Plant-like phosphofructokinase from Plasmodium falciparum belongs to a novel class of ATP-dependent enzymes

International Journal for Parasitology 39 (2009) 1441–1453

Contents lists available at ScienceDirect

International Journal for Parasitology journal homepage: www.elsevier.com/locate/ijpara

Plant-like phosphofructokinase from Plasmodium falciparum belongs to a novel class of ATP-dependent enzymes Binny M. Mony *,1, Monika Mehta 1,2, Gotam K. Jarori, Shobhona Sharma Department of Biological Sciences, Tata Institute of Fundamental Research, Homi Bhabha Road, Mumbai 400 005, Maharashtra, India

a r t i c l e

i n f o

Article history: Received 3 February 2009 Received in revised form 23 April 2009 Accepted 5 May 2009

Keywords: Plasmodium Apicomplexa Phosphofructokinase Glycolysis Enzyme Regulation

a b s t r a c t Malaria parasite-infected erythrocytes exhibit enhanced glucose utilisation and 6-phospho-1-fructokinase (PFK) is a key enzyme in glycolysis. Here we present the characterisation of PFK from the human malaria parasite Plasmodium falciparum. Of the two putative PFK genes on chromosome 9 (PfPFK9) and 11 (PfPFK11), only the PfPFK9 gene appeared to possess all the catalytic features appropriate for PFK activity. The deduced PfPFK proteins contain domains homologous to the plant-like pyrophosphate (PPi)dependent PFK b and a subunits, which are quite different from the human erythrocyte PFK protein. The PfPFK9 gene b and a regions were cloned and expressed as His6- and GST-tagged proteins in Escherichia coli. Complementation of PFK-deficient E. coli and activity analysis of purified recombinant proteins confirmed that PfPFK9b possessed catalytic activity. Monoclonal antibodies against the recombinant b protein confirmed that the PfPFK9 protein has b and a domains fused into a 200 kDa protein, as opposed to the independent subunits found in plants. Despite an overall structural similarity to plant PPi-PFKs, the recombinant protein and the parasite extract exhibited only ATP-dependent enzyme activity, and none with PPi. Unlike host PFK, the Plasmodium PFK was insensitive to fructose-2,6-bisphosphate (F-2,6-bP), phosphoenolpyruvate (PEP) and citrate. A comparison of the deduced PFK proteins from several protozoan PFK genome databases implicates a unique class of ATP-dependent PFK present amongst the apicomplexan protozoans. Ó 2009 Australian Society for Parasitology Inc. Published by Elsevier Ltd. All rights reserved.

1. Introduction Plasmodium falciparum, the causative agent of severe malaria, is estimated to cause nearly 1 million deaths annually (WHO Report, 2008). It is the intraerythrocytic stages of the parasite that lead to the major clinical manifestations of the disease. During its intraerythrocytic growth phase, the malaria parasite relies mainly on glycolysis for its energy requirements (Lang-Unnasch and Murphy, 1998; Agbenyega et al., 2000). The parasite-infected red blood cells (pRBCs) utilise glucose at a rate 50- to 100-fold higher than that of uninfected red blood cells (uRBCs) (Oelshlegel et al., 1975; Roth et al., 1982; Mehta et al., 2005). An increase in activity of several enzymes of the glycolytic pathway has also been observed in pRBCs (Roth et al., 1988). Phosphofructokinase (PFK) enzyme plays a crucial role in the regulation of glycolysis and hence it is important to evaluate the properties of this enzyme in P. falciparum. PFK (ATP-dependent; EC 2.7.1.11) is the regulatory enzyme that commits the utilisation of glucose to the glycolytic pathway. The

* Corresponding author. Tel.: +91 22 22782865; fax: +91 22 22804610. E-mail address: [email protected] (B.M. Mony). 1 These authors contributed equally to this work. 2 Present address: Department of Cell Biology, Albert Einstein College of Medicine, Bronx, NY, USA.

regulatory irreversible ATP-dependent PFK enzyme has been well-characterised in several prokaryotic and eukaryotic organisms (Bloxham and Lardy, 1973; Kemp and Gunasekera, 2002). The presence of a novel PFK that uses inorganic pyrophosphate (PPi) was reported for the first time in the protozoan amoeba Entamoeba histolytica in 1974 (Reeves et al., 1974). The PPi-dependent PFK (EC 2.7.1.90) was subsequently found to be widespread in photosynthetic plants and in several bacteria organisms (Carlisle et al., 1990; Mertens, 1993; Alves et al., 2001; Müller et al., 2001). Recently, an additional ATP-PFK enzyme activity, and the gene coding for the same, has been demonstrated in E. histolytica (Chi et al., 2001). Thus, in plants and in E. histolytica the ATP- and the PPiPFK activities co-exist. The plant PPi-PFKs are composed of two individual subunits a and b and are generally present as a hetero-tetramer or heterooctomer (a2b2 and a4b4), although enzymatically active oligomers with just the b subunit have been postulated in certain plant tissues (Yan and Tao, 1984; Wong et al., 1990; Turner and Plaxton, 2003). It has been reported through gene transcript analysis that different amounts of a and b subunits are present in different tissues and it is speculated that PPi-PFKs are regulated through differential a and b oligomerization (Wong et al., 1990; Suzuki et al., 2003). The plant PPi-PFKs are insensitive to the several allosteric regulators which modulate glycolysis in typical eukaryotic PFKs.

0020-7519/$36.00 Ó 2009 Australian Society for Parasitology Inc. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.ijpara.2009.05.011

1442

B.M. Mony et al. / International Journal for Parasitology 39 (2009) 1441–1453

However, plant PPi-PFKs exhibit allosteric behaviour towards very small concentrations of fructose-2,6-bisphosphate (F-2,6-bP) (Turner and Plaxton, 2003). There seems to be a paucity of information regarding the structure and activity of the PFK protein amongst the apicomplexan parasites. Amongst Plasmodium species, activity of PFK has been reported only in the murine model of malaria Plasmodium berghei and was shown to be ATP-dependent (Buckwitz et al., 1990). However, activity assays showed a predominant PPi dependence in the coccidian organisms Toxoplasma gondii and Cryptosporidium (Peng and Mansour, 1992; Denton et al., 1996). The Plasmodium databases, PlasmoDB (http://plasmodb.org/plasmo/) (Aurrecoechea et al., 2009) as well as the Malaria Parasite Metabolic Pathways (MPMP) database (Ginsburg, 2006) defining all the genes encoding glycolytic enzymes, annotate two PFK genes, on chromosome 9 (PfPFK9) (PlasmoDB gene ID: PFI0755c) and 11 (PfPFK11) (PlasmoDB gene ID: PF11_0294) as putative 6-phosphofructokinase and ATP-dependent PFK genes, respectively. No experimental information is available about the PFK proteins and activities in P. falciparum. In this paper we report the cloning and expression of the recombinant PfPFK9 domains as well as an analysis of the properties of the protein from the parasite extract. Despite the overall structural similarity with the PPi-dependent PFKs, the P. falciparum PFK protein exhibits ATP-dependent activity. Analysis of the deduced PFK sequences from several protozoan organisms demonstrates a novel class of ATP-dependent PFK present amongst the apicomplexan protozoans, distinct from the ATP-dependent genes of the protozoan kinetoplastids such as Leishmania and Trypanosoma (Michels et al., 1997; López et al., 2002).

2.4. Cloning and expression of PfPFK9 constructs in Escherichia coli The PFKa, PFKb and PFKbD fragments of 1.93, 1.98 and 1.0 Kb, respectively, were amplified by PCR from P. falciparum genomic DNA using primers in which restriction enzyme sites were included to facilitate directional cloning. The primers used were: 0

0

(1) PFKa F15 CAGT GCATGC ATGTCCGAATTACAAAC3 and PFKa 0 0 R1.5 CAGT GTCGAC GTTCATTCTTTTTCTCTG3 for PFKa cloning in the pQE vector. 0 0 (2) PFKb F15 GCCGGCATGCATGGATACCAAGAGTGG3 and PFKb 50 30 R1. GCCGCTGCAGCATATTCTTGTTAGTTAG for PFKb cloning in the pQE vector. 0 0 (3) PFKb F25 GCCGGTCGACATGGATACCAAGAGTGG3 and PFKb 50 30 R2. GCCGGCGGCCGCCTTGTTAGTTAG for PFKb cloning in the pGEX4T3 vector. 0 0 (4) PFKb F35 GCGGTCGACGGTGGTCCAGCTCCTGG3 and PFKb 0 0 for R3.5 GCCGGCGGCCGCAGCACATCTACCTTCATAACC3 PFKbD cloning in the pGEX4T3 vector. The PCR products were cloned in the expression vector pQE30 in the SalI and SphI (PFKa) and SphI and PstI sites (PFKb) and in the pGEX4T3 vector in the SalI and NotI sites (PFKb and PFKbD). In both constructs the tags were N-terminal to the PFK protein domains. The constructs were confirmed by sequencing. These constructs were transformed into E. coli strain BL21DE3. Transformed bacterial cultures were grown in Luria–Bertani media supplemented with appropriate antibiotics to A600 of 0.6 at 37 °C, and expression was induced with 0.3 mM Isopropyl b-D-1-thiogalactopyranoside (IPTG) at 28 °C for 5 h.

