Báo cáo khoa học: An a-proteobacterial type malate dehydrogenase may complement LDH function in Plasmodium falciparum Cloning and biochemical characterization of the enzyme

doi:10.1111/j.1432-1033.2004.04281.x

Eur. J. Biochem. 271, 3488–3502 (2004) (cid:1) FEBS 2004

An a-proteobacterial type malate dehydrogenase may complement LDH function in Plasmodiumfalciparum Cloning and biochemical characterization of the enzyme

Abhai K. Tripathi1, Prashant V. Desai2, Anupam Pradhan1, Shabana I. Khan1, Mitchell A. Avery1,2, Larry A. Walker1,3 and Babu L. Tekwani1 1National Center for Natural Product Research, Research Institute of Pharmacological Sciences, 2Department of Medicinal Chemistry, and 3Department of Pharmacology, School of Pharmacy, University of Mississippi, MS, USA

Pf MDH but the N-terminal glycine motif, which is involved in nucleotide binding, was similar to the GXGXXG signa- ture sequence found in Pf LDH and also in a-proteobacterial MDHs. Oxamic acid did not inhibit Pf MDH, while gossy- pol, which interacts at the nucleotide binding site of oxido- reductases and shows antimalarial activity, inhibited Pf MDH also. Treatment of a synchronized culture of P. falciparum trophozoites with gossypol caused induction in expression of Pf MDH, while expression of Pf LDH was reduced and expression of malate:quinone oxidoreductase remained unchanged. Pf MDH may complement Pf LDH function of NAD/NADH coupling in malaria parasites. Thus, dual inhibitors of Pf MDH and Pf LDH may be required to target this pathway and to develop potential new antimalarial drugs.

Keywords: gossypol; lactate dehydrogenase; malate dehy- drogenase; malate:quinone oxidoreductase; Plasmodium falciparum.

Malate dehydrogenase (MDH) may be important in car- bohydrate and energy metabolism in malarial parasites. The cDNA corresponding to the MDH gene, identiﬁed on chromosome 6 of the Plasmodium falciparum genome, was ampliﬁed by RT-PCR, cloned and overexpressed in Escherichia coli. The recombinant Pf MDH was puriﬁed to homogeneity and biochemically characterized as an NAD+(H)-speciﬁc MDH, which catalysed reversible inter- conversion of malate to oxaloacetate. Pf MDH could not use NADP/NADPH as a cofactor, but used acetylpyridine adenine dinucleoide, an analogue of NAD. The enzyme exhibited strict substrate and cofactor speciﬁcity. The highest levels of Pf MDH transcripts were detected in trophozoites while the Pf MDH protein level remained high in troph- ozoites as well as schizonts. A highly reﬁned model of Pf MDH revealed distinct structural characteristics of sub- strate and cofactor binding sites and important amino acid residues lining these pockets. The active site amino acid residues involved in substrate binding were conserved in

The enzymes associated with carbohydrate and energy metabolism in malarial parasites have attracted signiﬁcant attention as potential targets for new antimalarial drug discovery [1]. During asexual reproduction and growth within the host’s erythrocytes the parasite depends mainly on

the glycolytic pathway to obtain energy. Infected erythro- cytes consume almost 100 times more glucose than unin- fected erythrocytes [2]. Almost all of this increase in glucose utilization is the result of the synthesis of enzymes of glycolytic (and associated) pathways by the parasite. Fulminating malaria infections are characterized by hypo- glycemia and potentially lethal lactic acidosis [3]. Earlier studies and the recent release of a fully annotated map of the Plasmodium falciparum genome have shown the presence of a complete battery of enzymes of the Embden–Meyerhof– Parnas pathway of glycolysis and the tricarboxylic acid (TCA) cycle in malarial parasites [4]. However, the role of oxidative metabolism in the malaria parasite through the TCA cycle remains unclear. A recent report has indicated the operation of oxidative phosphorylation and the presence of an alternative NADH-Q oxidoreductase and malate:qui- none oxidoreductase in Plasmodium yoelii [5]. The oxidative metabolism in malaria parasite may be important for de novo pyrimidine biosynthesis rather than energy metabolism [1]. L-Malate dehydrogenase (MDH; EC 1.1.1.37) and L-lactate dehydrogenase (LDH; EC 1.1.1.27) are 2-keto- acid:NAD(P) oxidoreductases that are universally distri- buted in both eukaryotic and prokaryotic organisms [6,7]. LDH is particularly important in anaerobic metabolism,

Correspondence to B. L. Tekwani, National Center for Natural Product Research, School of Pharmacy, University of Mississippi, MS 38677, USA. Fax: +1 662 915 7062, Tel.: +1 662 915 7882, E-mail: btekwani@olemiss.edu Abbreviations: IPTG, isopropyl thio-b-D-galactoside; Pf MDH, Plas- modium falciparum malate dehydrogenase; Pf LDH, P. falciparum lactate dehydrogenase; Pf MQO, P. falciparum malate : quinone oxidoreductase; OAA, oxaloacetate; APAD, acetyl pyridine adenine dinucleotide; RBC, red blood cells; TCA, tricarboxylic acid. Enzymes: L-Malate dehydrogenase (MDH; EC 1.1.1.37); L-lactate dehydrogenase (LDH; EC 1.1.1.27). Notes: Part of the work reported in this paper was presented at the 51st Annual Meeting of American Society of Tropical Medicine & Hygiene, held at Denver, November 10–14, 2002 [Am. J. Trop. Med. Hyg. (2002) 67, 146]. The cDNA sequence reported in paper has been submitted to GenBank under accession no AY324107. (Received 2 March 2004, revised 2 June 2004, accepted 7 July 2004)

Malate dehydrogenase of P. falciparum (Eur. J. Biochem. 271) 3489

(cid:1) FEBS 2004

crystal structures of nearest structural homologues i.e. Pf LDH [14], E. coli MDH [22] and Chlorobium tepidum MDH [23]. High homology of Pf MDH with Pf LDH also prompted us to ask whether MDH can complement the function of LDH in the malaria parasite. This was studied by treating the P. falciparum cultures with gossypol, an inhibitor of LDH, which has also shown antimalarial action [24].

Materials and methods

the molecular and structural

P. falciparum (D6 strain) was grown in vitro in RPMI 1640 medium with A+ human red blood cells (RBCs) and A+ human serum as described previously [25]. For large-scale culture the parasite was grown in 75 cm2 culture ﬂasks, which can accommodate up to 200 mL of culture medium. The parasite was grown in 24 culture ﬂasks to (cid:1) 10–15% parasitemia. The cultures were highly synchronized with two/three cycles of sorbitol treatment [26]. The parasite cultures initiated with early ring stage were harvested at regular 8 h intervals starting from 0 to 40 h to isolate the parasite at different developmental stages of life cycle viz., early rings, late rings, early trophozoites, late trophozoites, early schizonts and late/mature schizonts. The RBC-free parasite was prepared by lysis of infected RBCs with saponin [27]. Genomic DNA and RNA were prepared by using a genomic DNA isolation kit (Qiagen) and trizol(cid:2) method (Invitrogen), respectively, as per the manufacturer’s protocol.

Cloning of PfMDH, PfLDH and PfMQO

primer

reverse for cloning of

while MDH may have a role in oxidative metabolism as well as other physiological functions, depending on its biochemi- cal characteristics and intracellular localization [8]. LDHs are cytosolic proteins but isoforms of MDH have been localized to the cytosol as well as different subcellular organelles such as mitochondria, chloroplast, peroxysomes and glyoxysomes [8]. LDHs and MDHs characterized from different organisms form a large superfamily [9]. A speciﬁc phylogenetic distribution of LDH, LDH-like MDH and dimeric MDH over the Archaeal, Bacterial and Eukaryal domains was observed. All LDHs and MDHs enzymes from apicomplexan parasites, which include Plasmodium spp., Toxoplasma gondii, Cryptosporidium parvum and Eimeria tenella, were found to be monophylectic within the (cid:1)LDH-like MDH(cid:2) group as a sister to alpha-proteobac- terial MDHs [10]. All of the apicomplexan LDHs, with the exception of LDH1 from Cryptosporidium parvum, form a separate clade from their MDH counterparts, indicating that these LDHs evolved from an ancestral apicomplexan MDH by gene duplication and functional conversion before expansion of apicomplexans [10,11]. Both LDH and MDH from various organisms have been characterized in consid- erable detail at levels [6,7,12,13]. However, only LDH from P. falciparum (Pf LDH) has been characterized in signiﬁcant detail, including determination of its crystal structure. The enzyme has also been exploited as a potential target for design and development of speciﬁc enzyme inhibitors as antimalarial agents [14–17]. A comparative kinetic and structural analysis of LDH from all four species of human malaria parasite has also been reported recently [18]. However, MDH, which catalyses reversible conversion of malate to oxaloacetate (OAA) using NAD(P) as a cofactor, has not been characterized in sufﬁcient detail from P. falciparum. It was shown previously that P. falciparum contained a single isozyme of MDH (Pf MDH) which was suggested to be localized to the cytosol of the parasite [19]. The avian malaria parasite P. lophurae was also found to contain a single isozyme of MDH which was distinct from host erythrocyte MDH [20]. Alternatively, P. berghei the rodent malaria parasite has been shown to have multiple MDH isozymes [21]. However, the presence of multiple MDH isoforms in mouse erythrocytes, the preference of P. berghei to infect reticulocytes, and the detection of low activity of MDH in P. berghei raises questions regarding the validity of this report. The recent release of the P. falciparum genome sequence has indicated the presence of a gene putatively encoding for MDH [4]. Initial analysis of the Pf MDH sequence revealed its higher homology with the MDH sequences from a-proteobacteria and also with LDHs. Pf MDH may therefore be classiﬁed as LDH-like MDH, similar to that from other apicomplexan parasites [10,11].

