- Email: [email protected]
Evaluation of a serological test system for the diagnosis of Sarcocystis cruzi infection in cattle using S. cruzi merozoite antigen
Veterinary Parasitology, 51 ( 1994 ) 181-189 0304-4017/94/$07.00 © 1994 - Elsevier Science B.V. All fights reserved
181
Evaluation of a serological test system for the diagnosis of Sarcocystis cruzi infection in cattle using S. cruzi merozoite antigen G. Savini*, J.D. Dunsmore, I.D. Robertson School of Veterinary Studies, Murdoch University, Murdoch, W.A. 6150, Australia (Accepted I 1 May 1993)
Abstract Fifty serum samples were examined by enzyme-linkedimmunosorbentassay (ELISA) for antibodies directed against crude antigens isolated from cystozoites and merozoites of Sarcocystis cruzi and Sarcocystis tenella/Sarcocystis arieticanis and merozoites of Toxoplasma gondii. Of these 50 samples, 25 were from cattle originatingfrom an area where the prevalence ofS. cruzi was found to be very low (9%) and in which no cystozoites were detected, and 25 were from cattle which were found by a digestion method to be heavily infected with S. cruzi. A very high correlation was observed between the parasitological data and the results obtained from the serological assays which used antigens from either cystozoites or merozoites of S. cruzi. The assay using the antigen derived from merozoites provided the best result for discriminating infected and non-infectedanimals. There was some cross-reactivity between the antigen derived from cystozoites of heterologous species of Sarcocystis and S. cruzi antibodies, and some cross-reactivity between antigen of T. gondii and antibodies to S. cruzi. The reproducibility of the assays was found to be high and similar results were observed when the sera were tested on two separate occasions. The unpurified S. cruzi merozoite antigen produced in vitro is relatively accurate in discriminating positive and negative animals and may be used for diagnosis in economically important hosts such as cattle and sheep. Key words: Sarcocystis cruzi; Cattle-Protozoa; Diagnosis-Protozoa
Introduction The cyst-forming coccidian parasites Toxoplasma gondii and Sarcocystis spp. are intraceUular protozoan organisms occurring in domestic animals and humans throughout the word. The parasites belong to the same family, Sarcocystidae (Levine, 1986) and have life cycles in which carnivores (predator) are the definitive hosts and herbivores (prey) the intermediate hosts. The importance of infection by T. gondii during pregnancy to the developing embryo has been recognised for some time in women as well as in domes*Corresponding author,
SSDI 0 3 0 4 - 4 0 1 7 ( 93 ) 0 0 5 4 1 - 6
182
G. Savini et al. / Veterinary Parasitology 51 (1994) 181-189
tic animals such as sheep and goats. However, the clinical and subclinical effects of Sarcocystis infections in domestic animals as well as the potential pathogenicity of Sarcocystis hominis and Sarcocystis suihominis, which require cattle and pigs, respectively, as intermediate hosts, are still being elucidated. Direct diagnostic methods such as digestion and histology, which readily detect the parasite in muscle tissues, have shown that Sarcocystis infection is very common in cattle worldwide. The infection has also been seen to stimulate the production of circulating antibodies which can be detected by various serological methods. In contrast, T. gondii has rarely been isolated from bovine tissues, although the frequent demonstration of antibodies in serum suggests that inapparent infection may be common (Dubey, 1986 ). It is accepted that soluble cystozoite antigens, prepared from cysts of Sarcocystis spp., react with heterologous species (O'Donoghue and Weyreter, 1983 ), although homologous species usually give rise to stronger reactions (O'Donoghue and Weyreter, 1984 ). The Sarcocystis antibodies that are present during the acute phase of infection with Sarcocystis cruzi in calves have been shown to cross-react with crude somatic Toxoplasma antigen prepared by either freeze-thawing or ultrasonication of merozoites (Uggla et al., 1987 ). The recent development of an in vitro system for growing numerous merozoites of Sarcocystis has made possible the preparation of new crude antigen from this stage of the parasite. The present study assesses and compares the performance of an enzyme-linked immunosorbent assay (ELISA) using crude antigens derived from merozoites and cystozoites of S. cruzi and Sarcocystis tenella a n d / o r Sarcocystis arieticanis and from merozoites of T. gondii for detecting antibodies to S. cruzi in the sera of cattle naturally infected with this parasite. Materials and methods
Sample collection Blood samples were collected from a total of 294 cattle slaughtered at three different abattoirs in Western Australia. Serum was obtained by centrifugation and stored in aliquots at - 20 ° C until use. A segment of muscle tissue (oesophagus) was collected from each carcass, subjected to pepsin digestion and examined microscopically for the presence of S. cruzi cystozoites and T. gondii bradyzoites as described by Bfttner et al. ( 1987 ) and Dubey and Beattie ( 1988 ).
