Detection of hog cholera virus antigens in experimentally-infected pigs using an antigen-capture ELISA

Veterinary Microbiology, 34 ( 1993 ) 2 3 3 - 2 4 8 Elsevier Science Publishers B.V., A m s t e r d a m

233

Detection of hog cholera virus antigens in experimentally-infected pigs using an antigen-capture ELISA A.D. Shannon a, C. Morrissy b, S.G. M a c k i n t o s h a a n d H . A . W e s t b u r y b aElizabeth Macarthur Agricultural Institute, Camden, N.S. W.., A ustralia bAustralian Animal Health Laboratory, East Geelong, Vic., Australia (Accepted 22 July 1992 )

ABSTRACT Shannon, A.D., Morrissy, C., Mackintosh, S.G. and Westbury, H.A., 1993. Detection of hog cholera virus antigens in experimentally-infected pigs using an antigen-capture ELISA. Vet. Microbiol., 34: 233-248. An antigen-capture ELISA was used to detect hog cholera virus (HCV) antigens in blood and tissues taken from pigs infected with 2 different strains of virus. Specific antigens were demonstrated in peripheral blood leucocytes (PBLs) and a wide range of tissue samples 4-6 days after infection of pigs with a moderate-high virulent HCV strain (Weybridge virus). Strong signal to noise ( S / N ) ratios were obtained in the ELISA for PBLs and lymphoid tissues such as spleen, tonsil and mesenteric lymph nodes at 5-7 days after infection with the Weybridge virus, S / N ratios varying between 8.119.7 for blood samples and 4.3-19.1 for spleen samples. High positive ELISA results were also obtained for duodenum and ileum samples ( S / N ratios 10.3-18.6) taken from these pigs, reflecting severe pathological changes observed in the gut at post mortem. In contrast, the antigen-capture ELISA gave strong positive results for PBLs and spleen samples only at 7-9 days after infection of pigs with a low virulent strain of HCV (New South Wales virus). The ELISA S / N ratios averaged 9.5 for PBLs and 8.9 for spleen samples in these animals. Although virus isolation detected infection earlier in the infected pigs, the ELISA returned positive results on PBLs and spleen samples around the time all of the animals first showed typical signs of classical swine fever. The technique does not require tissue culture and takes less than 36 h to return a definitive result.

INTRODUCTION

Hog cholera virus (or classical swine fever virus), a highly-infectious pestivirus, causes widespread disease epidemics with consequent severe economic losses to the pig industries on a world-wide basis. Hog cholera is present in at least 36 countries and the disease is enzootic in South America and Asia (Terpstra, 1991 ). In regions free of the disease, such as North America, Correspondence to: A.D.Shannon, Virology Section, EMAI, Private Mail Bag 8, C a m d e n NSW 2570, Australia.

0 3 7 8 - 1 1 3 5 / 9 3 / $ 0 6 . 0 0 © 1993 Elsevier Science Publishers B.V. All rights reserved.

234

A.D. SHANNON ET AL.

Australia and New Zealand, there are strict quarantine precautions to prevent entry of the virus into susceptible pig populations that would result in major epidemics of classical swine fever. In those countries where HCV is endemic, rigorous efforts are being made to control and eliminate the virus from pigs to prevent an ongoing economic drain on the industry. This is underlined by the policy of The Commission for the European Communities to eradicate residual HCV infections from all of its member states (Bendixen 1988 ). Rapid and accurate methods for the diagnosis of HCV infections in pigs are essential for the control and eradication programs being proposed. Differentiation of HCV infections from infections of pigs with the two related pestiviruses, bovine viral diarrhoea virus (BVDV) and border disease virus (BDV), is also required given the known cross-species transfer of pestiviruses (Terpstra and Wensvoort, 1988; Moennig, 1990). In this regard, the production of HCV-specific, group pestivirus-specific and ruminant pestivirus-specific monoclonal antibodies (Wensvoort et al., 1986,1989; Cay et al., 1989; Edwards et al., 1991 ) has enabled the accurate diagnosis of HCV infections in pigs and a distinction to be made from infections with the two ruminant pestiviruses. At present, these monoclonal antibodies can be used to detect specific pestivirus antigens in frozen tissue sections from infected animals using techniques such as indirect immunoperoxidase staining or direct immunofluorescence (cf Bielefeldt O h m a n n et al., 1981; Bel~ik et a1.,1989; Hewicker et al., 1990; Terpstra, 1991 ). Similarly, virus can be visualised in infected tissue cultures using these techniques after conventional virus isolation procedures (Meyling, 1984; Smith et al., 1988 ). Although these methods are both sensitive and specific, they are slow and cumbersome for use in routine diagnostic situations and do not take full advantage of the potential of monoclonal antibodies. We have recently described a rapid, specific and sensitive antigen-capture ELISA that detects BVDV antigens in blood and tissues of immunotolerant, persistently-infected cattle (Shannon et al., 1991 ). The method uses a strongly reactive goat polyclonal antiserum to trap antigen and a combination of 3 pestivirus-specific monoclonal antibodies as a detection system. The present paper describes an extension of the antigen-capture technique to the detection of HCV antigens in both blood and tissue samples taken from pigs experimentally infected with different isolates of HCV. The test is rapid and sensitive and is able to identify infected pigs at about the time they first show clinical signs of classical swine fever. MATERIALS AND METHODS

Animals and samples Three experiments were carried out on a total of 24 pigs. The animals were all commercial Large White × Landrace cross pigs, 4-6 weeks old. They were

DETECTION OF HOG CHLOERA VIRUS ANTIGENS

235

housed in the High Security Animal Wing at the Australian Animal Health Laboratory (AAHL) in individually isolated rooms. The pigs were fed a commercial pig starter ration ad libitum. Rectal temperatures were monitored prior to infection with hog cholera virus and at daily intervals after infection. Blood samples ( 10 ml in heparinised vacutainers) were collected from the anterior vena cava prior to infection and at intervals after infection (daily, or every second day). Animals exhibiting acute disease symptoms following infection were killed by an overdose of barbiturate and tissue specimens collected at post mortem. All animals showing clinical symptoms had been subjected to post mortem examination by 9 days post-infection. At post mortem, a range of tissue samples were collected for virus isolation and detection of antigen by the antigen-capture ELISA.