In this study Pf MDH was cloned from P. falciparum genomic DNA and total RNA by PCR and RT-PCR, overexpressed in Escherichia coli and the recombinant enzyme was puriﬁed to homogeneity. Functional charac- terization of the recombinant Pf MDH protein was carried out by analysis of its biochemical characteristics and enzyme kinetics. Expression of MDH in P. falciparum was found to be developmentally regulated during its growth and prolif- eration in human erythrocytes. A highly reﬁned homology model for Pf MDH was also constructed on the basis of

Pf MDH was cloned for functional and biochemical two genes Pf LDH and characterization. The other Pf MQO were also cloned to obtain the DNA probes for analysis of its expression in P. falciparum cultures. Pf LDH was also overexpressed in E. coli cultures to obtain the enzyme for evaluation of the inhibitors. The primers 5¢-ACTAAAATTGCTTTAATAGG primer (forward TAG-3¢ 5¢-TTATTTAATGTC and GAAAGC-3¢) full-length MDH ORF DNA and cDNA corresponding to the complete coding sequence were designed from the putative MDH gene (MAL6P1.242) identiﬁed on chromosome 6 of P. falcipa- rum (http://www.plasmodb.org). The DNA corresponding to full-length Pf MDH ORF was ampliﬁed by PCR using Taq DNA polymerase and P. falciparum genomic DNA as the template. The cDNA was prepared by RT-PCR using DNase-treated total RNA, isolated from mixed culture of P. falciparum as the template using a Qiagen RT-PCR kit. The control reaction without reverse transcriptase was run simultaneously to check for genomic DNA contamination and to ensure speciﬁc ampliﬁcation of cDNA. The DNA corresponding to full-length ORF sequence of Pf LDH (PF13-0141) and Pf MQO (MAL6P1.258) were also ampli- ﬁed using the appropriate primers (Pf LDH: forward 5¢-ATGGCACCAAAAGCAAAAATCG-3¢ and reverse 5¢-AGCTAATGCCTTCATTCTCTTAG-3¢; Pf MQO for- ward 5¢-ATGATATGTGTTAAAAATATTTTG-3¢ and reverse 5¢-TCATAAATAATTAACGGGATATTCG-3¢). Both PCR and RT-PCR products were cloned directly into an E. coli expression vector (pQE30) using a UA cloning kit

3490 A. K. Tripathi et al. (Eur. J. Biochem. 271)

(cid:1) FEBS 2004

was determined by a standard size exclusion chromatog- raphy procedure using Sephacryl S200.

Enzyme assays

(Qiagen). The ligation mixtures were used to transform E coli XL-1 blue cells and the bacterial colonies transformed with recombinant plasmids were selected on Luria–Bertani medium agar plates containing 100 lgÆmL)1 ampicillin. The presence of the DNA/cDNA inserts, and their orientation in the recombinant plasmids, was conﬁrmed by digestion of the plasmid minipreps with appropriate restriction enzymes and their analysis by electrophoresis on 0.8% (w/v) agarose gels. The plasmid containing Pf MDH ORF (DNA and cDNA) and Pf MQO cDNA in the correct orientation were sent to Laragen Inc. (http://www.laragen.com) for sequen- cing on both strands. The Pf MDH mRNA sequence for P. falciparum (Sierra Leone D6 strain) has been submitted to GenBank under accession no AAQ23154 and was found to be the same as that reported for a putative MDH sequence of a 3D7 P. falciparum strain. The recombinant pQE30Pf LDH and pQE30Pf MQO plasmids were also analysed in the same way.

Overexpression and puriﬁcation of recombinant proteins

time,

isopropyl

MDH catalyses reversible oxidation of malate to OAA utilizing NAD(H) or NADP(H) as cofactor. MDH was assayed spectrophotometrically for both oxidation of malate and reduction of OAA by recording the change in absorbance at 340 nm as described earlier [19]. The assays were set up in clear ﬂat-bottomed 96-well micro- plates. For the malate reduction assay, the reaction mixture contained in a total volume of 200 lL with glycine buffer (pH 10.2, 50 mM) malate (20 mM or as speciﬁed) and NAD or NADP (500 lM or as speciﬁed) and an appropriate amount of the enzyme. Similarly for the OAA reduction assay the reaction mixture contained in a total volume of 200 lL with phosphate buffer (pH 7.0, 50 mM), OAA (250 lM or as speciﬁed) and NADH or NADPH (200 lM or as speciﬁed). Appropriate controls without substrates or cofactor were also set up simultaneously. Each assay was set up in triplicate. The plates were read on a microplate reader in kinetic mode for 5 min and the change in absorbance per minute was recorded. Activity of the enzyme was calculated in terms of micromoles of NAD reduced or NADH oxidized. Similarly Pf LDH activity was also measured according to the spectrophotometric method as described earlier [16]. For determination of optimum pH for pfMDH activity the assays were performed at different pH ranging from 6.0 to 11.0 using phosphate (pH 6.0–8.0) and glycine (pH 8.5–11.0) buffers with saturating concentrations of the substrate and the cofactor. For enzyme kinetic studies, assays were performed at varying concentrations of substrates (OAA or malate) and the cofactors [NAD/ H or acetyl pyridine adenine dinucleotide (APAD/H)]. Kinetic constants were computed with GRAFIT 5. To determine substrate speciﬁcity of Pf MDH the enzyme activity was also determined using different 2-keto or 2-hydroxy carboxylic acids such as lactate, pyruvate, a-ketoglutarate, keto-malonate, oxo-butyrate and keto- adipic acid.

Analysis of expression of the enzymes in P.falciparum cultures

Expression of Pf MDH at the different stages of the asexual intra-erythrocytic cycle was evaluated by semiquantitative analysis of transcripts by Northern blotting and also by quantiﬁcation of the enzyme protein by Western blotting. Expression of Pf MDH, Pf LDH and Pf MQO in control and drug treated P. falciparum cultures was evaluated by Northern and Western blotting. Quantitative analysis of the transcripts and proteins was performed by using the NIH IMAGE (version 1.61) analysis program.

Northern blotting. The P. falciparum cultures at different stages of life cycle or the control and drug treated cultures were harvested by centrifugation. Total RNA was isolated directly from the infected RBCs by the Trizol(cid:2) method as per the manufacturer’s protocol. Purity and concentration of RNA was checked spectrophotometrically by reading the

For overexpression of Pf MDH, Pf LDH and Pf MQO the recombinant plasmids prepared from XL-1 blue cells were transformed into E. coli M15 (pREP 4) cells as recommended by Qiagen. The transformed colonies were selected on LB agar plates containing 100 lgÆmL)1 ampicillin and 25 lgÆmL)1 kanamycin. A 5 mL overnight culture of transformed E. coli cells was transferred to 500 mL fresh LB medium containing ampicillin and kanamycin and grown further at 37 (cid:3)C to D600 (cid:1) 0.5. At thio-b-D-galactoside (IPTG) was this added to the cultures to a ﬁnal concentration of 0.5 mM in order to induce overexpression of recombinant proteins. The cultures were grown for additional 5 h at 37 (cid:3)C with constant shaking. Cells were harvested by centrifugation at 5000 g for 15 min at 4 (cid:3)C. Harvested bacterial pellets were re-suspended in phosphate buffer containing 300 mM NaCl and 10 mM imidazole and lysed by sonication. The extracts were centrifuged at 4 (cid:3)C at 15 000 g for 30 min. Expression of recombinant protein was checked by SDS/PAGE and also by Western blotting. Signiﬁcant overexpression of Pf MDH and Pf LDH was achieved in E. coli cultures induced with IPTG. Expression of recombinant Pf MDH and Pf LDH could be achieved in soluble fractions. However, over- expression of Pf MQO could not be achieved in this system, even with the use of various concentrations of IPTG and varying culture conditions. The soluble frac- tions, which contained signiﬁcant amounts of the Pf MDH/Pf LDH protein, were used for puriﬁcation. The recombinant proteins contain a 6 · His tag which facilitates their puriﬁcation by afﬁnity chromatography using Ni–NTA chelating columns. The soluble bacterial extracts were passed through Ni–NTA agarose columns and the columns were washed with buffer containing 20 mM imidazole. Recombinant proteins were eluted with buffer containing 200 mM imidazole. The soluble extracts, column fractions and puriﬁed proteins were analysed by SDS/PAGE. Gels were stained with Coomassie brilliant blue R and also with silver stain, to check the purity of recombinant proteins. The oligomeric structure of the en- zymatically functional recombinant Pf MDH preparations

Malate dehydrogenase of P. falciparum (Eur. J. Biochem. 271) 3491

(cid:1) FEBS 2004

Western blotting as described above. Antimalarial activities of oxamic acid and gossypol were determined in vitro for P. falciparum cultures as described earlier [29].

Analysis of PfMDH sequence

Trypanosoma

absorbance at 260/280 nm. An equal amount of total RNA (20 lg) from each of the six stages or untreated and drug treated samples was separated on denaturing 1% (w/v) agarose gel. To check the concentration and purity of RNA the gels were stained with ethidium bromide and visualized on a UV transluminator. Equal intensity of ribosomal RNA bands in all of the lanes indicated that an equal amount of total RNA had been loaded. The RNA from agarose gels was transferred under vacuum to positively charged nylon membrane (Sigma) and the membranes were washed with 0.1% NaCl/Cit and 0.1% SDS for 30 min and prehybrid- ized in 10 mL prehybridization solution [3.6 mL sonicated salmon sperm DNA of a 10 mgÆmL)1 stock, 0.2 mL 50· Denhardt’s solution, 0.1 mL 10% (w/v) SDS, 3.64 mL 10 mgÆmL)1 and 3 mL 20· NaCl/Cit] for 5 h at 42 (cid:3)C with gentle shaking. Membranes were then hybridized with hybridization solution containing denatured radioactive Pf MDH or Pf LDH or Pf MQO probe. The recombinant plasmids containing Pf MDH/Pf LDH/Pf MQO cDNA inserts were digested with appropriate restriction enzymes and the inserts were puriﬁed on the agarose gel, using a gel extraction kit (Qiagen). The probes were prepared by labelling the inserts with [32P]dCTP using a random prime labelling kit (Amersham Biosciences). Hybridization was performed overnight at 42 (cid:3)C with gentle shaking. The membranes were then washed successively with 10· NaCl/ Cit, 1% (w/v) SDS (20 min), 1· NaCl/Cit, 0.5% (w/v) SDS (30 min) and 0.1· NaCl/Cit, 0.2% (w/v) SDS (45 min). The membranes were exposed to hyperﬁlm and the transcripts were visualized by autoradiography.