2. Materials and methods

2.5. Affinity purification of recombinant PfPFK9 protein domains

2.1. Plasmodium falciparum culture

For the His6-tagged proteins, cells from induced cultures were pelleted, resuspended in the His-purification base buffer (20 mM HEPES (pH 7.5), 500 mM NaCl, 20 mM imidazole, 5 mM MgCl2, 20 mM b-mercaptoethanol, 10% glycerol) with 1% Triton X-100 and bacteria protease inhibitors (Pepstatin, Leupeptin and phenylmethanesulphonylfluoride). For the GST-tagged proteins, the cell pellets from induced cultures were resuspended in the GST-purification base buffer (50 mM Tris–Cl (pH 8.0), 1 mM EDTA, 5 mM MgCl2, 150 mM KCl, 1 mM DTT and 25% glycerol) with 1% Triton X-100 and bacteria protease inhibitors. The cells were lysed by sonication and the cellular debris removed by high speed centrifugation. Briefly, the supernatant, containing the tagged protein was incubated with Ni–NTA agarose beads (Qiagen, Germany) for the His6-tagged proteins or Glutathione-agarose beads (Sigma, USA) for GST-tagged proteins at 4 °C for 1 h. The beads were washed with the His- or GST-purification base buffer and the bound proteins were eluted with 250 mM imidazole or 5 mM glutathione, respectively.

Parasites were maintained in culture as described earlier (Goswami et al., 1997). Human blood, from healthy adults with O+ blood group, was collected in acid citrate dextrose as the anticoagulant. After removing the leukocytes, the erythrocytes were washed and resuspended in complete RPMI (RPMI with 0.5% Albumax). Asexual stages of P. falciparum (3D7 strain) were cultured in vitro and maintained at 5% haematocrit in complete RPMI at 37 °C in a humidified chamber containing 5% CO2. 2.2. Plasmodium falciparum genomic DNA preparation pRBCs (10% parasitemia) were taken and washed with cold PBS. The RBCs were lysed by incubating with 0.05% saponin in PBS at 37 °C and the DNA was extracted using standard molecular methods (Schichtherle and Wahlgren, 2008). The DNA was visualised on 0.8% agarose gel to ensure that the preparation was unsheared and free of RNA. 2.3. Database searches and sequence analysis

2.6. Antibodies against PfPFK9

The PFK sequences of plants (Ricinus communis, Arabidopsis thaliana, Solanum tuberosum, Oryza sativa), human (Homo sapiens), yeast (Saccharomyces cerevisiae), spirochaetes (Borrelia burgdorferi, Treponema pallidum), protozoa (T. gondii, Leishmania donovani, Trypanosoma brucei, Theileria parva, Cryptosporidium parvum, E. histolytica), were obtained from the NCBI database (http:// www.ncbi.nlm.nih.gov/). The sequences and the expression data of the Plasmodium genes were obtained from PlasmoDB (Aurrecoechea et al., 2009). Sequences were aligned and analyzed for similarity using Biology WorkBench (http://seqtool.sdsc.edu/CGI/ BW.cgi/) or ClustalW (http://www.ebi.ac.uk/clustalw/).

Recombinant purified protein His6-PfPFKb was used for immunization of animals to raise the monoclonal antibodies. The hybridomas were generated with the service of Bioklone, Chennai, India. Briefly, about 50 lg of purified His6-PfPFKb emulsified with FCA was administered i.p. into 6 weeks old, female BALB/c mice. After four weekly injections, the mice were immunized monthly for two months. Two sets of fusions were carried out, of which 12 clones showed reactivity to His6-PfPFKb. Supernatants from these clones were tested for reactivity against the respective proteins using ELISA and the clone 2F4, which reacted with PfPFKb but not PfPFKa, was selected for subcloning.

B.M. Mony et al. / International Journal for Parasitology 39 (2009) 1441–1453

2.7. Immunoblotting Protein samples were prepared by boiling in SDS gel loading buffer for 10 min. To prepare Plasmodium samples, the pRBCs from an asynchronous culture were treated with 0.15% saponin in PBS for 10 min at 37 °C to liberate the parasites. The parasite cell pellet was washed at least three times with PBS or until there was no visible trace of haemoglobin in the supernatant. The parasite pellets were lysed in reducing sample buffer (50 mM Tris-pH 6.8, 100 mM DTT, 2% SDS, 0.1% bromophenol blue, 10% glycerol) and boiled for 10 min. The prepared samples were then run on SDS– PAGE gel of appropriate polyacrylamide concentration and transferred to methanol-activated Polyvinylidene Fluoride (PVDF) membrane (Millipore) in western trans-blot buffer using TransBlot Semi Dry Transfer Cell (Bio-Rad, USA). Membranes were blocked with 5% non-fat skim milk powder in PBS for 1 h and then probed with the desired antibodies. Primary antibody binding was detected by appropriate secondary antibodies conjugated to horseradish peroxidase (BD Biosciences, USA). The immunoblots were developed using 3,30 -Diaminobenzidine (Sigma) system, or the ECL PlusTM (Amersham) system. 2.8. Immunofluorescence assay (IFA) IFA was performed with P. falciparum (3D7 strain) as described earlier (Singh et al., 2002). The smears were fixed for 5 min using chilled methanol and permeabilized by 15 min of treatment with 0.05% saponin in PBS. Saponin (0.01%) in PBS was used as the washing buffer for the permeabilized IFA. The cells were blocked with 3% BSA in PBS for 1 h, followed by treatment with the monoclonal antibody culture supernatant. Secondary antibodies conjugated to Alexa-FluorÒ 488 (Molecular Probes, Invitrogen) were used at 1:2,000 dilution. The smears were treated with DAPI (0.1 lg/ml) and were observed under an Axioplan Zeiss Epifluorescence microscope. 2.9. Enzyme activity assays The ATP-PFK activity was followed spectrophotometrically by linking it to the oxidation of NADH to NAD+ in the presence of an excess of auxiliary enzymes (Beutler, 1984). The assay buffer was composed of 0.1 M Tris–HCl (pH 7.4), 0.5 mM EDTA (pH 8), 1 mM MgCl2, 90 mM KCl, 1 mM DTT, 0.2 mM NADH, excess of the auxiliary enzymes-aldolase (200 lg), triose phosphate isomerase (40 lg), glycerol 3-phosphate dehydrogenase (80 lg) (Boehringer Manheim) and 1 mM of one of the substrates-fructose 6-phosphate (F6P) or ATP in a total volume of 0.5 ml. After adding the cell lysate (total protein concentration 50 lg) or recombinant protein (pure protein concentration 4 lg), the decrease in absorbance was observed for 5 min at 340 nm to measure the background rate of conversion of NADH to NAD+. The reaction was then started by adding 1 mM of the missing substrate, typically F6P. The decrease in NADH was measured at 340 nm for 5 min with a Lambda 40 UV/ VIS Spectrophotometer (Perkin Elmer Instruments). One unit of enzyme activity is defined as the amount of enzyme that would convert 1 lmole of F6P to product in 1 min (Massey and Deal, 1973). The effects of ADP, AMP, phosphoenolpyruvate (PEP), citrate, F-2,6-bP and Mg2+ on the enzyme preparations were checked at concentrations of the modulators ranging from 0.05 to 1 mM, with the exception of F-2,6-bP, where concentrations ranged from 1 to 11 lM and Mg2+, where 0.5–3 mM were used. The enzyme preparation was incubated with the assay buffer and modulator for 5 min at 37 °C, after which the reaction was started by the addition of 1 mM F6P. The activity was measured over the linear range within 3–5 min. The commercial ATP-PFK from rabbit muscle (Boehringer Mannheim) and also Bacillus stearothermophillus

