an iron regulatory-like protein expressed in plasmodium falciparum displays aconitase activity

Molecular & Biochemical Parasitology 143 (2005) 29–38

An iron regulatory-like protein expressed inPlasmodium falciparumdisplays aconitase activity

Marcus Hodgesa, Emine Yikilmazb, George Pattersonb, Ishmael Kasvosvea,Tracey A. Rouaultb, Victor R. Gordeuka, Mark Loyevskyc,∗

a The Center for Sickle Cell Disease, Howard University, Washington, DC 20059, USAb The National Institute of Child Health and Human Development, National Institutes of Health, Bethesda, MD 20892, USA

c Sanaria, Inc., 12115 Parklawn Drive, Suite L, Rockville, MD 20852, USA

Received 22 March 2005; accepted 5 May 2005Available online 31 May 2005

Abstract

Plasmodium falciparumiron regulatory-like protein (PfIRPa) has homology to both mammalian iron regulatory proteins and aconitasesa operties atl ts that thel uorescencea e parasiticc se activitya y theh rn e andh n MTT-baseda emonstratea rdst©

K

1

s

ppbfmnb

and

all of

RNAt syn-zoite

ateafter

hasyticome

0d

nd is capable of binding RNA iron response elements. We examined the subcellular localization of PfIRPa and its enzymatic prow oxygen tension. Differential digitonin permeabilization of isolated trophozoites with subsequent Western blot analysis suggesocalization of PfIRPa is predominantly in the membranous compartments of the parasite, such as the mitochondrion. Immunoflnalysis showed that PfIRPa colocalizes with heat shock protein 60 (Hsp60), a mitochondrial marker, and is also present in thytosol/food vacuole. Under conditions favoring the formation of an iron–sulfur cluster, recombinant PfIRPa (rPfIRPa) had aconitas detected by a colorimetric NADPH-MTT (3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl-2H-tetrazolium bromide) assay. As assessed bydration ofcis-aconitate spectrophotometrically at 240 nm, rPfIRPa had high affinity forcis-aconitate (Km = 3.5�M) but a low turnoveumber (Kcat=3.3 s−1). The overall catalytic efficiency (Kcat/Km) of rPfIRPa was similar in magnitude to human cytosolic IRP1/aconitasuman mitochondrial aconitase. PfIRPa immunoprecipitated from parasite lysates also had aconitase activity, as assessed by assay. Our results provide evidence that PfIRPa localizes in the mitochondrion and in the cytosol/food vacuole and is able to dconitase activity. Further understanding of the role of PfIRPa/aconitase in the regulation ofP. falciparumhomeostasis may contribute towa

he development of novel antimalarial strategies against plasmodial species.2005 Elsevier B.V. All rights reserved.

eywords:Aconitase;Plasmodium falciparum; TCA cycle; Iron

. Introduction

The major energy supply for the asexual erythrocytictage ofPlasmodium falciparumis the process of glycoly-

Abbreviations: PfIRPa,Plasmodium falciparumiron regulatory-likerotein; rPfIRPa, recombinantPlasmodium falciparumiron regulatory-likerotein; MTT, 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl-2H-tetrazoliumromide; IRP1, iron regulatory protein 1; AESBF, aminoethylbenzene sul-

onyl fluoride; Hsp60, heat shock protein 60; LSCM, laser scanning confocalicroscopy; IPTG, isopropylthiogalactoside; ICDH, isocitrate dehydroge-ase; DTT, dithiotreitol; Hb, hemoglobin; mtDNA, mitochondrial deoxyri-onucleic acid∗ Corresponding author. Tel.: +1 301 770 3222; fax: +1 301 770 5554.E-mail address:[email protected] (M. Loyevsky).

sis [1], but the classical association between glycolysisthe tricarboxylic acid (TCA) cycle is not apparent[2]. Thesequencing of the plasmodial genome has revealed thatthe major TCA cycle enzymes are present inP. falciparum[3], and gene expression profiles have shown that the mlevels of the genes encoding these enzymes are almoschronously increased during the middle and late trophostages (20–36 h post-invasion of red blood cells)[2–4]. ThemRNA levels of the trigger enzyme of the TCA cycle, citrsynthase, are maximal 45 h post-invasion, i.e. 9–15 hmaximum mRNA levels of all other TCA cycle enzymes[2].However, it has not been determined if the TCA cyclea vital role in providing energy for the asexual erythrocparasites[5]. Progress in sequencing the plasmodial gen

166-6851/$ – see front matter © 2005 Elsevier B.V. All rights reserved.oi:10.1016/j.molbiopara.2005.05.004

30 M. Hodges et al. / Molecular & Biochemical Parasitology 143 (2005) 29–38

and the search for novel antimalarial targets stimulated thecloning and partial characterization of fiveP. falciparumgenes that may be involved in the TCA cycle: malate dehy-drogenase[6], the flavoprotein and iron–sulfur subunits ofsuccinate dehydrogenase[7,8], lactate-dehydrogenase-likemalate dehydrogenase[9], isocitrate dehydrogenase[5] anda gene with homology for iron regulatory protein/aconitasegenes of other species[10,11]. The intracellular localizationof these proteins inP. falciparumremains to be completed.

The connection of the TCA cycle with glycolysis in theasexual stages ofP. falciparumis not completely evident, asexpression has been recently detected for the� and� chainsof pyruvate dehydrogenase[12], but not for dihydrolipoamideS-acetyl transferase, the typical links between glycolysis andthe TCA cycle[2]. In addition, the expression of pyruvatedehydrogenase, which is typically cytosolic in eukaryoticsystems, has been mapped to the apicoplast of plasmodialparasites[12]. The expression of TCA cycle genes is syn-chronized with the expression of a large number of genesin the mitochondrial genome, including those involved inelectron transport[2]. These data are consistent with the pos-sibility that electron transport occurs in the mitochondria ofintraerythrocyticP. falciparumparasites, which is supportedby the detection of enzymatic activities of dihydroorotatedehydrogenase, cytochromec reductase, and cytochromecoxidase in asexual stage parasites[13]. Additionally, mito-c asiteso -t lf Ha hey tess nr tasea t) arep ddi-tm res-p ry-l ina tioni

s ina erater f them rmt ofc so le. Inh oni-t otein( otein1 t ap -

malian iron regulatory proteins and has the capacity to bindspecifically to putative RNA iron response elements presentin the plasmodial system. The purpose of the present studywas to determine the subcellular localization ofP. falciparumiron regulatory-like protein (PfIRPa) and to investigate itspossible aconitase function.

