key: cord-0897445-ro3x3qa3 authors: Ingram, George A.; Al-Yaman, Fadwa title: A comparative assessment of four serological methods used in the detection and measurement of anti-parasite antibodies in the serum of the amphibian, bufo viridis date: 1988-04-30 journal: International Journal for Parasitology DOI: 10.1016/0020-7519(88)90147-6 sha: bcea326da7ffbd034aa531671d14f0b907e561cc doc_id: 897445 cord_uid: ro3x3qa3 Abstract Antibodies against Crithidia fasciculata choanomastigotes were detected in green toad (Bufo viridis) sera by direct agglutination, indirect haemagglutination (IHA), complement-fixation test (CFT) and enzyme-linked immunosorbent assay (ELISA), Correlation coefficients (r) were calculated for comparisons between each of the techniques and regression formulae derived in order to convert antibody levels as determined by one immunological method to that of another. The highest mean titre obtained by ELISA was approximately 1.5–3.5 times greater than those obtained by the other techniques whilst CFT gave the lowest values. IHA and ELISA titres were affected by different preparations of the crithidial antigen extracts. Highly significant r values were determined for control sera when IHA was compared to ELISA (r > 0.79), and to both CFT and ELISA with immune animals (r > 0.96). ELISA would seem most applicable for screening other lower vertebrates for anti-parasite antibodies especially in areas of human disease prevalence. AMPHIBIANS are frequently parasitized by various protozoans present in body fluids and tissues and in some cases the parasites cause serious and debilitating diseases (Abrams, 1969; Roudabush & Coatney, 1937) . Trypanosomatid flagellates, in particular trypanosomes, have been reported in and isolated from the blood of frogs, toads and newts (Bardsley & Harmsen, 1973; Woo & Bogart 1986) . Throughout the investigations into trypanosomatid infection in humans and mammals, sera have been examined for both anti-parasite antibodies and parasite antigens using various immunodiagnostic techniques (Strickland & Hunter, 1982; Tiru & Hennessen, 1985) . However there is a dearth of information pertaining to the use of serodiagnostic methods to detect kinetoplastid infections and resultant antibody production in poikilothermic vertebrates with only reptiles having been studied (Dollahon, Hager & Hua, 1983; Ingram & Molyneux, 1983a , b. 1984a . In this paper we present the results of acomparative assessment of four serological tests (direct agglutination, indirect haemagglutination, complement-fixation test and ELISA) used to detect and to determine the levels of antibodies in the sera of green toads (B. viridis), used as experimental models. These amphibians had been injected with the choanomastigotes of C. fusciculutu, a trypanosomatid flagellate. The parasites were chosen because of their ease of culture under laboratory conditions and more importantly because of their presence in the blood and alimentary canal of ranid anurans under natural conditions (Smyth & Smyth, 1980) . To the authors' knowledge, this is the first report of the use of an immunoenzyme method to detect antibodies in amphibians. Furthermore, there are no previous data concerning anti-parasite antibody detection in amphibian serum. MATERIALS AND METHODS Injecrion und serrr. Before injection of the parasites. a small sample of blood was taken by cardiac puncture from each of the toads and inspected for any current infection (with naturally-occurring trypanosomatid flagellates) by smear, wet mount preparation and slope culture as described elsewhere (Ingram & Molyneux, 1983a. b ). In addition. blood smears were also examined by an immunoenzyme method reported previously by Ingram and Molyneux (1 Y84b, c) but using a rabbit anti-C., firscic~tkctcr serum/swine anti-rabbit immunoglobulins antiserum/ peroxidase-rabbit antiperoxidase/amino-ethylcarbazole substrate system. Furthermore, the gut contents and PBS extracts of randomly selected insects and worms respectively were also examined microscopically and by culture for the presence of trypanosomatids. The choanomastigotes in culture overlay were centrifuged at 250 X g for 5 min. The overlay was removed and the parasites were washed three times in phosphate buffered saline-PBS. pH 7.3 (I 45 mM-NaCI; 7.53 mM-Na?HPO, and I 30 mM-NaH,PO,. 2H,O in de-ionized water). The number of parasites was counted and the suspension adjusted with PBS to give the required dose for injection purposes.Toads were given a single intraperitoneal (IP) injection of 16 X IO" choanomastigotes in PBS. Control animals consisted of those given PBS IP and normal, uninjected toads. The toads were anaesthetized. bled, killed and their weights and lengths noted. The parasite-and PBS-injected animals were bled at 7-day intervals. The uninjected controls were sampled at random intervals throughout the duration of the experiment. In order to detect parasite infection. blood was examined in a similar manner to that obtained from prrinjected animals and the uninjected controls. The sera were isolated and stored at -20°C. Specificifv. The promastigotes and procyclics of Lershmania herfigi her@ and Trpanosoma brucei brucei respectively, both kinetoplastid flagellate species related to c'. ~uscic~ulrtu, were used in the antibody assays to examine for possible non-specific reactions in the toad sera. Antisera preparation. Rabbit anti-toad serum was produced by an immunization schedule as described previously (Ingram & Alexander. 