Charles M. Allen

Protein farnesyltransferase and protein prenylation in Plasmodium falciparum.

Comparison of the malaria parasite and mammalian protein prenyltransferases and their cellular substrates is important for establishing this enzyme as a target for developing antimalarial agents. Nineteen heptapeptides differing only in their carboxyl-terminal amino acid were tested as alternative substrates of partially purified Plasmodium falciparum protein farnesyltransferase. Only NRSCAIM and NRSCAIQ serve as substrates, with NRSCAIM being the best. Peptidomimetics, FTI-276 and GGTI-287, inhibit the transferase with IC50 values of 1 and 32 nm, respectively. Incubation of P. falciparum-infected erythrocytes with [3H]farnesol labels 50- and 22–28-kDa proteins, whereas [3H]geranylgeraniol labels only 22–28-kDa proteins. The 50-kDa protein is shown to be farnesylated, whereas the 22–28-kDa proteins are geranylgeranylated, irrespective of the labeling prenol. Protein labeling is inhibited more than 50% by either 5 μm FTI-277 or GGTI-298. The same concentration of inhibitors also inhibits parasite growth from the ring stage by 50%, decreases expression of prenylated proteins as measured with prenyl-specific antibody, and inhibits parasite differentiation beyond the trophozoite stage. Furthermore, differentiation specific prenylation of P. falciparum proteins is demonstrated. Protein labeling is detected predominantly during the trophozoite to schizont and schizont to ring transitions. These results demonstrate unique properties of protein prenylation in P. falciparum: a limited specificity of the farnesyltransferase for peptide substrates compared with mammalian enzymes, the ability to use farnesol to label both farnesyl and geranylgeranyl moieties on proteins, differentiation specific protein prenylation, and the ability of peptidomimetic prenyltransferase inhibitors to block parasite differentiation.

Protein prenyl transferase activities of Plasmodium falciparum

Prenylated proteins have been shown to function in important cellular regulatory processes including signal transduction. The enzymes involved in protein prenylation, farnesyl transferase and geranylgeranyl transferase, have been recent targets for development of cancer chemotherapeutics. We have initiated a systematic study of protein prenyl transferases of the malaria parasite, Plasmodium falciparum, to determine whether these enzymes can be developed as targets for antimalarial chemotherapy. We report here the identification of protein farnesyl transferase and protein geranylgeranyl transferase-I in the malaria parasite, P. falciparum. The farnesyl transferase has been partially purified from the cytosolic fraction through ammonium sulfate precipitation and Mono-Q chromatography. Farnesyl and geranylgeranyl transferase-I activities are present at all stages of P. falciparum intraerythrocytic development with maximum specific activity in the ring stage. Geranylgeranyl transferase-I specific activity is two times that of farnesyl transferase in the ring stage. Peptidomimetics and prenyl analogues of protein farnesyl transferase substrates were tested as in vitro inhibitors of partially purified P. falciparum prenyl transferase and of malaria parasite growth. The peptidomimetics were significantly more potent inhibitors than lipid substrate analogues of both the activity of Mono-Q purified enzyme and parasite growth in intraerythrocytic cultures. Exposure of the parasite to the peptidomimetic L-745,631 also showed significant inhibition of morphological development beyond the trophozoite stage. These studies suggest the potential of designing or identifying differential inhibitors of P. falciparum and mammalian prenyl transferases as an approach to novel malaria therapy.

Prenyl transferases from Micrococcus luteus: Characterization of undecaprenyl pyrophosphate synthetase☆