2. Materials and methods

2.1. Maintenance of P. falciparum growth in culture

P. falciparum(strain 3D7) was grown in culture flaskscontaining RPMI-1640 supplemented with 25 mM HEPES,23 mM sodium bicarbonate, 10 mM glucose, 10% (v/v) heat-inactivated human plasma (0+ or A+), and washed humanerythrocytes (A+) at 2–2.5% hematocrit. The growth mediumwas replaced daily and the cultures were gassed with a mix-ture of 90% N2, 5% CO2 and 5% O2 [20]. Ring stage synchro-nization was achieved by the lysis of cells containing matureparasites with alanine or sorbitol[21]. The morphologicalcharacteristics of the parasites and the degree of parasitemiawere assessed by microscopic inspection of thin blood smearsstained by Giemsa. The separation of parasites on Percollgradient cushions according to the stage of development wasperformed as described[22].

2s

izedt ites hate-b , and0 BS,3 Ma wasa mino ubesa rpm)o1 )w d2 s ofd d,i pa-r imals red for5 S-P losem nce,K nti-P 00),a body3 i-b )a

hondrial respiration has been reported in asexual parf P. falciparum[13,14] andP. berghei[15]. The produc

ion of NADH and FADH2 in the citric acid cycle is criticaor oxidative phosphorylation, as the oxidation of NADnd FADH2 is coupled to the formation of ATP. Althougarlier studies suggested that purifiedP. falciparumandP.oelii mitochondria do not oxidize NADH-linked substrauch as�-ketoglutarate, malate, or pyruvate[16], it has beeecently demonstrated that NADH-quinone oxidoreducnd a malate-quinone oxidoreductase (FAD-dependenresent inP. yoelii yoelii asexual stage parasites. The a

ion of ADP to the digitonin-permeabilized isolatedP. yoeliiitochondria containing malate induced an increase iniration, indicating the transition from resting to phospho

ating state respiration. Oligomycin inhibited respirationgreement with the operation of oxidative phosphoryla

n theP. yoeliimitochondria[17].It has been proposed that the TCA cycle enzyme

sexual malaria parasites may serve primarily to geneducing equivalents for maintaining the redox status oitochondrion[5], to produce ATP[17] or to provide otheetabolic intermediates such as succinyl-CoA[1] or glu-

amate[18]. Aconitase, which facilitates the conversionitrate to isocitrate through the intermediatecis-aconitate, ine of the essential enzymes associated with this cycigher eukaryotes, the catalytically inactive form of ac

ase has the capacity to function as an iron regulatory prIRP-1, also known as iron response element-binding pr) [19]. Previous studies[10,11] have demonstrated tharotein expressed inP. falciparumhas homology to mam

.2. Differential digitonin permeabilization ofaponin-isolated trophozoites

Asexual intraerythrocytic parasites were synchrono the ring stage[21] and harvested at the trophozotage. The cultures were washed once with phospuffered saline (PBS), the supernatants were discarded.5 ml-erythrocytic pellets were resuspended in 5 ml of P% sucrose, 10�g/ml leupeptin and aprotinin, and 10 mminoethylbenzene sulfonyl fluoride (AESBF). Saponindded to a final concentration of 0.1% on ice, and after 5f incubation the suspension was split into eppendorf tnd centrifuged for 30 s at a maximal speed (14,000n an Eppendorf 5415R centrifuge at 4◦C. Approximately00�l of trophozoite-containing pellets (about 109 parasitesere resuspended in the same buffer to 5× 108 parasites, an0�l-aliquots were subjected to different concentrationigitonin (100�M–10 mM). The tubes were briefly vortexe

ncubated at 37◦C for 2 min, and the pellets were seated from supernatants by 30 s-centrifugation at a maxpeed. Proteins in pellets and supernatants were denatumin at 95◦C in 2× SDS dye before loading on 10% SDAGE. The proteins were then transferred to nitrocelluembranes Protran BA83 (Scheicher & Schuell BioScieeene, NH). Western blotting was performed using afIRPa 3950 serum (1:1000), anti-Hsp60 serum (1:10nd anti-enolase serum (1:1000). The anti-PfIRPa anti950 has been described previously[10]. The enolase antody was derived fromCandida albicans(Takara Bio, Inc.nd the heat shock protein 60 antibody was derived fromHeli-

M. Hodges et al. / Molecular & Biochemical Parasitology 143 (2005) 29–38 31

cobacter virescens(Stressgen Biotechnologies). The pro-teins detected by these antibodies served as cytosolic andmitochondrial markers, respectively[27,28]. The sequencecomparison searches of the antibody recognition epitopesbetween plasmodial enolase and Hsp60 sequences and thesequences from the above source species showed more than90% identity (the data not shown).

2.3. Indirect immunofluorescence analysis

To determine the intracellular localization of PfIRPa,immunostaining of PfIRPa was combined with that ofHsp60, a mitochondrial marker. Infected red blood cell sus-pensions (approximately 107 cells/ml; 10–12% parasitemia)were washed once with RPMI-1640 and then fixed to poly-l-lysine-coated microscopic slides (Electron MicroscopySciences, Ft. Washington, PA) with 4% paraformaldehyde,0.05% glutaraldehyde for 10 min at room temperature. Theslides were washed three times by submerging in PBS (5 mineach). Non-specific epitopes of cells were quenched with 2%BSA in PBS for 1 h, followed by three washes in PBS (5 mineach). The cells were probed with primary antibodies (rabbitpolyclonal�-PfIRPa 3950, 1:100, and mouse monoclonal�-Hsp60, 1:200) in blocking solution for 1 h and washed threetimes with PBS. Cells were then probed with fluorescent sec-ondary antibodies (goat�-rabbit, alexa568, 1:500, and goat� boveT unt-i torL es-c nningc inga a568a tively.

2

E3)p edaI diTa ride( Paw wasi ow-t ingb le,3 1:5,vp icol-u 8.0,2 allyb h ane M

imidazole, 300 mM NaCl). The obtained protein is labileand sensitive to oxidation, therefore the following steps wereperformed in the Coy anaerobic chamber (Coy LaboratoryProducts, Grass Lake, MI). Imidazole, which interferes withsubsequent spectrophotometric determination of aconitaseactivity, was removed from rPfIRPa by argon ultrafiltration ofrPfIRPa-containing eluate using the Millipore ultrafiltrationunit, 30 kDa cut-off filter and a buffer containing 100 mMTris–HCl, pH 7.5, 0.1% Triton X-100, 150 mM NaCl.