1443

(BsPFK) (kind gift of Prof. Gregory D. Reinhart, Texas A&M University, USA) were used as the positive controls for enzyme as well as modulator assays. The PPi-PFK activity was assayed as previously described (Van Schaftingen et al., 1982). The commercial PPi-PFK from potato tubers (Sigma) was used as a positive control. For statistical analysis of the results, the means of experimental groups were compared using one-way ANOVA (GraphPad InStat, San Diego, CA), and statistical significance was determined at P < 0.05. 2.10. Complementation of E. coli DF1020 Escherichia coli DF1020 strain (F-, pro-82, glnV44(AS), LAM-, DpfkB201, recA56, endA1, D(rhaD-pfkA)200, thi-1, hsdR17) was obtained from the E. coli Genetic Stock Center, Yale University, USA. The cells were transformed with pQE30, PFKb/pQE30 and PFKa/ pQE30 by the simple and efficient (SEM) method (Inoue et al., 1990). The primary inoculum of the transformants was made in LB broth and the secondary inoculum in M9 medium, each containing 100 lg/ml of ampicillin and supplemented with 0.4% glucose. The culture was induced by the addition of 0.3 mM IPTG at the time of inoculation and growth was monitored for 30 h at 37 °C by recording A600 every 4 h. Escherichia coli DF1020 transformed with pQE30 served as the negative control. The PFK gene from B. stearothermophillus (BsPFK) (kind gift of Prof. Gregory D. Reinhart, TAMU, USA) was used as the positive control. The complementation study was similarly performed using E. coli RL257 (F-, [araD139], lacIp-4000(lacIQ), e14-, pfkB205(del-ins)::FRT, flhD5301, D(fruK-yeiR)725(fruA25), relA1, rpsL150(strR), rbsR22, pfkA203(del-ins)::FRT, D(fimB-fimE)632(::IS1), deoC1). The cells were harvested at the end of 30 h and the enzyme activity of the crude extract was assayed spectrophotometrically. 2.11. Phylogenetic tree construction and analysis The integrated phylogenetic analysis software, Bosque (Ramirez-Flandes and Ulloa, 2008) was used as the platform for analysis. For phylogenetic tree construction and inference, the PhyML algorithm using WAG as the substitution matrix was used. 3. Results 3.1. Deduced structures of the PfPFK genes Multiple sequence analysis of the deduced PFK genes from apicomplexan parasite genomes revealed that in each of the coccidian organisms T. gondii, C. parvum and Cryptosporidium hominis, two putative PFK genes are present, while only one putative PFK gene was detected for each of the piroplasmic T. parva, Theileria annulata and Babesia bovis. A comparison between the deduced open reading frames (ORFs) from apicomplexan parasites and representative samples from other species was compiled. Fig. 1 shows the overall sizes and the approximate homologous positions of PFK genes from several organisms with respect to the P. falciparum PfPFK9 gene. Escherichia coli possesses the smallest PFK gene, while each of the apicomplexan organisms possesses large ORFs for the putative PFK genes. The apicomplexan single large ORFs contain the conserved domains orthologous to the putative plant PPi-PFK a and b gene subunits. Plants possess several paralogues of ATP- and PPi-PFK genes (Bapteste et al., 2003; Mustroph et al., 2007). A representative PPi-gene for each of the b and a subunit from A. thaliana is shown in Fig. 1. It has been documented earlier that the protozoan kinetoplastid organisms such as Trypanosoma and Leishmania possess single genes for PFK (Michels et al., 1997; López et al., 2002). These and the two E. histolytica PFK proteins are comparable in size to the b subunit of the plant PPi-PFKs (Deng et al.,

1444

B.M. Mony et al. / International Journal for Parasitology 39 (2009) 1441–1453

(1570)

Pf PFK11 TgPFKI

(1225)

CpPFKI

(1327)

TgPFKII

(1320) (1426)

CpPFKII TpPFK

(1692)

BbPFK

(1337)

Pf PFK9

(1418) Pf PFKβ (661)

Pf PFKα (641)

At PFKβ (585)

At PFKα (614)

AtPFK TbPFK

(487)

HsPFK EcPFK

(784) (320)

Fig. 1. Schematic showing an orthologous-positional comparison of the deduced phosphofructokinase (PFK) protein structures present in different organisms with respect to the PFK9 protein of Plasmodium falciparum. The apicomplexan orthologues and the similar plant (Arabidopsis) pyrophosphate PFK regions are shown as , while the thin lines and indicate the approximate length of the gaps or extra insert sequences in the apicomplexan PFK sequences, respectively. The two regions of PfPFK9 that were cloned and expressed (PfPFKb and PfPFKa) are indicated with double-arrowheads. The ATP-dependent PFK genes are shown for Trypanosoma, human and Escherichia coli. The size (in number of amino acids) of each PFK protein/subunit is shown in parenthesis. Toxoplasma gondii TgPFKI: EEA99208, ToxoDB gene ID: 42.m00123; TgPFKII: EEA99567, ToxoDB gene ID: 49.m03242; Cryptosporidium parvum (Cp PFKI: XP_626715, Cp PFKII: XP_626418); Theileria parva (TpPFK: EAN32860); Babesia bovis (BbPFK: XP_001610135); P. falciparum (Pf PFK9: CAD51837, PlasmoDB gene ID: PFI0755c, Pf PFK11: (AAN35878, PlasmoDB gene ID: PF11_0294); Arabidopsis thaliana (AtPFKb: Q3E7M8, AtPFKa: BAF01310); Trypanosoma brucei (Tb: AAZ10361); Homo sapiens (Hs: Platelet PFK: Q01813); E. coli (Ec: P0A797).

1998). The human PFK genes are an intermediate size, with the regulatory domain fused to the catalytic one (Kemp and Gunasekera, 2002), and the sequence homology between the apicomplexan genes and the human genes was found to be only about 15%. The two special features noticed in the putative apicomplexan PFK genes are: (i) the fusion of a and b domains to produce larger units; and (ii) the presence of additional protein regions inserted within the ORFs (Fig. 1). The Plasmodium database predicts no introns for the PfPFK9 gene, which was verified through PCR amplification and sequencing from the genomic DNA and a cDNA library. Sequence of PfPFK9 regions from the cDNA library made from NF54 strain showed differences from the 3D7 strain sequence at positions 2522 and 2855 (both A to G), corresponding to amino acid changes in positions 841 (D to G) and 952 (N to S) both in the a region (Supplementary Fig. S1). Sequence alignment of the apicomplexan PFK gene orthologues showed that they are closer to the plant PFP (PPi-PFK) genes (Supplementary Fig. S2). A larger collation of protozoan sequences showed that although the kinetoplastid PFK genes are generally compared with the E. histolytica PPi-PFK gene, kinetoplastid PFK genes cluster specifically with the EhPFKII (ATP-dependent) rather than the PPi-dependent EhPFKI gene (Supplementary Fig. S3). The PfPFK11 gene ORF possesses orthologues in Plasmodium vivax, P. berghei and Plasmodium yoelii. However, a comparison of the important ATP/PPi, fructose 6-phosphate (F6P) and magnesium binding sites required for PFK activity (Carlisle et al., 1990) showed that neither the PfPFK11 gene nor its orthologues from other Plasmodium species contain these conserved features (Supplementary

Fig. S3 and Table 1). Thus, it is unlikely that PfPFK11 protein or its orthologues in other Plasmodium species would be able to function as a PFK enzyme. The specificity of the PFK protein towards preferential catalysis of PPi versus ATP as the phosphate donor has been assessed earlier and has been assigned mainly to the GGDD(PPi)/GGDG(ATP) and the KTIDG(PPi)/GSIDN(ATP) motifs (Fig. 2) (Shirakihara and Evans, 1988; Deng et al., 1999; Chi and Kemp, 2000). Through in vitro mutagenesis of the PPi-PFK gene of E. histolytica, it was shown that amino acids D/G at the fourth position of the GGDD/G motif and the starting K/G of the K/GT/SIDG/ND motif determined unambiguously as to whether an overall PPi-PFK protein will exhibit ATP- or PPi-dependent PFK activity (Chi and Kemp, 2000). However, the deciding amino acid appears to be the D/G of the GGDD/G motif, based on the current alignment (Fig. 2). The single mutation from D to G changed the preference from PPi to ATP by about 107-fold. Chi and Kemp (2000) also reported that mutating the corresponding reciprocal mutations in E. coli ATP-PFK sites from G to D eliminated the ATP-dependent activity, but did not exhibit the PPi-dependent PFK activity. Thus, with the GGDG motif, we postulated that Plasmodium PfPFK9 enzyme would show a substrate preference for ATP. In order to verify the same, the PfPFK9 gene fragments were cloned and expressed. 3.2. Cloning, expression and characterisation of the PfPFK9 gene domains Two fragments (1.98 and 1.93 Kbp) of the PfPFK9 gene, homologous to the plant b and the a subunits (Fig. 1), were amplified

1445

B.M. Mony et al. / International Journal for Parasitology 39 (2009) 1441–1453 Table 1 Comparison of catalytic residues of parasitic phosphofructokinases (PFKs). Length (aa)

Borrelia burgdoferi PFK I B. burgdoferi PFK II Entamoeba histolytica PFK I E. histolytica PFK II Trypanosoma cruzi PFK Trypanosoma brucei PFK Leishmania donovani PFK Leishmania major PFK Cryptosporidium parvum PFK I C. parvum PFK II Cryptosporidium hominis PFK I C. hominis PFK II Theileria parva PFK Theileria annulata PFK Babesia bovis PFK Toxoplasma gondii PFK I T. gondii PFK II Plasmodium falciparum PFK9 Plasmodium vivax PFK9 ho Plasmodium yoelii PFK9 ho Plasmodium berghei PFK9 ho Plasmodium knowlesi PFK9 ho Plasmodium chabudi PFK9 ho P. falciparum PFK11 P. vivax PFK11 ho P. yoelii PFK11 ho P. berghei PFK11 ho P. chabudi PFK11 ho

555 447 546 439 485 487 486 486 1,327 1,426 1,328 1,426 1,692 1,725 1,337 1,225 1,399 1,418 1,421 1,310 1,252 1,417 1,208 1,570 1,347 NA NA NA

