j ourna l homepage: www.e lsev ie r .com/ locate /par in t
Post-translational modification of the pyruvate phosphate dikinase fromTrypanosoma cruzi
Eglys González-Marcano a,⁎, Alfredo Mijares b, Wilfredo Quiñones a, Ana Cáceres a, Juan Luis Concepción a
a Laboratorio de Enzimología de Parásitos, Facultad de Ciencias, Universidad de Los Andes, La Hechicera, Mérida 5101, Venezuelab Laboratorio de Fisiología de Parásitos, Centro de Biofísica y Bioquímica, Instituto Venezolano de Investigaciones Científicas, Caracas 1020-A, Venezuela
In kinetoplastids such as Trypanosoma cruzi, glycolysis is compartmentalized in peroxisome-like organelles calledglycosomes. Pyruvate phosphate dikinase (PPDK), an auxiliary enzyme of glycolysis, is also located in theglycosomes. We have detected that this protein is post-translationally modified by phosphorylation and proteo-lytic cleavage. On western blots of T. cruzi epimastigotes, two PPDK forms were found with apparent MW of100 kDa and 75 kDa, the latter one being phosphorylated at Thr481, a residue present in a highly conserved re-gion. In subcellular localization assays the 75 kDa PPDK was located peripherally at the glycosomal membrane.Both PPDK forms were found in all life-cycle stages of the parasite. When probing for both PPDK forms duringa growth of epimastigotes in batch culture, an increase in the level of the 75 kDa form and a decrease of the100 kDa one were observed by western blot analysis, signifying that glucose starvation and the concomitantswitch of the metabolism to amino acid catabolism may play a role in the post-translational processing of thePPDK. Either one or both of the processes, phosphorylation and proteolytic cleavage of PPDK, result in inactiva-tion of the enzyme. It remains to be established whether the phenomenon exerts a regulatory function.
Trypanosoma cruzi, as well as other trypanosomatids, utilizes bothglucose and amino acids as carbon and energy source, with glucosebeing preferentially catabolized [1]. Trypanosomatids have a uniquelyorganized glucose metabolism, in which the major part of glycolysistakes place in glycosomes, a specialized form of peroxisomes [2].Glycosomes contain also an auxiliary glycolytic system, comprising anATP-producing and CO2-fixing phosphoenolpyruvate carboxykinase(PEPCK) and a malate dehydrogenase (MDH) that together catalyzethe production of malate from phosphoenolpyruvate (PEP) via the for-mation of oxaloacetate, and coupled to the oxidation of a reducedform of nicotinamide adenine dinucleotide (NADH) [3]. Malate is thenconverted to fumarate by fumarate hydratase (FH) and subsequentlyto succinate by fumarate reductase (FRD) coupled to the oxidation ofanother molecule of NADH.
In glycosomes, PEP can also be converted to pyruvate by pyrophos-phate (PPi)-dependent pyruvate phosphate dikinase (PPDK) and subse-quently to alanine by alanine dehydrogenase (ADH), generating onemolecule of each ATP and NAD+. From the perspective of redox andATP/ADP balances, the reaction of PPDK plus ADH is equivalent to that
of PEPCK plus MDH [4]. However, the activity of PEPCK/MDH branchin PEP metabolism is merely associated with the ATP/ADP and NADH/NAD+ ratios, while the PPDK/ADHbranch depends also on the availabil-ity of PPi and AMP. Therefore, a system for sorting PEP into either thePPDK/ADH or the PEPCK/MDH branch may be operational to maintainthe redox and ATP/ADP balances within the glycosomes.
All glycosomal proteins are synthesized in the cytosol and importedinto the organelle, as is the case for peroxisomal proteins. These proteinsare routed to their destination by signal peptide sequences known asperoxisome-targeting signals (PTS). PTS are classified into two maintypes; PTS1 which is a C-terminal tripeptide, SKL, or physicochemicalvariations of it such as AKL, while PTS2 is a nonapeptide located closeto the N-terminus with the consensus motif (RK)-(LIV)-X5-(QH)-(LA).PTS-containing proteins are recognized in the cytosol, transferred tothe peroxisomal membranes, and delivered into the matrix of the or-ganelle by the action of proteins called peroxins [5]. The peroxins ofthis import machinery and their molecular interactions are also con-served in glycosomes and have been studied in Trypanosoma brucei;orthologous genes are also present in the T. cruzi genome and thePPDK of both trypanosomatid species has been found to possess aPTS1 [6].
Several biosynthetic processes and the activation of fatty acids takeplace also inside glycosomes [7]. PPi, a common byproduct of these pro-cesses that is toxic to the cell, inhibits PEPCK activity [4]. On the otherhand, acetyl-CoA, a product of the β-oxidation of fatty acids that canalso occur in glycosomes, increases the PPDK activity [4]. This stronglysuggests that indeed a regulatory mechanism for this auxiliary pathway
81E. González-Marcano et al. / Parasitology International 63 (2014) 80–86
exists, making its activity dependent on themetabolism of the parasite.Such a mechanism has not yet been described for trypanosomes.
