Chagas' disease: polyspecificity of antibodies against Trypanosoma cruzi acidic antigens

Acta Tropica, 59(1995)93-103 © 1995 Elsevier Science B.V. All rights reserved 0001-706X/95/$09.50

93

ACTROP 00453

Chagas' disease: polyspecificity of antibodies against Trypanosoma cruzi acidic antigens P. Ordofiez a'*, S. Gea a, D. Iosa b, E.

Vottero-Cima

a

alnmunologia, Departamento Bioquimiea Cliniea, Faeultad de Ciencias Quimieas, Universidad Naeional de Cordoba, Cordoba, CC 61, Suc. 16, (5000) Cordoba, Argentina b Centro Privado de Medicina, Cordoba, Argentina Received 12 September 1994; revised 2 December 1994; accepted 8 December 1994 Abstract The aim of this work was to study the reactivity of chagasic patient sera against a panel of natural antigens and its relationship with the immune response against T. cruzi acidic antigens. The presence of IgG and IgM antibodies reactive with myosin, myoglobin, actin and thyroglobulin was investigated in sera with positive serology for Chagas' disease classified into groups (G) I, n=7, with normal electrocardiogram (ECG) and no signs or symptoms of the disease; GII, n = 7, with ECG abnormalities but without cardiomegaly and GIII, n =7, with cardiomegaly and congestive heart failure. Healthy individual sera were analyzed in parallel as controls. In the three groups of chagasic patients, a high proportion of sera exhibited an enhancement of IgG response anti actin ranging from 71 to 100%. IgM against this antigen was found positive in GI, 21%; GII and GIII, 57%. The antibodies binding to myosin and myoglobin were mainly of IgM type. When myosin was assayed, the frequency of reactive sera was gradually diminished as heart involvement increased: GI 57%, GII 28% and Gill 14%. Only IgG antibodies against thyroglobulin were detected in the three groups of chagasic patients ranging from 43 to 86%. IgG natural antibodies showed to be polyreactive, since a diminished reactivity against each one of the natural antigens assayed and against T. cruzi acidic antigens (FIV) was observed in the sera absorbed with any of the selected antigens irrespective of the absorbing ones. Moreover, the antibodies against FIV parasite's antigens purified by immunoabsorption showed a similar reactivity with FIV, myosin and actin, and a slight lower reactivity with thyroglobulin. These results indicate that in chagasic patients, the specific humoral response against FIV is associated with an increase of the natural autoantibodies along with their polyspecificity. Key words: Chagas' disease; Trypanosoma cruzi; Acidic antigens; Polyspecificantibodies

1. ln~oducfion D u r i n g the c h r o n i c phase o f C h a g a s ' disease, a b o u t 30% o f chagasic patients develop lesions in their n e u r o m u s c u l a r system associated with cardiopathies or digestive p r o b l e m s ( K o b e r l e , 1968; Said et al., 1985; Iosa et al., 1989). It was related with i m m u n o l o g i c a l a n d i n m u n o p a t h o l o g i c a l reactions ( L a g u e n s et al., 1981; H o n t e b e y r i e - J o s k o w i c z et al., 1987; Petry a n d Eisen., 1989). I n fact, d u r i n g either the c h r o n i c or the acute phase, v a r i o u s a u t o i m m u n e p h e n o m e n a can be observed. The e x a c e r b a t i o n o f n a t u r a l a u t o a n t i b o d i e s has been d e m o n s t r a t e d in this infection, *Corresponding author Fax." 5451694724/695101. SSDI 0001-706X(95)00088-3

