Quantitative polymerase chain reaction for Porcine circovirus-2 in swine feces in a Porcine circovirus disease-affected commercial herd and a nonaffected commercial herd

Abstract

This study examined if pigs in a Porcine circovirus disease (PCVD)-affected herd (n = 100) had shed more Porcine circovirus-2 (PCV-2) in their feces than pigs in a PCVD-nonaffected herd (n = 101), and if differences in shedding among production stages within and between the herds existed. The PCV-2 shedding was quantified by real-time polymerase chain reaction. The highest median PCV-2 shedding was found in the nursery of the PCVD-affected herd and in the grower of the PCVD-nonaffected herd. The PCV-2 shedding was significantly higher in earlier stages (newly weaned, nursery, and pregrower) in the PCVD-affected herd (Wilcoxon rank sum; P < 0.001) compared with the PCVD-nonaffected herd. Porcine circovirus-2 DNA was not detected in a significant proportion of lactating sows (parity ≥ 3) in the PCVD-nonaffected herd (Fisher’s exact test; P = 0.001). The results of this study suggest there may be an association between the presence of PCV-2 in the feces of lactating sows and increased PCV-2 shedding in younger pigs.

Introduction

First described in the mid-1990s (1), postweaning multisystemic wasting syndrome (PMWS) was found to clinically affect pigs 7 to 15 wk of age with wasting, enlarged lymph nodes, dyspnea, diarrhea, pallor, and jaundice. In 1998, Porcine circovirus-2 (PCV-2) was isolated from a pig in Saskatchewan showing clinical signs of PMWS (2). In the last decade, other clinical syndromes have been recognized in association with PCV-2 infection; these include the porcine dermatitis and nephropathy syndrome (PDNS) (3,4), congenital tremors (5), abortion (6), and reproductive disorders (7,8). For simplicity, diseases associated with PCV-2 infection are now collectively termed Porcine circovirus diseases (PCVD) (9,10), or Porcine circovirus associated disease (PCVAD) (11).

The general consensus among scientists is that PCV-2 is the etiological agent of PCVD; however, other infectious agents (12), stressors, or cofactors are required concurrently to exacerbate PCVD in pigs infected with PCV-2. Coinfection with porcine reproductive and respiratory syndrome virus (PRRSV) (13,14), mycoplasma (15), or porcine parvovirus (PPV) (16,17); immune stimulation (18,19); or production and weaning practices (20) can contribute to, or exacerbate, PCVD in PCV-2-infected pigs. The PCV-2 nucleic acid, or antigen, or both can be found in, and PMWS is characterized by, gross or microscopic lesions in multiple organs and tissues of infected pigs (1,2,21). The PCV-2 is shed in various secretions from both naturally and experimentally infected pigs, and in healthy versus PCVD-affected pigs via nasal, tonsillar, tracheobronchial, oropharyngeal, fecal, and urinary routes (22,23). Additionally, PCV-2 DNA is detected in semen from experimentally (24) and naturally (25) infected boars. An association exists between the severity of clinical disease, or PCVD, in affected pigs and the PCV-2 viral load that is found in tissues from affected animals (26,27); however, PCV-2 quantified from fecal samples has not been reported in PCVD-affected versus -nonaffected pigs, or across production stages. The objectives of this study were as follows: 1) to determine if pigs in a PCVD-affected herd shed more PCV-2 in their feces than did pigs in a PCVD-nonaffected herd; and 2) to determine if there are differences in the amount of PCV-2 shed in feces among production stages within and between a PCVD-affected and a PCVD-nonaffected commercial swine herd.