PPDK has been well studied in C4 plants, where it is connected tothe process of photosynthesis. It operates in the gluconeogenic path-way, producing PEP and PPi from ATP and pyruvate, therefore con-tributing to the CO2 fixation in plants [8,9]. The enzyme undergoesa light–darkness dependent regulation of its activity in vivo. This reg-ulation implicates a post-translational modification involving the re-versible phosphorylation of an active-site threonine residue in aconserved region of the polypeptide (Prosite AC: PS00370). Duringdarkness, the enzyme is inactivated by phosphorylation while it isre-activated upon dephosphorylation during daylight [10]. This reg-ulation is exerted through a bifunctional stromal regulatory protein(RP) that acts as both a kinase (in an ADP-dependent reaction) anda phosphatase [11]. This led us to investigate if T. cruzi PPDK mightundergo a similar regulatory modification.
2. Materials and methods
2.1. Parasites
Epimastigotes of T. cruzi strain EP (isolated from a patient withChagas disease in acute phase) were cultured at 28 °C in LIT medium(liver infusion-tryptose) supplementedwith 5% inactivated fetal bovineserum [12]. Trypomastigotes and amastigotes were obtained from Verocell cultures, as described in [13] with slight modifications.
2.2. Cloning and expression of recombinant T. cruzi PPDK
Two specific primers were designed based on a consensus sequenceof PPDK genes available in the NCBI database (GI:71666489 andGI:71658998) and used to amplify the complete T. cruzi gene by PCR;the oligonucleotides used were: 5′CGGCATATGGAATCCAAAAAGTTTGTTTAC3′ as sense primer (NdeI cutting site in bold) and 5′CGGAATTCCTACAGCTTGGCGGCGAT3′ as antisense primer (EcoRI cuttingsite in bold). Amplification was performed in 30 PCR cycles involvingdenaturation at 95 °C for 60 s, annealing at 60 °C for 120 s, and elonga-tion at 72 °C for 180 s with Platinum Pfu Taq DNA polymerase, and thecorresponding reactionbuffer suppliedby themanufacturer (Invitrogen®).The amplified product was purified from the agarose gel using the GelDNA purification kit (Promega®). The product was ligated to thepGEM-T-Easy vector (Promega®), the resulting plasmid was used totransform the TG-1 strain of Escherichia coli and plasmid DNA from pos-itive clones was purified using the SV miniprep-DNA Purification Sys-tem (Promega®) and subjected to sequencing.
The complete gene after being treatedwithNdeI and EcoRIwas ligat-ed to the expression vector pET28a(+). The resulting plasmid togetherwith the pRARE plasmid was used to doubly transform the BL21(DE3)strain of E. coli. Expression was performed in auto-induction medium(plus 450 mM NaCl) at ambient temperature (25 °C) for 30 h in thepresence of 50 μg·ml−1 of kanamycin and 60 μg·ml−1 of chloram-phenicol. The protein was purified by Immobilized Metal (Ni) AffinityChromatography (IMAC) using a ProBond™ resin (Stratagene®); purifi-cation steps were performed as recommended by the manufacturer,and the enzymewas shown to be properly purified and active. The spe-cific activity was determined by measuring spectrophotometrically theoxidation of NADH at 340 nm by coupling the PPDK reaction to that oflactate dehydrogenase (LDH), as described in [6].
2.3. Protein determination, SDS-PAGE and immunoblotting
Proteins were quantitatively assayed by the Lowry method as mod-ified by [14] with bovine serum albumin as standard. SDS-PAGE wasperformed in 10% polyacrylamide gels according to the method de-scribed in [15]. Proteins separated by SDS-PAGE were transferred to aHybond ECL nitrocellulose membrane (GE Healthcare Amersham) and
western blotting was performed as described elsewhere [16]. Mousemonoclonal anti-PPDK from T. brucei (kindly donated by Dr. FrédéricBringaud)was used at a dilution of 1:4000; purified polyclonal antibod-ies raised in rabbit against the phosphorylated Thr456 PPDK peptide ofZea mays (hereafter called anti-phospho peptide of Z. mays PPDK) andrabbit-raised purified polyclonal anti-PPDK from Z. mays (both kindlydonated by Dr. Chris Chastain) were used at a dilution of 1:3000 and1:5000, respectively, andmouse monoclonal anti-phosphorylated thre-onine was used at a dilution of 1:1000 (commercially obtained fromSigma-Aldrich). The secondary antibody incubation was performedwith peroxidase-conjugated goat anti-rabbit immunoglobulin and rab-bit anti-mouse immunoglobulin diluted at 1:5000 and 1:14000, respec-tively. Immunoblotswere processed using the ECL Pluswestern blottingdetection reagents (GE Healthcare Amersham) and the images were vi-sualized in a Molecular Imager (ChemiDoc™ XRS Plus — BioRad). Forquantitative analysis of western blots, the Image J program was used.
