Properties of Phosphoenolpyruvate Mutase, the First Enzyme in the Aminoethylphosphonate Biosynthetic Pathway in Trypanosoma cruzi

Phosphoenolpyruvate (PEP) mutase catalyzes the conversion of phosphoenolpyruvate to phosphonopyruvate, the initial step in the formation of many naturally occurring phosphonate compounds. The phosphonate compound 2-aminoethylphosphonate is present as a component of complex carbohydrates on the surface membrane of many trypanosomatids including glycosyl-inositolphospholipids of Trypanosoma cruzi . Using partial sequence information from the T. cruzi genome project we have isolated a full-length gene with significant homology to PEP mutase from the free-living protozoan Tetrahymena pyriformis and the edible mussel Mytilus edulis . Recombinant expression in Escherichia coli con-firms that it encodes a functional PEP mutase with a K m apparent of 8 (cid:1) M for phosphonopyruvate and a k cat of 12 s (cid:2) 1 . The native enzyme is a homotetramer with an abso-lute requirement for divalent metal ions and displays negative cooperativity for Mg 2 (cid:3) (S 0.5 0.4 (cid:1) M ; n (cid:4) 0.46). Immunofluorescence and sub-cellular fractionation indicates that PEP mutase has a dual localization in the cell. Further evidence to support this was obtained by Western analysis of a partial n M T. mutase, m M phosphono- pyruvate. K m for phosphonopyruvate was determined using this method by varying phosphonopyruvate concentration in the presence of 5 m M MgSO 4 or 5 m M magnesium acetate. Concentrations phosphono- pyruvate were standardized by spectrophotometric assay in the presence of excess PEP mutase. Protein concentration was determined based on the calculated extinction coefficient at 280 nm (34850 M (cid:2) 1 cm (cid:2) ). This method yields a 1.5-fold higher protein concentration than that determined by the bicinchoninic acid method (Pierce). One unit of enzyme activity is defined as 1 (cid:1) mol of NADH oxidized/min. Because Mg 2 (cid:1) is required for the coupled assay, the requirement for divalent metal ions was determined using a direct assay after the formation of phosphoenolpyruvate. Each 1-ml assay contained 50 m M (K (cid:1) ) HEPES, pH 8.0, 0.5 m M dithiothreitol, 450 n M T. cruzi recombinant PEP mutase, 1 m M phosphonopyruvate, and varying amounts of MgSO 4 . The production of phosphoenolpyruvate was assayed by the increase in absorbance at 233 nm using a

AEP was first identified as a minor component in acid hydrolysates of lipid and glycolipid extracts of the intracellular protozoan parasite Trypanosoma cruzi (5), the causative agent of Chagas' disease. Although dismissed as a minor component of lipopeptidophosphoglycan (now known as glycosylinositolphospholipid (GIPL)), subsequent studies indicate that AEP may be a universal component of the dense coat of glycoconjugates (mucins and GIPLs) that cover the surface of different stages of the parasite (6 -8). Mucins are highly glycosylated proteins that are anchored in the plasma membrane via a glycosylphosphatidylinositol (GPI) moiety. Although the lipid moiety of GPI anchors varies throughout the life cycle, most GPIs and GIPLs share a common Man 4 (AEP)GlcN-Ins-PO 4 core (6) (Fig. 1B). AEP can also substitute for phosphoethanolamine in the linkage between the GPI anchor and the polypeptide chain of mucins, although this is not obligatory. In contrast, an additional AEP substituent on the O-6 of glucosamine of T. cruzi GPI anchors appears to be universal (6) and has also been identified in glycolipids from other members of the Kinetoplastida, including the dixenic bat trypanosome Trypanosoma dionisii (9) and the monoxenic insect parasites Leptomonas samueli (10) and Herpetomonas samuelpessoai (11). However, the most abundant surface glycoconjugate of Leishmania (lipophosphoglycan, a GPI-anchored polymer of the repeating disaccharide phosphate units) does not contain AEP (12). Likewise, AEP is absent from the African trypanosome, Trypanosoma brucei (5), where ethanolamine is an integral component of the variant surface glycoprotein (13).
