NITROPHENIDE (MEGASUL™) BLOCKS EIMERIA TENELLA DEVELOPMENT BY INHIBITING THE MANNITOL CYCLE ENZYME MANNITOL-1-PHOSPHATE DEHYDROGENASE
Unsporulated oocysts of the protozoan parasite Eimeria tenella contain high levels of mannitol, which is thought to be the principal energy source for the process of sporulation. Biosynthesis and utilization of this sugar alcohol occurs via a metabolic pathway known as the mannitol cycle. Here, results are presented that suggest that 3-nitrophenyl disulfide (nitrophenide, Megasul™), an anticoccidial drug commercially used in the 1950s, inhibits mannitol-1-phosphate dehydrogenase (M1PDH), which catalyzes the committed enzymatic step in the mannitol cycle. Treatment of E. tenella-infected chickens with nitrophenide resulted in a 90% reduction in oocyst shedding. The remaining oocysts displayed significant morphological abnormalities and were largely incapable of further development. Nitrophenide treatment did not affect parasite asexual reproduction, suggesting specificity for the sexual stage of the life cycle. Isolated oocysts from chickens treated with nitrophenide exhibited a dose-dependent reduction in mannitol, suggesting in vivo inhibition of parasite mannitol biosynthesis. Nitrophenide-mediated inhibition of M1PDH was observed in vitro using purified native enzyme. Moreover, M1PDH activity immunoprecipitated from E. tenella-infected cecal tissues was significantly lower in nitrophenide-treated compared with untreated chickens. Western blot analysis and immunohistochemistry showed that parasites from nitrophenide-treated and untreated chickens contained similar enzyme levels. These data suggest that nitrophenide blocks parasite development at the sexual stages by targeting M1PDH. Thus, targeting of the mannitol cycle with drugs could provide an avenue for controlling the spread of E. tenella in commercial production facilities by preventing oocyst shedding.
Coccidiosis in poultry is caused by the apicomplexan parasites, Eimeria spp. Control and prevention of coccidiosis in commercial production facilities relies on prophylactic administration of anticoccidial drugs in the feed. Although the demand for coccidiostats has increased significantly during the last 20 yr, there have been no new anticoccidial drugs developed during this period. The majority of anticoccidial drugs currently marketed for use against Eimeria spp. parasites in poultry have poorly defined mechanisms of action. Furthermore, the extensive use of these agents in densely populated poultry facilities has led to the development of resistance against most of the anticoccidial drugs currently used to control avian coccidiosis (Dutton et al., 1985; Edgar, 1993; Chapman, 1993; Coombs et al., 1997). There is, therefore, a need to develop new and effective anticoccidial drugs by identifying and targeting novel biochemical pathways in these protozoan parasites.
A metabolic pathway termed the mannitol cycle has been identified in Eimeria. Mannitol is thought to be essential for the completion of the parasite's life cycle by serving as the primary energy source for the process of sporulation (Schmatz et al., 1988, 1989; Michalski et al., 1992; Allocco et al., 1999). The mannitol cycle is an attractive drug target in parasitic protozoa because it is absent in the avian and mammalian hosts of the parasite. In the first committed step of the biosynthetic portion of the Eimeria spp. pathway, fructose-6-phosphate (F6P) is converted to mannitol 1-phosphate (M1P) by mannitol-1-phosphate dehydrogenase (M1PDH). The next enzyme, mannitol-1-phosphatase (M1Pase), generates mannitol by cleaving the phosphate group from M1P. During catabolism, mannitol is reoxidized first to fructose, then phosphorylated to F6P by mannitol dehydrogenase and hexokinase, respectively (Schmatz et al., 1989). Each of the 4 enzymes involved in the mannitol cycle could potentially serve as a suitable chemotherapeutic target, but M1PDH is particularly suited for this purpose since it catalyzes the committed step in the pathway. Consequently, studies have been initiated aimed at the purification and biochemical characterization of Eimeria M1PDH, and identification of potential inhibitors of this enzyme.