1976 ) and the immunoglobulin tttres of the antisera obtained were estimated by either countercurrent electrophoresis or ELISA (Ingram & Molyneux, 1983a). Antigen extract. Parasite antigen extracts were produced in two ways for use in the immunological techniques. The choanomastigotes were centrifuged at 1700 X g for IO min in cold PBS. pH 7.3. The pellet formed was resuspended in PBS and then washed and centrifuged a further three times. Prior to use, the cells were subjected to either freezing and thawing (F&T) or sonication (Son) treatments. In the former case the pellet was resuspended in chilled PBS, broken up by mixing and the parasites F&T at 3-min periods for 20 min. The material wsas then centrifuged to remove cell debris and the supernatant protein concentration measured by the Lowry method. Alternatively, after addition of cold buffer, the pellet was disrupted by ultrasonication for three 4-min intervals whilst the mixture was kept chilled. The sonicated material was then left overnight at 5 "C to further remove any protein. It was then centrifuged at 2900 X g for I5 min and the amount of soluble protein in the supernatant determined. Agglutination assay Two-fold serial dilutions of toad sera. inactivated by heating at 3X"C for 20 min to destroy naturally-occurring complement activity, were preparcd with PBS. pH 7.3 (containing 24X mr+NaCI). To each dilution wsas added an equal volume of choanomastigotes (I 5 X IO cells ml-') and the mixtures incubated at 30°C for 45 min. The direct agglutination (DA) end point titre was regarded as that dilution in which visible agglutination was observed when compared to the PBS/parasite controls. Normal toad sera were examined for the presence of natural haemagglutinins against sheep crythrocytes (ShE) before commencing the IHA test as described by Weir (IY7X). ShE were tanned with 3.3 X It)-' mg ml-' tannic acid and coated with either F&T or Son ant&n extract containing 0.Y mg ml-' protein. Doubling dilutionr of inactivated sera were made and to each was added the same volume of 2% tanned and coated ShE.The test samples were incubated at 37°C for 45 min followed by{ overnight at 4C. The samples were then examined and the degree of haemapglutination assessed. Untanned, tanned and antigencoated ShE were used as controls. Antigen. toad antisera and either BVS or GPC were mixed together and fixation allowed to occur overnight at 5 "C. Sensitized ShE were then added and the mixtures incubated at 37°C for 60 min after which the end-points were scored. Following further incubation for 12 h at SC, the samples under test were re-examined. The complement-fixing antibody (CFA) titre was taken as that dilution which gave 50% hacmolysis. ELlSA. A modification of the method of Chandler. Cox. Premier & Hurrell (I YX2) was used. The concentrations of antigen. rabbit anti-toad antiserum and enzyme-conjugate used for the ELISA were determined by chequerboard titration. The F&T and Son antigen extracts (containing I mg ml-' protein) were prepared in 0.053 M-carbonate/bicarbonate coating buffer pH Y.6: C'. fi~scicrtlartr was adjusted to 2.5 X 10" cells ml-' in coating buffer and the three antigen preparations were separately dried onto the plates by overnight incubation at 30°C. After washing with PBS (containing O.OS% Tween 20) pH 7.3, a range of two-fold serial dilutions of control and immune sera were added to the appropriate wells and the plates were reincubated at 37°C for 30 min. They were washed again. rabbit anti-toad antiserum (ELISA titre I : 1024) at a dilution of 1 :7S was added and the plates were incubated as before. After re-washing, sheep anti-rabbit I@ immunoglobulin comugated to urease. diluted I in 120. was dispensed into the wells and the plates were similarly incubated. After a final three washes with PBS/Tween followed by four washes with distilled water, urease substrate (Sera-I.ab, Crawley. U.K.) was added to the wells. The plate\ were subjected to a final incubation at 37°C for 30~40 min and the reaction halted by the addition of I % thiomersal solution (v/v). The end point antibody titre was considered to be that dilution which was visually different from the appropriate reference controls included with each experimental run. The mean toad serum antibody titres, S.E.M. and ranges in the control and C. fasciculuta -injected groups, together with the number of sera-containing detectable antibodies, are given in Table 1 . The mean value, range in