Basigin is a receptor essential for erythrocyte invasion by Plasmodium falciparum

Abstract

Erythrocyte invasion by Plasmodium falciparum is central to the pathogenesis of malaria. Invasion requires a series of extracellular recognition events between erythrocyte receptors and ligands on the merozoite, the invasive form of the parasite. None of the few known receptor-ligand interactions involved are required in all parasite strains, indicating that the parasite is able to access multiple redundant invasion pathways. Here, we show that we have identified a receptor-ligand pair that is essential for erythrocyte invasion in all tested P. falciparum strains. By systematically screening a library of erythrocyte proteins, we have found that the Ok blood group antigen, basigin, is a receptor for PfRh5, a parasite ligand that is essential for blood stage growth. Erythrocyte invasion was potently inhibited by soluble basigin or by basigin knockdown, and invasion could be completely blocked using low concentrations of anti-basigin antibodies; importantly, these effects were observed across all laboratory-adapted and field strains tested. Furthermore, Ok(a-) erythrocytes, which express a basigin variant that has a weaker binding affinity for PfRh5, had reduced invasion efficiencies. Our discovery of a cross-strain dependency on a single extracellular receptor-ligand pair for erythrocyte invasion by P. falciparum provides a focus for new anti-malarial therapies.

Figure 1. BSG is an erythrocyte receptor for PfRh5

(a) PfRh5 was screened as either a prey (top panel) or a bait (bottom panel) against an erythrocyte receptor protein library using AVEXIS. BSG (protein 9) was identified as a receptor for PfRh5 in both bait-prey orientations. (b) Domain structure of the BSG isoforms (left); lollipops represent potential N-linked glycosylation sites. BSG regions were expressed as baits and used to map the PfRh5 binding site to the two membrane-proximal domains. Bar charts show mean ± SEM; n = 3. (c) Biophysical analysis of the PfRh5-BSG-S interaction using SPR. The indicated concentrations of purified PfRh5 were injected over immobilised BSG, and biophysical parameters derived from a 1:1 binding model (red line).

Figure 2. Soluble BSG, anti-BSG antibodies and BSG knockdown potently block erythrocyte invasion

(a) Erythrocyte invasion was inhibited by purified pentamerised BSG-S-Cd4d3+4-COMP-His ectodomains but not by the two non-binding BSG-S domains added individually or Cd4d3+4-COMP-His (control); strain = Dd2. (b) Cross-strain inhibition of invasion using pentamerised BSG-S. (c) Anti-BSG monoclonal antibodies, TRA-1-85 and MEM-M6/6, potently inhibited invasion of erythrocytes; strain = 3D7. (d) MEM-M6/6 concentrations ≥10 μg/ml prevented all detectable invasion by microscopic observation of cultures; strain = 3D7. (e, f) MEM-M6/6 inhibited invasion of synchronised P. falciparum culture-adapted lines (e) and unsynchronised field isolates (f). (g) Cell surface BSG is reduced in erythrocytes differentiated from hematopoietic stem cells transduced with lentiviruses containing shRNA targeting BSG (light blue line) relative to a control virus (pLKO, shaded); black line represents secondary antibody alone. (h) 3D7 and W2mef invasion was inhibited in BSG knockdown erythrocytes. A and B are replicates. Invasion efficiencies are mean ± SEM, n = 3.

Figure 3. The Ok(a⁻) BSG variant has reduced binding affinity for PfRh5 and Ok(a⁻) erythrocytes have reduced merozoite invasion frequencies

(a) Schematic of the membrane distal IgSF domain of BSG-S showing the location of naturally-occurring variants. (b) Equilibrium binding isotherms of PfRh5 binding to BSG-S variants. (c) Association (k_a) and dissociation (k_d) rate constants of PfRh5 binding to BSG-S and variants. Means ± SEM; n = 3. (d) Invasion of 3D7 and Dd2 strains in Ok(a⁻) blood cells are reduced relative to the Ok(a⁺) control. Mean ± SEM, n = 3; *, P ≤ 0.0003; #, P = 0.0349, unpaired one-tailed t test. A repeat is shown in Supplementary Fig. 9.