Characterization of E. tenella M1PDH using selective thiol reagents has led to the identification of a reactive cysteine residue in or near the active site of the enzyme. Compounds such as N-ethylmaleimide and 5,5′-dithiobis(2-nitrobenzoic acid) were found to be low-micromolar inhibitors of enzyme activity. This finding led to the speculation that the anticoccidial drug nitrophenide, a molecule potentially capable of selectively modifying thiol groups, could be exerting its anticoccidial activity by inhibiting M1PDH, and thus mannitol production in Eimeria. In the present report, it is demonstrated that nitrophenide is a relatively selective and apparently irreversible inhibitor of E. tenella M1PDH both in vitro and in vivo. M1PDH inhibition in vivo leads to a decrease in mannitol production and interruption of the parasite's life cycle, thus accounting for the anticoccidial activity of nitrophenide.
MATERIALS AND METHODS
Purification of M1PDH and enzyme assays
Freshly prepared unsporulated oocysts of E. tenella (5 × 109; 12 ml) were extracted in 10 ml of 10 mM N-(2-hydroxyethyl)piperazine-N′-(2-ethanesulfonic acid) (HEPES)-sodium pH 7.4, 1 mM ethylene glycol-bis(β-aminoethyl ether) N,N,N′,N′-tetraacetic acid, 1 mM dithiothreitol (DTT), 70 mM sucrose, and 210 mM mannitol (buffer A) using a Potter-Elvehjem homogenizer (525 rpm, 5 min). After centrifugation at 1,000 g for 10 min, the supernatant was saved and the pellet reextracted with buffer A and combined supernatants were centrifuged (40,000 g, 30 min; then 100,000 g, 60 min). The final supernatant was immediately loaded onto a HiTrap Q column (Cl− form; 5ml) equilibrated with 10 mM HEPES-sodium pH 7.4, 0.1 mM ethylenediaminetetraacetic acid (EDTA), and 1 mM DTT (buffer B). The column was washed with 10 ml of buffer B and eluted with 0–350 mM NaCl gradient increasing at 5 mM/ml. Fractions 59–68 (nominal [NaCl] = 240–290 mM), containing all M1PDH activity, were pooled, concentrated, and diluted into 25 mM bis-Tris chloride pH 6.0, 0.1 mM EDTA, and 1.0 mM DTT (buffer C) before loading onto a Mono Q HR 5/5 column (Cl− form; 1 ml). The column was washed with 100 mM NaCl in buffer C (5ml) and then eluted using a 100–250 mM NaCl gradient increasing at 3 mM/ml in buffer C. Fractions 32–35 (nominal [NaCl] = 147–159 mM), containing all of the protamine-activated M1PDH activity, were pooled, brought to 1 M (NH4)2SO4, and loaded onto a phenyl superose HR 5/5 column (1 ml) equilibrated with 50 mM Tris-chloride pH 7.0, 1.0 M (NH4)2SO4, 1 mM EDTA, and 1 mM DTT (buffer D). The column was washed with buffer D containing 0.8 M (NH4)2SO4 followed by elution with an 800–400 mM (NH4)2SO4 gradient decreasing at 20 mM/ml in buffer D. Fractions 30–36 (nominal [(NH4)2SO4] = 710–640 mM), containing all M1PDH activity, were pooled and concentrated. This nearly-homogeneous M1PDH was loaded onto a Superose 12 HR 10/30 column (25 ml) equilibrated and eluted with 50 mM Tris-chloride pH 7.5, 0.1 mM EDTA, and 1 mM DTT. Fractions 49–58, containing all of the M1PDH activity, were pooled and concentrated. The final specific activity of this enzyme preparation was 65.4 units/mg, the purity 97%, and the yield 35% after 150-fold purification.
M1PDH activity was measured in a solution containing 50 mM Tris pH 7.5, 1 mg/ml bovine serum albumin (BSA), 10% glycerol, and 40 μg/ml protamine chloride. Purified enzyme (0.1 μg) was preincubated in a volume of 500 μl for 5 min at 41 C with or without drug. The reaction was initiated by the addition of 500 μM F6P and 100 μM NADH final concentration. The reaction rate was determined by measuring NADH oxidation for 5 min at 340 nm with a Shimadzu UV-2101PC spectrophotometer at 41 C.
