Plasmodium falciparum, the causative agent of the fatal form of malaria, synthesizes GMP primarily from IMP and, hence, needs active GMPS (GMP synthetase) for its survival. GMPS, a G-type amidotransferase, catalyses the amination of XMP to GMP with the reaction occurring in two domains, the GAT (glutamine amidotransferase) and ATPPase (ATP pyrophosphatase). The GAT domain hydrolyses glutamine to glutamate and ammonia, while the ATPPase domain catalyses the formation of the intermediate AMP-XMP from ATP and XMP. Co-ordination of activity across the two domains, achieved through channelling of ammonia from GAT to the effector domain, is the hallmark of amidotransferases. Our studies aimed at understanding the kinetic mechanism of PfGMPS (Plasmodium falciparum GMPS) indicated steady-state ordered binding of ATP followed by XMP to the ATPPase domain with glutamine binding in a random manner to the GAT domain. We attribute the irreversible, Ping Pong step seen in initial velocity kinetics to the release of glutamate before the attack of the adenyl-XMP intermediate by ammonia. Specific aspects of the overall kinetic mechanism of PfGMPS are different from that reported for the human and Escherichia coli enzymes. Unlike human GMPS, absence of tight co-ordination of activity across the two domains was evident in the parasite enzyme. Variations seen in the inhibition by nucleosides and nucleotide analogues between human GMPS and PfGMPS highlighted differences in ligand specificity that could serve as a basis for the design of specific inhibitors. The present study represents the first report on recombinant His-tagged GMPS from parasitic protozoa.

INTRODUCTION

Plasmodium falciparum, a parasitic protozoan that is the causative agent of cerebral malaria, is responsible for approx. 2 million deaths annually [1]. Despite the availability of antimalarial drugs, the widespread emergence of drug-resistant parasites necessitates a quest for new therapies. Structure- and mechanism-based combinatorial approach for drug design has proved highly fruitful. All parasitic protozoa are auxotrophic for purine residues and use different reactions of the salvage pathway for the synthesis of purine nucleotides [2]. The purine salvage pathway of P. falciparum is a novel target for antimalarials, as the parasite lacks the de novo purine biosynthetic pathway [3]. During the intra-erythrocytic stages of P. falciparum, GMP is mainly formed from IMP through a two-step process involving the enzymes IMPDH (inosine monophosphate dehydrogenase) and GMPS (GMP synthetase). Parasites and human erythrocytes also possess HGPRT (hypoxanthine guanine phosphoribosyltransferase) and, therefore, phosphoribosylation of guanine by this enzyme could be an alternative route for GMP production. However, as free guanine levels in human blood are very low [4], GMP production through HGPRT would be insufficient for the rapidly multiplying parasite. Human blood contains higher levels of hypoxanthine [4], making this purine base the key precursor for the generation of both AMP and GMP. Also, both human erythrocytes and P. falciparum lack guanosine kinase, eliminating the possibility of generating GMP from guanosine. The potent antimalarial activities of mycophenolic acid, an inhibitor of IMPDH, and bredinin, an inhibitor of both IMPDH and GMPS, underscore the necessity of GMPS for the survival of intra-erythrocytic P. falciparum[5,6]. McConkey [7] reported the DNA sequence of PfGMPS (P. falciparum GMPS) and the absence of any introns in the gene.

GMPS (EC 6.3.5.2) catalyses the last step in GMP formation in the purine salvage pathway of P. falciparum. This enzyme belongs to the family of G-type GATs (glutamine amidotransferases) that utilizes glutamine as a source of ammonia for the amination of their substrates. Ammonia channelling, where ammonia is transferred from one domain to another within the protein, is a characteristic feature of amidotransferases [8,9]. GMPS catalyses the irreversible conversion of XMP into GMP, a reaction that requires ATP and involves the transfer of an amino group from glutamine to the C-2 carbon of XMP via an adenyl-XMP intermediate [1013]. The overall reaction catalysed by GMPS is:

 
formula
 
formula

GMPSs from both prokaryotes and eukaryotes consist of two domains that are associated with distinct functions. The C-terminal ATPPase (ATP pyrophosphatase) domain catalyses the condensation of XMP with ATP, and the N-terminal glutaminase domain liberates ammonia by the hydrolysis of glutamine. GMPS has been used as an anticancer and immunosuppressive drug target [14,15]. GMPSs have been biochemically characterized from human [1618], Escherichia coli [1923], Yoshida sarcoma ascites cells [24] and Ehrlich ascites cells [25]. The only crystal structure available to date for GMPS is that from E. coli [26,27]. Kinetic characterization of E. coli and human GMPSs show differences in the order of substrate binding. In E. coli GMPS, ATP is the first substrate to bind followed by XMP, while glutamine or ammonia binds randomly [20]. However, in the case of human GMPS, inhibition patterns obtained with decoyinine, an analogue of adenosine, with respect to the three substrates suggest that ATP may not be the first substrate to bind to the enzyme [17].

The present paper reports on the expression and biochemical and kinetic characterization of PfGMPS and, to our knowledge, is the first detailed study on GMPS from any parasitic protozoan. We have used product inhibition and initial velocity kinetics to elucidate the order of substrate binding, which shows differences from that reported for E. coli and human GMPSs, and in general adds to the understanding of the kinetic mechanism operating in this class of enzymes. We also report on the inhibition of PfGMPS activity by the irreversible inhibitors, acivicin [L-(αS,5S)-α-amino-3-chloro-4,5-dihydro-5-isoxazoleacetic acid] and DON (6-diazo-5-oxo-L-norleucine), which throws light on the regulation of activity across the two domains in PfGMPS. The present study also shows that guanine and its derivatives are inhibitors of PfGMPS activity, which could provide a lead in the design of antimalarial drugs acting through this enzyme.

MATERIALS AND METHODS

Restriction enzymes, Pfu DNA polymerase and T4 DNA ligase were obtained from Invitrogen and Bangalore Genei and were used according to the manufacturer's instructions. E. coli strain AT2465 was obtained from the E. coli Genetic Stock Center (Yale University, New Haven, CT, U.S.A.). Primers were custom-synthesized at Microsynth. All chemical reagents were of high quality and obtained from Sigma Chemical Co., Alexis Biochemicals or Merck India. Media components were from Himedia.

Cloning, complementation, expression and purification

Primers were designed using the annotated gene sequence of PfGMPS (PF10_0123) available in the P. falciparum genome database PlasmoDB (http://www.plasmodb.org) [28]. The primers 5′-GCAGGATCCATGGCAGAAGGAGAGGAATATGACAAGATTTTG-3′ and 5′-GTCCTCGAGCTGCAGTCATTCGATTCAATCGTTGCTGGTG-3′ that contain the restriction sites BamHI and PstI (underlined) were used for the amplification of the gene by PCR, using parasite genomic DNA as a template. The amplified fragment was cloned into the E. coli expression vector pQE30 and DNA sequencing of the clone pQE30PfGMPS with the N-terminal His tag confirmed complete identity with the PlasmoDB entry. Complementation studies were carried out using E. coli strain AT2465 (λ, e14, guaA21, relA1, spoT1 and thi1) transformed with the expression construct pQE30PfGMPS. Single colonies were inoculated into 5 ml of minimal media without guanine or guanosine, prepared according to the method of Hirst et al. [29], with 100 μg·ml−1 ampicillin and grown overnight. A 100 μl portion of this culture was spread on minimal-medium plates and examined for the appearance of colonies. An appropriate control with pQE30 vector was also grown at the same time. The E. coli strain AT2465 was also used for hyperexpression of PfGMPS. The hyperexpressed protein was purified using Ni-NTA (Ni2+-nitrilotriacetate) beads followed by Sephacryl 200 (37.5 cm×1.6 cm) size-exclusion chromatography. The purified protein was stored in 20 mM Tris/HCl (pH 7.4), 2 mM DTT (dithiothreitol), 10% (v/v) glycerol and 1 mM EDTA at –80 °C. Protein concentration was estimated by the method of Bradford [30]. The molecular mass was confirmed by MS by using a MALDI (matrix-assisted laser-desorption ionization) mass spectrometer (Ultraflex II; Bruker). The protein sample was dialysed against water, 1 μl of protein (3 μg) was mixed with sinapinic acid (0.1 g of sinapinic acid in 0.5 ml of 50% acetonitrile and 0.1% trifluoroacetic acid) in 1:1 ratio and spotted on the MALDI target plate and mass spectra were recorded in the positive ion mode.

