| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
 |
ABSTRACT |
 TOP ABSTRACT INTRODUCTIONSUBJECTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
(DBL-1 ) domain to the overall anti-VSA response. Thus, we conclude that both antibodies to rifin and, to a lesser extent, antibodies directed to conserved regions of the Plasmodium falciparum erythrocyte membrane protein 1 (PfEMP1) DBL-1 domain contribute to the overall antibody response to VSA. We also assessed the associations between these different antibody responses in a cohort of Gabonese children. We found marked differences related to the childrens’ history of presentation with either mild or severe malaria, but no consistent pattern that would indicate a specific role or influence of antibody responses to rifin. |
INTRODUCTION |
|
TOPABSTRACTINTRODUCTIONSUBJECTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
domain, a constituent of the so-called "head structure" common to all PfEMP1 variants, that is involved in the formation of rosettes and in cytoadherence.13–15 Recombinant DBL-1 domains can block rosette formation, as can antibodies in the sera of malaria patients, while peptides corresponding to the conserved regions of DBL-1 induce antibodies that identify putatively cross-reactive antibodies.16,17 A larger family of clonally variant molecules expressed by P. falciparum asexual stages, the rifin proteins, has recently been described. Rifins are encoded by the rif (repetitive interspersed family) multigene family and are inserted into the infected erythrocyte surface membrane.18,19 Unlike var genes, rif genes are only expressed at the late ring or early trophozoite stage of the asexual blood cycle, although the rifins usually appear on the surface of the infected erythrocytes at the same time as PfEMP1.20 This co-localization has prompted speculation that the expression and trafficking to the surface of the var and rif gene products are linked. Antibodies in the serum of semi-immune African adults show higher-level recognition of recombinant rifin proteins compared with non-immune African children, although in the latter higher levels of such antibodies are nevertheless associated with more rapid post-treatment parasite clearance kinetics and with an asymptomatic course of P. falciparum infection.21,22 An important feature that distinguishes rifins from PfEMP1 is the simultaneous transcription of several rif gene products by a single parasite, leading to the expression of multiple rifin proteins at the infected erythrocyte surface.18,19
In the study described here, we assessed the extent of the contribution of anti-rifin antibodies to the anti-VSA antibody profile detectable in a standardized flow cytometric assay. We used plasma samples from semi-immune Gabonese adults selected according to their known high or low reactivity with a panel of recombinant rifin proteins, and analyzed their reactivity to the VSA expressed by three different field isolates of P. falciparum. We made comparisons of anti-VSA responses following pre-adsorption of plasma samples with either recombinant rifin proteins or peptides corresponding to conserved regions of the DBL-1 domain of PfEMP1. In separate analyses, we also assessed the extent of the associations between anti-rifin, anti-DBL-1, and anti-VSA antibodies detected in plasma samples from cohorts of Gabonese children during and after hospitalization and treatment of an acute P. falciparum malaria attack.
|
SUBJECTS AND METHODS |
|
TOPABSTRACTINTRODUCTIONSUBJECTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
antibody levels in Gabonese children at an age when they are acquiring anti-malarial immunity. For this purpose, we compared antibody responses in plasma samples collected at three different time-points: acute phase (pre-treatment, n = 61 mild and n = 61 severe), convalescent phase (one month post-treatment, n = 22 mild and n = 15 severe), and healthy phase (at least six months post-treatment, n = 25 mild and n = 26 severe). The 1-95/C study was conducted at the Albert Schweitzer Hospital in Lambaréné and was initiated in 1995.25,26 The study included 100 children presenting with severe malaria who were admitted to the hospital and an equal number presenting with mild malaria that were pair-matched to severe malaria cases by sex, age, and area of residence. Inclusion criteria for severe malaria were a P. falciparum parasitemia > 1,000/µL, older than six months, not homozygous for hemoglobin S, severe anemia (hemoglobin level < 5g/dL), and/or hyperparasitemia (> 250,000 parasites/µL), with or without other signs of severe malaria, for example, loss of consciousness, hypoglycemia, lactic acidosis, or respiratory distress.27 The level of consciousness was determined using the Blantyre coma score.27 Mild malaria was defined as a parasitemia between 1,000 and 50,000 parasites/µL on admission, no schizontemia, circulating leukocytes containing malarial pigment < 50/µL, hemoglobin > 8 g/dL, platelets > 50/nL, leukocytes < 12/nL, lactate < 3mM, and glycemia > 50 mg/dL. Exclusion criteria for the mild malaria controls were signs of severe malaria, concomitant acute infection, prior hospitalization for any reason, and intake of antimalarial agents within the preceding week. All study participants or their parents or guardians gave informed consent. Ethical approval for the study was obtained from the ethics committee of the International Foundation of the Albert Schweitzer Hospital in Lambaréné.