Glycerol-3-phosphate dehydrogenase (G3PDH) (McLoughlin et al., 1978) from rabbit muscle (Sigma, St. Louis, Missouri) was assayed in a solution containing 50 mM triethanolamine-HCl, 1mM EDTA, and 1mM mercaptoethanol (pH 7.5). The enzyme (0.1 units) was incubated at 30 C for 5 min in the presence or absence of nitrophenide or its analogues. The reaction was initiated with 0.1 mM NADH and 0.3 mM dihydroxyacetone phosphate and the reaction rate was monitored at 340 nm as described for M1PDH.
Drug administration, isolation of oocysts, and sporulation assays
For experimental infections, 3-wk-old broilers (Peterson and Arbor Acre, Avian Services, Frenchtown, New Jersey) were inoculated with 5 × 104 (1 ml) E. tenella unsporulated oocysts (Merck strain LS-18) per bird by gavage. To test for drug effects on oocyst output, or mannitol production, or both, several concentrations of nitrophenide (up to 1,000 parts per million [ppm]) were tested by adding drug to the feed to achieve continuous dosing regimens. For single dosing levels, nitrophenide was administered continuously at 375 ppm in the feed.
Purification of oocysts from isolated cecal cores was accomplished by homogenization, pepsin digestion (pH 2.0), and sucrose flotation (Schmatz et al., 1988). Unsporulated oocysts were treated with 5.25% sodium hypochlorite (Clorox) for 10 min to eliminate bacterial and fungal contaminants, followed by extensive washes in water or buffer (Jackson, 1964). Oocysts were used immediately for gas chromatography–mass spectrometry (GC-MS) or enzyme purification. Sporulation of E. tenella oocysts was accomplished by vigorously shaking the parasites (1 × 107/ml) in phosphate-buffered saline (PBS) at 29 C for 48 hr. The relative percentage of sporulation was determined by phase-contrast microscopy. At the end of this process the sporulated oocysts were either used immediately or they were snap-frozen in an ethanol/dry ice bath and stored at −80 C for analysis of mannitol content.
Tissue preparation and parasite quantification
For quantitative analysis of parasite DNA, mannitol content, and enzyme activity, the ceca were removed from either infected or control birds (2 birds per time point) and homogenized in 15 ml of 10 mM Tris-HCl pH 7.0, 1 mM EDTA, and 0.07 mM phenylmethyl sulfonyl fluoride (PMSF). The samples were either used immediately or stored at −80 C. For DNA quantification, an aliquot of the above suspension was removed and sodium dodecyl sulfate (SDS) was added to a final concentration of 0.5%. The samples were then digested overnight at 55 C using 200 μg/ml proteinase K. DNA from a 600-μl aliquot of the digested homogenate was purified by exhaustive organic solvent extraction and the nucleic acids were precipitated with ethanol. The DNA was resuspended in 600 μl of water and quantified by spectrophotometric analysis. Serial dilutions were then prepared from each sample, the RNA was hydrolyzed, and the genomic DNA was denatured. The DNA samples were immobilized on nylon filters (Nytran; Schleicher and Schuell, Keene, New Hampshire) using a Minifold II slot blot apparatus. The filters were prehybridized in 6× SSPE (1× SSPE: 0.18 M NaCl, 10 mM NaH2PO4, 1 mM EDTA, pH7.4), 1.0% SDS, 50% formamide, 10× Denhardt's solution (1× Denhardt's: 0.02% Ficoll, 0.02% polyvinylpyrrolidone, 0.02% BSA-Pentax Fraction V), and 100 μg/ml denatured salmon sperm DNA at 42 C for 4–16 hr. Hybridization was carried out in essentially the same solution except that the Denhardt's was substituted with 10% dextran sulfate and radiolabeled probe was added to the hybridization mix. The DNA probe that was used consisted of an EcoR1 insert isolated from E. tenella cDNA clone SO7 (Liberator et al., 1989). Washes were performed under stringent conditions in 0.1× SSPE, 0.1% SDS at 65 C. To determine parasite numbers, signals were quantified directly from the filters using the PhosphorImager 400E (Molecular Dynamics, Sunnyvale, California). Radioactive emissions were converted to organism number by including a standard curve in each hybridization reaction, using genomic DNA prepared from E. tenella LS18 sporozoites.
