Characterization of Carbohydrate Binding Proteins in Trypanosoma cruzi
1995; Elsevier BV; Volume: 270; Issue: 19 Linguagem: Inglês
10.1074/jbc.270.19.11062
ISSN1083-351X
Autores Tópico(s)Research on Leishmaniasis Studies
ResumoTrypanosoma cruzi is an obligatory intracellular protozoan parasite that causes Chagas’ disease in humans and invades a great variety of mammalian cells. The nature of the ligand(s) and receptor components in both T. cruzi and target cells remains controversial, although it seems to involve an interaction with oligosaccharides. In an attempt to identify possible ligands on the parasite, we have searched for the presence of carbohydrate binding proteins (CBPs) in T. cruzi. By fluorescence-activated cell sorter analysis using a panel of fluoresceinated glyco- and neoglycopeptides with well characterized glycans, the presence of at least two different CBPs was identified on the surface of T. cruzi epimastigotes and trypomastigotes. The specificity of binding of the two CBPs seems to be mediated by galactose and mannose residues. The mannose- and galactose-mediated CBPs from epimastigotes and trypomastigotes were purified to homogeneity by affinity chromatography on immobilized thyroglobulin and identified as 60–70-kDa glycoproteins. Purified CBPs were able to specifically bind with high affinity to murine and human macrophages as well as other cell types susceptible to infection by T. cruzi but not to fat or neuronal cells. This binding was inhibited by the corresponding ligands. Moreover, the mannose-mediated CBP binding was completely abolished by α-mannosidase treatment of the cells. These results suggest a possible role for the CBPs in the recognition events between the parasite and target cells and/or in the interaction of the epimastigotes with the insect gut cells. Trypanosoma cruzi is an obligatory intracellular protozoan parasite that causes Chagas’ disease in humans and invades a great variety of mammalian cells. The nature of the ligand(s) and receptor components in both T. cruzi and target cells remains controversial, although it seems to involve an interaction with oligosaccharides. In an attempt to identify possible ligands on the parasite, we have searched for the presence of carbohydrate binding proteins (CBPs) in T. cruzi. By fluorescence-activated cell sorter analysis using a panel of fluoresceinated glyco- and neoglycopeptides with well characterized glycans, the presence of at least two different CBPs was identified on the surface of T. cruzi epimastigotes and trypomastigotes. The specificity of binding of the two CBPs seems to be mediated by galactose and mannose residues. The mannose- and galactose-mediated CBPs from epimastigotes and trypomastigotes were purified to homogeneity by affinity chromatography on immobilized thyroglobulin and identified as 60–70-kDa glycoproteins. Purified CBPs were able to specifically bind with high affinity to murine and human macrophages as well as other cell types susceptible to infection by T. cruzi but not to fat or neuronal cells. This binding was inhibited by the corresponding ligands. Moreover, the mannose-mediated CBP binding was completely abolished by α-mannosidase treatment of the cells. These results suggest a possible role for the CBPs in the recognition events between the parasite and target cells and/or in the interaction of the epimastigotes with the insect gut cells. The intracellular protozoan parasite, Trypanosoma cruzi, is the causative agent of South American trypanosomiasis or Chagas’ disease, a chronic and debilitating multisystemic disorder that affects about 25 million people in Latin America (1World Health Organization Weekly Epidemiol. Res. 1990; 65: 257-264PubMed Google Scholar). From a clinical point of view, the disease is characterized by an acute phase with high parasitemia and strong immunosuppression (2Beltz L.A. Kierszenbaum F. Immunology. 