|Home | About | Journals | Submit | Contact Us | Français|
Inhibitory antibodies specific for the 19-kDa fragment of merozoite surface protein 1 (MSP119) are a significant component of inhibitory responses in individuals immune to malaria. Nevertheless, conflicting results have been obtained in determining whether this antibody specificity correlates with protection in residents of areas where malaria is endemic. In this study, we examined sera collected from a population of semi-immune individuals living in an area of Vietnam with meso-endemicity during a 6-month period. We used two Plasmodium falciparum parasite lines that express either endogenous MSP119 or the homologous region from Plasmodium yoelii to measure the MSP119-specific inhibitory activity. We showed that (i) the level of MSP119-specific inhibitory antibodies was not associated with a delay in P. falciparum infection, (ii) MSP119-specific inhibitory antibodies declined significantly during the convalescent period after infection, and (iii) there was no significant correlation between the MSP119-specific inhibitory antibodies and the total antibodies measured by enzyme-linked immunosorbent assay. These results have implications for understanding naturally acquired immunity to malaria and for the development and evaluation of MSP119-based vaccines.
Infection of humans by Plasmodium falciparum remains one of the most deadly infectious diseases worldwide, leading to approximately 1 million deaths annually, predominantly in children under 5 years of age. It is the infection of red blood cells by asexual parasites that is associated with all clinical signs and symptoms and is responsible for malaria morbidity and mortality. In areas where malaria is endemic, immunity to this stage develops after repeated exposure and acts to prevent symptomatic illness and severe complications and to limit parasitemia (19). This immunity can be transferred passively among humans (6, 29), suggesting that antibodies are an important component of protective immunity. Attention has been devoted to mechanisms by which antibodies act to protect humans and to the identification of P. falciparum proteins that may be the targets of such protective antibodies and that may in turn induce such protective antibodies when administered in a vaccine.
The C-terminal 19-kDa fragment of merozoite surface protein 1 (MSP119) is a major target of protective antibodies against blood-stage infection and a leading candidate for inclusion in a subunit malaria vaccine. Studies with rodent and nonhuman primate models have shown that passive transfer of anti-MSP119 antibodies or immunization with recombinant MSP119 can provide significant protection against lethal challenge (8, 17, 18). Antibodies to MSP119, affinity purified from either immune human sera or monoclonal or polyclonal experimental sera, are capable of inhibiting parasite growth in vitro (3, 12, 27). However, the association between levels of MSP119-specific antibodies in humans and clinical immunity remains unclear. Using approaches such as enzyme-linked immunosorbent assay (ELISA), the levels of MSP119-specific antibodies have been quantified in many field studies, and correlations with protection have not been observed consistently (1, 11, 13, 16, 28, 30). ELISAs do not account for antibody affinity and fine specificity, which may be critical for functional activity. Monoclonal antibodies directed against MSP119 have been shown to have various effects on parasite growth, ranging from inhibition to enhancement. These specificities, as well as the presence of antibodies that block the action of inhibitory antibodies, have been detected in naturally acquired responses (15, 23). Thus, it remains unclear how antibody levels relate to inhibitory function in immune humans.
Relatively few field studies have examined the association between the subset of growth inhibitory antibodies and protective immunity due to methodological constraints on performing these assays in a reproducible and reliable manner (10, 20, 25). The recent development of paired transgenic P. falciparum lines that differ only in their MSP119 region has provided a tool with which to measure MSP119-specific inhibitory antibodies (24). By calculating the difference in the levels of inhibition of the two parasite lines in the presence of a particular serum, the inhibitory effect attributable to MSP119-specific antibodies can be determined. Using this assay, O'Donnell et al. demonstrated that MSP119-specific antibodies capable of inhibiting parasite growth were a major component of inhibitory responses in serum samples from individuals living in Papua New Guinea (24). However, such a finding does not establish if individuals with these particular inhibitory antibodies are immune and whether detection of these antibodies is an accurate correlate of protection. Two field studies using this functional assay reached conflicting conclusions: one study in western Kenya during a malaria epidemic revealed a correlation between MSP119-specific inhibitory antibodies and protection from P. falciparum infection (16), whereas a study conducted in Gambia showed that the MSP119-specific inhibitory antibodies were not associated with protection (7). It is important to resolve this discrepancy and to determine the generality of the observation and whether it applies outside Africa to different ethnic groups and differing levels of malaria transmission. It is also important to determine the kinetics of acquisition and maintenance of these specificities during infection.