Analytical gel filtration

Analytical gel filtration was performed on a Superdex-200 column (1 cm×30 cm) attached to an AKTA Basic HPLC system. The column was equilibrated with Tris/HCl (pH 7.4) and calibrated with β-amylase (200 kDa), alcohol dehydrogenase (150 kDa), BSA (66 kDa), carbonic anhydrase (29 kDa) and cytochrome c (12.4 kDa). Then, 30 μM (100 μl) PfGMPS was injected into the column and it was eluted at a flow rate of 0.5 ml·min−1 by using Tris/HCl (pH 7.4) as the eluent with detection at 280 nm. To monitor the stability of the oligomer, protein samples were pre-incubated for 30 min with 150 mM–1.5 M NaCl. The elution buffer contained the appropriate concentration of NaCl.

Assay for glutaminase activity of PfGMPS

The glutaminase activity of the GAT domain was monitored spectrophotometrically at 340 nm as the formation of NADH in which glutamate production was coupled with the reduction of NAD+ to NADH with the release of 2-oxoglutarate using glutamate dehydrogenase [31]. The assay volume of 0.1 ml containing 90 mM Tris/HCl (pH 8.5), 5 mM glutamine, 0.1 mM EDTA, 0.1 mM DTT and 6 μg of PfGMPS was incubated at 37 °C for 20 min with different combinations of 150 μM XMP, 2 mM ATP, 2 mM p[NH]ppA (adenosine 5′-[β,γ-imido]triphosphate) and 20 mM MgCl2. The reaction was terminated by heating the sample in a boiling water bath for 3 min, chilled on ice and centrifuged at 13600 g for 10 min. Levels of glutamate in the supernatants were determined using glutamate dehydrogenase (Sigma Chemical Co.). A total reaction volume of 0.3 ml containing 50 mM Tris/HCl (pH 8.0), 50 mM KCl, 1 mM EDTA, 0.5 mM NAD and 2 units of glutamate dehydrogenase was incubated for 90 min at 37 °C and clarified by centrifugation at 13600 g for 5 min. Conversion of NAD+ into NADH was monitored as the absorbance at 340 nm, and the amount of product formed was estimated using the molar absorption coefficient (ϵ) of 6220 M−1·cm−1. All assays were performed in duplicate and the experiment was repeated twice.

Assay for GMPS activity

PfGMPS activity was monitored spectrophotometrically using Hitachi U-2010 or U-2810 spectrophotometer. Reaction rates were monitored as decrease in absorbance at 290 nm due to conversion of XMP (ϵ 4800 M−1·cm−1) into GMP (ϵ 3300 M−1·cm−1). A Δϵ value of 1500 M−1·cm−1 was used to calculate the amount of product formed [10]. The standard assay consisted of 90 mM Tris/HCl (pH 8.5), 150 μM XMP, 2 mM ATP, 5 mM glutamine, 20 mM MgCl2, 0.1 mM EDTA and 0.1 mM DTT in a total reaction volume of 0.25 ml. Reactions were initiated with 6 μg of PfGMPS and monitored at 25 °C. The assay conditions were the same in all the studies unless otherwise mentioned. The effect of phosphate ions on PfGMPS activity was studied in a concentration range of 0.1–10 mM. Univalent (Na+, K+ and Li+) and bivalent (Ca2+, Mn2+, Co2+ and Zn2+) metal ions as chloride salts were examined for their effect on PfGMPS activity in the concentration range of 0–500 mM and 0–100 mM respectively. To monitor the effect of different natural and modified analogues of nucleosides, nucleotides and purine bases on PfGMPS activity, the concentrations of XMP, ATP and glutamine were fixed at 45 μM, 750 μM and 1.5 mM respectively. Specific activity was calculated from the initial velocity and data were fitted to different equations by using GraphPad Prism, version 4.0. The kinetic constants for XMP, ATP, glutamine and NH4+ [as (NH4)2SO4] were determined by varying the concentration of one substrate while keeping the other two at saturating concentrations. The concentrations of the substrates when fixed were: XMP, 0.15 mM; ATP, 2 mM; and glutamine, 5 mM. The initial velocity data were fitted to the Michaelis–Menten equation (eqn 1) and analysed by non-linear regression using the software GraphPad Prism.

 
formula
(1)

Initial velocity kinetics

All initial velocity measurements were made under standard assay conditions at varying concentrations of substrates. For this study, the method of Frieden [32] was used in which concentrations of two substrates were varied while maintaining the third substrate at a fixed saturating concentration. Of the two different substrates, the concentration of one was varied regularly while the second was maintained at different fixed concentrations. Each experiment was repeated three to four times to confirm reproducibility.

Non-linear analysis was performed by fitting the data to rate equations, (2) for sequential and (3) for Ping Pong bireactant mechanism. For the best fit, values for S.E.M. and 95% confidence intervals were taken as indicators. The nomenclature used in the equations is that of Cleland [33].

 
formula
(2)
 
formula
(3)

where v=initial velocity, Vmax=maximal velocity, [A] and [B] are concentrations of the substrates A and B, and KA and KB are their respective Michaelis–Menten constants. Kia is the dissociation constant of the substrate A. In eqn (2), A and B refer to substrates ATP and XMP respectively when glutamine was the fixed substrate. When glutamine was replaced with ammonia, A refers to either of the substrates ATP and XMP and B refers to NH4+. In eqn (3), A refers to either of the substrates ATP and XMP and B refers to glutamine.

Kinetics of inhibition by products and substrate analogues

Inhibition of PfGMPS activity by products (GMP, PPi, AMP and glutamate), non-hydrolysable ATP analogue p[NH]ppA and the glutamine analogue, glutamic acid-γ-methyl ester, were examined by varying one substrate at different fixed concentrations of the inhibitor. The other two substrates were kept at fixed concentrations. All experiments were repeated at least twice. The data were fitted to eqns (4–6) for competitive, uncompetitive and non-competitive inhibition respectively.

 
formula
(4)
 
formula
(5)
 
formula
(6)

where [S], Km and Ki are the substrate concentration, Michaelis–Menten constant and inhibition constant respectively. The other terms are the same as those in eqns (2) and (3).

Irreversible inhibition of PfGMPS

Inhibition by two irreversible glutamine analogues, acivicin and DON, was examined by pre-incubation of the enzyme with the inhibitors. Pre-incubation was performed both in the presence and absence of the substrates XMP and ATP in 50 mM Tris/HCl (pH 8.5), with 20 mM MgCl2, keeping the inhibitor concentration always in excess of the enzyme concentration. Pre-incubation in the absence of substrates was carried out both at 25 and 15 °C, and at 15 °C only in the presence of substrates as the inactivation was very rapid at 25 °C. In both the cases, aliquots containing 2.5 μg of the pre-incubated enzyme were taken at different time points and assayed under standard assay conditions. Eqn (7) describes the mechanism of inactivation in which the first step occurs reversibly and the second step is irreversible [34].

 
formula
(7)

where ϵ is inactivated enzyme, E0 is total enzyme, KI is dissociation constant for the initial reversible complex and k3 is first-order rate constant for the conversion of reversible complex into the irreversibly inhibited enzyme.