Protein sequences of various MDHs and LDHs were obtained from GenBank. All of the selected sequences along with Pf MDH sequence were aligned using CLUSTAL– WINDOWS interface with default parameters [30]. Misaligned or poorly aligned sequences were selected manually and realigned to obtain proper alignment of all sequences. PHYLIP format tree output was selected to obtain the bootstrap neighbor-joining tree. An unrooted phylogenetic tree was drawn using TREE VIEW. The accession number of the sequences used for the phylogenetic analysis were: P. falciparum MDH (NP_703844), P. falciparum LDH (CAD52397), P. yoelii MDH (EAA22943), P. yoelii LDH (EAA15666), Ricketssia prowozekii MDH (NP_220759), Mesorhizobium loti MDH (NP_105210), Sinorhizobium meliloti MDH (AAG41996), Rhizobium leguminosarum brucei MDH MDH (CAA05717), (AAK83037), human cytoplasmic MDH (P40925), mouse cytoplasmic MDH (P14152), maize cytoplasmic MDH (T02935), Arabidopsis thaliana MDH (AAM65532), Tricho- monas vaginalis MDH (2208292 A), T. vaginalis LDH (AAC72735), E. coli MDH (AAC76268), E. coli LDH (NP_418062), C. tepidum MDH (NP_662392), Leishmania major MDH (CAB55506), S. meliloti LDH (CAC49543), M. loti LDH (NP_107321), human LDH (P07864), maize LDH (P29038), and A. thaliana LDH (AAM64829).

Generation of a model structure of PfMDH

Computational studies were performed on a Silicon Graph- ics Octane 2 workstation, equipped with two parallel R12000 processors, V6 graphics board and 512 MB mem- ory. Comparative protein structure modelling was per- formed with the HOMOLOGY module of INSIGHTII 2000 (Accelrys Inc., San Diego, CA, USA). Energy minimiza- tions and molecular dynamics were accomplished in the DISCOVER module of INSIGHTII 2000. The geometric and local environmental consistency of the model was evaluated with the PROSTAT and PROFILES-3D [31] modules of INSIGHTII 2000 as well as the MATCHMAKER [32] module of SYBYL 6.9 (Tripos Associates Inc., St. Louis, MO, USA).

Western blotting. For quantiﬁcation of enzyme protein the parasites were isolated from infected RBCs by saponin lysis [27]. The parasite pellets were washed extensively with cold NaCl/Pi to remove RBC proteins and membrane contaminants. Finally, the parasite pellet was resuspended in NaCl/Pi containing 0.1% (v/v) Triton X-100 and the proteins were solublized by sonication. The soluble protein extracts were obtained after centrifugation of the lysates at 10 000 g for 15 min in an Eppendorf microcentrifuge tube. Protein concentrations in the clear supernatants were determined by the Bradford method [28]. Equal amount of total protein (100 lg) from each sample was loaded and separated by SDS/PAGE. Pre- stained molecular mass protein markers (Sigma-Aldrich) were also loaded in the ﬁrst lane. The proteins were transferred onto nitrocellulose membrane (Sigma) using semi-dry blot and membrane was probed with puriﬁed IgG fraction obtained from polyclonal antiserum raised against recombinant Pf MDH (Antibodies Inc.). Pf MDH protein was visualized using peroxidase conjugated sec- ondary antibody (anti-rabbit IgG).

A WU-BLAST 2.0 [33] PDB search was performed on the Pf MDH sequence with default parameters of BLASTP gapped alignment. Only two sequences, MDH from C. tepidum (40% sequence identity) and Pf LDH (39% sequence identity) passed the identity ﬁlter of 35% and hence were used to build the 3D model. A total of nine structurally conserved regions and seven structurally variable regions or loops were identiﬁed. The structurally conserved regions were built from the homologues whereas the coordinates for the loops were obtained by searching the PDB for regions of proteins that meet a deﬁned geometric criterion. The protocol uses an existing Ca carbon distance matrix to search for regions of proteins whose Ca distances best ﬁt those of the selected region of the protein being studied, while meeting the additional constraint of having the speciﬁed number of

To evaluate the effect of gossypol on expression of the enzymes, highly synchronized cultures of P. falciparum with 10–15% parasitaemia were treated at early trophozoite stages with gossypol at a concentration of 25 or 100 lgÆmL)1 in the culture medium. The cultures were harvested after 8 h and total RNA was isolated from the infected RBCs and analysed for expression of Pf MDH, Pf LDH and Pf MQO by Northern blotting as described above. Parasite was isolated from control and treated cultures and equal amount of protein was analysed by

3492 A. K. Tripathi et al. (Eur. J. Biochem. 271)

(cid:1) FEBS 2004

LDHs, rather than the signature sequence of GXXGXXG found in most of the MDHs. More detailed analysis of substrate and nucleotide binding sites has been described in the section on structural analysis of Pf MDH by homology modelling.

residues present between the regions of interest. Different rotamers for the residues that line the active pocket were also studied and the most energy stable rotamers were retained. The crude model so obtained was then reﬁned by minimization using an analogous approach reported previously [34].

Overexpression and biochemical characterization of PfMDH

Overexpression of Pf MDH cDNA in E. coli yielded a recombinant protein of (cid:1) 36 kDa as analysed by SDS/ PAGE (Fig. 3). The recombinant protein expressed in E. coli was larger then the predicted Pf MDH protein of 34 kDa because it also contained some extra amino acids derived from the bacterial expression vector, including a His6-tag at the N-terminal end, which facilitates its puriﬁ- cation. Most of the recombinant protein was recovered in the soluble fraction and was found to be functionally active as MDH, i.e. it catalysed reduction of OAA and oxidation of malate in the presence of NADH/NAD. The protein could be puriﬁed to apparent homogeneity, as analysed by SDS/PAGE, after single-step afﬁnity chromatography through a Ni–NTA agarose column. About 10 mg of the functionally active recombinant Pf MDH could be recov- ered from 1 L of the E. coli culture.

The structure of the Pf MDH in complex with OAA and NADH was obtained as follows. First, the substrate and the cofactor were placed at appropriate binding regions of Pf MDH based on their corresponding locations in the Pf LDH [14] and C. tepidum MDH structures [23]. The complex was then minimized using steepest descents for 1000 iterations followed by 2000 iterations of conjugate gradients. This was followed by molecular dynamics simulations for 25 ps at 300 K and ﬁnally minimization to a gradient of 0.001 kcalÆmol)1ÆA˚ )1 or less using conjugate gradients. During this reﬁnement, the side chains of the residues lining the substrate as well as the cofactor binding pockets were allowed to move freely whereas rest of the protein atoms were ﬁxed. Comparison of the Pf MDH model structure with crystal structures of Pf LDH [14], C. tepidum MDH [23], E. coli MDH [35] and T. gondii LDH [36] was achieved by determination of the rmsd between each pair. This is one of the most acceptable methods for determination of structural similarity among the proteins [37].

Results

Cloning and characterization of PfMDH mRNA and gene

Preparation and ampliﬁcation of cDNA corresponding to the complete encoding region of Pf MDH by RT-PCR yielded a single amplicon of 942 bases, which encodes a predicted protein of 313 amino acids with calculated molecular mass of 34 040 Da. The sequence of Pf MDH mRNA from P. falciparum (Sierra Leone D6 strain) was found to be the same as reported for the Pf MDH gene identiﬁed on chromosome 6 of P. falciparum (3D7 strain). The results conﬁrm that the PfMDH gene had no introns. Presence of Pf MDH mRNA in P. falciparum cultures also indicated that the parasite expressed the gene during the asexual reproduction cycle. The sequence of the Pf MDH protein was subjected to BLAST-P analysis with non- redundant GenBank protein database. Comparison of Pf MDH sequence with some representative MDH sequences are presented in Fig. 1. The sequence of Pf LDH was also included in this as Pf MDH showed signiﬁcant similarity with a-proteobacterial MDHs and also with LDHs characterized in bacteria and lower eukaryotes, particularly that from apicomplexan parasites. Pf MDH did not possess any target signal sequence and therefore represents a cytosolic MDH. This was further conﬁrmed by the observation that Pf MDH sequence shows higher homology with cysolic MDH than with organellar MDHs (Fig. 2). The Pf MDH protein sequence also contained several conserved motifs and amino acids residues that have been found to be important in MDH and LDH functions (Fig. 1). The N-terminal glycine motif which is involved in nucleotide binding function was identiﬁed as GSGQIG in Pf MDH and corresponded to the signature sequence GXGXXG, found in proteobacterial MDHs and

The homogeneous preparations of recombinant Pf MDH were used to characterize oligomeric status, enzyme kinetic properties, and substrate/cofactor speciﬁcities. Analysis of recombinant Pf MDH by size exclusion chromatography on Sepahcryl S200 did not provide conclusive information on the oligomeric state of the enzymatically active protein. Both dimeric as well as terameric forms of the proteins were eluted. Pf MDH may exist as a dimer of dimers. MDH catalyses interconversion of malate and OAA inside the cell. However, in vitro these reactions occur at a different pH. Therefore, before a detailed kinetic analysis of the enzyme could be conducted, the optimum pH for oxidation of malate and reduction of OAA were determined by per- forming the assays at varying pH (5.5–11.0) using different reaction buffers (Fig. 4). The optimum pH for oxidation of malate in the presence of NAD was 10.2 and for the reduction of OAA using NADH as the cofactor it was 7.0. Further kinetic characterization of enzyme was performed at the experimentally determined pH optima. Pf MDH catalysed the reduction of OAA with an efﬁciency six to eight times greater than that of oxidation of malate as determined by evaluation of Vmax and kcat values for malate/NAD and OAA/NADH (Table 1). The saturating concentrations of OAA and NADH were found to be (cid:1) 250 lM and (cid:1) 150 lM, respectively, while for malate and NAD these were (cid:1) 20 mM and (cid:1) 1.5 mM, respectively. Further increase in the concentration of the substrates or cofactor did not cause any inhibition in enzyme activity. Importantly Pf MDH did not show characteristic substrate inhibition. Pf MDH did not utilize NADP/NADPH as an alternate cofactor, having been tested up to maximum concentration of 100 mM. A surprising observation was the use of acetyl pyridine adenine dinucleotide (APAD/ APADH) as an alternate cofactor by Pf MDH, which showed almost comparable efﬁciency to that of NAD/ NADH (Table 1). Porcine heart MDH (M7383, Sigma- Aldrich) did not show any activity with APAD/APADH