titres and number of animals in which antibodies were detected were lowest for the CFT and highest by ELISA. Of the 29 control and 38 parasiteinjected toads, 93 and 92%, respectively contained antibodies detectable by ELISA. When all four immunological assays were applied to each individual serum, in all cases antibodies were detected by at least two or more of the methods used. However the background 'positive antibody' titres in the control animals ranged from 0 to 2-j depending on the technique employed. Therefore values higher than 2-' were considered to be positive for the parasiteinjected toads. Furthermore comparisons between control and immune sera for each method revealed significant differences in antibody titres (P< 0.01; Student's t-test). Moreover, 'antibody' levels were not significantly different between the two control groups (P> 0.05). The titres against L. hertigi and T. brucei varied from 0 to 2-j and from 0 to 2-1 in the control and immune sera, respectively. The results for each technique were compared in turn with those for the other methods and the half matrix of the Pearson product-moment correlation coefficients (r) calculated for all experimental animals ( Table 2 ). The significance of each r value was tested by the t-test. In the case of the control sera, r values ranging from 49 to SY% (P< 0.01) were calculated whilst overall higher correlations were determined for sera from immune animals which varied from 82 to 99% (all I'< 0.00 1). In order that the results of the present study can be used by other investigators for comparative purposes with different experimental models, regression formulae to convert the antibody titres as determined for each technique to those of another are given in Table 3 . The regression equations, based on the rectilinear relationship Y = mX + c, were only calculated for the highest mean titre found for each of the four immunological methods. The mean antibody titres against C. fasciculata and the number of control and immune sera containing detectable antibodies were the lowest for DA and CFT, intermediate for IHA and highest for the ELISA method (Table 1) . These findings may reflect the different sensitivities of the immunological techniques used (Voller & De Savigny, 1 Y8 1). The classical agglutination test has often been used for antibody titration in amphibian immunobiological studies (Cooper, 1976 ). Significant correlations (P < 0.0 1) were obtained for both control (range 4Y-76%) and immune sera (83-Y2%) when the DA titres were compared to the values found for the other methods. Although DA is simple to perform, preconditions of the test include antigen-type specificity, the non-immobilization and non-autoagglutination of parasites and usually the use of living cells. In the current study, loss of cell motility was observed in some instances and the possibility of inclusion of nonviable cells in others cannot be ruled out. Nevertheless caution must be taken in the interpretation of natural 'antibody' levels in view of the detection of 'antibodies' in normal BVS against L. hertigi and T. brucei with titres within the range found for C. fuscicdata. Positive results for normal BVS would suggest the presence of low amounts of specific immunoglobulins induced by a current infection with or previous exposure to C. fasciculata parasites. However, the low levels of naturally-occurring 'antibodies' in normal BVS may also have been stimulated by the environmental presence of micro-organisms or other trypanosomatid flagellates which non-specifically cross react with shared cell wall carbohydrate antigenic determinants (Andrews, Reilly, Ferris & Hanson, 1965; Schnaidman, Yoshida, Gorin & Travassos, 1986; Sharabi & Gilboa-Garber, 1980 ). In the present study, the low levels of 'antibody' in sera from the control groups are not likely to be caused by a current infection because no increase in titres were found throughout the 1 O-week duration of the experiment. The lack of detection of C. fasciculata in blood microscopically or by culture coupled with the finding of parasite antigen(s) in blood up to 2 weeks postinjection by the immunoenzyme method suggests that 8. Grids possesses an efficient immune system responsible for the rapid elimination of antigen. Therefore it was not possible in the work reported here to correlate the level of parasitaemia and antibody titres. However the exposure of B. Grids to C. fasciculata evoked a specific immune response and resulted in significantly increased levels of serum antibody. Under normal environmental situations, it is feasible that naturally-occurring immunoglobulins in BVS restrict parasite numbers to below a potential infectious threshold. The finding of serum antibody titres above those of normal background levels in amphibians or other animal hosts in similar habitats or areas endemic for certain diseases would indicate current infection with parasites, other infectious agents or pathogens within a population. No