Antibodies to Pf155, a major antigen of Plasmodium falciparum: seroepidemiological studies in Haiti*

Bulletin of the World Health Organization, 65 (3): 339-344 (1987) Antibodies to Pf155, a major antigen of Plasmodium falciparum: seroepidemiological studies in Haiti* P. DELORON,1 Y. T. DUVERSEAU,2 A. ZEVALLOS-IPENZA,3 R. MAGLOIRE,4 P. S. STANFILL,5 & PHUC NGUYEN-DINH6 The presence of malaria parasites and the serological antibody responses against whole Plasmodium falciparum and the PflSS antigen were studied in the population of a small rural locality in Haiti in December 1985. Only 7 (1.5%) of the individuals were found to be infected with P. falciparum, the only species observed. Antibodies to P. falciparum were detected in an ELISA in 38.2% of the sera, the positivity rates being age-related. Anti-Pf55 antibodies were detected in 12.5% and 13.6% of individuals by two different techniques used. The anti-Pf55 positivity rates increased only after 25 years ofage. No trends were detectedfor a clear-cut protective value ofPf155 antibodies against clinical malaria and further longitudinally conducted field surveys are needed to satisfactorily assess the potential protective effect of Pf155 antibodies. In the context of limited resources, a sound knowledge of local epidemiological situations is necessary for the appropriate application of malaria control measures. A potential additional control measure, which might become available in the not too distant future, is immune intervention. To advance the progress towards development of a malaria vac- cine and to help measure its potential impact, field studies are needed on the mechanisms involved in immune responses in persons exposed to malaria. Malaria is a major public health problem in Haiti, as in many other developing countries. In December 1985 the extent of malaria endemicity and the im- mune experience of the population of a small locality in Haiti was measured. The studies concentrated on the immune responses of the population to Pf155, a Plasmodium falciparum antigen with a relative molecular mass (Mr) of 155 000, which is deposited in the membrane of ring-infected erythrocytes (1). * From the Malaria Branch, Division of Parasitic Diseases, Center for Infectious Diseases, Centers for Disease Control (CDC), Public Health Service, U.S. Department of Health and Human Services, Atlanta, GA 30333, USA, and the Division d'Epidemio- logie et de Recherche, Service National des Endemies Majeures (SNEM), Port-au-Prince, Haiti. ' Guest Researcher, Malaria Branch, Division of Parasitic Diseases, CDC. Requests for reprints should be addressed to this author. 2 Medecin-Chef, Section de Recherche, SNEM. 3 Biologist, Section de Recherche, SNEM. 4 Medecin-Chef, Division d'Epidemiologie et de Recherche, SNEM. s Biologist, Malaria Branch, CDC. 6 Medical Officer, Malaria Branch, CDC. Pf155 is considered as a prime candidate for a vaccine against the asexual blood stages of P. falciparum, since antibodies against Pfl55 have been reported to inhibit merozoite invasion (2) and to correlate with clinical immunity (3). SUBJECTS AND METHODS Study area The studies were conducted in Laborde-1, about 24 km north-west of Les Cayes, the main city on the southern peninsula of Haiti. Situated at an elevation of 40 m, Laborde-l is composed of 145 inhabited houses clustered on both sides of the road between Les Cayes and Jeremie. According to the records of the Service National des Endemies Majeures (SNEM) of Haiti, there is here a high incidence of P. falci- parum malaria; perennial transmission occurs, with two peaks in June-August and November-January during the rainy periods. Residual household insecticide spraying was con- ducted in Laborde-l until 1983, but the principal control measure now consists of chloroquine distri- bution. The drug is prescribed mainly for presumptive treatment (single dose of 10 mg/kg body weight) of febrile cases by the primary health worker in the locality or by the local government outpatient clinic. Occasionally chloroquine has been given in mass drug administration; the last such administration in Laborde-l was in November 1984. 