Malate dehydrogenase of P. falciparum (Eur. J. Biochem. 271) 3493

(cid:1) FEBS 2004

MAPKAKIVLV-G-SGMIGGVMATLIVQKNLG---DVVLFDIVKNMPHGKALD----TSHT 51 P. falciparum(LDH) P. falciparum(MDH) ---MTKIALI-G-SGQIGAIVGELCLLENLG---DLILYDVVPGIPQGKALD----LKHF 48 Rickettsia prowazekii MKKNPKISLI-G-SGNIGGTLAHLISLKKLG---DIVLFDVSEGLPQGKALD----LMQA 51 Rhizobium leguminosarum MARN-KIALI-G-SGMIGGTLAHLAGLKELG---DIVLFDIADGIPQGKGLD----ISQS 50 Cryptosporidium parvum MR--KKISII-G-AGQIGSTIALLLGQKDLG---DVYMFDIIEGVPQGKALD----LNHC 49 ----MKITVI-G-AGNVGATTAFRLAEKQLAR--ELVLLDVVEGIPQGKALD----MYES 48 Chlorobium tepidum ----MK-VAVLGAAGGIGQALALLLKTQLPSG-SELSLYDIAPVTP-GVAVD-----LSH 48 E. coli MSNTCKRVAVTGAAGQIGYSLLPLIAAGRMLGFDQRVQLQLLDISPALKALEGIRAELMD 60 Trypanosoma brucei MSEPIR-VLVTGAAGQIAYSLLYSIGNGSVFGKDQPIILVLLDITPMMGVLDGVLMELQD 59 Homo sapiens : * :* :. : : * :: NVMAYSNCKVSGSNTYDDLAGADVVIVTAGFTKAPGKSDKEWNRDDLLPLNNKIMIEIGG 111 P. falciparum (LDH) STILGVNRNILGTNQIEDIKDADIIVITAGVQRKEGMT-----REDLIGVNGKIMKSVAE 103 P. falciprum (MDH) Rickettsia provazakii ATIEGSDIKIKGTNDYRDIEGSDAVIITAGLPRKPGMS-----RDDLISVNTKIMKDVAQ 106 Rhizobium legumunosarum SPVEGFDVNLTGASDYSAIEGADVCIVTAGVARKPGMS-----RDDLLGINLKVMEQVGA 105 Cryptosporidium parvum MALIGSPAKIFGENNYEYLQNSDVVIITAGVPRKPNMT-----RSDLLTVNAKIVGSVAE 104 GPVGLFDTKVTGSNDYADTANSDIVVITAGLPRKPGMT-----REDLLSMNAGIVREVTG 103 Chlorobium tepidum IPTAVKIKGFSGEDATPALEGADVVLISAGVARKPGMD-----RSDLFNVNAGIVKNLVQ 103 E. coli CSFPLLDGVVITDEPKVAFDKADIAILCGAFPRKPGME-----RRDLLQTNAKIFSEQGR 115 Trypanosoma brucei CALPLLKDVIATDKEDVAFKDLDVAILVGSMPRREGME-----RKDLLKANVKIFKSQGA 114 Homo sapiens . . * :: ... : . * **: * :. . HIKKNC-PNAFIIVVTNPVDVMVQLLHQHS---GVPKNKIIGLGGVLDTSRLKYYISQKL 167 P. falciparum (LDH) SVKLHC-SKAFVICVSNPLDIMVNVFHKFS---NLPHEKICGMAGILDTSRYCSLIADKL 159 P. falciparum (MDH) Rickettsia provazakii NIKKYA-QNAFVIVITNPLDIMVYVMLKES---GLPHNKVIGMAGVLDSSRFNLFLAKEF 162 Rhizobium leguminosarum GIKKYA-PNAFVICITNPLDAMVWALQKFS---GLPANKVVGMAGVLDSSRFRLFLAKEF 161 Cryptosporidium parvum NVGKYC-PNAFVICITNPLDAMVYYFKEKS---GIPANKVCGMSGVLDSARFRCNLSRAL 160 RIMEHS-KNPIIVVVSNPLDIMTHVAWQKS---GLPKERVIGMAGVLDSARFRSFIAMEL 159 Chlorobium tepidum QVAKTC-PKACIGIITNPVNTTVAIAAEVLKKAGVYDKNKLFGVTTLDIIRSNTFVAELK 162 E. coli VLGEVASPNCRVCVVGNPANTNALILLRESK--GKLNPRFVTALTRLDHNRATAQVAERA 173 Trypanosoma brucei ALDKYAKKSVKVIVVGNPANTNCLTASKSAP--SIPKENFS-CLTRLDHNRAKAQIALKL 171 Homo sapiens : . . : : : . . . * :: NVCPRDVN-AHIVGAHGNKMVLLKRYITVGGIPLQEFINNKLISDAELE-AIFDRTVNTA 225 P. falciparum (LDH) P. falciparum (MDH) KVSAEDVN-AVILGGHGDLMVPLQRYTSVNGVPLSEFVKKNMISQNEIQ-EIIQKTRNMG 217 Rickettsia prowazakii KVSVKNVN-SIVLGGHGDTMVPLLRYSTISGVPIPDLIKMGLSSNKNIE-KIIDRTKNGG 220 Rhizobium laguminosarum NVSVQDVT-AFVLGGHGDTMVPLARYSTVGGIPLTDLVTMGWVTKERLE-EIIQRTRDGG 219 Cryptosporidium parvum GVKPSDVS-AIVVGGHGDEMIPLTSSVTIGGILLSDFVEQGKITHSQIN-EIIKKTAFGG 218 GVSMQDVT-ACVLGGHGDAMVPVVKYTTVAGIPVADLIS-----AERIA-ELVERTRTGG 212 Chlorobium tepidum GKQPGEVE-VPVIGGHSGVTILPLLSQVPGVSFTEQEVADLTKRIQNAGTEVVEAKAGGG 221 E. coli RARVEEVKNCIIWGNHSGTQVPDVNSATVG--GK--PARAAVDNDAFFDNEFITIVQERG 229 Trypanosoma brucei GVTANDVKNVIIWGNHSSTQYPDVNHAKVKLQGKEVGVYEALKDDSWLKGEFVTTVQQRG 231 Homo sapiens :* * * . .. : LEIVNLHA—-SPYVAPAAAIIEMAESYLKDLKKVLICSTLLE-G-QYGHSD-IFGGTPVV 280 P. falciparum (LDH) P. falciparum (MDH) AEIIKLAK-ASAAFAPAAAITKMIKSYLYNENNLFTCAVYLN-G-HYNCSN-LFVGSTAK 273 Rickettsia prowazakii GEIVKLLKTGSAYYAPAASAIAMLESYLKDKRQILTCAAYLQ-G-EYDIHD-LYIGVPII 277 Rhizobium laguminosarum AEIVGLLKTGSAYYAPAASAIEMAESYLKDKKRVLPCAAHLS-G-QYGVKD-MYVGVPTV 276 Cryptosporidium parvum GEIVELLKTGSAFYAPAASAVAMAQAYLKDSKSVLVCSTYLT-G-QYNVNN-LFVGVPVV 275 AEIVNHLKQGSAFYSPATSVVEMVESIVLDRKRVLTCAVSLD-G-QYGIDG-TFVGVPVK 269 Chlorobium tepidum S------ATLSMGQAAARFGLSLVRALQGEQGVVECAYVE---G---DGQYARFFSQPLL 269 E. coli AEIMKLRGLSSALSAAKAIVDHVHDWMLGTPSGTHVSMAVYSDGNPYGVPGGLIFSFPVT 289 Trypanosoma brucei AAVIKARKLSSAMSAAKAICDHVRDIWFGTPEGEFVSMGVISDGNSYGVPDDLLYSFPVV 291 Homo sapiens : . : : . *. . . LGANGVEQVIELQ-LNSEEKAKFDEAIAETKRMKALA-------- 316 P. falciparum (LDH) P. falciprum (MDH) INNKG-AHPVEFP-LTKEEQDLYTESIASVQSNTQKAFDLIK--- 313 Rickettsia prowazakii IGKEGVIKVIELQ-LTEEEKILFYKSVTEVKKLIDTIQ------- 314 Rhizobium laguminosarum IGAGGVERIIEID-LNKTEKEAFDKSVGAVAGLCEACINIAPALK 320 Cryptosporidium parvum IGKNGIEDVVIVN-LSDDEKSLFSKSVESIQNLVQDLKSLNL--- 316 LGKNGVEHIYEIK-LDQSDLDLLQKSAKIVDENCKMLDASQG--- 310 Chlorobium tepidum LGKNGVEERKSIGTLSAFEQNALEGMLDTLKKDIALGEEFVNK-- 312 E. coli CSGGEWQIVSGLN-VTPAISERIKATTTELEEERREVSA------ 327 Trypanosoma brucei IKNKTWKFVEGLP-INDFSREKMDLTAKELTEEKESAFEFLSSA- 334 Homo sapiens

(results not shown). Pf MDH also showed strict substrate speciﬁcity as no enzyme activity could be detected with lactate, pyruvate, a-ketoglutarate, oxo-butyrate and

keto-adipic acid. However, a-keto malonate was utilized by Pf MDH at very high concentrations with a Km of 5.44 mM (Table 1).

Fig. 1. Multiple sequence alignment of MDHs from some representative organisms with deduced amino acid sequence of Pf MDH, for analysis of conserved motifs and amino acids important for enzyme function. MDH sequences [P. falciparum (AAQ23154.1), Rickettsia prowazaki (NP_220759), R. leguminosarum (CAA05717), C. parvum (AAP87358), C. tepidum (CAA56810), E. coli (NP_312136), T. brucei (AAK83037) and H. sapiens (NP_005908)] were aligned by CLUSTAL W (http://www.ebi.ac.uk/clustalw/index.html) using a default parameter. The sequence of Pf LDH (Q27743) was also included in the multiple sequence alignment as it shows signiﬁcant sequence similarity to Pf MDH. The N-terminal glycine motif (the nucleotide binding site) and the substrate binding motif have been boxed. Other conserved amino acids are indicated by : or *.

3494 A. K. Tripathi et al. (Eur. J. Biochem. 271)

(cid:1) FEBS 2004

Structural characteristics of PfMDH

ldh- P. falciparum ldh-P. yoelii

P.falciparum

ldh-MAIZE C.tepidum

P.y oelii

ldh-A.thaliana

R.prowazekii

ldh-Human

M.loti S.meliloti R.leguminosarum

T. brucei

E.coli

Human MOUSE

L.major

Maize A.thaliana

The model structure of Pf MDH was validated using several tools. The Ramachandran plot [38] showed a normal distribution of points with Phi (/) angles mostly restricted to negative values and Psi (w) values clustered in a few distinct regions with 95% of residues occupying the allowed region. An average value of )0.20 kT of the MATCHMAKER score suggested a reasonable 3D model. PROSTAT check for bond lengths, C-a chirality, amide torsion (x), Phi and Psi torsions for helices, Phi for prolines and side chain torsions (v1 and v2) showed no major deviation from the corres- ponding allowed values. PROFILES-3D analysis did not suggest any misfold and the overall self-compatibility score for the model was 137 as compared with a score of 64 or less; the latter would indicate an almost certainly incorrect structure.

ldh-M.loti

T. vaginalis

ldh-T.vagi nalis

ldh-S.meliloti

ldh-E.coli

0.1

Fig. 2. Phylogenetic tree showing the evolutionary relationship of Pf MDH with various MDHs and LDHs. The unrooted tree was con- structed by multiple alignments of all the sequences using CLUSTAL X interface with default parameters. Misaligned or poorly aligned sequences were manually selected and realigned before constructing the phylogenetic tree.