1987; 60: 309-315PubMed Google Scholar, 3Brener Z. Adv. Parasitol. 1980; 18: 247-292Crossref PubMed Scopus (141) Google Scholar) followed by a chronic phase with an autoimmune pathology (4Hudson L. Ann. Soc. Belge Med. Trop. 1985; 65: 71-77PubMed Google Scholar, 5Kierszenbaum F. J. Parasitol. 1986; 72: 201-211Crossref PubMed Scopus (62) Google Scholar). The parasite life cycle can be divided into four stages (6Brener Z. Annu. Rev. Microbiol. 1973; 27: 347-382Crossref PubMed Scopus (492) Google Scholar, 7de Souza W. Int. Rev. Cytol. 1984; 86: 197-283Crossref PubMed Scopus (242) Google Scholar). The parasite is taken in the blood meal of the insect as a trypomastigote, which differentiates into the epimastigotes that multiply extracellularly in the midgut of reduviid insects. In the hindgut, epimastigotes transform into infective nondividing metacyclic trypomastigotes, which are released in the feces. Metacyclic trypomastigotes are able to invade a wide variety of host mammalian cells, phagocytic and non-phagocytic (7de Souza W. Int. Rev. Cytol. 1984; 86: 197-283Crossref PubMed Scopus (242) Google Scholar, 8Pereira M.E.A. Wyler D.J. Modern Parasitology: Cellular, Immunological and Molecular Aspects. W. H. Freeman and Compan, New York1990: 64-78Google Scholar). Once inside the cells, the metacyclic forms escape from endocytic vacuoles to the cytoplasm where they transform into amastigotes, which multiply intracellularly. Upon rupture of host cells, they differentiate into trypomastigotes that circulate in the blood until they encounter appropriate target cells, and then they go through another intracellular cycle or are taken up by the insect again. This complex cycle of infection requires the involvement of mutual active recognition phenomena occurring between the parasite and the host cell, the adhesion or binding steps being an essential requisite for parasite penetration. During the last decade, evidence has been accumulating that implicates specific receptor-ligand interactions in the early events of parasite invasion of cells (8Pereira M.E.A. Wyler D.J. Modern Parasitology: Cellular, Immunological and Molecular Aspects. W. H. Freeman and Compan, New York1990: 64-78Google Scholar, 9Piras R. Piras M.M. Henriquez D. CIBA Found. Symp. 1983; 99: 31-51PubMed Google Scholar, 10Zingales B. Colli W. Curr. Top. Microbiol. Immunol. 1985; 117: 129-152PubMed Google Scholar, 11Ouaissi M.A. Cornette J. Capron A. Mol. Biochem. Parasitol. 1986; 19: 201-211Crossref PubMed Scopus (60) Google Scholar, 12Prioli R.P. Rosenberg I. Pereira M.E. Proc. Natl. Acad. Sci. U. S. A. 1987; 84: 3097-3101Crossref PubMed Scopus (13) Google Scholar, 13Prioli R.P. Rosenberg I. Shivakumar S. Pereira M.E. Mol. Biochem. Parasitol. 1988; 28: 257-263Crossref PubMed Scopus (21) Google Scholar, 14Schenkman S. Andrews N.W. Nussenzweig V. Robbins E.S. Cell. 1988; 55: 157-165Abstract Full Text PDF PubMed Scopus (56) Google Scholar, 15De Araujo-Jorge T.C. Mem. Inst. Oswaldo Cruz Rio J. 1989; 84: 441-462Crossref PubMed Google Scholar, 16Yoshida N. Mortara R.A. Araguth M.F. Gonzalez J.C. Russo M. Infect. Immun. 1989; 57: 1663-1667Crossref PubMed Google Scholar, 17Ortega-Barria E. Pereira M.E. Cell. 1991; 67: 411-421Abstract Full Text PDF PubMed Scopus (176) Google Scholar, 18Ramirez M.I. Ruiz R. Araya J.E. Da Silvera J.F. Yoshida N. Infect. Immun. 