2.5. Measurements of aconitase activity of rPfIRPausing an indirect MTT assay

To determine if rPfIRPa has aconitase activity a modifiedprocedure of Kaptain and colleagues was utilized[23].rPfIRPa was activated with an aconitase-regeneration buffercontaining 100 mM Tris–HCl, pH 7.5, 100�M Na-citrate,0.1% Triton X-100, 140�M Fe(NH4)2(SO4)2, 120�MNa2S, 1 mM DTT for 30 min. at 20◦C. These conditionsfavor iron–sulfur cluster formation. The aconitase detectionmix consisted of 150 mM Tris–HCl, pH 7.5, 8.6 mMcis-aconitate, 60 mM MgCl2, 0.04 IU of isocitrate dehydro-genase (ICDH) per 10 ml mix, 125�M NADP, 240�M MTT(3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl-2H-tetrazoliumbromide) (Sigma, St. Louis, MO), and 80�M phenazinemethosulfate (Sigma, St. Louis, MO). If aconitase activity isp henc . Inp ichr lfate.T olorf at5 in aC etricc U640s from4

2u

oto-m ionb ga to -i afterr ithtNo atedaat gt ram( tion

-mouse alexa488, 1:500) and washed as described ahe slides were dried and mounted with VectaShield mo

ng medium with or without the nuclear DAPI stain (Vecaboratories, Inc., Burlingame, CA) before taking fluorence images. Images were collected using laser scaonfocal microscopy (LSCM). LSCM was performed usZeiss LSM410 microscope. The fluorescence of alex

nd alexa488 was registered at 568 and 488 nm, respec

.4. Purification of recombinant PfIRPa (rPfIRPa)

Expression of His-tagged rPfIRPa in the Rosetta(DLysSE. colistrain (Novagen, Madison, WI) was performs described[11]. After inducing bacterial cells with 1 mM

PTG and growing overnight at 18◦C, the cells were lysen lysis buffer [0.1 mM dithiotreitol (DTT), 40 mM KCl, 1%riton X-100, 25 mM Tris–HCl, pH 8.0, 10�g/ml leupeptinnd aprotinin, 10 mM aminoethylbenzene sulfonyl fluoAESBF)] and the lysate containing recombinant PfIRas obtained after pelleting cellular debris. The lysate

mmediately passed through a Hi-Trap column and the flhrough was collected. His-bind resin (Novagen) in binduffer (50 mM Na-phosphate pH 8.0, 10 mM imidazo00 mM NaCl) was then added to the flow-through (/v), and the suspension was incubated at 4◦C on a rotatinglatform for 3 h. The suspension was loaded onto a minmn and washed with a buffer (50 mM Na-phosphate pH0 mM imidazole, 300 mM NaCl) to remove non-specificound proteins. Recombinant PfIRPa was eluted witlution buffer (50 mM Na-phosphate pH 8.0, 250 m

. resent,cis-aconitate is converted to isocitrate, which is tonverted to�-ketoglutarate by isocitrate dehydrogenasearallel to this reaction, NADP is reduced to NADPH, wheduces MTT in the presence of phenazine methosuhis last reduction is accompanied by change of a c

rom yellow to magenta with an absorbance maximum60 nm. The reaction components were assembledoy anaerobic chamber and sealed in a spectrophotomuvette. The absorbance was measured in a Beckman Dpectrophotometer, by scanning wavelengths ranging25 to 650 nm.

.6. Measurements of aconitase activity of rPfIRPasing UV spectrophotometry

Aconitase activity of rPfIRPa was assayed spectrophetrically at 20◦C and pH 7.5 in the Coy anaerobic staty monitoring the hydration ofcis-aconitate at 240 nm, usinmolar absorption coefficient of 3600 M−1 [24]. One uni

f activity corresponds with 1�mol of cis-aconitate diminshed per minute. Aconitase activity was measurede-activating of rPfIRPa under anaerobic conditions whe buffer containing 140�M Fe(NH4)2(SO4)2, 120�Ma2S, 1 mM DTT for 30 min at 20◦C [25]. The initial ratef cis-aconitate hydration was monitored using estimmounts of rPfIRPa in 500�l 100 mM Tris–HCl, pH 7.5nd various amounts ofcis-aconitate ranging from 2�mol

o 10 mmol. Km, Vmax, and Kcat were calculated usinhe Michaelis–Menten equation and the Prizm progGraphPad Software, San Diego, CA). The concentra


of rPfIRPa was estimated using Coomassie gel stainingand a standard Bradford assay (Bio-Rad, Hercules, CA).Considering rPfIRPa’s sensitivity to oxygen and increasedinstability upon additional manipulations, the histidinetag used to facilitate purification of this protein was notremoved during aconitase detection experiments of rPfIRPa.The initial design that positioned the histidine tag at theC-terminus of the protein[10] predicted that such placementwould not result in misfolding of the protein.

2.7. Immunoprecipitation of PfIRPa from parasitelysates and determination of aconitase activity

We used a modified procedure described by Kaptain andcolleagues[23]. Hemoglobin-free extracts of the parasiteswere obtained as described above and the pellets (75–100�l)were lysed by passing them 12 times through a 26 gauge-needle with lysis buffer [10 mM Tris–HCl, pH 7.4, 100�MNa-citrate, 1% Triton X-100, 0.5% Nonidet P-40, 150 mMNaCl and a protease inhibitor cocktail (1:100 dilution)]. Thelysate (250�l volume) was tumbled with 60�l of rabbitantisera-containing antibodies 3950, 3951, and 3952[10]for 1.5 h at 4◦C. Lysate from uninfected red blood cellstumbled with antiserum 3951 was used as a negative con-trol. This lysate was obtained by freezing-thawing of 125-�laliquots of uninfected red blood cell and adjusting the finalvo hed 3t dtT uffer( -t uffer( -t T)t etec-t Mc ,1 -2 is,M uis,M tw nga ntrol.

2u

andm ablek ri-o ients.T nt ofc d, ane elletso d by

freezing and thawing the cells three times. In addition, ring-and trophozoite-containing erythrocytes were passed 10–12times through a 26-gauge hypodermic needle. The lysateswere then placed into the Amicon ultracentrifugation tubeswith a 3 kDa cut-off and centrifuged at 4◦C for 1 h to obtainclear, hemoglobin-free lysates. Citrate concentrations weremeasured immediately in the lysates obtained. In the sec-ond method, the contents of uninfected and malaria-infectederythrocytes were extracted using 1 M HClO4, followed byimmediate neutralization of the supernatants by 1 M NaOH.There was no difference in the obtained intracellular contentsusing these two methods.

