Determination of Infectious Titer and Virulence of an Original US PEDv PC22A Strain
Porcine epidemic diarrhea (PED) is a highly contagious enteric disease of swine characterized by acute watery diarrhea, vomiting, dehydration, and weight loss. It was first observed in farm pigs in England in 1971, write Xinsheng Liu et al, Ohio State University.
4 December 2015
11 minute read
By: The causative agent, PED virus (PEDV), was identified in 1978. Subsequently, PEDV caused epidemics with pig mortality in European countries until the late 1980s. Thereafter PEDV was more often associated with endemic cases in Europe.
In Asia, epidemics with major losses in suckling pigs were first reported in 1982, continuing into the 1990s-2000s.
Since October 2010, severe PED epizootic outbreaks, affecting pigs of all ages but characterized by high mortality rates among suckling piglets, have been reported in China, causing significant economic losses.
In April 2013, PEDV emerged in US swine and spread rapidly throughout the country, leading to the death of 8 million pigs and economic losses between $900 million - $1.8 billion in 2013–2014.
The pathogenesis of the original US PEDV isolates has been evaluated in gnotobiotic (Gn), cesarean-derived, colostrum-deprived (CDCD), and conventional pigs.
In agreement with field observations, the original US PEDV isolates were highly virulent with 100% morbidity and 50-100% mortality in suckling piglets.
Intentional exposure of pregnant sows to PEDV via feedback of fecal material or/and the intestinal tracts of infected piglets stimulated maternal immunity and shorted the outbreaks on some farms. However, outbreaks of PEDV continue in the US. The development of effective vaccines is urgently needed to prevent PED.
To evaluate vaccine efficiency in vivo, challenge of piglets derived from immunized sows with a standardized and validated dose of PEDV is required. To date, information on the infectious titer of a PEDV challenge pool is limited and a standard PEDV inoculum is needed to assess PEDV vaccine efficacy.
In this study, the median pig diarrhea dose (PDD 50 ) of an original US PEDV, PC22A strain, was determined by using 4-day-old CDCD piglets and was confirmed using conventional suckling pigs. Also, the acute clinical signs and pathological lesions were studied. The animal use protocols were reviewed and approved by the Agricultural Animal Care and Use Committee, The Ohio State University.
Four PEDV-naïve sows (A, B, C, D) with no previous herd history of PED outbreaks and that tested PEDV seronegative (Veterinary Diagnostic Laboratory, University of Minnesota (UMN)) were identified.
They were subjected to cesarean section for the derivation of CDCD pigs (n = 53). On the day of birth, the piglets were fed bovine colostrum replacer (AgriLabs, St. Joseph, MO, USA) that was gradually replaced by whole milk (Parmalat, Parmalat USA Corp) by the third day.
Treatments with antibiotics (Penicillin (VET one, UK), gentamicin (VET one, UK) and neomycin (MED-PHARMEX, Pomona, CA, USA)), antitoxin (Clostridium antitoxin (Boehringer Ingelheim Vetmedica, St. Joseph)) and probiotics (probios (Chr. Hansen, Menomonie, WI, USA)), as prescribed by the clinician, were performed to prevent opportunistic infections of piglets throughout the experiment.
All piglets were randomly allocated into 8 dose-groups (G1-G8) and two mock control groups (Table 1). Each piglet was housed in individual steel cages and the same group of pigs was in cages installed in 3 levels (2–4 cages/level) above the floor. The cages were maintained in biological safety level II (BSL2) rooms. Two adjacent dose-groups of pigs were housed in different sides of the same room (5.4 × 3.0 m 2 ).
The virus pool of PEDV PC22A strain was prepared in a gnotobiotic (Gn) pig as described in our previous study. The second passage (P2) of tissue culture-adapted PC22A was plaque-purified and one plaque was selected and propagated once more (total P3) in Vero cells. It was used as inoculum (5 log 10 plaque-forming unit (PFU)/mL) to orally challenge a 19-day-old Gn pig.
The pig was euthanized at 1 day post-inoculation (dpi) when it had watery diarrhea. The small and large intestinal contents were collected at 1 dpi aspectically, mixed and stored in 1 mL aliquots at −80 °C as the virus pool.
The infectious and RNA titers of the virus pool were 7.75 log 10 PFU/mL and 13 log 10 genomic equivalents (GE)/mL, respectively, using plaque assay and quantitative real-time reverse transcription-PCR (RT-qPCR) as reported.