Dependence of PfGMPS activity on Mg2+ ions

The effect of varying MgCl2 concentration on PfGMPS activity was monitored at a fixed ATP concentration of 2 mM with XMP and glutamine at 0.15 and 5 mM respectively. To estimate the number of Mg2+-binding sites on PfGMPS, MgCl2 was fixed at 2 mM, while ATP concentration was varied. Standard assay conditions were used for activity measurements. Eqns (8) and (9) were solved simultaneously for the calculation of MgATP concentration [35,36].

 
formula
(8)
 
formula
(9)

[M]t, [ATP]t,, [MATP], [M] and [H] represent the total Mg2+, total ATP, total MgATP2−, free Mg2+ and free hydrogen ion concentrations respectively. K1, K2 and KH represent the dissociation constants for MgATP2−, MgHATP and HATP3− respectively and the values used were according to Nakamura and Lou [17].

RESULTS AND DISCUSSION

Functional complementation and preliminary characterization of recombinant PfGMPS

The Plasmodium genome database PlasmoDB carries a single annotation for GMPS that lacks introns. PfGMPS was cloned earlier from genomic DNA and the sequence was deposited in GenBank® (accession number AAF09184) [7]. IPTG (isopropyl β-D-thiogalactoside)-induced BL21DE3 cells transformed with pQE30PfGMPS (pQE30 expression vector carrying the PfGMPS gene) did not show expression of the recombinant protein on SDS/PAGE. Use of BL21DE3 cells harbouring the plasmids RIG or RIL, known to overcome codon bias [37], or use of M15 cells with pREP plasmid [38] also did not lead to expression of PfGMPS. However, pQE30PfGMPS complemented the gua A (GMPS) deficiency in the E. coli cells AT2465 (λ, e14, guaA21, relA1, spoT1 and thi1), which were grown on minimal medium without supplementation with guanine or guanosine (see Supplementary Figure 1A at http://www.BiochemJ.org/bj/409/bj4090263add.htm). This showed that the clone produced functional GMPS and, on SDS/PAGE, an overproduction of protein of the required molecular mass of 65 kDa was seen (Supplementary Figure 1B). Hence, the use of mutant AT2465 E. coli cells that lack the gua A gene not only confirmed the functionality of the clone, but also proved to be a good expression system for PfGMPS. Purification of recombinant PfGMPS from the induced cell lysate using Ni-NTA affinity matrix followed by size-exclusion chromatography yielded 20 mg of pure protein from 1 litre of culture. The purified His-tagged protein showed a molecular mass of 65253.9 Da by MALDI mass measurement, while the calculated molecular mass of the protein without the initiator methionine is 65254 Da (Supplementary Figure 1C). The subunit molecular mass of PfGMPS is comparable with the bacterial (58 kDa) and yeast proteins (59 kDa), but smaller than the human counterpart (77 kDa), which contains a long insertion in the ATPPase domain.

Examination of the subunit association of PfGMPS by size-exclusion chromatography gave a single protein peak at 12.4 ml corresponding to a molecular mass of 160 kDa (see Supplementary Figure 2A at http://www.BiochemJ.org/bj/409/bj4090263add.htm). The expected molecular mass of the dimer of PfGMPS is 131 kDa. On addition of NaCl over the concentration range 150 mM–1.5 M, the peak shifted slightly towards a higher retention volume of 12.7 ml, corresponding to a molecular mass of 144.5 kDa. However, there was no complete shift of the peak towards a monomer (14.7 ml). Hence, in the presence or absence of NaCl, the estimated molecular mass was slightly higher than that of a dimer. This possibly arises from an elongated dimeric form of PfGMPS. Like PfGMPS, E. coli GMPS is a dimer [19,26], while enzymes from higher eukaryotes including human have been shown to be monomers [11,16,17,25]. The presence of large insertions near the dimerization domain in GMPSs of higher eukaryotes seems to preclude them from forming dimers [27]. This reflects the diversity in oligomeric status of this class of enzymes and, hence, they may exhibit different modes of regulation and varied functions.

Tris/HCl was found to be the most suitable buffer for activity measurements, while the activity in 20 mM potassium phosphate (pH 7.4) was 4-fold lower. Addition of phosphate to the assay mixture in Tris/HCl (pH 8.5) led to inhibition of enzyme activity with an IC50 of 2.7 mM. Magnesium was essential for catalysis and could not be replaced effectively by other bivalent metal ions. Replacement of Mg2+ with Mn2+ and Ca2+ lead to 70 and 85% decrease in activity respectively. Zn2+ and Co2+ did not support activity. Univalent metal ions Na+, K+ and Li+ inhibited PfGMPS activity with Na+ having the highest effect (Supplementary Figure 2B). At 24 mM Na+, 25% decrease in activity was seen that increased to 90% in 200 mM NaCl solution. K+ and Li+ also affected the activity, but to a lesser extent. This possibly indicates the presence of high-affinity binding sites for Na+ on PfGMPS, which upon occupancy inhibit the catalytic activity.

The Km values from saturation kinetics for XMP, ATP, glutamine and ammonium ions are shown in Table 1. The Vmax value was found to be 450 nmol·min−1·mg−1. PfGMPS has the highest affinity towards XMP and its Km value is lower than that for human and E. coli GMPSs. The Km values for ATP and glutamine are comparable with that for the human protein but lower than that for E. coli GMPS [17,19]. PfGMPS was also found to directly utilize ammonia for the amination of XMP, although with a significantly higher Km, indicating that the preferred physiological substrate must be glutamine. Similar observations have been reported for other eukaryotic GMPSs [25]. PfGMPS does not show any co-operativity with respect to ATP, XMP and glutamine or NH4+, as all the v against [S] plots exhibited Michaelis–Menten behaviour and the Eadie–Hofstee plots were not parabolic. Unlike the E. coli and P. falciparum enzymes, human GMPS shows co-operativity with respect to XMP. The kcat of PfGMPS is 10-fold lower than that of the human enzyme. This may be a step towards the regulation of the synthesis of adenine and guanine nucleotides, as the P. falciparum genome has a very high proportion of AT compared with GC nucleotides. A possible support for this comes from the recent studies on P. falciparum CTP synthetase, which has been shown to have a very low specific activity of 76 nmol·min−1·mg−1 [39], suggesting lower activity with respect to CTP synthesis.

Table 1
Kinetic parameters of P. falciparum, human and E. coli GMPSs
 XMP ATP Glutamine (NH4)2SO4 
P. falciparum    
Km (μM) 16.8±2 260±38 472±69 5.4±0.8† 
kcat (s−10.43‡ − −  
kcat/Km (μM−1·s−12.5×10−2 0.16×10−2 0.09×10−2  
Human§     
Km (μM) 35.6±1.8 132±7 406±49  
 45.4±5.3 180±12 358±34  
kcat (s−15.4 − −  
 5.6 − −  
E. coli∥ (μM) 29 530 1000 1.0† 
 XMP ATP Glutamine (NH4)2SO4 
P. falciparum    
Km (μM) 16.8±2 260±38 472±69 5.4±0.8† 
kcat (s−10.43‡ − −  
kcat/Km (μM−1·s−12.5×10−2 0.16×10−2 0.09×10−2  
Human§     
Km (μM) 35.6±1.8 132±7 406±49  
 45.4±5.3 180±12 358±34  
kcat (s−15.4 − −  
 5.6 − −  
E. coli∥ (μM) 29 530 1000 1.0† 
*

The present study.

The Km of (NH4)2+ is expressed in millimolar.

The kcat values with respect to the three substrates were similar. The value is the average of the three kcat values.

§

Two values correspond to the two isoforms of human GMPS [17].

Sakamoto et al. [19].

Kinetic mechanism of PfGMPS

PfGMPS binds three substrates ATP, XMP and glutamine, catalyses the amination of XMP to GMP and releases the four products AMP, GMP, PPi and glutamate. The reaction is completed in a concerted manner in two separate domains. The steady-state kinetic mechanism of PfGMPS was elucidated from patterns obtained from the initial velocity and product inhibition plots that are diagnostic of a specific kinetic mechanism.