Antigen from cystozoites ofS. cruzi An isotonic digest medium was prepared as follows: 2.5 g of pepsin ( 1 : 10 000) was dissolved in 10 ml of distilled water, centrifuged at 2000 rev
G. Savini et al. / Veterinary Parasitology 51 (1994) 181-189
183
min-~ and the supernatant solution containing dissolved pepsin collected. Sodium chloride (8.5 g) and 8 ml of concentrated hydrochloric acid ( 14 N) were added and the solution was made up to 1 I with distilled water. A heavily infected bovine heart was t r i m m e d of fat and minced in a blender for a few seconds at top speed and 500 g were mixed with 2 1 of the digest m e d i u m in a water bath at 40 ° C for 40 min. The mixture was filtered through a 250/tin metallic sieve. The filtrate was centrifuged at 2400 rev m i n - 1 for 5 min and the pellet was washed three times in normal saline solution. The washed pellet was suspended in saline solution in a ratio of 1 : 3 and the pH adjusted to 7 with a solution of 10% sodium bicarbonate in distilled water. The suspension was mixed in a ratio of 1:2 with Percoll isotonic solution (Pharmacia, Uppsala, Sweden), and centrifuged in a conical centrifuge tube at 2400 rev min-1 for 10 min at room temperature. The pellet containing cystozoites was washed three times and suspended in normal saline solution. The cystozoites in suspension were then frozen at - 2 0 ° C, thawed four times and ultrasonicated three times for 20 s each time at 100 W. The suspension was centrifuged at 15 000 g for 30 min. The supernatant was then collected and used as antigen.
Antigen from cystozoites ofS. tenella and~or S. arieticanis The antigen was prepared from two heavily infected hearts of rams using the same procedure employed for S. cruzi.
Antigen from merozoites ofS. cruzi Sporocysts of S. cruzi were obtained by scraping the mucosa of the small intestine of dogs 15 days after they had ingested S. cruzi infected cattle meat and were shown to be passing sporocysts. Sporocysts were purified, concentrated and stored at 4°C in Hank's balanced salt solution (HBSS) with antibiotics before use as described by Dubey et al. (1989). Prior to excystation, sporocysts were washed twice by centrifugation in HBSS, treated with 2.6% sodium hypochlorite in HBSS for 30 min at 4°C and washed four times in HBSS. Excystation was achieved by incubating the sporocysts for 4-6 h in HBSS containing 5% bovine sodium taurocholate, 2% trypsin, 1.8 ml of a solution of 0.14 M sodium hydrogen carbonate in distilled water. Excysted sporozoites were washed and passed through a nylon wool column (Robbins Scientific, Sunnyvale, CA) to remove the debris and sporocyst walls. Sporozoites were suspended in culture m e d i u m (CM) consisting of RPMI 1640 with L-glutamine (Flow Laboratories, Irvine, U K ) plus 5% foetal calf serum, 50 U m l - ~ penicillin G and 50 #g m l - ~ dihydrostreptomycin (Sigma Chemicals, St. Louis, MO). Each of four 25 cm 2 tissue culture flasks containing a monolayer of bovine pulmonary artery endothelial cells (American Type Culture Collection, Rockville, M D ) was inoculated with 5 × 105 sporozoites and incubated at
184
G. Savini et al. / Veterinary Parasitology 51 (1994) 181-189
37°C under 5% CO2 and 95% air in a continuous flow C O 2 incubator. At 3 day intervals from 34 to 50 days after incubation, merozoites were harvested from each flask by vigorously rocking the CM back and forth 20 times, decanting the CM into a 50 ml centrifuge tube, adding 7 ml sterile phosphate buffer saline solution (PBS) to the flask and repeating the process. After harvesting the merozoites, fresh CM was added to each flask and the cultures were incubated as above. Merozoites obtained from the endothelial cells were washed three times in PBS, frozen and thawed five times and ultrasonicated three times for 20 s each time, at 100 W. The suspension was centrifuged at 15 000 g for 30 min and the supernatant was collected and used as antigen.
Antigen from merozoites of S. tenella and~or S. arieticanis The antigen was prepared using the same procedures employed for S. cruzi merozoites except that the sporocysts were treated with 5.2% sodium hypochlorite in HBSS at 4°C for 30 min and the excystation was achieved by incubating the sporocysts at 40°C for 4-6 h in HBSS containing 1% pepsin (1:3000), 17% ovine bile and 1% sodium hydrogen carbonate. The merozoites were collected at 2 day intervals from 22 to 34 days after cell inoculation using a similar monolayer of bovine pulmonary artery cells.
Antigen from merozoites ofT. gondii The antigen was kindly provided by the Institute of Parasitology, University of Zurich. To determine the stage of the Sarcocystis life cycle which produces the strongest reacting antigen, and to determine the cross-reactivity between heterologous species of Sarcocystis and T. gondii, 50 samples of cattle serum were examined by ELISA for antibodies directed against crude antigens isolated from cystozoites and merozoites of S. cruzi and S. tenella/S, arieticanis and T. gondii merozoites. Of these 50 samples, 25 were selected from cattle which were found not to have cystozoites. These cattle were approximately 1 year of age and originated from an area where the prevalence of S. cruzi had been found to be low (9%; Savini et al., 1992a). The remaining 25 sera were selected from cattle which were found by the digestion method to be heavily infected. No macroscopic cysts were found in the carcasses. It is likely that the species detected was S. cruzi, as this species has been found to be responsible for 99% of infection in Western Australia (Savini et al., 1992b). All 50 animals were found to be negative for T. gondii by the digestion method as described by Dubey and Beattie ( 1988 ). An ELISA using alkaline phosphatase conjugate was used in the present test. Vinyl 96-well assay plates (Linbro, Flow Laboratories, Irvine, UK) were coated overnight at 4°C with the optimal (0.5-5 #g m1-1 ) concentration of
18 5
G. Savini et al. / Veterinary Parasitology 51 (1994) 181-189
antigen in carbonate buffer (pH 9.6). The wells were subsequently filled with PBS containing 0.03% of Tween-20 and allowed to stand at 37 °C for 30 rain. To each well, 100/zl of diluted bovine serum ( 1/ 100 in PBS Tween) was added and the plates were incubated at 37°C for 2 h. Diluted rabbit antibovine IgG ( 100 gl), conjugated to alkaline phosphatase (Sigma Chemicals, St. Louis, MO), was added to each well and the plates were incubated at 37 ° C for 2 h. Phosphate substrate (4-nitrophenyl phosphate; 100 #l) was added and the plates were incubated for 30 rain at 37 °C. The plates were read on an ELISA reader with a 405 nm absorbance filter. Reference positive, negative, no antigen and no serum controls were included in triplicate on each plate. A preliminary test was performed to optimise the assay to achieve maximal discrimination between sera from S. cruzi infected and non-infected cattle. Optimal protein concentration of the antigen, serum dilution and conjugate dilution were determined by titration and comparisons of positive and negative reference sera. The assays were also checked for any inter-plate variation by expressing the test serum results as a percentage of the positive reference serum for each individual plate. These data allowed inter-plate standardisation, but they were found to be unnecessary in the majority of cases because little variation was observed and the number of test sera scored positive remained constant. The threshold (cut-offpoint) for discriminating a negative from a positive reaction was determined by calculating the mean absorbance of the 25 negative cattle and adding three standard deviations (SD). Results
Comparative results of parasitological and serological examinations of cattle for S. cruzi are shown in Figs. 1-5. A very high correlation was observed between the parasitological data and 0.6-
0.5-
•
0.4-
•
•
[ 0.3-
•
<
0.2.