Abstract Three prenyl transferases in Micrococcus luteus were recovered in the soluble fraction following cell disruption. Undecaprenyl pyrophosphate (C 55 -PP) synthetase chromatographed on DEAE-cellulose independently from geranylgeranyl-PP and octaprenyl-PP synthetases. Further purification of C 55 -PP synthetase resulted in an approximate 250-fold purification over the crude lysate. The molecular weight of the synthetase was estimated to be between 47,000 and 49,000 by Sephadex G-100 chromatography. The enzyme had a broad specificity toward the allylic pyrophosphate substrate. The reactivities of the allylic substrates increased with chain length, C 10 15 20 , except for trans -solanesyl-PP, which was unreactive. Moreover, the enzyme was active on allylic substrates having both cis - and trans -stereochemistry. Although C 55 -PP and C 50 -PP were the major products, some shorter chain products were also produced, when t,t -farnesyl pyrophosphate and Δ 3 sopentenyl pyrophosphate (IPP) were used as substrates. The stereochemistries of the products formed with C 55 -PP synthetase were established, using [ 14 C]IPP and 2 R -[2- 3 H] and 2 S -[2- 3 H]IPP. Each new isoprene unit added had a cis -configuration. The enzyme was inactive in the absence of added effectors. It was stimulated by Triton X-100, egg lecithin, and a whole phospholipid extract from M. luteus . Cardiolipin and deoxycholate were poor activators of the enzyme. The product chain length distribution observed with the phospholipid-activated enzyme showed highly favored production of the C 55 -PP product over the C 50 -PP product.

Variable product specificity of solanesyl pyrophosphate synthetase

Abstract The distribution of polyprenyl pyrophosphates synthesized by the action of solanesyl pyrophosphate synthetase from Micrococcus luteus is dramatically changed depending on the Mg++ concentration. When the metal ion concentration is higher than 5 mM, octaprenyl and solanesyl (nonaprenyl) pyrophosphate are synthesized predominantly. On the other hand, when the metal ion level is lower than 0.5 mM, a variety of polyprenyl pyrophosphates ranging in carbon chain length from C15 to C40 are formed. Heptaprenyl pyrophosphate is the longest of the products formed at 0.1 mM of Mg++.

Undecaprenyl pyrophosphate synthetase from Lactobacillus plantarum: A dimeric protein

a++Undecaprenyl pyrophosphate synthetase has been purified from Lactobacillus plantarum. It catalyzes the formation of a C55 polyprenyl pyrophosphate having isoprene residues with cis stereochemistry. The enzyme was shown to be an acidic protein (pI = 5.1), which can be partially purified by preparative gel electrophoresis and Blue-agarose column chromatography. The Kms of the enzyme for its substrates t,t-farnesyl pyrophosphate and isopentenyl pyrophosphate were determined to be 0.13 and 1.92 microM, respectively. The molecular weight of the enzyme was estimated by molecular sieve chromatography and gradient centrifugation to be 56,000 +/- 4000. Analysis by sodium dodecyl sulfate-polyacrylamide gel electrophoresis indicated that the protein was composed of a dimer of 30,000-Da subunits. The enzyme was inactivated by the arginine-specific reagents phenylglyoxal, butanedione and, cyclohexanedione, but this inactivation was not prevented by either of the substrates.

Inhibition of farnesyl transferases from malignant and non-malignant cultured human lymphocytes by prenyl substrate analogues

Cytosolic prenyl transferases from two human lymphoid tissue-derived cell lines, IM-9 and Molt-4 cells, are shown to isoprenylate recombinant p21H-ras. Isoprenylation was inhibited by an N-acetylated pentapeptide (N-Ac-Lys-Cys-Val-Leu-Ser), c,t-farnesyl diphosphate, c,t,t-geranylgeranyl diphosphate, t,t,t-geranylgeranyl diphosphate and a photolabile farnesyl diphosphate analogue. c,t-Farnesyl and t,t,t-geranylgeranyl monophosphates were also effective inhibitors of the Molt-4 enzyme but not the IM-9 enzyme.

Changes in dehydrodolichyl diphosphate synthase during spermatogenesis in the rat

The levels of dolichyl phosphate and 2,3-dehydrodolichyl diphosphate synthase were determined in seminiferous tubules of prepuberal rats to assess any changes occurring during early stages of spermatogenesis. Dolichyl phosphate increased in concentration two- to threefold from Day 10 to Day 23 after birth. A method was optimized to measure dehydrodolichyl diphosphate synthesis from delta 3-[14C]isopentenyl diphosphate and t,t-farnesyl diphosphate in homogenates of seminiferous tubules. Both dehydrodolichyl mono- and diphosphates were observed as products of the in vitro assay. The specific activity of tubular synthase increased twofold between Day 7 and Day 23 and decreased similarly between Day 23 and Day 60. Since there was a parallel increase in the concentration of tubular dolichyl phosphate and dehydrodolichyl diphosphate synthase activity during early stages of spermatogenesis, it is proposed that the level of dolichyl phosphate may be controlled at least in part by the regulation of de novo dehydrodolichyl diphosphate biosynthesis. The synthase was also solubilized from tubular membranes with deoxycholate and partially purified by chromatography.