Apart from the demonstration that 32 P can be incorporated into AEP (5), the biosynthesis of this intermediate has not been studied. The identification of pyruvate phosphate dikinase in T. cruzi (14), which catalyzes formation of phosphoenolpyruvate from pyruvate, prompted us to search for other genes in the biosynthetic pathway to AEP. Here we report the isolation and functional expression of PEP mutase, some of its kinetic properties, and its subcellular localization in T. cruzi.

EXPERIMENTAL PROCEDURES
Organisms and Reagents-Epimastigotes of T. cruzi (Y strain and Silvio clone X10Ϫ7) were cultured at 28°C in RTH/fetal calf serum medium (RPMI, trypticase, and hemin supplemented with 5% fetal calf serum) (15). Other trypanosomatids were cultured as described (15). The edible mussel Mytilus edulis was purchased from a local fishmonger in Dundee. Routine manipulations were conducted in Escherichia coli strain JM109 and overexpression in strain BL21(DE3)pLysS (Stratagene). All chemicals were of the highest grade available from Sigma, BDH and Roche Applied Science. Phosphonopyruvate (tris-cyclohexylammonium salt) was prepared as described (16,17).
Cloning of PEP Mutase Gene from T. cruzi-A BLAST search of conserved sequence regions of PEP mutase from M. edulis and Tetrahymena pyriformis against the T. cruzi genome data base yielded a partial sequence (accession number AQ445225.2) with significant homology to other PEP mutases. This partial PEP mutase gene fragment was amplified using oligonucleotides 5Ј-GGGGTTCCGCGACACAACG-AAGCG and 3Ј-CTGCATGGCGGCAATGCAGGCCC under the following conditions: denaturation at 95°C for 10 min followed by 30 cycles of denaturation at 95°C for 1 min, annealing at 50°C for 1 min, and extension at 72°C for 1 min followed by a final extension at 72°C for 10 min. The PCR product was cloned into the Novagen vector ZERO-BLUNT TOPO. The fluorescein-labeled PEP mutase PCR fragment was used as a probe in the initial isolation and characterization of this gene. A size-selected library (2-4 kilobases) was constructed in pUC18 with genomic DNA digested with HindIII and SacI. Positive clones were identified from a colony screen using the labeled probe and found to contain the same size insert. One of these was sequenced using primers designed to the known partial PEP mutase gene sequence, and the full-length gene was obtained.
For expression studies the open reading frame of PEP mutase was amplified by PCR as described above using the primers PEPM-pET-15bF, GGCTCGAGATGCGTCACTGCTGTGGTCTG, and PEPM-pET-15bR, GGCTCGAGTTATTTCTTCGGCAAGTACATTTCC, engineered with XhoI sites in the 5Ј and 3Ј primers respectively, for cloning into the Invitrogen E. coli expression vector pET15b.
Southern Blot Analysis-T. cruzi genomic DNA (5 g) was digested with selected restriction endonucleases (BamHI, SacI, HindIII, and NdeI), separated by gel electrophoresis using a 0.8% agarose gel, and transferred to a positively charged nylon membrane (Hybond N ϩ , Amersham Biosciences). Hybridization and signal detection were performed using the Gene Images labeling and detection kit (Amersham Biosciences) following the manufacturer's recommendations.
Western Blot Analysis-Total extracts of T. cruzi and other trypanosomatids (1 ϫ 10 6 parasites per lane) and M. edulis (5 g) were fractionated by electrophoresis on 4 -12% gradient SDS-PAGE gels (18). Immunoblot analysis was performed essentially as described (19) using polyclonal antisera to PEP mutase and GAPDH at a dilution of 1:50. Blots were developed by chemiluminescence following the manufacturer's instructions (ECL kit, Amersham Biosciences).