1993; 61: 3636-3641Crossref PubMed Google Scholar). This evidence was indirect and based largely on the modulation of parasite binding and/or penetration by a variety of different perturbations of either the host cell or the parasite cell surface. It is noteworthy that the literature always assumes that the host cell plays an active role (i.e. the receptor) in the recognition event. However, it is likely that parasites have developed mechanisms to facilitate infection by producing specific molecules (receptors) that recognize suitable complementary ligands on the host cell plasma membrane and not exclusively in the opposite way. Recent studies from our laboratory have clearly involved the β1 integrins as “receptors” for T. cruzi binding and internalization (19Fernandez M. Mu-oz-Fernandez M.A. Fresno M. Eur. J. Immunol. 1993; 23: 552-557Crossref PubMed Scopus (34) Google Scholar). In spite of several prospective candidates as “ligands” for target cell receptors reported during recent years, the nature of the receptor(s) or ligand(s) in both host cell and parasite remains controversial, although there is relevant evidence pointing to the involvement of interactions with glycans (20Villaita F. Kierszenbaum F. Biochem. Biophys. Res. Commun. 1984; 119: 228-235Crossref PubMed Scopus (35) Google Scholar, 21Villaita F. Kierszenbaum F. Biochim. Biophys. Acta. 1983; 736: 39-44Crossref PubMed Scopus (43) Google Scholar, 22Villaita F. Kierszenbaum F. Biochim. Biophys. Acta. 1985; 845: 216-222Crossref PubMed Scopus (19) Google Scholar, 23Villalta F. Kierszenbaum F. Mol. Biochem. Parasitol. 1985; 16: 1-10Crossref PubMed Scopus (20) Google Scholar, 24Villalta F. Kierszenbaum F. Mol. Biochem. Parasitol. 1987; 22: 109-114Crossref PubMed Scopus (11) Google Scholar, 25Schenkman S. Jiang M.S. Hart G.W. Nussenzweig V. Cell. 1991; 65: 1117-1125Abstract Full Text PDF PubMed Scopus (381) Google Scholar). The identification of T. cruzi surface molecules mediating the interaction between the parasite and the host cell is an obligatory step in understanding this process. In this report we identify a family of novel carbohydrate binding proteins (CBP) 1The abbreviations used are: CBPcarbohydrate binding proteinCHAPS3-[(3-cholamidopropyl)dimethylammonio]-1-propanesulfonic acidDMEMDulbecco’s modified Eagle’s mediumFITCfluorescein isothiocyanateFPLCfast protein liquid chromatographyGal-CBPgalactose-mediated CBPMan-CBPmannose-mediated CBPPBSphosphate-buffered salineBSAbovine serum albuminPAGEpolyacrylamide gel electrophoresisTricineN-[2-hydroxy-1,1-bis(hydroxymethyl)ethyl]glycineBHKbaby hamster kidneyNRKnormal rat kidney. present in metacyclic trypomastigote and epimastigote stages of T. cruzi. The purified proteins retain biological activity and exhibit properties that make them prospective candidates as receptors mediating early recognition events. carbohydrate binding protein 3-[(3-cholamidopropyl)dimethylammonio]-1-propanesulfonic acid Dulbecco’s modified Eagle’s medium fluorescein isothiocyanate fast protein liquid chromatography galactose-mediated CBP mannose-mediated CBP phosphate-buffered saline bovine serum albumin polyacrylamide gel electrophoresis N-[2-hydroxy-1,1-bis(hydroxymethyl)ethyl]glycine baby hamster kidney normal rat kidney. Parasites—The strain of T. cruzi used was originally obtained from a patient with Chagas’ disease at the Instituto Nacional de la Salud, Madrid, Spain. It was cloned and was named strain G (26Alcina A. Fresno M. Mol. Biochem. Parasitol. 1988; 29: 181-190Crossref PubMed Scopus (12) Google Scholar). Epimastigotes were continuously cultured in liver infusion-tryptose medium supplemented with 10% fetal calf serum as described previously (26Alcina A. Fresno M. Mol. Biochem. Parasitol. 1988; 29: 181-190Crossref PubMed Scopus (12) Google Scholar). Metacyclic trypomastigotes were obtained by metacyclogenesis induced by incubating late log epimastigotes in triatomine artificial urine medium plus 0.035% sodium carbonate for 96 h (27Goldenberg S. Contreras V.T. Bonaldo M.C. Salles J.M. Lima-Franco M.P.A. Lafaille J. Gonzales-Perdomo M. Linss J. Morel C.M. Mol. Biochem. Parasitol. 