94 (Unterkircher et al., 1993; Grauert et al., 1993). It may be due to nonspecific polyclonal activation that occurs during T. cruzi infection (Ortiz-Ortiz et al., 1980; Minoprio et al., 1986). Another mechanism proposed as a possible cause of exacerbation of pathological manifestations involves the existence of antigens cross-reactive between T. cruzi and mammalian cells (Khoury et al., 1979; McCormick and Rowland, 1989; Petry and Eisen, 1989; Gea et al., 1990), but some controversies still remain (Kierszembaum, 1985). In order to study these phenomena it is interesting to analyse the overlap between the populations of natural and parasite specific antibodies. We considered that clearer results would be obtained if a specific antibody population is studied rather than all of them. Hence, a cytosolic acidic fraction (pI 4-5) purified from epimastigotes, named FIV, was the selected parasite antigen, since sera from chagasic patients have antibodies highly reactive with FIV (Gea et al., 1987; Gruppi et al., 1988). Besides, antibodies directed against FIV crossreacted with Cruzipain, a purified parasite antigen with cystein proteinase activity (Cazzulo et al., 1990), whose isoelectric point and molecular weight are similar to those of FIV (Gea et al., 1987; Malchiodi et al., 1993). Cruzipain is an antigen virtually identical to the GP 57/51 (Scharfstein et al., 1986) which has recently received considerable attention as a source of a pathogenetic antigen (Meirelles et al., 1990). Moreover, antibodies reactive to FIV were crossreactive with human heart tissue (Gea et al., 1990) and sciatic nerve antigens (Gea et al., 1993). In this work, sera from chagasic patients (ChP) with different degrees of heart involvement as well as sera from healthy humans were tested by ELISA against a panel of self antigens: actin, myoglobin, myosin and thyroglobulin, proteins highly conserved during species evolution behaving as natural antigens. These antigens were chosen because they are major cell constituents and represent a large fraction of the self molecules able to induce antibodies. To ascertain the polyspecificity of the revealed natural antibodies, a pool of ChP sera was absorbed with actin, myosin or thyroglobulin and the residual reactivity against each one and FIV was tested. Also, the reactivity of antibodies against FIV eluted from FIV-immunoadsorbent was tested against actin, myosin and thyroglobulin.

2. Materials and methods 2.1. H u m a n sera

Twenty one individuals, living in Argentinian endemic areas, 20 to 40 years old, with positive serology for Chagas' disease (indirect hemagglutination and immunofluorescense tests) were included in this study. They were divided into three disease severity groups according to the Consejo Argentino de Enferrnedad de Chagas, 1988, criteria: Group (G) I patients, with no cardiac symptoms, normal electrocardiograms (ECG) and normal chest radiography (CXR) films; GII patients with normal CXR but with ECG abnormalities and GIII patients with abnormal ECG findings, signs and symptoms of congestive heart failure and cardiomegaly on CXR. Control sera were taken from seven individuals, from the same endemic area, 25 to 40 years old, with negative serology.

95

2.2. FIV Trypanosoma cruzi antigens Epimastigote forms (EPI) from Tulahuen strain of Trypanosoma cruzi were harvested from cultures in monophasic medium (Camargo et al., 1964). The EPI homogenate was centrifuged at 105 000 g. The supernatant (cytosol) was submitted to isoelectric focusing (IEF) on agarose gel at pH gradient 3 to 10. The agar zone corresponding to pH 4-5, was cut, homogenized, eluted with 0.05M carbonate buffer pH 9.6 and used as the partially purified antigen named FIV (Gea et al 1987). The final yield was about 5% of the initial protein content. Other antigens used were actin from chicken muscle, thyroglobulin from porcine, myosin from rabbit muscle, myoglobin from whale sperm recombinant purchased from Sigma (St Louis, Mo, USA).

2.3. ELISA Plates were coated with 100 pl of 50 ~tg/ml FIV T. cruzi antigens, or 10 lxg/ml of actin, thyroglobulin, myoglobin or myosin. Coating was performed overnight at 4°C in carbonate buffer pH 9.6. Before the assay, the free binding sites were blocked with 5% human serum albumin (HSA) in PBS for 30 min at 37°C and plates washed three times with PBS containing 0.05% Tween 20 (PBS-Tween). After washing, peroxidase-conjugate anti human IgG or anti human IgM was added and incubated for 30 min at 37°C. The reaction was developed with o-phenylendiamine. The optical densities were read at 492 nm in an ELISA reader. Test sera were considered positive if the mean absorbance value was > 2 standard deviations above the mean value for control sera assayed in parallel.

