The C-terminal extension of Leishmania pifanoi amastigote-specific cysteine proteinase Lpcys2: A putative function in macrophage infection

Page 1

Tp

MJBa

Mb

c

d

e

f

g

a

ARRAA

KLMCCI

1

cwiitbml

0d

Molecular & Biochemical Parasitology 162 (2008) 52–59

Contents lists available at ScienceDirect

Molecular & Biochemical Parasitology

he C-terminal extension of Leishmania pifanoi amastigote-specific cysteineroteinase Lpcys2: A putative function in macrophage infection

arcel Marín-Villaa,b,1, Diego A. Vargas-Inchausteguic,1, Suzana P. Chavesd, Antonio J. Temponea,uliana M.F. Dutraa, Maurilio J. Soarese, Tania Ueda-Nakamuraf, Sergio C.F. Mendoncag,artira Rossi-Bergmannd, Lynn Soongc, Yara M. Traub-Cseköa,∗

Laboratório de Biologia Molecular de Tripanosomatídeos e Flebotomíneos, Instituto Oswaldo Cruz, Fiocruz, Av. Brasil 4365,anguinhos, 21045-900 Rio de Janeiro, RJ, BrazilPECET-Universidad de Antioquia, Medellín, ColombiaDepartment of Microbiology and Immunology, The University of Texas Medical Branch, Galveston, TX, USAInstituto de Biofísica Carlos Chagas Filho, Universidade Federal do Rio de Janeiro, Rio de Janeiro, RJ, BrazilInstituto Carlos Chagas, Fiocruz, Curitiba, PR, BrazilCentro de Ciências da Saúde, Departamento de Análises Clínicas, Universidade Estadual de Maringá, Maringá, PR, BrazilLaboratório de Imunoparasitologia, Instituto Oswaldo Cruz, Fiocruz, Rio de Janeiro, RJ, Brazil

r t i c l e i n f o

rticle history:eceived 17 April 2007eceived in revised form 26 June 2008ccepted 11 July 2008vailable online 23 July 2008

eywords:eishmania

a b s t r a c t

Cysteine proteinases have been implicated in many aspects of protozoan parasite pathogenesis. Thesehydrolases are normally found as zymogens, and some classes in trypanosomatids possess a long C-terminal extension (CTE), for which no function has been assigned. In this paper we hypothesize that theCTE domain of Lpcys2, the abundant lysosomal cysteine proteinase of Leishmania pifanoi amastigotes, isinvolved in host cell infection. Confirming previous reports that this peptide is highly immunogenic inTrypanosoma cruzi, we detected antibodies against CTE in sera of leishmaniasis patients. We produced apolyclonal antibody specific to Lpcys2 CTE and determined that this antibody was capable of recognizing

acrophageysteine proteinase-terminal extension

nfection

both L. pifanoi and Leishmania amazonensis cysteine proteinases. Using this antibody, we detected a pre-dominant localization of Lpcys2 CTE in the lysosome and flagellar pocket of cultured axenic amastigotesof both parasite species; however, its location was shifted towards the surface of the parasites duringmacrophage infection. We examined the role of Lpcys2 CTE in macrophage infection and found a signif-icant reduction in the percentage of infected cells when macrophages were infected with L. pifanoi andL. amazonensis in the presence of anti-CTE antibody. This study suggests a role for leishmanial cysteineproteinases CTE at early stages of infection.

tc

hbco

. Introduction

Infection with Leishmania parasites can cause diverse clini-al diseases, ranging from cutaneous to visceral leishmaniasis,hich affects millions of people worldwide, mostly in develop-

ng countries [1]. Leishmania sp. are dimorphic parasites that existn a flagellate form in the insect vector. After being transmit-

ed to mammalian hosts, promastigotes are quickly phagocytosedy macrophages where they transform into amastigotes, whichultiply inside the phagolysosome and maintain infection. Since

eishmaniasis treatments presently available are not always effec-

∗ Corresponding author. Tel.: +55 21 3865 8110; fax: +55 21 2590 3495.E-mail address: [email protected] (Y.M. Traub-Csekö).

1 These authors contributed equally to this work.

m[a[os

nc

166-6851/$ – see front matter © 2008 Elsevier B.V. All rights reserved.oi:10.1016/j.molbiopara.2008.07.003

© 2008 Elsevier B.V. All rights reserved.

ive and pose serious side effects, finding alternative ways ofontrolling the disease is in demand.

Leishmania and other trypanosomatids, when compared toigher eukaryotes, present unique biological features which cane exploited with the intent of finding new chemotherapeuti-al/vaccine targets. One possibility is to target the initial eventsf parasite–host interaction, which may lead to disease establish-ent and are known to involve many different parasite molecules

reviewed in 2 and 3]. Parasite proteases are highly immunogenicnd have been exploited for serodiagnosis and as vaccine targets4,5]. Differences in relation to host protease homologues have

pened opportunities for chemotherapy with the development ofpecific inhibitors [6,7].