Soluble Expression of Recombinant PEP Mutase-A 6-liter culture of BL21(DE3)pLysS/pET15b-PepM, derived from a single colony, was grown at 37°C with vigorous agitation in Terrific Broth containing 50 g ml Ϫ1 carbenicillin and 12.5 g ml Ϫ1 chloramphenicol. When the culture reached an A 600 of ϳ0.8, the culture was cooled to 25°C, agitation was reduced to 100 rpm, and the culture was induced with a final concentration of 0.5 mM isopropyl-␤-D-thiogalactopyranoside. Cultures were grown for 16 h and harvested by centrifugation. Cells were washed with 20 mM Tris, pH 8.0, and lysed in 50 ml of breaking buffer (20 mM Tris, pH 8.0, 0.5 M NaCl, 5 mM MgCl 2 , 100 g ml Ϫ1 DNase I and protease inhibitor mixture, Roche Applied Science) by flash-freezing in an ethanol/dry ice bath followed by rapid thawing and bead beating. Cell debris was separated and discarded after centrifugation at 30,000 ϫ g for 30 min at 4°C.
Purification and Properties of Recombinant PEP Mutase Protein-The supernatant containing soluble protein was diluted 2-fold with 20 mM Bis-Tris propane, 20 mM Tris, pH 7.4, 0.5 M NaCl, passed through a 0.2-m Steriflip filter, and loaded onto a nickel-chelating Sepharose high performance column (Amersham Biosciences) pre-equilibrated with the same buffer. Protein was eluted using a linear gradient from 0 to 1 M imidazole, and active fractions were pooled, dialyzed against phosphate-buffered saline (10 mM phosphate buffer, 2.7 mM KCl, 137 mM NaCl, pH 7.4), and digested with human thrombin (50 g ml Ϫ1 ) for 2 h at 25°C to remove the hexahistidine (His 6 ) tag. The sample was then dialyzed for 2 h in 20 mM Bis-Tris propane, 20 mM Tris, pH 7.4, 1 mM EDTA, 1 mM dithiothreitol, and loaded onto a 6-ml ResourceQ column (Amersham Biosciences). Protein was eluted with a linear gradient from 0 to 0.5 M NaCl, and the active fractions were pooled and dialyzed against 50 mM (K ϩ ) Hepes, pH 8.0, containing 1 mM dithiothreitol and 0.01% sodium azide. Aliquots of 50 l were dispensed, rapidly frozen, and stored at Ϫ80°C. Under these conditions the enzyme lost less than 10% of its activity over 4 months.
Native M r was determined by Superdex 200 column chromatography (Amersham Biosciences) against gel filtration standards (Bio-Rad). Molecular mass was determined by matrix-assisted laser desorption ionization time-of-flight spectroscopy in linear mode using sinapinic acid as a matrix on a Voyager-DE STR mass spectrometer (PerSeptive Biosystems). Cross-linking experiments were conducted using the BS 3 crosslinker (bis[sulfosuccinimidyl]suberate, Pierce) following the manufacturer's instructions.
Kinetic Analysis of PEP Mutase-The pH optimum of the enzyme was determined over the pH range 5-10 using a Applied Science), 88 nM T. cruzi PEP mutase, and 1 mM phosphonopyruvate. The K m for phosphonopyruvate was determined using this method by varying phosphonopyruvate concentration in the presence of 5 mM MgSO 4 or 5 mM magnesium acetate. Concentrations of phosphonopyruvate were standardized by spectrophotometric assay in the presence of excess PEP mutase. Protein concentration was determined based on the calculated extinction coefficient at 280 nm (34850 M Ϫ1 cm Ϫ1 ). This method yields a 1.5-fold higher protein concentration than that determined by the bicinchoninic acid method (Pierce). One unit of enzyme activity is defined as 1mol of NADH oxidized/min. Because Mg 2ϩ is required for the coupled assay, the requirement for divalent metal ions was determined using a direct assay after the formation of phosphoenolpyruvate. Each 1-ml assay contained 50 mM (K ϩ ) HEPES, pH 8.0, 0.5 mM dithiothreitol, 450 nM T. cruzi recombinant PEP mutase, 1 mM phosphonopyruvate, and varying amounts of MgSO 4 . The production of phosphoenolpyruvate was assayed by the increase in absorbance at 233 nm using a ⌬⑀ ϭ 1.5 mM Ϫ1 cm Ϫ1 (20). Data were fitted by nonlinear regression analysis to either the Michaelis-Menten equation (for phosphonopyruvate) or the Hill equation (for Mg 2ϩ ) using the computer program GraFit.
