Naturally acquired IgG responses to Plasmodium falciparum do not target the conserved termini of the malaria vaccine candidate Merozoite Surface Protein 2

Abstract

Introduction: Malaria remains a significant burden, and a fully protective vaccine against Plasmodium falciparum is critical for reducing morbidity and mortality. Antibody responses against the blood-stage antigen Merozoite Surface Protein 2 (MSP2) are associated with protection from P. falciparum malaria, but its extensive polymorphism is a barrier to its development as a vaccine candidate. New tools, such as long-read sequencing and accurate protein structure modelling allow us to study the genetic diversity and immune responses towards antigens from clinical isolates with unprecedented detail. This study sought to better understand naturally acquired MSP2-specific antibody responses.

Methods: IgG responses against recombinantly expressed full-length, central polymorphic regions, and peptides derived from the conserved termini of MSP2 variants sequenced from patient isolates, were tested in plasma from travelers with recent, acute malaria and from individuals living in an endemic area of Tanzania.

Results: IgG responses towards full MSP2 and truncated MSP2 antigens were variant specific. IgG antibodies in the plasma of first-time infected or previously exposed travelers did not recognize the conserved termini of expressed MSP2 variants by ELISA, but they bound 13-amino acid long linear epitopes from the termini in a custom-made peptide array. Alphafold3 modelling suggests extensive structural heterogeneity in the conserved termini upon antigen oligomerization. IgG from individuals living in an endemic region, many who were asymptomatically infected, did not recognize the conserved termini by ELISA.

Discussion: Our results suggest that responses to the variable regions are critical for the development of naturally acquired immunity towards MSP2.

Keywords: antibody response; immune evasion; malaria vaccine; merozoite; natural immunity; polymorphic antigens; structural heterogeneity.

Figure 1

IgG responses towards full MSP2 variants by ELISA. Normalized absorbance of samples with detected IgG responses to at least one of the 10 tested antigens are shown in raincloud plots for Travelers (A), Tanzania 1994 (B) and Tanzania 1999 (C). Travelers are grouped by level of previous exposure to P. falciparum malaria and the Tanzania cohort is grouped by age. Boxplot lines denote median and CI.25-.75. IgG responses in normalized absorbance towards each of four MSP2 variants are shown for the same groups of travelers (D) and Tanzania samples (E). Normalized absorbance values were compared using unpaired Wilcoxon test where significance values are as follows: * < 0.05, ** < 0.01, *** < 0.001, **** < 0.0001. ns, non-significant.

Figure 2

Increasing breadth towards full MSP2 variants is associated with increased IgG responses. Breadth for each individual sample was plotted, where samples were considered “binding” if they passed the positivity cutoff based on negative controls for travelers (A) and Tanzania samples (C). Boxplots denote CI 0.95 and mean breadth. IgG responses were plotted against breadth to full MSP2 variants for travelers (B) and Tanzania samples (D). Pairwise comparisons performed with Wilcoxon test. Significant differences are marked with asterisks (* < 0.05, ** < 0.01, *** < 0.001, **** < 0.0001). ns, non-significant.

Figure 3

IgG responses towards truncated versions of MSP2 antigens. IgG responses towards truncated MSP2 variants by ELISA. IgG responses in normalized absorbance are shown for travelers (A), as well as Tanzania participants in the 1994 (B) and 1999 (C) surveys, respectively. Traveler samples are grouped by level of previous exposure to P. falciparum malaria while Tanzania samples are grouped by age. IgG responses in normalized absorbance towards each of four MSP2 variants are shown for the same groups of travelers (D) and Tanzania samples (E). Normalized absorbance values were compared using unpaired Wilcoxon test where significance is marked as follows: * < 0.05, ** < 0.01, *** < 0.001, **** < 0.0001. ns, non-significant.

Figure 4

Increasing breadth towards truncated MSP2 variants is associated with increased IgG responses. Breadth for each individual was plotted, where samples were considered “binding” if they passed the positivity cutoff based on negative controls for travelers (A) and Tanzania samples (B). Boxplots denote 0.95 CI and mean breadth. IgG responses in normalized absorbance to all four truncated variants were plotted against increasing breadth to truncated MSP2 variants for travelers (C) and Tanzania samples (D). Pairwise comparisons performed with Wilcoxon test and significance is show with asterisks (* < 0.05, ** < 0.01, *** < 0.001, **** < 0.0001). ns, non-significant.

Figure 5

IIgG responses towards conserved MSP2 termini by ELISA. IgG responses to N and C terminal peptides, in normalized absorbance, are shown for travelers (A), Tanzania 1994 (B) and Tanzania 1999 (C), where travelers samples are grouped by level of previous exposure to P. falciparum malaria and Tanzania samples are grouped by age. IgG responses in normalized absorbance towards both peptides are shown for the same groups of travelers (D) and Tanzania samples (E). Normalized absorbance values were compared using unpaired Wilcoxon Test where significance values are as follows: * < 0.05, *** < 0.001. ns, non-significant.

Figure 6

Differences between IgG responses towards full and truncated MSP2 variants. IgG responses for each sample, indicated with connecting lines, were plotted against each antigen tested for primary infected and previously exposed travelers (A). Responses were also plotted for children and adults in the 1994 (B) and 1999 (C) surveys in Tanzania. Comparisons of mean normalized absorbance were performed by unpaired Wilcoxon test (* < 0.05, ** < 0.01). ns, non-significant.

Figure 7

Binding to N- and C-terminus 13mer peptides by four primary infected Travelers. Number of positive, overlapping 13-mer peptides were mapped to each amino acid residue of the N- and C- termini on the peptide array for each of four primary infected travelers (A). IgG responses in normalized absorbance for each traveler against all ten MSP2 antigens (B). Dashed line indicates global mean negative cutoff. Positive overlapping 13mer peptides for each amino acid of the N- and C-termini for each traveler mapped onto AlphaFold2 predictions of the full MSP2-FC27 4, MSP2-FC27 1, MSP2-IC 1, and MSP2-IC 2 antigens used in the ELISA assays (C).

Figure 8

Structural heterogeneity in the conserved termini of MSP2 is associated to its oligomerization status. SDS-PAGE for (A) recombinant proteins and (B) synthetic peptides used in the characterization of antibody responses against MSP2 in the presence or absence of the reducing agent DTT, respectively. (C) Alphafold3 models for monomers and oligomers of the native and engineered MSP2 variants used in the study. AF3 models for monomeric and oligomeric forms of the synthetic peptides corresponding to the conserved N and C termini are also presented. Predicted molecular weight and pTM scores are shown for every model. Truncated MSP2 variants are marked as “var”. Tags engineered into the variants before expression are shown in black. Colors represent pLDDT scores calculated for every residue in the model.