Viruses Three different hog cholera virus (HCV) isolates were inoculated into pigs in the animal experiments and the same 3 viruses were used to infect cultures of PK-15 cells. Two of these isolates (Weybridge and Baker A ) were imported into Australia for use as reference strains for HCV diagnosis while the third (New South Wales) is the only existing field isolate obtained from pigs in Australia. The Weybridge isolate from the United Kingdom has a moderate to high pathogenicity for pigs, whereas the Baker A strain from the United States is non-pathogenic in pigs having been tissue-culture adapted by 30 passages in PK-15 cells. The New South Wales (NSW) isolate has a low pathogenicity in pigs. This particular virus was obtained from a pig during the 1960/ 1961 outbreak of chronic 'swine fever' (hog cholera) in Australia. All 3 isolates have been propagated in PK- 15 cells but both the Weybridge and NSW viruses were passaged through pigs prior to their use in the present experiments. Pig inoculations. A total of 13 young pigs in 2 experiments were infected with 1 ml of pig blood (diluted 1:10 in phosphate buffered saline, PBS) containing Weybridge strain HCV at a final titre of 104s TCIDso/dose. Five pigs were infected with 1 ml whole blood (undiluted) containing the NSW strain of HCV at the same titre of 104.5 TCIDso/dose. Two pigs in each of 2 experiments acted as uninfected controls. In a third experiment, 2 pigs were inoculated with 1 ml tissue culture supernatant ( 107° TCIDso/ml) harvested from PK- 15 cells infected with the attenuated Baker A strain. Tissue culture inoculations PK- 15 cells were infected with the 3 strains of HCV for monoclonal antibody (mAb) screening using the antigen-capture ELISA. Virus was adsorbed on monolayers at 80-90% confluency for 2-3 h at 37°C, the cells rinsed in serum-free medium, then incubated for 48 h at 37°C (5% CO2) in Eagles minimal essential m e d i u m (EMEM) supple-

236

A.D. SHANNON ET AL.

mented with 1% foetal calf serum (pestivirus free). The infected cells were scraped from the flasks, pelleted at 2000g, and extracted with 1% Nonidet P40 detergent (Sigma, NP-40, diluted in PBS) for 2 h at room temperature. Unlysed cell debris was removed by centrifugation for 10 min at 1200g and the supernatants assayed in the ELISA using mAbs singly or in combination.

Sample preparation Anticoagulated blood samples were centrifuged (10 min, 1200g) and the peripheral blood leucocytes (PBLs) harvested and transferred to 25 ml capped centrifuge tubes. Nonidet P-40 detergent extracts for use in the antigen-capture ELISA were prepared as previously described (Shannon et al., 1991 ). Briefly, 10 ml cold (4 ° C ) 0.17M NHaCI was added to lyse any contaminating red cells, the tubes mixed and allowed to stand for 10 min. All tubes were then filled with cold PBS (4°C) to dilute the NH4C1, mixed by gentle inversion, and the PBLs pelleted by centrifugation (5 min, 1200g). The resulting pellet was lysed by the addition of I ml 1% NP-40 followed by thorough vortexing. Samples were allowed to stand at room temperature (RT) for 2 h with occasional mixing. Cellular debris was removed by centrifugation ( 10 min, 1200g) and supernatants stored at 4 °C prior to assay. Tissues collected at post mortem were also extracted with NP-40 according to the methods of Shannon et al., 1991. Pieces of tissue (1-2g) were placed in 25 ml capped centrifuge tubes and roughly minced with fine-pointed scissors. Five ml 1% NP-40 was added to each sample and the tubes thoroughly vortexed before standing at RT for 2 h. The samples were vortexed every 30 min over this period before tissue debris was pelleted by centrifugation ( 10 min, 1200g). Supernatants were transferred to microcentrifuge tubes and stored at 4 °C until assayed. Where necessary, the supernatants were further clarified in a microcentrifuge before use (2 min, 10 000g, 4 °C). Tissue homogenates ( 10% in PBS) were also prepared by standard methods for all samples collected at post mortem and used for virus isolation.

