Phosphatase activity characterization on the surface of intactbloodstream forms of Trypanosoma brucei

Eloise Cedro Fernandes a;�, Jose¤ Mauro Granjeiro b, Eula¤zio Mikio Taga b,Jose¤ Roberto Meyer-Fernandes c, Hiroshi Aoyama d

a Departamento de Bioqu|¤mica, Universidade Federal de Sa‹o Paulo (UNIFESP), 04023900 Sa‹o Paulo, SP, Brazilb Departamento de Cie“ncias Biolo¤gicas, Faculdade de Odontologia de Bauru, Universidade de Sa‹o Paulo (USP), Bauru, SP, Brazil

c Departamento de Bioqu|¤mica, Instituto de Ciencias Biologicas, Universidade Federal do Rio de Janeiro (UFRJ), Rio de Janeiro, RJ, Brazild Departamento de Bioqu|¤mica, Instituto de Biologia, Universidade Estadual de Campinas (Unicamp), 13083-970 Campinas, SP, Brazil

Received 1 November 2002; accepted 24 January 2003

First published online 15 February 2003

Abstract

Procyclic forms of Trypanosoma brucei possess a phosphatase activity on their external cell surface. This activity, while itdephosphorylates [32P]phosphocasein, is inhibited weakly by NaF and tartrate but strongly by vanadate. In this work, we describe thepresence of an external phosphatase activity in intact bloodstream forms of T. brucei. With p-nitrophenyl phosphate (pNPP) as substrate,these intact cells produced 3^5 nmol pNP min31 mg31, linearly for up to at least 30 min. The activity was not significantly increased byMg2þ, Mn2þ, Ca2þ and Co2þ, but was inhibited by vanadate, NaF, p-chloromercuribenzoate and Zn2þ and was insensitive to okadaicacid. Membrane-enriched fractions of parasites contained an acid phosphatase activity, with a pH optimum in the range of 4.5^5.5. Thisactivity hydrolyzed phosphotyrosine (40 nmol phosphate min31 mg31) better than phosphothreonine or phosphoserine. Partialpurification of this phosphatase yielded a single activity band following gel electrophoresis, a Km value of 0.29 mM with pNPP and wasinsensitive to the Fe2þ/H2O2/ascorbate system.< 2003 Federation of European Microbiological Societies. Published by Elsevier Science B.V. All rights reserved.

Keywords: Phosphotyrosine phosphatase; Bloodstream forms; Trypanosoma brucei

1. Introduction

Phosphorylation and dephosphorylation are fundamen-tal pathways that regulate a wide variety of cellular events[1]. Tyrosine phosphorylation in vivo is regulated by pro-tein tyrosine kinases, which catalyze tyrosine phosphory-lation, and protein tyrosine phosphatases (PTPs), whichare responsible for dephosphorylation [2]. Cell membraneenzymes in which the active site faces the external mediumrather than the cytoplasm can be assayed using live cells

(reviewed in [3]). Of particular interest are ecto-enzymespresent on the outer surface of the membrane of blood-stream parasites since they can mediate parasite^host in-teractions.

Trypanosoma brucei is a protozoan parasite that passesthrough several extracellular cycles in its mammalian hostsand tsetse £y vector. These cycles involve changes in theparasite’s metabolism, morphology and membrane com-position. Several protein kinases have been described inT. brucei [4^7] with the appearance of the protein phos-phorylation networks probably representing an ancientevent in trypanosomes [8]. The di¡erentiation of the para-site from bloodforms to procyclic forms was apparentlyaccompanied by increased tyrosine phosphorylation [9]. Incontrast, little is known about protein phosphatases andtheir function in the life-cycle of T. brucei [10^12]. In othertrypanosomatids, such as Leishmania donovani [13], phos-phatase activity is uniformly distributed in the plasmamembrane and in the vicinity of the £agellar pocket.Some of these enzymes have been puri¢ed [14] and because

0378-1097 / 03 / $22.00 < 2003 Federation of European Microbiological Societies. Published by Elsevier Science B.V. All rights reserved.doi :10.1016/S0378-1097(03)00091-0

* Corresponding author. Present address: Centro de Gene¤tica, BiologiaMolecular e Fitoqu|¤mica, Instituto Agrono“mico de Campinas, Av. Theo-dureto de A. Camargo 1500, 13.075-630, Campinas, SP, Brazil. Tel. : +55(19) 32415188; Fax: +55 (19) 32423602.

E-mail address: eloise@cec.iac.br (E. Cedro Fernandes).