We previously examined a group of transmigrants who experienced sequential infections during settlement in an area where malaria is highly endemic and showed that the acquisition of the MSP119-specific inhibitory antibodies required two or more infections (22). This contrasts with the case for Australian travelers returning with P. falciparum infection, as nearly half of the individuals showed significant inhibitory antibodies after their first infection (14).
In this study, we examined a population of semi-immune individuals from southern central Vietnam who were drug cured and then intensively monitored over a 6-month period for infection. We measured the MSP119-specific inhibitory antibodies in the serum samples from these individuals and showed that they did not correlate with a delay in infection. We also investigated the kinetics of MSP119-specific inhibitory antibodies during infection, drug treatment, and convalescence, as well as their correlation with antibody levels measured by ELISA.
The serum samples examined in this study have been described previously (32). They were collected in 1994 from residents living in the Khanh-Nam Commune of Khanh-Hoa Province in southern central Vietnam. Three species of Plasmodium were endemic to this area, and blood smear positivity rates were between 25 and 30%, with P. falciparum, Plasmodium vivax, and Plasmodium malariae identified in 54, 42, and 4% of infections, respectively. Older children and adults living in this area were classified as semi-immune, as only about half of those with parasitemia reported having malaria symptoms, and when symptoms were present, they were usually mild. At the commencement of the survey in June 1994, blood samples (T0) were obtained with informed consent from 134 volunteers aged 9 to 55 years (mean age, 27.5 years). These volunteers were drug treated to radically cure infection (quinine sulfate, doxycycline hyclate, and primaquine phosphate) and were monitored daily for symptoms and weekly for parasitemia by finger prick for a period of 6 months. Individuals who developed patent parasitemia were treated with mefloquine (15 mg/kg of body weight), and blood samples were collected at the time of treatment (T1) and 28 days later (T28). A total of 137 serum samples were chosen for this study, including samples collected at T0 (n = 38), T1 (n = 37), and T28 (n = 35) from 38 individuals who acquired P. falciparum parasitemia and the samples collected at T0 from 27 individuals who remained smear negative.
The invasion inhibition assay was performed as described previously (24), except that a new transgenic parasite line, D10-PyMEGF, in which the endogenous MSP119 fragment was replaced with the homologous region from Plasmodium yoelii (E. Murhandarwati, unpublished data), was used instead of D10-PcMEGF, which was used in the earlier study. Both D10-PyMEGF and the isogenic parental line D10-PfM3′ had similar growth rates to that of the wild-type, D10, in vitro. Ring-stage parasites of lines D10-PyMEGF and D10-PfM3′ were synchronized by sorbitol lysis twice at 4-h intervals, and the assay was conducted after incubation for 24 h, when parasites had matured to schizonts. Into each well, 50 μl of infected red blood cells (at 1% parasitemia and 2% hematocrit) and 50 μl of preadsorbed, heat-inactivated serum (1/5 dilution) were added to give a final serum dilution of 1/10. Each serum was tested on both lines in triplicate. Pooled human nonimmune serum (negative control) and rabbit anti-P. yoelii MSP119 (anti-PyMSP119) and rabbit anti-P. falciparum MSP119 (anti-PfMSP119) antibodies (positive controls) were included in each assay. After incubation for 38 to 42 h, late-trophozoite-stage or schizont-stage parasites were harvested, fixed with 0.025% glutaraldehyde, and stained with 10 to 15 μg/ml propidium iodide (PI). The parasites were then subjected to a Beckman Coulter FC500 cytometry analyzer with a 488-nm laser for excitation and an FL-3 channel (670 nm) for emission. One hundred thousand cells were counted for each sample. Uninfected red blood cells stained with PI were used to set the gating, with the average area inside the gating being around 0.05%. Parasitemia was calculated in batch, using FC500 CXP analysis software. The mean parasitemia for triplicate wells was calculated, and invasion was expressed as a percentage of the mean parasitemia observed in parallel cultures of each parasite line in the presence of pooled human nonimmune sera. Net inhibition attributable to MSP119-specific antibodies was calculated by subtracting the inhibition identified in the D10-PyMEGF cultures from the inhibition in the D10-PfM3′ cultures. Inhibitory activity was considered positive when net inhibition was greater than 15%, with a significant difference in invasion rates between the two parasite lines (P < 0.05). Negative inhibition values of <−15% with P values of <0.05 were considered to represent invasion enhancement of P. falciparum infection. In this assay, rabbit antibodies raised to PfMSP119 inhibited D10-PfM3′ and D10-PyMEGF by 43% and 7%, respectively, resulting in a PfMSP119-specific inhibition of 40%. Rabbit anti-PyMSP119 antibodies inhibited D10-PfM3′ and D10-PyMEGF by −1% and 41%, respectively, giving a PyMSP119-specific inhibition of 42%.
