Veterinary Microbiology, 31 ( 1992 ) 89-99 Elsevier Science Publishers B.V., Amsterdam
89
Development of double sandwich ELISA for Clostridium perfringens beta and epsilon toxins Ahmed H. E1 Idrissi a and Gilbert E. Ward b aDOpartement de Microbiologie et Maladies Contagieuses, Institut Agronomique et VOt~rinaire Hassan II, B.P. 6202 Rabat-Instituts, Morocco bDepartment of Veterinary Pathobiology, University of Minnesota, St. Paul, MN 55108, USA (Accepted 30 August 1991 )
ABSTRACT El Idrissi, A. and Ward, G.E., 1992. Development of double sandwich ELISA for Clostridium perfringens beta and epsilon toxins. Vet. Microbiol., 31: 89-99. Specific, double-sandwich ELISAs for beta and epsilon toxins were developed by coating wells of microplates with specific sheep antitoxin IgG and using specific rabbit antitoxin IgG as detecting antibodies. The assay for beta toxin detected a minimum level of 8 ng/ml of purified toxin. The assay for epsilon toxin was capable of detecting 2 ng/ml of purified toxin. When applied to detect the toxin in intestinal contents using 50% fetal bovine serum as diluent the lowest amounts detected were about at least 30 ng/ml for beta toxin and 4 ng/ml for epsilon toxin. Clear differences in ELISA readings of both assays have been found between culture filtrates from toxin and non-toxin producing strains. These results suggested that both assays described in this study could detect their respective toxin in buffers, culture supernatants or in intestinal contents.
INTRODUCTION
Amongst the five types (A, B, C, D and E ) of Clostridium perfringens, types B, C and D in particular have been associated with enterotoxemias in sheep (Sterne, 1981 ). The major toxins produced by these types are alpha, beta and epsilon. The latter two are particularly important in laboratory diagnosis of sheep enterotoxemias. The most commonly used test to detect C. perfringens toxins is the mouse toxicity assay which is tedious and like all biological tests suffers several drawbacks. Consequently, enzyme linked immunosorbent assays (ELISA) have been developed in an effort to produce simpler, faster and more accurate in vitro tests (Weddell and Worthington, 1984; Martin et al., 1988). The reported procedures have limitations such as elaborate antibody and conjugate preparation protocols. The present study describes two double sandwich ELISA systems for spe0378-1135/92/$05.00
© 1992 Elsevier Science Publishers B.V. All rights reserved.
90
A.H. EL IDRISSI AND G.E. WARD
cific detection of C. perfringens beta and epsilon toxins in buffers, intestinal contents and culture supernatants. MATERIALS AND METHODS
Preparation of beta and epsilon antigens A purification procedure for beta toxin was made using the methods described by Sakurai and Duncan ( 1977 ) and Lawrence et al. ( 1978 ) with some modifications. Toxic cultures of C. perfringens type B (a strain known to produce mainly beta toxin) were kindly supplied by a local vaccine making company (BIOPHARMA, Rabat Morocco) where cultures were grown in automatic pH controlled fermenters. After 5 to 6 h growth, the culture was harvested and centrifuged at 15 000 g f o r 2 0 min at 4°C. A m m o n i u m sulphate ( 313 g/l) was added to the culture supernatant fluid and left overnight at 4 ° C. The precipitate was sedimented by centrifugation, dissolved in 0.01 M PBS pH 8 and dialyzed against the same buffer for 24 h. The a m m o n i u m sulphate fraction was applied to a DEAE-Sephacel (Pharmacia Fine Chemicals, Uppsala, Sweden ) column and an ionic gradient was developed with 0.4 M phosphate buffered saline (PBS) pH 8, running into 600 ml of 0.01 M PBS. The peak containing beta toxin was applied to a column containing commercial C. perfringens types A and D diagnostic sera (Wellcome Diagnostics, Research Triangle Park, NC, USA) coupled to Ch-Sepharose 48 (Pharmacia) according to the manufacturer's instructions. The effluent containing non-bound proteins was checked for beta toxin activity and homogeneity respectively by mouse toxicity and sodium dodecyl sulphate-polyacrylamide gel electrophoresis (SDS-PAGE) as described elsewhere (El Idrissi, 1991). Epsilon toxin was essentially produced as described by Habeeb ( 1969 ) and Worthington et al. ( 1973 ). Toxic cultures of C. perfringens type D were obtained as described above. The culture supernatant was fractionated with amm o n i u m sulphate (350 g/l) at 4 ° C. The precipitate was collected by centrifugation, and dissolved in 0.01 M phosphate buffer (pH 7.0) then dialyzed against the same buffer. The a m m o n i u m sulphate fraction was applied to a DEAE-Sephacel (Pharmacia) column which has been equilibrated with 0.005 M phosphate buffer, pH 7.2. The elution was made by the same buffer and the first peak was checked for toxin activity and homogeneity respectively by mouse assay and SDS-PAGE (El Idrissi, 1991 ).
