SGM Journals
BIOCHEMISTRY AND MOLECULAR BIOLOGY

Divergent polyamine metabolism in the Apicomplexa

Tuesday Cook, David Roos, Mary Morada, Guan Zhu, Janet S. Keithly, Jean E. Feagin, Gang Wu, Nigel Yarlett

  • 1Haskins Laboratories, Pace University, New York, NY 10038, USA
  • 2Department of Biology, University of Pennsylvania, Philadelphia, PA 19104, USA
  • 3Department of Veterinary Pathobiology, College of Veterinary Medicine and Biomedical Sciences, Texas A&M University, College Station, TX 77843, USA
  • 4Division of Infectious Diseases, David Axelrod Institute, Wadsworth Center, NYS Department of Health, Albany, NY 1220, USA
  • 5Seattle Biomedical Research Institute, 307 Westlake Ave N., Seattle, WA 9810, USA
  • 6Department of Chemistry and Physical Sciences, Pace University, New York, NY 10038, USA
  • Correspondence
    Nigel Yarlett
    nyarlett{at}pace.edu

    • Microbiology 2007; 153(4):1123–1130 · https://doi.org/10.1099/mic.0.2006/001768-0

    View at publisher PubMed

    Abstract

    The lead enzymes of polyamine biosynthesis, i.e. ornithine decarboxylase (ODC) and arginine decarboxylase (ADC), were not detected in Toxoplasma gondii [the limit of detection for ODC and ADC was 5 pmol min−1 (mg protein)−1], indicating that T. gondii lacks a forward-directed polyamine biosynthetic pathway, and is therefore a polyamine auxotroph. The biochemical results were supported by results obtained from data-mining the T. gondii genome. However, it was possible to demonstrate the presence of a highly active backconversion pathway that formed spermidine from spermine, and putrescine from spermidine, via the combined action of spermidine/spermine N1-acetyltransferase (SSAT) or spermidine N1-acetyltransferase (SAT) and polyamine oxidase (PAO). With spermine as the substrate, T. gondii SSAT had a specific activity of 1.84 nmol min−1 (mg protein)−1, and an apparent Km for spermine of 180 mM; with spermidine as the substrate, the SAT had a specific activity of 3.95 nmol min−1 (mg protein)−1, and a Km for spermidine of 240 mM. T. gondii PAO had a specific activity of 10.6 nmol min−1 (mg protein)−1, and a Km for acetylspermine of 36 mM. Furthermore, the results demonstrated that T. gondii SSAT was 50 % inhibited by 30 mM di(ethyl)norspermine. The parasite actively transported arginine and ornithine, which were converted via the arginine dihydrolase pathway to citrulline and carbamoyl phosphate, resulting in the formation of ATP via carbamate kinase. The lack of polyamine biosynthesis by T. gondii is contrasted with polyamine metabolism by other apicomplexans.

    Edited by: J. Tachezy

    INTRODUCTION

    Polyamines are essential for the growth and function of all eukaryotic cells. To date, only two orders of the Archaea, the Methanobacteriales and the Halobacteriales (Hamana & Matsuzaki, 1992), have been found that lack these cationic molecules. The apparent universal distribution of polyamines, and the complexity of compensatory mechanisms that are invoked to maintain polyamine homeostasis, suggest that these molecules are critical to cell survival (Wallace et al., 2003). The biosynthesis of polyamines typically starts with the conversion of arginine to putrescine by one of two distinct pathways: arginase coupled with ornithine decarboxylase (ODC), or arginine decarboxylase (ADC) coupled with agmatine iminohydrolase. Putrescine can be further converted into spermidine and spermine by spermidine synthase and spermine synthase, respectively. When polyamines are available, spermine or spermidine can be retro-converted into spermidine and putrescine by the consecutive action of spermine/spermidine N1-acetyltransferase (SSAT) and polyamine oxidase (PAO).

    The phylum Apicomplexa comprises a large group of intracellular parasitic protists, including Toxoplasma gondii, Plasmodium spp., Babesia spp., Cryptosporidium spp. and Eimeria spp., all of which have a multistage life cycle that includes merozoites, meronts and gamonts, which are formed within the host cell membrane-bound parasitophorous vacuole. The merozoites and sporozoites are found outside the host cell, and they invade new host cells (Entzeroth et al., 1998; Coombs et al., 1977). All apicomplexans examined to date, except for members of the Cryptosporidium genus, contain a non-photosynthetic plastid, termed the apicoplast, which was acquired by an ancestral apicomplexan from a member of the red or green algae, via a secondary endosymbiotic event (Fichera & Roos, 1997; Kohler et al., 1997; Cai et al., 2003). However, Cryptosporidium parvum possesses an unusual mitochondrial organelle (Keithly et al., 1997; Riordan et al., 2003; Slapeta & Keithly, 2004), as well as a number of unique molecular and biochemical features that differ from those found in other apicomplexans (Abrahamsen et al., 2004; Thompson et al., 2005).

