Large scale, international public health efforts to eliminate smallpox and malaria were undertaken after World War II. Although smallpox was successfully eradicated, malaria prevailed. A staggering 40% of the world lives at risk of infection, and more than a million people die of the disease each year; most are children under the age of five. Intensive efforts to develop a vaccine are underway, but treatment of malaria at present depends entirely on effective chemotherapy. Unfortunately, only a limited number of drugs is presently available, and resistance has emerged to all of them. The identification of promising new targets and antimalarials is recognized to be of critical importance by the international agencies committed to controlling malaria.

In the August Issue of Nature Medicine, Bonday et al. presented evidence that a host cell enzyme involved in heme biosynthesis (d-aminolevulinate dehydratase; ALAD) may be a new antimalarial target (1). At first glance, it is surprising that a host enzyme could be an effective target for targeting the malaria parasite. But drugs may be developed to target not the metabolic pathway itself, but the process by which this enzyme (and other macromolecules) moves between the host and parasite within the infected red cell.

Malaria parasites belong to the genus Plasmodium, phylum Apicomplexa. The blood stages of infection (responsible for all of the symptoms and pathologies associated with the disease) develop within a specialized, membrane bounded vacuole within the red cell. Parasites are therefore isolated from the erythrocyte cytoplasm by at least two membranes (Fig. 1). Despite being surrounded by a sea of host hemoglobin which it ingests and digests, the parasite is dependent on de novo heme synthesis. Bonday et al. have previously argued that Plasmodium imports host ALAD to facilitate this process (2), although it is also worth noting that sequences emerging from the P. falciparum genome sequencing project (3) include several putative parasite-encoded heme biosynthetic enzymes, including ALAD (see http://www.PlasmoDB.org). Whether these genes are expressed during blood stage infection — and the relative contribution of parasite and host enzymes to the total ALAD activity in the infected red cell — remains unknown. The present report identifies a protein from infected erythrocytes which binds to host ALAD and may function as a receptor mediating transport into the parasite.

The major pathway of macromolecular import from the cytoplasm of the red cell to the parasite occurs via the cytostome (4). This structure is formed by invagination of the parasitophorous vacuolar membrane (PVM) and parasite plasma membrane, and results in uptake of hemoglobin and a portion of the vacuole (Fig. 1, pathway 1). The resulting double membrane vesicle is delivered to the food vacuole. Here cargo proteins are degraded into peptides that may be subsequently translocated to the parasite cytoplasm. Whether and how ALAD is translocated to the parasite cytoplasm without being degraded in the food vacuole are open questions. An alternative pathway might involve translocation of ALAD across the PVM and parasite plasma membrane (Fig. 1, pathway 2). The PVM in Plasmodium and the related apicomplexan Toxoplasma gondii is permeant to small molecules, but the size exclusion limit is well below the size of the ALAD complex (5,6). To date there is no evidence for direct translocation of macromolecules across the parasite plasma membrane.

The most intriguing aspect of the work of Bonday et al. is evidence that fragments of recombinant ALAD added to the culture medium can enter infected red cells and inhibit intracellular replication of the human malaria parasite P. falciparum. The mechanism by which molecules enter into these cells (Fig. 1, pathway 3), especially given that parasite-infected erythrocytes are non-endocytic, has been area of intense study and debate. An early study suggested that extracellular fluorescent dextrans and immunoglobulins enter through tubular 'ducts' (Fig. 1, pathway 4) (ref. 7). These initial observations stimulated considerable controversy and may have been flawed (8,9), but there is now firm evidence that some macromolecules can enter the vacuole and reach the malaria parasite.

Membranous structures that extend from the PVM to the red cell membrane have long been observed in infected erythrocytes (10), and parasite-induced solute transport across the infected red cell membrane, intraerythrocytic tubovesicular membranes (TVM) and the PVM is well documented (11,12). But mechanisms responsible for the import of macromolecules remain elusive. This problem has been difficult to investigate due to the failure to detect host-derived integral membrane proteins in the PVM. The entry of lipids and glycosylphosphatidylinositol-anchored proteins into the PVM of Plasmodium and T. gondii led to suggestions that transmembrane host proteins may be excluded (13). However a recent study shows that transmembrane receptors and signaling proteins that associate with detergent-resistant membranes (a characteristic of microdomains or 'rafts' in the host cell membrane) can access the Plasmodium vacuole (14). It is possible that macromolecules associated with microdomain components may have specialized access to the malaria parasite.