Mycoplasma hyopneumoniae bacterins and porcine circovirus type 2 (PCV2) infection: Induction of postweaning multisystemic wasting syndrome (PMWS) in the gnotobiotic swine model of PCV2-associated disease

Abstract

Groups (5 to 15 per group) of gnotobiotic swine were infected oronasally with porcine circovirus type 2 (PCV2) at 3 days of age and then given 1 of 6 different commercial Mycoplasma hyopneumoniae (M. hyopneumoniae) bacterins as either a single dose (7 d of age, 1 application products) or 2 doses (7 and 21 d of age, 2 application product). Control groups received PCV2 alone (n = 9) or were infected with PCV2 and immunized twice with keyhole limpet hemocyanin (KLH) emulsified in incomplete Freund’s adjuvant (ICFA) (n = 7). Five of 7 (71%) PCV2-infected piglets immunized with KLH/ICFA developed mild or overt PMWS, whereas none of 9 piglets infected with PCV2 alone developed PMWS. Five of 12 (42%) piglets vaccinated with a commercial bacterin containing mineral oil adjuvant developed PMWS following vaccination. None of the PCV2-infected piglets in the other bacterin-vaccinated groups developed PMWS in this model of PCV2-associated disease. This difference in prevalence of PMWS in piglets given the mineral oil-adjuvanted M. hyopneumoniae bacterin and the other M. hyopneumoniae bacterin vaccination groups was statistically significant (P < 0.05).

Introduction

Porcine circovirus type 2 (PCV2) infection emerged as a viral pathogen of swine in Canada in the mid to late 1990s (1,2). Infection with PCV2 occurs in all swine-producing areas of the world and postweaning multisystemic wasting syndrome (PMWS) and related PCV2-associated diseases (PCVAD) are increasingly recognized as serious threats to global hog production (3,4). A primary manifestation of infection in young swine is PMWS, a multisystem, angiocentric, and often progressive granulomatous inflammatory disease of lymphoid tissues, liver, kidney, lungs, and gastrointestinal tract (1,3,4). Histologic lesions of PMWS are distinct from those of other swine infectious diseases and are diagnostic for PCV2/PMWS when intra-cytoplasmic viral inclusion bodies are present (1,3,4). In recent years, PCV2 infection has been associated with abortion and reproductive failure (5,6), as an important component of the porcine respiratory disease complex (7,8) and as a possible infectious component of porcine dermatitis and nephropathy syndrome (3,4).

Subclinical infection with PCV2 is commonplace in swine production units (3,4). In fact, the high prevalence of subclinical infection complicates understanding of the virulence of this virus for swine. Proof for pathogenic capability was obtained first with gnotobiotic swine coinfected with Porcine parvovirus (PPV) (9,10) and then in colostrum-deprived conventional piglets coinfected with PPV (11) or Porcine reproductive and respiratory syndrome virus (PRRSV) (12,13). Challenge inoculation with PCV2 cloned viral DNA (14) has reproduced lesions attributable to PCV2, confirming that this agent is the necessary infectious cause of PMWS. In spite of these successes, experimental reproduction of clinically expressed disease in conventional swine at a level that can be used to evaluate management, nutritional, and breed or genetic influences upon disease expression is difficult. The neonatal gnotobiotic piglet model of PCV2 infection is a reliable experimental model in which a high percentage of infected piglets develop PMWS (9,10,15).

Although the cellular tropisms for replication of this virus during the preclinical phase of the infection are still unclear, both histologic and immunohistochemical (IHC) studies have repeatedly shown that histiocytes, dendritic follicular cells, and macrophages accumulate infectious virus, viral DNAs, and viral nucleocapsid protein (3,4,15–19). These findings suggested that “activation” of the histiocytic lineage cells by coinfections potentiates PCV2 replication and hence disease expression in infected swine. In gnotobiotic swine, fatal PMWS occurs in a high percentage of PCV2-infected viremic piglets, provided that they are immunostimulated by injections with keyhole limpet hemocyanin (KLH) emulsified in a mineral oil-based adjuvant, namely incomplete Freund’s adjuvant (ICFA) (15). Field studies using artificial immunostimulants (20) and certain commercial swine bacterins (20–23) are associated with an increased incidence of PMWS in vaccinates; others have not found this association (24). Of the swine biologicals tested, the Mycoplasma hyopneumoniae bacterins have received the most scrutiny, since early-age vaccinations for protection against mycoplasmal enzootic pneumonia are routinely practised. As well, anecdotal reports from both producers and veterinarians indicate that an early-age vaccination program for protection against mycoplasma pneumonia may be a risk factor for development of PMWS. The objective of this study was to determine if selected commercial M. hyopneumoniae bacterins potentiate PCV2 replication in infected gnotobiotic swine and if this effect increases the severity of PCV2-specific lesions and clinical expression of disease as PMWS. Gnotobiotic swine (25) were used as these pigs are not infected with other swine pathogens or commensal microbes and the diet and environmental husbandry conditions can be precisely controlled.