    Despite the importance of polyamines to cell growth and multiplication, polyamine metabolism has not been thoroughly characterized in the Apicomplexa. Published data have shown that Plasmodium falciparum and Eimeria tenella possess ODC, whereas C. parvum has a plant-like pathway that utilizes ADC (Keithly et al., 1997). However, there have been no reports on polyamine metabolism in T. gondii. In this study, we investigated polyamine metabolism by T. gondii. Our results indicate that T. gondii lacks a forward-directed polyamine biosynthestic pathway. However, we were able to demonstrate a highly active polyamine retro-conversion pathway responsible for the conversion of spermine to spermidine and putrescine via SSAT and PAO. T. gondii SSAT was inhibited by di(ethyl)norspermine (DENSpm), which is a specific inhibitor of SSAT in other cells (Yarlett et al., 2000). These observations are also supported by the bioinformatic analysis of apicomplexan genomic data that indicates that polyamine metabolism is highly divergent among apicomplexans.

    METHODS

    Organisms.

    T. gondii tachyzoites (strain RH) were cultivated in human foreskin fibroblast cells, as described (Furtado et al., 1992). Parasites were harvested soon after emerging from host cells, passed through a 3 μm filter (Whatman no. 110612), centrifuged at 1000 g for 12 min, washed once with PBS, and pelleted with desktop centrifugation for 3 min. Oocysts of C. parvum (Iowa strain) were obtained from Bunch Grass Farms (Troy, Idaho, USA), and purified through discontinuous CsCl gradients at 1.4, 1.1 and 1.05 g ml−1 by centrifugation for 1 h at 16 000 g at 4 °C. They were then rinsed three times in distilled water, resuspended in 2.5 % (w/v) aqueous potassium dichromate, and stored at 4 °C for 2–10 days. On the day of use, 109 oocysts were washed in distilled water, surface sterilized on ice for 10 min with 10 % sodium hypochlorite, washed five times with distilled water, and resuspended in Hanks' balanced salt solution (HBSS). Excystation was performed by mixing equal volumes of excystation buffer [0.5 % (w/v) trypsin and 1.5 % (w/v) taurodeoxycholate in HBSS] and oocysts, and incubating for 1 h at 37 °C. To prepare free sporozoites from the chicken coccidium E. tenella (WIS), oocysts in PBS were broken by strong vortex with 1 mm glass beads in a 15 ml tube to release sporocysts. Excystation was then performed in a similar way to that for C. parvum, except that the incubation was conducted at 41 °C for 1.5 h. Mixed erythrocytic stage P. falciparum (strain C10; Hempelmann et al., 1981) was supplied by Dr Jean Feagin.

    Enzyme assays.

    Parasites were resuspended in 0.1 M KH2PO4/K2HPO4 buffer, pH 7.4, and extracts were prepared by 30 strokes in a Potter-Elvehjem at 4 °C. ADC was assayed in incubations containing 7 μCi (259 kBq) [1-14C]arginine (327 mCi mmol−1; 12.10 GBq mmol−1), 10 mM Tris/HCl, pH 6.0–8.0, 60 μM pyridoxal phosphate, 1 mM 2-mercaptoethanol, and varying amounts of arginine (0.06–2.0 mM), for 30 min at 37 °C. The 14CO2 released in 30 min was trapped on filter paper soaked with 1 M benzethonium hydroxide, and measured by scintillation (Smith, 1983). ODC was assayed as described for ADC, except that 1 μCi (37 kBq) [1-14C]ornithine (51.3 mCi mol−1; 1.90 GBq mmol−1) was used instead of [1-14C]arginine. SSAT and spermidine N1-acetyltransferase (SAT) were assayed in incubations containing 0.5 μCi (18.5 kBq) [1-14C]acetyl coenzyme A (60 mCi mmol−1; 2.22 GBq mmol−1), and supplemented with 60 μM acetyl coenzyme A, 0.1 mM Bicine, pH 7.0, and varying amounts of spermine (0.02–0.60 mM) or spermidine (0.02–0.50 mM), for 20 min. The reaction was stopped by the addition of 0.2 M hydroxylamine, and placing on ice. Labelled samples were placed in a boiling water bath for 3 min, and they were measured as for ODC and ADC (Keithly et al., 1997). PAO activity was determined by measuring the substrate-dependent formation of hydrogen peroxide in 10 mM glycine (pH 8.0) containing 10 mM 2,2′-azino-bis(3-ethylbenzothiazoline-6-sulfonic acid) diammonium salt (ABTS) (BioChemika), 1 mM N1-acetylspermine, 5 units horseradish peroxidase, and 50–100 μg protein. The peroxide released was detected by measuring the change in absorption of ABTS at 420 nm. Arginine deiminase was determined by measuring the colorimetric formation of citrulline at 37 °C. The assay contained 1 mM l-arginine, 40 mM MES, pH 6.0, and 0.07 mg protein, in a final volume of 1.0 ml. After 1 h, the reaction was stopped by the addition of 0.075 ml 100 % (w/v) TCA, and the amount of citrulline formed was determined using diacetyl monoxime, as described by Boyde & Rahmatullah (1980). Catabolic ornithine carbamoyl transferase was determined by measuring 14CO2 release from l-[14C-carbamoyl]citrulline. The reaction mixture contained 40 mM Tricine, pH 8.0, 0.1 mM l-citrulline, 17.2 mM l-[14C-carbamoyl]citrulline (57.7 mCi mmol−1; 2.13 GBq mmol−1) (DuPont NEN Life Science), and 0.07 mg protein, in a final volume of 1 ml. After incubation at 37 °C for 1 h, the reaction was stopped with 1 ml 40 % TCA, and incubated for a further 30 min. CO2 was trapped using filter paper soaked in benzethonium hydroxide. Carbamate kinase was determined in incubations containing 1 mM ADP, 20 mM MgSO4, 0.15 mM luciferin, 1 mg firefly lantern extract, and 1 mM carbamoyl phosphate, in 50 mM potassium phosphate buffer, pH 7.6. ATP formation was determined by monitoring the luminescence using a photomultiplier tube. Proteins were determined using the method of Lowry et al. (1951).