For the preparation of inocula for pigs, 1 mL of PC22A virus pool was diluted 1:10 in phosphate-buffered saline (PBS, pH 7.4; Sigma-Aldrich, St. Louis, MO), mixed well, and vortexed. The suspension was centrifuged at 2095 × g for 10 min at 4 °C, and the supernatant was collected as the 1:10 (10 −1 ) dilution. Subsequently, 10-fold serial dilutions (10 −2 -10 −10 ) were prepared in PBS.
Each experimental group (G1-G8) of piglets was inoculated orally with 3 mL of the 10-fold serially diluted (10 −3 -10 −10 ) virus at 4 days of age. The control groups received PBS (Table 1).
After inoculation, piglets were observed 4 times daily for clinical signs, including diarrhea. Rectal swabs were collected daily from all piglets and scored for fecal consistency: 0 = normal; 1 = pasty; 2 = semi-liquid; and 3 = liquid. Scores of 3 were considered as watery diarrhea.
Median pig diarrhea dose (PDD 50 ) was determined as the reciprocal of the virus dilution at which 50% of the pigs developed watery diarrhea at a given time point using the Reed and Muench method.
To reduce the risk of cross contamination among pigs, PEDV PC22A-inoculated CDCD piglets were euthanized at onset of watery diarrhea and subjected to necropsy examination. Duodenum, jejunum, ileum, cecum, colon and mesenteric lymph nodes were collected and fixed in 10% neutral buffered formalin for histopathological examinations as described previously.
For each jejunum section, ten villi and crypts were measured using a computerized image system (PAX-it software, PAXcam, Villa Park, IL, USA). Villous height and crypt depth ratios (VH:CD) were calculated.
Also, PEDV nucleocapsid (N) proteins were detected by immunohistochemistry (IHC) using mouse monoclonal antibody (SD6-29) (gift from Drs. Steven Lawson and Eric Nelson at South Dakota State University).
Because conventional suckling pigs are the targets for future vaccine studies, PEDV-naïve sow E was selected and the naturally delivered suckling piglets were inoculated orally with PC22A at 100 PDD 50 /pig at 4 days of age to verify the results from the CDCD pig experiments.
In addition, two PEDV-field exposed-recovered sows F and G were obtained from a farm with a recent PEDV outbreak (July 19, 2014) and subsequent exposure to live virus for 3 continuous days (July 20–22, 2014), at 73 to 75 days pre-farrowing. Serum samples of sows F and G tested positive for PEDV-specific IgG, IgA and virus neutralizing antibodies at 53 days post-outbreak (20 and 22 days pre-farrowing) (Table 2) by PEDV-specific cell culture immunofluorescence (CCIF) and plaque reduction virus neutralization (PRVN) assays as described.
Sows F and G delivered 13 and 10 piglets, respectively, by natural farrowing. Piglets and their sows were housed together in separate rooms for each litter. At 4 days of age, piglets of sow F and G were inoculated orally with 10 000 PDD 50 and 1000 PDD 50 , respectively. On 7 dpi and 9 dpi, respectively, the piglets and their sows were euthanized.
Porcine epidemic diarrhea virus RNA fecal shedding in rectal swab samples or intestinal contents was negative before virus inoculation and became positive on 1 dpi in G1-G6 CDCD piglets (10 −3 -10 −8 diluted virus), with titers ranging from 9.8-13.7 log 10 GE/mL (Table 1).
By 1 dpi, 100% of pigs of G1 to G5 and 40% (2/5) of G6 had diarrhea, and no pigs in G7 and G8 (10 −9 and 10 −10 diluted virus) and control groups 1 and 2 had diarrhea. The 2 pigs in control group 1 were housed in the same room as G5 pigs (with 10 −7 diluted virus). They were clinically healthy on 1 dpi but developed watery diarrhea on 2 dpi.
The two pigs of control group 1 shed viral RNA on 1 and 2 dpi, respectively. These results indicated that cross contamination of PEDV occurred between the two groups (G5 and control 1) of pigs housed in the same room.
The cut-off time point was set as 1 dpi for determination of the PDD 50 which was 7.83 PDD 50 /3 mL, corresponding to 7.35 log 10 PDD 50 /mL. It was similar to the cell culture infectious titer (7.75 PFU/mL) determined by plaque assay. No clinical signs were observed and no PEDV RNA shedding was detected by 3 dpi in the two pigs of control 2 group.
