Acta vet. scand. 1990,31.413-422.

Evaluation of ELISA for the Detection of Toxoplasma in Swine Sera

By Varpu Hirveld-Koski

National Veterinary Institute, Department of Hygiene, Helsinki, Finland.

Hirvelii-Koski, V: Evaluationof ELISA for the detectionof Toxoplasmaantibodies in swine sera. Acta vet. scand. 1990, 31, 413-422. - -linked immuno• sorbent (ELISA) is compared with the indirect fluorescent test (IFAT), the indirect haemagglutination test (IHAT) and the latex (LA) test for the detection of toxoplasma antibodies in swine sera. The 100 swine sera examined represent ELISA values from > 0 to 154 EIU. The agreement was highest (0.67) between ELISA and IFAT with an ELISA cut-off value of 30 EIU, and between ELISAand the LA test with an ELISAcut-off value of 50 EIU (0.74). All sera giving < 10 EIU were negative in the other tests, and all those with > 70 EIU were positive in I, 2 or all of the referencetests. In order to avoid false posi• tive results with ELISA,all sera giving 1(}..70 EIU should be confirmed with a test which has a good specificity, e.g. IFAT. ELISAis a sensitivetest and is highlysuitable for the screeningof largeamounts of samples, but it may be too complicated for screeningtoxoplasma antibodies in the laboratories of abattoirs.

IFA T ; IHA T; latex aggl uti nation test. Introduction plasma tissue cysts (Boch 1973, Dubey et al. is a worldwide parasite zoo• 1986). They are not visible, and carcasses nosis . Toxoplasma gondii undergoes a coc• harbouring cysts cannot be identified by cidian cycle in the intestinal epithelium of conventional meat inspection methods. Be• cats which results in the shedding of oocysts cause the parasitological methods used for in the cat faeces (Hutchison et al. 1970). the detection of toxoplasma cysts are very Other animals, including man, serve as in• laborious, the diagnosis of toxoplasmosis is termediate hosts, and may harbour toxoplas• usually based on . Even though not ma tissue cysts in their musculature or other every pig with a positive antibody titre har• organs. Toxoplasma infection can be trans• bours toxoplasma tissue cysts, the appear• mitted to humans by ingestion of either tis• ance of toxoplasma antibodies is associated sue cysts or oocysts. The foetus may be in• with the presence of viable T. gondii orga• fected congenitally through trophozoites, the nisms in the meat or other organs ofswine. parasitic form of T. gondii circulating in the The methylene blue dye test , indirect fluor• blood during the acute phase of infection. escent antibody test (IFAT), indirect hae• The reactivation of a latent toxoplasma in• magglutination test (lHAT), latex agglutina• fection in immunosuppressed patients can tion (LA) test and enzyme-linked immuno• lead to a clinical disease, e.g. toxoplasmic sorbent assay (ELISA) are used for the detec• encephalitis in patients with AIDS (Lufi & tion of toxoplasma antibodies. When used Remington 1988). for screening purposes, ELISA has some ad• Raw or inadequately cooked pork is con• vantages over the other methods. It can be sidered to be an important source of toxo- semiautomated and the reading ofthe results