The reactivity of human sera with recombinant PfMSP119 was examined by ELISA as described previously (31, 32). Briefly, an MSP119 (Wellcome allele)-glutathione S-transferase fusion protein was used to coat 96-well microtiter plates. Human serum samples were used at a 1:2,000 dilution, and the specific optical density (OD) values were calculated by subtracting the control OD values measured on glutathione S-transferase alone. Immunoglobulin G1 (IgG1) and IgG3 were measured by an isotype-specific ELISA (32).
Statistical analyses were performed using Graphpad Prism5 software (Graphpad Software Inc., San Diego, CA). Student's t test was used to compare the invasion rates between the two parasite lines in the presence of a particular serum. The log rank test was used to determine the contribution of inhibitory antibodies to delaying subsequent P. falciparum infection. Spearman's rank correlation test was used to assess the correlation between MSP119-specific antibodies (total Ig and IgG subclasses) and the MSP119-specific inhibitory antibodies in each serum sample. The chi-square test was used to compare proportions of positive sera in different groups, whereas Wilcoxon and Mann-Whitney tests were used to compare the invasion rates and inhibitory levels between groups for paired and unpaired data, respectively.
Each of the serum samples was examined for inhibition of growth of the two P. falciparum transgenic lines, D10-PfM3′ and D10-PyMEGF, which are isogenic except for the MSP119 domain. MSP119-specific inhibitory activity was expressed as the difference in inhibition between these two parasite lines, whereas the total inhibitory activity could also be determined using the measurement for D10-PfM3′, which expresses endogenous MSP119. Flow cytometry was used in preference to microscopic analysis to determine parasitemia, due to the high sensitivity and low variation of flow methods (25). Of the total of 137 serum samples tested, 16 (11.7%) were positive for MSP119-specific inhibitory activity, which was defined as a net inhibition of >15% with a significant difference in invasion rates between the two parasite lines. The levels of MSP119-specific inhibition ranged between 16% and 68% (Table (Table1).1). Thirteen of the sera (9.5%) had significant negative inhibition (net inhibition of <−15% with a P value of <0.05) ranging between −18 and −70%, indicating the presence of MSP119-specific enhancement activity. The total inhibitory activities in all serum samples measured were between −60% and 82%.
At the beginning of the study (T0), 14 of the 65 individuals (21.5%) had significant MSP119-specific inhibitory activity, whereas 6 of these individuals (9.2%) showed significant MSP119-specific enhancement (Table (Table1).1). No correlation was observed between the level of MSP119-specific inhibitory activity and the age of the individuals.
We classified the individuals who acquired P. falciparum parasitemia during the 6-month surveillance period as susceptible and the individuals who did not acquire parasitemia with any species of Plasmodium as potentially protected. The individuals who developed parasitemia with P. vivax or P. malariae were excluded because of the possibility that they were indeed susceptible to P. falciparum but were not detected due to an intervening infection with another species. At the beginning of the study (T0), the levels of total inhibitory activity measured on D10-PfM3′ were not significantly different between the susceptible and protected groups (P = 0.6603), nor were the levels of MSP119-specific inhibitory activity (Fig. (Fig.1).1). The proportion of individuals with positive MSP119-specific inhibitory antibodies in the potentially protected group (8 of 27 individuals [30%]) was higher than that in the susceptible group (6 of 38 individuals [16%]), but the difference was not statistically significant (P = 0.1810).
To determine whether the level of inhibitory activity, either total or MSP119 specific, could delay subsequent P. falciparum infection in the susceptible group, samples at T0 were ranked in the order of inhibitory activity from least to greatest, and the top and bottom quartiles were analyzed for their correlation with the infection-free intervals (time between T1 and T0). As shown in Fig. Fig.2,2, neither the total nor MSP119-specific inhibitory antibodies were associated with delayed P. falciparum infection.