2.4. Sub-organellar localization of PPDK in epimastigotes of T. cruzi
Highly purified intact glycosomeswere obtained fromepimastigotesin their exponential growth phase, by isopycnic centrifugation as de-scribed in [17]. During glycosome purification, buffer A was used(Tris–HCl 20 mM, pH 7.2, with sucrose 225 mM, KCl 20 mM, KH2PO4
10 mM, EDTA-Na2 1 mM and MgCl2 5 mM) always with a cocktail ofprotease inhibitors (Sigma), and 0.5mMof PPi and 1mMof potassiumfluoride (KF) as phosphatase inhibitors. Latencies were determined byassaying hexokinase activity, as in [18]. Sub-organellar fractions wereobtained by dividing the glycosomal preparation in three parts, eachone to be treated differently [19]: i) Osmotic shock. 10 mg of purifiedglycosomes were incubated on ice with 25 ml of cold deionized water(milli-Q) for 2 h, and centrifuged at 170,000 ×g for 1.5 h at 4 °C. Pro-teins from the fractions obtained were precipitated with trichloroaceticacid/deoxycholate (TCA/DOC) followed by precipitation with acetone;pellets were suspended in buffer C (Tris–HCl 10 mM, pH 7.4, withNaCl 150 mM) and homogenized. The protein concentrationwas deter-mined and 30 μg of protein were mixed with sample buffer to be usedfor SDS-PAGE. ii) Phase separation with Triton X-114. 2 mg of purifiedglycosomes were mixed with four volumes of buffer C and 2% TritonX-114 (v/v). This mixture was frozen at −80 °C for 1 h, thawed on iceand then maintained at ambient temperature until clouding occurred,immediately followed by centrifugation at 300 ×g for 5 min. Twophases were obtained, an aqueous phase (AP)with the hydrophilic pro-teins and a detergent phase (DP) with the amphiphilic proteins. Phaseswere separated and both washed twice with buffer C (plus NaCl to afinal concentration of 0.5 mM) and 2% Triton X-114 (v/v) for the APand 0.1% (v/v) for the DP; the process of clouding and centrifugationwas repeated. Proteins from theDPwere precipitatedwith cold acetoneand the obtained pellet suspended in sample buffer, while the AP wasdirectly mixed with sample buffer to be used for SDS-PAGE [20]. iii)Treatment with sodium carbonate. 10 mg of purified glycosomes weresuspended in 100 mM cold sodium carbonate at pH 11.5 to a volumeof 25 ml. This suspension was incubated for 2 h at 0 °C, thenultracentrifuged at 170,000 ×g for 1.5 h at 4 °C. Proteins of the two frac-tions obtained, the membrane fraction in the pellet and a soluble frac-tion in the supernatant, were precipitated with TCA/DOC and coldacetone, and the pellets were subsequently resuspended in buffer C.The protein concentration was determined and 30 μg of each fractionwas mixed with sample buffer to be used for SDS-PAGE [21].
2.5. Differential expression and activity of PPDK during growth of T. cruziepimastigotes
Three samples of cultured epimastigotes (30 ml each) were taken atdifferent growth phases, pre-exponential, exponential and stationary.Parasites of each samplewere counted using a hemocytometer. All sam-ples were centrifuged at 3000 ×g for 15 min and the pellets washed
Fig. 1. Alignment of PPDK sequences. Amino acid segments of PPDK from Arabidopsis thaliana, Zeamays, Trypanosoma cruzi and Trypanosoma bruceiwere alignedwith each other andwiththe sequence of the phosphorylatedpeptide (Prosite AC: PS00370). Conserved identical aminoacidswith the peptide sequence are enclosed in a black box, andpositionally identical aminoacids among the PPDKs sequences are enclosed in a white box and in bold characters. The phosphorylated threonine is indicated with a star ( ) and the catalytic histidine with an up-pointing triangle (▲).
Fig. 2. Identification of a processed PPDK in T. cruzi. Proteins frompurified glycosomeswereanalyzed bywestern blottingusing amonoclonal anti-PPDK from T. brucei (α-PPDKm), anti-phospho threonine (α-Thr-P) and anti-phospho peptide from Z. mays PPDK (α-Peptide).Two forms of PPDK were detected, one of 100 kDa and another of 75 kDa. Proteins wereseparated by SDS-PAGE, blotted and then analyzedwith the indicated antibodies.MW:Mo-lecular weight.