Acta vel. scand . vel, 31 no. 4 - 1990 414 V. Hir veld-Koski is objective and does not require skilled per• the indirect haemagglutination test (Toxo• sonnel, as does IFAT, for instance, which HA Kit, biolvlerieux, France). relies on a subjective interpretation of the No toxoplasma antibodies could be detected microscope reading. Like IFAT, ELISA can in the sera of these pigs before the first ino• distinguish between immunoglobulin clas• culation (!HA titres < 10). When the IgG ses, which the LA test cannot do. ELISA titres had reached a plateau 40-80 in pig I has been considered a suitable method for and 160-320 in pig 2, the pigs were anae• screening for swine toxoplasmosis, and its sthetized and bled. The sera were stored at possible use at abattoirs for the inspection of - 20'C after centrifugation. individual animals has also been discussed (van Knap en et al. 1982, Oliver et al. 1983, En zyme-linked immunosorbent assay Waltman et al. 1984). (ELISA) The purpose of this investigation was to eva• The serum of pig 2 was used as a positive luate ELISA as a screening method for the control, while a negative control serum was detection of toxoplasma antibodies in swine obtained by pooling 5 porcine sera which sera in connection with meat inspection at had been negative either by IFAT or !HAT abattoirs. and which had ELISA absorbance values of about 0.1 or less. The pool was stored at Material and methods -20'C. Test sera Microtitre plates coated with toxoplasma A total of 100 sera were chosen from a were prepared by Labsystems Oy group of 1849 swine sera obtained from Fin• (Helsinki, Finland) by the method described nish slaughter houses from clinically healthy by Turunen (1983). These flat-bottomed pigs. All 1849 sera were examined for microtitre plates were sensitized toxoplasma ant ibodies by ELISA. The 100 with this antigen preparation to the degree sera used in this study were chosen to repre• that the positive control serum (pig 2) and sent various ELISA values ranging from < the negative control serum gave absorbance o to 154 EIU which was the highest ELISA values of about 1.0 and 0.1, respectively. value measured. The ELISA results were The plates were stored at + 6'C. compared with those obtained by IFAT, The control sera and test sera were diluted !HAT and the LA test. I :200 in phosphate-buffered saline (PBS) (Labsystems OY) and added to the micro• Positive control sera titre plates in duplicate, 0.1 ml in each well. Posit ive control sera were produced in 2 The positive and negative control sera were pigs, about 18 kg each, which were inocula• included in each plate. The reagent blank ted subcutaneously with Toxoplasma gondii wells received 0.1 ml ofthe diluent buffer. (T.g.) trophozoites obtained from the peri• The plate was covered and incubated for 90 toneal exudate of mice infected intra peri• min at 37'C, after which the sera were dis• toneally with T.g. Pig I was inoculated with carded and the wells washed 3 times with 10,000, 2,000,000 and 75,000 trophozoites Tween 20 diluted I:500 with 0.9 % (w/v) and pig 2 with 20,000, 2,000,000 and NaCI. 150,000 trophozoites at 2-3 weeks intervals. conjugated sheep anti• Blood samples were taken weekly and the bodies to swine IgG (Orion Diagnostica, sera examined for toxoplasma antibodies by Helsinki, Finland) were diluted I:200 with

Acta vel. scand. vol. 31 no. 4 - 1990 Toxoplasma antibodies in swine sera 415

PBS, and 0.1 ml of this solution was added The slides were examined with a fluores• to each well. After incubation for 60 min at cence microscope. The reaction was consi• 37"C, the wells were emptied and washed as dered positive when yellow-green fluores• described above. cence extended around the entire periphery The solution of paranitrophenyl of most of the parasites in several fields of phosphate 20 mglml in diethanolamine buf• the smear. Incomplete peripheral or faint fer, pH 10.0 , was added (0.1 mllwell). After staining was deemed negative. All the samp• incubation for 30 min at 37"C, the reaction les were tested at least twice in different test was stopped by the addition of 0.1 ml of I runs . To calculate the geometric mean titre, M NaOH. The intensity of the colour reac• the common logarithms of the reciprocal tion was measured at 405 nm with a Titer• endpoint dilutions were averaged and the re• tek Multiskan spectrophotometer (Eflab). sult converted to its antilogarithm. A geo• The absorbance values of the parallel wells metric mean titre > 10 was considered posi• were averaged and the result converted to tive. "enzyme units" (EIU) by the following formula: Indirect haemagglutination test (IHAT) absorbance value) _ (abSOrbanCe ValUe) Toxo-Ha Kits (biolvlerieux, France) were of the sample of the negative used, following the manufacturers' instruc• ( control serum tions. All the samples were tested at least EIU:::: x 100 twice in different test runs . Geometric mean absorbance ValUe) ( absorbance ValUe) of the positive - of the negative titres ofthe test results were calculated as for ( control serum control serum IFAT , an a titre > 20 was considered posi• tive. Indirect fluorescent antibody test (IFAT) Toxoplasma antigen was prepared from pe• Latex agglutination (LA) test ritoneal exudate of mice infected 5 days pre• Toxolex kits (Orion Diagnostic, Helsinki, viously. The exudate was diluted I: I with Finland) were used in accordance with the 3 % formalin-PBS, and the suspension was manufacturers' instructions with the excep• left to stand for 2 h before being centrifuged tion that the reaction was studied for 10 min and washed 3 times. The sediment was re• instead of 6 min . All the samples were tested suspended in a sufficient amount of PBS and at least twice . The positive control sera were small drops of this transferred to microscope included in each test run , and gave positive slides. The slides were air dried and stored reactions after 3 min for pig 2 and 7 min for at - 20·C. pig I. The test was performed as described in the procedural guide of the CDC (Anon. 1980). Results The sera were tested at a series of dilutions Comparisons between the results for ELISA increasing in two-fold steps beginning at and the 3 reference tests are shown in Tables I: IO. Incubation was for 30 min at 37'C in I, 2 and 3, respectively. The levels of agree• moist chamber. Fluorescein-conjugated anti• ment between ELISA and the other tests are body to swine IgG (heavy and light chain• given in Fig. I. specific F (ab')2 fragment, Cappel Labora• The 29 sera with 10.0 EIU or less with this tories) diluted I:50 with PBS containing ELISA were negative in all the reference I: 10 Evans Blue 0.06 % solution was used. tests (Fig. 2), while the 19 sera with EIU