Production of antibodies Beta and epsilon toxoids were prepared by mixing either antigen preparation ( 1 m g / m l ) with an equal volume of 0.45% formalin and incubating the mixture at 37 °C for 48 hours. The beta toxoid solution was mixed with equal volume of Freund's complete adjuvant and 4 ml were injected intramuscu-
DOUBLE SANDWICH ELISA FOR CLOSTR1DIUM PERFRINGENS TOXINS
91
larly in two different sites in a sheep which had been checked for the absence of beta toxin antibodies. Three weeks later, the sheep was given a similar injection followed by 4 weekly intravenous injections ( 1,2, 3 and 4 ml oftoxoid without adjuvant). The sheep was bled one week after the last injection. The same protocol was followed to produce epsilon toxin antibodies. Antisera were also prepared in rabbits. Four ml of the above mixtures (toxoid + adjuvant) were injected into two rabbits, each of which received 2 ml intramuscularly at different sites. Three weeks later, both were given an intravenous injection of 0.25 mg followed by another 0.5 mg of the toxin antigen without adjuvant. Animals were bled 10 days after the last injection. Immunoglobulins were fractionated from sheep and rabbit antisera by ammonium sulphate precipitation using the procedure of Hebert et al. (1973) and purified by DEAE-Sephacel chromatography according to the instructions of the manufacturer (Pharmacia). IgG-containing fractions were pooled and concentrated using a Minicon B 15 concentrator (Amicon, Corp. Lexington, Mass. USA). The specificity of all antibody preparations was tested against other clostridial antigens ( C. perfringens alpha toxin, C. septicum and C. novyi) by crossed immunoelectrophoresis and immunodiffusion tests (El Idrissi, 1991 ). Specific antibody preparations were stored at - 7 0 ° C until used.
ELISA supplies and reagents The ELISA was performed in 96-well disposable flat-bottomed plates (Immulon 1, Dynatech Laboratories, Inc. Chantilly VA, USA). Goat anti-rabbit IgG (whole molecule) peroxidase conjugate and o-phenylenediamine (OPD) were obtained from Sigma Chemical Co., St. Louis Mo, USA. Optimal dilutions of beta toxin and epsilon toxin antibodies to be used in the ELISAS were determined by checkerboard titrations. Optimal length of time and temperatures of incubation of coating antibodies, antigens and detecting antibodies were also determined for each assay.
ELISA procedure The procedure of Voller et al. (1980), was adopted with some modifications. Unless otherwise stated methods and reagents are as described by Voller etal. (1980).
ELISA for beta toxin. Microplates were coated with 200/~1 of sheep antibeta toxin IgG (25/tg/ml) in 0.05 M carbonate buffer (pH 9.6). After overnight incubation at 4°C, the coated plates were washed 5 times with phosphate buffered saline containing 0.05% of Tween 20 (PBS-T). Test samples of toxin (200/zl per well) diluted in PBS-T were then added to the wells, and the plates were incubated for 2 h at 37°C. After 5 washes with PBS-T, 200/A of a 1 : 2000 dilution of rabbit anti-beta toxin IgG in PBS-T was added to each
92
A.H. EL IDRISSI AND G.E. WARD
well. After incubation for 2 h at 37°C the plates were washed as described above and 2 0 0 / t l of goat anti-rabbit IgG-peroxidase conjugate diluted at 1:2000 in PBS-T was added to each well. The plates were incubated for 1 h at room temperature, washed as above and 100/~1 of o-phenylenediamine/ H202 was added to each well. The plates were incubated in the dark for 30 min at room temperature and the reaction was stopped by the addition of 100 ~1 of HC1 to each well. The absorbance was measured in a microplate reader (Titerteck Multiskan, Flow laboratories, McLean VA, USA) at 405 nm. Negative controls included coating the wells with normal sheep IgG in place of sheep anti-toxin IgG, the addition of PBS-T in place of beta toxin (blank) and the substitution of normal rabbit antibodies for the rabbit anti-toxin. Net A405 was determined by substracting the A4o5 of the blank from the A4o5 in wells with toxin containing solutions.
