Microbes and Infection 9 (2007) 111e118

Original article

The binding of CCL2 to the surface of Trypanosoma cruziinduces chemo-attraction and morphogenesis

Lucy M. Yamauchi a, Julio C. Aliberti b, Marcelo D. Baruffi c, Ricardo W. Portela d,Marcos A. Rossi e, Ricardo T. Gazzinelli d, Jose R. Mineo f, Jo~ao S. Silva a,*

a Department of Biochemistry and Immunology, School of Medicine of Ribeir~ao Preto-USP,Av. Bandeirantes 3900, 14049-900 Ribeir~ao Preto, SP, Brazil

b Division of Molecular Immunology, Cincinnati Children’s Hospital Medical Center and University of Cincinnati College of Medicine,

3333 Burnet Avenue, 45229-3039 Cincinnati, OH, USAc Department of Clinical Analysis, Toxicology and Bromatology, School of Pharmaceutical Sciences of Ribeir~ao Preto-USP,

Av. Cafe s/n� 14040-903, Ribeir~ao Preto, SP, Brazild Department of Biochemistry and Immunology, Institute of Biological Science, Universidade Federal of Minas Gerais,

Av. Antonio Carlos, 6627, 31270-910 Belo Horizonte, MG, Brazile Department of Pathology, School of Medicine of Ribeir~ao Preto-USP, Av. Bandeirantes 3900, 14049-900 Ribeir~ao Preto, SP, Brazil

f Laboratory of Immunology, Institute of Biomedical Science, Universidade Federal de Uberlandia,

Av. Para, 1.720, 38400-902 Uberlandia, MG, Brazil

Received 23 May 2006; accepted 25 October 2006

Available online 8 December 2006

Abstract

Adhesion of Trypanosoma cruzi to host cells employs mechanisms which are complex and not completely understood. Upon infection, hostcells release pro-inflammatory cytokines and chemokines in the environment. These had been found to be involved with increasing parasite up-take as well as killing by macrophages and cardiomyocytes. In the present study, we focused on the interaction of murine b-chemokine CCL2with trypomastigote forms of T. cruzi. We found that this chemokine directly triggers the chemotaxis and morphogenesis of trypomastigoteforms of parasites. Binding assays showed that the interaction of CCL2 with molecules present in trypomastigote forms is abolished by the ad-dition of condroitin 6-sulphate, a glycosaminoglycan. Moreover, we also observed that the parasite glycoproteins are the major players in thisinteraction. In summary, our study demonstrates a host ligand/parasite receptor interaction that may have relevant implications in the tissue tro-pism of this important parasitic disease.� 2006 Elsevier Masson SAS. All rights reserved.

Keywords: Trypanosoma cruzi; CCL2; Morphogenesis; Glycoproteins

www.elsevier.com/locate/micinf

1. Introduction

Trypanosoma cruzi, the causative agent of Chagas’ disease,has a polymorphic life cycle. There are three developmentalstages: epimastigote, the insect vector multiplication form;amastigote, the vertebrate intracellular replication form; andtrypomastigote, the infective form. Initially, the metacyclic

Abbreviations: CCL2, chemokine (motif CC) ligand 2.

* Corresponding author. Tel.: þ55 16 360 23234; fax: þ55 16 363 36890.

E-mail address: jsdsilva@fmrp.usp.br (J.S. Silva).

1286-4579/$ - see front matter � 2006 Elsevier Masson SAS. All rights reserve

doi:10.1016/j.micinf.2006.10.016

trypomastigote forms, transmitted by infected triatomine vec-tor, invade cells in the dermal layers and mucosa, after fewdays of replicative cycles, bloodstream trypomastigotes are re-leased and they disseminate using the bloodstream as a vehicleto reach organs such as the heart and the gastrointestinal tract.To contact and invade a particular host cell type, bloodstreamtrypomastigote must cross the vascular endothelium andmigrate through the intricate network of extracellular matrixprotein of the vessel wall and target organs [1]. The abilityof T. cruzi to infect and replicate within a variety of cell typesis an essential feature of its life cycle. During the invasion

d.

112 L.M. Yamauchi et al. / Microbes and Infection 9 (2007) 111e118

process, trypomastigote forms associated with cell membranestrigger signaling pathways that lead to the localized recruitmentand fusion of host cell lysosomes with the plasma membraneand generation of a lysosome-derived parasitophorous vacuoleor lead to a lysosome-independent mechanism dependent ofphosphatidylinositol 3-kinase [2,3]. The signaling pathwaysactivated are transient calcium-release, cyclic AMP, phosphati-dylinositol (PI) 3-kinases, tyrosine phosphorylation and others[3]. The activation of host cell signaling and a modulation ofgene expression are being recognized as essential determinantsof infectivity and intracellular survival of this parasite.