    Enzyme inhibition studies.

    Extracts were incubated with 30 μM α-difluoromethylornithine (DFMO) or α-difluoromethylarginine (DFMA) for 30 min prior to determination of enzyme activity by addition of 1 μCi (37 kBq) [1-14C]ornithine plus 2 mM ornithine, or 7 μCi (259 kBq) [1-14C]arginine plus 2 mM arginine, as described for ODC and ADC activity (Yarlett et al., 1992). The inhibitory effect of DENSpm on SSAT was evaluated by measuring the enzyme activity in incubations containing 0.1 mM bicine (pH 7.0), 60 μM acetyl-CoA, 25 μM spermine and varying concentrations of DENSpm (10–100 μM). The assay was repeated with 75, 150 and 300 μM spermine and the amount of N1-acetylspermine produced after 30 min was determined as described previously (Yarlett et al., 2000).

    Uptake and interconversion of [14C]spermine, [14C]arginine and [14C]ornithine.

    T. gondii tachyzoites (104) were incubated in HBSS, pH 7.4, containing 0.09 μM (0.5 μCi; 18.5 kBq) [5,8-14C]spermine mixed with 2 mM spermine, 6 μM (2 μCi; 74 kBq) l-[U-14C]arginine mixed with 2 mM arginine, or 0.64 μM (20 μCi; 740 kBq) l-[2,3-3H]ornithine mixed with 2 mM ornithine, for 15 min at 37 °C. In some experiments, the incubation medium also contained 5 mM DFMO or 5 mM DFMA. Cells were sedimented by centrifugation at 14 000 g for 1 min, and the supernatant was removed. The cell pellet was resuspended in ice-cold 4 % TCA and vortexed, and the precipitated protein was removed by centrifugation at 14 000 g for 5 min. Amino acids and polyamines in the protein-free supernatant were separated by HPLC, and detected by dual analysis using radiometric and fluorescence detection, as described below.

    HPLC.

    Polyamines were separated by reverse-phase HPLC, using a series LC 410 pump (Perkin-Elmer) coupled to a C-18 10 μm column (4.5×250 mm), at a flow rate of 1 ml min−1. The method employed a 70 min discontinuous gradient starting with 85 % (v/v) buffer A: 2.5 g lithium citrate l−1, pH 2.65, containing 0.22 g octane sulfonic acid l−1 and 15 % (v/v) acetonitrile. Separation of polyamines used a gradient change in the buffer, as described (Yarlett & Bacchi, 1988). Standards and samples were derivatized prior to injection by mixing one part standard or sample with two parts 0.8 g o-pthalaldehyde l−1 (dissolved in 3 ml methanol and 30.9 g boric acid l−1 containing 24 g KOH l−1 and 1 ml 2-mercaptoethanol, pH 10.4). The derivatized compounds were subjected to dual analysis using a fluorescence monitor (λexcitation 320 nm, λemission 455 nm) coupled to a flow-through model 1B Radiometric detector (IN/US Systems) that mixed three parts scintillant (INFLOW ES) to one part sample. Areas under the peaks were determined using β-RAM computer software (IN/US Systems), version 1.62.

    Bioinformatic analysis.

    To validate our biochemical observations, we also data-mined the complete or nearly complete genome-sequencing data for T. gondii (, release 3.0; and ), P. falciparum (, release 4.4), C. parvum (, release 3.2), and E. tenella (). Because the divergence in apicomplexan polyamine metabolism is mainly found in the forward direction, we focused on the homology search of ODC and ADC genes that represent two distinct pathways to convert arginine to putrescine. The annotated ODC and ADC protein sequences from all major taxonomic groups were used as queries to perform a blast search of DNA and protein databases (where available). SSAT and PAO were searched using conserved sequences for these proteins in the lower eukaryotes. Hits from the apicomplexan databases were retrieved, and used as queries to search homologues in all non-redundant protein databases at the National Center for Biotechnology Information (NCBI) () to verify their true identities.