Abbreviations: EDTA, ethylenediaminetetraacetic acid; CDTA, trans-1,2-diaminocyclohexane-N,N,NP,NP-tetraacetic acid; pCMB, p-chloro-mercuribenzoate; pNPP, p-nitrophenyl phosphate ; pNP, p-nitrophenol

FEMSLE 10871 19-3-03

FEMS Microbiology Letters 220 (2003) 197^206

www.fems-microbiology.org

FEMSLE 10871 19-3-03

E. Cedro Fernandes et al. / FEMS Microbiology Letters 220 (2003) 197^206198

they are excreted, it has been suggested that they may beuseful as a marker of virulence [15]. The cloning and char-acterization of a protein phosphatase from Leishmaniachagasi showed that this protein was conserved amongLeishmania and eukaryotic serine/threonine protein phos-phatases [16].Since most of studies of Trypanosoma phosphatases

have been done either with crude trypanosome lysates orwith puri¢ed enzymes, we have investigated the behaviorof these enzymes in intact cells. The results presented hereindicate that intact and alive bloodstream forms ofT. brucei possess ecto-phosphatase activities able to hydro-lyze the phosphotyrosine analog p-nitrophenyl phosphate(pNPP) and phosphotyrosine. In addition, this membrane-associated acid phosphatase was found to be resistant toinactivation by oxygen metabolites, as also reported forLeishmania phosphatase [17].

2. Materials and methods

2.1. Parasites

Bloodstream forms of the monomorphic clone T. bruceibrucei MITat 1.4 were harvested at mid-exponential phasefrom infected rats and isolated by DEAE-cellulose (What-man DE-52) chromatography [18]. T. brucei brucei procy-clic forms ILTar 2 were grown at 28‡C in SDM-79 me-dium. The parasites were collected, washed and kept in 50mM Tris^HCl, pH 7.2, containing 20 mM KCl and 100mM sucrose.

2.2. Preparation of the membrane fraction

Parasite membrane-enriched fractions were prepared asdescribed by Seyfang and Duszenko [19] and stored at3130‡C. Protein concentrations were determined by thebiuret method [20] using bovine serum albumin as stan-dard.

2.3. Partial puri¢cation of the membrane phosphatase

Phosphatase was puri¢ed from cell concentrates as de-scribed by Seyfang and Duszenko [19]. Brie£y, a ghostfraction was obtained by hypotonic lysis, washed severaltimes and the protein fraction was homogenized in acetatebu¡er, pH 5.0, and Triton X-114. After di¡erential centri-fugation, the supernatant was collected and concentrated

under positive pressure with nitrogen in an Amicon ultra-¢ltration system using PM-10 membranes. Enzyme activitywas assayed by pNPP hydrolysis and purity was assessedby sodium dodecyl sulfate^polyacrylamide gel electropho-resis (SDS^PAGE).

2.4. Phosphatase assay

Unless otherwise speci¢ed, pNPP hydrolysis was as-sayed as previously described [21]. Reactions were initiatedby the addition of intact cells, membrane-enriched frac-tions or partially puri¢ed enzyme and stopped by theaddition of 1 ml of 1 M NaOH. trans-1,2-Diaminocyclo-hexane-N,N,NP,NP-tetraacetic acid (CDTA) or ethylene-diaminetetraacetic acid (EDTA) was used to measure ac-tivity in the absence of divalent cations. The p-nitrophenol(pNP) released was measured based on the increase in ab-sorbance at 425 nm (the extinction coe⁄cient used for thep-nitrophenolate ion was 1.75U104 M31 cm31). Speci¢cactivity was expressed as nmol of pNP released min31

mg31 of protein. The hydrolysis of phosphorylated aminoacids was measured under the same conditions as forpNPP. The free inorganic phosphate released was deter-mined according to Lowry and Lopez [22]. Phosphataseactivity was assayed in the absence of metal ions by in-cluding 1 mM CDTA. Phosphatase activity was also de-tected in gels as described by Gomez [23]. Brie£y, afterelectrophoresis in non-denaturing conditions, any activitywas detected by incubating the gels with 50 mM L-naph-tylphosphate and 1 mg Fast-blue BB ml31 in 100 mMacetate bu¡er, pH 5.0, at 37‡C.

2.5. Chemicals

All chemicals were obtained from Sigma Chemical (St.Louis, MO, USA).