Another event that is triggered during the invasion process isthe production of cytokines and chemokines by macrophagesand cardiomyocytes [4,5]. In the acute phase of T. cruzi experi-mental infection, parasitemia is controlled by interferon (IFN)-g and tumor necrosis factor (TNF)-a. These cytokines canactivate macrophages and cardiomyocytes to produce nitric oxide(NO), the main effector molecule that controls intracellularT. cruzi replication [6,7]. The macrophages and cardiomyocytesalso produce chemokines in vitro, such as MCP-1 (monocytechemo-attractant protein-1 or CCL2), RANTES (regulation onactivation, normal T-cell expressed and secreted or CCL5),MIP-1a (macrophage inflammatory protein-1a or CCL3) thatcan induce recruitment of leukocytes to the inflammatory focus.These chemokines also play an important role in the uptakeand killing of intracellular parasites by inducing NO synthaseactivation and enhancing NO production by macrophages andcardiomyocytes [4,5].

The main functions of the chemokines are promotion ofleukocyte recruitment to inflamed tissues, stimulation of leu-kocyte exocytosis, and induction of hematopoiesis [8,9].Most chemokines are produced under pathological conditionsby tissue cells and infiltrating leukocytes. CCL2 is producedby a variety of cell types, including lymphocytes, mononuclearphagocytes and vascular endothelial cells. It is expressed inresponse to diverse stimuli, including the cytokines platelet-derived growth factor, IFN-g, interleukin (IL)-1b, TNF-a,granulocyte/macrophage colony-stimulating factor and macro-phage colony-stimulating factor, as well as hypoxia and super-oxides. CCL2 interacts with a receptor, CCR2, which belongsto a large family of seven-transmembrane-spanning, the G pro-tein-coupled receptors on target cells [8,9]; additionally, it in-teracts with glycosaminoglycans (GAGs) on cell surfaces andextracellular matrix [10]. The interaction of CCL2 and CCR2stimulates monocytes by increasing cytosolic free calcium andthe respiratory burst [11]. This cellular response is sensitive topertussis toxin (PTx), which uncouples the receptor from its Gprotein and thus blocks signal transduction by receptors [12].

The initial contact between parasite and mammalian cell re-quires interaction of parasite molecules with host cell receptors.Trypomastigotes use mechanisms of the host’s cellular machin-ery to invade and survive; however, how parasites exploit themare still unclear. Interaction of parasites and host cells has beenstudied and the mammalian host signaling pathway has beenidentified. In the present work, we studied the effects of CCL2directly on trypomastigote forms of T. cruzi and the conse-quences of this interaction. We analyzed bindings of CCL2 on

different forms and two strains of T. cruzi. Besides, we alsostudied the interaction of CCL2 with parasite surface glycopro-teins, and the role of this chemokine during T. cruzi infection.

2. Materials and methods

2.1. Experimental animals

Female BALB/c mice, 5- to 6-week-old, were bred andmaintained under standard conditions in the animal house ofthe Department of Biochemistry and Immunology, School ofMedicine of Ribeir~ao Preto-USP (Ribeir~ao Preto, Brazil).

2.2. Parasites

The Y and Colombian strains of T. cruzi were used in theexperiments. For in vitro experiments, trypomastigotes andamastigotes were grown and purified from a monkey epithelialcell line (LLC-MK2) cultured in RPMI (Sigma, St. Louis, MO)supplemented with 5% of heat inactivated bovine serum (Hy-clone Lab, Logan, UT). Trypomastigotes were collected fromsupernatants of infected cells and amastigotes were collectedfrom 4-day infected cells, the cells were harvested and hada mechanical disruption. Both forms were centrifuged at150� g for 10 min, the supernatants were collected andcentrifuged at 1500� g for 30 min. The parasites were washedwith PBS buffer. This procedure eliminates the residual LLC-MK2 cells. Epimastigotes were grown in Schneider’s medium(Sigma), supplemented with 20% of bovine serum (Hyclone).For in vivo experiments, bloodstream trypomastigotes wereobtained from infected animals.

2.3. Spheromastigote-like genesis assay

Cell culture derived T. cruzi trypomastigotes (106 cells/ml)were collected, and incubated in the presence of CCL2 (rang-ing from 50 to 200 ng/ml) and medium (control). After the in-cubation time (from 0 to 180 min), the parasites were countedin a hemocytometer and the ratio of amastigote:trypomastigotewas calculated. The cell suspensions were also evaluated inFACSorter (B&D Co., San Jose, CA). The parasites incubatedwith increasing concentration of CCL2 and CXCL10/IP-10 (ascontrol, CT) for 3 h were washed and incubated with 5 mg/mlof the mAb 2C2 (anti-Ssp-4, kindly donated by Dr. R.A. Mor-tara, Universidade Federal de S~ao Paulo), which recognizesexclusively amastigote forms [13] for 30 min in PBS at4 �C. The cells were washed in cold PBS and incubated withgoat anti-mouse IgG-FITC (Vector, Bullingame, CA) for60 min at 4 �C in the dark. The parasites were washed andfixed with PBS/formaldehyde 1%, 10,000 parasites were ac-quired by flow cytometry, and analyzed according to the fre-quency of parasite labeled.

