Measurement of isotype-specific antibody responses to Aujeszky's disease virus in sera and mucosal secretions of pigs

Veterinary Immunology and lmmunopathology, 31 (1992) 95-113

95

Elsevier Science Publishers B.V., Amsterdam

Measurement of isotype-specific antibody responses to Aujeszky's disease virus in sera and mucosal secretions of pigs T.G. Kimman, R.A.M. Brouwers, F.J. Daus, J.T. van Oirschot and D. van

Zaane Central VeterinaryInstitute, Department of Virology, Lelystad, Netherlands (Accepted 28 February 1991 )

ABSTRACT Kimman, T.G., Brouwers, R.A.M., Daus, F.J., van Oirschot, J.T. and van Zaane, D., 1992. Measurement of isotype-specific antibody responses to Aujeszky's disease virus in sera and mucosal secretions of pigs. Vet. Immunol. Immunopathol., 31:95-113. Enzyme-linked immunosorbent assays (ELISAs) for the detection of porcine IgM, IgA, IgG 1 and IgG2 antibodies directed against Aujeszky's disease vires (ADV) are described. ADV-specific IgA and IgM were detected in an antibody capture assay, and ADV-specific IgG 1 and IgG2 were detected in an indirect double antibody sandwich assay. A selected set of samples was tested in the four ELISAs and in a 24 h virus neutralization assay. Comparison of the results showed that the ELISAs were isotype-specific, sensitive, and reproducible. Samples with ADV antibody of one isotype showed that ADV-specific IgGl, IgG2 and IgM were able to neutralize the virus in vitro. In vitro neutralization of virus can be enhanced by complement. ADV-specific IgA neutralized virus only weakly. ADV-infected cells activated complement in the absence of antibody. Specific IgG2 and IgM enhanced complement activation. Analysis of the time course of antibody responses after infection or vaccination revealed that the isotype-specific ELISAs are suitable to study the humoral antibody response of pigs to the virus in mucosal secretions. Wild-type virus (strain NIA-3) and an attenuated vaccine strain (Bartha) administered intranasaUy induced mucosal IgM and IgA responses to the virus. In contrast, a killed vaccine (Nobivac) administered intramuscularlyinduced only weak mucosal IgM responses. The attenuated vaccine strain primed for a mucosal IgA memory response evoked upon challenge infection with wild-type virus.

ABBREVIATIONS ACA, antibody capture assay; ADV, Aujeszky's disease virus; C3, complement component C3; ELISA, enzyme-linked immunosorbent assay; EMEM, Eagle's minimal essential medium; IDAS, indirect double antibody sandwich assay; Ig, immunoglobulin; MAb, monoclonal antibody; PBS, phosphate-buffered saline; PFU, plaque-forming unit; PI, post infection; PV, post vaccination; SPF, specific-pathogen-free. INTRODUCTION

Aujeszky's disease is an important disease in pigs. Vaccination against this disease is widely practiced in countries with an intensive pig husbandry. Most © 1992 Elsevier Science Publishers B.V. All fights reserved 0165-2427/92/$05.00

96

T.G. KIMMAN ET AL.

vaccines offer clinical protection against Aujeszky's disease, but they usually do not prevent virus excretion or the establishment of latency after challenge infection. Both the mechanisms and viral antigens responsible for protective immunity against Aujeszky's disease virus (ADV; syn. pseudorabies virus, herpesvirus suis 1 ) are poorly understood. Thus there is no rational basis for the development of better vaccines. Antibodies in serum seem to afford a certain degree of protection, but because the correlation between antibodies in serum and protection is poor (Andries et al., 1978; De Leeuw et al., 1982; Zuffa et al., 1982; Martin et al., 1986 ), serum antibodies do not appear to be of major importance in protection. Because the virus enters the pig via mucosal surfaces in the nose and throat, local defence mechanisms may be more protective. With certain vaccines, protection after intranasal administration is indeed better than after parenteral administration, especially in pigs with maternal antibodies against the virus (De Leeuw and Van Oirschot, 1985 ). In this report, we describe the development of methods to study the humoral immune response against ADV in sera and mucosal secretions of pigs, we present data on the biological functions of different antibody isotypes against the virus, and we present data on the time course of the humoral immune response in serum and at mucosae after vaccination and infection. MATERIALS AND METHODS

Monoclonal antibodies Monoclonal antibodies (MAbs) specific for swine IgG1, IgG2, IgA and IgM were produced and characterized as described (Van Zaane and Hulst, 1987 ). The following MAbs were selected from a large panel: 23.49.2 (anti-porcine IgG1 ), 34.2.1a (anti-porcine IgG2), 27.9.1 (anti-porcine IgA), and 28.4.1 (anti-porcine IgM ). MAbs directed against ADV were selected from a panel prepared as described earlier (Lukacs et al., 1985; Van Oirschot et al., 1988). In the IgG1 and IgG2 enzyme-linked immunosorbent assays (ELISAs), Mab 45.11.1a (mouse isotype IgG 1 ) was used as coating antibody. In the IgA and IgM ELISAs, MAb 45.24.1 a (mouse isotype IgG2a) was used as conjugated antibody. MAbs 45.11. la and 45.24. la were both directed against the glycoprotein glI of ADV as determined by radio-immunoprecipitation (Lukacs et al., 1985; courtesy of Dr. H.J. Rziha). The MAbs were purified from mouse ascites fluid by 50% ammonium sulphate precipitation, followed by dialysis overnight against phosphate-buffered saline (PBS, pH 7.4). MAbs were conjugated with horseradish peroxidase as described (Wilson and Nakane, 1978 ).

