| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
 |
ABSTRACT |
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
INTRODUCTION |
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The P. vivax Duffy binding protein (DBP) represents a valuable malaria subunit vaccine candidate for asexual blood stages. The protein is a conserved antigen expressed on the merozoite surface and it has been demonstrated to be a ligand for parasite invasion of the erythrocyte through a specific interaction with the Duffy blood group antigen (Fy), recently renamed the Duffy antigen receptor for chemokines (DARC) because of its role in chemokine binding.3 DARC is an indispensable receptor for P. vivax invasion of human erythrocyte, and individuals lacking DARC (i.e., Fy-negative individuals) are naturally protected from P. vivax–induced clinical malaria.4 The immune blockage of this receptor-ligand interaction through the induction of antibodies to PvDBP is therefore likely to protect humans against P. vivax blood infection.
PvDBP belongs to a family of erythrocyte binding proteins that includes the P. falciparum sialic acid binding protein, also known as 175 kD erythrocyte binding antigen, P. knowlesi DBP (PkDBP), and P. knowlesi ß and 
proteins. The receptor-binding domains of these proteins lie in conserved, N terminal, cysteine-rich regions (region II) that are found in all of these proteins. Region II of PvDBP (PvRII) contains approximately 350 amino acids with 12 conserved cysteine residues.3,5
Because of its potential use as a subunit malaria vaccine, methods to produce recombinant PvRII in its functional conformation by expression in Escherichia coli followed by in vitro refolding and purification have been recently established.6 Recombinant PvRII is recognized by antibodies of malaria-exposed individuals,7,8 and immunization studies have indicated that the protein is highly immunogenic in rabbits, and mice, and elicits high-titer binding inhibitory antibodies.6,9
Following natural exposure to P. vivax infection, individuals residing in malaria-endemic areas develop antibodies that block binding of PvDBP to DARC-positive erythrocytes.8 Studies in Papua New Guinea also show that cellular and humoral immune responses to PvRII develop progressively with age in residents from P. vivax–endemic areas, suggesting they are implicated in acquired immunity to P. vivax.8,10,11 Importantly, antibodies raised against the homologous binding domain of PkDBP (PkDBP-RII) block the invasion of human Duffy-positive erythrocytes by P. knowlesi in vitro, providing further support for developing a recombinant vaccine based on PvRII.12
Rabbits immunized with PvRII formulated in Freund’s adjuvant, as well as mice immunized with PvRII formulated with the human compatible adjuvants Montanide ISA720 and ASO2A, develop high-titer binding inhibitory antibodies that block the binding of Duffy positive erythrocytes to COS-7 cells expressing PvRII on their surface.6,9 However, the protective efficacy of immunization with PvRII against parasite challenge has not yet been demonstrated. In the present study, we assess the immunogenicity and protective efficacy of PvRII formulated in Freund’s adjuvant and the human adjuvant Montanide ISA 720 against blood-stage challenge with P. vivax.
|
MATERIALS AND METHODS |
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Aotus monkeys and immunization. Aotus lemurinus griseimembra monkeys from the Primate Center in Cali, Colombia were used for this study. Adult, naive male and female animals were selected for weight (> 800 grams), age (> 2 years old), and presence of normal liver and kidney functions before immunization. The animal experimental protocol was reviewed and approved by the Animal Ethical Committe of the Universidad del Valle. The study was conducted in compliance with National Institutes of Health guidelines for animal use.
A total of 24 monkeys (6 per group) were arbitrarily assigned to two experimental groups (A and C) and two controls groups (B and D). Group A was immunized with 50 µg of recombinant PvRII (rPvRII) formulated in completed Freund’s adjuvant and incomplete Freund’s adjuvant in a 50: 50 ratio. Group C was immunized with rPvRII formulated in Montanide ISA 720 in a 7:3 antigen:adjuvant ratio, as recommended by the adjuvant manufacturer (Seppic, Inc., Paris, France). Groups B and D were injected with distilled water containing no protein and emulsified with Freund’s adjuvant and Montanide ISA-720, respectively. Each animal was immunized by subcutaneous injection at five different sites on the back on days 0, 30, and 60.