ATPPase domain

This domain binds the substrates ATP and XMP in the presence of Mg2+ to form the activated AMP-XMP intermediate. To determine the order of substrate binding to the ATPPase domain, glutamine was held at saturating concentration, while XMP or ATP was varied at different fixed concentrations of the other. Plots of both 1/v against 1/[ATP] and 1/v against 1/[XMP] (Figures 1B and 1A) yielded intersecting line patterns with lines converging to the left of the y-axis. The point of intersection was on the x-axis in the former plot and above the x-axis in the latter plot. According to the rules of Cleland [33], this pattern rules out the Ping Pong mechanism and is indicative of either ordered or random sequential mechanism for the binding of ATP and XMP to this domain. The data did not fit the rapid equilibrium ordered mechanism but best fitted eqn (2) that represents both steady-state ordered and rapid-equilibrium random bisubstrate models. The kinetic constants obtained from this fit (Table 2) were in the same range as those obtained from saturation kinetic studies and, hence, validate the equation used.

Initial velocity (v) patterns for PfGMPS reaction

Figure 1
Initial velocity (v) patterns for PfGMPS reaction

Double reciprocal plots represent the initial velocity patterns with respect to XMP, ATP, glutamine and NH4+. In each experiment, one substrate was varied at regular intervals keeping the other at different fixed concentrations, while the third substrate was kept at a constant saturating concentration. Assay conditions were as described in the Materials and methods section. In (A, C, E), ATP is the fixed variable substrate with XMP, glutamine and NH4+ as the varied substrates respectively. The concentrations of ATP are indicated against each line. In (B, D, F), XMP is the fixed variable substrate with ATP, glutamine and ammonia as the varied substrates respectively. The concentration of XMP is indicated against each line. The concentration of ATP, XMP and Q when maintained at saturating concentrations were: 2, 0.15 and 5 mM respectively. The units for v are μmol·min−1·mg−1. All the experiments were performed three times and the reproducibility was within 15%. The results of one representative experiment are presented here.

Figure 1
Initial velocity (v) patterns for PfGMPS reaction

Double reciprocal plots represent the initial velocity patterns with respect to XMP, ATP, glutamine and NH4+. In each experiment, one substrate was varied at regular intervals keeping the other at different fixed concentrations, while the third substrate was kept at a constant saturating concentration. Assay conditions were as described in the Materials and methods section. In (A, C, E), ATP is the fixed variable substrate with XMP, glutamine and NH4+ as the varied substrates respectively. The concentrations of ATP are indicated against each line. In (B, D, F), XMP is the fixed variable substrate with ATP, glutamine and ammonia as the varied substrates respectively. The concentration of XMP is indicated against each line. The concentration of ATP, XMP and Q when maintained at saturating concentrations were: 2, 0.15 and 5 mM respectively. The units for v are μmol·min−1·mg−1. All the experiments were performed three times and the reproducibility was within 15%. The results of one representative experiment are presented here.

Table 2
Kinetic parameters and line patterns obtained from initial velocity studies on PfGMPS

I, intersecting; P, parallel.

  Apparent Michaelis constant 
Varied substrate* Fixed substrate† MgATP (μM) XMP (μM) Gln (μM) NH4+ (mM) Pattern 
ATP against XMP Gln 212±31 10.0±1.9   
XMP against ATP Gln 167±42 16.9±3.1   
Gln against XMP ATP  26.8±2.4 570±77  
Gln against ATP XMP 196±31  589±83  
NH4+ against XMP ATP  7.2±1.4  8.0±1.1 
NH4+ against ATP XMP 142±35   7.0±1.8 
  Apparent Michaelis constant 
Varied substrate* Fixed substrate† MgATP (μM) XMP (μM) Gln (μM) NH4+ (mM) Pattern 
ATP against XMP Gln 212±31 10.0±1.9   
XMP against ATP Gln 167±42 16.9±3.1   
Gln against XMP ATP  26.8±2.4 570±77  
Gln against ATP XMP 196±31  589±83  
NH4+ against XMP ATP  7.2±1.4  8.0±1.1 
NH4+ against ATP XMP 142±35   7.0±1.8 
*

In all of the pairs of varied substrates, the first one was varied continuously and the second at different fixed concentrations.

 All of the fixed substrates were kept at saturating concentrations as mentioned in the Materials and methods section.

To differentiate between the two mechanistic possibilities (steady-state ordered compared with rapid-equilibrium random), product inhibition studies using AMP, GMP and PPi were carried out. p[NH]ppA was also used to validate the results obtained with PPi. The results are summarized in Table 3 and Figure 2. AMP was found to be a weak inhibitor of PfGMPS activity, while PPi, GMP and p[NH]ppA were potent inhibitors. PPi showed competitive and non-competitive inhibition with respect to ATP and XMP respectively (Figures 2A and 2B), while GMP showed competitive and uncompetitive inhibition with respect to XMP and ATP respectively (Figures 2E and 2D). p[NH]ppA yielded patterns similar to those obtained with PPi, being competitive and non-competitive inhibitor with respect to ATP and XMP respectively (Supplementary Figures 3A and 3B at http://www.BiochemJ.org/bj/409/bj4090263add.htm). The slope and intercept replots from the primary non-competitive inhibition plots of PPi and p[NH]ppA yielded Kii and Kis values that were similar (results not shown). These results indicate that ATP can bind to the free enzyme, while XMP binding is conditional to the binding of ATP. Taken together, the initial velocity and product inhibition plots support steady-state ordered binding of the substrates ATP and XMP to the ATPPase domain, with ATP being the first substrate to bind followed by XMP.

Table 3
Apparent inhibition constants for inhibition of PfGMPS activity by products and substrate analogue

The invariant substrates were held at saturating concentration.

Variable substrate Inhibitor Ki (μM) Pattern 
XMP GMP 38.0±6.7 Competitive 
 PPi  443±46 Non-competitive 
 p[NH]ppA  611±69 Non-competitive 
ATP GMP  179±16 Uncompetitive 
 PPi 36.4±7.0 Competitive 
 p[NH]ppA  210±42 Competitive 
Glutamine GMP  390±79 Non-competitive 
 PPi  933±136 Non-competitive 
Variable substrate Inhibitor Ki (μM) Pattern 
XMP GMP 38.0±6.7 Competitive 
 PPi  443±46 Non-competitive 
 p[NH]ppA  611±69 Non-competitive 
ATP GMP  179±16 Uncompetitive 
 PPi 36.4±7.0 Competitive 
 p[NH]ppA  210±42 Competitive 
Glutamine GMP  390±79 Non-competitive 
 PPi  933±136 Non-competitive 

Product inhibition patterns for PfGMPS reaction

Figure 2
Product inhibition patterns for PfGMPS reaction

Double reciprocal plots show the effects of PPi and GMP as inhibitors with respect to the three substrates. In each plot, the concentration of one substrate was varied at different fixed concentrations of the inhibitor (indicated against each line), keeping the remaining two substrates at saturating concentration. Assays were performed under standard conditions described in the Materials and methods section. When saturating, the concentrations of ATP, XMP and glutamine were fixed at 2, 0.15 and 5 mM respectively. In (AC), PPi is the inhibitor with ATP, XMP and glutamine as the variable substrates respectively. In (DF), GMP is the inhibitor with ATP, XMP and glutamine as the variable substrates respectively. In (AF), the units for v were μmol·min−1·mg−1. Each experiment was performed at least twice and reproducibility was within 10%. Shown are results from one experiment.

Figure 2
Product inhibition patterns for PfGMPS reaction

Double reciprocal plots show the effects of PPi and GMP as inhibitors with respect to the three substrates. In each plot, the concentration of one substrate was varied at different fixed concentrations of the inhibitor (indicated against each line), keeping the remaining two substrates at saturating concentration. Assays were performed under standard conditions described in the Materials and methods section. When saturating, the concentrations of ATP, XMP and glutamine were fixed at 2, 0.15 and 5 mM respectively. In (AC), PPi is the inhibitor with ATP, XMP and glutamine as the variable substrates respectively. In (DF), GMP is the inhibitor with ATP, XMP and glutamine as the variable substrates respectively. In (AF), the units for v were μmol·min−1·mg−1. Each experiment was performed at least twice and reproducibility was within 10%. Shown are results from one experiment.