• ....
......................
o_ o o ~ ° o • go
•
•
•
•
o__.
•
0.10.0
Animals not found to be Infected with any species
Animals found to be Infected with S. cruzl only
Fig. 1. Detection of antibodies in cattle against S. cruzi merozoite crude antigen by ELISA.
186
G. Savini et al. / Veterinary Parasitology 51 (1994.) 181-189
O.6. 0.5. 0.4. 0.3. 0.2. ~+3SD .........................
~ , ~
0.1
.~ ~'~,~
• • gO
Oo
• ....
• -0
"o: o ~ o
• •
0.0 Animals not found t o be Infected with any species
Animals found to be Infected w i t h .~.. c r u z l o n l y
Fig. 2. Detection of antibodies in cattle against S. tenella and/or S. arieticanis merozoite crude antigen by ELISA.
0.6-
0.5 0.4
E 0.3 '
O0
0.2-
•
~+3SD •- ...................
•
•
•
~• oqr
o '~ • - - o - O - - L~--'X~-. . . . . . • •
0.1-
o
0.0
Animals not found t o be infected with any species
A n i m a l s f o u n d to be Infected w i t h .~.. c r u z l o n l y
Fig. 3. Detection of antibodies in cattle against S. cruzi cystozoite crude antigen by ELISA.
0.60.50.4-
0.3. 0.2.
_:o,
•Q
• ,,o
..... .~+..3s.?.......... ~_-.,o, ___o_ __.~. 0.1 0.0
,
~
~
~,
~ A
~,
n
Animals not found t o be Infected with any species
Animals found to be Infected w i t h S. c r u z l o n l y
Fig. 4. Detection of antibodies in cattle against S. tenella and/or S. arieticanis cystozoite crude antigen by ELISA.
G. Savini et al. / Veterinary Parasitology 51 (1994) 181-189
187
0.6. 0.5. 0.4[
0.3<
•
0.2o.~ - - . . . .
.~.,..3S_D . . . . . . . . . .
o.o
Animals not found to be infected with any species
Animals found to be Infected with S. cruzl only
Fig. 5. Detection of antibodies in cattle against T. gondii merozoite crude antigen by ELISA.
the results obtained from the serological assays which used antigens from either cystozoites or merozoites of S. cruzi. The assay using the antigen derived from merozoites provided the best result for discriminating infected and non-infected animals (Figs. 1 and 3 ). Antibodies against S. cruzi showed cross-reactivity to crude antigens of the heterologous species S. tenella/S, arieticanis and T. gondii (Figs. 2, 3 and 5 ). No false negative reactions were recorded in cattle when the merozoite antigen ofS. cruzi was used (Fig. 1 ). In contrast, false negatives were recorded in cattle which were subsequently found to have cystozoites when other antigens were used: two when using antigen from S. cruzi cystozoites (Fig. 3), five when using antigen from S. tenella/S, arieticanis cystozoites and T. gondii merozoites (Figs. 4 and 5) and 17 when using antigen from S. tenella/S. arieticanis merozoites (Fig. 2). Apart from the merozoite antigen ofS. tenella/S, arieticanis, where we were not able to repeat the test for lack of antigen, the reproducibility of the test was high and similar results were observed when the sera were tested on two separate occasions. Discussion Accurate diagnosis in the intermediate host could play a major role in the effective implementation of control of Sarcocystis infection. Conventional methods of diagnosing infections with Sarcocystis spp. by examining the muscle tissue for the presence of cysts or cystozoites are not suitable for use in large-scale screening programmes nor for use in diagnosing infections in live animals. Recent studies have shown encouraging results for serodiagnostic tests. Twenty-one days after experimental infection in cattle, Gasbarre et al. (1984) demonstrated a transient IgM response which lasted for several
188
G. Savini et al. / Veterinary Parasitology 51 (1994) 181-189
months. A persistent and often slowly rising IgG response was also detected approximately 30 days after infection. The IgG reached its maximum level approximately 3 months after infection. The presence of specific IgM thus coincided with clinical sarcocystosis in cattle, and could be used to confirm this diagnosis, whereas detection of specific IgG would be more useful for the diagnosis of latent or chronic infections. The apparent lack of antibody response to the early stages of Sarcocystis infection may be due to the fact that all previous studies have been based primarily on responses against antigens obtained from cystozoites of S. cruzi which may not detect antibodies directed against the earliest stages of the parasite (Dubey et al., 1989). Speer et al. (1986), Burgess et al. (1988) and Speer and Burgess (1988) detected several distinctive proteins in antigens from merozoites and cystozoites of S. cruzi. Thus, an assay employing merozoite antigens may be preferable for detecting an early humoral immune response. In the present study, antigens derived from merozoites of S. cruzi and S. tenella/S, arieticanis were