269

Braz J Med Biol Res 33(3) 2000

Signal transduction in Trypanosoma cruzi metacyclic formsBrazilian Journal of Medical and Biological Research (2000) 33: 269-278ISSN 0100-879X

Signal transduction induced inTrypanosoma cruzi metacyclictrypomastigotes during theinvasion of mammalian cells

Departamentos de 1Microbiologia, Imunologia e Parasitologia, and2Biofísica, Escola Paulista de Medicina, Universidade Federal de São Paulo,São Paulo, SP, Brasil

N. Yoshida1,S. Favoreto Jr.1,

A.T. Ferreira2 andP.M. Manque1

Abstract

Penetration of Trypanosoma cruzi into mammalian cells depends onthe activation of the parasite�s protein tyrosine kinase and on theincrease in cytosolic Ca2+ concentration. We used metacyclic trypo-mastigotes, the T. cruzi developmental forms that initiate infection inmammalian hosts, to investigate the association of these two eventsand to identify the various components of the parasite signal transduc-tion pathway involved in host cell invasion. We have found that i) boththe protein tyrosine kinase activation, as measured by phosphorylationof a 175-kDa protein (p175), and Ca2+ mobilization were induced inthe metacyclic forms by the HeLa cell extract but not by the extract ofT. cruzi-resistant K562 cells; ii) treatment of parasites with the ty-rosine kinase inhibitor genistein blocked both p175 phosphorylationand the increase in cytosolic Ca2+ concentration; iii) the recombinantprotein J18, which contains the full-length sequence of gp82, ametacyclic stage surface glycoprotein involved in target cell invasion,interfered with tyrosine kinase and Ca2+ responses, whereas the mon-oclonal antibody 3F6 directed at gp82 induced parasite p175 phospho-rylation and Ca2+ mobilization; iv) treatment of metacyclic forms withphospholipase C inhibitor U73122 blocked Ca2+ signaling and im-paired the ability of the parasites to enter HeLa cells, and v) drugs suchas heparin, a competitive IP3-receptor blocker, caffeine, which affectsCa2+ release from IP3-sensitive stores, in addition to thapsigargin,which depletes intracellular Ca2+ compartments and lithium ion,reduced the parasite infectivity. Taken together, these data suggest thatprotein tyrosine kinase, phospholipase C and IP3 are involved in thesignaling cascade that is initiated on the parasite cell surface by gp82and leads to Ca2+ mobilization required for target cell invasion.

CorrespondenceN. Yoshida

Departamento de Microbiologia,

Imunologia e Parasitologia

EPM, UNIFESP

R. Botucatu, 862, 6º andar

04023-062 São Paulo, SP

Brasil

Fax: +55-11-571-1095

E-mail: nyoshida.dmip@epm.br

Research supported by FAPESP

(No. 94/3496-4). P.M. Manque and

S. Favoreto Jr. were recipients

of CAPES fellowships.

Received August 3, 1999

Accepted January 3, 2000

Key words· Trypanosoma cruzi· Signal transduction· Metacyclic trypomastigotes· Protein kinase· Calcium response· Cell invasion

Introduction

To enter mammalian cells, the protozoanparasite Trypanosoma cruzi, the etiologicalagent of Chagas� disease in man, requires theactivation of signal transduction pathways

both in the parasite and the host cell (1-7). Inthe process of cell invasion, the engagementof cell surface receptors is thought to triggerthe signaling cascade that mobilizes varioussecond messengers. Mobilization of one ofsuch messengers, namely the calcium ion,

270

Braz J Med Biol Res 33(3) 2000

N. Yoshida et al.

promotes in the host cell the rearrangementof microfilaments (8), the recruitment oflysosomes to the site of T. cruzi entry (9) andparasite internalization (10). The flow ofinformation from the cell surface to the in-tracellular components, leading to an in-crease in cytosolic free Ca2+ concentration,possibly occurs in the following manner.Binding of the infective trypomastigote formsto the target cell induces in the latter theactivation of pertussis toxin-sensitive heter-otrimeric G protein (1). This, coupled to theactivity of phospholipase C, generates inosi-tol 1,4,5-triphosphate (IP3) that participatesin Ca2+ release from IP3-sensitive intracellu-lar stores (8).

As to the signaling events in T. cruziduring mammalian cell invasion, the avail-able data are still fragmentary. Upon contactwith target cells, Ca2+ mobilization is trig-gered in trypomastigotes. This Ca2+ response,which is required for parasite penetration, isnot elicited in noninfective epimastigoteforms (2,6). An early event in this cascademay include protein tyrosine kinase (PTK)activation, for treatment of trypomastigoteswith PTK inhibitor genistein reduces cellinvasion (3,7). Whether PTK activation isassociated with the Ca2+ response remains tobe established.

In metacyclic trypomastigotes, the insect-stage T. cruzi developmental forms that areinfective to mammalian hosts, the interac-tion of the surface glycoprotein gp82 with anas yet to be defined host cell receptor ap-pears to induce PTK activity (7). This mole-cule is also capable of inducing host cellCa2+ mobilization (5) and has been impli-cated in the process of cell invasion (11). Toadvance our understanding on the metacy-clic trypomastigote signal transduction path-way relevant for infection, experiments wereperformed i) to determine the possible asso-ciation between PTK activation and Ca2+

response, ii) to examine the involvement ofphospholipase C and IP3 in the parasite entryinto target cells, and iii) to confirm the role

of gp82 as the cell surface receptor thatmediates signaling to the parasite interior.

