- Email: [email protected]
Susceptibility of Plasmodium falciparum to glutamate dehydrogenase inhibitors—A possible new antimalarial target
Molecular & Biochemical Parasitology 172 (2010) 152–155
Contents lists available at ScienceDirect
Molecular & Biochemical Parasitology
Short communication
Susceptibility of Plasmodium falciparum to glutamate dehydrogenase inhibitors—A possible new antimalarial target Isabela M. Aparicio a , Alejandro Marín-Menéndez b , Angus Bell b , Paul C. Engel a,∗ a b
Conway Institute of Biomolecular and Biomedical Research, University College Dublin, Dublin, Ireland Department of Microbiology, School of Genetics & Microbiology, Moyne Institute of Preventive Medicine, Trinity College Dublin, Dublin, Ireland
a r t i c l e
i n f o
Article history: Received 3 February 2010 Received in revised form 31 March 2010 Accepted 8 April 2010 Available online 23 April 2010 Keywords: Plasmodium falciparum Antimalarial Glutamate dehydrogenase Anti-oxidant defences Enzyme inhibition Isophthalic acid
a b s t r a c t With the rapid spread of drug-resistant strains of Plasmodium falciparum, the development of new antimalarials is an urgent need. As malaria parasites live in a highly pro-oxidant environment, their anti-oxidant defences have frequently been suggested as candidate drug targets. A key point in such defences is the production of NADPH e.g. for maintaining anti-oxidant glutathione in the reduced state. Some authors have attributed this function in P. falciparum to a glutamate dehydrogenase, therefore proposed as a potential drug target. Here we show that isophthalic acid inhibits both Plasmodium GDH and bovine GDH but showing marked discrimination (70-fold lower Ki for the parasite GDH). Isophthalic acid impairs intra-erythrocytic growth of P. falciparum in vitro whilst o-phthalic acid, not a GDH inhibitor, shows no effect. This offers hope that with careful design or thorough screening it should be possible to find inhibitors with the necessary selectivity between parasite and human GDHs. © 2010 Elsevier B.V. All rights reserved.
Every year over 1 million deaths are attributed to falciparum malaria [1], and available drugs are gradually becoming ineffective, as new resistant strains emerge. The development of effective new drugs is thus of utmost importance for public health. Spending most of their mammalian cycle within red blood cells, Plasmodium species are subject to dangerous oxidative stress. The parasite digests host haemoglobin, freeing the reactive haem group. In this environment, reactive oxygen species (ROS) are liable to be formed mainly through Fenton-like reactions [2]. The parasite’s anti-oxidant defences include various peroxidases and thiol-dependent reductases [3]. It seems, however, that these mechanisms are barely adequate and are supplemented by a reliance on parallel defences of the host red cell, as evidenced, for example, by the prevalence of glucose 6-phosphate dehydrogenase deficiency in malarial areas. This suggests that the anti-oxidant defences are a suitable target for antimalarial attack. A central ele-
Abbreviations: DMIPA, dimethyl isophthalic acid; GDH, glutamate dehydrogenase; INT, 2-(4-iodophenyl)-3-(4-nitrophenyl)-5-phenyltetrazolium chloride; IPA, isophthalic acid; NAD(H), nicotinamide adenine dinucleotide oxidised/reduced; NADP(H), nicotinamide adenine dinucleotide phosphate oxidised/reduced; o-PA, orthophthalic acid; PES, phenazine ethosulfate; pf, Plasmodium falciparum; RBCs, red blood cells; ROS, reactive oxygen species. ∗ Corresponding author at: UCD School of Biomolecular and Biomedical Sciences, Conway Institute, University College Dublin, Belfield, Dublin 4, Ireland. Tel.: +353 1 716 6764; fax: +353 1 283 7211. E-mail address: [email protected] (P.C. Engel). 0166-6851/$ – see front matter © 2010 Elsevier B.V. All rights reserved. doi:10.1016/j.molbiopara.2010.04.002
ment of these defences is the supply of NADPH, e.g. for the reduction of glutathione, and it has been suggested that in Plasmodium falciparum a glutamate dehydrogenase (GDH) is the main generator of NADPH [4]. This enzyme is abundantly expressed by the parasite, and, since it is absent from the host red blood cell, it is actually a marker for the presence of the parasite [5]. Glutamate dehydrogenases are widely distributed enzymes that catalyse the reversible oxidative deamination of l-glutamate to give 2-oxoglutarate and ammonia, using as a co-factor either NAD(H), NADP(H) or both: l-Glutamate+NAD(P)+ ↔ 2-oxoglutarate + NH4 + +NAD(P)H + H+ P. falciparum NADP+ -GDH has been cloned and expressed in Escherichia coli [6] making possible inhibitor studies. However, since GDH is also important for the host, and has a well-conserved active site, an initial question is whether sufficiently selective inhibition can be achieved. Encouraging initial proof-of-principle experiments [7] showed remarkable selectivity in inhibition of bovine and clostridial GDHs (used as surrogates) by various compounds. Solved structures for parasite [8] and human [9] GDHs offer a valuable framework for inhibitor design. A useful starting point is provided by a wide range of compounds screened by Caughey et al. [10] 30 years before the solution of the first GDH structure [11,12]. The best inhibitors (of bovine GDH) were compounds incorporating an extended glutarate structure (e.g. isophthalic acid) or structurally similar molecules, e.g. bromofuroic acid, where
I.M. Aparicio et al. / Molecular & Biochemical Parasitology 172 (2010) 152–155
153
Fig. 1. Secondary plot of isophthalic acid inhibition for pfGDHa to calculate the inhibition constants. Ki was estimated by plotting Km /Vmax against IPA concentration. For clarity, only lower concentrations of IPA were displayed on the inset. Assays were performed accordingly to the previously described protocol [6], in the presence of increasing concentrations of IPA.