PORCINE ANTIBODY RESPONSES TO AUJESZKY'S DISEASE VIRUS

97

AD V antigen preparation Confluent monolayers of primary or secondary porcine kidney cells were grown in Eagle's minimal essential medium (EMEM) supplemented with 10% fetal calf serum and antibiotics. Cultures were inoculated with a stock preparation of ADV (strain NIA-3) at a multiplicity of infection of 10. When monolayers showed extensive cytopathic effects (usually 2 days after inoculation), cells and medium were frozen ( - 70°C) and thawed. The cellular lysate was clarified (30 min; 2000 ×g) and the supernatant was precipitated with 50% ammonium sulphate. The pellet was suspended in distilled water and dialyzed overnight against ELISA-buffer (PBS, pH 7.4, with sodium chloride added to a final concentration of 0.5 M and 0.5% (w/v) Tween 80). The dialysate was used as antigen and was stored in small aliquots at -20°C. Control antigen was similarly prepared from uninfected monolayers.

Isotype-specific ELISAs Two types of assays were selected: an indirect double antibody sandwich assay (IDAS) for the detection of ADV-specific IgG 1 and IgG2, and an antibody capture assay (ACA) for the detection of ADV-specific IgA and IgM. For the IDAS, micro-ELISA plates (M 129B, Dynatech, Greiner B.V., Alphen a/d Rijn, Netherlands) were coated with MAb directed against ADV in 50 mM NaHCO3 (pH 9.6) by incubation overnight at 37°C. Subsequent steps were incubation with antigen, test sample, enzyme-labelled MAb directed against IgG1 or IgG2, and substrate solution. For the ACA, plates were coated with MAb directed against IgA or IgM in PBS (pH 7.4) overnight at 37°C. Plates were then incubated with the test sample, antigen, enzyme-labelledMAb directed against ADV, and substrate solution. Antigen preparations, test sampies, and conjugates were diluted in ELISA buffer. Optimal concentrations of reagents were determined by checkerboard titration. Each reagent was added at a volume of 100 ~tl per well. Sample dilutions and conjugate solutions contained 1% ovalbumin (Sigma Chemical Co., St. Louis, MO) to reduce background staining. Serial twofold dilutions of samples were made in the plate starting at 1: 20 for sera and 1 : 5 for mucosal secretions. In each plate a standard positive and negative serum were included. The standard negative serum was obtained from a specific-pathogen-free (SPF) pig free of ADV antibodies. Standard positive sera were chosen during the initial phase of the development of the assays because of their high titre and high maximum E450 nm absorbance value. The serum samples were collected from pigs vaccinated against ADV, infected with ADV, or both (see below). Each sample was also tested in the lowest dilution with control antigen to check for non-specific binding. After each incubation step, plates were washed 10 times with 0.05% (w/v) Tween 80 in tap water. Antigen preparations, test samples, and conjugates

98

T.G. KIMMAN ETAL.

were incubated at 37 °C for 1 h. Each ELISA procedure was completed by the addition of the substrate solution, which consisted of 0.1 mg m l - 1 3,3' 5,5'tetramethylbenzidine (Sigma) in 0.1 M sodium acetate buffer (pH 6.0) and 0.005% freshly added H202. Color development was stopped after 10 min by adding 100/zl of 2 M H2SO 4. Absorbance was measured with a Titertek-multiskan spectrophotometer using a 450 n m filter. The titer of a sample was expressed as l°log of the reciprocal of the highest dilution yielding an E450 nm value/> 0.2. Titers were corrected with the dilution factor (see below).

ELISA for detection of antibodies directed against gI A blocking ELISA that detects antibodies directed against one or two epitopes of the glycoprotein gI of ADV was performed as described (Van Oirschot et al., 1988).

AD V neutralization test A 24 h virus neutralization (VN) test was performed as described (Bitsch and Eskildsen, 1976; De Leeuw et al., 1982). Titers are expressed as 1°log of the reciprocal of the highest final serum dilution inhibiting the viral cytopathic effect in 50% of the cell cultures. C o m p l e m e n t in the samples was inactivated at 56 °C for 30 min before use. Complement-dependent neutralization was tested by adding complement (using heat-inactivated complement as control) to a final concentration of 10% 30 min before the incubation of the secondary porcine kidney cells. Commercial rabbit complement was purchased (Sera-lab, Sanbio B.V., Uden, Netherlands). Porcine complement was obtained from a sow from the institute's SPF unit. This sow had no antibodies to ADV. Blood from the sow was collected, allowed to clot for 1 h at room temperature and then centrifuged. The resulting serum was stored in 1 ml portions at - 70 ° C.

Complement-mediated cell lysis Complement-mediated cell lysis was examined in a 51Cr release assay. Monolayers of secondary porcine kidney cells in microtitre plates (Greiner) were simultaneously infected with ADV strain NIA-3 and labelled with 10 /~Ci 51Cr for 2 h. The monolayers were then washed four times with cell culture medium. Seven hours after the start of the infection, the monolayers were washed again and incubated with 50/tl of a heat-inactivated ( 30 min, 56 ° C) ADV antibody source (i.e. i m m u n e serum, negative serum, buffer, or affinity-purified antibody directed against ADV of one isotype). After further incubation for 1 h, 50 pl rabbit complement was added to a final concentration of 10% complement. Control release was determfned in wells incubated with

PORCINE ANTIBODY RESPONSES TO AUJESZKY'S DISEASE VIRUS

99

heat-inactivated complement. Maximum release was determined by adding 10% Triton X-100. After a 2 h incubation, medium was collected and 51Cr release was determined in a gamma scintillation counter. All incubations were done at 37 °C in a humidified CO2 incubator. Each variable was tested in five replicates. The percentage of complement-mediated release was calculated as: Mean sample release- mean control release Mean maximum release-mean control release × 100%

Affinity purification ofAD V antibody isotypes To examine the biological functions of antibody isotypes directed against ADV, serum and mucosal samples with high levels of ADV antibodies of a particular isotype were selected and pooled. These samples were collected from pigs infected with ADV strain NIA-3 or vaccinated intranasally with ADV strain Bartha. The samples were deprived of ADV antibodies of other isotypes by precipitation with 50% saturated ammonium sulphate and batchwise immunoaffinity chromotography. For that purpose, MAbs directed against IgG1 (23.49.2), IgG2 (34.1.1a), or IgM (28.4.1) were coupled to activated Sepharose 4B according to the manufacturer's manual (Pharmacia, Woerden, Netherlands).