Acta vel. scand. vol. 31 no. 4 • 1990 416 V. Hirveld-Koski

Table I . Results of enzyme-linked immunosorbent assay (ELISA) and the in• direct fluorescent antibody test (I FAT) with respect to 100 porcine sera exam• ined for toxoplasma antibodies.

ELISA IFAT titre (georn. mean) result (EIU) < 10 10-20 21-40 41-80 81- 160 161-320 321...{)40

< 0 151 0.1-10.0 14 10.1-20.0 II I 20.1-30.0 12 4 2 2 30.1-40.0 2 3 2 I 40.1-50.0 II 50.1-60.0 2 I 60.1-70.0 I 70.1-80.0 80.1-90.0 90.1-100.0 100.1-110.0 5 110.1-120.0 2 120.1-130.0 2 130.1-140.0 140.1-150.0 2 150.1-160.0

1 Number of sera.

Table 2. Results of enzyme-linked immunosorbent assay (ELISA) and the indirect haemagg1utination test (!HAT) with respect to 100 porcine sera examined for toxoplasma antibodies.

ELISA IHAT titre (geom. mean) result (EIU) < 20 21-40 41-80 81-160 161-320 321-640

< 0 151 0.1-10.0 14 10.1-20.0 13 20.1-30.0 20 30.1-40.0 7 40.1-50.0 4 50.1-60.0 4 60.1-70.0 4 70.1-80.0 0 80.1-90.0 3 90.1-100.0 I 100.1-110.0 4 110.1-120.0 2 120.1-130.0 130.1-140.0 140.1-150.0 150.1-160.0

I Number of sera.

Acta vet. scand . vol. 31 no. 4 - 1990 Toxoplasma antibodies in swine sera 417

Table 3. Results of enzyme-linked immunosor• ranging from < 0 to 20.0 and 89 % from bent assay (ELISA) and the latex agglutination < 0 to 30.0. (LA) test with respect to 100 porcine sera exam• The inter-test reproducibility of ELISA was ined for toxoplasma antibodies. evaluated by running the low positive con• ELISA LA test result trol serum from pig I 32 times on separate result days, and the intra-test reproducibility by (EIU) negative positive examining this same low positive control se• <0 15' rum 32 times within the same test runs. The 0.1-10.0 14 mean, standard deviat ion, minimum and 10.1-20.0 12 I maximum in the inter-test runs were 43.3, 20.1-30.0 18 2 9.8, 27.3 and 69.8 EIU and those the intra• 30.1-40.0 8 test runs 43.2, 5.1, 32.9 and 53.8 EIU, re• 40.1-50.0 2 I spectively. 50.1-60.0 2 2 60.1-70.0 I 3 Discussion 70.1-80.0 I I The results obtained by the different me• 80.1-90.0 I 2 90.1-100.0 I thods developed for detecting toxoplasma 100.1-110.0 5 antibodies are influenced by the stage of in• 110.1-120.0 2 fection and the antigen preparation used 120.1-130.0 2 (Camargo et al. 1978, Lovgren et al. 1987, 130.1-140.0 Verhofstede et al. 1987, Mineo et al. 1980). 140.1-150.0 2 IFAT employs whole cells of the parasite as 150.1-160.0 I an antigen, while the antigenic material in