Mg2+

PPi/ATP

Fructose-6-phosphate

D 175

N 177

K 201

G 205

D 174

D 206

D 204

M 249

E 310

Y 420

R 423

D G D G G G G G D G D G G G G D G G G G G G G S D D D NA

N Q N L Q Q Q Q N N N N N N N N N N N N N N N N Q S S NA

K K K K K K K K K K K K K K K K K K K K K K K T K T T K

A N G N N N N N G G G G G G G G G G G G G G G N N N N N

D D D D D D D D D D D D D D D D D D D D D D D V V V V NA

D D D D D D D D D D D D D D D D D D D D D D D E E E E E

D D D D D D D D D D D D D D D D D D D D D D D Y C H H H

M M M M M M M M M M M M M M M M M M M M M M M M I I V V

E E E E E E E E E E E E E E E E E E E E E E E D E E NA E

Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y K K N N N

R R R R R R R R R R R R R R R R R R R R R R R T T T T T

Residue numbers refer to their positions in E. histolytica PFKI. Plain text indicates the presence of residues which are identical to those found in E. histolytica pyrophosphate (PPi)-PFK (PFKI), while the bold text denotes ATP-PFK-specific residues. NA, complete sequence not available for analysis; aa, amino acid.

and cloned in an His6-tag expression vector. The recombinant fusion proteins were purified using an Ni–NTA column and observed to be at the expected sizes (75 and 73 kDa) in the SDS–PAGE (Fig. 3A). Monoclonal anti-PfPFKb antibodies were generated, amongst which 2F4G10 was specific for PfPFK9b and did not cross-react with PfPFKa (Fig. 3A). However, it did recognise the orthologues of sizes 195 and 190 kDa from the rodent malaria parasites P. yoelii and P. berghei, respectively (Fig. 3B). The P. falciparum PfPFK9 appeared to be considerably labile and could not be detected even when the P. falciparum total protein content was comparable with that of P. berghei (Fig. 3B). However, using the more sensitive ECL technique, PfPFK9 could be detected at a size of 210 kDa (Fig. 3C). Stability of purified and recombinant PPiPFK protein is generally a problem in many eukaryotes (Turner and Plaxton, 2003). In a recent study reporting the localisation of 18 (mainly glycolytic) enzymes of T. gondii, the authors were able to generate recombinant protein for 14 of the enzymes, but for the two PFK proteins, they were able to express only the amino-terminal half of the ATP-PFK (Fleige et al., 2007). Nevertheless, the absence of any lower protein bands in the immunoblots showed that the Plasmodium PFK proteins are synthesised as a whole and are not cleaved to produce a and b domains (Fig. 3B and C). The observed sizes of 210, 195 and 190 kDa are somewhat larger than the monomeric sizes of 159, 147 and 141 kDa as deduced for P. falciparum, P. yoelii and P. berghei, respectively, from the DNA sequence. Such discrepancies are often found with large proteins that are not as globular as the SDS–PAGE markers. Results of the immunofluorescence assays of the asexual stages (rings, trophozoites and schizonts) of P. falciparum using antiPfPFKb monoclonal antibodies are shown in Fig. 3D. The antibodies showed no cross-reactivity with uRBCs. PFK was localised mainly

in the cytoplasm and unlike enolase of P. falciparum (Pal-Bhowmick et al., 2007), no staining was observed in the nucleus. In several stages, especially in the trophozoite stages, punctate staining was observed, suggesting the presence of particulate complexes. 3.3. Assay of enzyme activity of the recombinant and native PfPFKs Although in plants it is assumed that the b subunit is catalytic and a is regulatory, no cloning and recombinant protein study has been carried out to prove this. We carried out a complementation test, transforming E. coli strain DF1020, a PFK deletion mutant with PfPFKb and a constructs. The deletion mutant was leaky and even the cells transformed with the vector grew slowly. However, a definitive growth advantage to cells containing PfPFK9b was observed indicating the b domain to be the more active domain (Fig. 4A). A complementation study was also performed with a tighter mutant, E. coli RL257, in which both pfkA and pfkB genes were deleted (Lovingshimer et al., 2006). The transformants showed no growth in minimal media, while the positive control, B. stearothermophillus PFK (BsPFK) showed a normal growth pattern. This suggested that the Plasmodium PFK gene domains are either not appropriately structured to function efficiently in a prokaryotic system or that these domains were not sufficient for the catalytic function. Therefore the PFK enzyme activities of the purified recombinant proteins were tested. The purified recombinant His-tagged PfPFK b domain showed consistent but low enzyme activity. In order to obtain better folded domains of the protein, the b domain and a smaller bD region were expressed as GST-fusion proteins and purified using glutathioneagarose chromatography (Fig. 4B and C). A comparison between enzyme activities of the purified His6-and GST-tagged-recombi-

1446

B.M. Mony et al. / International Journal for Parasitology 39 (2009) 1441–1453

Fig. 2. Multiple phosphofructokinase (PFK) protein sequence alignment showing the phosphate donor specificity-determining residues in the PFKs from different organisms, marked by . The conserved GGDD (pyrophosphate (PPi)-PFK)/GGDG(ATP-PFK) and the KTIDG(PPi-PFK)/GSIDN(ATP-PFK) motifs are indicated by red bars. The Plasmodium falciparum PFK, PfPFK9, sequence is marked by an arrowhead.

nant proteins showed a distinct ATP-dependent activity in the b domain (Fig. 4D). The bD domain, containing all the active site residues, showed maximum activity amongst the recombinants, but it also showed the largest variations in different protein preparations. The His6-tagged b-pQE protein showed lower activity but with less variation (Fig. 4D). The His6-tagged a-pQE protein showed the least activity, which was still significant compared with the control purified GST protein. The recombinant PFK proteins were also assayed with PPi as substrate but they did not show any detectable activity. These results suggest that the PfPFK9b domain is the catalytic subunit which uses ATP as substrate, and that it could function catalytically as an isolated domain. The PFKa domain also exhibited a very small, but distinct, amount of catalytic activity. A comparison of the substrate binding sites showed that the PFKa domain possessed most of the magnesium binding sites, but do not possess some of the canonical phosphate binding sites. In purified PPiPFK activities in plants there is no unequivocal demonstration of the presence of an active a subunit, but the regulatory subunit a

in yeast has been studied extensively (Klinder et al., 1998; Bär et al., 2000). It has been shown that in vivo, cells with single null mutations of the PFK1 (a) and PFK2 (b) genes are both viable and show glycolytic activity (Klinder et al., 1998), indicating that both subunits were capable of catalysis. However purification of just the active regulatory a subunit was very difficult and was possible only through severe denaturation and refolding steps (Bär et al., 2000). This indicated that in isolation, the conformation of the a-subunit is not stably oriented towards catalytic activity. The activity results demonstrate that in Plasmodium the b domain constitutes the catalytic domain, while the regulatory properties might be conferred on PfPFK9 by the a domain. We were not able to express the entire PfPFK9 in E. coli, nor did we observe any differentially enhanced or modulated activity by mixing the purified recombinant PfPFK9b and PfPFK9a proteins. We therefore assayed both the P. falciparum whole protein extract, containing the fused b and a domains in the native form, as well as the purified recombinant b-GST protein, for their kinetic properties and sensitivity to various modulators (Figs. 5 and 6, Tables 2 and 3).

B.M. Mony et al. / International Journal for Parasitology 39 (2009) 1441–1453

1447

diates have not been detected in the RBC stages of Plasmodium, some of the enzymes do exist, and it has been speculated that other stages of Plasmodium may have an operational TCA cycle (Mather and Vaidya, 2008). A recent study on the characterisation of the pyruvate kinase (PyK) enzyme from P. falciparum reported a definitive inhibitory effect of citrate on the PyK enzyme (Chan and Sim, 2005). Therefore we also assessed the effect of citrate on Plasmodium PFK activity. However, unlike PyK, citrate did not show any effect on PfPFK activity. We have earlier reported the pH sensitivity of P. falciparum activity (Mehta et al., 2005), and similar to the P. berghei enzyme, it had shown a lower sensitivity to pH changes as opposed to the extreme sensitivity of the RBC enzyme to an acidic environment (Mehta et al., 2005; Buckwitz et al., 1988). Thus the P. falciparum enzyme was insensitive to several PFK modulators and exhibited kinetic properties similar to the partially purified preparation of the enzyme from P. berghei (Buckwitz et al., 1988, 1990). A more extensive kinetic analysis of the native and recombinant proteins will provide further insights into the allosteric nature of the P. falciparum enzyme. 3.4. Evolutionary relationship between protozoan PFKs

Fig. 3. Immunoblot analysis of phosphofructokinase of Plasmodium falciparum and its cellular localisation (A) Coomassie stained SDS–PAGE and immunoblot analysis of purified His-tagged PfPFKb (the b domain of the P. falciparum phosphofructokinase) (3 lg) and PfPFKa (the a domain of the P. falciparum phosphofructokinase) (1 lg) proteins using monoclonal antibody (mAb) 2F4G10 supernatant. (B) Coomassie stained SDS–PAGE and immunoblot analysis of cell extracts of P. falciparum (F) (30 lg), Plasmodium yoelii (Y) (10 lg) and Plasmodium berghei (B) (30 lg), using mAb 2F4G10 developed using the 3,30 -Diaminobenzidine (DAB) system. (C) Immunoblot of cell extract of P. falciparum (30 lg) developed using the ECL PlusTM (Amersham) system. (D) Immunofluorescence assay of fixed P. falciparum asexual rings (R), trophozoites (T), schizonts (S) and secondary antibody control (C). Each panel shows identical fields stained with DAPI (a), anti-PfPFKb mAb 2F4G10 supernatant (b) and a merge of the two (c). The control panel (C) shows P. falciparum rings stained with DAPI (a) and the secondary antibody (b).