Sub-cellular Fractionation-The following procedures were performed at 4°C. T. cruzi epimastigotes were centrifuged and washed twice in STE buffer (0.32 M sucrose, 25 mM Tris-HCl, and 1 mM EDTA, pH 7.8). Cells were mixed with silicon carbide to form a paste and disrupted by grinding with a pestle and mortar (21). Grinding was continued until 90% of the cells were lysed as viewed by phase-contrast microscopy. The suspension was diluted 5-10-fold in STE buffer and briefly centrifuged for 3 min at 100 ϫ g. The pellet was washed once in STE buffer and centrifuged. The combined supernatants were centrifuged for 10 min at 1000 ϫ g to remove nuclei and unbroken cells. The resultant supernatant was then centrifuged at 14,500 ϫ g for 10 min. The pellet obtained contained the large granule fraction. The supernatant was spun for 1 h at 139,000 ϫ g; the resulting supernatant contained the cytosolic fraction. The pellet was dissolved in STE buffer and contained the small granule or microsomal fraction.
Production of T. cruzi PEP Mutase Antibody-Antiserum was raised in mice against recombinant T. cruzi PEP mutase (100 g). The initial injection was emulsified in complete Freund adjuvant and the second in incomplete Freund adjuvant.
Immunolocalization Studies-Mid-log phase epimastigotes were airdried onto microscope slides and fixed in 4% paraformaldehyde in PBS (0.15 M NaCl, 5 mM potassium-phosphate buffer, pH 7.4) for 10 min at room temperature followed by methanol at Ϫ20°C for 2 min. Slides were then incubated in PBS, 1% saponin, and 1 mg ml Ϫ1 heat-treated RNase for 30 min followed by blocking in 5% fetal calf serum, PBS for 5 min. The slides were then incubated in anti-T. cruzi PEP mutase diluted 1:50 in PBS for 1 h at room temperature in a dark humid chamber. After washing in PBS, slides were incubated for 1 h in fluorescein isothiocyanate-conjugated goat anti-mouse secondary antibody diluted 1:50 in PBS. Slides were washed again, incubated in 4,6-diamidino-2-phenylindole (DAPI) (1 g ml Ϫ1 ) for 2 min followed by a further wash in PBS, and mounted in Mowiol containing phenylenediamine (1 g ml Ϫ1 ).
For double-labeling experiments using the anti-GAPDH (rabbit) and anti-PEP mutase (mouse), slides were treated as above except the primary antibody was a mixture of 1:100 and 1:50 dilution, respectively, of each antibody in PBS, and the secondary was a mixture of anti-mouse fluorescein isothiocyanate and anti-rabbit TRITC.
Double-labeling experiments of monoclonal antibody to vacuolar type protein pyrophosphatase (V-H ϩ -PPiase) and anti-PEP mutase used Zenon TM One mouse IgG 1 labeling kit (Molecular Probes) to label anti-V-H ϩ -PPiase, as instructed by the manufacturers. After staining with the anti-V-H ϩ -PPiase, slides were stained with anti-PEP mutase as above.
Mitochondrial labeling of T. cruzi was conducted using MitoTracker Red 580 (Molecular Probes) as instructed by the manufacturers. Slides were double-labeled with anti-PEP mutase as described above.

Cloning of the T. cruzi PEP Mutase Gene-A TBLASTN
search of the EBI data base with the T. pyriformis protein sequence identified a T. cruzi genomic survey sequence (AQ445225.2) showing significant homology to PEP mutase. The partial sequence of the T. cruzi PEP mutase gene (PEPM) was amplified by PCR and used as a hybridization probe for a Southern blot analysis of genomic DNA digested with a range of restriction enzymes. The resulting restriction map indicated that PEPM is single-copy per haploid genome (not shown). A size-selected library was constructed with genomic DNA digested with HindIII/SacI, and a 2.5-kb fragment was cloned and sequenced. This contained a single full-length open reading frame (GenBank TM /EMBL/DDBJ accession number AJ414690) that showed significant homology to PEP mutase from other organisms (Fig. 2). The T. cruzi protein showed higher sequence identity to the eukaryotic proteins from M. edulis (65.4%) and T. pyriformis (62%) than those from prokaryotes Mesorhizobium loti (51.3%) and Streptomyces hygroscopicus (42.7%). Based on crystallographic data for the M. edulis PEP mutase (22), the active site amino acids are conserved in all species including T. cruzi PEP mutase (Trp-47, Ser-49, Asp-61, Asp-88, Asp-90, Glu-117, Lys-123, and Arg-163). The residues forming the oxyanion hole are also conserved in all the PEP mutase homologues (Gly-50 and Leu-51). Furthermore, the Mg 2ϩ binding residue is Asp-88, which is also invariant among the PEP mutase homologues.