Materials and methods

Animals and collection of samples

Approximately 100 pooled fecal samples were collected from each of 2 commercial swine facilities in Saskatchewan practising high-biosecurity procedures. Both facilities were farrow-to-finish operations housing Landrace and Large White sows, or their respective crosses. Porcine circovirus-2 vaccines were not used in either facility prior to collecting samples for this study. Both barns were infected endemically with PCV-2; however, the non-affected herd (300 sows) had no clinical evidence or diagnosis of PCVD and a long-standing postweaning mortality of < 2%, while the PCVD-affected herd (1200 sows) was experiencing a PCVD-related mortality of 14.5% during the period of sampling. Based on prior testing of lymphoid tissue, the PCV-2 genotypes present in each barn were determined, as per a previously described method (28). In limited testing of lymphoid tissue, concurrent PCV-2a and PCV-2b were detected in the PCVD-nonaffected herd, whereas PCV-2b alone was detected in the PCVD-affected herd. The PMWS was recognized as the primary clinical manifestation of PCVD in the affected herd; the diagnosis was based on observed jaundice, unthriftiness, and dyspnea, and confirmed by PCV-2 immunohistochemical staining and the presence of typical histological lesions in the tissues of selected pigs (Prairie Diagnostic Services, Saskatoon, Saskatchewan). Based on clinical records and serologic testing, there was no evidence of PRRSV, Mycoplasma hyopneumoniae, swine influenza virus H1N1, Actinobacillus pleuropneumoniae, or Haemophilus parasuis in either herd; however, the PCVD-affected herd had experienced sporadic disease associated with A. suis and Streptococcus suis.

Each barn was sampled in a cross-sectional manner with approximately 50 fecal samples being collected from each of the feeding and breeding herds. From the feeding herd, approximately 10 samples (range = 10 to 11) were collected from each of 5 evenly spaced production weeks between weaning and market age [newly weaned (NW), nursery (NU), pregrower (PG), grower (GR), and finisher (FIN)] (Table 1). From the breeding herd, sampling was stratified by age with approximately 10 samples (range = 7 to 14) collected from each of virgin (VG) and bred (BG) gilts, young (S ≤ 2) and old (S ≥ 3) sows (based on a parity of ≤ 2 or ≥ 3), and breeding boars (B) (Table 1). The barns were sampled on different days: PCVD-nonaffected herd (n = 101 samples; 18 Sept 06) and PCVD-affected herd (n = 100 samples; 17 May 06). However, all samples within a barn were collected on the same day. The bred gilts, sows, and boars were housed in crates or stalls; therefore, the fecal samples collected from these animals represented the individual animal. By contrast, all other animals were housed in pens containing a variable number of pigs (Table 2). Fecal samples obtained from pens comprised multiple individual fecal samples and were considered pools. By using sterile gloves for each collection, samples representing separate defecation events from multiple pigs in a single pen were pooled into a sterile bag and transported in an insulated box to the laboratory for processing on the same day. Upon arrival, pooled samples were homogenized, and 1 g of feces from each bag was stored in individual sterile 15 mL conical tubes and frozen at −70°C until DNA extraction.

Table 1.

The representative ages and the number of pens tested for each production stage in 2 commercial swine facilities (PCVD-nonaffected herd; PCVD-affected herd) tested for PCV-2 in pooled fecal samples by quantitative real-time PCR

	Number of pens
Production stage	Age of pigs	PCVD-nonaffected	PCVD-affected
Newly weaned (NW)	3–4 weeks	10	11
Nursery (NU)	7–8 weeks	10	10
Pregrower (PG)	11–12 weeks	10	10
Grower (GR)	15–16 weeks	10	10
Finisher (FIN)	22–24 weeks	10	11
Virgin gilt (VG)	6 months	10	10
Bred gilt (BG)	8–10 months	10	10
Lactating sow parity ≤ 2 (S ≤ 2)	11–15 months	10	10
Lactating sow parity ≥ 3 (S ≥ 3)	15+ months	14	11
Boar (B)	12–24 months	7	7
Total number of pooled fecal samples by barn		n = 101	n = 100

(S ≤ 2) — young sow; (S ≥ 3) — old sow; PCV-2 — Porcine circovirus-2; PCR — polymerase chain reaction; PCVD — Porcine circovirus disease

Table 2.