3. Results

To ascertain the subcellular localization of PfIRPa, wepermeabilized infected cells with digitonin, which solubilizes

Fig. 1. Subcellular localization of PfIRPa/aconitase based on differentialdigitonin permeabilization of saponin-isolated trophozoites. Synchronizedto the ring stage intraerythrocytic parasites were subjected to differentialpermeabilization by 0–10 mM digitonin (digitonin concentration indicatedon top) and were subsequently fractionated by centrifugation. Supernatant(S) and pellet (P) fractions were run on 10% (w/v) SDS-polyacrylamidegels and were analyzed by Western blotting using anti-PfIRPa 3950 serum(1:1000), anti-Hsp60 serum (1:1000), and anti-enolase serum (1:1000). Eno-lase served as a cytosolic marker and heat shock protein Hsp60 served asa mitochondrial marker. Western blots are shown inFig. 1A and the corre-sponding quantitative data obtained by densitometric scanning are displayedin Fig. 1B.The released fraction (% protein release) is calculated by divid-ing the amount of specific protein in the supernatant by the total amount ofspecific protein in the supernatant and pellet of the respective sample.

olume to 250�l with deionized water before adding 60�lf antiserum 3951. Protein A-agarose beads were was

imes with the lysis buffer, and 30�l of beads were addeo the lysate/antisera mixture and tumbled for 1 h at 4◦C.he beads were washed five times with the washing b100 mM Tris–HCl, pH 7.5, 100�M Na-citrate, 0.1% Trion X-100) and treated with an aconitase-regeneration b100 mM Tris–HCl, pH 7.5, 100�M Na-citrate, 0.1% Trion X-100, 2�M ferrous ammonium sulfate, 1 mM DTo activate potential aconitase activity. The aconitase dion mix consisted of 150 mM Tris–HCl, pH 7.5, 8.6 mis-aconitate, 60 mM MgCl2, 0.04 IU of ICDH per 10 ml mix25�M NADP, 240�M MTT (3-(4,5-dimethylthiazol-2-yl),5-diphenyl-2H-tetrazolium bromide) (Sigma, St. LouO), and 80�M phenazine methosulfate (Sigma, St. LoO). The beads were pelleted, andA560 of the supernatanas measured. TheA560value of a tumbled sample containiprebleed control serum was used as a background co

.8. Measurements of citrate concentrations inninfected and malaria-infected erythrocytes

Citrate concentrations were measured in uninfectedalaria-infected erythrocytes using a commercially avail

it from R-Biopharm (Marshall, MI) after separating vaus stages of intraerythrocytic parasites on Percoll gradwo methods were used to isolate the intracellular conteells and the results were averaged. In the first methoqual volume of deionized water was added to packed pf uninfected and malaria-infected erythrocytes, followe


the cholesterol-rich plasma membrane at low concentrations,whereas higher concentrations are required to solubilizeorganellar membranes[26,27]. Enolase[28] and the mito-chondrial heat shock protein Hsp60[29] were used as markersfor the cytosolic and mitochondrial fractions, respectively,as described before[29]. Saponin-isolated trophozoiteswere incubated with increasing concentrations of digitonin,followed by centrifugal separation of soluble and particulatefractions. The amount of PfIRPa/aconitase and of the markerproteins in each fraction was quantified by densitometricscanning of Western blots (Fig. 1). Saponin treatment oftrophozoite-infected red blood cells was necessary to removehemoglobin, which interferes with the Western assay. Atthe same time, use of saponin with subsequent washesbefore digitonin permeabilization could result in the loss ofunknown amounts of cytosolic proteins, as evidenced by thepresence of the residual bands in the supernatants of the threeproteins studied at 0 concentration of digitonin (Fig. 1). Theabundant glycolytic cytosolic plasmodial protein, enolase,was released from isolated trophozoites at 0.1 mM digitonin,whereas a mitochondrial marker, Hsp60, and PfIRPa exitedto the supernatant almost in parallel at higher digitoninconcentration.

To visualize the localization of PfIRPa in parasitizedred blood cells, cell suspensions were probed with primaryantibodies (rabbit polyclonal�-PfIRPa 3950, 1:100 andmouse monoclonal�-Hsp60, 1:200) followed by fluorescentsecondary antibodies (goat�-rabbit, alexa568 and goat�-mouse, alexa488). The images were taken at the excitationwavelengths of 568 and 488 nm for alexa568 (red) andalexa488 (green), respectively. The images were processedand analyzed using Adobe Photoshop 7.0.Fig. 2 showslight (A) and fluorescent (B–D) microscopic images of atrophozoite-infected cell probed with the aforementionedantibodies. Upon immunofluorescent staining with thePfIRPa-specific antisera, the localization of PfIRPa (Fig. 2C)coincided with that of Hsp60 (Fig. 2D) and also appearsin the cytosol/food vacuole. The cytosolic localization ofPfIRPa/aconiatse was also implied from our previous work[10]. Mitochondrial colocalization of the antibody signals,indicated by a yellow fluorescence, was clearly observed(Fig. 2B). Double immunofluorescent staining with PfIRPaanti-serum 3950 and antibodies detecting the mitochondrialmatrix protein Hsp60 confirmed successful antibody accessto the mitochondrial matrix under the given permeabilizationconditions.

FaawflE

ig. 2. Fluorescence-assisted localization of PfIRPa in the trophozoite-infecntibodies [(rabbit anti-PfIRPa 3950) and (mouse anti-Hsp60)] followed by flulexa488)] as described in Section2. The mitochondrial compartment of the ias stained in red (Fig. 2C). Light microscopic image of the same trophozoiteuorescence (yellow,Fig. 2B) is indicative of a mitochondrial localization of PM—erythrocyte plasma membrane, PPM—parasite plasma membrane, PV

ted erythrocyte. Trophozoite-infected red blood cells were probedwith primaryorescent secondary antibodies [(goat anti-rabbit, alexa568) and (goat anti-mouse

ntraerythrocytic trophozoite was stained in green (Fig. 2D), whereas PfIRPa-infected red blood cell is shown inFig. 2A. Colocalization of red and green

fIRPa, but the protein is also visualized in the parasite cytosol/food vacuole.M—parasitophorous vacuole membrane, H—hemozoin, M—mitochondrion.


Fig. 3. Coomassie staining (lanes 1–4, 11) and Western blot analysis (lanes5–10, 12) ofE. coli lysate proteins (10�g/lane) transformed with the pET28(+)-PfIRPa plasmid. Lanes: 1-MW markers; 2, 5, 8-uninduced trans-formed Rosetta(DE3)pLysSE. coli; 3, 6, 9-transformedE. coli induced with0.2 mM IPTG, 4, 7, 10-transformedE. coli induced with 1 mM IPTG; 11,12–2.5�g of purified protein. Arrow indicates the position of the full-lengthPfIRPa. Lanes 5–7, 12-anti-PfIRPa antibody; lanes 8–10-anti-His antibody.

