Isoprenoids play important roles in all living organisms as components of structural cholesterol, steroid hormones in mammals, carotenoids in plants, and ubiquinones. Significant differences occur in the length of the isoprenic side chains of ubiquinone between different organisms, suggesting that different enzymes are involved in the synthesis of these side chains. Whereas in Plasmodium falciparum the isoprenic side chains of ubiquinone contain 7–9 isoprenic units, 10-unit side chains are found in humans. In a search for the P. falciparum enzyme responsible for the biosynthesis of isoprenic side chains attached to the benzoquinone ring of ubiquinones, we cloned and expressed a putative polyprenyl synthase. Polyclonal antibodies raised against the corresponding recombinant protein confirmed the presence of the native protein in trophozoite and schizont stages of P. falciparum. The recombinant protein, as well as P. falciparum extracts, showed an octaprenyl pyrophosphate synthase activity, with the formation of a polyisoprenoid with eight isoprenic units, as detected by reverse-phase HPLC and reverse-phase TLC, and confirmed by electrospray ionization and tandem MS analysis. The recombinant and native versions of the enzyme had similar Michaelis constants with the substrates isopentenyl pyrophosphate and farnesyl pyrophosphate. The recombinant enzyme could be competitively inhibited in the presence of the terpene nerolidol. This is the first report that directly demonstrates an octaprenyl pyrophosphate synthase activity in parasitic protozoa. Given the rather low similarity of the P. falciparum enzyme to its human counterpart, decaprenyl pyrophosphate synthase, we suggest that the identified enzyme and its recombinant version could be exploited in the screening of novel drugs.
Malaria, one of the most important infectious diseases in the world, kills over one million people each year, and its incidence is increasing in several regions of the world. Lethal forms of the disease are caused by Plasmodium falciparum, and the increasing resistance of this parasite to virtually all current drugs calls for the identification of new therapeutic targets and the development of new drugs . In order to avoid the rapid emergence of resistance, combinations of antimalarial drugs acting on different points of the same metabolic pathway are believed to increase therapeutic success . An important target for the development of new antimalarial drugs is isoprenoid biosynthesis, which occurs via the 2-C-methyl-D-erythritol-4-phosphate pathway [3–7] in P. falciparum; in contrast, mammalian cells, certain eubacteria, archaea and fungi synthesize isoprenoid precursors through the mevalonate pathway . Several products of isoprenoid metabolism have also been identified in P. falciparum: prenylated proteins [3,5], dolichols  and ubiquinones . All of these products are derived from linear prenyl diphosphates, which are synthesized by sequential condensations of IPP (isopentenyl pyrophosphate) with allylic prenyl diphosphates. The condensation is catalysed by a family of enzymes known as prenyltransferases or polyprenyl diphosphate synthases [11–13]. Short-chain isoprenoids (C10 to C20) are used as precursors for many different compounds, such as the farnesylated and geranylgeranylated proteins , while long-chain isoprenoids (C30 to C50) seem to be used for the synthesis of the isoprenoid side chain of ubiquinone .
Amino acid sequence comparisons of polyprenyl diphosphate synthases have revealed the existence of similar, highly conserved regions containing two aspartate-rich domains (DDXXD) (domains II and VI), which are thought to be the binding sites for the diphosphate moieties of IPP and the allylic substrate [15–17]. Divergence in the N-terminal regions is assumed to represent signal sequences responsible for differential subcellular localizations [18,19].
OPPs (octaprenyl pyrophosphate synthase) belongs to a prenyltransferase family that catalyses the condensation reactions of FPP (farnesyl pyrophosphate) with five molecules of IPP to produce C40 OPP (octaprenyl pyrophosphate) . OPPs enzymes are responsible for the biosynthesis of side chains attached to ubiquinones in Escherichia coli [20,21]. Other organisms contain ubiquinones with different side chain lengths that are synthesized by specific prenyltransferases. For example, the sizes of the ubiquinone side chain are C30 in Saccharomyces cerevisiae, C45 in rats and C50 in humans, and these are synthesized by hexaprenyl pyrophosphate synthase, solanesyl pyrophosphate synthase and decaprenyl pyrophosphate synthase respectively [22,23]. The presence of an active isoprenoid pathway for the biosynthesis of isoprenic chains of CoQ in P. falciparum has also been demonstrated .
Herein, we cloned the gene encoding a putative P. falciparum OPPs and characterized its gene product. In parallel, we also characterized the partially purified native version of the enzyme from schizont stages, and compared its kinetic properties with those of the recombinant version. Finally, inhibition assays demonstrated the effects of the terpene nerolidol on the recombinant version of the enzyme.
