Thick-smear Plasmodium falciparum antigen from in vitro cultures for the indirect fluorescent antibody test

MALARIA Thick-smear Plasmodium falciparum antigen from in vitro cultures for the indirect fluorescent antibody test VIJAYAMMA THOMAS 1 & J. T. PONNAMPALAM 2 An important prerequisite for serological studies is the availability of specific antigens with which a high, consistently reproducible degree oftest sensitivity can be obtained. In the present study, thickfilms ofP. falciparum antigen were madefrom in vitro cultures in order to evaluate this antigen in terms of the sensitivity and reproducibility of the indirect fluorescent antibody test, as compared with an antigen prepared from washed infected erythrocytes of Aotus trivirgatus monkeys. The test was performed in 132 serum samples from 4 different sources. The results showed that the sensitivity and reproducibility of the two antigens compared well. No binding of donors' antibody to antigen was noticed within the 24-hour period of in vitro culture. The indirect fluorescent antibody (IFA) test has been used extensively to study the production and persistence of antibodies to malaria in man (1). The many uses of serological techniques in the field of malaria were described by Voller & Bruce- Chwatt (7) and by Draper et al. (2). Specific antigens with a high degree of consistently reproducible sensitivity are among the most impor- tant prerequisites for any study in serology. The preparation of thick-smear antigen from washed infected erythrocytes of A. trivirgatus monkeys has been described by Sulzer et al. (5). However, many human strains of P. falciparum are difficult to estab- lish in this species of monkey (8), which, moreover, is not found in Malaysia. The possibility of prepar- ing suitable P. falciparum antigen from in vitro cultures has been raised (9). A recent report by Lopez-Antufnano (3) was seen only after this work was completed. He used thick- smear antigen prepared from in vitro cultures and obtained constantly reproducible results, which were comparable with those obtained when a P. falci- parum antigen prepared from A. trivirgatus cells was used. However, no details of these results were given in the report. 1 Lecturer, Department of Parasitology, Faculty of Medicine, University of Malaya, Kuala Lumpur, Malaysia. Requests for reprints should be addressed to Dr Thomas. ' Senior Malariologist, Institute for Medical Research, Jalan Pahang, Kuala Lumpur, Malaysia. The present tests were carried out to prepare thick-smear P. falciparum antigen from in vitro cultures and to evaluate its sensitivity and reproduci- bility in the IFA test, as compared with those of a known lot of antigen prepared from washed infected erythrocytes of A. trivirgatus monkeys. MATERIALS AND METHODS In vitro culture of P. falciparum The in vitro culture techniques described by Rieck- mann et al. (4) were used. About 5-10 ml of blood containing moderate numbers of ring forms were taken from each of a group of patients who had not received antimalarial treatment during the previous week. If the infection was very heavy at the time when the blood was drawn, the trophozoites did not reach maturity in in vitro cultures. Only cultures in which most of the rings had grown to be early schizonts were used to prepare antigen. Preparation of antigen from in vitro culture The technique used to make thick-smear antigen was that of Sulzer et al. (5). On removal from the incubator, the culture containing P. falciparum schizonts was centrifuged and the supernatant was removed. The cells were resuspended in 10 times their volume of phosphate-buffered physiological saline (PH 7.6) and centrifuged. The saline was then 3318 -33- BULL. WORLD HEALTH ORGAN., Vol. 52, 1975 V. THOMAS & J. T. PONNAMPALAM discarded. Resuspension and centrifugation were repeated at least five times. The cells were then resuspended in phosphate-buffered saline to their original volume. Thick smears were prepared from the suspension and stained, and the number of P. falciparum schizonts per high-power field was determined. If too many schizonts were present, the cell suspension was diluted with phosphate-buffered saline. If there were too few parasites, the suspension was concentrated by centrifugation and removal of excess saline. About 15-20 schizonts per high-power field in a thick smear were considered to be a desirable number. Antigen slides were prepared by dispensing one drop of the suspended cells in each of the 6-mm wells on a microscope slide. The drop was spread so as to make a thick smear within the circle. Care was taken that the cells did not settle in the Pasteur pipette. The thick smears were then allowed to dry in the air at room temperature for about 1 h. Properly dried slides were wrapped in tissue paper and stored at -70°C until needed. Performance of tests To obtain comparable results, both series of anti- gen slides (prepared from A. trivirgatus monkeys and from in vitro cultures) were tested simultaneously for all tests and control serum dilutions. Slides of the first