Material and Methods

T. cruzi, mammalian cells and cell invasionassay

T. cruzi strain CL (12) was used through-out this study. Parasites were maintained bycyclic passage in mice and in axenic cul-tures. Liver infusion tryptose medium (13)and Grace�s medium (Gibco, Gaithersburg,MD, USA) were used, respectively, to growparasites and to obtain cultures enriched inmetacyclic forms. Metacyclic trypomasti-gotes were purified by passage through aDEAE-cellulose column, as described (14).HeLa cells, the human carcinoma-derivedepithelial cells, and human leukemic K562cells (15) were grown at 37oC in Dulbecco�sminimum essential medium (DMEM) supple-mented with 10% fetal calf serum (FCS),streptomycin (100 µg/ml) and penicillin (100U/ml) in a humidified 5% CO2 atmosphere.Mammalian cell invasion assays were car-ried out as previously described (16) by seed-ing 106 metacyclic forms onto each well of24-well plates containing 13-mm diameterround glass coverslips coated with 2 x 105

HeLa cells. After 1-h incubation, the tripli-cate coverslips were washed in phosphate-buffered saline (PBS) and stained with Gi-emsa.

Detection of tyrosine phosphorylated T. cruziproteins

In standard experiments, 5 x 107 para-sites were incubated for 20 min at 37oC inthe absence or presence of a mammalian cellextract, equivalent to 160 µg/ml protein, in atotal volume of 200 µl. After washing withPBS, the parasites were disrupted at 4oC in alysis solution containing phosphatase andprotease inhibitors (50 mM Tris-HCl, pH7.4, 150 mM NaCl, 1 mM EDTA, 1% Triton

271

Braz J Med Biol Res 33(3) 2000

Signal transduction in Trypanosoma cruzi metacyclic forms

X-100, 1 mM NaVO4, 1 mM NaF, 1 mMphenylmethylsulfonyl fluoride, 1 mM iodo-acetamide, 25 µg/ml leupeptin, and 25 µg/mlantipain). Lysates were dissolved in loadingbuffer and subjected to electrophoresis in a10% SDS-polyacrylamide gel, under reduc-ing conditions, and the proteins were trans-ferred to a nitrocellulose membrane. Fol-lowing blockage with PBS containing 5%defatted milk, the nitrocellulose membranewas incubated with anti-phosphotyrosineantibodies (mouse monoclonal antibody fromSigma Chemical Co., St. Louis, MO, USA)diluted in PBS-milk for 1 h at room temper-ature. After several washes in PBS contain-ing 0.05% Tween 20, the membrane wasincubated with peroxidase-conjugated anti-mouse IgG. The final reaction was devel-oped by chemiluminescence using the ELCWestern blotting detection reagent andHyperfilm-MP from Amersham (Bucking-ham, UK).

Preparation of mammalian cell extract

The cell extract used in phosphorylationand Ca2+ signaling experiments was pre-pared as follows: after several washes inPBS, HeLa or K562 cells were sonicated onice with an ultrasonic processor XL (2 pulsesof 30 s each) in the presence of proteaseinhibitors (1 mM phenylmethylsulfonyl fluo-ride, 1 mM iodoacetamide, 25 µg/ml leupep-tin, and 25 µg/ml antipain). After ascertain-ing cell disruption under a phase contrastmicroscope, the sonicated preparation wascentrifuged at 12,000 g for 5 min and thesupernantant was collected and immediatelyused for experiments or stored at -80oC untiluse.

Determination of intracellular Ca2+

concentration

To measure T. cruzi cytosolic free Ca2+,[Ca2+]i, parasites (7 x 107 cells/ml) werewashed in Tyrode solution (137 mM NaCl,

2.7 mM KCl, 12 mM NaHCO3, 0.36 mMNaH2PO4, 0.53 mM MgCl2, 1.36 mM CaCl2,and 5.5 mM glucose) and incubated withfura-2/AM (Sigma) for 3 h at room tempera-ture, and non-incorporated fura-2 was washedout. Fluorescence was read with a fluoro-photometer SPEX AR-CM system (Spex In-dustries, Edison, NJ, USA) with dual wave-length excitation (340 and 380 nm) and emis-sion at 505 nm. The increase in parasite[Ca2+]i after the addition of a mammaliancell extract to 2.5 ml of parasite suspensionwas calculated as described by Grynkiewiczet al. (17). For each preparation we deter-mined Rmax and Rmin, which correspond tothe fluorescence ratio at 340 and 380 nm inthe presence of saturating Ca2+ after treat-ment with 50 µM digitonin, and in the ab-sence of Ca2+, upon addition of 10 mMethylene glycol-bis(ß-aminoethyl ether)tetraacetic acid (EGTA), respectively. Someexperiments were carried out with parasitesin buffer A, pH 7.2, without Ca2+ and con-taining 116 mM NaCl, 5.4 mM KCl, 0.8 mMMgSO4, 5.5 mM glucose, and 50 mM N-(2-hydroxyethyl)piperazine-N�-(2-ethanesul-fonic acid) (HEPES).