General molecular biology reagents were obtained from Gibco BRL/Invitrogen (Rockville, MD, U.S.A.). Nerolidol, biochemical reagents and standards were obtained from Sigma Chemical Co. Percoll® was purchased from Pharmacia (Uppsala, Sweden). [1-14C]IPP ammonium salt (55.0 Ci·mmol−1), [1-(n)-3H]FPP triammonium salt (17.0 Ci·mmol−1) and [1-(n)-3H]GGPP (geranylgeranyl pyrophosphate) triammonium salt (16.5 Ci·mmol−1) were obtained from Amersham-Pharmacia Biotech. Geranyl pyrophosphate ammonium salt, FPP ammonium salt and GGPP ammonium salt were obtained from Sigma Chemical Co. Albumax I was from Gibco BRL Life Technologies. All solvents were of analytical grade or better. Hyperfilm™ MP films (Amersham-Pharmacia Biotech) were used for autoradiography.
Cultures of P. falciparum (isolate NF54; clone 3D7) were grown using a modification  of the method of Trager and Jensen . Development and multiplication of cultures was monitored by microscopic evaluation of Giemsa-stained thin smears. Ring-infected (0–20 h forms), trophozoite-infected (20–40 h forms) and schizont-infected (40–48 h forms) erythrocytes were purified on 40/70/80% discontinuous Percoll® gradients  (30 min at 25 °C and 10000 g), yielding an upper band of schizont stages (40%), a band of trophozoite stages (70–80% interface), and a pellet for the ring stage and uninfected erythrocytes.
Nucleic acid extraction, PCR amplification, and cloning and sequencing of the gene encoding PfOPPs (
P. falciparum OPPs)
Total RNA isolation was performed as described previously . Parasite DNA was isolated from the parasite pellet by proteinase K digestion and phenol/chloroform extraction . P. falciparum genomic DNA was submitted to PCR amplification using primers designed according to the sequence of the putative gene PFB0130w (http://www.PlasmoDB.org; GenBank® accession number AAC71816), introducing BamHI restriction sites (sense, 5′-CCGGATCCATGGTTCACCTAAGTAAAAG-3′; antisense, 5′-CCGGATCCTCATTTGAGGTTTCTTGATAAC-3′). PCR mixtures included 200 ng of extracted DNA, 0.1 μM of each primer, 200 μM dNTP, PCR buffer containing 1.5 mM MgCl2, 50 mM KCl and 0.05 unit/μl Taq DNA polymerase (Invitrogen). The amplification conditions were: initial denaturation at 95 °C for 5 min; 30 cycles of 94 °C for 30 s, 52 °C for 1 min and 72 °C for 30 s; and a final incubation at 72 °C for 10 min. The 1.6 kb amplicon obtained was cloned into pGEM®T-easy vector (Promega), according to the manufacturer's instructions. Three clones were sequenced in both directions using Big Dye terminator chemistry on an ABI3100 sequencer (Applied Biosystems). Identities were checked using the BLAST tool at NCBI (http://www.ncbi.nlm.nih.gov/BLAST/). Similarity between amino acid sequences was calculated using the Point program (http://www.geocities.com/alvesjmp/software.html).
RT-PCR (reverse transcription–PCR)
RT-PCR was performed on cDNAs from ring, trophozoite and schizont stages of P. falciparum. Total RNA from these stages was isolated by the Trizol method (Invitrogen) and treated sequentially three times with DNase I (Life Technologies/Invitrogen). An aliquot of each RNA sample was reverse-transcribed using SuperScript II (Invitrogen). To exclude the possibility of sample contamination with genomic DNA, an aliquot from each sample was treated in the same way, except that no reverse transcriptase was added. PCR conditions (30 cycles) were as described above. OPPs-specific primers (sense, 5′-ATGGTTCACCTAAGTAAAAG-3′; inner antisense, 5′-GGATTACACATAGGTATGGC-3′) amplified a 400 bp fragment from cDNA. Primers specific for the P. falciparum myosin gene (sense, 5′-TTACATGTTGCATCTATGAGTG-3′; antisense, 5′-TCTCAATTTTTAAATCAATTGACATCTTTAATG-3′) were included in a control reaction. PCR products were separated on 1% agarose/TAE (40 mM Tris/HCl, pH 7.8, 20 mM sodium acetate, 1 mM EDTA) gels. The RT-PCR products were cloned into the pGEM®T-easy vector (Promega) and sequenced as above.
Expression and purification of a recombinant version of PfOPPs
A fragment of the PFB0130w gene was excised using EcoRI from the pGEM-PFB0130w plasmid and ligated into EcoRI-cut pGEX2T (Amersham Pharmacia Biotech). The resulting plasmid was transformed in BL21-CodonPlus™ (DE3)-Ril (Stratagene) bacteria. Protein expression was induced with 1 mM isopropyl β-D-thiogalactoside for 3 h at 37 °C, after which bacterial cells were harvested and resuspended in lysis buffer [7 mM Na2HPO4, 1 mM KH2PO4, pH 7.2, 137 mM NaCl, 3 mM KCl (PBS), 0.1% (v/v) Triton X-100, 0.05 mg/ml lysozyme and 0.2 mM PMSF]. Lysis was completed by three freeze/thaw cycles and genomic bacterial DNA was sheared by 10 passages through a G21 needle. Recombinant proteins were then purified using glutathione–Sepharose beads (Amersham-Pharmacia), following the manufacturer's instructions. The purified protein was dialysed extensively against PBS, checked for purity by SDS/PAGE  and quantified by the Bradford method .