antigen-the sensitivity, specificity, and repro- ducibility of which are known-were supplied by Dr A. J. Sulzer, Parasitology Branch, Center for Disease Control, Atlanta, GA, USA. The test methods used were those described by Sulzer et al. (5). The antigen slides were removed from stor- age, placed in staining racks until they had reached room temperature, and washed for 5-7 min in distilled water on a slowly moving rotator. After the slides had been washed, they were removed from the bath and allowed to dry. The serum samples to be tested were diluted with phosphate-buffered saline (pH 7.6) in 4-fold steps, starting with a dilution of 1: 4. Then 0.05 ml of each dilution was dispensed on a labelled antigen mount. The antigen slides with various dilutions of anti- serum were placed in a moist chamber, covered, and incubated at 37°C for 30 min. After incubation, the excess serum was removed from each slide with a gentle stream of phosphate- buffered saline. The slides were placed in a staining rack in saline on a slowly moving slide rotator for 15 min. They were then dried and an appropriate dilution of anti-human-gamma-globulin conjugate (0.05 ml) containing Evans blue counter-stain was distributed over all the antigen mounts. The slides were covered and incubated again at 37°C for 30 min in the moist chamber. The excess conjugate was removed by a gentle stream of saline and by washing the slides for a minute or so. The slides were dried. A drop of buffered glycerol pH 9 was placed on each mount and covered with a coverslip. All tests were evaluated on a Leitz Ortho- lux fluorescence microscope equipped with BG 12 and UG 1 exciter filters and a Leitz 470 ocular (suppression) filter. All readings were made by the first author. Serum samples A total of 132 serum samples, including known positive and negative sera and control samples, were tested simultaneously with the two antigens. Of the 132 samples, 53 were obtained from patients with known falciparum infections who had shown vari- ous levels of parasitaemia. Nineteen were obtained from Orang Asli (aborigines) who had presented themselves at Gombak Orang Asli Hospital with various complaints. These patients came from vari- ous areas where malaria is endemic and where they were constantly exposed to infection with P. falci- parum and other species of malaria parasite. None of them was slide-positive to malaria when the serum was obtained. Twelve samples were from patients who had presented themselves at the University Hospital with one complaint or another. There were 28 samples from second-year medical students and other healthy people, and 20 control samples. In each experiment, one positive serum sample of known titre, one negative sample, and a phosphate- buffered saline control sample were included. RESULTS The results of the tests performed with the two antigens are given in Table 1. Of the 132 samples tested, 48 showed negative reactions to both antigens and 81 gave positive titres of 1: 16 and above. The highest titre obtained was 1: 4 096. Of the remaining 3 samples, 2 gave a positive titre of 1: 4 with both antigens, whereas the last sample showed a negative reaction with the antigen from in vitro cultures but a positive titre of 1: 4 with the other antigen. It was noticed that the thick, washed-cell smears from in vitro cultures did not lyse as easily as those from A. trivirgatus monkeys. Unlysed cells showed a greater degree of nonspecific background fluores- 34 MALARIA Table 1. Comparison of I FA titres with P. falciparum antigens derived from A. trivirgatus monkeys (A) and from in vitro cultures (B) on 132 serum samples from various sources Slide-piositive Medical Uiest oto Titres Slide-Pdonors Aborigines students Hospital samples a Total A B A B A B A B A B A B negative - - 2 2 26 27 10 10 10- 10- 48 49 1 :4 - - 1 1 2 1 - - - - 3 2 1 :16 4 3 2 2 - - 1 2 - - 7 7 1 :64 23 20 1 2 - - 1- -- 25 22 1 : 256 17 17 4 6 - - - - 1 + 2+ 22 25 1 :1 024 7 10 6 3 - - - - 8+ 8+ 21 21 1 :4096 2 3 3 3 - - - - 1+ - 6 6 Total 53 53 19 19 28 28 12 12 20 20 132 132 a -= negative; + = positive. cence. Therefore complete lysis was essential for the tests. All 53 serum samples from slide-positive persons were positive in the test. The lowest titre was 1: 16 and the highest was 1: 4 096 (Table 1). Thus the lowest significant diagnostic titre found in these tests was 1: 16-a result similar to that obtained by Sulzer et al. (5), who observed that a titre of 1 : 16 provided the most favourable balance of sensitivity and specificity in their results. Of the 19 serum samples from aborigines living in areas where malaria is endemic, all except 3 showed positive titres of 1: 16 and above. One of these 3 samples showed low reactivity, with a titre of 1: 4. Three of the positive serum samples produced high titres (1: 4 096) with both antigens (Table 1). None of these patients showed P. falciparum in the blood at the time of serum collection. This was to be expected, since they had lived, since their childhood, in areas where malaria is endemic. All but 2 of the 28 