ELISA for epsilon toxin. The ELISA for epsilon toxin was performed in the same manner as described for beta toxin, with the following exceptions. The plates were coated with sheep anti-toxin IgG at 10/tg/ml. Rabbit anti-toxin IgG was used at a dilution of 1 : 8000 in PBS-T. Conjugate was added and the reaction was quantified as described above. Detection of toxins in intestinal contents. To demonstrate that the ELISA systems could detect beta and epsilon toxins in intestinal material, gut contents were extracted and spiked with toxins using the m e t h o d described by Weddell and Worthington ( 1984 ) with slight modifications. The toxin (20/~g in 1 ml of PBS-T) was added to intestinal content (9 ml) collected from a slaughtered healthy sheep which had been to be shown negative for both toxins by mouse assay. The intestinal specimen was vortexed well and incubated for 30 min at room temperature. The specimen was then centrifuged at 15 000 g for 30 min and the supernatant fluid was passed through 0.45/lm membrane filters (Millipore, Corp. Bedford MA, USA). The filtrate was diluted 1:4 in PBS containing 50% fetal bovine serum and 0.05% Tween 20 (FBS-T) and assayed in each ELISA system by using doubling dilutions of the extract in FBS-T. Non-spiked samples tested at 1 : 4 dilution in FBS-T were used as negative control. Similar doubling dilutions of the toxin in PBS-T were used as positive control. Detection of toxins in culture supernatants. The filtrates of culture supernatants were used for the detection of toxins by ELISA. Test strains were grown anaerobically in the trypticase-glucose-yeast (TGY) broth (Difco Labs, Detroit MI, USA) at 37°C for 6 h. The cultures were centrifuged ( 15 000 g for 20 rain), and the supernatants were filtrated through a 0.45/~m membrane filter (Millipore corp. ). The filtrates were checked for toxicity by the mouse assay and the minimal lethal dose (the highest dilution causing 100% mortal-
93
DOUBLE SANDWICH ELISA FOR CLOSTRIDIUM PERFRINGENS TOXINS
ity within 3 days) was determined. All test samples were diluted 1 : 4 in PBST and were then examined by both ELISAs as described above. The T G Y broth m e d i u m diluted at 1:4 in PBS-T was considered as negative control. Each supernatant sample was run in triplicate with two wells per sample. RESULTS
Preparation of antigens and antibodies The beta toxin preparation used for immunization was not homogeneous (it showed more than one band by SDS-PAGE analysis) but did not contain any trace of alpha toxin as shown by crossed immunoelectrophoresis and agar gel immunodiffusion tests (data not shown). However, the m e t h o d adopted to purify the epsilon toxin yielded a highly purified preparation. The resulting antibodies both in sheep and rabbit were monospecific and neutralized their respective toxins.
ELISA for beta and epsilon toxins Specific double sandwich ELISAs for beta and epsilon toxins were developed by coating wells o f microplates with specific sheep antitoxin IgG and using specific rabbit antitoxin IgG as detecting antibodies. The response of each toxin in its ELISA was dose dependent (Figures 1 and 2 ). The assay for beta toxin detected a m i n i m u m level of 8 n g / m l o f purified toxin. In all experiments, the linear segment occurred either between 31 and 250 n g / m l or more frequently between 31 and 125 n g / m l as shown in Fig. 1. The assay for 10
8 0.8
•
Control Positive Ne
J
E ~
0.6
D ~ 0.4' N_ © 0.2
0.0
7.;
15:6
31:2
62,5 125 Toxin (ng/ml)
250
500
Fig. 1. Detection of beta toxin in intestinal contents by the double-sandwich ELISA. Control (purified toxin in PBS). Positive sample=intestinal material spiked with beta toxin (20 ~g/ ml) and diluted 1:4 in 50% (FBS). Negative sample = non-spiked intestinal material diluted 1 : 4 in 50O/oFBS.
94
A.H. EL IDRISSIAND G.E. WARD 15
E
c
•
Control
•
Negative
1.0-
D
~o 0 5 O
0.0
. 1.9
. 39
.
