Cysteine proteinase in Trypanosoma cruzi: immunocytochemical

localization and involvement in parasite-host cell interaction

THAIS SOUTO-PADRON, OSCAR E. CAMPETELLA, JUAN J. CAZZULO

and WANDERLEY DE SOUZA

Laboratdrio de Ultra-estrutura Celular e Microscopia Eletrdnica, Departamento de Parasitologia e Bioflsica Celular,Institute de Biofisica Carlos Chagas Filho, Universidade Federal do Rio de Janeiro, Ilha do Fundao, 21949, Rio de Janeiro, Brasiland Instituto de Investigaciones Bioquimicas, Fundacion Campomar, Antonio Machado 151, 1405 Buenos Aires, Argentina

Summary

A monospecific polyclonal antibody obtained againsta cysteine proteinase isolated from epimastigotes ofTrypanosoma cruzi was used for the immunocyto-chemical localization of the protein by electron mi-croscopy and to analyse the role played by cysteineproteinase in the process of T. cruzi-host cell interac-tion. Cytoplasmic structures that correspond to el-ements of the endosomal-lysosomal (reservosome)system found in epimastigote, amastigote and trypo-mastigote forms reacted intensely with colloidalgold-labelled antibodies using on-section indirectlabelling. The surface of most of the tissue culture-derived trypomastigotes was not labelled. However,the flagellar pocket of this form was labelled. Allepimastigotes obtained from axenic cultures andamastigote-like forms found in the supernatant ofvertebrate cells heavily infected with T. cruzi had

their surface intensely labelled, indicating also thesurface localization of the protein. Incubation of theparasites in the presence of a sub-agglutinating con-centration of the anti-cysteine proteinase antibodyled to a marked increase in their uptake by macro-phages. In contrast, addition of the F(ab')2 portion ofthe same antibody significantly reduced the uptakeof the parasites by the macrophages. The resultsobtained strongly suggest an important participationof cysteine proteinase in the process of T. cruzi-macrophage interaction.

Key words: Trypanosoma cruzi-macrophage interaction,cysteine proteinase, lysosomal enzyme, immunocytochemistry,electron microscopy.

Introduction

Studies carried out in recent years have shown thatTrypanosoma cruzi infects vertebrate cells by a process ofendocytosis following an initial step of parasite-host cellrecognition (see review by Zingales and Colli, 1985;Araujo-Jorge, 1990). In this process components located onthe surface of the parasite and the host cell play animportant role.

There is some evidence suggesting the involvement ofparasite proteinases in the process of T. cruzi-host cellinteraction. Treatment of infective forms with proteinasesincreases their interaction with host cells (Piras et al.1985). Addition to the interaction medium of alpha-2-macroglobulin, a glycoprotein found in high concentrationin the plasma of vertebrates that acts as a generalproteinase inhibitor, markedly inhibits the infection ofmacrophages and fibroblasts by T. cruzi (Araujo-Jorge etal. 1986).

We have recently isolated and characterized a lysosomalcysteine proteinase, for which we propose the trivial name'cruzipain' (Cazzulo et al. 19906), from epimastigotes of T.cruzi. This enzyme: (1) is able to degrade bovine serumalbumin, hemoglobin and azocasein at pH 5.0 (Bontempi etal. 1989); (2) contains amino acid sequences presentingconsiderable homology with papain and cathepsin L (Caz-Journal of Cell Science 96, 485-490 (1990)Printed in Great Britain © The Company of Biologists Limited 1990