serum samples from medical students did not react with either antigen. One sample reacted with a titre of 1: 4 when tested with the antigen from A. trivirgatus monkeys, but was negative to the antigen from in vitro cultures. The remaining serum sample reacted with a titre of 1: 4 to both antigens. No reaction to either antigen was shown by 10 of the 12 serum samples from patients at University Hospital, which had been sent for tests including that for malaria. One of the two positive serum samples showed a positive reaction, with a titre of 1 : 16, and the other produced a titre of 1: 64 with the first antigen. Both gave a titre of 1: 16 with the antigen from in vitro cultures. The numbers of positive titres obtained simulta- neously with the two antigens for the 132 serum samples tested were as follows: 100 (52 positive and 48 negative reactions) produced identical titres with both antigens; 16 gave a titre 4 times as high with the antigen from A. trivirgatus monkeys as they did with the antigen derived from in vitro cultures, and the reverse was the case with the remaining 16 serum samples. These results show that the sensitivity of the antigen made from in vitro cultures was com- parable to that of antigen prepared from washed blood cells of infected A. trivirgatus monkeys. The reproducibility of titres with thick-smear anti- gens from in vitro cultures on known positive, negative, and other serum samples was studied. The results (Table 2) show good reproducibility of titres for 8 serum samples in 4-10 replicates. The variation of titres in most cases was ± one 4-fold dilution, although there was a ± one 16-fold difference. It was noticed that, if all possible variables of the tests were strictly controlled, the results were mostly within these limits. In none of the tests did a serum specimen known to be negative give positive results. Similarly, there was no fluorescence in phosphate- buffered saline control samples. The antigen there- fore seems to be sensitive, with a reasonably good degree of reproducibility. 35 V. THOMAS & J. T. PONNAMPALAM Table 2. Reproducibility of IFA titres based on several replicates of known positive, negative, and test serum samples with antigen from in vitro cultures Test samples No. of Reproducibility of titres with antigen from in vitro culture replicates negative 1 :14 1 :16 1 : 64 1 :256 1 :1024 1 :4096 phosphate- buffered saline 10 10 neg. control 10 10 serum 0031 4 4 serum 0084 4 4 pos. control 10 1 2 7 serum 0028 4 - 1 1 3 serum 0032 4 3 1 serum 0044 4 1 2 1 serum 0051 4 1 3 DISCUSSION One of the problems inherent in using antigen made from cultures might be the possible in vitro reaction of the antigen donor's antibody and para- sites as shown by Sulzer et al. (5). These authors found that, if the parasitized cells of P. vivax were washed free of the donors' serum within 30 minutes of the specimens' being drawn, antigen prepared from the cells was always satisfactory. On the other hand, if the whole blood was allowed to stand for several hours before washing, the plasmodia in the antigen preparations produced brilliant fluorescence in the saline solutions and negative controls. With P. brasilianum, however, positive reactions in nega- tive serum samples were noticed only after 3 days' storage of unwashed cells. Such positive reactions with saline solutions or negative control samples were not found in the present series of tests. It may be that the in vitro culture period of 24 h was not long enough for the antibody to have passed the red cell membrane. It is also possible that P. vivax, which causes the enlargement and consequent stretching of the membrane, permits a faster reaction of the antibody with the intraerythrocytic parasite in a matter of hours if it is not removed from the blood. Furthermore, the time required for in vitro binding of donors' antibody by Plasmodium may vary with the species, and 24 h may be too short a period for P. falciparum. The thick-smear antigen from washed cells has other advantages, as described by Sulzer et al. (5). It has been shown by Targett (6) that schizonts of P. falciparum as an antigen gave titres 2 or 3 times as high as those observed with ring forms and tropho- zoites. However, as schizonts of P. falciparum are seldom found in the peripheral blood of human donors, it is not possible to obtain schizonts for antigen from this source. A. trivirgatus monkeys, which can act as host only to certain adapted strains of P. falciparum, are not found in the region where this study was carried out. To import these monkeys and to maintain them in healthy condition is expen- sive. Since P. falciparum infections are still common in human beings living in areas where malaria is endemic, the preparation of P. falciparum antigen from in vitro cultures is practical and easy. ACKNOWLEDGEMENTS The authors thank Dr A. J. Sulzer, Center for Disease Control, Atlanta, GA, USA, for supplying the P. falciparum antigen slides, the conjugate, and the known positive and negative control sera. They are indebted to him for reading the