Loading antimalarial drugs into noninfected red blood cells: an undesirable roommate for Plasmodium

The malaria parasite, Plasmodium spp., is a delicate unicellular organism unable to survive in free form for more than a couple of minutes in the bloodstream. Upon injection in a human by its Anopheles mosquito vector, Plasmodium sporozoites pass through the liver with the aim of invading hepatocytes. Those which succeed spend inside their host cell a recovery time before replicating and entering the blood circulation as fragile merozoites, although their exposure to host defenses is extraordinarily short. Quick invasion of red blood cells (RBCs) in a process lasting just a few minutes allows the parasite to escape immune system surveillance. For most of its erythrocytic cycle the pathogen feeds mainly on hemoglobin as it progresses from the early blood stages, termed rings, to the late forms trophozoites and schizonts. Early stages are ideal targets for antimalarial therapies because drugs delivered to them would have a longer time to kill the parasite before it completes its development. However, only 6 h after invasion does the permeability of the infected erythrocyte to anions and small nonelectrolytes, including some drugs, start to increase as the parasite matures [1]. During this maturation process the parasite hydrolyzes hemoglobin in a digestive vacuole, which is the target of many amphiphilic drugs that freely cross the RBC membrane and accumulate intracellularly. As a result, most antimalarials start affecting the infected cell relatively late in the intraerythrocytic parasite life cycle, when their effect is probably often too short to be lethal to Plasmodium.

Malaria-infected erythrocytes: an elusive target

Several strategies to improve the activity of antimalarial drugs concern their encapsulation in nanocarriers targeted to parasitized RBCs (pRBCs), an approach that requires the existence of specific pRBC markers. 200-nm liposomes studded with heparin or antibodies raised against pRBCs have been shown to bind late forms with high selectivity [2,3], improving the activity of encapsulated antimalarial drugs up to tenfold [2,4]. In addition to the inconvenient late-stage targeting, such liposomal delivery models will also have to overcome the obstacle of timing nanocarrier administration to the precise moment of the parasite's life cycle when trophozoites and schizonts are present. The relatively short blood half-life of liposomes (in the best cases, <10 h for polyethylene glycol-coated stealth liposomes) guarantees that if injected at the wrong moment (too soon or too late), they will not last the 48 h needed to ensure that they are present for the pathogen's next cycle. In another display of cunningness, Plasmodium leaves virtually no external signal on the parasitized cell, and only after spending half its life inside the erythrocyte does the parasite export a significant number of receptors and transporters to the host cell plasma membrane. Most of these externally recognizable clues are present in the parasite genome as multiple variants that can be clonally expressed [5], which further complicates delivery approaches designed to specifically target pRBCs. A receptor-independent alternative for the nanovector-mediated delivery of antimalarial drugs to Plasmodium blood stages can be provided by the tubulovesicular network induced in the host cell by the pathogen during its intraerythrocytic growth, which confers pRBC accessibility to a wide range of particles up to diameters of 70 nm [6]. Indeed, polymeric nanovectors were observed to penetrate trophozoites and schizonts [7,8], possibly in a significant fraction through the tubulovesicular network, although entry of nanoparticles into early ring stages has not been unambiguously observed so far.

Is there an ideal carrier for blood-circulating drugs?

Antimalarial drug carriers should provide optimal compound half-lives in circulation, adequate clearance mechanisms, restriction of unintended drug effects in non-target cells, specific delivery to the correct tissue, and a timely initiation and termination of the therapeutic action. Considering the need to target intraerythrocytic Plasmodium as early in its life cycle as possible and the lack of strategies currently out there for shuttling drugs into pRBCs, it is imperative that these issues are addressed and that alternative approaches are explored. A solution to the aforementioned problems in the design of pRBC-targeted nanocarriers can perhaps be provided by one of the most adequate vascular carriers, RBCs themselves [9]. Human erythrocytes have a life span in the blood of up to 120 days, which makes them attractive carriers for intravascular delivery because they prolong drug circulation. In addition, their large size (approximately 7 μm across and around 2 μm thick) significantly restricts unintended extravasation and in principle allows for a much larger encapsulation capacity than liposomes. Other interesting features of RBCs as drug carriers are their biocompatibility and the existence of natural mechanisms for their safe elimination from the body. Actually, delivery of antimalarials to noninfected RBCs has been previously carried out in chemotherapeutic investigations, in order to examine the effects on later invading parasites. In one such study, RBCs were pretreated with the drugs halofantrine, lumefantrine, piperaquine, amodiaquine and mefloquine, which were observed to diffuse into and remain within the erythrocyte, inhibiting downstream growth of Plasmodium [10]. However, it should be noted that the loading of drugs into noninfected RBCs has not yet been explored in detail as a clinically feasible therapeutic strategy against malaria, in part because of a number of restrictions that must be taken into consideration.