Antigen-capture ELISA The presence of specific HC viral antigens in extracts of blood and tissue samples from all pigs was determined using an antigen-capture ELISA similar to that described for the detection of specific BVDV antigens in the blood and tissues of infected cattle (Shannon et al., 1991 ). Briefly, a high titred ( > 1:100 000) goat polyclonal antiserum was bound to wells of a microtitre plate. In a separate plate, a combination of 3 monoclonal antibodies specific for pestiviruses were reacted with HCV antigens from detergent-extracted blood and tissue samples. The monoclonal-antigen mixture was then allowed to bind with the polyclonal antiserum immobilised on the ELISA plate. Bound monoclonals were detected using biotinylated goat anti-mouse IgG followed by a streptavidin biotinylated horseradish-peroxidase complex and tetrame-

DETECTION OF HOG CHLOERA VIRUS ANTIGENS

237

thylbenzidine (TMB) chromagen. All samples were tested in parallel with an irrelevant mAb (Belov and Whalley, 1988 ) to determine background binding. A detailed description of the ELISA is as follows (cf Shannon et al. 1991 ): Wells of microtitre ELISA plates (Greiner High Binding F, Nunc Maxisorp or Dynatech M 129B) were coated with 100/zl of goat polyclonal antiserum diluted 1:200 in 0.1 M carbonate buffer, pH 9.6, and incubated overnight at RT (20-22 °C) in a humidified chamber. The o p t i m u m dilution of the goat polyclonal antiserum used to coat the ELISA plates was estimated by a checkerboard titration. In parallel, wells of microtitration plates (Titretek, Linbro, or equivalent low binding plates) were blocked by the addition of 200/d/well of 5% skim milk p o w d e r + 5% normal goat serum in carbonate buffer, pH 9.6 (blocking solution 1 ). These plates were also incubated overnight at RT and will be referred to as primary binding plates. Polyclonal-coated ELISA plates were stored at 4°C for periods of up to 4 weeks before use while the blocked primary binding plates were stored for similar periods at - 2 0 ° C. All subsequent incubations were carried out under humidified conditions and at least 3 washing steps were undertaken between all stages on an automatic plate washer (Titretek Microplate Washer 120, Flow) using PBS/0.5% Tween 20 (PBST). The coated ELISA plates were blocked for 90 min at 37°C with 200/A of blocking solution 1. After incubation, these plates were set aside at RT without washing for a further period of 2-4 h. The primary binding plates were washed and 4 wells received 60/~l/well of each test antigen preparation. Duplicate wells then received 60/~l of positive mAb mixture and the remaining duplicate wells 60/11 of negative mAb. These plates were incubated for 90 rain at 37 °C, followed by 2-4 h at RT. The coated and blocked ELISA plates were washed and 100/~1 of the mAb-antigen mixture transferred from the primary binding plates to the corresponding wells of the ELISA plates. The binding plates were discarded and the ELISA plates incubated overnight at RT. After washing 5 times, the ELISA plates were reblocked with 150 /~l of blocking solution 2 (50/0 skim m i l k + 5°/o normal goat serum in PBS) for 90 min at 37°C. Bound monoclonal antibody was detected using an avidin-biotin complex (ABC) amplification technique. Plates were washed 3 times and 100/~l of biotinylated goat anti-mouse IgG (Amersham), appropriately diluted in PBST containing 1% gelatin (PBSGT), was added to all wells and incubated for 90 min at 37°C. After washing 5 times, 100/~1 of streptavidin biotinylated horseradish-peroxidase complex (Amersham) appropriately diluted in PBSGT was added to each well and incubated for 30 min at 37°C. Before the addition of chromogen, plates were washed 10 times in PBST. A stock solution of TMB chromogen (3.5 m g / m l in methanol, 3'-3', 5'-5'-tetramethylbenzidine, Sigma) was diluted 1:100 in 0.1 M citrate acetate buffer, pH 6.0, just before use. After addition of 0.5/~l/ml 300/0 H202, 100 #l of the activated chromogen (substrate) was added to each well and the plates incu-

238

A.D. SHANNON ET AL.

bated in the dark for 10 min at RT before the reaction was stopped by the addition of 100/d/well 1 M H2SO4. The optical densities (OD) were measured at 450 n m on a semi-automatic plate reader (Easy Reader, SLTLabinstruments ). The same mixture of 3 mAbs (2NB2, P4G 11 and N2B 12) used previously to detect cattle pestivirus antigens was employed in the first series of experiments but antibody N2B12 was replaced by the group pestivirus-specific monoclonal, P3C 12, in the subsequent 2 experiments to improve the sensitivity of the technique for HCV antigen. Signal to noise ( S / N ) ratios were calculated for each individual sample by dividing the mean optical density obtained with the pestivirus-specific mAbs by the mean optical density obtained with the irrelevant mAb. This enabled comparisons to be made between pigs and between viruses. For blood samples and all tissues other than liver and kidney, S / N ratios >_2.0 are positive in the antigen-capture ELISA. In the case of liver and kidney samples, ratios >2.3 are positive (Shannon et al., 1991). Virus isolation Parallel assays for the presence of infectious HCV in blood and tissue samples were carried out using conventional techniques on PK- 15 ceils. The presence of virus after 2-5 days in culture was determined by indirect immunofluorescence or immunoperoxidase staining using a hyperimmune pig anti-HCV polyclonal antiserum or the group-specific antipestivirus mAbs, 2NB2 and P4G 11. H C V antibody assays Antibody against HCV in the serum of 2 pigs at 21 days post infection with the Baker A strain was measured using modifications of the neutralisation peroxidase-linked assays (NPLA) described by Terpstra et al. (1984) and Hyera et al. ( 1987 ). The Baker A strain of HCV and group pestivirus-specific monoclonal antibodies were employed in this test. RESULTS Screening of a limited number of monoclonal antibodies raised against BVDV (Shannon et al., 1991 ) for activity against 3 isolates of HCV was carried out in the antigen-capture ELISA using NP-40 extracts of infected pig tissues in the case of the Weybridge strain or infected PK-15 cell extracts in the case of the NSW and Baker A strains. The results (Table 1 ) show that mAb N2B 12 did not react with any of the 3 HCV isolates while having a strong reactivity in the ELISA with the BVDV positive cattle spleen. The group-specific mAb P3C 12, on the other hand, showed moderate ELISA reactivity with all 3 HCV isolates as well as strong activity against the BVDV