I Number of sera. ELISA, IHAT and the LA test consists of the soluble parts of disrupted trophozoites. In spite of this, ELISA has been reported to values of 70.0 or more were positive in I, 2 correlate well with IFAT for detecting toxo• or all of the reference tests (Fig. 2). Two of plasma IgG antibodies (Walls et al. 1977, these sera were positive only in IFAT, giving Balsari et al. 1980, Violand et al. 1982, Fi EIU values of 85.0 and 119.1 and IFAT lice et al. 1983). IFAT detects antibodies titres of40 and 300, respectively . which appear at the early stage of infection Of the 52 sera giving EIU values from 10.1 against antigenic components of the mem• to 70.0 EIU , approximately half (27) were brane of intact trophozoites (Karim & Lud negative in all the reference tests, and none lam 1975), whereas ELISA antibodies ap• of these sera was positive in all of them (Fig. pear later in the course of infection (Oliver 2). Seven (13 %) of these sera were positive et al. 1983). This may be one explanation in 2 reference tests, IFAT and the LA test, for the situation seen in Table I, where and 18 (35 %) in I reference test, 16 of them some sera with low ELISA values have rel• IFAT and 2 the LA test. atively high IFAT titres. These presumably Altogether 56 sera were negative in the represent sera from an early stage of infec• IFAT, IHAT and LA tests, giving ELISA tion . EIU values ranging from 0 to 62.0 (Fig. 2), a The LA test is simple and rapid to perform, mean of 13.7 EIU and a median of9.1 EIU. but does not distinguish immunoglobulin 70 % of the negative sera gave EIU values classes as ELISA and IFAT do. The 3 sera

Acta vel. scand. vol, 31 no. 4 • 1990 418 V. Hirveld-Koski

AgreeMent

• •••••••• •• r ..•••• , , , .o . B.8 f • • • • • , • ::: :::: ::· . .: ..: : B.7 ...... ;.. .. -:- ": : .. ·.. ..., .. i . ! B.6

B.5

....,·.· . ., . B.4 ·:::.. ·.··..... B.3

B.2 ...+ B . 1 ::::r::·:::·::r:.::::::I:::::::::::I:::::::::::I:::::::::I:::::::::I:::::::] ·I .• .• .• ., ,• 1 1 1

IB 2B 3B 4B 5B 6B 7B 8B 9B IBB Figure I . Agreement between ELISA and IFAT, IHAT and the LA test. The extent of the agreement is assessed according to Martin et al. 1987.

with > 70 EIU which were negative in the in swine tKatsube et al. 1972, Boch et Neu • LA test (Table 3) are difficult to explain. rohr 1982). Two of them presumably represent recent The reproducibility of ELISA has been con• infections. sidered good (Woodward 1982, Lin et al. The sensitivity of IHAT was lower than that 1980, Walls et al. 1977, Tomasi et al. 1986, of ELISA and the other tests. None of the van Loon & van der Veen 1980), although sera with < 70 EIU was positive in IHAT, some authors report occasional inter-test but 2 sera with 110.1-120.0 EIU were still variation (Woodward 1982). The intra-test negative in IHAT (Table 2). IHAT has been reproducibility was not very good here . The reported to show fluctuation in titres during standard deviation was 5.1 (coefficient of the course of infection when detecting toxo• variation 11.8 %) and major differences be• plasma antibodies in swine sera (Oliver et al. tween 2 wells in the same microtitre plate 1983); to be extremely difficult to read (Wil• were found in some cases (minimum 32.0, son et af. 1987); and to be a poor test for maximum 53.8 EIU). The standard devia• diagnosing swine toxoplasmosis (Wallman tion in the inter-test runs was 9.8 EIU (co• et al. 1984). On the other hand, it is men• efficient of variation 22.6 %), figures that tioned elsewhere as being a useful test for would presumably have been lower if the the detection of latent toxoplasma infection number ofsamples tested had been higher.