4787 -339- 340 P. DELORON ET AL. Population investigated Individuals were surveyed over a 7-day period in December 1985 during the reported transmission season. Information was obtained on the occupants in each house regarding their age, sex, recent history of travel, fever or other illnesses and intake of anti- malarial drugs. Axillary temperature was recorded and capillary blood specimens were obtained by fingerprick for examination of thick smears, deter- mination of erythrocyte volume fraction (haemato- crit), and serological studies. For the latter studies, the capillary blood collected in heparinized micro- haematocrit tubes was centrifuged; the tubes were then cut and the plasma fraction collected (total vary- ing from 30 to 200 1l) and stored at - 20 °C for most of the remaining time. Microscopic observations Thick blood smears were stained with Giemsa and examined for parasites; smears where no parasites were seen in 100 oil-immersion fields (corresponding to 2000 leukocytes) were classified as negative. Anti-P. falciparum enzyme-linked immunosorbent assay The technique for the enzyme-linked immuno- sorbent assay (ELISA) described by Spencer et al. (4) was followed. The antigen preparation consisting of a whole parasite extract obtained by saponin lysis and sonication of in-vitro cultured P. falciparum (West African 1 strain) was dispensed in microtitration plates' at a concentration of 1.5 Ag protein per well. The wells were sequentially incubated with 200 yd of the following: (a) plasma diluted 1:100 in phosphate buffered saline (PBS), pH 7.2, containing 0.05% Tween 20 (30 minutes at 37 °C); (b) horseradish peroxidase (HPO)-conjugated rabbit IgG anti- human immunoglobulin Gb diluted at 1:2000 (same incubation conditions); and (c) o-phenylenediamine and 3% hydrogen peroxide as substrate (30 minutes at room temperature in the dark). The absorbance at X = 492 nm was determined for each well on a multi- scan ELISAc reader; samples with an absorbance of 0.3 or greater were considered as positive. Anti-Pf55 assays Monolayers of glutaraldehyde-fixed and air-dried P. falciparum-infected erythrocytes from cultures (clone B3 of strain Honduras-i /CDC) were used as a Immulon II, Dynatech Laboratories Inc., Alexandria, VA, USA. Use of trade names is for identification only and does not imply endorsement by the Public Health Service or by the U.S. Department of Health and Human Services. b Miles Laboratories Inc., Elkhart, IN, USA. ' Titertek Multiscan from Flow Laboratories Inc., McLean, VA, USA. antigen to detect the presence of anti-Pf155 anti- bodies. Two previously described techniques were used: modified immunofluorescence assay (MIFA) and cell-ELISA (CELISA) (1). Modified indirect immunofluorescence assay. Multitest microscope slides previously coated with infected erythrocytes (1-3%7 parasitaemia) were layered sequentially for 30 min with the following: (a) plasma diluted in Tris Hanks buffer (TBH: 10 ml of 0.15 mol/l Tris, pH 7.2, 90 ml of 0.9% NaCl, 100 ml of Hanks' balanced salt solution) (1); (b) goat anti- bodies against human immunoglobulins labelled with biotind (30 ug/ml); and (c) fluorescein isothio- cyanate-conjugated avidind (50 jtg/ml). Parasite nuclei were stained by ethidium bromide (10 itg/ml). Plasmas were studied at 1: 4 dilution; samples giving a positive reaction were titrated further with 4-fold stepwise dilutions. Cell-ELISA. Wells of microtitration plates were coated with either uninfected or infected erythrocytes (7% parasitaemia), and unbound sites were then blocked by overnight incubation with 0.5% casein in PBS at room temperature. Plasmas were reacted in three wells (two wells coated with infected erythro- cytes and one well with uninfected erythrocytes). Each plasma was tested at 1: 5000 dilution in PBS containing 0.05% bovine serum albumin, conditions which had previously been determined as optimal (P. Deloron et al., unpublished data). Each well was filled sequentially with 100 1u of the following: (a) diluted plasma (two hours at room temperature); (b) HPO-conjugated goat IgGs against human immuno- globulins' diluted 1:1000 in PBS (same conditions); and (c) o-phenylenediamine and 3% hydrogen peroxide as substrate (20 minutes at room tempera- ture in the dark). The absorbance at 492 nm was determined as described. Samples were considered as positive when a difference in absorbance of at least 0.4 OD was observed between the mean value of the two wells containing infected erythrocytes and the value of the well containing uninfected erythrocytes. All positive reactions were confirmed by repeating the assay with negligible variations of the results. Samples giving discordant results between MIFA and CELISA were further studied by repeating both assays and, in selected