Research ArticleIMMUNOLOGY

IgM in human immunity to Plasmodium falciparum malaria

See allHide authors and affiliations

Science Advances  25 Sep 2019:
Vol. 5, no. 9, eaax4489
DOI: 10.1126/sciadv.aax4489
  • Fig. 1 Induction of IgM and IgG following controlled human P. falciparum infection.

    (A) Malaria-naïve healthy individuals underwent inoculation with 1800 or 2800 viable P. falciparum 3D7-parasitized RBCs, and peripheral parasitemia was measured by qPCR. Participants were treated with anti-malarial drugs at day 7 or 8 of infection. Blood samples from 60 volunteers (from 10 independent cohorts) were collected before infection (day 0), at peak infection (day 7 or 8), and 14 or 15 and 28 to 36 days after inoculation; in analysis, these time points are grouped as days 0, 7, 14, and 30. Data are representative of 10 individuals from four cohorts. (B) IgM and IgG levels were measured against MSP2, AMA1, and EBA175 (RII) at days 0, 7, 14, and 30 following controlled human P. falciparum infection in malaria-naïve adults (n = 40). Antibody responses are expressed as optical density (OD) after background subtraction of no-serum wells. Positivity threshold was calculated as means + 3SDs of responses at day 0 (dotted line). P values are Wilcoxon matched-pairs signed-rank test. (C) IgM and IgG responses to several merozoite antigens were measured at day 30 following infection in CHMI participants (n = 20). Prevalence of antibody responses was considered as a positive response if antibody levels were greater than means + 3SDs of responses at day 0. Data are proportion positive, and error bars are upper 95% confidence interval calculated by the Wald method. P value is calculated by Fisher’s exact test. (D) Magnitude of responses relative to day 0 was assessed as arbitrary units (AU = OD day 30/means + 3SDs of responses at day 0). P values are Wilcoxon matched-pairs signed-rank test. (E) Affinity of IgM and IgG to MSP2 was assessed by the dissociation of antibodies using increasing concentrations of ammonium thiocynate. Data are means ± SEM of eight individuals tested in two assays in duplicate, expressed as the percentage of OD in PBS. P values indicate Sidak’s multiple comparison test of IgM verses IgG from two-way analysis of variance (ANOVA), taking into account repeat measures for each individual (*P < 0.05 and **P < 0.01).

  • Fig. 2 Induction and waning kinetics of merozoite-specific IgM during naturally acquired P. falciparum malaria.

    (A and B) IgM, IgG1, and IgG3 binding to intact merozoites (A) and recombinant MSP2 (B) was assessed among individuals with clinical malaria (n = 50) and uninfected endemic community controls (n = 30) in individuals in Sabah, Malaysian Borneo. Antibody reactivity is shown as seroprevalence and media [plus interquartile range (+IQR)] OD. Seroprevalence (left) was calculated using a cutoff defined as the mean OD + 3SDs of unexposed Australian naïve controls. (C and D) IgM, IgG1, and IgG3 binding to intact merozoites (C) and recombinant MSP2 (D) was assessed in patients with clinical malaria in Sabah, Malaysian Borneo, and at 7 (n = 26) and 28 days (n = 32) after anti-malarial treatment. Solid lines and gray area are generated from LOWESS smooth calculations in R software. Raw data are indicated as points. (E) Affinity of IgM and IgG to MSP2 was assessed by the dissociation of antibodies using increasing concentrations of ammonium thiocynate. Data are means ± SEM of eight individuals tested in two assays in duplicate, expressed as the percentage of OD in PBS. P values indicate Sidak’s multiple comparison test of IgM versus IgG from two-way ANOVA, taking into account repeat measures for each individual (**P < 0.01 and ***P < 0.001).

  • Fig. 3 Maintenance of P. falciparum specific IgM and IgG.

    (A and B) IgM and IgG binding to recombinant MSP2 was measured in children from Ngerenya, Kenya who had no detectable parasite infection in October 2003, May 2004, and October 2004, during a period of rapidly declining transmission (n = 50). (A) IgM and IgG prevalence at the three time points (seropositive defined as greater than means + 3SDs of reactivity of malaria-naïve donors). P is Chi-square test. (B) Magnitudes of IgM and IgG reactivity over three time points. P value is calculated by nonparametric ANOVA and Friedman test. Dotted line indicates positive threshold calculated from means + 3SDs of reactivity of malaria-naïve donors. (C) IgM and IgG binding to recombinant MSP2 during clinical malaria in Australian returned travelers (n = 10). Dotted line indicates the seropositive threshold. (D) Magnitudes of IgM and IgG were measured at different time points following acute illness and treatment (n = 10). Colored lines correspond to different individuals. Dotted line indicates seropositive threshold. (E) There was no significant difference in the change in IgG and IgM levels at 30 to 60 days (n = 8) or between 180 to 300 days (n = 5) after disease compared to clinical time point (measured 0 to 7 days following treatment). Dotted line indicates no change. P values were calculated using the Wilcoxon matched-pairs signed-rank test.

  • Fig. 4 Merozoite-specific IgM inhibits invasion in a complement-dependent manner.