Hematologic parameters (hemoglobin level, hematocrit, liver and kidney functions, blood urea nitrogen, aspartate aminotransferase, and alanine aminotransferase) were evaluated before and during the immunization process to check the safety of the vaccine candidate. After challenge, if animals developed a hematocrit below 25%, they were treated with a combination of sulfadoxine-pyrimethamine.13
Antibody responses. Plasma fractions were separated by centrifugation from whole blood collected from immunized and control monkeys before each vaccination and before challenge. Plasma fractions were kept at –70°C until used for detection of specific antibodies to PvRII using an enzyme-linked immunosorbent assay (ELISA) antibodies to parasites using an immunofluorescence assay (IFA).
Enzyme-linked immunosorbent assay. Plates (Maxisorp; Nunc, Roskilde, Denmark) were coated for two hours at 37°C with 100 µL/well of rPvRII at a concentration of 2 µg/mL in coating buffer (15 mM Na2CO3, 35 mM NaHCO3, pH 9.6). The plates were then blocked with 2.5% bovine serum albumin (BSA) in phosphate-buffered saline (PBS), pH 7.4, overnight at 4°C, washed with PBS 1x pH 7.2 tween 20 at 0.05% and incubated for one hour at 37°C in duplicate with 10-fold or two-fold serial dilutions of plasma from immunized Aotus monkeys containing 1% BSA. The plates were washed with PBS 1x pH 7.2 tween 20 at 0.05% incubated for one hour with peroxidase-conjugated rabbit anti-Aotus IgG (produced at the Malaria Vaccine and Drug Development Center, Cali, Colombia and conjugated at Absea Biotechnology Ltd., Beijing, People’s Republic of China) diluted 1:1,000, and incubated with 100 µL of the substrate peroxidase solution. Absorbance was read at 630 nm (OD630) in a multi-channel spectrophotometer (MRX; Dynex Technologies, Inc., Chantilly, VA). To define the cut-off value, we tested the antigen using sera from 10 Aotus monkeys with no history of malarial infection or vaccination. The ELISA results were considered positive when absorbance of the test sera was greater than or equal to the mean value of Aotus control plasma OD630 plus 3 standard deviations.
Enzyme-linked immunosorbent assay–based binding assay. To study the ability of Aotus antibodies to block binding of PvRII an ELISA-based binding assay was carried out as described elsewhere.14 Briefly, rPvRII at a concentration of 0.1 µg/mL was pre-incubated for one hour at room temperature with five-fold serial dilutions of individual Aotus serum samples from days 60 and 90 after immunization in duplicate and allowed to bind to ELISA microplates coated with recombinant Duffy antigen at a concentration of 1 µg/mL. Fifty percent inhibition titers were determined from curves obtained by plotting percentage binding at different dilutions of sera compared with binding with no sera.
Immunofluorescence assay. The IFA was performed using air-dried smears of asexual blood stages of the P. vivax Salvador I strain derived from an infected Aotus monkey. Parasites were purified using a modification of a method described elsewhere.15 Briefly, whole infected blood was collected and leukocytes were removed by passage through a loosely packed CF 11 cellulose powder column previously washed with RPMI 1640 medium. The effluent was centrifuged at 500 x g for 10 minutes at room temperature and resuspended at a 50% hematocrit. We examined thick and thin Giemsa-stained blood films to reconfirm parasite stage. The parasite fraction was enriched using a 47% Percoll gradient and the top cell layer was removed, washed three times with RPMI medium without serum and cultured for 40 hours using the candle jar culture. The IFA slides were prepared with mature schizonts and were stored at –70°C until use. After de-hemoglobinization and fixation in acetone for five minutes, slides were incubated with test plasma diluted 1:50 in PBS for one hour and washed with PBS. Fluorescein-conjugated affinity-purified rabbit anti-Aotus IgG (produced at Malaria Vaccine and Drug Development Center and conjugated at Absea Bio-technology, Ltd.) was added, and incubated for one hour at room temperature. After additional washings with PBS 1x pH 7.2, the slides were overlaid with 30% glycerol and examined with an ultraviolet epifluorescence microscope.