Similar to the recombinant protein, only ATP-dependent and no PPi-dependent activity could be detected with the parasite preparation. The Km values for the PFK activity in the parasite extract were found to be 0.02 ± 0.007 mM for ATP and 0.014 ± 0.003 mM for F6P (Table 2). The recombinant b-GST protein also showed similar Km values for these substrates (Table 2). These values are comparable with those obtained for the P. berghei PFK and exhibit five- to 10-fold higher affinities compared with human RBC PFK values (Layzer, 1975; Buckwitz et al., 1988, 1990). Also, similar to the P. berghei enzyme, the P. falciparum enzyme showed biphasic behaviour towards ATP and was inhibitory at concentrations >1 mM ATP, while with F6P as the substrate, a hyperbolic activity was observed (Fig. 5). Although the b-GST protein showed very similar behaviour with F6P, at high concentrations of ATP the activity was extremely variable from preparation to preparation (data not shown), making it difficult to interpret the b-GST activity at higher ATP concentrations. The effects of other known allosteric modulators of PFK enzyme, such as ADP, citrate, PEP and F-2,6-bP (Buckwitz et al., 1990; Uyeda, 1979) on native and recombinant PFK activities were studied (Fig. 6 and Table 3). It was observed that activity remained unaffected in the presence of most of these modulators with the exception of ATP and ADP. These were found to be inhibitory at concentrations >1.0 mM for ATP and >0.1 mM for ADP (Figs. 5 and 6). Thus the P. falciparum enzyme definitely exhibited allosteric behaviour for ATP and ADP, although it was insensitive to PEP, citrate and F-2,6-bP. There has been debate as to whether a Krebs cycle operates in Plasmodium species. Although TCA cycle interme-

PFK is an important element of the glycolytic pathway, which exists in the eukaryotic, prokaryotic and archaeal life forms, and it has therefore been argued that the evolution of this protein ought to be quite constrained. However, studies of the PFK protein and the deduced genes from several organisms show quite the reverse (Müller et al., 2001; Bapteste et al., 2003). The most recent phylogenetic tree shows as many as seven subgroups of PFK (Bapteste et al., 2003). In this analysis, the apicomplexan organisms, certain protozoan organisms and plant genes were placed together in the ‘LONG’ clade. This analysis did not include several apicomplexan PFK genes such as B. bovis, T. gondii, C. hominis, T. parva and T. annulata. With a search through the current genome database for these organisms, we could identify two PFK genes each for T. gondii, C. parvum and C. hominis, while only one each from B. bovis, T. parva and T. annulata were found. A phylogenetic tree with 39 sequences of PFK genes, mainly from the apicomplexan parasites together with some representatives of the other groups is shown in Fig. 7. We have not included several bacteria and archaebacteria spp. which constitute the groups III, B2 and SHORT clade (Müller et al., 2001; Bapteste et al., 2003). We found that the ATP-dependent enzymes cluster together in three categories, two of which are canonical eukaryotic (E) and bacteria (B) sequences, and one is the X-category (Müller et al., 2001) of the ATP-dependent protist or plant PFKs. It is the LONG clade that gets clustered into two clades, that of PPi-dependent and the ATP-dependent genes (Fig. 6). The branching pattern is supported by high bootstrap values. It is observed that each of the apicomplexan organisms possesses at least one PFK, which clusters with PfPFK9 with key features of GGDG175, a sequence that signifies ATP-dependent activity (Chi and Kemp, 2000). Whenever two PFK genes are observed in the genome, these appear to split into PPi- and ATP-dependent structures, e.g. T. gondii, C. parvum and C. hominis. The large sizes of the proteins of the apicomplexan PFKs, PPi- or ATP-dependent proteins, separate them from PFKs from any other organism. 4. Discussion The results presented in this paper demonstrate that the b domain of the P. falciparum PFK gene present on chromosome 9 codes for an active ATP-dependent PFK activity. The a domain, which shows greater homology with the regulatory a-subunit of plant PPi-PFKs, showed marginal activity. Unlike the plant PFKs, where

1448

B.M. Mony et al. / International Journal for Parasitology 39 (2009) 1441–1453

Fig. 4. Experiments demonstrating the b region of phosphofructokinase from Plasmodium falciparum to be the active domain (A) Growth curve of Escherichia coli DF1020 strain transformed with PfPFKb, (the b region of the P. falciparum phosphofructokinase), PfPFKa (the a region of the P. falciparum phosphofructokinase) or pQE30 (His-tag vector) plasmids, represented as the percentage of PfPFKb-containing cells after 30 h of growth. Data shown is averaged over three independent experiments. (P < 0.05). (B) Depiction of the positions and sizes of various recombinant PfPFK9 protein regions and the predicted conserved PFK domain (information obtained from the Conserved Domains Database) (Marchler-Bauer et al., 2005) with respect to the PfPFK9 gene. (C) Coomassie stained SDS–PAGE showing purified His-tagged recombinant proteins PfPFKb (75 kDa) and PfPFKa (73 kDa) (a) and GST-tagged recombinant proteins PfPFKb-GST (101 kDa) and PfPFKbD-GST (64 kDa) (b). (D) Specific activities of the His-tagged (b-pQE and a-pQE) and GST-tagged (b-GST and bD-GST) PfPFK recombinant proteins. Purified GST (pGEX) served as the control protein. Data from three protein preparations were used to calculate the average specific activity. (P < 0.05; P < 0.001).

the a and the b domains constitute separate subunits coded by separate genes, in Plasmodium species the PfPFK9 gene orthologues contain the two domains fused and the entire large protein is expressed in the asexual stages. Plant PFK proteins have been characterised mainly through biochemical purification and no heterologous expression results are available. Very recently seven ATP-PFK proteins of A. thaliana have been characterised through heterologous cloning and expression (Mustroph et al., 2007; Winkler et al., 2007). However, no data exists for any plant PPi-PFK subunits to demonstrate unequivocally that the b domain is the catalytic domain. To our knowledge, the P. falciparum a and the b

domains thus form the first of such PPi-PFK orthologues to be expressed in a heterologous system where we demonstrate that the b domain is indeed the catalytic domain. Amongst all genomic databases assessed to date, the PFK ORFs are much larger in size in apicomplexan organisms. In Plasmodium species the existence of a large PFK protein of about 200 kDa is demonstrated in this paper. It has been hypothesised that gene fusion leading to the formation of multi-domain proteins is a major route of protein evolution, as it provides a mechanism for the physical association of different catalytic domains or catalytic and regulatory structures (Yanai et al., 2002). Mammalian PFK is itself an

1449

B.M. Mony et al. / International Journal for Parasitology 39 (2009) 1441–1453

A

B 1.2

1.0

1.0 0.8

V/Vmax

0.8

V/Vmax

0.6 0.4

0.6 0.4

0.2

0.2 0.0

0.0 0.0

0.5

1.0

1.5

2.0

2.5

3.0

0.0

0.5

1.0

1.5

2.0

ATP Concentration (mM)

F6P Concentration

Fig. 5. Dependence of phosphofructokinase activity from Plasmodium falciparum cell extracts on (A) F6P and (B) ATP concentrations. Assays were performed as described in Section 2 at a fixed concentration of 1 mM ATP and F6P in (A and B), respectively.

120

Percent activity

110

120

A

Table 2 Km values for fructose 6-phosphate (F6P) and ATP/pyrophosphate (PPi) for phosphofructokinases (PFKs) from different sources.

B

110

100 90

100

80

80

90

70

70

60 50 40

*

60

**

*

50 40

30

30

20

20

10 0

10 0

0

0 .1

0 .3

0 .7

0

1

0.05

160

Percent activity

140

0.1

0.3

0.5

1

Citrate (mM)

ADP (mM) 160

C

D

120

100

100

80

80

60

60

40

40

20

20

0

0

0

3

5

7

F-2,6-bP ( μ M)

11

PO4 donor

Km ATP/PPi (lM)

Km F6P (lM)

Bacillus stearothermophillusa Homo sapiens (RBC)b Ricinus communisc Ricinus communisd Entamoeba histolyticae Toxoplasma gondiif Plasmodium bergheig Plasmodium falciparum (parasite lysate) Plasmodium falciparum (recombinant b protein)

ATP ATP ATP PPi PPi PPi ATP ATP ATP

120 120 8.7 17 17 33 20 20 25

23 450 16 300 40 270 32 14 13

The values in bold letters refer to those determined in the present study. RBC, red blood cell; PO4, phosphate. a Zhu et al. (1995). b Layzer (1975). c Knowles et al. (1990). d Kombrink et al. (1984). e Bruchhaus et al. (1996). f Denton et al. (1996). g Buckwitz et al. (1990).