1987; 16: 315-327Google Scholar). Transformation of the parasite was assessed by resistance to complement lysis using horse serum. After 96 h in metacyclogenesis medium, the parasites were washed with PBS and resuspended to 1 × 108/ml in PBS. Then 500-μl samples of parasites were mixed with equal volumes of 70% (v/v) fresh serum and incubated for 1 h at 37 °C. The mixture was washed in PBS, and the surviving parasites (metacyclics) were counted in a hemocytometer. For metabolic labeling, the parasites were washed twice in PBS and preincubated at 108 cells in 2 ml of methionine-free DMEM containing 5% dialyzed fetal calf serum for 1 h. Then 1 mCi of [35S]methionine labeling mix (Amersham) was added to the culture and incubated at 28 °C for 16–18 h. At the end of the labeling period the parasites were washed twice in PBS and resuspended at the desired density. Cell Culture and Metabolic Labeling—LLC-MK2, Vero, J774, and BHKR 14 cells were maintained in DMEM supplemented with 10% heat-inactivated fetal calf serum, 100 units/ml penicillin, and 100 μg/ml streptomycin. Unless otherwise noted, cells were maintained in plastic tissue culture flasks (150 cm 2P. Bonay and M. Fresno, manuscript in preparation.) and incubated with 5% CO2 and 95% air at 37 °C. For metabolic labeling, a semi-confluent plate (80 cm2) of LLC-MK2 cells was washed twice with PBS and preincubated with methionine-free DMEM containing 5% dialyzed fetal calf serum for 2–3 h at 37 °C. At the end of the preincubation period, 350 μCi/ml [35S]methionine labeling mix were added to the plate and incubated further at 37 °C for 16–18 h. Glycopeptide Preparation and Fluorescent Tagging—The glycopeptides containing the glycan structures shown in Fig. 1 used in this study were purified by proteolytic cleavage of the intact protein and further purification of the glycopeptide by lectin affinity chromatography and gel filtration. Glycopeptides A, B, and C were obtained from bovine fetuin (28Dorland L. Van-Halbeek H. Vliegenthart J. Lis H. Sharon N. J. Biol. Chem. 1981; 256: 7708-7711Abstract Full Text PDF PubMed Google Scholar) by Pronase digestion, sialidase digestion, and hydrolysis with Streptococcus pneumoniae β-galacosidase (100 milliunits in a volume of 100 μl of 20 mM citrate/phosphate buffer, pH 6.0) at 37 °C overnight, respectively. Structure D was obtained from α1-acid glycoprotein (29Conchie J. Strachan I. Carbohydr. Res. 1978; 63: 193-213Crossref Scopus (35) Google Scholar), and structure E was obtained from the latter by digestion with Arthrobacter ureafaciens sialidase (50 milliunits in 100 μl of 150 mm sodium acetate buffer, pH 5.5) at 37 °C overnight. Structure F was isolated from soybean agglutinin (30Yoshima H. Matsumoto A. Mizuochi T. Kawasaki T. Kobata A. J. Biol. Chem. 1981; 256: 8476-8484Abstract Full Text PDF PubMed Google Scholar). Glycopeptide G was obtained by exhaustive digestion of ovalbumin with Pronase and purification by concanavalin A-Sepharose (31Takasaki S. Kobata A. Biochemistry. 1986; 25: 5706-5715Crossref Scopus (123) Google Scholar). Glycopeptide H was obtained from fibrinogen (32Townsend R.R. Hilliker E. Li Y.T. Laine R.A. Bell W.R. Lee Y.C. J. Biol. Chem. 1982; 257: 9704-9710Abstract Full Text PDF PubMed Google Scholar). Glycopeptide I was derived from H by sialidase treatment or from human transferrin (33Bayard G.S.B. Fournet B. Strecker G. Bouquelet S. Montreuil J. FEBS Lett. 1975; 50: 296-299Crossref PubMed Scopus (302) Google Scholar). Glycopeptide J was obtained from porcine thyroglobulin as described before (34Yamamoto K. Tsuji T. Irimura T. Osawa T. Biochem. J. 1981; 195: 701-713Crossref PubMed Scopus (111) Google Scholar). Glycopeptide K was obtained from hen egg ovalbumin (35Tai T. Yamashita K. Ito S. Kobata A. J. Biol. Chem. 1977; 252: 6687-6694Abstract Full Text PDF PubMed Google Scholar), and glycopeptide L was purified from sheep IgG (36Rademacher T.W. Homans S.W. Parekh R.B. Dwek R.A. Biochem. Soc. Symp. 1986; 51: 131-148PubMed Google Scholar). The purified glycopeptides were labeled with fluorescein isothiocyanate (FITC) as described (37Harlow E.D. Lane D. Antibodies: A Laboratory Manual. Cold Spring Harbor Laboratory, Cold Spring Harbor, NY1988: 354-355Google Scholar). Flow Cytometry Analysis—T. cruzi epimastigotes or metacyclic trypomastigotes were washed twice in DMEM with 2% glucose and 1% bovine serum albumin. For each assay, 1–2 × 106 parasites were resuspended in 100 μl of the same medium containing the fluoresceinated glycopeptide for 1 h at 16 °C in the dark. The parasites were washed three times with DMEM with 2% BSA before being resuspended in the same medium containing 1% paraformaldehyde, and the fluorescence was analyzed in an EPICS cytofluorimeter. Preparation of Neoglycoproteins—The mannosyl-BSA neoglycoproteins were prepared according to Monsigny et al. (38Monsigny M. Roche A.C. Midoux P. Biol. Cell. 1983; 51: 187-196Crossref Scopus (199) Google Scholar) by using p-aminophenylthiocyanate mannopyranoside at 10:1 and 25:1 molar ratios (BSA-Man10 and BSA-Man25). Purification of CBPs—The carbohydrate binding proteins of T. cruzi were purified as follows. [35S]Methionine-labeled trypomastigotes of T. cruzi obtained by metacyclogenesis (1011 cells/5 ml) were lysed by mixing in a vortex with 5 ml of 2 × Buffer A (1% CHAPS in 25 mm HEPES, pH 7.2, 10% glycerol, and a mixture of protease inhibitors composed of 8 mm phenylmethylsulfonyl fluoride, 15 mm leupeptin, 10 μg/ml aprotinin, 15 mm pepstatin, 50 μg/ml Nα-p-tosyl-L-lysine chloromethyl ketone, and 50 μg/ml tosylphenylalanyl chloromethyl ketone) for 30 min on ice. The mixture was clarified by centrifugation at 150,000 × g for 45 min. The pellet was extracted again in the same way with 5 ml of the detergent solution and centrifuged under the same conditions. The supernatants were pooled and used immediately or kept frozen at –70 °C for no longer than a week. The pool of supernatants (~9 ml) was mixed by rocking with 7 ml of the affinity matrix (asialofetuin and asialothyroglobulin Affi-Gel) at 4 °C for 16 h and poured into a column (1 × 5 cm). The column was washed with buffer A containing 1% CHAPS until the A280 and the radioactivity in the effluent fell to background levels. Then the column was eluted with 30 ml of 1 m NaCl to remove nonspecifically bound proteins. The bound proteins were eluted by washing the column sequentially with 30 ml of buffer A containing 0.5 m of the following: α-methylmannoside, lactose, and GlcNAc. Finally the column was regenerated by eluting stepwise with 0.5 m EDTA and 1 m guanidine HCl. Fractions of 2.5 ml were collected, and their A280 was monitored. The radioactivity in a 25-μl aliquot was determined by scintillation counting. Fractions around the peaks of A280 and radioactivity were pooled, dialyzed, and reapplied to a small sized column (0.5 × 2 cm) and eluted in a similar way as before. In some instances the eluted material was relieved of minor contaminants by FPLC on a Mono Q (Pharmacia Biotech Inc.) column (0.5 × 5 cm), using 25 mm Tris/HCl, pH 7.9, and a NaCl gradient. The purified CBPs elute at 280–350 mm NaCl under such conditions. The fractions eluted from the affinity matrix and obtained from the FPLC column were analyzed by SDS-PAGE by the method of Laemmli (39Laemmli U.K. Nature. 