manuscript critically and making very valuable suggestions. Thanks are due also to Dr H. T. Ong, Orang Asli Hospital, Gombak, Selangor, Malaysia, for providing serum samples from aborigenes. The first author is very grateful to Profes sor A. S. Dissanaike, Department of Parasitology, University of Malaya, for his continued interest and 36 MALARIA 37 support, without which this study could not have been carried out. She is grateful also to the University of Malaya for the grant that made this research possible. Thanks are due also to the China Medical Board for supplying various items of equipment, including a - 70°C freezer that was indispensable for the study. RtSUME' EMPLOI DE FROTTIS EPAIS D'ANT1GtNE DE PLASMODIUM FALCIPARUM PROVENANT DE CULTURES IN VITRO POUR L'EPREUVE D'IMMUNOFLUORESCENCE INDIRECTE Cette 6tude a pour objet d'evaluer l'utilite, comme antigene pour les 6preuves d'immunofluorescence indi- recte, de frottis 6pais de cellules lav6es de P. falciparum, obtenues & partir de cultures in vitro. Pour ces dernieres, on a utilis6 les techniques de Rieckmann et al. (1968). Apres 24 h de culture, les cellules parasitees ont ete lav6es a plusieurs reprises a l'eau physiologique tampon- n6e au phosphate, puis des lames d'antigene ont et6 prepar6es selon la m6thode d6crite par Sulzer et al. (1969). Cent trente-deux echantillons de s6rum, provenant de 4 sources diff6rentes, et comprenant des 6chantillons positifs et negatifs connus, ont 6t6 6prouves. Dans toutes les 6preuves, les anticorps anti-Plasmodium ont et6 recher- ch6s simultan6ment avec l'antigene obtenu a partir de la culture in vitro et avec l'antigene pr6par6 i partir d'6ry- throcytes parasites et laves du singe Aotus trivirgatus. Sur les 132 6chantillons 6prouv6s, la r6action a 6te n6ga- tive dans 48 cas en pr6sence des deux types d'antigene, et elle a donn6 des titres positifs de 1:16 et davantage dans 81 cas. Le titre le plus 6leve 6tait de 1: 4096. Sur les 3 echantillons restants, on a not6 deux reactions posi- tives a 1: 4 en pr6sence des deux types d'antigene, et, pour le dernier 6chantillon, une reaction n6gative en pr6- sence de l'antigene cultiv6 in vitro et une r6action positive a 1: 4 en presence de l'antigene provenant du singe A. trivirgatus. Sur les 132 6chantillons, 100 ont donn6 des titres iden- tiques en presence des deux lots d'antigene; 16 autres ont presente un titre 4 fois superieur avec l'antigene pro- venant du singe A. trivirgatus, et les 16 derniers, un titre 4 fois superieur avec I'antigene provenant de la culture in vitro. On a 6tudie sur 8 echantillons, en 4 a 10 exem- plaires, la reproductibilit6 des r6sultats obtenus avec les frottis 6pais d'antigene provenant de la culture in vitro. Dans la plupart des cas, la variation des titres etait de ± 1 dilution quadruple mais, dans un cas, la diff6rence 6tait de ±16 fois. On a constat6 que, si toutes les variables possibles des epreuves 6taient strictement controlees, la majorit6 des resultats se situaient dans ces limites. La sen- sibilit6 de l'antigene provenant de la culture in vitro 6tait donc bonne et soutenait favorablement la comparaison avec celle de l'antigene provenant du singe A. trivirgatus. La reproductibilite des r6sultats obtenus avec le nouveau lot d'antig6ne etait bonne 6galement. Aucune adsorption d'anticorps des donneurs sur l'anti- gene n'a ete notee au cours des 24 h correspondant a la culture in vitro. REFERENCES 1. COLLINS, W. E. & SKINNER, J. C. The indirect fluorescent antibody test for malaria. American jour- nal of tropical medicine and hygiene, 21: 690-695 (1972). 2. DRAPER, C. C. ET AL. The epidemiologic interpreta- tion of serologic data in malaria. American journal of tropical medicine and hygiene, 21: 696-703 (1972). 3. LOPEZ-ANTUIqANO, F. J. Falciparum malaria antigen slides for indirect immunofluorescence test made from in vitro cultures. Transactions of the Royal Society of Tropical Medicine and Hygiene, 68: 257 (1974). 4. RIECKMANN, H. H. ET AL. Effects of chloroquine, quinine and cycloguanil upon the maturation of asexual erythrocytic forms of two strains of Plasmo- dium falciparum in vitro. American journal of tropical medicine and hygiene, 17: 661-671 (1968). 5. SULZER, A. J. ET AL. Indirect fluorescent-antibody tests for parasitic diseases. American journal oftropical medicine and hygiene, 18: 199-205 (1969). 6. TARGETT, G. A. T. Antibody response to Plasmodium falciparum malaria. Comparisons of immunoglobulin concentrations, antibody titres and the antigenicity of different asexual forms of the parasite. Clinical and experimental immunology, 7: 501-517 (1970). 7. VOLLER, A. & BRUCE-CHWATT, L. J. Serological malaria surveys in Nigeria. Bulletin of the World Health Organization, 39: 883-897 (1968). 8. VOLUR, A. & O'NEILL, P. Immunofluorescence method suitable for large-scale application to malaria. Bulletin ofthe World Health Organization, 45: 524-529 (1971). 9. WORLD HEALTH ORGANIZATION. Cultivation tech- niques for the erythrocytic stages of malaria parasites. Bulletin ofthe World Health Organization, 47: 357-373 (1972).