Materials and methods

Piglets

Eight date-mated pregnant sows were purchased from a single closed confinement swine operation to ensure that derived piglets were term gestation, healthy, and free of other known swine infectious diseases. A total of 75 piglets were derived by cesarian section, delivered into sterile pen tub isolation units (2–6 piglets per unit), and raised on sterile sow milk replacement liquid diet, as previously described (25). All animal experiments were approved by the Institutional Animal Use and Care Committee of The Ohio State University and are consistent with guidelines of the Canadian Council on Animal Care.

Porcine circovirus type 2 (PCV2)

A tissue culture-propagated isolate of PCV2 (Stoon 1010) (9) was the original source of infectious virus for this study. The virus was propagated in PCV-free PK15 cells (26) and confirmed to be PCV2, as described previously (3,9,10). A viral pool consisting of a 10% (w/v) lymphoid tissue and liver homogenate was prepared from the 3rd in vivo pass of this virus in gnotobiotic piglets, aliquoted into 1.0-mL amounts, and frozen at −70°C until use. The inoculum contained 4.3 × 10⁸ units of infectious virus per mL when titrated on PK15 monolayers (15). All 75 piglets were inoculated by the oronasal route with 1.0-mL aliquots of PCV2 inoculum at 3 d of age; experiments were terminated when the piglets were 34–44 d of age, 31–41 d after infection with PCV2.

Commercial Mycoplasma hyopneumoniae bacterins

Five single dose commercial [Pfizer Animal Health, New York, New York, USA (PAH); Fort Dodge Animal Health, Fort Dodge, Iowa, USA (FDAH-1); Boehringer Ingelheim Vetmedica, St. Joseph, Missouri, USA (BI); Intervet, Millsboro, Delaware, USA (IV); and Schering-Plough Animal Health, Union, New Jersey, USA (SPAH)] and one 2-dose commercial [Fort Dodge Animal Health (FDAH-2)] bacterin that are recommended for use as aids in the protection of swine from mycoplasma-associated pneumonia were evaluated. All contained inactivated Mycoplasma hyopneumoniae bacterin emulsified in proprietary adjuvant and were administered to PCV2-infected piglets by IM injection(s) at 7 d of age (1-dose products) or at 7 and 21 d of age (2-dose product). As negative and positive controls, respectively, piglets were infected with PCV2 alone (n = 9) or with PCV2 and immunized at 5 and 12 d of age with keyhole limpet hemocyanin emulsified in incomplete Freund’s adjuvant (KLH/ICFA) (n = 7), as described previously (15).

Experimental design, and histologic and immunohistochemical (IHC) studies

Each litter was divided into subgroups of PCV2-infected piglets for vaccination(s). Piglets received PCV2 only (n = 9), PCV2 and KLH/ICFA (n = 7), PCV2 and PAH (n = 12), PCV2 and FDAH-1 (n = 9), PCV2 and FDAH-2 (n = 13), PCV2 and BI (n = 15), PCV2 and SPAH (n = 5), PCV2 and IV (n = 5). At the termination of the observation period (34–44 d of piglet age; that is, 31 to 41 d after infection with PCV2) or when moribund with PMWS, pigs were euthanized and a standard set of samples (bronchial lymph nodes; thymus; superficial inguinal, axillary and mesenteric lymph nodes; tonsil; lung; liver; kidney; spleen; ileum; and bone marrow) were collected from each pig and fixed in 100% cold ethanol for 24 h; samples were trimmed, embedded in paraffin, sectioned, and stained with hematoxylin and eosin, or by IHC for PCV2 nucleocapsid protein as described previously (15,19,27,28).

Quantitative scoring of gross and histologic lesions and severity and distribution of viral nucleocapsid protein by IHC

Each piglet from each group was scored subjectively for clinical, gross, histologic, and IHC findings, using previously delineated criteria (19). The gross pathologic changes observed were assigned grades as follows: minimal lymphadenopathy = 1; moderate lymphadenopathy = 2; severe lymphadenopathy = 3; thymic atrophy = 1; ascites = 1; icterus = 1; pale liver = 1; tan and atrophic liver = 2; gastroesophageal ulceration with hemorrhage = 1; and wasting and emaciation = 1. The gross finding scores from each pig were added together and the group median value was then calculated for each vaccination group. Histologic changes (systemic lymphoid depletion, granulomatous inflammation, and syncytial giant cells) characteristic for PMWS in bronchial and peripheral lymph nodes were each scored as absent = 0, minimal = 1, moderate = 2, or severe = 3. The thymus was scored on the basis of medullary reticuloendothelial cell proliferation, cortical thymocyte depletion, and syncytial giant cells, using the same severity scale. Similarly, lymphocytic interstitial pneumonia, hepatitis, and nephritis were scored as minimal = 1, moderate = 2, or severe = 3. The extent of granulo-matous inflammation (when present in these organs) was scored on the same severity scale. The total histologic score from each pig was used to determine median values for each vaccination group. The distribution and severity of PCV2 nucleocapsid protein as an index of the extent of viral infection (19) was scored on a similar distribution and severity scale. The total IHC score from each pig was used to determine the median values for each vaccination group. Based on a combination of gross and histologic lesions and IHC data for the distribution and severity of PCV2 nucleocapsid in tissues, piglets were assigned into 1 of 3 broad disease categories: subclinical PCV2 infection, mild PCV2-associated PMWS, or overt PCV2-associated PMWS (19).

