Am J Trop Med Hyg., 73(2), 2005, pp 244–255 Copyright © 2005 by The American Society of Tropical Medicine and Hygiene COMPARISON OF IgG REACTIVITIES TO PLASMODIUM VIVAX MEROZOITE INVASION ANTIGENS IN A BRAZILIAN AMAZON POPULATION TUAN M TRAN, JOSELI OLIVEIRA-FERREIRA, ALBERTO MORENO, FATIMA SANTOS, SYED S YAZDANI, CHETAN E CHITNIS, JOHN D ALTMAN, ESMERALDA V-S MEYER, JOHN W BARNWELL, AND MARY R GALINSKI* Emory Vaccine Center & Yerkes National Primate Research Center, Emory University, Atlanta, Georgia; Department of Immunology, Institute Oswaldo Cruz, Oswaldo Cruz Foundation, Rio de Janeiro, Brazil; Fundaỗóo Nacional de Saỳde, Ministry of Health, Rondụnia, Brazil; Malaria Research Group, International Centre for Genetic Engineering and Biotechnology (ICGEB), New Delhi, India; Malaria Branch, Division of Parasitic Diseases, National Center for Infectious Diseases, Centers for Disease Control and Prevention, Atlanta, Georgia; Department of Medicine, Division of Infectious Diseases, Emory University, Atlanta, Georgia Abstract Naturally acquired antibody reactivity to two major Plasmodium vivax vaccine candidates was investigated in 294 donors from three malaria-endemic communities of Rondônia state, Brazil Antibody recognition of recombinantly expressed antigens covering five different regions of P vivax reticulocyte binding protein (PvRBP1) and region II of P vivax Duffy binding protein (PvDBP-RII) were compared Positive IgG responses to these antigens were significantly related to the level of malaria exposure in terms of past infections and years of residence in the endemic area when corrected for age The highest prevalence of anti-PvRBP1 total IgG antibodies corresponded to the amino acid regions denoted PvRBP1431-748 (41%) and PvRBP1733-1407 (47%) Approximately one-fifth of positively responding sera had titers of at least 1:1,600 Total IgG responses to PvDBP-RII were more prevalent (67%), of greater magnitude, and acquired more rapidly than those to individual PvRBP1 antigens Responses to both PvRBP1 and PvDBP-RII were biased toward the cytophilic subclasses IgG1 and IgG3 These data provide the first insights on acquired antibody responses to PvRBP1 and a comparative view with PvDBP-RII that may prove valuable for understanding protective immune responses to these two vaccine candidates as they are evaluated as components of multitarget blood-stage vaccines DARC are not susceptible to infection.10 The PvDBP erythrocyte binding domain, designated region II (PvDBP-RII), has been mapped to an ∼350 amino acid cysteine-rich region near the amino-terminus.11 Region II has significantly higher diversity than the rest of the PvDBP and appears to be under strong selective pressure.12 In addition, naturally occurring antibodies from humans exposed to malaria have been demonstrated to inhibit the in vitro binding of erythrocytes to region II, suggesting an antiparasitic role of the naturally acquired immune response.13 The functional and immunologic significance of PvDBP-RII has supported its candidacy as a vaccine Two other P vivax candidate antigens are the reticulocyte binding proteins (PvRBP1 and PvRBP2), which were identified based on their ability to preferentially adhere to reticulocyte-enriched populations of erythrocytes.14,15 Evidence suggests that the two PvRBPs form a complex at the apical pole of the merozoite and confer the reticulocyte-specificity of P vivax blood-stage infections.14,16 PvRBP1 and PvRBP2 are both Type I integral membrane proteins of approximately 330 kDa, each consisting of a large extracellular domain, a single transmembrane domain, and a short cytoplasmic tail Homologues with similar structures have subsequently been identified and characterized in P falciparum.17–19 Global diversity analysis of these genes has revealed a cluster of nonsynonymous polymorphisms within the first 750 amino acids of the PvRBP1 extracellular domain,20 suggesting that this region may be the target of immune selection Despite evidence that the PvRBPs play an important role during the P vivax invasion of reticulocytes, the immunogenicity of these proteins has yet to be studied Furthermore, evidence for naturally occurring antibodies reactive against PvRBPs has not been demonstrated in a malaria-exposed population Given the large size of PvRBP1, the identification of functional subdomains and protective B-cell epitopes would prove valuable in the development of a PvRBP1 subunit vaccine Toward these goals, we have screened plasma samples from INTRODUCTION Plasmodium vivax accounts for an estimated 80 million cases occurring throughout most of Asia, Oceania, southeastern Europe, Central and South America, and parts of Africa.1 Despite its wide geographic distribution, P vivax has long been overshadowed by the burden caused by Plasmodium falciparum, which causes more than million deaths per year in sub-Saharan Africa alone.2 The disproportionate morbidity and mortality figures are a major reason why research on P vivax has lagged far behind that of P falciparum As of now, more than 23 different P falciparum vaccines are currently undergoing clinical trials compared with only two P vivax vaccine trials.3,4 However, the phylogenetic and antigenic disparity between the two species implies that effective falciparum vaccines will not be cross-protective against vivax infections In light of rising drug resistance in P vivax and its increasing prevalence, it is imperative to advance the development of more P vivax antigens as vaccine candidates.5 Malaria blood-stage vaccines aim to disrupt interactions between receptors on Plasmodium merozoites and their erythrocytic ligands by eliciting inhibitory antibodies that target the parasitic receptors One such target unique to P vivax is the Duffy binding protein (PvDBP),6,7 a Type I integral membrane protein whose homologues in Plasmodium knowlesi and P falciparum have been localized by immunoelectron microscopy to the micronemes of merozoites.8,9 PvDBP mediates the invasion of erythrocytes via interactions with its erythrocytic ligand, the Duffy antigen receptor for chemokines (DARC).6 Invasion of human erythrocytes by P vivax requires this interaction, as individuals deficient in * Address correspondence to Mary R Galinski, Emory Vaccine Center, Yerkes National Primate Research Center, Emory University, 954 Gatewood Rd., Atlanta, Georgia 30329 E-mail: galinski@rmy emory.edu 244 245 ANTIBODY RESPONSES TO VIVAX INFECTIONS a cross section of three epidemiologically distinct communities in the Brazilian Amazon state of Rondônia for the presence of IgG antibodies against five recombinant fragments spanning the length of the PvRBP1 extracellular domain For comparative purposes, we also tested the same plasma samples for antibodies against a recombinant, refolded PvDBP-RII.21 We used malaria infection and exposure histories obtained from individual donors to assess the relationship between exposure and naturally acquired antibody responses to PvRBP1 and PvDBP-RII Previous seroepidemiological studies in hyperendemic areas of Africa have shown cytophilic antibody subclasses IgG1 and IgG3 to be associated with acquired protection to falciparum malaria.22 Furthermore, in vitro studies have demonstrated that cytophilic antibodies mediate their antiparasitic effect by engaging with Fc␥ receptors on monocytes.23,24 We therefore evaluated the IgG subclass reactivities in positive responders and determined the subclass distribution for both PvRBP1 and PvDBPRII This study provides the first insights into immune responses made against these two blood-stage vaccine candidates upon natural exposure to P vivax MATERIALS AND METHODS Study site and subjects Cohort studies were conducted in populations living in three communities in the malaria endemic region of Rondônia state, Brazil, where vivax malaria accounts for more than 70% of all malaria cases in the past years (Brazilian Ministry of Health, 2004, unpublished data) Samples and survey data were collected during the dry months of June and July of 2001, coinciding with the period of increased malaria transmission The two communities of Colina and Ribeirinha have been described previously.25–27 Briefly, Colina consists primarily of transmigrants from nonendemic areas of Brazil who have lived in the region for 10 years or more They reside in rural settlements alongside unpaved roads that traverse the transnational highway BR-364 approximately 50–100 km southeast of the capital, Porto Velho Ribeirinha encompasses