Cysteine proteinases have been extensively studied in Leishma-ia [reviewed in 8–10]. Amastigotes in the Leishmania mexicanaomplex, when compared to their promastigote counterparts, are

Page 2

ochem

kcwoiasmoavtmst

mofstsaaptAetdiprtuip

2

2

2(3aPMpbitpfmGEp

2

fi

[R31iscwripwvra

2

a1p(r(a(twTdadn

paCwShSTsId

2

7adh9lCwt

M. Marín-Villa et al. / Molecular & Bi

nown to contain high levels of cysteine proteinase activity, indi-ating a possible role for these proteinases in parasite survivalithin the hostile environment of the macrophage [11]. Inhibitors

f cysteine proteinases affect these parasites dramatically, impair-ng the in vitro transformation of promastigotes into amastigotes,nd diminishing macrophage infection [12]. Immunolocalizationtudies have shown that these enzymes are found in large lysoso-al organelles called megasomes [13,14], characteristic of parasites

f the L. mexicana complex. Knock-out experiments show that anbundant cysteine proteinase from L. mexicana may be acting as airulence factor [15,16]. When the copies of the genes coding forhis proteinase were removed, the parasite’s capacity for invading

acrophages was diminished, although these parasites may stillurvive in culture. Early studies have confirmed the efficacy of cys-eine protease inhibitors in the treatment of Leishmania major [17].

Our studies on cysteine proteinases from Leishmania pifanoi, aember of the L. mexicana complex, have led to the identification

f a multi-copy gene, Lpcys2, highly expressed in amastigoteorms [18]. The sequence of Lpcys2 predicts a hydrophobic leaderequence (pre-region), a proenzyme segment, and the charac-eristic C-terminal extension (CTE) domain. Immunolocalizationtudies by electron microscopy using an anti-catalytic domainntibody showed preferential labeling of the megasomes, whilen anti-CTE antibody was abundantly detected in the flagellarocket [14]. The unusually long CTE from cysteine proteinases ofrypanosomatids was first detected in Trypanosoma brucei [19].mong the functions suggested previously in the literature for thisxtension is a role in processing [20] or cellular targeting. Never-heless, our previous results indicate no role in targeting for thisomain [9]. Although CTE is found in other trypanosomatids [21],

t has no assigned function. In an attempt to understand the rolelayed by the leishmanial CTE domain in infection, we generatedecombinant CTE, that was used to detect anti-CTE antibodies inhe sera of patients, and anti-CTE polyclonal antibodies that weresed in immunolocalization studies and in macrophage infection

nhibition assays. This study suggests a role for leishmanial cysteineroteinases CTE at early stages of infection.

. Materials and methods

.1. Parasites

L. pifanoi (MHOM/VE/60/Ltrod) promastigotes were cultured at5 ◦C in M199 medium supplemented with 10% fetal bovine serumFBS) (Hyclone, Logan, UT). Axenic amastigotes were cultured at1 ◦C in F-29 medium as previously described [22]. Leishmaniamazonensis (RAT/BA/74/LV78) clone 12-2 (a gift from Dr. Kwang-oo Chang, Department of Microbiology and Immunology, Chicagoedical School, USA) was maintained as axenic amastigotes in com-

lete Grace’s insect cell culture medium (Invitrogen GIBCO, Carls-ad, CA), supplemented with 20% FBS pH 5.0 at 33 ◦C [24]. Parasite

nfectivity was maintained by passing parasites in highly suscep-ible BALB/c mice via subcutaneous inoculation of 107 stationary-hase parasites into footpads and subsequent harvest of parasitesrom lesions, according to published protocols [23]. Infected

acrophages for the EM study were obtained by infecting J774-8 cell lines grown overnight at 37◦C/5% CO2 in Dulbecco Modifiedssential Medium (DMEM) 10% FBS with L. pifanoi or L. amazonensisromastigotes for 24 h at 34◦C/5% CO2 at a 1:20 proportion.

.2. Recombinant peptide and antibody production

A DNA fragment encoding the CTE of Lpcys2 was ampli-ed from the genomic clone EMBL3-1971-b from L. pifanoi

igiot

ical Parasitology 162 (2008) 52–59 53

18] using the primers F5′ggatccgcacccagacccgtgatg and5′aagcttctacgtgtagtgacaggt. The amplification conditions were:5 cycles of denaturation at 94 ◦C for 30 s, annealing at 58 ◦C formin, and extension at 72 ◦C for 1 min. The PCR product was cloned

n pGEM-T Easy vector (Promega). The insert was sequenced, andub-cloned into the expression vector pQE-30 (Qiagen). Escherichiaoli M15pREP4 was transformed with this construct and inducedith 1 mM isopropyl �-d-thiogalactoside (IPTG). His-tagged

ecombinant protein was purified using His-SelectTM cobalt affin-ty gel (Sigma) under denaturing conditions. Rabbit and mouseolyclonal antibodies were prepared by subcutaneous inoculationith 100 �g of recombinant peptide in Freund’s complete adju-

ant. Animals were boosted 2 and 4 weeks later with 100 �g ofecombinant peptide plus Freund’s incomplete adjuvant and bledt 6 weeks.

.3. Western blot

L. pifanoi and L. amazonensis axenic amastigotes were washednd solubilized in lysis buffer (1 M Tris–HCl pH 8.0, 150 mM NaCl,% Triton X-100) containing the protease inhibitors PMSF and leu-eptin at 50 and 2 �g/ml, respectively. Similar amounts of protein30 �g) were separated by 15% SDS–PAGE. Parasite proteins orecombinant CTE were transferred onto nitrocellulose membranesHybondTM ECLTM, Amersham Pharmacia) for 1 h at 100 V at 4 ◦Cnd blocked overnight at 4 ◦C with 5% non-fat milk in TBS-TweenTBS-T) pH 7.6. The membranes were incubated with a 1:100 dilu-ion of the anti-CTE polyclonal antibody in 0.1% TBS-T for 1 h at RT,ashed 5 times, for 5 min each with TBS-0.1% Tween pH 7.6 (TBS-

). Peroxidase-conjugate goat anti-mouse IgG (Bio-Rad) at a 1:2000ilution in the blocking buffer with 3% non-fat milk in TBS-T wasdded with agitation for 1 h at RT, followed by TBS-T washes asescribed before. Membranes were developed using a chemilumi-escence Super Signal kit (PIERCE).