Evidence for Prenylation-Dependent Targeting of a Ykt6 SNARE in Plasmodium falciparum

Ykt6 proteins are the most versatile fusogens in eukaryotic cells, and the only SNAREs that can be both prenylated and acylated at a C-terminal CAAX motif. Unlike yeast and mammalian cells where a single Ykt6 gene is expressed, the Plasmodium falciparum genome encodes two Ykt6 proteins. We have investigated the expression and prenylation of the Ykt6 orthologue, PfYkt6.1 in intra-erythrocytic stages of P. falciparum. PfYkt6.1 localized to the parasite Golgi and other unidentified cytoplasmic compartments, and was partly cytosolic (∼50% in early trophozoites). The membrane-association of PfYkt6.1 was dependent on the presence of a conserved C-terminal CAAX motif (CCSIM). By expressing full-length and mutant proteins in Escherichia coli, we have shown that PfYkt6.1 indeed serves as substrate for prenylation by P. falciparum farnesyltransferases. Surprisingly, PfYkt6.1 could also be geranylgeranylated by parasite extracts independent of the C-terminal amino acid residue. Deletion of the CAAX motif inhibited both farnesylation and geranylgeranylation activities. Additionally, the PfYkt6.1 heptapeptide KQCCSIM, corresponding to the C-terminal CAAX sequence, inhibited the parasite farnesyltransferase activity with an IC(50) of 1 μM. Our findings underscore the importance of CAAX motif-derived peptidomimetics for antimalarial drug development.

Dehydrodolichyl diphosphate synthetase from rat seminiferous tubules.

Homogenates of seminiferous tubules from rat testes catalyzed the incorporation of label from [14C]isopentenyl diphosphate into a variety of polyprenyl products. Long chain polyprenyl mono- and diphosphates were formed as major products when undesirable side reactions were minimized. The long chain polyprenyl diphosphate synthetase was measured as a sum of the mono- and diphosphate derivatives formed and was dependent on the addition of t,t-farnesyl diphosphate, isopentenyl diphosphate, and divalent cation. The highest activity was associated with the membranous fractions, whereas activity was negligible in the cytosolic fraction. The products of this prenyl transferase were labile to acid and yielded petroleum ether soluble products which indicated that the alpha-isoprene unit was unsaturated. Hydrolysis of either the polyprenyl mono-or diphosphates with a testicular phosphatase in the absence of NaF yielded C75, C80, C85, and C90 polyprenols. The chain lengths of the products of the synthetase suggest that this enzyme is responsible for the de novo biosynthesis of dehydrodolichyl diphosphates which are precursors of the dolichyl derivatives found in testes.

Lipid activation of undecaprenol kinase from Lactobacillus plantarum.

Extraction of membranes of Lactobacillus plantarum with Triton X-100/glycerol solubilized up to 80% of the undecaprenol kinase activity. Fractionation of the extract by gel chromatography separated endogenous phospholipid from the enzyme but simultaneously inactivated the enzyme. The kinase was reactivated by reconstitution with various synthetic phosphatidylcholines and purified L. plantarum phospholipids. Ditetradecanoylphosphatidylcholine and lysylphosphatidylglycerol were the best activators. Furthermore, the optimal environment for enzyme stimulation was provided by different defined molar ratios of Triton X-100/phospholipid. The ratios for the phospholipids tested ranged from 1.25 to 6.3. Similar substrate specificity and kinetic constants were observed for both the solubilized and reconstituted enzymes suggesting that no fundamental changes in the enzyme activity occurred during the delipidation-reconstitution process.