Purification of recombinant protein J18

The recombinant protein J18 (18) con-taining the entire sequence of the metacyclictrypomastigote surface molecule gp82 fusedto glutathione S-transferase (GST) was puri-fied from IPTG-induced Escherichia colisuspensions according to the procedure de-tailed in Ref. 19, followed by electroelutionas described (20).

Heparin loading by reverse permeabilization

The method of Kobayashi et al. (21) wasused, with slight modifications. Metacyclictrypomastigotes were incubated in the fol-lowing sequence of solutions kept at 2oC: I)20 min in 10 mM EGTA, 120 mM KCl, 5mM ATP, 2 mM MgCl2, 20 mM N-tris

272

Braz J Med Biol Res 33(3) 2000

N. Yoshida et al.

(hydroxymethyl)methyl-2-aminoethanesul-fonic acid (TES), pH 6.8, at 2oC; II) 90 minin 120 mM KCl, 5 mM ATP, 2 mM MgCl2,20 mM TES, pH 6.8, at 2oC; III) 30 min in120 mM KCl, 5 mM ATP, 10 mM MgCl2, 20mM TES, pH 6.8, at 2oC; IV) 30 min in 140mM NaCl, 5 mM KCl, 10 mM MgCl2, 5.6mM glucose, 2 mM MOPS, pH 7.1, at 22oC.To load heparin, 1 mg/ml heparin was addedto the solutions used in steps I-IV. The para-sites were warmed to 22-25oC and CaCl2

was added at 15-min intervals to give thefollowing concentrations: 0.001, 0.01, 0.1and 1.6 mM. After loading with heparin,parasite viability was ascertained by the ethi-dium bromide test (22). Control parasiteswere treated in the same manner except thatheparin was not added.

Treatment of parasites with different drugs

Metacyclic trypomastigotes were treatedwith 1 µM thapsigargin at 37oC for 30 min.Control parasites were incubated under thesame conditions in the presence of 0.1%

DMSO, equivalent to the DMSO concentra-tion in the thapsigargin-treated samples. Caf-feine treatment was performed by incubat-ing the parasites at 37oC for 30 min with 10mM of the drug. In both cases, the parasiteswere washed in buffer A before use. Toinhibit phospholipase C, the parasites weretreated for 4 min at 37oC in the presence of10 µM U73122, or were incubated with 100µM neomycin for 4 h at 37oC, washed inPBS and then used for the invasion assay.Parasites used as controls were incubatedunder the conditions described above in theabsence of U73122 or neomycin.

Results

Both p175 phosphorylation and Ca2+

mobilization are induced in metacyclictrypomastigotes by HeLa cells but not bythe K562 cell extract

We observed that the activation of T.cruzi PTK, as measured by the specific in-crease in the phosphorylation levels of atrypomastigote 175-kDa protein (p175), isinduced by exposure of parasites to culturedHeLa cells or to the sonicated extract. As theHeLa cell extract can be stored frozen with-out losing its phosphorylation-inducing ac-tivity, for practical reasons we have used thesonicated preparations in our study. The firstassay to determine the association betweenPTK activation and Ca2+ response was car-ried out by i) incubating metacyclic trypo-mastigotes with HeLa cell or the K562 cellextract for varying periods of time, and thenprocessing for immunoblot analysis usinganti-phosphotyrosine antibodies, and ii) add-ing HeLa cell or K562 cell extract to fura-2-loaded parasites and measuring the increasein [Ca2+]i. As shown in Figure 1, the HeLacell extract triggered an increase in the para-site cytosolic Ca2+ concentration in a mannercompatible with the increase in the levels ofp175 phosphorylation. By contrast, the ex-tract of K562 cells, which are resistant to T.

+ HeLa

+ K562

p175

p175

3020100

[Ca2

+] i

(nM

)

150

100

500 200 400 600 800

Time (s)

HeLa

K562

Figure 1 - Phosphorylation ofp175 and Ca2+ mobilization in-duced in T. cruzi metacyclic try-pomastigotes. A, Parasites wereincubated at 37oC for the indi-cated periods of time in the pres-ence of HeLa cell or T. cruzi-re-sistant K562 cell extract, equiva-lent to 160 µg/ml protein. Afterwashing in PBS, the parasiteswere lysed with detergent andthe total lysates were subjectedto SDS-PAGE and analyzed byimmunoblotting using anti-phos-photyrosine antibodies. B, Soni-cated extract of HeLa or K562cells, containing an equivalent of25 µg protein, was added to fura-2-loaded parasites at the indi-cated time (arrow) and [Ca2+]iwas measured. Results repre-sentative of three experimentsare presented.

A

B

Time (min)

273

Braz J Med Biol Res 33(3) 2000

Signal transduction in Trypanosoma cruzi metacyclic forms

cruzi infection, did not show any PTK- orCa2+ signal-inducing activity. The resultswere also negative when a two times moreconcentrated K562 cell extract was used.

