Long-lived antibody and B Cell memory responses to the human malaria parasites, Plasmodium falciparum and Plasmodium vivax

Abstract

Antibodies constitute a critical component of the naturally acquired immunity that develops following frequent exposure to malaria. However, specific antibody titres have been reported to decline rapidly in the absence of reinfection, supporting the widely perceived notion that malaria infections fail to induce durable immunological memory responses. Currently, direct evidence for the presence or absence of immune memory to malaria is limited. In this study, we analysed the longevity of both antibody and B cell memory responses to malaria antigens among individuals who were living in an area of extremely low malaria transmission in northern Thailand, and who were known either to be malaria naïve or to have had a documented clinical attack of P. falciparum and/or P. vivax in the past 6 years. We found that exposure to malaria results in the generation of relatively avid antigen-specific antibodies and the establishment of populations of antigen-specific memory B cells in a significant proportion of malaria-exposed individuals. Both antibody and memory B cell responses to malaria antigens were stably maintained over time in the absence of reinfection. In a number of cases where antigen-specific antibodies were not detected in plasma, stable frequencies of antigen-specific memory B cells were nonetheless observed, suggesting that circulating memory B cells may be maintained independently of long-lived plasma cells. We conclude that infrequent malaria infections are capable of inducing long-lived antibody and memory B cell responses.

Figure 1. Antibody responses to P. falciparum antigens and tetanus toxoid.

Antibody titres against tetanus toxoid (A) and P. falciparum antigens (B–F) among City naïve (circle), Rural 1 (triangle) or Rural 2 (inverted triangle) subjects at the time of recruitment were determined by indirect ELISA. Each symbol represents the antibody titre of one individual. Solid lines show the median antibody titres in each group. The Mann Whitney U test was used to analyse differences in the levels of antibodies or memory B cells among groups. Figures G–L show the percentages of all rural (i.e. Rural 1 plus Rural 2) subjects who had antibody titres above the cut-off for each antigen at the time of recruitment and 12 months later. Fischer's exact test was used to analyse differences in the proportion of seropositives at recruitment compared to 12 months later but no significant differences were observed. The antibody titres for each seropositive subject over the 12 months of the study are shown in figures M–R. Dotted lines show cut-off values calculated from a mixture model as described in materials and methods.

Figure 2. Antibody responses to P. vivax antigens.

A–C show antibody titres against P. vivax antigens among City naïve (circle), Rural 1 (triangle) or Rural 2 (inverted triangle) subjects at the time of recruitment. Each symbol represents the antibody titre of one individual. Solid lines show the median antibody titres in each group. The titres of antibodies of seropositive rural (i.e. Rural 1 plus Rural 2) subjects at the time of recruitment and at each time point during the 12 months of study are shown in figures D–F. Dotted lines show cut-off values calculated from a mixture model as described in materials and methods.

Figure 3. Avidity indices for anti-PfAMA-1 and PfMSP-1₁₉ antibodies.

Avidity indices for PfAMA-1 (A) and PfMSP-1₁₉ (B) at the time of recruitment were compared between Rural 1 and Rural 2 groups. Avidity indices for antibodies to both antigens were compared at the time of recruitment and at 12 months later for Rural 1 (C and D) and Rural 2 (E and F) subjects. Lines show the mean (95% CI) for each group. An unpaired Student's t-test test was used to analyse differences between rural groups. A paired t-test was used to confirm that there are no differences between indices at time of recruitment and 12 months later.

Figure 4. Longevity of anti-malarial antibody responses.

The titres of antibodies specific to PfSE (A), PfAMA-1 (B) and PfMSP-1₁₉ (C) in relation to time since last clinical infection in P. falciparum exposed individuals (Rural 2 only) were determined by analyzing longitudinal data with a mixed-effects model. Each symbol represents the antibody level at each time point of one individual. The regression analysis was adjusted for inclusion of multiple data points from the same individual. Solid lines represent best fit regression lines estimating the rates of decline of antibody concentrations over time and the dashed lines represent the 95% CI. Horizontal dotted lines indicate the cut-off as defined in Materials and Methods and Figure 1.

Figure 5. B cell memory responses to malaria antigens and tetanus toxoid.

B cell memory responses to P. falciparum antigens (A–D), P. vivax antigens (E and F) and tetanus toxoid (I) at the time of recruitment were determined by ELISPOT assay and are presented as the percentage of all IgG-secreting cells that are specific for each malaria antigen. Each symbol represents the memory B cell numbers for one individual. The longevity of the memory B cell responses specific to PfAMA-1 (G) and PfMSP-1₁₉ (H) were determined by analyzing longitudinal data with a mixed-effects model. Solid lines represent best fit regression lines estimating the rates of decline of memory B cell numbers over time and the dashed lines represent the 95% CI.

Figure 6. Correlation between ELISA and ELISPOT responses for each antigen.

For malaria antigens, data are shown for Rural 2 subjects: 33 tested against P. falciparum antigens (A–D) and 26 tested against P. vivax antigens (E and F). For TT (G), data are shown for all subjects (City naive, Rural 1 and Rural 2) whose PBMC were available (n = 67). The number (and percentages) of subjects who were double positive (top left), ELISA positive but ELISPOT negative (bottom left), ELISA negative but ELISPOT positive (top right), or double negative (bottom right) are shown.