1993, 61(3):898. Infect. Immun. R P Schenkman, F Vandekerckhove and S Schenkman by Trypanosoma cruzi. Mammalian cell sialic acid enhances invasion http://iai.asm.org/content/61/3/898 Updated information and services can be found at: These include: CONTENT ALERTS more» cite this article), Receive: RSS Feeds, eTOCs, free email alerts (when new articles http://journals.asm.org/site/misc/reprints.xhtml Information about commercial reprint orders: http://journals.asm.org/site/subscriptions/ To subscribe to to another ASM Journal go to: on May 22, 2014 by guest http://iai.asm.org/ Downloaded from on May 22, 2014 by guest http://iai.asm.org/ Downloaded from
Transcript
1993, 61(3):898. Infect. Immun.
R P Schenkman, F Vandekerckhove and S Schenkman by Trypanosoma cruzi.Mammalian cell sialic acid enhances invasion
http://iai.asm.org/content/61/3/898Updated information and services can be found at:
These include:
CONTENT ALERTS more»cite this article),
Receive: RSS Feeds, eTOCs, free email alerts (when new articles
http://journals.asm.org/site/misc/reprints.xhtmlInformation about commercial reprint orders: http://journals.asm.org/site/subscriptions/To subscribe to to another ASM Journal go to:
ROCILDA P. F. SCHENKMAN,l FILIP VANDEKERCKHOVE,2 AND SERGIO SCHENKMAN1*
Department ofMicrobiology, Immunology and Parasitology, Escola Paulista de Medicina, Rua Botucatu862/8 andar, 04023 Sdo Paulo, Brazil, 1 and Michael Heidelberger Division of Immunology, Department
ofPathology, New York University Medical Center, New York, New York 100162
Received 24 August 1992/Accepted 14 December 1992
We have used a Chinese hamster ovary cell mutant (Lec2) that express much less sialic acid on the surfacethan the parental cell line (ProS) to investigate whether sialic acid plays a role during cell invasion byTrypanosoma cruzi. Trypomastigotes derived from a tissue culture (corresponding to bloodstream trypomastig-otes) and metacyclic trypomastigotes (corresponding to infective stages of the insect vector) invaded the Lec2mutant less efficiently than the parental cell line. Invasion of the Lec2 mutant cells could be restored to the Pro5level by resialylation of the mutant cells with T. cruzi trans-sialidase and sialyllactose. Conversely, pretreatmentof the Pro5 parental cells with bacterial neuraminidase decreased invasion. These results indicate that sialicacid associated with the host cell contributes to invasion by T. cruzi.
Trypanosoma cruzi, an obligatory intracellular protozoanparasite that causes Chagas' disease in humans, invadesmany cell types and infects hosts belonging to severalmammalian species (4, 6, 29). The stages of T. cruzi thatactively enter mammalian cells are the metacyclic trypo-mastigotes, found in the hind gut of Reduviidae insects, andthe bloodstream trypomastigotes (20).
Several observations suggest that sialic acid plays a role incell invasion by bloodstream and metacyclic forms. Asreviewed by Pereira (9), he and coworkers identified a
neuraminidase on the surface of T. cruzi trypomastigotes andshowed that the inhibition of this enzyme by monoclonalantibodies (MAbs) (13) or by serum lipoproteins (14) in-creased the entry of trypomastigotes. They further foundthat the addition of exogenous bacterial neuraminidase ab-rogated this enhancement. On the basis of these findings,they suggested that in order to invade, T. cruzi recognizes a
sialic acid-containing receptor and that the removal of thiscarbohydrate by the parasite neuraminidase negatively con-trols the level of infection.We have found that the neuraminidase described by
Pereira is in fact a trans-sialidase (TS) (16, 19, 24). Thetransfer of sialic acid was detected originally in noninfectiveepimastigotes by Previato et al. (12) and later in trypomasti-gotes by Zingales et al. (28). The enzyme transfers a(2,3)-linked sialic acid from host glycoproteins and glycolipids toparasite acceptors consisting of terminal ,-galactosyl units.In the absence of acceptors, TS transfers sialic acid from a
donor to water in a typical sialidase reaction (16).Trypomastigotes growing in medium without sialoglyco-
proteins contain low levels of sialic acid but are still able toinvade cells. During invasion, the sialic acid is transferredfrom the host cell to the parasite, resulting in the sialylationof a trypomastigote surface antigen, recognized as stage-specific epitope 3 (Ssp-3) (1, 19). On the basis of the findingthat MAbs to Ssp-3 prevent cell invasion (15, 17), we havereasoned that the transfer of sialic acid from the host cell tothe parasite could be required for invasion. To test thishypothesis, we have studied T. cruzi invasion of a mutant
* Corresponding author.