haemolytic anti-complementary activities were detected in normal BVS against the antigen extracts unlike previous reports for amphibians (Gigli & Austen, 197 1) . Whereas the CFT is frequently nonspecific and inconvenient for handling many samples, it can utilize crude, soluble parasite antigen extracts. As with DA cross reactivity can often lead to false positive results. When the results obtained by CFT were compared with each other and with IHA and ELISA, the lowest range of correlations, although significant and similar to those determined for the DA comparisons, were found for control BVS (50-77%; IgM which is an efficient complement fixer and also a good agglutinin (Atwell & Marchalonis, 1976; Yamaguchi. Kurashige & Mitsuhashi, 1973) . However, the 'antibodies' in normal BVS would be present in limited amounts, fix complement less effectively and hence result in low background CFA values. This could also account for the lowest correlations found when the CFA titres were compared to those of the other methods for the controls. The source of complement seems to be important for the efficient fixation of toad antibodies. The use of toad serum as homologous complement source gave a higher mean value and usually slightly higher individual endpoint titres in both control and immunized animals compared to the antibody levels obtained with heterologous GPC. Homologous serum has proved a reliable source of complement for the fixation of immunoglobulins in other anurans (Alexander & Steiner, 1980; Sekizawa, Fujii & Katagiri, 1984) . However the use of commercially available GPC is also known to initiate good fixation in amphibian species (Lallone, Chambers & Horton, 1984 : Romano, Geczy & Steiner, 1973 ). In the current study, 50% correlation (P < 0.0 1) and 92"/,, (P< 0.00 1) were found for control and immune sera. respectively when the different complement sources were compared. Low (SO-77%, controls) and high (86-94X, immunized) but significant (I'< 0.01) r values were determined in comparisons between IHA and CFT. IHA is prone to lack specificity in some cases and, in contrast to the CFT, usually requires highly purified antigen preparations but is easier to perform. ELISA gave the highest percentage positive titres in all the samples examined and appears to be the most sensitive of the techniques used to detect anticrithidial antibodies in toad sera. ELISA was easy to peform, specific, An important factor which affects the values of the antibody titrcs is the preparation of the antigen extract (Crouch & Raybould, 1983; Pappas. Hajkowski, Cannon & Hockmeyer. 1984) . Son antigen preparations gave higher titres than F&T antigen extracts. However, the coating of ELISA plates with whole cells produced the highest values. In the work reported here, similar batches of antigen were used thus reducing potential variations due to different preparations. It is of interest to note that a comparison between IHA and ELISA control titres revealed 79-89% (P < 0.00 I ) significant correlations and similar numbers of positive animals. This implies that the above two techniques have similar sensitivities in the screening of normal BVS for anti-parasite antibodies. Nevertheless the method of antigen preparation ma> not be a salient criterion for antibody estimation in immune sera because correlation values of over 89% (f'< 0.001) were found when IHA titres were compared to those of ELISA. ELISA seems appropriate for use in serodiagnostic surveillance programmes. applied to different amphibian species or other aquatic and semi-aquatic lower vertebrates, to detect antibodies stimulated against diverse parasite environmental pathogens. Furthermore this technique would be of value in screening lower vertebrates for potential carriers or reservoirs of infective kinetoplastid tlagellates or indeed other pathogenic micro-organisms. Such information may reflect the health status of animals within a population and aid in the determination of specific epidemiological and aetiological features of zoonoses and epizootics prevalence. Ac~k~f~~*,lef~~.lg~l?~e~~t,s-'I'his work was undertaken whilst one of us (GAI) was on an EEC Funded Project contracted to the British Council. in part subcontracted to the University of Salford, for the development of the Faculty of Science at Yarmouk University. Jordan. GA1 would like to thank Dr S. K. Abdul-Hafez for the use of his laboratory facilities for part of the work, Dr N. Ishmail for assistance with the bleeding of the animals and Dr Ellen Lee of the University of Salford Computing Centre for advice on the use of the computer for statistical analysis of the results. Diseax\ in an amphihian colony The first component of complement from the bullfrog Runa cutrsheianu: functional properties of Cl and isolation of subcompon-entClq Leptospiral agglutinins in sera from southern Illinois Herpetofauna Immune response in XenopuA iuevis and immunochemical properties of the serum antibodies. Immunology 24: 109-l 1x.