Veterinary Microbiology, 31 ( 1992 ) 169-180 Elsevier Science Publishers B.V., Amsterdam
169
Serological assay for swine erysipelas using nitrocellulose particles impregnated with an immunodominant 65 kDa antigen from
Erysipelothrix rhusiopathiae J.C. Chin, B. Turner and G.J. Eamens Immunology & Microbiology, Elizabeth Macarthur Agricultural Institute, NSW Agriculture & Fisheries, PMB 8, Camden NSW 2570, A ustralia (Accepted 16 October 1991 )
ABSTRACT Chin, J.C., Turner, B. and Eamens, G.J., 1992. Serological assay for swine erysipelas using nitrocellulose particles impregnated with an immunodominant 65 kDa antigen from Erysipelothrix rhusiopathiae. Vet. Microbiol., 31: 169-180. Pigs (n = 10) that were experimentally challenged with an arthritogenic isolate of Erysipelothrix rhusiopathiae (strain VRS 229; serotype 1a) developed arthritis in at least one of twelve major limb joints, lmmunoblots using sera obtained from these pigs at necropsy revealed a major band of immunoreactivity against a subunit polypeptide of apparent molecular mass 65 kDa. The usefulness of the 65 kDa immunodominant subunit as an assay reagent in an ELISA test was examined by presentation of antigen impregnated onto nitrocellulose particles (AINP). This was prepared by electrotransfer of bacterial polypeptides from SDS-PAGE gels to nitrocellulose. Protein bands were visualized by staining with amido black and a strip of nitrocellulose bearing the 65 kDa band was excised and extracted with formic acid. Nitrocellulose particles impregnated with the 65 kDa antigen (65AINP) were precipitated from solution by neutralization with ammonium hydroxide. 65-AINP was suspended in water and the optimum dilution for ELISA assay was determined by titration to be 0.1 A65o units. Sera from all pigs challenged with VRS 229 reacted against the 65-AINP antigen in the ELISA assay while sera from control, and experimental pigs prior to challenge, failed to do so. The 65-AINP antigen could also be used efficaciously to quantify serological reactivity of pigs experimentally infected with other strains ofE. rhusiopathiae representing the three major serotypes ( 1a, lb and 2 ) that are most commonly associated with swine erysipelas infections. Mouse immunizations with 65-AINP also confirmed that nitrocellulose particles bearing the immunodominant subunit antigen will elicit murine antibodies that are monospecific against this determinant.
INTRODUCTION
The advent of ELISA (enzyme linked immunosorbent assay) technology has greatly facilitated methods for the detection of antibodies against pathogenic organisms. A simple indirect ELISA test for example would involve: 0378-1135/92/$05.00
© 1992 Elsevier Science Publishers B.V. All rights reserved.
170
J.C. CHIN ET AL.
firstly, the binding of antigen to a solid-phase, followed by the addition of antisera and the corresponding enzyme-conjugated anti-species specific second antibody. The availability of a stable antigen devoid of contaminating and non-specific determinants is an important prerequisite for the development of a specific ELISA test. At present, many commercial kits for indirect or direct ELISA tests utilize native antigens (undenatured by heat and mercaptoethanol treatment) extracted from the pathogen of interest as the target antigen (Fister et al., 1989), or as a control positive reagent (Coleman et al., 1989 ). ELISA tests which use whole cells or organisms represent another example where primarily native epitopes are presented for binding (Chin and Plant, 1989). These antigens are not only easier to prepare but are also intended to present in situ conformational epitopes for antibody binding. In contrast, western blotting techniques (Towbin and Gordon, 1984) are designed to detect antibody binding to antigens which have been denatured to their respective subunits by boiling and mercaptoethanol treatment. While denaturation of an antigen may alter its structural or conformational integrity, it is conceivable that not all immunoreactive determinants are necessarily changed. If this were indeed the case, then it is more likely that antibodies binding to subunit antigens preferentially recognise component linear epitopes. The specificity of such antibodies against a linear array of amino acids can be further characterised by ELISA using peptide mosaics synthesized by solid phase technology (Maeji et al., 1990). Although immunodominant and immunoreactive subunit antigens can be readily detected by immunoblotting, it is frequently not possible with this technique to assay quantitatively, serum antibodies against a specific subunit antigen. However, nitrocellulose particles bearing subunit antigen have been extracted from immunoblots and used to stimulate lymphocytes in vitro (Abou-Zeid et al., 1987 ). Lo et al. (1990) demonstrated that ELISA assays can be conducted with protein blots bearing subunit antigen without the need for further extraction. This paper examines the feasibility of extracting an immunodominant and immunoreactive subunit antigen of Erysipelothrix rhusiopathiae (Chin and Eamens, 1986) from a nitrocellulose matrix following electrotransfer from SDS-PAGE gels, and the utility of the subunit antigen impregnated nitrocellulose particles (AINP) as an ELISA reagent. MATERIALS AND METHODS
Growth of bacteria An arthritogenic isolate of E. rhusiopathiae (VRS 229, serotype 1a) was cultured from lyophilised stock on blood agar prior to inoculation into horse meat infusion broth supplemented with proteose-peptone (2% w/v), glucose (0.2% w/v) and horse serum (10% v/v) (HMS broth; Chin and Eamens, 1986). Cultures were incubated for 16 h in 10% COz in air at 35°C, then
SEROLOGICAL ASSAY FOR SWINE ERYSIPELAS
171
killed by the addition of phenol to a final concentration of 1% (v/v). Cells were recovered by centrifugation (3000 g, 20 min) and washed thrice in phosphate-buffered saline (PBS; 25 mM phosphate, 0.15 M NaC1, pH 7.2). Washed bacteria were kept at - 7 0 ° C as a frozen pellet (approximately 50
mg). Pig challenge Seventeen 5-week-old Large White × Landrace pigs were purchased from a commercial piggery with no previous history of swine erysipelas infections. Ten pigs were then challenged (Group Ch) intravenously 6, 7 and 8 weeks later with 2, 2 and 1 ml respectively ofE. rhusiopathiae ( 108 cfu ml-1 ) suspended in PBS. The remaining pigs were not challenged (Group UnCh) and served as controls. These were kept in pens distant from Group Ch pigs. In additional experiments, groups of pigs (minimum n = 3 ) were also challenged by essentially the same procedure with strains (Chin and Eamens, 1986 ) representing serotype 1a (strains VRS 225 ); serotype 1b (strains VRS 226, VRS 293, VRS 223) and serotype 2 (strains VRS 227, VRS 252, VRS 228 ) respectively.
Blood collection and postmortem examinations Pigs were bled at weekly intervals starting from week 0 (ie. when they were 5 weeks old) and ending on week 13, at which time the animals were euthanized and necropsied. Serum processed from blood collected on week 13 is referred to in the text as terminal sera. Altogether, 12 joints were examined from each pig and these included the scapulo-humeral, humeroradio-ulnar, and carpal joints from each foreleg; the coxo-femoral, femoro-tibial and tarsal joints from each hindleg. Joints were considered to be arthritic if there was evidence of increased volume and turbidity of joint fluid and if the synovial membrane was grossly thickened. Synovial membrane of all joints was also cultured on selective medium (HMS agar supplemented with kanamycin, neomycin and vancomycin at 80, 10 and 5 mg/100 ml respectively) for the presence ofE. rhusiopathiae. Serotyping of isolates was performed according to Eamens et al. (1988).