(Lec2) derived from Chinese hamster ovary cells and resis-tant to wheat germ agglutinin (22). This mutant has littlesialic acid on its surface because of a defect in the transfer ofCMP-sialic acid into the Golgi complex (5).
MATERIALS AND METHODS
Trypanosomes. T. cruzi trypomastigotes, strain Y (21),were derived from supernatants of cultures of LLCMK2cells (ATCC CCL-7; American Type Culture Collection,Rockville, Md.) grown in low-glucose Dulbecco's modifiedEagle's medium with penicillin and streptomycin (DMEM;GIBCO, Grand Island, N.Y.) and 10% fetal bovine serum
(FBS) at 37°C in 5% CO2. Subconfluent cultures ofLLCMK2cells were infected with 5 x 106 trypomastigotes. Freeparasites were removed after 24 h, and the cultures weremaintained in 10% FBS-DMEM. 10% FBS-DMEM wasremoved on the third day following infection, the monolay-ers were washed twice with phosphate-buffered saline(PBS), and DMEM containing 0.2% bovine serum albumin(BSA) (ultrapure; Boehringer Mannheim, Indianapolis, Ind.)and 20 mM N-2-hydroxyethylpiperazine-N'-2-ethanesul-fonic acid (HEPES) (Sigma) (pH 7.4) (0.2% BSA-DMEM)was added. The parasites in the cell supernatants were useddirectly in some experiments, whereas in others, the slendertrypomastigotes were purified from contaminating amasti-gotes or intermediate forms. For this purification, parasitesuspensions were centrifuged at 2,000 x g for 5 min andincubated at 37°C. The motile trypomastigotes swam upfrom the pellet into the medium and were collected with thesupernatant after 2 h. The contamination of this fraction withamastigotes or intermediate forms was less than 1%. Meta-cyclic trypomastigotes, strain CL (2), were obtained fromaged cultures of epimastigotes in liver infusion tryptosemedium containing 10% FBS at 28°C. When the culturesstarted to differentiate, the parasites were centrifuged andresuspended in Grace's medium (GIBCO). After 5 to 7 days,50 to 60% of the parasites were metacyclic trypomastigotesand were separated from epimastigotes by passage through aDE-52 column (Whatman, Maidstone, United Kingdom).Enzymes and other reagents. TS was affinity purified from
the trypomastigote supematants as described previously
(16). In brief, the trypomastigotes released from infectedLLCMK2 cells were centrifuged for 10 min at 3,200 x g, andthe supernatants were filtered through 0.45-,um-pore-sizefilters to remove remaining parasites or any other cellulardebris. The filtered supernatants were concentrated by pre-cipitation with 50% ammonium sulfate, and the precipitateswere resuspended in PBS, dialyzed against PBS, and passedthrough a Tresyl-Sepharose (Schleicher & Schuell) columncontaining immobilized MAb 39 (prepared in accordancewith manufacturer instructions). The column was washedwith PBS and then with 10 mM sodium phosphate (pH 6.5),and TS was eluted with 3.5 M MgCl2-10 mM sodiumphosphate (pH 6.0). The fractions eluted from the columnwere immediately filtered through Sephadex G-25 equili-brated with 20 mM Tris-HCl (pH 8.0) to remove the MgCl2.The enzyme was stored in a sterile manner at 4°C. MAb 39reacts specifically with TS that is more than 95% pure, asjudged by sodium dodecyl sulfate-polyacrylamide gel elec-trophoresis silver staining, as shown previously (20). Vibriocholerae neuraminidase (protease free), biotin-labeledMaackia amurensis lectin (MAA), and sialyllactose frombovine colostrum [65% a(2,3)-linked sialic acid] were ob-tained from Boehringer. Clostndium perfringens neuramin-idase (type X, with less than 0.002 U of protease activity permg of casein) was obtained from Sigma. Anti-T. cruzi rabbitantiserum was prepared as described previously (17). Hep-arin sulfate was kindly provided by Helena Nader, EscolaPaulista de Medicina, Sao Paulo, Brazil.
