T Cell Response in Rheumatic Fever: Crossreactivity Between Streptococcal M Protein Peptides and Heart Tissue Proteins

Transcription

1 Current Protein and Peptide Science, 2007, 8, T Cell Response in Rheumatic Fever: Crossreactivity Between Streptococcal M Protein Peptides and Heart Tissue Proteins Luiza Guilherme 1,2 *, Kellen C. Faé 1,2, Sandra E. Oshiro 1,2, Ana C. Tanaka 1, Pablo M.A. Pomerantzeff 1 and Jorge Kalil 1,2,3 1 Heart Institute (InCor), School of Medicine, University of São Paulo; 2 Institute for Immunology Investigation, Millennium Institute; 3 Clinical Immunology and Allergy, Department of Clinical Medicine, University of São Paulo, School of Medicine, São Paulo, Brazil Abstract: Molecular mimicry between streptococcal and human proteins has been proposed as the triggering factor leading to autoimmunity in rheumatic fever (RF) and rheumatic heart disease (RHD). In this review we focus on the studies on genetic susceptibility markers involved in the development of RF/RHD and molecular mimicry mediated by T cell responses of RHD patients against streptococcal antigens and human tissue proteins. We identified several M protein epitopes recognized by peripheral T cells of RF/RHD patients and by heart tissue infiltrating T cell clones of severe RHD patients. The regions of the M protein preferentially recognized by human T cells were also recognized by murine T cells. By analyzing the T cell receptor (TCR) we observed that some V families detected on the periphery were oligoclonal expanded in the heart lesions. These results allowed us to confirm the major role of T cells in the development of RHD lesions. Keywords: Rheumatic fever, rheumatic heart disease, Streptococcus pyogenes, T cell response, autoimmunity, molecular mimicry, M protein, heart proteins. INTRODUCTION Acute Rheumatic Fever (acute RF) results from an autoimmune response triggered by Streptococcus pyogenes infection. Although the acute illness is accompanied by important clinical symptoms, including carditis, polyarthritis, chorea and cutaneous disease, the major clinical and public-health effects derive from the long-term damage to heart valves that characterizes rheumatic heart disease (RHD). According to World Health Organization data [1] at least 15.6 million people have RHD; 40 to 60% of individuals who acquire acute RF every year go on to develop RHD, and 233,000 deaths are directly attributable to acute RF or RHD every year [1]. The incidence of acute RF exceeds 50 per 100,000 children in some developing countries. The highest reported rates, found among the indigenous populations of Australia and New Zealand, are of about 500 per 100,000 children [2]. In Brazil, the incidence of acute RF in the 1990 s was 360 cases per 100,000 children [3], but these numbers decreased in the last years, according to the Brazilian Ministry of Health. Even though the pathogenesis of RF is not completely understood, it is clear that an exacerbated immune response to bacterial antigens in susceptible hosts leads to autoimmune attack to several tissues and, in RHD patients, this triggers an inflammatory response to heart tissue, probably caused by molecular mimicry between group-a streptococcus antigens and heart tissue proteins. *Address correspondence to this author at Laboratório de Imunologia, Instituto do Coração (HC-FMUSP). Av. Dr. Eneas de Carvalho Aguiar, 44-9 andar São Paulo, SP, Brazil; Tel: ; ; Fax: ; luizagui@usp.br HOST GENETIC FACTORS In the 19th century, familial aggregation suggested that acute RF and RHD have a genetic background [4]. Numerous studies have associated specific genetic markers with RF and RHD. Many of these studies focus on the Major Histocompatibility Complex (MHC) region, with special emphasis on HLA class II polymorphisms and have shown associations between particular alleles with susceptibility to the disease in different populations. The first reports were done based on serological HLA class II typing that provide the definition of few alleles. Later the allele definition was improved by molecular typing that brought more accurate definition of the specific allele association with clinical features of the disease. Among the HLA class II alleles studied, HLA-DR7 was the one most consistently associated with the disease [5, 6, 7, 8, 9]. In RHD patients from Latvia, the presence of DR7 with DQB1*0302 and DQB* alleles seems to be associated with the development of multiple valvular lesions (MVL) and mitral valve regurgitation (MVR), respectively [9]. The presence of DR7 with different DQ-A alleles (DQA*0102 and DQA*0401) was also associated with mitral valve regurgitation (MVR) in Egyptian RHD patients [8]. In the Brazilian population, the HLA-DR7 and DR53 alleles were found to be in strong association with RF/RHD among mulatto Brazilian patients [5, 7]. HLA-DR4 and DR9 were found to be associated with RF also in American Caucasians, Arabians, and in Indians from Kashmir [10, 11, 12, 13]. Other HLA class II antigens such as DR1, DR2, DR3, and DR6 were also found to be associated with RF/RHD in other populations [11, 14-18]. In Japanese RHD patients, susceptibility to mitral stenosis seems to be in part controlled by one or more genes in the HLA-DQ region, in close linkage disequilibrium with HLA-DQA*0104 and /07 $ Bentham Science Publishers Ltd.