Parasite isolates and cultivation. The three P. falciparum isolates used in this study (laboratory reference codes cys007, cys002, and cym033) were obtained from pediatric patients recruited in a separate outpatient study carried out in 1997 at the Albert Schweitzer Hospital. They were adapted to in vitro culture conditions using standardized methods, as previously described.5,28 Briefly, infected erythrocytes were resuspended in complete medium supplemented with 10% heat-deactivated, pre-screened malaria non-immune AB+ serum (from the blood bank of University Hospital, Tübingen, Germany), and were then incubated in an atmosphere of 5% CO2, 5% O2, and 90% N2. Fresh O+ erythrocytes depleted of lymphocytes were periodically added. Isolates were initially expanded over a short period of 8–10 multiplication (48 hours) cycles, after which identical stabilates of cultures containing mostly asexual ring forms were cryopreserved for later use in flow cytometric assays.5
Flow cytometric measurement of IgG antibodies with specificity for the VSA of P. falciparum. The IgG responses with specificity for the VSA of P. falciparum expressed on trophozoite-infected erythrocytes (T-IE) was performed using a standardized flow cytometric assay described in detail elsewhere.5,29 Briefly, frozen aliquots of parasites were thawed and cultured until a sufficient number were available for assays. Parasite cultures were then synchronized and enriched by flotation on plasmagel (Fresenius, Louviers, France), the parasitemia was adjusted to 10–15% (5% hematocrit), and plasma samples (diluted 1:50 in phosphate-buffered saline [PBS]-1% bovine serum albumin [BSA]) added to micro-volumes of T-IE suspension in the wells of 96-well culture plates (Costar, Corning, NY). After incubation for 30 minutes at room temperature, T-IE were washed and incubated for an additional 30 minutes with mouse anti-human IgG (1:100 in PBS-1% BSA). After another washing step, antibodies bound to the surface of T-IE were detected by addition of fluorescein isothiocyanate-conjugated goat anti-mouse IgG (1:100 in PBS-1% BSA) containing 50 µg/mL of ethidium bromide. The T-IE were washed once, and re-suspended in PBS for analysis on a fluorescence-activated cell sorter (FACS) (FACScan flow cytometer; Becton Dickinson, Heidelberg, Germany). The flow cytometry data were analyzed using the CELLQuest 3.3 software (Becton Dickinson).
Samples were segregated into infected (T-IE) and uninfected (E) erythrocytes using forward and side scatter parameters, and a gate defining fluorescing (ethidium bromide-stained) cells further segregated parasite-infected cells. Using mean fluorescence intensity (MFI) values, the amount of IgG specifically bound to the surface of T-IE was estimated by application of the following formula: MFI = (MFIT-IE test –MFIE test) – MFIT-IE NC – MFIE NC), where NC represents a negative control pool of plasma from non-exposed Europeans. A threshold value of positivity was established for each isolate using a panel of 50 individual plasma samples from malaria non-exposed Europeans (University Hospital, Tu Tübingen, Germany), such that test samples were considered anti-VSA IgG responders when the MFI calculated with the above equation was greater than the mean plus two standard deviation of the values obtained with these control samples. The MFI threshold established in this way for the three isolates were cys002 = 0.0, cys007 = 4.9, and cym033 = 1.7.