2.4. Absorption studies Actin, myosin, thyroglobulin (10 ~tg/ml) and FIV (50 pg/ml) were adsorbed on nitrocellulose membrane, then washed with PBS, blocked with PBS-HSA 3% and incubated overnight with 5 ml of a pool of sera from all groups of chagasic patients diluted 1/10 in blocking solution. The unabsorbed antibodies were collected by three washes with PBS-0.05% Tween and two washes with 0.5M CINa-0.05% Tween (Hall et al., 1984). Original sera and the supernatant of absorbed sera were assayed by ELISA against natural and FIV antigens. Anti FIV antibodies from the pooled sera before mentioned or from controls, purified by immunoabsorption on FIV-nitrocellulose, were eluted with a solution of 1M Tris (pH 10.5). The eluted materials were dialyzed against PBS overnight at 4°C and tested by ELISA against natural and FIV antigens.

3. Results

3.1. Reactivity of sera from chagasic patients against natural antigen A quantitative determination of autoantibodies by ELISA in the sera of chagasic patients GI, GII and GIII and controls was carried out against a panel of natural antigens: actin, myosin, myoglobin and thyroglobulin. A variable level of the IgG

96 a n d I g M a g a i n s t these antigens was o b s e r v e d in the different groups. A s c a n be seen in Fig. 1, the actin was the a n t i g e n t h a t d e t e c t e d the highest levels o f I g G reactivity ( F i g . 1A) a n d the highest p e r c e n t a g e s o f positive sera ( F i g . 1B) in all g r o u p s o f chagasic patients: G I , 71%; G I I , 100% a n d G i l l , 86% . T h e response a g a i n s t

Actin

Myogtobin

My0~

Thyroglobulin

Q~jO l

080. 0'R, 060 050, 0~.0

~0

:

.

(1~0 " ' - ~ . . . . . . .

*-. . . . . . . . t

6:

Gil

6h5

G~

Gn

6~

i Gi

~ G11

J Gill

i

•

i

i

61

511

i 61[1

A)

% 120

100

80

60

40

20

0

~

actln

i

myosin

myoglobln

thyroglobulln

(B)

Fig. 1. IgG reactivity of sera from group I, II and III chagasic patients against natural antigens. (A) Results are expressed as the mean OD at 492 nm of duplicates determined by ELISA. Each symbol represents the value obtained with each serum assayed at 1:25 dilution. The dotted lines for each antigen indicate the mean OD obtained with the control sera plus 2 standard deviations. (B) The bars indicate percentages of positive sera for actin, myosin, myoglobin and thyroglobulin in GI II, GII m and Gill []. Sera were considered positive when their values were above the dotted lines.

97 thyroglobulin was exclusively of IgG type and this reactivity was detected in the three groups; GIII having the highest frequency of positive sera. Only 1 out of 7 patients of GI and 1 out of 7 patients of GIII showed IgG reactive with myosin. IgG against myoglobin was not detected. As regards the IgM response against natural antigens, the highest reactivity levels were observed with myoglobin and myosin (Fig. 2A). With myosin the percentages of positive sera in GI, GII and Gill were 57%, 28% and 14%. With myoglobin they were 71% for GI and GII, and 28% for GIII. Assaying the IgM reactivity against actin, some sera gave positive results irrespective of the clinical state (Fig. 2A-2B).