    RESULTS

    Enzyme analysis

    T. gondii was found to be incapable of converting arginine to putrescine, whereas P. falciparum, E. tenella and C. parvum use different pathways for this process. Extracts of T. gondii had no detectable ODC and ADC activity (Table 1⇓). The addition of pyridoxal phosphate (0.06 mM) or Mg2+ (0.1 mM) at concentrations shown to be necessary for these enzymes in other cells did not result in detectable enzyme activity when assayed over a pH range of 4.5–8.5. Positive controls included lysates of Trichomonas vaginalis and Saccharomyces cerevisiae expressing an Escherichia coli ADC. These were used at 1/10 of the protein content for T. gondii (not shown). The activity of these enzymes was examined in extracts from other apicomplexans. In agreement with previous observations, ODC was detected in extracts of P. falciparum and E. tenella, while ADC activity was detected in extracts of C. parvum (Table 1⇓). The ADC and ODC detected in the respective apicomplexans were specifically inhibited by fluorinated analogues of the parent amino acid. The ADC activity in C. parvum was 50 % inhibited by 30 μM DFMA, but was unaffected by the same concentration of DFMO (Table 1⇓). In contrast, the ODC activity observed in P. falciparum and E. tenella was 84 and 56 % inhibited by 30 μM DFMO, respectively, and 58 and 12 % inhibited by 30 μM DFMA, respectively (Table 1⇓).

    Table 1.

    Effect of DFMO and DFMA on ADC and ODC activity in the apicomplexa

    ODC and ADC activity was determined as described in Methods. Parasite extracts were incubated for 30 min with 30 μM DFMA or DFMO prior to the addition of l-[1-14c]ornithine or l-[1-14C]arginine, as described in Methods. One unit of enzyme activity is defined as 1 μmol product formed min−1 (mg protein)−1. nd, Not detected; limit of detection, 5 pmol min−1 (mg protein)−1.

    Genomic data support the notion that the forward-converting pathway is divergent among apicomplexans

    Our biochemical observations agree with the genomic data. By data-mining various apicomplexan genome sequences, we failed to detect any ODC and ADC homologues encoded by the T. gondii genome (ToxoDB, release 3.0, with 10× coverage), but we identified ODC genes from P. falciparum which has a well characterized bifunctional ODC/S-adenosyl-l-methionine decarboxylase (Wrenger et al., 2001) and E. tenella which has ODC activity (Keithly et al., 1997; The Sanger Institute, with >8× coverage). No ADC homologues were detected from the T. gondii, P. falciparum and E. tenella genomes, except for the lysine decarboxylase (LDC) homologues that typically display significant sequence similarity with ADC and ODC. No ODC homologue was present in the C. parvum genome, which agrees with the absence of ODC activity in this parasite. Despite the consistent enzymic detection of ADC in C. parvum (Keithly et al., 1997; this study), we failed to find any ADC or even LDC homologues in the C. parvum genome. It is possible that the ADC gene in C. parvum is highly divergent from known ADC genes that have so far been found in bacteria and plants only, or that this parasite may possess another uncharacterized pathway that displays activity similar to ADC. The T. gondii SSAT gene identified with sequence 12 290 from the T. gondii genome database. blast search of this sequence indicated the highest homology with the SSAT gene from C. parvum (Yarlett et al., 2007). A putative PAO gene was also identified in the T. gondii genome database, and it had the contiguous sequence chr 0-995297-63. The T. gondii PAO gene is highly divergent, but it encodes a protein of similar molecular size (508 aa residues) to that observed for PAO from other organisms.

    T. gondii possesses highly active back-converting activity to synthesize putrescine

    Although neither ODC nor ADC activity was detected in T. gondii, SSAT and SAT activities were clearly detected in extracts of T. gondii by measuring the acetylation of exogenously supplied spermine and spermidine, respectively (Table 2⇓). The enzymes had a pH optimum of 8.0 for both substrates, and exhibited Michaelis–Menten kinetics. Hanes–Woolf analysis of the SSAT activity for spermine resulted in a linear plot with maximal activity at a saturating spermine concentration of 1.84 μM min−1 (mg protein)−1, and an apparent KM for spermine of 180 μM (Table 2⇓). The maximal activity with spermidine was 3.95 μM min−1 (mg protein)−1, and an apparent KM for spermidine of 240 μM (Table 2⇓). T. gondii SSAT was competitively inhibited by DENSpm, with a calculated Ki of 30 μM. These results indicate that T. gondii may rely on polyamine uptake and retro-conversion to satisfy its polyamine requirements. PAO, the enzyme transforming N1-acetylspermine to spermidine, and N1-acetylspermidine to putrescine, was detected in extracts of T. gondii tachyzoites. As described for other cells, the SSAT is the rate-limiting step in the reaction.

