Brief communication

Chagasic patients develop a type 1 immune response to

Trypanosoma cruzi trans-sialidase

MARCELO RIBEIRAO1, VERA LUCIA PEREIRA-CHIOCCOLA1,2, LAURENT RENIA 3,ABI LIO AUGUSTO FRAGATA FILHO2, SERGIO SCHENKMAN1 & MAURICIO M.RODRIGUES1

1Departamento de Microbiologia, Imunologia e Parasitologia, Universidade Federal de Sa˜o Paulo-Escola Paulista de Medicina,Rua Botucatu 862, Sa˜o Paulo 04023–062, SP Brazil,2Laboratorio de Xenodiagno´stico, Instituto Dante, Pazzanese de Cardiologia do Estado de Sa˜o Paulo, Avenue Dante Pazzanese 500,Sao Paulo, SP Brazil and3U445 INSERM, Instituit Cochin de Gene´tique Moleculaire, Laboratoire d’Immunologie des Pathologies Infectieuses et Tumorales,UniversiteReneDescartes, Hoˆpital Cochin 27, Rue du Fbg Saint Jacques, 75014, Paris, France

SUMMARY

Infective forms ofTrypanosoma cruzi, the parasite thatcauses Chagas’ disease, express on their surface anenzyme denominatedtrans-sialidase (TS). The presentstudy was designed to evaluate the naturally acquiredimmune responses to a bacterial recombinant protein repre-senting the catalytic domain of TS in chronically infectedchagasic individuals. The cellular immune response wasmeasured by in-vitro T-cell proliferation and by interferon(INF)-g, interleukin (IL)-4 and IL-10 production in responseto a whole-parasite homogenate and the recombinantprotein. The peripheral blood mononuclear cells of 78·6%of 28 chagasic patients responded to the recombinantprotein as estimated by T-cell proliferation. With respectto cytokine production, 88% of the cells of the chagasicindividuals produced IFN-g on stimulation with the recom-binant protein. In contrast, IL-4 or IL-10 were minimallyproduced in response to TS. The cellular immune responsewas specific because most healthy individuals never exposedto T. cruzi failed to react with this recombinant protein.The plasma of 71·4% or 100% of chagasic patients had IgGantibodies as determined by ELISA or by the presence ofTS inhibitory antibodies, respectively. We conclude thatthe catalytic domain of TS is recognized by IFN-g producingtype 1 cells and antibodies in a large proportion of patientsinfected withT. cruzi.

Keywords Trypanosomes, sialidases, cell-mediatedimmune response

INTRODUCTION

Trypanosoma cruzitrypomastigotes express on their surfacean enzyme denominatedtrans-sialidase (TS). This enzymeis required for the parasite to obtain sialic acid from hostglycoproteins (Schenkmanet al. 1991b). Serological studieshave shown that TS is highly immunogenic during naturalhuman infection withT. cruzi. Almost all individualsdevelop antibodies that recognize an epitope composed ofamino acid repetitions located at the carboxy-terminalregion of TS (Cazzulo & Frasch 1992). Chagasic patientsalso generate antibodies specific for theN-terminal region ofTS that strongly inhibit its enzymatic activity (Leguizamonet al. 1994, Pereira-Chioccolaet al. 1994). We extendedthese findings by demonstrating that immunization witheither a plasmid DNA encoding the catalytic domain ofTS or a recombinant protein produced by the same genesignificantly reduce the parasitemia and the acute phasemortality during T. cruzi infection in mice (Costaet al.1998, Pereira-Chioccolaet al. 1999).

Since TS can be a target for protective immune responsesagainstT. cruzi infection in rodents, the present investi-gation was designed to complement these studies by eval-uating the type of cell-mediated immune response (CMI)developed by chagasic patients to the catalytic domain ofthe enzyme. We used as an antigen the same bacterialrecombinant protein that we have demonstrated earlier to

Parasite Immunology, 2000:22: 49–53

q 2000 Blackwell Science Ltd 49

Correspondence: Mauricio M.RodriguesReceived: 16 November 1998Accepted for publication: 21 July 1999

elicit a significant degree of immunity against experimentalT. cruzi infection. Human CMI was estimatedin vitro byantigen-specific proliferation and by interferon (IFN)-g,interleukin (IL)-4 and IL-10 production.