66

45 36 29 24

1 2 3

4

5

6

7

The model structure (Fig. 5) reveals characteristic fea- tures of a 2-hydroxyacid dehydrogenase like MDH and LDH including the classical Rossmann fold [39] constituting the NADH binding pocket. The structural comparison of some of the relevant 2-hydroxyacid dehydrogenases is provided in Table 2. The overall structure of Pf MDH is quite distinct from that of E. coli MDH as evidenced by an rmsd of 5.3 between the Ca coordinates which is not surprising considering the sequence identity of only 26% between the two enzymes. On the other hand the structure appears to be similar to that of Pf LDH with a rmsd of 0.8 between the Ca atoms (Fig. 5). Both the cofactor and the substrate binding regions are seen to overlap closely in the two structures. However, a signiﬁcant difference is observed between the structures of the substrate speciﬁcity loops (residues 78–94 of Pf MDH) as clearly seen in Fig. 5. This is primarily due to the insertion of ﬁve residues into the loop in the case of Pf LDH (Fig. 1). The substrate speciﬁcity loop of Pf MDH shares a sequence identity of 65% with that of the MDH from C. tepidum and E. coli. As expected, the structure of this loop in Pf MDH closely resembles that of ligand bound activated E. coli MDH structures. The model structure of the Pf MDH complexed with NADH and OAA (Fig. 5) appears to be stabilized by several hydrogen bonds between the enzyme, the substrate and the cofactor in addition to hydrophobic interactions. Important residues lining the cofactor binding pocket indicates that Gln11, Asp32, Thr76, Ala77, Val117 and M142 are involved in hydrogen bonding interaction with NADH. Structurally equivalent residues in the case of Pf LDH and E. coli MDH structures show similar interactions.

The binding of OAA to the enzyme appears to be stabilized by strong electrostatic interactions. The catalytic His–Asp pair (His174 and Asp147 in Pf MDH) conserved in the 2-hydroxyacid dehydrogenase family that functions as proton relay system appears to be oriented in a fashion similar to that in the Pf LDH (Fig. 6) and the E. coli MDH structures. The substrate forms hydrogen bonds with Arg81, Arg87, Asn119, His174 and Arg150 as shown in Fig. 6A. Pf LDH appears to form a similar hydrogen bonding network with oxamate except that there is no corresponding residue in the binding pocket equivalent to Arg81 of Pf MDH (Fig. 6B). It is well established that the extra conserved arginine (Arg81), present on the substrate speciﬁcity loops of all MDHs, provides the complementary

Fig. 3. Overexpression and puriﬁcation of recombinant Pf MDH. E. coli cultures were induced by 0.5 mM IPTG for 4 h at 37 (cid:3)C. Recombinant Pf MDH was puriﬁed by Ni–NTA columns (Qiagen) according to the manufacturer’s protocol. Diﬀerent protein samples were analysed by SDS/PAGE and staining with coomassie Brilliant blue R. Lane 1, Molecular mass marker proteins; lane 2, soluble ex- tracts of recombinant E. coli lysate loaded on a Ni–NTA agarose column; lane 3, unadsorbed proteins in the bacterial lysate passed through a Ni–NTA agarose column; lanes 4 and 5, fractions obtained by washing the column with the buﬀer containing a low concentration (10 mM) of imidazole; lane 6, recombinant Pf MDH protein eluted from a Ni–NTA agarose column in buﬀer containing 200 mM imi- dazole; lane 7, second eluate of the column with buﬀer containing 200 mM inmidazole.

Malate dehydrogenase of P. falciparum (Eur. J. Biochem. 271) 3495

(cid:1) FEBS 2004

OAA reduction

4000

Expression of PfMDH during asexual intraerythrocytic schizogony

3000

2000

1000

0

5

6

7

8

9

10 Malate Oxidation

750 y t i v i t c a H D M

500

Highly synchronized P. falciparum cultures were harvested at different developmental stages and expression of Pf MDH was evaluated by analysis of transcripts using Northern blotting (Fig. 7A) and the enzyme protein by Western blotting techniques (Fig. 7B). MDH in P. falcip- rum cultures was expressed as a single transcript of approximately 1.6 kb. The level of Pf MDH transcript was highest in late rings and early trophozoites which declined further in schizonts and was very low in early rings. Western blotting analysis, with anti-Pf MDH IgG, detected only a single protein band of (cid:1) 34 kDa in P. falciparum lysates. The level of Pf MDH protein was equally high in the trophozoite and schizont stages of P. falciparum while these were very low in the early and late ring stages. The results indicate that expression of Pf MDH is developmentally regulated. Despite signiﬁcant the polyclonal antibodies against structural similarity, Pf MDH did not cross react with Pf LDH.

Effect of the inhibitors on PfMDH and PfLDH

250

0

6

8

10

12

pH

charge that neutralizes the additional carboxylate group on the substrate and therefore confers substrate speciﬁcity to MDH.

Oxamic acid, which is known to inhibit LDH by interacting with the substrate binding site, inhibited Pf LDH as well as bovine heart LDH with almost the same efﬁciency (Fig. 8 and Table 3). Pf MDH, porcine heart MDH (mitochon- drial) and porcine heart MDH (cytosolic) were not inhibited by oxamic acid up to a concentration of 2 mM (Table 3 and Fig. 8). Oxamic acid (up to 1 mM) did not show any effect on the growth of P. falciparum in in vitro culture. Gossypol, which is known to inhibit Pf LDH by interacting with the inhibited Pf LDH with a 50% nucleotide binding site, inhibitory concentration (IC50) of 3.1 lM while bovine heart LDH was much less sensitive to inhibition with gossypol. MDHs including Pf MDH and porcine heart MDHs (both mitochondrial and cytosolic) were inhibited by gossypol with almost the same sensitivity (Fig. 8 and Table 3). Gossypol inhibited the growth of P. falciparum with an IC50 of 11.5 ± 2.1 lM. Berberine, a speciﬁc inhibitor of cytosolic as well as organellar MDHs, did not inhibit Pf MDH activity up to a concentration of 10 mM (data not shown).

Fig. 4. Assay of Pf MDH activity at diﬀerent pH. The assay was per- formed at diﬀerent pH in phosphate (pH 5.5–8.00) and glycine (pH 8.00–11.00) buﬀers as described in the Materials and methods section. The amount of enzyme used for assay of malate oxidation (using NAD cofactor) was ﬁve times higher when compared to that used for the assay of OAA reduction (using NADH cofactor). The MDH activity is presented as lmoles of NAD reduced or (NADH oxidized)Æmin)1Æ(mg protein))1.

Table 1. Kinetic characterization of recombinant Pf MDH. Kinetic characteristics of the recombinant Pf MDH were determined as described in Materials and methods. The speciﬁc activity is expressed as lmol NADH oxidized or (NAD reduced)Æmin)1Æmg)1 enzyme protein. Speciﬁc activities were calculated from the activity of enzyme at saturating concentration of diﬀerent substrates and cofactors. Values are given as mean ± SD of at least three observations.

)1Æs)1)

Substrate/cofactor Speciﬁc activity Km (mM) kcat (1Æs)1) kcat/Km (M

Reduction (pH 7.5)

Oxaloacetate (NADH) NADH (OAA) APADH (OAA) 0.030 ± 0.001 0.036 ± 0.002 0.022 ± 0.001 5.44 ± 0.57 960 ± 68 950 ± 79 1010 ± 150 674 ± 35 3657 ± 450 4870 ± 398 4935 ± 238 3899 ± 390 32 · 106 26 · 106 45 · 106 0.12 · 106

a-Keto-malonate (NADH) Oxidation (pH 10.2) Malate (NAD) NAD (malate) APAD (malate) 1.350 ± 0.024 0.152 ± 0.013 0.370 ± 0.019 250 ± 19 150 ± 20 175 ± 21 571 ± 38 507 ± 43 562 ± 56 0.18 · 106 0.98 · 106 0.47 · 106

3496 A. K. Tripathi et al. (Eur. J. Biochem. 271)

(cid:1) FEBS 2004

Comparative expression of PfMDH, PfLDH and PfMQO and effect of gossypol

The effect of gossypol on the expression of Pf LDH and Pf MDH in P. falciparum cultures was evaluated. Highly synchronized cultures of P. falciparum with (cid:1) 15% para- sitaemia were treated with 25 and 100 lgÆmL)1 gossypol

Fig. 5. Model structure of Pf MDH superimposed on the crystal structure of Pf LDH. Pf LDH is shown as a white ribbon, Pf MDH as a yellow ribbon, NADH and oxamate bound to Pf LDH as cyan sticks, NADH and OAA bound to Pf MDH as magenta sticks. To highlight the diﬀerences the substrate speciﬁcity loop is coloured green in the case of Pf LDH and red in the case of Pf MDH.

Table 2. Structural comparison of 2-hydroxyacid dehydrogenases rela- ted to Pf MDH. Root mean square distance (rmsd) for the Ca atoms is used to compare the protiens’ structures [37].

C. tepidum MDHc E. coli MDHd T. gondii LDHe Pf MDHa Pf LDHb

and the cultures were harvested after 8 h. Control cultures without treatment were set up in parallel. Comparative expression of Pf MDH, Pf LDH and Pf MQO were eval- uated using the Northern blotting technique (Fig. 9A). The control P. falciparum cultures (late trophozoite stages) exhibited the highest expression of Pf LDH followed by the expression of Pf MQO which was signiﬁcantly lower than Pf LDH but was still considerably higher compared to Pf MDH. The cultures treated with 25 lgÆmL)1 of gossypol showed signiﬁcant induction of expression of Pf MDH, while at this concentration expression of Pf LDH and Pf MQO remained unaltered. Treatment with 100 lgÆmL)1 gossypol caused further induction of Pf MDH expression while expression of Pf LDH signiﬁcantly decreased with this treatment (Fig. 9A). The Pf MQO expression level remained unchanged upon treatment with 100 mgÆmL)1 gossypol. Induction of Pf MDH due to gossypol treatment was further conﬁrmed by Western analysis (Fig. 9B).