1.63 0.16

0.22

0.22

0.31 0.19 0.20 0.22

16.3 1.5

1.7

1.6

1.3 1.3 1.5 1.5

1.25 0.16

0.74

0.63

0.55 0.38 0.40 0.20

OD

OD ~

S/N 2

P3C 12

N2B 12

12.5 1.5

6.2

4.5

2.4 2.5 3.1 1.3

S/N

1.44 0.21

2.20

2.21

2.13 2.06 2.12 0.43

OD

14.4 1.9

18.3

15.8

9.3 13.7 16.3 2.9

S/N

2NB2 + P4G 11

Pestivirus-specific monoclonal antibodies

~ELISA optical density measured at 450 nm 2 Signal to Noise ratios; S/N ratios >_2.0 are positive in the antigen-capture ELISA 3 Assayed on tissues taken from an infected pig 6 days p.i.

Positive cattle spleen Negative cattle spleen

B VD V controls

Infected PK-I 5 cells

Baker A H C V

Infected PK-15 cells

NS W H C V

Spleen Tonsil Ileum Heart

14eybridge H C V 3

Samples tested (N P-40 extracts )

1 S1 0.11

1.98

2.02

1.76 1.98 1.81 0.30

OD

18.1 1.0

16.5

14.4

7.7 13.2 13.9 2.0

S/N

2NB2 + P4G 11 +N2BI2

Antigen-capture ELISA screening of anti-pestivirus monoclonal antibodies against 3 HCV isolates

TABLE 1

1.15 0.13

2.07

2.15

2.05 2.02 2.06 0.32

OD

11.5 1.2

17.3

15.4

8.9 13.4 15.8 2.1

S/N

2NB2+P4GII +P3CI2

0.10 0.11

0.12

0.14

0.23 0.15 0.13 0.15

OD

Irrelevant monoclonal antibody

t~3

m 7

g

z

>.

<

0/8 0/8 6/8 8/8 8/8 4/4 5/5 2/2 nt nt

8 8 13 11 10 4 9 2 nt nt

Number of animals 1.0 0.9 1.0 1.1 3.3 9.2 14.4 15.5 nt nt

(0.7-1.4) (0.8-1.0) (0.8-1.3) (0.9-1.4) (1.2-7.0) 2 (8.1-10.3) (8.9-16.8) (11.3-19.7)

S / N ratio ~ Mean (Range) 0/5 nt 0/5 0/5 nt 3/5 nt 4/4 3/3 1/ 1

12.2

1

nt

1.2 1.2 nt 1.5 nt 7.6 11.1

1.3

S / N ratio Mean

5 nt 5 5 nt 5 nt 4 3

Number of animals

ELISA

Virus isolation (No. positive/ No. tested)

Virus isolation (No. postive/ No. tested)

ELISA

New South Wales virus

Weybridge virus

~Signal to noise ratio; ratios > 2.0 are positive in the ELISA. 2At day 4 p.i., 6/10 Weybridge virus-infected pigs ELISA positive; at day 5 p.i., 4 / 4 ELISA positive. 3At day 5 p.i., 1/5 NSW virus-infected pigs EL1SA positive; at day 7 p.i., 4 / 4 ELISA positive. hi, Not tested.

0 1 2 3 4 5 6 7 8 9

Day postinfection

Virus isolation and antigen-capture ELISA results on white blood cells from pigs infected with 2 strains of HCV

TABLE 2

(5.0-10.7) (8.5-12.8) (12.2)

(1.1-2.3) 3

(1.1-1.3) (1.1-1.3)

(1.1-1.4)

(Range)

1',3

m ,q >

z z oz

>

0

DETECTION OF HOG CHLOERA VIRUS ANTIGENS

241

positive control. Replacing mAb N2B 12 with mAb P3C 12 in the combinations of 3 monoclonals routinely used in the ELISA slightly improved the sensitivity of the technique for the detection of HCV, as shown by the higher S / N ratios obtained for all 3 HC viruses using the combination 2 N B 2 + P 4 G 1 1 + P 3 C 1 2 (Table 1). These 3 mAbs, while raised in mice against BVDV, are group specific and have been shown to react with 40 other HCV isolates (Shannon, Bloemraad and Wensvoort, unpublished data). Although replacing a BVDV-specific antibody with a group-specific antibody improved the sensitivity of the ELISA for HCV antigens, it is evident from the results shown in Table 1 that the use of the 2 strong group-pestivirus specific mAbs, 2NB2 and P4G 11, was sufficient to clearly detect HCV antigens in the ELISA. Indeed, the combinations of 2 and 3 mAbs used in this study detected Weybridge HCV antigen in heart muscle of an infected pig at 6 days post infection (p.i.). This tissue was included as a low positive control in these experiments since it contained less virus than tissues such as spleen, tonsil and ileum (data not shown ). Analyses of blood samples taken from pigs at various times after infection with 2 different strains of HCV are shown in Table 2. The Weybridge HCV strain was more pathogenic in the pigs, resulting in a significant elevation in temperature ( + 1.3 °C) 48 h p.i. and severe clinical symptoms of disease by 4 days after infection. By 6 days p.i., the majority of the pigs were moribund and body temperatures were elevated by an average of + 2 ° C . In contrast, pigs infected with the NSW strain did not show a significant temperature rise until 5-6 days p.i. ( + 1.2°C) and they exhibited mild to moderate clinical symptoms beginning at 6-7 days after infection. On day 8 p.i., the NSW HCVinfected pigs were reluctant to move and body temperatures had risen to a mean of 41.1 ° C ( + 1.7 ° C). The results of virus isolation and antigen-capture ELISA detection of specific HCV antigens in the blood of infected pigs over the 7-9 day period following infection showed a similar difference between the 2 groups (Table 2). All of the pigs infected with Weybridge HCV were positive by virus isolation at day 3 p.i. while the S / N ratios in the antigencapture ELISA were positive for 6 of the 10 infected animals at day 4. The ELISA results were strongly positive on days 5 and 6 p.i. for all of these animals, S / N ratios ranging from a low of 8.1 to a high of 19.7 (Table 2 ). Pigs infected with NSW HCV showed a slower rate of virus replication with 3 of the 5 infected pigs positive by virus isolation on day 5 p.i. and only one animal ELISA positive at this time with a low S / N ratio of 2.3. However, by 7 days p.i., the ELISA results were strongly positive for all NSW HCV-infected pigs with S / N ratios varying between 5.0 and 12.8 (Table 2). In addition to analysing blood samples, animals infected with both Weybridge and NSW strains of HCV were killed at various times after infection and a wide range of tissues collected at post mortem. All tissues were processed and tested in the antigen-capture ELISA and, where possible, the re-