Analysis of mannitol content by GC-MS
Frozen tissue samples, composed of homogenized whole ceca with their contents, were lyophilized overnight in preweighed tubes and dry weights were noted before rehydration with water. For an internal standard, 500 μg of α-methylglucoside was added to each sample (Wong et al., 1990). The samples were vortexed for 10 min with 3-mm glass beads, sonicated, and heated to 100 C to ensure complete disruption of mature oocysts and release of intracellular components. Protein was precipitated overnight in 80% acetone at −20 C, followed by centrifugation at 100,000 g for 30 min. The supernatant was collected and dried down under nitrogen in a 70 C water bath. To prepare trimethylsilyl (TMS) derivatives, 1 ml of Tri-Sil® reagent (Pierce, Rockford, Illinois) was added and the samples were heated for 1 hr at 80 C.
The TMS-derivatized samples were analyzed by GC-MS on a Hewlett Packard 5890 gas chromatograph fitted with a 5972 mass selective detector. The sample (1 μl) was injected at a split ratio of 1:100 on a 0.25 mm × 30 m HP-5 capillary column (0.25-μm film thickness, Hewlett Packard) with helium as carrier gas at a linear velocity of 26.7 cm/sec. The temperatures were 250 C for the injector, 280 C for the detector, and 145 C for the oven at the start. Oven temperature was held constant for 3 min, then ramped to 215 C at a rate of 10 C/min, 216 C at 0.1 C/min, 280 C at 70 C/min, and then held constant for 4 min. The mass selective detector was set to scan mode for a range of 60–650 amu, a threshold of 50, and a sampling number of 2 (4 scans/0.1 amu, 78 scans/min). The retention time and mass spectrum of authentic TMS-mannitol was used to identify mannitol from tissue samples. For quantification of tissue mannitol levels, the instrument was calibrated against standard curves of authentic TMS-mannitol containing α-methylglucoside as the internal standard (Schmatz et al., 1988).
Antibodies to M1PDH and Western blot analysis
Purified M1PDH from E. tenella was run on and excised from preparative SDS-polyacrylamide gels (stained with 0.05% Coomassie Blue R250) and used to elicit polyclonal antiserum in pathogen-free New Zealand White rabbits. In the primary immunization, 68 μg of protein was administered in Freund's complete adjuvant. Rabbits were boosted 4 times at 20-day intervals with 34 μg of purified protein in incomplete Freund's adjuvant and serum was collected 1 wk after each boost.
For Western blot analysis, cecal tissue extracts were prepared by homogenization in 50 mM Tris-HCl pH 7, 1 mM EDTA, and 0.1 mM PMSF. Nucleic acids were removed from the protein samples by precipitation with 0.2% w/v protamine sulfate, followed by incubation on ice for 20 min and centrifugation at 100,000 g for 30–60 min. The protein content of the supernatant was quantified by the dye binding assay (Bradford, 1976). Protein samples were electrophoresed on 10% SDS-polyacrylamide gels (Laemmli, 1970) and electrophoretically transferred onto nitrocellulose membranes (Towbin et al., 1976). Blots were probed with a 1:100 dilution of preimmune rabbit serum or with antiserum to E. tenella M1PDH. Specific antibody binding was detected with 1.4 × 107 counts per minute/blot of [125I]-labeled goat anti-rabbit IgG and quantified directly from the filters using the PhosphorImager 400E (Molecular Dynamics, Sunnyvale, California).
Immunoprecipitation
Protein-A agarose beads were preincubated with an excess of rabbit antisera raised to E. tenella M1PDH or preimmune serum overnight at 4 C or for 30 min at room temperature with constant agitation in PBS. The beads were washed 3 times with 50 mM Tris/150 mM NaCl pH 8.5 to remove any unbound proteins. Bound antibody was cross-linked to the agarose beads with dimethylpimelimidate (Harlow and Lane, 1988). To immunoprecipitate M1PDH, cross-linked antibody was incubated with cecal extracts in enzyme assay buffer for 2 hr at 4 C and complexes were washed 3 times with M1PDH assay buffer. For assay of immunoprecipitated M1PDH, 100 μl of immune protein-A agarose beads were added to the reaction mixture and incubated for 5 min at 41 C. The reaction was initiated with 2.5 mM F6P and 200 μM NADH and the absorbance change was monitored at 340 nm.