zulo et al. 1989a); and (3) is a high-mannose type ofglycoprotein, as shown by binding to concanavalin A-Sepharose, endo-AT-acetyl-glucosaminidase H sensitivity,and determination of oligosaccharide chain composition(Cazzulo et al. 1989a, 1990a). The purified protein has beenused to obtain polyclonal antibodies (Campetella et al.1990). In the present paper we describe the results ob-tained using these antibodies, as well as antibodies thatrecognize the cytosolic enzyme NADP-linked glutamatedehydrogenase (Cazzulo et al. 19896), for the immunocyto-chemical localization of the proteins in thin sections of theepimastigote, amastigote and trypomastigote forms of T.cruzi. F(ab')2 portions of the antibodies are used inexperiments on T. cruzi—macrophage interaction. Ourobservations show that: (1) cysteine proteinase is locatedin lysosomes of epimastigotes but is also expressed on thesurface of epimastigotes and amastigote-trypomastigotetransitional forms; and (2) addition of anti-proteinaseantibodies to the interaction medium significantly inhibitsthe ingestion of parasites by macrophages.

Materials and methods

Antigens and seraCysteine proteinase and NADP-linked glutamate dehydrogenase

485

were purified as previously described (Cazzulo et al. 1989a,6). Toobtain polyclonal antisera rabbits were immunized by intramus-cular injection (hind legs) of 50 /<g of purified enzyme in 1 ml ofphosphate-buffered saline (PBS), emulsified (1:1, v/v) with com-plete Freund's adjuvant (GIBCO). Three more doses were injectedsubcutaneously, emulsified with incomplete Freund's adjuvant(GIBCO), at 20-day intervals. The presence of specific antibodieswas detected by the method of Outchterlony (titre 1/160); only oneband, with identical migration, was observed when either thepurified enzyme or the cell-free extract was used as antigen. Thiswas confirmed using Western blots, as described previously(Campetella et al. 1990). F(ab')2 fragments were obtained bystandard methods (Stanworth and Turner, 1978).

ParasitesThe Y strain of Trypanosoma cruzi was used for the localization ofthe antigens. Epimastigote forms were cultivated for 5 days at38 °C in a liver-infusion trypticase medium containing 10% fetalcalf serum (Camargo, 1964). Trypomastigote and amastigoteforms were obtained from the supernatant of Vero cell culturesinfected with bloodstream trypomastigote forms of T. cruzi aspreviously described in detail (Carvalho and De Souza, 1983).

Electron microscopy and immunocytochemistryThe parasites were collected by centrifugation (2000 g for 10 minat 4°C), washed twice with phosphate-buffered saline and fixed for60 min in a solution containing 0.1 % glutaraldehyde, 2% formal-dehyde (freshly prepared from paraformaldehyde) in 0.1 M phos-phate buffer, pH7.2. After fixation the parasites were washedtwice with PBS, dehydrated in methanol and embedded inLowicryl K4M at -20°C (Bendayan et al. 1987). Thin sectionswere collected on 300-mesh nickel grids, incubated subsequentlyfor 60 min at room temperature in a PBS solution, pH8.0,containing 5 % non-fat milk and 0.01 % Tween 20 (PBS-MT), andthen in the same solution containing the polyclonal antibody.After incubation the grids were washed three times with PBScontaining 1 % bovine serum albumin and 0.01 % Tween 20,pH 8.0, and incubated for 60 min at room temperature in PBS-MTsolution containing gold-labelled goat anti-rabbit IgG (E-Y Lab-oratories, USA) diluted 1:20. Control grids were incubated in thepresence of non-immune rabbit serum, sera that recognizes otherparasite proteins (Souto-Padr6n et al. 1989) or only in thepresence of the gold-labelled antibody. After incubation the gridswere washed with PBS and distilled water, stained with uranylacetate and lead citrate and observed in a JEOL 100 CX or Zeiss902 transmission microscope.

MacrophagesCells were collected from peritoneal cavities of uninfected Swissmice after injection of 4 ml of Hank's balanced salt solution(HBSS). Samples of 300-400^1 were plated in Falcon multiwelltissue culture plates (15 mm diameter/well) with coverslips andmaintained in a humidified 5% CO2 atmosphere at 37 °C. Afterincubation for 30 min, the non-adherent cells were removed andmacrophages were cultivated for 24 h in 199 medium sup-plemented with 10 % heat-inactivated fetal calf serum.