one carboxyl group is replaced by an electronegative halogen atom. In this paper we extend the characterization of the NADPHdependent GDH from P. falciparum (over-expressed in E. coli) and explore the selectivity of inhibition by isophthalate in comparison with bovine GDH, a close homologue of the human enzyme. We have also studied the inhibition of GDH activity in crude parasite extracts and measured IC50 values for isophthalate and related compounds using cultured parasites. Sequencing of the P. falciparum genome [13] has revealed the existence of two other GDH genes. For clarity, the previously described GDH [6,8] is called GDHa and the coding gene gdha. For the screening of potential inhibitors, it was necessary to obtain high yields of pure enzyme, and to improve expression we re-cloned the gdha gene into pET21a (Novagen). This gave an approximately 50fold increase in soluble expression when compared to the pQE60 system (Qiagen) previously used [6]. We also optimized a cheaper and robust two-step method of purification, using dye-affinity and hydrophobic interaction, which resulted in ∼60% recovery of the pfGDHa purified to homogeneity, as confirmed by SDS-PAGE and Western blot (not shown). The kinetic parameters of the recombinant pfGDHa at 25 ◦ C were in good agreement with those previously published [6]. However, in order to work in physiological conditions, we also obtained the kinetic parameters at 37 ◦ C. No significant difference was observed in the apparent Km , when compared to figures for 25 ◦ C [6], but the Vmax and therefore catalytic efficiency of pfGDHa was significantly higher (about 1.6-fold) at 37 ◦ C in comparison to 25 ◦ C. Also pfGDHa shows remarkable stability at higher temperatures, keeping 100% of its activity for at least 24 h at 37 ◦ C (not shown). In face of these observations, the subsequent analysis could be performed at physiological temperature, more consistent with the development of a stable and effective drug. Previous studies of competitive inhibitors from our group [7] have demonstrated differences in affinity constants (Ki ) of up to 10-fold between Clostridium symbiosum GDH (csGDH) and bovine liver GDH (boGDH), used as surrogates for pfGDHa and human liver GDH, with 49 and 97% identity, respectively. Accordingly, we selected isophthalic acid (IPA), which showed the lowest Ki for csGDH (∼30 M against ∼300 M for boGDH), to test with pfGDHa, which shares much higher identity with csGDH than with boGDH. IPA was proposed to compete with l-glutamate in the GDH active site [10], and molecular modelling simulations [7] indeed show IPA binding to the l-glutamate pocket. To determine the Ki (affinity constant for the free enzyme) of IPA for pfGDHa, the apparent Km (Km app ) of the parasite
enzyme for l-glutamate was determined with varying concentrations of IPA. The Ki was determined as 6.2 M by plotting Km app /Vmax against concentration of IPA (Fig. 1). With boGDH, in agreement with earlier studies [7,10], the Ki for IPA was 426 M, corresponding to ∼70-fold discrimination between the plasmodial and mammalian enzymes, suggesting that it may be possible to design a selective inhibitor of P. falciparum GDH. The next step was to confirm the effect of IPA on the native NADP(H)-dependent GDH activity of the parasite. We could not detect NADP(H)-dependent GDH activity directly in parasite extracts by spectrophotometric assays, possibly owing to interference by other enzymes that use NADPH e.g. oxidases. Also, it was not possible to purify enough enzyme for our measurements from small amounts of parasite crude extract. Therefore, we sought to detect GDH activity in samples separated by native-PAGE. A major single band of l-glutamate-dependent dehydrogenase activity was detected (Fig. 2A, box I) with NADP+ ; reaction with NAD+ was not significant (not shown). No obvious difference was observed between crude extracts prepared by freeze–thawing or by Triton X-100 (not shown), consistent with placement of pfGDHa in the parasite’s cytosol [14]. The activity-staining band was also recognised by anti-GDHa antiserum (Fig. 2A, box II). The reaction was dramatically inhibited by IPA (Fig. 2B) even at a concentration 10 times below that of l-glutamate. Densitometry (not shown), indicated 60–80% inhibition with 0.1 mM IPA and 80–100% with 1 mM, but this can only be a rough estimate. However, from the experimental Ki , we can calculate 87% inhibition by 0.1 mM IPA and 98% by 1 mM inhibitor, in fair agreement with the densitometry estimates. The susceptibility of P. falciparum growth to IPA was tested next. Growth inhibition was clearly observed although only with concentrations of IPA 10- to 20-fold higher than the Ki determined for the purified enzyme (Fig. 3). The IC50 was estimated to be ∼340 ± 80 M at low O2 (Fig. 3A) and ∼450 ± 6 M at high O2 (Fig. 3B) tensions. No differences in growth rate of the untreated cultures in low and high O2 were detected. We attributed this relatively low effectiveness to the low permeability of these compounds through biological membranes, especially those of red blood cells [15,16]. To test this hypothesis, dimethyl isophthalic acid (DMIPA) was tested in P. falciparum cultures under the same conditions (Fig. 3). In Aspergillus this diester is easily absorbed and metabolised down to isophthalic acid, which can impair the fungal growth by inhibiting glutamate dehydrogenase [16]. We reasoned that a similar situation could arise in Plasmodium.