1970; 227: 680-685Crossref PubMed Scopus (207537) Google Scholar). Sequencing of CBP—FPLC-purified trypomastigote Man-CBP (25 μg) was cleaved with 3-bromo-3-methyl-2-(2-nitrophenylmercapto)-3-4-indole-Skatole as described (40Crimmins D.L. McCourt D.W. Thoma R.S. Scott M.G. Macke K. Schwartz B.D. Anal. Biochem. 1990; 187: 27-38Crossref PubMed Scopus (59) Google Scholar), and the peptides resulting were separated by SDS-PAGE in 16% T, 2.67% C gels (1 mm) using the Tris-Tricine buffer (41Schaegger H. von Jagow G. Anal. Biochem. 1987; 166: 368-379Crossref PubMed Scopus (10505) Google Scholar) but omitting the 10% T intermediate gel. The peptides separated were electroblotted onto Immobilon P, detected by Coomassie (42Matsudaira P. J. Biol. Chem. 1987; 262: 10035-10038Abstract Full Text PDF PubMed Google Scholar) or fluorescamine (43Vandekerckhove J. Bauw G. Puype M. Van Damme J. Van Montagu M. Eur. J. Biochem. 1985; 152: 9-19Crossref PubMed Scopus (215) Google Scholar) staining, excised, and placed into the sequencer cartridge. Automated sequencing was done on a model 473A sequencer (Applied Biosystems, Foster City, CA). Iodination of CBP—The labeling with 125I was carried out in plastic 0.5-ml tubes catalyzed by IODO-GEN (l,3,4,6-tetrachloro-3α,6α-diphenylglycoluril, Pierce) as described by the manufacturer. Binding of CBPs to Cells in Culture—LLC-MK2, Vero, BHK, and NRK cells were grown in 96-well plates as described above. The monolayers were washed twice and incubated with different concentrations of 125I-labeled CBPs at 16 or 37 °C during 1 h. After the incubation the cells were washed twice with PBS and solubilized with 50 μl of 1 n NaOH, and the bound radioactivity was counted in a γ counter. For the inhibition experiments, the 125I-labeled CBPs were incubated for 30 min at 4 °C with the inhibitors at 100 μg/ml prior to adding them to the cells. Cell Attachment Assay—Solutions of purified Man-CBP, Gal-CBP, or fibronectin in PBS were used to coat 96-well microtiter plates by incubating for 16 h at 4 °C. The wells were washed twice with PBS, blocked with periodate-treated 5% BSA (5% BSA solution in PBS treated with 0.5 mm sodium periodate for 30 min in the dark and then dialyzed against PBS) for 16 h at room temperature, and used during the next 24 h. 35S-Labeled LLC-MK2 cells that were dispersed from confluent monolayers by EDTA treatment and washed twice with DMEM with 2% glucose and 1% bovine serum albumin were added to each well and incubated for 1 h at 37 °C. The wells were gently washed twice with 250 μl of warm DMEM. The cells remaining in the plate were solubilized by addition of 200 μl of 1 n NaOH and counted by liquid scintillation. Plasma Membrane Preparation—Plasma membrane fractions from cultured cells were prepared according to Reinhardt et al. (44Reinhardt R. Bridges R.J. Rummel W. Lindemann B. J. Membr. Biol. 1987; 95: 47-59Crossref PubMed Scopus (45) Google Scholar). Aliquots of the purified membranes were separated by SDS-PAGE and transferred to Immobilon P as described above. Inhibition of T. cruzi Attachment to Vero Cells by the CBPs—This was assessed by preincubating Vero cells (5 × 104 cells/well) in 24-well plates with increasing amounts of purified CBPs in DMEM at 4 °C for 4 h. Controls were carried out by incubating the cells at the same concentration of BSA in DMEM. At the end of the preincubation period, 10 μl of a suspension of metacyclic trypomastigotes (1 × 107 parasites/ml) were added to the cell monolayer and incubated at 4 °C for 2 h. The cells were then rinsed twice with DMEM, fixed, and stained with Giemsa (21Villaita F. Kierszenbaum F. Biochim. Biophys. Acta. 1983; 736: 39-44Crossref PubMed Scopus (43) Google Scholar). The number of trypomastigotes attached per 100 cells and the percentage of Vero cells containing attached trypomastigotes were microscopically determined by screening 200 