Treatment of metacyclic trypomastigotes withgenistein inhibits p175 phosphorylation andCa2+ response

The next assay to determine the associa-tion between PTK activation and Ca2+ re-sponse was performed by i) incubating meta-cyclic trypomastigotes, untreated or treatedwith 250 µM genistein for 30 min at 37oC, inthe absence or in the presence of HeLa cellextract for 20 min at 37oC, and then process-ing for anti-phosphotyrosine immunoblot-ting, and ii) adding the HeLa cell extract tofura-2-loaded parasites, untreated or treatedwith genistein, and measuring [Ca2+]i. Fig-ure 2 shows that genistein, an inhibitor ofprotein tyrosine kinase (23) capable of in-hibiting metacyclic trypomastigote entry intoHeLa cells by ~75% (7), blocked both thep175 tyrosine phosphorylation and Ca2+ mo-bilization, suggesting that the two events areassociated.

p175 phosphorylation and Ca2+ response inmetacyclic forms are mediated by gp82

To confirm the assumption that the meta-cyclic stage surface glycoprotein gp82, whichbinds to HeLa cells in a receptor-dependentmanner (11), plays a role in signal transduc-tion, the following assay was performed.The PTK activation and Ca2+ response weredetermined upon incubation of parasites withHeLa cell extract in the absence or in thepresence of J18, a recombinant protein con-taining the full-length gp82 sequence whichbinds to HeLa cells in the same manner asthe native gp82 (19). In the PTK assay, theconcentration of J18 in the reaction mixturewas 250 µg/ml. For [Ca2+]i measurement, 50µl of HeLa cell extract containing 250 µg/mlof J18 was added to fura-2-loaded parasites.

[Ca2

+] i

(nM

)

150

100

0 200 400 600 800

Time (s)

HeLa

A

B

50

HeLa/J18

3F6

p175

- + +

HeL

a

HeL

a/J1

8

+

3F6

Figure 3 - Involvement of the T.cruzi metacyclic trypomastigotesurface molecule gp82 withp175 tyrosine phosphorylationand Ca2+ response. A, Parasiteswere incubated in the absence(-) or in the presence of HeLacell extract, unmixed (+HeLa) ormixed with recombinant proteinJ18 which contains the full-length gp82 sequence (+HeLa/J18), or were incubated with amonoclonal antibody directed atgp82 (+3F6). After 20 min at37oC, the parasites were pro-cessed for anti-phosphotyrosineimmunoblotting. B, HeLa cell ex-tract, mixed or not with J18, ormonoclonal antibody 3F6 wasadded at the indicated time (ar-row) to fura-2-loaded metacyclictrypomastigotes and [Ca2+]i wasmeasured.

p175

- + - +

Meta* Meta

[Ca2

+] i

(nM

)

150

100

500 200 400 600 800

Time (s)

Meta

Meta*

HeLa

A

B

Figure 2 - Inhibition of T. cruzimetacyclic trypomastigote p175phosphorylation and Ca2+ signal-ing by genistein. A, Metacyclicforms, untreated (Meta) ortreated with 250 µM of proteintyrosine kinase inhibitor genis-tein (Meta*), were incubated inthe absence (-) or in the pres-ence (+) of HeLa cell extract at37oC for 20 min, and then pro-cessed for anti-phosphotyrosineimmunoblotting. B, HeLa cellextract was added at the indi-cated time (arrow) to fura-2-loaded parasites, untreated(Meta) or treated with 250 µMgenistein (Meta*), and [Ca2+]iwas measured. Results repre-sentative of three experimentsare shown.

As shown in Figure 3, J18 inhibited bothp175 phosphorylation- and Ca2+ signaling-inducing activity of the HeLa cell extract.This result raises the possibility of isolatingthe signaling molecules from the HeLa cellextract using a gp82 affinity column. Whenmetacyclic trypomastigotes were exposed tomonoclonal antibody 3F6 directed at gp82,

274

Braz J Med Biol Res 33(3) 2000

N. Yoshida et al.

at 10 µg/ml, both PTK activation and Ca2+

mobilization were triggered (Figure 3), sup-porting the idea of the involvement of gp82in the signaling cascade.

Metacyclic trypomastigote Ca2+ required forhost cell invasion is released from intracellu-lar stores possibly in an IP3-mediated manner

In a series of experiments, we determinedwhether target cell invasion by metacyclicforms was dependent on Ca2+ released fromthe parasite intracellular compartments, morespecifically from IP3-susceptible stores. First,the metacyclic forms either untreated ortreated with thapsigargin, a sesquiterpene

lactone that depletes intracellular Ca2+ storesin many mammalian cell types by specificinhibition of endoplasmic reticulum Ca2+-ATPase (24), were incubated with HeLacells for 1 h at 37oC in DMEM containing2% FCS. Following washes in PBS, HeLacells were stained with Giemsa and the num-ber of intracellular parasites was counted ina total of 500 cells. As shown in Figure 4A,thapsigargin-treated parasites entered HeLacells in significantly lower numbers as com-pared to untreated controls. Consistent withthis observation, Ca2+ mobilization wasbarely detectable in thapsigargin-treated para-sites upon stimulation with monoclonal anti-body 3F6, in contrast to untreated controls(Figure 5).