154
I.M. Aparicio et al. / Molecular & Biochemical Parasitology 172 (2010) 152–155
Fig. 2. Isophthalic acid inhibits native NADP(H)-dependent pfGDHa. (A) P. falciparum crude extracts were resolved by electrophoresis on 10% polyacrylamide gels under native conditions (native-PAGE) and then incubated in standard assay buffer [6] in the presence of 1 mM l-glutamate, 100 M NADP+ , 0.3 mg/mL of INT and 0.03 mg/mL PES (activity staining; box I) or pfGDHa was revealed by Western blot with anti-pfGDHa antiserum (box II). (B) Activity staining gels were performed in the presence of saturating concentrations of IPA. In A, purified E. coli GDH was used as a positive control.
Fig. 3. Isophthalic acid reduces intra-erythrocytic growth of P. falciparum 3D7 strain. Susceptibility tests were carried out in low O2 tensions (∼16% in the candle jar technique) (A), or in high O2 tensions (∼21%) (B). IPA, isophthalic acid; DMIPA, dimethyl isophthalic acid; o-PA, orthophthalic acid. In brief, parasites were cultivated asynchronously as described previously [21] in 96-well plates for 72 h in the presence of increasing concentrations of inhibitors. The parasites’ viability was determined by assaying the activity of lactate dehydrogenase (PfLDH) [22]. For the experiments with IPA and DMIPA, n = 6. For those with o-PA, n = 3. Vertical bars represent standard deviations.
I.M. Aparicio et al. / Molecular & Biochemical Parasitology 172 (2010) 152–155
DMIPA did not affect the activity of the purified pfGDHa (data not shown), but it proved to be toxic for P. falciparum cultures to a similar extent to IPA (IC50 of 250 ± 40 M at low O2 and of 760 ± 120 M at high O2 ). The statistical analysis showed no difference between the effects of IPA and DMIPA, either in low or high O2 tensions. This was not completely unexpected, as it has been reported that the permeability of the red blood cell membrane is altered during P. falciparum infection and that an extensive tubovesicular network is formed connecting the parasite cytoplasm and that of the host cell [17]. The authors have also shown that the uptake of l-glutamate by the infected RBC is higher than that of the non-infected cell, and proposed that this could be explored for the delivery of antimalarials structurally similar to this amino acid. In order to rule out the possibility that the toxicity of IPA or DMIPA was a general feature of organic acids, the IPA analogue, orthophthalate (o-PA) was tested. With the purified enzyme it gave Ki of ∼5.2 mM, so that it is ∼1000-fold less effective against pfGDHa than IPA. As expected, o-PA had no significant effect on the parasite growth (Fig. 3A and B), regardless of low or high O2 tensions, showing IC50 values between 10 M and ∞. This is consistent with the view that the inhibition of the parasite growth by IPA was caused by the inhibition of its glutamate dehydrogenase(s). Enzymes related to Plasmodium’s anti-oxidant defences, especially those responsible for the maintenance of the NADP+ :NADPH ratio are promising drug targets. The source of NADPH for the parasite is not completely clear, but, since pfGDHa has been proposed to be the main source of NADPH, GDH inhibitors seemed worth exploring. Mammals possess two GDH isoforms: a housekeeping gene, predominantly expressed in the liver, and a neural-specific gene [18,19]. GDH is, however, absent from mammalian red blood cells. Although the parasite and the human liver GDHs share only 23% sequence identity, most of the similarity is concentrated around the catalytic site, and so an important question to answer was whether or not it was possible to produce inhibitors that would successfully discriminate between the mammalian and protozoan enzymes. Isophthalic acid shows good discrimination and is a tight inhibitor of the recombinant pfGDHa. It also specifically reduces parasite growth/viability, whereas o-PA, which does not inhibit pfGDHa, does not affect parasite growth inside the RBCs. This suggests that IPA has a specific target within the parasite, which we propose to be its glutamate dehydrogenase activity. The inhibitory activity of IPA against the malaria parasite was, admittedly, lower than expected in view of the Ki observed for the purified enzyme. However, recent metabolomic studies with RBC stages of P. falciparum have shown that the intracellular concentration of l-glutamate is about 10 mM [20]. Since IPA is not an irreversible inhibitor of GDHs, such a high concentration of competing substrate would inevitably raise the concentration of the inhibitor required to impair parasite growth. Although we have focussed in this study on the abundant pfGDHa that has been proposed as a/the major source of NADPH for glutathione reduction in P. falciparum, one should not lose sight of the fact that two other GDH genes have been identified in the genome. Both enzymes have been cloned in our group and their characterization is currently under way. One is a second NADP(H)dependent enzyme, putatively targeted to the parasite’s apicoplast. The other is a putative NAD(H)-dependent GDH, showing significant homology with the large NAD(H)-dependent fungal GDHs. The possibility that one (or both) of these two new enzymes, rather than the cytoplasmic pfGDHa is(are) the actual target(s) of IPA cannot be discarded at this stage. Several apicoplast-targeted proteins characterized so far have been demonstrated to be essential for the parasite survival inside the RBC, and the same could be true for the apicoplast-targeted NADP(H)-dependent GDH. The affinity of IPA for these other GDHs has not yet been assessed. Overall, our results indicate that P. falciparum glutamate dehydrogenase(s) are good