Antibodies against the recombinant version of PfOPPs
Three Balb/C mice were inoculated intraperitoneally with the purified recombinant version of PFB0130w–GST (glutathione S-transferase) fusion protein. As a control, three mice were inoculated with the same quantities of recombinant GST. The immunogen consisted of 60 μg of recombinant fusion protein in complete Freund's adjuvant (Sigma, St. Louis, MO, U.S.A.) for primary immunization, and the same concentration of recombinant polypeptide in Freund's incomplete adjuvant for two booster injections, 21 days and 42 days after the primary immunization. After 52 days, blood samples from the retro-ocular plexus were obtained, under anaesthesia, and sera were prepared.
Metabolic labelling and immunoprecipitation assays
Asynchronous cultures of P. falciparum with parasitaemias of approx. 10% were labelled for 18 h with 25 μCi/ml L-[35S]methionine (>1000 Ci/mmol; Amersham) in methionine-depleted (10 mM) RPMI 1640 medium. Each stage (ring, trophozoite or schizont) was then purified as described above, followed by lysis of cells in two pellet volumes of ice-cold 10 mM Tris/HCl, pH 7.2, 150 mM NaCl, 2% (v/v) Triton X-100, 1 mM PMSF, 5 mM iodoacetamide, 1 mM tosyl-lysylchloromethane (‘TLCK’) and 1 μg/ml leupeptin. After incubation at 4 °C for 15 min, lysates were centrifuged for 30 min at 4 °C and 10000 g, and supernatants were stored in liquid nitrogen. The samples were immunoprecipitated as described by Kessler  and modified by Moura et al. . Immunoprecipitated proteins were analysed by SDS/PAGE and autoradiography.
Partial purification of native PfOPPs
The partial purification of native PfOPPs was conducted only for schizont stage parasites. The parasite pellet obtained from 1 litre of culture (50 culture flasks with a parasitaemia of approx. 20%) was purified on discontinuous Percoll® gradients as described above. The schizont pellet fraction was resuspended in 10 ml of buffer containing 50 mM Tris/HCl, pH 7.5, 1 mM EDTA, 0.2 mM PMSF and 0.1 mM leupeptin at 4 °C. The suspension was sonicated with a microtip for four cycles (15 s each). The lysate was centrifuged for 1 h at 4 °C and 100000 g. The supernatant (S-100) was subjected to 0–50% ammonium sulphate precipitation. The precipitate was resuspended in 3 ml of buffer containing 50 mM Tris/HCl, pH 7.5, 1 mM dithiothreitol, 20 μM ZnCl2 and 20 mM NaCl, and dialysed against the same buffer at 4 °C. This fraction was then used immediately in enzymatic reactions or stored at −80 °C.
Measurement of enzymatic activity and product analysis of native PfOPPs
Enzymatic activity was measured by determination of the amount of [1-14C]IPP incorporated into butanol-extractable polyprenyl diphosphates. The standard assay mixture contained, in a final volume of 0.3 ml, HKMT buffer (100 mM Hepes, pH 7.5, 50 mM KCl, 0.5 mM MgCl2, 0.1% Triton X-100), 10 μM FPP, 0.33 μCi/ml [1-14C]IPP and a suitable amount of enzyme solution (recombinant or partially purified native PfOPPs), and was incubated at 30 °C. The protein concentration used in all experiments was determined by its absorbance at 280 nm (ϵ280=20280 M−1·cm−1). Portions of the reaction mixture were withdrawn periodically (within 10% substrate depletion) and the reaction was stopped by adding 10 mM EDTA (final concentration). The products were then extracted with 1 ml of butan-1-ol, and the radioactivity in the butan-1-ol extract was measured in c.p.m. mode with a Beckman 5000 β-radiation scintillation counter. The initial rate was calculated by determining the quantities of product formed or IPP consumed at each time point by counting the 14C radioactivity in the butanol phase (product) and in the aqueous phase (IPP) .
Kinetic parameters for recombinant and native PfOPPs
Each PfOPPs was assayed at eight different concentrations (1–200 μM) of IPP or FPP, while the other substrate (FPP or IPP respectively) was kept constant at 200 μM. Km values (means±S.E.M.) were obtained by the least-squares method using Enzfitter software (Elsevier-Biosoft, Cambridge, U.K.). Radiometric assays were performed by modifications of the methods of Hirooka et al.  and Sagami et al. .