2.4. CCL2 binding assay

The binding assay was performed with biotin-conjugated-CCL2 (R&D Systems, Minneapolis, MN). The parasites

113L.M. Yamauchi et al. / Microbes and Infection 9 (2007) 111e118

(1� 105 forms) were collected, washed in cold PBS and pre-treated with purified mouse IgG (10 mg/106 parasites) for15 min followed by incubation with biotin-conjugatedCCL2, biotin-conjugated soybean trypsin inhibitor (STI), bio-tinylated CCL2 previously incubated with neutralizing anti-body anti-CCL2 (25 mg/ml) (CCL2þ anti-CCL2) or with2.5- or 100-fold of unlabeled CCL2. The parasites were incu-bated for 1 h at 4 �C, and then, the avidineFITC added and in-cubated for an additional 30 min at 4 �C in the dark. Theparasites were washed and fixed with PBS/formaldehyde1%, 10,000 parasites were acquired by flow cytometry, and an-alyzed according to the frequency of parasite labeled and themean of fluorescence intensity of the samples.

2.5. Inhibition of CCL2 binding assay

Trypomastigote forms (1� 105 forms) were collected,washed in cold PBS and pretreated with purified mouse IgG(10 mg/106 parasites) for 15 min followed by incubation withbiotin-conjugated CCL2 alone or with increasing concentra-tion of condroitin 6-sulphate (10e1000 mg/ml) (Sigma) for1 h at 4 �C, and then the avidineFITC added and incubatedfor an additional 30 min at 4 �C in the dark. The parasiteswere washed and fixed with PBS/formaldehyde 1%, 10,000parasites were acquired by flow cytometry, and analyzed ac-cording to the frequency of positive cells.

2.6. Binding of CCL2 to glycoprotein ofparasite membrane

For measurement of CCL2 binding to parasite glycopro-teins, a chemiluminescent ELISA test was developed. Briefly,the plate was coated with glycoinositolphospholipid of epi-mastigotes (EGIPL), mucin of trypomastigote (TMuc), epi-mastigotes (Emuc) and BSA (0.25 mg/well) in carbonatebuffer (pH 9.6) overnight at 4 �C. The plate was washed, theremaining binding sites blocked with carbonate buffer/bovineserum albumin 2% for 2 h at 37 �C and the plate washed threetimes with PBS/Tween 20 (0.05%). Biotin-conjugated CCL2(2.5 mg/ml, R&D Systems) and STI (5 mg/ml) were diluted(1:40) in PBS/Tween/BSA 1%, transferred to the plate, incu-bated for 1 h at 37 �C, and subsequently washed with PBS/Tween. Then, the streptavidin-conjugated peroxidase (1:2000)was diluted in PBS/Tween and incubated for 30 min at 37 �C.The plate was washed, and then the detection was performedusing ECL plus (Amersham Pharmacia Biotech, LittleChalfont, UK), the reaction was measured in a luminometer(Lumicount, Packard, Dawner’s Grove, IL).

2.7. Chemotaxis assay in vivo

To determine the role of CCL2 as a chemotactic factor forT. cruzi trypomastigotes, the air pouch assays were performedin infected BALB/c mice during the peak of parasitemia(around 7e9 days after infection). The air pouch was madeby injection of sterile air subcutaneously in the serosa cavityin the back of mice. After 3 days another air injection was

performed keeping the volume of the cavity. During thepeak of parasitemia, evaluated as described elsewhere [14],CCL2 ranging from 50 to 200 ng/ml (R&D Systems),CXCL9/MIG (100 ng/ml), medium (control), was injected in-side the air pouches. After 6 h, the air pouches were washedwith cold PBSeheparin (25 U/ml, Roche, Rio de Janeiro,RJ, Brazil), the cells were pelleted and suspended in 1 ml ofPBS. The counting of parasites and leukocytes were per-formed in a hemocytometer.

2.8. Statistical analysis

The results are presented as the means plus standard devi-ations. The number of individual experiments and the level ofsignificance are indicated in each figure legend. Statisticalanalysis was generally performed with ANOVA by using thePrism software (Graph pad Software, San Diego, CA, USA).

3. Results

3.1. Chemokine CCL2 induces transformation oftrypomastigote to spherical forms (amastigote-like)

We previously showed that peritoneal macrophages pro-duce chemokines when infected with T. cruzi trypomastigoteforms, which induce increased uptake of these forms [4].When the supernatants of stimulated macrophages were in-jected in the air pouches of infected mice, we found accumu-lation of leukocytes and, surprisingly, parasites in theexudates. Since we found more spherical than trypomastigoteforms in the exudates harvested from the air pouches, we hy-pothesized that the interaction of chemokines with trypomas-tigote forms could induce the transformation into sphericalforms. When trypomastigotes were incubated with CCL2 orCCL5, we observed that they transformed to spherical forms(transformation rate increased 3.6- and 2.4-fold, respectively).CCL3 and CCL4/MIP-1b presented transformation ratios sim-ilar to the control (medium). We next evaluated the morpho-logic differentiation of trypomastigotes after incubation withincreasing concentrations of CCL2. The addition of the che-mokine induced, in a dose-dependent manner, increased levelsof morphologic transformation compared with the control me-dium only (Fig. 1A). To evaluate if the spherical form of par-asites was amastigote-like, we incubated them with mAb 2C2,which recognizes Ssp-4 (an amastigote stage-specific marker),and the binding was quantified by flow cytometry. The resultsshowed that the spherical forms exhibited the Ssp-4 epitope onthe surface (Fig. 1B), suggesting that the chemokine interac-tion induces the morphogenesis of trypomastigote to amasti-gote-like form. Besides, when trypomastigotes wereincubated with CXCL10/IP-10 (CT), no positive staining tomAb 2C2 was detected, indicating a specificity of CCL2 in-duced transformation.