3. Results

A phosphatase activity not dependent on divalent cati-ons has been described on the external surface of intact,live procyclic forms of T. brucei (0.67 nmol pNP min31

mg31)[21]. As shown here, infectious, intact and livebloodstream forms of this parasite also have externalphosphatase activity able to hydrolyze pNPP at pH 8.0and 37‡C (3^5 nmol pNP min31 mg31). Fig. 1A comparesthe phosphatase activities of intact T. brucei procyclic

6

Fig. 1. Comparison of the phosphatase activity in procyclic and bloodstream forms of T. brucei. A: Time-dependent hydrolysis of pNPP by intact para-sites of procyclic forms at pH 7.2 and 30‡C without metals (line a) and bloodstream forms at pH 8.0 and 37‡C (line b), assayed as described in Section2. B: Phospho-amino acid hydrolysis by bloodstream forms of T. brucei. 1 mg of bloodstream membrane fraction was incubated with 60 mM sodiumcitrate, pH 5.5, containing 44 mM NaCl, 20 mM KCl, 55 mM glucose and 10 mM of phospho-amino acids, at 37‡C. Dephosphorylation of 10 mMP-Tyr was measured at pH 5.5 or at pH 8.0; higher activity was observed at pH 5.5 (inset). C: Phosphatase activity towards pNPP (0.01^30 mM) wasmeasured during 30 min using 1 mg of protein. Line a, intact bloodstream forms at 37‡C; line b, membrane fraction from bloodstream forms, 37‡C;line c, intact procyclic forms without metals at 30‡C. Inset shows the double reciprocal plot for line b.

FEMSLE 10871 19-3-03

E. Cedro Fernandes et al. / FEMS Microbiology Letters 220 (2003) 197^206 199

Fig. 2. Characterization of phosphatases in intact T. brucei. A: E¡ect of pH on T. brucei phosphatase activity. 1 mg of membrane fraction of blood-stream forms was incubated for 30 min at 37‡C with 10 mM pNPP at di¡erent pH values, as described in Section 2. B: E¡ect of divalent cations onphosphatase activity of intact procyclic and bloodstream forms of T. brucei. The assays were done at pH 7.2 and 30‡C (procyclics) or pH 8.0 and 37‡C(bloodstreams) using 1 mg of T. brucei protein ml31 and 10 mM of pNPP. C: E¡ect of classic phosphatase inhibitors on phosphatase activity of intactprocyclic and bloodstream forms. Intact procyclic and bloodstream forms (1 mg ml31) were incubated at 37‡C in the presence of 1 mM of di¡erentphosphatase inhibitors as described in Section 2. Other concentrations are indicated in the ¢gure.

FEMSLE 10871 19-3-03

E. Cedro Fernandes et al. / FEMS Microbiology Letters 220 (2003) 197^206200

forms and intact, infective bloodstream forms of T. bruceitoward pNPP. Fig. 1B shows the ability of membranefractions to hydrolyze phosphothreonine (1.3 V 0.1 nmolphosphate min31 mg31), phosphoserine (4.0 V 1.3 nmolphosphate min31 mg31) and especially phosphotyrosine(20.0 V 0.7 nmol phosphate min31 mg31) at pH 5.5. Thephosphatase activity towards phospho-amino acids couldnot be measured using intact infectious cells (see Section4). To con¢rm the acidic nature of the enzyme(s) towardsphosphotyrosine, this activity was tested at acidic and al-kaline pH and was found to be greater at pH 5.5 (Fig. 1B,inset). The phosphatase activity of intact, infectious blood-stream forms showed no saturation with pNPP up to atleast 30 mM pNPP (Fig. 1C, line a). The bloodstreammembrane fractions showed a Michaelis^Menten behaviorfor pNPP hydrolysis, with a Vmax of 3.5 nmol pNP mg31

min31 and a Km of 1.5 V 0.4 mM, at pH 8.0 (Fig. 1C, lineb). When tested at pH 5.5, Km was 0.40V 0.12 mM pNPP(curves not shown). Line c represents the phosphatase ac-tivity of intact procyclic forms in the presence of increas-ing amounts of substrate. The inset shows the double re-ciprocal transformation for line b.To avoid the possibility that secreted phosphatases or

cytoplasmic phosphatases released from dead parasitescould be responsible for the pNPP hydrolysis, the samereactions were run with pNPP in reaction medium inwhich live cells had been pre-incubated for 60 min andthen removed by centrifugation. No pNPP hydrolysiswas detected in the resulting supernatants (data notshown). The viability of bloodstream cells was assessedby incubating parasites at the same concentration in me-dium for the same time and then counting the cells in aNeubauer chamber. All cells remained viable and 80%retained their mobility. Rats inoculated with these cellsdeveloped sleeping sickness within 3 days, con¢rmingthat the parasite virulence was also retained.Since the bloodstream forms of T. brucei cannot survive

