Antibodies to a full-length VAR2CSA immunogen are broadly strain-transcendent but do not cross-inhibit different placental-type parasite isolates

Abstract

The high molecular weight, multidomain VAR2CSA protein mediating adhesion of Plasmodium falciparum-infected erythrocytes in the placenta is the leading candidate for a pregnancy malaria vaccine. However, it has been difficult so far to generate strong and consistent adhesion blocking antibody responses against most single-domain VAR2CSA immunogens. Recent advances in expression of the full-length recombinant protein showed it binds with much greater specificity and affinity to chondroitin sulphate A (CSA) than individual VAR2CSA domains. This raises the possibility that a specific CSA binding pocket(s) is formed in the full length antigen and could be an important target for vaccine development. In this study, we compared the immunogenicity of a full-length VAR2CSA recombinant protein containing all six Duffy binding-like (DBL) domains to that of a three-domain construct (DBL4-6) in mice and rabbits. Animals immunized with either immunogen acquired antibodies reacting with several VAR2CSA individual domains by ELISA, but antibody responses against the highly conserved DBL4 domain were weaker in animals immunized with full-length DBL1-6 recombinant protein compared to DBL4-6 recombinant protein. Both immunogens induced cross-reactive antibodies to several heterologous CSA-binding parasite lines expressing different VAR2CSA orthologues. However, antibodies that inhibited adhesion of parasites to CSA were only elicited in rabbits immunized with full-length immunogen and inhibition was restricted to the homologous CSA-binding parasite. These findings demonstrate that partial and full-length VAR2CSA immunogens induce cross-reactive antibodies, but inhibitory antibody responses to full-length immunogen were highly allele-specific and variable between animal species.

Figure 1. Immunization with DBL4-6 and DBL1-6 VAR2CSA recombinant proteins induces antibodies which target multiple domains in VAR2CSA.

Mouse and rabbit plasma were screened by ELISA against different VAR2CSA recombinant proteins. (A) Rabbit and mouse plasma against the full-length DBL1-6 recombinant protein were screened against individual VAR2CSA domains, (B) rabbit and mouse plasma against DBL4-6 recombinant protein were screened against individual VAR2CSA domains. The end point titration of the corresponding preimmune plasma in ELISA ranged from 0 to 200 at OD 0.1. R1, rabbit #1; R2, rabbit #2; M1, mouse #1; M2, mouse 2; M3, mouse #3.

Figure 2. Immunization with DBL4-6 and DBL1-6 VAR2CSA recombinant proteins induces antibodies against the surface of IEs-CSA.

Fluorescence of IEs labeled by rabbit plasma raised against 3D7 DBL1-6 (A) or 3D7 DBL4-6 (B) or mouse plasma raised against 3D7 DBL1-6 (C) or 3D7 DBL4-6 (D) is shown compared to the corresponding preimmune plasma (shaded histograms). In (A) and (B), blue histograms represent labeling with polyclonal plasma, green histograms show the corresponding purified IgG. In (C) and (D), individual mouse plasma were tested; blue (mouse 1), green (mouse 2), and orange (mouse 3).

Figure 3. Antibodies against the full length protein inhibit 3D7 DBL1-6 VAR2CSA recombinant protein binding to CSA-containing proteoglycans.

Purified antibodies were assessed for their capacity to inhibit protein binding to the CSA-containing proteoglycan decorin. Protein A purified rabbit IgG were preincubated at different concentrations with full-length recombinant VAR2CSA protein prior to addition to wells containing the CSA-containing proteoglycan decorin. Percent inhibition was calculated relative to the corresponding control protein A purified IgG raised against varO NTS-DBL1 domain. R1, rabbit #1; R2, rabbit #2.

Figure 4. Immunization with DBL1-6 VAR2CSA recombinant protein induces strain-specific adhesion blocking antibodies.

(A) Adhesion blocking properties of individual mouse plasma (1∶5 dilution) or pooled mouse plasma (1∶5 dilution) were tested against the 3D7csa and 7G8csa parasite lines. (B) Adhesion blocking properties of immune plasma (1∶5 dilution) or purified IgG (0.5 mg/ml) from rabbit #1 immunized with the DBL4-6 (grey bars) or DBL1-6 recombinant proteins (black bars) were tested against several different CSA-binding parasite lines and two CD36-binding negative control parasite lines. (C) Dose-dependency adhesion inhibition properties of rabbit #2 anti-3D7 DBL1-6 plasma and purified IgG tested against 3D7csa parasites. Rabbit #2 anti-3D7 DBL1-6 plasma (1∶5 dilution) and purified IgG (0.5 mg/ml) adhesion blocking properties were tested on four heterologous CSA-binding parasite lines and a CD36-binding negative control parasite line. For antibody adhesion inhibition, CSA-binding parasite lines were tested for adhesion to CSA and CD36 binding parasite lines were tested for adhesion to CD36. Percent inhibition was calculated relative to the corresponding preimmune rabbit plasma/purified IgG or a pool of three preimmune mouse plasma. Adhesion inhibition is the mean of 4 independent spots for all assays.