Toxoplasma gondii Purine Nucleoside Phosphorylase Biochemical Characterization, Inhibitor Profiles, and Comparison with the Plasmodium falciparum Ortholog*

Purine nucleoside phosphorylase (PNP) is an important component of the nucleotide salvage pathway in apicomplexan parasites and a potential target for drug development. The intracellular pathogen Toxoplasma gondii was therefore tested for sensitivity to immucillins, transition state analogs that exhibit high potency against PNP in the malaria parasite Plasmodium falciparum. Growth of wild-type T. gondii is unaffected by up to 10 μm immucillin-H (ImmH), but mutants lacking the (redundant) purine salvage pathway enzyme adenosine kinase are susceptible to the drug, with an IC50 of 23 nm. This effect is rescued by the reaction product hypoxanthine, but not the substrate inosine, indicating that ImmH acts via inhibition of T. gondii PNP. The primary amino acid sequence of TgPNP is >40% identical to PfPNP, and recombinant enzymes exhibit similar kinetic parameters for most substrates. Unlike the Plasmodium enzyme, however, TgPNP cannot utilize 5′-methylthio-inosine (MTI). Moreover, TgPNP is insensitive to methylthio-immucillin-H (MT-ImmH), which inhibits PfPNP with a \batchmode \documentclass[fleqn,10pt,legalpaper]{article} \usepackage{amssymb} \usepackage{amsfonts} \usepackage{amsmath} \pagestyle{empty} \begin{document} \(K_{i}^{{\ast}}\) \end{document} of 2.7 nm. MTI arises through the deamination of methylthio-adenosine, a product of the polyamine biosynthetic pathway, and its further metabolism to hypoxanthine involves PfPNP in purine recycling (in addition to salvage). Remarkably, analysis of the recently completed T. gondii genome indicates that polyamine biosynthetic machinery is completely lacking in this species, obviating the need for TgPNP to metabolize MTI. Differences in purine and polyamine metabolic pathways among members of the phylum Apicomplexa and these parasites and their human hosts are likely to influence drug target selection strategies. Targeting T. gondii PNP alone is unlikely to be efficacious for treatment of toxoplasmosis.

The phylum Apicomplexa consists of Ͼ5000 species of obligate intracellular parasites and is responsible for many important diseases in humans and other animals. Malaria (caused by Plasmodium) is a serious global problem with mortality rates in excess of 1 million a year (1). Toxoplasma gondii is a chronic infection estimated to affect ϳ30% of the world's population and poses a significant threat to immunocompromised individuals and congenitally infected children (2). The emergence of drug-resistant malaria parasites and complications associated with long-term treatment of chronic toxoplasmosis underscore the need for new chemotherapeutic agents.
Focusing on differences between host and parasite metabolism provides an attractive strategy for identifying potential drug targets. One metabolic discrepancy between apicomplexan parasites and their mammalian hosts is the lack of de novo purine biosynthesis in the former, making them completely reliant on host cells for these essential nutrients (3). Apicomplexan purine salvage pathways have been explored using a combination of biochemical, genetic, and genomic studies (4 -11), providing complete transport and metabolic maps for several species (8). Comparative analysis reveals two alternative, and functionally redundant, salvage routes for purine assimilation by Eimeria and Toxoplasma. Adenosine kinase (AK) 3 converts the nucleoside adenosine into the nucleotide AMP, and hypoxanthine-xanthine-guanine phosphoribosyl transferase (HXGPRT) converts guanylate nucleobases into nucleotides, including GMP (see "Discussion"). Cryptosporidium and Theileria rely on AK alone, while only HXGPRT is present in Plasmodium, but the ability to enzymatically interconvert AMP and IMP provides all of these parasites with a supply of all necessary purine nucleotides.