3.2. CCL2 binds to T. cruzi

We next evaluated if chemokines could bind to other formsof parasites. The three parasite-developmental stages

114 L.M. Yamauchi et al. / Microbes and Infection 9 (2007) 111e118

(epimastigote, trypomastigote and amastigote) were incubatedwith biotin-conjugated murine CCL2 or soybean trypsin inhib-itor (STI) as a specificity control. The binding was analyzed byflow cytometry and the data expressed as percentage of CCL2binding as well as the mean of fluorescence intensity (MFI).We observed that CCL2 bound to all three forms of parasites(Fig. 2A), although the intensity of fluorescence was higher intrypomastigote forms than in the others (Fig. 2B), indicatingmore interaction of CCL2 with trypomastigote forms. On the

1 10 100 1000 100000

50

100

150

200

2C2-FITC (MFI)

Cel

ls

CT (100ng/ml)

CCL2 (1ng/ml)

CCL2 (10ng/ml)

CCL2 (100ng/ml)

B

0 30 60 120 180Time after stimuli (min)

0

10

20

30

40

Sphe

rom

astig

otes

/ml (

10E-4

)

0

50

100

200

CCL2 (ng/ml)

A

Fig. 1. CCL2 induces transformation of trypomastigote forms in spherical

forms (amastigote-like) in vitro. Trypomastigote forms (1� 106 forms/ml, Y

strain) obtained from cultures of LLC-MK2-infected cells were incubated

with different concentrations of CCL2, medium alone, or CXCL10 (CT) in

a humid chamber at 37 �C and 5% of CO2. (A) The number of spheromasti-

gotes in the cultures was counted in a hemocytometer several times after the

chemokines were added. Each point represents the mean of triplicate samples.

(B) After 3 h, the parasites were incubated with monoclonal antibody anti-

Ssp-4 (2C2) for 1 h, washed and incubated with anti-mouse IgG-FITC for

1 h. The parasites were washed, fixed with PBS/formaldehyde 1% and

acquired by flow cytometry. The data represent 10,000 events.

other hand, the STI-labeled protein did not interact with anyform of T. cruzi, suggesting a specific chemokine interactionwith the parasite. The neutralization of CCL2 with mAb in-hibited the binding to all forms (Fig. 2). We also showedthat both strains of T. cruzi, Y and Colombian, bind toCCL2 with similar intensity (Fig. 3A). Again, no interactionsbetween STI and trypomastigote forms of these two strainswere observed and the addition of neutralizing antibodiesanti-CCL2 blocked such interaction, regardless of the parasitestrain. Additionally, we observed that labeled CCL2 could bedisplaced with the addition of 100-fold excess of unlabeledCCL2 when the Y and Colombian strain parasites were used(Fig. 3B). However, when 2.5-fold excess of unlabeledCCL2 was used, only the Colombian strain parasites havinglabeled CCL2 displaced. These results suggest that the interac-tion of CCL2 and trypomastigote forms of the Colombianstrain is weaker than that of the Y strain (Fig. 3B). Altogether,the binding of CCL2 and parasites seems to be specific, sincethere was no interaction between STI and parasites.

3.3. Condroitin 6-sulphate inhibits CCL2 binding totrypomastigote forms of T. cruzi

Since the majority of chemokines are basic proteins,whereas the extracellular domains of their receptors are acidic,there is the possibility that CCL2 is binding to trypomastigotesby receptor independent interaction, we next evaluated if con-droitin 6-sulphate (C6S) was able to inhibit the interaction ofCCL2 with T. cruzi. We found that C6S caused an inhibition ofthe CCL2 binding in a dose-dependent manner. When biotiny-lated CCL2 was incubated with trypomastigotes, 60.45% ofthe parasites bound to the chemokine (Fig. 4, CCL2). How-ever, with the addition of C6S the population of positive par-asites decreased in a dose-dependent manner (Fig. 4). With10 mg/ml of C6S, a defined population of parasites CCL2þ/

high was observed. When the concentrations of C6S were in-creased (100 and 1000 mg/ml), the intensity of fluorescencein the parasite decreased. Therefore, the interaction of CCL2with the parasite has a low affinity, since C6S can displaceit in a dose-dependent manner. With heparin, another glycos-aminoglycan, we obtained similar results (data not shown).No interaction of STI with parasites was observed (Fig. 4,STI). Overall, the data suggest that CCL2 interact with nega-tive charged molecules on the parasite surface.