    Table 2.

    Activities of polyamine metabolizing enzymes in representative members of the Apicomplexa

    Enzyme analysis was performed as described in Methods. One unit of enzyme activity is defined as the amount of enzyme required to generate 1 μmol product min−1 (mg protein)−1; Km values are expressed in μM.

    Spermidine is the major product of spermine by T. gondii

    Incubations of T. gondii tachyzoites (104) with a mixture of 2.02 mM [5,8-14C]spermine (0.25 μCi mmol−1; 9 kBq mmol−1) for 30 min, and HPLC analysis of the products, revealed that spermine was readily transported by the parasite (retention time, 28 min), and that spermidine was the major product (retention time, 32 min) of spermine metabolism by T. gondii tachyzoites (Fig. 1⇓). A minor signal corresponding to N1-acetylspermine was also observed at a retention time of 5 min (Fig. 1⇓).

    Figure image not available in archive
    Fig. 1.

    HPLC analysis of T. gondii tachyzoites after incubation with [14C]spermine. (a) T. gondii tachyzoites incubated with 2 mM (1 μCi μmol−1; 37 kBq μmol−1) [5,8-14C]spermine for 30 min, as described in Methods, took up 1.77 μmol spermine per 104 tachyzoites (28 min), and converted it to 0.29 μmol N1-acetylspermine per 104 tachyzoites (5 min) and 1.33 μmol spermidine per 104 tachyzoites (32 min), confirming the enzymic analysis with T. gondii tachyzoite extracts, and thedata-mining results. (b) Standards: 2 mM (2 μCi μmol−1; 74 kBq μmol−1) [1,4-14C]putrescine (39 min), 2 mM (0.30 Ci μmol−1; 11.1 GBq μmol−1) [1,7-3H]spermidine (32 min), 2 mM (1 μCi μmol−1; 37 kBq μmol−1) [5,8-14C]spermine (28 min).

    Citrulline, ornithine and carbamoyl phosphate are the major products in arginine metabolism in T. gondii

    The enzymes of the arginine dihydrolase pathway were detected in extracts of T. gondii. In common with this pathway in other cells, the activities of arginine deiminase and ornithine carbamoyl transferase were approximately equal, whereas that for carbamate kinase was significantly higher: 3.0±0.16, 2.1±0.09 and 980±22.0 μmol product min−1 (mg protein)−1 (mean±sd), respectively.

    Polyamine content of T. gondii

    The polyamine content of T. gondii tachyzoites whole-cell homogenates was determined (Table 3⇓). The concentrations of the physiological polyamines putrescine, spermidine and spermine were similar to those found in other protists that have been examined (Bacchi & Yarlett, 1995). In agreement with the enzymic analysis, determination of polyamine levels in cells incubated with l-[2,3-3H]ornithine or l-[U-14C]arginine did not produce detectable quantities of 3H-labelled or 14C-labelled polyamines, respectively. Additionally incubation with these polyamine precursors did not significantly alter basal polyamine levels in these cell extracts (Table 3⇓), and total polyamine pools remained constant for all treatments: control, 369–799 nmol (mg protein)−1; ornithine, 324–772 nmol (mg protein)−1; arginine, 340–718 nmol (mg protein)−1. These results corroborate the enzyme analysis indicating that T. gondii is auxotrophic for polyamines.

    Table 3.

    Polyamine content of T. gondii

    Values [nmol polyamine (mg protein)−1] were determined using whole-cell homogenates. Analysis was performed using control homogenates with no addition, homogenates incubated with 10 mM ornithine, and homogenates incubated with 10 mM arginine. Values are means±sd; the number of experiments is in parentheses.

    Analysis of products from incubations containing l-[U-14C]arginine revealed that arginine is rapidly transported, and converted into citrulline, ornithine and carbamoyl phosphate (Table 4⇓). These results demonstrate that the major products of arginine metabolism are citrulline, and equimolar amounts of ornithine and carbamoyl phosphate, consistent with the presence of an arginine dihydrolase pathway in this parasite. The addition of 5 mM DFMO to the incubation medium caused a 25 % increase in arginine uptake, with a concomitant 45 % increase in the amount of citrulline formed; however, 85 and 77 % reductions in ornithine and carbamoyl phosphate, respectively, were observed, indicating that DFMO interferes with ornithine carbamoyl transferase activity (Table 4⇓). The addition of 5 mM DFMA to the incubation medium caused a 45 % reduction in arginine, indicating that DFMA effectively competes with the transport of arginine into the parasite. The reduced arginine resulted in an equimolar (43 %) reduction in the amount of citrulline formed; additionally, 68 and 40 % reductions in ornithine and carbamoyl phosphate, respectively, were observed (Table 4⇓). Cells incubated with l-[2,3-3H]ornithine demonstrated that ornithine was also rapidly transported into the parasite, and that an anabolic function for ornithine carbamoyl transferase was present that converts ornithine back to citrulline (Table 4⇓). The addition of 5 mM DFMO to the incubation mix resulted in a 90 % reduction in ornithine, indicating that DFMO effectively competes with the transport of ornithine into the parasite, resulting in a 70 % reduction in citrulline (Table 4⇓). The addition of 5 mM DFMA to the incubation medium did not have a significant effect upon the intracellular citrulline and ornithine concentrations (Table 4⇓).