Purine nucleoside phosphorylase (PNP) converts inosine to hypoxanthine and guanosine to guanine, providing an important source of nucleobases for HXGPRT (3). PNPs have been examined in a variety of species (12) and may be grouped into two main families: trimeric forms (such as the human enzyme) are typically ϳ31 kDa and prefer 6-oxopurines (e.g. inosine, guanosine), and hexameric PNPs (such as the Escherichia coli * This work was supported in part by research and training grants from the National Institutes of Health and by United States Army Research Grant W81XWH-05-2-0025 (to K. K.). The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. enzyme) are ϳ26 kDa and active against both 6-oxopurines and 6-aminopurines (e.g. adenosine). P. falciparum PNP (Pf PNP) has been characterized in detail, and although its amino acid sequence is most similar to hexameric PNPs, its substrate is distinct from either family (10,13). In particular, Pf PNP is able to utilize 5Ј-methothioinosine (MTI), produced by the action of adenosine deaminase on methothioadenosine (MTA), a byproduct of polyamine metabolism. The involvement of Pf PNP in both purine and polyamine pathways makes this enzyme an attractive drug target, and rationally designed PNP inhibitors (immucillins) inhibit both parasite and host erythrocyte enzymes to produce purine-less death of P. falciparum parasites (11,14). This report describes the cloning, recombinant expression, and characterization of T. gondii PNP (TgPNP) and comparison with Pf PNP. Although broadly similar, TgPNP lacks activity against methylthio-purines. Surprisingly, examination of the T. gondii genome (15) indicates the absence of polyamine biosynthetic machinery in this parasite.

MATERIALS AND METHODS
Parasites, Host Cells, Chemicals, and Reagents-RH strain T. gondii tachyzoites and adenosine kinase knock-out (AK Ϫ ) mutants (7) were maintained by serial passage in primary human foreskin fibroblasts (16 Sensitivity of T. gondii to ImmH and Rescue by Purine Nucleobases-The growth of intracellular T. gondii (wild-type and AK Ϫ mutants) was measured in 24-well plates containing confluent human foreskin fibroblast cell monolayers by the incorporation of [ 3 H]uracil into acid-precipitable material. Parasites were grown for 24 h in the presence of varying concentrations of ImmH (0 -100 M), after which the cultures were subjected to a 4-h pulse of [ 3 H]uracil (5 Ci; 20 Ci mmol Ϫ1 ) and plates were processed as previously described (16). For metabolic rescue experiments, AK Ϫ parasites were inoculated into 96-well plates containing 10 M ImmH plus various concentrations of hypoxanthine or xanthine (0 -100 M) for 4 days. The disruption of host cell monolayers (an indicator of parasite viability) was measured by crystal violet staining and optical density measurements at 650 nm (16).
Genomic Analysis and Cloning of T. gondii Nucleoside Phosphorylases-T. gondii genome sequence data (10-fold coverage) are available at //ToxoDB.org (15), and TBLASTN (WU-BLAST 2.0) (17) was used to search predicted protein sequences for similarity to the PNP, uridine phosphorylase (UdP), methylthio-adenosine phosphorylase (MTAP), methylthio-adenosine nucleosidase, and polyamine biosynthetic pathway enzymes from various organisms. Based on the most significant match in the T. gondii genome obtained when PNP sequences were used as query, the following primers were constructed for 5Ј-and 3Ј-RACE (rapid amplification of cDNA ends) using the SMART TM RACE cDNA amplification kit (Clontech, Palo Alto CA): 5Ј-GCTGCCCGGGTACTTCG-ATCGCC-3Ј (sense primer for 3Ј-RACE); 5Ј-GGCACAGAC-CGAGGACACCGGAC-3Ј (antisense primer for 5Ј-RACE). T. gondii tachyzoite cDNA was synthesized from total cellular RNA prepared using the RNeasy RNA extraction kit (Qiagen, Valencia CA), and the complete TgPNP open reading frame was amplified using sense primer 5Ј-acatgcATGCAGGGCATGG-AAGTTCAGCCTC-3Ј and antisense primer 5Ј-cgggatccGTA-CTGGCGACGCAGATTC-3Ј (uppercase indicates native coding sequence; restriction sites underlined). The PCR product was gel purified (Qiagen Gel Extraction kit), digested with SphI and BglII, ligated into appropriately digested pQE-70 plasmid (Qiagen), and its sequence verified. The resulting construct (pQE-TgPNP-His 6 ) encodes TgPNP in-frame with a C-terminal His 6 tag under the control of an isopropyl-1-thio-␤-D-galactopyranoside-inducible promoter. A second putative nucleoside phosphorylase was also identified and cloned from T. gondii (see supplemental materials) and has been shown to function as a UdP (not shown).
Nucleoside phosphorylase amino acid sequences obtained from GenBank TM were aligned with T. gondii nucleoside phosphorylases (TgPNP and TgUdP) using ClustalX (18). Unambiguously aligned sequences were used to construct phylogenetic trees using the neighbor-joining method (19) and subjected to bootstrap analysis with 1000 replicates (20).