Acta vet. scand. vol, 31 no. 4·1990 Toxoplasma antibodies in swine sera 419

ELISA (EIU ) 158.1 - 168.8 o negative in a l l tests 14EL 1 - 158.8 .. 131L 1 - 148.8 121L 1 - 138.8 g:gpositive in 118,1 - 128.8 one test -+- 188.1 118.8 98.1 - 188.8. S8.1 98.8 II positive in tHO tests 78.1 - S8.8 68.1 - 78.8 58.1 68.8 • posi tive in thr ee tests 48.1 - 58,8 yy 38.1 - 48,8

28,1 - 38.8 18,1 - 28.8 8,1 - 18,8 1 nUMber < EI 1 of sera 8 2 4 6 S 1E1 12 14 16 1S 2E1 Figure 2. Distribution of the results of IFAT, !HAT and the LA test with respect to 100swine sera in relation to the ELISA results. The column between 10 EIU includes sera giving 0.1-10.0 EIU, that columnbetween EIUsera with 10.1- 20.0 EIU etc.

The qua lity of the polystyrene microtitre test sample to that ofthe negative control se• plates may effect the reproducibility of rum , or comparisons with a standard curve ELISA results (Chessum & Denmark 1978, for positive control sera (Burre//s & Dawson van Loon & van der Veen 1980). Chessum & 1982). This variety makes it difficult to Denmark found that the antigen attached it• compare the results obtained by different self unevenl y to the plates in some batches, authors (Wilson et al. 1987). The need for readings for the same sample on some plates standardization ofboth the test itself and the differing more than 10 % from the arithme• mode of expressing the results has been tic mean in most of the wells. Use of such emphasized (Uggla 1986, van Loon & van plates for diagnostic serology would lead to der Veen 1980). The availability of good erratic results. ELISA is highly sensitive , but control sera of known toxoplasma antibody it also seems to be sensitive to all kinds of titres for each species would be ofgreat help changes in procedure, which may influence in attempting to render the results of dif• the reproducibilit y of the results (Voller et ferent selfmade ELISA systems comparable. al. 1976). Even though an imals with a low positive ELISA test results can be expressed in vari• titre may harbour toxoplasma tissue cysts, it ous ways, e.g. as end-point titres , absorbance is true tha t the higher the ant ibody titre is, values, ratios of the optical density of the the greater is the possibility ofa latent infec-

Acta vet. scand. vol. 31 no. 4 • 1990 420 V. Hirveld-Koski tion (Work 1967, Katsube et al. 1972, Hel• rinary Department of the Ministry of Agriculture lesnes et al. 1978). Results < 10 EIU in the and Forestry, the Finnish Cultural Foundation, present ELISA procedure were negative, and the Finnish Veterinary Science Foundation and results > 70 EIU were positive, as ascer• Oy Hoechst Fennica Ab. tained from the reference tests (Fig. 2). References About half of the sera with 10.1-70.0 EIU Anonymous: Serodiagnosis of toxoplasmosis, ru• were positive in IFAT or the LA test or bella, cytomegalic inclusion disease, herpes both, and the rest were negative. The agree• simplex. series No.5, Procedu• ment between ELISA and IFAT was best ral guide. Center for Disease Control, Atlanta with a cut-off of 30 EIU and that between 1980. ELISA and the LA test with a cut-off of 50 Balsari A. Poli G. Molina V. Dovis M. Petruz• EIU and there was some fluctuation in zelli E. Boniolo A. Rolleri E: ELISA for toxo• agreement values between ELISA and IHAT plasma antibody detection: a comparison with (Fig. I). It is difficult to find a satisfactory other serodiagnostic tests. J. c1in. Pathol. 1980, 33. 640-643 . cut-off for ELISA which will produce as few Boch J. Neurohr B: Vorkommen latenter Toxo• false negative and false positive results as plasma-Infektionen bei Schweinen in Slid• possible. When this test is used for screening deutschland und deren Nachweis mit IFAT toxoplasma antibodies in porcine sera, all und !HAT. (Latent toxoplasma infection in sera giving 10.1-70.0 EIU should be con• pigs in southern Germany and diagnosis using firmed with a test which has a good specifi• the indirect fluorescent antibody test and the city, e.g. IFAT. indirect haemagglutination test). Tierarztl, Although ELISA has been reported to be a Umschau 1982,37.820-826. rapid, sensitive and easily performed test for Boch J: Toxoplasma - und Sarcocystis-Infektio• routine serology and for screening of toxo• nen der Haustiere. (Toxoplasma and sarcocy• plasma antibodies (Walls et at. 1977, Ca• stis in domestic animals). Wien. Tierarztl, Mschr. 1973,60.337-341. margo et al. 1978, Woodward 1982, Walt• Burrels C. Dawson AMcL: ELISA methodology: man et al. 1984, Pallangyo et al. 1985, variations in technical procedures. In: Ward• Uggla 1986), it has also been considered ley RC, Growther JR (eds.): The ELISA: technically more difficult than LA, IHAT or enzyme-linked immunosorbent assay in vete• [FAT (Wilson et al. 1987). The performance rinary research and diagnosis. Martinus Nij• of ELISA needs special equipment and a hoff Publishers. London 1982, p, 1-9. skilled staff. These may be readily available Camargo ME. Ferreira AW. Mineo JR, Takiguti in laboratories working daily with serology, CK. Nakahara OS: Immunoglobulin G and but scarcely in abattoirs, the laboratories of immunoglobulin M enzyme-linked immuno• which are equipped mainly for the perform• sorbent assays and defined toxoplasmosis sero• logical patterns. Infect. Immun. 1978, 21. 55 ing ofsensory and bacteriological analysis. -58 . Chessum BS. Denmark JR: Inconstant ELISA. Acknowledgement Lancet 1978,21.161. I wish to thank Ms. Anneli Salonen and Ms. Ritva Dubey JP. Murrell KD. Fayer R. Schad GA: Di• Kauppi for their skilful technical assistance. My stribution of Toxoplasma gondii tissue cysts warm thanks to Prof. Jorma Him, Prof. Jaakko in commercial cuts of pork. J. Amer. vet. med. Tuomi, Dr. Pirjo Veijalainen and lie. vet. med. Ass. 1986,188.1035-1037. Matti Aho for their criticism of the manuscript. Felice G. Meroni V. Carnevale G, Olliaro P. Ca• This work was supported financially by the Vete- rosi G: Comparison of ELISA and indirect im-