    Table 4.

    Intermediates of the arginine dihydrolase pathway in T. gondii

    Values (nmol ml−1) were determined in whole-cell homogenates incubated with [U-14C]arginine or [U-14C]ornithine, in the presence and absence of 5 mM DFMO or 5 mM DFMA, as described in Methods. Values are means±sd for triplicate samples.

    DISCUSSION

    The polyamines spermidine and spermine are essential to the growth and development of all known cells. These molecules play a critical role in cell proliferation and differentiation, and biosynthesis of macromolecules. Because of the significance of these molecules to cell survival, they are obvious targets for the development of chemotherapeutic agents designed to block cell growth. The lead and rate-controlling enzyme of polyamine biosynthesis in the majority of eukaryotic cells is either ODC or ADC, producing putrescine, to which an aminopropyl group is sequentially added forming first spermidine, and then spermine (Cohen, 1998; Bacchi & Yarlett, 2002). T. gondii is a member of the apicomplexa, a diverse group of intracellular parasites that have in common a multistage life cycle (Bush et al., 2001). In this study, we demonstrate that T. gondii lacks ODC and ADC activity, and therefore is auxotrophic for polyamines. These results agree with the observed growth requirement of T. gondii for host polyamines reported by others (Moraes et al., 2004; Seabra et al., 2004), but distinguish T. gondii from other members of the apicomplexa that have the ability to synthesize polyamines (Keithly et al., 1997; Wrenger et al., 2001). T. gondii has previously been shown to have a high-affinity putrescine transporter, demonstrating the ability to scavenge host-derived putrescine (Seabra et al., 2004). We show that the parasite also has the ability to retro-convert spermine to spermidine, and spermidine to putrescine, via an active SSAT. This ability would not only provide the parasite with a mechanism for production of spermidine from spermine, but also provide a mechanism for control of intracellular polyamines by acetylation, and subsequent removal from the cell (Wallace et al., 2003; Fig. 2⇓). The various members of the apicomplexa have different methods for polyamine biosynthesis: E. tenella and P. falciparum have an active ODC (Keithly et al., 1997; Wrenger et al., 2001), whereas C. parvum has an ADC (Keithly et al., 1997); these differences may be the result of different intracellular localizations of these parasites. Most information is available for the malaria parasite P. falciparum, which has been shown to possess a bifunctional ODC-adenosylmethionine decarboxylase that is responsible for polyamine biosynthesis (Wrenger et al., 2001). The activity of the P. falciparum enzyme has been shown to be significantly blocked by putrescine analogues that exhibited 1300-fold higher activity toward the parasite compared with DFMO (Das Gupta et al., 2005). Depletion of putrescine by these analogues results in an increase in spermine, and this is concluded to be due to the activity of a non-specific spermidine synthase (Das Gupta et al., 2005; Haider et al., 2005).

    Figure image not available in archive
    Fig. 2.

    Polyamine and arginine interconversion by T. gondii tachyzoites. Arginine, ornithine and putrescine transporters; Based upon the results shown in Table 4, DFMA is taken up via the arginine transporter, and DFMO is taken up via the ornithine transporter. These fluorinated amino acid analogues act as non-functional substrates for enzymes of the arginine dihydrolase pathway, resulting in the metabolic changes noted in the text. In the absence of these analogues, arginine is metabolized to ornithine and carbamoyl phosphate via the arginine dihydrolase pathway. 1, Arginine deiminase; 2, ornithine carbamoyl transferase (catabolic); 3, ornithine carbamoyl transferase (anabolic); 4, carbamate kinase. Spermine is taken up from the host cell, and metabolized to spermidine by the combined action of SSAT (5) and PAO, and to putrescine by the combined action of SAT (6) and PAO.

    It has been reported that T. gondii has a higher infection rate in younger embryos, which exhibit a greater ODC activity, and have a higher intracellular putrescine content, indicating the important contribution of host cell polyamines to parasite growth and development (Moraes et al., 2004). Previous studies have indicated that the specific irreversible inhibitor DFMO is only effective at inhibiting growth of T. gondii when used at concentrations that are toxic to the host-cell monolayers (Hofflin et al., 1985; Derouin & Chastang, 1988; Moraes et al., 2004). These results contrast with the intestinal parasite Eimeria, which has an active ODC. Growth of Eimeria is blocked by DFMO administration at concentrations that are non-toxic to the host (Hanson et al., 1982; San Martin-Nuñez et al., 1988).