Expression and Purification of TgPNP-E. coli strain M15[pREP4] (Qiagen) was transformed with the pQE-TgPNP-His 6 plasmid and grown at 37°C in 100 g/ml ampicillin and 25 g/ml kanamycin. Expression of His-tagged protein was induced with 1 mM isopropyl-1-thio-␤-D-galactopyranoside when the culture A 600 reached 0.6, and cells were harvested 5 h later by centrifugation at 4000 ϫ g for 20 min.
Isopropyl-1-thio-␤-D-galactopyranoside-induced, TgPNPtransformed E. coli cells were resuspended and sonicated in lysis buffer (50 mM NaH 2 PO 4 , 300 mM NaCl, 10 mM imidazole, 1 mg ml Ϫ1 lysozyme) containing a protease inhibitor mixture (Sigma). The lysate was then cleared by centrifugation at 10,000 ϫ g for 20 min. 1 ml of nickel-nitrilotriacetic acid-agarose (Qiagen) was added to the cleared lysate, mixed gently for 1 h, packed into a column, and washed twice with 4 ml of wash buffer (50 mM NaH 2 PO 4 , 300 mM NaCl, 30 mM imidazole). Tagged protein was then released with elution buffer (50 mM NaH 2 PO 4 , 300 mM NaCl, 250 mM imidazole) and estimated to be Ͼ95% pure based on denaturing polyacrylamide gel electrophoresis and staining with Coomassie Blue. Protein concentration was measured using a protein assay kit (Bio-Rad, Hercules, CA).
Enzyme Assays-All PNP assays were performed with purified enzyme in 50 mM KPO 4 , pH 7.4. Phosphorylysis of inosine, 2-deoxyinosine, and 5Ј-methylthio-inosine was measured in a coupled assay with 115 milliunits ml Ϫ1 xanthine oxidase to convert hypoxanthine into uric acid. Uric acid formation was followed by spectrophotometric measurement at 293 nm (E 293 ϭ 12.9 mM Ϫ1 cm Ϫ1 ). Guanosine phosphorylysis was monitored by measuring the disappearance of guanosine at 258 nm (E 258 ϭ 5.2 mM Ϫ1 cm Ϫ1 ). Aden-osine and 5Ј-methylthio-adenosine phosphorylase activities were measured by following the disappearance of the substrate at 274 nm (E 256 ϭ 1.9 mM Ϫ1 cm Ϫ1 ). Uridine phosphorylase activity was measured by following the conversion of uridine to uracil at 272 nm (E 260 ϭ 2.9 mM Ϫ1 cm Ϫ1 ).
For inhibition assays, excess substrate (0.5 mM) was used in combination with inhibitor ranging from 0 to 10 M; the concentration of inhibitor was always at least 10-fold greater than the enzyme concentration. The rapidly reversible inhibition of PNP was analyzed by fitting to the equation where v 0 is the initial reaction rate, k cat the maximal catalytic rate, S the substrate concentration, K m the Michaelis constant, I the inhibitor concentration, and K i the dissociation constant for the enzyme-inhibitor complex. Because transition state mimics typically exhibit a slow-onset tight binding inhibition (21), reaction rates were measured continuously to monitor for a second phase with a markedly different steady state rate v s , in which case K * i (the dissociation constant for steady state following slow-onset inhibition) was assessed by fitting to v s ϭ (

RESULTS
Immucillin-H Inhibits the Growth of AK Ϫ Parasites but Not Wild-type T. gondii-Immucillins are transition state inhibitors that exhibit potent activity against mammalian and P. falciparum PNPs (14,22). ImmH kills P. falciparum in culture with an IC 50 of 35 nM, with the mode of action dependent on inhibition of both Plasmodium and erythrocyte PNP (14). In contrast, wild-type RH strain T. gondii parasites (WT) are insensitive to 100 M ImmH (Fig. 1A). Parasites lacking the purine salvage enzyme adenosine kinase (AK Ϫ ) exhibit a dose-dependent inhibition of growth, however, with an IC 50 value of 23 Ϯ 9 nM. (Inhibition of host cell PNP is also expected at this concentration, as ImmH inhibits human PNP with a K i of 72 pM.) The difference in sensitivities between wild-type and AK Ϫ parasites is consistent with previous observations on the redundancy of purine salvage pathways in T. gondii (8): this parasite can survive by salvaging either adenosine (via AK) or hypoxanthine (via HXGPRT). AK Ϫ parasites are dependent on HXGPRT, which may depend on PNP activity as a source of nucleobases.
