Detection of porcine reproductive and respiratory syndrome virus type 2 in air emissions from grow-finish barns

INTRODUCTION

Porcine reproductive and respiratory syndrome virus (PRRSV), first isolated in the early 1990s and classified as PRRSV type 1 (PRRSV-1; predominant in Europe) and PRRSV type 2 (PRRSV-2; predominant in the Americas and Asia), is the causative agent of PRRS in pigs, and is endemic in most pork-producing countries.¹ Clinical expression of PRRSV infection in breeding animals includes reproductive failure, with late-term lost litters and fetal death, weak-born piglets and increased stillbirths. Neonatal, nursery and growing pigs may also develop respiratory disease with dyspnoea, fever, inappetence and lethargy, resulting in morbidity and mortality, with clinical severity dependent on the PRRSV variant, age of the pigs and co-infections, among other unidentified factors.^1–3

Air sampling in PRRSV research has allowed the characterisation of the role of aerosolised virus in the epidemiology of the disease.^4–6 However, definitive determination of the relative risk and frequency of aerosol transmission of PRRSV across long distances remains under debate.^{7, 8} For instance, Dee et al. recovered  PRRSV from air samples over 4.5 km from a source farm in four of 306 air samples collected over 50 days.⁹ In two other studies using the same collection device, the frequency of aerosolised PRRSV detection by RT-PCR outside pig barns ranged from 0/251 to 80/217,^{5, 10} while no positive air samples were detected as close as 5 m from the exhaust from a PRRSV-positive pig population.¹⁰

Oral fluids have been used in the routine monitoring and surveillance of pig populations and have become among the most commonly used sampling methods for PRRSV, influenza A virus and porcine epidemic diarrhoea virus.^11–15 Although a non-invasive and convenient sampling method, oral fluids require a high sample size and extended rope exposure time to minimise the probability of a false-negative result at the onset of virus shedding,¹⁴ with the necessity of access to the interior of farms, representing a potential risk to biosecurity. In contrast, the collection of air emissions exhausted from barns has the potential to be an efficient and biosecure sampling method, since entering the area where pigs are housed is not needed, and samples can likely capture a large proportion of the exhausted farm air without requiring highly specialised equipment or supplies. Nevertheless, the use of air emission samples to detect the onset of PRRSV infections in pig barns has not been compared to the traditional methods already in place, such as oral fluid testing. Therefore, this study was conducted to evaluate the feasibility of using air emission samples to detect the onset of PRRSV infection in growing pig populations. The detection of PRRSV in air emissions as a function of pig age and duration of detection from onset was also explored.

MATERIALS AND METHODS

Sample collection

The initial sample collection was conducted when the sample population was approximately 10 weeks of age in barn A and 3 weeks of age in barns B and C. Oral fluid (20 ropes) and air emission (2‒4 cotton cloths) samples were collected every 2 weeks at each of the three sites. To collect air emission samples, investigators used a 71 × 71 cm cotton cloth (Mainstays, Walmart, Arkansas) on a PVC frame (Figure 1). The frame was then placed and secured 1 m from each of four exhaust fans at barns A and C and two exhaust fans at barn B, which corresponded to the room selected for oral fluid collection inside each barn. Oral fluids were collected by conveniently placing one strand of a 90 cm three-strand, 100% cotton rope (Rocky Mount Cord Company, North Carolina) in 20 locations. Ropes were suspended from the horizontal metal rods dividing the pens at the appropriate height to allow the pigs to bite and chew on them. Ropes were placed at the same location for each subsequent collection event. Between 30 minutes and 1 hour after placement, the liquid from each of the 20 ropes was collected by inserting the rope into a one-quart plastic Ziploc bag (SC Johnson, Wisconsin) with a clean, latex-gloved hand, compressing the rope within the bag, decanting it into a 15 mL conical tube and storing it at ‒20°C. Upon completion of the oral fluid sample collection (approximately 2 hours), each of the air emission samples obtained using cloth was retrieved by cutting the plastic fasteners to release the cloth, folding and inserting the cloth into a one-gallon Ziploc bag. Approximately 50 mL of a phosphate-buffered saline solution (Quality Biological, Maryland) was added to the air emission samples, and the bag was sealed, massaged for 10‒20 seconds and repeatedly compressed to release the liquid from the cloth. The liquid was decanted from the bag into a 50 mL conical tube and refrigerated for transfer and storage at ‒20°C.