Statistical analysis

The Fisher’s exact test was used to compare the occurrence of PMWS between experimental groups. The Kruskal-Wallis test was used to evaluate the distribution of differences in score variables among challenge and vaccination groups for the gross, histologic, and IHC data. Protected post-hoc Wilcoxon rank-sum tests were used to measure the significance of differences in score variables between pairs of treatment groups where the Kruskal-Wallis test was significant. Differences between groups were considered statistically significant (P < 0.05).

Results

Subclinical PCV2 infection

Sixty-five piglets remained clinically asymptomatic throughout the experiment and were thus subclinically infected with PCV2. Piglets were alert and promptly consumed their diet at each feeding. Lymphadenopathy, if present, was histologically manifest as lymphofollicular and T-cell hyperplasia. As well, occasional hyperplastic foci of histiocytes and follicular dendritic cells in the centers of lymphoid follicles were seen. There was a general absence of PCV2 antigen in hepatocytes and in pulmonary and renal epithelial cells; histologic lesions in these organs consisted of multifocally distributed lymphoplasmacytic infiltrates. By IHC staining, viral nucleocapsid was largely confined to foci of proliferating histiocytes and inflammatory cell infiltrates in lymphoid tissues, liver, kidney, and lung.

Mild postweaning multisystemic wasting syndrome

Five of 75 PCV2-infected piglets survived the experimental observation period (35–44 d) but were “unthrifty” at its termination. In addition to mild wasting (roughened hair coat and emaciation), and anorexia, several exhibited subcutaneous edema, ascites, and mild icterus at necropsy. Lymphadenopathy, particularly of the bronchial and aortic lymph node chain, was prominent and characterized by proliferation and replacement of depleted T-cell zones and lymphoid follicles by histiocytes and macrophages. Syncytial giant cells were present and at least 1 lymph node set (usually bronchial) contained PCV2 cytoplasmic viral inclusion bodies in histiocytes and macrophages. In the liver, many hepatocytes exhibited cytoplasmic evidence of degeneration (acute cellular swelling and hydropic degeneration); a few hepatocyte nuclei were pyknotic. There was substantial infiltration of hepatic and renal parenchyma with lymphocytes and histiocytes, which often formed small granu-lomatous inflammatory foci. When present, pulmonary lesions were similar to those found in the liver and kidney. In lymphoid tissues, PCV2 nucleocapsid protein was widely distributed in macrophages, histiocytes, and syncytial giant cells. Occasional hepatocytes contained cytoplasmic viral nucleocapsid protein.

Overt postweaning multisystemic wasting syndrome

Five of 75 PCV2-infected piglets died or were euthanized before the end of the experimental period. Clinically, these piglets exhibited sudden onset anorexia and depression. Wasting appeared within 1 or 2 d after the onset of anorexia: generalized subcutaneous edema and ascites (4 of 5), clinically evident icterus (3 of 5). Two of 5 PMWS-affected piglets vaccinated with PAH had grossly evident hepatic atrophy. Histologic changes typical of PMWS, specifically granulomatous hepatitis with associated necrosis and loss of hepatocytes, were present in all 5 pigs (1,3,4,19). Viral inclusion bodies were seen in macrophages, and syncytial giant cells were found in the liver; all lymphoid tissues were severely depleted of lymphocytes, which were replaced by granulomatous inflammation that contained abundant and strongly staining PCV2 nucleocapsid protein and viral inclusion bodies.

Tabulated data

The results of these experiments are summarized in Tables 1 through 4. All 9 piglets infected with PCV2 and not vaccinated remained clinically normal. In contrast, 5 of 7 piglets infected with PCV2 and immunized with KLH/ICFA developed mild or overt PMWS. Of the piglets infected with PCV2 and vaccinated with Mycoplasma hyopneumoniae bacterins, only piglets (5 of 12) from one vaccination group (PAH) developed mild or overt PMWS. The distribution of gross changes associated with PCV2 infection is presented in Table 1. Piglets with PMWS exhibited severe generalized lymphadenopathy and most developed thymic atrophy. Regarding the latter, all 5 PMWS-affected piglets in the KLH/ICFA vaccinated group exhibited thymic atrophy, as did 6 of 12 piglets vaccinated with PAH bacterin; 1 of 5 piglets vaccinated with IV was similarly affected. Extra-lymphoid lesions detected in PMWS-affected piglets included ascites, icterus, and various degrees of liver damage (pale and tan to atrophy). Two infected piglets vaccinated with PAH had severe gastroesopha-geal ulceration with hemorrhage into gastric lumens.

Table 1.