2 40 Current Protein and Peptide Science, 2007, Vol. 8, No. 1 Guilherme et al. DQB1*05031 [19]. Alleles HLA-DQA*0501 and DQB*0301 in linkage disequilibrium with DRB1*1601 (DR2) were associated with RHD in a Mexican Mestizo population, and HLA-DR16 frequency was significantly increased in patients with multivalvular lesions [20]. In addition, associations with polymorphisms in genes coding for cytokines and other molecules directly involved in the control of immune response have been also described. Polymorphisms of transforming growth factor-beta 1 [21], immunoglobulin [22] and TNF-alpha [23] genes were associated with susceptibility to RF development. Recently, a striking association has been shown between a polymorphism in the Toll-like receptor (TLR-2) gene and occurrence of RF among Turkish children [24]. Taken together, the studies above mentioned provide strong evidence for the involvement of host genetic factors in disease susceptibility. Several genes are likely to predispose an individual to developing RF. This is compatible with what is seen in multifactorial disorders, which include a number of other autoimmune diseases, and which are probably caused by a combination of factors, including host susceptibility, infectious agent characteristics, and environmental factors. IMMUNOLOGICAL MECHANISMS INVOLVED IN RHD PATHOGENESIS It is possible that the autoimmune response behind acute RF be triggered by molecular mimicry between antigens of group-a streptococcus and specific human tissues. The M protein and N-acetylglucosamine are the most studied and well-characterized of these bacterial antigens, mainly due to their ability to elicit the production of crossreactive antibodies and cell-mediated immunity, which may lead to host tissue destruction. As to host target antigens, crossreactivity has been reported with antigens expressed in the joints (arthritis), heart (carditis), and central nervous system (chorea) (reviewed by Cunningham et al. [25]). HUMORAL IMMUNE RESPONSE For several decades, investigators have pursued the theory that antibodies from RF patients may cause carditis [26]. However, the concept of an involvement of autoimmune reactions in the pathogenesis of RF was introduced only in the 1960 s by Kaplan [27, 28]. This author showed that rabbit antisera against group-a streptococci reacted with human heart preparations. Following this discovery, several investigators have attempted to identify the crossreactive antigens that induced this humoral response. On the whole, it was demonstrated that the streptococcal M5 protein was capable of eliciting heart-reactive antibodies, thus implying that the M protein was one of the streptococcal components responsible for crossreactivity [29]. Anti-M protein antibodies were subsequently shown to crossreact with vimentin and cardiac myosin, suggesting that these proteins were the target autoantigens recognized in the heart [30, 31, 32, 33, 34]. Using anti-myosin antibodies purified by affinity from acute rheumatic fever patient sera, the authors identified crossreactive epitopes from myosin and the M5/M6 proteins [35]. Another streptococcal antigen capable of eliciting crossreactive antibodies is the N-acetylglucosamine carbohydrate. Studies conducted by Goldstein and colleagues showed that antibodies to this antigen crossreacted with glycoproteins present in the heart valves that contain N-acetylglucosamine [36]. It was also demonstrated that N-acetylglucosamine (GlcNAc) antibodies from RF patients crossreacted with cardiac myosin and laminin [37]. These antibodies showed cytotoxic activity against human endothelial cell-lines and reacted with human valvular endothelium and underlying basement membrane [37]. These data support the hypothesis