utilised. As far as we are aware, this is the first report detecting antibodies to merozoites of S. cruzi in naturally infected cattle. The sensitivity of the ELISA using merozoites of S. cruzi proved to be superior to the ELISA using other antigens. A highly immunogenic protein seems to be present in the merozoite antigen of S. cruzi and whether it may also detect antibodies against merozoites at earlier stages of infection needs to be investigated. Sera from cattle naturally infected with S. cruzi gave similar ELISA values when tested against either cystozoite antigens ofS. tenella/S, arieticanis or S. cruzi but not when tested against merozoite antigens of S. tenella/S, arieticanis or S. cruzi. It would appear that the antigen preparation from cystozoites contained cross-reacting determinants that rendered the ELISA assay species non-specific while its specificity improved when using merozoite antigens. The specificity of the test should improve further with the use of more accurate procedures for the extraction of antigen from merozoites of Sarcocystis and that may facilitate the identification of the species and stage of the infection. The sensitivity and specificity of serological tests for the diagnosis of bovine toxoplasmosis has been evaluated (Dubey et al., 1985 ). In our study, it is clear that cross-reactions between T. gondii antigen and anti-S, cruzi antibodies occur and would be enhanced by calculating the cut-off threshold by determining the mean value of negative animals and adding only 2 SD, as is frequently done in serological tests. Thus, infection with S. cruzi may play an important role when assessing antibody levels to Toxoplasma in cattle using a crude merozoite antigen of T. gondii. Other investigators have described similar cross-reactions between anti-
G. Savini et al. / VeterinaryParasitology51 (1994) 181-189
189
gens derived from T. gondii and antibodies to S. cruzi in experimentally infected animals (Uggla et al., 1987 ). In conclusion, the results presented here suggest that the unpurified merozoite antigen of S. cruzi produced in vitro is accurate in discriminating between animals infected and non-infected with S. cruzi and may be used in the diagnosis of acute infection of Sarcocystis in economically important hosts such as cattle and sheep.
References B6ttner, A., Charleston, W.A.G., Pomroy, W.E. and Rommel, M., 1987. The prevalence and identity of Sarcocystis in cattle in New Zealand. Vet. Parasitol., 24:157-168. Burgess, D.E., Speer, C.A. and Reducker, D.N., 1988. Identification of antigens of S. cruzi sporozoites, merozoites and bradyzoites with monoclonal antibodies. J. Parasitol., 74:828-832. Dubey, J.P., 1986. A review oftoxoplasmosis in cattle. Vet. Parasitol., 22:177-202. Dubey, J.P. and Beattie, C.P., 1988. Toxoplasmosis of Animals and Man. CRC Press, Boca Raton, FL, pp. 1-40. Dubey, J.P., Desmonts, G., Antunes, F., McDonald, C. and Walls, K.W., 1985. Serologic evaluation of cattle inoculated with Toxoplasma gondii: comparison of Sabin-Feldman dye test and other agglutination tests. Am. J. Vet. Res., 46: 1085-1088. Dubey, J.P., Speer, C.A. and Fayer, R., 1989. Sarcocystosis of Animals and Man. CRC Press, Boca Raton, FL, pp. 1-215. Gasbarre, L.C., Suter, P. and Fayer, R., 1984. Humoral and cellular immune responses in cattle and sheep inoculated with Sarcocystis. Am. J. Vet. Res., 45:1592-1596. Levine, N.D., 1986. The taxonomy of Sarcocystis (Protozoa: Apicomplexa) species. J. Parasitol., 72: 372-382. O'Donoghue, P.J. and Weyreter, H., 1983. Detection of Sarcocystis antigen in the sera of experimentally infected pigs and mice by an immunoenzymatic assay. Vet. Parasitol., 12:13-29. O'Donoghue, P.J. and Weyreter, H., 1984. Examinations on the serodiagnosis of Sarcocystis infections II. Class specific immunoglobulin responses in mice, pigs and sheep. Zentralbl. Bakteriol. Parasitenkd. Infektionskr. Hyg. Abt. 1, Orig. Reihe A, 257: 168-184. Savini, G., Dunsmore, J.D., Robertson, I.D. and Seneviratna, P., 1992a. Epidemiology of Sarcocystis spp. in cattle of Western Australia. Epidemiol. Infect., 108" 107-113. Savini, G., Dunsmore, J.D., Robertson, I.D. and Seneviratna, P., 1992b. Sarcocystis spp. in cattle of Western Australia. Aust. Vet. J., 68:201-202. Speer, C.A. and Burgess, D.E., 1988. In vitro development and antigen analysis of Sarcocystis. Parasitol. Today, 4: 46. Speer, C.A., Withmire, W.M., Reducker, D.N. and Dubey, J.P., 1986. In vitro cultivation of meronts ofS. cruzi. J. Parasitol., 72: 677. Uggla, A., Hilali, M. and Lovgren, K., 1987. Serological responses in S. cruzi infected calves challenged with T. gondii. Res. Vet. Sci., 43: 127-129.