155
candidate targets for the development of new antimalarials, and, with suitable design and screening, the necessary selectivity may be achievable. IPA can be used as a prototype for drug design, since it is less potent against mammalian GDHs, discriminating well between the plasmodial and bovine enzymes. With the availability of the structures of plasmodial and human enzymes, rational drug design will be employed in the near future in order to deduce structures likely to give improved affinity and specificity for the parasite enzyme(s). Acknowledgements I.M.A. was supported by a post-graduate research scholarship from the Centre for Synthesis and Chemical Biology of the Conway Institute under the PRTLI Scheme of the Irish Higher Education Authority. This work was also supported by a Fellowship grant to PCE from Science Foundation Ireland. References [1] WHO. World malaria report; 2008. [2] Buffinton GD, Hunt NH, Cowden WB, Clark IA. Malaria: a role for reactive oxygen species in parasite killing and host pathology. Free Rad Cell Dam Dis;1986:201–20. London: Richelieu. [3] Müller S. Redox and antioxidant systems of the malaria parasite Plasmodium falciparum. Mol Microbiol 2004;53:1291–305. [4] Vander Jagt DL, Hunsaker LA, Kibirige M, Campos NM. NADPH production by the malarial parasite Plasmodium falciparum. Blood 1989;74:471–4. [5] Vander Jagt DL, Intress C, Heidrich JE, Mrema JE, Rieckmann KH, Heidrich HG. Marker enzymes of Plasmodium falciparum and human erythrocytes as indicators of parasite purity. J Parasitol 1982;68:1068–71. [6] Wagner JT, Lüdemann H, Färber PM, Lottspeich F, Krauth-Siegel RL. Glutamate dehydrogenase, the marker protein of Plasmodium falciparum. Cloning, expression and characterization of the malarial enzyme. Eur J Biochem 1996;258:813–9. [7] Song Y. To what extent is it possible to design selective glutamate dehydrogenase inhibitors? MSc thesis, University College Dublin; 2005. [8] Werner C, Stubbs MT, Krauth-Siegel RL, Klebe G. The crystal structure of Plasmodium falciparum glutamate dehydrogenase, a putative target for novel antimalarial drugs. J Mol Biol 2005;349:597–607. [9] Smith TJ, Schmidt T, Fang J, Wu J, Siuzdak G, Stanley CA. The structure of apo human glutamate dehydrogenase details subunit communication and allostery. J Mol Biol 2002;318:765–77. [10] Caughey WS, Smiley JD, Hellerman L. l-Glutamic acid dehydrogenase: structural requirements for substrate competition: effect of thyroxine. J Biol Chem 1957;224:591–607. [11] Rice DW, Hornby DP, Engel PC. Crystallization of an NAD+ -dependent glutamate dehydrogenase from Clostridium symbiosum. J Mol Biol 1985;181:147–9. [12] Baker PJ, Britton KL, Engel PC, et al. Subunit assembly and active site location in the structure of glutamate dehydrogenase. Proteins 1992;12:75–86. [13] Gardner MJ, Hall N, Fung E, et al. Genome sequence of the human malaria parasite Plasmodium falciparum. Nature 2002;419:498–511. [14] Ling IT, Cooksley S, Bates SA, Hempelmann E, Wilson RJ. Antibodies to the glutamate dehydrogenase of Plasmodium falciparum. Parasitology 1986;92:313–24. [15] Aubert L, Motais R. Molecular features of organic anion permeability in ox red blood cell. J Physiol 1975;246:159–79. [16] Choudhury R, Noor S, Varadarajalu LP, Punekar NS. Delineation of an in vivo inhibitor for Aspergillus glutamate dehydrogenase. Enzyme Microb Technol 2008;42:151–9. [17] Lauer SA, Rathod PK, Ghori N, Haldar K. A membrane network for nutrient import in red cells infected with the malaria parasite. Science 1997;276:1122–5. [18] Plaitakis A, Metaxari M, Shashidharan P. Nerve tissue-specific (GLUD2) and housekeeping (GLUD1) human glutamate dehydrogenases are regulated by distinct allosteric mechanisms: implications for biological function. J Neurochem 2000;75:1862–9. [19] Plaitakis A, Zaganas I. Regulation of human glutamate dehydrogenases: implications for glutamate, ammonia and energy metabolism in brain. J Neurosci Res 2001;66:899–908. [20] Teng R, Junankar PR, Bubb WA, Rae C, Mercier P, Kirk K. Metabolite profiling of the intraerythrocytic malaria parasite Plasmodium falciparum by 1 H NMR spectroscopy. NMR Biomed 2009;22:292–302. [21] Fennell BJ, Al-shatr ZA, Bell A. Isotype expression, post-translational modification and stage-specific production of tubulins in erythrocytic Plasmodium falciparum. Int J Parasitol 2008;38:527–39. [22] Makler MT, Ries JM, Williams JA, et al. Parasite lactate dehydrogenase as an assay for Plasmodium falciparum drug sensitivity. Am J Trop Med Hyg 1993;48:739–41.