SDS-PAGE Protein electrophoresis was conducted in a mini-gel system (Hoeffer, CA, USA) under reducing conditions as described by Laemmli (1970). Frozen bacteria (50 mg) were suspended in 200/tl of SDS/mercaptoethanol in Trisglycine buffer (pH 8.3) and heated at 100°C for 5 min. The suspension was centrifuged for 5 min in an Eppendorf centrifuge (Model 541 4S) and 10 pl aliquots of the supernatant were loaded onto each lane of a 10% (w/v) acrylamide running gel with a 4% (w/v) stacking gel. Preparative gels were run in a similar fashion except that 200/11 of the boiled supernatant was loaded di-
172
J.c. CHIN ET AL.
rectly onto the stacking gel formed without a comb. Electrophoresis was run at constant 120 V until the bromophenol blue tracking dye had migrated to the bottom of the gel. Where necessary, polypeptide bands in the acrylamide gel were visualized by staining with Coomassie Brilliant Blue (CBB).
Imm unoblotting Polypeptides were transferred electrophoretically from acrylamide gels to nitrocellulose (NC) paper in a BioRad Trans-Blot apparatus (Bio-Rad Laboratories) under conditions essentially similar to that described by Towbin et al., ( 1979 ) except that isopropanol was used instead of methanol. A vertical strip of the electroblot was excised for immunoblotting while polypeptides on the rest of the nitrocellulose were stained for 30 s in Amido Black [0. 1% (w/ v) in destaining solution (EtOH: Acetic acid: H 2 0 ; 3 : 1 : 6 v/v/v) ]. The stained electroblot was washed repeatedly in destaining solution followed by water until color fast. For immunoblotting, the vertical strip of nitrocellulose was first blocked by incubation in high salt-Tween [25 m M Tris-HC1, pH 8.9; 0.15 M NaC10.5% (v/v) Tween 20), washed in TSTw (25 m M Tris-HC1, pH 7.4; 0.015 M NaC1; 0.01% (v/v) Tween 20) and then reacted with serum diluted in TSTw. Antibody binding to antigens on the nitrocellulose was carried out with a Decaprobe apparatus (Hoefer, CA, USA). Porcine antibodies which bound to polypeptides on the nitrocellulose were located by reaction with an alkaline phosphatase conjugated monoclonal antibody with specificities against all porcine immunoglobulins (IgG, IgM and IgA). Murine Igs were detected with alkaline phosphatase-conjugated goat anti-mouse (BioRad). The immunoblot was aligned next to the electroblot and a horizontal amido black-stained band with corresponding apparent molecular mass 65 kDa, was excised.
Antigen impregnated nitrocellulose particles Nitrocellulose particles impregnated with the 65 kDa antigen (65-AINP) were prepared essentially as described by Judd (1988). Briefly, NC strips containing amido black-stained 65 kDa bands were treated with formic acid (26 M; AR Grade) ( 1 ml per 0.2 g NC strip). The bluish acid extract was removed and cooled on ice. AINP was then precipitated from the acid extract by slow, dropwise addition of 5 vols of a m m o n i u m hydroxide ( 17 M; AR Grad). The bluish-tinged precipitate was pelleted by centrifugation and washed exhaustively with water until the supernatant was colourless. A stock suspension of the fluffy pellet was then diluted to an arbitrary concentration of 4 A65 o and kept at 4°C in sterile water. Frequent sonication with a microprobe was necessary to attain a uniform suspension following prolonged storage. In some experiments aimed at estimating antibody binding to different subunit antigens of E. rhusiopathiae, preparative electroblots, after amido black staining, were cut horizontally into 20 strips (9 cm by 3 m m wide).
SEROLOGICAL ASSAYFOR SWINE ERYSIPELAS
17 3
These were then washed exhaustively in water before treatment with formic acid.
AINP ELISA For ELISA, AINP suspensions were diluted to 0.1 A65o units in TSM (TS containing 10% (v/v) methanol). 50/d aliquots were loaded into microtiter wells of a U-bottomed polyvinyl plate. The particles were sedimented by centrifugation for 10 min at 500 g. The supernatant was decanted and non-specific binding sites blocked by overnight incubation at 4 °C in blocking buffer (TS containing 2% (w/v) skim milk ). Plates were then centrifuged and washed twice with TSTw before the addition of serum for 1 h at 37°C. Plates were washed again before the addition of enzyme conjugate.
Mouse immunization A 1 ml suspension of 65-AINP (0.4 A65o) was emulsified with an equal volume of Freunds complete adjuvant. Each of 4 Balb/c mice were immunized intraperitoneally with 0.5 ml of the emulsion. Two further booster doses in incomplete Freunds adjuvant were given by the same route at fortnightly intervals. Blood was collected one week after the last injection from the retroorbital plexus of anaesthetized mice and serum antibody responses were monitored by ELISA against 65-AINP. RESULTS
Experimental challenge of pigs The response of pigs to intravenous challenge with E. rhusiopathiae is depicted in Table 1. All 10 pigs developed characteristic urticarial skin lesions between 2-5 d post-challenge. Lameness was also evident in some of the pigs between 2-4 weeks after challenge (weeks 8-10). At necropsy (week 13 ), all 10 challenged pigs revealed at least one affected joint per pig with increased volume and turbidity of joint fluid as well as a thickening of the synovial TABLE1 Analysis of arthritis and infected joints in unchallenged pigs and pigs challenged with E. rhusiopathiae Group
No. pigs with Arthritis
UnCh (n=7) Ch
Mean No. Joints ~/pig with Infection
Arthritis
Infection
1
0
0.3
0
10
10
2.3
2.1
(n=lO) 12 major limb joints were examined from each pig.
174
J.c. CHIN ET AL.
membrane. The mean number of arthritic joints per pig in the challenged group was 2.3. E. rhusiopathiae of homologous serotype was isolated from affected joints of challenged pigs. Only one of seven pigs in the unchallenged group showed synovitis (mean 0.3 arthritic joints per pig) but bacteria could not be cultured from the affected joint.
SDS-PAGE and immunoblotting Figure 1 (lane b) depicts the polypeptide banding profile ofE. rhusiopathiae following SDS-PAGE and staining with Coomassie brilliant blue. At least nine prominently stained bands with molecular sizes - 65, 63, 59, 53, 50, 47,
a
b
c
d
m
~ 5 i~i!~¸
Fig. 1. SDS-PAGE and immunoblot profiles ofErysipelothrix rhusiopathiae. Lane a depicts from top to bottom, polypeptide standards of apparent molecular mass -92, 66, 45, 31 and 21 kDa. Lane b represents polypeptides of E. rhusiopathiae resolved by SDS-PAGE and stained with Coomassie Brilliant Blue. Lane c depicts an immunoblot of a vertical strip cut from nitrocellulose af~.c~ electrotransfer fror~ a preparative SDS-PAGE gel. Pooled sera from Group Ch pigs were reacted with the electrotransfer at a dilution of 1 in 500. Lane d depicts a strip of electrophoretically transferred polypeptides after staining with amido black.