riverine communities along the banks of the Madeira River and its tributaries approximately 30–60 km northeast of the capital Families in Ribeirinha have lived in the malaria-endemic region for more than 25 years, and many are native to the region The third group lives in Buritis, a newly developed municipality situated approximately 300 km due south of the capital Most individuals from this group consist of transmigrants who have resided in the endemic region for less than 10 years The study population consisted of 87 donors from Colina, 177 donors from Ribeirinha, and 30 donors from Buritis Age ranged from years to 85 years, but most donors were between 20 to 40 years of age Written informed consent was obtained from all adult donors or from parents of donors in the case of minors The study was reviewed and approved by the Fundaỗóo Oswaldo Cruz Ethical Committee Epidemiologic survey Donors giving informed consent answered questions from an epidemiologic survey Questions on the survey related to demographics, time of residence in the endemic area, personal and family histories of malaria, use of malaria prophylaxis, presence of malaria symptoms, and personal knowledge of malaria Survey data was entered into a database created with Epi Info 2002 (Centers for Disease Control and Prevention, Atlanta, GA) Demographic and epidemiologic data are summarized in Table Collection of human blood samples and malaria diagnosis Intravenous blood was collected in sodium heparin Vacutainer tubes (Becton Dickinson, Franklin Lakes, NJ) Cellular components were separated from plasma by centrifugation, and aliquots of plasma were frozen and stored at −20°C until use for antibody analysis Thin and thick blood smears of all donors were examined for malaria parasites The numbers of donors positive for P vivax and P falciparum at the time of phlebotomy are listed in Table Parasitemia for smearpositive donors was determined by counting the number of parasites against a predetermined number of white blood cells TABLE Characteristics of three communities in Rondoˆnia, Brazil Study population (N) Characteristics Buritis (30) Colina (87) Ribeirinha (177) Statistical analysis¶ P Gender, male (%) Median age in years (range) Transmigrants* (%) Median number of years residing in endemic area Males Females Median number of previous malaria episodes Males Females Months since previous malaria episode† Males Females Patent malaria infection by blood smear (%) Plasmodium vivax Asymptomatic§ Plasmodium falciparum Asymptomatic§ 12 (41.4) 24 (9–62) 27 (90%) 7.0 7.0 8.0 3.5 4.0 3.0 1.6–5.6 0.8–8.2 1.2–6.8 11 (37%) (30%) (44%) (6.7%) (50%) 60 (68.2) 35 (11–75) 64 (74%) 21.0 20.0 24.5 8.0 10.0 5.0 10.6–24.2 10.6–27.2 5.4–33.6 (8.0%) (5.7%) (100%) (2.3%) (100%) 96 (54.2) 32 (10–85) 20 (11%) 28.0 29.0 28.0 4.0 4.0 4.5 11.2–19.7 13.0–27.6 7.4–17.4 19 (10.7%) 13 (7.3%) 12 (92%) (3.4%) (67%) 2 ס8.57 KW ס3.7 2 ס135 KW ס51.3 KW ס26.0 KW ס27.6 KW ס11.9 KW ס7.1 KW ס3.6 F ס9.4‡ F ס5.4‡ F ס3.5‡ 2 ס18.0 2 ס17.5 2 ס8.8 2 ס1.30 2 ס1.27 0.014 NS < 0.0001 < 0.0001 < 0.0001 < 0.0001 0.003 0.029 NS 0.0001 0.005 0.035 0.0001 < 0.001 0.012 NS NS * Transmigrants are defined as individuals who migrated into the endemic area from a nonendemic state of Brazil † Back-transformed 95% confidence intervals of transformed data shown Only donors with a positive history for malaria were included ‡ Raw data was transformed using the function log (x + 1) to produce a normal distribution before one-way analysis of variance § Asymptomatic individuals were defined as blood smear positive donors who did not report malaria-like symptoms at the time of collection and during their next day follow-up ¶ KW, Kruskal-Wallis statistic; F, F statistic; NS, not statistically significant Statistical significance was set as P ס0.05 246 TRAN AND OTHERS in thick blood smears, and, from this, the number of parasites per l of blood was calculated.28 Infected donors who were asymptomatic at the time of phlebotomy were followed up the next day to confirm absence of malaria symptoms All smear-positive donors were subsequently treated for P vivax or P falciparum per the regimen recommended by the Brazilian Ministry of Health Protein expression and purification Five overlapping fragments spanning the PvRBP1 extracellular domain and PvDBP-RII were expressed as recombinant proteins in Escherichia coli and subsequently purified as described below The expression and purification of recombinant, refolded PvDBP-RII have been described previously.21 In addition, thioredoxin, used as a control for nonspecific IgG reactivity, was expressed and purified from Escherichia coli BL21(DE3) (Novagen, Madison, WI) containing the plasmid pET32b (Novagen) using standard procedures suggested by the manufacturer Recombinant proteins were used to detect the prevalence of antibodies in plasma samples by enzyme-linked immunosorbant assays (ELISA) Five overlapping regions of Pvrbp1 spanning nucleotides 67 to 7832 of the coding sequence were amplified from P vivax Belem strain genomic DNA using the Expand High Fidelity polymerase chain reaction (PCR) system (Roche, Indianapolis, IN) The regions (shown in Figure 1) were selected based on global diversity data20 and protein stability index as determined by ExPASy ProtParam program.29 The primers, cloning sites, and expression vectors used for each construct are listed in Table The PCR product PvRBP123-458 was cloned into pDONOR2.1 (Invitrogen, Carlsbad, CA) before subcloning into the expression vector pDEST17, yielding the recombinant plasmid PvRBP123-458/pDEST17 Escherichia coli BL21(SI) (Invitrogen) was transformed with plasmid PvRBP123-458/pDEST17 The PCR product PvRBP1431-748 was digested with EcoR I and cloned into the expression vector pET32b to yield plasmid PvRBP1431-748/pET32b The PCR products PvRBP1733-1407 and PvRBP11392-2076 were digested with Kpn I and cloned into the expression vector pET32b to yield plasmids PvRBP1 733-1407 /pET32b and PvRBP11392-2076/pET32b, respectively The PCR product PvRBP12038-2611 was cloned into the Kpn I site of expression vector pET29b (Novagen) to yield PvRBP12038-2611/pET29b plasmid Each plasmid was transformed into E coli BL21(DE3) All cloned constructs were verified for proper orientation, reading frame, and fidelity to the template PvRBP123-458 was expressed as an N-terminal hexa-histidine (6-His) fusion protein; PvRBP1431-748, PvRBP1733-1407, and PvRBP11392-2076 as N-terminal thioredoxin/6-His double fusion proteins; and PvRBP12038-2611 as a C-terminal 6-His fusion protein For PvRBP1 431-748 , PvRBP1 733-1407 , and PvRBP11392-2076, Luria-Bertania media containing 100 g/mL ampicillin (LB-amp) was inoculated and cultured at 37°C until the optical density at 600 nm (OD600) reached 0.6-0.8 Cells were induced to produce protein with mM isopropyl-1-thio-galactopyranoside (IPTG) for hours at 37°C The soluble 6-His recombinant proteins were purified with Ni-NTA agarose (Qiagen, Valencia, CA) per the manufacturer’s instructions before gel filtration chromatography Protein samples were loaded on a Sephacryl S-300 (Amersham Pharmacia Biotech, Piscataway, NJ) gel filtration column equilibrated with 20 mM Tris-HCl, pH 8.0, and eluted with 20 mM TrisHCl buffer, pH 8.0, containing 500 mM NaCl Fractions from the major peak based on absorbance280 were pooled, dialyzed overnight against phosphate-buffered saline, pH 7.4 (PBS; Cellgro, Herndon, VA), concentrated by ultrafiltration, and stored at −80°C The insoluble recombinant proteins PvRBP123-458 and PvRBP12038-2611 were isolated as inclusion bodies, reconstituted by step dialysis, and purified by gel filtration chromatography Escherichia coli BL21(SI) containing plasmid PvRBP123-458/pDEST17 was cultured in salt-free LB-amp at 37°C until an OD600 of 0.8–1.0, and protein expression was FIGURE Expression of the extracellular domain of Plasmodium vivax reticulocyte binding protein (PvRBP1) as overlapping recombinant fragments in Escherichia coli The five antigens span residues 23 through 2611 of the Belem strain nascent polypeptide sequence For details on the expression and purification of the proteins, see “Materials and Methods.” N סamino terminus, C סcarboxyl terminus, trx סthioredoxin 247 ANTIBODY RESPONSES TO VIVAX INFECTIONS TABLE Primers used to amplify PvRBP1 polymerase chain reaction products for cloning into expression vectors* Construct PvRBP123–458 PvRBP1431–748 PvRBP1733–1407 PvRBP11392–2076 PvRBP12038–2611 Forward primer 5ЈGGGGACAAGTTTGTACAAAA AAGCAGGCTCCGAAAATG