Experimental animals Dutch Landrace pigs were from the SPF herd of the Central Veterinary Institute. The pigs were born to unvaccinated sows. Before the start of the experiments, sows and pigs had no antibodies directed against ADV. During the experiments, each group was housed in a separate isolation room.

Vaccination and infection At 10 weeks of age, three pigs were vaccinated once intranasally with 106 plaque-forming units (PFU) of the live-attenuated ADV vaccine strain Bartha K-61 (Bartha, 1961 ). Three pigs were vaccinated once intramuscularly with the inactivated ADV vaccine Nobivac (Intervet, Boxmeer, Netherlands). Five pigs were left unvaccinated. Three months later all pigs were challenged intranasally with l05 PFU of the virulent ADV strain NIA-3 (McFerran and Dow, 1975). Clinical signs of Aujeszky's disease, rectal temperatures, and body weights were recorded beginning several days before challenge to 10 days after challenge. Virus excretion after challenge was determined by titration of saliva samples on secondary porcine kidney cells as described (De Leeuw et al., 1982).

100

T.G. KIMMAN ETAL.

Sampling procedures Blood, saliva, tears, feces, and nasal and lung secretions were collected 1 day before vaccination and before challenge, three times a week for 2 weeks after vaccination and challenge, and thereafter once a week. Saliva samples were further collected daily during 10 days after challenge. Saliva samples were collected using swabs which were extracted with 4 ml of tissue culture m e d i u m (EMEM) supplemented with 4% fetal calf serum. The weight of the saliva collected was measured after centrifugation of the swab in a special container. Tears were collected with cotton tips (Medical Wire and Equipment Co. (Bath) Ltd, Corsham, U K ) which were extracted in 1 ml of ELISA buffer supplemented with 1% fetal calf serum. Fetal calf serum was added to reduce proteolysis. Approximately 1 g of faeces was collected and mixed with 4 ml of ELISA buffer containing 10% fetal calf serum. Before being tested, the feces suspension was clarified by centrifugation. Lung and nasal secretions were collected while pigs were anaesthetized via inhalation of N20 and Halothane. A 10 ml syringe with rubber adaptor to fit the nostril was used to flush one nostril twice with 7 ml of PBS containing 1% methylene blue (E680 n m = 1.0). Under these conditions no washing fluid was lost in the oropharyngeal cavity. Lung wash fluid was collected as described by Van Leengoed and K a m p (1989). Briefly, the pigs were held in a horizontal position on their abdomens with their heads slightly elevated. A laryngoscope was used to introduce a catheter (outer diameter 2.3 m m ) into the main bronchus until it stuck. Twenty milliliters of PBS with methylene blue was introduced into the lung and sucked out 1 min later. Dilution factors of the nasal and lung secretions were calculated from the E680 nm values of the wash fluids before and after sampling, using the formula: Dilution factor -

E680 n m in wash fluid E680 n m in introduced methylene blue solution

The dilution factor (~°log) was 0.4-1.3 for saliva, 0.2-1.3 for nasal washings and 0.04-0.69 for lung washings. Approximately 75% of the samples had dilution factors differing not more than 0.4. The dilution factor for tears could not be determined reliably, but is at least 1.0. Samples were stored at - 7 0 ° C. RESULTS

Development ofAD V and isotype-specificELISAs Non-specific adherence was observed when semi-purified ADV antigen was coated to microtiter plates or when clarified ADV antigen from freeze-thawed infected cells was immuno-adsorbed on microtiter plates coated with poly-

101

PORCINE ANTIBODY RESPONSES TO AUJESZKY'S DISEASEVIRUS TABLE 1 Characteristics o f isotype-specific MAbs MAb

23.49.2 34.2.1 a 28.4.1 27.9.1

Mouse isotype IgG1 IgG 1 IgG 1 IgG 1

Titer I against porcine IgG 1

IgG2

IgM

IgA

6 3 -

6 -

5 -

2 (4)2 6

Specificity

Epitope

IgGl IgG2 IgM IgA

A B A B

ITitre in an indirect ELISA against purified porcine Ig isotypes IgG 1, IgG2, IgM and IgA, expressed as I°1og (from Van Zaane and Hulst, 1987). 2Shallow titration curve with low E450 n m at high MAb concentration.

clonal rabbit antisera directed against ADV (results not shown). These problems were solved by using both MAbs against ADV and MAbs against porcine immunoglobulin isotypes. Two types of assays were finally developed. An IDAS assay appeared optimal to detect ADV-specific IgG1 and IgG2, and an ACA appeared optimal to detect ADV-specific IgA and IgM. Both assays limit the competition between isotypes (see Discussion). From a panel of 23 MAbs directed against ADV, one coating MAb for the IDAS and one conjugated MAb for the ACA were selected. These two MAbs were both directed against glycoprotein glI of ADV. MAbs directed against porcine immunoglobulins were selected from a large panel. The characteristics of the selected isotypespecific MAbs are summarized in Table 1.

Specificity, sensitivity, and reproducibility of the isotype-specific ELISAs The isotype-specificity of the selected MAbs against porcine immunoglobulins is shown in Table 1. Each MAb reacted only with its homologous isotype or, in case of MAb 34.2. la, with a titer difference of at least 3 1°1ogunits also with heterologous isotypes. The isotype-specificity of the IDAS and ACA ELISAs is shown in Table 2. Selected samples with high titers in one isotype-specific ELISA were negative in the assays for other isotypes. Samples with more than one isotype did not react uniformly in the assays. The ELISAs appeared to be more sensitive than the neutralization assay, because titers were higher in the ELISAs than in the neutralization assay (Table 3). Reproducibility of the assays was estimated by including reference sera on every plate in each assay. When tested on the same day on different plates, or on different days, these sera gave exactly the same titre or a titer difference of only a single step in 96.8% of the samples (n = 248 ).