242

A.D. SHANNON ET AL.

suits were confirmed by virus isolation undertaken on the individual tissues. While all tissues were readily analysed by the ELISA, a large number of tissues proved toxic in cell culture and virus was not able to be demonstrated using this method. The results of applying the antigen-capture ELISA to 14 tissues collected at 2 days, 4 days and 6-7 days after infection of pigs with the Weybridge strain of HCV are presented in Table 3. All of the tissues collected from 2 animals at 2 days after infection were negative in the ELISA. At 4 days p.i., spleen and tonsil samples from 2 pigs were positive in the antigen-capture ELISA with both spleen samples showing moderately high S / N ratios of 5.2 and 6.8 (Table 3 ). Ileum and colon samples taken from one of these pigs were also positive in the ELISA. By days 6 and 7 after Weybridge virus infection, the majority of the tissue samples tested from 4 pigs were strongly positive in the ELISA with the highest S / N ratios obtained for abdominal tissues, including pancreas, duodenum, ileum and colon, and lymphoid tissues such as spleen, tonsil and mesenteric lymph node (Table 3). Brain tissue from only one of the 4 animals was positive in the ELISA and kidney tissue from one animal was negative, showing that these tissues tended to give variable responses. It can be seen (Table 3) that all tissues from 2 uninfected control pigs (not in TABLE 3 Antigen-capture ELISA results on tissues from 8 Weybridge virus infected pigs and 2 uninfected pigs Tissue NP-40 extract

Brain Tonsil Lung Heart muscle Spleen Liver 4 Kidney 4 Pancreas MesentericLN Duodenum Ileum Colon Bladder Urcter

ELISA signal to noise ratios ~ 2 days p.i. (2) 2 Mean (Range)

4 days p.i. (2) Mean (Range)

6-7 days p.i. (4) Mean (Range)

Uninfected controls (2)

1.2 (1.1-1.4) 1.0(1.0-1.0) 1.0 (1.0-1.0) 1.1 (I.0-1.2) 1.3(1.1-1.4) 1.1 (0.9-1.3) 0.9 (0.7-1.0) 0.9 (0.7-1.0) 0.9 (0.9-0.9) 1.0 (0.8-1.3) 0.9 (0.9-0.9) 0.8(0.5-1.1) 1.0 (0.8-1.3) 1.0 (0.8-1.2)

0.9 (0.8-1.0) 3.2 (2.5-4.0) 1.6(1.6-1.6) 1.1 (0.9-1.2) 6.0(5.2-6.8) 1.0(0.8-1.1) 0.7 (0.6-0.7) 1.2 (1.0-1.3) 1.1 (1.0-1.1) 0.9 (0.9-0.9) 3.7 (1.9-5.5) 1.8(1.4-2.2) 1.0 (0.9-1.0) 1.0 (0.9-1.1)

2.3 8.3 9.5 2.9 9.7 7.2 6.6 11.6 11.3 13.0 16.0 9.9 4.4 8.6

1.0; 1.03 0.8;1.0 0.9;1.0 1.1;1.1 1.0; 1.1 1.0;1.0 1.0; 1.1 1.0;1.2 0.9; 1.1 1.0; 1.0 0.9;0.9 0.8;1.0 1.0;1.0 1.0:1.1

(0.9-6.0) (6.1-13.6) (7.8-13.1) (I.2-4.8) (4.3-19.1) (4.6-10.5) (1.8-10.7) (8.3-14.2) (4.6-15.3) (10.3-17.6) (11.4-18.6) (6.6-13.1) (1.9-10.9) (4.4-11.4)

~Signal to noise ratios >_2.0 are positive in the ELISA for all tissues except liver and kidney. 2Number of animals tested shown in brackets. 31ndividual results for 2 control animals presented. 4Signal to noise ratios > 2.3 are positive for kidney and liver samples.