. 7.8
. 15.6
. 31.2~
625
125
250
500
Toxin (ng/ml)
Fig. 2. Detection of epsilon toxin in intestinal contents by the double-sandwich ELISA. Control (purified toxin in PBS ). Positive sample = intestinal material spiked with epsilon toxin (20/~g/ m l ) a n d diluted 1:4 in 50% FBS. Negative sample = non-spiked intestinal material diluted 1:4 in 50% FBS. TABLE1 Reproducibility of beta toxin ELISA in replicates
Beta toxin (ng/ml) 1
Day
ELISA values _+SD (OD 405 n m )
Intraassay (%CV) 2
50
l 2 1 2
0.289_+0.037 0.268 _+0.030 0.581 _+0.044 0.603_+0.053
12.8 11.2 7.6 8.8
100
Interassay (%CV) 11.9 8.1
~Two beta toxin levels were assayed in replicates of 6 on two consecutive days. 2CV = Coefficient of variation.
epsilon toxin was capable of detecting 2 ng/ml of purified toxin and was linear between 7.8 and 125 ng/ml (Fig. 2). To estimate the intra- and inter-assay coefficient of variation (%CV) in each assay, 2 concentration levels of each toxin were tested in six replicates on two separate days using the same preparative lots of reagents. As seen in Table 1 the intra-assay and inter-assay %CV in the beta assay are below 15% for a concentration of 50 ng/ml and less than 10% for a concentration o f 100 ng/ml. In the epsilon assay all %CV estimates are less than 10% (Table 2 ). Detection of toxins in intestinal contents Using 50% FBS as diluent in both assays, quantitative analysis of the toxinintestinal contents mixture also gave typical dose response curves in both as-
DOUBLE SANDWICH ELISAFOR CLOSTRID1UMPERFRINGENSTOXINS
95
TABLE2 Reproducibility of epsilon toxin toxin ELISA Epsilon toxin (ng/ml)'
Day
50
1 2 1 2
100
ELISA values _+SD (OD405 n m )
Intraassay
Interassay
(%CV) 2
(%cv)
0.944 + 0.064 0.929-+ 0.051 1.187 _+0.069 1.128_+0.055
6.8 5.5 5.8 4.9
6.1 5.3
'Two epsilon toxin levels were assayed in replicates of 6 on two consecutive days. zCV = Coefficient of variation. TABLE3 ELISA values of culture supernatants of test strains (beta toxin assay) Strain
MLD/ml
ELISA values + SD
1520 160 2800
0.589 + 0.048 0.393 _+0.034 0.824 _+0.059
320 50 32
0.411 _+0.038 0.363 _+0.027 0.311 _+0.030
C. perfringens, type B ATCC 3626 NCTC 3110 Vaccinal strain C. perfringens type C ATCC 3628 CPTC 88/4 CPTC 89/4 C. perfringens type A ATCC 3629 C. perfringens type D ATCC 3629 TGY broth (negative control)
0.121 _+0.023
0
0.107 -+0.021 0.087 -+ 0.018
MLD = minimal lethal dose ATCC = American type collection culture NCTC = national collection of type culture,London. CPTC = Clostridium perfringens type C (laboratory strain)
says. Reactivity was relatively low, particularly in the beta assay as compared to control curves (Figs. 1 and 2). The lowest amounts detected were at least 30 ng/ml of beta toxin and 4 n g / m l of epsilon toxin. These amounts correspond to absorbances greater than the mean absorbance of a 1:4 dilution of negative samples (6 determinations) plus 3 standard deviations.