82 E. González-Marcano et al. / Parasitology International 63 (2014) 80–86
twice with normal buffer (Tris–HCl 70 mM, pH 7.4, NaCl 140 mM, KCl100 mM). The parasites were suspended in sample buffer and a cocktailof protease and phosphatase inhibitorswere added, the volumewas cal-culated to obtain a final concentration of 109 cells·ml−1. For each phaseof growth, proteins corresponding to 7 × 106 parasites per lane wereused for SDS-PAGE and analyzed by western blot using monoclonalanti-PPDK from T. brucei and anti-phospho peptide of Z. mays PPDK.
2.6. In vitro phosphorylation of the recombinant T. cruzi PPDK
Extracts from darkness and light treatedmaize leaveswere obtainedas described in [10]. The recombinant PPDK phosphorylation assayswere performed as in [10] with slight variations. A 10 ml assay mixturewas composed of 0.5 mg of purified recombinant T. cruzi PPDK and0.2 mg of leaf extract in incubation buffer (bicine–KOH 50 mM,pH 8.3, MgCl2 10 mM, DTT 5 mM, BSA 1 mg·ml−1, ADP 1 mM, ATP0.2 mM, PEP 0.6 mM, protease and phosphatase inhibitors). Incubationwas performed at 30 °C for 1.5 h and samples were taken every 30 min.To terminate the reaction in each sample, an equal volume of samplebuffer was added followed by heating at 100 °C for 5 min; but prior tothis process the specific activity of PPDK was assayed. All sampleswere evaluated by western blotting using as probes monoclonal anti-T. brucei PPDK, anti-phospho threonine and anti-phospho peptidefrom Z. mays PPDK.
3. Results
3.1. Sequence analysis of T. cruzi PPDK
The amino acid sequence of T. cruzi PPDK was aligned with those ofArabidopsis thaliana, Z. mays, T. brucei and the phosphorylation peptidesequence obtained from Prosite (AC: PS00370). The alignment wasmade with MultAlin version 5.4.1, using the Symbol comparison tableBlosum62; and the output layout was edited with ESPript 2.2. In thealignment in Fig. 1 can be observed that the phosphorylation peptideis entirely conserved in all sequences analyzed [22]. Therefore, a possi-ble phosphorylation of the PPDK from T. cruzi was subsequentlyexplored.
3.2. T. cruzi PPDK is phosphorylated at threonine 481
The possible phosphorylation of the T. cruzi PPDK was explored bywestern blotting using purified glycosomes. As shown in Fig. 2, theexpected signal at 100 kDa with the anti-PPDK antibody was observed.Surprisingly also a strong signalwas detected at 75 kDawith all antibod-ies used (anti-PPDK, anti-phospho threonine andanti-phospho peptide),but the 100 kDa protein did not react with the anti-phospho threonineand anti-phospho peptide antibodies. As this might indicate that thereis a novel formof 75 kDa PPDK that is phosphorylated,we set out to con-firm this result by evaluating whole extracts obtained from leaves of Z.mays incubated in light and darkness using the same antibodies, plusan anti-PPDK from Z. mays. In Fig. 3, it is noticeable that, as expected,PPDK is present in both plant extracts (light and dark incubated), butonly the extract from the dark incubated leaves exhibits a signal whenprobed with the anti-phospho peptide antibody. Consequently, the rec-ognition of a 75 kDa protein in the glycosomes by this latter antibody as
well as by themonoclonal anti-PPDK, confirms the presence in T. cruzi ofa PPDK that is phosphorylated at the threonine residue 481 within theconserved peptide. Importantly, this phosphorylated PPDK form has aconsiderably lower molecular weight than that reported for the enzyme[6,22].
3.3. Sub-organellar localization of PPDK in T. cruzi
In previous studies performed in our laboratory, it was found thatpart of PPDK remained associated with the glycosomal membrane frac-tion after disruption of the organelle (data not shown). Therefore, wewere interested in determining inmore detail the localization of the en-zyme in the organelle, to confirm a possible dual distribution over ma-trix and membrane. As a strategy, three disruption procedures wereused; osmotic shock is a treatment that disrupts the organelles, releas-ing soluble proteins from the matrix while maintaining peripheral andintegral proteins in the membrane fraction; Triton X-114 (with0.5 mM NaCl) treatment results in separation of all soluble matrix pro-teins and some ionically bound peripheral proteins to the aqueousphase, while integral and peripheral (non-ionically bound) proteins re-main in the detergent phase; on the other hand sodium carbonate is astrong chaotropic reagent which strips the membranes from it non-integrally insertedproteins, therefore releasing all soluble and peripher-al proteins.