cells/monolayer. Identification of Carbohydrate Binding Proteins on the Surface of T. cruzi—In an attempt to look for carbohydrate binding proteins expressed on the surface of T. cruzi, we devised a fluorescence-activated cell sorter assay and studied the binding of a panel of fluoresceinated derivatives of glycopeptides with well characterized glycans (Fig. 1) to live G strain trypomastigotes and epimastigotes under conditions that minimized the internalization of the glycopeptides. Fig. 2 shows the flow cytometry analysis of FITC glycans binding to live T. cruzi trypomastigotes. The assay is specific and selective, as shown by the negligible binding of an FITC-aglycone derivative (Fig. 2a, BSA). When a FITC neoglycoprotein (BSA-Man10) was used, there was an evident increase in the fluorescence bound to the cells (Fig. 2a, BSA-man10) that increased concomitantly with the density of the mannosyl residues on the protein as compared with using BSA-Man25 (Fig. 2a, BSA-man25). Having demonstrated the specific binding of a neoglycoprotein to live T. cruzi, we decided to test the binding of some natural glycans. A fully sialylated triantennary glycan (structure A) bound to more than 90% of the parasites (Fig. 2b, A). Removing the terminal sialyl residues and exposing the galactosyl residues (structure B) results in an increase in the fluorescence intensity (Fig. 2b, B) suggesting either a stronger binding or a greater number of receptors. By contrast, removing the terminal galactosyl residues (structure C) reduces the binding to a level even lower than the fully sialylated counterpart (Fig. 2b, C). Similar results were obtained when tetra-antennary glycans (structures D and E; Fig. 2c, D and E) or biantennary glycans (not shown) were tested. However, it was noticed that complex asialylated bi- (structure I; Fig. 2e, I) or triantennary (structure B; Fig. 2b, B) glycans bound to the cells more strongly than tetra-antennary glycans (structure E, Fig. 2, E) like those found in human α1-acid glycoprotein (compare Fig. 2, b, c, and e). From the above results, it can be concluded that in general asialoglycopeptides bound more strongly to the cells than the fully sialylated counterparts. The presence of core-substituted fucose does not seem to significantly affect the binding, as seen by comparing the results obtained with Glycans I and J (Fig. 2e). High mannose glycopeptides (structures F and G) also were able to bind to the cells as shown in Fig. 2d. Interestingly, a better binding was observed with the lower mannose oligomer glycan (Man6, structure G) than with the higher mannose oligomer (Man9, structure F). By contrast, the structures K and L did not bind to the parasites, the fluorescence intensity being similar to that of the nonglycosylated control (Fig. 2f). The common structural feature of these two glycans is a bisecting N-acetylglucosamine residue bound with β1–4 to the core β1–4-linked mannose. It is interesting to note that this determinant abolishes the binding to T. cruzi as shown by comparing structures I and L, which are otherwise identical (Fig. 2, e and f). The binding was specific as indicated by the displacement of the fluorescence using the non-labeled glycopeptide as an inhibitor (not shown). Furthermore, glutaraldehyde or trypsin treatment of the intact parasite completely abolishes the binding of any glycan (results not shown). Moreover, the fact that non-labeled high mannose glycopeptides were not able to compete the binding of complex asialobi- or asialotriantennary glycopeptides (not shown) suggests the presence on the membranes of T. cruzi of at least two families of carbohydrate binding proteins with different specificities. Isolation and Purification of the CBPs—To purify the CBPs we