Contents lists available at ScienceDirect
Molecular & Biochemical Parasitology
Short communication
Susceptibility of Plasmodium falciparum to glutamate dehydrogenase inhibitors—A possible new antimalarial target Isabela M. Aparicio a , Alejandro Marín-Menéndez b , Angus Bell b , Paul C. Engel a,∗ a b
Conway Institute of Biomolecular and Biomedical Research, University College Dublin, Dublin, Ireland Department of Microbiology, School of Genetics & Microbiology, Moyne Institute of Preventive Medicine, Trinity College Dublin, Dublin, Ireland
a r t i c l e
i n f o
Article history: Received 3 February 2010 Received in revised form 31 March 2010 Accepted 8 April 2010 Available online 23 April 2010 Keywords: Plasmodium falciparum Antimalarial Glutamate dehydrogenase Anti-oxidant defences Enzyme inhibition Isophthalic acid
a b s t r a c t With the rapid spread of drug-resistant strains of Plasmodium falciparum, the development of new antimalarials is an urgent need. As malaria parasites live in a highly pro-oxidant environment, their anti-oxidant defences have frequently been suggested as candidate drug targets. A key point in such defences is the production of NADPH e.g. for maintaining anti-oxidant glutathione in the reduced state. Some authors have attributed this function in P. falciparum to a glutamate dehydrogenase, therefore proposed as a potential drug target. Here we show that isophthalic acid inhibits both Plasmodium GDH and bovine GDH but showing marked discrimination (70-fold lower Ki for the parasite GDH). Isophthalic acid impairs intra-erythrocytic growth of P. falciparum in vitro whilst o-phthalic acid, not a GDH inhibitor, shows no effect. This offers hope that with careful design or thorough screening it should be possible to find inhibitors with the necessary selectivity between parasite and human GDHs. © 2010 Elsevier B.V. All rights reserved.
Every year over 1 million deaths are attributed to falciparum malaria [1], and available drugs are gradually becoming ineffective, as new resistant strains emerge. The development of effective new drugs is thus of utmost importance for public health. Spending most of their mammalian cycle within red blood cells, Plasmodium species are subject to dangerous oxidative stress. The parasite digests host haemoglobin, freeing the reactive haem group. In this environment, reactive oxygen species (ROS) are liable to be formed mainly through Fenton-like reactions [2]. The parasite’s anti-oxidant defences include various peroxidases and thiol-dependent reductases [3]. It seems, however, that these mechanisms are barely adequate and are supplemented by a reliance on parallel defences of the host red cell, as evidenced, for example, by the prevalence of glucose 6-phosphate dehydrogenase deficiency in malarial areas. This suggests that the anti-oxidant defences are a suitable target for antimalarial attack. A central ele-
Abbreviations: DMIPA, dimethyl isophthalic acid; GDH, glutamate dehydrogenase; INT, 2-(4-iodophenyl)-3-(4-nitrophenyl)-5-phenyltetrazolium chloride; IPA, isophthalic acid; NAD(H), nicotinamide adenine dinucleotide oxidised/reduced; NADP(H), nicotinamide adenine dinucleotide phosphate oxidised/reduced; o-PA, orthophthalic acid; PES, phenazine ethosulfate; pf, Plasmodium falciparum; RBCs, red blood cells; ROS, reactive oxygen species. ∗ Corresponding author at: UCD School of Biomolecular and Biomedical Sciences, Conway Institute, University College Dublin, Belfield, Dublin 4, Ireland. Tel.: +353 1 716 6764; fax: +353 1 283 7211. E-mail address: [email protected] (P.C. Engel). 0166-6851/$ – see front matter © 2010 Elsevier B.V. All rights reserved. doi:10.1016/j.molbiopara.2010.04.002
ment of these defences is the supply of NADPH, e.g. for the reduction of glutathione, and it has been suggested that in Plasmodium falciparum a glutamate dehydrogenase (GDH) is the main generator of NADPH [4]. This enzyme is abundantly expressed by the parasite, and, since it is absent from the host red blood cell, it is actually a marker for the presence of the parasite [5]. Glutamate dehydrogenases are widely distributed enzymes that catalyse the reversible oxidative deamination of l-glutamate to give 2-oxoglutarate and ammonia, using as a co-factor either NAD(H), NADP(H) or both: l-Glutamate+NAD(P)+ ↔ 2-oxoglutarate + NH4 + +NAD(P)H + H+ P. falciparum NADP+ -GDH has been cloned and expressed in Escherichia coli [6] making possible inhibitor studies. However, since GDH is also important for the host, and has a well-conserved active site, an initial question is whether sufficiently selective inhibition can be achieved. Encouraging initial proof-of-principle experiments [7] showed remarkable selectivity in inhibition of bovine and clostridial GDHs (used as surrogates) by various compounds. Solved structures for parasite [8] and human [9] GDHs offer a valuable framework for inhibitor design. A useful starting point is provided by a wide range of compounds screened by Caughey et al. [10] 30 years before the solution of the first GDH structure [11,12]. The best inhibitors (of bovine GDH) were compounds incorporating an extended glutarate structure (e.g. isophthalic acid) or structurally similar molecules, e.g. bromofuroic acid, where
I.M. Aparicio et al. / Molecular & Biochemical Parasitology 172 (2010) 152–155
153
Fig. 1. Secondary plot of isophthalic acid inhibition for pfGDHa to calculate the inhibition constants. Ki was estimated by plotting Km /Vmax against IPA concentration. For clarity, only lower concentrations of IPA were displayed on the inset. Assays were performed accordingly to the previously described protocol [6], in the presence of increasing concentrations of IPA.