When T. cruzi glycosomes were treated by osmotic shock, about 95%of the specific activity of PPDK was released to the soluble fraction;however, the western blot showed that a signal for the 100 kDa PPDKform still remained in the membrane fraction, where also all signal forthe 75 kDa phosphorylated PPDK was detected (Fig. 4). This resultmay indicate that part of the total enzyme expressed in the parasiteis located at the organelle membrane. This was confirmed whenglycosomes were treated with Triton X-114, as a small quantity of the100 kDa PPDK band remained in the detergent phase and the 75 kDaphosphorylated PPDK was detected mainly in this fraction (Fig. 4).
Fig. 3. Evaluation of antibodies specificity. The antibodies anti-PPDK T. cruzi (polyclonal), anti-PPDK Z. mays and anti-phospho peptide from Z. mays PPDK were used with samples of anextract from dark-incubated leaves (DE), an extract from light-incubated leaves (LE), purified glycosomes of T. cruzi (G) and recombinant PPDK of T. cruzi (R). Proteins were separated bySDS-PAGE, blotted and then analyzed with the indicated antibodies. MW: Molecular weight.
83E. González-Marcano et al. / Parasitology International 63 (2014) 80–86
Taken together, these results point toward the presence of a 100 kDaPPDK with a dual location in the glycosomes, one part – being thegreatmajority – soluble in thematrix and the other part non-soluble as-sociated with themembrane, while the 75 kDa phosphorylated PPDK islocalized exclusively at the membrane of this organelle.
The T. cruzi PPDK sequence does not contain any predicted trans-membrane segments; therefore the enzyme found in the membranefraction must most likely be peripherally located at the glycosomal sur-face. This was verified by treating purified glycosomeswith sodium car-bonate. Indeed this resulted in the release of all PPDK (75 kDa and100 kDa) to the aqueous solution (Fig. 4), thus confirming its peripherallocalization in the glycosome.
3.4. Expression of PPDK in the three developmental forms of T. cruzi
In order to verify differential PPDK expression and the occurrence ofits post-translational modification process in the three life-cycle stagesof T. cruzi, epimastigotes, trypomastigotes and amastigotes, a westernblot using the same number of parasites per lane was prepared. Whenprobing with anti-PPDK, we found that the expression of the enzymevaries between the different developmental forms, showing the highestlevel in epimastigotes, then in trypomastigotes and the lowest inamastigotes (Fig. 5). This result differs from that reported in [22],where the authors did not observe significant variation in the expres-sion of PPDK between all three developmental forms. This discrepancy
Fig. 4. Sub-organellar localization of T. cruzi PPDK. Purified glycosomes were disrupted by diffedium carbonate. The different fractions obtained by each of the methods were separated by SDphospho threonine (α-Thr-P) and anti-phospho peptide from Z.mays PPDK (α-peptide). The saand non-soluble fraction after treatment (M). MW: Molecular weight.
might be a consequence of the different strains of T. cruzi used in eachstudy. Additionally, when using the antibody specific for the phosphor-ylated peptide and the anti-phosphorylated threonine antibody, a dif-ferential expression was observed among the three forms (Fig. 5), buta signal for the phosphorylated 75 kDa PPDK was observed in all threestages. This result then indicates that the post-translational modifica-tion of PPDK takes place in all developmental forms of T. cruzi.
3.5. Expression of PPDK along a growth curve of epimastigotes of T. cruzi
When three samples taken at three different phases of growth (pre-exponential, exponential and stationary) were analyzed by westernblot, it was observed that the expression of the 100 kDa PPDK slightlydecreaseswhen the culture reaches the stationary phase, but its specificactivity decreases about 3.25 times to an almost undetectable level(from 0.013 to 0.004 U·mg−1). Conversely, in the western blot, thephosphorylated 75 kDa PPDK signal increases significantly (Fig. 6-a).This result seems to reflect that glucose is consumed and becomesexhausted in the medium, causing the parasites to switch toward anamino acid as source of energy supply. PPDK may be redundant forthis latter form of metabolism and consequently the expression andspecific activity of the 100 kDa PPDK in the glycosome decreases. Con-currently there is an increase of itsmodified, inactive form that becomeslocated to the membrane (Fig. 6-b).
rent methods; osmotic shock, phase separation with Triton X-114 and treatment with so-S-PAGE, blotted and then analyzed with monoclonal anti-PPDK T. brucei (α-PPDKm), anti-mples analyzedwere glycosomes before treatment (G), soluble fraction after treatment (S)
Fig. 5. Expression of PPDK in different life-cycle stages of T. cruzi. Western blot probedwith the antibodies anti-PPDK T. brucei (α-PPDKm), anti-phospho threonine (α-Thr-P),anti-phospho peptide from Z. mays PPDK (α-peptide), and anti-tubulin from T. cruzi (α-Tubulin). Proteins from a total of 7 × 106 cells per lane were separated by SDS-PAGE,blotted and then analyzed using the indicated antibodies. Cultures of amastigotes (A),trypomastigotes (T) and epimastigotes (E) were used. MW: Molecular weight.