0 0.80 0.85 0.80 0 0.93 0.85 0.93 0 5.25 3.30 2.50 0.95 0.79 2.11 Pf MDH Pf LDH C. tepidum MDH

Discussion

a Model structure. b Dunn et al. [14]. c Dalhus et al. [23]. d Hall and Banaszak [35]. e Kavanagh et al. [36].

During asexual intraerythrocytic schizogony, the malaria parasite obtains its energy mainly by utilization of glucose

E. coli MDH 5.25 T. gondii LDH 0.95 3.30 0.79 2.50 2.11 0 1.74 1.74 0

Malate dehydrogenase of P. falciparum (Eur. J. Biochem. 271) 3497

(cid:1) FEBS 2004

therefore should be localized to the cytosol. The results presented here, and in a recent report [40], clearly demon- strate that this gene is functional in malaria parasites and encodes an NAD+-dependent MDH. Another gene (PF13– 0144) identiﬁed on chromosome 13 of P. falciparum as a putative oxidoreductase, shows 49% identity with 59% of Pf MDH and also with Pf LDH (http://www.plamodb.org). This gene, however, seems to be truncated at the N-terminal end. Pf LDH also is localized on chromosome 13 of P. falciparum.

through glycolysis [1], which also supplies precursors for several other pathways i.e. nonmevalonate pathway of isoprenoids biosynthesis, fatty acid biosynthesis, purine salvage, pyrimidine biosynthesis, shikimate pathway and synthesis of GPI anchors [4]. The parasite, however, is equipped with the complete battery of TCA cycle enzymes, although the operation of a mitochondrial TCA cycle and its role in energy generation in malaria parasites is still debatable. A recent report has indicated the operation of oxidative phosphorylation in P. yoelii, a rodent malaria parasite [5]. MDH is an important link between glycolysis, the TCA cycle and the aspartate malate/OAA shuttle [7]. The differential MDH functions are achieved through multiple MDH isoenzymes with different subcellular local- izations [8]. However, the P. falciparum genome contains only one full length MDH gene on chromosome 6. The Pf MDH gene did not posses any targeting signal and

The biochemical properties of recombinant Pf MDH are similar to those reported earlier for the enzyme puriﬁed from P. falciparum extracts [19]. The important character- istic of Pf MDH is its high homology with bacterial MDHs and also with LDHs. Pf MDH also shows some distinct enzymatic characteristics, i.e. strict substrate speciﬁcity, insensitivity to inhibition by high concentration of the

Fig. 6. Important residues interacting with (A) OAA in Pf MDH and (B) oxamate in Pf LDH [9]. The ligands are coloured by atom types whereas the protein residues are shown as yellow sticks. Hydrogen bonds are depicted as white dotted lines. Hydrogen atoms are not shown for clarity. The residues of the Pf LDH structure are numbered in accordance with Fig. 1.

3498 A. K. Tripathi et al. (Eur. J. Biochem. 271)

(cid:1) FEBS 2004

1.9 kb

A

1.39 kb 36.6 KDa

Table 3. Inhibition of P. falciparum and porcine MDH and LDH by oxamate and gossypol. The enzyme assays were performed with vary- ing concentrations of the inhibitors, at saturating concentrations of the OAA (for MDH) or pyruvate (for LDH) and NADH. Values given are mean ± SD of at least three observations.

B

28 KDa

Inhibitor IC50 (lM)

ER

LR

ET

LT

ES

LS

Enzyme Oxamate Gossypol

Pf MDH Porcine mitochondrial MDH Porcine cytosolic MDH P. falciparum LDH Porcine heart LDH > 2000 > 2000 > 2000 58.0 ± 8.9 78.4 ± 17.9 1.5 ± 0.5 2.4 ± 1.1 2.9 ± 0.9 3.2 ± 0.7 84.5 ± 9.5

Fig. 7. Expression of Pf MDH in P. falciparum cultures during intra- erythrocytic schizogony. (A) Northern blot analysis. Equal amounts (20 lg) of total RNA, isolated from infected RBCs harvested at six time points from a highly synchronized culture of P. falciparum were electrophoresed on agarose gel, transferred to positively charged nylon membrane and hybridized to a radiolabelled Pf MDH probe. Mem- branes were exposed overnight at )80 (cid:3)C with radiography ﬁlm to detect the transcripts. (B) Western blot analysis. The soluble protein lysates prepared from the RBC parasite preparation were separated by SDS/PAGE (10% acrylamide) and transferred to a nitrocellulose membrane. The Pf MDH enzyme protein was detected by blotting the membranes with anti-Pf MDH sera (1 : 100) and detection by the peroxidase method. ER, Early ring; LR, late ring; ET, early troph- ozoite; LT, late trophozoite; ES, early schizont; LS, late schizont.

120 Gossypol

100

80 )

60 %

40 i

20

substrate and utilization of APAD as an alternate cofactor. Some cytosolic MDHs show broader substrate speciﬁcity and are able to utilize other dicarboxylic keto acids also as the substrates [41] and MDHs are inhibited by high concentrations of OAA [42]. Pf MDH was therefore different from these MDHs in its biochemical characteris- tics. Pf LDH was reported earlier to use APAD as an alternate cofactor. This distinct characteristic of the enzyme has been used extensively in a Pf LDH assay-based malaria diagnostic test, Optimal(cid:2) [43]. The Malstat(cid:2) reagent, which is used for selective assay of Pf LDH in P. falciparum cultures, is also used for quantiﬁcation of growth of the malaria parasite in an in vitro antimalarial assay [44,45]. Pf LDH utilizes APAD(H) with much higher efﬁciency than NAD(H), while Pf MDH utilizes APAD(H) and NAD(H) with almost equal efﬁciency. A commercially available NAD+(H) speciﬁc MDH from porcine heart (Sigma M 2634) did not utilize APAD(H) even up to 100 mM.

i

0

0.1

1

10 1000

120 100 Oxamate

( g n n a m e r y t i

v

100

80

60

40 i t c a e m y z n E

20

0

1

10 1000

10000

100 Inhibitor Concentration (µµµµM)

Pf MDH contains almost all of the amino acids residues, which are typically conserved in cytosolic MDHs [46]. R81, R87 and R150, which line the anion binding site of the substrate binding pocket, are positioned to both stabilize and orient the substrate for catalysis. The D147 and H174 pair, which corresponds to the D150 and H177 pair of E. coli MDH, may function as a proton relay system for catalysis [22]. It has been proposed that the presence of an extra arginine residue (R81 in the case of Pf MDH) provided substrate speciﬁcity to the MDH [22,23,47]. The structure of the corresponding region in Pf LDH is highly different, which forms a distant loop due to the insertion of ﬁve amino acids [14,15]. The N-terminal glycine motif GXGXXG is similar to the cofactor binding motif found in most of the LDHs and a-proteobacterial MDHs [12,13,48]. In addition, G10, Q11, D32, T76, A77, V117 and M142 may be involved in a hydrogen bonding interaction with NADH. The hydrogen bond between the carbonyl oxygen of M142 and one of the carboxyamide hydrogens is interesting as a similar hydrogen bond is observed with the corresponding residue L150 (numbered as L163 in the crystal structure) in the case of the Pf LDH crystal structure. This hydrogen bond is considered to be a unique charac- teristic of Pf LDH [14], as it is not seen in most other LDH structures. Even in case of E. coli MDH structures [22,23,35] a corresponding hydrogen bond with V146 is not observed. Comparison of the Pf LDH crystal structure with the Pf MDH model structure indicates that the

Fig. 8. Inhibition of mammalian and Pf LDH and Pf MDH activities by (A) gossypol and (B) oxamic acid. LDHs were assayed in the presence of a saturating concentration of pyruvate and NADH, while MDHs were assayed in the presence of saturating concentrations of OAA and NADH. For inhibition studies the reaction mixtures containing the enzyme and the inhibitors were preincubated for 10 min before initi- ating the reaction by addition of NADH. Bovine heart LDH (j/solid line); Pf LDH (d/broken line); Pf MDH (h/solid line); porcine heart MDH (mitochondrial) (s/solid line); porcine heart MDH (cytosolic) (n/solid line).

Malate dehydrogenase of P. falciparum (Eur. J. Biochem. 271) 3499

(cid:1) FEBS 2004

A

250 0 0 1 G

200 MDH

l

150 5 2 G

) l o r t n o c f o %

o r t n o C

LDH

5 2 G

0 0 1 G

100 0 0 1 G

50 MQO

( n o i s s e r p x E

0 MDH

LDH

MQO

Control

G25

G100

) l

300 G100

B

250 o r t n o c

200 G25

f o %

150 MDH

Control

100

50 ( n o i s s e r p x E

Control

G25

G100

0

nucleotide binding pockets of Pf LDH and Pf MDH are similar and explain the use of APAD(H) as an alternate cofactor by both of the enzymes.

Expression of Pf MDH seems to be developmentally regulated in contrast to Pf LDH which is consistently expressed at very high levels throughout the asexual intraerythrocytic development of the malaria parasite [49]. Transcription of Pf MDH is initiated during the ring stage, peaking at the early trophozoite stage and decreas- ing in late schizonts, while the Pf MDH protein level was equally high in trophozoites and schizonts but was markedly lower in rings. Both anabolic as well as catabolic activities peak during the trophozoite stage, subside in late schizonts and are minimal the in ring stage [49,50]. Expression of Pf MDH therefore correlates with the metabolic proﬁle of the parasite. Under normal physiological conditions the expression of Pf MDH is markedly lower than that of Pf LDH. Expression of Pf MQO, another enzyme involved in the oxidation of malate and suggested to be localized to mitochondria, was also signiﬁcantly higher as compared to Pf MDH. Recently biochemical evidence has been presented dem- onstrating the presence of a rotenone insensitive ma- late : quinone oxidoreductase in P. yoelii [5].

with very high concentration of the enzyme protein and the substrates. As protons are the product of this reaction, the free energy difference is highly positive at physiological pH. To have the reaction proceed in the direction of oxidation of malate, the ratio of OAA to malate should be very small. Malate oxidation therefore could be observed at pH > 9.5 and only with high concentrations of malate [51]. A gene encoding a unique MDH known as malate : quinone oxidoreductase (MQO) was also identiﬁed on chromosome 6 of P. falciparum (http://www.plasmodb.org). MQO cata- lyses the conversion of malate to OAA without NAD+ and uses quinone as the electron acceptor instead of NAD(P). This dramatically reduces the free energy of the reaction. MQO was originally identiﬁed in Crynebacterium gluta- micum (and subsequently reported in several other bacteria including Mycobacterium [52,53]. However in the malaria parasite Pf MQO is functional only in combination with Pf MDH, which is the sole source of malate. Pf MQO may also be part of the suggested TCA cycle in the malaria parasite. Degradation of large amounts of haemoglobin by malaria parasite during intraerythrocytic proliferation makes it highly vulnerable to oxidative damage [1]. To avoid this oxidative damage, the parasite maintains a low level of oxygen tension. Under these conditions oxidation of malate by NAD+ dependent MDH will be highly unfa- vourable. MQO can provide an alternative route for oxidation of malate by the malaria parasite.