Cell invasion experiments. Lec2 cells (ATCC CRL 1736)and the parental Pro5 cells (ATCC CRL 1781) were plated in24-well plates with or without 12-mm-diameter glass cover-slips and containing alpha-MEM (GIBCO) plus 10% FBS.When plated directly on plastic, 5 x 104 cells were plated perwell and used after 48 h. When glass coverslips were used,1.5 x 104 cells were plated per well and used after 72 h toachieve identical spreading for both cell lines. Prior toinfection, the cells and the parasites were washed twice with0.2% BSA-alpha-MEM and treated as indicated in eachexperiment. The parasites were incubated with the cells for15 to 120 min at various concentrations in a final volume of0.25 to 0.5 ml. The infections were quantitated either bycounting microscopically the number of parasites stained byimmunofluorescence and associated with the cells or bymeasuring the TS activity of intracellular parasites after thefifth day of infection. In the first method, after incubation ofthe parasites with the cells, the coverslips were washed threetimes with medium 199 and one time with PBS and fixed with4% paraformaldehyde in PBS. The fixed cells were thenpermeabilized with 0.1% Triton X-100 in PBS, and theparasites were visualized after being stained by indirectimmunofluorescence with a polyclonal anti-T. cruzi anti-body. The total number of parasites and the number ofinfected cells were counted for at least 500 cells in triplicateexperiments. When intracellular TS activity was used as ameasure of infection, cells were washed five times with 0.2%BSA-DMEM and incubated in 10% FBS-alpha-MEM for 5days at 37°C. Cell monolayers were washed twice with PBS,removed from the culture wells by trypsin-EDTA treatment,suspended in DMEM-BSA, and transferred to Eppendorftubes. After centrifugation, the cells were washed with PBSand lysed in 100 ,ul of 50 mM Tris-HCl (pH 7.4) containing1% Nonidet P-40, 0.1 M EDTA, 1 mM phenylmethylsulfonylfluoride, and 0.5 ,ug each of antipain, leupeptin, and pepsta-tin (Sigma) per ml. TS activity was determined by incubationof the lysates in 20 mM HEPES buffer (pH 7.2) in thepresence of sialyllactose and D-glucose-[1- 4C]lactose (60
c)
c)0
a)-4-)0
0'I
60
40
20
0100
50
v)
60 a,C)w00
40>\0a)
20 V)S0
200
U)
C)
2 co
o0
CD
* _4
C)
Q)
0CHO-Pro5 CHO-Lec2
FIG. 1. T. cnrzi invades ProS cells better than Lec2 cells. Tissueculture trypomastigotes grown in 0.2% BSA-DMEM (A) and meta-cyclic trypomastigotes (B) at 2 x 10' parasites per ml of 0.2%BSA-alpha-MEM were incubated at 37°C for 30 min with ProS andLec2 cells. The percentage of infected cells (open bars) and thenumber of parasites per 100 cells (hatched bars), quantitated by theimmunofluorescence method, are shown. The difference in infectionof Pro5 and Lec2 cells was statistically significant (P < 0.05) for bothtypes of measurements (percentage of cells infected with tissueculture trypomastigotes and with metacyclic trypomastigotes: P =0.0063 and P = 0.010, respectively; number of parasites per 100 cellsinfected with tissue culture trypomastigotes and with metacyclictrypomastigotes: P = 0.0041 and P = 0.0002, respectively).
mCi/mmol) (Amersham, Arlington Heights, Ill.) as describedpreviously (16). This method provides results very similar tothose provided by the first method. All experiments wereperformed in triplicate, and the results are expressed as themean + the standard error of mean. P values were calculatedby use of an unpaired two-tailed t test.