Microscopically, different stages of villous atrophy (Figures 1A-D) were observed in the same group of CDCD piglets receiving the same virus dose at 1–3 dpi. In some piglets, the length of villi decreased slightly and the morphology of villous epithelial cells was generally intact (Figures 1A-B).
In other cases, significant shortening of villi along with exfoliation and vacuolation of enterocytes were observed (Figure 1C). As the severity of villous atrophy increased, the villi were blunted and fused throughout the entire small intestine (Figure 1D).
However, there were no significant differences in the degree of villous atrophy, and PEDV antigen scores among pig groups receiving different doses.
Overall, the majority (40/49, 82%) of PEDV-inoculated CDCD piglets had mean jejunum VH:CD ratios <2. In contrast, the mean jejunum VH:CD ratios of the 2 control pigs of control group 2 were 7.16 ± 1.25 and 7.80 ± 0.60, respectively.
Signal of PEDV antigens were in brown color and detected in the entire villous epithelial cells in the middle jejunum of PEDV PC22A-inoculated CDCD piglets (A-D). Shortening and fusion of villi along with exfoliation of enterocytes were observed (C, D). Depending on the progress of PEDV infection, PEDV-infected villous epithelial cells still formed villi (B), detaching, swelling and undergoing necrosis (C), or highly attenuated (D). PEDV antigens extended to the villus/crypt border and were sporadically located in the crypt cell layer (CCL) in ileum obtained from CDCD (E) and conventional (F) piglets. The majority of the residual epithelial cells were PEDV-positive. In addition, PEDV antigens were detected in few mononuclear cells in dome area in ileum obtained from conventional, PEDV-inoculated piglets derived from two PEDV-exposed and recovered sows (G). No PEDV antigen was detected in the jejunum of a mock-inoculated CDCD piglet (H). Original magnifications: × 40 (A), × 200 (B-H).
No PEDV N proteins were detected in tissues collected from the 2 pigs of control group 2 by IHC (Figure 1H). However, for the PEDV-inoculated pigs, the entire small intestine was positive for PEDV N proteins, which were detected mainly in the cytoplasm of enterocytes located both on the tip and lateral walls of the villi extending to the villus/crypt border (Figures 1A-B).
Less frequently, PEDV-positive cells were observed in or near the crypts of Lieberkühn (5/49, 10%) (Figures 1A-B). The highest IHC signal intensity was detected in some cases with relatively higher VH:CD ratios (2.54 to 5.17) (Figures 1A-C). As the severity of villous atrophy increased, the total number of villous epithelial cells and PEDV-positive cells decreased dramatically (Figure 1D). Scattered foci of PEDV antigens were detected in the colonic epithelia in 6/49 (12%) PEDV-inoculated CDCD pigs (data not shown).
The 12 suckling piglets of PEDV-naive sow E had watery diarrhea on 1 dpi with a high level of PEDV RNA shedding (11.8-13.0 log 10 GE/mL). The piglets that died or were euthanized between 1 and 5 dpi had mean VH:CD ratios ranging from 0.48 ± 0.12 to 1.57 ± 0.32, except for one piglet that was euthanized at 5 dpi that showed a relatively higher mean VH:CD ratio (2.29 ± 0.64).
Patterns of PEDV antigen distribution were similar to those observed in CDCD piglets (data not shown). Degeneration of crypt epithelial cells containing abundant PEDV N antigen was noted in one (1/8) conventional piglet euthanized at 5 dpi (Figure 1F).
The piglets of sows F and G did not have diarrhea by 7 dpi and 9 dpi, respectively. Some piglets of sows F (3/13, 23%) and G (8/10, 80%) did not shed PEDV RNA in feces, and the remaining piglets shed virus at low titers (4.9-8.9 log 10 GE/mL) (Table 1).
The serum samples of sows F and G collected at euthanasia (7 or 9 dpi) had similar PEDV-specific antibody titers as the pre-farrowing titers, consistent with lack of PEDV re-stimulation of the sows due to piglet protection (Table 2). Milk samples of sow F and G collected on 1 dpi had much higher PEDV-specific antibody titers (1024 (IgG), 512 (IgA) and 1351 (VN) for Sow F and 1024 (IgG), 128 (IgA) and 675 (VN) for sow G, respectively) than serum samples (Table 2).
On 7 or 9 dpi, no significant histopathological lesions were observed in piglets of sows F and G. PEDV N proteins were detected in individual mononuclear cells in intestinal submucosa/Peyer’s patches but not in intestinal villous epithelial cells (Figure 1G).