The immunogenicity of CTE was evaluated using sera from oneatient with diffuse cutaneous leishmaniasis (S72) caused by L.mazonensis, six patients with visceral leishmaniasis (AFS, JAS, JJD,ESS, EJB and MPMB) caused by Leishmania chagasi, eleven patientsith cutaneous leishmaniasis (S15, S19, S24, S50, S51, S53, S57,

62, S64, S65 and S76,) caused by Leishmania braziliensis, and sixealthy volunteers. Recombinant CTE (50 �g) was subjected to 17%DS–PAGE and transferred to a nitrocellulose membrane as above.he recombinant CTE containing membrane was incubated withera in a 1:50 dilution, peroxidase-conjugated goat anti-humangG in a dilution of 1:2000 and washed, according to the methodsescribed above.

.4. Transmission electron microscopy

Cells were collected, fixed for 1 h in 0.1 M cacodylate buffer pH.2 containing 0.5 glutaraldehyde, 4% formaldehyde and 1% picriccid [25]. Cells were dehydrated in ethanol at 4 ◦C and embed-ed in Lowycryl as described previously [26]. Polymerization of theydrophilic resin was conducted at −20 ◦C under UV radiation for6 h. Ultra-thin sections were collected on nickel grids. Sub-cellular

ocalization was performed using antibodies against recombinantTE or anti-Lpcys2 catalytic domain [14]. Sections were blockedith 5% non-fat milk, 0.01% Tween 20 and incubated for 1 h with

he polyclonal antibodies diluted 1:100 in blocking buffer. Label-

ng was detected by incubation for 1 h with 10 nm gold-labeledoat anti-rabbit antibody (BB International or Sigma) diluted 1:40n blocking buffer. Ultra-thin sections were stained with 5% aque-us uranyl acetate and lead citrate and observed in a Zeiss EM-10Cransmission electron microscope (Zeiss, Oberkochen, Germany)

Page 3

5 iochemical Parasitology 162 (2008) 52–59

os57LwFbTdtblPm

2

oomawm5pp1e[b(aatamaawgmsmdae

moarPd

Fig. 1. Western blot showing the processing of the CTE from axenic amastigotesoam

acmtece

P

3

3

aa

FpS

4 M. Marín-Villa et al. / Molecular & B

perating at 80 kV. Alternatively, after fixation the pellets weretained en bloc in 2% aqueous uranyl acetate, dehydrated through0% and 75% ethanol and after incubation in a 1:2 mixture of5% ethanol and LR White, then pure LR White, embedded intoR White resin medium grade (Structure Probe, West Chester, PA)hich was polymerized at 60 ◦C for 24 h. Ultra-thin sections on

ormvar-carbon coated nickel grids were incubated on blockinguffer (0.1% bovine serum albumin (BSA), 0.01 M glycine in 0.05 Mris-buffered saline (TBS)) for 15 min, then on primary antibodyiluted in diluting buffer (1% BSA in 0.05 M TBS) for 1 h at roomemperature and then overnight at 4 ◦C. After washing in blockinguffer they were reacted with secondary antibody labeled with col-

oidal gold (AuroProbe EM, Amersham Life Science, GE Healthcare,iscataway, NJ) diluted 1:20 in diluting buffer as recommended byanufacturer.

.5. Macrophage infection and inhibition assay

The infectivity of both promastigotes and axenic amastig-tes of L. pifanoi to macrophages was evaluated in the presencef anti-Lpcys2 CTE and catalytic domain antibodies. Peritonealacrophages were harvested from Swiss mice with DMEM and

llowed to adhere at 2 × 106 cells/well for 1 h at 37 ◦C in 24-ell culture plates. After removal of non-adherent cells, isolatedacrophages were incubated overnight in DMEM containing

% normal mouse serum (NMS). Meanwhile, cultured L. pifanoiromastigotes and axenic amastigotes were washed twice withhosphate-buffered saline (PBS) to remove serum and labeled with25 �M of the green fluorescent dye carboxy fluorescein diac-tate (CFSE) (Molecular Probes®) for 5 min at room temperature27]. After two washes to remove excess dye, parasites were incu-ated with NMS, anti-CTE or anti-catalytic domain mouse serum1:100) for 30 min. Macrophages were then immediately infectedt 34 ◦C with the antibody-promastigotes or amastigotes mixturest a cell ratio of 1:7 and 1:10, respectively. After 5 h of infection,he cells were washed with PBS to remove non-internalized par-sites, lysed with 200 �l distilled water and transferred to blackicrotiter plates. The fluorescence was read at 485 nm excitation

nd 528 nm emission using a Bio-Tek plate reader fluorometer,nd expressed as fluorescent units [28]. Alternatively, macrophagesere infected with L. pifanoi amastigotes as described above and

rown at 34 ◦C for 1, 6, 12 and 24 h. Samples were then fixed withethanol, dried on air and stained with Giemsa (1:10) for 1 h. The

lides were mounted in Permount and observed by bright fieldicroscopy with 100×magnification, and 500 cells were counted to

etermine the infection index. Results were expressed as a percent-ge of infected macrophages, using the average of two independentxperiments.