Rifin proteins and PfEMP1-derived DBL-1 peptides.
Recombinant rifin proteins (a pool of four different recombinant rifin proteins) were used for anti-rifin antibody adsorption. Proteins were purified on affinity columns to 95% homogeneity, using procedures described in detail elsewhere.21 Three synthetic DBL-1 peptides, corresponding to highly conserved regions of the PfEMP1 DBL-1 head-structure domain, were deduced from "pile-up" displays of published PfEMP1 sequences as previously described,17 and were used individually in ELISAs and as a pool for anti-DBL-1 antibody adsorption of plasma samples. The three peptides were obtained commercially at > 97% purity (ThermoHybaid, Ulm, Germany) coupled to BSA. Their individual amino acid sequences are as follows: P1: GACAPYRRLHLCD; P2: LARSFADIGDIVRGKDLY; and P3: VPQYLRWFEE-WAEDFCRK.
The levels of IgG antibody activity with specificity for the individual peptides were determined with a standardized ELISA as previously described, but incorporating the following minor technical modifications.30 Peptides were used for coating of MaxisorpTM 96-well microtiter plates (Nunc, Roskilde, Denmark) at a concentration of 10 µg/mL, and the subsequent blocking step consisted of a six-hour incubation in a constant-temperature water bath at 25°C. Plasma samples were diluted 1:100 in PBS-0.5% Tween 20 (Sigma, Deisenhofen, Germany)-1% BSA. Peroxidase-conjugated goat anti-human IgG (Fc-specific; Sigma) was added at a 1:50,000 dilution for peptide P1 and at 1:10,000 for peptodes P2 and P3.
Pre-adsorption of anti-rifin and anti-DBL-1 antibodies from plasma samples was carried out as follows. Ninety-six-well microtiter plates (Costar) were coated with 100 µL of pooled rifin proteins or pooled DBL-1 peptides at concentrations of 4, 8, and 16 mu;g/mL in coating buffer (15 mM Na2CO3, 35 mM NaHCO3, pH 9.5) and incubated overnight at 4°C. Purified protein derivative of Mycobacterium tuberculosis (PPD; Statens Seruminstitut, Copenhagen, Denmark) at a concentration of 16 µg/mL was used as a control protein, and control wells with no protein/peptide received coating buffer alone at this stage. After four washes (PBS containing 0.5% Tween 20) 200 µL of blocking buffer (PBS containing 4% BSA, Fraction V; Serva, Heidelberg, Germany) was added to each well and plates incubated for five hours at room temperature. An additional washing step was followed by addition to each well of 100 µL of test plasma (diluted 1:50 in washing buffer containing 1% BSA) and incubation overnight at 4°C. The plates were then centrifuged for two minutes at 1,200 rpm, and plasma-containing supernatants were collected for subsequent FACS analysis.
Statistical analysis. Data were analyzed using the Stat-viewTM (SAS, Cary, NC) software program. For paired and unpaired comparisons of continuous variables, respectively, the non-parametric Wilcoxon sign rank and Mann-Whitney U test tests were used. Contingency tables with continuity corrections were used to compare proportions within and among groups. Correlations between continuous variables were assessed with the non-parametric Spearman rank test. A two-sided P value < 0.05 was considered to indicate statistical significance in all cases.
|
RESULTS |
|
TOPABSTRACTINTRODUCTIONSUBJECTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
domain of PfEMP1 to the anti-VSA antibody response in the plasma of cohorts of African adults and children that are detected with a standardized flow cytometric assay using a panel of locally-collected wild parasite isolates adapted to in vitro culture conditions over a limited time period.
Anti-VSA antibody activity of semi-immune adults as a function of antibody responses to recombinant rifin proteins.