3.2. Study of polyspecificity of natural antibodies In order to study the polyspecificity of natural antibodies, pools of sera selected because of their high reactivities with each of the natural antigens, were absorbed with actin, myosin, or thyroglobulin fixed onto nitrocellulose. The original sera and the supernatants of absorbed sera were assayed by ELISA against all the natural antigens selected for this study and FIV T. cruzi antigens. In the experimental conditions used, the original homologous binding activities were not totally eliminated. Decreased reactivity against actin, myosin, thyroglobulin, and FIV was observed with all absorbed sera regardless of the absorbing antigen (Fig. 3). A decrease of about 50% in the reactivity against actin (Fig. 3A) and 63% against myosin was observed (Fig. 3B). The residual anti thyroglobulin reactivity ranged between 35% and 67% (Fig. 3C). Noteworthy, on assaying the reactivity against FIV T. cruzi antigens, the highest inhibition of anti parasite antibodies was observed with sera absorbed with myosin (78%). After absorption with thyroglobulin, the diminution was 63% and 40% after absorption with actin (Fig. 3D). These results indicate that a cross-reactivity exists between natural and FIV antigens. Moreover, as can be seen in Fig. 4, the antibodies anti FIV purified by immunoabsorption and assayed by ELISA against myosin, actin, thyroglobulin and FIV antigens, showed a reactivity against myosin and actin similar to the reactivity detected with FIV which was about 5 fold higher than that obtained with controls. The reactivity of anti FIV with thyroglobulin was somewhat lower than with the before mentioned antigens.

4. Discussion

In the present report the antibody response induced in humans alter infection with T. cruzi was analyzed by ELISA against a panel of self antigens. The IgG and IgM responses were studied in three groups of ChP with different degrees of heart involvement. In all groups we found an increase of both IgG and IgM antibodies to actin. All the reactivity revealed against thyroglobulin was of the IgG type, while the myoglobin fixed only IgM antibodies and myosin fixed mainly IgM. The percentages of patients showing IgG reactivity against thyroglobulin appeared to increase according to the severity of the disease. On the other hand, the levels and the percentages of sera with IgM antibodies against myosin and myoglobin were higher in GI (asymptomatic group) than in Gill (the group most clinically affected). However, a clear association between clinical state of Chagas' disease and the

98 MyosJn

~tin

1

My0globin

"f hyrogl0bulin

.

0.90. 080. 070 (160 060

,

i

~,o . . . . . . . . : ...... ~ .

.

.

.

_._

.....

._ .....

..,_-

O2O

G]

i

,

Gll

Gl.l~

i

,

J

i

GI

i

Gll

Gll]

i

i G1

i GII

Gm

i

i

I

GI

i

EH

1

Gill

(A)

% 80

60

40

20

actin

myosin

myoglobin

thyroglobulin

(8) Fig. 2. IgM reactivity of sera from group I, II and III chagasic patients against natural antigens. (A) and (B) are expressed as indicated in Fig. 1 A and B.

reactivity against natural antigens was not found. Enhanced response of natural IgG antibodies has been observed in patients with other parasitic infections (Howard et al., 1987). Moreover, IgM and IgG natural antibodies have also been reported in the sera of animals chronically infected with T. cruzi (Ternynck et al., 1990; Spinella et al., 1992) and Leishmania (Pateraki et al., 1983). It is likely that the

99 OD anti actin

1

OD anti myosin

o.a] (B)

OD anti Thyroglobulin

OD anti FIV

2[(0)

Fig. 3. Polyspecificity of natural antibodies. Reactivity of IgG antibodies anti actin (A); anti myosin (B); anti tbyroglobulin (C) and anti FIV parasite antigens (D) determined by ELISA using pooled chagasic patients sera assayed at 1:25 dilution an, or their supernatants at the same dilution [] after absorption on (a) actin, (m) myosin or (t) thyroglobulin-nitroeellulose. The percentages of reactivity diminution are indicated.

100 OD 2 - -

1.5

0.5

myosin

actin

thyroglobulin

FIV

Fig. 4. Reactivity of anti FIV antibodies purified by immunoabsorption on FIV-nitrocellulose from sera pooled of chagasic patients • and of healthy individuals as control [3. The eluted antibodies were tested by ELISA at dilution 1:25 using as antigen myosin, actin, thyroglobulin and FIV parasite antigens.