175
SEROLOGICAL ASSAY FOR SWINE ERYSIPELAS
41, 32 and 30.5 kDa were visible. The banding profile of E. rhusiopathiae after electrophoretic transfer to nitrocellulose and staining with amido black is shown in lane d. Although less intensely stained, the banding patterns on nitrocellulose were essentially similar to that seen in the polyacrylamide gel. Immunoreactive polypeptides on the unstained vertical nitrocellulose strip were identified by reaction with pooled terminal sera (week 13) collected from challenged pigs. Sera were pooled because preliminary analyses revealed that terminal serum from all 10 pigs yielded essentially similar immunoblot profiles. As shown in Fig. 1 (lane c), the most prominently reactive band corresponded to a 65 kDa band. Terminal sera from unchallenged pigs as well as pre-challenge sera from Group Ch pigs did not react against the 65 kDa subunit antigen in immunoblots.
ELISA reactivity of nitrocellulose particles bearing subunit antigen transferred by electroblotting The ELISA reactivity of nitrocellulose particles bearing subunit antigen was assessed by cutting an amido black-stained preparative blot into horizontal strips (9 cm across by 3 m m wide). Altogether, nitrocellulose particles bearing subunit antigen corresponding to each of 20 X 3 mm strips were extracted and used as ELISA reagents at a dilution of 0.075 A65 o units. As shown in Fig. 2, pooled sera from pigs challenged with E. rhusiopathiae showed strong reactivity against strips 4-6, and additional reactivities against strips 8 + 9; 11 + 12 ELISA O.D. (492 nm) 0.8 i
0.6
0.4
0.2
1
2
3
4
5
6
7 8 9 10 11 12 13 14 15 16 17 18 19 20 S t r i p No. (3 m m / s e g r n e n t ) Group UnCh
~
Group Ch
Fig. 2. ELISA response to antigen impregnated nitrocellulose particles prepared by electrotransfer to nitrocellulose from a preparative polyacrylamide gel. The nitrocellulose was cut into 3 mm strips after electrotransfer. AINPs were extracted from each strip and used as the assay reagent in an AINP-ELISA. Separately pooled sera from Group Ch and Group UnCh pigs were used in the analyses. Histograms represent mean values of triplicate determinations.
176
J.C. CHIN ET AL.
and 18 + 19. Pre-challenge bleeds failed to react and showed background values around 0.2 ELISA O.D. units. Similarly, terminal sera from unchallenged pigs failed to react against particulate antigens extracted from any of the strips. 65-AINP immunodominant subunit antigen as an ELISA reagent The usefulness of nitrocellulose particles impregnated with the 65 kDa subunit antigen (65-AINP) was then evaluated as an ELISA reagent. The optimal dilution of 65-AINP for ELISA was first established using pooled prechallenge and terminal Group Ch pig sera. Fig, 3 shows that the optimal dilution for a particular batch of 65 AINP was 0.1 A650units. Figure 4 shows the serological response of individual Group Ch pigs at weeks 0 and 13 against 65-AINP antigen coupled to ELISA plates at a dilution of 0.1 A65o. Pre-challenge responses were consistently at background levels while terminal sera showed strong reactivity ranging from 0.7-1.3 ELISA O.D. units. None of the sera tested from unchallenged pigs reacted against 65-AINP. Figure 5 depicts reactivity of sera from pigs which have been challenged with strains of E. rhusiopathiae representing the three major serotypes most frequently isolated from pigs with swine erysipelas. Strong reactivity was detected in all infected pigs irrespective of the strain or serotype of the challenge organism even though the 65-AINP antigen was derived from strain VRS 229 (serotype 1a). Immunogenicity of 65-AINP Figure 4 also depicts the serological response of four mice after three intraperitoneal doses of 65-AINP. Compared to pre-immune serum (0.2 ELISA ELISA O.D. (492 nm) 2
1.5
0.5 I
0.4
0.2
0.1 0.05 0.025 0.0125 65-AINP Turbidity (650 nm) [
: Group UnCh
~
0.006
0.001
Group Ch
Fig. 3. O p t i m i z a t i o n o f 65-AINP dilution for use in ELISA. Sera from each group of pigs were pooled.
177
SEROLOGICALASSAYFOR SWINEERYSIPELAS ELISA O.D. (492 nm) 1.4-~ 1:2
-
1
I i
0.8 0.6
.... m m
0.4-
........ .....
0.2-
iii
.......
.
.
.
.
.
o P1
~]
P2
P3
P4
P5
P6 P7 P8 P9 PIO M1 I n d i v i d u a l animal r e s p o n s e s
M2 M3 M4
Group Ch Prebleed
[~
Group Ch Terminal
Mice Prebleed
m
Mice Postvaccination
Fig. 4. ELISA response against nitrocellulose particles bearing denatured 65 kDa subunit antigen in individual Group Ch pigs ( P I - P 1 0 ) before and after experimental challenge with E. rhusiopathiae. The serological response of 4 mice previously hyperimmunized with 65-AINP is represented by M l-M4. 1.5 225
~
~
o.oN/ 0
la
229
226
la
lb
Serotype
227
lb
lb
of challenge
2
2
2
strain
Fig. 5. ELISA response of terminal sera collected from pigs (n = 3 ) which had been challenged with E. rhusiopathiae strains representing serotype la (VRS 225 and 229); serotype lb (VRS 226, 293 and 223) and serotype 2 (VRS 227, 252 and 228). The first bar represents the mean response of all pigs before experimental challenge. Each bar depicts the mean response + SEM.
O.D. units), all mice responded well to immunization (range of response from 0.7 to 1.2 ELISA O.D. units). DISCUSSION
Several antigenic preparations have been used for the ELISA detection of
178
J.C. CHIN ETAL.