at pH values below 7.4, it was necessary to use membrane-enriched fractions to examine the e¡ect of pH values onthe phosphatase activity. Fig. 2A shows high bloodstreamphosphatase activity at pH 4.5^5.5 and an abrupt fall inactivity above or below this range. Fig. 2B compares thee¡ect of divalent cations that caused an opposed e¡ect onthe phosphatase activity of intact bloodstream and procy-clic forms. Since the e¡ects of Zn2þ have been describedfor other membrane phosphatases [24], we examined theinhibition by Zn2þ in more detail : the activity in intactbloodstream parasites was inhibited by 50% by 0.05 mMZn2þ whereas 3 mM Zn2þ inhibited the activity by 90%.Table 1 shows the in£uence of other divalent cations on

the phosphatase activity of intact procyclic and blood-stream forms of T. brucei. CDTA and EDTA (1 mM)were ¢rst tested in the control assays. Mn2þ and Co2þ

increased the phosphatase activity of intact procyclicforms more than in bloodstream intact forms. Ca2þ, anessential divalent cation for Ca-calmodulin phosphatases,stimulated the phosphatase activity of intact bloodstreamforms by 50% but had no e¡ect on that of intact procyclicforms up to a concentration of 3 mM. Fe2þ, Sr2þ, Ni2þ

and Ba2þ did not cause signi¢cant alterations (6 1%) inthe control activity of both bloodstream and procyclicforms, even at a concentration of 5 mM (data not shown).Fig. 2C compares the e¡ect of phosphatase inhibitors

on the phosphatase activities of live procyclic forms in theabsence of divalent metals and of live bloodstream formswithout the addition of cations to the assay. At 1 mM,NaF inhibited 40% of the enzyme activity in intact procy-clic forms and 50% of the control phosphatase activity ofthe bloodstream forms. Levamizole (1 mM), a well-knowninhibitor of alkaline phosphatase, had little e¡ect on thephosphatase activity of intact procyclic or bloodstreamforms. A higher concentration of levamizole increasedthe phosphatase activities of both forms (data not shown).Tetramizole (1 mM) inhibited 40% of the procyclic phos-

Table 1E¡ect of divalent cations on phosphatase activity present in intact procyclic and bloodstream forms of T. brucei

Additions Final concentration Percentage of control activity

(mM) Intact procyclic forms Intact bloodstream forms

None ^ 100 100CDTA 1 50 80EDTA 1 50 80Mg2þ 1 250 110

3 300 120Mn2þ 1 300 150

5 400 150Co2þ 1 200 110

3 270 140Ca2þ 1 100 130

5 150 145Fe2þ 1 100 90

3 100 85

The assays were done as described for each form in Section 2, using 1 mg of intact parasites. The results are the mean of three experiments, correctedfor appropriate blanks.

FEMSLE 10871 19-3-03

E. Cedro Fernandes et al. / FEMS Microbiology Letters 220 (2003) 197^206 201

phatase activity but had no e¡ect on the phosphatase ac-tivity of bloodstream forms; at higher concentrations, tet-ramizole activated the enzyme (data not shown). Tartrate,an inhibitor of the secreted phosphatase of L. donovani[25], did not a¡ect the enzyme in intact procyclic formsbut decreased the activity in intact bloodstream forms to70% of the control activity. p-chloromercuribenzoate(pCMB; 1 mM) an inhibitor of SH group-dependent en-zymes, decreased the procyclic phosphatase activity to50% and, at the same concentration, lowered the enzymeactivity of bloodstream forms to less than 10% of the

control activity (50% inhibition was achieved with 1 WMpCMB; data not shown). Okadaic acid (10 WM), a toxinused to identify serine^threonine phosphatase, did not af-fect this phosphatase activity. Vanadate, a potent inhibitorof acid phosphatases and phosphotyrosine phosphatases,caused 90% inhibition of the phosphatase activity in bothforms of intact parasite. The IC50 for enzyme inhibitionwas 0.5 WM for the procyclic forms and 0.7 WM for thebloodstream forms.After puri¢cation, the membrane-bound acid phospha-

tase was subjected to gel electrophoresis with subsequent

94

68

43

30

24

a b c d e kDa

Fig. 3. Phosphatase activity and kinetic parameters of the enzyme puri¢ed from T. brucei bloodstream forms. A: Non-denaturating PAGE; lane a, gelslice with 30 Wg of protein was incubated with L-naphtylphosphate as substrate at pH 5.0 [23]. Under the same non-denaturing conditions, proteinstaining revealed two bands at pH 5.0 (lane b, 30 Wg of protein; lane c, 50 Wg of protein) and pH 7.5 (lane d, 50 Wg of protein). Lane e, SDS^PAGEshowing a major band at 95^100 kDa. B: E¡ect of incubation time on the release of pNP. The activity was assayed using 10 mM pNPP at 37‡C, as de-scribed in Section 2. C: E¡ect of pH on the enzyme puri¢ed from T. brucei, bloodstream forms. The activity was assayed using 10 mM pNPP at 37‡C,as described in Section 2. D: E¡ect of pNPP concentration on enzyme behavior, showing the Michaelis^Menten behavior. Inset : double reciprocal plot,with a Km of 0.29 mM for pNPP.

