Guiding the Immune Response to a Conserved Epitope in MSP2, an Intrinsically Disordered Malaria Vaccine Candidate

Abstract

The malaria vaccine candidate merozoite surface protein 2 (MSP2) has shown promise in clinical trials and is in part responsible for a reduction in parasite densities. However, strain-specific reductions in parasitaemia suggested that polymorphic regions of MSP2 are immuno-dominant. One strategy to bypass the hurdle of strain-specificity is to bias the immune response towards the conserved regions. Two mouse monoclonal antibodies, 4D11 and 9H4, recognise the conserved C-terminal region of MSP2. Although they bind overlapping epitopes, 4D11 reacts more strongly with native MSP2, suggesting that its epitope is more accessible on the parasite surface. In this study, a structure-based vaccine design approach was applied to the intrinsically disordered antigen, MSP2, using a crystal structure of 4D11 Fv in complex with its minimal binding epitope. Molecular dynamics simulations and surface plasmon resonance informed the design of a series of constrained peptides that mimicked the 4D11-bound epitope structure. These peptides were conjugated to keyhole limpet hemocyanin and used to immunise mice, with high to moderate antibody titres being generated in all groups. The specificities of antibody responses revealed that a single point mutation can focus the antibody response towards a more favourable epitope. This structure-based approach to peptide vaccine design may be useful not only for MSP2-based malaria vaccines, but also for other intrinsically disordered antigens.

Keywords: disordered protein; malaria; merozoite surface protein 2; peptide vaccines; structural vaccinology.

Figure 1

SPR competition assay using alanine-scan 16-residue peptides of 3D7-MSP2_207–222 with (A) mAb 9H4 and (B) mAb 4D11. Schematic representation of disulfide-bonded epitope sequence for both (C) 9H4 and (D) 4D11 indicating the location of key residues. Black indicates alanine mutants with no effect on binding, the green shows the wild-type binding of 3D7-MSP2_207–222, red indicates the alanine mutations that decreased the binding and blue indicates mutations that increased binding.

Figure 2

(A) Schematic representation of dimeric 4D11 epitope peptides: dimer—D1, linear—L1, backbone cyclised—C1, wild-type sequence 17-residue peptide—A1 and its single point mutant—K209A, which has had the residue Lys209, which is important for 9H4 binding, replaced with alanine. (B) Crystal structure of 4D11-bound homodimer peptide (PDB ID: 5TBD) shows close proximity between the N- and C-termini of the peptides; green dashed lines indicate distance between the peptide termini.

Figure 3

(A) Direct ELISA indicates that all peptide-KLH conjugates are recognised by mAb 4D11, (B) K209A-KLH is unable to bind to mAb 9H4, indicating that the 9H4 epitope had been removed successfully.

Figure 4

Mouse sera, taken two weeks after final immunisation with antigen, contain antibodies that recognise their corresponding peptide-BSA conjugate or in the case of the MSP2 mix, recombinant 3D7 and FC27 MSP2 coated on the ELISA plate. Each point is the mean of two replicate wells for individual mouse sera. The line indicates group median. Endpoint titre was calculated using a cut-off value of three standard deviations greater than OD₄₀₅ for naïve mouse serum (~0.09).

Figure 5

Antibody specificities of individual mice from each group were determined by peptide array (peptides A–I), numbers 1677–1712 indicate individual mice.

Figure 6

Merozoite ELISA using pooled mouse sera for each immunogen against (A) FC27 merozoites and (B) 3D7 merozoites.