Total IgG antibody responses to the variant surface antigens expressed by three different parasite isolates (cys002, cys007, and cym033) were measured with a standardized flow cytometric assay. Using this assay, we first addressed the question of whether pre-adsorption of plasma samples with recombinant rifin proteins would give an indication of the extent of the contribution, if any, of antibodies to rifin to the anti-VSA response measured in this way. For this purpose, we used plasma samples from a panel of semi-immune Gabonese adults, selected according to their known high (17 individuals) or low (7 individuals) anti-rifin antibody responses measured by an ELISA. Samples were pre-incubated with either 16 µg/mL of recombinant rifin proteins, PPD, or with no protein and then incubated with T-IE of each isolate. The results are shown in Figure 1
. The anti-VSA antibody activity of samples incubated with PPD was identical to that of samples incubated without protein (Figure 1A). Samples incubated with pooled recombinant rifin proteins exhibited significantly reduced anti-VSA activity with respect to all three wild isolates (Figure 1B–D). The magnitude of this reduction varied according to the individual isolates, but overall the MFI of pre-adsorbed samples was reduced by ~50% relative to the unadsorbed samples (Table 1). According to the classification of individuals into high or low anti-rifin antibody responders, no consistent pattern was evident with respect to reduced anti-VSA activity with the three isolates (Table 1), but the reduction of responses to isolate S007 was significantly greater in samples from those with low ELISA-derived anti-rifin activity (P = 0.039, by Mann-Whitney U test). In a separate experiment, plasma samples were pre-incubated with rifin proteins in solution (16 µg/mL in PBS), following which the reductions in anti-VSA activity detected were similar to those observed after pre-adsorption using plate-bound proteins.
|
|
antibodies: dose-dependent reduction of anti-VSA activity.
Since the reductions in anti-VSA antibody activity directed to the three different parasite isolates resulting from adsorption of antibodies to rifin were similar, we opted to use only a single parasite isolate (cys007) for subsequent experiments. To confirm that the effects of pre-incubation were indeed antigen-specific, we next used different concentrations (4, 8, and 16 µg/mL) of 1) pooled recombinant rifin proteins and 2) pooled PfEMP1 DBL-1 domain conserved region peptides for adsorption of antibodies from the same set of plasma samples from semi-immune Gabonese adults described earlier. The results are shown in Figure 2, and are presented as the median levels of the relative inhibition of anti-VSA antibody responses according to the different protein/peptide concentrations used for pre-adsorption. For both rifin proteins and DBL-1 peptides, there was a clear dose-dependency of the anti-VSA response such that the higher the concentration of proteins/peptides used, the lower the anti-VSA antibody activity detected by flow cytometry (Figure 2). At equivalent concentrations, the relative inhibition of anti-VSA activity was greater with the pool of recombinant rifin proteins compared with the pool of DBL-1 peptides.
|
, and anti-VSA antibody levels in non-immune Gabonese children.
Since the experiments described earlier with semi-immune adult plasma samples indicated that antibodies directed to both rifin proteins and the PfEMP1 DBL-1 domain do contribute to the overall anti-VSA response detectable by cytometry, we wished to assess the extent to which the levels of these antibodies with differing specificities might be correlated in non-immune Gabonese children. For this purpose, the responses at different times during and after acute P. falciparum malaria episodes in groups of Gabonese children segregated according to the clinical severity of their presentation at admission were assessed and compared. The profile of anti-DBL-1 peptide IgG antibody responses is shown in Figure 3. The kinetics of these responses differed according to the individual peptides, but the proportion of those with detectable peptide-specific antibody responses did not differ according to the childrens’ clinical status at admission (Figure 3). The highest level recognition was detected with peptide 3 when children were healthy and parasite free (Figure 3C). The magnitudes of anti-peptide responses at the separate time-points did not differ according to the childrens’ clinical status.
|
IgG antibody levels were positively and statistically significantly correlated in acute phase samples regardless of the clinical severity of presentation (Figure 4). No such associations were evident in either convalescent or healthy phase samples of either group. We also assessed associations between flow cytometrically-determined anti-VSA IgG antibody levels and ELISA-derived anti-rifin IgG antibody levels in the same Gabonese childrens’ plasma samples. In this case, we found no significant associations in acute-phase samples of either group, but there was a statistically significant negative correlation in convalescent phase samples from children in the group that presented with mild malaria but not in the group with severe malaria (Figure 5).