Detection of toxins in culture supernatants Filtrates of toxigenic strains of C. perfringens types B and D (positive test samples ) were examined by ELISA for the presence of beta and epsilon toxins respectively in comparison with non- or low-toxin producing strains (negative test samples) and TGY broth (negative control). The results shown in
96
A.H. El. 1DRISSI AND G.E. WARD
TABLE 4 E LISA values of culture supernatants of test strains (epsilon toxin assay)
Strain
MLD/ml
ELISA values +- SD
1020 650 6000 480 120 80
0.635 +_0.057 0.389_+ 0.045 1.182 _+0.069 0.311 _+0.033 0.241 +_0.030 0.222 _+0.034
320 140
0.291 _+0.031 0.238 _+0.028
C perfringens type D A T C C 3629 N C T C 8346
Vaccinal strain CPTD 88/4 CPTD 87/7 C P T D 89/11 C. pe
89
Development of double sandwich ELISA for Clostridium perfringens beta and epsilon toxins Ahmed H. E1 Idrissi a and Gilbert E. Ward b aDOpartement de Microbiologie et Maladies Contagieuses, Institut Agronomique et VOt~rinaire Hassan II, B.P. 6202 Rabat-Instituts, Morocco bDepartment of Veterinary Pathobiology, University of Minnesota, St. Paul, MN 55108, USA (Accepted 30 August 1991 )
ABSTRACT El Idrissi, A. and Ward, G.E., 1992. Development of double sandwich ELISA for Clostridium perfringens beta and epsilon toxins. Vet. Microbiol., 31: 89-99. Specific, double-sandwich ELISAs for beta and epsilon toxins were developed by coating wells of microplates with specific sheep antitoxin IgG and using specific rabbit antitoxin IgG as detecting antibodies. The assay for beta toxin detected a minimum level of 8 ng/ml of purified toxin. The assay for epsilon toxin was capable of detecting 2 ng/ml of purified toxin. When applied to detect the toxin in intestinal contents using 50% fetal bovine serum as diluent the lowest amounts detected were about at least 30 ng/ml for beta toxin and 4 ng/ml for epsilon toxin. Clear differences in ELISA readings of both assays have been found between culture filtrates from toxin and non-toxin producing strains. These results suggested that both assays described in this study could detect their respective toxin in buffers, culture supernatants or in intestinal contents.
INTRODUCTION
Amongst the five types (A, B, C, D and E ) of Clostridium perfringens, types B, C and D in particular have been associated with enterotoxemias in sheep (Sterne, 1981 ). The major toxins produced by these types are alpha, beta and epsilon. The latter two are particularly important in laboratory diagnosis of sheep enterotoxemias. The most commonly used test to detect C. perfringens toxins is the mouse toxicity assay which is tedious and like all biological tests suffers several drawbacks. Consequently, enzyme linked immunosorbent assays (ELISA) have been developed in an effort to produce simpler, faster and more accurate in vitro tests (Weddell and Worthington, 1984; Martin et al., 1988). The reported procedures have limitations such as elaborate antibody and conjugate preparation protocols. The present study describes two double sandwich ELISA systems for spe0378-1135/92/$05.00
© 1992 Elsevier Science Publishers B.V. All rights reserved.
90
A.H. EL IDRISSI AND G.E. WARD
cific detection of C. perfringens beta and epsilon toxins in buffers, intestinal contents and culture supernatants. MATERIALS AND METHODS
Preparation of beta and epsilon antigens A purification procedure for beta toxin was made using the methods described by Sakurai and Duncan ( 1977 ) and Lawrence et al. ( 1978 ) with some modifications. Toxic cultures of C. perfringens type B (a strain known to produce mainly beta toxin) were kindly supplied by a local vaccine making company (BIOPHARMA, Rabat Morocco) where cultures were grown in automatic pH controlled fermenters. After 5 to 6 h growth, the culture was harvested and centrifuged at 15 000 g f o r 2 0 min at 4°C. A m m o n i u m sulphate ( 313 g/l) was added to the culture supernatant fluid and left overnight at 4 ° C. The precipitate was sedimented by centrifugation, dissolved in 0.01 M PBS pH 8 and dialyzed against the same buffer for 24 h. The a m m o n i u m sulphate fraction was applied to a DEAE-Sephacel (Pharmacia Fine Chemicals, Uppsala, Sweden ) column and an ionic gradient was developed with 0.4 M phosphate buffered saline (PBS) pH 8, running into 600 ml of 0.01 M PBS. The peak containing beta toxin was applied to a column containing commercial C. perfringens types A and D diagnostic sera (Wellcome Diagnostics, Research Triangle Park, NC, USA) coupled to Ch-Sepharose 48 (Pharmacia) according to the manufacturer's instructions. The effluent containing non-bound proteins was checked for beta toxin activity and homogeneity respectively by mouse toxicity and sodium dodecyl sulphate-polyacrylamide gel electrophoresis (SDS-PAGE) as described elsewhere (El Idrissi, 1991). Epsilon toxin was essentially produced as described by Habeeb ( 1969 ) and Worthington et al. ( 1973 ). Toxic cultures of C. perfringens type D were obtained as described above. The culture supernatant was fractionated with amm o n i u m sulphate (350 g/l) at 4 ° C. The precipitate was collected by centrifugation, and dissolved in 0.01 M phosphate buffer (pH 7.0) then dialyzed against the same buffer. The a m m o n i u m sulphate fraction was applied to a DEAE-Sephacel (Pharmacia) column which has been equilibrated with 0.005 M phosphate buffer, pH 7.2. The elution was made by the same buffer and the first peak was checked for toxin activity and homogeneity respectively by mouse assay and SDS-PAGE (El Idrissi, 1991 ).