CAGAAGAAGACATAAG 3Ј 5ЈTCCGAATTCGAACGAACC AATTAAGAAAGCATAT 3Ј 5ЈGGGGGTACCGAAATGGAA GAGAAGCTACAGGAT 3Ј 5ЈGAGGGTACCGGGGAATTT ACAAAGGCTGAAGGT 3Ј 5ЈCACGGTACCCTAGACAAT ACAGAGGAGCTTGAAAA 3Ј Reverse primer 5ЈGGGGACCACTTTGTACAAGA AAGCTGGGTCCTATTTTCC TAACGTGTCTATAATGG 3Ј 5ЈCCACGAATTCTTCTATAATT TAGCAAAAGTTTCTTTGGC 3Ј 5ЈAGGGGTACCTTAGGCACTT GCGTTTTTCTCTTC 3Ј 5ЈGGGGGTACCCTAATAGGTG TTCATTTCGTTCTCCATTTC 3Ј 5ЈGAGGGTACCCTGTACTAG TTGGCCTATTTCTCG 3Ј Cloning site Expression vector att recombination pDEST17 EcoRI pET32b Kpn I pET32b Kpn I pET32b Kpn I pET29b * PvRBP1, Plasmodium vivax reticulocyte binding protein Cloning sites on primer sequences indicated by italics and underscoring induced with 0.3 M NaCl for hours For PvRBP12038-2611, LB containing 50 g/mL kanamycin (LB-kan) was inoculated and grown at 37°C to an OD600 of 0.8–1.0, and protein expression was induced with mM IPTG for hours Bacteria pellets were harvested and resuspended in 50 mM Tris-HCl, pH 8.0 containing 25% (w/v) sucrose, mM ethylenediaminetetraacetic acid (EDTA), and 10 mM dithiothreitol (DTT) before freezing at −80°C The resuspension was thawed, and mg/mL lysozyme, mM MgCl2, mg/mL DNase, 1% Triton-X 100, and 10 mM DTT were added Cells were lysed by sonication and pelleted by centrifugation before resuspension in 50 mM Tris-HCl, pH 8.0, containing 0.5% Triton X-100, 100 mM NaCl, mM EDTA, 0.1% azide, and mM DTT Sonication, centrifugation, and resuspension steps were repeated four more times, with Triton X-100 excluded in the last wash The washed pellets were dissolved in 25 mM 2-[Nmorpholino] ethane sulfonic acid, pH 6.0, containing M urea, 10 mM EDTA, 0.1 mM DTT, and these inclusion bodies were stored at −80°C PvRBP123-458 and PvRBP12038-2611 inclusion bodies were reconstituted by 1:2 dilution with 50 mM glycine, pH 8.0 containing 10% w/v sucrose, mM EDTA, mM reduced gluthathione, 0.1 mM oxidized glutathione, and M urea and dialyzed for 24 hours at 4°C against a 50× volume of the same buffer The dialysis buffer was exchanged with a 50x volume of 20 mM Tris-HCl, pH 8.0, containing 10% w/v sucrose, mM EDTA, 0.1 mM reduced glutathione, 0.01 mM oxidized glutathione and dialyzed for 24 hours at 4°C Samples were concentrated to an appropriate volume and further purified by gel filtration chromatography as described above Purity and stability of all recombinant proteins were verified by SDS-PAGE (Figure 2) and Western blot analysis using an anti-penta-His monoclonal antibody (mAb; Qiagen; supplemental Figure 1) Measurement of antibody levels Plasma samples from study participants were screened for total IgG reactivity against the five PvRBP1 constructs, PvDBP-RII, and purified thioredoxin using ELISA for total IgG antibodies Maxisorp 96-well plates (Nunc, Rochester, NY) were coated with 200 ng of antigen overnight at 4°C in PBS, pH 7.4 Plates were washed four times with PBS-0.05% Tween 20 (PBS-T) and blocked with blocking buffer (PBS-T containing 5% skim milk) for hours at 37°C Plates were washed once and incubated for hour at 22°C with plasma diluted 1:100 in blocking buffer After washing four times, the plates were incubated for hour at 22°C with peroxidase-conjugated goat anti-human IgG (␥ chain specific) (KPL, Gaithersburg, MD) diluted 1:3000 in blocking buffer Plates were washed four times and incubated for 30 minutes at 22°C with 2, 2’-azinodi(3-ethylbenzthiazoline-6-sulfonate) substrate (ABTS, KPL) per the manufacturer’s instructions The ODs were measured at 405 nm, and the threshold for positivity was an OD value set at SD above the mean OD of 30 residents of either the United States or Brazil who have never been exposed to malaria transmission For the three thioredoxin fusion proteins, background antibody reactivity to thioredoxin was subtracted from raw ODs before further analysis Average cutoffs were 0.607 (PvRBP1 - ), 0.723 (PvRBP1 - ), 0.289 (PvRBP1733-1407), 0.272 (PvRBP11392-2076), 0.616 (PvRBP12038-2611), and 0.388 (PvRII) All samples were tested in duplicate All plates included the same positive reference plasma and PvRBP1431-748 to standardize development time For positive responders, end-point titers were determined using 1:2 serial dilutions starting at 1:100 The end points were determined as the highest titers in which the OD was greater than the threshold for positivity at a control FIGURE Polyacrylamide gel of recombinant Plamodium vivax reticulocyte binding protein (PvRBP1) and P vivax Duffy binding protein region II (PvDBP-RII) proteins Approximately g of each protein were electrophoresed on a 10% polyacrylamide gel and stained with Coomassie R-250 to visualize the bands The minor bands below PvRBP1733-1407, PvRBP11392-2076, PvRBP12038-2611 represent degradation products See “Materials and Methods” for details on expression and purification of proteins Molecular weights in kilodaltons (kDa) are indicated Refer to supplemental Figure for Western blot analysis of these recombinant proteins 248 TRAN AND OTHERS FIGURE Seroprevalence of IgG antibodies against Plasmodium vivax reticulocyte binding protein (PvRBP1) recombinant antigens and P vivax Duffy binding protein region II (PvDBP-RII) for Buritis, Colina, and Ribeirinha communities as determined by enxyme-linked immunosorbent assay Total IgG seroprevalence refers to the frequency of IgG positive responders to an antigen in a population A positive response was defined as a response having an optical density405 that was greater than three standard deviations above the mean of nonexposed controls The solid black bars represent seroprevalence for all three communities combined 2 analyses were performed to determine if a statistically significant difference existed between the seroprevalence of the three populations for each antigen The level of significance is indicated by *(P < 0.05), **(P < 0.01), and ***(P < 0.001) plasma dilution of 1:100 An ELISA to detect the IgG subclasses of the anti-DBP and anti-RBP antibodies was also performed for positive responders Plates were coated with antigen, blocked, and incubated with plasma diluted 1:100 as in the ELISA for total IgG After washing, plates were incubated for hour at 22°C with mouse mAbs to human IgG subclasses diluted in blocking buffer according to the manufacturer’s specifications The mAbs were from clones HP6001 for IgG1, HP-6002 for IgG2, HP-6050 for IgG3, and HP-6023 for IgG4 (Sigma, St Louis, MO) and have been used previously to characterize IgG subclass reactivity in the Ribeirinha population After incubation, plates were washed and incubated for hour at 22°C with peroxidase-labeled goat anti-mouse antibody (KPL) diluted 1:1,000 in blocking buffer Plates were washed, incubated with ABTS, and the OD measured as described above Subclass-specific prevalence for each antigen was determined from OD values using SD above the appropriate mean OD of 24 nonexposed controls as the cutoff for positivity (Table 3) In addition, standard curves TABLE Optical density cutoffs for IgG subclass-specific positivity Subclass Antigen IgG1 IgG2 IgG2 IgG4 PvRBP123–458 PvRBP1431–748 PvRBP1733–1407 PvRBP11392–2076 PvRBP12038–2611 PvDBP-RII 0.114 0.119 0.061 0.077 0.122 0.180 0.255 0.542 0.123 0.227 0.560 0.525 0.093 0.214 0.057 0.043 0.089 0.110 0.117 0.151 0.089 0.054 0.113 0.156 PvRBP1, Plasmodium vivax reticulocyte binding protein 1; PvDBP-RII, P vivax Duffy binding protein region II Cutoff values represent the mean + SD of subclass-specific optical densities for 24 nonexposed controls were prepared to enable conversion of OD values to concentration (g/mL) values, thus allowing for the comparison of different subclasses Purified human IgG kappa myeloma proteins from each of the four subclasses (The Binding Site, San Diego, CA) were coated overnight at 4°C in PBS at 100 L per well onto 96-well plates in 1:2 serial dilutions from 24 g/mL to 2−11 g/mL After washing four times, plates were incubated with the appropriate human IgG subclass-specific mAb, washed four times, incubated with peroxidase-labeled goat anti-mouse antibody, washed a final four times, developed with ABTS, and measured as described above for subclass-specific ELISA Subclass-specific OD values were converted to concentration values (g/mL) using sigmoidal curve-fit equations derived from subclass-specific standard curves (supplemental Figure 2) Statistical analysis Differences in medians for the study population data were tested by non-parametric KruskalWallis test where appropriate Normalizing transformations were performed on raw data before testing