Jammed traffic impedes parasite growth
Plasmodia are pathogenic protozoa that cause malaria in humans and other mammals, e.g., rodents. After infection through the bite of a mosquito, the parasites undergo a first developmental cycle in the liver and soon enter the blood stage by retreating into erythrocytes. During this phase, parasites proliferate rapidly and multiply up to 32-fold within 48 h. Synchronized bursting of red cells accompanied by high fever marks the end of one growth cycle. Within minutes, the released parasites invade new red cells to start another round. In a recent issue of PNAS, Liu et al. (1) demonstrate that the lack of a glycerol facilitator from the aquaporin (AQP) superfamily, i.e., AQP9, increases the resistance of infected mice: 2 weeks after infection all knockout mice were still alive, whereas 50% of the wild-type animals had died. AQPs constitute an ancient family of tetrameric channel proteins with individual pores in each monomer. In humans, 13 isoforms are known that facilitate transport of water and/or small, uncharged solutes across lipid membranes according to the prevailing osmotic or chemical gradients (2). The study by Liu et al. not only provides important insight into the interconnection of solute transport pathways between plasmodia and red cells, it may also prompt novel therapeutic approaches that target the parasite indirectly and thus have the potential to overcome the increasingly severe problem of resistance development.
Glycerol Uptake Across Three Consecutive Membranes
Inside red blood cells, three membranes shield the malaria parasite from the host's immune system, i.e., the red blood cell membrane, the parasitophorous vacuole membrane, which is derived from the red cell membrane and forms during invasion, and finally the parasite's own plasma membrane (see Fig. 1). Its sheltered lifestyle, however, makes the parasite dependent on the provision of nutrients and metabolic precursors by the host red cell. One property of red cells is their high glycerol uptake rate. In humans, this rate is accounted for by the presence of AQP3, which together with AQP7, AQP9, and AQP10 belongs to the aquaglyceroporin solute facilitator branch of the AQP protein family (2, 3). A physiological role for glycerol permeability in red cells, however, is not yet established. Mouse red cells are equally well permeable for glycerol, but AQP3 was not detected. Liu et al. (1) identified AQP9 in mouse reticulocytes and mature erythrocytes by Western blotting and further showed that disruption of the gene indeed results in red cells with low glycerol permeability. The presence of different AQPs in human and mouse red cells is an unusual example for a functional replacement of one AQP isoform by another because AQPs typically have a very distinct and isoform-specific expression pattern throughout the body. Outside red cells, AQP9 is present at the blood–brain barrier and in liver (4) where it functionally pairs up with the adipocyte aquaglyceroporin AQP7 (5). Together, they facilitate shuttling of the lipolytic product glycerol from fat tissue into the liver to fuel hepatic glucose production. Liu et al. put a genuine twist to the study of AQP9 knockout mice by asking whether proliferation of malaria parasites may be compromised in red cells devoid of the aquaglyceroporin.
Proposed metabolic solute pathways in a Plasmodium-infected erythrocyte. Deletion of mouse AQP9 (orange) or the Plasmodium aquaglyceroporin (green) leads to reduced parasite proliferation and virulence. RBCM, red blood cell membrane; PVM, parasitophorous vacuole membrane; PPM, parasite plasma membrane; Hb, hemoglobin; Rh-protein, ammonium transporter of the Rhesus protein family.
Rapid biosynthesis of glycerolipids is a prerequisite for the parasite's high proliferation rate. Radiotracer studies showed glycerol uptake by plasmodia and integration into the extending membrane long before the discovery of AQPs (6). From an economic point of view, it seems reasonable for the parasite to use readily available glycerol from the host serum (Fig. 1) rather than metabolically generating glycerol from glucose, its sole energy source. Indeed, plasmodia express a single aquaglyceroporin with high glycerol permeability (7) that may very well complement the red cell AQP9 to form a continuous glycerol uptake pathway (Fig. 1). The presence of aquaglyceroporins in the plasmodial and red cell membrane, respectively, is established. It is not clear whether the parasitophorous vacuole membrane carries AQP9 or AQP3, respectively. There is also evidence for wide and rather un selective “nutrient channels” in this membrane. Yet, the molecular basis of such pores is unknown (8).