Which are the limitations of erythrocytes as drug carriers?

A significant limiting factor for the use of RBCs as antimalarial carriers is that when present at therapeutically active concentration, the drug has to be innocuous for the cell physiology, which might not be an unsurmountable obstacle given the reduced metabolic activity of eryrthrocytes. However, loading of some antimalarial drugs like clotrimazole had been observed to predispose RBCs to oxidative damage [11], an undesirable scenario because oxidized RBCs are rapidly taken up by hepatic reticuloendothelial system macrophages. Another obstacle for the incorporation of antimalarial drugs into RBCs is drug loading itself, since most currently available protocols use a harsh ex vivo isolation of erythrocytes followed by drug loading through diffusion [9]. In a clinical setting, perhaps RBC-targeted immunoliposomes can come to rescue, although the incapacity of mature erythrocytes to endocyte [12] calls for the development of specific targeted drug delivery strategies independent from the receptor-mediated endocytic pathway. Moreover, the physicochemical properties of each particular antimalarial drug will constrain the nanovector composition and the corresponding drug delivery mechanism. As an example, the optimal approach for delivery of membrane-impermeable hydrophilic drugs such as fosmidomycin would be immunoliposomal fusion with the RBC membrane, which requires the incorporation of special fusogenic agents into highly fluid vesicles. Including negatively charged phospholipids in the liposome formulation has been found to be crucial for the delivery of trehalose into RBCs in vitro [13], but nanovector fusion can be inhibited by components found in plasma [14], and charged vesicles are quickly complexed by serum proteins that target them for clearance from circulation [15]. A possible solution consists of incorporating stealth agents onto the nanovector surface like polyethylene glycol chains or gangliosides, which neutralize vesicle charge and significantly reduce unspecific interaction events, although they can also interfere with fusion if excessive amounts are used.

The capacity of amphiphilic antimalarial drugs (which comprise the extensive aminoquinoline and artemisinin drug derivative families) to easily cross lipid bilayers demands a careful design of their targeting liposomes. Active loading techniques based on pH gradients across liposome membranes [16] are required to efficiently encapsulate the fully ionized species of amphiphilic drugs, in combination with a saturated lipid-enriched bilayer capable of maintaining a proton gradient. As a consequence of the reduced fluidity of the resulting membrane, fusion events with targeted cells are significantly inhibited; sustained drug delivery while the liposome is docked onto the RBC is the most likely mechanism through which such nanovectors operate. This process would be mediated by a depletion of the liposomal proton gradient by means of temperature, liposome–cell interaction events [17] and lipid transference to plasma components [18], and might be highly effective for the delivery of weak basic drugs such as those from the aminoquinoline family. These compounds, positively charged at neutral pH, will theoretically accumulate inside the cell and become entrapped by virtue of the electrochemical gradient created by the phospholipid asymmetry in RBC membranes [19], which maintains a negatively charged intracellular membrane lining. Liposomal nanovectors are also efficient carriers for hydrophobic drugs like lumefantrine and halofantrine, which can be delivered to RBCs following a sustained release process by an exchange mechanism of hydrophobic material between the apposed membranes of liposome and erythrocyte [20]. Since the liposomes adsorbed on RBC surfaces would probably sufficiently modify cell shape to target it for removal through spleen filtration, a compromise between stable drug containment and lipid bilayer fusion will have to be reached through the adequate liposome formulation, with the objective of achieving liposome–RBC merging before spleen removal while avoiding rapid drug leaking from liposomes.

It is reasonable to predict that the nanovector design limitations exposed above can be satisfactorily dealt with, and that some of the future antimalarials yet to be discovered will be harmless for erythrocytes, thus allowing for the loading in these cells of drug amounts that are lethal for Plasmodium. If so, the pathogen might encounter its enemy at home, right at the very moment of entering the host cell, which would have devastating effects for the parasite and significantly compromise its survival capacity. Such a strategy could be likely developed into a prophylactic treatment against erythrocyte infection.