one carboxyl group is replaced by an electronegative halogen atom. In this paper we extend the characterization of the NADPHdependent GDH from P. falciparum (over-expressed in E. coli) and explore the selectivity of inhibition by isophthalate in comparison with bovine GDH, a close homologue of the human enzyme. We have also studied the inhibition of GDH activity in crude parasite extracts and measured IC50 values for isophthalate and related compounds using cultured parasites. Sequencing of the P. falciparum genome [13] has revealed the existence of two other GDH genes. For clarity, the previously described GDH [6,8] is called GDHa and the coding gene gdha. For the screening of potential inhibitors, it was necessary to obtain high yields of pure enzyme, and to improve expression we re-cloned the gdha gene into pET21a (Novagen). This gave an approximately 50fold increase in soluble expression when compared to the pQE60 system (Qiagen) previously used [6]. We also optimized a cheaper and robust two-step method of purification, using dye-affinity and hydrophobic interaction, which resulted in ∼60% recovery of the pfGDHa purified to homogeneity, as confirmed by SDS-PAGE and Western blot (not shown). The kinetic parameters of the recombinant pfGDHa at 25 ◦ C were in good agreement with those previously published [6]. However, in order to work in physiological conditions, we also obtained the kinetic parameters at 37 ◦ C. No significant difference was observed in the apparent Km , when compared to figures for 25 ◦ C [6], but the Vmax and therefore catalytic efficiency of pfGDHa was significantly higher (about 1.6-fold) at 37 ◦ C in comparison to 25 ◦ C. Also pfGDHa shows remarkable stability at higher temperatures, keeping 100% of its activity for at least 24 h at 37 ◦ C (not shown). In face of these observations, the subsequent analysis could be performed at physiological temperature, more consistent with the development of a stable and effective drug. Previous studies of competitive inhibitors from our group [7] have demonstrated differences in affinity constants (Ki ) of up to 10-fold between Clostridium symbiosum GDH (csGDH) and bovine liver GDH (boGDH), used as surrogates for pfGDHa and human liver GDH, with 49 and 97% identity, respectively. Accordingly, we selected isophthalic acid (IPA), which showed the lowest Ki for csGDH (∼30 M against ∼300 M for boGDH), to test with pfGDHa, which shares much higher identity with csGDH than with boGDH. IPA was proposed to compete with l-glutamate in the GDH active site [10], and molecular modelling simulations [7] indeed show IPA binding to the l-glutamate pocket. To determine the Ki (affinity constant for the free enzyme) of IPA for pfGDHa, the apparent Km (Km app ) of the parasite
enzyme for l-glutamate was determined with varying concentrations of IPA. The Ki was determined as 6.2 M by plotting Km app /Vmax against concentration of IPA (Fig. 1). With boGDH, in agreement with earlier studies [7,10], the Ki for IPA was 426 M, corresponding to ∼70-fold discrimination between the plasmodial and mammalian enzymes, suggesting that it may be possible to design a selective inhibitor of P. falciparum GDH. The next step was to confirm the effect of IPA on the native NADP(H)-dependent GDH activity of the parasite. We could not detect NADP(H)-dependent GDH activity directly in parasite extracts by spectrophotometric assays, possibly owing to interference by other enzymes that use NADPH e.g. oxidases. Also, it was not possible to purify enough enzyme for our measurements from small amounts of parasite crude extract. Therefore, we sought to detect GDH activity in samples separated by native-PAGE. A major single band of l-glutamate-dependent dehydrogenase activity was detected (Fig. 2A, box I) with NADP+ ; reaction with NAD+ was not significant (not shown). No obvious difference was observed between crude extracts prepared by freeze–thawing or by Triton X-100 (not shown), consistent with placement of pfGDHa in the parasite’s cytosol [14]. The activity-staining band was also recognised by anti-GDHa antiserum (Fig. 2A, box II). The reaction was dramatically inhibited by IPA (Fig. 2B) even at a concentration 10 times below that of l-glutamate. Densitometry (not shown), indicated 60–80% inhibition with 0.1 mM IPA and 80–100% with 1 mM, but this can only be a rough estimate. However, from the experimental Ki , we can calculate 87% inhibition by 0.1 mM IPA and 98% by 1 mM inhibitor, in fair agreement with the densitometry estimates. The susceptibility of P. falciparum growth to IPA was tested next. Growth inhibition was clearly observed although only with concentrations of IPA 10- to 20-fold higher than the Ki determined for the purified enzyme (Fig. 3). The IC50 was estimated to be ∼340 ± 80 M at low O2 (Fig. 3A) and ∼450 ± 6 M at high O2 (Fig. 3B) tensions. No differences in growth rate of the untreated cultures in low and high O2 were detected. We attributed this relatively low effectiveness to the low permeability of these compounds through biological membranes, especially those of red blood cells [15,16]. To test this hypothesis, dimethyl isophthalic acid (DMIPA) was tested in P. falciparum cultures under the same conditions (Fig. 3). In Aspergillus this diester is easily absorbed and metabolised down to isophthalic acid, which can impair the fungal growth by inhibiting glutamate dehydrogenase [16]. We reasoned that a similar situation could arise in Plasmodium.