Pre-treatment of parasites with anti-serum or F(ab')2fragmentsTrypomastigote and epimastigote forms obtained from the super-natant of Vero cells or from axenic media, respectively, werewashed twice in 199 medium at 4°C without serum in Eppendorfmicrofuge tubes and then treated with either a 1:200 dilution ofthe anti-cysteine proteinase serum or lO^gml"1 of the F(ab')2portion of the purified antibody in medium 199 without fetal calfserum for 60 min at 4°C.

The T. cruzi-macrophage interaction and evaluation ofresultsThe interaction of T. cruzi and macrophages was allowed to occurfor 60 min at 37 °C in medium 199 without fetal calf serum. Theinitial parasite: cell ratio was adjusted to 10:1 by counting the

number of macrophages per well in an inverted microscope andthe number of parasites in a Neubauer chamber. Fixation was inBouin's fixative and staining with Giemsa. The percentage ofparasites associated with macrophages and the mean number ofparasites per macrophage were calculated. We also obtained theassociation index that is the product of the two indices mentionedabove. The results are expressed as mean±standard deviations.

Results

Cysteine proteinaseLabelling of trypomastigotes. Typical trypomastigote

forms obtained from the supernatant of VERO cells andidentified by the morphology of the kinetoplast DNAnetwork were immunocytochemically labelled with anti-bodies associated with colloidal gold and analysed byelectron microscopy. The results showed a faint labellingof the cell and flagellar membranes (Figs 1-2). In contrast,we found an intense labelling of several intracellularvacuoles located close to the flagellar pocket region as wellas within the pocket (Figs 2-3).

Labelling ofamastigotes. Amastigote forms can be mor-phologically characterized by their rounded to oval shapeand by the structure of the kinetoplast DNA network,which appears as a slightly concave disk formed by afilamentous material arranged in a tightly packed row offibers perpendicularly oriented in relation to the longi-tudinal axis of the protozoan.

All amastigote forms obtained from the supernatant ofVERO cells showed an intense labelling of the surface(Figs 1, 3). We also found an intense labelling of intracellu-lar vesicles distributed throughout the cytoplasm of theparasite, in the membrane lining the flagellar pocket aswell as on the flagellar membrane.

Labelling of epimastigotes. Epimastigote forms ob-tained from axenic culture showed a very intense labellingof the membrane lining the cell body and the flagellum(Figs 4-5). Gold particles could be also observed at theregion of adhesion between cell body and the flagellum,inside the flagellar pocket and even in the cytostome. Anintense labelling of several intracellular vesicles locatedclose to the flagellar pocket or between the nucleus and thekinetoplast was observed (Fig. 4).

Glutamate dehydrogenaseUsing an antiserum that recognizes a NADP-linked gluta-mate dehydrogenase labelling was observed throughoutthe cytoplasm of the parasite (Fig. 6). Although all formsof the T. cruzi life cycle were labelled, the labelling wasmore intense in epimastigotes.

Effect of antibodies against cysteine proteinase on the T.cruzi-macrophage interaction. Incubation of epimastigoteand trypomastigote forms in the presence of a sub-aggluti-nating concentration of the antiserum (1:200) led to anincrease in the association index of about 4.4 and 9.5times, respectively. In both cases the high associationindices were due to an increase in the percentage ofmacrophages that ingested parasites and in the meannumber of parasites ingested by the macrophage (Table 1).

The ability of the F(ab')2 portion of the anti-cysteineproteinase antibodies to inhibit the ingestion of T. cruzi bymacrophages was determined. As shown in Table 2 thepresence of the antibody inhibited by about 60% theingestion of epimastigote and trypomastigote forms by themacrophage. The F(ab')2 portion of an unrelated antibodyused as a control, did not affect parasite interiorization(not shown).