Next, we tested the effect of caffeine,which inhibits IP3-induced Ca2+ release indifferent cell types (25,26). Caffeine-treatedparasites and untreated controls were seededonto HeLa cells and incubated as describedabove. The infectivity of caffeine-treatedparasites was significantly decreased (Fig-ure 4B) and this fact was correlated with agreatly reduced Ca2+ response (Figure 5). Inanother assay, we used parasites subjected toreversible permeabilization to incorporateheparin, a blocker of IP3-activated Ca2+ re-lease in smooth muscle cells (21,27), asdescribed in the Methods section. After in-cubation with HeLa cells for 1 h at 37oC inDMEM-2% FCS, the number of intracellu-lar parasites was counted in 500 Giemsa-stained cells. As compared to controls, theheparin-loaded parasites displayed a reducedcell invasion rate (Figure 4C). We also deter-mined whether the invasive capacity of Li+-treated parasites was reduced. Lithium isassumed to block the recycling of IP3, buttreatment with Li+ may eventually result inloss of the IP3 signal. For instance, exposureof agonist-stimulated cerebral cortex slicesor adrenal glomerulosa to Li+ for longer than5 min has been reported to suppress IP3

generation (28,29). Metacyclic trypomasti-gotes treated with 10 mM LiCl for 45 min at

[Ca2

+] i

(nM

)

300

200

100

0

3F6

c

b

a

0 200 400 600

Figure 5 - Inhibitory effect ofthapsigargin and caffeine on T.cruzi metacyclic trypomastigoteCa2+ mobilization. Monoclonalantibody 3F6, directed at themetacyclic stage surface mole-cule gp82, was added at the in-dicated time (arrow) to fura-2-loaded parasites, either un-treated (c) or pretreated withthapsigargin (a) or caffeine (b) at37oC for 30 min, and [Ca2+]i wasmeasured.

Time (s)

% In

vasi

on

100

50

0

% In

vasi

on

100

50

0

% In

vasi

on

100

50

0

% In

vasi

on

100

50

0

- -+ +Thapsigargin Caffeine

- -+ +Heparin Lithium

A

C

Figure 4 - Effect of treatment ofT. cruzi metacyclic trypomasti-gotes with thapsigargin or drugsthat affect the IP3 signal on HeLacell invasion. Parasites were: A,incubated at 37oC for 30 min inthe absence or in the presenceof 1 µM thapsigargin; B, treatedor not for 30 min at 37oC with 10mM caffeine; C, loaded or notwith 1 mg/ml heparin as de-scribed in the Methods section,or D, treated or not with 10 mMLi+ at room temperature for 45min, and then seeded onto HeLacells. After 1-h incubation at37oC, the number of intracellularparasites was counted in a totalof 500 Giemsa-stained cells andthe percentage of invasion deter-mined. The values are means ±SD of three independent assays.

B

D

275

Braz J Med Biol Res 33(3) 2000

Signal transduction in Trypanosoma cruzi metacyclic forms

room temperature before addition to HeLacells were internalized in significantly lowernumbers than the untreated controls (Figure4D). The viability of the parasites was notaffected by treatment with thapsigargin, caf-feine, lithium or heparin loading. They re-tained their normal morphology and motilityand were fully viable, as evaluated by Trypanblue staining and the ethidium bromide test.

Phospholipase C inhibitor blocks Ca2+

response and reduces parasite infectivity

IP3 is generated by phospholipase C-me-diated hydrolysis of phosphatidylinositol 4,5-biphosphate (30). As our results indicatedthe participation of IP3 in signal transductionrequired for cell invasion by metacyclicforms, we reasoned that inhibition of theparasite phospholipase C would affect itsability to enter host cells. To test that hypo-thesis, the parasites were treated with phos-pholipase C inhibitor U73122 (31) and thenassayed for HeLa cell invasion. As shown inFigure 6A, U73122-treated parasites enteredHeLa cells in significantly lower numbersthan untreated controls. In addition toU73122, the effect of neomycin, anotherphospholipase C inhibitor (32), was exam-ined. Although to a lesser extent than U73122,neomycin also inhibited the parasite infec-tivity (Figure 6B). Consistent with the inhi-bition of IP3-mediated Ca2+ release by thedrug, no significant Ca2+ response was de-tected in U73122-treated parasites (Figure6C). However, we have noted a higher basallevel of cytosolic Ca2+ concentration inU73122-treated parasites, suggesting also thepossibility of a nonspecific effect of the drugin releasing Ca2+ from intracellular stores(33).

The inhibitory effect of different drugs isreversible

To test whether the effect of the variousdrugs was reversible, metacyclic trypomas-

tigotes were treated with the drug, washed inPBS and then left to rest for 1 h at roomtemperature before seeding onto HeLa cells.Unlike the parasites used for the cell inva-sion assay immediately after drug treatment,the metacyclic forms incubated for 1 h in theabsence of thapsigargin, caffeine, Li+ orU73122 entered HeLa cells in numbers com-parable to those of untreated controls, indi-cating the reversible effect of these com-pounds. The reversibility of heparin actionwas not determined, but its reversible an-tagonism of IP3-activated Ca2+ release hasbeen demonstrated in other cell types (27).