Expression and Purification of Recombinant T. cruzi PEP Mutase-T. cruzi PEPM was sub-cloned into the expression vector pET15b containing an N-terminal His 6 tag and expressed in E. coli. The recombinant protein was purified on a nickelchelating Sepharose high performance column, digested with thrombin to remove the His 6 tag, and further purified by anionic exchange chromatography (Fig. 3A). The final yield of recombinant protein was ϳ3 mg liter Ϫ1 of culture. Matrix-assisted laser desorption ionization time-of-flight analysis of recombinant PEP mutase revealed a nominal molecular mass of 33,832 Da that correlates well with the predicted molecular mass of 33,482 Da after cleavage with thrombin. Migration on SDS-PAGE shows an M r of ϳ36,800 (Fig. 3A), similar to that reported for T. pyriformis (33,000) (4) and M. edulis (34, 000) (20). On gel filtration chromatography T. cruzi PEP mutase migrates as a single symmetrical peak corresponding to an M r of 86,000 (n ϭ 2.6, mean of 2 experiments), suggestive of either a homodimer (as reported for the T. pyriformis enzyme (23)) or a novel trimeric species. To test this possibility the native enzyme was cross-linked with BS 3 and analyzed by SDS-PAGE (Fig 3C). The resulting gel indicates that the majority of the PEP mutase is recovered as a homotetramer with only trace amounts of trimer and dimer. Identical results were obtained when PEP mutase was held constant (1.5 mg ml Ϫ1 ) and cross-linker was varied (0.25-2.0 mM) over 0.5-4 h (data not shown). Crystallographic studies on the T. cruzi enzyme also reveal a tetrameric arrangement 2 as reported for the enzyme from M. edulis (22). The reason for the anomalous behavior on gel filtration is not known.
Kinetic Characterization of Recombinant PEP Mutase Protein-The pH profile for PEP mutase enzyme activity follows a bell-shaped curve with a pH optimum of 8 and apparent pK a values of 7.05 Ϯ 0.09 and 9.12 Ϯ 0.09 (data not shown). Under optimal conditions (pH 8.0, 1 mM phosphonopyruvate, 5 mM The requirement for divalent metal ions (1 mM, all as chloride salts) was determined by measuring the thermodynamically favored direction of the reaction, i.e. formation of phosphoenolpyruvate at 233 nm. The enzyme displayed a pronounced requirement for divalent metal ions with Ͻ1% activity in the absence of Mg 2ϩ . The extent of activation was identical to that reported for the M. edulis enzyme (20), with the order being Mg 2ϩ Ͼ Co 2ϩ Ͼ Mn 2ϩ Ͼ Zn 2ϩ Ͼ Ni 2ϩ , with Ca 2ϩ showing no activation (not shown). Sulfate and other oxyanions have been reported to be inhibitory with the T. pyriformis enzyme (25). However, with the T. cruzi PEP mutase SO 4 2Ϫ is not inhibitory when compared with Cl Ϫ anion. In kinetic studies where MgSO 4 was the variable substrate, doublereciprocal plots were nonlinear (not shown). However, the data fitted well to the Hill equation yielding an S 0.5 of 0.40 Ϯ 0.04 M, n ϭ 0.46 Ϯ 0.03, k cat 19 Ϯ 0.4 s Ϫ1 , consistent with a negative co-operative effect between subunits (Fig. 4B). Co-operative behavior has not been reported with Mg 2ϩ for PEP mutase from M. edulis (K m 4 M) (20) or T. pyriformis (K m 6 M) (23).