486 T. Souto-Padrdn et al.

* < ! .

Figs 1-3. Thin sections of amastigote and trypomastigote forms of Trypanosoma cruzi incubated in the presence of anti-cysteineproteinase and subsequently in the presence of gold-labelled antibodies. The surface of amastigotes (a) is heavily labelled. Very fewparticles are seen on the surface of trypomastigotes (t). In Fig. 2 labelling of vesicles located close to the flagellar pocket (fp) as wellas within the pocket (arrow) of trypomastigotes is evident. Fig. 1, X24 000; Fig. 2, X34000; Fig. 3, X30000.

Cysteine proteinase in T. cruzi 487

Figs 4-5. Localization of cysteine proteinase in epimastigotes of T. cruzi. Intense labelling of the membrane enclosing the cell body(Fig. 4) and the flagellum (Fig. 5) as well as of the components of the endosome-lysosome system (1) is evident. Fig. 4, x 27 000;Fig. 5, x 30 000.Fig. 6. Localization of NADP-linked glutamate dehydrogenase in epimastigotes of T. cruzi. Labelling is observed throughout thecytoplasm, k, kinetoplast; n, nucleus, x 25 000.

Table 1. Effect of the whole anti-cysteine proteinase antibody on the T. cruzi-macrophage interaction

Cell treated

Epimastigote

Trypomastigote

• Refers to data obtainedt Antiserum was used at

Serum

Control*Antiserumt

Control*Antiserumf

% Macrophageswith associated

parasites

33.8±0.790.2±3.4

5.5+1.338.5±7.5

in experiments in which no serum was added,a sub-agglutinating concentration (1:200).

Mean numberof associated

parasites/macrophages

1.71+0.022.8+0.18

1.03+0.041.4+0.16

Associationindex

57.8±0.37255.4±25.1

5.71±1.354.5±13.7

Variation

4.4 x

9.5X

Table 2. Effect of the F(ab')2 portion of the anti-cysteine proteinase antibody on T. cruzi—macrophage interaction

Cell treated

% Macrophagewith associated

parasites*

Mean numberof associated

parasites/macrophagesAssociationindex (AI)

% Decreasein the AI

EpimastigoteControlt

TrypomastigoteControllOugml"1!

39.73±14.2516.85+12.98

11.72+5.404.69+2.32

1.77±0.31.58±0.49

1.06±0.061.05±0.04

73.79±38.7231.25±28.46

12.78+6.185.01±2.60

62.82±26.90

61.66±2.51

* Refers to adhered and ingested parasites.t Refers to data obtained in experiments in which no serum was added. Similar results were obtained using the F(ab'>2 portion of an immunoglobulin

that recognizes another parasite protein.% Refers to concentration of antiserum.

488 T. Souto-Padrdn et al.

Discussion

Our present immunocytochemical observations using amonospecific polyclonal antibody generated against a lyso-somal cysteine proteinase purified from epimastigotes ofT. cruzi show that both the lysosomes and the surface ofthis form, as well as those of trypomastigote-amastigotetransition forms, were intensely labelled. Labelling ofcytoplasmic vacuoles mainly located close to the flagellarpocket region, and which correspond to lysosomes (seereview by De Souza, 1984), is in agreement with ourprevious study of sub-cellular fractions, which indicatesthat the major proteinase activity is located in lysosomes(Bontempi et al. 1989). It is interesting to note that thecysteine proteinase was also located in association withthe membrane that encloses the cell body and the flagel-lum of epimastigotes, as well as the surface oftrypomastigote-amastigote transition forms and amasti-gote-like forms found in the supernatant of vertebrate cellcultures infected with T. cruzi. In contrast, very few or nogold particles, indicative of the presence of cysteine pro-teinase, were seen in association with the surface of mostof the typical trypomastigotes, identified by the organiz-ation of the kinetoplast—DNA network. However, thelysosomes and the flagellar pocket of this form wereintensely labelled. It is known that the flagellar pocket isthe only site where endocytic and exocytic processes takeplace in trypanosomatids. The presence of gold particleswithin the flagellar pocket suggests that cysteine protein-ase is secreted through this region. Despite the fact thatthe antiserum reacted in Western blots of the wholeparasite extracts only with the enzyme (Campetella et al.1990), we can not discount the possibility that the immu-nological reactivity at the parasite surface could be due tocross-reaction with other parasite antigen(s), since azoca-seinase activity could not be shown at the surface of intactepimastigote forms (J. J. Cazzulo, unpublished results). Itmust be taken into account that the previous subcellularlocalization studies (Bontempi et al. 1989) were performedto determine enzyme activity, whereas the present studiesare based on immunological reactivity. Control exper-iments with the anti-glutamate dehydrogenase serashowed that this enzyme is cytosolic in all parasite formstested, in good agreement with previous results (Cannataand Cazzulo, 1984).