For L. amazonensis infection experiments, peritonealacrophages derived from C57BL/6 mice were grown for 12 h

n the top of glass cover slides in 24-well plates at 37 ◦C. L.mazonensis (LV78) axenic amastigotes were washed twice,esuspended at a concentration of 8 × 106 parasites in 1 ml ofBS, and pre-incubated with anti-CTE antibody at the indicatedilutions for 30 min at 25 ◦C. Macrophages were infected at

tpphs

ig. 2. Immunogenicity of CTE. Recombinant CTE (50 �g) was run in SDS–PAGE. The gel watients. 1: (S72) diffuse leishmaniasis patient; 2–7: (AFS, JAS, JJD, CESS, EJB and MPMB) v76) cutaneous leishmaniasis patients; 18–22: healthy volunteers; 23: secondary peroxid

f L. pifanoi and L. amazonensis. Anti-CTE antibody was used against: 1, L. pifanoixenic amastigotes; 2, L. amazonensis axenic amastigotes. Arrows: Precursor Lpcys2olecules. (*) CTE.

1:4 cell-to-parasite ratio for various times at 32 ◦C. Finally,over slides were washed, stained (Quick-Diff), and mounted onicroscope slides. Slides were analyzed at a 400× magnifica-

ion, and the number of infected cells was counted. Results arexpressed as percentage of infected macrophages (# infectedells/# total cells × 100), and pooled from two independentxperiments.

Data were analyzed by a one-way analysis of variance, GraphPadrism 4.0 (*p < 0.05, **p < 0.01, ***p < 001).

. Results

.1. Detection of processed CTE domain

The presence of the CTE domain was visualized by Western blotnalysis using cell lysates from L. pifanoi and L. amazonensis axenicmastigotes (Fig. 1). The anti-CTE polyclonal antibody recognizedhe expected CTE domain. The processed CTE was observed in L.

ifanoi as a doublet of approximately 15 and 12 kDa, as describedreviously [14], and of slightly larger size in L. amazonensis. Theigher molecular weight precursor molecule and processed precur-ors were more evident in L. pifanoi than in L. amazonensis (Fig. 1).

as transferred to nitrocellulose membrane to react with serum from leishmaniasisisceral leishmaniasis patients; 8–17: (S15, S50, S51, S53, S55, S57, S62, S64, S65 andase-labeled antibody alone.

Page 4

M. Marín-Villa et al. / Molecular & Biochemical Parasitology 162 (2008) 52–59 55

F icroscL cellull g is pc cleus;

3

afsSloSd

3

iama

FahP

ig. 3. Immunolocalization of CTE in axenic and intracellular parasites. Electron mpcys2 CTE antibody. Labeling is seen in the megasome and flagellar pocket. (C) Intraabeling. The arrowhead indicates the parasitophorous vacuole membrane: labelinytoplasm; FP, flagellar pocket; K, kinetoplast; L, lipid droplet; M, megasome; N, nu

.2. Anti-CTE antibody in human serum samples

Serum from patient S72 with diffuse leishmaniasis caused by L.mazonensis, and all six sera (AFS, JAS, JJD, CESS, EJB and MPMB)rom patients with visceral leishmaniasis caused by L. chagasitrongly reacted with the recombinant CTE (Fig. 2). Four (S62, S64,65 and S76) out of ten sera samples from patients with cutaneous

eishmaniasis showed a positive reaction, whereas sera from thether cutaneous leishmaniasis patients (S15, S50, S51, S53, S55 and57), healthy donors, and the peroxidase labeled conjugate aloneid not recognize CTE.

pmms

ig. 4. Immunolocalization of Lpcys2 CTE and catalytic domain in intracellular parasitesmazonensis amastigotes for 6 h. (A) Sections were stained with rabbit anti-CTE polycloneads). (B) Sections were stained with anti-Lypcys2 catalytic domain polyclonal antibody,V, parasitophorous vacuole. Bars = 0.5 �m.

opy of axenic L. amazonensis (A) and L. pifanoi (B) amastigotes reacted with anti-ar L. pifanoi amastigote infecting a macrophage. Arrows indicate Leishmania surfaceresent at the parasite membrane, but not at the vacuole membrane. HC, host cellLu, lumen of the parasitophorous vacuole. Bars = 1 �m.

.3. Subcellular localization of amastigote Lpcys2 CTE

The anti-CTE antibody was used in immunolocalization stud-es by transmission electron microscopy. In axenic amastigotes,bundant labeling was seen in the flagellar pocket and also in theegasome of both L. amazonensis and L. pifanoi parasites (Fig. 3A

nd B), while no labeling was seen on the cell surface. When L.

ifanoi promastigotes were put in contact with J774-G8 culturedacrophages for 24 h, and the infected cells were processed foricroscopy, the label was seen in the flagellar pocket and on the

urface of recently transformed amastigotes inside macrophage

. Peritoneal macrophages were obtained from C57BL/6 mice and infected with L.al antibody, showing the majority of labeling on the surface of the parasite (arrowshowing the majority of labeling within the parasite (arrows). N, parasite nucleus;