by one-way analysis of variance where appropriate Differences in proportions were evaluated by 2 test Relationships between years of residence in the endemic area and number of past malaria infections or months since last known malaria episode were assessed with Spearman’s rank correlation Multivariate logistic regression was used to assess the relationship between antigen-specific total IgG responses and the independent variables of gender, age, years of residence in the endemic area, number of past malaria episodes, and months since last known malaria infection For comparison of IgG subclassspecific antibody responses, 2 test was used to evaluate differences in prevalence Normalizing transformations were performed on subclass concentrations before one-way analysis of variance (for evaluating differences in subclass distri- ANTIBODY RESPONSES TO VIVAX INFECTIONS 249 FIGURE Seroprevalence of IgG antibodies against Plasmodium vivax reticulocyte binding protein (PvRBP1) and P vivax Duffy binding protein region II (PvDBP-RII) in 294 residents of Rondônia, Brazil Donors were grouped by (A) years of residence in the endemic area and (B) past number of malaria episodes Servoprevalence is the proportion of donors responding to the different antigens among the total for each group Responses to the five PvRBP1 antigens are shown as filled shapes Anti-PvDBP-RII responses are shown as open squares bution within responders or within controls for each antigen) or Student’s t test (for evaluating subclass-specific differences between responders and controls for each antigen) Spearman’s rank correlation coefficients were calculated to determine the relationships between IgG subclass levels and months since last known malaria episode Analyses were done using Epi Info 2002 and Prism 3.0 (GraphPad Software, San Diego, CA) RESULTS Characteristics of Buritis, Colina, and Ribeirinha Table shows the characteristics of the three communities studied Among our sample set, the communities differ significantly in gender ratio, with Colina donors being disproportionately male Median age is similar between the three communities The proportion of transmigrants is highest in Buritis (90%) and lowest in Ribeirinha (11%) The communities differ significantly in years of residence in the endemic area, number of past malaria episodes, months since last malaria episode, and proportion of patent malaria infections as listed Data was stratified by gender to account for male bias among Colina residents Relationship between years of residence in the endemic area and malaria history For this study, we used years of residence in the endemic area and number of past malaria episodes as reported by donors as indices of malaria exposure Absence of recent symptomatic malaria infection, measured as the number of months since a donor’s last known malaria episode, was used as a crude approximation of protection The number of past malaria episodes did not correlate significantly with years of residence in the endemic area (Spearman r ס0.086, P ס0.14, N ס294) Among donors with a positive history of malaria, years of residence in the endemic area correlated positively with the number of months since a donor’s last malaria episode (Spearman r ס 0.29, P < 0.0001, N ס258) Anti-PvRBP1 and anti-PvDBP-RII seroprevalence in three communities of Rondônia Naturally acquired antibody responses to PvDBP-RII have been shown to have high 250 TRAN AND OTHERS prevalence in regions of vivax endemicity.13,30–32 Thus, antiPvDBP-RII IgG reactivity was used in this study to estimate the rate of exposure to P vivax Anti-PvDBP-RII seroprevalence was 67% among all 294 donors and 43%, 70%, and 70% for Buritis, Colina, and Ribeirinha communities, respectively In comparison, seroprevalence for any of the five PvRBP1 fragments was 66% among all donors and 57%, 58%, and 72% for Buritis, Colina, and Ribeirinha, respectively Of these five PvRBP1 fragments, the highest frequencies of positive responses among all donors were for PvRBP1431-748 (41%) and PvRBP1733-1407 (47%) Total IgG reactivity to PvRBP123-458 and PvRBP1431-748 differed significantly among the three communities (2 ס10.08, P ס0.0065 and 2 ס 17.20, P ס0.0002, respectively), with increased prevalence to both in Colina and Ribeirinha The results are summarized in Figure Although there was a trend for anti-PvRBP1 IgG antibodies to be more prevalent in males than females, this difference was not statistically significant Magnitude of IgG responses To compare the relative magnitude of antibody responses, we determined IgG end-point titers for positive responders to each antigen The geometric means of IgG titers for each antigen were 196.0 for PvRBP1 23-458 , 371.1 for PvRBP1 431-748 , 455.8 for PvRBP1 733-1407 , 394.4 for PvRBP1 1392-2076 , 243.1 for PvRBP12038-2611, and 488.8 for PvDBP-RII Positive responders to each antigen were subdivided by their total IgG antibody titers into low responders (titers < 1,600) and high responders (titers Ն 1,600) High responders were frequently observed for PvDBP-RII (27% of all PvDBP-RII positive responders), PvRBP11392-2076 (24%), PvRBP1733-1407 (22%), and PvRBP1 - (20%), while PvRBP1 - and PvRBP12038-2611 had a lower proportion of high responders (4.4% and 8.7%, respectively) The maximum total IgG titer achieved for each antigen was 6,400 for PvRBP123-458 (N ס1), 12,800 for PvRBP1431-748 (N ס1), 25,600 for PvRBP1733-1407 (N ס1), 12,800 for PvRBP11392-2076 (N ס2), 3,200 for PvRBP12038-2611 (N ס2), and 51,200 for PvDBPRII (N ס2) Maximum titers were obtained in donors from either Ribeirinha or Colina but never from Buritis Multivariate logistic regression analysis of total IgG antibody reactivity according to gender, age, and indices of malaria exposure The contributions of five independent variables related to malaria exposure to total IgG antibody responses against each of the six vivax recombinant proteins were determined using stepwise, multivariate logistic regression The variables initially tested were gender, age, years of residence in the malaria endemic area, number of past malaria episodes, and months since last known malaria episode The initial analysis showed that gender did not affect IgG antibody responses to any of the recombinant antigens and was subsequently removed from the final regression analysis (Table 4) The analysis demonstrated that IgG responses to the recombinant antigens were generally age independent but exposure dependent, except for responses against PvRBP123-458, which appeared independent of both age and exposure Individuals living in the malaria-endemic area for more than 30 years were 3.3 to 4.6 times more likely to react to PvRBP1431-748, PvRBP2038-20611, and PvDBP-RII than the most recently migrated individuals Donors who reported having greater than 15 malaria infections were 4.4 to times more likely to respond to PvRBP1431-748, PvRBP1733-1407, PvRBP11392-2076, PvRBP12038-2611, and PvDBP-RII than self- reported malaria-naïve individuals Donors who have had a recent malaria infection (within two months preceding sample collection) were significantly more likely to respond to PvRBP12038-2611 and PvDBP-RII than those who have been malaria-free within this same period Seroprevalence and malaria exposure All 294 donors from Rondônia were stratified according to years of residence in the endemic area (< 11, 11–20, 21–30, and > 30 years) and number of past malaria episodes (0, 1–5, 6–15, and > 15 malaria infections) The prevalence of total IgG antibodies against each antigen was plotted for each group (Figure 4) Seroprevalence generally increased with years of residence in the endemic area and number of past malaria episodes for all antigens This trend of rising seroprevalence was similar among the five PvRBP1 antigens However, the prevalence of anti-PvRBP1 antibody responses was shown to increase at a slower rate than that of anti-PvDBP-RII antibody responses Many donors who reported never having malaria responded to PvDBP-RII (11 of 31), PvRBP1 (16 of 31), or either antigen (21 of 31) IgG subclass distribution of anti-PvDBP-RII and antiPvRBP1 antibodies We assessed the overall subclass distribution of the IgG antibody responses to each antigen using two different comparative analyses Firstly, we determined subclass-specific prevalence of total IgG positive responders for each antigen using background-subtracted OD values and OD cutoffs determined from ODs of controls Secondly, antigen-specific IgG1, IgG2, IgG3, and IgG4 concentrations