Acta vet. scand. vol. 31 no. 4 - 1990 Toxoplasma antibodies in swine sera 421

munofluorescence in the detection of IgG and gondii seroepidemiological surveys. Tohoku IgM anti toxoplasma antibodies. Boll. 1st sie• J. expoMed. 1985, /47. 349-356. roter. Milan . 1983,62.445-450. Toma si JP. Barka N. Stadtsbaeder S: Serodiagno• Hellenes I. Mohn SF. Melhuus B: Toxoplasma sis of human G and M immunoglobulins to gondii in swine in south-eastern Norway. Toxoplasma gondii by ELISA using whole Acta vet. scand. 1978, 19. 574-587. tach yzoites as : a comparative study Hutchison WM. Dunachie JF. Siim JC. Work K: with the indirect haemagglutination (IHA) Coccidian-like nature of Toxoplasma gondii. and (IFA) tests. Med. Brit. med. J. 1970, I. 142-144. Microbiol. Immunol. 1986,175.261-269. Karim KA. Ludlam GB: The relationship and Turunen HJ: Detection of soluble antigens of significance of antibody titres as determined Toxoplasma gondii by a four-layer modifica• by various serological methods in glandular tion of an enzyme immunoassay. J. elin . and ocular toxoplasmosis. J. elin . Pathol. Microbiol. 1983,17.768-773. 1975,28,42-49. Uggla A: Toxoplasma gondii in farm animals. Katsube Y. Hagiwara T, Imaizumi K. Hanaki T. Some immunodiagnostic methods and their Nobuto K: Reliability of the dye and modified potential use. Thesis. Swedish University of tests for the latent infection Agricultural Sciences, Uppsala 1986, pp. 56. of Toxoplasma. Jap. J. vet. Sci. 1972, 34, van Loon AM. van der Veen J: Enzyme-linked 123-133. immunosorbent assay for quantitat ion of toxo• Lin TM. Halbert SP. O'Connor GR: Standardized plasma antibodies in human sera. J. elin. quantitative enzyme-linked immunoassay for Pathol. 1980, 33. antibodies to Toxoplasma gondii. J. elin . van Knap en F. Franchimont JH. van der Lugt G: Microbiol. 1980, /I. Prevalence of antibodies to toxoplasma in Lufi BJ. Remington JS: Toxoplasmic encephali• farm animals in the Netherlands and its impli tis. J. infoDis. 1988,157. 1-6. cation for meat inspection. Vet. Quarterly Lovgren K. Uggla A. Morein B: A new approach 1982,4. 101-105. to the preparation of a Toxoplasma gondii Verhofstede C. Sabbe L. van Renterghem L: membrane antigen for use in ELISA. J. vet. Ability of enzyme-linked immunosorbent as• Med. B 1987,34. 274-282. says to detect early immunoglobulin G ant i• Martin SW. Meek AH. Willeberg P: Veterinary bodies to Toxoplasma gondii. Eur. J. elin . Epidemiology. Iowa State University Press, Microbiol. 1987,6.147-151. Ames, Iowa 1987,343 p. Violand SA . Mitchell TG. Kleeman KT: Compari• Mineo JR . Camargo ME. Ferreira AW: Enzyme• son of an enzyme-linked immunoassay and a linked immunosorbent assay for antibodies to quantitative indirect fluorescent antibody test Toxoplasma gondii polysaccharides in human for detecting antibodies to Toxoplasma gondii. toxoplasmosis. Infect. Immun. 1980, 2 7. 283• J. elin. Microbiol. 1982,16.341-344. 287 . Voller A. Bidwell DE. Bartlett A. Fleck DG. Per• Oliver DG. Dreesen DW. Prickett MD. Waltman kins M. Oladehin B: A enzyme• WD. Poy DS. Blue JL: Comparative study of immunoassay for toxoplasma antibody. J. elin. enzyme-linked immunoassay with standard se• Pathol. 1976, 29. rological tests for the diagnosis of toxoplas• Walls KW. Bullock SL, English DK: Use of the mosis in swine. Proc . 3rd Int. Symp. of enzyme-linked immunosorbent (ELISA) and World Association of Veterinary Laboratory its microadaptation for the serodiagnosis of Diagnosticians, Ames, Iowa 1983, pp. 163• toxoplasmosis. J. elin . Microbiol. 1977,5.273 170. -277. Pallangyo KJ. Suzuki H. Fukumoto Y. Matsu• Waltman WD. Dreesen DW. Prickett MD. Blue moto K: One point dilution enzyme-linked im• JL. Oliver DG: Enzyme-linked immunosor• munosorbent assay (ELISA) for Toxoplasma bent assay for the detection of toxoplasmosis