102 TABLE

T.G. K I M M A N ET AL. 2

Isotype-specificity

of the ELISAs

from SPF pigs vaccinated

for the detection

of antibodies

to ADV. All samples

against ADV and then infected with ADV strain NIA-Y

with asterisks were enriched Materials and Methods) Sample

for antibodies

of a particular

IgG1

IgG2

were collected

Samples

indicated

isotype by affinity chromatography

IgM

(see

IgA

1 Serum*

1.61

_2

_

2 Serum

2.3

-

4.0

_ -

3 Serum

4.6

4.2

-

-

4 Serum*

-

3.1

-

-

5 Serum

-

-

4.5

-

6 Lung fluid

-

-

>14.6

-

7 Serum*

-

-

3.5

8 Tears

-

-

-

3.4

9 Tears

-

-

-

3.6

10 S a l i v a *

-

-

-

2.2

~Titer in isotype-specific 2Negative

ELISA expressed

( i . e . < 1.3 i n s e r u m

and <0.7

a s J°log.

in mucosal

samples).

TABLE3 Comparison

of sensitivity of isotype-specific

Sample

VN test

ELISAs with a 24 h neutralization IgG 1

IgG2

assay IgM

IgA

1 Serum

2.7

3.7

3.4

-

-

2 Serum 3 S e r u m *~

3.7 0.9

4.2 1.6

3.5 -

>/4.6 -

-

4 Serum

3.6

4.0

4.0

3.2

3.2

5 Serum*

3.0

-

3.1.

-

-

6 Serum

0.8

-

-

2.6

-

7 Serum*

1.8

-

-

3.5

-

8 Tears

0.9

-

-

2.6

-

9 Lung wash

0.6

-

-

-

1.2

10 L u n g w a s h

1.7

1.5

-

-

3.6

11 L u n g w a s h

0.5

-

-

-

2.1

12 T e a r s

0.6

-

-

-

2.1

13 T e a r s

0.9

-

-

-

2.9

14 S a l i v a *

0.6

-

-

-

2.2

~See T a b l e 2 f o r d e t a i l s .

Functional and gI binding activity of A D V antibody isotypes (Table 4) Selected serum and mucosal samples with high levels of A D V antibodies of a particular isotype were deprived of A D V antibodies of other isotypes by affinity chromatography. (This approach was more successful than attempts

103

PORCINE ANTIBODY RESPONSESTO AUJESZKY'SDISEASEVIRUS TABLE 4

Functional and gI-binding activity of ADV antibody isotypes. Samples 4-7, l0 and I l are pooled samples from pigs infected with ADV strain NIA-3 (gI positive). Samples 8 and 9 are pooled samples from pigs vaccinated with ADV strain Bartha (gI negative). Samples 4-11 were deprived of other isotypes by affinity chromatography (see Materials and Methods ) Sample Isotype

1 2 3 4 5 6 7 8 9 10

11

No serum Ned. serum Pos. serum IgG1 IgG2 IgA IdA IgM IgM IgM IgM

Isotypespecific

Virus neutralization test

ELISA

Titre Increase Increase by rabbit by porcine complement complement

1.65 3.1 1.0 2.2 1.8 1.8 2.9 3.5

-t 2.13 0.9 3.0 0.3 0.6 0.9 0.9 1.5 1.8

1.35 1,2 1.2 >t2.85 2.1 >/2.25 /> 1.95

0.45 0.6 0.6 0.3 0.9 1.8 0.6 0.6

gI-blocking Complement-mediated ELISA lysis: percentage 5~Cr release

>t0.74 0,6 2.5

-

-

Infected cells

Uninfected cells

34.82 22.4 46.0 20.1 36.6 ND 6 25.6 ND 39.9 49.9 ND

14.2 8.1 5.8 9,0 0.3 ND 5.0 ND 18.9 -2.5 ND

INegative. 2Results represent the mean of two to three separate experiments (each sample was tested in five replicates ). Each sample was tested at a 1 : 10 dilution. 3Titer in virus neutralization assay. 4Titer in gI-specific blocking ELISA. STiter in homologous isotype-specific ELISA.

6ND, not determined.

to isolate immunoglobulin isotypes.) IgG1, IgG2 and IgM ADV-specific antibodies strongly neutralized. In contrast, IgA neutralized virus only weakly. This observation was confirmed by testing sequentially collected saliva samples of two Bartha-vaccinated pigs that showed a memory IgA response upon challenge with ADV strain NIA-3 (see below). Neutralizing antibodies were either not detected or at low levels only (not shown). Rabbit complement increased IgG 1- and IgG2-mediated neutralization approximately l 0-fold and increased IgM-mediated neutralization approximately 100-fold (Table 4). Porcine complement increased IgG l- and IgG2-, and IgM-mediated neutralization approximately four-fold. It was difficult to interpret the influence of antibody isotypes on complement-mediated cytotoxicity, because ADV-infected cells activate more complement than uninfected cells (or because ADV-infected cells are more susceptible to complement-mediated cytotoxicity than uninfected cells). Nonetheless, both IgG2 and IgM appear to enhance complement-mediated cytotoxicity. Binding of glycoprotein gI of ADV was only observed in samples contain-

104

T.G. KIMMANETAL.

ing lgG1 or IgG2. However, the samples containing IgA and IgM originated from pigs that were infected with the gI-expressing NIA-3 strain, and these pigs all developed a serum antibody response to gI.