Glutaminase domain

This domain hydrolyses glutamine to glutamate and NH3, with the latter product channelled to the ATPPase domain. The effect of varying glutamine at different fixed concentrations of ATP with XMP at saturating concentration (Figure 1C) or at different fixed concentrations of XMP with ATP saturating (Figure 1D) resulted in parallel patterns in double reciprocal plots. Similar patterns were obtained when either XMP or ATP was the variable substrate at different fixed concentrations of glutamine, with the invariant substrate at saturating concentration (results not shown). Parallel plots suggest the occurrence of an irreversible step between the binding of XMP or ATP and glutamine and, hence, a Ping Pong mechanism. In the case of E. coli GMPS, the parallel initial velocity pattern in 1/v against 1/Gln plots at different fixed concentrations of ATP has been proposed to indicate that XMP binds between ATP and glutamine [20]. This possibility is ruled out in the case of PfGMPS, as 1/v against 1/Gln plots at different fixed concentrations of ATP remained parallel even at subsaturating concentration of XMP. The pattern in this plot should have been intersecting if XMP bound to the enzyme in between the binding of ATP and glutamine. We also find that both PPi and GMP are non-competitive inhibitors with respect to glutamine (Figures 2C and 2F), suggesting that ATP and XMP bind independently of glutamine. Recently, Abbott et al. [40] have also shown that in E. coli GMPS, removal of the GAT domain does not affect the binding of XMP and ATP to the ATPPase domain. Initial velocity measurements where glutamine was replaced with ammonia in the reaction also indicated that in PfGMPS, binding of XMP is not between the binding of ATP and ammonia, as 1/v against 1/NH4+ plots with ATP or XMP as fixed variable were intersecting (Figures 1E and 1F). It is interesting to note that in the case of E. coli GMPS, plots of 1/v against 1/NH4+ at different fixed concentrations of ATP resulted in parallel lines, while 1/v against 1/NH4+ plots at different fixed concentrations of XMP were intersecting.

The above results suggested that the parallel pattern in all the glutamine plots (1/v against 1/Gln) could be due to the presence of an irreversible step involving the release of products AMP, PPi or glutamate, before the attack of the adenyl-XMP intermediate by ammonia, leading to the formation of GMP. The parallel pattern arising from the release of AMP is ruled out, as adenyl-XMP intermediate would have to remain bound to the enzyme until attack by ammonia at the C-2 of XMP. The results are also not indicative of release of PPi because a competitive inhibition pattern was seen when PPi was varied with respect to ATP, and plots of 1/v against 1/NH4+ at different fixed ATP or XMP concentrations were intersecting, supporting the absence of PPi release before the binding of NH3 (Figures 1E and 1F). This leaves only the possibility of glutamate release before the reaction of ammonia with adenyl-XMP intermediate. Glutamate and its analogue, γ-glutamyl methyl ester did not inhibit PfGMPS activity (results not shown), further suggesting that the enzyme cannot retain glutamate once it is formed. Coupled enzyme assays using glutamate dehydrogenase indicated glutamate release even in the absence of substrates (Table 4), suggesting that glutamate release could occur before GMP formation, thereby giving rise to a parallel pattern in double reciprocal plots. This parallel pattern is akin to the two-site Ping Pong mechanism [41,42] where the coupling of two reactions leads to a parallel pattern in double reciprocal plots (when a substrate from one site is varied with respect to a substrate from the other site) due to the release of a product from one site. Our results show that the parallel initial velocity plots with PfGMPS are due to the release of glutamate before the attack of adenyl-XMP intermediate by ammonia. However, these studies do not indicate whether glutamate release is conditional for the movement of ammonia to the ATPPase domain.

Table 4
Effect of ligand binding (ATP, XMP and p[NH]ppA) to the ATPPase domain on glutaminase activity in the GAT domain

The experiment was performed twice with each assay in duplicate. The values from the two experiments agreed within 10%. Results are means for the duplicates from one experiment. The amount of glutamate formed was estimated using glutamate dehydrogenase and the protocol followed was as described in the Materials and methods section. The amount of glutamate produced from glutamine in the presence of ATP, XMP and Mg2+ was taken as 100%.

Substrates Activity (%) 
ATP, XMP, Gln, Mg2+ 100 
ATP, Gln, Mg2+ 58 
XMP, Gln, Mg2+ 57 
Gln, Mg2+ 56 
Gln 59 
p[NH]ppA, XMP, Gln, Mg2+ 77 
Substrates Activity (%) 
ATP, XMP, Gln, Mg2+ 100 
ATP, Gln, Mg2+ 58 
XMP, Gln, Mg2+ 57 
Gln, Mg2+ 56 
Gln 59 
p[NH]ppA, XMP, Gln, Mg2+ 77 

The kinetic scheme for PfGMPS (1) deduced from the above results involves ordered binding of ATP and XMP, with ATP binding first followed by XMP. While glutamine binding is independent of the binding of ATP and XMP, the glutaminase assays suggest a level of co-ordination of activity between the two domains. Our kinetic scheme also shows the order of release of products, where glutamate is the first product to get released, followed by AMP that is a weak inhibitor of PfGMPS. 1 also suggests that the release of GMP is before PPi, both of which are competitive inhibitors of their respective substrates, as the value of Km/Ki for XMP/GMP (0.56) is lower than that for ATP/PPi (4.9), suggesting the enzyme's higher affinity for PPi. This supports the suggestion by von der Saal et al. [20] that PPi is the last product to be released. PPi and GMP have also been shown to be competitive inhibitors of ATP and XMP respectively for GMPS from other sources [17,43,44].

Kinetic scheme for PfGMPS

Mg2+ requirement for PfGMPS activity

Figure 3(A) shows the dependence of PfGMPS activity on Mg2+ ions. The data best fitted the equation for positive co-operativity, with a Hill coefficient (h) of 4.4 and half-maximal activity at 2.09±0.03 mM of Mg2+. This suggested the presence of multiple binding sites for Mg2+ on PfGMPS. Maximum activity was reached at 5.5 mM of Mg2+, which is 2.7-fold the concentration of ATP present in the reaction, suggesting that apart from Mg2+ being complexed with ATP, an additional binding site necessary for activity is present on each subunit of PfGMPS. To further confirm this observation, ATP was varied with Mg2+ at a fixed concentration of 2 mM and the resultant plot is shown in Figure 3(B). It is evident from the Figure that maximum activity was not achieved at 2 mM of MgATP, but at a lower concentration of 1.2 mM. When free ATP concentration exceeded 2 mM, activity decreased as a consequence of decrease in the concentration of free Mg2+. This suggests that either free ATP competes for the MgATP-binding site or that the enzyme has an additional requirement for free magnesium. To confirm whether ATP is a competitive inhibitor of MgATP, Mg2+ concentration was varied at different fixed concentrations of MgATP, keeping XMP at saturating concentration. The plot of 1/v against 1/[Mg2+] showed an intersecting pattern, which rules out the possibility of ATP being a competitive inhibitor of MgATP, as the expected plot in that case would be parallel [17]. The decrease in activity on increase in ATP concentration above that of Mg2+ is thus a consequence of the fact that PfGMPS binds free Mg2+ apart from MgATP (Figure 3B). Co-operative binding of Mg2+ has also been reported in the case of human GMPS [17]. Requirement of Mg2+ in excess of ATP concentration has also been observed in GMPS from E. coli [20] and human [17]. Although kinetic evidence for multiple Mg2+-binding sites is available, the specific role of the different Mg2+ ions has not been elucidated.