RESULTS AND DISCUSSION
T. cruzi trypomastigotes derived from LLCMK2 cells andmetacyclic trypomastigotes growing in liver infusion tryp-tose medium were incubated for 30 min with ProS andmutant Lec2 cells. As shown in Fig. 1, tissue culturetrypomastigotes and metacyclic forms invaded Lec2 cellsless efficiently, as reflected by the total number of parasitesassociated with the cells and by the percentage of infectedcells.To verify whether the absence of sialic acid was respon-
sible for the lower level of infection of Lec2 cells, wesialylated the cells by incubating them with affinity-purifiedT. cruzi TS in the presence of an appropriate sialic acid
FIG. 2. Sialylation of Lec2 cells by TS. Lec2 cells were treatedat 37°C for 40 min with 5 ,g of affinity-purified TS per ml and 1 mMsialyllactose in 0.2% BSA-alpha-MEM (A and B) or with mediumalone (C and D). The cells were washed and incubated withbiotin-labeled MAA in 1% BSA-TBS containing 1 mM CaC12 and 1mM MnCl2 after fixation in 4% paraformaldehyde-PBS. Boundlectin was detected after reaction with 2 jig of FITC-streptavidin per
ml by use of a fluorescence microscope (B and D). The fluorescencepictures were photographed under identical conditions of exposure,
development, and printing. Magnification, x363.
donor, sialyllactose. The transfer of sialic acid to the Lec2cells was corroborated by the incubation of TS-treated cellswith biotin-labeled MAA and then with fluorescein isothio-cyanate (FITC)-streptavidin. We found that MAA, whichrecognizes a(2,3)-sialyl-galactosyl units (26), bound to Lec2cells treated with TS and sialyllactose but not to untreatedcells, as seen by use of FITC-streptavidin (Fig. 2). As shownin Fig. 3A, the level of invasion of Lec2 cells that had beenpretreated with TS and sialyllactose was similar to the levelof invasion of ProS cells. This effect was not due to thepresence of TS, since pretreatment with TS but withoutsialyllactose had no effect on the invasion of Lec2 cells (Fig.3B). Similar results were obtained when the TS activity ofparasites 5 days after invasion was measured. As shown inTable 1, TS activity on the fifth day after infection was
significantly higher in Lec2 cells that had been resialylatedby pretreatment with TS and sialyllactose.The level of invasion of the Lec2 cell line was consistently
lower than that of the Pro5 cell line, and the deficiency forthe Lec2 cells was overcome following sialylation. Thesimplest explanation for these results is that sialylation of theLec2 cells was responsible for the increase in invasion.Nevertheless, we cannot exclude the possibility that TS inthe presence of sialyllactose had an effect other than thesialylation of Lec2 cells. For example, TS could act as an
adhesive protein and increase the attachment of trypo-mastigotes to Lec2 cells; TS is a complex molecule thatcontains, in addition to the catalytic site, type III modules ofa fibronectin domain that could interact with the target cellsurface (10). To obtain further evidence for the participationof sialic acid in invasion, we pretreated the Lec2 and ProScell lines with neuraminidase before invasion under twodifferent sets of conditions. In the first treatment, the cellswere incubated with 0.1 U of V. cholerae neuraminidase perml for 40 min at pH 7.0, while in the second treatment, thecells were incubated with the same concentration of neur-
aminidase but at pH 6.5 for 50 min. As shown in Fig. 4A,treatment at pH 7.0 diminished infection of ProS cells only.Incubation with neuraminidase at a lower pH also decreased
60
v)
a)
C.)a)
0
40
20
0
a)C)
0
0
._)
0
Cl
0
Medium TS TS + SLIncubation ofCHO-Lec 2
FIG. 3. Resialylation of Lec2 cells restores invasion by T. cruzi.(A) Cells plated on 12-mm-diameter glass coverslips and grown in10% FBS-alpha-MEM were washed three times with medium with-out serum and treated for 30 min with 10 ,ug of TS per ml and 1 mMsialyllactose in 0.2% BSA-alpha-MEM (hatched bars) or with me-dium alone (open bars). The cells were washed with medium withoutserum and incubated for 30 min with 2 x 107 trypomastigotes per mlof 0.2% BSA-alpha-MEM. The number of parasites associated withthe cells was determined by the immunofluorescence method. (B)As in panel A, except that Lec2 cells were pretreated with mediumalone, TS alone, or TS plus sialyllactose (SL). Open bars corre-spond to the percentage of infected cells, and hatched bars corre-spond to the number of parasites per 100 cells. The difference ininfection of Lec2 cells treated with TS and sialyllactose and un-
treated cells was statistically significant (P = 0.0034).