Page 5

56 M. Marín-Villa et al. / Molecular & Biochem

Fig. 5. Effect of anti-CTE antibody on macrophage infection with L. pifanoi and L.amazonensis. Macrophages (2 × 106 well−1) were infected for 5 h with (A) promastig-otes or (B) axenic amastigotes of L. pifanoi, that were pre-incubated with normal,anti-CTE or anti Lpcys2 catalytic domain mouse serum (1:100), as indicated. Par-asite loads were expressed as fluorescence units (UF). Mean ± S.D. in relation tonormal serum (n = 6 for promastigotes; n = 5 for amastigotes). (C) C57BL/6 peritonealmacrophages were infected with L. amazonensis axenic amastigotes at a 1:4 cell-to-p(t

psvLctLw

3

LcaCwbiAwetwraosamlo[idswapwneacabo

4

nllLrbtaWlcfawaisand megasome labeling observed by immunolocalization exper-

arasite ratio for 5 h. Macrophages were stained, and the percentage of infected cells# infected cells/# total cells × 100) was measured. Shown are pooled results fromwo independent experiments (average ± S.D.). *p < 0.05; **p < 0.01.

hagosomes (Fig. 3C). Decreased labeling was seen in the phago-ome membrane or dispersed in the lumen of the parasitophorousacuole. When mice peritoneal macrophages were infected with. amazonensis amastigotes for 6 h and stained with anti-CTE poly-lonal antibody, the majority of labeling was seen on the surface of

he parasite (Fig. 4A). On the other hand, sections stained with anti-ypcys2 catalytic domain antibody showed the majority of labelingithin the parasite (Fig. 4B).

irdo

ical Parasitology 162 (2008) 52–59

.4. Inhibition of infection by anti-CTE antibodies

In initial experiments, when macrophages were infected with. pifanoi promastigotes in the presence of anti-CTE or anti-Lpcys2atalytic domain antibody, after 5 h neither antibody significantlyltered infectivity, in spite of the slight reduction seen with anti-TE (Fig. 5A). On the other hand, when macrophages were infectedith L. pifanoi amastigotes pre-incubated with the anti-CTE anti-

ody, a significant reduction in the number of parasites was seenn infected cells in relation to normal serum (p < 0.01) (Fig. 5B).

less pronounced inhibition was achieved when the amastigotesere pre-incubated with anti-catalytic antibody (Fig. 5B). Similar

xperiments were done using L. amazonensis axenic amastigoteso infect peritoneal macrophages, with or without pre-incubationith anti-CTE antibodies (Fig. 5C). Again, a diminished infection

ate was observed when macrophages were infected with axenicmastigotes in the presence of anti-CTE antibody. No effect wasbserved when promastigotes were used for infection (data nothown). The reduced number of parasites was due to decreasedmastigote internalization and not to enhanced killing by theacrophages during the 5-h infection period, since the intracel-

ular growth of parasites of the L. mexicana complex has beenbserved only 15–18 h after infection with culture amastigotes27,29], and as indicated by our further results. In further exper-ments, where different times after infection were investigated,iminished infection rate was observed with both L. amazonen-is and L. pifanoi as early as 1 h post infection when macrophagesere infected with axenic amastigotes in the presence of anti-CTE

ntibody (Fig. 6). The inhibitory effect of the anti-CTE antibodyersisted up to 24 h post-infection. Minimal effect was observedhen promastigotes were used in the L. amazonensis infection (dataot shown). As shown in Fig. 6A, when the parasite burden wasvaluated, pre-incubation of L. amazonensis amastigotes with thenti-CTE antibody induced a decrease in the percentage of cellsontaining high parasite numbers (≥9 parasites/cell, p < 0.001) andn increase in the number of cells presenting no or a low num-er of parasites (≤3 parasites/cell, p < 0.001). Similar results werebserved with L. pifanoi (Fig. 6B).

. Discussion

In previous immunolocalization studies using cultured Leishma-ia, we observed that an anti-Lpcys2 catalytic domain antibody

abeled mostly the megasome, and an anti-CTE mAb while alsoabeling the megasome, localized heavily to the flagellar pocket of. pifanoi amastigotes, suggesting a possible processing after Lpcys2eaches the lysosome [14]. No surface labeling was observed usingoth antibodies. To extend these findings and examine the func-ional role of the CTE domain, we produced an anti-CTE polyclonalntibody, which also recognized the L. amazonensis proteinase.hen this antibody was tested in Western blots using L. pifanoi

ysates, we observed the expected high-molecular-weight pre-ursor molecules, and also two bands relative to the processedorms of the CTE, as previously observed [14]. In L. amazonensis,

strong double-band of a slightly larger size was also detected,hile precursor molecules were also visible, but not as strongly

s in L. pifanoi. The heavy CTE bands evident in the Western blotsndicate an accumulation of these molecules, and suggest hightability, which is compatible with the intense flagellar pocket

ments using the anti-CTE antibody. This is in agreement witheports of Gonzalez et al. [30] that developed an ELISA assay toetect Trypanosoma cruzi cruzipain and CTE in serum and urinef infected mice. Levels of CTE were about 10 times higher than

Page 6

M. Marín-Villa et al. / Molecular & Biochem

Fig. 6. Effect of anti-CTE antibody in L. pifanoi and L. amazonensis amastigote infec-tion of macrophages at various times of infection. Macrophages were infected with(A) L. amazonensis or (B) L. pifanoi axenic amastigotes that were pre-treated withindicated antibodies or control sera. The total number of infected cells was deter-mpdv

tt

epw

owat

ipshtetsawotfn

tidnp

tcppCflHmwmrao

tEotiasmsiitCmaLpio

ined at 1, 6, 12 and 24 h post-infection, and parasite loads were determined at 6 host-infection. *p < 0.05, **p < 0.01 and ***p < 0.001 indicate statistically significantifferences between control and anti-CTE treated groups by a two-way analysis ofariance. ns: not significant.

hose of cruzipain in these fluids, indicating the high stability of

his domain.