porcine anti-E, rhusiopathiae antibodies. A high molecular weight carbohydrate-enriched and heat-stable antigen c o m m o n to serotypes 1 and 2 strains ofE. rhusiopathiae was used successfully as an ELISA reagent (Chin and Eamens 1986; Eamens et al., 1989). Others (Kirchhoff et al., 1985) have resorted to the use of SDS to generate a total lysate of the bacterium and used the crude extraction mixture directly in an ELISA test. Although it is likely that the antigens used in these assays may be partially denatured by heat or detergent, their identity has not been characterised. More recently, Lachmann and Deicher ( 1986 ) using immunoblotting procedures, identified several immunologically active proteins with molecular masses of 78, 72, 68 and 48 kDa, and a non-proteinaceous component with apparent mass 14-22 kDa. The present study has shown by immunoblotting that all challenged and infected pigs react against a major denatured subunit antigen of 65 kDa molecular mass. Immunoreactivity was retained even when the 65 kDa antigen was extracted from nitrocellulose and presented as subunit antigen impregnated particles (Fig. 2). This result permitted the use of 65-AINPs as an ELISA reagent for the detection ofanti-E, rhusiopathiae antibodies in challenged and control pigs. As shown in Fig. 4, terminal sera from challenged pigs reacted against 65-AINP in contrast to terminal sera from control pigs. Even the control pig that was culture-negative but showed signs of synovitis, did not react against 65-AINP. The versatility of 65-AINP as an ELISA reagent was confirmed when serum from pigs challenged with other strains of E. rhusiopathiae representing serotypes la, lb and 2 were shown to react against this subunit antigen (Fig. 5 ). It would therefore appear that the 65 kDa antigen is a c o m m o n determinant present in the three major serotypes of E. rhusiopathiae most frequently isolated from pigs with swine erysipelas (Eamens et al., 1988). As with all ELISA reagents, the assay antigen should first be titrated so that the o p t i m u m dilution may be established. This was found to range between 0.08-0.12 A65o units. It must be stressed that the choice of A65o is an arbitrary one and is dependent upon the absorbance of nitrocellulose particles bearing polypeptides stained with amido black. The A65 o measurement therefore estimates the density of nitrocellulose particles used in the assay and indirectly, the amount of subunit antigen bound to nitrocellulose particles. Nevertheless, provided that standard reference positive and negative sera are used, optimal AINP concentrations for ELISA assays can be readily established for different batches of subunit antigen. In agreement with Judd (1989), we have also found that the immunoreactivity and antigenicity of subunit determinants was not affected by staining with amido black. The use of subunit AINPs as an ELISA reagent represents a significant improvement to the use of crude antigenic preparations. For example, it has been recently reported that ovine antibodies are produced against subunit antigens from a cytosolic subfraction ofBrucella ovis (Chin and Pang-Turner
SEROLOGICAL ASSAY FOR SWINE ERYSIPELAS
179
1990) but remain unreactive against native antigens. This class of antibody reactivity also appears to correlate with the presence of a productive or active infection in affected rams. Conceptually, AINPs may therefore provide a more specific diagnosis than crude antigenic preparations. We have found that AINPs retain their activity even after 6 m o n t h s storage at 4 ° C. They can be easily transported into the field and used for serological assays. More importantly, AINPs may provide a facile m e t h o d for generating monospecific antisera against an i m m u n o d o m i n a n t subunit antigen after it has been identified by immunoblotting. This is exemplified by the response in mice i m m u n i z e d with 65-AINP. Spleen cells from one of the four mice have also been fused with S p 2 / 0 myelomas to produce hybridomas which excrete monoclonal antibodies specific against the 6 5 kDa antigen o f E. rhusiopathiae (unpublished results). The AINP procedure can therefore easily replace the use of slices of subunit antigen e m b e d d e d in polyacrylamide gels ( K n u d s e n 198 5 ) or surgical implantation of antigen loaded nitrocellulose strips (Coghlan and Hanausek, 1990) to generate subunit specific antibodies. ACKNOWLEDGEMENTS This work was supported in part by a grant (Project DAN 31 P) from the Australian Pig Industry Research Council and the New South Wales Swine C o m p e n s a t i o n Fund.
REFERENCES Abou-Zeid, C., Filley, E., Steel, J. and Rook, G.A.W., 1987. A simple new method for using antigens separated by polyacrylamidegel electrophoresis to stimulate lymphocytesin vitro after converting bands cut from western blots into antigen-bearing particles. J. Immunol. Methods, 8: 5-8. Coghlan, L.G. and Hanausek, M., 1990. Subcutaneous immunization of rabbits with nitrocellulose paper strips impregnated with microgramquantities of protein. J. Immunol. Methods, 129: 135-138. Coleman, P., Varitek, V., Mushawar, I.K., Marchlewicz,J., Safford, J., Hansen, J., Kurpiewski, G. and Grier, T., 1989. Test Pack Chlamydia, a new rapid assay for the direct detection of Chlamydia trachomatis. J. Clin. Microbiol., 27:2811-2814. Chin, J.C. and Eamens, G.J., 1986. Immunoreactivity of fractionated antigens obtained from an arthritogenic isolate ofErysipelothrix rhusiopathiae. Aust. Vet. J., 63: 355-358. Chin, J.C. and Plant, J., 1989. The temporal ELISAresponse of rams to Brucella ovis following experimental infection or vaccination. Res. Vet. Sci., 46: 73-78. Chin, J.C. and Pang-Turner, B., 1990. Profiles of serological reactivity against cytosolubleantigens ofBrucella ovis in experimentallyinfected rams. J. Clin. Microbiol., 28: 2647-2652. Eamens, G.J., Turner, M.J. and Catt, R.E., 1988. Serotypes of Erysipelothrix rhusiopathiae in Australian pigs, small ruminants, poultry, and captive wild birds and animals. Aust. Vet. J., 65: 249-252. Eamens, G.J., Chin, J.C. and Nicholls, P.J., 1989. Comparison of inoculation regimes for the
180
J.C. CHIN ET AL.
experimental production of swine erysipelas arthritis. I1. Serological findings in a gel diffusion precipitin test and enzyme-linked immunosorbent assay. Aust. Vet. J., 66:216-220. Fister, R.D., Weymouth, L.A., McLaughlin, J.C., Ryan, R.W. and Tilton, R.C., 1989. Comparative evaluation of three products for the detection ofBorrelia burgdoferi antibody in human serum. J. Clin. Microbiol., 27: 2834-2837. Judd, R.C., 1988. Purification of outer membrane proteins of the Gram-negative bacterium Neisseria gonorrhoeae. Anal. Biochem., 173:307-316. Kirchhoff, H., Dubenkropp, H., Kerlen, G., Steffens, H-W., Hermanns, W., Trautwein, G. and Bohm, K.H., 1985. Application of the indirect enzyme immunoassay for the detection of antibodies against Erysipelothrix rhusiopathiae. Vet. Microbiol., 10: 549- 559. Knudsen, K.A., 1985. Proteins transferred to nitrocellulose for use as immunogens. Anal. Biochem., 147: 285-288. Lachmann, P.G. and Deicher, H., 1986. Solubilization and characterization of surface antigenic components ofErysipelothrix rhusiopathiae T28. Infect. Immun., 52:818-822. Laemmli, U.K., 1970. Cleavage of structural proteins during assembly of the head of bacteriophage T4. Nature (Lond.), 277: 680-685. Lo, W-C., Miller, M.A. and Montelaro, R.C., 1990. Analysis of antibody reactivities in ELISA using protein blots as antigen substrates: S-ELISA. J. Immunol. Methods, 128:17-25. Maeji, N.J., Bray, A.M. and Geysen, H.M., 1990. Multi-pin peptide synthesis strategy for T cell determinant analysis. J. Immunol. Methods, 134: 23-33. Towbin, H. and Gordon, J., 1984. Immunoblotting and dot immunoblotting - current status and outlook. J. Immunol. Methods, 72:313-340. Towbin, H., Staehlin, T. and Gordon, J., 1979. Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheets: procedure and some applications. Proc. Natl. Acad. Sci. U.S.A., 76: 4350-4353.