DETECTION OF HOG CHLOERA VIRUS ANTIGENS

243

contact with infected pigs) were negative in the ELISA. Of interest in this regard was the finding that a further uninfected control pig maintained in contact with 5 Weybridge virus-infected pigs showed a very strong positive ELISA result for spleen tissue ( S / N ratio 12.2 ) collected 9 days after the contact pigs were infected. This animal did not show a significant rise in body temperature until 6-7 days after infection of the contact animals and blood samples collected from the animal remained negative throughout the 9 day period. Antigen-capture ELISA results for the same series of tissues collected at 8 and 9 days after infection of 5 pigs with the NSW strain of HCV are shown in Table 4. Also presented are the ELISA results for a control, uninfected pig maintained in contact for 9 days. The ELISA S / N ratios for all tissues were, on average, considerably lower than those seen following Weybridge virus infection (cf Tables 3 and 4) but a significant number of tissues were strongly positive by day 9 after infection with the NSW virus. These included spleen, ileum and colon, with the highest S / N ratios observed for spleen samples from all 5 infected pigs (Table 4). Interestingly, tonsil samples were only positive in 2 of these 5 pigs, unlike the results for the Weybridge virus-infected pigs 67 days p.i. (see Table 3 ). This pattern was also observed for other tissues such TABLE 4 Antigen-capture ELISA results for tissues from 5 NSW virus infected pigs and 1 uninfected contact pig Tissue NP-40 extract

Brain Tonsil Lung Heart muscle Spleen Liver 3 Kidney 3 Pancreas Mesenteric LN Duodenum Ileum Colon Bladder Ureter

ELISA signal to noise ratios ~ 8 days p.i. (3) 2 Mean (Range)

9 days p.i. (2) Mean (Range)

Uninfected contact at 9 days (1)

1.1 2.6 3.2 1.3 9.1 0.7 0.4 0.8 2.1 2.0 4.2 2.4 1.3 1.4

1.3 2.5 5.0 1.4 8.5 1.5 0.5 1.2 2.2 3.7 8.2 4.7 1.3 1.4

1.1 1.0 1.0 1.2 0.9 0.6 0.4 1.1 1.0 1.0 1.0 0.8 1.2 1.1

(1.1-1.1) (1.0-4.9) (1.6-4.9) ( 1.0-1.5 ) (7.8-10.0) (0.4-1.1) (0.3-0.5) (0.7-1.0) (1.4-3.2) (0.9-3.9) (0.9-7.2) (1.1-3.8) (1.1-1.5) ( 1.2-1.6 )

(1.3-1.3) (1.9-3.2) (2.1-7.9) ( 1.1-1.7 ) (6.3-10.7) (1.5-1.5) (0.5-0.5) (1.0-1.3) (1.9-2.5) (2.2-5.1) (6.4-10.0) (3.8-8.5) (0.8-1.7) ( 1.2-1.6 )

'Signal: Noise ratios >_2.0 are positive in the ELISA for all tissues except liver and kidney. 2Number of animals tested shown in brackets. 3Signal:noise ratios > 2.3 are positive in the ELISA for kidney and liver samples.

244

A.D. S H A N N O N E T A L .

as lung, liver, pancreas, d u o d e n u m and ureter where strong positive results were obtained following Weybridge virus infection but negative, or weak positive, reactions only obtained following NSW virus infection. It is noteworthy that virus was isolated from the spleen samples of all 5 of the NSW infected pigs but from tonsilar tissue from only 3 of these animals. Similarly, virus isolation results tended to parallel the ELISA results for all of the other tissues taken from the NSW virus-infected animals. All tissues from the uninfected pig were negative in the ELISA (Table 4 ). This animal showed no significant rise in body temperature over the 9 days it was in contact with the virusinfected group of pigs and all blood samples collected over the period remained negative in the ELISA. Significantly, virus was not isolated from either blood or tissue samples taken from this animal. A further experiment was undertaken with 2 pigs infected with the Baker A tissue-culture adapted HCV strain. Following infection, one animal showed a transient increase in body temperature ( + 1.0°C) for 24 h at 7 days p.i. No clinical signs were evident but virus was isolated from the blood of both animals on days 7 and 8 p.i. Virus titrations showed an average of 1021 TCIDso/ ml in blood collected on these 2 days. Antigen-capture ELISA results were consistently negative on the blood samples taken from both animals. By 21 days p.i., both animals had developed specific anti-HCV antibodies to a titre of 1:1028, as demonstrated by the NPLA (results not shown). DISCUSSION

Hog cholera virus infections in pigs spread rapidly due to the highly infectious nature of the virus, particularly with the more virulent strains (cf Terpstra, 1991 ). Strains of low to moderate virulence also pose a problem since they may give rise to chronic infections where animals can shed virus intermittently or even continuously until death (Mengeling and Packer, 1969). To control and eradicate outbreaks of hog cholera (classical swine fever), there is a need to have rapid and accurate laboratory diagnostic tests to assist field veterinary services with the task of eliminating infection from the pig population. The results presented in this paper show that an antigencapture ELISA technique, originally developed for detection of pestivirus infections in cattle (Shannon et al., 1991 ), can be used to detect specific HCV antigens in blood and tissues of pigs acutely infected with HC viruses of different virulence. The antigen-capture ELISA is based on reacting specific HCV antigen with a combination of 3 broadly reactive, group-pestivirus specific mAbs that are all directed against the conserved p 120/p75 nonstructural protein encoded by the viral RNA (Meyers et al., 1989; M o o r m a n n et al., 1990; Thiel et al., 1991; Shannon and Mackintosh, unpublished data). The method allows priority binding of the monoclonals to the antigens before the antigen-anti-