The expression and purification of recombinant PfIRPa asdescribed in Section2 resulted in the obtaining of an almostpure protein displaying one band on a Coumassie gel (Fig. 3,lane 11) and a major predominant band upon labeling withrabbit PfIRPa antiserum 3950 ([10], Fig. 3, lane 12). Theprotein is not stable and tends to precipitate upon exposureto the ambient air and upon freezing-thawing.

Fig. 4. Recombinant PfIRPa displays aconitase activity. The reaction com-ponents were assembled in a Coy anaerobic chamber as described in Section2 and the absorbance was measured in a sealed spectrophotometric cuvette,by scanning from 425 to 650 nm in a Beckman DU640 spectrophotometerafter anaerobic incubation at 22◦C for 30 min. Curve 1: 0.015 U/ml aconi-tase (Sigma, St. Louis, MO); curve 2: rPfIRPa activated with Fe-S-DTT( trol:e ingH ve 6:r nase.

The absorbance of an aconitase-detection mixture con-taining recombinant PfIRPa was measured at 560 nm afterincubation for 30 min. at 22◦C. Fig. 4 shows that rPfIRPadisplayed aconitase activity under conditions favoring theassembly of an iron–sulfur cluster. The transfection control,buffer control and reaction control (Fig. 4, traces 4–6) didnot show any appreciable absorbance at 560 nm, whereasrPfIRPa that was not exposed to elemental iron and reducingconditions (Fig. 4, trace 3) showed only trace activity.

To confirm that recombinant PfIRPa has aconitase activity,we used a UV spectrophotometric assay at 240 nm[24]. Theinitial rates ofcis-aconitate hydration were plotted againstdifferent concentrations ofcis-aconitate (Fig. 5), from whichthe values forKm, Vmax, andKcat were calculated. Resultsare presented inTable 1 along with the values reportedfor human IRP1/aconitase, human mitochondrial aconitaseand bacterial aconitases A and B. Recombinant PfIRPa

Fig. 5. Kinetic analysis of recombinant PfIRPa withcis-aconitate as sub-strate. Histidine-tagged PfIRPa was expressed inE. coli and purified asdescribed in Section2. Kinetic measurements with activated recombinantprotein were performed withcis-aconitate as substrate, and disappearanceof cis-aconitate was monitored spectrophotometrically at 240 nm. Usinga molar absorption coefficient of 3600 M−1, the initial velocities ofcis-aconitate hydration in units/mg protein were plotted againstcis-aconitateconcentration. TheKm was calculated from the nonlinear regression of theMichaelis–Menten graph. The bottom graph shows the transformation ofdata into the linear form in Lineweaver–Burke inverse coordinates.

rPFIRP+); curve 3: no activation (rPfIRPa-); curve 4: transfection conluate from RosettaE. coli transfected with the pET28a+ vector but lackis-tagged PfIRPa; curve 5: eluate control: imidazole elution buffer; cur

eaction control: aconitase detection mix lacking isocitrate dehydroge


Table 1Kinetic properties of various aconitases withcis-aconitate as a substrate

Parameter PfIRPaa HsmAcnb HscAcn/IRP1b ECAcnAc ECAcnBc

Km (�M) 3.5± 1 6.5± 1.4 31± 14 57± 3.6 18± 2.0Kcat (s−1) 3.3± 0.5 8.0± 0.3 24.0± 0.6 19.7± 0.5 69.9± 2.5Kcat/Km (s−1 �M−1) 1.0 1.2 0.8 0.3 3.9

Activities were measured withcis-aconitate as substrate in 100 mM Tris–HCl, pH 7.4, using 2.5�g rPfIRPa and 500-�l cuvette. HsmAcn—human mitochondrialaconitase, HscAcn/IRP1—human cytosolic aconitase/IRP1, ECAcnA—E. coliaconitase A, ECAcnB-E. coliaconitase B.

a Mean± S.D. of three triplicate measurements.b Mean± S.E. of three triplicate measurements[44].c Mean± S.D. of three triplicate measurements as previously calculated[25].

showed high affinity forcis-aconitate (Km = 3.5�M) but alow turnover number (Kcat= 3.3 s−1) compared to humancytosolic IRP1/aconitase, human mitochondrial aconitase,andE. coli aconitases A and B. The overall catalytic effi-ciency of PfIRPa, measured as theKcat/Km ratio, was of thesame order as human cytosolic IRP1/aconitase and humanmitochondrial aconitase.

PfIRPa present in parasite hemoglobin-free lysates pre-pared from trophozoite-stage parasites was immunoprecipi-tated on protein A-agarose beads using three available rabbitantisera-containing antibodies 3950, 3951, and 3952[10].Afterwards, the MTT-based assay similar to that describedabove was used to probe for the presence of aconitase activ-ity. A pre-immune control serum and lysate from uninfectederythrocytes were used as background controls. Aconitaseactivity was detected with antisera 3951 and 3952, but not

F sated wereo bitP bleeds itha beadsw elyw acti-v bufferd ss itional3pa

Fig. 7. Citrate concentrations in uninfected, ring, and trophozoite-infectederythrocytes. Citrate concentrations were measured using a kit from R-Biopharm after separating various stages of intraerythrocytic parasites onPercoll gradients. Parasite lysates were obtained either by freezing-thawingof cells three times with the subsequent ultracentrifugation through the 3 kDacut-off Amicon tubes, or by extracting the parasite contents using 1 M HClO4

followed by immediate neutralization with 1 M NaOH. The results of sevenmeasurements (±S.D.) are shown.

with antiserum 3950 (Fig. 6). The inability to demonstrateactivity with antiserum 3950 might be related to steric hin-drances produced by the latter antibody on the active centerof the immobilized protein.

Citrate concentrations were measured in uninfected ery-throcytes and in malaria-infected erythrocytes separated onPercoll gradients. The concentration of citrate (mean± S.D.,n= 7) was 56.1± 26.2�mol/l in uninfected erythro-cytes, 38.7± 8.9�mol/l in ring-stage erythrocytes and2.5± 3.7�mol/l in trophozoite-stage erythrocytes (Fig. 7).

4. Discussion

Our previous cloning and expression of plasmodialIRP (PfIRPa) in a bacterial expression system facilitatedits characterization as an RNA-binding protein[11]. Ourpresent results indicate that PfIRPa is localized partiallyin the plasmodial mitochondrion and partially in thecytosol/food vacuole, and that under appropriate condi-

ig. 6. PfIRPa immunoprecipitated from intraerythrocytic parasite lyisplays aconitase activity. Hemoglobin-free extracts of the parasitesbtained as described in Section2, and lysates were tumbled with rabfIRPa-antisera-containing antibodies 3950, 3951, and 3952 or preerum for 1.5 h at 4◦C. Lysate from uninfected red blood cells tumbled wntiserum 3951 was used as a negative control. Protein A-agaroseere then tumbled with lysates for 1 h at 4◦C. The beads were extensivashed with a Na-citrate-containing buffer, and aconitase activity wasated for 30 min at room temperature with an aconitase-regenerationescribed in Section2. The aconitase detection mix includedcis-aconitate aubstrate. The results shown in the figure were obtained after an add
0 min incubation in the aconitase detection mix. At this time, the beads wereelleted, andA560 of the supernatant was measured. Each bar represents anverage of two duplicate measurements of one experiment performed.
tions this molecule undergoes the necessary changes todisplay aconitase activity. The reduction of intracellular


citrate concentration upon maturation of malaria parasitesin infected red blood cells suggests thatP. falciparumparasites may demonstrate a significant degree of aconitaseactivity.