3.4. CCL2 binds to parasite glycoprotein

CCL2 binds to glycosaminoglycan through ionic interac-tion between two basic amino acid residues to the negativelycharged proteoglycans [15]. Since the parasites are coveredby glycoproteins, and mucins are the most common [16,17],we analyzed whether these glyco-conjugated proteins couldbe the target of host-derived CCL2. To evaluate it, a chemilu-minescence-based ELISA was performed using T. cruzi mem-brane glycoproteins. The results showed that CCL2 bound tomucins of trypomastigote and epimastigote forms, as well asGIPL of epimastigote forms, but not to BSA (Fig. 5). No

115L.M. Yamauchi et al. / Microbes and Infection 9 (2007) 111e118

CCL2 STI CCL2+

anti-CCL2

CCL2 STI CCL2+

anti-CCL2

0

10

20

30

40

50EpiTrypoAma

A

Bind

ing

of C

CL2

(%)

0

300

600

900B

*

Fluo

resc

ence

inte

nsity

Fig. 2. CCL2 binds to different forms of T. cruzi. Epimastigote, trypomastigote and amastigote forms of T. cruzi Y strain were pretreated with purified mouse IgG

(10 mg/106 parasites) for 15 min followed by incubation with biotin-conjugated CCL2, biotin-conjugated soybean trypsin inhibitor (STI), biotinylated CCL2 pre-

viously incubated with neutralizing antibody anti-CCL2 (25 mg/ml) (CCL2þ anti-CCL2). The parasites were incubated for 1 h, the avidineFITC added and in-

cubated for an additional 30 min. The parasites were washed and fixed with PBS/formaldehyde 1%, 10,000 parasites were acquired by flow cytometry, and

analyzed according to the frequency of parasite labeled (A) and the mean fluorescence intensity of the samples (B). The significant differences ( p< 0.05) between

experimental groups are indicated by asterisks. This experiment was performed three times and shown is a representative experiment.

interaction of mucin and GIPL with STI was observed. More-over, neither CCL2 nor STI bound to BSA. These data suggestthat CCL2 bind to molecules on the surface of trypomastigoteprobably through negatively charged protein.

3.5. CCL2 induces accumulation of parasites in the airpouch of T. cruzi-infected mice

We next asked if the chemokines play a role in the para-sites’ migration in vivo. Therefore, increasing concentrationsof CCL2 or medium (control) were injected in the air pouchesof infected mice and the leukocytes and parasites’ migrationquantified. The injection of CCL2 into the air pouches of in-fected mice caused the migration of parasites and leukocytes.This phenomenon was clearly dependent on the concentrationof chemokine injected (Fig. 6). Fewer parasites and leukocyteswere found in the exudates from medium-injected air pouches(Fig. 6A). Differently, the injection of CXCL9/MIG resulted insignificant migration of leukocytes, but not parasites (Fig. 6B).

These results suggest that CCL2, but not CXCL9, is able to in-duce accumulation of parasites in the air pouches.

4. Discussion

Chemokines make up an important defense mechanism inthe immune response, serving both as the sparks that initiateresponses to infection and as the fuel that feeds subsequent in-flammation. The importance of the chemokine system to hostdefense is reflected by its extensive inherent redundancy andthe elaborate mechanisms used by microbes to undermine oremploy its function. Microbial organisms have been successfulin subverting the chemokine system in a number of differentways, often using mechanisms derived from host pathways.In general, these mechanisms can be divided into four maingroups: production of a microbial protein able to directly inter-act with the chemokine system; altered expression of chemo-kines or receptors; blockage of chemokine receptor signalingpathways; or sabotage of chemokine proteins [18]. Chemo-kines are also involved in metastasis, chemo-attracting tumor

2.5 1000

25

50

75YColombian

CCL2 unlabeled

0

250

500

750

1000

***

CCL2 STI CCL2+

anti-CCL2

2.5 100

CCL2 unlabeled

CCL2 STI CCL2+

anti-CCL2

A

Bind

ing

of C

CL2

(%)

B

Fluo

resc

ence

inte

nsity

Fig. 3. CCL2 binds to trypomastigote forms of Y and Colombian strains. Labeled mouse CCL2 was incubated with trypomastigotes as described in the legend to

Fig. 2. Additionally, the parasites were incubated with 2.5- and 100-fold of not labeled CCL2, followed by the same protocol described. The frequency of parasites

stained (A) as well as by the mean fluorescence intensity (B) was determined using flow cytometry. The significant differences ( p< 0.001) between experimental

groups are indicated with three asterisks. This experiment was performed twice and shown is a representative experiment.