FEMSLE 10871 19-3-03

E. Cedro Fernandes et al. / FEMS Microbiology Letters 220 (2003) 197^206202

detection of activity in situ [23] : under non-denaturingconditions, the puri¢ed enzyme showed a single band ofphosphatase activity (Fig. 3A, lane a) coincident with theband detected by Coomassie blue staining (lanes b and cwith 30 and 50Wg of protein, respectively). Electrophoresisunder the same conditions, but at pH 8.0, revealed twomain bands, suggesting that this heterogeneity may be dueto protein glycosylation (lane d, 30 Wg of protein). SDS^PAGE showed a major band of approximately 95^100kDa (lane e, 50 Wg of protein). Fig. 3B shows that theenzymatic reaction was linear with time, at least up to60 min. The activity was also linearly dependent on theamount of protein used (data not shown). The puri¢edenzyme had a pH optimum between 4.5 and 5.5 (Fig.3C) with an apparent Km of 0.29V 0.03 mM pNPP (Fig.3D) and a Vmax of 0.74 Wmol pNPP min31 mg31 (Fig. 3D,inset).As with the intact parasites, the puri¢ed enzyme was not

a¡ected by Ca2þ, Mg2þ and Co2þ (Fig. 4A) or by 5 mMEDTA or CDTA (data not shown). However, Zn2þ andCu2þ (1.0 mM) inhibited the activity by 40% and 60%,respectively. Of the inhibitors tested (Fig. 4B), sodiumvanadate, NaF, molybdate and, to a lesser extent,pCMB were the strongest inhibitors. High concentrationsof tartrate, cysteine, dithiothreitol and guanosine, de-scribed as activators of low molecular mass PTPs, didnot a¡ect the enzyme activity. The IC50 for vanadate atpH 5.0 was 0.8 mM (Fig. 5B, inset). Among the otherphosphatase substrates tested (Fig. 4C), pyrophosphateand phosphotyrosine provided the best results ; glycero-phosphate and phosphoserine were also hydrolyzed at alesser extent; no hydrolysis was observed when 6-phos-phoglucose was tested (data not shown).To test the hypothesis proposed by Gottlieb et al. [13],

we examined the resistance of puri¢ed enzyme to toxicoxygen metabolites produced by a Fe2þ/H2O2/ascorbatesystem. Fig. 5 shows that ectophosphatase puri¢ed frombloodstream forms of T. brucei remained active towardspNPP after a 15 min exposure. Under the same conditions,a low molecular mass PTP, extracted from bovine kidney[26], used as a control, lost 50% of its activity.

4. Discussion

To our knowledge, this is the ¢rst demonstration of aphosphatase activity on the surface of intact procyclic andbloodstream forms of T. brucei. The present results main-tained their attention on the behavior of the phosphataseactivity of the entire and live parasites because we wereinterested in the possible role of this enzyme in the cell^cellinteractions. That phosphatase activity could not be mea-sured in the supernatants of cells, even after a 60 minincubation, indicates that the pNPP hydrolysis seen wasnot caused by broken cells. As also shown by phase con-trast microscopy, control cells retained their motility. Sim-

ilarly, the phosphatase activity detected in live procyclicand bloodstream forms showed a linear increase in prod-uct formation with time, during at least 30 min. If cell lysiswere occurring, the increase in pNP-formation wouldprobably have been non-linear since more and more cellswould be dying during the assay.Comparison of the kinetic parameters of the procyclic

and bloodstream forms indicated that: (i) the phosphataseactivity was more intense in live bloodstream forms (3^5nmol of pNP min31 mg31) than in live procyclic forms inthe absence of metals (0.67 nmol of pNP min31 mg31) orin the presence of Mg2þ 3 mM (1.5 nmol of pNP min31

mg31) ; (ii) in both forms, the phosphatase activity wasunequivocally acid, although the activities in (i) were mea-sured at pH 7.2^8.0; (iii) in contrast to live bloodstreamforms, the phosphatase activity in intact procyclic formsand in the membrane-enriched fraction showed a Michae-lis^Menten behavior, with a Km of 0.40 mM pNPP at pH5.5 and 2.4 mM pNPP at pH 8.0; (iv) the bloodstreamform phosphatase activity was resistant to inactivation byoxygen metabolites; and (v) the greatest distinction be-tween these phosphatase activities was the capacity ofbloodstream forms to hydrolyze P-Tyr whereas no activitytoward P-amino acids was detected in the procyclic forms.Some of the parameters described here were similar to