The number of pigs housed per pen for each production stage in a PCVD-nonaffected herd and a PCVD-affected herd. Pooled fecal samples from each production stage in each barn were tested for PCV-2 by quantitative real-time PCR

Barn	NW	NU	PG	GR	FIN	VG	BG	S ≤ 2	S ≥ 3	B
PCVD-nonaffected	12	12	5–6	5–6	12	20	1	1	1	1
PCVD-affected	20–25	20–25	20–25	20–25	20–25	4–5	1	1	1	1

PCVD — Porcine circovirus disease; PCV-2 — Porcine circovirus-2; PCR — polymerase chain reaction; NW — newly weaned; NU — nursery; PG — pregrower; GR — grower; FIN — finisher; VG — virgin gilt; BG — bred gilt; S ≤ 2 — lactating sow parity ≤ 2; S ≥ 3 — lactating sow parity S ≥ 3; B — boar

DNA extraction and quantitative polymerase chain reaction (PCR)

Fecal samples were extracted by using a commercial kit (QIAamp DNA stool mini kit; Qiagen, Mississauga, Ontario), following the manufacturer’s instructions for DNA isolation from larger amounts of stool. Samples consisting of 1 g of homogenized feces were removed from −70°C and 10 mL of stool lysis buffer (ASL Buffer, QIAamp DNA stool mini kit; Qiagen) was added immediately, prior to thawing. Extraction modifications included pipetting 1.5 mL, instead of 2 mL, of lysate from 10 mL of homogenized sample in stool lysis buffer. Extracted DNA was stored at −70°C until quantitative real-time PCR was performed.

A previously described (29), quantitative SYBR green (DNA binding dye) real-time PCR assay and plasmid standard curve (Genbank accession EU126886) for PCV-2 was used to determine the viral copy number in pooled fecal samples; it was reported as copy number of PCV-2 in 1 g of feces. The assay was developed to detect the current genotypes of PCV-2 with equal efficiency; it does not differentiate between genotypes within PCV-2. Each extracted fecal sample was tested in duplicate, and each quantitative PCR 96-well plate contained a plasmid standard curve in duplicate (range 6.60 × 10⁶ to 6.60 × 10⁻¹ copies per well). Each plate contained a negative control in duplicate, and duplicate values were averaged for all samples only when the difference between samples had a threshold cycle (Ct) value < 1. If the Ct was > 1, the extracted sample was repeated in duplicate. A cutoff value representing the detection limit of the assay was extrapolated from a plasmid standard curve, as previously described (29), and was set at 1–10 copies of PCV-2 per gram of homogenized feces.

Statistical analysis

Since PCV-2 shedding in feces was non-normally distributed, nonparametric tests were used for statistical comparison. The Wilcoxon rank sum test was used to compare production stage group median PCV-2 shedding between PCVD-nonaffected and PCVD-affected pigs. To control for multiple comparisons, a Bonferroni correction was applied to adjust the P-value at which comparisons were considered significantly different (P = 0.005 for 10 comparisons).

Within each barn, the median PCV-2 shedding in feces for the production stage with the highest observed load was compared with that of the preceding production stage, using a Wilcoxon rank sum test. For groups where there was an observed difference in the presence of pigs not shedding detectable virus in the feces, the numbers of shedders and nonshedders were compared, using a Fisher’s exact test. Statistical analysis was performed, using a commercial statistics package (Statistix 8; Analytical Software, Tallahassee, Florida, USA). The box and whisker plots in Figure 1 were created by using a separate statistics package (Stata 10; StataCorp LP; College Station, Texas, USA).

Figure 1.