116 L.M. Yamauchi et al. / Microbes and Infection 9 (2007) 111e118

Fig. 4. Condroitin 6-sulphate inhibits CCL2 binding to trypomastigote forms of T. cruzi. The trypomastigote forms (Y strain) were incubated in the presence of

biotin-labeled STI, mouse CCL2, and CCL2 with condroitin 6-sulphate (10, 100, and 1000 mg/ml) followed by incubation with FITC-labeled avidin. In total,

10,000 parasites were acquired and the frequency of parasites stained was determined using flow cytometry. This experiment was performed twice and shown

is a representative experiment.

cells to organs far away from the original cancer. The chemo-kine receptors CXCR4 and CCR7/MIP-4 are highly expressedin human breast cancer cells, malignant breast tumors and me-tastases indicating that chemokines and their receptors playa critical role in determining the metastasis destination of tumorcells [19]. All these examples showed that chemokines arepotential targets and this system could easily be exploited bymicro-organisms as seen in viruses that encode chemokine- and

TMuc EGILP EMuc BSA0

5000

10000

15000

20000

25000CCL2STI

***

***

***

Rel

ativ

e Lu

min

esce

nce

Uni

t

Fig. 5. CCL2 binds to glycoproteins of parasite surface. Plate was coated with

glycoinositolphospholipid of epimastigote (GIPL), mucin of trypomastigote,

mucin of amastigote, and BSA (0.25 mg/well). CCL2 biotinylated or STI bio-

tinylated were added, incubated for 2 h, washed, followed by the addition of

streptoavidin peroxidase. The detection was performed with ECL plus and

the reaction was measured in a lumiometer (RLU, relative light unit). The sig-

nificant differences ( p< 0.001) between CCL2 binding to different glycopro-

teins and BSA are indicated with three asterisks. The bars represent

mean� SD of triplicate samples and are representative of one experiment

performed.

chemokine receptor-like protein [18] and Schistosoma man-soni eggs which secrete a chemokine binding protein withanti-inflammatory activity [20].

b-Chemokines such as CCL2, CCL3, CCL4 and CCL5 areproduced by macrophages and cardiomyocytes infected withT. cruzi, and their biological effects are the recruitment of leu-kocyte subsets to the site of infection and controlling the par-asite growth [4,5,21]. Besides, we have shown that CCL2 isproduced in higher quantities by infected cells, and it is themost potent stimulator of microbicidal activity by macro-phages and cardiomyocytes [4,5]. Here, we showed thatCCL2 has an important role in morphogenesis and chemotaxisin vivo of trypomastigote forms of T. cruzi. CCL2 was capableof chemo-attracting spheromastigotes into the air pouches ofinfected mice. Additionally, when we examined if anotherchemokine could also chemo-attract this parasite, we observedthat CXCL9 (a-chemokine) was able to cause the migration ofleukocytes but not of parasites. Another observation showedthat dextran, known to trigger vascular edema due to histaminerelease [22], was not capable of inducing significant levels ofsuch forms in the air pouch (data not shown). These observa-tions suggest that CCL2 could interact with trypomastigoteforms of T. cruzi and drive their migration to different tissues,exploiting host systems to their own benefit. The morpholog-ical change process observed is chemokine concentration-dependent. Moreover, the resulting spheromastigote formsare recognized by the mAb 2C2, suggesting that the interac-tion with CCL2 resulted in morphogenesis, since this antibodyonly recognizes Ssp-4, a molecule expressed on amastigotesurface [13,14]. Although the morphogenesis can be triggeredby pH, proteasome, and protein phosphatase A [23e25], here

117L.M. Yamauchi et al. / Microbes and Infection 9 (2007) 111e118

Fig. 6. The b-chemokine CCL2 induces accumulation of spherical forms (amastigote-like) in the air pouch of Trypanosoma cruzi-infected mice. BALB/c mice

were infected with 1� 104 bloodstream derived trypomastigotes of T. cruzi (Y strain) and the air pouch was made by injection of sterile air into the serosa cavity

in the back of these animals. During the parasitemia peak, medium (Med) or chemokines were injected into the air pouch cavity. After 6 h, the air pouch cavities

were washed with cold PBSeheparin and the parasite counting evaluated in a hemocytometer. (A) Addition of CCL2 in different concentrations (range 50e200 ng/

ml) and medium into the air pouches. (B) Addition of CCL2 (100 ng/ml), CXCL9 (100 ng/ml) and medium into the air pouches. The bars represent mean� SD of

triplicate samples and are representative of three independent experiments performed.

we suggest that CCL2 can also induce it; although the properintracellular mechanisms remain unsolved. The chemo-attrac-tion and morphogenesis could be independent processes. How-ever, as we showed before, the uptake of parasites is enhancedin the presence of chemokines [4]. Additionally, mice infectedwith amastigote forms exhibit higher parasitemia than thoseinfected with trypomastigote forms, due to differences in theimmune response triggered by these two forms and also be-cause these two forms use different mechanisms to attachand invade host cells [26]. Therefore, it is possible that thechemokines attract and facilitate the parasite uptake and, atthe same time, induce the morphogenesis process, facilitatingthe escape from the immune system. The morphogenesis hap-pens inside the phagocytic cells, although it could occur in theintertitium and circulation [13]. In fact, amastigote forms canbe found outside of cells [13]. In our model, the finding ofspheromastigotes is enhanced, since the transformation occursin exudates, where the binding of parasites and phagocytes ismore difficult.