|
|
|
DISCUSSION |
|
TOPABSTRACTINTRODUCTIONSUBJECTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Our findings with well-characterized samples from semi-immune African adults in a standardized flow cytometric assay show that antibodies to anti rifin that can be pre-adsorbed with recombinant rifin proteins may represent up to 50% of the total antibody activity with specificity for the parasite-derived variant antigens inserted into the infected erythrocyte membrane. The data showing rifin protein dose dependency of the reduction in anti-VSA antibody activity measured in this way substantiates this idea. We interpret the substantial reduction of anti-VSA activity observed after such pre-adsorption of plasma samples shown initially by ELISA to contain only small amounts of antibodies to rifin as a possible indication that the level of antibodies measurable in a given assay may not always directly reflect the functional capacity of such antibodies. Others have recently reported similar apparent discrepancies in studies of anti-VSA antibody activity.37 Since the isolates used in these assays represent an unselected heterologous sample drawn from the parasite population in the locality of residence of these individuals, it is plausible to suppose that such antibodies comprise a mixture of specificities for both rifin variant-specific as well as cross-reactive epitopes, as is reported in studies of antibodies to VSA as well as of antibodies to PfEMP1.33,35–41 The fact that the four recombinant rifin proteins used here each comprise a part of the conserved and the majority of the variable regions of these molecules is consistent with this supposition.21 The similarly modified profile of anti-VSA antibody activity we obtained here after pre-incubation of the same semi-immune adult plasma samples with a pool of DBL-1 domain conserved region peptides lends further credence to the idea that distinct conserved epitopes of PfEMP1 may induce cross-reactive antibodies.
Antibodies to rifin in the cohort of Gabonese children studied here are sustained over a long period of time, but although high level antibody responses are associated with faster post-treatment parasite clearance kinetics, they do not represent a marker for anti-malarial protection per se and do not differ in either their magnitude or their kinetics in groups segregated according to the clinical severity of malaria at admission.22 The positive correlation between the anti-rifin and the anti-DBL-1 antibody response we observed here in samples from the acute attack suggests simultaneous infection-related enhancement of these responses that most likely reflects the on-going dynamics of parasite multiplication, death, and antigen processing. The absence of similar associations in the post-treatment samples may at least partly reflect the marked down-modulation of convalescent phase responses to the DBL-1 peptide 3, the most immunogenic of the three peptides we used. In contrast, the inverse correlation between anti-rifin and anti-VSA antibody responses we observed during convalescence in those that presented with mild malaria may be indicative of differential regulation of B cell-mediated responses to these antigens following drug-induced clearance of parasites. High convalescent phase IgG2 anti-VSA antibody responses in this group show a non-significant trend towards an association with delayed reinfection (Yone C and others, unpublished data). Those in this group with high convalescent phase anti-rifin responses and correspondingly low anti-VSA responses may thus be at a disadvantage with respect to subsequent reinfection. We have also shown that high convalescent phase anti-VSA IgG1 antibody isotype responses are strongly associated with longer delays to reinfection, but only in the group that presented with severe malaria (Yone C and others, unpublished data). The absence of an association between anti-rifin and anti-VSA antibody responses in this group of children reinforces the idea that for reasons that remain unclear, B cell-mediated responses to these erythrocyte surface-expressed antigens display marked inter-individual variability.
In summary, the results show clearly that antibodies with specificity for both rifins and conserved epitopes of the PfEMP1 DBL-1 domain comprise a substantial part of the anti-VSA antibody repertoire detectable in a standardized flow cytometric assay in the plasma of semi-immune African adults. They also suggest that the antibody responses to these highly polymorphic parasite antigens in non-immune African children vary both at the individual level and as a function of the presence or absence of active P. falciparum infection.
Received October 27, 2003. Accepted for publication February 11, 2004.