DETECTION OF HOG CHLOERA VIRUS ANTIGENS

245

body mixture is reacted with an immobilised polyclonal antibody on ELISA plates. Sensitivity is further enhanced by the use of an avidin-biotin complex (ABC) technique to detect the bound mAbs. Although the mAbs used in the ELISA were raised against BVDV, extensive analysis of their reactivities has shown that they detect the 3 pestiviruses, BVDV, BDV and HCV (Shannon et al., 1991 ). The present antigen-capture ELISA can therefore be expected to detect infections in pigs with any of these viruses. In this regard, it has been reported that ruminant pestivirus infections in pigs cause disease that is clinically indistinguishable from chronic hog cholera (Terpstra and Wensvoort, 1988 ) and it is necessary to differentiate these infections from those due to HCV. Using monoclonal antibodies with varying specificities for the 3 different pestiviruses, this is now possible using a peroxidase-linked assay on fixed, infected cells in a microtitre plate system (Edwards et al., 1991 ). However, different panels of mAbs used in the current antigen-capture ELISA can achieve the same result. Indeed, preliminary testing has shown that the use of 2 combinations of mAbs in the ELISA can correctly identify BVDV infections in pigs using tissue samples taken from infected animals (Shannon and Wensvoort, unpublished data). The antigen-capture ELISA detected specific HCV antigens in blood leucocytes and in a wide range of tissue samples at days 4-6 after infection of pigs with a moderate-high virulent strain of HCV. This was around the time the animals first exhibited typical symptoms of disease. It was of interest that the highest readings in the ELISA were obtained for the white blood cells and for lymphoid organs such as spleen, tonsil and mesenteric lymph node, similar to the findings for cattle (Shannon et al., 1991 ). It has recently been reported (Su~a et al., 1992 ) that pigs infected with the highly pathogenic HCV strain, Alfort, develop a dramatic depletion preferentially of B lymphocytes in the circulatory system as well as in lymphoid tissues. These effects were observed around 9 days p.i. and are likely to have occurred in our pigs infected with Weybridge HCV. However, both PBL and lymphoid tissue extracts tested from these animals returned high S/N ratios around the time lymphocyte depletion would be expected. It is evident that those lymphocytes remaining in the circulation and in the lymphoid tissues were infected and contained sufficient HCV antigen to be detected in the antigen-capture ELISA. Very high readings were also obtained for duodenum and ileum in the Weybridge HCV-infected pigs, reflecting severe pathological changes observed in these organs at post mortem. Virus isolation confirmed that high levels of HCV proliferation had occurred in the lymphoid organs and in the gut of all of these animals. In contrast, the results obtained in the antigen-capture ELISA for blood leucocytes and tissues from animals infected with the low virulent strain of HCV (NSW isolate) showed that infection could only be reliably identified using PBL or spleen samples at 7-9 days after infection. Virus isolation and

246

A.D. SHANNON ET AL.

ELISA results for other tissues gave variable results at this time. It was surprising to find that tonsil samples from these pigs were unreliable in indicating infection, both by virus isolation and in the ELISA, in view of the results previously reported for low virulent strains of HCV by Mengeling and Packer (1969). These authors found that virus was isolated earlier and persisted longer in tonsillar tissue in pigs infected with a low virulence strain when compared to tissues such as spleen. Irrespective, it is evident from the results obtained in the present study that blood leucocyte preparations reliably detect HCV infection using the antigen-capture ELISA. Assaying several tissues collected from individual animals also gave dependable results in the ELISA, even in the case of infections with HCV strains of low virulence. On our findings, spleen is the tissue of choice to include in the range of tissues to be assayed in the ELISA. A comparison of the results obtained by virus isolation techniques and by the antigen-capture ELISA for blood and tissue samples demonstrated that virus isolation is more sensitive than the ELISA and can detect infection earlier in the course of the disease. Notwithstanding, virus isolation takes from 3-5 days to return a definitive result whereas the ELISA can be performed in less than 36 h. The results obtained with the ELISA were positive for all animals before the results of virus isolation were known in the case of all 18 animals infected with 2 HCV isolates of low and moderate-high virulence. The ELISA therefore has an advantage in outbreaks of classical swine fever where early results of laboratory testing are of importance in controlling the spread of infection. Admittedly, direct immunofluorescence or immunoperoxidase staining of cryostat sections is still the most rapid way of confirming infection at the beginning of a suspected outbreak, but these methods are not applicable to large numbers of samples and negative results may still require verification by virus isolation (Terpstra, 1991 ). The antigen-capture ELISA, on the other hand, is ideally suited to handling large numbers of blood or tissue samples in a short time, does not require sophisticated equipment and can be carried out without the use of tissue culture facilities. These factors combine to make the method suitable for use in endemic hog cholera areas, such as the South East Asia region, where access to tissue culture may be limited in regional laboratories. In Australia, where hog cholera has been eradicated, we aim to add the new test to virus isolation and immunoperoxidase staining techniques for use in the event of a future outbreak of disease (Shannon, 1991 ). Further work is in progress to assess the use of HCV-specific mAbs in the antigen-capture ELISA to improve sensitivity and specificity. ACKNOWLEDGEMENTS

The authors wish to thank the staff of the secure large animal facility of the Australian Animal Health Laboratory for their assistance with the care of the

DETECTION OF HOG CHLOERA VIRUS ANTIGENS

247

animals used in the experiments. The study was undertaken as part of the Australian exotic disease preparedness strategy and partly funded by the Exotic Animal Diseases Preparedness Consultative Committee (EXANDIS).