for total IgG positive responders were determined from OD values using subclass-specific standard curves and the results compared with the concentrations of controls using parametric statistics Both analyses are summarized in Table With the exception of PvRBP1392-2076 responses, the results of both analyses were comparable Using prevalence as the basis of comparison, the predominant responses to PvRBP123-458, PvRBP12038-2611, and PvDBP-RII were IgG3, while IgG1 predominated in responses to PvRBP1431-748, PvRBp1733-1407, and PvRBP11392-2076 When concentrations were compared, positive responses to PvRBP1431-748 and PvRBP1733-1407 showed increases in IgG1 levels relative to nonexposed controls, while positive responses to PvRBP1 - and PvRBP12038-2611 showed statistically significant increases for all IgG subclasses Positive responses to PvDBP-RII showed increased IgG1, IgG3, and IgG4 levels relative to controls In contrast, responses to PvRBP11392-2076 showed increased IgG2 and IgG4 but decreased IgG3 relative to controls IgG subclass levels and months since last known malaria episode Time (in months) since a donor’s last known malaria episode was used as a crude approximation of the donor’s level of protection from malaria To assess if IgG subclass levels correlated with this approximation, we calculated Spearman’s correlation coefficients between IgG subclass concentrations and months since last known malaria episode for each antigen Only donors who were identified as positive responders to a particular antigen were included in the analyses for that antigen Donors who either had detectable levels of parasitemia or reported themselves as malaria-naïve were excluded from analyses Months since last known malaria episode positively correlated with IgG3 concentration for PvRBP1733-1407 (r ס0.20, P < 0.05, N ס110) and with IgG2 concentration for PvRBP1733-1407 (r ס0.22, P < 0.02, N ס 110) and PvRBP11392-2076 (r ס0.34, P < 0.002, N ס82) N 0.418 0.346 0.743 0.974 0.036 0.010 0.732 0.371 0.037 0.072 0.587 0.189 Reference 0.66 (0.24–1.80) 0.60 (0.20–1.75) 0.83 (0.28–2.46) Reference 1.02 (0.33–3.16) 3.49 (1.09–11.21) 4.64 (1.45–14.86) Reference 0.84 (0.32–2.22) 1.61 (0.56–4.62) 3.83 (1.09–13.48) Reference 0.50 (0.23–1.06) 0.81 (0.38–1.73) 0.59 (0.27–1.30) 0.659 0.927 0.141 0.227 0.074 0.083 0.308 0.311 0.910 P 0.947 0.923 0.081 P PvRBP1431–748 Odds ratio (95% CI) PvRBP1733–1407 Reference 0.74 (0.36–1.51) 0.79 (0.38–1.67) 0.57 (0.26–1.34) Reference 1.06 (0.41–2.78) 1.85 (0.65–5.25) 4.40 (1.23–15.67) Reference 0.70 (0.26–1.87) 1.31 (0.48–3.61) 1.61 (0.59–4.36) Reference 0.76 (0.31–1.90) 1.23 (0.47–3.23) 1.64 (0.61–4.46) Odds ratio (95% CI) 0.402 0.536 0.158 0.901 0.245 0.022 0.477 0.602 0.350 0.561 0.680 0.330 P PvRBP11392–2076 Reference 0.69 (0.33–1.47) 0.79 (0.37–1.71) 0.74 (0.33–1.65) Reference 1.74 (0.59–5.10) 2.06 (0.65–6.55) 4.40 (1.19–16.25) Reference 0.45 (0.16–1.29) 1.82 (0.63–5.26) 1.12 (0.39–3.20) Reference 0.57 (0.20–1.63) 0.84 (0.28–2.48) 1.29 (0.43–3.87) Odds ratio (95% CI) 0.339 0.555 0.460 0.311 0.219 0.026 0.137 0.268 0.831 0.299 0.749 0.651 P PvRBP12038–2611 Reference 0.37 (0.17–0.81) 0.36 (0.16–0.80) 0.36 (0.16–0.83) Reference 3.20 (0.91–11.28) 4.11 (1.10–15.46) 5.30 (1.21–23.24) Reference 1.00 (0.30–3.32) 2.24 (0.67–7.44) 3.36 (1.03–10.95} Reference 1.05 (0.36–3.10) 0.92 (0.29–2.92) 3.79 (1.21–11.89) Odds ratio (95% CI) 0.013 0.013 0.017 0.070 0.036 0.027 0.994 0.189 0.044 0.930 0.884 0.023 P PvDBP-RII Reference 0.31 (0.13–0.73) 0.30 (0.12–0.73) 0.28 (0.11–0.69) Reference 2.27 (0.88–5.91) 5.37 (1.83–15.79) 6.98 (1.79–27.17) Reference 1.21 (0.44–3.33) 2.25 (0.79–6.43) 3.30 (1.17–9.35) Reference 0.69 (0.27–1.80) 0.49 (0.18–1.34) 0.88 (0.30–2.57) Odds ratio (95% CI) 0.007 0.008 0.006 0.092 0.002 0.005 0.708 0.129 0.025 0.451 0.165 0.821 P * PvRBP1, Plasmodium vivax reticulocyte binding protein 1; PvDBP-RII, P vivax Duffy binding protein region II Donors were grouped as indicated according to the independent variables of age (in years), years of residence in endemic area, number of past malaria episodes, and months since last malaria episode Multivariate logistic regression was performed comparing the likelihood of positive responses in each group to the reference group for each antigen Likelihood is measured as odds ratios and 95% confidence intervals (CI) Statistical significance was set at P ס0.05 Age (in years) 9–19 65 Reference 20–35 98 1.05 (0.29–3.83) 36–50 66 0.94 (0.24–3.66) 51–85 65 3.25 (0.87–12.20) Years of residence in endemic area 0–10 29 Reference 11–20 85 0.73 (0.18–2.92) 21–30 74 0.94 (0.24–3.65) > 30 106 2.65 (0.72–9.73) Number of past malaria episodes 31 Reference 1–5 143 2.20 (0.61–7.90) 6–15 94 3.47 (0.89–13.6) > 15 26 3.83 (0.84–17.5) Months since last malaria episode 0–1 64 Reference 2–18 71 0.63 (0.26–1.54) 19–60 64 0.63 (0.25–1.55) > 60 or never infected 89 0.95 (0.39–2.33) Independent variables PvRBP123–458 Odds ratio (95% CI) Dependent variable (antigen response) TABLE Multivariate logistic regression of PvRBP1 and PvDBP-RII IgG antibody responses* ANTIBODY RESPONSES TO VIVAX INFECTIONS 251 252 TRAN AND OTHERS TABLE Prevalence (% positive) and 95% confidence intervals (g/mL) of IgG subclass-specific responses to PvRBP1 and PvDBP-RII in responders positive for total IgG* Antibody subclass† Antigen PvRBP123–458 Prevalence Responders Controls P† PvRBP1431–748 Prevalence Responders Controls P† PvRBP1733–1407 Prevalence Responders Controls P† PvRBP11392–2076 Prevalence Responders Controls P† PvRBP12038–2611 Prevalence Responders Controls P† PvDBP-RII Prevalence Responders Controls P† N IgG1 IgG2 IgG3 IgG4 P‡ 68 23 54.4% 5.556–6.485 4.769–5.110 < 0.0001 39.7% 1.781–2.196 1.301–1.542 < 0.0001 64.7% 7.893–9.091 6.317–6.631 < 0.0001 44.1% 0.456–0.546 0.354–0.411 < 0.0001 0.02 < 0.0001 < 0.0001 120 24 42.5% 4.677–8.375 1.968–3.221 < 0.0001 12.5% 1.194–1.702 0.885–1.567 NS 20.0% 4.529–6.124 3.828–5.623 NS 18.3% 0.244–0.356 0.193–0.297 NS < 0.0001 < 0.0001 < 0.0001 138 24 34.1% 2.301–3.540 1.268–2.009 0.0004 27.5% 0.804–1.104 0.548–0.940 NS 28.3% 2.904–3.963 2.218–3.443 NS 15.2% 0.109–0.159 0.062–0.154 NS 0.004 < 0.0001 < 0.0001 99 24 23.2% 1.936–2.944 1.816–2.642 NS 4.0% 1.954–2.812 0.615–1.096 < 0.0001 18.2% 1.245–1.888 1.977–2.999 0.002 12.1% 1.306–1.811 0.094–0.156 < 0.0001 < 0.001 < 0.001 < 0.0001 103 24 73.8% 7.722–9.074 4.960–5.330 < 0.0001 36.9% 2.108–2.609 1.527–1.965 0.0007 83.5% 8.850–9.814 6.333–6.596 < 0.0001 66.0% 0.603–0.707 0.390–0.425 < 0.0001 < 0.0001 < 0.0001 < 0.0001 187 24 55.3% 6.680–7.770 5.447–6.184 < 0.0001 31.5% 1.559–1.785 1.504–1.960 NS 72.1% 7.564–8.432 6.297–6.716 < 0.0001 55.8% 0.952–1.122 0.381–0.4394 < 0.0001 < 0.0001 < 0.0001 < 0.0001 * PvRBP1, Plasmodium vivax reticulocyte binding protein 1; PvDBP-RII, P vivax Duffy binding protein region II Prevalence refers to the percentage of responders determined as positive for a particular subclass based on optical densities (OD) from subclass-specific enzyme-linked immunosorbent assays Responders are donors previously determined as having a positive total immunoglobulin G (IgG) response to the indicated antigen Controls are non-malaria-exposed donors OD values were converted to g/mL using standard curves for each IgG subclass (see “Materials and Methods” for details) Log or reciprocal transformations were performed as appropriate to normalize the distribution of data points before parametric statistical analyses Back-transformed 95% confidence intervals of means in g/ml are shown † Student’s t test was used to compare differences between responders and controls within each IgG subclass for a single antigen Bold typeface indicates the statistically greater of the two groups ‡ For each antigen, differences in proportions of subclass-specific positive responders (subclass-specific seroprevalence) were compared by 2 test One-way analysis of variance was used to compare differences between IgG subclass levels within responders or within controls for each antigen Statistical significance was set at P ס0.05 NS, not statistically significant DISCUSSION Naturally acquired antibody responses against PvRBP1, one of the proteins implicated in conferring reticulocyte specificity of P vivax invasion of red blood cells, have not been studied to date This study attempts to characterize the acquired antibody response to PvRBP1 and PvDBP in malaria-endemic populations of Rondônia state in the Brazilian Amazon Malaria transmission in this area is low but continuous throughout the year, with sharp