To determine whether ImmH exerts its effect by targeting T. gondii PNP (as opposed to the host enzyme or another target), we assayed the ability of hypoxanthine and xanthine to rescue drug-treated AK Ϫ parasites, as shown in Fig. 1B. Both of these nucleobases were able to rescue AK Ϫ parasites from the inhibitory effects of 10 M ImmH. For example, 25 M hypoxanthine restored parasite viability to 75% of the levels observed without ImmH treatment, and 25 M xanthine restored parasite viability to 60% of control levels. In contrast, incubation with up to 100 M concentrations of the nucleoside inosine was unable to rescue AK Ϫ parasites from the effects of ImmH (not shown). These data strongly suggest that ImmH acts by inhibiting T. gondii PNP directly.
Identification of T. gondii PNP-Similarity searches using sequences from other species to interrogate the T. gondii genome data base identified a single significant match to the P. falciparum (10) and E. coli (23) PNP genes, but not to human PNP (24). Further analysis of protein ortholog groups (25) also identified another putative nucleoside phosphorylase in the T. gondii genome based on significant similarity to E. coli uridine phosphorylase (26) (see supplemental Fig. S1). Preliminary enzymatic characterization (not shown) indicates that this gene exhibits UdP activity.
The full-length cDNA sequence for TgPNP was obtained by 5Ј-and 3Ј-RACE and predicts a 247-amino acid protein of molecular mass 26,803 Da. Alignment of TgPNP with related enzymes shows 39% average similarity to the hexameric PNPs (Family 1, including bacterial PNPs and UdPs) versus 21% similarity to trimeric PNPs, (Family 2, including mammalian PNPs and eukaryotic MTAPs). TgPNP exhibits 41% sequence identity to Pf PNP and 27% identity to EcPNP (Fig. 2). Phylogenetic analysis also associates the apicomplexan PNPs with Family 1, as TgPNP and Pf PNP cluster more closely with UdPs than with any of the PNPs (Fig. 3). TgUdP is an outlier, not grouping strongly with other UdPs or PNPs.
Neither TgPNP nor Pf PNP (10) exhibits the complete consensus motif for either family of phosphorylases. Sequence alignment (Fig. 2) shows that eight of the sixteen residues known to be involved in substrate binding in the active site of EcPNP (23) are conserved and five more are conservatively sub- stituted. The remaining three sites are conserved within the Apicomplexa but differ from EcPNP. Asp-206 in Pf PNP and Asp-207 in TgPNP probably correspond to Asp-204 in EcPNP, which has been proposed to be the general acid/base for N7 protonation of the substrate purine ring (13).
Based on biochemical and structural studies, it has been established that PfPNP exhibits activity against MTI, a metabolite not found in humans (11,13). The crystal structure of PfPNP complexed with the transition state analog MT-ImmH reveals that the methylthio group nestles within a hydrophobic pocket formed by Val-66, Tyr-160, and Met-183 from one subunit and His-7 and Val-73 from a neighboring subunit of the PfPNP hexamer. It is interesting to note that despite the overall similarity of apicomplexan PNPs, three conservative substitutions (Ile-68, Ile-75, and Phe-162) render the methylthio binding pocket of TgPNP more similar to EcPNP, which does not possess activity against MTI. 4 Expression, Purification, and Biochemical Characterization of TgPNP-Recombinant TgPNP engineered to contain a C-terminal hexahistidine tag was overexpressed and purified from bacteria using a nickel-nitrilotriacetic acid column under native conditions (Fig. 4). On a denaturing polyacrylamide gel, the protein ran as a single band close to its predicted molecular mass of 27 kDa. The protein was stored in elution buffer (see "Materials and Methods") at 4°C with minimal loss of activity over a period of 3 months.