Earlier, the research group around Liu (9) generated a rodent Plasmodium berghei knockout strain that lacks the endogenous aquaglyceroporin (ΔPbAQP). The strain quite suggestively displays similarly reduced virulence. Additionally and importantly, the ΔPbAQP parasites proliferate only half as fast as wild-type parasites. Having said that, under in vitro culturing conditions, i.e., with ample glucose available, proliferation of wild-type plasmodia appears independent of the presence or absence of glycerol (unpublished observation). Glycerol uptake may thus be relevant only when the glucose supply is limited as during later stages of the disease in which patients are likely to become hypoglycemic. Nevertheless, the discrepancy between in vitro and in vivo effects and the clearly milder phenotype in AQP9 knockout red cells pose the question of whether restriction from serum glycerol alone can explain the observed growth defect in ΔPbAQP parasites.
Waste Release via the Plasmodium AQP and Red Cell Channels/Transporters?
The Plasmodium aquaglyceroporin efficiently conducts various other physio logical solutes, such as polyols, urea, methylglyoxal, and ammonia (7, 10–12). The permeability profile of AQP9 also includes urea, even purines and pyrimidines, and ammonia (13, 14). Which metabolite will actually pass the aqua glyceroporins in a physiological setting obviously depends on the availability and the presence of transmembrane gradients in the current metabolic situation.
Blood-stage plasmodia are metabolically extremely active. They consume glucose 100 times faster than uninfected red cells and further engage in massive hemoglobin proteolysis. As a result, toxic side products and waste products can be expected in large quantities, e.g., methylglyoxal from glycolysis, urea from degradation of arginine, and ammonia from conversion of amino acids to α-keto acids (Fig. 1). Methylglyoxal is chemically highly reactive and interferes with protein and DNA function by covalent modification (11). Ammonia (NH3) changes the intracellular pH because of chemical equilibrium with its protonated form ammonium (NH4 +; pKa 9.23). At neutral pH, ≈99% of the ammonia molecules will be protonated; protonation increases further at lower pH (Fig. 1). Taking into account the ammonia production rate (0.8 fmol per parasite per hour), the parasite volume (20 fl), and the IC50 (2.8 mM), one can calculate that if there was no release of ammonia from the parasite, it would intoxicate itself within only minutes (12).
Blood-stage plasmodia are metabolically extremely active.
Small, uncharged metabolites, such as urea or methylglyoxal, could follow an outward directed gradient via the same combination of the Plasmodium AQP plus AQP9 (11) or red cell urea transporters (refs. 1 and 15; Fig. 1). The case gets more convoluted when pH gradients interfere with chemical gradients as laid out above for the ammonia/ammonium equilibrium. One aspect of the Plasmodium physiology seems particularly geared to extrude ammonia, i.e., the active acidification of the parasitophorous vacuole by proton ATPases (ref. 16; Fig. 1). A proton gradient across the parasite membrane generates an outward directed gradient for uncharged ammonia because in the acidic vacuole, ammonia is immediately converted into ammonium. The Plasmodium AQP increases ammonia transmembrane permeability ≈10-fold (12). Charged ammonium, in turn, should not pass the parasite membrane because the genome does not encode typical ammonium transporters. Further export of ammonium into the serum may be accomplished via red cell ammonium transporters of the Rh family or in the uncharged ammonia form via mouse AQP9 or human AQP3 (Fig. 1). Generally, the red cell membrane seems to provide alternative solute pathways, whereas transport across the parasite membrane is restricted to a smaller number of channel proteins. The more severe phenotype of the parasite ΔPbAQP knockout as compared with the red cell AQP9 knockout is indicative of a less redundant set of transmembrane channels in plasmodia.
The recent results by Liu et al. (1) in combination with the ΔPbAQP knockout phenotype illuminate one pathway of an intertwined system of channel proteins from both plasmodia and red cells that facilitates transport of solutes from the serum into the parasite's cytosol or vice versa. They show that if one component fails, either in the parasite or in the red cell, flow will be compromised with consequences for the parasite. In view of the increasingly rapid spreading of drug-resistant Plasmodium strains, this is an important notion from which novel therapeutic concepts may be derived that do not directly aim at the parasite but at its immediate environment. If it is possible to shut off vital supply or waste lines in the red cell, it might leave the parasite helpless and without any means to adapt.
Footnotes
- *E-mail: ebeitz{at}pharmazie.uni-kiel.de
-
Author contributions: E.B. wrote the paper.
-
The author declares no conflict of interest.
-
See companion article on page 12560 in issue 30 of volume 104.
- © 2007 by The National Academy of Sciences of the USA
