INTRODUCTION
Up to half a billion people contract malaria each year, and nearly one million die of the disease. The development of resistance to antimalarial drugs is a major obstacle to the treatment of malaria and has made many of the available drugs ineffective. Reports of emerging resistance to artemisinins in Western Cambodia (
6,
17–
18) warn that the malaria burden may increase, especially if new intervention strategies are not introduced. There is therefore an urgent need to discover new antimalarial drugs that act via novel drug targets. Robust validation of novel targets is critical to this process (
30), and for technical and clinical reasons, this has been studied predominantly in the asexual erythrocytic stage of plasmodial parasites. Characterizing the nature of new targets during different stages of the parasite's life cycle can define their essentiality and therefore help in choosing those that are critical to the survival of most stages. Targets that indicate parasites are vulnerable at asexual liver stages of infection that rapidly amplify parasite numbers or that could block transmission have advantages over those that may make parasites vulnerable at a more limited range of stages of the parasite's life cycle (
10).
Predominantly, malarial parasites use glycolysis to generate ATP for their energy requirements, with the more efficient tricarboxylic acid (TCA) cycle being largely disconnected from the energy-generating process, at least during the asexual blood stage (
20,
27,
33). Without intracellular energy stores during most of their life cycle, they are dependent on a constant supply of
d-glucose from their hosts (reviewed in reference
21). The
Plasmodium falciparum hexose transporter, PfHT, is the primary
d-glucose transporter in
P. falciparum parasites, enabling the uptake of this essential nutrient across the parasite plasma membrane (
34). Using a selective
d-glucose derivative (compound 3361) as a competitive inhibitor, PfHT was validated chemically as an antimalarial target (
8). Compound 3361 kills asexual blood-stage
P. falciparum parasites
in vitro, with a 50% inhibitory concentration (IC
50) of 16 μM, as well as suppressing the rodent malarial parasite,
Plasmodium berghei,
in vivo (
8). The importance of PfHT and PbHT (the
P. berghei orthologue) has been further supported by the demonstration that disruption of the corresponding genes renders asexual blood-stage parasites nonviable (
29). Furthermore, using a GFP tag, it was reported that PbHT is expressed throughout the parasite's development inside the mosquito vector (
29). This suggests PbHT may be functionally important during insect stages.
Here we have investigated additional life cycle stages during which plasmodial hexose transporters might be essential, using compound 3361. The aims of the current investigation were the following: (i) to examine the expression of PbHT during liver-stage development, using the PbHT-GFP-expressing
P. berghei parasite generated previously (
29), (ii) to determine the chemotherapeutic potential of PbHT at the liver stage, using compound 3361 (
8), and (iii) with the same compound, to ascertain if targeting of plasmodial hexose transporters can block transmission.
DISCUSSION
In comparison to a range of different drug classes acting against the replicative stages of drug-sensitive parasites in the blood, there are relatively few active against liver and sexual stages (
24). Yet effective intervention at the liver stage can cure infection before it causes symptoms, eliminating the risk of progression to severe or fatal disease associated with blood stages, while intervention at the sexual stage can stop transmission. Only the following drugs act against liver stages of infection: pyrimethamine, proguanil, atovaquone, primaquine, and tafenoquine. Currently methodologies limit the application of conventional screening approaches for testing compound libraries on hepatocyte and transmission stages of infection, although new methodologies are under development (
5,
22). It is therefore valuable to consider developing drugs against high-biological-value targets that are essential for survival of these stages but which may be less amenable for adaptation to high-throughput screening assays.
d-Glucose uptake mediated by PfHT and PbHT is essential for the survival of erythrocytic stages of
P. falciparum and
P. berghei parasites (
8,
29). This may be unsurprising, since they need large amounts of
d-glucose due to their reliance upon glycolysis for ATP production. The reason for this reliance is not fully understood, but blood-stage asexual parasites have evolved a novel carbon metabolic pathway in which glycolysis is largely disconnected from an intact TCA cycle (a far more efficient ATP production process) that has become branched rather than cyclical (
20). Whether this applies to other life cycle stages awaits determination. While these data confirm the importance of
d-glucose (and its transport) during blood stages, there are relatively few studies that have investigated energy metabolism in other life cycle stages. The use of a relatively specific inhibitor of the key hexose transporter encoded by
Plasmodium spp. (PxHT) together with tractable animal and
in vitro models of infection (such as the recently developed liver-stage infection models [
22–
23]) has established stages of infection using hexose.
The data presented here demonstrate liver-stage expression of PbHT-GFP, which is consistent with the identification of the
P. yoelii orthologue of PbHT (PyHT) in a recent proteomic study of liver-stage parasites (
32). Data also localize the tagged hexose transporter to the parasite surface (being present in the parasite plasma membrane and/or the parasitophorous vacuolar membrane) predominantly over the first 48 h of intrahepatic development, in keeping with its localization during other life cycle stages (
29,
34). At these time points, the internal PbHT-GFP signal observed might be PbHT in the process of production and trafficking (and/or mislocalized). At 67 h postinvasion, the GFP fluorescence pattern observed here in parasites is similar to that observed in cytomere-stage parasites stained using an antibody against merozoite surface protein 1 (
31). This protein is localized to the parasite plasma membrane (rather than the parasitophorous vacuole membrane), which invaginates at the cytomere stage to surround nuclei and leads to merozoite formation. This suggests that PbHT-GFP is also localized to the parasite plasma membrane, although further localization studies with additional reagents would be required to confirm this conclusion definitively.