    (A) Isolated IgM and IgG fractions at 50 μg/ml from malaria-exposed donors (Sabah, Malaysian Borneo) or naïve donors (Australia) were tested in merozoite invasion inhibition assays (at 50 μg/ml) with normal serum (NS) or heat-inactivated serum (HIS) from malaria-naïve donors as a source of complement. Data are expressed as the percentage of invasion in heat-inactivated serum, with no antibody control. Data are means ± SD of two assays performed in duplicate. Significant differences are indicated as P < 0.05 by Mann-Whitney U test. *P < 0.05. (B) Isolated IgM and IgG fractions were tested for deposition of complement components (C1q, C3b, and C5b-C9) on intact merozoites or recombinant MSP2 by ELISA with 10% normal sera from malaria-naïve donors as a source of complement. Data represent the level of complement binding measured in OD values minus deposition observed using naïve IgM and IgG. (C) IgM and IgG fractions were tested for deposition of complement components on recombinant MSP2 by ELISA with 10% C1q-depleted sera and C1q-depleted sera reconstituted with purified C1q (10 μg/ml). Data are means ± SEM of two independent assays performed in duplicate. (D) The level of C1q-fixing antibodies and IgM and IgG1/IgG3 isotypes to merozoites was quantified in 50 individuals with a clinical malaria episode from Sabah. IgM, IgG1, and IgG3 were correlated with C1q fixation. (E) To compare directly between C1q fixation and IgM or IgG1/IgG3, antibody levels were converted to arbitrary units (the percentage of the highest measured response within cohort) and ranked according to levels of C1q fixation. In some individuals with relatively high C1q-fixing antibodies (defined as >40% of highest responder), IgM dominated the antibody response targeting the merozoite.

  • Fig. 5 Association between IgM and IgG levels to the merozoite and malaria risk.

    IgM and IgG levels to the merozoite surface were measured in plasma from a cohort of 199 children in PNG. (A) IgM and IgG were higher in older children (>9 years, n = 109). (B) High IgM and IgG binding were higher in children with concurrent P. falciparum infection (n = 134) than in uninfected children, as determined by PCR. (C) Correlation between IgM and IgG among children (Spearman rho and P) is indicated. (D to G) Children were stratified into three equal-sized groups based on tertiles according to low (including those classified as “negative”), medium, or high IgM and IgG responders, as determined by OD values for each sample. (D and E) High IgM and IgG binding to the merozoite was strongly associated with reduced risk of clinical malaria episodes. (F) Children who had at least one infection during the follow-up period (n = 190) were allocated into group defined as protected (always asymptomatic; n = 112) or susceptible (experiencing at least one malaria episode; n = 78) in follow up. P values calculated by Wilcoxon rank sum between groups are indicated.

  • Table 1 Association between IgM and IgG to the merozoite surface and risk of clinical malaria.

    The cohort (n = 199) was stratified into three equal groups according to low, medium, or high levels (defined by tertiles) of IgM to the merozoite surface. Unadjusted hazard ratios (uHRs) and HRs adjusted for the predetermined confounders of age and location of residence [adjusted HRs (aHRs)] are presented. IgM levels were compared between low-versus-medium (LvM) and low-versus-high (LvH) groups for the risk of symptomatic malaria (fever plus >5000 parasites/μl). Calculations were based on the first episode only.

    uHR [95% CI]PaHR [95% CI]
    age/location
    P
    IgMLvM0.44
    [0.23–0.81]
    0.0090.62
    [0.32–1.18]
    0.15
    LvH0.38
    [0.20–0.81]
    0.0020.47
    [0.25–0.89]
    0.02
    IgGLvM0.51
    [0.29–0.91]
    0.0210.58
    [0.32–1.06]
    0.08
    LvH0.24
    [0.12–0.49]
    <0.0010.3
    [0.14–0.64]
    0.002

Supplementary Materials

  • Supplementary material for this article is available at http://advances.sciencemag.org/cgi/content/full/5/9/eaax4489/DC1

    Fig. S1. IgM and IgG antibody induction to individual merozoite antigens following primary P. falciparum infection in naïve adults.

    Fig. S2. IgM, IgG1, and IgG3 during clinical P. falciparum malaria and following treatment in patients from Sabah.

    Fig S3. Purification of IgG and IgM fractions.

    Fig S4. Merozoite lysis with IgM and IgG fractions.

    Fig S5. C1q-fixing antibodies in Sabah individuals.

    Table S1. Proportion of responses above positive threshold.

    Table S2. Cohort characteristics of patients with clinical malaria from Sabah, Malaysia.

    Table S3. Australia resident returned travelers.

    Table S4. Prevalence and levels of IgM and IgG to the merozoite surface in the longitudinal cohort of PNG.

    Table S5. Associations between IgM, IgG, and C1q to the merozoite surface and odds of susceptibility to malaria in PNG children.

  • Supplementary Materials

    This PDF file includes:

    • Fig. S1. IgM and IgG antibody induction to individual merozoite antigens following primary P. falciparum infection in naïve adults.
    • Fig. S2. IgM, IgG1, and IgG3 during clinical P. falciparum malaria and following treatment in patients from Sabah.
    • Fig S3. Purification of IgG and IgM fractions.
    • Fig S4. Merozoite lysis with IgM and IgG fractions.
    • Fig S5. C1q-fixing antibodies in Sabah individuals.
    • Table S1. Proportion of responses above positive threshold.
    • Table S2. Cohort characteristics of patients with clinical malaria from Sabah, Malaysia.
    • Table S3. Australia resident returned travelers.
    • Table S4. Prevalence and levels of IgM and IgG to the merozoite surface in the longitudinal cohort of PNG.
    • Table S5. Associations between IgM, IgG, and C1q to the merozoite surface and odds of susceptibility to malaria in PNG children.

    Download PDF

    Files in this Data Supplement:

Stay Connected to Science Advances


Editor's Blog

Navigate This Article