the invasion of Lec2 cells (Fig. 4B), suggesting that even thesmall amounts of sialic acid present on the surface of Lec2cells (10% of the amounts present on the surface of parentalcells [23]) could mediate invasion by T. cruzi.The fact that the invasion of Lec2 cells is still significant
even after neuraminidase treatment suggests that other mol-ecules mediate attachment and/or invasion of T. cruzi inde-pendently or in combination with sialic acid. For example,matrix components have been implicated in attachment ofand invasion by T. cruzi, and it is known that trypomasti-gotes bind to collagen (8) and fibronectin (25). Trypomasti-gotes also express a heparin-binding molecule, named pen-etrin, that is able to prevent the invasion of T. cruzi and,when expressed in Escherichia coli, confers invasive capa-bilities on the bacteria (7). Therefore, in addition to sialic
a Trypomastigotes released from LLCMK2 cells were preincubated with 1mM sialyllactose (SL) (+) or medium alone (-) and incubated with Chinesehamster ovary cells also pretreated with 1 mM SL and 10 jig of TS per ml ormedium alone for 2 h. Parasites and cells were washed and coincubated for 15min at 37'C. Unbound parasites were aspirated, the cells were washed fivetimes with 0.2% BSA-alpha-MEM, and the infectivity was measured on thebasis of the intracellular production of TS 5 days postinfection. Results aregiven as the mean ± the standard error of the mean for triplicate experiments.Similar results were obtained in two independent experiments. The value forLec2 cells alone was 175 ± 45 cpm. The difference between the invasion ofLec2 cells and that of Lec2 cells pretreated with TS and SL was significant inthe case of nonsialylated trypomastigotes (P = 0.004) or sialylated trypo-mastigotes (P = 0.003). The difference between sialylated and nonsialylatedtrypomastigotes was not significant (P > 0.05).
acid, heparan sulfate could mediate adhesion and invasion.Under our assay conditions, however, preincubation oftissue culture trypomastigotes, metacyclic trypomastigotes,or the target cells with up to 50 ,g of heparan sulfate per mlhad no effect on the invasion of either Lec2 or ProS cells(data not shown).