Species Distribution and Intracellular Location-Western analysis using antibodies raised against the T. cruzi recombinant protein indicated that PEP mutase is constitutively expressed in the insect epimastigote, trypomastigote, and mammalian amastigote stages of the life cycle (data not shown). The antiserum also detects a 34-kDa band in M. edulis extracts but not in more closely related trypanosomatid species, suggesting that PEP mutase is absent in T. brucei, Crithidia fasciculata, Leishmania major, and Leishmania donovani (Fig. 5A). Two other trypanosomatids, T. dionisii and Herpetomonas muscarum, which contain AEP glycolipids, also show a band corresponding to T. cruzi PEP mutase (Fig. 5A).
Subcellular fractionation studies and immunolabeling experiments indicate that PEP mutase has a dual location in the cell. Western blots of subcellular fractions show that the enzyme is present in the cytosol and large granule fraction, which is enriched in mitochondria, glycosomes, and other large vesicular organelles (Fig. 5B). This dual location is unlikely to be a preparative artifact since antiserum to T. cruzi trypanothione synthetase (26) showed a largely cytosolic location.
Immunofluorescence staining shows a diffuse cytosolic staining together with a punctate pattern in the cells (Fig. 6). This distribution is reminiscent of glycosomal staining even though the amino acid sequence does not show a type I or type II glycosomal targeting signal (27). Staining with anti-serum to glycosomal GAPDH is also punctate, but examination of the merged image indicates that anti-PEP mutase does not co-localize exactly with the glycosome (Fig. 6A). Double-labeling with anti-PEP mutase antibody and Mitotraker indicates that PEP mutase is not located to the mitochondrion either (Fig. 6B). The presence of glycoconjugates in acidocalcisomes has been reported recently (30). However, double labeling of slides with anti-PEP mutase and anti-H ϩ -PPiase, an acidocalcisome marker, clearly do not co-localize (Fig. 6C). These results suggest that PEP mutase is not specifically localized to the glycosome, mitochondria, or acidocalcisome, but to some other unidentified organelle. DISCUSSION Although AEP was first identified in T. cruzi by 32 P-labeling studies (5), the biosynthetic pathway has not been investigated in detail. Our current study has identified a functional PEP mutase in T. cruzi, which is expressed in all stages of the life cycle. The kinetic properties closely resemble those of M. edulis and T. pyriformis, suggesting that the biosynthetic pathway is similar to other microorganisms (Fig. 1). Attempts to identify candidate genes for the subsequent step in the pathway (phosphonopyruvate decarboxylase) in the partially completed T. cruzi genome data base proved negative. Pyruvate phosphate dikinase, which could catalyze formation of PEP from pyruvate, has been localized to the glycosome (28), but PEP mutase does not perfectly co-localize with this organelle. Furthermore, pyruvate phosphate dikinase must serve additional roles in T. brucei since AEP and PEP mutase appear to be absent from this parasite (5). Additional studies are required to resolve the unusual subcellular localization of this enzyme.
Phosphonolipids constitute 23% of total phospholipids in T. pyriformis, and it has been proposed that surface phosphonolipids are important for protection against phospholipases secreted by itself or by other organisms (29). However, this is unlikely to be the case for T. cruzi since phosphonolipids constitute a minor fraction (0.34%) of total phospholipids (5). However, the occurrence of PEP mutase in trypanosomatids correlates exactly with those parasites that have been reported to possess AEP glycolipids, and thus, one function of the pathway could be to supply AEP for glycolipid synthesis. Although AEP and ethanolamine phosphate can be used interchangeably to link the GPI anchor to the polypeptide chain of mucins, the glucosamine moiety is exclusively substituted by AEP (6). Thus, the universality of AEP modification of the 6-O position of glucosamine in GIPLs and mucins suggests an essential role for this moiety in these surface glycoconjugates. Because they are thought to play a role in attachment, invasion, and intracellular survival in the parasite, the AEP moiety may be an important determinant in one or more of these functions. Likewise, GPI anchors from T. cruzi, which show pronounced proinflammatory activity, play an important role in activation of innate immunity during infection (7,8). Conceivably, the AEP substituent could protect these molecules from degradation in the gut of the insect vector or the cytoplasm of vertebrate host cells. Gene knockout studies are planned to examine these possibilities and to evaluate whether this pathway could represent a therapeutic target since it is absent in the mammalian host.