The distribution and severity of gross findings in gnotobiotic piglets infected with porcine circovirus type 2 (PCV2) at 3 d of age and then vaccinated with various commercial Mycoplasma hyopneumoniae bacterins once at 7 d of age (1-dose products) or twice at 7 and 21 d of age (2-dose product)

		Generalized lymphadenopathy						Liver
Piglet infection and vaccination groups	Piglets per group	Minimal	Moderate	Severe	Thymic atrophy	Ascites	Icterus	Pale, Tan	Tan, Atrophic	Other
Negative (PCV2 only) and positive (KLH/ICFA-vaccinated) PCV2 infection groups
PCV2 only	9	8	1	—	—	—	—	—	—	—
KLH/ICFAa	7	—	4	3	5	4	3b	4	—	—
PCV2 infection and Mycoplasma hyopneumoniae bacterin vaccination groups
PAHa	12	—	4	8	6	2	—	8	2	3c
FDAH-2a	9	5	4	—	1	—	—	—	—	—
FDAH-1a	13	9	4	—	—	—	—	—	—	—
BIa	15	11	4	—	—	—	—	—	—	—
SPAHa	5	4	1	—	—	—	—	1	—	—
IVa	5	1	3	1	1	—	—	1	—	—

^aAbbreviations used: KLH/ICFA = keyhole limpet hemocyanin emulsified in incomplete Freund’s adjuvant; PAH = Pfizer Animal Health, New York, New York, USA; FDAH-1 = Fort Dodge Animal Health, Fort Dodge, Iowa, USA, 1-dose product; FDAH-2 = Fort Dodge Animal Health, 2-dose product; BI = Boehringer Ingelheim Vetmedica, St. Joseph, Missouri, USA; SPAH = Schering-Plough Animal Health, Union, New Jersey, USA; IV = Intervet, Millsboro, Delaware, USA

^bThese 3 animals had clinically evident icterus prior to necropsy evaluation

^c2 piglets in this group had severe gastroesophageal ulceration with hemorrhage; a 3rd pig was severely emaciated

Table 4.

A summary of statistical analyses (P < 0.05) of the differences between vaccination groups by clinical score (PMWS prevalence), gross lesions, histologic lesions of PMWS and IHC staining estimates of tissue viral burden in gnotobiotic piglets infected with porcine circovirus 2 (PCV-2) and then vaccinated with various commercial Mycoplasma hyopneumoniae bacterins

Piglet infection and vaccination groups	PMWS Prevalence (%)1	Median gross lesion score2 (min and max)	Median histologic lesion score3 (min and max)	Median immunochemical (IHC) score4 (min and max)
Negative (PCV2 only) and positive (KLH/ICFA-vaccinated) PCV2 infection groups
PCV2 only	0 (0/9)ab	1.0 (1.0–5.0)a	3.5 (2.0–9.0)a	9.7 (5.3–19.5)a,b
KLH/ICFA5	71.0 (5/7)c	7.0 (3.0–10.0)b	13.5 (7.5–24.0)b,c	24.0 (13.5–34.0)b
PCV2 infection and Mycoplasma hyopneumoniae bacterin vaccination groups
PAH5	42.0 (5/12)a,c	5.0 (1.5–9.0)b	13.0 (9.0–19.0)b	12.2 (5.5–18.3)a,b
FDAH-25	0 (0/9)a,b	1.5 (0.0–3.5)a	4.0 (2.5–13.0)a,c	6.0 (4.5–15.5)a
FDAH-15	0 (0/13)b	1.0 (0.0–3.0)a	4.5 (0.5–10.5)a,c	6.1 (3.7–13.2)a
BI5	0 (0/15)b	2.0 (0.0–4.0)a	4.0 (1.5–12.0)a	7.0 (3.0–17.0)a
SPAH5	0 (0/5)a,b	2.5 (1.5–2.5)a,b	6.5 (3.0–7.0)b	6.5 (5.0–16.5)a,b
IV5	0 (0/5)a,b	4.0 (2.0–5.0)a,b	4.5 (3.5–8.0)b	14.0 (7.0–18.0)a,b

¹The percent prevalence of PMWS (mild and overt) was determined by the number of pigs with mild or overt PMWS divided by the total number of pigs in each experimental group

²The gross lesion score for each piglet in each group was obtained by scoring each lesion as follows: [ascites: yes = 1, no = 0, icterus: yes = 1, no = 0, bronchial lymphadenopathy (0 to 3), peripheral lymphadenopathy (0 to 3), thymus: atrophy = 1, no = 0, liver: no change = 0, pale = 1, tan = 1, gastric ulcer: yes = 1, no = 0, wasting: yes = 1, no = 0, hepatic atrophy: yes = 1, n = 0]. The total gross score for each piglet in each group were added together and the median score was determined for each experimental group

³The histologic lesion score for each piglet was obtained by grading each PMWS-associated histologic lesion observed on a scale of absent (0) to severe (3). Histologic features considered were: bronchial lymph nodes, (reticuloendothelial cell hyperplasia, lymphocyte depletion and syncytial giant cell formation), peripheral lymph nodes (reticuloendothelial cell hyperplasia, lymphocyte depletion and syncytial giant cell formation), thymus (reticuloendothelial cell hyperplasia, T lymphocyte depletion, and syncytial giant cell formation), and inflammatory lesions (0 to 3) in the liver, lungs, and kidney). The total histologic score for each piglet in each infection and vaccination group were added together and the median score was determined for each experimental group