140

120

Organismreferences

Table 3 Effect of modulators on recombinant enzyme (b domain). 0

0.1

0.3

0.5

0.7

1

PEP (mM)

Fig. 6. The effect of ADP (A), citrate (B), fructose 2,6-bisphosphate (C) and phosphoenolpyruvate (D) on phosphofructokinase activity present in Plasmodium falciparum cell lysates (n = 3). The activity is expressed as the percentage of activity observed for the control in each case (P < 0.05).

example of the catalytic and regulatory domains fused in the same gene, both apparently derived from the bacteria PFK (Kemp and Gunasekera, 2002). In Plasmodium, there have been reports of catalytic and regulatory gene fusions. For instance, the calciumdependent protein kinase from P. falciparum (PfCPK) is a multidomain protein composed of an N-terminal kinase domain fused to a C-terminal CaM-like calcium-binding domain (Zhao et al., 1994). While such fusions confer definitive association of the catalytic and regulatory regions, it could compromise the oligomerization regulation of the protein function. In plants, different expression levels of the a and b subunits have been observed in different tissues, indicating a control on the multimeric forms of the PFK protein (Wong et al., 1990; Suzuki et al., 2003).

Modulator

Concentration

Percent activitya (standard error)

PEP

0.5 mM 1.0 mM

119 (28.7) 123 (21.1)

Citrate

0.5 mM 1.0 mM

69 (12.2) 99 (28.5)

F-2,6-b-P

0.5 lM 1.0 lM

139 (24.3) 112 (36.8)

AMP

0.5 mM 1.0 mM

111 (52.5) 140 (47.6)

Mg2+

0.5 mM 1.0 mM 3.0 mM

265 (79.8) 498 (38.5) 377 (9.4)

a

Activity without modulator is taken as 100%.

Examination of the group II PFK gene structures has been conducted recently with several PFK gene structures available in the NCBI database (Müller et al., 2001; Bapteste et al., 2003). It has been established that both eukaryotic and eubacteria spp. populate this group. The clade X contains the plant and kinetoplastid

1450

B.M. Mony et al. / International Journal for Parasitology 39 (2009) 1441–1453

Fig. 7. Phylogenetic tree showing phosphofructokinases (PFK) of different types from different organisms. Bootstrap values, based on 100 replicates, are indicated at the branch points. Ath PFPb: PFKb Arabidopsis thaliana, NP_172664; Rco PFPb: PFKb Ricinus communis (Castor bean), CAA83683; TgPFKI: Toxoplasma gondii, EEA99208, ToxoDB gene ID: 42.m00123; CparPFKI: Cryptosporidium parvum, XP_626715; ChomPFKI: Cryptosporidium hominis, EAL37572; BbPFKI: Borrelia burgdorferi, CAA11968; EhPFKI: Entamoeba histolytica, AAC04465; PyPFK9ho: Plasmodium yoelii, EAA20618, PlasmoDB gene ID: PY01321; PbPFK9ho: Plasmodium berghei, CAH94255, PlasmoDB gene ID: PB000520.00.0; PvPFK9ho: Plasmodium vivax, XP_001614690, PlasmoDB gene ID: Pv099200; PfPFK9: Plasmodium falciparum CAD51837, PlasmoDB gene ID: PFI0755c; TgPFKII: T. gondii, EEA99567, ToxoDB gene ID: 49.m03242; CparPFKII: C. parvum, XP_626418; ChomPFKII: C. hominis, EAL35989; Tpar: Theileria parva EAN32860; BbovPFK: Babesia bovis, XP_001610135; Tann: Theileria annulata, CAI74406; PyPFK11ho: P. yoelii, EAA18023, PlasmoDB gene ID: PY05918; PbPFK11ho: P. berghei, CAI00470 + CAH99238, PlasmoDB gene ID: PB000972.03.0 + PB000087.03.0; PfPFK11: P. falciparum, AAN35878, PlasmoDB gene ID: PF11_0294; PvPFK11ho: P. vivax, XP_001615400, PlasmoDB gene ID: Pv092035; Hsap P: Homo sapiens, Q01813 (PFK type C); Cfamil: Canis familiaris (Dog), P52784 (PFK, muscle type); Dmela: Drosophila melanogaster, P52034; Aoryz: PFKb, Aspergillus oryzae, Q9HGZ0; ScPFKb: PFKb Saccharomyces cerevisiae, P16862; Celeg: Caenorhabditis elegans, Q27483; Dradio: Deinococcus radiodurans, Q9RWN1; Ecoli: Escherichia coli, P0A797 (PFK isozyme 1); Gstearo: Geobacillus stearothermophillus, P00512; Ath PFK: A. thaliana, AAL90928; Osat PFK: Oryza sativa, BAF04187; BbPFKII: B. burgdorferi, AAC67070; EhPFKII: E. histolytica EAL48335; Ldono: Leishmania donovani, AAK31633; Tbruc: Trypanosoma brucei, AAZ10361; PfreudPFKI: Propionibacterium freudenreichii subspecies shermanii, P29495; Atum: Agrobacterium tumefaciens, AAK87863; TmariPFKI: Thermotoga maritime, AAD35377.

B.M. Mony et al. / International Journal for Parasitology 39 (2009) 1441–1453

PFKs which are distinctly ATP-dependent and quite different from the PPi-PFK structure. Thus the protozoan PFKs appear to derive both the ATP- and PPi-dependent structures from plants, distinct from other eukaryotic species. In R. communis (castor), the organisation of the a and b subunit genes was analyzed and it was concluded that these PPi-dependent PFK genes were not derived from the ATP-PFK gene, but that all the three evolved from a common ancestral gene (Todd et al., 1995). In the case of apicomplexans, it is not clear whether the basic PFK gene was subject to gene duplication and then diverged as the catalytic b and regulatory a domains, or whether the genes diverged first and then fused together to provide the larger protein. The presence of the large fused ORFs in all the apicomplexan organisms suggest that these must have originated from a fused ancestral PFK gene. Our analysis demonstrates that Plasmodium contains only ATPdependent activity. This fits with the mutational assignment of the ATP- versus PPi-PFK activity through the G175 motif (Chi and Kemp, 2000). The ATP-PFKs of the apicomplexans possess the G175N177G205 structure while those belonging to other protists or Borrelia exhibit a G175Q177N205 (or G175L177N205 in the case of E. histolytica) motif (Table 1). All of the PPi-dependent structures showed a D175N177G205 motif except B. burgdorferi, which shows a D175N177A205 motif. Of all the apicomplexan G175-PFK gene structures, the Plasmodium protein is the only one which demonstrates ATP-dependent activity (Sander et al., 1982; this paper). The phylogenetic tree (Fig. 7) shows that the apicomplexan PFK proteins seem to belong to a unique class of the Type II PFK proteins. Based on the sequence similarities and the G/D175 characteristics of the deduced structures, we postulate that the PFKs from the other Plasmodium species, as well as T. gondii PFKII and C. parvum PFKII, would exhibit ATP-dependent activity (Type IIb category). The Type IIb category thus seems to be populated exclusively by apicomplexan organisms, while Type IIa has representations from plants, amoeba, spirochaete B. burgdorferi and apicomplexans. It is observed that in the second PFK gene, PfPFK11 in P. falciparum, almost all of the crucial catalytic motifs of this enzyme are missing (Table 1). It is possible that originally, together with other apicomplexan organisms, the two PfPFK genes served the function of PPi- and ATP-dependent activities and later the PfPFK11 gene lost this capability. The transcriptome and proteomic data indicate that this protein is not expressed in the asexual stages, but the mRNA is expressed relatively highly in the gametocytic stages, while the reverse situation exists with PfPFK9 protein (Aurrecoechea et al., 2009). The presence of homologous gene ORFs of PfPFK11 in other species of Plasmodium (percent identity scores of 38–46) indicates a possible function, which seems to pertain specifically to the gametocytes. Through over-expression studies of the PfPFK11 protein in Plasmodium asexual stages, it would be interesting to assess whether it plays a dominant negative role towards PfPFK9 activity. It would also be of interest to generate reagents specific for PfPFK11 and measure the PFK enzyme activity and the concentrations of each of PfPFK9 and PfPFK11 proteins in the sexual and asexual stages of Plasmodium. Currently it appears that all living organisms possess at least one ATP-dependent PFK. The co-existence of PPi- and ATP-PFKs occurs in plants, protists and in several bacteria species. Certain bacteria, which use multiple sources of carbon for growth, exhibit a very remarkable regulation of ATP-dependent versus PPi-dependent enzymes based on the carbon source. The immediate induction of ATP-dependent enzyme activity in the presence of glucose (Alves et al., 2001) indicates that in vivo glucose-metabolism is closely connected to ATP, and not PPi, as the phosphate donor for PFK. It is interesting to note that Plasmodium, Babesia and Theileria are obligatory parasites, which do not exist in a free form. These organisms would have access to glucose as a carbon source