In the experiments described above, we used trypomasti-gotes released by LLCMK2 cells growing in medium con-taining BSA. Under these conditions, the parasites areweakly labeled by anti-Ssp-3 antibody and contain smallamounts of sialic acid (19). The amounts of sialic acid greatlyincrease when the parasites are released into medium con-taining serum or are pretreated with sialyllactose. In Table 1,we compare invasion by parasites with low levels of sialicacid with invasion by those fully sialylated. Trypomastigotespreincubated with sialyllactose still invaded Pro5 cells betterthan Lec2 cells. Similar results were obtained in manyexperiments when invasion was estimated after 30 min ofcontact between cells and parasites or when we used para-sites released by LLCMK2 cells grown in medium containingFBS (data not shown). In all of these situations, only a smallincrease in invasion was obtained for sialylated parasites.These results indicate that complete sialylation and theexpression of large amounts of Ssp-3 are not primary re-quirements for invasion. The same conclusion was obtainedin other experiments in which sialic acid was transferredfrom Ssp-3 to a sialic acid acceptor in the incubationmedium. Preincubation of trypomastigotes in 10 mM lactoseor the addition of lactose during invasion did not interferewith entry for either the Lec2 or the ProS cell line. Similarresults were obtained with other cell lines, such as BALB/C3T3 fibroblasts and LLCMK2 cells (data not shown).These conclusions are in apparent contradiction with prior
findings. Cell invasion is inhibited by Fab fragments ofMAbs against Ssp-3 (17, 19), which is the major sialic acidacceptor of trypomastigotes. Also, MAbs to the glycoconju-gates of 35- and 50-kDa antigens, which are the major sialicacid acceptors of metacyclic trypomastigotes (15a, 27),prevent cell invasion by metacyclic forms (15). That Ssp-3-containing molecules are involved in invasion is also sup-ported from experiments in which trypomastigotes were
UV
a)c)
0CI)
0U)
S0*e-0v)
Ev
150
100
50
urIl r-
A
I
50 -
25T
0 CO e
CHO-Pro5 CHO-Lec2
FIG. 4. Neuraminidase treatment of target cells diminishes inva-sion by T. cruzi. Pro5 and Lec2 cells were treated with 0.1 U ofazide-free V. cholerae neuraminidase per ml in 0.2% BSA-alpha-MEM (pH 7.0) for 40 min (A) or in the same medium but at pH 6.5for 50 min (B). In the control experiments, the cells were treatedwith the same medium and buffer lacking neuraminidase. The cellswere then washed and incubated with 2.5 x 107 trypomastigotes perml in 0.2% BSA-alpha-MEM for 30 min. Open bars correspond tothe infection of untreated cells, and hatched bars correspond to theinfection of cells treated with neuraminidase. The difference in theinvasion of Pro5 cells treated or not treated with neuraminidase wassignificant (P = 0.0071 in panel A and P = 0.011 in panel B). Thedifference in the invasion of Lec2 cells treated or not treated withneuraminidase was not significant in panel A (P = 0.47) but wassignificant in panel B (P = 0.01).
opsonized with anti-Ssp-3 antibodies and then incubatedwith nonphagocytic cells that had been engineered to ex-press Fc receptors. In sharp contrast to opsonization withantibodies against other parasite surface antigens, opsoniza-tion of trypomastigotes with anti-Ssp-3 antibodies did notenhance but rather inhibited invasion (18). Others have alsoprovided data that suggest that sialylated molecules in theparasites have a role in invasion. For example, Piras et al.(11) have shown that incubation of trypomastigotes withsialic acid-containing glycoproteins increases their invasive-ness, whereas Couto et al. (3) have found sialic acid in an85-kDa glycoprotein of trypomastigotes that is involved ininvasion.One possible explanation for these contradictory results is
that sialic acid-dependent mechanisms for parasite invasionrequire minimal amounts of sialic acid, which could beprovided by the LLCMK2 cells in which the trypomastigoteswere grown. Alternatively, the Ssp-3 precursor moleculesfrom the parasite surface may accept sialic acid from and
donate sialic acid to the host during invasion. This sialic acidexchange could facilitate the movement of the parasites intothe cells and prevent an irreversible attachment. This hy-pothesis would explain why cells with less sialic acid are
invaded to a lesser degree. In the presence of exogenous
acceptors, such as lactose, the transfer would occur from thetarget cell both to the exogenous saccharide and to the Ssp-3precursor, thus explaining the lack of inhibition of penetra-tion upon the addition of lactose to the incubation medium.It follows that the effect of antibodies to Ssp-3 could be toprevent the exchange of sialic acid. However, the directdemonstration that sialic acid transfer is necessary for inva-sion by T. cruzi requires the development of powerful TSinhibitors or the construction of T. cruzi strains lacking TS.
ACKNOWLEDGMENTS
We thank Nobuko Yoshida for providing metacyclic forms. Weare also grateful to Nobuko Yoshida and Victor Nussenzweig forsuggestions and for reading the manuscript.