⁴The immunohistochemical (IHC) staining score for each piglet was obtained by grading the severity and distribution of PCV2 nucleocapsid protein distribution on a severity scale of absent (0) to severe (3) and applied to the following tissue sections: bronchial lymph nodes, peripheral lymph nodes, thymus, liver, kidney and lung from each piglet in each experimental group. The total IHC score for each piglet in each infection and vaccination group were added together and the median group score was determined for each experimental group

⁵Abbreviations used: KLH/ICFA = keyhole limpet hemocyanin emulsified in incomplete Freund’s adjuvant; PAH = Pfizer Animal Health, New York, New York, USA; FDAH-1 = Fort Dodge Animal Health, Fort Dodge, Iowa, USA, 1-dose product; FDAH-2 = Fort Dodge Animal Health, 2-dose product; BI = Boehringer Ingelheim Vetmedica, St. Joseph, Missouri, USA; SPAH = Schering-Plough Animal Health, Union, New Jersey, USA; IV = Intervet, Millsboro, Delaware, USA

^abcValues within the same vertical column with different superscripted letters are significantly different (P < 0.05) from each other

The distribution and severity of histologic lesions are given in Table 2. All PMWS-affected pigs, either immunized with KLH/ICFA or vaccinated with PAH bacterin, exhibited systemic lymphoid depletion and reticuloendothelial hyperplasia characteristic of PMWS. Similar but less severe lymphoid lesions were seen in some, but not all, piglets in other vaccination groups. Lymphocytic and plasmacytic cellular infiltrates into the liver were a regular feature of infection in all infected and vaccinated piglet groups. In a subset of both the KLH/ICFA- and PAH bacterin-vaccinates, multifocal to diffuse granulomatous inflammatory lesions were present in liver and kidneys. Nonsuppurative to mild granulomatous interstitial pneumonia was a prominent feature of disseminated disease in 5 of 7 KLH/ ICFA-immunized piglets and in all 12 of the PAH bacterin- vaccinates. Other vaccination groups also had histologic evidence of interstitial pneumonia, but the severity of this histologic change was less than in the 2 PMWS-affected piglet groups.

Table 2.

The incidence of histologic changes associated with postweaning multisystemic wasting syndrome (PMWS) in gnotobiotic piglets infected with porcine circovirus type 2 (PCV2) at 3 d of age and then vaccinated with various commercial Mycoplasma hyopneumoniae bacterins once at 7 d of age (1-dose products) or twice at 7 and 21 d of age (2-dose product)

				Hepatitis		Nephritis
Piglet infection and vaccination groups	Piglets per group	Systemic lymphoid depletion	Multifocal interstitial pneumonia	Lympho-plasmacytic	Granulomatous	Lympho- plasmacytic	Granulomatous
Negative (PCV2 only) and positive (KLH/ICFA-vaccinated) PCV2 infection groups
PCV2 only	9	2	1	9	0	3	0
KLH/ICFAa	7	7	5	3	4	3	1
PCV2 infection and Mycoplasma hyopneumoniae bacterin vaccination groups
PAHa	12	12	12	12	2	11	3
FDAH-2a	9	3	8	8	1	2	0
FDAH-1a	13	4	8	8	0	2	0
BIa	15	2	12	12	1	4	2
SPAHa	5	4	5	5	0	0	0
IVa	5	2	5	5	0	0	0

^aAbbreviations used: KLH/ICFA = keyhole limpet hemocyanin emulsified in incomplete Freund’s adjuvant; PAH = Pfizer Animal Health, New York, New York, USA; FDAH-1 = Fort Dodge Animal Health, Fort Dodge, Iowa, USA, 1-dose product; FDAH-2 = Fort Dodge Animal Health, 2-dose product; BI = Boehringer Ingelheim Vetmedica, St. Joseph, Missouri, USA; SPAH = Schering-Plough Animal Health, Union, New Jersey, USA; IV = Intervet, Millsboro, Delaware, USA

In general, the distribution and intensity of PCV2 nucleo-capsid protein detected by IHC staining paralleled histologic findings (Table 3). Moderate to severe grades (scores of 2 and 3) of IHC staining were detected most regularly in the bronchial lymph nodes of all KLH/ICFA-immunized pigs and the majority (9 of 12) of the PAH bacterin-vaccinates. Several piglets in the other vaccination groups had bronchial lymph node IHC staining scores of 3. A similar, but less consistent pattern of IHC staining was seen in other thymuses and peripheral lymph nodes (data not shown for thymus). Liver sections from PMWS-affected piglets contained abundant PCV2 antigen. As well, one half (6 of 12) of PAH bacterin vaccinates had moderate to extensive amounts of PCV2 nucleocapsid protein within the hepatic granulomatous and lymphocytic inflammatory cell infiltrates. In these pigs, spread of PCV2 beyond lymphoid tissues was also reflected in the appearance of PCV2-positive epithelial cells in the kidney and lung.