154
I.M. Aparicio et al. / Molecular & Biochemical Parasitology 172 (2010) 152–155
Fig. 2. Isophthalic acid inhibits native NADP(H)-dependent pfGDHa. (A) P. falciparum crude extracts were resolved by electrophoresis on 10% polyacrylamide gels under native conditions (native-PAGE) and then incubated in standard assay buffer [6] in the presence of 1 mM l-glutamate, 100 M NADP+ , 0.3 mg/mL of INT and 0.03 mg/mL PES (activity staining; box I) or pfGDHa was revealed by Western blot with anti-pfGDHa antiserum (box II). (B) Activity staining gels were performed in the presence of saturating concentrations of IPA. In A, purified E. coli GDH was used as a positive control.
Fig. 3. Isophthalic acid reduces intra-erythrocytic growth of P. falciparum 3D7 strain. Susceptibility tests were carried out in low O2 tensions (∼16% in the candle jar technique) (A), or in high O2 tensions (∼21%) (B). IPA, isophthalic acid; DMIPA, dimethyl isophthalic acid; o-PA, orthophthalic acid. In brief, parasites were cultivated asynchronously as described previously [21] in 96-well plates for 72 h in the presence of increasing concentrations of inhibitors. The parasites’ viability was determined by assaying the activity of lactate dehydrogenase (PfLDH) [22]. For the experiments with IPA and DMIPA, n = 6. For those with o-PA, n = 3. Vertical bars represent standard deviations.
I.M. Aparicio et al. / Molecular & Biochemical Parasitology 172 (2010) 152–155
DMIPA did not affect the activity of the purified pfGDHa (data not shown), but it proved to be toxic for P. falciparum cultures to a similar extent to IPA (IC50 of 250 ± 40 M at low O2 and of 760 ± 120 M at high O2 ). The statistical analysis showed no difference between the effects of IPA and DMIPA, either in low or high O2 tensions. This was not completely unexpected, as it has been reported that the permeability of the red blood cell membrane is altered during P. falciparum infection and that an extensive tubovesicular network is formed connecting the parasite cytoplasm and that of the host cell [17]. The authors have also shown that the uptake of l-glutamate by the infected RBC is higher than that of the non-infected cell, and proposed that this could be explored for the delivery of antimalarials structurally similar to this amino acid. In order to rule out the possibility that the toxicity of IPA or DMIPA was a general feature of organic acids, the IPA analogue, orthophthalate (o-PA) was tested. With the purified enzyme it gave Ki of ∼5.2 mM, so that it is ∼1000-fold less effective against pfGDHa than IPA. As expected, o-PA had no significant effect on the parasite growth (Fig. 3A and B), regardless of low or high O2 tensions, showing IC50 values between 10 M and ∞. This is consistent with the view that the inhibition of the parasite growth by IPA was caused by the inhibition of its glutamate dehydrogenase(s). Enzymes related to Plasmodium’s anti-oxidant defences, especially those responsible for the maintenance of the NADP+ :NADPH ratio are promising drug targets. The source of NADPH for the parasite is not completely clear, but, since pfGDHa has been proposed to be the main source of NADPH, GDH inhibitors seemed worth exploring. Mammals possess two GDH isoforms: a housekeeping gene, predominantly expressed in the liver, and a neural-specific gene [18,19]. GDH is, however, absent from mammalian red blood cells. Although the parasite and the human liver GDHs share only 23% sequence identity, most of the similarity is concentrated around the catalytic site, and so an important question to answer was whether or not it was possible to produce inhibitors that would successfully discriminate between the mammalian and protozoan enzymes. Isophthalic acid shows good discrimination and is a tight inhibitor of the recombinant pfGDHa. It also specifically reduces parasite growth/viability, whereas o-PA, which does not inhibit pfGDHa, does not affect parasite growth inside the RBCs. This suggests that IPA has a specific target within the parasite, which we propose to be its glutamate dehydrogenase activity. The inhibitory activity of IPA against the malaria parasite was, admittedly, lower than expected in view of the Ki observed for the purified enzyme. However, recent metabolomic studies with RBC stages of P. falciparum have shown that the intracellular concentration of l-glutamate is about 10 mM [20]. Since IPA is not an irreversible inhibitor of GDHs, such a high concentration of competing substrate would inevitably raise the concentration of the inhibitor required to impair parasite growth. Although we have focussed in this study on the abundant pfGDHa that has been proposed as a/the major source of NADPH for glutathione reduction in P. falciparum, one should not lose sight of the fact that two other GDH genes have been identified in the genome. Both enzymes have been cloned in our group and their characterization is currently under way. One is a second NADP(H)dependent enzyme, putatively targeted to the parasite’s apicoplast. The other is a putative NAD(H)-dependent GDH, showing significant homology with the large NAD(H)-dependent fungal GDHs. The possibility that one (or both) of these two new enzymes, rather than the cytoplasmic pfGDHa is(are) the actual target(s) of IPA cannot be discarded at this stage. Several apicoplast-targeted proteins characterized so far have been demonstrated to be essential for the parasite survival inside the RBC, and the same could be true for the apicoplast-targeted NADP(H)-dependent GDH. The affinity of IPA for these other GDHs has not yet been assessed. Overall, our results indicate that P. falciparum glutamate dehydrogenase(s) are good