181
Evaluation of a serological test system for the diagnosis of Sarcocystis cruzi infection in cattle using S. cruzi merozoite antigen G. Savini*, J.D. Dunsmore, I.D. Robertson School of Veterinary Studies, Murdoch University, Murdoch, W.A. 6150, Australia (Accepted I 1 May 1993)
Abstract Fifty serum samples were examined by enzyme-linkedimmunosorbentassay (ELISA) for antibodies directed against crude antigens isolated from cystozoites and merozoites of Sarcocystis cruzi and Sarcocystis tenella/Sarcocystis arieticanis and merozoites of Toxoplasma gondii. Of these 50 samples, 25 were from cattle originatingfrom an area where the prevalence ofS. cruzi was found to be very low (9%) and in which no cystozoites were detected, and 25 were from cattle which were found by a digestion method to be heavily infected with S. cruzi. A very high correlation was observed between the parasitological data and the results obtained from the serological assays which used antigens from either cystozoites or merozoites of S. cruzi. The assay using the antigen derived from merozoites provided the best result for discriminating infected and non-infectedanimals. There was some cross-reactivity between the antigen derived from cystozoites of heterologous species of Sarcocystis and S. cruzi antibodies, and some cross-reactivity between antigen of T. gondii and antibodies to S. cruzi. The reproducibility of the assays was found to be high and similar results were observed when the sera were tested on two separate occasions. The unpurified S. cruzi merozoite antigen produced in vitro is relatively accurate in discriminating positive and negative animals and may be used for diagnosis in economically important hosts such as cattle and sheep. Key words: Sarcocystis cruzi; Cattle-Protozoa; Diagnosis-Protozoa
Introduction The cyst-forming coccidian parasites Toxoplasma gondii and Sarcocystis spp. are intraceUular protozoan organisms occurring in domestic animals and humans throughout the word. The parasites belong to the same family, Sarcocystidae (Levine, 1986) and have life cycles in which carnivores (predator) are the definitive hosts and herbivores (prey) the intermediate hosts. The importance of infection by T. gondii during pregnancy to the developing embryo has been recognised for some time in women as well as in domes*Corresponding author,
SSDI 0 3 0 4 - 4 0 1 7 ( 93 ) 0 0 5 4 1 - 6
182
G. Savini et al. / Veterinary Parasitology 51 (1994) 181-189
tic animals such as sheep and goats. However, the clinical and subclinical effects of Sarcocystis infections in domestic animals as well as the potential pathogenicity of Sarcocystis hominis and Sarcocystis suihominis, which require cattle and pigs, respectively, as intermediate hosts, are still being elucidated. Direct diagnostic methods such as digestion and histology, which readily detect the parasite in muscle tissues, have shown that Sarcocystis infection is very common in cattle worldwide. The infection has also been seen to stimulate the production of circulating antibodies which can be detected by various serological methods. In contrast, T. gondii has rarely been isolated from bovine tissues, although the frequent demonstration of antibodies in serum suggests that inapparent infection may be common (Dubey, 1986 ). It is accepted that soluble cystozoite antigens, prepared from cysts of Sarcocystis spp., react with heterologous species (O'Donoghue and Weyreter, 1983 ), although homologous species usually give rise to stronger reactions (O'Donoghue and Weyreter, 1984 ). The Sarcocystis antibodies that are present during the acute phase of infection with Sarcocystis cruzi in calves have been shown to cross-react with crude somatic Toxoplasma antigen prepared by either freeze-thawing or ultrasonication of merozoites (Uggla et al., 1987 ). The recent development of an in vitro system for growing numerous merozoites of Sarcocystis has made possible the preparation of new crude antigen from this stage of the parasite. The present study assesses and compares the performance of an enzyme-linked immunosorbent assay (ELISA) using crude antigens derived from merozoites and cystozoites of S. cruzi and Sarcocystis tenella a n d / o r Sarcocystis arieticanis and from merozoites of T. gondii for detecting antibodies to S. cruzi in the sera of cattle naturally infected with this parasite. Materials and methods
Sample collection Blood samples were collected from a total of 294 cattle slaughtered at three different abattoirs in Western Australia. Serum was obtained by centrifugation and stored in aliquots at - 20 ° C until use. A segment of muscle tissue (oesophagus) was collected from each carcass, subjected to pepsin digestion and examined microscopically for the presence of S. cruzi cystozoites and T. gondii bradyzoites as described by Bfttner et al. ( 1987 ) and Dubey and Beattie ( 1988 ).
Antigen from cystozoites ofS. cruzi An isotonic digest medium was prepared as follows: 2.5 g of pepsin ( 1 : 10 000) was dissolved in 10 ml of distilled water, centrifuged at 2000 rev
G. Savini et al. / Veterinary Parasitology 51 (1994) 181-189
183
min-~ and the supernatant solution containing dissolved pepsin collected. Sodium chloride (8.5 g) and 8 ml of concentrated hydrochloric acid ( 14 N) were added and the solution was made up to 1 I with distilled water. A heavily infected bovine heart was t r i m m e d of fat and minced in a blender for a few seconds at top speed and 500 g were mixed with 2 1 of the digest m e d i u m in a water bath at 40 ° C for 40 min. The mixture was filtered through a 250/tin metallic sieve. The filtrate was centrifuged at 2400 rev m i n - 1 for 5 min and the pellet was washed three times in normal saline solution. The washed pellet was suspended in saline solution in a ratio of 1 : 3 and the pH adjusted to 7 with a solution of 10% sodium bicarbonate in distilled water. The suspension was mixed in a ratio of 1:2 with Percoll isotonic solution (Pharmacia, Uppsala, Sweden), and centrifuged in a conical centrifuge tube at 2400 rev min-1 for 10 min at room temperature. The pellet containing cystozoites was washed three times and suspended in normal saline solution. The cystozoites in suspension were then frozen at - 2 0 ° C, thawed four times and ultrasonicated three times for 20 s each time at 100 W. The suspension was centrifuged at 15 000 g for 30 min. The supernatant was then collected and used as antigen.