REFERENCES Bel~ik, K., Gimeno, E.J. and Bel~ik, S., 1989. Demonstration of bovine viral diarrhoea virus antigens in cell cultures and in paraffin-embedded tissue sections by the peroxidase-antiperoxidase (PAP) technique using monoclonal antibodies. Acta Vet. Scand., 30:231-233. Belov, L. and Whalley, J.M., 1988. Virus specific polypeptides ofcaprine arthritis- encephalitis virus recognised by monoclonal antibodies to virion proteins p24 and p14. J. Gen. Virol., 69:1097-1103. Bendixen, H.J., 1988. Control of classical swine fever. In: B.Liess (Editor), Classical Swine Fever and Related Viral Infections. Martinus Nijhoff, Boston, pp. 217-232. Bielefeldt Ohmann, H., Holm Jensen, M., Sorensen, K.J. and Dalsgaard, K., 1981. Demonstration of bovine viral diarrhea virus antigen in cryostat- and paraffin-sections of bovine tissues by the immunoperoxidase technique. Acta Path. Microbiol. Scand. Sect., C, 89:281-285. Cay, B., Chappuis, G., Coulibaly, C., Dinter, Z., Edwards, S., Greiser-Wilkc, I., Gunn, M., Have, P., Hess, G., Juntti, N., Liess, B., Mateo, A., McHugh, P., Moennig, V., Nettleton, P. and Wensvoort, G., 1989. Comparative analysis of monoclonal antibodies against pestiviruses: report of an international workshop. Vet Microbiol., 20:123-129. Edwards, S., Moennig, V. and Wensvoort, G., 1991. The development of an international reference panel of monoclonal antibodies for the differentiation of hog cholera virus from other pestiviruses. Vet. Microbiol., 29:101-108. Hewicker, M., W6hrmann, T., Fernandez, A., Trautwein, G., Liess, B. and Moennig, V., 1990. Immunohistological detection of bovine viral diarrhoea virus antigen in the central nervous system of persistently infected cattle using monoclonal antibodies. Vet. Microbiol., 23: 203210. Hyera, J.M.K., Leiss, B. and Frey, H.R., 1987. A direct neutralizing peroxidase-linked antibody assay for detection and titration of antibodies to bovine viral diarrhoea virus. J. Vet. Med. B 34, 227-239. Mengeling, W.L. and Packer, R.A., 1969. Pathogenesis of chronic hog cholera: host response. Am. J. Vet. Res., 30: 409-417. Meyers, G., Riimenapf, T. and Thiel, H-J., 1989. Molecular cloning and nucleotide sequence of the genome of hog cholera virus. Virology, 171: 555-567. Meyling, A., 1984. Detection of BVD virus in viremic cattle by an indirect immunoperoxidase technique. Curr. Top. Vet. Med. Anim. Sci., 29: 37-46. Moennig, V., 1990. Pestiviruses: a review. Vet. Microbiol., 23: 35-54. Moormann, R.J.M., Warmerdam, P.A.M., Van der Meer, B., Schaaper, W.M.M., Wensvoort, G. and Hulst, M.M., 1990. Molecular cloning and nucleotide sequence of hog cholera virus strain Brescia and mapping of the genomic region encoding envelope protein El. Virology, 177: 184-198. Shannon, A.D., 1991. An antigen-capture EL1SA for pestivirus detection in pigs and cattle. Report on Meeting of National Swine Fever Laboratories within the European Economic Communities, 27-28 May, 1991. Commission of the European Communities, DirectorateGeneral for Agriculture, VI/B/II.2; Publication No. VI/5458/91 -EN, pp. 63-65. Shannon, A.D., Richards, S.G., Kirkland, P.D. and Moyle, A., 1991. An antigen- capture ELISA detects pestivirus antigens in blood and tissues of immunotolerant carrier cattle. J. Virol. Methods, 34: 1-12.

248

A.D. SHANNON ET AL.

Smith, G.H., Collins, J.K., Carman, J. and Minocha, H.C., 1988. Detection of cytopathic and noncytopathic bovine viral diarrhea virus in cell culture with an immunoperoxidase test. J. Virol. Methods, 19: 319-324. Suga, M., KSnig, M., Saalmtiller, A., Reddehase, M.J. and Thiel, H-J., 1992. Pathogenesis of classical swine fever: B-lymphocyte deficiency caused by hog cholera virus. J. Virol., 66: 1171-1175.

Terpstra, C., 1991. Hog cholera: an update of present knowledge. Br. Vet. J., 147: 397-406. Terpstra, C. and Wensvoort, G., 1988. Natural infections of pigs with bovine viral diarrhoea virus associated with signs resembling swine fever. Res. Vet. Sci., 45:137-142. Terpstra C, Bloemraad, M. and Gielkens, A.L.J., 1984. The neutralizing peroxidase-linked assay for detection of antibody against swine fever virus. Vet. Microbiol., 9:113-120 Thiel, H-J., Stark, R., Weiland, E., RiJmenapf, T. and Meyers, G., 1991. Hog cholera virus: molecular composition of virions from a pestivirus. J. Virol., 65:4705-4712. Wensvoort, G., Terpstra, C., De Kluijver, E.P., Kragten, C. and Warnaar J.C., 1989. Antigenic differentiation of pestivirus strains with monoclonal antibodies against hog cholera virus. Vet. Microbiol., 21: 9-20. Wensvoort, G., Terpstra, C., Boonstra, J., Bloemraad, M. and Van Zaane, D., 1986. Production of monoclonal antibodies against swine fever virus and their use in laboratory diagnosis. Vet. Microbiol., 12: 101-108.