Discussion

In the present study we analyzed the sig-nal transduction events in T. cruzi metacy-clic trypomastigotes required for mamma-lian cell invasion. Previous studies haveshown the requirement of protein tyrosinekinase activation, that results in p175 phos-phorylation, and the increase in cytosolicCa2+ concentration (5-7). Our results indi-cate that Ca2+ mobilization in metacyclic

Figure 6 - Inhibition of HeLa cellinvasion and Ca2+ response bytreatment of T. cruzi metacyclictrypomastigotes with phospho-lipase C inhibitor. Parasites weretreated or not with: A, 10 µMU73122 for 4 min at 37oC, andB, 100 µM neomycin at 37oC for4 h and then washed. Followingincubation of parasites withHeLa cells at 37oC for 1 h, per-cent invasion was determinedby counting the number of intra-cellular parasites in a total of 500Giemsa-stained cells. C, HeLacell extract was added to fura-2-loaded parasites, either un-treated (control) or treated with10 µM U73122 for 4 min at37oC, and [Ca2+]i was meas-ured. The values are means ±SD of three assays.

[Ca2

+] i

(nM

)

150

100

50

% In

vasi

on

100

50

0

% In

vasi

on

100

50

0- -+ +U73122 Neomycin

HeLa

U73122

Control

0 200 400 600

Time (s)

C

A B

276

Braz J Med Biol Res 33(3) 2000

N. Yoshida et al.

forms is dependent on PTK activation, sothat the inhibition of the enzyme activitywith the specific inhibitor genistein also in-hibits the Ca2+ response (Figure 2).

The signaling cascade in metacyclic try-pomastigotes is presumably initiated by rec-ognition of the surface glycoprotein gp82 byits target cell receptor. Thus, if this recogni-tion is prevented, for instance by the pres-ence of gp82 recombinant protein, PTK acti-vation and Ca2+ response cannot be induced(Figure 3). On the other hand, binding of themonoclonal antibody 3F6 to gp82, possiblyby mimicking the parasite ligand-host cellreceptor interaction, triggers p175 phospho-rylation and Ca2+ mobilization (Figure 3).The next event in the metacyclic stage signaltransduction pathway, following engagementof gp82, is probably PTK activation. Howgp82, which is anchored to the plasma mem-brane via the glycosylphosphatidylinositol(GPI) moiety (34), associates with tyrosinekinase is not known. In mammalian cells, atleast two possibilities of association of GPI-anchored surface molecules with tyrosinekinases have been reported. The neuron sur-vival factor neurturin, for instance, wasshown to signal through multicomplex re-ceptors that consist of receptor tyrosine ki-nase and a member of a GPI-linked family ofreceptors that determine ligand specificity(35). Protein tyrosine kinases have beenfound in complexes immunoprecipitatedfrom T cells with antibodies directed at GPI-linked proteins (36,37). Interaction of GPI-anchored proteins and kinases could also bemediated by transmembrane linker proteins(38).

Activation of phospholipase C, that leadsto IP3 generation and Ca2+ mobilization, ispossibly the next step in the metacyclic try-pomastigote-signaling pathway. We havefound that treatment of parasites with thephospholipase C inhibitor U73122 blocked

the Ca2+ response and the ability to invadeHeLa cells (Figure 6). Whether differentphospholipase C isozymes are present in T.cruzi, and in this case which form is acti-vated during target cell invasion, is one ofthe many questions that remain to be investi-gated. Also to be clarified is the mode ofassociation of protein PTK or p175 withphospholipase C.

The increase in cytosolic Ca2+ necessaryfor metacyclic trypomastigote entry into hostcells may originate from thapsigargin-sensi-tive intracellular Ca2+ stores in an IP3-medi-ated manner. First, treatment of parasiteswith thapsigargin, which depletes intracellu-lar Ca2+ stores by inhibiting endoplasmicreticulum Ca2+-ATPase (24), significantlyreduced both the Ca2+ response (Figure 5)and the ability to enter HeLa cells (Figure 4).Second, drugs that interfere with IP3-medi-ated Ca2+ release, such as heparin, caffeineand lithium, reduced the parasite infectivity(Figure 4). Although our data suggest thatIP3-mediated Ca2+ release occurs in metacy-clic trypomastigotes, this event has not beendetected in studies with other T. cruzi devel-opmental forms in spite of the presence ofinositol phosphates (39,40). The basis forsuch discrepancies has yet to be determined.

Although a number of questions still re-main to be answered, taken together, ourresults suggest that interaction of the meta-cyclic trypomastigote surface molecule gp82with its host cell receptor triggers a signalingcascade leading to sequential activation ofprotein tyrosine kinase and phospholipaseC, with consequent generation of IP3 thatpromotes Ca2+ mobilization necessary forparasite internalization.

Acknowledgments

We thank Dr. Maria E. Oshiro for valu-able suggestions.