This work was supported by grants from the Fundac,o de Amparoa Pesquisa do Estado de Saio Paulo, Sao Paulo, Brazil, the ConselhoNacional de Desenvolvimento Cientifico e Tecnologico, Brazil, theMacArthur Foundation, and the UNDP/World Bank/WHO SpecialProgramme for Research and Training in Tropical Diseases. F.V. isa research assistant of the Belgian National Fund for ScientificResearch.
REFERENCES1. Andrews, N. W., K.-S. Hong, E. S. Robbins, and V. Nussen-
zweig. 1987. Stage-specific surface antigens expressed duringthe morphogenesis of vertebrate forms of Trypanosoma cruzi.Exp. Parasitol. 64:474-484.
2. Brener, Z., and E. Chiari. 1963. Variacoes morfologicas obser-vadas em diferentes amostras de Trypanosoma cruzi. Rev. Inst.Med. Trop. Sao Paulo 5:220-224.
3. Couto, A. S., M. F. Gonqalves, W. Colli, and R. M. de Le-derkremer. 1990. The N-linked carbohydrate chain of the 85-kilodalton glycoprotein from Trypanosoma cruzi trypomasti-gotes contains sialyl, fucosyl and galactosyl (alpha 1-3)galactoseunits. Mol. Biochem. Parasitol. 39:101-107.
4. de Souza, W. 1984. Cell biology of Trypanosoma cruzi. Int. Rev.Cytol. 86:197-283.
5. Deutscher, S. L., P. Nuwayhid, P. Stanley, E. I. B. Briles, andC. B. Hirschberg. 1984. Translocation across Golgi-vesiclemembranes: a CHO glycosylation mutant deficient in CMP-sialic acid transport. Cell 39:295-301.
6. Kirchhoff, L. V. 1990. Trypanosoma species (Americantrypanosomiasis, Chagas disease): biology of trypanosomes, p.2077-2084. In G. L. Mandell, R. G. Douglas, Jr., and J. E.Bennett (ed.), Principles and practice of infectious diseases, 3rded. Churchill Livingstone, New York.
7. Ortega-Barria, E., and M. E. A. Pereira. 1991. A novel T. cruziheparin-binding protein promotes fibroblast adhesion and pen-etration of engineered bacteria and trypanosomes into mamma-lian cells. Cell 67:411-421.
8. Ouaissi, M. A., J. Cornette, D. Afchain, A. Capron, H. Gras-Masse, and A. Tartar. 1986. Trypanosoma cruzi infection inhib-ited by peptides modeled from a fibronectin cell attachmentdomain. Science 234:603-607.
9. Pereira, M. E. A. 1990. Cell biology of Trypanosoma cruzi, p.64-78. In D. Wyler (ed.), Modern parasite biology. W. H.Freeman & Co., New York.
10. Pereira, M. E. A., J. S. Mejia, E. Ortega-Barria, D. Matzilevich,and R. P. Prioli. 1991. The Trypanosoma cruzi neuraminidasecontains sequences similar to bacterial neuraminidases, YWTDrepeats of the low density lipoprotein receptor, and type IIImodules of fibronectin. J. Exp. Med. 174:179-191.
11. Piras, M. M., D. Henriquez, and R. Piras. 1987. The effect offetuin and other sialoglycoproteins on the in vitro penetration of
Trypanosoma cruzi trypomastigotes into fibroblastic cells. Mol.Biochem. Parasitol. 22:135-143.
12. Previato, J. O., A. F. Andrade, M. C. Pessolani, and L. Men-donsa-Previato. 1985. Incorporation of sialic acid into Trypano-soma cruzi macromolecules: a proposal for a new metabolicroute. Mol. Biochem. Parasitol. 16:85-96.
13. Prioli, R. P., J. S. Mejia, and M. E. Pereira. 1990. Monoclonalantibodies against Trypanosoma cruzi neuraminidase revealenzyme polymorphism, recognize a subset of trypomastigotes,and enhance infection in vitro. J. Immunol. 144:4384-4391.