Production of antibodies Beta and epsilon toxoids were prepared by mixing either antigen preparation ( 1 m g / m l ) with an equal volume of 0.45% formalin and incubating the mixture at 37 °C for 48 hours. The beta toxoid solution was mixed with equal volume of Freund's complete adjuvant and 4 ml were injected intramuscu-
DOUBLE SANDWICH ELISA FOR CLOSTR1DIUM PERFRINGENS TOXINS
91
larly in two different sites in a sheep which had been checked for the absence of beta toxin antibodies. Three weeks later, the sheep was given a similar injection followed by 4 weekly intravenous injections ( 1,2, 3 and 4 ml oftoxoid without adjuvant). The sheep was bled one week after the last injection. The same protocol was followed to produce epsilon toxin antibodies. Antisera were also prepared in rabbits. Four ml of the above mixtures (toxoid + adjuvant) were injected into two rabbits, each of which received 2 ml intramuscularly at different sites. Three weeks later, both were given an intravenous injection of 0.25 mg followed by another 0.5 mg of the toxin antigen without adjuvant. Animals were bled 10 days after the last injection. Immunoglobulins were fractionated from sheep and rabbit antisera by ammonium sulphate precipitation using the procedure of Hebert et al. (1973) and purified by DEAE-Sephacel chromatography according to the instructions of the manufacturer (Pharmacia). IgG-containing fractions were pooled and concentrated using a Minicon B 15 concentrator (Amicon, Corp. Lexington, Mass. USA). The specificity of all antibody preparations was tested against other clostridial antigens ( C. perfringens alpha toxin, C. septicum and C. novyi) by crossed immunoelectrophoresis and immunodiffusion tests (El Idrissi, 1991 ). Specific antibody preparations were stored at - 7 0 ° C until used.
ELISA supplies and reagents The ELISA was performed in 96-well disposable flat-bottomed plates (Immulon 1, Dynatech Laboratories, Inc. Chantilly VA, USA). Goat anti-rabbit IgG (whole molecule) peroxidase conjugate and o-phenylenediamine (OPD) were obtained from Sigma Chemical Co., St. Louis Mo, USA. Optimal dilutions of beta toxin and epsilon toxin antibodies to be used in the ELISAS were determined by checkerboard titrations. Optimal length of time and temperatures of incubation of coating antibodies, antigens and detecting antibodies were also determined for each assay.
ELISA procedure The procedure of Voller et al. (1980), was adopted with some modifications. Unless otherwise stated methods and reagents are as described by Voller etal. (1980).
ELISA for beta toxin. Microplates were coated with 200/~1 of sheep antibeta toxin IgG (25/tg/ml) in 0.05 M carbonate buffer (pH 9.6). After overnight incubation at 4°C, the coated plates were washed 5 times with phosphate buffered saline containing 0.05% of Tween 20 (PBS-T). Test samples of toxin (200/zl per well) diluted in PBS-T were then added to the wells, and the plates were incubated for 2 h at 37°C. After 5 washes with PBS-T, 200/A of a 1 : 2000 dilution of rabbit anti-beta toxin IgG in PBS-T was added to each
92
A.H. EL IDRISSI AND G.E. WARD
well. After incubation for 2 h at 37°C the plates were washed as described above and 2 0 0 / t l of goat anti-rabbit IgG-peroxidase conjugate diluted at 1:2000 in PBS-T was added to each well. The plates were incubated for 1 h at room temperature, washed as above and 100/~1 of o-phenylenediamine/ H202 was added to each well. The plates were incubated in the dark for 30 min at room temperature and the reaction was stopped by the addition of 100 ~1 of HC1 to each well. The absorbance was measured in a microplate reader (Titerteck Multiskan, Flow laboratories, McLean VA, USA) at 405 nm. Negative controls included coating the wells with normal sheep IgG in place of sheep anti-toxin IgG, the addition of PBS-T in place of beta toxin (blank) and the substitution of normal rabbit antibodies for the rabbit anti-toxin. Net A405 was determined by substracting the A4o5 of the blank from the A4o5 in wells with toxin containing solutions.