those reported by Seyfang and Duszenko [19], includingthe acidic nature of the enzyme and the sensitivity to in-hibition by tartrate, NaF and vanadate. Other parameters,like the metal interference, could not be discussed becausethere were no data about metal in that work.This is not the ¢rst time that acid phosphatases have

been described attached to the external surface of themembrane of a bloodstream parasite [11,27,28]. Tosombaet al. [28] used cytochemical procedures to demonstratethe presence of an acid phosphatase on the surface mem-brane of T. congolense. The low optimum pH and thesurface location of these enzymes suggest a role in anacidic microenvironment and/or a close relationship withlysosomal digestion and the £agellar pocket membrane.Indeed, procyclic forms cause a marked acidi¢cation ofthe medium after 36 h in culture medium.To emphasize the stage speci¢city of the phosphatase in

bloodstream forms, the experiments were done simulta-neously with live procyclic and bloodstream forms. Theprocyclic forms were unable to hydrolyze phosphotyrosineand the procyclic phosphatase was dependent on divalentmetals, particularly Mg2þ, in a manner similar to typePP2C tyrosine phosphatase [1]. On the other hand, thephosphatase of bloodstream forms was unable to hydro-lyze phosphoserine and phosphothreonine and was inhib-ited by divalent cations such as Cu2þ and Cd2þ, a ¢ndingcon¢rmed by using puri¢ed enzyme (Fig. 3A). The addi-tion of 1 mM EDTA decreased the phosphatase activityby 50% in the procyclic forms and by 80% in intact blood-stream forms.The Km (0.4 mM) for the phosphotyrosine analog

FEMSLE 10871 19-3-03

E. Cedro Fernandes et al. / FEMS Microbiology Letters 220 (2003) 197^206 203

pNPP, obtained using bloodstream membrane fractions atpH 5.5, was the same as that described by Schell [29] forthe £agellum bag of T. brucei. The speci¢c activities ofthese two enzymes were very similar: 43.4 mmol pNPP

min31 mg31 [29] and 43.6 mmol pNPP min31 mg31 (thiswork).Trypanosomes do not secrete acid phosphatases,

although exceptions have been reported for T. cruzi [30]

A B

C

Fig. 4. Dephosphorylation of 10 mM pNPP by puri¢ed enzyme in the presence of metals and phosphatase inhibitors. A: 1 mg of protein was incubatedfor 15 min at 37‡C as described in Section 2. The concentrations of the metals used are indicated in the legend. B: 1 mg of protein was incubated for15 min at 37‡C in the presence of di¡erent inhibitors, as described in Section 2. The inhibitor concentrations are indicated in the legend. The e¡ect ofdi¡erent concentrations of vanadate on the puri¢ed phosphatase is indicated in the inset. C: Potential substrates for the puri¢ed enzyme. The enzymeactivity was determined using 10 mM of each substrate and an incubation time of 60 min. The inset shows the Michaelis^Menten behavior of the en-zyme towards P-Tyr.

FEMSLE 10871 19-3-03

E. Cedro Fernandes et al. / FEMS Microbiology Letters 220 (2003) 197^206204

and L. donovani [13]. The phosphatase of intact blood-stream forms showed a pattern of general inhibition notvery di¡erent from that described for the procyclic forms,i.e. inhibition by NaF, vanadate, and tartrate, but not bylevamizole and tetramizole. pCMB markedly inhibited thephosphatase of bloodstream forms, indicating that thisenzyme was SH-dependent, with a cysteine residue essen-tial for enzyme activity and/or stability.Using the Lowry and Lopez method [22] and phospho-

tyrosine as substrate, the free phosphate released by livebloodstream parasites was observed to disappear quicklyin the reaction medium, giving negative results when com-pared to the assay blank. On the other hand, with increas-ing phosphate concentrations, the live bloodstream formsreduced the inorganic phosphate concentration at a rate of5 pmol phosphate min31 (data not shown). This behaviorwas not observed with the procyclic forms.Gottlieb and Dwyer [13] and Saha et al. [17] tested the

resistance of Leishmania phosphatase to inactivation byoxygen metabolites. The toxicity of H2O2 to bloodstreamforms of T. brucei and the inhibitory e¡ect on parasiteoxygen consumption and pyruvate output have been de-scribed [31]. Since T. brucei bloodstream forms lack cata-lase and glutathione peroxidase (for review, see [31]), analternative H2O2-detoxi¢cation pathway has been pro-posed [32,33]. Because hemo£agellate parasites are ex-posed to toxic oxygen metabolites derived from drug me-tabolism or immune mechanisms, it is necessary todetermine whether the parasite enzymes are resistant to

oxidative stress. Such a resistance could help to explainthe survival of the parasite in the host circulatory systemwithout a need to invade host cells to escape the host’simmune system. Just as the parasite developed an alterna-tive system of escape, the variant surface glycoproteins, it ispossible that the escape mechanism of the oxidative stressalso brings us new surprises.