The cytosolic versus mitochondrial localization of anumber of TCA cycle proteins inP. falciparumis a matterof controversy. At least three of the TCA-cycle plasmodialenzymes were previously found to be localized in the cytosol:P. falciparummalate dehydrogenase[6] and aconitase[10]andP. knowlesiICDH [30]. Our search of the sequencedplasmodial genome (PlasmoDB) for the presence of a secondaconitase inP. falciparum failed to yield additional can-didates. Our previous immunofluorescence data suggestedthat PfIRPa/aconitase is significantly localized within theparasite cytosol in the trophozoite-infected erythrocyte[10].However, selective permeabilization of saponin-isolatedtrophozoites shows that PfIRPa exited to the supernatantalmost in parallel to a mitochondrial marker, Hsp60, athigher concentrations of digitonin (Fig. 1). Additionally,further improvements in immunofluorescence technique(Fig. 2) in our present study indicate that PfIRPa/aconitaseis colocalized with the mitochondrial marker, Hsp60, inthe mitochondrion of the asexual erythrocytic parasite, withanother portion of PfIRPa/aconitase being located in thecytosol/food vacuole. Such localization is also demonstratedin another protozoan parasite,T. brucei, in which a singlea riona oft sureos rkerp entp -o isl par-e thee edingt selyb selyb nin-t as itw tarica izedi blep Pac oft rti theRr bilei

tedf tasea donewA trate

the presence of aconitase activity in these lysates[10],PfIRPa has striking sequence similarity towards aconitasesfrom various species. Our lack of finding aconitase activityin our previous work may have been related to ineffectiveimmunoprecipitation or inhibition of aconitase activity bythe antibody. In addition, the iron–sulfur cluster indispens-able for aconitase activity[24] may not have been assembledbecause of superoxide radicals present in the incubationmedium, when reconstitution of aconitase activity wasattempted in ambient air[32].

In contrast to our previous work, further purification ofPfIRPa (Fig. 3) and testing of its properties in anaerobicconditions in the present study allowed us to demonstratethat this malarial protein can indeed function as an aconitase(Figs. 4 and 5, Table 1). The comparison of the kineticproperties of rPfIRPa/aconitase with those of several otheraconitases (Table 1) indicate that rPfIRPa/aconitase has veryhigh affinity for cis-aconitate (Km is 3.5�M), low turnovernumber, and catalytic efficiency (Kcat/Km) that is betweenthat of human mitochondrial aconitase and cytosolic (IRP1)aconitase, but lower than that ofE. coli aconitase B. Thesecomparisons are only valid for recombinant PfIRPa. Thekinetic properties of innate PfIRPa were not assessed andmight not be identical, based on the comparison of kineticproperties of native[33] and recombinant[5] isocitratedehydrogenase (ICDH) fromP. falciparum. The low Kv ntwtT thants n beu hostc e. Att asitec , isa

en-s , asc ationr coni-t ani trateo P( ).S ar toP isp entlys gralp edi aryf em sef sn sn rom

conitase functions and localizes both in the mitochondnd in the cytosol[27]. Upon saponin permeabilization

rophozoite-infected red blood cells and further expof the permeabilized parasites to digitonin (Fig. 1), PfIRPaeems to exit the cells similar to the mitochondrial marotein (Hsp60), without much of the cytosolic componresent. However, our previous data[10] and immunoflurescence staining (Fig. 2) show that PfIRPa/aconitase

ocalized in the cytosol/food vacuole as well. This apnt discrepancy may emerge from the differences inmployed techniques: saponin permeabilization prece

he digitonin treatment can result in a loss of the looound cytosolic pool of PfIRPa/aconitase. This looound fraction of the protein would not be seen in sapo

reated parasites with additonal digitonin treatment,as washed from the isolated parasites. In turn, the gluldehyde/paraformaldehyde fixation of unpermeabil

nfected red blood cells would firmly retain all the availaools of PfIRPa/acoinitase. The cytosolic function of PfIRould conceivably include the RNA-binding propertieshis protein[11]. The dual function of PfIRPa we repon this study implies that the switch from aconitase toNA-binding conformation is present inP. falciparum. It

emains to be established if this switch is mediated by laron, as in higher eukaryotes[31].

In our previous work, PfIRPa was immunoprecipitarom Hb-free lysates and the reconstitution of aconictivity was attempted in these lysates, as previouslyith human IRP1/aconitase[23] and described in Section2.lthough we reported that we were unable to demons

malue of PfIRPa/aconitase forcis-aconitate is consisteith the rather low concentrations of citrate (2.5± 3.7�M)

hat we found in trophozoite-infected erythrocytes (Fig. 7).hese concentrations are approximately 22 times less

hose found in uninfected erythrocytes (56.1± 26.2�M). Iteems possible that the aconitase activity of PfIRPa cased by the intraerythrocytic parasite to metabolize theitrate from the ring stage until the late schizogony staghis latter stage, approximately 45 h post-invasion, paritrate synthase, the trigger enzyme of the TCA cyclectivated[2].

The mitochondrial TCA cycle begins with the condation of oxaloacetate and acetyl-CoA to form citrateatalyzed by citrate synthase. The dehydration–rehydreactions catalyzed by aconitase occur subsequently. Aase reversibly converts citrate to isocitrate throughntermediate,cis-aconitate. Isocitrate serves as a subsf isocitrate dehydrogenase (ICDH), which uses NAD+

or NAD+) as a cofactor to produce NADPH (NADHearches conducted in PlasmoDB indicate that, similfIRPa/aconitase, only one NADPH-dependent ICDHresent in the plasmodial genome. Hence, it was recuggested that ICDH in malaria could either be an inteart of the mitochondrial TCA cycle or else be involv

n providing NADPH for reductive reactions necessor the parasite’s survival[5]. Aconitase activity thereforight be necessary to supply isocitrate to ICDH. BecauP.alciparum ICDH is clearly NADP+-dependent and showo activity with NAD+ [5,18], one possibility is that it iot an essential part of the parasite’s TCA cycle. F


this viewpoint, both ICDH and aconitase could be onlymarginally involved in energy metabolism but rather beimportant for maintenance of the mitochondrion’s redoxstate. The fact that ICDH transcript and protein levels are up-regulated in oxidatively stressed parasites seems to supportthis view[5].