In T. cruzi, it has been demonstrated that the cruzipain CTExposes highly immunogenic epitopes that are absent in the nativeroteinase [31]. In patients with chronic Chagas disease, most seraere found to react with the CTE, thus suggesting the presence

tatba

ical Parasitology 162 (2008) 52–59 57

f immunodominant B-cell epitopes within this protein domain. Itas suggested that cruzipain consists of an enzymatic domain andnon-enzymatic immunodominant domain, which corresponds to

he CTE [31].Our results also suggest that the Leishmania CTE domain is

mmunogenic, since antibodies were seen in sera of leishmaniasisatients. All sera samples from patients with visceral leishmania-is were positive, which is expected, since these patients normallyave high levels of anti-Leishmania antibodies. The same is true forhe diffuse cutaneous disease caused by L. amazonensis which gen-rates a strong humoral response. On the other hand, only some ofhe patients with cutaneous leishmaniasis caused by L. brazilien-is reacted with CTE. This is not unexpected, since anti-Leishmaniantibody titers in these patients many times are low [32]. Althoughe could not find any clinical feature associated with the negative

r positive results, more studies would be necessary to determinehe different response among patients. Also, a positive reaction wasound in the sera of BALB/c mice infected with L. amazonensis (dataot shown).

The observation of a possible migration of the CTE moleculeso the parasite surface, upon contact with macrophages, led us tonvestigate a role for these molecules in vitro. We detected a markedecrease in macrophage infection by both L. pifanoi and L. amazo-ensis axenic amastigotes, when infection was performed in theresence of anti-CTE antibody.

Souto-Padrón et al. [33] produced a polyclonal antibody againsthe T. cruzi cysteine proteinase that was used for immunolo-alization by electron microscopy and studying the role of theroteinase in parasite–host cell interactions. This antibody wasroduced against the purified proteinase, which contained theTE. Lysosomal labeling was seen in all forms of T. cruzi, andagellar pocket labeling was seen in cultured trypomastigotes.eavy-surface labeling was observed in amastigote forms fromammalian infected cells. When the antibody was incubatedith the parasites prior to infection, a reduced uptake by theacrophages was observed. These results show an interesting cor-

elation to our results described here, which could be indicative ofconserved mechanism of immunomodulation by the CTE domainn the macrophage, both in Leishmania and T. cruzi.

Leishmania has the reputation of being a stealthy parasite, inhat it evades the initial activation of the immune response [34].ntrance into macrophages is mediated by the endocytosis ofpsonized parasites, which are recognized by membrane recep-ors present on the macrophage surface [3]. One molecule involvedn parasite–macrophage interaction is gp63, an abundant met-lloprotease expressed on the parasite surface of all Leishmaniapecies, which participates in the interaction of parasites withacrophages and the complement system [35,36]. Overexpres-

ion of gp63 in L. amazonensis caused increased infection andntracellular parasite survival [37], while gp63-deficient parasitesnfecting macrophages or BALB/c mice, showed a diminished infec-ion and survival [38]. Other molecules clearly involved with theR3 macrophage receptor are L. infantum PSA-2 protein and the L.ajor proteophosphoglican (PPG) [39]. These authors showed that

nti-PSA-2 and PPG antibodies inhibited macrophage invasion by. major, this inhibition being more notable in amastigotes thanromastigotes. These observations are consistent with our find-

ngs, indicating a possibility of other as-yet-unknown moleculesf Leishmania with a role in macrophage infection.

Available evidence suggests a role of Leishmania cysteine pro-

einases in virulence. When the abundant cysteine proteinases generray of L. mexicana was disrupted, there was a marked pheno-ype affecting virulence, with infectivity to macrophages reducedy 80% [15]. Later studies employing these deficient mutants inhamster model reported evidence of a delayed disease onset

Page 7

5 iochem

wtafhwipoCin

amttoaimpiC

A

OGfVlmrm

R

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

8 M. Marín-Villa et al. / Molecular & B

ith smaller lesions, lower parasite burden and lower infectivityo human mononuclear phagocytic cells [40]. These results are ingreement with our observations, although there might be otheractors at play in these responses. The previously observed effectas been interpreted as being due to enzyme activity, while heree suggest a role for a specific domain of the cysteine proteinase

n infection, for the first time. Leishmania has been shown to sup-ress macrophage responses, such as phagocytosis and productionf iNOS [41,42]. Also, Leishmania were shown to block protein kinaseactivity in macrophages [43], and blocking this macrophage activ-

ty prior to infection induced an increase of intracellular amastigoteumbers [44].