that crossreactive antibodies in rheumatic carditis cause injury to the endothelium and underlying matrix of the valve. The deposition of autoantibodies in the heart tissue is probably the initial inflammatory event that triggers cellular infiltration. In agreement with this hypothesis, it was shown that valvular endothelium from RF patients showed increased expression of vascular cell adhesion molecule-1 (VCAM-1) [38], which facilitates T cell infiltration through the endothelium into the valves, leading to chronic inflammation. However, a direct role of crossreactive antibodies in RHD has never been shown. CELLULAR IMMUNE RESPONSE Only 25 years after the putative role of antibodies in the development of RF was described did the role of cellular immune responses in RF begin to be investigated. Peripheral T lymphocytes from patients with acute RF showed high response levels to streptococcal cell-wall and membrane antigens [39, 40]. Further studies reported the presence of cytotoxic T lymphocytes in the blood of acute RF patients [41] and that stimulation with the M protein could induce cytotoxic lymphocytic responses [42]. We showed that peripheral T cell response to M5 peptides could discriminate between the M protein recognition patterns of severe and mild RHD patients and healthy subjects [43]. Peptides M5(81-96) and M5(91-103) were recognized by 46.0% of severe RHD patients and 8.6% of healthy subjects (P=0.0005), and 24.3% of severe RHD and 3.0% of healthy subjects (P=0.01), respectively (Table 1) [43]. Peptides M5(11-25) and M5( ) were preferentially recognized by mild RHD patients when compared to healthy subjects (P=0.008 and P=0.01), respectively (Table 1). Reactivity to peptide M5( ) discriminates between severe and mild RHD patients (P=0.03). In contrast, M5( ) was preferentially recognized by both mild and severe RHD patients, and not by healthy subjects (P=0.04) (Table 1) [43]. It is interesting to note that some streptococcal M5 human T-cell epitopes were also recognized by murine T lymphocytes [46, 47]. Peptides M5(1-25) and M5(81-96), recognized by RHD patients, aligned peptides M5(1-35) and NT5 (59-76), respectively, which are recognized by murine T cells [46, 47]. Peptide M5( ), recognized by human T cells, aligned with four overlapping peptides (B1B2, B2, B2B3A and B3A), amino acid residues 137 to 193, recognized by murine T cells (Table 2). In the murine model, these peptides crossreacted with cardiac myosin [47]. These data strongly suggest that certain regions of the M5 protein are more involved in triggering crossreactivity with self antigens.

3 T Cell Response in Rheumatic Fever Current Protein and Peptide Science, 2007, Vol. 8, No Table 1. Epitopes from the N-Terminal Region of Streptococcal M5 Protein Preferentially Recognized by Peripheral Blood T Lymphocytes from Different Clinical Forms of Rheumatic Heart Disease (RHD) M5 protein RHD Patients (% Reactivity) Severe RHD Mild RHD Healthy subjects P values a P = a 8.6 P = a P = a P = * 31.3 b P = a 3.0 P = * 31.3 a P = a 24.2 a 5.7 P = 0.04 a P value compared with healthy individuals; b P value compared with mild RHD patients; Number of individuals tested for each peptide: severe RHD (n = 37), mild (n = 33) and healthy subjects (n = 35); * Peptides tested on different numbers of subjects: severe RHD (n = 50), mild/sydenham chorea (n = 54) and healthy subjects (n = 48); Peptides were synthesized as mers, based on the previously described streptococcal M5 protein sequence [43, 44, 45]. Peptide sequences are presented in Table 2. T cells reactivity was evaluated by proliferation assay as previously described [46]. Table 2. Epitopes from the N-Terminal Region of Streptococcal M5 Protein Recognized by Human and Murine T Cells