Identification by RP (reverse-phase)-TLC of dephosphorylated polyisoprenoid products
Polyprenyl diphosphates were extracted from a 1 ml reaction and treated with potato acid phosphatase to convert the poly-prenyl pyrophosphates into their corresponding polyprenols. Hydrolysates were extracted with n-pentane and analysed by RP-TLC (HPTLC RP-18; Merck) developed with acetone/water (19:1, v/v). Authentic standard polyprenols were visualized with iodine vapour, and the plates were exposed in the Storage phosphor screen for approx. 10 days and digitized using a Storm 820 apparatus (Amersham). Image analyses were performed using Image Quant software (Molecular Dynamics).
Purification and identification of the first intermediate (GGPP) by RP-HPLC
The n-butanol fraction was submitted to analytical RP-HPLC separations, which were performed on an Ultrasphere ODS Beckman column (4.6 mm×26 cm) with a Gilson HPLC 322 pump connected to a Gilson 152 variable UV/visible detector at 214 nm. Fractions were collected in 1 ml/min intervals with a Gilson FC203B fraction collector, and UNIPOINT™ System Software was used for the operation and registration of the chromatograms. Unless otherwise stated, 25 mM NH4HCO3, pH 8.0, was used to dissolve samples for analysis by RP-HPLC. The gradient elution system used was: solvent A, 25 mM NH4HCO3, pH 8.0; solvent B, 100% (v/v) acetonitrile. A linear gradient was run from 0% to 100% B over a period of 40 min, after which 100% B was then pumped through for an additional 5 min. All solvents were filtered and degassed before use, and the samples were passed through a 0.45 μm filter before injection, essentially as described by Zhang and Poulter .
The resulting fractions were vacuum evaporated and resuspended in scintillation liquid. Radioactivity quantification (c.p.m.) was performed in a Beckman 5000 β-radiation scintillation counter.
Purification and identification of dephosphorylated polyisoprenoid products (≥C40) by RP-HPLC
After the enzymatic reaction with either the recombinant or the native partially purified PfOPPs, the polyprenyl diphosphates were extracted with butan-1-ol and evaporated under a stream of nitrogen. The resulting polyprenyl diphosphates were treated with potato acid phosphatase according to the method of Fujii et al. , and submitted to analytical RP-HPLC separation performed in the same Gilson apparatus as described above, on a Phenomenex Luna C18 column (250 mm×4.6 mm). Methanol/water (9:1, v/v) (solvent A) and hexane/propan-2-ol/methanol (1:1:2, by vol.) (solvent B) were used as solvents. A linear gradient was run from 5% to 100% B over a period of 25 min; 100% B was then pumped through for an additional 5 min. The flow rate was 1.5 ml/min, using a method adapted from that of Low et al. .
Q-TOF (quadrupole–time-of-flight)-MS investigation of the first intermediate product (GGPP)
The samples purified by RP-HPLC with the retention time of the GGPP standard were analysed by ESI (electrospray ionization)-MS with direct infusion of an equimolar acetonitrile/water solution using a Harvard Syringe pump (model 11) and a Q-TOF mass spectrometer (Waters-Micromass) in the negative ion mode. Main conditions were as follows: flow rate, 5 μl/min; dissolvation temperature, 100 °C; block temperature, 100 °C; capillary voltage, 3000 V; cone voltage, 30 V. The ESI-MS experiments were performed using the high-resolution TOF mass analyser scanning in the range m/z 50–1000. For the acquisition of ESI tandem mass spectra (ESI-MS/MS), the precursor ion was selected using the quadrupole analyser, and dissociated by collisions with argon in the hexapole collision cell at energies varying from 15 to 45 eV. The product ion mass spectra were acquired by TOF-MS analysis.
Ion-trap MS investigation of dephosphorylated polyisoprenoid products
Ion-trap ESI-MS and ESI-MS/MS data were collected using a Finnigan LCQ-Duo ion-trap mass spectrometer (Thermo-Finnigan, San Jose, CA, U.S.A.). The samples of dephosphorylated polyisoprenoids purified by RP-HPLC corresponding to the C40, C45 and C55 standards were resuspended in 30 μl of chloroform/methanol (1:1, v/v) containing 2 mM lithium iodide. The samples (10 μl) were infused directly into the ESI probe across a 10 μl loop, using an Omnifit N2 pressure system (Omnifit Ltd., Cambridge, U.K.) with 69 kPa (10 lb/in2) pressure at a flow rate of 10 μl/min, applying the same diluents as used for the sample. The C40, C45 and C55 dephosphorylated polyisoprenoids were run in the ESI positive-ion mode, with spray voltage, capillary voltage and capillary temperature set at 4.52 kV, 17 V and 250.0 °C respectively, and mass spectra were acquired in SIM (selected ion monitoring) mode in the m/z ranges 570±10 for C40, 637±3 for C45 and 773±5 for C55. For ESI-MS/MS, a relative collision energy of 40% (2 eV) was applied, and the sheath (N2) and collision helium gas pressures were 27.6 kPa (4 lb/in2) and 0.2 Pa (1.5 mtorr) respectively. The MS/MS spectra were acquired in full ion mode in the m/z ranges 500–600 for C40, 500–700 for C45 and 700–800 for C55. These parameters were optimized for the highest intensity of the [M+Li]+ ion.