    It is significant that the total polyamine content of T. gondii remains constant after incubation with radiolabelled arginine or ornithine, and that no label is incorporated into parasite intracellular polyamines. Radiolabelled arginine was, however, converted via the arginine dihydrolase pathway into citrulline, and equimolar amounts of ornithine and carbamoyl phosphate (Fig. 2⇑). Interestingly, the specific irreversible inhibitor of arginine decarboxylase, DFMA, significantly reduced the uptake of arginine, resulting in a proportional reduction of the products of the arginine dihydrolase pathway. DFMO did not interfere with the uptake of arginine, but significantly reduced the synthesis of radiolabelled ornithine from arginine, indicating that DFMO exhibits an inhibitory effect towards ornithine carbamoyl transferase. DFMO did significantly compete with the uptake of ornithine, and caused a concomitant reduction in citrulline formation. The transport and metabolism of host-derived arginine via the ADH pathway would have an important role in parasite ATP formation (Fig. 2⇑). It is also proposed that this mechanism would divert host arginine away from NO synthesis, and hence be responsible for protection from this host-derived cytotoxic defence mechanism (Seabra et al., 2004). It is likely that the parasite ADH pathway is a rational target for the development of chemotherapeutic agents to treat disease caused by T. gondii.

    Acknowledgments

    This research was supported by grant AI 49785 (N. Y.). The following apicomplexan genome databases were searched for polyamine metabolic genes: T. gondii at , which contains genomic data provided by the Institute for Genomic Research [supported by the National Institutes of Health (NIH) grant AI05093], the Sanger Institute (Wellcome Trust), and Washington Univeristy (NIH grant 1R01AI045806-01A1); P. falciparum at , funded by NIH grant 1R01AI058515-01; C. parvum at , funded by NIH contract N01-AI-40037 (HHSN266200400037C); and E. tenella at , funded by the Biotechnology and Biological Sciences Research Council (UK).