Antigen from cystozoites ofS. tenella and~or S. arieticanis The antigen was prepared from two heavily infected hearts of rams using the same procedure employed for S. cruzi.
Antigen from merozoites ofS. cruzi Sporocysts of S. cruzi were obtained by scraping the mucosa of the small intestine of dogs 15 days after they had ingested S. cruzi infected cattle meat and were shown to be passing sporocysts. Sporocysts were purified, concentrated and stored at 4°C in Hank's balanced salt solution (HBSS) with antibiotics before use as described by Dubey et al. (1989). Prior to excystation, sporocysts were washed twice by centrifugation in HBSS, treated with 2.6% sodium hypochlorite in HBSS for 30 min at 4°C and washed four times in HBSS. Excystation was achieved by incubating the sporocysts for 4-6 h in HBSS containing 5% bovine sodium taurocholate, 2% trypsin, 1.8 ml of a solution of 0.14 M sodium hydrogen carbonate in distilled water. Excysted sporozoites were washed and passed through a nylon wool column (Robbins Scientific, Sunnyvale, CA) to remove the debris and sporocyst walls. Sporozoites were suspended in culture m e d i u m (CM) consisting of RPMI 1640 with L-glutamine (Flow Laboratories, Irvine, U K ) plus 5% foetal calf serum, 50 U m l - ~ penicillin G and 50 #g m l - ~ dihydrostreptomycin (Sigma Chemicals, St. Louis, MO). Each of four 25 cm 2 tissue culture flasks containing a monolayer of bovine pulmonary artery endothelial cells (American Type Culture Collection, Rockville, M D ) was inoculated with 5 × 105 sporozoites and incubated at
184
G. Savini et al. / Veterinary Parasitology 51 (1994) 181-189
37°C under 5% CO2 and 95% air in a continuous flow C O 2 incubator. At 3 day intervals from 34 to 50 days after incubation, merozoites were harvested from each flask by vigorously rocking the CM back and forth 20 times, decanting the CM into a 50 ml centrifuge tube, adding 7 ml sterile phosphate buffer saline solution (PBS) to the flask and repeating the process. After harvesting the merozoites, fresh CM was added to each flask and the cultures were incubated as above. Merozoites obtained from the endothelial cells were washed three times in PBS, frozen and thawed five times and ultrasonicated three times for 20 s each time, at 100 W. The suspension was centrifuged at 15 000 g for 30 min and the supernatant was collected and used as antigen.
Antigen from merozoites of S. tenella and~or S. arieticanis The antigen was prepared using the same procedures employed for S. cruzi merozoites except that the sporocysts were treated with 5.2% sodium hypochlorite in HBSS at 4°C for 30 min and the excystation was achieved by incubating the sporocysts at 40°C for 4-6 h in HBSS containing 1% pepsin (1:3000), 17% ovine bile and 1% sodium hydrogen carbonate. The merozoites were collected at 2 day intervals from 22 to 34 days after cell inoculation using a similar monolayer of bovine pulmonary artery cells.
Antigen from merozoites ofT. gondii The antigen was kindly provided by the Institute of Parasitology, University of Zurich. To determine the stage of the Sarcocystis life cycle which produces the strongest reacting antigen, and to determine the cross-reactivity between heterologous species of Sarcocystis and T. gondii, 50 samples of cattle serum were examined by ELISA for antibodies directed against crude antigens isolated from cystozoites and merozoites of S. cruzi and S. tenella/S, arieticanis and T. gondii merozoites. Of these 50 samples, 25 were selected from cattle which were found not to have cystozoites. These cattle were approximately 1 year of age and originated from an area where the prevalence of S. cruzi had been found to be low (9%; Savini et al., 1992a). The remaining 25 sera were selected from cattle which were found by the digestion method to be heavily infected. No macroscopic cysts were found in the carcasses. It is likely that the species detected was S. cruzi, as this species has been found to be responsible for 99% of infection in Western Australia (Savini et al., 1992b). All 50 animals were found to be negative for T. gondii by the digestion method as described by Dubey and Beattie ( 1988 ). An ELISA using alkaline phosphatase conjugate was used in the present test. Vinyl 96-well assay plates (Linbro, Flow Laboratories, Irvine, UK) were coated overnight at 4°C with the optimal (0.5-5 #g m1-1 ) concentration of
18 5
G. Savini et al. / Veterinary Parasitology 51 (1994) 181-189
antigen in carbonate buffer (pH 9.6). The wells were subsequently filled with PBS containing 0.03% of Tween-20 and allowed to stand at 37 °C for 30 rain. To each well, 100/zl of diluted bovine serum ( 1/ 100 in PBS Tween) was added and the plates were incubated at 37°C for 2 h. Diluted rabbit antibovine IgG ( 100 gl), conjugated to alkaline phosphatase (Sigma Chemicals, St. Louis, MO), was added to each well and the plates were incubated at 37 ° C for 2 h. Phosphate substrate (4-nitrophenyl phosphate; 100 #l) was added and the plates were incubated for 30 rain at 37 °C. The plates were read on an ELISA reader with a 405 nm absorbance filter. Reference positive, negative, no antigen and no serum controls were included in triplicate on each plate. A preliminary test was performed to optimise the assay to achieve maximal discrimination between sera from S. cruzi infected and non-infected cattle. Optimal protein concentration of the antigen, serum dilution and conjugate dilution were determined by titration and comparisons of positive and negative reference sera. The assays were also checked for any inter-plate variation by expressing the test serum results as a percentage of the positive reference serum for each individual plate. These data allowed inter-plate standardisation, but they were found to be unnecessary in the majority of cases because little variation was observed and the number of test sera scored positive remained constant. The threshold (cut-offpoint) for discriminating a negative from a positive reaction was determined by calculating the mean absorbance of the 25 negative cattle and adding three standard deviations (SD). Results
Comparative results of parasitological and serological examinations of cattle for S. cruzi are shown in Figs. 1-5. A very high correlation was observed between the parasitological data and 0.6-
0.5-
•
0.4-
•
•
[ 0.3-
•
<
0.2.