Dependence of PfGMPS activity on Mg2+ concentration

Figure 3
Dependence of PfGMPS activity on Mg2+ concentration

All the assays were performed using standard conditions described in the Materials and methods section with EDTA omitted from all reactions. (A) Activity of PfGMPS with respect to Mg2+, at a fixed concentration of ATP (2 mM). The data best fitted the positive co-operativity equation (v=Vmax [S]h/[S]h+Kh0.5, where h is the Hill coefficient and K0.5 is the substrate concentration at which half of the sites are occupied). (B) Effect of varying ATP concentration on PfGMPS activity at fixed concentration of Mg2+ (2 mM). ●, activity as the formation of GMP in nmol·min−1; ■, [MgATP2−]; ♦, free [ATP]; ▲, free [Mg2+]. All experiments were repeated three times to confirm reproducibility. Shown are results from one representative experiment.

Figure 3
Dependence of PfGMPS activity on Mg2+ concentration

All the assays were performed using standard conditions described in the Materials and methods section with EDTA omitted from all reactions. (A) Activity of PfGMPS with respect to Mg2+, at a fixed concentration of ATP (2 mM). The data best fitted the positive co-operativity equation (v=Vmax [S]h/[S]h+Kh0.5, where h is the Hill coefficient and K0.5 is the substrate concentration at which half of the sites are occupied). (B) Effect of varying ATP concentration on PfGMPS activity at fixed concentration of Mg2+ (2 mM). ●, activity as the formation of GMP in nmol·min−1; ■, [MgATP2−]; ♦, free [ATP]; ▲, free [Mg2+]. All experiments were repeated three times to confirm reproducibility. Shown are results from one representative experiment.

Interdomain cross-talk

GMPS has two domains that catalyse two different reactions with ammonia generated by the GAT domain being channelled to the ATPPase domain to form GMP. Glutamine hydrolysis in the GAT domain, independent of the binding of XMP and ATP to the ATPPase domain, would be a physiologically wasteful process and, hence, coupling of GAT activity with the activity of ATPPase domain would be expected. The cross-talk occurring between the two domains in PfGMPS was examined by two methods: first, by measuring the glutaminase activity of the GAT domain and, secondly, by inactivation of the GAT domain by GAT-specific inhibitors. Table 4 summarizes the effect of substrate binding to the ATPPase domain on glutaminase activity. It is evident from the data that in PfGMPS there is a high basal level of glutaminase activity that is not altered by the binding of XMP or ATP alone to the ATPPase domain but once the ATPPase domain is fully liganded, GAT activity reaches a maximum. Replacement of ATP by p[NH]ppA led to significant increase in glutaminase activity that was lower when compared with the complete reaction having both ATP and XMP. This shows that although the adenyl-XMP intermediate is not formed, complete occupancy of the catalytic pocket in the ATPPase domain is sufficient to stimulate GAT activity. This also suggests that the formation of the reaction intermediate in ATPPase domain signals the GAT domain for complete activity, leading to coupling of the two reactions in the two domains. The level of leaky glutaminase activity is significantly higher than that for other GATs. Under similar assay conditions, human GMPS [18], imidazole glycerol phosphate synthase [45] and NAD synthetase [46] exhibited tight regulation with very low background glutaminase activity in the absence of complete liganding of the acceptor domain. However, asparagine synthetase has been shown to have leaky glutaminase activity [41,47]. It is interesting to note that while, in the case of PfGMPS, liganding of ATPPase domain with p[NH]ppA, XMP and Mg2+ led to activation of the glutaminase domain, these ligands had no effect on the activity of the GAT domain in human GMPS [17]. The in vivo significance of the partial/weak domain regulation in PfGMPS is unclear at the present stage. The presence of other modulators under in vivo conditions that prevent the leaky glutaminase activity cannot be ruled out.

DON and acivicin, two antibiotics from Streptomyces, are known irreversible inhibitors of E. coli [23,48] and human GMPSs [18]. These inhibitors covalently modify the catalytic cysteine residue in the GAT domain, thereby inactivating the enzyme. Measurement of inactivation rates as a function of substrate binding to the ATPPase domain permits evaluation of the extent of cross-talk between the domains. PfGMPS is inactivated by both DON and acivicin and the rate of inactivation is enhanced by the presence of substrates in the ATPPase domain. In the absence of substrates, 30–50% loss of activity was seen at 25 °C over 30 min; however, nearly complete inactivation was seen in 10 min when the substrates were present. In order to evaluate the effect of substrate binding on the efficiency of inactivation, the pre-incubation temperature was lowered to 15 °C to slow down the rate of inactivation. Plots of ln % activity against time show a biphasic trend indicating the existence of two steps in the inactivation kinetics by acivicin (Figures 4A and 4C) and DON (Figures 4B and 4D). These plots indicate the occurrence of an initial phase with rapid loss of activity followed by a second phase with slower inhibition rates. The secondary plot of the slope (kapp) of the lines in Figure 4 against [I] was linear (results not shown), indicating that saturation was not achieved at the concentrations of inhibitor used. Hence, eqn (7), ln(ϵ/E0)=k3t[I]/([I]+KI), which describes the inactivation kinetics, was simplified to ln(ϵ)/E0=k3t*I/KI to obtain an estimate of the second-order rate constant (k3/KI) from the plots in Figure 4. This value for acivicin in the absence of substrates (0.18±0.01 min−1·mM−1) increased by 8-fold to 1.49±0.34 min−1·mM−1 in the presence of the substrates.

Inactivation kinetics of PfGMPS by acivicin and DON

Figure 4
Inactivation kinetics of PfGMPS by acivicin and DON

The inactivation is represented as ln % residual activity as a function of time. (A, B) The plots in the absence of substrates with acivicin and DON respectively at 15 °C. (C, D) Inactivation with acivicin and DON in the presence of substrates at 15 °C. The concentration of inhibitors used in all these plots is presented in the inset of each plot. The concentration of substrates used in pre-incubation reaction was that of standard assay. The reproducibility was checked by repeating the experiments three times with separate batches of enzymes.

Figure 4
Inactivation kinetics of PfGMPS by acivicin and DON

The inactivation is represented as ln % residual activity as a function of time. (A, B) The plots in the absence of substrates with acivicin and DON respectively at 15 °C. (C, D) Inactivation with acivicin and DON in the presence of substrates at 15 °C. The concentration of inhibitors used in all these plots is presented in the inset of each plot. The concentration of substrates used in pre-incubation reaction was that of standard assay. The reproducibility was checked by repeating the experiments three times with separate batches of enzymes.

In the case of DON, the difference is 5-fold with the values 0.19±0.01 and 0.95±0.08 min−1·mM−1 obtained in the presence and absence of substrates respectively. Such a biphasic behaviour in the inactivation kinetics has been observed in the case of E. coli GMPS [23]. Unlike PfGMPS, acivicin was found to be a more potent inhibitor than DON in the case of E. coli GMPS. Zyk et al. [49] have suggested that binding of XMP is sufficient to bring about conformational changes in E. coli GMPS. However, our results do not agree with this finding as the presence of XMP alone was insufficient for activation of GAT domain.

Inhibition by nucleosides, nucleotides and purine bases

Inhibition studies on PfGMPS using various analogues of nucleosides and nucleotides permitted differentiation of the ligand binding specificities of the human and parasite enzymes. PfGMPS was inhibited by GMP (the product of the reaction) with a low Ki value of 38.0±6.7 μM, while IMP and AMP were weak inhibitors (Table 5). Our studies show that guanine derivatives, 8-azaguanine, 6-thioguanine, guanosine, GDP and GTP, all inhibit PfGMPS moderately. Adenosine was also found to inhibit PfGMPS, while xanthosine and inosine failed to show any effect. Psicofuranine (9-β-D-psicofuranosyl-adenine), an analogue of adenosine, exhibited a weak inhibitory effect, while chloroadenosine and decoyinine [9-β-D-(5, 6-psicofuranose-enyl)-adenine] had no effect on PfGMPS activity. McConkey [7] has shown the inhibition of growth of intra-erythrocytic parasites in culture by psicofuranine, albeit at a high concentration of inhibitor (IC50=300 μM). Decoyinine and psicofuranine are specific inhibitors of GMPS from other sources with IC50 values for the human enzyme being 46.5 and 17.3 μM respectively. Also, xanthosine and chloroadenosine have been shown to inhibit GMPS from Ehrlich ascites cells [44]. This indicates that the P. falciparum enzyme has binding attributes different from other GMPSs, highlighting the possibility of designing species-specific inhibitors.