MATERIALS AND METHODS

After the patients gave verbal consent, 10 ml of venousblood was obtained from each individual into heparinizedtubes and used as a source of peripheral blood mononuclearcells or plasma. The first group of individuals consistedof healthy adult volunteers selected among blood donorsfrom the city of Sa˜o Paulo, Brazil. They were serologicallynegative at the time the blood samples were collected. Thetotal number of volunteers was 15, their average age was29·4þ 5·5 years and 46·6% of the subjects were males. Thesecond group consisted of adults from different areas inBrazil where Chagas’ disease is transmitted. The total num-ber of volunteers was 28, their average age was 41·1þ

7·8 years and 53% of the subjects were males. Their clinicaldiagnosis was performed by one of us, Dr Abı´lio A. FragataFilho from Instituto Dante Pazzanese de Cardiologia (Sa˜oPaulo, Brazil). The patients were primarily diagnosed onthe basis of clinical examination, electrocardiogram, chest-X-ray, and conventional serological methods performed onthree different occasions. Individuals were considered to bechagasic when they tested positive for all three serologicaltests. In 39·2% (11/28) of individuals,T. cruziwas success-fully isolated by xenodiagnosis and haemoculture. At thetime their blood sample was collected, these individuals hadnot received any treatment for Chagas’ disease.

Trypomastigote homogenate (Tc antigens) were gener-ated from trypomastigotes of the Y strain ofT. cruzi. Afterwashing in cold PBS, trypomastigotes (108 per ml) isolatedas described by Schenkmanet al. (1991a) were lysed in coldPBS containing 6M urea and centrifuged at 30 000g. Thesupernatant was dialysed against PBS and sterile filtered.Native TS was purified from culture supernatants ofT. cruzi-infected cells as described by Schenkmanet al. (1992). Therecombinant TS was produced inEscherichia colitrans-formed with plasmid pTS-cat7 (Ribeira˜o et al. 1997).

Plasmas or cells from individuals were tested for reactiv-ity with the native or recombinant TS by proliferation assayand ELISA essentially as described by Soareset al. (1997).For the proliferation assay, arithmetic mean cpm for each setof triplicate wells was calculated. The stimulation indices(SI) were determined as the ratio of the arithmetic mean cpmof cultures stimulated with the antigen and the arithmeticmean cpm of nonstimulated cultures. In parallel, as positivecontrols, peripheral blood mononuclear cells (PBMC) werecultivated in the presence of the mitogen ConcanavalinA (25mg/ml, Sigma, St Louis, MO, USA) for 72 h. The

cpm obtained after stimulation with this mitogen was41 771þ 21 095 for chagasic patients and 58 280þ 20 089for healthy individuals. IFN-g, IL-4 and IL-10 concentra-tions were estimated in duplicate samples using commer-cially available ELISA kits (Genzyme, Cambridge, MA,USA). The assay was performed following manufacturerrecommendations. The presence of antibodies that inhibitTS activity was detected essentially as described by Schenk-man et al. (1994). Kruskal–Wallis one-way analysis ofvariance and comparison of proportions were performedusing the True Epistat software package (Dr T.L.Gustafson,Richardson, TX, USA).

RESULTS AND DISCUSSION

To determine whether the recombinant TS elicited antigen-specific proliferative responses, PBMC were collected from28 individuals in the chronic phase of Chagas’ disease. Cellswere tested for [3H]TdR uptake upon culture in the presenceof different concentrations of the recombinant protein (two-fold dilutions ranging from 1·25 to 20mg/ml). The intensityof the response generated by the recombinant protein wascompared with different concentrations of Tc antigens(threefold dilutions ranging from 0·03 to 10mg/ml). Indivi-duals who showed SI$ 2·5 for at least one concentration ofthe antigens, compared to PBMC cultures without antigen,were considered as positive responders. We found that thePBMC of 78·6% of the chagasic patients presented with SIof 2·5 or higher when cultivated in the presence of therecombinant TS. In contrast, none of the healthy individualsreached SI of 2·5 upon stimulation with the same antigen.Both the frequency of responders and SIs to recombinant TSwere significantly higher among chagasic patients whencompared to healthy individuals (Table 1,P<0·05, compar-ison of proportions;P<0·001, Kruskal–Wallis one-wayanalysis of variance). When we compared the proliferationresponse of chagasic individuals to the recombinant TS orTc antigens, we observed that the intensity of the response toTc antigens was significantly higher than the response to therecombinant protein (P<0·001).