Although Leishmania cysteine proteinase has been identifiednd studied for more than 30 years [45], there is still limited infor-ation on its functional domains in host–parasite interactions. In

his study, we have focused on the effects on infection of a cys-eine proteinase CTE domain. Our results for the surface locationf CTE, the inhibition of amastigote infectivity to macrophages bynti-CTE antibodies, as well as the presence of anti-CTE antibodiesn some leishmaniasis patients and in mice infected with Leish-

ania, collectively indicate that the CTE domain of the cysteinerotease of Leishmania may serve as a critical molecule involved

n parasite entrance into cells. They also suggest that production ofTE-blocking antibodies by the host might help to control infection.

cknowledgements

This work was supported by Papes IV-FIOCRUZ, Institutoswaldo Cruz-Fiocruz and by a National Institutes of Healthrant AI43003 to L. Soong. Marcel Marín V. received fellowships

rom COLCIENCIAS and Instituto Oswaldo Cruz-FIOCRUZ. Diegoargas-Inchaustegui was supported by the James W. McLaugh-

in Fellowship Fund. We thank Dr. Vsevolod L. Popov for electronicroscopy experiments, Dr. Diane McMahon-Pratt for critical

eading of the manuscript and Mardelle Susman for assisting inanuscript preparation.

eferences

[1] Lainson R, Rangel EF. Lutzomyia longipalpis and the eco-epidemiology of Amer-ican visceral leishmaniasis, with particular reference to Brazil. Mem InstOswaldo Cruz 2005;100:811–27.

[2] Denkers EY, Butcher BA. Sabotage and exploitation in macrophages parasitizedby intracellular protozoans. Trends Parasitol 2005;21:35–41.

[3] Rittig MG, Bogdan C. Leishmania–host-cell interaction: complexities and alter-native views. Parasitol Today 2000;16:292–7.

[4] Redmond DL, Knox DP. Further protection studies using recombinant forms ofHaemonchus contortus cysteine proteinases. Parasite Immunol 2006;28:213–9.

[5] Raina OK, Yadav SC, Sriveny D, Gupta SC. Immuno-diagnosis of bubaline fas-ciolosis with Fasciola gigantica cathepsin-L and recombinant cathepsin L 1-Dproteases. Acta Trop 2006;98:145–51.

[6] Cazzulo JJ. Proteinases of Trypanosoma cruzi: potential targets for thechemotherapy of Chagas disease. Curr Top Med Chem 2002;2:1261–71.

[7] Sajid M, McKerrow JH. Cysteine proteases of parasitic organisms. Mol BiochemParasitol 2002;120:1–21.

[8] Traub-Cseko YM, Almeida RW, Boukai LK, Costa-Pinto D, Duboise SM,McMahon-Pratt D. Cysteine proteinases of Leishmania. Ciencia e Cultura1993;45:339–42.

[9] Costa-Pinto D, Trindade LS, McMahon-Pratt D, Traub-Cseko YM. Cellulartrafficking in trypanosomatids: a new target for therapies? Int J Parasitol2001;31:537–44.

10] Mottram JC, Coombs GH, Alexander J. Cysteine peptidases as virulence factorsof Leishmania. Curr Opin Microbiol 2004;7:375–81.

11] North MJ, Coombs GH. Proteinases of Leishmania mexicana amastigotesand promastigotes: analysis by gel electrophoresis. Mol Biochem Parasitol1981;3:293–300.

12] North MJ. Comparative biochemistry of the proteinases of eukaryotic microor-

ganisms. Microbiol Rev 1982;46:308–40.

13] Pupkis MF, Tetley L, Coombs GH. L. mexicana: amastigote hydrolases in unusuallysosomes. Exp Parasitol 1986;62:29–39.

14] Duboise SM, Vannier-Santos MA, Costa-Pinto D, et al. The biosynthesis, pro-cessing, and immunolocalization of Leishmania pifanoi amastigote cysteineproteinase. Mol Biochem Parasitol 1994;68:119–32.

[

[

ical Parasitology 162 (2008) 52–59

15] Mottram JC, Souza AE, Hutchinson JE, Carter R, Frames MJ, Coombs JG. Evi-dence from disruption of the lmcpb gene array of Leishmania mexicana thatcysteine proteinases are virulence factors. Proc Natl Acad Sci USA 1996;93:6008–13.

16] Descoteaux A. Leishmania cysteine proteinases: virulence factors in quest of afunction. Parasitol Today 1998;14:220–1.

17] McKerrow JH, Engel JC, Caffrey CR. Cysteine protease inhibitors as chemother-apy for parasitic infections. Bioorg Med Chem 1999;7:639–44.

18] Traub-Cseko YM, Duboise SM, Boukai LK, McMahon-Pratt D. Identification oftwo distinct cysteine proteinase genes of Leishmania pifanoi axenic amastig-otes using the polymerase chain reaction. Mol Biochem Parasitol 1993;57:101–16.

19] Mottram JC, North MJ, Barry JD, Coombs GH. A cysteine proteinase cDNA fromTrypanosoma brucei predicts an enzyme with an unusual C-terminal extension.FEBS Lett 1989;258:211–5.

20] Paul W, Amiss J, Try R, Praekelt U, Scott R, Smith H. Correct processing of thekiwifruit protease actinidin in transgenic tobacco requires the presence of theC-terminal propeptide. Plant Physiol 1995;108:261–8.

21] Aslund L, Henriksson J, Campetella O, Frasch ACC, Petersson U, Cazzullo JJ.The C-terminal extension of the major cysteine proteinase (cruzipain) fromTrypanosoma cruzi. Mol Biochem Parasitol 1991;45:345–8.

22] Pan AA, Duboise SM, Eperon S, et al. Developmental life cycle of Leishmania-cultivation and characterization of cultured extracellular amastigotes. JEukaryot Microbiol 1993;40:213–23.