ELISA for epsilon toxin. The ELISA for epsilon toxin was performed in the same manner as described for beta toxin, with the following exceptions. The plates were coated with sheep anti-toxin IgG at 10/tg/ml. Rabbit anti-toxin IgG was used at a dilution of 1 : 8000 in PBS-T. Conjugate was added and the reaction was quantified as described above. Detection of toxins in intestinal contents. To demonstrate that the ELISA systems could detect beta and epsilon toxins in intestinal material, gut contents were extracted and spiked with toxins using the m e t h o d described by Weddell and Worthington ( 1984 ) with slight modifications. The toxin (20/~g in 1 ml of PBS-T) was added to intestinal content (9 ml) collected from a slaughtered healthy sheep which had been to be shown negative for both toxins by mouse assay. The intestinal specimen was vortexed well and incubated for 30 min at room temperature. The specimen was then centrifuged at 15 000 g for 30 min and the supernatant fluid was passed through 0.45/lm membrane filters (Millipore, Corp. Bedford MA, USA). The filtrate was diluted 1:4 in PBS containing 50% fetal bovine serum and 0.05% Tween 20 (FBS-T) and assayed in each ELISA system by using doubling dilutions of the extract in FBS-T. Non-spiked samples tested at 1 : 4 dilution in FBS-T were used as negative control. Similar doubling dilutions of the toxin in PBS-T were used as positive control. Detection of toxins in culture supernatants. The filtrates of culture supernatants were used for the detection of toxins by ELISA. Test strains were grown anaerobically in the trypticase-glucose-yeast (TGY) broth (Difco Labs, Detroit MI, USA) at 37°C for 6 h. The cultures were centrifuged ( 15 000 g for 20 rain), and the supernatants were filtrated through a 0.45/~m membrane filter (Millipore corp. ). The filtrates were checked for toxicity by the mouse assay and the minimal lethal dose (the highest dilution causing 100% mortal-
93
DOUBLE SANDWICH ELISA FOR CLOSTRIDIUM PERFRINGENS TOXINS
ity within 3 days) was determined. All test samples were diluted 1 : 4 in PBST and were then examined by both ELISAs as described above. The T G Y broth m e d i u m diluted at 1:4 in PBS-T was considered as negative control. Each supernatant sample was run in triplicate with two wells per sample. RESULTS
Preparation of antigens and antibodies The beta toxin preparation used for immunization was not homogeneous (it showed more than one band by SDS-PAGE analysis) but did not contain any trace of alpha toxin as shown by crossed immunoelectrophoresis and agar gel immunodiffusion tests (data not shown). However, the m e t h o d adopted to purify the epsilon toxin yielded a highly purified preparation. The resulting antibodies both in sheep and rabbit were monospecific and neutralized their respective toxins.
ELISA for beta and epsilon toxins Specific double sandwich ELISAs for beta and epsilon toxins were developed by coating wells o f microplates with specific sheep antitoxin IgG and using specific rabbit antitoxin IgG as detecting antibodies. The response of each toxin in its ELISA was dose dependent (Figures 1 and 2 ). The assay for beta toxin detected a m i n i m u m level of 8 n g / m l o f purified toxin. In all experiments, the linear segment occurred either between 31 and 250 n g / m l or more frequently between 31 and 125 n g / m l as shown in Fig. 1. The assay for 10
8 0.8
•
Control Positive Ne
J
E ~
0.6
D ~ 0.4' N_ © 0.2
0.0
7.;
15:6
31:2
62,5 125 Toxin (ng/ml)
250
500
Fig. 1. Detection of beta toxin in intestinal contents by the double-sandwich ELISA. Control (purified toxin in PBS). Positive sample=intestinal material spiked with beta toxin (20 ~g/ ml) and diluted 1:4 in 50% (FBS). Negative sample = non-spiked intestinal material diluted 1 : 4 in 50O/oFBS.
94
A.H. EL IDRISSIAND G.E. WARD 15
E
c
•
Control
•
Negative
1.0-
D
~o 0 5 O
0.0
. 1.9
. 39
.
