A full-length recombinant Plasmodium falciparum PfRH5 protein induces inhibitory antibodies that are effective across common PfRH5 genetic variants

Abstract

The lack of an effective licensed vaccine remains one of the most significant gaps in the portfolio of tools being developed to eliminate Plasmodium falciparum malaria. Vaccines targeting erythrocyte invasion - an essential step for both parasite development and malaria pathogenesis - have faced the particular challenge of genetic diversity. Immunity-driven balancing selection pressure on parasite invasion proteins often results in the presence of multiple, antigenically distinct, variants within a population, leading to variant-specific immune responses. Such variation makes it difficult to design a vaccine that covers the full range of diversity, and could potentially facilitate the evolution of vaccine-resistant parasite strains. In this study, we investigate the effect of genetic diversity on invasion inhibition by antibodies to a high priority P. falciparum invasion candidate antigen, P. falciparum Reticulocyte Binding Protein Homologue 5 (PfRH5). Previous work has shown that virally delivered PfRH5 can induce antibodies that protect against a wide range of genetic variants. Here, we show that a full-length recombinant PfRH5 protein expressed in mammalian cells is biochemically active, as judged by saturable binding to its receptor, basigin, and is able to induce antibodies that strongly inhibit P. falciparum growth and invasion. Whole genome sequencing of 290 clinical P. falciparum isolates from across the world identifies only five non-synonymous PfRH5 SNPs that are present at frequencies of 10% or more in at least one geographical region. Antibodies raised against the 3D7 variant of PfRH5 were able to inhibit nine different P. falciparum strains, which between them included all of the five most common PfRH5 SNPs in this dataset, with no evidence for strain-specific immunity. We conclude that protein-based PfRH5 vaccines are an urgent priority for human efficacy trials.

Fig. 1

Surface plasmon resonance demonstrates biochemical activity of recombinant full-length 7G8 PfRH5-Cd4. (A) 7G8 PfRH5-Cd4-biotin was specifically captured in a flow cell of a streptavidin-coated chip using an approximate molar equivalent of Cd4-biotin immobilized in a neighbouring flow cell used as a reference. Serial dilutions of purified basigin-Cd4-6H were injected until equilibrium was reached (solid bar, inset) and reference-subtracted data were used to plot the binding responses once equilibrium had been achieved. An equilibrium dissociation constant was calculated using non-linear regression fitting of a simple Langmuir binding isotherm to the data (solid line). Note that the binding was saturable and therefore specific and a K_D of 0.9 ± 0.13 μM (mean ± SEM) was calculated from two independent experiments (see Supplementary Table 2). A representative experiment using 7G8 is shown, equilibrium constants for 3D7 and GB4 were calculated in the same way and are reported in Supplementary Table 2. (B) 3D7 PfRH5-Cd4-d3+4-6H was purified using Ni²⁺-NTA-Sepharose and resolved as two bands (∼85 and 65 kDa) using using Coomassie staining after SDS-PAGE under reducing conditions. The 85 kDa protein, which is consistent with the expected size of the full-length PfRH5-Cd4-d3+4-6 H protein, could be separated from the smaller 65kDa band using size exclusion chromatography (post-SEC).

Fig. 2

Immunoreactivities to recombinant PfRH5 7G8 and GB4 are indistinguishable from the 3D7 variant used as an antigen. (A) The 3D7, 7G8 and GB4 PfRH5 variants (colored lines, as indicated in the key) were expressed as Cd4-biotin tagged recombinant proteins, normalized, and specifically captured on a streptavidin-coated microtitre plate. Each variant was equally immunoreactive to a polyclonal antiserum that was raised using the 3D7 variant. Immunoreactivity to the Cd4 tag was removed by preadsorption of the antiserum against Cd4; the preadsorbed antiserum was shown to lack anti-Cd4 immunoreactivity (black line). (B) Recombinant PfRH5 contains heat-labile epitopes. The 3D7 variant of PfRH5-Cd4-biotin was again captured on a streptavidin-coated plate either without treatment or after denaturation by heat (80 °C, 10 min). The two proteins were tested for their immunoreactivity to the anti-RH5 polyclonal antiserum that had been preadsorbed against Cd4 (shown as a control). Data points are means; error bars represent SEM; n = 3.

Fig. 3

Polyclonal antibodies raised against 3D7 PfRH5 inhibit both 3D7 and FCR1 P. falciparum parasites. (A) Total IgG purified from a rabbit immunized with the 3D7 variant of PfRH5 was used in invasion assays with the 3D7 and FCR1 P. Falciparum strains. Growth inhibition was observed to a similar extent for both strains with an IC50 ratio for FCR1 of 1.6 (95% CI 1.17–2.18). (B) Total IgG purified from a rabbit immunized with the 3D7 AMA variant was used in erythrocyte invasion assays with the 3D7 or FCR1 P. Falciparum strains and showed strain-specific growth inhibition. (C) Total IgG purified from unimmunized rabbits was used in erythrocyte invasion assays with the 3D7 or FCR1 P. falciparum strains and had no inhibitory effect on erythrocyte invasion. Each IgG concentration/strain combination was assayed in three wells in a 96 well plate; data points are means of all three wells, with standard deviation error bars shown.

Fig. 4

Polyclonal antibodies raised against 3D7 PfRH5 inhibit P. falciparum strains that collectively include all common PfRH5 variants. (A) Total IgG purified from a rabbit immunized with the 3D7 variant of PfRH5 was used in invasion assays with Dd2, 7G8, GB4 and 3D7P. falciparum strains. IC50 ratios were 0.5, 1.0 and 1.4 for Dd2, 7G8 and GB4, respectively. 95% CI for IC50s were 0.47–0.63 (Dd2), 0.71–1.36 (7G8) and 1.18–1.71 (GB4). (B) Total IgG purified from a rabbit immunized with the 3D7 variant of PfRH5 was used in erythrocyte invasion assays with two P. falciparum isolates from Kenya (K7336 and KCC103), and the 3D7 P. falciparum strain. IC50 ratios were 1.1 and 0.5 for K7336 and KCC103, respectively. 95% CI for IC50s were 0.81–1.42 (K7336) and 0.35–0.76 (KCC103). (C) Total IgG purified from a rabbit immunized with the 3D7 variant of PfRH5 was used in erythrocyte invasion assays with two P. falciparum isolates from Cambodia (PH21 and PH22) and the 3D7 P. falciparum strain. IC50 ratios were 0.8 and 1.3 for PH21 and PH22 respectively. 95% CI for IC50s were 0.59–1.35 (PH21) and 0.93–1.76 (PH22). Each IgG concentration/strain combination was assayed in three wells in a 96 well plate; data points are means of all three wells, with standard deviation error bars shown.