169
Serological assay for swine erysipelas using nitrocellulose particles impregnated with an immunodominant 65 kDa antigen from
Erysipelothrix rhusiopathiae J.C. Chin, B. Turner and G.J. Eamens Immunology & Microbiology, Elizabeth Macarthur Agricultural Institute, NSW Agriculture & Fisheries, PMB 8, Camden NSW 2570, A ustralia (Accepted 16 October 1991 )
ABSTRACT Chin, J.C., Turner, B. and Eamens, G.J., 1992. Serological assay for swine erysipelas using nitrocellulose particles impregnated with an immunodominant 65 kDa antigen from Erysipelothrix rhusiopathiae. Vet. Microbiol., 31: 169-180. Pigs (n = 10) that were experimentally challenged with an arthritogenic isolate of Erysipelothrix rhusiopathiae (strain VRS 229; serotype 1a) developed arthritis in at least one of twelve major limb joints, lmmunoblots using sera obtained from these pigs at necropsy revealed a major band of immunoreactivity against a subunit polypeptide of apparent molecular mass 65 kDa. The usefulness of the 65 kDa immunodominant subunit as an assay reagent in an ELISA test was examined by presentation of antigen impregnated onto nitrocellulose particles (AINP). This was prepared by electrotransfer of bacterial polypeptides from SDS-PAGE gels to nitrocellulose. Protein bands were visualized by staining with amido black and a strip of nitrocellulose bearing the 65 kDa band was excised and extracted with formic acid. Nitrocellulose particles impregnated with the 65 kDa antigen (65AINP) were precipitated from solution by neutralization with ammonium hydroxide. 65-AINP was suspended in water and the optimum dilution for ELISA assay was determined by titration to be 0.1 A65o units. Sera from all pigs challenged with VRS 229 reacted against the 65-AINP antigen in the ELISA assay while sera from control, and experimental pigs prior to challenge, failed to do so. The 65-AINP antigen could also be used efficaciously to quantify serological reactivity of pigs experimentally infected with other strains ofE. rhusiopathiae representing the three major serotypes ( 1a, lb and 2 ) that are most commonly associated with swine erysipelas infections. Mouse immunizations with 65-AINP also confirmed that nitrocellulose particles bearing the immunodominant subunit antigen will elicit murine antibodies that are monospecific against this determinant.
INTRODUCTION
The advent of ELISA (enzyme linked immunosorbent assay) technology has greatly facilitated methods for the detection of antibodies against pathogenic organisms. A simple indirect ELISA test for example would involve: 0378-1135/92/$05.00
© 1992 Elsevier Science Publishers B.V. All rights reserved.
170
J.C. CHIN ET AL.
firstly, the binding of antigen to a solid-phase, followed by the addition of antisera and the corresponding enzyme-conjugated anti-species specific second antibody. The availability of a stable antigen devoid of contaminating and non-specific determinants is an important prerequisite for the development of a specific ELISA test. At present, many commercial kits for indirect or direct ELISA tests utilize native antigens (undenatured by heat and mercaptoethanol treatment) extracted from the pathogen of interest as the target antigen (Fister et al., 1989), or as a control positive reagent (Coleman et al., 1989 ). ELISA tests which use whole cells or organisms represent another example where primarily native epitopes are presented for binding (Chin and Plant, 1989). These antigens are not only easier to prepare but are also intended to present in situ conformational epitopes for antibody binding. In contrast, western blotting techniques (Towbin and Gordon, 1984) are designed to detect antibody binding to antigens which have been denatured to their respective subunits by boiling and mercaptoethanol treatment. While denaturation of an antigen may alter its structural or conformational integrity, it is conceivable that not all immunoreactive determinants are necessarily changed. If this were indeed the case, then it is more likely that antibodies binding to subunit antigens preferentially recognise component linear epitopes. The specificity of such antibodies against a linear array of amino acids can be further characterised by ELISA using peptide mosaics synthesized by solid phase technology (Maeji et al., 1990). Although immunodominant and immunoreactive subunit antigens can be readily detected by immunoblotting, it is frequently not possible with this technique to assay quantitatively, serum antibodies against a specific subunit antigen. However, nitrocellulose particles bearing subunit antigen have been extracted from immunoblots and used to stimulate lymphocytes in vitro (Abou-Zeid et al., 1987 ). Lo et al. (1990) demonstrated that ELISA assays can be conducted with protein blots bearing subunit antigen without the need for further extraction. This paper examines the feasibility of extracting an immunodominant and immunoreactive subunit antigen of Erysipelothrix rhusiopathiae (Chin and Eamens, 1986) from a nitrocellulose matrix following electrotransfer from SDS-PAGE gels, and the utility of the subunit antigen impregnated nitrocellulose particles (AINP) as an ELISA reagent. MATERIALS AND METHODS
Growth of bacteria An arthritogenic isolate of E. rhusiopathiae (VRS 229, serotype 1a) was cultured from lyophilised stock on blood agar prior to inoculation into horse meat infusion broth supplemented with proteose-peptone (2% w/v), glucose (0.2% w/v) and horse serum (10% v/v) (HMS broth; Chin and Eamens, 1986). Cultures were incubated for 16 h in 10% COz in air at 35°C, then
SEROLOGICAL ASSAY FOR SWINE ERYSIPELAS
171
killed by the addition of phenol to a final concentration of 1% (v/v). Cells were recovered by centrifugation (3000 g, 20 min) and washed thrice in phosphate-buffered saline (PBS; 25 mM phosphate, 0.15 M NaC1, pH 7.2). Washed bacteria were kept at - 7 0 ° C as a frozen pellet (approximately 50
mg). Pig challenge Seventeen 5-week-old Large White × Landrace pigs were purchased from a commercial piggery with no previous history of swine erysipelas infections. Ten pigs were then challenged (Group Ch) intravenously 6, 7 and 8 weeks later with 2, 2 and 1 ml respectively ofE. rhusiopathiae ( 108 cfu ml-1 ) suspended in PBS. The remaining pigs were not challenged (Group UnCh) and served as controls. These were kept in pens distant from Group Ch pigs. In additional experiments, groups of pigs (minimum n = 3 ) were also challenged by essentially the same procedure with strains (Chin and Eamens, 1986 ) representing serotype 1a (strains VRS 225 ); serotype 1b (strains VRS 226, VRS 293, VRS 223) and serotype 2 (strains VRS 227, VRS 252, VRS 228 ) respectively.
Blood collection and postmortem examinations Pigs were bled at weekly intervals starting from week 0 (ie. when they were 5 weeks old) and ending on week 13, at which time the animals were euthanized and necropsied. Serum processed from blood collected on week 13 is referred to in the text as terminal sera. Altogether, 12 joints were examined from each pig and these included the scapulo-humeral, humeroradio-ulnar, and carpal joints from each foreleg; the coxo-femoral, femoro-tibial and tarsal joints from each hindleg. Joints were considered to be arthritic if there was evidence of increased volume and turbidity of joint fluid and if the synovial membrane was grossly thickened. Synovial membrane of all joints was also cultured on selective medium (HMS agar supplemented with kanamycin, neomycin and vancomycin at 80, 10 and 5 mg/100 ml respectively) for the presence ofE. rhusiopathiae. Serotyping of isolates was performed according to Eamens et al. (1988).