Fig. 6. Expression of PPDK along a growth curve of T. cruzi epimastigotes. From a batchgrown epimastigote culture, samples were taken at three stages: pre-exponential (1), ex-ponential (2) and stationary (3). The cells were disrupted using sample buffer, and theprotein from a total of 7 × 106 cells per lane were separated by SDS-PAGE, blotted andanalyzed with monoclonal anti-T. brucei PPDK (α-PPDKm) and anti-phospho peptidefrom Z. mays PPDK (α-peptide) (a). The film obtained by exposure of the western blotwas scanned, and the signals were quantified and plotted (b). MW: Molecular weight.
Fig. 7. In vitro phosphorylation of T. cruzi recombinant PPDK.Western blot of the samplesobtained after incubation of the recombinant PPDKwith an extract obtained from leaves ofZ. maysmaintained in the dark. Samples were taken after 0 min (T0), 30 min (T1), 60 min(T2) and 90 min (T3) of incubation. Proteins were separated by SDS-PAGE, blotted andthen probed with monoclonal anti-T. brucei PPDK (α-PPDKm), anti-phospho threonine(α-Thr-P), and anti-phospho peptide from Z. mays PPDK (α-peptide). All signals obtainedcorrespond to the 100 kDa PPDK. MW: Molecular weight.
84 E. González-Marcano et al. / Parasitology International 63 (2014) 80–86
3.6. In vitro phosphorylation of the recombinant T. cruzi PPDK
Searching similarities with the DUF299 protein system from C4plants [10] and to verify if the T. cruzi PPDK phosphorylation inactivatesthe enzyme, the recombinant T. cruzi PPDK was incubated with an ex-tract of Z. mays leaves maintained in the dark. During this incubation,the T. cruzi PPDK becomes phosphorylated (Fig. 7) and its specific ac-tivity decreases progressively in time about 1.6 fold (from 0.4 to0.25 U·ml−1). This result confirms that the PDRP (PPDK regulatoryprotein) from Z. mays recognizes the phosphorylation motif in T. cruziPPDK; and as it happens in C4 plants, in T. cruzi PPDK is inactivatedwhen phosphorylated.
4. Discussion
Pyruvate phosphate dikinase is an enzyme found in plants andmany microorganisms. It has been mainly studied in plants, whereit participates in the process of photosynthesis generating PEP, thesubstrate of phosphoenolpyruvate carboxylase (PEPC), an enzymethat fixes CO2 [8]. In all trypanosomatids analyzed so far, PPDK is aglycosomal enzyme that produces pyruvate, ATP and inorganic phos-phate from PEP, AMP and PPi [6]. At present, in T. cruzi, PPDK is theonly known enzyme with pyrophosphatase (PPiase) activity in thisorganelle, with the observation that its activity is predominantly inthe glycolytic direction, while in T. brucei under some specific condi-tions it maywork in the gluconeogenic direction producing PEP [4,6].According to this notion, glycosomal PPDKwould be an essential linkbetween glycolysis, fatty acid β-oxidation and several biosyntheticpathways, because as a PPi-hydrolyzing enzyme it makes the PPi-producing reactions thermodynamically possible and eliminates atthe same time the toxic PPi [4]. These considerations of a possiblecrucial role of PPDK in the metabolism of these parasites emphasizethe importance of the present study.
In this work, an unusual post-translational modification of the PPDKin T. cruzi is described; it is phosphorylated and proteolytically cleaved,giving rise to a phosphorylated 75 kDa PPDK form that localizes exclu-sively to the glycosomal membrane. The possibility that the 100 kDaand the 75 kDa PPDKs are products of two different genes can be firmlyexcluded, because the two genes present in the genome are virtuallyidentical, and there is no evidence for the existence of an intron in eitherof the gene copies that could explain the presence of an additionalsmaller product [22]. Additionally, when the recombinant PPDK wasphosphorylated in vitro (Fig. 7), it remained at the same MW of100 kDa, confirming that an aberrant migration of the phosphorylatedPPDK in the SDS-PAGE cannot be invoked either. We therefore consider
85E. González-Marcano et al. / Parasitology International 63 (2014) 80–86
the occurrence of a proteolytic cleavage of the 100 kDa PPDK as themost likely explanation of the result.
Here we demonstrate that PPDK is phosphorylated at a threonineresidue (Thr481) within a highly conserved motif. In Z. mays, PPDK isphosphorylated/dephosphorylated by a bifunctional protein namedPPDK regulatory protein (PDRP) containing a specific domain from theDUF299 family, which phosphorylates and dephosphorylates PPDK ina light-dependent reaction [23]. This specific family of proteins seemsto be absent in all protists including the Trypanosomatidae [24], there-fore it is uncertain which type of protein participates in the phosphory-lation and truncation of PPDK in T. cruzi. Moreover, the exact regulationof the process is still unknown and it also remains to be established inwhich compartment of the cell this modification takes place.
