Nonimmune immunoglobulin binding and multiple adhesion characterize Plasmodium falciparum-infected erythrocytes of placental origin

Abstract

The harmful effects of pregnancy-associated malaria (PAM) are engendered by the heavy sequestration of Plasmodium falciparum-parasitized RBCs in the placenta. It is well documented that this process is mediated by interactions of parasite-encoded variant surface antigens and placental receptors. A P. falciparum erythrocyte membrane protein 1 variant, VAR2CSA, and the placental receptor chondroitin sulfate A (CSA) are currently the focus of PAM research. A role for immunoglobulins (IgG and IgM) from normal human serum and hyaluronic acid as additional receptors in placental sequestration have also been suggested. We show here (i) that CSA and nonimmune IgG/IgM binding are linked phenotypes of in vitro-adapted parasites, (ii) that a VAR2CSA variant shown to bind CSA also harbors IgG- and IgM-binding domains (DBL2-X, DBL5-epsilon, and DBL6-epsilon), and (iii) that IgG and IgM binding and adhesion to multiple receptors (IgG/IgM/HA/CSA) rather than the exclusive binding to CSA is a characteristic of fresh Ugandan placental isolates. These findings are of importance for the understanding of the pathogenesis of placental malaria and have implications for the ongoing efforts to develop a global PAM vaccine.

Fig. 1.

Assessment of the nonimmune Ig-binding properties of 3D7 VAR2CSA domains by FACS. (A) Domain structure of 3D7 VAR2CSA. IgG- and IgM-binding domains as well as previously identified CSA-binding domains are indicated. (B) Data shown are flow-cytometric analyses of IgG- and IgM-binding properties of CHO-745 transfectants expressing various domains of 3D7var2CSA demonstrated as overlay histograms. Data are representative of more than three experiments. (C) The mean level of domain expression of each transfectant during all experimental runs is summarized and expressed as mean fluorescence intensity (MFI). In each transfectant, >90% of the cells stably expressed their respective domains (data not shown). ∗, CSA-binding domains were previously identified by Gamain et al. (17) using CHO-745 transfectants. APC-SA, allophycocyanin–streptavidin.

Fig. 2.

Receptor-adhesion profile of Ugandan P. falciparum placental isolates. Freshly eluted pRBCs from infected placentas (n = 15) were examined for binding to various receptors by using ex vivo placental adhesion/inhibition assays. Binding to CSA and human IgG was assessed by preincubating the pRBCs with saturating concentrations of the soluble receptors before placental binding. The involvement of HA was assessed by enzymatically removing the receptor subset from the placental section before pRBC adhesion using Streptomyces HAse. (A) The diagram illustrates the level of pRBC adhesion to placental section for each isolate after treatment with IgG alone, CSA alone, a combination of both (IgG plus CSA), HAse alone, and a combination of all three treatments (IgG plus CSA plus HAse). Adhesion level is expressed as the percentage of positive control (no treatment). Each isolate was tested in duplicate or triplicate for each receptor or receptor combination. (B) The consensus receptor-adhesion profile of each isolate is indicated. ∗, Not tested for the IgG plus CSA plus HAse combination.

Fig. 3.

Nonimmune IgG- and IgM-binding profile of fresh Ugandan placental isolates (n = 22) assessed by live SIFA. Each data point represents the Ig-, IgG-, or IgM-binding level of one placental isolate. Each placental isolate was tested directly upon elution from placenta (Pre) and after incubation with malaria naïve serum (Post) by using anti-Ig, -IgG, or -IgM antibodies. ∗, pairwise comparison of Ig-binding level of each isolate before vs. after serum, P < 0.05.

Fig. 4.

Ig-binding profile of fresh placental isolates. Correlation between data obtained with two different assays, ex vivo placental adhesion/inhibition (IgG) and live SIFA (Ig, IgM, and IgG).