CCL2 binds to the different forms of T. cruzi, but more withtrypomastigote forms, and with parasites of the Colombian andY strains. Moreover, the affinity of binding with trypomastigoteforms of the Colombian strain is lower than that of the Y strain,since less unlabeled CCL2 is necessary to displace labeledCCL2 from Colombian trypomastigote forms. Therefore, thechemokine chemo-attraction may explain the variation in para-site tropism described [14]; if a chemokine binds to a strain withdistinctive specificity, it could result in differences in tissue rec-ognition, suggesting that chemokines may play a role in thepathogenesis of T. cruzi infection. We have shown that CCL2 in-teracted with molecules on the parasite surface, and it can be dueto ionic interactions as observed in glycosaminoglycan-chemo-kine binding. CCL2 C-terminal is thought to tether the chemo-kine to the luminal side of the endothelial cells of the bloodvessel, via the ionic interaction between the two juxtaposed ba-sic amino acid residues [15] with the negatively charged

extracellular proteoglycans. In fact, condroitin 6-sulphate andheparin inhibited the interaction of CCL2 to trypomastigoteforms even at low concentrations, suggesting that the interactionbetween CCL2 and parasite surface is, certainly, weak. There-fore, like GAGechemokine-cell interactions, T. cruzi trypomas-tigotes could hold themselves in the blood vessel and transversethe endothelial cells.

The molecule candidates for this interaction are glycoinosi-tol phospholipids (GIPL) and mucin-like glycoproteins (GPImucin) which are involved in the adhesion, invasion and pro-tection of the parasite [27,28]. Our results clearly showed thatCCL2 binds to mucin of trypomastigote, epimastigote, andGILP of epimastigote forms. Nevertheless, it is possible thatthe disaccharide group in the carboxy-terminal region in mu-rine CCL2 binds to lectins or galactose-binding proteins pres-ent on the parasite surface. In accordance, some lectins areexpressed in T. cruzi, such as calreticulin and 67 kDa lectin-like glycoprotein [29,30].

The results presented here suggest that chemokines coulddrive parasites to certain tissues, which may explain the tro-pism previously described [14]. A thorough understanding ofthe mechanisms that drive the parasite to the tissue and triggerits morphogenesis certainly will be helpful for the comprehen-sion of Chagas’ disease pathogenesis.

Acknowledgments

This work was supported by The Millennium Institute forVaccine Development and Technology (CNPq-420067/2005-1)and FAPESP.

References

[1] E. Ortega-Barria, M. Pereira, A novel T. cruzi heparin-binding protein

promotes fibroblast adhesion and penetration of engineered bacteria

and trypanosomes into mammalian cells, Cell 67 (1991) 411e421.

118 L.M. Yamauchi et al. / Microbes and Infection 9 (2007) 111e118

[2] L.O. Andrade, N.W. Andrews, The Trypanosoma cruziehost-cell inter-

play: location, invasion, retention, Nat. Rev. Microbiol. 3 (2005) 819e823.

[3] B.A. Burleigh, A.M. Woolsey, Cell signalling and Trypanosoma cruzi in-

vasion, Cell. Microbiol. 4 (2002) 701e711.

[4] J.C.S. Aliberti, F.S. Machado, J.T. Souto, A.P. Campanelli,

M.M. Teixeira, R.T. Gazzinelli, J.S. Silva, b-Chemokines enhance para-

site uptake and promote nitric oxide-dependent microbiostatic activity in

murine inflammatory macrophages infected with Trypanosoma cruzi,Infect. Immun. 67 (1999) 4819e4826.

[5] F.S. Machado, G.A. Martins, J.C.S. Aliberti, F.L.A.C. Mestriner,

F.Q. Cunha, J.S. Silva, Trypanosoma cruzi-infected cardiomyocytes pro-

duce chemokine and cytokine that trigger potent nitric oxide-dependent

trypanocidal activity, Circulation 102 (2000) 3003e3008.

[6] R.T. Gazzinelli, S.H. Oswaldo, L. James, A. Sher, The microbicidal ac-

tivity of interferon-g treated macrophages against Trypanosoma cruzi in-

volves an L-arginine-dependent, nitrogen oxide-mediated mechanism

inhabitable by interleukin-10 and transforming growth factor-b, Eur. J.

Immunol. 22 (1992) 2501e2506.

[7] G.N.R. Vespa, F.Q. Cunha, J.S. Silva, Nitric oxide is involved in control

of Trypanosoma cruzi-induced parasitemia and directly kills the parasite

in vitro, Infect. Immun. 62 (1994) 5177e5182.

[8] M. Baggiolini, Chemokines and leukocyte traffic, Science 392 (1998)

565e568.

[9] B.J. Rollins, Chemokines, Blood 90 (1997) 909e928.

[10] J. Middleton, A.M. Patterson, L. Gardner, C. Schumutz, B.A. Ashton,

Leukocyte extravasation: chemokine transport and presentation by the

endothelium, Blood 100 (2002) 3853e3860.

[11] B.J. Rollins, A. Watz, M. Baggiolini, Recombinant human MCP-1/JE in-

duces chemotaxis, calcium flux, and the respiratory burst in human

monocytes, Blood 78 (1991) 1112e1116.

[12] M.I. Simon, M.P. Strathmann, N. Gautam, Diversity of G protein in sig-

nal transduction, Science 252 (1991) 802e808.