samples, by Western immuno- blotting. Western immunoblotting Western blots were performed essentially as de- scribed by Tsang et al. (5). Plasmas were diluted 1:100 in PBS containing 0.3% Tween 20 and in- d Vector Laboratories Inc., Burlingame, CA, USA. ' Cappel Laboratories, Cochranville, PA, USA. Pfl55 ANTIGEN OF PLASMODIUM FALCIPARUM cubated overnight at 4 °C with nitrocellulose strips containing electrophoretically transferred polypep- tides from sodium dodecyl sulfate polyacrylamide gel (3.3-20/o gradient) electrophoresis (SDS-PAGE) separation of a merozoite-enriched parasite prepar- ation (1) (isolate Indochina-i/CDC). Bound anti- bodies were detected by incubation with HPO-conju- gated goat antibodies to human IgG in PBS contain- ing 0.3% Tween 20 (1 hour at room temperature), followed by staining with diaminobenzidine and hydrogen peroxide. RESULTS During the study, 136 houses were visited and 642 persons were interviewed, from whom 628 usable blood samples were obtained; 51% of the individuals were male and 49% were female. The age distribution of the population was as follows: < 1 year of age (3.1%), 1-4 years (11.7%), 5-14 years (26.6%), 15-24 years (20.4%), and >25 years (38.27o). Parasitological findings A total of 618 thick blood smears were examined. Nine subjects (1.5%) were found to be infected with malaria parasites, P. falciparum being the only species identified. Two of the 9 infected subjects had only gametocytes in their peripheral blood and were afebrile (axillary temperature <37.5 °C). The 7 others were found to have asexual parasites; 5 were afebrile (one 13-year-old girl and 4 males, aged 3, 14, 16 and 21 years) and 2 were febrile (a 7-year-old girl and a 54-year-old man). Fever was noted in 18 of the non-parasitaemic individuals. Anti-P. falciparum ELISA Six hundred and five samples were available for testing. A positive reaction was obtained in 38.21o of the plasmas. The positivity rate was 5.9%, 16%, 29.8%, 34.2% and 55.6% in the age groups < 1, 1-4, 5-14, 15-24 and > 25 years, respectively; the differ- ence between each group was significant (P< 0.001) (Fig. 1). The mean OD titre increased with age from 0.15 ± 0.04 in the youngest group to 0.38 ± 0.11 in the oldest group. Anti-PfS55 assays Six hundred and sixteen samples were available for MIFA studies. Antibodies against parasite antigens in the membrane of ring-infected erythrocytes were detected by immunofluorescence in 12.5% of the samples. The positivity rate was 5.90/o, 4%, 5.5% and 9.8%, respectively, in the four younger age groups, 601 50- e.. 40-i 0 IE U. 30 0 z 20 U t0w A. 10. no A III < 1 1-4 5-14 AGE (yw) I I 15-24 > 25 Fig. 1. Antimalarial antibodies detected by whole- parasite ELISA (LII) and anti-Pf 155 antibodies detected by MIFA (0) and by CELISA (M) in the population of Laborde-1. and 21.4% in the oldest one. A significant difference was observed between the four younger groups and the oldest one (P< 0.001) but not among the younger groups. Six hundred and nine samples were available for CELISA studies. Antibodies were detected in 13.6% of the plasmas. The positivity rate was 5.9%, 9.3%, 7.9% and 8.2%, respectively, in the four younger age groups and 22.3/o in the oldest one (P<0.001). Although the anti-Pf155 antibody positivity rates detected by MIFA and CELISA were similar within Table 1. Anti-Pf 155 antibodies in 609 subjects in Laborde-1, Haiti: correlation between the results of MIFA (reciprocal titres) and CELISA (OD/492nm) per- formed as described in the text CELISA (OD/492nm) MIFA titres <0.4 0.4-0.5 0.5-0.7 >0.7 Total Negative 491 37 5 1 534 4-64 29 7 7 2 45 256-1024 8 3 7 2 20 > 4096 0 0 3 7 10 Total 528 47 22 12 609 - b ' 341 P. DELORON ET AL. Table 2. Anti-Pf155 antibodies in 609 subjects in Laborde-1, Haiti: results of MIFA and CELISA by age group Percentage of sera MIFA + MIFA - MIFA + MIFA - Age (years) No. examined CELISA - CELISA + CELISA + CELISA - <1 17 0 0 5.9 94.1 1-4 75 2.7 8 1.3 88 5-14 163 3.1 5.5 2.5 89 15-24 121 9.1 6.6 0.8 83.5 >25 233 7.7 8.2 14.6 69.5 Total 609 6.1 7.1 6.2 80.6 each age group, discrepancies between the results obtained by the two techniques were observed and confirmed by repeated assays in 13.2% of the samples: 6.1% were MIFA+ /CELISA-; 7.1% were MIFA- /CELISA+. High MIFA reciprocal titres were associated overall with positive reactions by CELISA, and high CELISA OD values were associ- ated with a positive MIFA (Table 1). Further analysis showed no