followed a general strategy widely used for the purification of lectins. We used affinity chromatography on a mixed bed of asialofetuin and asialothyroglobulin glycopeptides immobilized on Affi-Gel 10 to afford the purification of the majority of different CBPs present followed by sequential elution with various agents. A CHAPS extract of 35S-labeled metacyclic trypomastigotes was applied to the column. After washing the column with high salt (1 m NaCl) to remove nonspecifically bound proteins, the column was eluted sequentially with 0.5 m of the following solutions: α-methylmannoside, GlcNAc, and lactose (Fig. 3). The elution with high salt did not release any radioactivity from the column. However, when the eluting agent was α-methylmannoside or lactose, a sharp peak of radioactivity came off from the column in each case. Eluting the column with 0.5 m GlcNAc released a smaller and broader peak. Washing the column with 0.5 m EDTA or 1 m guanidine HCl did not release any additional peaks (not shown). Quite similar results were obtained when an epimastigote extract was passed through the column and eluted in the same way (Fig. 3). However, contrary to what was obtained with trypomastigotes, eluting the epimastigote extract-loaded column with 0.5 m GlcNAc did not result in any radioactivity being released from the column. The purified trypomastigote and epimastigote CBPs retained their ability to bind to the affinity matrix when reapplied to a new column and could be eluted only by the original eluting agent (i.e. the Man-CBP, eluted by washing the column with α-methylmannoside, could not be eluted from the affinity matrix by lactose (not shown)). Neither could lactose inhibit the rebinding of the Man-CBP to the affinity matrix. Analysis of the eluates by reducing SDS-PAGE and autoradiography revealed a single band of molecular mass around 68–70 kDa strikingly similar in each case as shown in Fig. 4A for the trypomastigote (panel 1) and epimastigote (panel 2) extracts. Silver staining of the gels also revealed a single band of the same mobility whether the gels were run in the presence or absence of 1% β-mercaptoethanol (data not shown). Fig. 4B shows a silver-stained gel of the FPLC-purified Man-CBP from trypomastigotes (lane T) and epimastigotes (lane E). Similar purity was obtained for all the CBPs isolated either from the trypomastigote or epimastigote stage. It is important to mention that, after washing the column with the last saccharide, no additional radioactivity came off of the column when the sequence of elution was repeated, thus indicating that each initial step released the specific CBP completely. It is interesting that the proteins eluted with quite distinct agents apparently showed an identical or very similar molecular size. The Man-CBP obtained from trypomastigotes was cleaved with 3-bromo-3-methyl-2-(2-nitrophenylmercapto)-3-4-indole-Skatole, and the peptides were separated by SDS-PAGE and transferred to a polyvinylidene difluoride membrane. Some peptides were cut from the membrane and subjected to sequence analysis. The internal amino acid sequence obtained from one of those peptides was DLAKSYAFLGL, as shown in Fig. 5, compared with the sequences of other carbohydrate binding proteins obtained from the GenBank™. Interaction of CBPs with Cells—The purified trypomastigotes Man-CBP and Gal-CBP were able to bind to live LLC-MK2 (Fig. 6A, black circles (Man-CBP) and white circles (Gal-CBP)) and glutaraldehyde-fixed Vero, LLC-MK2, J774, NRK, and BHK cells (data not shown) at 16 °C (to prevent internalization). As shown in Fig. 6A, the binding was dose-dependent and saturable for both CBPs. Interestingly, when the cells were pretreated with α-mannosidase to remove the terminal mannose
Referência(s)