Clinical protection from falciparum malaria correlates with neutrophil respiratory bursts induced by merozoites opsonized with human serum antibodies

Abstract

Background: Effective vaccines to combat malaria are urgently needed, but have proved elusive in the absence of validated correlates of natural immunity. Repeated blood stage infections induce antibodies considered to be the main arbiters of protection from pathology, but their essential functions have remained speculative.

Methodology/principal findings: This study evaluated antibody dependent respiratory burst (ADRB) activity in polymorphonuclear neutrophils (PMN) induced by Plasmodium falciparum merozoites and antibodies in the sera of two different African endemic populations, and investigated its association with naturally acquired clinical protection. Respiratory bursts by freshly isolated PMN were quantified by chemiluminescence readout in the presence of isoluminol, which preferentially detects extra-cellular reactive oxygen species (ROS). Using a standardized, high throughput protocol, 230 sera were analyzed from individuals of all age groups living in meso- (Ndiop) or holo-endemic (Dielmo) Senegalese villages, and enrolled in a cross-sectional prospective study with intensive follow-up. Statistical significance was determined using non-parametric tests and Poisson regression models. The most important finding was that PMN ADRB activity was correlated with acquired clinical protection from malaria in both high and low transmission areas (P = 0.006 and 0.036 respectively). Strikingly, individuals in Dielmo with dichotomized high ADRB indexes were seventeen fold less susceptible to malaria attacks (P = 0.006). Complementary results showed that ADRB activity was (i) dependent on intact merozoites and IgG opsonins, but not parasitized erythrocytes, or complement, (ii) correlated with merozoite specific cytophilic IgG1 and IgG3 antibody titers (P<0.001 for both), and (iii) stronger in antisera from a holo-endemic compared to a meso-endemic site (P = 0.002), and reduced in asymptomatic carriers (P<0.001).

Conclusions/significance: This work presents the first clearly demonstrated functional antibody immune correlate of clinical protection from Plasmodium falciparum malaria, and begs the question regarding the importance of ADRB by PMN for immune protection against malaria in vivo.

Figure 1. Characterization and activity of merozoite preparations.

(A) Confocal imaging of a merozoite preparation showing a simple transmission image superimposed with immuno-fluorescence staining by a 1∶500 dilution of rabbit polyclonal anti-baculovirus recombinant PfMSP1p19 , and a 1∶1000 dilution of Alexa Fluor 488 conjugated goat anti-rabbit IgG (H+L) (Molecular Probes). The insert at upper right shows control immuno-fluorescence staining on the same preparation using a 1∶500 dilution of rabbit polyclonal anti-[baculovirus-null infected insect cells] and the Alexa Fluor 488 conjugate. (B) Immuno-fluorescence staining with anti-PfMSP1p19 as in (A) above showing two images from a confocal 3-D projection, either a direct view perpendicular to the slide (left) or at a 90° angle (right). The images are enlarged 5x compared to (A). (C) Chemiluminescence using HIS and merozoites prepared from (i) synchronized mature schizont cultures and percoll fractionation (blue trace) or (ii) centrifuged culture supernatants (orange trace). (D) ADRB indexes determined with a single pool of PMN using fresh (orange), frozen (green) or sonicated (3×5″ at 80 watts) (blue) merozoites and immune (HIS, S2, S3, or S4) or non-immune (NIS, S5, S6) sera.

Figure 2. PMN respiratory burst activity depends on merozoites and immune antibodies.

ADRB indexes determined as described in Materials and Methods using a single pool of PMN: with immune (HIS) or non-immune (NIS) sera only (blue); with merozoites and immune (HIS, S2, S3, or S4) or non-immune (NIS, S5, S6) sera (green) or the same with fMLP activated PMN (yellow); with normal (light red), parasitized (dark red) RBC, or lysed parasitized RBC (3x freeze-thawed) (grey) in the presence of immune or non-immune sera.

Figure 3. ADRB chemiluminescence profiles can be standardized.

(A-C) Chemiluminescence profiles from 3 different ADRB experiments (3 PMN pools) tested without serum (blank, green trace), or in the presence of NIS (grey trace), HIS (red trace) and the same individual sample serum (S1, blue trace). The ADRB index for S1, calculated as described in Materials and Methods using HIS as a positive internal standard, is shown for each experiment.

Figure 4. Serum IgG, but not complement, mediates ADRB activity.

ADRB indexes were determined individually using 7 different immune sera either untreated (lane 1), after inactivation of complement (lane 2), after IgG depletion (lane 3), or after purification of IgG (lane 4). Distribution and median values are indicated by vertical and horizontal lines respectively.

Figure 5. Comparison of ADRB index profiles for Dielmo and Ndiop.

ADRB index values for Dielmo and Ndiop were analyzed as a function of parasite carriage at the time of sampling. Statistical significance was determined using the Mann-Whitney test and significant P values are noted.

Figure 6. Chemiluminescence activity correlates with anti-merozoite IgG1 and IgG3 isotype titers.

Sera from 88 aparasitemic Ndiop residents were stratified into 2 groups with ADRB indexes <250 (blue, 59 sera) or ADRB ≥250 (red, 29 sera). Stratified ADRB indexes are shown as a function of (A) total anti-merozoite IgG titer (1∶5000 dilution), and (B) separate anti-merozoite IgG1 and IgG3 titers (1∶200 dilution for both). Statistically significant differences, as detected with the Mann-Whitney test, are indicated.

Figure 7. Dichotomized ADRB indexes have an age-dependent impact on clinical episodes at both low and high transmission sites.

Analyses of associations between ADRB index and the number of clinical episodes experienced during a 5.5-month follow-up period, gave the best Poisson regression model with a cut-off of 250. The cumulative mean no. of clinical episodes/person/day at monthly intervals across the follow-up period were plotted for different age groups as a function of ADRB index <250 (blue trace) or ≥250 (red trace). (A) Ndiop: age groups: <15 years, 15–29 years, and ≥30 years (Table 1); (B) Dielmo: age groups: <7 years, 7–14 years, ≥15 years (Table 3).