Immunohistochemistry
Cecal tissues from E. tenella-infected chickens with and without nitrophenide treatment were obtained at 144 hr postinfection, fixed in 3% paraformaldehyde/PBS, paraffin embedded, and sectioned. Sections were deparaffinized, rehydrated, and blocked with 10% normal swine serum overnight at room temperature. Sections were washed 3 times in PBS and incubated (1 hr at room temperature) with 1:100 dilution of either preimmune serum or antiserum to E. tenella M1PDH that was preabsorbed with a lyophilized cecal preparation from uninfected chicken. Slides were developed with peroxidase–antiperoxidase (PAP) (Dako PAP Kit), a procedure that utilizes soluble PAP immune complexes instead of conjugated antibodies, which results in higher sensitivity. The slides were counterstained with Mayer's hematoxylin and mounted with DAKO glycergel C563. Bright field images were taken on a Leitz microscope fitted with an Orthomat 35 mm camera.
RESULTS
Inhibition of M1PDH by nitrophenide and effects on oocyst output
Purified M1PDH is inhibited by an array of reagents capable of specifically modifying cysteine thiol groups and can be protected from inactivation by substrates. These data suggested the presence of a reactive cysteine thiol group in or near the active site of the enzyme. Examination of the available anticoccidial agents led to the identification of nitrophenide, a likely thiol reagent. This observation led us to examine the effects of nitrophenyl disulfides on the activity of M1PDH in vitro. M1PDH was sensitive to nitrophenyl disulfide compounds with IC50 values (inhibitor concentration required to reduce dehydrogenase activity by 50%) of 3, 68, and 256 μM for meta, ortho, and para isomers respectively (Table I). Interestingly, the most potent isomer was the meta isomer, corresponding to nitrophenide. To assess whether inhibition of M1PDH by nitrophenide was selective for this enzyme, the 3 nitrophenyl disulfide isomers were also evaluated against G3PDH, an enzyme that catalyzes a reaction similar to that catalyzed by M1PDH and that is known to be sensitive to general sulfhydryl reagents (Smith and MacQuarrie, 1979). G3PDH was not inhibited by any of the nitrophenyl disulfide isomers when tested at 350 μM (Table I). When meta, ortho, and para isomers of nitrophenyl disulfides were tested for their ability to control coccidiosis in chickens, only the most potent M1PDH inhibitor, i.e., nitrophenide, exhibited activity (250–500 ppm). Nitrophenide in the feed exhibited a dose-dependent reduction in oocyst output, with a 90% reduction at 500 ppm (Fig. 1). Despite this, parasite-induced lesions in nitrophenide-treated and untreated E. tenella-infected chickens were similar.
Citation: Journal of Parasitology 87, 6; 10.1645/0022-3395(2001)087[1441:NMBETD]2.0.CO;2
Effects of nitrophenide treatment on mannitol content and sporulation in E. tenella oocysts
M1PDH catalyzes the first, committed step of the mannitol cycle. We predicted that the inhibition of M1PDH by nitrophenide in vivo would result in diminished mannitol content due to inhibition of E. tenella-mediated mannitol biosynthesis. This prediction was evaluated by examining mannitol levels in oocysts from chickens that were infected with E. tenella and subsequently treated with different concentrations of nitrophenide in the feed. Mannitol content was reduced in a dose-dependent manner (nearly 80% reduction at 500 ppm) in oocysts shed from nitrophenide-treated chickens at day 7 postinfection (Fig. 2). Previous studies have established that mannitol is an essential component of the developmental transition from the unsporulated oocyst to the sporozoite stage of E. tenella (Michalski et al., 1992; Allocco et al., 1999). Therefore, the consequence of nitrophenide treatment on this developmental process was evaluated. The sporulation efficiency of oocysts recovered from nitrophenide-treated infected chickens exhibited a dose-dependent reduction in development, with up to 40% reduction at the highest dose (500 ppm) tested (Fig. 2). When sporulated oocysts from nitrophenide-treated infected chickens were inspected by phase-contrast microscopy, significant abnormalities were observed (Fig. 3). Normal sporulation produces 4 sporocysts, each containing 2 sporozoites, but nitrophenide treatment resulted in distortions in both numbers and morphology of sporocysts. In some cases, it appeared that the zygote failed to proceed beyond first division or produced many segments resulting in 8 or more daughter cells (Fig. 3).