Subsequently, we determined whether human PBMCproduced IFN-g, IL-4 and IL-10 upon stimulation with therecombinant TS or Tc antigens. We considered as respon-ders only individuals whose IFN-g concentration was higherthan 16 pg/ml. Employing this parameter, we found thatrecombinant TS stimulated 88% of the chagasic patientsPBMC to secrete IFN-g. In contrast, only 13·3% of PBMCfrom healthy individuals produced IFN-g upon stimulationwith this recombinant protein (P<0·05, comparison ofproportions). In addition, the IFN-g production observedamong chagasic patients was significantly higher than inhealthy individuals (Table 1,P<0·001). Also of note was

M.Ribeirao et al. Parasite Immunology

q 2000 Blackwell Science Ltd,Parasite Immunology, 22, 49–5350

the fact that the percentage of responders to the Tc antigenswas 100%, and the magnitude of the IFN-g response to theparasite extract was significantly higher than the response tothe recombinant TS (P<0·001).

IL-4 concentrations in culture supernatants of cells fromchagasic and healthy individuals stimulated with eitherrecombinant TS or Tc antigens were below the limit ofdetection (5 pg/ml). In control cultures stimulated for 3 dayswith Concanavalin A, IL-4 could be easily detected in thesupernatants.

We also estimated the concentration of a second type2 cytokine, IL-10. We compared the level of IL-10 in super-natants of PBMC from normal and chagasic individuals whenstimulated with recombinant TS or Tc antigens. Very lowconcentrations of IL-10 were observed when PBMC fromchagasic individuals were stimulated with TS. These levels ofIL-10 were lower from those detected in the normal healthyindividual group (P¼ 0·022).

Finally, we confirmed that these individuals had IgGantibodies to TS. As determined by ELISA, we foundIgG antibodies to native TS in 96·4% of the chagasicpatients. When we used the recombinant TS as antigen,which contains only the catalytic domain of the enzyme,71·4% of the chagasic patients had specific IgG (Figure 1a).The frequency of individuals with IgM was very smallin both cases. Specific antibodies to TS or Tc antigens

were IgG1, IgG2 and IgG3. However, the levels of IgG2were significantly higher for both antigens (P<0·001, datanot shown).

By using the TS inhibition assay, we found that plasmafrom 100% of the chagasic patients significantly inhibitedthe enzymatic activity of recombinant TS. The inhibitionof TS activity ranged from 54% to 100%. In contrast,control plasmas obtained from 15 healthy individualsdid not inhibit more than 34% of TS enzymatic activity(Figure 1b,P<0·001).

In the present study, we report that the CMI exhibitedby humans to atrans-sialidase expressed on the surface ofT. cruzi trypomastigotes is mainly mediated by type 1 cellswhich secrete IFN-g, but not IL-4 or IL-10. The CMI ofdifferent chagasic patients to recombinant TS and to Tcantigens were extremely variable. A possible explanationfor the variability could be the fact that we studied chagasicpatients with different clinical forms of the disease. How-ever, in our limited group of patients, we did not find astatistically significant correlation between the forms of thedisease and the magnitude of T-cell proliferation or IFN-g

production. Such diversity in T-cell response has beenobserved in many other studies and possibly reflects avariety of factors including the genetic composition of thehosts and the parasites (Moratoet al. 1986, Cetronet al.1993, Dutraet al. 1996, Barros-Mazonet al. 1997).

