Inhibitors of the Nonmevalonate Pathway of Isoprenoid Biosynthesis as Antimalarial Drugs

See allHide authors and affiliations

Science  03 Sep 1999:
Vol. 285, Issue 5433, pp. 1573-1576
DOI: 10.1126/science.285.5433.1573


A mevalonate-independent pathway of isoprenoid biosynthesis present in Plasmodium falciparum was shown to represent an effective target for chemotherapy of malaria. This pathway includes 1-deoxy-d-xylulose 5-phosphate (DOXP) as a key metabolite. The presence of two genes encoding the enzymes DOXP synthase and DOXP reductoisomerase suggests that isoprenoid biosynthesis in P. falciparum depends on the DOXP pathway. This pathway is probably located in the apicoplast. The recombinant P. falciparumDOXP reductoisomerase was inhibited by fosmidomycin and its derivative, FR-900098. Both drugs suppressed the in vitro growth of multidrug-resistant P. falciparum strains. After therapy with these drugs, mice infected with the rodent malaria parasiteP. vinckei were cured.

Malaria is one of the leading causes of morbidity and mortality in the tropics, with 300 million to 500 million estimated clinical cases and 1.5 million to 2.7 million deaths per year. Nearly all fatal cases are caused by Plasmodium falciparum. Because the parasite's resistance to conventional antimalarial drugs such as chloroquine is growing at an alarming rate, new efficient drugs are urgently needed.

In all organisms studied so far, the biosynthesis of isoprenoids such as sterols and ubiquinones depends on the condensation of different numbers of isopentenyl diphosphate units (1). In mammals and in fungi, isopentenyl diphosphate is derived from the mevalonate pathway. This pathway depends on the condensation of three molecules of acetyl coenzyme A (acetyl CoA) into 3-hydroxy-3-methylglutaryl coenzyme A (HMG-CoA), which is reduced to mevalonate by HMG-CoA reductase. Mevalonate is further converted into isopentenyl diphosphate with mevalonate 5-phosphate as an intermediate. Previous studies revealed very low HMG-CoA reductase activity in P. falciparum(2), and attempts to establish HMG-CoA reductase inhibitors as antimalarial drugs failed (3), suggesting the absence of a mevalonate pathway in P. falciparum.

Recently, in some eubacteria, algae, and plants, the 1-deoxy-d-xylulose 5-phosphate (DOXP) pathway was described as an alternative nonmevalonate pathway for the early steps in the biosynthesis of isoprenoids. In higher plants, plastidic isoprenoids such as carotenoids are formed through the DOXP pathway (4). The sterols, in turn, are formed in the cytosol through the mevalonate pathway. The DOXP pathway is characterized by the condensation of glyceraldehyde 3-phosphate and pyruvate to DOXP and its conversion to 2-C-methyl-d-erythritol 4-phosphate by the enzymes DOXP synthase and DOXP reductoisomerase (5, 6).

Searching for genes encoding enzymes of the DOXP pathway in P. falciparum, we identified similarities between known bacterial and blue algal protein sequences of the enzyme DOXP reductoisomerase and sequences on chromosome 14 of P. falciparum, obtained from the P. falciparum genome project (7). The high degree of similarity suggests the existence of a functional DOXP reductoisomerase in P. falciparum. The complete gene encoding the DOXP reductoisomerase of P. falciparum was consequently cloned (Fig. 1). In addition, we identified aP. falciparum gene that was very similar to the DOXP synthase of different species of bacteria and plants. Expression of these genes in the erythrocytic stages of P. falciparum (the stage responsible for the clinical manifestation of the disease) was confirmed by reverse transcriptase–polymerase chain reaction (8).

Figure 1

Alignment of the deduced amino acid sequence of the P. falciparum DOXP reductoisomerase with sequences from the homologous enzymes from other organisms (18). Pfal,P. falciparum (GenBank accession number AF1118131); Ecol,E. coli (GenBank accession number AF035440 and Swiss-Prot accession number P45568); Bsub, Bacillus subtilis(EMBL accession number Z99112); Syne, Synechocystis sp. PCC6803 (Swiss-Prot accession numbers P73067 and Q55663). Black and gray outlines indicate identical and similar amino acid residues, respectively.

In contrast to the protein sequences of the DOXP reductoisomerases from bacteria and blue algae, a unique NH2-terminal extension of the corresponding homologous enzyme of P. falciparum was identified (Fig. 1). The first 30 amino acids of the NH2-terminal extension of the P. falciparum DOXP reductoisomerase resemble an endoplasmic reticulum signal peptide, whereas the following 44 amino acids exhibit the characteristics of plastidial targeting sequences (9). Because Toxoplasma gondii, in contrast to P. falciparum, can be transfected efficiently by means of established vectors, we transfectedT. gondii with a construct containing the NH2-terminal sequence of the P. falciparum DOXP reductoisomerase fused to green fluorescent protein (GFP) (10). GFP accumulated in a restricted region of the cell that colocalizes with extranuclear apicoplast DNA (Fig. 2). The apicoplast is a plastidlike organelle acquired by members of the phylum Apicomplexa by secondary endosymbiosis of an alga. The metabolic function of the apicoplast is not known, but it is essential for the survival of the parasites (11).