SDS-PAGE Protein electrophoresis was conducted in a mini-gel system (Hoeffer, CA, USA) under reducing conditions as described by Laemmli (1970). Frozen bacteria (50 mg) were suspended in 200/tl of SDS/mercaptoethanol in Trisglycine buffer (pH 8.3) and heated at 100°C for 5 min. The suspension was centrifuged for 5 min in an Eppendorf centrifuge (Model 541 4S) and 10 pl aliquots of the supernatant were loaded onto each lane of a 10% (w/v) acrylamide running gel with a 4% (w/v) stacking gel. Preparative gels were run in a similar fashion except that 200/11 of the boiled supernatant was loaded di-
172
J.c. CHIN ET AL.
rectly onto the stacking gel formed without a comb. Electrophoresis was run at constant 120 V until the bromophenol blue tracking dye had migrated to the bottom of the gel. Where necessary, polypeptide bands in the acrylamide gel were visualized by staining with Coomassie Brilliant Blue (CBB).
Imm unoblotting Polypeptides were transferred electrophoretically from acrylamide gels to nitrocellulose (NC) paper in a BioRad Trans-Blot apparatus (Bio-Rad Laboratories) under conditions essentially similar to that described by Towbin et al., ( 1979 ) except that isopropanol was used instead of methanol. A vertical strip of the electroblot was excised for immunoblotting while polypeptides on the rest of the nitrocellulose were stained for 30 s in Amido Black [0. 1% (w/ v) in destaining solution (EtOH: Acetic acid: H 2 0 ; 3 : 1 : 6 v/v/v) ]. The stained electroblot was washed repeatedly in destaining solution followed by water until color fast. For immunoblotting, the vertical strip of nitrocellulose was first blocked by incubation in high salt-Tween [25 m M Tris-HC1, pH 8.9; 0.15 M NaC10.5% (v/v) Tween 20), washed in TSTw (25 m M Tris-HC1, pH 7.4; 0.015 M NaC1; 0.01% (v/v) Tween 20) and then reacted with serum diluted in TSTw. Antibody binding to antigens on the nitrocellulose was carried out with a Decaprobe apparatus (Hoefer, CA, USA). Porcine antibodies which bound to polypeptides on the nitrocellulose were located by reaction with an alkaline phosphatase conjugated monoclonal antibody with specificities against all porcine immunoglobulins (IgG, IgM and IgA). Murine Igs were detected with alkaline phosphatase-conjugated goat anti-mouse (BioRad). The immunoblot was aligned next to the electroblot and a horizontal amido black-stained band with corresponding apparent molecular mass 65 kDa, was excised.
Antigen impregnated nitrocellulose particles Nitrocellulose particles impregnated with the 65 kDa antigen (65-AINP) were prepared essentially as described by Judd (1988). Briefly, NC strips containing amido black-stained 65 kDa bands were treated with formic acid (26 M; AR Grade) ( 1 ml per 0.2 g NC strip). The bluish acid extract was removed and cooled on ice. AINP was then precipitated from the acid extract by slow, dropwise addition of 5 vols of a m m o n i u m hydroxide ( 17 M; AR Grad). The bluish-tinged precipitate was pelleted by centrifugation and washed exhaustively with water until the supernatant was colourless. A stock suspension of the fluffy pellet was then diluted to an arbitrary concentration of 4 A65 o and kept at 4°C in sterile water. Frequent sonication with a microprobe was necessary to attain a uniform suspension following prolonged storage. In some experiments aimed at estimating antibody binding to different subunit antigens of E. rhusiopathiae, preparative electroblots, after amido black staining, were cut horizontally into 20 strips (9 cm by 3 m m wide).
SEROLOGICAL ASSAYFOR SWINE ERYSIPELAS
17 3
These were then washed exhaustively in water before treatment with formic acid.
AINP ELISA For ELISA, AINP suspensions were diluted to 0.1 A65o units in TSM (TS containing 10% (v/v) methanol). 50/d aliquots were loaded into microtiter wells of a U-bottomed polyvinyl plate. The particles were sedimented by centrifugation for 10 min at 500 g. The supernatant was decanted and non-specific binding sites blocked by overnight incubation at 4 °C in blocking buffer (TS containing 2% (w/v) skim milk ). Plates were then centrifuged and washed twice with TSTw before the addition of serum for 1 h at 37°C. Plates were washed again before the addition of enzyme conjugate.
Mouse immunization A 1 ml suspension of 65-AINP (0.4 A65o) was emulsified with an equal volume of Freunds complete adjuvant. Each of 4 Balb/c mice were immunized intraperitoneally with 0.5 ml of the emulsion. Two further booster doses in incomplete Freunds adjuvant were given by the same route at fortnightly intervals. Blood was collected one week after the last injection from the retroorbital plexus of anaesthetized mice and serum antibody responses were monitored by ELISA against 65-AINP. RESULTS
Experimental challenge of pigs The response of pigs to intravenous challenge with E. rhusiopathiae is depicted in Table 1. All 10 pigs developed characteristic urticarial skin lesions between 2-5 d post-challenge. Lameness was also evident in some of the pigs between 2-4 weeks after challenge (weeks 8-10). At necropsy (week 13 ), all 10 challenged pigs revealed at least one affected joint per pig with increased volume and turbidity of joint fluid as well as a thickening of the synovial TABLE1 Analysis of arthritis and infected joints in unchallenged pigs and pigs challenged with E. rhusiopathiae Group
No. pigs with Arthritis
UnCh (n=7) Ch
Mean No. Joints ~/pig with Infection
Arthritis
Infection
1
0
0.3
0
10
10
2.3
2.1
(n=lO) 12 major limb joints were examined from each pig.
174
J.c. CHIN ET AL.
membrane. The mean number of arthritic joints per pig in the challenged group was 2.3. E. rhusiopathiae of homologous serotype was isolated from affected joints of challenged pigs. Only one of seven pigs in the unchallenged group showed synovitis (mean 0.3 arthritic joints per pig) but bacteria could not be cultured from the affected joint.
SDS-PAGE and immunoblotting Figure 1 (lane b) depicts the polypeptide banding profile ofE. rhusiopathiae following SDS-PAGE and staining with Coomassie brilliant blue. At least nine prominently stained bands with molecular sizes - 65, 63, 59, 53, 50, 47,
a
b
c
d
m
~ 5 i~i!~¸
Fig. 1. SDS-PAGE and immunoblot profiles ofErysipelothrix rhusiopathiae. Lane a depicts from top to bottom, polypeptide standards of apparent molecular mass -92, 66, 45, 31 and 21 kDa. Lane b represents polypeptides of E. rhusiopathiae resolved by SDS-PAGE and stained with Coomassie Brilliant Blue. Lane c depicts an immunoblot of a vertical strip cut from nitrocellulose af~.c~ electrotransfer fror~ a preparative SDS-PAGE gel. Pooled sera from Group Ch pigs were reacted with the electrotransfer at a dilution of 1 in 500. Lane d depicts a strip of electrophoretically transferred polypeptides after staining with amido black.