[13] N.W. Andrews, K.S. Hong, E.S. Robbins, V. Nussenzweig, Stage-specific

surface antigens expressed during the morphogenesis of vertebrate forms

of Trypanosoma cruzi, Exp. Parasitol. 64 (1987) 474e484.

[14] R.C. Melo, Z. Brener, Tissue tropism of different Trypanosoma cruzi

strains, J. Parasitol. 64 (1978) 476e482.

[15] L. Chakravarty, L. Rogers, T. Quach, S. Breckenridge, P.E. Kolattukudy,

Lysine 58 and histidine 66 at the C-terminal alpha-helix of monocyte

chemoattractant protein-1 are essential for glycosaminoglycan binding,

J. Biol. Chem. 273 (1998) 29641e29647.

[16] A. Acosta-Serrano, I.C. Almeida, L.H. Freitas-Junior, N. Yoshida,

S. Schenkman, The mucin-like glycoprotein super-family of Trypanosoma

cruzi: structure and biological roles, Mol. Biochem. Parasitol. 114 (2001)

143e150.

[17] A.C.C. Frasch, Functional diversity in the trans-sialidase and mucin fam-

ilies in Trypanosoma cruzi, Parasitol. Today 16 (2000) 282e286.

[18] A. Liston, S. McColl, Subversion of the chemokine world by microbial

pathogens, Bioessays 25 (2003) 478e488.

[19] A. Muller, B. Homey, H. Soto, N. Ge, D. Catron, M.E. Buchanan,

T. McClanahan, E. Murphy, W. Yuan, S.N. Wagner, J.L. Barrera,

A. Mohar, E. Verastegui, A. Zlotnik, Involvement of chemokine receptors

in breast cancer metastasis, Nature 410 (2001) 50e56.

[20] P. Smith, R.E. Fallon, N.E. Mangan, C.M. Walsh, M. Saraiva,

J.R. Sayers, A.N.J. McKEnzie, A. Alcami, P.G. Fallon, Schistosoma man-soni secretes a chemokine binding protein with anti-inflammatory activ-

ity, J. Exp. Med. 202 (2005) 1319e1325.

[21] M.M. Teixeira, R.T. Gazzinelli, J.S. Silva, Chemokines, inflam-

mation and Trypanosoma cruzi infection, Trends Parasitol. 18 (2002)

262e265.

[22] J.M. Harris, D.K. Luscombe, Carbohydrates and inflammation the local

effects of dextran, ovomucoid and zymosan in dextran-resistant rats,

Int. Arch. Allergy Appl. Immunol. 28 (1965) 50e59.

[23] S. Tomlinson, F. Vandekerckhove, U. Frevert, V. Nussenzweig, The in-

duction of Trypanosoma cruzi trypomastigote to amastigote transforma-

tion by low pH, Parasitology 110 (1995) 547e554.

[24] J. Gonzalez, F.J. Ramalho-Pinto, U. Frevert, J. Ghiso, S. Tomlinson,

J. Scharfstein, E.J. Corey, V. Nussensweig, Proteasome activity is re-

quired for the stage-specific transformation of a protozoan parasite,

J. Exp. Med. 184 (1996) 1909e1918.

[25] J. Gonzalez, A. Cornejo, M.R.M. Santos, E.M. Cordero, B. Gutierrez,

P. Porcile, R.A. Mortara, H. Sagua, J.F. da Silveira, J.E. Araya, A novel

protein phosphatase 2A (PP2A) is involved in the transformation of hu-

man protozoan parasite Trypanosoma cruzi, Biochem. J. 374 (2003)

647e656.

[26] V. Ley, N.W. Andrews, E.S. Robbins, V. Nusseszweig, Amastigotes of

Trypanosoma cruzi sustain an infective cycle in mammalian cells,

J. Exp. Med. 168 (1988) 649e659.

[27] V.L. Pereira-Chiaccola, A. Acosta-Serrano, I.C. Almeida,

M.A. Fergunson, T. Souto-Padron, M.M. Rodrigues, L.R. Travassos,

S. Schenkman, Mucin-like molecules form a negatively charged coat

that protects Trypanosoma cruzi trypomastigotes from killing by human

anti-a-galactosyl antibodies, J. Cell Sci. 113 (2000) 1299e1307.

[28] R.C. Ruiz, V.L. Rigoni, J. Gonzalez, N. Yoshida, The 35/50 kDa surface

antigen of Trypanosoma cruzi metacyclic trypomastigotes, an adhesion

molecule involved in host cell invasion, Parasite Immunol. 15 (1993)

121e125.

[29] C. Labriola, J.J. Cazzulo, A.J. Parodi, Trypanosoma cruzi calreticulin is

a lectin that binds monoglucosylated oligosaccharides but not protein

moieties of glycoproteins, Mol. Biol. Cell 10 (1999) 1381e1394.

[30] A.M. Silber, I.S. Marcipar, C. Roodveldt, P. Cabeza Meckert, R. Laguens,

A.J. Marcipar, Trypanosoma cruzi identification of a galactose-binding

protein that binds to cell surface of human erythrocytes and is involved

in cell invasion by the parasite, Exp. Parasitol. 100 (2002) 217e225.