Acknowledgements

The authors thank Dr. Roberto Docampo for the try-panosomes and Dr. Silvia Moreno for helpful comments.This work was supported by FAPESP (FundacXa‹o de Am-paro a' Pesquisa do Estado de Sa‹o Paulo) and CNPq(Conselho Nacional de Pesquisa e Desenvolvimento Tec-nolo¤gico).

References

[1] Cohen, P. (1982) The role of protein phosphorylation in neural andhormonal control of cellular activity. Nature 296, 613^619.

[2] DePierre, J.W. and Karnovsky, M.L. (1973) Plasma membrane ofmammalian cells. A review of methods for their characterizationand isolation. J. Cell Biol. 56, 275^286.

[3] Zhang, Z.-Y. (2001) Protein tyrosine phosphatases: prospects fortherapeutics. Curr. Opin. Chem. Biol. 5, 416^423.

[4] Keith, K., Hide, G. and Tait, A. (1990) Characterization of proteinkinase C-like activities in T. brucei. Mol. Biochem. Parasitol. 43, 107^116.

[5] Parsons, M., Valentine, M., Deans, J., Schieven, G.L. and Ledbetter,J.A. (1990) Distinct patterns of tyrosine phosphorylation during thelife cycle of T. brucei. Mol. Biochem. Parasitol. 45, 241^248.

[6] Wheeler, A. and Shapiro, S.Z. (1992) Evidence of tyrosine kinaseactivity in the protozoan parasite T. brucei. J. Protozool. 39, 413^416.

[7] Hide, G., Graham, T., Buchanan, N., Tait, A. and Keith, K. (1994)Trypanosoma brucei : characterization of protein kinases that arecapable of autophosphorylation in vitro. Parasitology 108, 161^166.

[8] Grellier, P., Blum, J., Santana, J., Bylen, E., Mouray, E., Sinov, V.,Teixeira, A.R.L. and Schevel, J. (1999) Involvement of calyculinA-sensitive phosphatase(s) in the di¡erentiation of Trypanosoma cruzitrypomastigotes to amastigotes. Mol. Biochem. Parasitol. 98, 239^252.

[9] Parsons, M., Ledbetter, J.A., Schieven, G.L., Nel, A.E. and Kanner,S.B. (1994) Developmental regulation of pp44/46, tyrosine-phosphor-ylated proteins associated with tyrosine/serine kinase activity in Try-panosoma brucei. Mol. Biochem. Parasitol. 63, 69^78.

[10] Hendriks, E., van Deursen, F.J., Wilson, J., Sarkar, M., Timms, M.and Matthews, K.R. (2000) Life-cycle di¡erentiation in Trypanosomabrucei : molecules and mutants. Biochem. Soc. Trans. 28, 531^536.

[11] Bakalara, N., Santarelli, X., Davis, C. and Baltz, T. (2000) Puri¢ca-tion, cloning and characterization of an acid ectoprotein phosphatasedi¡erentially expressed in the infectious bloodstream forms of Trypa-nosoma brucei. J. Biol. Chem. 275, 8863^8871.

[12] Chaudhuri, M. (2001) Cloning and characterization of a novel serine/threonine protein phosphatase type 5 from Trypanosoma brucei. Gene266, 1^13.

[13] Gottlieb, M. and Dwyer, D.M. (1981) L. donovani surface membraneacid phosphatase activity of promastigotes. Exp. Parasitol. 52, 117^128.

Fig. 5. Tolerance of the puri¢ed phosphatase to oxidative stress. E¡ectof the Fe2þ/H2O2/ascorbate system on the activity of acid phosphatasepuri¢ed from bloodstream forms and on bovine kidney low-molecular-mass PTP. The enzyme was pre-incubated for di¡erent periods with 0.2mM Fe2þ and 10 mM ascorbate, in 0.1 M acetate bu¡er, pH 5.0 at37‡C. Oxidation was initiated by adding 14.5 mM H2O2. At the indi-cated times, the reaction was stopped by adding 1 mM EDTA. The re-sidual activity was determined using pNPP as substrate, as described inSection 2 and a reaction time of 60 min. Control activity (100%) corre-sponded to activity in the absence of oxidation.