Box and whisker plots of PCV-2 shed in feces by pen quantified by real-time PCR (approximately 10 pooled pens per stage; range = 7 to 14) for each production stage in a PCVD-nonaffected and a PCVD-affected commercial swine herd. Production stages significantly different between barns were NW, NU, and PG (P < 0.001), and VG (P = 0.0002). [Horizontal lines represent the median; upper and lower edges of the boxes represent the 75th and 25th percentile, respectively; and lines (whiskers) attached to the top and bottom of the boxes extend to the maximum and minimum data points that are up to 1.5 times the interquartile range (1.5 IQR) from the 75th or 25th percentile, respectively. Outliers are data points that are more than 1.5 IQR from either the 25th or 75th percentile and are denoted as ‘x’].

PCV-2 — Porcine circovirus-2; PCR — polymerase chain reaction; PCVD — Porcine circovirus disease; NW — newly weaned; NU — nursery; PG — pregrower; GR — grower; FIN — finisher; VG — virgin gilt; BG — bred gilt; S ≤ 2 — lactating sow parity ≤ 2; S ≥ 3 — lactating sow parity ≥ 3; B — boar.

Results

Since an interaction was present between production stage and barn with respect to the PCV-2 shed in feces, separate barn comparisons were made for each production stage. The observed median PCV-2 shed in feces was higher in the PCVD-affected herd compared with the PCVD-nonaffected herd in the NW, NU, PG, GR, S ≤ 2, S ≥ 3, and B stages, but it was significantly higher only in the 3 youngest (NW, NU, PG) stages (P < 0.001; Figure 1). Conversely, the observed median PCV-2 shed in feces was higher in the PCVD-nonaffected herd in the FIN, VG, and BG stages, but it was significantly higher only in the VG stage (P = 0.0002; Figure 1). The highest observed PCV-2 shed in feces occurred in a later production stage in the PCVD-nonaffected herd (GR) than in the PCVD-affected herd (NU), and the PCV-2 shed was significantly higher than in the immediately preceding stages (P ≤ 0.0001) for each herd.

For most production stages, there was detectable viral DNA in all or almost all fecal samples. More sows (S ≤ 2 and S ≥ 3) in the PCVD-nonaffected herd did not shed PCV-2 in feces, or the level of PCV-2 shedding was below the detection limit of the quantitative PCR assay used in this study, when compared with the PCVD-affected herd (Figure 2). However, this difference was significant only for the older sow group (S ≥ 3; P = 0.001).

Figure 2.

The percent of pooled fecal samples that was negative for PCV-2 by quantitative real-time PCR for each production stage in a PCVD-nonaffected and a PCVD-affected commercial swine herd. The number of sows that were negative for PCV-2 in feces of the total tested is shown. Production stage significantly different between barns (S ≥ 3; P = 0.001) is denoted with an asterisk (*).

PCV-2 — Porcine circovirus-2; PCR — polymerase chain reaction; PCVD — Porcine circovirus disease; NW — newly weaned; NU — nursery; PG — pregrower; GR — grower; FIN — finisher; VG — virgin gilt; BG — bred gilt; S ≤ 2 — lactating sow parity ≤ 2; S ≥ 3 — lactating sow parity ≥ 3; B — boar.

Discussion

This study evaluated the PCV-2 shed in pig feces among production stages in a PCVD-affected compared with a PCVD-nonaffected herd, using quantitative real-time PCR. Quantities of PCV-2 DNA were significantly higher in younger production stages in the PCVD-affected herd (NW, NU, and PG) than in the PCVD-nonaffected herd; however, gestational sows and boars (BG, S, and B) had similar median PCV-2 shed in feces between herds. Possible explanations for the differences in viral shedding profiles of young pigs between these herds may be associated with early PCV-2 exposure or the infectious dose, or both, to which piglets are exposed in the farrowing crate by sows shedding PCV-2. Sows shedding PCV-2 in feces into the surrounding environment may expose piglets to the virus via the oronasal route (23). In addition, vertical transmission of PCV-2 from the sow to piglets can occur in utero (3,7), and PCV-2 is shed in colostrum (30). It is feasible that sows with a higher systemic PCV-2 viral load may expose piglets to a higher infectious dose of PCV-2, either in utero or during the neonatal period, but this has not been reported.

