Molecular basis of severe malaria

Malaria is an ancient scourge that continues to plague many regions of the developing world. Despite its long history, many aspects of the disease remain difficult to explain. For example, why some individuals experience relatively mild symptoms and others progress to severe and sometimes lethal disease is not understood. Similarly, the molecular basis for immunity to severe malaria is poorly defined. Three studies in PNAS (1–3) cast significant light on this subject.

Malaria parasites reside within circulating erythrocytes, where they modify the host cell cytoskeleton and membrane, placing a protein called Plasmodium falciparum erythrocyte membrane protein 1 (PfEMP1) on the cell surface (4). This protein binds to receptors on the surface of vascular endothelial cells, thereby sequestering the infected cells within various tissues. This property can have severe consequences for the human host, causing inflammation, circulatory obstruction, and organ-specific damage (4). Infected individuals readily make antibodies that efficiently recognize PfEMP1, disrupting adhesion and leading to destruction of the infected erythrocytes. To avoid this fate, parasites have evolved a multicopy gene family called var. Each var gene encodes a variant form of PfEMP1 (5–7). By switching expression between different var genes, parasites can avoid the host’s antibody response and establish chronic infection. Significant work has demonstrated that different forms of PfEMP1 display different adhesive properties (8), leading to a simple paradigm: The organs in which infected cells sequester and, consequently, the severity of the disease are determined by which var genes are expressed (Fig. 1). Further, if specific PfEMP1 types are associated with severe disease, it might be possible to develop intervention strategies that specifically target these PfEMP1 types. This paradigm was validated with the discovery of a specific var gene that confers adhesion of infected cells within the placenta of pregnant women (9, 10). With the identification of a placenta-specific endothelial receptor (chondroitin sulfate A) (11) that is bound by a single PfEMP1 type (called VAR2CSA) (9, 10), the molecular interaction responsible for placental sequestration was illuminated. Subsequent studies showed that women who have suffered from placental malaria develop antibodies against VAR2CSA and appear to be immune to similar infections during subsequent pregnancies (12, 13), thus providing a strong rationale for development of a VAR2CSA-based vaccine.

Fig. 1.

P. falciparum-infected erythrocytes adhere within the microcapillary beds of various tissues and organs, causing a range of symptoms and disease severity. Parasites can express different types of PfEMP1, which bind to alternative host endothelial receptors, thereby determining where in the body the infected erythrocytes sequester and the severity of disease.

The success in identifying the basis for pregnancy-associated malaria led researchers to search for similar PfEMP1/host receptor interactions underlying other severe malaria syndromes, particularly cerebral malaria. Different parasite isolates often carry almost completely different var gene repertoires; thus, within any geographical region, the variability within the var gene family is extensive (14). A notable exception is VAR2CSA, which is largely conserved. The extreme diversity of PfEMP1 forms makes identification of specific types associated with severe disease or cerebral malaria difficult. However, computational analysis of var gene sequences did enable the gene family to be divided into three basic types referred to as A, B, and C groups (15, 16). Subsequent field studies identified expression of group A var genes as being more frequently associated with severe disease (17, 18).

The three reports in PNAS (1–3) identify a specific class of PfEMP1 that is associated with the incidence of cerebral malaria and severe disease. The studies by Avril et al. (1) and Claessens et al. (2) identify narrow subsets of group A var genes that were expressed by parasites selected to adhere with high affinity to endothelial cells derived from human brain tissue. When the encoded PfEMP1s were classified according to their domain architecture, they were found to possess one of two specific combinations of binding domain cassettes, referred to as DC8 or DC13, at their N-terminal ends (19). Interestingly, parasites expressing these var genes also bound to endothelial cells derived from nonbrain tissues; however, they did not bind intracellular adhesion molecule 1 (ICAM1), a host surface molecule previously proposed to be the endothelial receptor bound by infected erythrocytes in the brain. Both studies also found that serum from African children who had experienced severe malaria could recognize the selected parasite lines and disrupt binding to brain endothelial cells. PfEMP1 expressed by field isolates that cause severe malaria thus shares B-cell epitopes with laboratory strains selected for binding to brain endothelium. Limited diversity in B-cell epitopes in PfEMP1 expressed by parasite strains that cause severe disease suggests that it may be possible to elicit antibody responses that protect against severe malaria. These observations provide a rationale for the development of vaccines to protect against severe malaria.

Using a different approach, the study by Lavstsen et al. (3) provides complementary data. Working with field samples, the authors compared var gene transcripts expressed by parasites isolated from patients suffering from either severe or uncomplicated malaria. Similar to the studies of Avril et al. (1) and Claessens et al. (2), they identify group A var genes with the DC8 or DC13 architecture as being associated with severe disease. Interestingly, they found that both cerebral malaria and severe anemia were associated with expression of DC8 or DC13 encoding var genes. Together, these three studies suggest that conserved domains constituting DC8 or DC13 should be considered as the basis for syndrome-specific vaccines to protect against severe malaria.

The next step is to identify the host endothelial receptors to which this limited class of PfEMP1 molecules bind. Avril et al. (1) tested for binding to many of the known endothelial receptors previously identified as adhesive targets of infected erythrocytes; however, none of these appear to be involved. Identification of the host receptor(s) will provide a more complete picture of the molecular interactions responsible severe malaria and elevate our understanding of cerebral malaria to that previously achieved for pregnancy-associated disease. Identification of the endothelial receptor may also enable development of small molecule inhibitors to block the interaction and prevent severe malaria. It will also be critical to determine if antibodies raised against recombinant DC8 or DC13 recognize diverse P. falciparum isolates collected from patients with severe malaria. Such wide recognition of P. falciparum isolates capable of causing severe malaria may provide an approach to develop a vaccine to protect against severe malaria. Together, these studies mark a significant advance in our comprehension of the molecular basis of malaria pathogenesis and provide avenues for the development of novel therapeutic or prophylactic strategies against severe P. falciparum malaria.