Figure 2

Targeting of GFP to the apicoplast of T. gondii by the leader peptide of the P. falciparum DOXP reductoisomerase. Toxoplasma gondii cells were transiently transfected with a construct encoding the leader peptide fused to the NH2-terminus of GFP. The localization of GFP was analyzed by direct fluorescence (green). Nuclear and apicoplast DNA were counterstained with 4′,6′-diamidino-2-phenylindole (blue). Color images were (left and middle) obtained independently and (right) overlaid.

Recombinant P. falciparum DOXP reductoisomerase was produced in Escherichia coli (Fig. 3A) (12). The conversion of DOXP to 2-C-methyl-d-erythritol 4-phosphate by the recombinant enzyme was determined in an assay based on the NADPH (the reduced form of nicotinamide adenine dinucleotide phosphate) dependency of the reaction. This conversion was inhibited by fosmidomycin and its derivative, FR-900098, in a dose-dependent manner (Fig. 3B) (13). Fosmidomycin has recently been shown to inhibit the DOXP reductoisomerase from bacteria and plants (14).

Figure 3

Inhibition of the recombinant P. falciparum DOXP reductoisomerase by fosmidomycin and FR-900098. (A) The P. falciparum DOXP reductoisomerase was expressed in E. coli with an NH2-terminal His tag. Crude bacterial lysates (lane 1) and the purified enzyme (lane 2) were analyzed by SDS-PAGE. (B) The enzymatic activity of the recombinant enzyme was monitored by NADPH oxidation in the presence of different concentrations of fosmidomycin (open circles) and FR-900098 (solid circles).

To investigate whether the DOXP pathway is essential for the survival of the parasites, we treated P. falciparum cultures with fosmidomycin and FR-900098. The viability of P. falciparum was determined by [3H]-hypoxanthine incorporation in the presence of different concentrations of the drugs (15). Both drugs inhibited the growth of P. falciparum in submicromolar concentrations (Fig. 4). Three P. falciparum strains with different sensitivity patterns to chloroquine and pyrimethamine were efficiently inhibited (Table 1).

Figure 4

In vitro antimalarial activity of fosmidomycin and FR-900098. Plasmodium falciparum–infected erythrocytes were cultured in the presence of different amounts of fosmidomycin (open circles) or FR-900098 (solid circles). The viability of the parasites was determined by [3H]-hypoxanthine incorporation. The antimalarial activity was measured with the P. falciparum strains (A) HB3, (B) A2, and (C) Dd2. Mean values from three to four independent experiments are shown.

Table 1

Sensitivity of different P. falciparumstrains to fosmidomycin, FR-900098, chloroquine, and pyrimethamine. Concentrations causing half-maximal inhibition (IC50) were determined in vitro with the indicated P. falciparum strains (15). Mean values and standard deviations from three to four independent experiments are shown.

View this table:

Because the DOXP pathway is absent in mammals and fosmidomycin and FR-900098 are known to have low toxicity (lethal doses of >8000 and >5000 mg/kg after oral and subcutaneous administration, respectively, in rats), their antimalarial activity was tested in an animal model (13). Mice infected with the rodent malaria parasiteP. vinckei, which is fatal if untreated, were treated with the drugs from day 1 after infection to 4 days later (16). The drugs were well tolerated, even at the highest dosage of 300 mg/kg. Untreated control animals developed parasitemias of >60% on day 5 and died on day 7 after infection. Animals treated intraperitoneally with dosages of >10 mg/kg of fosmidomycin or 5 mg/kg of FR-900098 were apparently free of parasites (Table 2). After treatment with 2 mg/kg of FR-900098 or 5 mg/kg of fosmidomycin, parasitemias were <1%. Animals treated orally with 50 or 100 mg/kg of either fosmidomycin or FR-900098 were apparently free of parasites, and parasitemias were <1% after treatment with 20 mg/kg of either drug. Recrudescence was observed when the treatment was terminated after 4 days. Mice treated with 30 mg/kg of fosmidomycin or FR-900098 over a period of 8 days were totally cured. The efficiency of these drugs at low dosages and their activity after oral administration demonstrate that fosmidomycin and FR-900098 are potential antimalarial drugs.

Table 2

In vivo antimalarial activity of fosmidomycin and FR-900098. The in vivo antimalarial activity was determined by a 4-day suppressive test with P. vinckei–infected mice (16). The indicated dosages of fosmidomycin and FR-900098 in PBS were administrated intraperitoneally or orally three times per day over a period of 4 days. Parasitemias were determined on day 5 by Giemsa-stained blood smears. Mean values and standard deviations from three to four independent experiments, each containing four to six mice per dosage, are shown.

View this table:

In summary, we provide evidence for the presence of the DOXP pathway in the apicoplast of P. falciparum and for the inhibition of one of the key enzymes of this pathway, DOXP reductoisomerase, with fosmidomycin and FR-900089. In addition, we have demonstrated the antimalarial activity of these drugs in vitro and in vivo. In former studies with volunteers and patients suffering from bacterial infections, fosmidomycin was well tolerated (17). The efficacy of these drugs against multidrug-resistant parasites and their low manufacturing costs and high stability make them very attractive as a potential new class of antimalarial drugs.

  • * To whom correspondence should be addressed. E-mail: hassan.jomaa{at}


Stay Connected to Science

Navigate This Article