14. Prioli, R. P., I. Rosenberg, and M. E. Pereira. 1990. High- andlow-density lipoproteins enhance infection of Trypanosomacruzi in vitro. Mol. Biochem. Parasitol. 38:191-198.
15. Ruiz, R. C., V. L. Rigoni, J. Gonzalez, and N. Yoshida. The35/50 kDa surface antigen of Trypanosoma cruzi metacyclictrypomastigotes, an adhesion molecule involved in host cellinvasion. Parasite Immunol., in press.
15a.Schenkman, S., et al. Submitted for publication.16. Schenkman, S., L. P. de Carvalho, and V. Nussenzweig. 1992.
Trypanosoma cruzi trans-sialidase and neuraminidase activitiescan be mediated by the same enzymes. J. Exp. Med. 175:567-575.
17. Schenkman, S., C. Diaz, and V. Nussenzweig. 1991. Attachmentof Trypanosoma cruzi trypomastigotes to receptors at restrictedcell surface domains. Exp. Parasitol. 72:76-86.
18. Schenkman, S., T. Kurosaki, J. V. Ravetch, and V. Nussenzweig.1992. Evidence for the participation of the Ssp-3 antigen in theinvasion of non-phagocytic mammalian cells by Trypanosomacruzi. J. Exp. Med. 175:1635-1641.
19. Schenkman, S., J. Man-Shiow, G. W. Hart, and V. Nussenzweig.1991. A novel cell surface trans-sialidase of Trypanosoma cruzigenerates a stage-specific epitope required for invasion of mam-malian cells. Cell 65:1117-1125.
20. Schenkman, S., E. S. Robbins, and V. Nussenzweig. 1991.Attachment of Trypanosoma cruzi to mammalian cells requiresparasite energy, and invasion can be independent of the targetcell cytoskeleton. Infect. Immun. 59:645-654.
21. Silva, L. H. P., and V. Nussenzweig. 1953. Sobre uma cepa deTrypanosoma cruzi altamente virulenta para o camundongobranco. Folia Clin. Biol. 20:191-203.
22. Stanley, P., and L. Siminovitch. 1977. Complementation be-tween mutants of CHO cells resistant to a variety of plantlectins. Somatic Cell Genet. 3:391-405.
23. Stanley, P., T. Sudo, and J. P. Carver. 1980. Differentialinvolvement of cell surface sialic acid in wheat germ agglutininbinding to parental and wheat germ agglutinin-resistant Chinesehamster ovary cells. J. Cell Biol. 85:60-69.
24. Umera, H., S. Schenkman, V. Nussenzweig, and D. Eichinger.1992. Only some members of a gene family in Trypanosomacruzi encode proteins which express both trans-sialidase andneuraminidase activities. EMBO J. 11:3837-3844.
25. Velge, P., M. A. Ouaissi, J. Cornette, D. Afchain, and A.Capron. 1988. Identification and isolation of Trypanosoma cruzitrypomastigote collagen-binding proteins: possible role in cell-parasite interaction. Parasitology 97:255-268.
26. Wang, W.-C., and R. D. Cummings. 1988. The immobilizedleukoagglutinin from the seed of Maackia amurensis binds withhigh affinity to complex-type Asn-linked oligosaccharides con-taining terminal sialic acid linked (x-2,3 to penultimate galactoseresidues. J. Biol. Chem. 263:4576-4585.
27. Yoshida, N., R. A. Mortara, M. F. Araguth, J. C. Gonzalez, andM. Russo. 1989. Metacyclic neutralizing effect of monoclonalantibody 10D8 directed to the 35- and 50-kilodalton surfaceglycoconjugates of Trypanosoma cruzi. Infect. Immun. 57:1663-1667.
28. Zingales, B., C. Carniol, R. M. de Lederkremer, and W. Colli.1987. Direct sialic acid transfer from a protein donor to glycolip-ids of trypomastigote forms of Trypanosoma cruzi. Mol. Bio-chem. Parasitol. 26:135-144.
29. Zingales, B., and W. Colli. 1985. Trypanosoma cruzi: interactionwith host cells. Curr. Top. Microbiol. Immunol. 117:129-152.