Volume 22, Number 1, January 2000 Type 1 immune response toTrypanosoma cruzi trans-sialidase

q 2000 Blackwell Science Ltd,Parasite Immunology, 22, 49–53 51

Table 1 In-vitro cell mediated immune-response of Chagasic patients and healthy volunteers to recombinant TS and Tc antigens

Cell-mediated immune response Normal healthy volunteers Chagasic patients P-value1 Chi-squared valuea

Proliferative response (n)b 15 28Rec TSMedian SI (range) 1·3 (0·76–2·47) 3·76 (0·6–12·25) >0·001 19·89

Tc antigensMedian SI (range) 1 (0·56–2·99) 14·15 (0·6–23·66) >0·001 21·51

Interferon-g secretion (n) 15 25Rec TSc

Median pg/ml (range) 0 (0–136) 84 (0–620) >0·001 18·14

Tc Antigensd

Median pg/ml (range) 0 (0–78) 1680 (52–2434) >0·001 27·90

IL-10 secretion (n) 14 25Rec TSe

Median pg/ml (range) 19·4 (0–106) 12·2 (0–89) 0·022 5·22

Tc antigensd

Median pg/ml (range) 29·4 (0–71) 31·1 (0–125) 0·45 0·55

aKruskal-Wallis one-way analysis of variance.b The stimulation indices (SI) were determined as the ratio of the arithmetic mean cpm of culturesstimulated with the antigen and the arithmetic mean cpm of nonstimulated cultures. The background counts obtained in the absence of antigenranged from 636 to 5386 cpm.c Recombinant TS was used in a final concentration of 20mg/ml. d Tc antigens was used in final concentration of10mg/ml.

The fact that upon stimulation with recombinant TS,PBMC from 88% of chagasic patients secreted IFN-g mayargue in favour of the participation of these cells in protec-tive immunity. IFN-g has been shown to be a critical factor

for host survival during experimentalT. cruzi infection(Holscheret al. 1998). Our earlier studies on immunizationwith a recombinant protein and naked DNA vaccines basedon the catalytic domain of TS also indicate that immunity toTS can reduce the acute phase parasitemia and protect miceagainst a lethalT. cruziinfection (Costaet al. 1998, Pereira-Chioccolaet al. 1999). Immunity observed in the mousemodel is, at least in part, mediated by IFN-g producing Tcells as a TS-specific CD4 T cell clone activates macro-phages to eliminate the parasitein vitro (Rodrigueset al.1999). In addition to protective immune responses, wecannot completely exclude the possibility that CMI to TSparticipates in the inflammatory response observed in someof the individuals in the chronic phase of Chagas’ disease.

In summary, the presence of type 1 T cells in humanssupports the notion that the immune response to TS shouldbe further investigated during Chagas’ disease.

ACKNOWLEDGEMENTS

This work was supported by grants from FAPESP, PADCT,CNPq, FINEP, PRONEX (Brazil) and INSERM re´seauNORD-SUD (contract 4N002C-France).

REFERENCES

Barros-Mazon S., Guariento M.E. & Abrahansohn I.A. (1997) IL-12enhances proliferation of peripheral blood mononuclear cells fromChagas’ disease patients toTrypanosoma cruziantigen.ImmunologyLetters57, 39–45

Cazzulo J.J. & Frasch A.C.C. (1992) SAPA/trans-sialidase and cruzi-pain: two antigens fromTrypanosoma cruzicontain immuno-dominant but enzymatically inactive domains.FASEB Journal6,3259–3264

Cetron M.S., Basilio F.P., Moraes A.P.et al. (1993) Humoral andcellular immune response of adults from northeastern Brazil withchronic Trypanosoma cruziinfection: depressed cellular immuneresponse toT. cruzi antigen among Chagas’ disease patients withsymptomatic versus indeterminate infection.American Journal ofTropical Medicine and Hygiene49, 370–382

Costa F., Franchin G., Pereira-Chioccola V.Let al. (1998) Immuniza-tion with a plasmid DNA containing the gene oftrans-sialidasereducesTrypanosoma cruziinfection in mice.Vaccine16, 768–774

Dutra W.O., da Luz Z.M., Canc¸ado J.R.et al. (1996) Influence ofparasite presence on the immunologic profile of peripherial bloodmononuclear cells from chagasic patients after specific drug therapy.Parasite Immunology18, 579–585

Holscher C., Ko¨lher G., Muller U. et al. (1998) Defective nitric oxideeffector function leads to extreme susceptibility ofTrypanosomacruzi-infected mice deficient in gamma interferon receptor orinducible nitric oxide synthase.Infection and Immunity66,1208–1215