277

Braz J Med Biol Res 33(3) 2000

Signal transduction in Trypanosoma cruzi metacyclic forms

References

1. Tardieux I, Nathanson MH & Andrews NW(1994). Role in host cell invasion of Try-panosoma cruzi-induced cytosolic freeCa2+ transients. Journal of ExperimentalMedicine, 179: 1017-1022.

2. Moreno SNJ, Silva J, Vercesi AE &Docampo R (1994). Cytosolic-free calciumelevation in Trypanosoma cruzi is requiredfor cell invasion. Journal of ExperimentalMedicine, 180: 1535-1540.

3. Vieira MCF, Carvalho TU & Souza W(1994). Effect of protein kinase inhibitorson the invasion process of macrophagesby Trypanosoma cruzi. Biochemical andBiophysical Research Communications,203: 967-971.

4. Yakubu MA, Majumder S & KierszenbaumF (1994). Changes in Trypanosoma cruziinfectivity by treatments that affect cal-cium ion levels. Molecular and Biochemi-cal Parasitology, 66: 119-125.

5. Dorta ML, Ferreira AT, Oshiro MEM &Yoshida N (1995). Ca2+ signal induced byTrypanosoma cruzi metacyclic trypomas-tigote surface molecules implicated inmammalian cell invasion. Molecular andBiochemical Parasitology, 73: 285-289.

6. Ruiz RC, Favoreto S, Dorta ML, OshiroMEM, Ferreira AT, Manque PM & YoshidaN (1998). Infectivity of Trypanosoma cruzistrains is associated with differential ex-pression of surface glycoproteins with dif-ferential Ca2+ signaling activity. Biochemi-cal Journal, 330: 505-511.

7. Favoreto S, Dorta ML & Yoshida N (1998).Trypanosoma cruzi 175-kDa protein ty-rosine phosphorylation is associated withhost cell invasion. Experimental Parasitol-ogy, 89: 188-194.

8. Rodriguez A, Rioult MG, Ora A & AndrewsN (1995). A trypanosome-soluble factorinduces IP3 formation, intracellular Ca2+

mobilization and microfilament rearrange-ment in host cells. Journal of Cell Biology,129: 1263-1273.

9. Tardieux I, Webster P, Ravesloot J, BoronW, Lunn JA, Huser JE & Andrews NW(1992). Lysosome recruitment and fusionare early events required for Trypanoso-ma invasion of mammalian cells. Cell, 71:1117-1130.

10. Andrews NW (1995). Lysosome recruit-ment during host cell invasion by Trypa-nosoma cruzi. Trends in Biology, 5: 133-137.

11. Ramirez MI, Ruiz RC, Araya JE, Franco daSilveira J & Yoshida N (1993). Involve-ment of the stage-specific 82-kilodalton

adhesion molecule of Trypanosoma cruzimetacyclic trypomastigotes in host cellinvasion. Infection and Immunity, 61:3636-3641.

12. Brener Z & Chiari E (1963). Variaçõesmorfológicas observadas em diferentesamostras de Trypanosoma cruzi. Revistado Instituto de Medicina Tropical de SãoPaulo, 5: 220-224.

13. Camargo EP (1964). Growth and differen-tiation in Trypanosoma cruzi. I. Origin ofmetacyclic trypanosomes in liquid media.Revista do Instituto de Medicina Tropicalde São Paulo, 6: 93-100.

14. Teixeira MMG & Yoshida N (1986). Stage-specific surface antigens of metacyclic try-pomastigotes of Trypanosoma cruzi iden-tified by monoclonal antibodies. Molecu-lar and Biochemical Parasitology, 18: 271-282.

15. Lozzio CB & Lozzio BB (1975). Humanchronic myelogenous leukemia cell-linewith positive Philadelphia chromosome.Blood, 45: 321-334.

16. Yoshida N, Mortara RA, Araguth MF,Gonzalez J & Russo M (1989). Metacyclicneutralizing effect of monoclonal antibody10D8 directed to the 35 and 50 kilodaltonsurface glycoconjugates of Trypanosomacruzi. Infection and Immunity, 57: 1663-1667.

17. Grynkiewicz G, Poenie M & Tsien RY(1985). A new generation of Ca2+ indica-tors with greatly improved fluorescence.Journal of Biological Chemistry, 260:3440-3450.

18. Araya JE, Cano MI, Yoshida N & Franco daSilveira J (1994). Cloning and character-ization of a gene for the stage-specific 82kDa surface antigen of metacyclic trypo-mastigotes of Trypanosoma cruzi. Molec-ular and Biochemical Parasitology, 65:161-169.

19. Santori FR, Dorta ML, Juliano L, JulianoMA, Franco da Silveira J, Ruiz RC &Yoshida N (1996). Identification of a do-main of Trypanosoma cruzi metacyclic try-pomastigote surface molecule gp62 re-quired for attachment and invasion ofmammalian cells. Molecular and Bio-chemical Parasitology, 78: 209-216.

20. Wheatley M (1993). Peptide mapping andthe generation and isolation of sequen-tiable peptides from receptors. In: HulmeEC (Editor), Receptor Biochemistry. APractical Approach. IRL Press at OxfordUniversity Press, Oxford, UK, 213-261.