increases in malaria incidence during the dry months of June, July, and August.33,34 Distinct rural communities of Rondônia represent the various temporal cohorts of transmigrants who have settled in the region during the past 40 years as a result of government social programs.35 The seroepidemiology of malaria infection in these communities has been studied extensively for P falciparum; however, studies on antibody responses to specific P vivax antigens have been limited to small fragments of the long-standing vaccine candidates merozoite surface protein (PvMSP1) and the circumsporozoite protein (PvCSP).25–27,36–42 We investigated three communities that have comparable median age but are clearly epidemiologically distinct, differing in time of residence in the endemic area, number of past malaria episodes, and number of patent vivax malaria infections (Table 1) Male bias in the Colina community may be a confounding variable, especially in regard to analyses with number of past malaria episodes The effect of gender bias must be considered when interpreting comparative analyses between the communities since men in these communities typically engage in outdoor activities (farming, fishing, logging, etc.) during peak transmission hours and thus have greater overall exposure than women The communities listed in order of increasing time of residence in the malaria endemic area are 1) Buritis, 2) Colina, and 3) Ribeirinha Curiously, residents of Ribeirinha reported having the least number of past malaria episodes, despite being natives of the endemic area This finding is probably a limitation of donorreported data Most likely, natives underestimate past malaria histories because they not recall childhood infections Comparable times since last malaria episode between Colina and Ribeirinha (Table 1) not only argues against alternative explanations such as lower malaria transmission or enhanced acquired clinical immunity in the Ribeirinha community, but even suggests that the transmigrant Colina population is acquiring a level of clinical immunity similar to that of the native Ribeirinha population This is not surprising given that the median number of years of residence in the endemic re- ANTIBODY RESPONSES TO VIVAX INFECTIONS gion for the Colina population approaches that of Ribeirinha The community of recent transmigrants (Buritis) has the lowest seroprevalence of antibodies against all of the antigens tested when compared with the established communities of Colina and Ribeirinha Although no significant relationships were found between infection status and seropositivity for any of the antigens (data not shown), the higher frequency of patent P vivax infections, relative absence of asymptomatic infections, and lowered P vivax antibody responses observed in the Buritis residents may be indicative of their nonimmune status It has long been known that acquired clinical immunity to falciparum malaria depends on repeated exposure to the parasite, a conclusion based on the observation that effective natural immunity to P falciparum is restricted to areas of high level transmission.43 The presence of asymptomatic cases in Amazonian communities with long-term malaria exposure reported here and elsewhere suggests that a parallel phenomenon occurs in areas of low P vivax transmission.44,45 This is further supported by our finding that the time that has passed since a donor’s last malaria episode positively correlates with the time the donor has resided in the endemic area The biological basis for such clinical immunity has yet to be elucidated, but it is likely that the effect is partially mediated by antibodies against P vivax blood stage antigens We demonstrate that IgG antibody responses to PvDBP and PvRBP1 are dependent on indices of malaria exposure such as time of residence in the endemic area and number of past malaria episodes Our results coincide with other studies investigating the antibody responses to malaria antigens in areas of low transmission, supporting the view that natural immunologic boosting occurs even with less frequent transmission patterns.25,30,46 Studies on nonimmune transmigrants to hyperendemic areas of Indonesia suggest that immunity to falciparum malaria is age-dependent and that adults acquire protective immunity to chronic infection more rapidly than children.47 We were not able to investigate whether the clinical immunity we observed was age-dependent as our study was limited primarily to adults However, incidental crossreactivity of malaria-naïve plasma with Plasmodium antigens, a major component of the age-dependent hypothesis, is suggested in our study by 1) high OD values among a few nonexposed North American and Brazilian controls for several of the antigens tested, resulting in unexpectedly high OD cutoffs and 2) the high frequency of seropositivity (67%) among exposed donors who reported themselves as malaria-naïve An alternative explanation for seropositivity in this latter group may be unrecalled or subclinical P vivax infections, as mentioned previously Studies on IgG responses against P falciparum antigens have consistently shown that cytophilic subclasses IgG1 and IgG3 play an important role in a protective antibody response in humans.22,48–51 The proposed mechanism involves parasitic inhibition effected by engagement of Plasmodium-specific cytophilic Ig antibodies to Fc␥ receptors on the surface of monocytes.23,24,52 To date, however, only a small number of studies have investigated cytophilic antibody responses against P vivax.53,54 In this study, we have demonstrated that positive IgG responses to both PvRBP1 and PvDBP-RII show increased levels of the cytophilic subclasses relative to nonexposed controls Specifically, positive responses to PvRBP1431-748 and PvRBP733-1407 were predominantly IgG1, 253 while positive responses to PvRBP123-458, PvRBP12038-2611, and PvDBP-RII demonstrated increases in IgG3 Curiously, positive responses to PvRBP11392-2076 demonstrated discordant subclass distributions between our two analyses, with one analysis suggesting a predominance of non-cytophilic antibodies Although we demonstrate a general predominance of cytophilic IgG antibodies, the correlation to clinical immunity is rather tenuous The cross-sectional design of this study limited our investigation to retrospective malaria histories, and the best approximation of an individual’s protection was the estimated amount of time that had passed since their last malaria episode, a measurement that may be confounded by low or absent exposure to the parasite We were only able to observe modest, positive correlations between subclass reactivities against the PvRBP1733-1407 and PvRBP11392-2076 antigens and the amount of time since the last infection, only one of which involved a cytophilic subclass Prospective studies on humoral immune responses or biologic studies addressing the ability of P vivax antibodies to inhibit merozoite invasion or development will provide more direct evidence with regards to their protective efficacy This study attempts to determine the target of antibody responses to PvRBP1 by screening for IgG antibodies to recombinant fragments spanning the extracellular region of the protein The prevalence of IgG responses to individual PvRBP1 antigens and PvDBP-RII approached 50% and 70%, respectively, comparable to levels previously observed for the PvCSP repeat region and the N-terminal region of PvMSP1 36,39 Our results highlight PvRBP1 431-748 and PvRBP1733-1407 as targets of a natural antibody response, as these regions demonstrate both broad and high-titer responses Although responses to PvRBP1733-1407 were more prevalent than those to PvRBP1431-748, the latter antigen spans only 318 amino acids of the 2749 residues comprising the PvRBP1 extracellular domain (compared with 675 amino acids for PvRBP1733-1407) Furthermore, of the 25 total polymorphic residues in PvRBP1, 12 (48%) reside in the region spanned by PvRBP1431-748.20 These data taken together suggest that PvRBP1431-748 may be under immunologic selection pressure Finer mapping studies are needed to further define the B cell epitopes contained within this region The leading strategy for malaria vaccine design calls for a multiple target approach against several blood-stage antigens.55 Direct comparisons of the natural antibody response to these antigens provide valuable insight as to how such a vaccine might work We have compared the antibody responses for two promising P vivax blood-stage