Inhibition of PfOPPs activity
The terpene nerolidol was preincubated at 10, 50 or 100 nM with TPfOPPs (truncated recombinant PfOPPs) for 30 min at 30 °C (a parallel control reaction was assayed without nerolidol). After preincubation, the reaction was performed by fixing the [1-14C]IPP concentration at 200 μM (corresponding to 3.3 μCi/ml) and varying the FPP concentration (2, 5, 10 and 50 μM) in HKMT buffer. The reactions were performed as described above. The products were then extracted with 1 ml of butan-1-ol and the radioactivity in the extract was measured in the c.p.m. mode with a Beckman 5000 β-radiation scintillation counter. Inhibition data were fitted to the competitive inhibition model using the software Enzfitter for Ki calculations (mean±S.E.M.) based on double-reciprocal Lineweaver–Burk plots.
A control assay with FTS (S-farnesylthiosalicylic acid) was performed at three different concentrations: 7 μM (in vivo IC50 value), 14 μM and 21 μM. Nerolidol was diluted in methanol and FTS was diluted in chloroform at 0.1 M (stock solution) essentially as described in Rodrigues Goulart et al. .
Identification and cloning of a putative
P. falciparum polyprenyl synthase
In the P. falciparum genome, a gene encoding a putative P. falciparum polyprenyl synthase was predicted and annotated on chromosome 2 (PFB0130w). The alignment of this sequence with those of OPPs enzymes of Thermotoga maritima, OPPs of Escherichia coli, solanesyl pyrophosphate synthase of Mucor circinelloides, decaprenyl diphosphate synthase of Schizosaccharomyces pombe, and GGPP synthase and FPP synthase of Thermoplasma volcanium revealed a degree of similarity. Primers specific for the PFB0130w gene were designed and used to amplify by PCR the corresponding gene, and its sequence was confirmed by DNA sequencing. The deduced amino acid sequences of the putative P. falciparum polyprenyl synthase and several trans-prenyltransferases showed sequence identity, with two common DDXXD motifs. These conserved aspartate-rich regions have been associated with OPPs activity .The P. falciparum polyprenyl synthase has also a phenylalanine at position 132, (corresponding to the position found in T. maritima OPPs), which was determined as being responsible for chain isoprenoid elongation to C40 in the T. maritima enzyme (Figure 1A). The P. falciparum polyprenyl synthase sequence shows 18.67% similarity with the sequence of the corresponding human enzyme (similarity matrix not shown).
Alignment, transcription, expression and immunoprecipitation of PfOPPs
The putative PfOPPs is transcribed and expressed in intraerythrocytic stage parasites
RT-PCR assays from purified ring, trophozoite and schizont stage parasites showed the presence of PFB0130w transcripts in ring and trophozoite stages, but not in schizont stages, suggesting that the gene is transcribed in a stage-specific manner (Figure 1B). To detect the protein corresponding to the PFB0130w gene, we expressed it as a recombinant GST fusion protein. We cloned a truncated version of the gene in the vector pGEX2T, and the resulting protein (putative TPfOPPs) was then used to immunize Balb/C mice. Notably, production of a full-length protein was impossible (results not shown). The truncated version showed a tendency for either premature arrest of translation or proteolysis, resulting in a co-purified GST portion (Figure 1C).
Using the putative TPfOPPs antiserum, we were able to immunoprecipitate a protein of the expected size of 63 kDa from parasites metabolically labelled with L-[35S]methionine. As demonstrated in Figure 1(D), only the putative TPfOPPs antiserum, but not the anti-GST serum, was able to immunoprecipitate the 63 kDa protein from trophozoite and in schizont stages.
Demonstration of PfOPPs activity and analysis of its substrate specificity by RP-TLC
In order to identify if P. falciparum extracts or the recombinant peptide showed any enzymatic activity, in vitro assays were performed with 10 μM IPP and 50 μM [3H]GGPP (7.9 μCi/ml) or 50 μM [3H]FPP (0.39 μCi/ml) as an allylic substrate under the optimum conditions described above. The reaction products were extracted, dephosphorylated and submitted to RP-TLC. A polyprenol of eight isoprenic units (C40), typical of OPPs activity, and another with nine isoprenic units (C45) were observed when [3H]FPP was used as a substrate with either TPfOPPs or native PfOPPs (Figure 2, lanes 1 and 3). When [3H]GGPP was used as substrate, no products were observed (Figure 2, lanes 2 and 4). Similar results were obtained when [1-14C]IPP and FPP or GGPP were used for analysis of substrate specificity (results not shown). Importantly, no synthesis of polyprenols occurred when either TPfOPPs or native PfOPPs was added (Figure 2, lane 5), confirming an enzymatic activity of both the semi-purified P. falciparum extract (Figure 2, lane 3) and the recombinant polypeptide (Figure 2, lane 1), and also supporting the preference of the PfOPPs for FPP as the main substrate. No band was detected when geranyl pyrophosphate was used in the assay instead of FPP (results not shown). Importantly, the dephosphorylation procedure causes a loss of short-chain isoprenoids, since the non-elongated, radioactive substrate GGPP was not present even in lanes 2 and 4 of the RP-TLC (co-migrating with the standard geranylgeraniol).