increase of both classes of autoantibodies resulted from the polyclonal stimulation induced by the infection. Besides, it could also arise from a permanent stimulation of the host by self constituents from infected and damaged host cells or perhaps by cross-reactive antigens. The molecular mimicry of mammalian target organs by a multitude of parasite molecules may be responsible for stimulating self-directed lymphocyte clones. Much of this molecular mimicry has been demonstrated by cross-reactive antibodies (Petry and Eisen, 1989). The polyclonal cell activation induced in mice by T. cruzi infection has been previously demonstrated (Minoprio et al., 1988). It seems that T cells like TH2 cooperating for multiple B cell response are induced. In fact, the T-cell line derived from chronically T. cruzi infected mice was able to induce polyclonal B cell activation when transferred to naive animals (Spinella et al., 1990). The role of polyreactive antibodies is still controversial. A pathogenetic role is assigned to them, since it has been demonstrated that these autoantibodies are correlated with clinical symptoms for some autoimmune diseases (Ternynck et al., 1990). On the other hand, a beneficial role of natural autoantibodies modulating some autoimmune diseases has been proposed (Rossi et al., 1989). In this context, the high level of anti-galactosyl antibodies present in human acute chagas' disease was proposed as first line of protection against early infection (Gazzinelli., 1991). In the same way the IgM antibody levels against myosin seem to be inversely correlated with heart involvement in agreement with Unterkircher et al., 1993. Being the polyspecificity a characteristic of natural antibodies (Avrameas and Ternynck, 1993), it was further analyzed. This study revealed that at least a part of the IgG reactive with actin, myosin or thyroglobulin is absorbed by any of these natural antigens as well as by the partially purified FIV parasite antigen. The absorptions with each antigen removed a variable percentage of reactivity against the homologous and the other antigens assayed. Similar findings were also found on analyzing antibodies from mice chronically infected with T. cruzi (Ternynck et al., 1990) and

101

in an acute case of human Chagas' disease (Grauert et al., 1993). Noteworthy, we observed a high percentage of inhibition of IgG anti-FIV antibodies when the sera were absorbed with myosin (78%), thyroglobulin (63%) or actin (40%). Moreover, the IgG anti-FIV antibodies purified by immunoabsorption from a pool of chagasic patients sera were able to bind to FIV antigens and also to actin, myosin and thyroglobulin. These findings indicate an important cross-reactivity between natural antigens and the selected T. cruzi antigens. It is know that the natural polyreactive autoantibodies have in their variable regions a particularly high number of positivily charged amino acids (Conger et al., 1992). As a consequence, these antibodies can interact preferentially with constituents bearing negative charges (Gonzalez-Quintal et al., 1990) between them, the electronegative T. cruzi antigens. The antibodies anti actin, myosin and thyroglobulin purified with each immunoabsorbent were also assayed in a similar way, but their reactivities were very low perhaps due to a denaturalization of the eluted natural antibodies. It would be conceivable that the polyclonal activation was the first barrier of the defense mechanism for early T. cruzi infections as it is for other infections. However, the cross-reactivity mechanism may take over in the outbreak of autoimmune reactions, long time after the initial infection has elapsed.

Acknowledgments This work was supported by the Consejo de Investigaciones Cientificas y Tecnologicas de la Provincia de Cordoba (CONICOR) and Consejo Nacional de Investigaciones Cientificas y Tecnicas (CONICET).