Time course of isotype-specific antibody responses after infection and vaccination Mucosal antibody responses in individual pigs were qualitatively approximately the same in all secretions, but they were stronger and occurred more regularly in saliva and tears than in lung or nasal wash fluid. Antibodies were not detected in feces. The responses in serum, saliva, and tears after vaccination, infection, or both are summarized in Tables 5-9. The responses of one pig, vaccinated intranasally with ADV strain Bartha and subsequently challenged with ADV strain NIA-3, are shown in Fig. 1 as a representative example. Table 5 shows antibody responses after infection of unvaccinated pigs. ADVspecific IgM appeared at about 6-8 days post infection (PI). IgA appeared TABLE 5

Primary antibody responses against ADV after intranasal infection of unvaccinated pigs (n = 5 ) Specimen and antibody Serum IgG 1

lgG2 IgM IgA ~ Saliva IgM

IgA2

Tears IgM IgA3

Mean day PI of first detection (range)

Mean day PI of reaching peak titer (range)

Mean no. of days detectable (range)

Mean peak titer (range)

10.2 (8-13) 15 (10-21) 7.6 (6-8) (10)

19.4 (13-27) 21.8 (13-34) 10.2 (8-15) (13)

Persistent

3.3 (2.5-4.5) 2.8 (1.8-4.0) 4.3 (3.7-6.2) (2.7)

6.8 (6-7) 14.5 (9-20)

8.2 (7-10) 14.5 (9-20)

8 (8-8) 9-7 (8-13)

11.6 (8-15) 11 (10-13)

1Excluding four non-responding pigs. -'Excluding three non-responding pigs. 3Excluding two non-responding pigs.

Persistent

(13-~90) (~90)

( 8 - ~ 21 ) (1->/27)

(8->_-90) ( 1-/> 90)

2.1 (1.0-4.0) 1.2 (0.7-1.7)

2.2 . (1.2-3.4) 1.8 (1.2-2.4)

105

PORCINE ANTIBODY RESPONSES TO AUJESZKY'S DISEASE VIRUS

titre (1°log)

titre (t°fog)

600

G.---E) IgG1 ¢ ¢lgG2 [3-,.-E] IgM

5 00

Serum

Saliva

4.00

C

+

3.00 2.00 1.00 0.00

6.00

Tears

Lung w a s n

Nasal wash

Faeces

5,00 4O0

0,00 600

-

500

-

400 3.00 - + 2.00

-~/~

|

1.00 000 ~

•

•

•

m

i clays post vacCination

i

i

i

i

I 192

days post vaccination

Fig. 1. Time course of antibody activity against ADV in serum, tears, nasal wash fluid, saliva, lung wash fluid, and feces of one pig after intranasal vaccination (V) with the attenuated ADV strain Bartha and intranasal challenge infection (C) with the wild-type ADV strain NIA-3.

somewhat later at about 8-10 days PI, but not uniformly in all pigs. IgG 1 and IgG2 responses were detected almost exclusively in serum from 8 and 10 days PI onward, respectively. IgM and IgA were detected in the various samples for strongly variable periods. IgG l and IgG2 responses persisted. Table 6 and Fig. 1 show antibody responses after intranasal vaccination with live attenuated ADV strain Bartha. These responses strongly resembled

106

T.G, KIMMAN ET AL.

TABLE 6 Primary antibody responses against ADV after intranasal vaccination with ADV strain Bartha ( n = 3 ) Specimen and antibody

Mean day PV of first detection (range)

Mean day PV of reaching peak titer (range)

Mean no. of days detectable (range)

Mean peak titer (range)

12 (10-13) 15.3 (13-20) 8 (8-8) 13 (13-13)

29.3 (20-48) 46 (42-48) 15.3 (13-20) 27 (27-27)

Persistent

3.4 (3.2-3.5) 3.1 (3.0-3.2) 4.4 (4.2-4.5) 2.9 (2.7-3.0)

7 (7-7) 11.5 (10-13)

9.3 (9-10) 30.5 (13-48)

18.7 (14-21)

8 (8-8) 12 (10-13)

12 (10-13) 20 (13-34)

54.3 (35-87)

Serum

IgG1 IgG2 IgM IgAI

Saliva IgM

lee' Tears IgM

IgA

Persistent 22.3 (20-27) 17.5 (15-20)

(50->t90)

(50-990)

2.5 (2.4-2.6) 1.6 ( 1.0-2.1 )

2.8 (2.4-31) 1.2 (1.0-1.5)

~Excluding one non-responding pig.

antibody responses after infection with the wild-type strain NIA-3, but they started somewhat later and reached their peak values somewhat later. In tears and saliva, IgA was usually detected for long periods. Intramuscular vaccination with the killed vaccine Nobivac (Table 7) induced serum IgM, IgG1, and IgG2 responses. The IgG1 and IgG2 responses started and peaked somewhat later than after infection, but reached similar titers. This route of vaccination also induced mucosal IgM, but no mucosal IgA. Table 8 and Fig. 1 show antibody responses after infection of Bartha-vaccinated pigs. Strong and rapid serum IgG1, IgG2, and IgA responses were observed. IgGl and IgG2 persisted at a high titre, whereas the titer of the IgA antibodies fell gradually to a lower level. In secretions, strong and rapid IgA responses were invariably detected. IgM responses were also detected in secretions and serum, but usually for short periods. Table 9 shows antibody responses after infection of Nobivac-vaccinated pigs. Strong and rapid IgG 1 and IgG2 responses were detected similar to those detected after infection of Bartha-vaccinated animals. Serum and mucosal

PORCINE ANTIBODY RESPONSES TO AUJESZKY'S DISEASEVIRUS

107

TABLE 7 Primary antibody responses against ADV after intramuscular vaccination with killed vaccine Nobivac ( n = 3 ) Specimen and antibody ~

Mean day PV of first detection (range)

Mean day PV of reaching peak titer (range)

Mean no. of days detectable (range)

Mean peak titer (range)

15 (11-20) 16 (14-20) 7 (7-7)

65.7 (47-75) 56.3 (33-75) 10.3 (9-11 )

Persistent

14 (14-14)

3.1 (3.0-3.2) 3.2 (3.0-3.4) 3.7 (3.7-3.7)

7 (7-7)

11 ( 11-11 )

8 (8-8)

1.4 (1.0-1.7)

7 (7-7)

12.5 (11-14)

20.5 (14-27)

1.4 (1.4-1.4)