• ....
......................
o_ o o ~ ° o • go
•
•
•
•
o__.
•
0.10.0
Animals not found to be Infected with any species
Animals found to be Infected with S. cruzl only
Fig. 1. Detection of antibodies in cattle against S. cruzi merozoite crude antigen by ELISA.
186
G. Savini et al. / Veterinary Parasitology 51 (1994.) 181-189
O.6. 0.5. 0.4. 0.3. 0.2. ~+3SD .........................
~ , ~
0.1
.~ ~'~,~
• • gO
Oo
• ....
• -0
"o: o ~ o
• •
0.0 Animals not found t o be Infected with any species
Animals found to be Infected w i t h .~.. c r u z l o n l y
Fig. 2. Detection of antibodies in cattle against S. tenella and/or S. arieticanis merozoite crude antigen by ELISA.
0.6-
0.5 0.4
E 0.3 '
O0
0.2-
•
~+3SD •- ...................
•
•
•
~• oqr
o '~ • - - o - O - - L~--'X~-. . . . . . • •
0.1-
o
0.0
Animals not found t o be infected with any species
A n i m a l s f o u n d to be Infected w i t h .~.. c r u z l o n l y
Fig. 3. Detection of antibodies in cattle against S. cruzi cystozoite crude antigen by ELISA.
0.60.50.4-
0.3. 0.2.
_:o,
•Q
• ,,o
..... .~+..3s.?.......... ~_-.,o, ___o_ __.~. 0.1 0.0
,
~
~
~,
~ A
~,
n
Animals not found t o be Infected with any species
Animals found to be Infected w i t h S. c r u z l o n l y
Fig. 4. Detection of antibodies in cattle against S. tenella and/or S. arieticanis cystozoite crude antigen by ELISA.
G. Savini et al. / Veterinary Parasitology 51 (1994) 181-189
187
0.6. 0.5. 0.4[
0.3<
•
0.2o.~ - - . . . .
.~.,..3S_D . . . . . . . . . .
o.o
Animals not found to be infected with any species
Animals found to be Infected with S. cruzl only
Fig. 5. Detection of antibodies in cattle against T. gondii merozoite crude antigen by ELISA.
the results obtained from the serological assays which used antigens from either cystozoites or merozoites of S. cruzi. The assay using the antigen derived from merozoites provided the best result for discriminating infected and non-infected animals (Figs. 1 and 3 ). Antibodies against S. cruzi showed cross-reactivity to crude antigens of the heterologous species S. tenella/S, arieticanis and T. gondii (Figs. 2, 3 and 5 ). No false negative reactions were recorded in cattle when the merozoite antigen ofS. cruzi was used (Fig. 1 ). In contrast, false negatives were recorded in cattle which were subsequently found to have cystozoites when other antigens were used: two when using antigen from S. cruzi cystozoites (Fig. 3), five when using antigen from S. tenella/S, arieticanis cystozoites and T. gondii merozoites (Figs. 4 and 5) and 17 when using antigen from S. tenella/S. arieticanis merozoites (Fig. 2). Apart from the merozoite antigen ofS. tenella/S, arieticanis, where we were not able to repeat the test for lack of antigen, the reproducibility of the test was high and similar results were observed when the sera were tested on two separate occasions. Discussion Accurate diagnosis in the intermediate host could play a major role in the effective implementation of control of Sarcocystis infection. Conventional methods of diagnosing infections with Sarcocystis spp. by examining the muscle tissue for the presence of cysts or cystozoites are not suitable for use in large-scale screening programmes nor for use in diagnosing infections in live animals. Recent studies have shown encouraging results for serodiagnostic tests. Twenty-one days after experimental infection in cattle, Gasbarre et al. (1984) demonstrated a transient IgM response which lasted for several
188
G. Savini et al. / Veterinary Parasitology 51 (1994) 181-189
months. A persistent and often slowly rising IgG response was also detected approximately 30 days after infection. The IgG reached its maximum level approximately 3 months after infection. The presence of specific IgM thus coincided with clinical sarcocystosis in cattle, and could be used to confirm this diagnosis, whereas detection of specific IgG would be more useful for the diagnosis of latent or chronic infections. The apparent lack of antibody response to the early stages of Sarcocystis infection may be due to the fact that all previous studies have been based primarily on responses against antigens obtained from cystozoites of S. cruzi which may not detect antibodies directed against the earliest stages of the parasite (Dubey et al., 1989). Speer et al. (1986), Burgess et al. (1988) and Speer and Burgess (1988) detected several distinctive proteins in antigens from merozoites and cystozoites of S. cruzi. Thus, an assay employing merozoite antigens may be preferable for detecting an early humoral immune response. In the present study, antigens derived from merozoites of S. cruzi and S. tenella/S, arieticanis were