References Avrameas S. and Ternyck T. (1993) The natural autoantibody system. Between hypothesis and facts. Mol. Immunol. 30, 12, 1133. Camargo E.P. (1964) Growth and differentiation in T. cruzi origen of metacyclic trypanosomes in liquid media. Rev. Inst. Med. Trop. S. Panlo 6, 93-100. Cazzulo, J., Cazzulo Frank, M., Martinez, J. and Franke de Cazzuto, B. (1990) Some kinetics properties of a cysteine proteinase (cruzipain) from T. cruzi. Biochim. Biophys. Acta. 1037, 186-191. Conger, J.D., Sage H.J and Corley R.B. (1992) Correlation of antibody multireactivity with variable region primary structure among murine anti-erythrocyte autoantibodies. Eur. J. Immun. 22, 783-790. Consejo Argentino de Enfermedad de Chagas "Dr. Salvador Mazza". (1988) Rev. Arg. Cardiol. 56, 108-118. Gazzinelli, R.T. (1991) Natural anti-Gal antibodies prevent rather than cause, autoimmunity in human Chagas' disease. Res. Immunol. 142, 164-167. Gea, S., Gruppi, A., Basso, B., Menso, E. and Vottero-Cima, E. (1990) Antibodies to T. cruzi cytosol acidic antigens (FIV) in Chagas' disease recognize parasite cell surface and human heart epitopes. J. Clin. Lab. Immunol. 31, 183-187. Gea, S., Ordofiez, P., Cerban, F., Iosa, D., Chizzolini, C. and Vottero-Cima, E. (1993) Chagas'cadioneuropathy: Association of anti T cruzi and anti sciatic nerve antibodies. Am. J. Trop. Med. Hyg. 49(5), 581-588. Gea, S., Rodriguez, P., and Vottero-Cima, E. (1987) Characterization of Trypanosoma cruzi antigens recognized by sera from patients with chronic Chagas' disease. Int. Arch. Allergy. Appl. Immunol. 84, 410-413. Gonzalez-Quintial R., Baccala R., Alzari P.M., Nahmias C., Mazza G., Fougerau M. and Avrameas S.