23] Bates PA, Robertson CD, Tetley L, Combs GH. Axenic cultivation and char-acterization of Leishmania mexicana amastigotes-like forms. Parasitology1992;105:193–202.

24] Dutta S, Ray D, Kolli BK, Chang KP. Photodynamic sensitization of Leishma-nia amazonensis in both extracellular and intracellular stages with aluminumphthalocyanine chloride for photolysis in vitro. Antimicrob Agents Chemother2005;49:4474–84.

25] Berryman MA, Rodewald RD. An enhanced method for post-embeddingimmunocytochemical staining which preserves cell membranes. J HistochemCytochem 1990;38:159–70.

26] Bendayan M, Nanci A, Kan W. Effect of tissue processing on colloidal gold cyto-chemistry. J Histochem Cytochem 1987;35:983–96.

27] Qi H, Popov V, Soong L. Leishmania amazonensis-dendritic cell interactionsin vitro and the priming of parasite-specific CD4+ T cells in vivo. J Immunol2001;167:4534–42.

28] Boeck P, Bandeira Falcao CA, Leal PC, et al. Synthesis of chalcone analogues withincreased antileishmanial activity. Bioorg Med Chem 2006;14:1538–45.

29] Puentes F, Diaz D, Hoya RD, et al. Cultivation and characterization of stableLeishmania guyanensis complex axenic amastigotes derived from infected U937cells. Am J Trop Med Hyg 2000;63:102–10.

30] Gonzalez G, Sunnemark D, Orn A, Gronvik KO. Detection of cruzipain, themajor cysteine proteinase from Trypanosoma cruzi and its C-terminal exten-sion in biological fluids during experimental infection in mice. Scand J Immunol1996;44:122–8.

31] Martinez J, Campetella O, Frasch AC, Cazzulo JJ. The reactivity of sera fromchagasic patients against different fragments of cruzipain, the major cysteineproteinase from Trypanosoma cruzi, suggests the presence of defined antigenicand catalytic domains. Immunol Lett 1993;35:191–6.

32] Mendonca SC, Souza WJ, Nunes MP, Marzochi MC, Coutinho SG. Indirectimmunofluorescence test in New World leishmaniasis: serological and clinicalrelationship. Mem Inst Oswaldo Cruz 1988;83:347–55.

33] Souto-Padrón T, Campetella OE, Cazzulo JJ, Souza W. Cysteine proteinasesin Trypanosoma cruzi: immunocytochemical localization and involvement inparasite–host cell interaction. J Cell Sci 1990;98:485–90.

34] Reiner SL, Zheng S, Wang ZE, Stowring L, Locksley RM. Leishmania promastigotesevade interleukin 12 (IL-12) induction by macrophages and stimulate a broadrange of cytokines from CD4C T cells during initiation of infection. J Exp Med1994;179:447–56.

35] Chang KP, Chaudhuri G, Fong D. Molecular determinants of Leishmania viru-lence. Annu Rev Microbiol 1990;44:499–529.

36] Brittingham A, Morrison CJ, McMaster WR, McGwire BS, Chang KP, MosserDM. Role of the Leishmania surface protease gp63 in complement fixa-tion, cell adhesion, and resistance to complement-mediated lysis. J Immunol1995;155:3102–11.

37] Chen DQ, Kolli BK, Yadava N, et al. Episomal expression of specific senseand antisense mRNAs in Leishmania amazonensis: modulation of gp63 levelin promastigotes and their infection of macrophages in vitro. Infect Immun2000;68:80–6.

38] Thiakaki M, Kolli B, Chang KP, Soteriadou K. Down-regulation of gp63 level inLeishmania amazonensis promastigotes reduces their infectivity in BALB/c mice.Microbes Infect 2006;8:1455–63.

39] Kedzierski L, Montgomery J, Bullen D, et al. A leucine-rich repeat motif ofLeishmania parasite surface antigen 2 binds to macrophages through the com-plement receptor 3. J Immunol 2004;172:4902–6.

40] Saravia NG, Escorcia B, Osorio Y, Valderrama L, Arteaga L, Travi BL. Pathogenic-

ity and protective immunogenicity of cysteine proteinase-deficient mutants ofLeishmania mexicana in non-murine models. Vaccine 2006;24:4247–59.

41] Olivier M, Tanner CE. Susceptibilities of macrophage populations to infectionin vitro by Leishmania donovani. Infect Immun 1987;55:467–71.

42] Proudfoot L, Nikolaev AV, Feng GJ, et al. Regulation of the expression ofnitric oxide synthase and leishmanicidal activity by glycoconjugates of Leish-

Page 8

ochem

[

M. Marín-Villa et al. / Molecular & Bi

mania lipophosphoglycan in murine macrophages. Proc Natl Acad Sci USA1996;93:10984–9.

43] Olivier M, Brownsey RW, Reiner NE. Defective stimulus-response couplingin human monocytes infected with Leishmania donovani is associated withaltered activation and translocation of protein kinase C. Proc Natl Acad Sci USA1992;89:7481–5.

[

[

ical Parasitology 162 (2008) 52–59 59

44] Moore KJ, Labrecque S, Matlashewski G. Alteration of Leishmania donovaniinfection levels by selective impairment of macrophage signal transduction.J Immunol 1993;150:4457–65.

45] Zuckerman A. Current status of the immunology of blood and tissue Protozoa.I Leishmania Exp Parasitol 1975;38:370–400.