Table 3.

The incidence and severity of immunohistochemical (IHC) staining for PCV2 viral nucleocapsid protein in selected tissues^a from gnotobiotic piglets infected with porcine circovirus type 2 PCV2 at 3 d of age and then vaccinated with various commercial Mycoplasma hyopneumoniae bacterins once at 7 d of age (1-dose products) or twice at 7 and 21 d of age (2-dose product)

		Bronchial lymph nodes				Peripheral lymph nodes				Liver				Kidney and lung
Piglet infection and vaccination groups	Piglets per group	0	1	2	3b	0	1	2	3b	0	1	2	3b	0	1	2	3b
Negative (PCV2 only) and positive (KLH/ICFA-vaccinated) PCV2 infection groups
PCV2 only	9	—	4	3	2	—	6	3	—	—	6	3	—	5	2	2	—
KLH/ICFAc	7	—	—	1	6	—	1	2	4	—	1	3	3	—	1	3	3
PCV2 infection and Mycoplasma hyopneumoniae bacterin vaccination groups
PAHc	12	—	3	5	4	—	5	5	2	2	4	6	—	5	3	4	—
FDAH-2c	9	—	6	3	—	—	7	2	—	1	5	3	—	8	—	1	—
FDAH-1c	13	—	10	3	—	1	5	5	2	9	4	—	—	10	3	—	—
BIc	15	1	5	8	1	—	13	2	—	6	8	1	—	12	3	—	—
SPAHc	5	—	—	4	1	—	3	2	—	—	3	2	—	3	1	1	—
IVc	5	—	4	1	—	1	3	1	—	3	1	1	—	3	1	1	—

^aThe IHC staining data from the thymus is omitted from this table but was used to calculate the group mean IHC staining scores for statistical analyses in Table 4

^bThe amount and distribution of PCV2 nucleocapsid protein was assessed subjectively whereas a score of “0” was recorded when nucleocapsid protein was not found in the tissue. A score of “1” was recorded when individual cells or groups of cells widely scattered throughout the tissue section, a score of “2” was recorded when confluent foci of nucleocapsid-positive cells with rare to occasional syncytial giant cell formation was observed and a score of “3” was recorded when the amount of viral nucleocapsid was strongly and widely distributed throughout the tissue sections and syncytial giant cells were commonly observed. This scale has been described previously (19)

^cAbbreviations used: KLH/ICFA = keyhole limpet hemocyanin emulsified in incomplete Freund’s adjuvant; PAH = Pfizer Animal Health, New York, New York, USA; FDAH-1 = Fort Dodge Animal Health, Fort Dodge, Iowa, USA, 1-dose product; FDAH-2 = Fort Dodge Animal Health, 2-dose product; BI = Boehringer Ingelheim Vetmedica, St. Joseph, Missouri, USA; SPAH = Schering-Plough Animal Health, Union, New Jersey, USA; IV = Intervet, Millsboro, Delaware, USA

Differences between experimental groups

As shown in Table 4, the prevalence of PMWS in KLH/ICFA-immunized pigs was significantly (P < 0.05) greater than in all other PCV2 infection and vaccination groups, except those PCV2-infected piglets vaccinated with the PAH bacterin (P < 0.17). The prevalence of PMWS in the PAH product vaccinates was also significantly greater (P > 0.04) than in the FDAH-1, FDAH-1, and BI vaccine products. There was no significant difference (P > 0.05) in the PMWS status in PCV2 only-infected piglets or in PCV2-infected piglets vaccinated with FDAH-1, FDAH-2, BI, SPAH, and IV bacterins. Both gross and histologic scores directly reflected clinical findings and the median values for these indices of infection were the highest in the piglet groups immunized with KLH/ICFA (7.0 and 13.5, respectively) or vaccinated with the PAH bacterin (5.0 and 13.0, respectively) (Table 4). Median gross findings values for KLH/ICFA and PAH vaccinates were significantly higher than those vaccinated with FDAH-1, FDAH-2, and BI products. As well, the median histologic value generated in piglets vaccinated with the PAH bacterin was significantly higher than the values obtained from piglets infected with PCV2 only or in infected piglets vaccinated with either of the FDAH products or the BI bacterin. Group median gross and histologic scores in piglets infected with PCV2 only were not significantly different (P > 0.05) from those values obtained from piglets infected with PCV2 and vaccinated with FDAH-1, FDAH-2, BI, SPAH, and IV formulations. Conversely, IHC staining values for the KLH/ICFA-positive control group were significantly higher than those infected piglet groups vaccinated with FDAH-1, FDAH-2, and BI bacterins (Table 4). Although median IHC staining scores for both IV and PAH product-vaccinates were higher than the other vaccination groups, these median values were not statistically significant at the 0.05 level. The tissue section IHC staining scores were moderately correlated with histologic (P = 0.65, P = 0.0001) scores except for high IHC staining scores for piglets vaccinated with IV.