utilised. As far as we are aware, this is the first report detecting antibodies to merozoites of S. cruzi in naturally infected cattle. The sensitivity of the ELISA using merozoites of S. cruzi proved to be superior to the ELISA using other antigens. A highly immunogenic protein seems to be present in the merozoite antigen of S. cruzi and whether it may also detect antibodies against merozoites at earlier stages of infection needs to be investigated. Sera from cattle naturally infected with S. cruzi gave similar ELISA values when tested against either cystozoite antigens ofS. tenella/S, arieticanis or S. cruzi but not when tested against merozoite antigens of S. tenella/S, arieticanis or S. cruzi. It would appear that the antigen preparation from cystozoites contained cross-reacting determinants that rendered the ELISA assay species non-specific while its specificity improved when using merozoite antigens. The specificity of the test should improve further with the use of more accurate procedures for the extraction of antigen from merozoites of Sarcocystis and that may facilitate the identification of the species and stage of the infection. The sensitivity and specificity of serological tests for the diagnosis of bovine toxoplasmosis has been evaluated (Dubey et al., 1985 ). In our study, it is clear that cross-reactions between T. gondii antigen and anti-S, cruzi antibodies occur and would be enhanced by calculating the cut-off threshold by determining the mean value of negative animals and adding only 2 SD, as is frequently done in serological tests. Thus, infection with S. cruzi may play an important role when assessing antibody levels to Toxoplasma in cattle using a crude merozoite antigen of T. gondii. Other investigators have described similar cross-reactions between anti-
G. Savini et al. / VeterinaryParasitology51 (1994) 181-189
189
gens derived from T. gondii and antibodies to S. cruzi in experimentally infected animals (Uggla et al., 1987 ). In conclusion, the results presented here suggest that the unpurified merozoite antigen of S. cruzi produced in vitro is accurate in discriminating between animals infected and non-infected with S. cruzi and may be used in the diagnosis of acute infection of Sarcocystis in economically important hosts such as cattle and sheep.
References B6ttner, A., Charleston, W.A.G., Pomroy, W.E. and Rommel, M., 1987. The prevalence and identity of Sarcocystis in cattle in New Zealand. Vet. Parasitol., 24:157-168. Burgess, D.E., Speer, C.A. and Reducker, D.N., 1988. Identification of antigens of S. cruzi sporozoites, merozoites and bradyzoites with monoclonal antibodies. J. Parasitol., 74:828-832. Dubey, J.P., 1986. A review oftoxoplasmosis in cattle. Vet. Parasitol., 22:177-202. Dubey, J.P. and Beattie, C.P., 1988. Toxoplasmosis of Animals and Man. CRC Press, Boca Raton, FL, pp. 1-40. Dubey, J.P., Desmonts, G., Antunes, F., McDonald, C. and Walls, K.W., 1985. Serologic evaluation of cattle inoculated with Toxoplasma gondii: comparison of Sabin-Feldman dye test and other agglutination tests. Am. J. Vet. Res., 46: 1085-1088. Dubey, J.P., Speer, C.A. and Fayer, R., 1989. Sarcocystosis of Animals and Man. CRC Press, Boca Raton, FL, pp. 1-215. Gasbarre, L.C., Suter, P. and Fayer, R., 1984. Humoral and cellular immune responses in cattle and sheep inoculated with Sarcocystis. Am. J. Vet. Res., 45:1592-1596. Levine, N.D., 1986. The taxonomy of Sarcocystis (Protozoa: Apicomplexa) species. J. Parasitol., 72: 372-382. O'Donoghue, P.J. and Weyreter, H., 1983. Detection of Sarcocystis antigen in the sera of experimentally infected pigs and mice by an immunoenzymatic assay. Vet. Parasitol., 12:13-29. O'Donoghue, P.J. and Weyreter, H., 1984. Examinations on the serodiagnosis of Sarcocystis infections II. Class specific immunoglobulin responses in mice, pigs and sheep. Zentralbl. Bakteriol. Parasitenkd. Infektionskr. Hyg. Abt. 1, Orig. Reihe A, 257: 168-184. Savini, G., Dunsmore, J.D., Robertson, I.D. and Seneviratna, P., 1992a. Epidemiology of Sarcocystis spp. in cattle of Western Australia. Epidemiol. Infect., 108" 107-113. Savini, G., Dunsmore, J.D., Robertson, I.D. and Seneviratna, P., 1992b. Sarcocystis spp. in cattle of Western Australia. Aust. Vet. J., 68:201-202. Speer, C.A. and Burgess, D.E., 1988. In vitro development and antigen analysis of Sarcocystis. Parasitol. Today, 4: 46. Speer, C.A., Withmire, W.M., Reducker, D.N. and Dubey, J.P., 1986. In vitro cultivation of meronts ofS. cruzi. J. Parasitol., 72: 677. Uggla, A., Hilali, M. and Lovgren, K., 1987. Serological responses in S. cruzi infected calves challenged with T. gondii. Res. Vet. Sci., 43: 127-129.