Parasite challenge and parasitemia follow-up. One month after last immunization (on day 90), all monkeys were intravenously challenged with 105 P. vivax Salvador I blood-stage parasites. Plasmodium vivax Salvador I was kindly provided by Dr. William Collins (Centers for Disease Control and Prevention, Atlanta, GA) and maintained as a frozen stabilate in liquid nitrogen until use. An Aotus monkey was infected with the thawed parasites and used as a parasite donor for the challenge studies. We initially assessed parasitemia by thick blood smear until it was detectable. The parasitemia was then quantified by scoring the frequency of infected erythrocytes in thin blood smears stained with Giemsa. Parasitemia was checked three times a week for 39 days (i.e., up to day 129) when all animals were treated with 25 mg of sulfadoxine-pyrimethamine.
Statistical analysis. We evaluated three efficacy outcomes: prepatent period, maximum parasitemia, and total parasitemia estimated as area under the curve (AUC). The AUC was calculated using the trapezoid rule. Samples that tested positive for parasitemia by thick smear that were not quantifiable by thin smear were treated as 0% in the AUC calculation. Differences between experimental and control groups (A versus B, C versus D) were tested using a two-sided Wilcoxon rank sum test (Stata version 8.0; Stata Corporation, College Station, TX).
|
RESULTS |
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
). Substantial increases in the specific anti-rPvRII end point ELISA titers (2–5.6 x 106) were observed in all animals immunized (day 90) with rPvRII formulated in Freund’s adjuvant (group A). Animals immunized with PvRII formulated in Freund’s adjuvant developed approximately four-fold higher endpoint titers than animals immunized with PvRII formulated in Montanide ISA 720 (0.6–20 x 104, group C). No reactivity against PvRII was found in the corresponding control groups (B and D). Sera from three monkeys from group A (079*032, 301*630, and 276*614) showed reactivity to the parasite, as shown by the IFA at a serum dilution of 1:50.
|
). High-binding inhibition titers in the Freund’s group (group A) are associated with a longer pre-patent period (Table 3). The 50% binding inhibition titers for the Montanide ISA 720 group (group C) ranged from < 1:10 to 1:99 on day 60 with moderate increases on day 90. However, these values were much lower than those observed for the Freund’s adjuvant group (group A) on day 60 and were insufficient to provide any protection. Due to small amounts of sera available, day 90 sera from group A could not be tested in binding inhibition assays.
|
|
and Figure 1, all animals from both the experimental and control groups became infected after parasite challenge. Overall pre-patency was longer and parasitemia levels were lower for vaccinated monkeys in group A compared with control group B. Five of the six animals in group B had a pre-patent period of seven days by thick blood smear, whereas in experimental group A, all animals were negative (Table 2). The median number of days to patency was 7 days for group B and 16 days for group A (Table 4). The median peak parasitemia was 0.25% for group B and 0.1% for group A. The median AUC for group B was nearly four times greater than the median AUC for group A (Table 4). However, only the differences in pre-patency distributions between groups A and B reached statistical significance (P < 0.05).
|
|
). All animals from the corresponding control group (group D) had detectable parasitemias on day 7. The median number of days to patency was eight days for group C and seven days for group D (Table 3). The median peak parasitemia was 0.1% for group C and 0.2% for group D. The median AUC for group D was nearly twice as high as the median AUC for group C (Table 4). In contrast to groups A and B, which were immunized with Freund’s adjuvant, the differences between experimental and control groups C and D immunized with Montanide ISA 720 was not as large. No statistically significant differences were observed between groups C and D. Other clinical outcomes. Animals immunized with Freund’s adjuvant developed granulomas, but the severity of the toxic reactions was smaller than that observed in studies using other malaria proteins, as well as in a pilot study using the same protein.