1451

throughout their existence, while other protists which may spend some time in the environment as a cyst may not. For such free-living parasites it may be advantageous to possess both the ATP- and PPi-dependent PFKs. This is perhaps why T. gondii, C. parvum and C. hominis, which have a cyst form at some point in their life cycle, possess both of the PFKs, while Plasmodium, Babesia and Theileria species can make do with only one ATPdependent PFK activity. The plant PPi-dependent PFK enzymes catalyze a reversible reaction and operate near equilibrium (Nielsen et al., 2004). F-2,6-bP regulation of the eukaryotic ATP-PFKs is a hallmark of the regulation of glycolysis (Kemp and Gunasekera, 2002) but in plants this regulation is exerted only on the PPi-dependent protein and not on the cytosolic ATP-dependent protein (Nielsen et al., 2004). The enzyme activities in both directions are enhanced by very low (nanomolar) levels of F-2,6-bP and are supposed to regulate the flow of carbon through photosynthetic, catabolic and anabolic phases (Van Praag, 1997). The apicomplexan PPi-dependent enzymes of T. gondii and C. parvum are not sensitive to F-2,6-bP (Denton et al., 1996). It is hypothesised that the F-2,6bP binds at the interface between the a and b subunits (Van Praag, 1997; Nielsen et al., 2004). In the cases of T. gondii and C. parvum the a and b units are fused, just as in Plasmodium; hence the interface may be very distinct from the plant heteromeric structures. A database search failed to identify the fructose-6-phosphate, 2-kinase/fructose-2,6-bisphosphatase (FBPase) gene (responsible for the generation of F-2,6-bP) amongst all apicomplexans. The Plasmodium genome also does not seem to contain the gene for FBPase, which catalyzes the reverse reaction of PFK, leading to gluconeogenesis. It appears that glycolysis is a robust energy utilisation pathway in Plasmodium and is not subject to fine regulation as long as it receives glucose from its environment. The T. gondii database, however, shows the presence of FBPase, indicating that the lack of the gene for FBPase in Plasmodium is not due to a problem with the search program. The lack of FBPase is consistent with the observation that gluconeogenesis is virtually non-existent in the asexual stages of P. falciparum. The Plasmodium PFK protein is homologous to plant PFKs and shows only 15% similarity with human PFK proteins. With such major structural differences from the host PFK protein, the Plasmodium PFK protein is a potential drug target. Through heterologous expression, we have demonstrated that PfPFK9b region from P. falciparum is the functional ATP-dependent catalytic domain of PFK. Elucidation of the structure of this important glycolytic PfPFK9 protein and a search for selective inhibitors to function as suitable drugs would enable us to better control this disease. Acknowledgements We thank the E. coli Genetic Stock Center, Yale University, USA, for providing us with the E. coli DF1020 strain and Prof. Gregory D. Reinhart, Texas A&M University, USA, for the B. stearothermophillus PFK (BsPFK) gene construct and the E. coli RL257 strain. We are indebted to Prof. Nirbhay Kumar, Johns Hopkins School of Public Health, Baltimore, USA and Pavani Neeraja, Tata Institute of Fundamental Research, Mumbai, India for inputs in discussions and experiments. We also wish to acknowledge Bioklone, Chennai, for the great interactive service provided for specific hybridoma selection for PFKb. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.ijpara.2009.05.011.

1452

B.M. Mony et al. / International Journal for Parasitology 39 (2009) 1441–1453

References Agbenyega, T., Angus, B.J., Bedu-Addo, G., Baffoe-Bonnie, B., Guyton, T., Stacpoole, P.W., Krishna, S., 2000. Glucose and lactate kinetics in children with severe malaria. J. Clin. Endocrinol. Metab. 85, 1569–1576. Alves, A.M., Euverink, G.J., Santos, H., Dijkhuizen, L., 2001. Different physiological roles of ATP- and PP(i)-dependent phosphofructokinase isoenzymes in the methylotrophic actinomycete Amycolatopsis methanolica. J. Bacteriol. 183, 7231–7240. Aurrecoechea, C., Brestelli, J., Brunk, B.P., Dommer, J., Fischer, S., Gajria, B., Gao, X., Gingle, A., Grant, G., Harb, O.S., Heiges, M., Innamorato, F., Iodice, J., Kissinger, J.C., Kraemer, E., Li, W., Miller, J.A., Nayak, V., Pennington, C., Pinney, D.F., Roos, D.S., Ross, C., Stoeckert Jr., C.J., Treatman, C., Wang, H., 2009. PlasmoDB: a functional genomic database for malaria parasites. Nucleic Acids Res. 37, D539– D543. Bapteste, E., Moreira, D., Philippe, H., 2003. Rampant horizontal gene transfer and phospho-donor change in the evolution of the phosphofructokinase. Gene 318, 185–191. Bär, J., Golbik, R., Hubner, G., Kopperschlager, G., 2000. Denaturation of phosphofructokinase-1 from Saccharomyces cerevisiae by guanidinium chloride and reconstitution of the unfolded subunits to their catalytically active form. Biochemistry 39, 6960–6968. Beutler, E., 1984. In: Red Cell Metabolism: A Manual of Biochemical Methods. Grune & Stratton, Philadelphia, pp. 8–45. Bloxham, D.P., Lardy, H.A., 1973. Phosphofructokinase. In: Boyer, P.D. (Ed.), The Enzymes, vol. 8. Academic Press, New York, NY, pp. 239–278. Bruchhaus, I., Jacobs, T., Denart, M., Tannich, E., 1996. Pyrophosphate-dependent phosphofructokinase of Entamoeba histolytica: molecular cloning, recombinant expression and inhibition by pyrophosphate analogues. Biochem. J. 316, 57– 63. Buckwitz, D., Jacobasch, G., Gerth, C., Holzhutter, H.G., Thamm, R., 1988. A kinetic model of phosphofructokinase from Plasmodium berghei. Influence of ATP and fructose-6-phosphate. Mol. Biochem. Parasitol. 27, 225–232. Buckwitz, D., Jacobasch, G., Gerth, C., 1990. Phosphofructokinase from Plasmodium berghei: influence of Mg2+, ATP and Mg2(+)-complexed ATP. Biochem. J. 267, 353–357. Carlisle, S.M., Blakeley, S.D., Hemmingsen, S.M., Trevanion, S.J., Hiyoshi, T., Kruger, N.J., Dennis, D.T., 1990. Pyrophosphate-dependent phosphofructokinase: conservation of protein sequence between the alpha- and beta-subunits and with the ATP-dependent phosphofructokinase. J. Biol. Chem. 265, 18366– 18371. Chan, M., Sim, T.S., 2005. Functional analysis, overexpression, and kinetic characterization of pyruvate kinase from Plasmodium falciparum. Biochem. Biophys. Res. Commun. 326, 188–196. Chi, A., Kemp, R.G., 2000. The primordial high energy compound: ATP or inorganic pyrophosphate? J. Biol. Chem. 275, 35677–35679. Chi, A.S., Deng, Z., Albach, R.A., Kemp, R.G., 2001. The two phosphofructokinase gene products of Entamoeba histolytica. J. Biol. Chem. 276, 19974–19981. Deng, Z., Huang, M., Singh, K., Albach, R.A., Latshaw, S.P., Chang, K.P., Kemp, R.G., 1998. Cloning and expression of the gene for the active PPi-dependent phosphofructokinase of Entamoeba histolytica. Biochem. J. 329, 659–664. Deng, Z., Roberts, D., Wang, X., Kemp, R.G., 1999. Expression, characterization, and crystallization of the pyrophosphate-dependent phosphofructo-1-kinase of Borrelia burgdorferi. Arch. Biochem. Biophys. 371, 326–331. Denton, H., Brown, S.M., Roberts, C.W., Alexander, J., McDonald, V., Thong, K.W., Coombs, G.H., 1996. Comparison of the phosphofructokinase and pyruvate kinase activities of Cryptosporidium parvum, Eimeria tenella and Toxoplasma gondii. Mol. Biochem. Parasitol. 76, 23–29. Fleige, T., Fischer, K., Ferguson, D.J., Gross, U., Bohne, W., 2007. Carbohydrate metabolism in the Toxoplasma gondii apicoplast: localization of three glycolytic isoenzymes, the single pyruvate dehydrogenase complex, and a plastid phosphate translocator. Eukaryot. Cell 6, 984–996. Ginsburg, H., 2006. Progress in in silico functional genomics: the malaria metabolic pathways database. Trends Parasitol. 22, 238–240. Goswami, A., Singh, S., Redkar, V.D., Sharma, S., 1997. Characterization of P0, a ribosomal phosphoprotein of Plasmodium falciparum: antibody against aminoterminal domain inhibits parasite growth. J. Biol. Chem. 272, 12138–12143. Inoue, H., Nojima, H., Okayama, H., 1990. High efficiency transformation of Escherichia coli with plasmids. Gene 96, 23–28. Kemp, R.G., Gunasekera, D., 2002. Evolution of the allosteric ligand sites of mammalian phosphofructo-1-kinase. Biochemistry 41, 9426–9430. Klinder, A., Kirchberger, J., Edelmann, A., Kopperschlager, G., 1998. Assembly of phosphofructokinase-1 from Saccharomyces cerevisiae in extracts of singledeletion mutants. Yeast 14, 323–334. Knowles, V.L., Greyson, M.F., Dennis, D.T., 1990. Characterization of ATP-dependent fructose 6-phosphate 1-phosphotransferase isozymes from leaf and endosperm tissues of Ricinus communis. Plant Physiol. 92, 155–159. Kombrink, E., Kruger, N.J., Beevers, H., 1984. Kinetic properties of pyrophosphate: fructose-6-phosphate phosphotransferase from germinating castor bean endosperm. Plant Physiol. 74, 395–401. Lang-Unnasch, N., Murphy, A.D., 1998. Metabolic changes of the malaria parasite during the transition from the human to the mosquito host. Annu. Rev. Microbiol. 52, 561–590. Layzer, R.B., 1975. Phosphofructokinase from human erythrocytes. Methods Enzymol. 42, 110–115.