175
SEROLOGICAL ASSAY FOR SWINE ERYSIPELAS
41, 32 and 30.5 kDa were visible. The banding profile of E. rhusiopathiae after electrophoretic transfer to nitrocellulose and staining with amido black is shown in lane d. Although less intensely stained, the banding patterns on nitrocellulose were essentially similar to that seen in the polyacrylamide gel. Immunoreactive polypeptides on the unstained vertical nitrocellulose strip were identified by reaction with pooled terminal sera (week 13) collected from challenged pigs. Sera were pooled because preliminary analyses revealed that terminal serum from all 10 pigs yielded essentially similar immunoblot profiles. As shown in Fig. 1 (lane c), the most prominently reactive band corresponded to a 65 kDa band. Terminal sera from unchallenged pigs as well as pre-challenge sera from Group Ch pigs did not react against the 65 kDa subunit antigen in immunoblots.
ELISA reactivity of nitrocellulose particles bearing subunit antigen transferred by electroblotting The ELISA reactivity of nitrocellulose particles bearing subunit antigen was assessed by cutting an amido black-stained preparative blot into horizontal strips (9 cm across by 3 m m wide). Altogether, nitrocellulose particles bearing subunit antigen corresponding to each of 20 X 3 mm strips were extracted and used as ELISA reagents at a dilution of 0.075 A65 o units. As shown in Fig. 2, pooled sera from pigs challenged with E. rhusiopathiae showed strong reactivity against strips 4-6, and additional reactivities against strips 8 + 9; 11 + 12 ELISA O.D. (492 nm) 0.8 i
0.6
0.4
0.2
1
2
3
4
5
6
7 8 9 10 11 12 13 14 15 16 17 18 19 20 S t r i p No. (3 m m / s e g r n e n t ) Group UnCh
~
Group Ch
Fig. 2. ELISA response to antigen impregnated nitrocellulose particles prepared by electrotransfer to nitrocellulose from a preparative polyacrylamide gel. The nitrocellulose was cut into 3 mm strips after electrotransfer. AINPs were extracted from each strip and used as the assay reagent in an AINP-ELISA. Separately pooled sera from Group Ch and Group UnCh pigs were used in the analyses. Histograms represent mean values of triplicate determinations.
176
J.C. CHIN ET AL.
and 18 + 19. Pre-challenge bleeds failed to react and showed background values around 0.2 ELISA O.D. units. Similarly, terminal sera from unchallenged pigs failed to react against particulate antigens extracted from any of the strips. 65-AINP immunodominant subunit antigen as an ELISA reagent The usefulness of nitrocellulose particles impregnated with the 65 kDa subunit antigen (65-AINP) was then evaluated as an ELISA reagent. The optimal dilution of 65-AINP for ELISA was first established using pooled prechallenge and terminal Group Ch pig sera. Fig, 3 shows that the optimal dilution for a particular batch of 65 AINP was 0.1 A650units. Figure 4 shows the serological response of individual Group Ch pigs at weeks 0 and 13 against 65-AINP antigen coupled to ELISA plates at a dilution of 0.1 A65o. Pre-challenge responses were consistently at background levels while terminal sera showed strong reactivity ranging from 0.7-1.3 ELISA O.D. units. None of the sera tested from unchallenged pigs reacted against 65-AINP. Figure 5 depicts reactivity of sera from pigs which have been challenged with strains of E. rhusiopathiae representing the three major serotypes most frequently isolated from pigs with swine erysipelas. Strong reactivity was detected in all infected pigs irrespective of the strain or serotype of the challenge organism even though the 65-AINP antigen was derived from strain VRS 229 (serotype 1a). Immunogenicity of 65-AINP Figure 4 also depicts the serological response of four mice after three intraperitoneal doses of 65-AINP. Compared to pre-immune serum (0.2 ELISA ELISA O.D. (492 nm) 2
1.5
0.5 I
0.4
0.2
0.1 0.05 0.025 0.0125 65-AINP Turbidity (650 nm) [
: Group UnCh
~
0.006
0.001
Group Ch
Fig. 3. O p t i m i z a t i o n o f 65-AINP dilution for use in ELISA. Sera from each group of pigs were pooled.
177
SEROLOGICALASSAYFOR SWINEERYSIPELAS ELISA O.D. (492 nm) 1.4-~ 1:2
-
1
I i
0.8 0.6
.... m m
0.4-
........ .....
0.2-
iii
.......
.
.
.
.
.
o P1
~]
P2
P3
P4
P5
P6 P7 P8 P9 PIO M1 I n d i v i d u a l animal r e s p o n s e s
M2 M3 M4
Group Ch Prebleed
[~
Group Ch Terminal
Mice Prebleed
m
Mice Postvaccination
Fig. 4. ELISA response against nitrocellulose particles bearing denatured 65 kDa subunit antigen in individual Group Ch pigs ( P I - P 1 0 ) before and after experimental challenge with E. rhusiopathiae. The serological response of 4 mice previously hyperimmunized with 65-AINP is represented by M l-M4. 1.5 225
~
~
o.oN/ 0
la
229
226
la
lb
Serotype
227
lb
lb
of challenge
2
2
2
strain
Fig. 5. ELISA response of terminal sera collected from pigs (n = 3 ) which had been challenged with E. rhusiopathiae strains representing serotype la (VRS 225 and 229); serotype lb (VRS 226, 293 and 223) and serotype 2 (VRS 227, 252 and 228). The first bar represents the mean response of all pigs before experimental challenge. Each bar depicts the mean response + SEM.
O.D. units), all mice responded well to immunization (range of response from 0.7 to 1.2 ELISA O.D. units). DISCUSSION
Several antigenic preparations have been used for the ELISA detection of
178
J.C. CHIN ETAL.