Recent studies have shown that rounded forms found inthe supernatant of vertebrate cell cultures infected with T.cruzi are infective in vivo and in vitro (Carvalho and DeSouza, 1983; Piras et al. 1985; Ley et al. 1988). These formshave a kinetoplast DNA network similar to that seen inamastigote forms. However, in our opinion these data donot provide enough evidence for us to consider them asidentical to intracellular multiplicative amastigote forms(see review by De Souza, 1984). Our previous immunocyto-chemical observations on the localization of neuramini-dase show that, while a very light labelling of the surfaceof intracellular amastigotes was seen, intense labelling ofextracellular amastigote-like forms was observed (Souto-Padr6n et al. 1990). It is possible that the intracellularrounded forms represent amastigotes from heavilyinfected cells that were liberated, after rupture of the hostcell, when they were in the initial stage of the process ofintracellular transformation of amastigote into trypomas-tigote, a process that takes place at the end of theintracellular life cycle of T. cruzi (see review by De Souza,1984). Further comparative immunocytochemical studiesanalysing the distribution of neuraminidase, cysteine

proteinase and Ssp-4 (Andrews et al. 1987) in intracellularand extracellular amastigotes are necessary to clarify thispoint.

Some evidence has been obtained in recent years toindicate that parasite proteinases may play an importantrole in the process of parasitic protozoan-host cell interac-tion. In the case of Leishmania it is now clear that a majorsurface glycoprotein, which presents proteinase activity,known as gp63, is involved in the parasite-macrophagerecognition process (Russell and Wilhelm, 1986). In thecase of Trypanosoma cruzi the following data suggest aninvolvement of parasite proteinases in the modulation ofthe process of its interaction with host cells: (1) treatmentof trypomastigote forms with trypsin renders them muchmore infective to vertebrate cells (Nogueira et al. 1980;Araujo-Jorge and De Souza, 1984; Piras et al. 1985);(2) addition of proteinase inhibitors to the interactionmedium significantly decreases the penetration of infec-tive trypomastigotes, but not of non-infective epimasti-gotes, into host cells (Araujo-Jorge et al. 1986).

In view of the apparent localization of cysteine protein-ase on the surface of epimastigote and infective amasti-gote-like forms and its possible secretion through theflagellar pocket region, we decided to determine whetherthis protein plays some role in the process of T. cruzi-hostcell interaction. The experiments were carried out withresident mouse peritoneal macrophages in view of the factthat this system has been well-characterized previously(see review by Araujo-Jorge, 1990) and that a largenumber of these cells associate with parasites during ashort interaction time, in contrast to other cell types. Ourobservations show that previous incubation of epimasti-gote and trypomastigote forms in the presence of the wholeanti-cysteine proteinase antibody markedly increased theingestion of the parasites by the macrophages, probablydue to typical Fc receptor-mediated phagocytosis. Thisresult confirms the immunocytochemical observationsshowing the surface localization of the protein. In contrast,incubation of the parasites in the presence of the F(ab')2portion of the anti-cysteine proteinase IgG significantlyblocked the ingestion of the parasites by macrophages. Inview of the fact that a large number of trypomastigotes didnot express the protein at the cell surface, as evaluated byimmunocytochemistry, we would expect that the F(ab')portion of the antibody exerted a higher inhibitory effectin the ingestion of epimastigotes compared with that oftrypomastigotes. However, the inhibitory effect observedwas basically the same. In view of the fact that almost alltrypomastigotes showed the presence of cysteine protein-ase in association with the flagellar pocket, it is possiblethat the protein is released upon interaction of the para-site with the host cell.