    References

    1. Abrahamsen, M. S., Templeton, T. J., Enomoto, S., Abrahante, J. E., Zhu, G., Lancto, C. A., Deng, M., Liu, C., Widmer, G. & other authors (2004). Complete genome sequence of the apicomplexan, Cryptosporidium parvum. Science 304, 441–445.
    2. Bacchi, C. J. & Yarlett, N. (1995). Polyamine metabolism. In Biochemistry and Molecular Biology of Parasites, pp. 119–131. Edited by J. J. Marr & M. Muller. New York: Academic Press.
    3. Bacchi, C. J. & Yarlett, N. (2002). Polyamine metabolism as a chemotherapeutic target in protozoan parasites. Mini Rev Med Chem 2, 553–563.
    4. Boyde, T. R. & Rahmatullah, M. (1980). Optimization of conditions for the colorimetric detection of citrulline using diacetyl monoxime. Anal Biochem 107, 424–431.
    5. Bush, A. O., Fernandez, J. C., Esch, G. W. & Seed, J. R. (2001). Parasitism: the Diversity and Ecology of Animal Parasites, pp. 66–94. Cambridge: Cambridge University Press.
    6. Cai, X., Fuller, A. L., McDougald, L. R. & Zhu, G. (2003). Apicoplast genome of the coccidian Eimeria tenella. Gene 321, 39–46.
    7. Cohen, S. S. (1998). A Guide to the Polyamines, pp. 122–230. New York: Oxford University Press.
    8. Coombs, G. H., Denton, H., Brown, S. M. A. & Thong, K.-W. (1977). Biochemistry of the coccidia. Adv Parasitol 39, 141–226.
    9. Das Gupta, R., Krause-Ihle, T., Bergmann, B., Müller, I. B., Khomutov, A. R., Müller, S., Walter, R. D. & Lüersen, K. (2005). 3-Aminooxy-1-aminopropane and derivatives have an antiproliferative effect on cultured Plasmodium falciparum by decreasing intracellular polyamine concentrations. Antimicrob Agents Chemother 49, 2857–2864.
    10. Derouin, F. & Chastang, C. (1988). Enzyme immunoassay to assess effect of antimicrobial agents on Toxoplasma gondii in tissue culture. Antimicrob Agents Chemother 32, 303–307.
    11. Entzeroth, R., Mattig, F. R. & Werner-Meier, R. (1998). Structure and function of the parasitophorous vacuole in Eimeria species. Int J Parasitol 28, 1015–1018.
    12. Fichera, M. & Roos, D. S. (1997). A plastid organelle as a drug target in apicomplexan parasites. Nature 390, 407–409.
    13. Furtado, G. C., Cao, Y. & Joiner, K. A. (1992). Lamnin on Toxoplasma gondii mediates parasite binding to the β-1 integrin receptor α-6 β-1 on human foreskin fibroblasts of Chinese hamster ovary cells. Infect Immunol 60, 4925–4931.
    14. Haider, N., Eschbach, M.-L., de Souza Dias, S., Gilberger, T.-W., Walter, R. D. & Lüersen, K. (2005). The spermidine synthase of the malaria parasite Plasmodium falciparum: molecular and biochemical characterization of the polyamine synthesis enzyme. Mol Biochem Parasitol 142, 224–236.
    15. Hamana, K. & Matsuzaki, S. (1992). Polyamines as a chemotaxonomic marker in bacterial systematics. Crit Rev Microbiol 18, 261–283.
    16. Hanson, W. L., Bradford, M. M., Chapman, W. L., Waits, V. B., McCann, P. P. & Sjoerdsma, A. (1982). α-Difluoromethylornithine: a promising lead for preventative chemotherapy for coccidiosis. Am J Vet Res 43, 1651–1653.
    17. Hempelmann, E., Ling, I. & Wilson, R. J. (1981). S-antigens and isozymes in strains of Plasmodium falciparum. Trans R Soc Trop Med Hyg 75, 855–858.
    18. Hofflin, J. M., Guptill, D. R., Araujo, F. G. & Remmington, J. S. (1985). Difluoromethylornithine and formycin B in toxoplasmosis. J Infect Dis 152, 1101.
    19. Keithly, J. S., Zhu, G., Upton, S. J., Woods, K. M., Martinez, M. P. & Yarlett, N. (1997). Polyamine biosynthesis in Cryptosporidium parvum and its implications for chemotherapy. Mol Biochem Parasitol 88, 35–42.
    20. Kohler, S., Delwiche, C. F., Denny, P. W., Tilney, P., Webster, P., Wilson, R. J. M., Palmer, J. D. & Roos, D. S. (1997). A plastid of probable green algal origin in apicomplexan parasites. Science 275, 1485–1489.
    21. Lowry, O. H., Rosebrough, N. J., Farr, A. L. & Randall, R. J. (1951). Protein measurement with the Folin phenol reagent. J Biol Chem 193, 265–275.
    22. Moraes, A. M. M., Pessôa, C. N., Vommaro, R. C., De Souza, W., de Mello, F. G. & Hokoç, J. N. (2004). Cultured embryonic retina systems as a model for the study of underlying mechanisms of Toxoplasma gondii infection. Invest Ophthamol Vis Sci 45, 2813–2821.
    23. Riordan, C. E., Ault, J. G., Langreth, S. G. & Keithly, J. S. (2003). Cryptosporidium parvum Cpn60 targets a relict organelle. Curr Genet 44, 138–147.
    24. San Martin-Nuñez, B. V., Ordoñez-Escudero, D. & Alunda, J. M. (1988). Preventative treatment of rabbit coccidiosis with α-difluoromethylornithine. Vet Parasitol 30, 1–10.
    25. Seabra, S. H., Da Matta, R. A., de Mello, F. G. & de Souza, W. (2004). Endogenous polyamine levels in macrophages are sufficient to support growth of Toxoplasma gondii. J Parasitol 90, 455–460.
    26. Slapeta, J. & Keithly, J. S. (2004). Cryptosporidium parvum mitochondrial-type HSP70 targets homologous and heterologous mitochondria. Eukaryot Cell 3, 483–494.
    27. Smith, T. A. (1983). Arginine decarboxylase (oat seedlings). Methods Enzymol 94, 176–180.
    28. Thompson, R. C., Olson, M. E., Zhu, G., Enomoto, S., Abrahamsen, M. S. & Hijjawi, N. S. (2005). Cryptosporidium and cryptosporidiosis. Adv Parasitol 59, 77–158.
    29. Wallace, H. M., Fraser, A. V. & Hughes, A. (2003). A perspective of polyamine metabolism. Biochem J 376, 1–14.
    30. Wrenger, C., Lüersen, K., Krause, T., Müller, S. & Walter, R. D. (2001). The Plasmodium falciparum bifunctional ornithine decarboxylase, S-adenosyl-l-methionine decarboxylase, enables a well balanced polyamine synthesis without domain-domain interaction. J Biol Chem 276, 29651–29656.
    31. Yarlett, N. & Bacchi, C. J. (1988). Effect of dl-α-difluoromethylornithine on polyamine synthesis and interconversion in Trichomonas vaginalis grown in semi-defined medium. Mol Biochem Parasitol 31, 1–9.
    32. Yarlett, N., Goldberg, B., Moharrami, M. A. & Bacchi, C. J. (1992). Inhibition of Trichomonas vaginalis ornithine decarboxylase by amino acid analogs. Biochem Pharmacol 44, 243–250.
    33. Yarlett, N., Martinez, M. P., Goldberg, B., Kramer, D. L. & Porter, C. W. (2000). Dependence of Trichomonas vaginalis upon polyamine backconversion. Microbiol 146, 2715–2722.
    34. Yarlett, N., Wu, G., Waters, W. R., Harp, J. A., Wannemuehler, M. J., Morada, M., Athanasopoulos, D., Martinez, M. P., Upton, S. J. & other authors (2007). Cryptosporidium parvum spermidine/spermine N1-acetyltransferase exhibits different characteristics to the host enzyme. Mol Biochem Parasitol (in press).