The serum samples taken at T0, T1, and T28 from the 38 individuals who acquired P. falciparum parasitemia were analyzed to examine the changes in inhibitory antibodies during the 6-month survey period. As shown in Fig. Fig.3,3, the total inhibitory activities measured on D10-PfM3′ were not significantly different between T0 and T1 but increased significantly at T28. In contrast, MSP119-specific inhibitory activity decreased after infection, with the levels at T28 being significantly lower than those at T0 and T1. The proportion of individuals with positive MSP119-specific inhibitory antibodies also decreased to 0% (0/35 individuals) at T28, compared to 16% (6/38 individuals) at T0 and 5% (2/37 individuals) at T1.
We previously analyzed anti-MSP119 antibodies by ELISA, using the same set of serum samples (31, 32). Anti-MSP119 antibodies were detected at a high prevalence and at high levels, and these were predominantly of the IgG1 subclass, with a lesser IgG3 response. The anti-MSP119 antibodies, either total Ig or the individual Ig subclasses, were not correlated with subsequent infection with P. falciparum (31, 32). We analyzed the association between MSP119-specific inhibitory antibodies and total antibodies measured by ELISA for all serum samples, including those from T0, T1, and T28. No significant correlation was observed between inhibitory activity and Ig, IgG1, or IgG3 (Fig. (Fig.44).
We have shown in this study that MSP119-specific inhibitory antibodies are present in semi-immune individuals living in southern central Vietnam. The proportion and levels of the inhibitory antibodies are modest compared to those observed in other areas of malaria endemicity reported by other research groups (7, 16, 24). In a cross-sectional survey of 187 Gambian children and adults, the median level of MSP119-specific inhibitory activity was ~25% (7). A similar level was also reported for groups of immune adults living in areas of endemicity in Papua New Guinea (24). In another population of semi-immune individuals (both children and adults) who lived in a highland area of western Kenya, the 75th percentile of MSP119-specific inhibitory activity in 76 serum samples taken during a malaria epidemic was 42.5% (16). This difference may reflect differences in exposure and number of infections or may represent a genetic difference leading to differing immune responses, as the ethnic groups in the different studies differ significantly. There are also methodological differences between the studies, for example, the control parasite lines used to measure specific inhibitory activity (wild-type D10 versus D10-Pf3′). The method of parasitemia determination, the stage when parasitemia was counted, and the definition of positive values all differed. To aid in cross-comparison studies in the future, it would be worthwhile to develop a panel of sera with differing levels of inhibition that could be provided to investigators to calibrate their assays. In the absence of such standardization, a degree of caution must be attached to comparisons between studies.
In this study, time to infection as a measure of immunity depended on a high level of natural challenge of all participants. The level of endemicity measured in this area is such that we would expect multiple infectious bites of all individuals, but it must be kept in mind that in natural challenge experiments there is a degree of uncertainty related to the magnitude and timing of challenge. Keeping this caveat in mind, we did not find a correlation between the MSP119-specific inhibitory activity and delayed infection in this study population. This is in contrast to the result reported by John et al. for a treatment time-to-infection study conducted over a 10-week period in western Kenya during a malaria outbreak (16). However, another study conducted with a group of Gambian children and adults also reported a lack of correlation between MSP119-specific inhibitory activity and the prevalence or density of parasitemia (7). Our results are consistent with the latter study and suggest that MSP119-specific inhibitory antibodies do not contribute greatly to the immune state in this study population. Some individuals lacking MSP119-specific inhibitory antibodies remained uninfected throughout the study, whereas others were infected although they had significant levels of MSP119-specific inhibitory antibodies. This underlines the complexity of naturally acquired protection, which is likely to involve multiple immune responses to multiple antigens and to be mediated through multiple effector mechanisms (19). Increasing evidence suggests that antibody action via the Fc interaction with monocytes/macrophages plays an important role in protective immunity (2, 4, 21, 26, 33). Thus, noninhibitory anti-MSP19 antibodies may still prevent parasite growth in vivo if, once bound, they are able to cooperate with monocytes/macrophages. It is also possible that MSP119-specific antibodies are not of key importance in naturally acquired immunity to P. falciparum infection.