López, C., Chevalier, N., Hannaert, V., Rigden, D.J., Michels, P.A., Ramirez, J.L., 2002. Leishmania donovani phosphofructokinase: gene characterization, biochemical properties and structure-modeling studies. Eur. J. Biochem. 269, 3978–3989. Lovingshimer, M.R., Siegele, D., Reinhart, G.D., 2006. Construction of an inducible, pfkA and pfkB deficient strain of Escherichia coli for the expression and purification of phosphofructokinase from bacterial sources. Protein Expr. Purif. 46, 475–482. Marchler-Bauer, A., Anderson, J.B., Cherukuri, P.F., DeWeese-Scott, C., Geer, L.Y., Gwadz, M., He, S., Hurwitz, D.I., Jackson, J.D., Ke, Z., Lanczycki, C.J., Liebert, C.A., Liu, C., Lu, F., Marchler, G.H., Mullokandov, M., Shoemaker, B.A., Simonyan, V., Song, J.S., Thiessen, P.A., Yamashita, R.A., Yin, J.J., Zhang, D., Bryant, S.H., 2005. CDD: a conserved domain database for protein classification. Nucleic Acids Res. 33, D192–D196. Massey, T.H., Deal Jr., W.C., 1973. Unusual, metabolite-dependent solubility properties of phosphofructokinase. J. Biol. Chem. 248 (1), 56–62. Mather, M.W., Vaidya, A.B., 2008. Mitochondria in malaria and related parasites: ancient, diverse and streamlined. J. Bioenerg. Biomembr. 40, 425–433. Mehta, M., Sonawat, H.M., Sharma, S., 2005. Malaria parasite-infected erythrocytes inhibit glucose utilization in uninfected red cells. FEBS Lett. 579, 6151–6158. Mertens, E., 1993. ATP versus pyrophosphate: glycolysis revisited in parasitic protests. Parasitol. Today 9, 122–126. Michels, P.A., Chevalier, N., Opperdoes, F.R., Rider, M.H., Rigden, D.J., 1997. The glycosomal ATP-dependent phosphofructokinase of Trypanosoma brucei must have evolved from an ancestral pyrophosphate-dependent enzyme. Eur. J. Biochem. 250, 698–704. Müller, M., Lee, J.A., Gordon, P., Gaasterland, T., Sensen, C.W., 2001. Presence of prokaryotic and eukaryotic species in all subgroups of the PP(i)-dependent group II phosphofructokinase protein family. J. Bacteriol. 183, 6714–6716. Mustroph, A., Sonnewald, U., Biemelt, S., 2007. Characterisation of the ATPdependent phosphofructokinase gene family from Arabidopsis thaliana. FEBS Lett. 581, 2401–2410. Nielsen, T.H., Rung, J.H., Villadsen, D., 2004. Fructose-2,6-bisphosphate: a traffic signal in plant metabolism. Trends Plant Sci. 9, 556–563. Oelshlegel Jr., F.J., Sander, B.J., Brewer, G.J., 1975. Pyruvate kinase in malaria host– parasite interaction. Nature 255, 345–347. Pal-Bhowmick, I., Vora, H.K., Jarori, G.K., 2007. Sub-cellular localization and posttranslational modifications of the Plasmodium yoelii enolase suggest moonlighting functions. Malar. J. 6, 45. Peng, Z.Y., Mansour, T.E., 1992. Purification and properties of a pyrophosphatedependent phosphofructokinase from Toxoplasma gondii. Mol. Biochem. Parasitol. 54, 223–230. Ramirez-Flandes, S., Ulloa, O., 2008. Bosque: integrated phylogenetic analysis software. Bioinformatics 24, 2539–2541. Reeves, R.E., South, D.J., Blytt, H.J., Warren, L.G., 1974. Pyrophosphate: D-fructose 6phosphate 1-phosphotransferase: a new enzyme with the glycolytic function of 6-phosphofructokinase. J. Biol. Chem. 249, 7737–7741. Roth Jr., E.F., Raventos-Suarez, C., Perkins, M., Nagel, R.L., 1982. Glutathione stability and oxidative stress in P. falciparum infection in vitro: responses of normal and G6PD deficient cells. Biochem. Biophys. Res. Commun. 109, 355–362. Roth Jr., E.F., Calvin, M.C., Max-Audit, I., Rosa, J., Rosa, R., 1988. The enzymes of the glycolytic pathway in erythrocytes infected with Plasmodium falciparum malaria parasites. Blood 72, 1922–1925. Sander, B.J., Lowery, M.S., Kruckeberg, W.C., 1982. Plasmodium berghei acidinsensitive phosphofructokinase in infected mouse erythrocytes. Exp. Parasitol. 53, 11–16. Schichtherle, M., Wahlgren, M., 2008. Small-scale genomic DNA isolation from Plasmodium falciparum. In: Moll, K., Ljungström, I., Perlmann, H., Scherf, A., Wahlgren, M. (Eds.), Methods in Malaria Research. MR4/ATCC, Manassas, pp. 199–200. Shirakihara, Y., Evans, P.R., 1988. Crystal structure of the complex of phosphofructokinase from Escherichia coli with its reaction products. J. Mol. Biol. 204, 973–994. Singh, S., Sehgal, A., Waghmare, S., Chakraborty, T., Goswami, A., Sharma, S., 2002. Surface expression of the conserved ribosomal protein P0 on parasite and other cells. Mol. Biochem. Parasitol. 119, 121–124. Suzuki, J., Mutton, M.A., Ferro, M.I., Lemos, M.V., Pizauro, J.M., Mutton, M.J., Di Mauro, S.M., 2003. Putative pyrophosphate phosphofructose 1-kinase genes identified in sugar cane may be getting energy from pyrophosphate. Genet. Mol. Res. 2, 376–382. Todd, J.F., Blakeley, S.D., Dennis, D.T., 1995. Structure of the genes encoding the alphaand beta-subunits of castor pyrophosphate-dependent phosphofructokinase. Gene 152, 181–186. Turner, W.L., Plaxton, W.C., 2003. Purification and characterization of pyrophosphate- and ATP-dependent phosphofructokinases from banana fruit. Planta 217, 113–121. Uyeda, K., 1979. Phosphofructokinases. Adv. Enzymol. 48, 193–244. Van Praag, E., 1997. Use of 3-D computer modelling and kinetic studies to analyse grapefruit pyrophosphate-dependent phosphofructokinase. Int. J. Biol. Macromol. 21, 307–317. Van Schaftingen, E., Lederer, B., Bartrons, R., Hers, H.G., 1982. A kinetic study of pyrophosphate: fructose-6-phosphate phosphotransferase from potato tubers. Application to a microassay of fructose 2,6-bisphosphate. Eur. J. Biochem. 129, 191–195. Winkler, C., Delvos, B., Martin, W., Henze, K., 2007. Purification, microsequencing and cloning of spinach ATP-dependent phosphofructokinase link sequence and function for the plant enzyme. FEBS J. 274, 429–438. World Health Organization, 2008. World Malaria Report 2008. WHO, Geneva.

B.M. Mony et al. / International Journal for Parasitology 39 (2009) 1441–1453 Wong, J.H., Kiss, F., Wu, M.X., Buchanan, B.B., 1990. Pyrophosphate fructose-6-p 1phosphotransferase from tomato fruit: evidence for change during ripening. Plant Physiol. 94, 499–506. Yan, T.F., Tao, M., 1984. Multiple forms of pyrophosphate: D-fructose-6-phosphate 1-phosphotransferase from wheat seedlings. Regulation by fructose 2,6bisphosphate. J. Biol. Chem. 259, 5087–5092. Yanai, I., Wolf, Y.I., Koonin, E.V., 2002. Evolution of gene fusions: horizontal transfer versus independent events. Genome Biol. 3. research0024.

1453

Zhao, Y., Pokutta, S., Maurer, P., Lindt, M., Franklin, R.M., Kappes, B., 1994. Calciumbinding properties of a calcium-dependent protein kinase from Plasmodium falciparum and the significance of individual calcium-binding sites for kinase activation. Biochemistry 33, 3714–3721. Zhu, X., Byrnes, M., Nelson, J.W., Chang, S.H., 1995. Role of glycine 212 in the allosteric behavior of phosphofructokinase from Bacillus stearothermophillus. Biochemistry 34, 2560–2565.