targets, PvRBP1 and PvDBP In addition to the differences in subclass distribution, we show that their IgG responses differ in two important characteristics First, anti-PvDBP-RII IgG responses appear more prevalent and of higher magnitude than anti-PvRBP1 responses Second, the rate of IgG antibody acquisition differs between the two antigens Individuals seem to require less exposure to build up detectable levels of antibodies to PvDBP-RII compared with PvRBP1, implying immunodominance of PvDBP-RII over PvRBP1; the lone exception is the response against PvRBP12038-2611, which shows a rate of increase similar to that of the anti-PvDBP-RII response, albeit at lower prevalence and magnitude We quantified the rapid acquisition of anti-PvDBP-RII antibodies by logistic regression, showing that donors with more than five 254 TRAN AND OTHERS malaria episodes were more than five times as likely to respond to this antigen compared with malaria-naïve individuals (Table 4) However, the regression also suggests that the anti-PvDBP-RII response may be short-lived Donors who have not had a recent malaria infection (within two months preceding sample collection) were significantly less likely to respond to PvDBP-RII than their recently infected counterparts In contrast, with the exception of PvRBP12038-2611, the prevalence of responses to PvRBP1 antigens was independent of time since the last known malaria episode The evidence implies that although anti-PvRBP1 IgG responses may be slowly acquired relative to anti-PvDBP-RII responses, the longevity of such responses may be greater in the absence of clinical malaria episodes In conclusion, we have characterized the IgG antibody responses to five distinct regions spanning the extracellular domain of PvRBP1 We show that both PvRBP1 with PvDBPRII elicit predominantly cytophilic IgG responses but these responses differ in magnitude, breadth, and rate of acquisition, with PvDBP-RII eliciting the more robust and rapid antibody response in three communities of the Brazilian Amazon These data provide information on the characteristics of acquired humoral immunity to two key vivax antigens in populations exposed to malaria transmission that could be beneficial in the development of a multitarget, blood-stage vaccine for P vivax Prospective studies are needed to determine whether synergistic, additive, or antagonistic interactions exist between IgG antibody responses to PvDBP-RII and PvRBP1 as well as their roles in clinical immunity Received October 7, 2004 Accepted for publication March 18, 2005 Note: Supplemental figures and appear online at www.ajtmh.org Acknowledgments: The authors thank the Fundaỗóo Nacional de Saude (Brazilian Ministry of Health) and the Secretary of Health of Rondônia for providing fieldwork support and K.B Anderson for help with statistical analyses We are grateful to all donors who participated in this study Financial support: This research is supported by the National Institute of Allergy and Infectious Diseases, National Institutes of Health, grant no R01AI247-18 Tuan M Tran was also a recipient of an American Society of Tropical Medicine and Hygiene Benjamin H Kean Fellowship in Tropical Medicine Chetan E Chitnis is a Wellcome Trust International Senior Research Fellow and Howard Hughes International Research Scholar Authors’ addresses: Tuan M Tran, Alberto Moreno, John D Altman, Esmeralda V-S Meyer, and Mary R Galinski, Emory Vaccine Center, Yerkes National Primate Research Center, Emory University, 954 Gatewood Rd., Atlanta GA 30329 Joseli Oliveira-Ferreira, Laboratorio de Pesquisas em Malaria, Fundaỗóo Oswaldo Cruz FIOCRUZ, Av Brasil, 4365 – Manguinhos, Rio de Janeiro, RJ, Brazil Fatima Santos, Fundaỗóo Nacional de Saỳde, Av George Teixeira S/N, Porto Velho, Rondônia, Brazil Syed S Yazdani and Chetan E Chitnis, Malaria Research Group, International Centre for Genetic Engineering and Biotechnology (ICGEB), Aruna Asaf Ali Marg, New Delhi 110067, India John W Barnwell, Malaria Branch, Division of Parasitic Diseases, National Center for Infectious Diseases, Centers for Disease Control and Prevention, MS F-36, Atlanta, GA 30341 Reprint requests: Mary R Galinski, Emory Vaccine Center at Yerkes, Emory University, 954 Gatewood Rd., Atlanta, GA 30329, Telephone: 404-727-7214, Fax: 404-727-8199, E-mail: galinski@rmy emory.edu REFERENCES Mendis K, Sina BJ, Marchesini P, Carter R, 2001 The neglected burden of Plasmodium vivax malaria Am J Trop Med Hyg 64: 97–106 Breman JG, 2001 The ears of the hippopotamus: manifestations, determinants, and estimates of the malaria burden Am J Trop Med Hyg 64: 1–11 Moorthy VS, Good MF, Hill AV, 2004 Malaria vaccine developments Lancet 363: 150–156 Ballou WR, Arevalo-Herrera M, Carucci D, Richie TL, Corradin G, Diggs C, Druilhe P, Giersing BK, Saul A, Heppner DG, Kester KE, Lanar DE, Lyon J, Hill AV, Pan W, Cohen JD, 2004 Update on the clinical development of candidate malaria vaccines Am J Trop Med Hyg 71: 239–247 Sina B, 2002 Focus on Plasmodium vivax Trends Parasitol 18: 287–289 Wertheimer SP, Barnwell JW, 1989 Plasmodium vivax interaction with the human Duffy blood group glycoprotein: identification of a parasite receptor-like protein Exp Parasitol 69: 340–350 Fang XD, Kaslow DC, Adams JH, Miller LH, 1991 Cloning of the Plasmodium vivax Duffy receptor Mol Biochem Parasitol 44: 125–132 Adams JH, Hudson DE, Torii M, Ward GE, Wellems TE, Aikawa M, Miller LH, 1990 The Duffy receptor family of Plasmodium knowlesi is located within the micronemes of invasive malaria merozoites Cell 63: 141–153 Sim BK, Toyoshima T, Haynes JD, Aikawa M, 1992 Localization of the 175-kilodalton erythrocyte binding antigen in micronemes of Plasmodium falciparum merozoites Mol Biochem Parasitol 51: 157–159 10 Miller LH, Mason SJ, Clyde DF, McGinniss MH, 1976 The resistance factor to Plasmodium vivax in blacks The Duffyblood-group genotype, FyFy N Engl J Med 295: 302–304 11 Chitnis CE, Miller LH, 1994 Identification of the erythrocyte binding domains of Plasmodium vivax and Plasmodium knowlesi proteins involved in erythrocyte invasion J Exp Med 180: 497–506 12 Cole-Tobian J, King CL, 2003 Diversity and natural selection in Plasmodium vivax Duffy binding protein gene Mol Biochem Parasitol 127: 121–132 13 Michon P, Fraser T, Adams JH, 2000 Naturally acquired and vaccine-elicited antibodies block erythrocyte cytoadherence of the Plasmodium vivax Duffy binding protein Infect Immun 68: 3164–3171 14 Galinski MR, Medina CC, Ingravallo P, Barnwell JW, 1992 A reticulocyte-binding protein complex of Plasmodium vivax merozoites Cell 69: 1213–1226 15 Galinski MR, Xu M, Barnwell JW, 2000 Plasmodium vivax reticulocyte binding protein-2 (PvRBP-2) shares structural features with PvRBP-1 and the Plasmodium yoelii 235 kDa rhoptry protein family Mol Biochem Parasitol 108: 257–262 16 Galinski MR, Barnwell JW, 1996 Plasmodium vivax: merozoites, invasion of reticulocytes and considerations for vaccine development Parasitol Today 12: 20–29 17 Rayner JC, Galinski MR, Ingravallo P, Barnwell JW, 2000 Two Plasmodium falciparum genes express merozoite proteins that are related to Plasmodium vivax and Plasmodium yoelii adhesive proteins involved in host cell selection and invasion Proc Natl Acad Sci U S A 97: 9648–9653 18 Rayner JC, Vargas-Serrato E, Huber CS, Galinski MR, Barnwell JW, 2001 A Plasmodium falciparum homologue of Plasmodium vivax reticulocyte binding protein (PvRBP1) defines a trypsin-resistant erythrocyte invasion pathway J Exp Med 194: 1571–1581 19 Triglia T, Thompson J, Caruana SR, Delorenzi M, Speed T, Cowman AF, 2001 Identification of proteins from Plasmodium falciparum that are homologous to reticulocyte binding proteins in Plasmodium vivax Infect Immun 69: 1084–1092 20 Rayner JC, Tran TM, Corredor V, Huber CS, Barnwell JW, Galinski MR, 2005 Dramatic difference in diversity between Plasmodium falciparum and Plasmodium vivax reticulocyte binding-like genes Am J Trop Med Hyg 72: 666–674 21 Singh S, Pandey K, Chattopadhayay R, Yazdani SS, Lynn A, ANTIBODY RESPONSES TO VIVAX INFECTIONS 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 Bharadwaj A, Ranjan A, Chitnis C, 2001 Biochemical, biophysical, and functional characterization of bacterially expressed and refolded receptor binding domain of Plasmodium vivax duffy-binding protein J Biol Chem 276: 17111–17116 Bouharoun-Tayoun H, Druilhe P, 1992 Plasmodium falciparum malaria: evidence for an isotype imbalance which may be responsible for delayed acquisition of protective immunity Infect