detectable difference between the age groups in terms of the correlation between MIFA and CELISA results (Table 2). Of the samples positive by both MIFA and CELISA, 92.1% were also positive by standard ELISA. Of the samples negative by both MIFA and CELISA, 30.3% were positive by standard ELISA. Fig. 2. Western blots of 37 plasma samples collected in Laborde-1. Plasma samples from subjects with anti-Pf155 antibodies detectable by both MIFA and CELISA (group A), by MIFA only (group B), or by CELISA only (group C) were reacted with nitrocellulose strips containing electrophoretically separated Plasmodium falciparum antigens. Arrow indicates the location of the Pf155 band. 342 Pfl55 ANTIGEN OF PLASMODIUM FALCIPARUM Western blot Discrepancies between the MIFA and CELISA results were investigated by performing Western blots on 37 plasmas that gave positive results by either or both assays. The plasmas reacted with several poly- peptides electroblotted from parasite preparations. A 155 x 103 Mr polypeptide was recognized by all the 11 plasmas positive by both MIFA and CELISA. In the case of 16 plasmas positive by MIFA only, this poly- peptide was recognized strongly by 8 plasmas, more weakly by 6 plasmas, and was not detected by 2 plasmas. In the case of 10 plasmas positive by CELISA only, the polypeptide was recognized strongly by 2 plasmas, weakly by 3 plasmas, and was not detected by the remaining 5 plasmas (Fig. 2). DISCUSSION The overall slide and serology positivity rates con- firm the indication by the SNEM of active malaria transmission in Laborde-1, and can be compared to those found (J. G. Breman, personal communi- cation) in rural northern Haiti in September 1983. These authors observed by indirect immunofluor- escence a seropositivity rate (40.7%) similar to that of Laborde-1 (38.2/o), but the slide positivity rate found in northern rural Haiti (6.1%) was higher than that noted in Laborde-l (1.5%). This latter finding might be due to the fact that transmission in Laborde-l had decreased just before the survey was conducted. Alternatively, antimalarial drugs might be more readily available in Laborde-l than in northern rural Haiti, with the Laborde-l inhabitants seeking treatment earlier for symptoms suggestive of malaria. Antibodies against P. falciparum antigens found in the membrane of ring-infected erythrocytes were detected in 12.5-13.6%o of the plasmas, depending on the technique used. Most of the plasmas containing these antibodies also had antibodies against whole P. falciparum antigens, as detected by ELISA. This relationship was not reciprocal, since a substantial proportion of plasmas positive by whole-parasite ELISA did not contain detectable anti-Pfl55 anti- bodies. Although the serological profiles obtained by MIFA and CELISA were similar, discrepancies between the results of these two tests for anti-Pfl55 antibodies were not exceptional. Spencer et al. (6) found a similar phenomenon for whole-parasite malaria antibodies in comparative ELISA/IFA studies. Out of 261 sera from individuals in Viet Nam, 5.4% were IFA+/ELISA- and 17.6% were IFA- /ELISA+; out of 351 sera from individuals in Honduras, 6.6% were IFA+ /ELISA- and 22.8% IFA- /ELISA+. The authors attributed these dis- crepancies to the fact that different antibodies were measured by the two serological tests, an explanation that might also apply to MIFA and CELISA. The overall results of MIFA and CELISA suggest that, in addition to Pfl55, antibodies to other malaria antigens might be detected by MIFA only or by CELISA only. The exact nature and cellular location of such antigens remain to be determined. With Western blot detection techniques as a reference standard, MIFA appears to be more specific for anti- Pfl55 antibodies than CELISA (P<0.05). Although a progressive age-related increase in antibody titres can be detected by whole-parasite ELISA, a significant increase in prevalence of anti- Pf155 antibodies is detectable only after 25 years of age. In addition, no trends were detected for a clear- cut protective value against clinical malaria of anti- Pf155 antibodies, which is different from the findings reported from a holoendemic area in Liberia (3). Of the 7 parasitaemic subjects, two individuals, one febrile and one afebrile, had antibodies detectable by both MIFA and CELISA (MIFA reciprocal titres, 1/64 and 1/4096; CELISA OD, 0.72 and 0.75, respectively), a rate similar to that found in the non- parasitaemic population. Further field investi- gations, conducted longitudinally, will be required for satisfactorily ascertaining the potential protective effect of anti-Pf155 antibodies. ACKNOWLEDGEMENTS We thank the personnel of the Service National des Endemies Majeures (SNEM) of Haiti, and particularly Marc M. Day, Gerard M. Jeanty, Ilem Aristil and Remus Brionne for their excellent technical assistance. We gratefully acknowledge the enthusiastic support of Dr Mario Alvarez, Director of the SNEM, in the conduct of this investigation. This project was supported by the U.S. Agency for International Development and the World Health Organization. 