FEMSLE 10871 19-3-03

E. Cedro Fernandes et al. / FEMS Microbiology Letters 220 (2003) 197^206 205

[14] Remaley, A.T., Das, S., Campbell, P.I., LaRocca, G.M., Pope, M.T.and Glew, R.H. (1985) Characterization of L. donovani acid phos-phatases. J. Biol. Chem. 260, 880^886.

[15] Singla, N., Khuller, G.K. and Vinayak, V.K. (1992) Acid phospha-tase activity of promastigotes of L. donovani a marker of virulence.FEMS Microbiol. Lett. 94, 221^226.

[16] Burns, J.M., Parsons, M., Rosman, D.E. and Reed, S.G. (1993) Mo-lecular cloning and characterization of a 42-kDa protein phosphataseof L. chagasi. J. Biol. Chem. 268, 17155^17161.

[17] Saha, A.K., Das, S., Glew, R.H. and Gottlieb, M. (1985) Resistenceof leishmanial phosphatases to inactivation by oxygen metabolites.J. Clin. Microbiol. 22, 329^332.

[18] Cross, G.A.M. (1984) Structure of the variant glycoproteins and sur-face coat of T. brucei. Proc. R. Soc. Lond. 307, 3^12.

[19] Seyfang, A. and Duszenko, M. (1993) Functional reconstitution ofthe T. brucei plasma-membrane D-glucose transporter. Eur. J. Bio-chem. 214, 593^597.

[20] Gornall, A.G., Bardawill, C.J. and David, M.M. (1949) Determina-tion of serum proteins by means of the biuret reaction. J. Biol. Chem.177, 751^766.

[21] Fernandes, E.C., Meyer-Fernandes, J.R., Silva-Neto, M.A.C. andVercesi, A.E. (1997) Trypanosoma brucei : ecto-phosphatase presenton the surface of intact procyclic forms. Z. Naturforsch. C 52, 351^358.

[22] Lowry, H.O. and Lopez, J.A. (1946) The determination of inorganicphosphate in the presence of labile phosphate esters. J. Biol. Chem.162, 421^428.

[23] Gomez, J.J. (1978) An improved method for phosphatase detection.Ann. Enzymol. 32, 44^47.

[24] Cherno¡, J. and Li, H.-C. (1985) A major phosphotyrosyl-protein

phosphatase from bovine heart is associated with a low molecular-weight acid phosphatase. Arch. Biochem. Biophys. 240, 135^145.

[25] Lovelace, J.K. and Gottlieb, M. (1986) Comparison of extracellularacid phosphatases from various isolates of Leishmania. Ame J. Trop.Med. Hyg. 35, 1121^1128.

[26] Granjeiro, J.M., Taga, E.M. and Aoyama, H. (1997) Puri¢cation andcharacterization of a low-molecular-weight bovine kidney acid phos-phatase. Ann. Acad. Bras. Cien. 69, 451^460.

[27] McLaughlin, J. (1986) The association of distinct acid phosphataseswith the £agella pocket and surface membrane fractions obtainedfrom bloodstream forms of T. rhodesiense. Mol. Cell. Biochem. 70,177^184.

[28] Tosomba, O.M., Coetzer, T.H.T. and Londale-Eccles, D. (1996) Lo-calisation of acid phosphatase activity on the surface of bloodstreamforms of T. congolense. Exp. Parasitol. 84, 429^438.

[29] Schell, D., Stierhof, Y.D. and Overath, P. (1990) Puri¢cation andcharacterization of a tartrate-sensitive acid phosphatase of Trypano-soma brucei. FEBS Lett. 271, 67^70.

[30] Nakamura, K.H., Tachibana, H. and Kanada, Y. (1985) Alterationon cell surface acid phosphatase concomitant with the morphologicaltransformation in T. cruzi. Comp. Biochem. Physiol. 81B, 815^817.

[31] Penketh, P.G. and Klein, R.A. (1986) Hydrogen peroxide metabolismin T. brucei. Mol. Biochem. Parasitol. 20, 111^121.

[32] Henderson, G.B., Fairland, A.H. and Cerami, A. (1987) Trypano-thione-dependent peroxide mechanism in C. fasciculata and T. brucei.Mol. Biochem. Parasitol. 24, 39^45.

[33] El-Sayed, N.M.A., Alarcon, C.M., She⁄eld, V.C. and Donelson, J.E.(1995) cDNA expressed sequence tags of T. rhodesience provide newinsights into the biology of the parasite. Mol. Biochem. Parasitol. 44,116^118.

FEMSLE 10871 19-3-03

E. Cedro Fernandes et al. / FEMS Microbiology Letters 220 (2003) 197^206206