In conclusion, our observations support the idea thatproteinases of T. cruzi play some role in the process of T.cruzi—host cell interaction. Further studies are necessaryto determine whether they act on the surface of theparasite, the host cell or both.

The authors thank Mrs Marlene Cazuza and Mr AntonioOliveira for technical assistance, and Mrs Alba Valeria Peres forsecretarial assistance. This work has been supported by UNDP/World Bank/WHO Special Programme for Research and Trainingin Tropical Diseases, Financiadora de Estudos e Projetos (FINEP),Conselho Nacional de Desenvolvimento Cientifico e Tecnol<5gico(CNPq), Fundacao de Amparo a Pesquisa do Rio de Janeiro(FAPERJ) and Secretaria de Ciencia Y Tecnica de la RepublicaArgentina.

Cysteine proteinase in T. cruzi 489

ReferencesANDREWS, N. W., HONG, L. S., ROBBINS, E. S. AND NUSSENZWEIG, V.

(1987). Stage specific surface antigens expressed during themorphogenesis of vertebrate forms of Trypanosoma cruzi. Expl Parasit.64, 474-480.

ARAUJO-JORGE, T. C. (1990). The biology of Trypanosoma cruzi-macrophage interaction. Mem. Inst. Oswaldo Cruz, (in press).

ARAUJO-JORGE, T. C. AND DE SOUZA, W. (1984). Effect of carbohydrates,periodate and enzymes in the process of endocytosis of Trypanosomacruzi by macrophages. Ada trop. 41, 17-28

ARAUJO-JORGE, T. C, SAMPAIO, E. P. AND DE SOUZA, W. (1986).Trypanosoma cruzi: Inhibition of host cell uptake of infectivebloodstream forms by alpha-2-macroblobulin. Zentbl. Bakt. ParasitKdeAbt. II72, 323-329.

BENDAYAN, M., NANCI, A. AND YAN, F. W. K. (1987). Effect of tissueprocessing on colloidal gold cytochemistry. J. Histochem. Cytochem. 35,983-996.

BONTEMPI, E., MANTINEZ, J. AND CAZZULO, J. J. (1989). Subcellularlocalization of a cysteine proteinase from Trypanosoma cruzi. Molec.biochem. Parasit. 33, 43-48.

CAMARGO, E. P. (1964). Growth and differentiation of Trypanosoma cruziin liquid media. I. Origin of metacyclic trypanosomes in liquid media.Rev. Inst. Med. Trop. Sao Paulo 6, 93-100.

CAMPETELLA, 0., MARTINEZ, J. AND CAZZULO, J. J. (1990). A majorcysteine proteinase is developmentally regulated in Trypanosomacruzi. FEMS Microbwl. Lett, (in press).

CANNATA, J. J. B. AND CAZZULO, J. J. (1984). Glycosomal andmitochondrial malate dehydrogenases in epimastigotes ofTrypanosoma cruzi. Molec. biochem. Parasit. 11, 37-49.

CARVALHO, T. U. AND DE SOUZA, W. (1983). Separation of amastigotesand trypomastigotes of Trypanosoma cruzi from cultured cells. Zentbl.Bakt. ParasitKde Abt. 169, 571-575.