Analysis of substrate specificity and demonstration of PfOPPs activity by RP-TLC
Identification by RP-HPLC and MS of the first intermediate product of PfOPPs catalysis
In the next step, we characterized early intermediates of the synthase reactions. To this end, enzymatic reactions were performed with 10 μM [1-14C]IPP and 50 μM FPP in the presence of TPfOPPs (Figure 3A1) or native PfOPPs (Figure 3A2). The products produced by both versions of PfOPPs were then submitted to RP-HPLC purification. After extraction and RP-HPLC analysis, a major peak with a retention time of 23 min, coinciding with standard GGPP, was observed for the labelled products. To confirm that FPP was the substrate for OPPs, we conducted the same reaction with geranyl pyrophosphate as substrate. Under these conditions no peak was detected at this retention time for either TPfOPPs or native PfOPPs (results not shown), confirming the RP-TLC results (Figure 2).
Identification by RP-HPLC and MS of the first intermediates in the reaction catalysed by PfOPPs
In order to identify the molecules with a retention time of 23 min in the RP-HPLC analysis, we conducted the same enzymatic reaction with non-radioactive substrates (IPP and FPP) in the presence of TPfOPPs, and analysed the products by ESI-Q-TOF-MS. As indicated by the great similarity between the ESI tandem mass spectra shown in Figures 3(B1) and 3(B2), the eluting compound was confirmed to be GGPP. As seen in both spectra, deprotonated GGPP of m/z 449.5 dissociated upon 35 eV energy collisions mainly by processes driven by the pyrophosphate group (Figure 3B3). Under the same conditions, the MS/MS spectrum of GGPP presented the same fragments (Figure 3B1), confirming its identity with the standards (Figure 3B2).
Identification by RP-HPLC and MS of the production of polyisoprenoids of ≥C40 by PfOPPs
To determine if PfOPPs synthesizes products of C40 or larger, assays were carried out with 10 μM [1-14C]IPP and 50 μM FPP in the presence of TPfOPPs (Figure 4A1) or partially purified native PfOPPs (Figure 4A2). The products were analysed in their dephosphorylated forms. Three major peaks with retention times of 25, 26 and 30 min coincided with the authentic C40, C45 and C55 polyisoprenoid standards respectively. Peaks with retention times between 5 and 20 min may correspond to smaller polyisoprenoids, and were not characterized further. The structures of the purified compounds detected as peaks 1, 2 and 3 were then investigated by MS. The ESI mass and tandem mass spectra shown in Figures 4(B1) and 4(B2) reveal the presence of a C40 polyisoprenoid (monoisotopic mass 562.7 Da), represented by the major single-charged lithium adduct ([M+Li]+) ion species at m/z 569.4 in reactions with both TPfOPPs (Figure 4B1-II) and native PfOPPs (Figure 4B1-III). This same ion species was detected when the authentic C40 polyisoprenoid standard (Figure 4B1-I) was analysed. The molecular identity was confirmed by comparing the ESI-MS/MS spectrum of the ion of m/z 568.9 [M+Li]+ produced by TPfOPPs (Figure 4B2-II) and native PfOPPs (Figure 4B2-III) with the ESI-MS/MS spectrum of the standard (Figure 4B2-I), revealing the same dissociation profile, with a major fragment ion at m/z 551.2 [M+Li−H2O]+. Accordingly, Figures 4(C1) and 4(C2) show the presence of a C45 polyisoprenoid (monoisotopic mass 631.0 Da), represented by the major single-charged lithium adduct ([M+Li]+) ion species at m/z 637.6 in reactions with TPfOPPs (Figure 4C1-II). This same ion species was detected when the authentic C45 polyisoprenoid standard (Figure 4C1-I) was analysed. The molecular identity was confirmed by comparing the ESI-MS/MS spectrum of the ion of m/z 637.3 of [M+Li]+ from the TPfOPPs reaction (Figure 4C2-II) with the ESI-MS/MS spectrum of the standard (Figure 4C2-I), revealing the same dissociation profile with the major fragment ion of m/z 619.4 [M+Li−H2O]+. Probably due to the low quantities of the C45 polyisoprenoid in the reaction with native PfOPPs, we were unable to detect this molecule.