As in plants, the phosphorylationmotif in T. cruzi PPDK is recognizedby the PDRP from a leaf extract of Z. mays, resulting in inactivation of theenzyme, probably by the same mechanism as described for plants [25].Nevertheless, the further proteolytic cleavage of the 100 kDa PPDK ei-ther from the C-terminal or the N-terminal end of the protein will irre-versibly inactivate the enzyme, as a result of the loss of the protons andmagnesium, or the nucleotide (ATP/AMP) binding sites, respectively.Although we observed the phosphorylation of the recombinant T. cruziPPDK in vitro, we cannot establish the order by which the phosphoryla-tion and the cleavage of the protein occur.
In T. cruzi, the phosphorylation-dependent regulatorymechanism aswell as the proteolytic cleavage of PPDK is unlikely to be light–darknesscontrolled as in plants. As shown in samples taken during the growthcurve of epimastigotes, there are changes in the expression of the differ-ent PPDK forms when the metabolism changes from glucose-based toamino acid-based energy and carbon source, indicating that this regula-tory system may be a metabolism-dependent process in the parasite.This notion was supported by the observation that there were almostno changes in the levels of both PPDKs (75 and 100 kDa) when the nu-trients, including glucose, in the epimastigote culture were regularlyreplenished by replacing the medium during its batch growth (datanot shown). Nevertheless, specific details of this regulatory mechanismremain to be determined.
The most surprising finding of this work was to discover that the75 kDa phosphorylated PPDK form is associated to the glycosomemembrane, as is also the case for a portion of the 100 kDa non-phosphorylated PPDK. Bioinformatic analysis of the PPDK protein se-quence showed that it does not have any predicted transmembranesegment(s). Indeed our experiments indicate that it is bound periph-erally to the glycosomal membrane, probably interacting with an in-tegral membrane protein in order to be stabilized at this localization.At this moment, questions remain about the function of themembrane-associated PPDK, and also about to which protein it isassociated. Further experiments are necessary in order to addressthese questions.
Recently, it has been described for Saccharomyces cerevisiae thatPEX11, a peroxin belonging to the family of peroxisomemembrane pro-teins (PMPs), is phosphorylated as part of a regulatory system for per-oxisome proliferation [26,27]. Furthermore, in T. brucei a phosphatasesignaling cascade with enzymes in the glycosomes has been described,that is involved in the regulation of the parasite's differentiation [28,29].Hence, the existence of a similar reversible phosphorylationmechanismfor the regulation of T. cruzi PPDKmight be possible; as we found that inpurified glycosomes the phosphorylation signal of PPDK is lostwhen nophosphatase inhibitors are used. Although the specific mechanism ofthe process is not known at this moment, we have shown in thispaper that PPDK in T. cruzi undergoes two forms of post-translationalmodifications, phosphorylation and proteolytic cleavage, resulting ininactivation of the enzyme. We clearly demonstrated for the first timethat T. cruzi PPDK, an auxiliary enzyme of the glycolytic pathway inglycosomes of trypanosomes, is regulated by its phosphorylation linkedto changes of metabolism, and that this process takes place in all life-cycle stages of this parasite.
Acknowledgments
This work was performed in a collaboration between the VenezuelanScientific Research Institute (IVIC) and the Los Andes University (ULA),financially supported by the Science, Technology and Innovation Na-tional Fund (FONACIT) within a Biotechnology scholarship, and theMisión Ciencia Project No. 2007001425. We thank Dr. Chris Chastainfor kindly providing the anti-PPDK of Z. mays and anti-phospho peptideof PPDK antibodies, and Dr. Frédéric Bringaud for providing the mono-clonal anti-T. brucei PPDK antibody. We are grateful to Dr. Paul Michelsand Dr. Melisa Gualdrón-López for their helpful comments and revisionof the manuscript.
References
[1] Silber AM, Tonelli RR, Lopes CG, Cunha-e-Silva N, Torrecilhas ACT, Schumacher RI,et al. Glucose uptake in the mammalian stages of Trypanosoma cruzi. Mol BiochemParasitol 2009;168:102–8.
[2] Gualdrón-López M, Brennand A, Hannaert V, Quiñones W, Cáceres AJ, Bringaud F,et al.When, how andwhy glycolysis became compartmentalised in theKinetoplastea.A new look at an ancient organelle. Int J Parasitol 2012;42:1–20.
[3] Urbina JA. Intermediary metabolism of Trypanosoma cruzi. Parasitol Today1994;10:107–10.
[4] Acosta H, Bringaud F, Cáceres A, Quiñones W, Dubourdieu M, Concepción JL. Pyruvatephosphate dikinase and pyrophosphate metabolism in the glycosome of Trypanosomacruzi epimastigotes. Comp Biochem Physiol B BiochemMol Biol 2004;138:347–56.