Pf LDH has been exploited as a potential target for new antimalarial drug discovery. Pf LDH activity is inhibited by oxamic acid [14,15] ) a substrate analogue – and gossypol [18,24] which interacts with cofactor binding site of the enzyme. The lack of inhibition of Pf MDH by oxamate

The optimum pH for reduction of OAA by Pf MDH was found to be 7.0 while the optimum pH for malate oxidation was highly alkaline (pH 10.2). The saturating concentration for OAA was only 250 lM while for malate it was quite high (20 mM). Malate oxidation at physiological pH is of interest, but at physiological pH the reaction reaches steady state very soon producing only undetectable changes. At pH 7 and 7.5 no oxidation of malate could be detected, even

Fig. 9. Eﬀect of treatment of P. falciparum cultures with gossypol on comparative expression of Pf MDH, Pf LDH and Pf MQO. Highly syn- chronized cultures with approximately 15% parasitaemia were exposed to 25 and 100 lgÆmL)1 gossypol at the early trophozoites stage. The cultures were harvested after 8 h of exposure at the late trophozoites stage. Expression the enzymes was checked by (A) analysis of equal amounts of RNA by Northern blotting and also by (B) analysis of equal amounts of proteins by Western blotting. G25 and G100 indicate the P. falciparum cultures treated with 25 and 100 lgÆmL)1 gossypol, respectively. P. falciparum cultures without treatment (control) were also processed under similar conditions. The bars represent relative density of RNA/protein band as compared to controls based on quantitative image analysis.

3500 A. K. Tripathi et al. (Eur. J. Biochem. 271)

(cid:1) FEBS 2004

Glucose

Glycolysis

PEP

Pathways

Pyruvate

Oxaloacetate

X

NADH LDH MDH

MQO

NAD+

Lactate

malate

Cytosol

Mitochondria

induction of expression of Pf MDH caused by ever, treatment with a Pf LDH inhibitor indicates that besides its role in production of malate/OAA for further metabolic reactions, Pf MDH may also complement the Pf LDH function of NAD/NADH coupling reactions and regener- ation of NAD+ [14,15,40]. In the absence of complete oxidation of glucose by the malaria parasite regeneration of NAD+ is necessary for uninterrupted utilization of glucose and energy generation through glycolysis. Severe malaria infection is usually associated with hypoglycemia and lactic acidosis [3]. Despite the necessity of glycolysis for survival of the malaria parasite and the distinct molecular character- istics of the parasite enzyme, the inhibitors of Pf LDH have not yielded the expected results as potential antimalarial agents. It would be useful therefore to develop dual inhibitors of Pf LDH and Pf MDH for complete blockade of energy generation through glucose utilization in the malaria parasite. A comparative analysis of substrate and cofactor binding pockets of Pf MDH and Pf LDH would be important in the design of common inhibitors. Our results thus indicate distinct biochemical and structural character- istics of Pf MDH and also its possible importance in the energy metabolism of the malaria parasite. Detailed inves- tigations will be required to speciﬁcally pinpoint it role.

Acknowledgements

Fig. 10. The scheme indicating correlation of Pf LDH, Pf MDH and Pf MQO functions in P. falciparum. Thickness of the line/arrow indi- cates relative abundance of the enzyme/enzymatic reaction in the parasite. The broken lines/arrows show inhibition and the double lines/ arrows indicate induction of the enzyme/enzymatic reaction in the presence of gossypol the LDH/MDH inhibitor. The scheme partic- ularly depicts the role of Pf LDH in NAD/NADH coupling, which may be completed by Pf MDH under the conditions suppression of Pf LDH expression.

References

This work was supported by CDC Cooperative agreements U50/ CCU418839 and UR3/CCU418652. Partial support was also obtained from United States Department of Agriculture (USDA)-ARS, under scientiﬁc cooperative agreement no. 58-6408-20009. We are thankful to Dr Rafael Balana Fouce for his critical reading of the manuscript and useful suggestions. The information on the sequence of putative MDH and MQO was obtained from PlasmoDB (http://www.plasmodb.org), which is supported by the Burroughs Wellcome Fund.

1. Sherman, I.W. (1998) Malaria: Parasite Biology, Pathogenesis and Protection (Sherman, I.W., ed.), pp. 135–143. ASM Press, Washington DC.

2. Roth, E. Jr (1990) Plasmodium falciparum carbohydrate metabo- lism: a connection between host cell and parasite. Blood Cells 16, 453–460.

3. Holloway, P.A., Krishna, S. & White, N.J. (1991) Plasmodium berghei: lactic acidosis and hypoglycaemia in a rodent model of severe malaria; eﬀects of glucose, quinine, and dichloroacetate. Exp. Parasitol. 72, 123–133.

reconﬁrms that the substrate binding pockets of Pf LDH and Pf MDH are signiﬁcantly different. Inhibition of Pf LDH with gossypol was selective as bovine heart LDH was inhibited with much lower potency. However Pf MDH did not show much selectivity for inhibition by gossypol. Both malarial and mammalian MDHs were inhibited by gossypol with similar potency. Earlier reports also showed inhibition of oxidoreductases, including MDHs from human tissues [54] and from Trypanosoma cruzi [55], by gossypol with almost the same potency. Gossypol also inhibits growth of P. falciparum in vitro. Treatment of a highly synchronized P. falciparum cultures (at the early trophozoite stage) with gossypol did not cause any induc- tion in Pf LDH expression; rather, expression of Pf LDH in treated cultures was considerably reduced at higher doses of gossypol. Fig. 10 shows a correlation between LDH and MDH and MQO expression in the malaria parasite under normal physiological conditions and also under conditions of suppression of the expression of LDH. Almost 85% of the glucose utilized by malaria parasites is converted to lactate [1]. Therefore, constitutive expression of Pf LDH may be already at a maximum level. Gossypol treatment suppressed the expression of Pf LDH, rather than causing any induction. A concomitant induction in expression of Pf MDH in cultures treated with gossypol indicates that this enzyme is regulated in the parasite according to its physiological and metabolic requirements; it would be of interest to determine the mechanism of this regulation. Under normal physiological conditions, expression of Pf MDH is signiﬁcantly lower than that of Pf LDH. Pf MDH therefore may not have a primary role in energy generation through glucose utilization by the malaria parasite under normal physiological circumstances. How-

4. Gardner, M.J., Hall, N., Fung, E., White, O., Berriman, M., Hyman, R.W., Carlton, J.M., Pain, A., Nelson, K.E., Bowman, S., Paulsen, I.T., James, K., Eisen, J.A., Rutherford, K., Selzberg, S.L., Craig, A., Kyes, S., Chan, M.S., Nene, V., Shallom, S.J., Suh, B., Peterson, J., Angiuoli, S., Pertea, M., Allen, J., Selengut, J., Haft, D., Mather, M.W., Vaidya, A.B., Martin, D.M.A., Fair- lamb, F., Fraunholz, M.J., Roos, D.J., Ralph, S.A., McFadden, G.I., Cummings, L.M., Subramaniyan, M., Mungall, C., Venten, J.C., Carucci, D.J., Hoﬀman, S.L., Newbold, C., Davis, R.W., Frasen, C.M. & Barrell, B. (2002) Genome sequence of the human malaria parasite Plasmodium falciparum. Nature 419, 498–511. 5. Uyemura, S.A., Luo, S., Vieira, M., Moreno, S.N.J. & Doc- amopo, R. (2004) Oxidative phosphorylation and rotenone- insensitive malate- and NADH-quinone oxidoreductases in Plasmodium yoelii yoelii mitochondria in situ. J. Biol. Chem. 279, 385–393.

Malate dehydrogenase of P. falciparum (Eur. J. Biochem. 271) 3501

(cid:1) FEBS 2004

6. Minarik, P., Tomaskova, N., Kollarova, M. & Antalik, M. (2002) Malate dehydrogenases – structure and function. Gen. Physiol. Biophys. 21, 257–265.

23. Dalhus, B., Saarinen, M., Sauer, U.H., Eklund, P., Johansson, K., Karlsson, A., Ramaswamy, S., Bjork, A., Synstad, B., Naterstad, K., Sirevag, R. & Eklund, H. (2002) Structural basis for thermo- philic protein stability: structures of thermophilic and mesophilic malate dehydrogenases. J. Mol. Biol. 318, 707–721. 7. Goward, C.R. & Nicholls, D.J. (1994) Malate dehydrogenase: a model for structure, evolution, and catalysis. Protein Sci. 3, 1883– 1888.

24. Royer, R.E., Deck, L.M., Campos, N.M., Hunsaker, L.A. & Vander Jagt, D.L. (1986) Biologically active derivatives of gossy- pol: synthesis and antimalarial activities of peri-acylated gossylic nitriles. J. Med. Chem. 29, 1799–1801. 8. Musrati, R.A., Kollarova, M. & Mernik, N. & Mikulasova, D. (1998) Malate dehydrogenase: distribution, function and proper- ties. Gen. Physiol. Biophys. 17, 193–210. 25. Trager, W. & Jenson, J.B. (1976) Human malaria parasites in continuous culture. Science 193, 673–675. 9. Madern, D. (2002) Molecular evolution within 1-malate and 1-lactate dehydrogenase super-family. J. Mol Evol. 54, 825–840.

26. Lambros, C. & Vanderberg, J.P. (1979) Synchronization of Plas- modium falciparum erythrocytic stages in culture. J. Parasitol. 65, 418–420. 10. Zhu, G. & Keithly, J.S. (2002) Alpha-proteobacterial relationship of apicomplexan lactate and malate dehydrogenases. J. Eukaryot. Microbiol. 49, 255–261.

27. Elandalloussi, L.M. & Smith, P.J. (2002) Preparation of pure and intact Plasmodium falciparum plasma membrane vesicles and partial characterization of the plasma membrane ATPase. Malar. J. 1, 1–7. 11. Madern, D., Cai, X., Abrahamsen, M.S. & Zhu, G. (2004) Evo- lution of Cryptosporidium parvum lactate dehydrogenase from malate dehydrogenase by a very recent event of gene duplication. Mol. Biol. Evol. 21, 489–497.

