Identification, molecular cloning and functional characterization of an octaprenyl pyrophosphate synthase in intra-erythrocytic stages of Plasmodium falciparum

Abstract

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.

Figure 1. Alignment, transcription, expression and immunoprecipitation of PfOPPs

(A) Alignment of amino acid sequences of prenyltransferases. Black and white lettering on a grey background indicates identical and similar amino acids residues respectively. The two conserved DDXXD motifs are indicated. The star indicates a phenylalanine residue characteristic of OPPs activity. Aligned sequences are for the putative polyprenyl synthase of P. falciparum (Falciparum; PlasmoDB gene PFB0130w), OPPs enzymes of Thermotoga maritima (Maritima; PDB code 1V4E) and Escherichia coli (Coli; GenBank® accession number P19641), solanesyl pyrophosphate synthase of Mucor circinelloides (Mucor; GenBank® accession number CAD42868.1), decaprenyl diphosphate synthase of fission yeast (Yeast; GenBank® accession number O43091), and GGPP synthase (GGPPsther; GenBank® accession number NP_394768.1) and FPP synthase (FPPsther; GenBank® accession number BAB59406.1) of Thermoplasma volcanium. (B) RT-PCR detection of putative PfOPPs transcripts. B1, control with myosin-specific oligonucleotides; B2, cDNA from different stages [ring (R), trophozoite (T) and schizont (S)] was amplified with sense and antisense PfOPPs primers. Genomic DNA of P. falciparum (isolate NF54; clone3D7) was used as a positive control (g); +, and − denote the presence and absence respectively of reverse transcriptase. Fragment sizes are indicated on the right. (C) Expression assays of the putative PfOPPs by SDS/12.5%-PAGE. TPfOPPs, TPfOPPs–GST fusion protein; GST, control. (D) Immunoprecipitation of parasites metabolically labelled with L-[³⁵S]methionine using putative TPfOPPs antiserum. Proteins extracts from ring (R), trophozoite (T) and schizont (S) stages were separated by discontinuous Percoll® gradients, and immunoprecipitated using sera of mice immunized with GST alone as a control (1) or TPfOPPs (2), or mock-immunized as a negative control (3). The arrows indicate the protein with the expected molecular mass immunoprecipitated from trophozoite and schizont extracts using the putative TPfOPPs–GST antiserum.

Figure 2. Analysis of substrate specificity and demonstration of PfOPPs activity by RP-TLC

The enzymatic reaction, the acid hydrolyses and TLC were performed as described in the Experimental section. Lanes 1 and 3, [³H]FPP as substrate; lanes 2 and 4, [³H]GGPP as substrate; lane 5, negative control reaction without enzymes ([³H]FPP as substrate). Lanes 1 and 2, TPfOPPs; lanes 3 and 4, native PfOPPs. The positions of various prenol plus geranylgeraniol and farnesol standards are indicated on the left.

Figure 3. Identification by RP-HPLC and MS of the first intermediates in the reaction catalysed by PfOPPs

HPLC was used to purify the first product (GGPP) in the reaction catalysed by TPfOPPs (A1) and partially purified native PfOPPs (A2). GGPP, detected as a peak with the retention time of 23 min in the RP-HPLC analysis, was then characterized by Q-TOF MS. The ESI mass spectrum (B1) confirms the detection of the polyisoprenoid GGPP (as the deprotonated molecule of m/z 449.5) as the first intermediate in reactions catalysed by TPfOPPs. As shown by ESI tandem mass spectra (B1 and B2 for the standard), GGPP dissociates upon 35 eV collisions by routes driven by the pyrophosphate group (B3).

Figure 4. Identification by RP-HPLC and MS of polyisoprenoids of C₄₀, C₄₅ and C₅₅ formed in the reaction catalysed by PfOPPs

RP-HPLC was used to purify the dephosphorylated products (C₄₀, C₄₅ and C₅₅) of the reactions catalysed by TPfOPPs (A1) and partially purified native PfOPPs (A2). Arrows indicate the elution positions of authentic isoprenoid standards: 1, C₄₀; 2, C₄₅; 3, C₅₅. ESI ion-trap MS was then used to analyse the dephosphorylated polyisoprenoid products (C₄₀, C₄₅ and C₅₅). (B1, B2) ESI mass spectra show a C₄₀ polyisoprenoid as a major singly charged [M+Li]⁺ ion of m/z 568.3 in both the TPfOPPs (B1-II) and native PfOPPs (B1-III) reactions, as for the C₄₀ polyisoprenoid standard (B1-I). The molecular structure of this product was confirmed by comparison of the tandem mass spectrum of the [M+Li]⁺ ions of m/z 568.9 formed in both the TPfOPPs (B2-II) and native PfOPPs (B2-III) reactions with that of the standard (B2-I). The major fragment ion in these spectra is [M+Li−H₂O]⁺ of m/z 551.2. Panels (C1) and (C2) show the presence of a C₄₅ polyisoprenoid, represented by the major singly charged lithium adduct ([M+Li]⁺) ion species at m/z 637.6 in reactions with TPfOPPs (C1-II). This same ion species was detected when the authentic polyisoprenoid standard of C₄₅ (C1-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−H₂O]⁺ formed in the TPfOPPs reaction (C2-II) with that of the standard (C2-I), revealing the same dissociation profile with a major fragment ion of m/z 619.4 [M+Li−H₂O]⁺. (D1, D2) ESI mass spectra show a C₅₅ polyisoprenoid as a major singly charged [M+Li]⁺ ion of m/z 773.2 from both the TPfOPPs (D1-II) and native PfOPPs (D1-III) reactions, as for the C₅₅ polyisoprenoid standard (D1-I). The molecular structure of the product was confirmed by comparison of the tandem mass spectrum of the [M+Li]⁺ ions of m/z 773.5 formed in both the TPfOPPs (D2-II) and native PfOPPs (D2-III) reactions with that of the standard (D2-I). The major fragment ion in these spectra is [M+Li−H₂O]⁺ of m/z 755.6.

Figure 5. Lineweaver–Burk plots of activity data for native PfOPPs and TPfOPPs with different concentrations of IPP or FPP

Intercepts at the x-axis denote similar Michaelis constants for the native and recombinant forms of the enzyme, for binding of both substrates.

Figure 6. Inhibition of TPfOPPs by nerolidol

(A) Fractional inhibition by 0, 50 and 100 nM nerolidol in the presence of different concentrations of FPP (1, 2, 5, 10 and 50 μM). The data are shown as Lineweaver–Burk plots. (B) Replot of inhibition data. Points are experimental data, and curves were obtained by fitting the data to competitive inhibition equations (least-squares method).