155
candidate targets for the development of new antimalarials, and, with suitable design and screening, the necessary selectivity may be achievable. IPA can be used as a prototype for drug design, since it is less potent against mammalian GDHs, discriminating well between the plasmodial and bovine enzymes. With the availability of the structures of plasmodial and human enzymes, rational drug design will be employed in the near future in order to deduce structures likely to give improved affinity and specificity for the parasite enzyme(s). Acknowledgements I.M.A. was supported by a post-graduate research scholarship from the Centre for Synthesis and Chemical Biology of the Conway Institute under the PRTLI Scheme of the Irish Higher Education Authority. This work was also supported by a Fellowship grant to PCE from Science Foundation Ireland. References [1] WHO. World malaria report; 2008. [2] Buffinton GD, Hunt NH, Cowden WB, Clark IA. Malaria: a role for reactive oxygen species in parasite killing and host pathology. Free Rad Cell Dam Dis;1986:201–20. London: Richelieu. [3] Müller S. Redox and antioxidant systems of the malaria parasite Plasmodium falciparum. Mol Microbiol 2004;53:1291–305. [4] Vander Jagt DL, Hunsaker LA, Kibirige M, Campos NM. NADPH production by the malarial parasite Plasmodium falciparum. Blood 1989;74:471–4. [5] Vander Jagt DL, Intress C, Heidrich JE, Mrema JE, Rieckmann KH, Heidrich HG. Marker enzymes of Plasmodium falciparum and human erythrocytes as indicators of parasite purity. J Parasitol 1982;68:1068–71. [6] Wagner JT, Lüdemann H, Färber PM, Lottspeich F, Krauth-Siegel RL. Glutamate dehydrogenase, the marker protein of Plasmodium falciparum. Cloning, expression and characterization of the malarial enzyme. Eur J Biochem 1996;258:813–9. [7] Song Y. To what extent is it possible to design selective glutamate dehydrogenase inhibitors? MSc thesis, University College Dublin; 2005. [8] Werner C, Stubbs MT, Krauth-Siegel RL, Klebe G. The crystal structure of Plasmodium falciparum glutamate dehydrogenase, a putative target for novel antimalarial drugs. J Mol Biol 2005;349:597–607. [9] Smith TJ, Schmidt T, Fang J, Wu J, Siuzdak G, Stanley CA. The structure of apo human glutamate dehydrogenase details subunit communication and allostery. J Mol Biol 2002;318:765–77. [10] Caughey WS, Smiley JD, Hellerman L. l-Glutamic acid dehydrogenase: structural requirements for substrate competition: effect of thyroxine. J Biol Chem 1957;224:591–607. [11] Rice DW, Hornby DP, Engel PC. Crystallization of an NAD+ -dependent glutamate dehydrogenase from Clostridium symbiosum. J Mol Biol 1985;181:147–9. [12] Baker PJ, Britton KL, Engel PC, et al. Subunit assembly and active site location in the structure of glutamate dehydrogenase. Proteins 1992;12:75–86. [13] Gardner MJ, Hall N, Fung E, et al. Genome sequence of the human malaria parasite Plasmodium falciparum. Nature 2002;419:498–511. [14] Ling IT, Cooksley S, Bates SA, Hempelmann E, Wilson RJ. Antibodies to the glutamate dehydrogenase of Plasmodium falciparum. Parasitology 1986;92:313–24. [15] Aubert L, Motais R. Molecular features of organic anion permeability in ox red blood cell. J Physiol 1975;246:159–79. [16] Choudhury R, Noor S, Varadarajalu LP, Punekar NS. Delineation of an in vivo inhibitor for Aspergillus glutamate dehydrogenase. Enzyme Microb Technol 2008;42:151–9. [17] Lauer SA, Rathod PK, Ghori N, Haldar K. A membrane network for nutrient import in red cells infected with the malaria parasite. Science 1997;276:1122–5. [18] Plaitakis A, Metaxari M, Shashidharan P. Nerve tissue-specific (GLUD2) and housekeeping (GLUD1) human glutamate dehydrogenases are regulated by distinct allosteric mechanisms: implications for biological function. J Neurochem 2000;75:1862–9. [19] Plaitakis A, Zaganas I. Regulation of human glutamate dehydrogenases: implications for glutamate, ammonia and energy metabolism in brain. J Neurosci Res 2001;66:899–908. [20] Teng R, Junankar PR, Bubb WA, Rae C, Mercier P, Kirk K. Metabolite profiling of the intraerythrocytic malaria parasite Plasmodium falciparum by 1 H NMR spectroscopy. NMR Biomed 2009;22:292–302. [21] Fennell BJ, Al-shatr ZA, Bell A. Isotype expression, post-translational modification and stage-specific production of tubulins in erythrocytic Plasmodium falciparum. Int J Parasitol 2008;38:527–39. [22] Makler MT, Ries JM, Williams JA, et al. Parasite lactate dehydrogenase as an assay for Plasmodium falciparum drug sensitivity. Am J Trop Med Hyg 1993;48:739–41.