Citation: Journal of Parasitology 87, 6; 10.1645/0022-3395(2001)087[1441:NMBETD]2.0.CO;2
Citation: Journal of Parasitology 87, 6; 10.1645/0022-3395(2001)087[1441:NMBETD]2.0.CO;2
Kinetics of nitrophenide effects on E. tenella and parasite population dynamics
To establish the effects of nitrophenide on the developmental stages and mannitol content in E. tenella in vivo, we examined the kinetics of oocyst output and mannitol content over a 10-day period after infection. Oocyst shedding from untreated birds was biphasic, with peaks at days 7 and 9. The second phase of oocyst shedding can be attributed to heterogeneity. The inoculum may have become contaminated with a slower-developing variant of E. tenella because of multiple chicken passages. Nitrophenide treatment (375 ppm) depressed oocyst shedding by greater than 90% between days 6 and 10 (Fig. 4A). Daily monitoring of mannitol content in nitrophenide-treated infected chicken ceca versus nonmedicated controls revealed a pattern that mimicked the kinetics of oocyst shedding (Fig. 4B vs. Fig. 4A). After inoculation with oocysts of E. tenella, the numbers of parasites in infected chickens increased with subsequent rounds of asexual divisions (days 1–6) and then plateaued during transformation into sexual stages (days 6–8) in which there was no significant replication. Nitrophenide treatment (375 ppm) does not affect the parasite load during these stages of development (Fig. 4C), despite the clear inhibition of mannitol biosynthesis in the same birds. From these observations it was concluded that the decrease in mannitol content in nitrophenide-treated chickens does not arise from a simple decrease in overall parasite load in the host. These findings are consistent with those displayed in Table I in which there was a reduction in oocyst output due to nitrophenide treatment without a corresponding reduction in lesion size and severity.
Citation: Journal of Parasitology 87, 6; 10.1645/0022-3395(2001)087[1441:NMBETD]2.0.CO;2
M1PDH expression in nitrophenide-treated infected chickens
Results obtained in the present study suggest that inhibition of M1PDH may be responsible for the reduction in mannitol biosynthesis and consequently, reduction in E. tenella oocyst shedding in nitrophenide-treated chickens. Direct evidence for such a relationship was obtained by quantification of M1PDH expression levels and M1PDH activity using immunoprecipitation and Western blot assays. Both mannitol levels and M1PDH enzyme activity are decreased greater than 95% in cecal tissue from nitrophenide-treated chickens (Fig. 5). In contrast, equivalent amounts of M1PDH were present in nitrophenide-treated infected and control chickens (Fig. 5), establishing that the enzyme is present, but inhibited in tissues from drug-treated birds. Biochemical (Fig. 5) and immunohistochemical (Fig. 6) data on M1PDH expression in nitrophenide-treated versus untreated infected chickens are concordant and serve to establish that drug treatment does not affect enzyme expression.
Citation: Journal of Parasitology 87, 6; 10.1645/0022-3395(2001)087[1441:NMBETD]2.0.CO;2
Citation: Journal of Parasitology 87, 6; 10.1645/0022-3395(2001)087[1441:NMBETD]2.0.CO;2
DISCUSSION
Significance of the mannitol cycle
The discovery of the mannitol cycle in E. tenella (Schmatz et al., 1988, 1989; Schmatz, 1997) and subsequent characterization of the various enzymes involved in this pathway enabled the identification of novel, interesting targets for chemotherapeutic intervention against avian coccidiosis. The mannitol cycle is also present in other pathogenic apicomplexan protozoans of humans such as Toxoplasma gondii and Cryptosporidium parvum, providing an opportunity for broader therapeutic potential. The absence of the mannitol biosynthetic pathway in hosts of protozoan parasites fulfills 1 of the major criteria of a good target for the discovery of antiparasitics. Furthermore, previous studies have demonstrated that mannitol is essential for the developmental process of sporulation. Although all 4 enzymes involved in the mannitol cycle could serve as potential targets for the design of specific inhibitors to block mannitol biosynthesis, M1PDH may represent the ideal choice since it catalyzes the first committed step in this pathway.