RESULTS

Oral fluid and air emission samples were PRRSV negative by RT-PCR for the first six collection events at site A (10‒22 weeks of age) and for the first four collections at sites B and C (3‒9 weeks of age). The first PRRSV-positive results at each site were obtained simultaneously in air emissions and oral fluid samples (Figure 2).

The logistic regression results are presented as estimated probabilities and confidence intervals (CIs) at representative values. The estimated probability of detecting PRRSV in air emissions increased from 3% (95% CI: 0.5‒21%) to 32% (95% CI: 17‒53%) when 50% and 100% of oral fluids tested RT-PCR positive, respectively (Figure 3a). The estimated probability of detecting PRRSV in air emissions decreased as the mean Ct values in oral fluids increased, from 26% (95% CI: 14‒43%) with a mean Ct of 30 to 7% (95% CI: 2‒17%) with a mean Ct of 35 (Figure 3b).

DISCUSSION

Air emission samples obtained from the exhausted air of grow-finish barns were detected to be RT-PCR-positive for PRRSV at the same time as oral fluid samples collected inside also tested positive for the first time, following disease development in the studied pig population. The detection of PRRSV in air emission samples was repeatable in all three barns sampled in this investigation. The results of this study highlight the potential use of air emissions as an alternative means to detect the onset of PRRSV infections. While oral fluid sampling is   convenient to monitor and surveil growing pigs, an alternative method that allows for more frequent sample collection in less time, and without requiring personnel to enter the barn, was explored. Air emission sampling, as performed in this investigation, offered a biosecure, low-cost and highly feasible option for the initial detection of PRRSV in a pig population.

It is important to note that although PRRSV was detected in both oral fluid and air emission samples at the same collection event, the ability to detect PRRSV in air emissions by RT-PCR decreased at subsequent collection events. The detection of PRRSV in air emissions appeared to be a function of the Ct value of oral fluids, which reflects the relative concentration of virus in the population. While not part of this study, subsequent sequencing of the PRRSV identified the variant L.1.C at two of the sites. The detection of aerosolised virus in air emissions could be a function of the specific PRRSV type and variant present. The experimental design of this study limited air emission collection to a subset of the rooms at each site and to a time of approximately 2 hours per collection event. Furthermore, determination of the exact onset of PRRSV infection at the pen level was limited by the constraint of collections occurring every 2 weeks, which is a common frequency used for monitoring and surveillance of growing pig populations in North America. Additional work is warranted to explore the feasibility of using air emission samples to detect PRRSV using alternative collection times and locations.

Under the conditions of this study, air emission samples were equally as sensitive as oral fluids for the initial detection of PRRSV in recently infected pigs. These findings may inform the design of PRRSV monitoring and surveillance protocols to accurately determine the onset of infection while maintaining high biosecurity measures. Additionally, given that a large proportion of growing pig populations eventually become PRRSV positive, such sites may contribute significantly to the regional viral burden, in turn increasing the risk of introducing PRRSV into nearby breeding herds and perpetuating the infection cycle. These findings may provide guidance and justification for sound biocontainment strategies if regional PRRSV control programmes are to be successful.

AUTHOR CONTRIBUTIONS

Mark Schwartz conceptualised the study. Mark Schwartz, Montserrat Torremorell and Maria Pieters secured funding. Mark Schwartz and Casondra Snow recruited study sites and collected and processed samples. Mark Schwartz and Aaron Rendahl performed data visualisation and analysis. Mark Schwartz drafted the manuscript with critical review and editing by Montserrat Torremorell, Aaron Rendahl and Maria Pieters. All the authors read, revised and approved the final manuscript.

CONFLICT OF INTEREST STATEMENT

The authors declare no conflict of interest regarding the publication of this paper.

ETHICS STATEMENT

The study was approved by, and the procedures were conducted in accordance with, the University of Minnesota Institutional Animal Care and Use Committee (IACUC protocol number 2206-40112A).

DATA AVAILABILITY STATEMENT

All the data generated or analysed during this study are available from the corresponding author upon request.