Acta vel. scand. vol. 31 no. 4 • 1990 422 V. Hirveld-Koski

in swine: Interpreting assay results and com• Sammanfattning paring with other serologic tests. Amer. J. vet. Evaluering av ELISA jOrundersokning av Res. 1984,45. 1719-1724. antikroppar mot Toxoplasma gondii hos svin. Wilson M. Ware DA. Walls KW: Evaluation of ELISA (enzyme-linked immunosorbent assay) commercial serodiagnostic kits for toxoplas• jamfors med IFAT (indirect fluorescent antibody mosis. J. c1in. Microb iol. 1987, 25. 2262• test), IHAT (indirect haemagglut ination test) och 2265. LA (latex agglutination) test for att mara antikrop• Woodward BC: Evaluation of an enzyme-linked par mot Toxoplasma gondii i sera av 100 svin. immunosorbent assay specific for human Alia sera med < 10 EIU var negativa med andra immunoglobulin G as a screening test for tester och alia med > 70 EIU var positiva med I, detecting anti-Toxoplasma antibodies. J. c1in. 2 eller 3 av de andra testema. Halften av sera med Microbiol. 1982,16.367-372. 10-70 EIU var positiva med andra tester och half• Work K: Isolation of Toxoplasma gondii from ten var negativa. Det basta "cut-off' vardet for the flesh of sheep, swine and cattle . Acta ELISA var 30 EIU jamfort med IFAT och 50 EIU Pathol. Microbiol. scand. 1967, 7/ ,296-306. jamfort med LA test. Om man viII undvika fel positiva resultat med ELISA, borde alia sera med 10-70 EIU testas med t. ex. IFAT . ELISA ar sensitiv och lamplig for undersokning av stora mangder provo Det kan linda vara tek• niskt for svart att utfora pa slakterilaboratorier, vars kapacitet ar inriktad mest pa bakteriologiska och sensoriska undersokningar,

(Received September 8, 1989; accepted January 4, 1990).

Reprints may be requested from: Varpu Hirvela-Koski, National Veterinary Institute, Regional Laboratory Oulu, P. O. Box 517, SF-90 10I Oulu, Finland.

Acta vet. scand. vol. 31 no. 4 • 1990