Thus, the group median gross, histologic, and IHC staining scores were consistent with the diagnosis of mild/overt PMWS in 5 of 7 KLH/ICFA-immunized piglets and in 5 of 12 PAH-vaccinates. Conversely, these same group median data values support the contention that, in this experimental model for induction of PMWS, vaccination(s) of PCV2-infected gnoto-biotic piglets with FDAH-1, FDAH-2, BI, and SPAH bacterins did not potentiate PMWS.

Discussion

The objective of these experiments was to determine if commonly used Mycoplasma hyopneumoniae bacterins potentiate PCV2 replication in vivo and hence induction of PMWS when given parenterally to 7 d (1-dose formulations) or 7 and 21 d (2-dose formulation) old PCV2-infected and viremic gnotobi-otic swine. The design was such that any vaccination-associated effect, if present, would be maximally expressed in vaccinates. Important to the evaluation of the bacterins was inclusion of a method for inducing PMWS (15), so that the effects of vaccinations could be compared with the KLH/ICFA-immunized positive control group.

For PCV2, the chief determinant of virulence appears to be the infectious virus burden contained within affected tissues/ organs (15,19,29). While kinetics of infection are not yet precisely determined, the more rapidly that high levels of virus are achieved within lymphoid tissues, the more likely it is that clinical PMWS will develop. The morphologic hallmarks of this phenomenon are generalized lymphadenopathy characterized by reticuloendothelial cell hyperplasia and concomitant T- and B-cell depletion (1,3,4,9,15,17–19). Experimentally, infectious virus levels below 10^5–6 infectious units per gram of tissue are expressed clinically as subclinical infections, whereas infectious titers above 10^7–8 infectious units are characteristic of PMWS (15,19). These titration data correlate to the predominant distribution of PCV2 wherein containment to lymphoid tissues is characteristic of subclinical infection and PMWS is characterized by not only extensive infection of lymphoid tissues but also an expanding viral tropism to various epithelial cell types (3,4,15,19,28).

The viral tissue burden may be elevated by increasing the rate of the virion replication cycle, increasing encapsidation efficiency of viral DNAs, or by increasing the numbers of cells that are permissive to viral replication. For PCV2, it is difficult to envision an increased rate of replication beyond that of the 22–24 h shown in vitro (30). Encapsidation is an intrinsic property of the single-stranded DNA and nucleocapsid protein (30) and, again, it is difficult to envision heightened encapsidation efficiency, since in vitro studies have shown that the virion assembles itself as components are produced (30,31). Thus, it is most likely that the amount of infectious virus in tissues is elevated by an increased number of virus-permissive cells within target organs. While the cell types permissive for viral replication have not been precisely identified, it is probable that histiocytic lineage cells are permissive for PCV2, as they are strongly positive for viral DNAs and viral nucleocapsid protein (3,4,15–19) in the early stages of the infection. Morphologic evidence for cytolytic PCV2 infection, either in vitro (32) or in vivo, has been difficult to obtain and most believe that the absence of viral cytopathic effects (cpe) is a characteristic of permissive PCV2 infection. If the primary permissive cell type for PCV2 replication is the immature histiocyte and if the infection occurs in the absence of viral cpe, infected histiocytes will continue to proliferate and differentiate. As well, virus-positive phagocytic mononuclear cells will ingest virion aggregates and cellular debris containing infectious virus, further increasing the amount of infectious virus contained within them. Thus, the most likely explanation for the high levels of infectious virus in PMWS-affected tissue organs is that the number of virus-permissive cells is increased as they differentiate into phagocytic cells.

Successful experimental reproduction of PMWS in PCV2-infected swine appears to be dependent upon cofactors for clinical disease expression. These include concurrent infections with other swine viral pathogens, such as PPV (9–11), PRRSV (12,13,33), and Mycoplasma hyopneumoniae (34). Experimentally, infection with these pathogens, immunosuppression (28), or immunizations with KLH/ICFA (15), are known to potentiate PCV2 infection and result in PMWS. In young conventional swine, vaccination with Actinobacillus pleuropneumoniae bacterins has the same potentiating effect (23). Importantly, several (20–22), but not all (24), field studies have suggested that artificial immunostimulation (20), as well as vaccination with a Mycoplasma hyopneumoniae bacterin combined into a mineral oil-based adjuvant, potentiate expression of disease in PCV2-infected swine (20–22). Of the identified cofactors, immunosuppression (28) is the most easily explained, as piglets treated with cyclosporine did not seroconvert to PCV2 and it is likely that inhibition of immune responses by cyclosporine facilitated disseminated infection (28). The event shared in common by all of the other potentiating factors appears to be activation of the immune system. During the inductive phases of an immune response, uncommitted lymphocytes and antigen-presenting histiocytes/macrophages receive proliferative and activation cytokine signals that are released during encounters with antigens. These cytokines act not only upon cells committed to antigen-specific proliferation but also upon adjacent “innocent bystander” cells. In fact, it has long been noted that manifestations of antiviral immunity are increased during convalescence from unrelated virus infections. Certainly, in those instances where infectious agents are cofactors, lymphoreticular hyperplasia during convalescence suggests that one mechanism of potentiation involves increasing the number of PCV2- permissive histiocytes or macrophages in lymphoreticular tissues. In this regard, the timing of the immunostimulating event(s) in relationship to the stage of PCV2 infection in swine is likely of great importance. Immunostimulation after viremia and when piglets have developed immunity or immunostimulation before PCV2 infection is unlikely to have the same effect as immunostimulation during active (viremic) PCV2 infection.