Nitrophenide targets the mannitol cycle
The presence of a cysteine thiol group in or near the catalytic site is a common feature of many dehydrogenases. Consistent with this, initial biochemical characterization of M1PDH suggested that thiol reagents inhibited this enzyme in an active site-directed manner. The current studies demonstrate that nitrophenide, an aryl disulfide thiol reagent marketed as an anticoccidial in the 1950s (Waletsky et al., 1949; Peterson and Hymas, 1950), is in fact an inhibitor of M1PDH and mannitol biosynthesis. As a thiol reagent, nitrophenide is unlikely to be entirely selective for M1PDH versus other proteins. Nevertheless, some evidence for selectivity has been obtained. First, the meta isomer of nitrophenide was 23- and 85-fold more active against M1PDH than the ortho and para isomers respectively, and this correlated with activity against E. tenella in vitro and in vivo. Second, nitrophenide has no activity against G3PDH, a dehydrogenase enzyme that is sensitive to general sulfhydryl reagents (Smith and MacQuarrie, 1979). Although nitrophenide is not particularly potent against M1PDH and E. tenella in vivo, its ability to inhibit mannitol biosynthesis provides a likely explanation for its antiparasitic mechanism of action and provides a new opportunity for evaluating M1PDH as an anticoccidial target (Waletsky et al., 1949; Peterson and Hymas, 1950).
Role of mannitol in parasite development
It has been demonstrated that in chickens treated with nitrophenide doses less than 1,000 ppm, only a small proportion of the oocysts that are shed are capable of sporulation and that a significant number of these contain developmental abnormalities. Specific effects on the development of sexual stages, specifically the zygote, and abnormal oocyst development due to nitrophenide treatment, have been previously reported (Brackett and Bliznick, 1949). During the development of E. tenella, mannitol biosynthesis is upregulated 250-fold during oogenesis, resulting in a tremendous accumulation of this sugar in oocysts (Allocco et al., 1999). Delayed medication experiments (data not shown) in which drug treatment was initiated at day 5 postinfection produced the same results as continuous treatment from day 1. This result provided additional evidence for the specificity of nitrophenide for the sexual stages of the parasite life cycle. These and other observations led us to speculate that mannitol has an essential role in oogenesis and sporulation and this may explain the nitrophenide-induced malformations that have been observed (Brackett and Bliznick, 1949). The observation that nitrophenide effects are restricted to life cycle stages in which mannitol is synthesized and utilized is highly significant. The lack of activity of nitrophenide against the asexual stages of E. tenella allows for their free replication and development, resulting in significant pathology despite drug administration. Nevertheless, these observations provide strong evidence that reduction in mannitol content in oocysts is a direct result of M1PDH inhibition rather than an indirect consequence of parasite killing via an unrelated mechanism.
M1PDH as a drug target
Administration of nitrophenide or another M1PDH inhibitor to E. tenella-infected chickens would not prevent proliferation of asexual stages of the parasite, but would prevent or reduce shedding of oocysts by infected poultry. Given that E. tenella oocysts are responsible for transmission of coccidiosis from one host to the next, prevention of oocyst shedding or blocking their sporulation could provide an effective avenue to prevent the spread of the disease. Targeting the sexual stages of other pathogens such as C. parvum, in which oocyst cycling is critical for passage of the infection from host to host, offers some promise for control of these organisms using inhibitors of mannitol biosynthesis. Unfortunately, targeting the mannitol cycle in other apicomplexan protozoans such as T. gondii may be of limited value since the sexual cycle occurs in the intermediate rather than the definitive host, unless mannitol plays a role in the asexual stages. In conclusion, the present study identifies the mannitol cycle, specifically M1PDH, as the primary target for the anticoccidial agent nitrophenide and points to its potential as a target for the development of new strategies to prevent the transmission of coccidiosis.