Serum

lgG 1 IgG2 IgM

Persistent

Saliva

IgM Tears

IgM2

INo IgA responses were detected. 2Excluding one non-responding pig. TABLE 8 Secondary antibody responses against ADV after intranasal challenge infection of Bartha-vaccinated pigs ( n = 3 ) Specimen and antibody

Mean day PI of first detection (range)

Mean day PI of reaching peak titer (range)

Mean no. of days detectable (range)

Mean peak titer (range)

9.7 (8-13) 8 (8-8) 10 (10-10) 6.3 (3-8)

15.7 (13-17) 15 (13-17) 13 (13-13) 16.3 (15-17)

Persistent

4.6 (4.5-4.6) 4.3 (4.0-4.5) 3.2 (3.2-3.2) 3.6 (3.0-4.2)

8.3 (7-9) 7 (7-7)

9.7 (9-10) 13.3 (7-20)

8 (5-14)

10.3 (8-13) 10.3 (8-13)

13.7 (8-20) 13.3 (7-20)

Serum

IgG1 IgG2 IgM 1 IgA

Persistent 8.5 (6-11 ) (20->-92)

Saliva

IgM IgA

(72->_-92)

1.5 (1.2-1.7) 2.4 (2.1-2.6)

Tears

IgM IgA

1Excluding one non-responding pig.

19 (5-27) (66->.92)

1.3 (0.7-2.0) 3.1 (2.1-3.6)

108

T.G. KIMMANET AL.

TABLE 9 Secondary antibody responses against ADV after intranasal challenge infection of Nobivac-vaccinated pigs ( n = 3 ) Specimen and antibody~ Serum IgG 1

IgG2 IgA2

Saliva IgA 2

Mean day PI of first detection (range)

Mean day PI of reaching peak titer (range)

Mean no. of days detectable (range)

Mean peak titer (range)

8.7 (8-10) 8 (8-8) 9 (8-10)

22.3 (13-27) 15.3 (13-20) 10 ( 10-10)

Persistent

(4->/27)

4.3 (4.2-4.4) 4.4 (4.0-4.6) 2.8 (1.8-3.7)

20 (20-20)

20 (20-20)

I ( 1-1 )

0.9 (0.7-1.0)

14.3 (10-20)

15.3 (13-20)

(>t27)

1.4 (1.2-1.9)

Persistent

Tears

IgA

~No IgM responses were detected. 2Excluding one non-responding pig. TABLE 10 Clinical and virological findings after challenge infection with ADV strain NIA-3 Group

No.

Fever ~ (days)

Growth retardation (days)

Virus excretion (days)

Unvaccinated

52

Barthavaccinated 3 Nobivacvaccinate&

3

6 (5-9) 0.3 (0-1 ) 5.7 ( 3-8 )

14.8 (9-18) 1.7 (0-5) ND

5.4 (4-8) 2.7 (0-5) 0.7 (0-1 )

3

Results represent mean (range) no. of days. Rectal temperature > 40 ° C. 2A sixth pig died at Day 6 PI. Results of this pig are not shown. 3Vaccinated once intranasally with 105 PFU of the Bartha strain K-61. 4Vaccinated once intramuscularly with the inactivated vaccine Nobivac.

IgA responses were also detected, but these developed slower and reached lower peak titers than after infection o f Bartha-vaccinated pigs. IgM responses were not detected.

PORCINE ANTIBODY RESPONSES TO AUJESZKY'SDISEASEVIRUS

109

Clinical and virological observations after infection (Table I0) Non-vaccinated pigs developed typical signs of Aujeszky's disease after infection, i.e. fever, loss of appetite, and dullness from Day 2 PI onwards; some pigs vomited. They sneezed, developed a mucopurulent discharge and some had neurological signs, such as tremors and ataxia. One pig died at Day 6 PI (results of this pig are not shown). The other pigs started to recover from Day 8 PI onwards. The vaccinated pigs developed only mild signs of disease after infection including slight dullness and reduced appetite for 2-4 days. These pigs excreted virus for shorter periods than the unvaccinated pigs. DISCUSSION

This report describes ELISAs for the detection of porcine IgM, IgA, IgGl, and IgG2 antibodies directed against ADV. MAbs were used as isotype detecting or catching antibody, because conventional antisera, notably against IgM, may cause non-specific binding (Rodak et al., 1987). In an indirect ELISA with purified porcine Ig isotypes that were coated to microtiter plates, the isotype-specific MAbs reacted only with their homologous isotype, or with a titer difference of at least 3 l°log units also with heterologous isotypes (Mab 34.2.1 a). The isotype specificity of the final tests was confirmed by the observation that samples with high levels of ADV-specific antibodies of a particular isotype reacted only in the homologous assay, and because samples that contained antibodies of more than one isotype did not show a uniform pattern of reactivity in the other ELISAs. The IDAS appeared optimal for detecting ADV-specific IgG1 and IgG2. The ACA appeared optimal for detecting ADV-specific IgM and IgA. Similar experiences were obtained in isotype-specific assays for bovine rotavirus (Van Zaane and IJzerman, 1984 ) and bovine respiratory syncytial virus (Kimman et al., 1987a). The IDAS is probably optimal for detecting specific IgG 1 and IgG2 because it circumvents competition between antigen-specific and nonspecific IgG1/IgG2 (intra-isotype competition). In contrast, the ACA is probably optimal for detecting specific IgM and IgA, because it circumvents competition between isotypes (inter-isotype competition). However, because in different secretions total IgA concentrations may vary, as well as the extent to which the IgA is extracted, the assays should be interpreted with caution. Another advantage of the ACA for the detection of specific IgM is the absence of false-positive reactions caused by rheumatoid factors (Chantier and Diment, 1981; Butler et al., 1990; James, 1990). The MAbs used to detect or to catch the ADV antigen were directed against the glycoprotein gII of the virus. The ELISAs therefore cannot discriminate between pigs infected with wild-type strains of ADV and pigs vaccinated with

110

T.G. KIMMAN ETAL.

