§ (−0.02, 0.01)	0.02‡ (0.01, 0.04)	0.00§ (−0.02, 0.02)
14	32	0.04 (0.02, 0.06)	0.04 (0.01, 0.08)	0.02 (−0.01, 0.04)	0.03 (0.00, 0.06)
Late sample†
0	67	7.21 (6.57, 7.86)	7.23 (6.58, 7.87)	7.16 (6.52, 7.79)	7.22 (6.57, 7.87)
2	67	7.58 (6.93, 8.24)	7.59 (6.94, 8.24)	7.70 (7.03, 8.37)	7.61 (6.93, 8.29)
4	67	7.93 (7.17, 8.70)	7.91 (7.15, 8.67)	8.07 (7.32, 8.82)	8.08 (7.31, 8.85)
6	67	7.54 (6.81, 8.27)	7.66 (6.93, 8.39)	7.74 (7.02, 8.46)	7.64 (6.91, 8.38)
14	20	6.74 (5.30, 8.18)	7.16 (5.81, 8.51)	7.55 (6.21, 8.89)	7.07 (5.76, 8.38)

Numbers in parentheses are 95% confidence intervals. Early samples = samples collected ⩽7 d post-vaccination; late samples = samples collected ⩾14 d post-vaccination.

^*PRRS OF Ab test, IDEXX Laboratories.

^†Ingelvac PRRS MLV, Boehringer Ingelheim Vetmedica.

^{‡, §}Treatment effect by 0, 2, 4, 6, 14 DPT was analyzed separately for oral fluid samples collected prior to and after the expected appearance of detectable antibody (Kruskal–Wallis test). Within DPT, differences (p < 0.05) between treatments are indicated by symbols (paired symbols represent similarity; unpaired symbols represent significant difference between treatments). Subsequent analysis of the complete dataset detected no significant interaction between storage time (DPT) and treatment (repeated-measure multiple-comparison test with Tukey adjustment).

Analysis of the complete dataset using a repeated-measures multiple-comparison test with Tukey adjustment found no significant difference in ELISA response between treatments (NC, A–C) at each storage time (0, 2, 4, 6, 14 DPT). Likewise, no interaction was detected between treatment and storage time on ELISA S/P ratios. Analysis within treatments showed differences (p < 0.05) in S/P responses among days post-treatment, but the direction of change was inconsistent and not compatible with antibody degradation or inactivation. The datasets used and/or analyzed during our study are available from the corresponding author on reasonable request.

Discussion

The use of swine OF specimens began with the isolation of PRRSV from buccal swabs.³⁴ Thereafter, it was shown that ELISA-detectable PRRSV antibodies appeared concurrently in serum and OF matrices, and the use of a pen-based OF sample was explored under experimental conditions.²⁴ The appearance of ELISA-detectable antibody in oral fluids collected in boars 7 d after administration of a modified-live vaccine has been reported previously.²² In the field, a strong temporal association in the ontogeny of OF and serum antibody has been found,¹³ and the concurrent appearance of serum and OF PRRSV antibody isotypes (IgM, IgG, IgA) under both experimental and field conditions has been demonstrated.^11,12 Overall, the antibody results of our study were consistent with earlier reports. In particular, the design of the study (i.e., 17 individually penned pigs sampled over a period of 50 d) allowed for a precise estimate of the earliest appearance of ELISA-detectable antibody because a bleeding rotation schedule provided paired serum samples and OF specimens every day during the expected period of seroconversion (3–11 DPV).

Chitosan is a cationic polymer derived by high alkaline partial deacetylation from chitin, the biopolysaccharide structural component in the exoskeletons of arthropods.²⁸ Chitosan’s chemical properties vary depending on the degree of deacetylation, polymer length, and product purity but it is generically considered nontoxic, biocompatible, biodegradable, and amenable to a wide variety of clarification applications.^1,28 For example, chitosan was used at a concentration of 3,000 ppm to treat textile wastewater,¹⁸ 300 ppm to clarify fruit juice,⁴ 200 ppm to clarify cell culture media for antibody recovery,²⁹ 5 ppm to clarify beer,⁷ and as low as 0.6 ppm when combined with alum for water treatment.³⁵ The objective of our study was to evaluate the effect of chitosan-based clarification of OFs on the performance of a commercial PRRSV ELISA: would chitosan reduce ELISA-detectable antibody levels through coagulation–flocculation and/or precipitation, or interfere with the antibody-binding functions required for detection by ELISA?

Our data suggest that chitosan did not affect the concentration of ELISA-detectable antibody in OF samples and did not interfere with the antibody-binding functions. Statistically significant differences in mean S/P ratios were observed among treatments in “early” samples (samples collected ⩽7 DPV). However, these differences were not clinically significant (Table 1; i.e., represented minor difference in S/P values), and no such effect was observed among treatments in “late” samples (samples collected ⩾14 DPV). Further, analysis of the effect of treatment over time (samples held at 4°C and tested at 2, 4, 6, and 14 DPT) found no significant change in ELISA S/P values among treatments or over time. Although there are no prior reports of testing after storage at 4°C after 14 d, the results were consistent with previous reports examining antibody stability in OFs.^19,23

Providing a cleaner OF sample may lead to wider acceptance by users and holds the potential to move OF-based surveillance to the next level by providing a sample well-suited to testing in high-throughput laboratories. Within the constraints of our study, formulation B would be the preferred method for clarification of swine OFs for ELISA because it required the fewest components. The cost of formula B components sufficient for treating one OF sample (1 mL) was USD $0.032 (€ 0.026). Future efforts should include the evaluation of other flocculants and coagulants, and/or combinations of flocculants–coagulants (e.g., chitosan in combination with others). Likewise, it will be necessary to evaluate the compatibility of OF clarification to other pathogens, antibody assays, and nucleic acid detection technologies.