Adjuvanted bacterins have been used for infectious disease prophylaxis in swine and other domestic animal species for many years. Both clinical protection data and objective measures of immunity have determined that these bacterins promote increased antibody titers and T-cell immune responses. Delivery of immunogens is best accomplished by incorporation into an indigestible or poorly digestible adjuvant, as exemplified by mineral or plant oil-aqueous antigen emulsions (35,36), or aluminum salts (38). These preparations preferentially target antigen-presenting cells (APC). Adjuvants may also provide a slow-release form of antigen, thereby increasing opportunities for induction of immunity. Additional adjuvant effects may include up-regulation of MHC class II antigen on APC and stimulated secretion of a cytokines that promote T- and B-cell activation and proliferation (35,38). It is difficult to identify specific mechanisms of action for the adjuvanted bacterins evaluated in this study, as this information is not readily available in the literature. The bacterin component of these products is likely equivalent and M. hyopneumoniae membranes contained within the bacterins have in vitro mitogenic effects upon porcine lymphocytes (39,40). Regardless of any bacterin-associated stimulating effects, our data clearly demonstrate that enhanced PCV2 replication and thus potentiation of PCV2-associated disease occurred only in the bacterin (PAH) that contains a mineral oil-based adjuvant. In fact, there was no statistically significant difference between the adverse effects seen with vaccination using this product and the positive control (KLH/ICFA) group, even though fewer piglets given the PAH bacterin developed PMWS than did those given KLH/ICFA. In contrast, neither PMWS nor clinically evident signs of PCV2 infection were seen in any of the other PCV2-infected piglet groups vaccinated with FDAH-1, FDAH-1, BI, IV, and SPAH bacterins. The only histologic difference related to vaccination between these vaccination groups and piglets infected only with PCV2 was accumulation of vaccine material and associated focal granulomatous inflammation in some peripheral lymph nodes draining immunization sites in some piglets (data not shown).

In commercial swine production facilities, most piglets are exposed to PCV2 in the immediate postnatal period, but they do not become infected until after maternal antibody titers decline. Infection coincides with the management practice of segregated early weaning and with mixing of pigs from different litters in preparation for entry into either replacement stock or feed and finish programs. In Europe, it is a common practice to vaccinate swine with these bacterins at 1–2 wk of age; relatively few European hogs are raised under the North American type of intensive management system. In the latter, swine are vaccinated at 5–6 wk of age. Thus, the experimental design used here (early vaccinations of PCV2-viremic piglets), while not typical of US practice, represents a rigorous test of the safety of these bacterins regarding potentiation of PMWS. Under these conditions, the only vaccine product that was identified with heightened risk for PMWS, was that similar to the positive control piglet group that was immunized with KLH/ICFA, which contains a mineral oil-based adjuvant.

Finally, the most pressing issue is weighing the adverse consequences of delaying or not vaccinating swine at risk for the development of mycoplasmal (enzootic) pneumonia versus the risk of induction of PMWS in some of the vaccinates (41). Clearly, a recommendation to not vaccinate is neither responsible nor practical. Immunizations of gravid sows may provide temporary protection against enzootic pneumonia in suckling piglets by lactogenic immune mechanisms (42). In another field trial conducted on a farm with multiple infectious disease problems, including PMWS, investigators compared a single dose bacterin at 22 d of age with a 2-dose bacterin given at 7 and 22 d of age and demonstrated that the 1-dose vaccine adequately controlled mycoplasma pneumonia and reduced the prevalence of PMWS when compared with the 2-dose early-age vaccination regimen (43). Yet another option is to vaccinate pigs at 8–12 wk of age, after the expected highest prevalence of PCV2 viremia. These different approaches indicate that tailoring a vaccine prophylaxis program to farm-specific circumstances is warranted (41). Swine producers are thus faced with choices that are dictated by farm-specific circumstances. If a facility is experiencing substantial PCV2-related disease, one aspect of control may be to discontinue use of mineral oil-based products and, instead, employ other bacterins. Alternatively, it may be that most of the losses associated with respiratory disease can be prevented by vaccinating swine later in the production cycle, wherein vaccinations are separated by several weeks or more from the probable age for acquisition of PCV2 infection. Our data suggest that, whatever the dynamics of the specific circumstances, mineral oil-adjuvanted bacterins carry with them heightened potential for induction of PMWS, whereas the other adjuvanted bacterins tested in this study appear to have minimal or no potentiating effects upon the subsequent expression of PMWS in PCV2-infected vaccinates.