In addition to maintaining the redox status of the mito-chondrion [5] to favor production of ATP via oxidativephosphorylation [17] or through alternative oxidationpathways[17,34], other important proposed functions of theTCA cycle in asexual parasites include synthesis of severalcritical metabolic intermediates, such as the precursors ofheme biosynthesis, succinyl-CoA[1] or glutamate[18].Regardless of the actual role(s) of the TCA cycle in asexualparasites, aconitase activity is essential for the functioningof this cycle. Further work is needed to establish whetherplasmodial aconitase activity can be inhibited selectively andwhether silencing of PfIRPa/aconitase is lethal for asexualparasites.

Many TCA cycle proteins may play additional cellularroles beyond their catalytic roles within this pathway[35]. Several TCA cycle proteins, including aconitase, haverecently been identified in nucleoids with mitochondrial DNA(mtDNA) and have been reported to play an active role in sta-bilizing mtDNA in yeast[36]. Yeast isocitrate dehydrogenasebinds to mitochondrially encoded mRNAs and appears to reg-u tsi g am res-p inc ehy-d defeca ids,s ot e oft enase[

achg thea unc-t nes,ia cts,a twop , pre-d wasd on ofa efects[

lariap andt thisa ands ucha thed

Acknowledgements

The authors acknowledge the grant support provided inpart by grants RO1-A144857-05 from the National Insti-tute of Allergy and Infectious Diseases, UH1-HL03679-07from The National Heart, Lung and Blood Institute and theOffice of Research on Minority Health, and Howard Univer-sity GCRC grant MO1-RR10284-06. The authors are gratefulto Drs. John R. Guest and John C. Wootton for fruitful dis-cussions and insightful help. The technical assistance of Ms.Tiffany N. Johnson is appreciated.

References

[1] Sherman IW. In: Sherman IW, editor. Malaria: Parasite Biology,Pathogenesis, and Protection. Washington, DC: American Societyfor Microbiology; 1998.

[2] Bozdech Z, Llinas M, Pulliam BL, Wong ED, Zhu J, DeRisi JL.The transcriptome of the intraerythrocytic developmental cycle ofPlasmodium falciparum. PLoS Biol 2003;1:E5.

[3] Gardner MJ, Hall N, Fung E, et al. Genome sequence of the humanmalaria parasitePlasmodium falciparum. Nature 2002;419:498–511.

[4] Le Roch KG, Zhou Y, Blair PL, et al. Discovery of gene functionby expression profiling of the malaria parasite life cycle. Science2003;301:1503–8.

[5] Wrenger C, Muller S. Isocitrate dehydrogenase ofPlasmodium fal-ciparum. Eur J Biochem 2003;270:1775–83.

[6] Lang-Unnasch N. Purification and properties ofPlasmod-sitol

se inof

avo-sitol

ttana-sitol

[lac--

[ inive

[ a-itol

[ ad-

plast.

[ a P,sex-

[ ito-tagest

[ pho-aria

late their translation[37]. In turn, RNA binding also affecsocitrate dehydrogenase catalytic activity, suggestin

echanism for coordinating expression of mitochondrialiratory complexes and TCA cycle oxidative metabolismS.erevisiae[38]. In humans, the fumarase and succinate drogenase genes act as tumor suppressors. Fumarasere associated with dominantly inherited uterine fibrokin leiomyomata, and renal cell cancer[39], whereas twypes of brain tumors are caused by mutations in threhe genes encoding subunits of succinate dehydrog40–42].

When disruptive mutations were introduced into eene of the yeast TCA cycle at a time, mutations inconitase gene led to the largest abnormality in cellular f

ions as compared to mutations in other TCA cycle gencluding changes in mRNA levels of 54 genes[43]. Thebnormalities included altered heme levels, mtDNA defend a slower growth rate. These 54 genes fell intoredominant categories. Expression of aerobic genesominantly of the mitochondrial respiratory complexes,iminished by defects in aconitase, whereas expressinaerobic and hypoxic genes was elevated with these d

43].Further studies of PfIRPa/aconitase-knockout ma

arasites could help to elucidate the role of aconitasehe TCA cycle in the asexual erythrocytic stages ofpicomplexan. Better understanding of the functionpecific properties of plasmodial TCA cycle enzymes ss PfIRPa/aconitase could facilitate the search forevelopment of new antimalarials.

ts

ium falciparum malate dehydrogenase. Mol Biochem Para1992;50:17–25.

[7] Takeo S, Kokaze A, Ng CS, et al. Succinate dehydrogenaPlasmodium falciparummitochondria: molecular characterizationthe SDHA and SDHB genes for the catalytic subunits, the flprotein (Fp) and iron–sulfur (Ip) subunits. Mol Biochem Para2000;107:191–205.

[8] Suraveratum N, Krungkrai SR, Leangaramgul P, PrapunwaP, Krungkrai J. Purification and characterization ofPlasmodium falciparum succinate dehydrogenase. Mol Biochem Para2000;105:215–22.

[9] Chan M, Sim TS. Functional characterization of an alternativetate dehydrogenase-like] malate dehydrogenase inPlasmodium falciparum. Parasitol Res 2004;92:43–7.

10] Loyevsky M, LaVaute T, Allerson CR, et al. An IRP-like protefrom Plasmodium falciparumbinds to a mammalian iron-responselement. Blood 2001;98:2555–62.

11] Loyevsky M, Mompoint F, Yikilmaz E, et al. Expression ofrecombinant IRP-likePlasmodium falciparumprotein that specifically binds putative plasmodial IREs. Mol Biochem Paras2003;126:231–8.

12] Foth BJ, Stimmler LM, Handman E, Crabb BS, Hodder AN, McFden GI. The malaria parasitePlasmodium falciparumhas only onepyruvate dehydrogenase complex, which is located in the apicoMol Microbiol 2005;55:39–53.

13] Learngaramkul P, Petmitr S, Krungkrai SR, PrapunwattanKrungkrai J. Molecular characterization of mitochondria in aual and sexual blood stages ofPlasmodium falciparum. Mol CellBiol Res Commun 1999;2:15–20.

14] Krungkrai J, Burat D, Kudan S, Krungkrai S, Prapunwattana P. Mchondrial oxygen consumption in asexual and sexual blood sof the human malarial parasite,Plasmodium falciparum. SoutheasAsian J Trop Med Public Health 1999;30:636–42.