Table 5
Inhibition of PfGMPS by nucleosides, nucleotides and their derivatives

Hypoxanthine, chloroadenosine, xanthosine, inosine and decoyinine did not inhibit PfGMPS at a concentration of 0.5 mM. Results are means of two measurements that agreed within 5%.

Inhibitor (0.5 mM) Inhibition (%) 
Guanosine 50 
GDP 42 
8-Azaguanine 40 
GTP 40 
Adenosine 37 
6-Thioguanine 27 
Psicofuranine 25 
AMP 22 
IMP 10 
Inhibitor (0.5 mM) Inhibition (%) 
Guanosine 50 
GDP 42 
8-Azaguanine 40 
GTP 40 
Adenosine 37 
6-Thioguanine 27 
Psicofuranine 25 
AMP 22 
IMP 10 

Conclusions

In summary, the present study has given a detailed insight into the kinetic mechanism operating in PfGMPS and contributes to further understanding of this class of enzymes. Our results show the occurrence of two-site Ping Pong mechanism in which the two individual reactions in the ATPPase and GAT domains are coupled to carry out the synthesis of GMP in a concerted manner. PfGMPS sequence exhibits 42% identity with E. coli GMPS and only 20% with its human counterpart and the results presented in this paper also indicate closer biochemical similarity to the bacterial enzyme. The differences in inhibition by nucleosides and analogues between the P. falciparum and human enzyme highlight the utility of the parasite enzyme as a potential drug target. PfGMPS has a unique insertion of 20 amino acid residues in the GAT domain which is absent even in GMPSs from other species of Plasmodium. The role of this insertion is yet to be deduced.

This work was financially supported by CSIR (Council of Scientific and Industrial Research), CSIR-NMITLI (New Millennium India Technology Leadership Initiative), Government of India. J. Y. B. is a recipient of a senior research fellowship from CSIR, Government of India.

Abbreviations

     
  • ATPPase

    ATP pyrophosphatase

  •  
  • DON

    6-diazo-5-oxo-L-norleucine

  •  
  • DTT

    dithiothreitol

  •  
  • GAT

    glutamine amidotransferase

  •  
  • GMPS

    GMP synthetase

  •  
  • HGPRT

    hypoxanthine guanine phosphoribosyltransferase

  •  
  • IMPDH

    inosine monophosphate dehydrogenase

  •  
  • MALDI

    matrix-assisted laser-desorption ionization

  •  
  • Ni-NTA

    Ni2+-nitrilotriacetate

  •  
  • PfGMPS

    Plasmodium falciparum GMPS

  •  
  • p[NH]ppA

    adenosine 5′-[β,γ-imido]triphosphate

References

References
1
Roll Back Malaria, World Health Organization, UNICEF
World Malaria Report 2005
2005
Geneva
World Health Organization
 