Acknowledgments: We are indebted to the study participants for their willing and enthusiastic cooperation and to the staff of the Albert Schweitzer Hospital in Lambaréné. We also thank Anselme Ndzengé and Marcel Nkeyi for technical assistance. The 1/95-C study was initiated in 1995 and inclusion into the study was completed in 1996. Follow-up surveillance continued until February 2002. We acknowledge the important contribution to the data included in this manuscript of the following members of the study team: Ruprecht Schmidt-Ott, Leopold G. Lehman, Doris Luckner, Bernhard Greve, Peter Matousek, Klaus Herbich, Daniela Schmid, Milena Sovric, Birgit Bojowald, Hanna Rudloff, Andreas Schindler, and Michel A. Missinou. Mohamed S. Abdel-Latif expresses his gratitude to Dr. Mo Quen Klinkert for her support.
Financial support: This study was supported by the fortune programme of the Medical Faculty of the University of Tuebingen (863-0-1), by the European Commission (QLK2-CT-2002-01197), and by the Deutsche Forschungsgemeinschaft (DFG, Ku775/12-1).
Authors’ address: Mohamed S. Abdel-Latif, Gerardo Cabrera, Carsten Köhler, Peter G. Kremsner, and Adrian J. F. Luty, Department of Parasitology, Institute for Tropical Medicine, University of Tübingen, Wilhelmstrasse 27, 72074 Tübingen, Germany.
Reprint requests: Adrian J. F. Luty, Department of Parasitology, Institute for Tropical Medicine, University of Tübingen, Wilhelmstrasse 27, 72074 Tübingen, Germany, Telephone: 49-7071-2980228, Fax: 49-7071-295189, E-mail: adrian.luty{at}uni-tuebingen.de.
|
REFERENCES |
|
TOPABSTRACTINTRODUCTIONSUBJECTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  S. Liu, J. Mu, H. Jiang, and X.-z. Su Effects of Plasmodium falciparum Mixed Infections on In Vitro Antimalarial Drug Tests and Genotyping Am J Trop Med Hyg, August 1, 2008; 79(2): 178 - 184. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. E. Blythe, X. Y. Yam, C. Kuss, Z. Bozdech, A. A. Holder, K. Marsh, J. Langhorne, and P. R. Preiser Plasmodium falciparum STEVOR Proteins Are Highly Expressed in Patient Isolates and Located in the Surface Membranes of Infected Red Blood Cells and the Apical Tips of Merozoites Infect. Immun., July 1, 2008; 76(7): 3329 - 3336. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. del Pilar Crespo, T. D. Avery, E. Hanssen, E. Fox, T. V. Robinson, P. Valente, D. K. Taylor, and L. Tilley Artemisinin and a Series of Novel Endoperoxide Antimalarials Exert Early Effects on Digestive Vacuole Morphology Antimicrob. Agents Chemother., January 1, 2008; 52(1): 98 - 109. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. P. ALKER, P. LIM, R. SEM, N. K. SHAH, P. YI, D. M. BOUTH, R. TSUYUOKA, J. D. MAGUIRE, T. FANDEUR, F. ARIEY, et al. PFMDR1 AND IN VIVO RESISTANCE TO ARTESUNATE-MEFLOQUINE IN FALCIPARUM MALARIA ON THE CAMBODIAN-THAI BORDER Am J Trop Med Hyg, April 1, 2007; 76(4): 641 - 647. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J.-P. Guthmann, L. Pinoges, F. Checchi, S. Cousens, S. Balkan, M. van Herp, D. Legros, and P. Olliaro Methodological Issues in the Assessment of Antimalarial Drug Treatment: Analysis of 13 Studies in Eight African Countries from 2001 to 2004 Antimicrob. Agents Chemother., November 1, 2006; 50(11): 3734 - 3739. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. G. BREMAN, M. S. ALILIO, and A. MILLS CONQUERING THE INTOLERABLE BURDEN OF MALARIA: WHAT'S NEW, WHAT'S NEEDED: A SUMMARY Am J Trop Med Hyg, August 1, 2004; 71(2_suppl): 1 - 15. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
) and statistical significance of the associations (p), as assessed by the non-parametric Spearman rank correlation test corrected for ties, are shown. Regression lines include 95% confidence levels (dotted lines).