15] Uyemura SA, Luo S, Moreno SN, Docampo R. Oxidative phosrylation, Ca2+ transport, and fatty acid-induced uncoupling in malparasites mitochondria. J Biol Chem 2000;275:9709–15.


[16] Fry M, Beesley JE. Mitochondria of mammalianPlasmodiumspp.Parasitology 1999;102(Pt 1):17–26.

[17] Uyemura SA, Luo S, Vieira M, Moreno SN, Docampo R. Oxida-tive phosphorylation and rotenone-insensitive malate- and NADH-quinone oxidoreductases inPlasmodium yoelii yoeliimitochondriain situ. J Biol Chem 2004;279:385–93.

[18] Chan M, Sim TS. RecombinantPlasmodium falciparumNADP-dependent isocitrate dehydrogenase is active and harbours a unique26 amino acid tail. Exp Parasitol 2003;103:120–6.

[19] Klausner RD, Rouault TA. A double life: cytosolic aconitase as aregulatory RNA binding protein. Mol Biol Cell 1993;4:1–5.

[20] Trager W, Jensen JB. Continuous culture of human malaria parasites.Science (Washington, DC) 1976;193:673–5.

[21] Lambros C, Vanderberg JP. Synchronization ofPlasmodium falci-parumerythrocytic stages in culture. J Parasitol 1979;65:418–20.

[22] Fernandez V, Treutiger CJ, Nash GB, Wahlgren M. Multiple adhesivephenotypes linked to rosetting binding of erythrocytes inPlasmodiumfalciparummalaria. Infect Immun 1998;66:2969–75.

[23] Kaptain S, Downey WE, Tang C, et al. A regulated RNA-bindingprotein also possesses aconitase activity. Proc Natl Acad Sci USA1999;88:10109–13.

[24] Kennedy MC, Emptage MH, Dreyer JL, Beinert HJ. The roleof iron in the activation-inactivation of aconitase. Biol Chem1983;258:11098–105.

[25] Jordan PA, Tang Y, Bradbury AJ, Thomson AJ, Guest JR. Biochem-ical and spectroscopic characterization of Escherichia coli aconitases(AcnA and AcnB). Biochem J 1999;344(Pt 3):739–46.

[26] Zhang JW, Luckey C, Lazarow PB. Three peroxisome proteinpackaging pathways suggested by selective permeabilization ofyeast mutants defective in peroxisome biogenesis. Mol Biol Cell1993;4:1351–9.

[ evel--1 isa

[ cto-r-Biol

[ Rte-

[ cific-ool

[31] Rouault T, Klausner R. Regulation of iron metabolism in eukaryotes.Curr Top Cell Regul 1997;35:1–19.

[32] Gardner PR, Fridovich I. Inactivation-reactivation of aconitase inEscherichia coli. A sensitive measure of superoxide radical. J BiolChem 1992;267:8757–63.

[33] Vander Jagt DL, Hunsaker LA, Kibirige M, Campos NM. NADPHproduction by the malarial parasitePlasmodium falciparum. Blood1989;74:471–4.

[34] Mi-Ichi F, Takeo S, Takashima E, et al. Unique properties of respi-ratory chain inPlasmodium falciparummitochondria. Adv Exp MedBiol 2003;531:117–33.

[35] Jeffery CJ. Moonlighting proteins. Trends Biochem Sci1999;24:8–11.

[36] Kaufman BA, Newman SM, Hallberg RL, Slaughter CA, PerlmanPS, Butow RA. In organello formaldehyde crosslinking of proteinsto mtDNA: identification of bifunctional proteins. Proc Natl AcadSci USA 2000;97:7772–7.

[37] Elzinga SD, Bednarz AL, van Oosterum K, Dekker PJ, Griv-ell LA. Yeast mitochondrial NAD-dependent isocitrate dehydroge-nase is an RNA-binding protein. Nucleic Acids Res 1993;21:5328–31.

[38] Anderson SL, Minard KI, McAlister-Henn L. Allosteric inhibition ofNAD-specific isocitrate dehydrogenase by a mitochondrial mRNA.Biochemistry 2000;39:5623–9.

[39] Tomlinson IP, Alam NA, Rowan AJ, et al. Germline mutations in FHpredispose to dominantly inherited uterine fibroids, skin leiomyomataand papillary renal cell cancer. Nat Genet 2002;30:406–10.

[40] Baysal BE, Ferrell RE, Willett-Brozick JE, et al. Mutations inSDHD, a mitochondrial complex II gene, in hereditary paragan-glioma. Science 2000;287:848–51.

[41] Niemann S, Muller U. Mutations in SDHC cause autosomal domi-

[ ci-milialenet

[ nnlicpres-Cell

[ JM.con-buthem

27] Saas J, Ziegelbauer K, von Haeseler A, Fast B, Boshart M. A dopmentally regulated aconitase related to iron-regulatory proteinlocalized in the cytoplasm and in the mitochondrion ofTrypanosombrucei. J Biol Chem 2000;275:2745–55.

28] Polidori E, Saltarelli R, Ceccaroli P, et al. Enolase from the emycorrhizal fungusTuber borchii Vittad.: biochemical characteization, molecular cloning, and localization. Fungal Genet2004;41:157–67.

29] Hausler T, Stierhof YD, Wirtz E, Clayton C. Import of a DHFhybrid protein into glycosomes in vivo is not inhibited by the folaanalogue aminopterin. J Cell Biol 1996;132:311–24.

30] Sahni SK, Saxena N, Puri SK, Dutta GP, Pandey VC. NADP-speisocitrate dehydrogenase from the simian malaria parasitePlasmodium knowlesi: partial purification and characterization. J Protoz1992;39:338–42.

nant paraganglioma, type 3. Nat Genet 2000;26:268–70.42] Astuti D, Latif F, Dallol A, et al. Gene mutations in the suc

nate dehydrogenase subunit SDHB cause susceptibility to fapheochromocytoma and to familial paraganglioma. Am J Hum G2001;69:49–54.

43] McCammon MT, Epstein CB, Przybyla-Zawislak B, McAlister-HeL, Butow RA. Global transcription analysis of Krebs TricarboxyAcid Cycle mutants reveals an alternating pattern of gene exsion and effects on hypoxic and oxidative genes. Mol Biol2003;14:958–72.

44] Brazzolotto X, Gaillard J, Pantopoulos K, Hentze MW, MoulisHuman cytoplasmic aconitase (Iron regulatory protein 1) isverted into its [3Fe–4S] form by hydrogen peroxide in vitrois not activated for iron-responsive element binding. J Biol C1999;274:21625–30.

an iron regulatory-like protein expressed in plasmodium falciparum displays aconitase activity

Documents