2
Ullman
B.
Carter
D.
Hypoxanthine-guanine phosphoribosyltransferase as a therapeutic target in protozoal infections
Infect. Agents Dis.
1995
, vol. 
4
 (pg. 
29
-
40
)
3
Gardner
M. J.
Hall
N.
Fung
E.
White
O.
Berriman
M.
Hyman
R. W.
Carlton
J. M.
Pain
A.
Nelson
K. E.
Bowman
S.
, et al. 
Genome sequence of the human malaria parasite Plasmodium falciparum
Nature
2002
, vol. 
419
 (pg. 
498
-
511
)
4
Traut
T. W.
Physiological concentrations of purines and pyrimidines
Mol. Cell. Biochem.
1994
, vol. 
140
 (pg. 
1
-
22
)
5
Webster
H. K.
Whaun
J. M.
Antimalarial properties of bredinin: prediction based on identification of differences in human host–parasite purine metabolism
J. Clin. Invest.
1982
, vol. 
70
 (pg. 
461
-
469
)
6
Susan
A. Q.
Jagt
D. L.
Reyes
P.
In vitro susceptibilities of Plasmodium falciparum to compounds which inhibit nucleotide metabolism
Antimicrob. Agents Chemother.
1990
, vol. 
34
 (pg. 
1393
-
1398
)
7
McConkey
G. A.
Plasmodium falciparum: isolation and characterisation of a gene encoding protozoan GMP synthase
Exp. Parasitol.
2000
, vol. 
94
 (pg. 
23
-
32
)
8
Zalkin
H.
The amidotransferases
Adv. Enzymol. Relat. Areas Mol. Biol.
1993
, vol. 
66
 (pg. 
203
-
309
)
9
Raushel
F. M.
Thoden
J. B.
Holden
H. M.
Enzymes with molecular tunnels
Acc. Chem. Res.
2003
, vol. 
36
 (pg. 
539
-
548
)
10
Moyed
H. S.
Magasanik
B.
Enzymes essential for the biosynthesis of nucleic acid guanine; xanthosine 5′-phosphate aminase of Aerobacter aerogenes
J. Biol. Chem.
1957
, vol. 
226
 (pg. 
351
-
363
)
11
Lagerkvist
U.
Biosynthesis of guanosine 5′-phosphate. II. Amination of xanthosine 5′-phosphate by purified enzyme from pigeon liver
J. Biol. Chem.
1958
, vol. 
233
 (pg. 
143
-
149
)
12
Abrams
R.
Bentley
M.
Guanosine 5′-phosphate formation from xanthosine 5′-phosphate and L-glutamine
Arch. Biophys. Biochem.
1959
, vol. 
79
 (pg. 
91
-
110
)
13
Fukuyama
T. T.
Formation of an adenyl xanthosine monophosphate intermediate by xanthosine 5′-phosphate aminase and its inhibition by psicofuranine
J. Biol. Chem.
1966
, vol. 
241
 (pg. 
4745
-
4749
)
14
Sweeny
M. J.
Mycophenolic acid and its mechanism of action in cancer and psoriasis
Jpn. J. Antibiot.
1977
, vol. 
30
 (pg. 
85
-
92
)
15
Ishikawa
H.
Mizoribine and mycophenolate mofetil
Curr. Med. Chem.
1999
, vol. 
6
 (pg. 
575
-
597
)
16
Lou
L.
Nakamura
J.
Tsing
S.
Nguyen
B.
Chow
J.
Straub
K.
Chan
H.
Barnett
J.
High-level production from a baculovirus expression system and biochemical characterization of human GMP synthetase
Protein Expression Purif.
1995
, vol. 
6
 (pg. 
487
-
495
)
17
Nakamura
J.
Lou
L.
Biochemical characterization of human GMP synthetase
J. Biol. Chem.
1995
, vol. 
270
 (pg. 
7347
-
7353
)
18
Nakamura
J.
Straub
K.
Wu
J.
Lou
L.
The glutamine hydrolysis function of human GMP synthetase: identification of an essential active site cysteine
J. Biol. Chem.
1995
, vol. 
270
 (pg. 
23450
-
23455
)
19
Sakamoto
N.
Hatfield
G. W.
Moyed
H. S.
Physical properties and subunit structure of xanthosine 5′-phosphate aminase
J. Biol. Chem.
1972
, vol. 
247
 (pg. 
5880
-
5887
)
20
von der Saal
W.
Crysler
C. S.
Villafranca
J. J.
Positional isotope exchange and kinetic experiments with Escherichia coli guanosine-5′-monophosphate synthetase
Biochemistry
1985
, vol. 
24
 (pg. 
5343
-
5350
)
21
Patel
N.
Moyed
H. S.
Kane
J. F.
Xanthosine-5′-phosphate amidotransferase from Escherichia coli
J. Biol. Chem.
1975
, vol. 
250
 (pg. 
2609
-
2613
)
22
Deras
M. L.
Chittur
S. V.
Davisson
V. J.
N2-hydroxyguanosine 5′-monophosphate is a time-dependent inhibitor of Escherichia coli guanosine monophosphate synthetase
Biochemistry
1999
, vol. 
38
 (pg. 
303
-
310
)
23
Chittur
S. V.
Klem
T. J.
Shafer
C. M.
Davisson
V. J.
Mechanism for acivicin inactivation of triad glutamine amidotransferases
Biochemistry
2001
, vol. 
40
 (pg. 
876
-
887
)
24
Hirai
K.
Matsuda
Y
Nakagawa
H.
Purification and characterization of GMP synthetase from Yoshida sarcoma ascites cells
J. Biochem. (Tokyo)
1987
, vol. 
102
 (pg. 
893
-
902
)
25
Spector
T.
Studies with GMP synthetase from Ehrlich ascites cells: purification, properties, and interactions with nucleotide analogs
J. Biol. Chem.
1975
, vol. 
250
 (pg. 
7372
-
7376
)
26
Tesmer
J. J.
Stemmler
T. L.
Penner-Hahn
J. E.
Davisson
V. J.
Smith
J. L.
Preliminary X-ray analysis of Escherichia coli GMP synthetase: determination of anomalous scattering factors for a cysteinyl mercury derivative
Proteins
1994
, vol. 
18
 (pg. 
394
-
403
)
27
Tesmer
J. J.
Klem
T. J.
Deras
M. L.
Davisson
V. J.
Smith
J.
The crystal structure of GMP synthetase reveals a novel catalytic triad and is a structural paradigm for two enzyme families
Nat. Struct. Biol.
1996
, vol. 
3
 (pg. 
74
-
86
)
28
Bahl
A.
Brunk
B.
Crabtree
J.
Fraunholz
M. J.
Gajria
B.
Grant
G. R.
Ginsburg
H.
Gupta
D.
Kissinger
J. C.
Labo
P.
, et al. 
PlasmoDB: the Plasmodium genome resource: a database integrating experimental and computational data
Nucleic Acids Res.
2003
, vol. 
31
 (pg. 
212
-
215
)
29
Hirst
M.
Haliday
E.
Nakamura
J.
Lou
L.
Human GMP synthetase: protein purification, cloning, and functional expression of cDNA
J. Biol. Chem.
1994
, vol. 
269
 (pg. 
23830
-
23837
)
30
Bradford
M. M.
A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding
Anal. Biochem.
1976
, vol. 
72
 (pg. 
248
-
254
)
31
Patel
N.
Moyed
H. S.
Kane
J. F.
Properties of xanthosine 5′-monophosphate-amidotransferase from Escherichia coli
Arch. Biochem. Biophys.
1977
, vol. 
178
 (pg. 
652
-
661
)
32
Frieden
C.
Glutamic dehydrogenase. III. The order of substrate addition in the enzymatic reaction
J. Biol. Chem.
1959
, vol. 
234
 (pg. 
2891
-
2896
)
33
Cleland
W. W.
The kinetics of enzyme-catalyzed reactions with two or more substrates or products: I. nomenclature and rate equations
Biochim. Biophys. Acta
1963
, vol. 
67
 (pg. 
104
-
137
)
34
Kitz
R.
Wilson
I. B.
Esters of methanesulfonic acid as irreversible inhibitors of acetylcholinesterase
J. Biol. Chem.
1962
, vol. 
237
 (pg. 
3245
-
3249
)
35
Morrison
J. F.
Approaches to kinetic studies on metal activated enzymes
Methods Enzymol.
1979
, vol. 
63
 (pg. 
257
-
294
)
36
Robertson
J. G.
Villafranca
J. J.
Characterization of metal ion activation and inhibition of CTP synthetase
Biochemistry
1993
, vol. 
32
 (pg. 
3769
-
3777
)
37
Baca
A. M.
Hol
W. G. J.
Overcoming codon bias: a method for high-level overexpression of Plasmodium and other AT-rich parasite genes in Escherichia coli
Int. J. Parasitol.
2000
, vol. 
30
 (pg. 
113
-
118
)
38
Gottesman
S.
Halpern
E.
Trisler
P.
Role of sulA and sulB in filamentation by lon mutants of Escherichia coli K-12
J. Bacteriol.
1981
, vol. 
48
 (pg. 
265
-
273
)
39
Yuan
P.
Hendriks
E. F.
Fernandez
H. R.
O'Sullivan
W. J.
Stewart
T. S.
Functional expression of the gene encoding cytidine triphosphate synthetase from Plasmodium falciparum which contains two novel sequences that are potential antimalarial targets
Mol. Biochem. Parasitol.
2005
, vol. 
143
 (pg. 
200
-
208
)
40
Abbot
J. L.
Newell
J. M.
Lightcap
C. M.
Olanich
M. E.
Loughlin
D. T.
Weller
M. A.
Lam
G.
Pollack
S.
Patton
W. A.
The effects of removing the GAT domain from E. coli GMP synthetase
Protein J.
2006
, vol. 
25
 (pg. 
483
-
491
)
41
Fresquet
V.
Thoden
J. B.
Holden
H. M.
Raushel
F. M.
Kinetic mechanism of asparagine synthetase from Vibrio cholerae
Bioorg. Chem.
2004
, vol. 
32
 (pg. 
63
-
75
)
42
Barden
R. E.
Fung
C. H.
Utter
M. F.
Scrutton
M. C.
Pyruvate carboxylase from chicken liver: steady state kinetic studies indicate a ‘two-site’ ping-pong mechanism
J. Biol. Chem.
1972
, vol. 
247
 (pg. 
1323
-
1233
)
43
Spector
T.
Beacham
L. M.
Guanosine monophosphate synthetase from Escherichia coli B-96: inhibition by nucleosides
J. Biol. Chem.
1975
, vol. 
250
 (pg. 
3101
-
3107
)
44
Spector
T. J.
Jones
T. E.
Krenitsky
T. A.
Harvey
R. J.
Guanosine monophosphate synthetase from Ehrlich ascites cells: multiple inhibition by pyrophosphate and nucleosides
Biochim. Biophys. Acta
1976
, vol. 
452
 (pg. 
597
-
607
)
45
Myers
R. S.
Jensen
J. R.
Deras
I. L.
Smith
J. L.
Davisson
V. J.
Substrate-induced changes in the ammonia channel for imidazole glycerol phosphate synthase
Biochemistry
2003
, vol. 
42
 (pg. 
7013
-
7022
)
46
Wojcik
M.
Seidle
H. F.
Bieganowski
P.
Brenner
C.
Glutamine-dependent NAD+ synthetase: how a two-domain, three-substrate enzyme avoids waste
J. Biol. Chem.
2006
, vol. 
281
 (pg. 
33395
-
33402
)
47
Tesson
A. R.
Soper
T. S.
Ciustea
M.
Richards
N. G.
Revisiting the steady-state kinetic mechanism of glutamine-dependent asparagine synthetase from Escherichia coli
Arch. Biochem. Biophys.
2003
, vol. 
413
 (pg. 
23
-
31
)
48
Zalkin
H.
Truitt
C. D.
Characterization of the glutamine site of Escherichia coli guanosine 5′-monophosphate synthetase
J. Biol. Chem.
1977
, vol. 
252
 (pg. 
5431
-
5436
)
49
Zyk
N.
Citri
N.
Moyed
H. S.
Conformative response of xanthosine 5′-phosphate aminase
Biochemistry
1969
, vol. 
8
 (pg. 
2787
-
2794
)

Supplementary data