Identification by RP-HPLC and MS of polyisoprenoids of C40, C45 and C55 formed in the reaction catalysed by PfOPPs
Similarly, in Figures 4(D1) and 4(D2) the presence of a C55 polyisoprenoid (monoisotopic mass 767.2 Da) is shown, represented by the peak of m/z 773.2 [M+Li]+ for the reactions with both TPfOPPs (Figure 4D1-II) and PfOPPs (Figure 4D1-III), as well as the similar ionization of an authentic polyisoprenoid C55 standard (Figure 4D1-I). The molecular identity was confirmed by comparison of the MS/MS spectrum of the ion of m/z 773.4 of [M+Li]+ of TPfOPPs (Figure 4D2-II) and PfOPPs (Figure 4D2-III) with the MS/MS spectrum of the standard (Figure 4D2-I), revealing the same dissociation in the major fragment ion of m/z 755.6 [M+Li−H2O]+ and another minor ion of m/z 743.5 [M+Li−30]+.
Kinetic parameters of TPfOPPs and native PfOPPs
To characterize the enzymatic activity of TPfOPPs and partially purified native PfOPPs, we conducted kinetic characterization and measured the products of reactions catalysed by the two enzymes. Addition of TPfOPPs or PfOPPs to reaction mixtures containing [1-14C]IPP and FPP led to the formation of polyisoprenoids, whereas without this addition no product was observed.
The Lineweaver–Burk plot in Figure 5 shows the similarity between the Michaelis constants for TPfOPPs (IPP, 2.4±0.2 μM; FPP, 1.1±0.1 μM) and PfOPPs (IPP, 3.1±0.4 μM; FPP, 2.9±0.7 μM). Given the kinetic similarity between TPfOPPs and native PfOPPs, subsequent inhibition assays were conducted only with the recombinant version.
Lineweaver–Burk plots of activity data for native PfOPPs and TPfOPPs with different concentrations of IPP or FPP
Inhibition of PfOPPs by the terpene nerolidol
To elucidate if PfOPPs can be inhibited by allylic isoprene analogues, we conducted enzymatic reactions using FPP and [1-14C]IPP as substrates in the presence of 10, 50 and 100 nM nerolidol. The products of these reactions were then analysed by HPLC and β-scintillation counting. In a parallel reaction, FTS was used as a control at concentrations of 7 μM (the in vivo IC50 value), 14 μM and 21 μM (results not shown). Catalysis by TPfOPPs in the presence of 200 μM [1-14C]IPP and 1 μM FPP was inhibited up to 68% when 100 nM nerolidol was added. A maximal inhibition of 57% was observed at an inhibitor concentration of 50 nM, whereas when nerolidol was present at 10 nM, no consistent inhibition was observed (results not shown). The Ki of nerolidol, as calculated from Lineweaver–Burk plots, was 15±6 nM (Figures 6A and 6B) and the type of inhibition was clearly competitive within this inhibitor concentration range.
Inhibition of TPfOPPs by nerolidol
Due to their absence in humans, the enzymes of the 2-C-methyl-D-erythritol-4-phosphate pathway of isoprenoid precursor biosynthesis have been identified as an interesting target for the development of novel antibiotics, as well as antimalarial and herbicidal agents. In view of the differences observed in the length of the isoprenoid side-chain compounds encountered in humans and the malaria parasite Plasmodium falciparum, we asked if the enzymes that finally lead to the synthesis of medium- and/or long-chain isoprenoids could also be exploited as drug targets. Herein, we demonstrate the existence of an OPPs in P. falciparum. This is the first report that directly demonstrates an OPPs activity in parasitic protozoa and the inhibition of this activity by the terpene nerolidol.
The deduced amino acid sequences of PfOPPs and several trans-prenyltransferases showed some amino acid sequence similarity, with the presence of two common DDXXD motifs that are characteristic of OPPs activity among various organisms. There was significant divergence in N-terminal regions, which are assumed to be signal sequences responsible for different subcellular localizations .
The first DDXXD motif is responsible for binding of and interaction with FPP, and the second motif for IPP binding [15,38]. In the archaeon T. maritima, several key amino acids were identified, on the basis of the conformational structure of OPPs, that play critical roles in regulating the chain length of the final products. These include Ala-76 and Ser-77 for C20 products. Phe-132 was found to be crucial for the limited elongation of isoprenic chains to the correct C40 product. In the P. falciparum version of the enzyme, only Phe-132 seems to be conserved. Despite limited identity between T. maritima OPPs and PfOPPs, Phe-132 may also play this role in PfOPPs, since de Macedo et al.  and Rodrigues Goulart et al.  detected isoprenoids from C35 to C45 from parasites metabolically labelled with [1-14C]sodium acetate in in vitro cultures. Notably, the P. falciparum version of the protein sequence shows only 18.67% similarity with the human version, perhaps explaining the differences in isoprenic side-chain lengths between P. falciparum and humans.