vaccine strains with deletions in either gI, gIII, or gX (Kit et al., 1987; Marchioli et al., 1987; Moormann et al., 1987). MAbs against porcine Ig isotypes appeared also suited for affinity chromatography. In this way, samples with ADV antibody of one isotype were prepared. Results showed that porcine IgGl, IgG2, and IgM can neutralize ADV and that this neutralization can be enhanced by complement. These results do not agree with those of Rodak et al. (1987), who detected neutralizing activity (in a 1 h neutralization assay) in the IgG, but not in the IgM class. IgA had only weak neutralizing activity. This observation agrees with results of other studies in other viral-host systems (McIntosh et al., 1978; Kimman et al., 1987a). The low or non-neutralizing nature of IgA may be caused by differences in sensitivity of the assays or by low avidity of IgA. Although some conflicting results have been reported (Hiemstra et al., 1987), our study and most other studies have demonstrated that IgA does not apparently activate complement. IgM and IgG2 appeared to enhance complement-mediated cytotoxicity, but IgG1 (probably because of low concentration) and IgA did not. It was difficult to interpret the complement-mediated cytotoxicity assay, because ADV-infected cells appeared to activate more complement (in the absence of antibody) than uninfected cells. Numerous virus-infected cells can activate complement in the absence of antibody, either via the alternative or classical pathway (Joseph et al., 1975; Perrin et al., 1976; Kimman et al., 1989), but this has not been reported for ADV-infected cells. This observation should be confirmed by detecting complement component C3 on the surface of ADV-infected cells. Alternatively, ADV-infected cells may be more susceptible to complement-mediated cytotoxicity than uninfected cells. In contrast to herpes simplex virus type 1-infected cells, ADV-infected cells reportedly lack a C3b receptor, which can inhibit complement activation (Friedman et al., 1986). Our observations raise questions about the biological function of IgA in vivo. IgA apparently neutralizes virus in vitro only weakly. Previous studies have shown that IgA cannot mediate antibody-dependent cell-mediated cytotoxicity in human, murine, and porcine systems (Martin and Wardley, 1987 ). IgA might be important in vivo to trap virus in secretions. The mucosal antibody responses in individual pigs after infection or vaccination were highly similar in all mucosal secretions, although they were stronger in saliva and tears. Thus, the mucosal immune system in the eye and in the upper respiratory and upper alimentary tract of pigs apparently reacts as a whole. Responses were not detected in feces. However, after a respiratory infection of young calves with bovine respiratory syncytial virus, IgA was also detected in feces (Kimman et al., 1987b). IgA is probably proteolytically digested in the alimentary tract (T.G. Kimman, unpublished observations, 1989). The ELISAs developed are suitable to study the mucosal and serum immune responses against ADV in pigs. The mucosal antibody responses of pigs

PORCINE ANTIBODY RESPONSES TO AUJESZKY'S DISEASE VIRUS

111

infected with the wild-type ADV strain NIA-3 or vaccinated intranasaUy with the attenuated strain Bartha closely resembled each other. Both IgM and IgA were detected in mucosal secretions, IgA was usually detected for longer periods. In contrast, parenteral vaccination with an inactivated vaccine (Nobivac) induced a weak mucosal ( a n d serum) IgM response, but no IgA response. A similar observation was m a d e by Rodak et al. (1987) using another inactivated vaccine. After challenge infection of Bartha-vaccinated pigs, mucosal m e m o r y responses were clearly induced, but they were less obvious after challenge infection o f Nobivac-vaccinated pigs. Both Bartha and Nobivacvaccinated pigs were partially protected against the challenge infection. Because the n u m b e r o f pigs used in this study was small, however, we cannot state conclusively that mucosal m e m o r y responses induced protection. In another study, in which the efficacy o f different ADV vaccines was tested, we found no apparent correlation between protection and either serum IgA responses after challenge, or serum IgG levels at the m o m e n t o f challenge, or serum IgG responses after challenge (not shown). We are presently investigating the protective value o f m e m o r y responses at mucosae. We conclude that the ELISAs developed in this study to detect antibodies directed against ADV are isotype-specific and sensitive and thus offer a useful methodology to further study local and systemic i m m u n e responses against ADV in pigs. ACKNOWLEDGMENTS This work was supported by a grant from the Commission o f the European Communities, contract no. 3910. The authors thank Dr. L. van Leengoed, Dr. J.M.A. Pol, and the animal technicians for helping with the lung lavages.

REFERENCES Andries, K., Pensaert, M.B. and Vandeputte, J., 1978. Effect of experimental infection with pseudorabies (Aujeszky's disease) virus on pigs with maternal immunity from vaccinated sows. Am. J. Vet. Res., 39: 1282-1285. Bartha, A., 1961. Experimental reduction of virulence of Aujeszky's disease virus. Ungarische Tier~irztl. Monatschr., 16: 42-45. Bitsch, V. and Eskildsen, M., 1976. A comparative examination of swine sera for antibody to Aujeszkyvirus with the conventional and a modified virus-serum neutralization test and a modified complement fixation test. Acta Vet. Scand., 17: 142-145. Butler, J.E., Heo, Y., Adams, P. and Richerson, H.B., 1990. The antigen-linkednature of microtiter ELISAsrequires partial depletion of IgG to permit reliabledetermination of rabbit serum IgA antibody activity. Mol. Immunol., 27:319-325. Chantler, S. and Diment, J.A., 1981. Current status on specific IgM antibody assays. In: A. Voller, A. Barlett and D. Midwell (Editors), Immunoassays for the 80s. MTP Press, Lancaster, pp. 417-430.

112

T.G. KIMMAN ET AL.