It is interesting to note the fluctuation in MSP119-specific inhibitory activity during and after infection. At the time of infection (T1), both total and MSP119-specific inhibitory activities remained at similar levels to those found at T0. However, by 28 days after infection, the MSP119-specific inhibitory activity decreased significantly, although the total inhibitory activity increased substantially. In fact, none of the 35 sera had significant MSP119-specific inhibitory activity, with some sera now enhancing rather than inhibiting parasite growth. The increase in total inhibitory activity may be artifactual in part, as patients in the study were treated with mefloquine at T1 and the half-life of this drug of 2 to 3 weeks may have meant that there were significant levels still present at T28. If so, residual drug would inhibit both transgenic lines, leading to an increase in total inhibition. However, it is also possible that other antimalarial antibody specificities were boosted, leading to the observed increases. It is important, however, that the measurement of MSP119-specific inhibitory antibodies would not be confounded by the presence of the drug because the use of matched transgenic parasite lines would control for this case. We have previously shown the anti-MSP119 antibodies measured by ELISA to be increased significantly at T28 compared to T0 or T1 (31). The simplest interpretation is that the total anti-MSP119 antibodies are indicative of recent infection and boosting of the antibody responses, whereas the specific inhibitory antibodies are consumed during the convalescent period after infection, by either binding or agglutinating parasites. The mechanism for how this happens is obscure. Since the antibodies are made in response to the same molecule, one might expect that the trend would be the same for all classes of functional antibody. Perhaps neutral antibodies do not make contact with MSP119 in the intact merozoite and are not susceptible to depletion. Certainly, blocking or neutral antibodies directed to MSP119 are a common phenomenon in naturally acquired antibodies (7, 15, 23). Perhaps reinfection by another strain of P. falciparum can boost the total anti-MSP119 antibodies but not boost inhibitory antibodies. Consistent with this possibility are studies showing that antibodies induced by one allele of MSP119 cross-reacted with different alleles, but this cross-reactivity did not extend to inhibitory activity (9). In this study, time points beyond 28 days were not taken, and it is thus not possible to determine the kinetics and persistence of MSP119-specific inhibitory antibodies in this study population. Such information would be of interest, and future experiments may be able to provide this information.
We have previously shown for this set of serum samples that MSP119-specific antibodies are predominantly IgG1 and, to a lesser extent, IgG3 (31, 32); both of these are cytophilic subclasses with opsonizing and complex-fixing properties (5). However, there is no correlation between subclass or total Ig and delayed infection. In this report, we further showed that levels of MSP119-specific antibodies measured by ELISA do not correlate with their inhibitory activity. Although the inhibitory antibodies were measured for the MAD20 allele of MSP119, whereas the total antibodies were measured for the Wellcome allele, this is unlikely to be a complicating factor for data interpretation. Our previous studies using almost 100 sera have shown that ELISA values for antibodies measured with these two alleles are highly correlated (r = 0.96; P < 0.001) (22). The lack of correlation between total and inhibitory antibodies is consistent with results from other studies (7, 16) and has implications for the development of MSP119-based vaccines. If such vaccines are to protect via growth inhibition, it would appear that they would function by a mechanism that is not particularly common in immune populations who have developed immunity following repeated infections. It is starting to appear more likely that anti-MSP119 immunity may have some substantial component of Fc-dependent inhibition. It remains of considerable interest to determine the correlation between antibody levels measured by serology, invasion inhibitory activity, and protection among recipients protected with MSP119-based vaccines. Presently, such data can be obtained more readily with model systems, and we are currently addressing this by examining the inhibitory activity in sera from mice immunized with P. yoelii MSP119 and its relationship with in vivo protective efficacy measured by parasite challenge (Murhandarwati, unpublished data).
This work was supported by the National Health and Medical Research Council (NHMRC) of Australia and the National Institutes of Health (grant DK-32094). The field component of this study was conducted as a collaboration between the Institute for Malariology, Parasitology and Entomology, Hanoi, Vietnam, and U.S. Naval Medical Research Unit 2, Jakarta, Indonesia, in accordance with U.S. Navy regulations governing the protection of human subjects in medical research. E.E.H.M. is a recipient of an AusAID scholarship.
All protocols involving human subjects were reviewed and approved by institutional review boards in accordance with the U.S. Navy regulations (SECNAVINST 3900.39B) governing the use of human subjects in medical research.
We thank the Australian Red Cross Blood Service for the provision of human blood and serum.
The opinions and assertions herein are the private ones of the authors and are not to be construed as official or as reflecting the views of the U.S. Navy or the Department of Defense.
Editor: J. F. Urban, Jr.
Published ahead of print on 20 July 2009.