A slight decrease in the levels of hemoglobin was observed in three (monkey 355*782, 9.7 g/dL; 301*630, 11.1 g/dL; and 079*032, 12.4 g/dL) of the six animals immunized with the rPvRII formulated in Freund’s adjuvant at day 90 in comparison with only one monkey that showed a similar decrease in the control group (077*552, 9 0.6 g/dL). However, these changes remained in the normal range (13.1–23.4 g/dL) and were not clinically significant. In contrast, animals immunized with the rPvRII formulated in Montanide ISA-720 had normal hemoglobin levels during the immunization period. We observed normal leukocyte values in the groups immunized with Montanide, but the number of leukocytes increased in some of the animals immunized with Freund’s adjuvant.
All animals in this trial were subjected to liver and kidney function tests to determine the safety of the vaccine, particularly when formulated in Montanide ISA-720. No clinically significant differences in hematologic, liver, or kidney parameters were observed during the immunization period in either experimental or control group.
|
DISCUSSION |
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
We found that the Freund’s formulation of the protein was more immunogenic than the formulation with Montanide ISA-720. More importantly, vaccination with rPvRII formulated in Freund’s adjuvant partially protects Aotus monkeys from experimental parasite challenge. All parameters examined were strongly suggestive of efficacy. However, except for the longer pre-patency period in the rPvRII-Freund’s group, the differences between vaccinated and control groups were not statistically significant due to the small sample sizes.
Immunization of Aotus with rPvRII formulated in Freund’s adjuvant not only substantially delayed the appearance of patent parasitemia at a statistically significant level (P = 0.004), but also decreased the magnitude of parasitemia in most animals. These results are consistent with those of an earlier pilot study in which we observed complete protection in three of five Aotus monkeys immunized with 100 µg of rPvRII formulated in Freund’s adjuvant. In both experiments, immunization with this diluted formulation of Freund’s adjuvant induced granuloma, a toxic effect that will not be acceptable in humans. Here, we have observed cutaneous lesions such as inflammation at the sites of inoculation, and granuloma formation without further complications. Bacteriologic examination failed to demonstrate the presence of acid-fast microorganisms and other opportunistic microorganisms.
With the aim of assessing the protective efficacy of a vaccine formulation composed of rPvRII and an adjuvant suitable for human use, we selected Montanide ISA720. In previous studies, this adjuvant has been used with rPvRII to immunize mice9 and with other proteins to immunize monkeys.17 Similar to the results found in the mice study,9 we report that Aotus sera raised against rPvRII are able to block the binding of rPvRII to recombinant DARC in vitro. In the group immunized with the Montanide ISA720 formulation, the binding inhibition titers were related to the end point titers, with both titers increasing from day 60 to day 90. In the group immunized with the PvRII-Freund’s formulation, Monkey 301*630 showed the highest 50% binding inhibition titer (1:42,600), which was associated with a delay in patency (day 21). However, the antibody titers achieved could not block the development of parasitemia in monkeys directly challenged by inoculation with live P. vivax parasites.
High antibody titers are necessary for effective blocking of parasite invasion and growth in vivo. Monkey 378*513, which developed the earliest patency in the Freund’s group (group A), had a low binding inhibition titer on day 60. Only half of the sera from group A (Freund’s group) could recognize native PvDBP in the parasite by IFA, whereas none of the sera from group C (Montanide group) could recognize native PvDBP. Inability to detect PvDBP in merozoites may be due to technical difficulties in the IFAs given the small amounts of PvDBP present. Since immunoglobulin subtyping reagents are not available for Aotus monkeys, it was not possible to study any possible correlation of immunoglobulin subtypes with protection.