Among the various substrates tested against recombinant TgPNP, the highest catalytic efficiencies (k cat /K m ) were observed for phosphorylysis of inosine (1.98 ϫ 10 5 mol Ϫ1 s Ϫ1 ) and guanosine (3.83 ϫ 10 5 mol Ϫ1 s Ϫ1 ) as shown in Table 1. The kinetic parameters determined for these substrates were most similar to those observed in Pf PNP, with a turnover rate ϳ10fold lower than reported for Homo sapiens PNP and EcPNP. Deoxynucleosides, which serve as substrates for mammalian PNPs, showed very poor catalytic efficiency using either parasite PNP, and virtually no activity was detected against adenosine or uridine (substrates for EcPNP and EcUdP, respectively). Although MTI is readily transformed by Pf PNP, this purine is not an effective substrate for TgPNP, with a catalytic efficiency Ͻ0.5% than that for inosine. MTA is not a substrate for any of these enzymes. Overall, the substrate specificity of TgPNP is distinct from both mammalian and bacterial PNPs and also from the closely related P. falciparum ortholog.
Inhibition of TgPNP by Immucillins-Immucillins mimic the transition state structure for PNP and strongly inhibit both mammalian and Plasmodium PNPs (10,22). These compounds usually exhibit a two-step mechanism, with modest inhibition of the initial reaction rates followed by the slow onset of a tighter binding, pow-  , and E. coli (Ec) were aligned using ClustalX (17). Black shading indicates identity (and gray shading similarity) of two or more sequences. Amino acids known to be associated with the EcPNP active site are marked: squares indicate binding to the ribose sugar, circles bind to the phosphate group, and triangles are involved in binding to the nucleobase. Open and closed symbols represent amino acids contributed by different subunits in the PNP hexamer. Stars indicate amino acids known to be associated with a hydrophobic cavity in Pf PNP that can accommodate the 5Ј-methylthio group of MTI. erful inhibition phase (22). ImmH and ImmG (analogs of inosine and guanosine, respectively; see Fig. 5) show significant inhibition of TgPNP with equilibrium dissociation constants (K * i ) of 2.03 and 1.89 nM ( Table 2). This corresponds to K m /K * i ratios of 6450 for ImmH and 4970 for ImmG, indicating that these inhibitors bind to the enzyme considerably more tightly than do the substrates. 5Ј-deoxy-ImmH inhibits parasite PNPs somewhat less strongly with a K * i of 10 nM corresponding to the relatively high K m for deoxy-inosine (Table 1). All three of these inhibitors exhibit biphasic inhibition, with inhibition at equilibrium 68-to 195-fold stronger than the initial inhibition. All three also bind even more strongly to human PNP (for which they were designed), with K * i for ImmH and ImmG in the picomolar range. Taking advantage of the unusual substrate specificity of the Plasmodium enzyme, MT-ImmH was found to bind Pf PNP Ͼ100-fold more strongly than human PNP (13). MT-ImmH is a poor inhibitor of TgPNP, however, with K * i of 290 nM, ϳ142 times the inhibition constant for ImmH (Table 2). This observation is consistent with the observation that MTI is not a substrate for TgPNP (Table 1).
Comparative Genomics of Purine and Polyamine Metabolism in T. gondii and P. falciparum-We recently carried out a comparative analysis of purine salvage pathways in five apicomplexan species and the ciliate Tetrahymena thermophila (8). Purine auxotrophy is common in intracellular pathogens, and T. gondii possesses redundant pathways, including both AK and HXGPRT (Fig. 6). Other organisms (such as T. thermophila) are also capable of salvaging adenine via phosphoribosylation to produce AMP. P. falciparum maintains a more stripped down version of these pathways, retaining HXGPRT but not AK; in contrast, Theileria parasites have retained only AK.
Most organisms are also capable of recycling purines via the action of MTAP or methylthio-adenosine nucleosidase on MTA (a byproduct of polyamine metabolism and a dead-end metabolite) producing adenine. These enzymes are lacking in apicomplexan parasites, but the ability of P. falciparum PNP to utilize MTI as a substrate provides an alternative route (11). The T. gondii genome encodes neither MTAP nor methylthioadenosine nucleosidase, raising the question of how methythiopurines are recycled or detoxified, given that TgPNP does not act on MTI. Remarkably, in contrast to Plasmodium, which possesses a robust polyamine synthetic pathway, including S-adenosylmethionine decarboxylase, ornithine decarboxylase, and spermidine synthase activities (27), T. gondii genome lacks all of these genes. Consistent with this observation, T. gondii is insensitive to the ornithine decarboxylase inhibitor difluoromethylornithine at concentrations as high as 10 mM (see supplemental Fig. S2). The T. gondii genome also lacks arginine decarboxylase or agmatinase, which could provide an alternate route for polyamine biosynthesis (28). In sum, Toxoplasma appears to be the first eukaryote known to have completely dispensed with de novo polyamine biosynthesis, presumably in favor of salvaging these metabolites from the host cell.