Leguizamon M.S., Campetella O., Russomando G.et al. (1994) Anti-bodies inhibiting Trypanosoma cruzitrans-sialidase activity insera from human infections.Journal of Infectious Disease170,1570–1574

M.Ribeirao et al. Parasite Immunology

q 2000 Blackwell Science Ltd,Parasite Immunology, 22, 49–5352

100

80

60

40

20

0

% o

f p

osi

tive

ser

a

Native TS Rec TS

IgM

IgG

(a)

100

80

60

40

20

0

Inh

ibit

ion

of T

S a

ctiv

ity

(%)

Healthyindividuals

(a)

Healthyindividuals

Chagasicpatients

Figure 1 Recognition of native and recombinant TS by plasmasamples from chagasic patients. (a) The percentage of responders wascalculated as those plasma samples that presented OD492 values 3 SDabove the average OD492 obtained for plasma samples from 15healthy individuals never exposed toT. cruzi. The cutoff values fornative TS were 0·10 or 0·16 OD492 for IgM and IgG, respectively.The cut-off values for recombinant TS were 0·11 or 0·10 OD492 forIgM and IgG, respectively. (b) Inhibition of TS activity wasperformed as described in the Methods section. All plasmas weretested at a final concentration of 25% (v/v). Each circle representsone individual. The bars represent the mean inhibition of each group.Statistical comparisons are provided in the Results section.

Morato M.J., Brener Z., Canc¸ado J.R.et al. (1986) Cellular immuneresponses of chagasic patients to antigens derived from differentTrypanosoma cruzistrains and clones.American Journal of TropicalMedicine and Hygiene35, 505–511

Pereira-Chioccola. V.L., Costa F., Ribeira˜o M. et al. (1999) Compari-son of antibody and protective immune responses againstTrypano-soma cruzi infection elicited by immunization with a parasiteantigen delivered as naked DNA or recombinant protein.ParasiteImmunology21, 103–110

Pereira-Chioccola V.L., Schenkman S. & Kloetzel J.K. (1994)Sera from chronic chagasic patients and rodents infected withTrypanosoma cruziinhibit trans-sialidase by recognizing its amino-terminal and catalytic domain.Infection and Immunity62, 2973–2978

Ribeirao M., Pereira-Chioccola V.L., Eichinger D.et al. (1997)Temperature differences fortrans-glycosylation and hydrolysisreaction reveal an acceptor binding site in the catalyticmechanism ofTrypanosoma cruzi trans-sialidase. Glycobiology7, 1237–1246

Rodrigues M.M., Ribeira˜o M., Pereira-Chioccola V.L.et al. (1999).Predominance of CD4 Th1 and CD8 Tc1 cells revealed by charac-terization of the cellular immune response generated by a DNA

vaccine containing aTrypanosoma cruzigene.Infection and Immu-nity 67, 3855–3863

Schenkman S., Diaz C. & Nussenzweig V. (1991a) Attachment ofTrypanosoma cruzitrypomastigotes to receptors at restricted cellsurface domain.Experimental Parasitology72, 76–86

Schenkman S., Man-Shiow J., Hart G.W.et al. (1991b). A novel cellsurface trans-sialidase ofTrypanosoma cruzigenerates a stage-specific epitope required for invasion of mammalian cells.Cell 65,1117–1125

Schenkman S., De Carvalho L.P. & Nussenzweig V. (1992)Trypano-soma cruzi trans-sialidase and neuraminidase activities can bemediated by the same enzymes.Journal of Experimental Medicine175,567–575

Schenkman S., Chaves L.B., Pontes de Carvalho L.et al. (1994) Aproteolytic fragment ofTrypanosoma cruzi trans-sialidase lackingthe carboxy-terminal domain is active, monomeric and generatesantibodies that inhibit enzymatic activity.Journal of BiologicalChemistry269,7970–7975

Soares I.S., Levitus G., Souza J.M.et al. (1997) Acquired immuneresponse to the N- and C-terminal regions ofPlasmodium vivaxmerozoite surface protein 1 in individuals exposed to malaria.Infection and Immunity65, 1606–1614

Volume 22, Number 1, January 2000 Type 1 immune response toTrypanosoma cruzi trans-sialidase

q 2000 Blackwell Science Ltd,Parasite Immunology, 22, 49–53 53