The transcription of PfOPPs was found to occur in the ring and trophozoite stages of the parasite, and the polypeptide was mainly found in schizont-stage parasites, as detected by immunoprecipitation assays. These results are in accordance with previous transcriptome data from Le Roch et al.  and proteomics data published by Nirmalan et al.  and Florens et al. . Additionally, our group demonstrated that CoQ7–CoQ9 are biosynthesized predominantly in the trophozoite and schizont stages [10,37].
To confirm that the recombinant and native P. falciparum proteins possessed OPPs-like enzymatic activity, we characterized both versions. To define substrate specificity, the condensation products of the reaction of [3H]FPP with unlabelled IPP were analysed by RP-TLC, and corresponded to a C40 polyisoprenoid and longer polymerization products. These products were observed when using both versions of PfOPPs. Similar to the T. maritima and E. coli enzymes, no products were detected when [3H]GGPP was used as a substrate .
The OPPs activity was confirmed by purification of the synthesized products by RP-HPLC. When the products were analysed in their dephosphorylated forms, three major peaks, corresponding to C40, C45 and C55 polyisoprenoids, were detected. These results were confirmed by MS and tandem MS analysis. In this analysis we used lithium iodide for the ionization of polyisoprenoids instead of sulphates or derivatization to t-butyldimethylsilyl ethers [42,43], since the yield of derivated dolichols and polyisoprenoids using the latter derivatization compounds is very low.
The atypical C55 product detected in enzymatic reactions may have been synthesized due to the presence of an excess of precursor molecules. Similar products were also found during in vitro experiments with native OPPs from T. maritima . Again, both the recombinant and native versions of the P. falciparum enzyme showed the same product spectrum of C40, C45 and C55 isoprenoids.
We tried to express the complete sequence of PfOPPs in several expression vectors, but the polypeptide was not expressed, suggesting that the 3′-end of the gene contains regulatory sequences, as seen in the dihydrofolate reductase gene . The portion of the polypeptide expressed as a GST-fusion protein contains the important regions for substrate binding and catalytic activity. Additionally, parameters such as the spectrum of products formed by the recombinant version showed marked similarity to that of the native, partially purified PfOPPs from schizont-stage parasites.
The chemical structure of the terpene nerolidol resembles that of the substrate FPP. Previously, our group demonstrated that P. falciparum parasites treated with nerolidol showed a reduced ability to synthesize CoQ in all intra-erythrocytic stages, and inhibited P. falciparum growth in vitro by decreasing the progression of the ring to the trophozoite stage . This drug also suppressed Leishmania amazonensis promastigotes in vitro and intracellular amastigote growth by inhibiting the synthesis of dolichol and ergosterol . In the present study, a clear competitive inhibitory action of nerolidol on PfOPPs was detected. In contrast, when we used FTS at a concentration three times higher than the IC50, product formation was the same as in the control reactions without this drug. These results reinforce the specificity of PfOPPs for FPP.
Another aspect of the inhibition assay was to verify the viability of the enzyme as an antimalarial target. Several observations support this idea. First, the isoprenoid side chains attached to the benzoquinone ring in mammalian cells are longer than those in P. falciparum; thus the enzyme involved in side-chain synthesis must be different. Accordingly, the sequence similarity between the human decaprenyl pyrophosphate synthase and the plasmodial OPPs is very low. Secondly, the CC50 (concentration at which the number of cells or cell proliferation is reduced by 50%) of nerolidol in mammalian cells was determined to be approx. 180 times higher than the IC50 in P. falciparum , and the free plasmodial enzyme can be inhibited efficiently by low concentrations of this terpene.
In conclusion, the present study demonstrated that P. falciparum has an OPPs activity. Further structural studies to compare the active sites of human and malarial enzymes may reveal possible inhibitors specific for PfOPPs.
We thank Dr Walter R. Terra (Department of Biochemistry, Institute of Chemistry, University of São Paulo) for numerous helpful comments and suggestions, Dr Tadeuz Chojnaki (Institute of Biochemistry and Biophysics, Polish Academy of Sciences, Warsaw, Poland) for the gift of authentic prenol/prenyl phosphate standards, Maria Belen Cassera for help in HPLC purification, Evandro Luiz Duarte and Antonio Carlos Franco da Silveira for help with ImageQuant analyses and Dr Silvano Wendel (Sírio Libanês Hospital, NESTA, São Paulo, Brazil) for the gift of erythrocytes. R.T., F.L.D. and F.A.G. are the recipients of a FAPESP scholarship. This research was supported by grants from the Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP), the Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq), PRONEX and UNDP/World Bank/WHO.
octaprenyl pyrophosphate synthase
Plasmodium falciparum OPPs
truncated recombinant version of P. falciparum OPPs