Immun 60: 1473–1481 Bouharoun-Tayoun H, Attanath P, Sabchareon A, Chongsuphajaisiddhi T, Druilhe P, 1990 Antibodies that protect humans against Plasmodium falciparum blood stages not on their own inhibit parasite growth and invasion in vitro, but act in cooperation with monocytes J Exp Med 172: 1633–1641 Bouharoun-Tayoun H, Oeuvray C, Lunel F, Druilhe P, 1995 Mechanisms underlying the monocyte-mediated antibodydependent killing of Plasmodium falciparum asexual blood stages J Exp Med 182: 409–418 Banic DM, de Oliveira-Ferreira J, Pratt-Riccio LR, Conseil V, Goncalves D, Fialho RR, Gras-Masse H, Daniel-Ribeiro CT, Camus D, 1998 Immune response and lack of immune response to Plasmodium falciparum P126 antigen and its aminoterminal repeat in malaria-infected humans Am J Trop Med Hyg 58: 768–774 Jacobson KC, Thurman J, Schmidt CM, Rickel E, Oliviera de Ferreira J, Ferreira-da-Cruz MF, Daniel-Ribeiro CT, Howard RF, 1998 A study of antibody and T cell recognition of rhoptry-associated protein-1 (RAP-1) and RAP-2 recombinant proteins and peptides of Plasmodium falciparum in migrants and residents of the state of Rondonia, Brazil Am J Trop Med Hyg 59: 208–216 Banic DM, Goldberg AC, Pratt-Riccio LR, De Oliveira-Ferreira J, Santos F, Gras-Masse H, Camus D, Kalil J, Daniel-Ribeiro CT, 2002 Human leukocyte antigen class II control of the immune response to p126-derived amino terminal peptide from Plasmodium falciparum Am J Trop Med Hyg 66: 509– 515 Shute GT, 1988 The microscopic diagnosis of malaria Wernsdorfer WH, McGregor SI, eds Malaria: Principles and Practice of Malariology New York: Churchill Livingstone, 781–814 Guruprasad K, Reddy BV, Pandit MW, 1990 Correlation between stability of a protein and its dipeptide composition: a novel approach for predicting in vivo stability of a protein from its primary sequence Protein Eng 4: 155–161 Michon PA, Arevalo-Herrera M, Fraser T, Herrera S, Adams JH, 1998 Serologic responses to recombinant Plasmodium vivax Duffy binding protein in a Colombian village Am J Trop Med Hyg 59: 597–599 Fraser T, Michon P, Barnwell JW, Noe AR, Al-Yaman F, Kaslow DC, Adams JH, 1997 Expression and serologic activity of a soluble recombinant Plasmodium vivax Duffy binding protein Infect Immun 65: 2772–2777 Xainli J, Cole-Tobian JL, Baisor M, Kastens W, Bockarie M, Yazdani SS, Chitnis CE, Adams JH, King CL, 2003 Epitopespecific humoral immunity to Plasmodium vivax Duffy binding protein Infect Immun 71: 2508–2515 Camargo LM, Noronha E, Salcedo JM, Dutra AP, Krieger H, Pereira da Silva LH, Camargo EP, 1999 The epidemiology of malaria in Rondonia (Western Amazon region, Brazil): study of a riverine population Acta Trop 72: 1–11 Salcedo JM, Camargo EP, Krieger H, Silva LH, Camargo LM, 2000 Malaria control in an agro-industrial settlement of Rondonia (Western Amazon region, Brazil) Mem Inst Oswaldo Cruz 95: 139–145 Singer BH, de Castro MC, 2001 Agricultural colonization and malaria on the Amazon frontier Ann N Y Acad Sci 954: 184–222 Oliveira-Ferreira J, Nakaie CR, Daniel-Ribeiro C, 1992 Low frequency of anti- Plasmodium falciparum circumsporozoite repeat antibodies and rate of high malaria transmission in endemic areas of Rondonia State in northwestern Brazil Am J Trop Med Hyg 46: 720–726 Mertens F, Levitus G, Camargo LM, Ferreira MU, Dutra AP, Del Portillo HA, 1993 Longitudinal study of naturally acquired humoral immune responses against the merozoite surface protein of Plasmodium vivax in patients from Rondonia, Brazil Am J Trop Med Hyg 49: 383–392 255 38 Ferreira MU, Kimura ES, Camargo LM, Alexandre CO, da Silva LH, Katzin AM, 1994 Antibody response against Plasmodium falciparum exoantigens and somatic antigens: a longitudinal survey in a rural community in Rondonia, western Brazilian Amazon Acta Trop 57: 35–46 39 Levitus G, Mertens F, Speranca MA, Camargo LM, Ferreira MU, del Portillo HA, 1994 Characterization of naturally acquired human IgG responses against the N-terminal region of the merozoite surface protein of Plasmodium vivax Am J Trop Med Hyg 51: 68–76 40 Balthazar-Guedes HC, Ferreira-Da-Cruz MF, MontenegroJames S, Daniel-Ribeiro CT, 1995 Malaria diagnosis: identification of an anti-40-kDa polypeptide antibody response associated with active or recent infection and study of the IgG/IgM ratio of antibodies to blood-stage Plasmodium falciparum antigens Parasitol Res 81: 305–309 41 Ferreira-da-Cruz MF, Deslandes DC, Oliveira-Ferreira J, Montenegro-James S, Tartar A, Druilhe P, Daniel-Ribeiro CT, 1995 Antibody responses to Plasmodium falciparum sporozoite-, liver- and blood-stage synthetic peptides in migrant and autochthonous populations in malaria endemic areas Parasite 2: 23–29 42 Oliveira-Ferreira J, Pratt-Riccio LR, Arruda M, Santos F, Daniel Ribeiro CT, Carla Golberg A, Banic DM, 2004 HLA class II and antibody responses to circumsporozoite protein repeats of P vivax (VK210, VK247 and P vivax-like) in individuals naturally exposed to malaria Acta Trop 92: 63–69 43 Riley E, 1994 Malaria Kierszenbaum F, ed Parasitic Infections and the Immune System San Diego: Academic Press, xii, 254 44 Camargo EP, Alves F, Pereira da Silva LH, 1999 Symptomless Plasmodium vivax infections in native Amazonians Lancet 353: 1415–1416 45 Alves FP, Durlacher RR, Menezes MJ, Krieger H, Silva LH, Camargo EP, 2002 High prevalence of asymptomatic Plasmodium vivax and Plasmodium falciparum infections in native Amazonian populations Am J Trop Med Hyg 66: 641–648 46 Braga EM, Barros RM, Reis TA, Fontes CJ, Morais CG, Martins MS, Krettli AU, 2002 Association of the IgG response to Plasmodium falciparum merozoite protein (C-terminal 19 kD) with clinical immunity to malaria in the Brazilian Amazon region Am J Trop Med Hyg 66: 461–466 47 Baird JK, 1998 Age-dependent characteristics of protection v susceptibility to Plasmodium falciparum Ann Trop Med Parasitol 92: 367–390 48 Druilhe P, Khusmith S, 1987 Epidemiological correlation between levels of antibodies promoting merozoite phagocytosis of Plasmodium falciparum and malaria-immune status Infect Immun 55: 888–891 49 Oeuvray C, Bouharoun-Tayoun H, Gras-Masse H, Bottius E, Kaidoh T, Aikawa M, Filgueira MC, Tartar A, Druilhe P, 1994 Merozoite surface protein-3: a malaria protein inducing antibodies that promote Plasmodium falciparum killing by cooperation with blood monocytes Blood 84: 1594–1602 50 Nguer CM, Diallo TO, Diouf A, Tall A, Dieye A, Perraut R, Garraud O, 1997 Plasmodium falciparum- and merozoite surface protein 1-specific antibody isotype balance in immune Senegalese adults Infect Immun 65: 4873–4876 51 Taylor RR, Allen SJ, Greenwood BM, Riley EM, 1998 IgG3 antibodies to Plasmodium falciparum merozoite surface protein (MSP2): increasing prevalence with age and association with clinical immunity to malaria Am J Trop Med Hyg 58: 406–413 52 Druilhe P, Bouharoun-Tayoun H, 2002 Antibody-dependent cellular inhibition assay Methods Mol Med 72: 529–534 53 Kirchgatter K, del Portillo HA, 1998 Molecular analysis of Plasmodium vivax relapses using the MSP1 molecule as a genetic marker J Infect Dis 177: 511–515 54 Soares IS, da Cunha MG, Silva MN, Souza JM, Del Portillo HA, Rodrigues MM, 1999 Longevity of naturally acquired antibody responses to the N- and C-terminal regions of Plasmodium vivax merozoite surface protein Am J Trop Med Hyg 60: 357–363 55 Good MF, 2001 Towards a blood-stage vaccine for malaria: are we following all the leads? Nat Rev Immunol 1: 117–125 1 SUPPLEMENTAL FIGURE Anti-penta-His Western blot of recombinant Plasmodium vivax reticulocyte binding protein (PvRBP1) and P vivax Duffy binding protein region II (PvDBP-RII) proteins Approximately g of each protein were run on a 10% polyacrylamide gel and transferred to a nitrocellulose membrane Western blot was performed using an anti-penta-His monoclonal antibody (Qiagen, Valencia, CA) according to the manufacturer’s specifications See “Materials and Methods” for details on expression and purification of proteins Molecular weights in kilodaltons (kDa) are indicated SUPPLEMENTAL FIGURE Immunoglobulin G subclass calibration curves Ninety-six well plates were coated with serial dilutions of purified human IgG1, IgG2, IgG3, and IgG4 myeloma proteins (The Binding Site, San Diego, CA) Enzyme-linked immunosorbent assay was performed with the corresponding anti-IgG subcalss monoclonal antibody (Sigma, St Louis, MO) as described in “Materials and Methods.” Points represent experimental data points Lines correspond to sigmoidal curves representing the best-fit equation for each subclass