Maternal antibody protection provided to piglets in the neonatal period may be extremely variable among sows (31,32). The amount or concentration and the virus neutralizing capability of the antibodies provided to piglets from the sow affect both the duration of immunity in the young pig and the ability to reduce or prevent viral replication. Both of these conditions contribute to effectively reduce the PCV-2 viral load in the piglet; however, the immune status of the sows or their respective piglets in this study was not known. Passive antibodies provide protection in early life (NW, NU, and, possibly, PG); when maternal antibodies wane, PCV-2 viremia is detected (33). The report that the amount of PCV-2 shed in feces is correlated with the systemic or tissue PCV-2 viral load (22) supports the observation that a significant increase in the PCV-2 shed in feces occurred in the GR stage in the PCVD-nonaffected herd, after protective maternal antibodies had waned. This suggests that sufficient protection was afforded to the piglet until degradation of maternal antibodies at approximately 10 wk of age, at which time, PCV-2 shedding in feces was observed to increase significantly. It can only be postulated that passive protection afforded to the piglet by maternal antibodies; the dose of PCV-2 that piglets are exposed to in early life, or both contribute to the resulting viral load in the pig once it reaches the GR stage.

In the PCVD-affected herd, the PCV-2 shed in feces was highest in the NU stage. The PCV-2-specific immunity of the sows, or their respective progeny, was not tested in this study; however, by pooling fecal samples representing each production stage within a PCVD-affected and a PCVD-nonaffected herd, it was determined that significantly more sows in the PCVD-affected herd shed detectable levels of PCV-2 in their feces. This observation suggests that sows shedding PCV-2 in their feces may contribute to the earlier exposure and infection of piglets in the farrowing crate, as seen by the PCV-2 fecal shedding profile of young pigs in the PCVD-affected herd (Figure 1). The corollary is that the majority of sows in a PCVD-nonaffected herd may not shed detectable levels of PCV-2 in their feces; therefore, piglet exposure is reduced and delayed until the mixing of pigs in the NU stage. A significant increase in PCV-2 DNA is then observed in the feces of pigs in the GR stage, as a result of this delay in exposure in the PCVD-nonaffected herd.

To substantiate the effect of sows shedding PCV-2 in their feces on piglets in the farrowing crate, future studies, using serially collected samples of blood and feces from individual animals, are underway. Quantifying PCV-2 in the feces of piglets from sows with known immune and fecal shedding status would further elucidate this association, and piglets from individual sows in a population should be followed through NW, NU, PG, GR, and FIN production stages. Additionally, it is not known if pen density (Table 2) and the resulting increase in the mixing of pigs contributed to the increase in PCV-2 fecal shedding in younger pigs in the PCVD-affected herd in this study. Future studies should include herds consisting of similar pen densities by production stage of the pigs.

The PCV-2 shed in feces across production stages in PCVD-affected and PCVD-nonaffected herds have not been reported previously. Based on the shedding profile between pigs in a PCVD-affected and a PCVD-nonaffected herd, quantitative real-time PCR of PCV-2 in feces may be a useful tool to evaluate viral load in pigs. Future applications for PCV-2 quantitative PCR in pig feces may include measures for vaccine efficacy, treatment of PCVD, or changes in production and management practices.

More pigs, particularly lactating sows, in the PCVD-nonaffected herd had no PCV-2 DNA detected in feces by quantitative real-time PCR when compared with the PCVD-affected herd. As a result, there may be an association between the presence of PCV-2 in the feces of lactating sows, increased quantity of the virus in the feces of younger pigs, and mortality in PCVD-affected herds. Further study with serial sampling in PCVD-affected and PCVD-nonaffected farms is required to elucidate this potential relationship.