. 7.8
. 15.6
. 31.2~
625
125
250
500
Toxin (ng/ml)
Fig. 2. Detection of epsilon toxin in intestinal contents by the double-sandwich ELISA. Control (purified toxin in PBS ). Positive sample = intestinal material spiked with epsilon toxin (20/~g/ m l ) a n d diluted 1:4 in 50% FBS. Negative sample = non-spiked intestinal material diluted 1:4 in 50% FBS. TABLE1 Reproducibility of beta toxin ELISA in replicates
Beta toxin (ng/ml) 1
Day
ELISA values _+SD (OD 405 n m )
Intraassay (%CV) 2
50
l 2 1 2
0.289_+0.037 0.268 _+0.030 0.581 _+0.044 0.603_+0.053
12.8 11.2 7.6 8.8
100
Interassay (%CV) 11.9 8.1
~Two beta toxin levels were assayed in replicates of 6 on two consecutive days. 2CV = Coefficient of variation.
epsilon toxin was capable of detecting 2 ng/ml of purified toxin and was linear between 7.8 and 125 ng/ml (Fig. 2). To estimate the intra- and inter-assay coefficient of variation (%CV) in each assay, 2 concentration levels of each toxin were tested in six replicates on two separate days using the same preparative lots of reagents. As seen in Table 1 the intra-assay and inter-assay %CV in the beta assay are below 15% for a concentration of 50 ng/ml and less than 10% for a concentration o f 100 ng/ml. In the epsilon assay all %CV estimates are less than 10% (Table 2 ). Detection of toxins in intestinal contents Using 50% FBS as diluent in both assays, quantitative analysis of the toxinintestinal contents mixture also gave typical dose response curves in both as-
DOUBLE SANDWICH ELISAFOR CLOSTRID1UMPERFRINGENSTOXINS
95
TABLE2 Reproducibility of epsilon toxin toxin ELISA Epsilon toxin (ng/ml)'
Day
50
1 2 1 2
100
ELISA values _+SD (OD405 n m )
Intraassay
Interassay
(%CV) 2
(%cv)
0.944 + 0.064 0.929-+ 0.051 1.187 _+0.069 1.128_+0.055
6.8 5.5 5.8 4.9
6.1 5.3
'Two epsilon toxin levels were assayed in replicates of 6 on two consecutive days. zCV = Coefficient of variation. TABLE3 ELISA values of culture supernatants of test strains (beta toxin assay) Strain
MLD/ml
ELISA values + SD
1520 160 2800
0.589 + 0.048 0.393 _+0.034 0.824 _+0.059
320 50 32
0.411 _+0.038 0.363 _+0.027 0.311 _+0.030
C. perfringens, type B ATCC 3626 NCTC 3110 Vaccinal strain C. perfringens type C ATCC 3628 CPTC 88/4 CPTC 89/4 C. perfringens type A ATCC 3629 C. perfringens type D ATCC 3629 TGY broth (negative control)
0.121 _+0.023
0
0.107 -+0.021 0.087 -+ 0.018
MLD = minimal lethal dose ATCC = American type collection culture NCTC = national collection of type culture,London. CPTC = Clostridium perfringens type C (laboratory strain)
says. Reactivity was relatively low, particularly in the beta assay as compared to control curves (Figs. 1 and 2). The lowest amounts detected were at least 30 ng/ml of beta toxin and 4 n g / m l of epsilon toxin. These amounts correspond to absorbances greater than the mean absorbance of a 1:4 dilution of negative samples (6 determinations) plus 3 standard deviations.
Detection of toxins in culture supernatants Filtrates of toxigenic strains of C. perfringens types B and D (positive test samples ) were examined by ELISA for the presence of beta and epsilon toxins respectively in comparison with non- or low-toxin producing strains (negative test samples) and TGY broth (negative control). The results shown in
96
A.H. El. 1DRISSI AND G.E. WARD
TABLE 4 E LISA values of culture supernatants of test strains (epsilon toxin assay)
Strain
MLD/ml
ELISA values +- SD
1020 650 6000 480 120 80
0.635 +_0.057 0.389_+ 0.045 1.182 _+0.069 0.311 _+0.033 0.241 +_0.030 0.222 _+0.034
320 140
0.291 _+0.031 0.238 _+0.028
C perfringens type D A T C C 3629 N C T C 8346
Vaccinal strain CPTD 88/4 CPTD 87/7 C P T D 89/11 C. pe