21. Kobayashi S, Kitazawa T, Somlyo AV &

Somlyo AP (1989). Cytosolic heparin in-hibits muscarinic and a-adrenergic Ca2+

release in smooth muscle. Journal of Bio-logical Chemistry, 264: 17997-18004.

22. Vercesi AE, Hoffmann ME, Bernardes CF& Docampo R (1991). Regulation of intra-cellular homeostasis in Trypanosoma cru-zi. Effects of calmidazolium and trifluoper-azine. Cell Calcium, 12: 361-369.

23. Akiyama T, Ishida J, Nakagawa S,Ogawara H, Watanabe S, Itoh N, ShibuyaM & Fukami Y (1987). Genistein, a specif-ic inhibitor of tyrosine-specific protein ki-nases. Journal of Biological Chemistry,262: 5592-5595.

24. Thastrup O, Cullen PJ, Drobak BK, HanleyMR & Dawson AP (1990). Thapsigargin, atumor promoter, discharges intracellularCa2+ stores by specific inhibition of theendoplasmic reticulum Ca2+-ATPase. Pro-ceedings of the National Academy of Sci-ences, USA, 87: 2466-2470.

25. Brown GR, Sayers LG, Kirk CJ, MichellRH & Michelangelli F (1992). The openingof the inositol 1,4,5-triphosphate-sensitiveCa2+ channel in rat cerebellum is inhib-ited by caffeine. Biochemical Journal, 282:309-312.

26. Hirose K, Iino M & Endo M (1993). Caf-feine inhibits Ca2+-mediated potentiationof inositol 1,4,5-triphosphate-inducedCa2+ release in permeabilized vascularsmooth muscle cells. Biochemical andBiophysical Research Communications,194: 726-732.

27. Ghosh TK, Eis PS, Mullaney JM, Ebert CL& Gill DL (1988). Competitive, reversible,and potent antagonism of inositol 1,4,5-triphosphate-activated calcium release byheparin. Journal of Biological Chemistry,263: 11075-11079.

28. Balla T, Enyedi P, Hunyadi L & Spät A(1984). Effect of lithium on angiotensin-stimulated phosphatidylinositol turnoverand aldosterone production in adrenal glo-merulosa cells: a possible causal relation-ship. FEBS Letters, 171: 179-182.

29. Batty I & Nahorski SR (1985). Differentialeffects of lithium on muscarinic receptorstimulation of inositol phosphates in ratcerebral cortex slices. Journal of Neuro-chemistry, 45: 1514-1521.

30. Berridge MJ (1993). Inositol triphosphateand calcium signalling. Nature, 361: 315-325.

31. Bleasdale JE, Thakur R, Gremban RS,Bundy GL, Fitzpatrick FA, Smith RJ &Bunting S (1990). Selective inhibition of

278

Braz J Med Biol Res 33(3) 2000

N. Yoshida et al.

receptor-coupled phospholipase C-de-pendent processes in human plateletsand polymorphonuclear neutrophils. Jour-nal of Pharmacology and ExperimentalTherapeutics, 255: 756-768.

32. Ma L & Michel WC (1998). Drugs affect-ing phospholipase C-mediated signaltransduction block the olfactory cyclicnucleotide-gated current of adult zebra-fish. Journal of Neurophysiology, 79:1183-1192.

33. Taylor C & Broad LM (1998). Pharmacolo-gical analysis of intracellular Ca2+ signal-ling: problems and pitfalls. Trends in Phar-macological Sciences, 19: 370-375.

34. Cardoso de Almeida MI & Heise N (1993).Proteins anchored via glycosylphosphati-dylinositol and solubilizing phospholipases

in Trypanosoma cruzi. Biological Re-search, 26: 285-312.

35. Buj-Bello A, Adu J, Piñon LGP, Horton A,Thompson J, Rosenthal A, Chinchetru M,Buchman VL & Davies AM (1997).Neurturin responsiveness requires a GPI-linked receptor and the ret receptor ty-rosine kinase. Nature, 387: 721-724.

36. Stefanova I, Horejsi V, Ansotegui IJ,Knapp W & Stockinger H (1991). GPI-an-chored cell surface molecules complexedto protein tyrosine kinase. Science, 254:1016-1019.

37. Thomas PM & Samelson LE (1992). Theglycosylphosphatidylinositol anchored Thy-1 molecule interacts with the p60fyn pro-tein tyrosine kinase in T cells. Journal ofBiological Chemistry, 276: 12317-12322.

38. Brown D (1993). The tyrosine kinase con-nection: how gpi-anchored proteins acti-vate T cells. Current Opinion in Immunol-ogy, 5: 349-354.

39. Moreno SNJ, Vercesi AE, Pignataro OP &Docampo R (1992). Calcium homeostasisin Trypanosoma cruzi amastigotes: pres-ence of inositol phosphates and lack ofinositol 1,4,5-triphosphate-sensitive cal-cium pool. Molecular and BiochemicalParasitology, 52: 251-262.

40. Docampo R, Moreno SNJ & Vercesi AE(1993). Effect of thapsigargin on calciumhomeostasis in Trypanosoma cruzi trypo-mastigotes and epimastigotes. Molecularand Biochemical Parasitology, 59: 305-314.