This study suggests that to achieve protection, antibodies to DBP, particularly the fraction able to cross-react with the native protein, must be at sufficiently high levels to efficiently block erythrocyte invasion by P. vivax merozoites. The partial protection achieved in the Freund’s group suggests that a potent adjuvant will be required for use in humans. PvDBP is expressed in small quantities for a limited time starting late in schizogeny and is localized intracellularly in micronemes.18 It is likely to be on the apical surface only briefly once the invasion process is initiated. Antibodies directed against PvDBP would need to be in abundant amounts and have high binding rates and high binding affinities to efficiently block the binding of P. vivax merozoites to DARC.
In addition, although a monkey model is usually not used to test safety of malaria vaccine candidates, in this study we found that certain parameters such as hematologic, liver, and kidney functions could be evaluated and thus provided useful information on the effect of the protein during immunization. In this regard, the use of Montanide ISA-720 eliminated concerns related to the potential toxicity of vaccination with a protein that binds the chemokine receptor DARC on erythrocytes. Animals from group C immunized with rPvRII formulated in Montanide ISA-720 and those from group that D that only received the adjuvant did not show any detectable toxic effect.
A recent clinical study using ASO-2A adjuvant (Glaxo-SmithKline, Research Triangle Park, NC) and a recombinant hybrid protein derived from the P. falciparum circumsporozoite protein reported high levels of antibodies that protected human volunteers from natural malaria infection.19 Moreover, other adjuvants such as Montanide ISA-51, which are suitable for human use, can induce strong antibody responses likely to efficiently block parasite reinvasion.20 We conclude that the rPvRII vaccine candidate is immunogenic, and partially protective in Aotus monkeys, and that it has promising potential for further use in clinical trials. However, to achieve higher levels of protective antibodies for this promising protein, new and more powerful adjuvants suitable for human need to be identified.
Received April 12, 2005. Accepted for publication August 10, 2005.
Acknowledgments: We thank Claudia Salas (Fundación Centro de Primates) for technical support, Mario Chen (Family Health International) for assistance in statistical analyses, and Elisabeth Robinson for critically reading the manuscript. Chetan E. Chitnis is an International Research Scholar of the Howard Hughes Medical Institute and International Senior Research Fellow of The Wellcome Trust, United Kingdom.
Financial support: This work was supported by the National Institute of Allergy and Infections Diseases (Tropical Medicine Research Centers contract no. AI49486-02, Family Health International Contract NIH/NO1-AI-05403, and a grant to Chetan E. Chitnis from the World Health Organization Special Program in Tropical Diseases Research.
* Address correspondence to Myriam Arévalo-Herrera, Malaria Vaccine and Drug Testing Center, Cra 35 No. 4A-53, Cali, Colombia. E-mail: marevalo{at}immuno.org 
Authors’ addresses: Myriam Arévalo-Herrera, Angélica Castellanos, and Socrates Herrera, Instituto de Inmunología de la Universidad del Valle, Calle 4B 36-00, Edificio de Microbiología, Tercer Piso, Facultad de Salud, Universidad del Valle Sede San Fernado, AA 25574, Cali, Colombia, Telephone: 57-2-558-1931, Fax: 57-2-557-0449 and Malaria Vaccine Development Center, AA 26020, Cali, Colombia, Telephone: 57-2-558-3937, Fax: 57-2-556-0141, E-mail: marevalo{at}inmuno.org. Syed S. Yazdani, Ahmad Rushdi Shakri, Chetan E. Chitnis, Malaria Research Group, International Centre for Genetic Engineering and Biotechnology, Campus Aruna Asaf Ali Marg, 110067, New Delhi, India, Telephone: 91-11-2619-5007, Fax: 91-11-2616-2316. Rosalie Dominik, Family Health International, 2224 E. NC Highway 54, Durham, NC 27713, Telephone: 919-544-7040, Fax: 1-919-544-7261.
Reprint requests: Myriam Arévalo-Herrera, Malaria Vaccine and Drug Testing Center, Cra 35 No. 4A-53, Cali, Colombia, E-mail: marevalo{at}inmuno.org
|
REFERENCES |
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
