Multiple human serum components act as bridging molecules in rosette formation by Plasmodium falciparum-infected erythrocytes

Rosetting, the binding of parasitized erythrocytes to 2 or more uninfected erythrocytes, is an in vitro correlate of disease severity in Plasmodium falciparum malaria. Although cell ligands and receptors have been identified and a role for immunoglobulin M has been suggested, the molecular mechanisms of rosette formation are unknown. The authors demonstrate unequivocally that rosette formation by P falciparum-infected erythrocytes is specifically dependent on human serum, and they propose that serum components act as bridging molecules between the cell populations. Using heparin treatment and Percoll density gradient centrifugation, they have developed an assay in which parasitized erythrocytes grown in serum-containing medium and optimally forming rosettes are stripped of serum components. These infected cells were no longer able to form rosettes when mixed with erythrocytes and incubated in serum-free medium. Rosette formation was restored by the addition of serum or certain serum fractions obtained by concanavalin A (conA) affinity, anti-IgM affinity, anion exchange, and gel filtration chromatography. The authors clearly demonstrate that multiple serum components-IgM and at least 2 others-are involved in rosette formation. Those others consist of 1 or more acidic components of high-molecular mass that binds to conA (but that is not thrombospondin, fibronectin, or von Willebrand's factor) and of at least 1 more basic, smaller component that does not bind to conA. Data on the size and number of rosettes formed support the authors' hypothesis that multiple bridges are involved in this complex cellular interaction. These findings have important implications for the understanding of pathogenic adhesive interactions of P falciparum and host susceptibility to severe malaria. (Blood. 2000;95:674-682)