De Leeuw, P.W. and Van Oirschot, J.T., 1985. lntranasal vaccination of pigs against Aujeszky's disease: comparison with inactivated vaccines in pigs with low maternal antibody titers. Res. Vet. Sci., 39: 34-38. De Leeuw, P.W., Wijsmuller, J.M., Zantinga, J.W. and Tielen, M.J.M., 1982. Intranasal vaccination of pigs against Aujeszky's disease. 1. Comparison of intranasal or parenteral vaccination with an attenuated vaccine in 12 week old pigs from immunized dams. Vet. Q., 4: 4956. Friedman, H.M., Glorioso, J.C., Cohen, G.H., Hastings, J.C., Harris, S.L. and Eisenberg, R.J., 1986. Binding of complement component C3b to glycoprotein gC of herpes simplex virus type 1: mapping of gC-binding sites and demonstration of conserved C3b binding in lowpassage clinical isolates. J. Virol., 60: 470-475. Hiemstra, P.S. Goner, A., Stuurman, M.E., Van Es, L.A. and Daha, M.R., 1987. Activation of the alternative pathway of complement by human serum IgA. Eur. J. Immunol., 17: 321326. James, K., 1990. Immunoserology of infectious diseases. Clin. Microbiol. Rev., 3:132-152. Joseph, B.S., Cooper, N.R. and Oldstone, M.B.A., 1975. Immunologic injury of cultured ceils infected with measles virus. I. Role of IgG antibody and the alternative pathway of complement. J. Exp. Med., 141: 761-774. Kimman, T.G., Westenbrink, F., Straver, P.J. and Van Zaane, D., 1987a. Isotype-specific ELISAs for the detection of antibodies to bovine respiratory syncytial virus. Res. Vet. Sci., 43: 180-187. Kimman, T.G., Westenbrink, F., Schreuder, B.E.C. and Straver, P.J., 1987b. Local and systemic antibody response to bovine respiratory syncytial virus infection and reinfection in calves with and without maternal antibodies. J. Clin. Microbiol., 25: 1097-I 106. Kimman, T.G., Daha, M.R., Brinkhof, J.M.A. and Westenbrink, F., 1989. Activation of complement by bovine respiratory syncytial virus-infected cells. Vet. Immunol. Immunopathol., 21:311-325. Kit, S., Sheppard, M., Ichimura, H. and Kit, M., 1987. Second-generation pseudorabies vaccine with deletions in thymidine kinase and glycoprotein genes. Am. J. Vet. Res., 48: 780-793. Lukacs, N., Thiel, H.J., Mettenleiter, T.C. and Rziha, H.J., 1985. Demonstration of three major species of pseudorabies virus glycoproteins and identification of a disulfide-linked glycoprotein complex. J. Virol., 53: 166-173. Marchioli, C.C., Yancey, R.J., Wardley, R.C., Thomsen, D.R. and Post, L.E., 1987. A vaccine strain ofpseudorabies virus with deletions in the thymidine kinase and glycoprotein X genes. Am. J. Vet. Res., 48: 1577-1583. Martin, S. and Wardley, R.C., 1987. Local humoral and cellular responses in Aujeszky's disease virus infection in pigs. Res. Vet. Sci., 42: 170-174. Martin, S., Wardley, R.C. and Donaldson, A.I., 1986. Functional antibody responses in pigs vaccinated with live and inactivated Aujeszky's disease virus. Res. Vet. Sci., 41:331-335. McFerran, J.B. and Dow, C., 1975. Studies on immunisation of pigs with the Bartha strain of Aujeszky's disease virus. Res. Vet. Sci., 19:17-22. Mclntosh, K., Masters, H.B., Orr, I., Chao, R.K. and Barkin, R.M., 1978. The immunological response to infection with respiratory syncytial virus in infants. J. Inf. Dis., 138: 24-32. Moormann, R.J.M., De Rover, A.P., Briaire, J., Berns, A.J.M., Gielkens, A.L.J. and Van Oirschot, J.T., 1987. Generation of a safe, highly immunogenic PRV vaccine strain through inactivation of the viral thymidine kinase gene. Proc. 12th Int. Herpesvirus Workshop, Philadelphia, p. 255. Perrin, L.H., Joseph, B.S., Cooper, N.R. and Oldstone, M.B.A., 1976. Mechanism of injury of virus-infected cells by antiviral antibody and complement: participation of IgG, F(ab' )2, and the alternative complement pathway. J. Exp. Med., 143: 1027-1041. Rodak, L., Smid, B., Valicek, L. and Jurak, E., 1987. Four-layer enzyme immunoassay (EIA)

PORCINE ANTIBODY RESPONSES TO AUJESZKY'S DISEASE VIRUS

113

detection of differences in IgG, IgM and IgA antibody response to Aujeszky's disease virus in infected and vaccinated pigs. Vet. Microbiol., 13:121-133. Van Leengoed, L.A.M.G. and Kamp, E.M., 1989. Bronchoalveolar lavage in live pigs. Vet. Q., I l: 65-72. Van Oirschot, J.T., Houwers, D.J., Rziha, H.J. and Van Zaane, D., 1988. Development of an ELISA for detection of antibodies to glycoprotein I of Aujeszky's disease virus: a method for the serological differentiation between infected and vaccinated pigs. J. Virol. Methods, 22: 191-206. Van Zaane, D. and Hulst, M.M., 1987. Monoclonal antibodies against porcine immunoglobulin isotypes. Vet. Immunol. Immunopathol., 16: 23-36. Van Zaane, D. and IJzerman, J.J., 1984. Monoclonal antibodies against bovine immunoglobulins and their use in isotype-specific ELISAs for rotavirus antibody. J. Immunol. Methods, 72: 427-441. Wilson, M.B. and Nakane, P.K., 1978. Recent developments in the periodate method of conjugating horseradish peroxidase (HRPO) to antibodies. In: W. Knapp, K. Holubar and G. Wick (Editors), Immunofluorescence and Related Staining Techniques. Elsevier North Holland, Amsterdam, pp. 215-225. Zuffa, A., Salaj, J. and Cernik, K., 1982. Immunity of pigs vaccinated by live or inactivated Aujeszky vaccines against experimental infection. Zentralbl. Veterinaermed., Reihe B, 29: 663-675.