porcine anti-E, rhusiopathiae antibodies. A high molecular weight carbohydrate-enriched and heat-stable antigen c o m m o n to serotypes 1 and 2 strains ofE. rhusiopathiae was used successfully as an ELISA reagent (Chin and Eamens 1986; Eamens et al., 1989). Others (Kirchhoff et al., 1985) have resorted to the use of SDS to generate a total lysate of the bacterium and used the crude extraction mixture directly in an ELISA test. Although it is likely that the antigens used in these assays may be partially denatured by heat or detergent, their identity has not been characterised. More recently, Lachmann and Deicher ( 1986 ) using immunoblotting procedures, identified several immunologically active proteins with molecular masses of 78, 72, 68 and 48 kDa, and a non-proteinaceous component with apparent mass 14-22 kDa. The present study has shown by immunoblotting that all challenged and infected pigs react against a major denatured subunit antigen of 65 kDa molecular mass. Immunoreactivity was retained even when the 65 kDa antigen was extracted from nitrocellulose and presented as subunit antigen impregnated particles (Fig. 2). This result permitted the use of 65-AINPs as an ELISA reagent for the detection ofanti-E, rhusiopathiae antibodies in challenged and control pigs. As shown in Fig. 4, terminal sera from challenged pigs reacted against 65-AINP in contrast to terminal sera from control pigs. Even the control pig that was culture-negative but showed signs of synovitis, did not react against 65-AINP. The versatility of 65-AINP as an ELISA reagent was confirmed when serum from pigs challenged with other strains of E. rhusiopathiae representing serotypes la, lb and 2 were shown to react against this subunit antigen (Fig. 5 ). It would therefore appear that the 65 kDa antigen is a c o m m o n determinant present in the three major serotypes of E. rhusiopathiae most frequently isolated from pigs with swine erysipelas (Eamens et al., 1988). As with all ELISA reagents, the assay antigen should first be titrated so that the o p t i m u m dilution may be established. This was found to range between 0.08-0.12 A65o units. It must be stressed that the choice of A65o is an arbitrary one and is dependent upon the absorbance of nitrocellulose particles bearing polypeptides stained with amido black. The A65 o measurement therefore estimates the density of nitrocellulose particles used in the assay and indirectly, the amount of subunit antigen bound to nitrocellulose particles. Nevertheless, provided that standard reference positive and negative sera are used, optimal AINP concentrations for ELISA assays can be readily established for different batches of subunit antigen. In agreement with Judd (1989), we have also found that the immunoreactivity and antigenicity of subunit determinants was not affected by staining with amido black. The use of subunit AINPs as an ELISA reagent represents a significant improvement to the use of crude antigenic preparations. For example, it has been recently reported that ovine antibodies are produced against subunit antigens from a cytosolic subfraction ofBrucella ovis (Chin and Pang-Turner
SEROLOGICAL ASSAY FOR SWINE ERYSIPELAS
179
1990) but remain unreactive against native antigens. This class of antibody reactivity also appears to correlate with the presence of a productive or active infection in affected rams. Conceptually, AINPs may therefore provide a more specific diagnosis than crude antigenic preparations. We have found that AINPs retain their activity even after 6 m o n t h s storage at 4 ° C. They can be easily transported into the field and used for serological assays. More importantly, AINPs may provide a facile m e t h o d for generating monospecific antisera against an i m m u n o d o m i n a n t subunit antigen after it has been identified by immunoblotting. This is exemplified by the response in mice i m m u n i z e d with 65-AINP. Spleen cells from one of the four mice have also been fused with S p 2 / 0 myelomas to produce hybridomas which excrete monoclonal antibodies specific against the 6 5 kDa antigen o f E. rhusiopathiae (unpublished results). The AINP procedure can therefore easily replace the use of slices of subunit antigen e m b e d d e d in polyacrylamide gels ( K n u d s e n 198 5 ) or surgical implantation of antigen loaded nitrocellulose strips (Coghlan and Hanausek, 1990) to generate subunit specific antibodies. ACKNOWLEDGEMENTS This work was supported in part by a grant (Project DAN 31 P) from the Australian Pig Industry Research Council and the New South Wales Swine C o m p e n s a t i o n Fund.
REFERENCES Abou-Zeid, C., Filley, E., Steel, J. and Rook, G.A.W., 1987. A simple new method for using antigens separated by polyacrylamidegel electrophoresis to stimulate lymphocytesin vitro after converting bands cut from western blots into antigen-bearing particles. J. Immunol. Methods, 8: 5-8. Coghlan, L.G. and Hanausek, M., 1990. Subcutaneous immunization of rabbits with nitrocellulose paper strips impregnated with microgramquantities of protein. J. Immunol. Methods, 129: 135-138. Coleman, P., Varitek, V., Mushawar, I.K., Marchlewicz,J., Safford, J., Hansen, J., Kurpiewski, G. and Grier, T., 1989. Test Pack Chlamydia, a new rapid assay for the direct detection of Chlamydia trachomatis. J. Clin. Microbiol., 27:2811-2814. Chin, J.C. and Eamens, G.J., 1986. Immunoreactivity of fractionated antigens obtained from an arthritogenic isolate ofErysipelothrix rhusiopathiae. Aust. Vet. J., 63: 355-358. Chin, J.C. and Plant, J., 1989. The temporal ELISAresponse of rams to Brucella ovis following experimental infection or vaccination. Res. Vet. Sci., 46: 73-78. Chin, J.C. and Pang-Turner, B., 1990. Profiles of serological reactivity against cytosolubleantigens ofBrucella ovis in experimentallyinfected rams. J. Clin. Microbiol., 28: 2647-2652. Eamens, G.J., Turner, M.J. and Catt, R.E., 1988. Serotypes of Erysipelothrix rhusiopathiae in Australian pigs, small ruminants, poultry, and captive wild birds and animals. Aust. Vet. J., 65: 249-252. Eamens, G.J., Chin, J.C. and Nicholls, P.J., 1989. Comparison of inoculation regimes for the
180
J.C. CHIN ET AL.
experimental production of swine erysipelas arthritis. I1. Serological findings in a gel diffusion precipitin test and enzyme-linked immunosorbent assay. Aust. Vet. J., 66:216-220. Fister, R.D., Weymouth, L.A., McLaughlin, J.C., Ryan, R.W. and Tilton, R.C., 1989. Comparative evaluation of three products for the detection ofBorrelia burgdoferi antibody in human serum. J. Clin. Microbiol., 27: 2834-2837. Judd, R.C., 1988. Purification of outer membrane proteins of the Gram-negative bacterium Neisseria gonorrhoeae. Anal. Biochem., 173:307-316. Kirchhoff, H., Dubenkropp, H., Kerlen, G., Steffens, H-W., Hermanns, W., Trautwein, G. and Bohm, K.H., 1985. Application of the indirect enzyme immunoassay for the detection of antibodies against Erysipelothrix rhusiopathiae. Vet. Microbiol., 10: 549- 559. Knudsen, K.A., 1985. Proteins transferred to nitrocellulose for use as immunogens. Anal. Biochem., 147: 285-288. Lachmann, P.G. and Deicher, H., 1986. Solubilization and characterization of surface antigenic components ofErysipelothrix rhusiopathiae T28. Infect. Immun., 52:818-822. Laemmli, U.K., 1970. Cleavage of structural proteins during assembly of the head of bacteriophage T4. Nature (Lond.), 277: 680-685. Lo, W-C., Miller, M.A. and Montelaro, R.C., 1990. Analysis of antibody reactivities in ELISA using protein blots as antigen substrates: S-ELISA. J. Immunol. Methods, 128:17-25. Maeji, N.J., Bray, A.M. and Geysen, H.M., 1990. Multi-pin peptide synthesis strategy for T cell determinant analysis. J. Immunol. Methods, 134: 23-33. Towbin, H. and Gordon, J., 1984. Immunoblotting and dot immunoblotting - current status and outlook. J. Immunol. Methods, 72:313-340. Towbin, H., Staehlin, T. and Gordon, J., 1979. Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheets: procedure and some applications. Proc. Natl. Acad. Sci. U.S.A., 76: 4350-4353.