Compound 3361 not only kills
P. falciparum parasites in blood culture and
P. berghei in vivo but also inhibits
P. berghei development in liver stages of infection in a human cell line. The potency of antimalarial activity of compound 3361 for hepatic stages of
P. berghei (11 μM) is highly comparable to what we have observed for blood stages of
P. falciparum infection (16 μM) (
8), consistent with compound 3361 targeting PbHT. Note that these data also compare favorably with the
Ki values for the effect of compound 3361 on
d-glucose transport via heterologously expressed PfHT and PyHT, which are 53 and 80 μM, respectively (
8–
9). The latter hexose transporter shares 96% amino acid sequence identity with PbHT, suggesting that similar results would be obtained for PbHT.
Compound 3361 may alter host cell
d-glucose homeostasis (or other unrelated mechanisms) sufficiently to kill the intracellular parasite but not the host cell. However, this is a less likely explanation because compound 3361 did not affect 2-DOG uptake into uninfected Huh-7 cells. Under the conditions used, uptake was linear (extrapolating to the origin) and concentrative (5-fold over 20 min). While it is not possible to rule out completely a faster initial transport step (for example, due to the loss of unphosphorylated [
14C]2-DOG during processing), this is consistent with 2-DOG being phosphorylated at the same rate as being transported into the Huh-7 cell cytosol (i.e., transport is rate limiting). These data suggest that compound 3361 does not interfere with either endogenous
d-glucose transporters (facilitative glucose transporters, GLUT1 and -2, in Huh-7 cells, predominantly [
11]) or kinases (hexokinases II and IV in Huh-7 cells, predominantly [
19]). This is consistent with the previous findings that compound 3361 is a weak inhibitor of mammalian GLUT1 and -5 (
Ki values > 1 mM) and has no affect on parasite hexokinase activity at concentrations up to 200 μM (
8,
26). Interestingly, the effect of compound 3361 was not sensitive to external
d-glucose. However, this may not be surprising if compound 3361 targets the intracellularly localized PbHT and given the role that hepatocytes play in tightly regulating intracellular
d-glucose levels as part of their role in systemic glucose homeostasis (
2,
4). As is the case with erythrocytes, these data suggest that delivery of inhibitor to the parasite surface (its proposed site of action) is not materially impeded in infected liver cells, consistent with the highly lipophilic nature of the inhibitor.
The data failed to show any effect of compound 3361 on parasite invasion (as opposed to development), since the presence of the inhibitor in the Huh-7 cell culture either 1 h before or 2 h after sporozoite addition made no difference to subsequent development. While this is far from conclusive, it may suggest that the hexose transporter is not essential for sporozoite function.
Plasmodium-infected mosquitoes are known to upregulate an endogenous
d-glucose transporter gene in their salivary glands, suggesting that
d-glucose is competed for by salivary gland-localized sporozoites (
25). Also, previous studies, including our own PbHT-GFP studies, have demonstrated expression of PbHT in sporozoites (
14,
29). However,
P. berghei sporozoites remain motile, and thus energized, in the absence of
d-glucose if one of a number of amino acids is present (
15). This suggests that under physiological conditions inhibition of
d-glucose uptake by compound 3361 is unlikely to affect sporozoite motility and invasion.
These studies have also explored the dependency of some of the sexual stages of parasite development on the delivery of
d-glucose, since our own expression studies (
29) and previous proteomic analysis (
12; A. M. Talman, J. H. Prietro, S. Marques, M. Lawniczak, R. Frank, A. Ecker, R. S. Stanway, S. Krishna, C. Morin, G. K. Christophides, J. R. Yates III, and R. E. Sinden, unpublished data) have demonstrated the presence of PbHT in transmission stages studied. We have shown here that compound 3361 can inhibit early sexual stages of parasite development. Indeed, ookinete development was hampered prefertilization and to a lesser extent postfertilization. In both cases, there was no clear evidence for competition by excess
d-glucose reversing inhibition of ookinete development. Compound 3361 also inhibited exflagellation, and in this case, inhibition was sensitive to the
d-glucose concentration. These data suggest that
d-glucose transport is essential for the cellular events of microgametogenesis and agree with previous observations that
d-glucose maintains the viability (ability to complete microgametogenesis) of male gametocytes in
P. gallinaceum (
16). They suggest
d-glucose as the key metabolite that powers male gamete motility. The relatively high concentrations of compound 3361 that are needed to inhibit sexual compared with blood and liver stages may have several explanations. These include a reduced requirement for
d-glucose, reduced access to the target, and the lack of an appropriate preincubation period with inhibitor or “off-target” effects that kill this stage so that PbHT is not required for survival. With regard to the former, it is worth noting that the concentrations of compound 3361 required to inhibit sexual stages of development are those predicted to maximally block transport of
d-glucose via plasmodial hexose transporters (
8–
9,
26).
In conclusion, here we have demonstrated that PbHT is expressed during intrahepatic parasite development and is localized to the parasite surface. Furthermore, our data are consistent with the hypothesis that targeting plasmodial hexose transporters could be used for causal prophylaxis. We have also demonstrated that transmission-stage parasites are susceptible to a selective inhibitor of plasmodial hexose transporters, although we were only able to demonstrate P. berghei transmission blocking activity, using 1 mM compound 3361. Nevertheless, our findings indicate that d-glucose transport may be a suitable target for mammal-to-mosquito-transmission-blocking drugs.