343 344 P. DELORON ET AL. RESUME ANTICORPS DIRIGES CONTRE PfI55, UN DES PRINCIPAUX ANTIGENES DE PLASMODIUM FALCIPARUM: RtSULTATS SERO-tPIDtMIOLOGIQUES D'UNE ENQUETE MENEE EN HAITI En decembre 1985, on a evalue l'ampleur de 1'endemicite palustre et l'etat immunitaire de la population d'une petite localite de Haiti, Laborde-l. Lors de visites A domicile etalees sur une semaine, 628 habitants ont e interroges et des prises de sang effectuees en vue de l'examen de frottis epais, de la determination du volume globulaire (hemato- crite) et d'etudes serologiques. Sept sujets (soit 1,5%) etaient infectes par Plasmodiumfalciparum, la seule espece de plasmodies observ&e. La methode immuno-enzymatique en phase solide (ELISA) destinee A mettre en evidence les anticorps diriges contre P. falciparum a donne lieu A une reaction positive dans 38,2% des echantillons de serum. On a observe une augmentation, en rapport avec l'Age, des taux de positivite anticorpale. En outre, on a evalue la reponse immunitaire au Pf155, un antigene de P. falciparum (Mr 155 000), depose dans la membrane d'hematies infectees par des schizontes, au moyen d'une reaction d'immunofluo- rescence modifiee (MIFA) et d'une methode immuno- enzymatique A mediation cellulaire (CELISA), toutes deux fondees sur l'utilisation d'une monocouche d'hematies infectees par P. falciparum, fixees par le glutaraldehyde et sechees A l'air. Une reaction positive a e obtenue dans 12,501o des echantillons avec 1'epreuve d'immunofluo- rescence (MIFA) et 13,60o avec la technique immuno- enzymatique (CELISA). De meme qu'avec la methode ELISA, le pourcentage des seropositifs en MIFA et en CELISA augmentait avec l'Age, mais cette augmentation survenait plus tard, apres l'age de 25 ans. Dans 13,2%o des echantillons, les resultats obtenus avec les deux epreuves MIFA et CELISA divergeaient et, pour certains echan- tillons, on a pousse les investigations en recourant A la technique d'immunotransfert. Cette derniere technique, utilisee comme epreuve de reference, a montre que les serums qui ont presente une reaction positive avec l'epreuve d'immunofluorescence (MIFA) contenaient plus souvent des anticorps anti-Pf 155 que les serums reagissant avec la technique CELISA. On n'a cependant observe aucune tendance des anticorps anti-Pf 155 A assurer une protection franche contre la maladie clinique. II sera necessaire de faire d'autres etudes longitudinales sur le terrain pour evaluer 1'eventuelle action protectrice de ces anticorps. REFERENCES 1. PERLMANN, H. ET AL. Antibodies in malarial sera to parasite antigens in the membrane of erythrocytes infected with early asexual stages of Plasmodium falci- parum. Journal of experimental medicine, 159: 1686-1704 (1984). 2. WAHLIN, B. ET AL. Human antibodies to a Mr 155 000 Plasmodium falciparum antigen efficiently inhibit merozoite invasion. Proceedings of the National Academy of Sciences USA, 81: 7912-7916 (1984). 3. WAHLGREN, M. ET AL. Anti-Plasmodium falciparum antibodies acquired by residents in a holoendemic area of Liberia during development of clinical immunity. American journal of tropical medicine and hygiene, 35: 22-29 (1986). 4. SPENCER, H. C. ET AL. The enzyme-linked immuno- sorbent assay (ELISA) for malaria: I. The use of in-vitro cultured Plasmodium falciparum as antigen. American journal oftropical medicine and hygiene, 28: 927-932 (1979). 5. TSANG, V. C. W. ET AL. Enzyme-linked immunoelectro- transfer blot techniques (EITB) for studying the speci- ficities of antigens and antibodies separated by gel electrophoresis. Methods in enzymology, 92: 377-391 (1983). 6. SPENCER, H. C. ET AL. The enzyme-linked immuno- sorbent assay (ELISA) for malaria: II. Comparison with the malaria indirect immunofluorescent antibody test (IFA). American journal of tropical medicine and hygiene, 28: 933-936 (1979).