12. Irmia, A., Vellieux, F.M.D., Madern, D., Zaccai, G., Karshikoﬀ, A., Tibbelin, G., Ladenstein, R., Lien, T. & Birkeland, N.K. (2004) The 2.9A resolution crystal structure of malate dehydro- genase from Archaeoglobus fugidus: mechanism of oligomerization and thermal stabilization. J. Mol. Biol. 335, 343–356.

28. Bradford, M.M. (1976) A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal. Biochem. 72, 248–254. 29. Muhammad, I., Dunbar, D.C., Khan, S.I., Tekwani, B.L., Bedir, E., Takamatsu, S., Ferreira, D. & Walkerm, L.A. (2003) Anti- parasitic alkaloids from Psychotria klugii. J. Nat. Prod. 66, 962– 967.

13. Irimia, A., Ebel, C., Madern, D., Richard, S.B., Cosenza, L.W., Zaccai, G. & Velliex, M.D. (2003) The oligomeric status of Haloarcula marismortui malate dehydrogenase are modulated by solvent components as shown by crustallographic and biochemical studies. J. Mol. Biol. 326, 859–873. 30. Thompson, J.D., Gibson, T.J., Plewniak, F., Jeanmougin, F. & Higgins, D.G. (1997) The CLUSTAL_X windows interface: ﬂex- ible strategies for multiple sequence alignment aided by quality analysis tools. Nucleic Acids Res. 25, 4876–4882.

14. Dunn, C.R., Banﬁeld, M.J., Barker, J.J., Hingham, C.W., More- ton, K.M., Turgut-Balik, D., Brady, R.L. & Holbrook, J.J. (1996) The structure of lactate dehydrogenase from Plasmodium falci- parum reveals a new target for anti-malarial design. Nat. Struct. Biol. 3, 912–915. 31. Luthy, R., Bowie, J.U. & Esenberg, D. (1992) Assessment of protein models with three-dimensional proﬁles. Nature 356, 83–85. 32. Godzik, A., Kolinski, A. & Skolnick, J. (1992) Topology ﬁnger- print approach to the inverse protein folding problem. J. Mol. Biol. 227, 2227–2238.

33. Gish, W. (1996-2002) http://blast.wustl.edu 34. Sabnis, Y.A., Desai, P.V., Rosenthal, P.J. & Avery, M.A. (2003) Probing the structure of falcipain-3, a cysteine protease from Plasmodium falciparum: comparative protein modeling and docking studies. Protein Sci. 12, 501–509.

15. Hewitt, C.O., Sessions, R.B., Daﬀorn, T.R. & Holbrook, J.J. (1997) Protein engineering tests of a homology model of Plasmo- dium falciparum lactate dehydrogenase. Protein Eng. 10, 39–44. 16. Gomez, M.S., Piper, R.C., Hunsaker, L.A., Royer, R.E., Deck, L.M., Makler, M.T. & Vander Jagt, D.L. (1997) Substrate and cofactor speciﬁcity and selective inhibition of lactate dehydro- genase from the malarial parasite Plasmodium falciparum. Mol. Biochem. Parasitol. 90, 235–246. 35. Hall, M.D. & Banaszak, L.J. (1993) Crystal structure of a ternary complex of Escherichia coli malate dehydrogenase citrate and NAD at 1.9 A resolution. J. Mol. Biol. 232, 213–223.

36. Kavanagh, K.L., Elling, R.A. & Wilson, D.K. (2004) Structure of Toxoplasma gondii LDH1: Active sites diﬀerences from human lactic dehydroegenases and structural basis for eﬃcient APAD use. Biochemistry 43, 879–889. 17. Winter, V.J., Cameron, A., Tranter, R., Sessions, R.B. & Brady, R.L. (2003) Crystal structure of Plasmodium berghei lactate dehydrogenase indicates the unique structural diﬀerences of these enzymes are shared across the Plasmodium genus. Mol. Biochem. Parasitol. 131, 1–10.

37. Carugo, O. (2003) How root mean square distance (r.m.s.d.) val- ues depend on the resolution of protein structure that are com- pared. J. Appl. Cryst. 36, 125–128.

18. Brown, W.M., Yowell, C.A., Hoard, A., Vander Jagt, T.A., Hunsaker, L.A., Deck, L.M., Rorey, R.E., Piper, R.C., Dame, J.B., Makler, M.T. & Vander Jagt, D.L. (2004) Comparative structural analysis and kinetic properties of lactate dehydro- genases from the four specied of human malarial parasites. Bio- chemistry 43, 6119–6229. 38. Edsall, J.T., Flory, P.J., Liquori, A.M., Kendrew, J.C., Nemethy, G., Ramchandran, G.N. & Scheraga, H.A. (1996) A proposal of standard conventions and nomenclature for the description of polypeptide conformation. Biopolymers 4, 121–129. 39. Rao, S.T. & Rossman, M.G. (1973) Comparison of super-sec- ondary structures in proteins. J. Mol. Biol. 76, 241–256. 19. Lang-Unnasch, N. (1992) Puriﬁcation and properties of Plasmo- dium falciparum malate dehydrogenase. Mol. Biochem. Parasitol. 17, 17–26.

20. Sherman, I.W. (1966) Malic dehydrogenase heterogeneity in malaria (Plasmodium lophurae and P. berghei). J. Protozool. 13, 344–349. 40. Chan, M. & Sim, T.S. (2004) Functional characterization of an alternative [lactate dehydrogenase-like] malate dehydrogenase in Plasmodium falciparum. Parasitol. Res. 92, 43–47.

41. Friedrich, C.A., Ferrell, R.E., Siciliano, M.J. & Kitto, G.B. (1988) Biochemical and genetic identity of alpha-keto acid reductase and cytoplasmic malate dehydrogenase from human erythrocytes. Ann. Human Genet. 52, 25–37. 21. Momen, H., Atkinson, E.M. & Homewood, C.A. (1975) An electrophoretic investigation of the malate dehydrogenase of mouse erythrocytes infected with Plasmodium berghei. Int. J. Biochem. 6, 533–535.

22. Bell, J.K., Yennawar, H.P., Wright, S.K., Thompson, J.R., Viola, R.E. & Banaszak, L.J. (2001) Structural analyses of a malate dehydrogenase with a variable active site. J. Biol. Chem. 276, 31156–31162. 42. Vessal, M. & Tabei, S.M. (1996) Partial puriﬁcation and kinetic properties of cytoplasmic malate dehydrogenase from ovine liver Echinococcus granulosus protoscolices. Comp Biochem. Physiol. B Biochem. Mol Biol. 113, 757–763.

3502 A. K. Tripathi et al. (Eur. J. Biochem. 271)

(cid:1) FEBS 2004

D.J. & Winzeler, E.A. (2003) Discovery of gene function by expression proﬁling of the malaria parasite life cycle. Science 301, 1503–1508. 43. Moody, A.H. & Chiodini, P.L. (2002) Non-microscopic method for malaria diagnosis using OptiMAL IT, a second-generation dipstick for malaria pLDH antigen detection. Br. J. Biomed. Sc. 59, 228–231.

50. Bozdech, Z., Zhu, J., Joachimiak, M.P., Cohen, F.E., Pulliam, B. & De Risi, J.L. (2003) The transcriptome of the intraerythrocytic developmental cycle of Plasmodium falciparum. Genome Biol. 4, R9. 44. Piper, R., Lebras, J., Wentworth, L., Hunt-Cooke, A., Houze, S., Chiodini, P. & Makler, M. (1999) Immunocapture diagnostic assays for malaria using Plasmodium lactate dehydrogenase (pLDH). Am. J. Trop. Med. Hyg. 60, 109–118.

51. Molenaar, D., van der Rest, M.E., Drysch, A. & Yucel, R. (2000) Functions of the membrane-associated and cytoplasmic malate dehydrogenases in the citric acid cycle of Corynebacterium gluta- micum. J. Bacteriol. 182, 6884–6891. 45. Makler, M.T., Ries, J.M., Williams, J.A., Bancroft, J.E., Pipe, R.R.C., Gibbins, B.L. & Hinrichs, D.J. (1993) Parasite lactate dehydrogenase as an assay for Plasmodium falciparum drug sen- sitivity. Am. J. Trop. Med. Hyg. 48, 739–741.

52. Kather, B., Stingl, K., van der Rest, M.E., Altendorf, K. & Molenaar, D. (2000) Another unusual type of citric acid cycle enzyme in Helicobacter pylori: the malate: quinone oxido- reductase. J. Bacteriol. 182, 3204–3209.

46. Chapman, A.D., Cortes, A., Daﬀorn, T.R., Clarke, A.R. & Brady, R.L. (1999) Structural basis of substrate speciﬁcity in malate dehydrogenases: crystal structure of a ternary complex of porcine cytoplasmic malate dehydrogenase, alpha-ketomalonate and tetrahydoNAD. J. Mol. Biol. 285, 703–712.

47. Wright, S.K. & Viola, R.E. (2001) Alteration of the speciﬁcity of malate dehydrogenase by chemical modulation of an active site arginine. J. Biol. Chem. 276, 31151–31155. 53. Kretzschmar, U., Ruckert, A., Jeoung, J.H. & Gorisch, H. (2002) Malate: quinone oxidoreductase is essential for growth on ethanol or acetate in Pseudomonas aeruginosa. Microbiol. 148, 3839–3847. 54. Burgos, C., Gere, Z., de Burgos, N.M., Rovai, L.E. & Blanco, A. (1986) In vitro inhibition by gossypol of oxidoreductases from human tissues. Biochem. Pharmacol. 35, 801–804.

48. Charnock, C. (1997) Structural studies of malate dehydrogenases (MDHs): MDHs in Brevundimonas species are the ﬁrst reported MDHs in Proteobacteria which resemble lactate dehydrogenases in primary structure. J. Bacteriol. 179, 4066–4070. 55. Gere, Z., de Burgos, N.M., Burgos, C., Montamat, N.E., Rovai, L.E. & Blanco, A. (1984) Inhbition by gossypol of oxidoreductases from Trypanosoma cruzi. Biochem. Pharmacol. 33, 955–959. 49. Le Roch, K.G., Zhou, Y., Blair, P.L., Grainger, M., Moch, J.K., Haynes, J.D., De La Vega, P., Holder, A.A., Batalov, S., Carucci,