102 (1990) Poly (Glu6° Ala3° Tyr 1°) (GAT)-induced IgG monoclonal antibodies cross-react with various self and non-self antigens through the complementarity determining regions. Comparison with IgM monoclonal polyreactive natural antibodies. Eur. J. Immunol. 20, 2383-2387. Grauert M., Houdayer M. and Hontebeyrie-Joskowicz M. (1993) Trypanosoma cruzi infection enhances polyreactive antibody response in an acute case of human Chagas' disease. Clin. Exp. Immunol. 93, 85-92. Gruppi, A., Moretti, E., Basso, B. and Vottero-Cima, E. (1988) Exoantigenos del T. cruzi. Epitopes reactivos con Igs de pacientes con enfermedad de Chagas. Medicina (Bs. As.) 48, 161-166. Hall R., Hyde J.E., Goman M., Simmous D.L., Hope A., Mackay M., Scarfe J., Merkli B., Richle A. and Stocker J. (1984) Major surface antigen gene of a human malaria parasite cloned and expressed in bacteria. Nature 311, 379-390. Hontebeyrie-Joskowicz, M., Said, G., Marchal, G., and Eisen, H. (1987) L3T4 T cells able to mediate parasite-specific delayed type hipersensivity play a role in the pathology of experimental Chagas' disease. Eur. J. Immunol. 17, 1027-1033. Howard M.K., Gull K. and Miles M.A. (1987) Antibodies to tubulin in patients with parasitic infections. Clin. Exp. Immunol. 68, 78-85. Iosa, D., De Quatro, V., De-Pink Lee, D., Elkayam, V. and Palmero, H. (1989) Plasma norepinephrine in Chagas' cardioneuromyopathy: a marker of progressive dysautonomia. Am. Heart. J. 117, 882-887. Khoury, E., Ritacco, V., Cossio, P., and Arana, R. (1979) Circulating antibodies to peripheral nerve in American trypanosomiasis (Chagas' disease). Clin. Exp. Immunol. 36, 8-13. Kierszembaum, F. (1985) Is there autoinmune in Chagas' disease? Parasitol. Today 1, 4-10. Koberle, F. (1968) Chagas' disease and Chagas' syndromes: The pathology of American Trypanosomiasis. Adv. Parasitol. 6, 63-116. Laguens, R.P., Cabeza-Merkert, P.C., Chambo, J.G. and Gelp, R. (1981) Chronic Chagas' disease in the mouse. Transfer of the heart disease by means of immunocompetent cells. Medicina (Bs.As) 41, 40-43. Louzir, H., Ternynck, T., Gorgi, Y., Tahar S., Ayed, K. and Avrameas, S. (1992) Autoantibodies and circulating immune complexes in sera from patients with hepatitis B virus related chronic liver disease. Clin. Immunol. Immunopathol. 62, 160-167. Malchiodi, E.L., Chiaramonte, M.G., Martinez, J.A., Zwirner, N.W., Margni, R.A. and Cazzulo, J.J. (1993) Identity of the major cysteine proteinase Cruzipain from Trypanosoma cruzi and an antigen (Ag 163B6) isolated with a monoclonal antibody. Immunol. Lett. 35, 59-62. Mc.Cormick, T.S., and Rowland, E. (1989) Trypanosoma cruzi: cross-reactive anti-heart autoantibodies produced during infection in mice. Exp. Parasitol. 69, 393-401. Meirelles, M.N.L., Juliano, L., Carmona, E., Costa, E., Silva, S.G., Lima, A.T., Arnholdt, A.V., Leme, V.M., Guimares, E.S., Berro, O.J. and Scharfstein, J. (1990) Functional and antigenic properties of the major cystein proteinase (GP 57/51) of T. cruzi. Mem. Inst. Oswaldo Cruz. Rio J. 85, 533-538. Minoprio P., Burlen O., Pereira P., Guilbert R., Andrade L., Hontebeyrie- Joskowicz M. and Coutinho A. (1988) Most B cells in acute T. cruzi infection lack parasite specificity. Scan. J. Immunol. 28, 553-551. Minoprio, P.M., Eisen, H., Forni, L.D., Imperio Lima, M.R., Joscowicz, M. and Coutinho, R. (1986) Polyclonal responses to murine T. cruzi infection. Quantitation of both T and B responses. Scan. J. Immunol. 24, 661-668. Ortiz-Ortiz, L., Elliot, L., Parks, D., Rodriguez, M. and Weigle, W. (1980) Polyclonal B lymphocyte activation during T. cruzi infection. J. Immunol. 124, 121-125. Pateraki E., Portocala R., Labrouse H. and Guesdon J.L. (1983) Anti actin and anti tubulin antibodies in canine visceral leishmaniasis. Infect. Immun. 42, 496-500. Petry, K. and Eisen, H. (1989) Chagas' disease: A model for the study of autoinmune disease. Parasitol. Today 5, 111-116. Rossi F., Dietrich G. and Kazatckkine M. (1989) Anti-idiotypes against autoantibodies in normal immunoglobulins evidence for net-work regulation of human autoimmune responses. Immunol. Rev. 110, 135-149. Said, G., Joscowicz, M., Barreira, A.A. and Eisen, H. (1985) Neurophathy associated with experimental Chagas' disease. Ann. Neurol. 18, 676-683. Scharfstein, J., Schechter, M., Senna, M., Peralta, J., Mendonca-Previato, L. and Miles, M. (1986) Trypanosoma cruzi: Characterization and isolation of a 57/51.000 mw surface glycoprotein expressed by epimastigotes and bloodstream trypomastigotes. J. Immunol. 137, 1336-1341. Spinella S., Liegeard P., Hontebeyrie-Joskowicz M. (1992) Trypanosoma cruzi: predominance of IgG2a in nonspecific humoral response during experimental Chagas' disease. Exp. Parasitol. 74, 46-56.

103 Spinella, S., Milon, G. and Hontebeyrie-Joskowicz, M. (1990) A CD4TH 2 cell line isolated from mice chronically infected with Trypanosoma cruzi induces IgG2 polyclonal response in vivo. Eur. J. Immunol. 20, 1045-1051. Ternynck, T., Bleux, C., Gregoire, J., Avrameas, S. and KaneUopoulos-Langevin C. (1990) Comparison between autoantibodies arising during Z cruziinfection in mice and natural autoantibodies. J. Immunol. 144, 1504-1511. Unterkircher, C., Avrameas, S. and Ternynck, T. (1993) Autoantibodies in the sera of Trypanosoma cruz#infected individuals with or without clinical Chagas' disease. J. Clin. Lab. Analysis 7, 60-69.