[5] Girzalsky W, Saffian D, Erdmann R. Peroxisomal protein translocation. BiochimBiophys Acta 1803;2010:724–31.
[6] Bringaud F, Baltz D, Baltz T. Functional andmolecular characterization of a glycosomalPPi-dependent enzyme in trypanosomatids: pyruvate, phosphate dikinase. Proc NatlAcad Sci U S A 1998;95:7963–8.
[8] Hatch MD, Slack CR. A new enzyme for the interconversion of pyruvate andphosphopyruvate and its role in the C4 dicarboxylic acid pathway of photosynthesis.Biochem J 1968;106:141–6.
[9] Reeves RE. A new enzyme with the glycolytic function of pyruvate kinase. J BiolChem 1968;243:3202–4.
[10] Chastain CJ, Botschner M, Harrington GE, Thompson BJ, Mills SE, Sarath G, et al.Further analysis of maize C4 pyruvate, orthophosphate dikinase phosphoryla-tion by its bifunctional regulatory protein using selective substitutions of theregulatory Thr-456 and catalytic His-458 residues. Arch Biochem Biophys2000;375:165–70.
[11] Chastain CJ, Fries JP, Vogel JA, Randklev CL, Vossen AP, Dittmer SK, et al. Pyruvate, or-thophosphate dikinase in leaves and chloroplasts of C3 plants undergoes light–dark-induced reversible phosphorylation. Plant Physiol 2002;128:1368–78.
[12] Camargo E. Growth and differentiation in Trypanosoma cruzi. I. Origin of metacyclictrypanosomes in liquid media. Rev Inst Med Trop Sao Paulo 1964;6:93–100.
[13] Benaim G, Sanders JM, Garcia-Marchán Y, Colina C, Lira R, Caldera AR, et al.Amiodarone has intrinsic anti-Trypanosoma cruzi activity and acts synergisticallywith posaconazole. J Med Chem 2006;49:892–9.
[14] Schacterle GR, Pollack R. A simplified method for the quantitative assay of smallamounts of protein in biologic material. Anal Biochem 1973;51:654–5.
[15] Laemmli UK. Cleavage of structural proteins during the assembly of the head of bac-teriophage T4. Nature 1970;227:680–5.
[16] Ausubel F, Brent R, Kingston R, Moore D, Seidman J, Smith J, et al. Current protocolsin molecular biology. New York: Wiley; 1989.
[18] Sanz-Rodríguez CE, Concepción JL, Pekerar S, Oldfield E, Urbina JA. Bisphosphonatesas inhibitors of Trypanosoma cruzi hexokinase: kinetic and metabolic studies. J BiolChem 2007;282:12377–87.
[19] Quiñones W, Urbina J, Dubordieu M, Concepción JL. The glycosome membrane ofTrypanosoma cruzi epimastigotes: protein and lipid composition. Exp Parasitol2004;106:135–49.
[20] Bordier C. Phase separation of integral membrane proteins in Triton X-114 solution.J Biol Chem 1981;256:1604–7.
[21] Fujiki Y, Hubbard AL, Fowler S, Lazarow PB. Isolation of intracellular membranes bymeans of sodium carbonate treatment: application to endoplasmic reticulum. J CellBiol 1982;93:97–102.
[22] Maldonado RA, Fairlamb AH. Cloning of a pyruvate phosphate dikinase fromTrypanosoma cruzi. Mol Biochem Parasitol 2001;112:183–91.
[23] Burnell JN, Hatch MD. Regulation of C4 photosynthesis: purification and propertiesof the protein catalyzing ADP-mediated inactivation and Pi-mediated activation ofpyruvate, Pi dikinase. Arch Biochem Biophys 1985;237:490–503.
[25] Herzberg O, Chen CC, Kapadia G, McGuire M, Carroll LJ, Noh SJ, et al. Swiveling-domain mechanism for enzymatic phosphotransfer between remote reaction sites.Proc Natl Acad Sci U S A 1996;93:2652–7.
86 E. González-Marcano et al. / Parasitology International 63 (2014) 80–86
[26] Thoms S, Erdmann R. Dynamin-related proteins and Pex11 proteins in peroxisomedivision and proliferation. FEBS J 2005;272:5169–81.
[27] Knoblach B, Rachubinski RA. Phosphorylation-dependent activation of peroxi-some proliferator protein PEX11 controls peroxisome abundance. J Biol Chem2010;285:6670–80.
[28] Szöör B, Wilson J, McElhinney H, Tabernero L, Matthews KR. Protein tyrosine phos-phatase TbPTP1: a molecular switch controlling life cycle differentiation in trypano-somes. J Cell Biol 2006;175:293–303.
[29] Szöör B. Trypanosomatid protein phosphatases. Mol Biochem Parasitol2010;173:53–63.