DISCUSSION
Purine nucleoside phosphorylase is an important component of the purine salvage pathway in several apicomplexan para-  sites. This report describes the identification and biochemical characterization of T. gondii PNP and analysis of its contribution to the parasite's purine economy. AK and HXGPRT provide redundant routes of purine salvage in T. gondii (8). Adenosine is the major source of purine utilized by T. gondii (4) and is normally phosphorylated by AK to produce AMP. Parasites lacking AK activity are viable but completely dependent on HXGPRT (8). The flux of inosine is second only to adenosine in T. gondii (4), serving as the major source of hypoxanthine for HXGPRT (8). Parasites lacking AK activity are therefore strongly inhibited by the PNP inhibitor ImmH (Fig. 1). This is analogous to the situation in P. falciparum, which naturally lacks AK activity and is killed by similar concentrations of ImmH (14). The rescue of ImmH-treated T. gondii with hypoxanthine and xanthine (but not inosine) confirms that TgPNP is the relevant drug target.
TgPNP shows high sequence similarity to Pf PNP and Family 1 hexameric PNPs (bacterial PNPs and UdPs) but lower simi-larity to Family 2 trimeric PNPs (including mammalian enzymes) (Figs. 2 and 3). Inosine and guanosine are major substrates for TgPNP, but the protein does not possess activity against the PfPNP substrate MTI, 2Ј-deoxynucleosides (substrates for mammalian PNPs), or adenosine or uridine (substrates for bacterial PNPs and UdPs) ( Table 1). This substrate specificity is also reflected in unique aspects of TgPNP sequence. Five of the ten residues known to be important for base binding in EcPNP (22) are not conserved in TgPNP (Fig. 2), providing a possible explanation for the lack of TgPNP activity against adenosine. Furthermore, the differences observed in the residues known to be important for the formation of a hydrophobic pocket that accommodates the methylthio group in Pf PNP (13) may explain why MTI is a poor substrate for TgPNP.
Several immucillins inhibit TgPNP strongly (Table 2), exhibiting a slow onset mechanism with equilibrium inhibition constants in the nanomolar range. Measured inhibition profiles are consistent with the observed substrate specificity: the inosine and guanosine analogs ImmH and ImmG are far more effective than MT-ImmH against TgPNP.
In P. falciparum, which lacks MTAP, the polyamine biosynthesis byproduct MTA is converted to MTI by adenosine deaminase, which is further broken down to hypoxanthine and 5Ј-methylthio-ribose-1-phosphate by PNP, underlining the   importance of these enzymes in both polyamine and purine metabolism (Fig. 6). The absence of either an MTAP gene in T. gondii or the ability of TgPNP to utilize MTI poses an interesting conundrum: how is MTA recycled in this parasite? Examination of the T. gondii genome indicates that this parasite is entirely lacking in polyamine biosynthetic machinery in distinct contrast to P. falciparum, which possesses genes encoding a bifunctional ornithine decarboxylase-S-adenosylmethionine decarboxylase and spermidine synthase. Lack of ornithine decarboxylase activity in T. gondii cell extracts (29) and insensitivity of the parasite to difluoromethylornithine (supplemental data) provide further support for this observation. An alternative route for polyamine biosynthesis via ADC and agmatinase is present in plants and bacteria and has also been proposed for the apicomplexan parasite, Cryptosporidium parvum (28), but these genes are also missing from the T. gondii genome. The absence of polyamine pathway enzymes and, by extension, their MTA and MTI products obviates the need for TgPNP activity against MTI. In aggregate, the available data suggest that T. gondii is incapable of de novo polyamine biosynthesis and must depend on the transport of these crucial metabolites from the host cell and/or extracellular media, making polyamine transport a potential target for anti-parasitic chemotherapy.
It is interesting to observe that orthologous enzymes may exhibit significantly divergent biochemical characteristics even when operating in the same metabolic pathway and that these differences can affect inhibitor specificity and efficacy. Thus, although TgPNP and Pf PNP are similar enzymes performing similar functions in related parasites, Pf PNP is a more promising drug target. The striking differences in the composition of purine and polyamine metabolic pathways between P. falciparum and T. gondii probably reflect the adaptation of parasite metabolic machinery to specific environmental conditions. These studies serve to further validate comparative genomic analysis of metabolic pathways.