CARVALHO, T. U. AND DE SOUZA, W. (1986). Infectivity of amastigotes ofTrypanosoma cruzi. Rev. Inst. Med. Trop. Sao Paulo 28, 205-212.

CAZZULO, J. J., CAZZULO FRANKE, M. C, MARTINEZ, J. AND FRANKE DECAZZULO, B. M. (19906). Some kinectic properties of a cysteineproteinase (cruzipain) from Trypanosoma cruzi. Biochim. biophys. Ada(in press).

CAZZULO, J. J., Couso, R., RAIMONDI, A., WERNSTEDT, C. AND HELLMAN,U. (1989a). Further characterization and partial amino acid sequenceof a cysteine proteinase from Trypanosoma cruzi. Molec. biochem.Parasit. 33, 33-42.

CAZZULO, J. J., HELLMAN, U., COUSO, R. AND PARODI, A. J. A. (1990a).Amino acid and carbohydrate composition of a lysosomal cysteine

proteinase from Trypanosoma cruzi. Absence of phosphorylatedmannose residues. Molec. biochem. Parasit. (in press).

CAZZULO, J. J., NOWICKI, C, SANTOME, A., WERNSTEDT, C. AND HELLMAN,

U. (19896). Amino acid composition and amino terminal sequence ofthe NADP-linked glutamate dehydrogenase from Trypanosoma cruzi.FEMS Microbiol. Lett. 56, 215-220.

DE SOUZA, W. (1984). Cell biology of Trypanosoma cruzi. Int. Rev. Cyotol.86,197-285.

LEY, V., ANDREWS, N. W., ROBBINS, E. S. AND NUSSENZWEIG, V. (1988).Amastigotes of Trypanosoma cruzi sustain an infective cycle inmammalian cells. J. exp. Med. 168, 645-655.

NOGUEIRA, N., CHAPLAN, S. AND COHN, Z. (1980). Trypanosoma cruzi:factors modifying ingestion and fate of blood form trypomastigotes. J.exp. Med 152, 447-451.

PIRAS, M. M., HENRIQUEZ, D. AND PIRAS, R. (1985). The effect ofproteolytic enzymes and protease inhibitors on the interaction betweenTrypanosoma cruzi and fibroblasts. Molec. biochem. Parasit. 14,151-163.

PIRAS, M. M., PIRAS, R. AND HENRIQUEZ, D. (1982). Changes inmorphology and infectivity of cell culture-derived trypomastigotes ofTrypanosoma cruzi. Molec. biochem. Parasit. 6, 67-81.

RUSSELL, D. G. AND WILHELM, H. (1986). The involvement of GP63, themajor surface glycoprotein, in the attachment of Leishmaniapromastigotes to macrophages. J. Immun. 136, 2613-2630.

SOUTO-PADR6N, T , HARTH, G. AND DE SOUZA, W. (1990).Immunocytochemical localization of neuraminidase in Trypanosomacruzi. Infect. Immun. 50, 586-592.

SOUTO-PADR6N, T., REYES, M. B., LEGUIZAMON, S., CAMPETELLA, O. E.,FRASH, A. C. C. AND DE SOUZA, W. (1989). Trypanosoma cruzi proteinswhich are antigenic during human infections are located in definedregions of the parasite. Eur. J. Cell Biol. 50, 272-278.

STANWORTH, D. R. AND TURNER, M. W. (1978). Immunochemical analysisof immunoglobulins and their subunits. In Handbook of ExperimentalImmunology, Immunochemistry (ed. D. M. Weir), vol. 1, pp. 61-102.Blackwell Scientific Publications, Oxford.

ZINGALES, B. AND COLLI, W. (1985). Trypanosoma cruzi interaction withhost cells. In Curr. Top. Microbiol. Immunol, vol. 117, pp. 129-152.Springer Verlag, Berlin, Heidelberg.

(Received 17 January 1990 - Accepted, in revised form, 4 April 1990)

490 T. Souto-Padrdn et al.