Streptococcal M5 epitopes Amino acid sequences Human T cells * 1-25 TVTRGTISDPQRAKEALDKYELENH DKLKQQRDTLSTQKET LKQQRDTLSTQKETLEREVQN YNNETLKIKNGDLTKELNK NGDLTKELNKTRQELANKQQ NEKALNELLEKTVKD ELLEKTVKDKIAKEQENKET ETIGTLKKILDETVK LDETVKDKLAKEQKSKQNI Murine T cells ** 1-35 AVTRGTINDPQRAKEALDKYELENHDLKTKNEGLK (NT4) GLKTENEGLKTENEGLKTE (NT5) KKEHEAENDKLKQQRDTL (NT6) QRDTLSTQKETLEREVQN (B1B2) VKDKIAKEQENKETIGTL (B2) TIGTLKKILDETVKDKIA (B2B3A) KDKIAKEQENKETIGTLK (B3A) IGTLKKILDETVKDKLAK Overlapping sequences are underlined and epitopes recognized by both human and murine T cells are in bold type; * M5 protein epitopes recognized by human T cells were described by Guilherme et al. [43, 48]; ** M5 protein epitopes recognized by murine T cells were described by Cunningham et al. [47], except for epitope 1-35 that was described by Robinson et al. [46]; Peptide sequences based on the streptococcal M5 protein sequence described by Manjula et al.. [44] and Phillips et al.. [45], except for peptide sequences and that were based on the streptococcal M5 protein sequence described by Robinson et al. [46].

4 42 Current Protein and Peptide Science, 2007, Vol. 8, No. 1 Guilherme et al. Studies conducted by Raizada et al. (1983) and Kemeny et al. (1989) [49, 50] demonstrated the presence of T cells in the heart lesions of RHD patients. These authors showed an intense inflammatory infiltrate in rheumatic valvular tissue, with predominance of CD4 + T cells and macrophages, with only occasional B cells. However, the functional role of these infiltrating T cells was first demonstrated by our group. We characterized molecular mimicry between streptococcal M protein and heart tissue proteins at the T cell level [48]. In this work, we isolated and characterized the cellular reactivity of heart-infiltrating T cell clones cultured from heart tissue fragments obtained from severe RHD patients undergoing valve replacement. Heart-infiltrating T cell clones simultaneously recognized streptococcal M5 synthetic peptides and heart tissue-derived proteins, indicating crossreactive epitopes. Among the heart-infiltrating T cell clones studied, 7.4% (12/163) recognized at least one of the M5 peptides tested Fig. (1) [43, 48]. M5 peptides (1-20, 11-25, 62-82, 81-96, and ) were preferentially recognized. Of the heart tissue-derived proteins identified by molecular weight, 6.1% (10/163) and 3.0% (5/163) of intralesional T cell clones recognized valve and myocardium-derived proteins, respectively Fig. (1) [43, 48]. In order to verify if the pattern of reactivity observed in the site of the heart lesions was shared with T cells from the periphery we also analyzed peripheral T cell clone responses to M protein and aortic valve-derived proteins. Peripheral T cell clones showed 8.7% (2/23) reactivity to the overlapping M5(81-96) and M5(83-103) peptides, which were also recognized by intralesional T cell clones Fig. (1). Interestingly, we also found that certain heart-tissue proteins (e.g., the kda proteins) were recognized by both intralesional and peripheral T cell clones [43, 48]. In addition, other heart proteins from the myocardium and valvular tissue were recognized Fig. (1), suggesting that several autoantigens are recognized, probably through molecular mimicry and epitope spreading. The fact that the same antigens were recognized both in the periphery and in the heart suggests that specific T cell populations migrated from the periphery to the heart lesion, probably driven by antigen recognition. Confirming this hypothesis, we have shown that CD4 + T cell clones obtained from mitral valveinfiltrating T cells crossreact with mitral valve proteins isolated by molecular weight (MW) and isoelectric point (pi), including the 56-53kDa/pI6.76 and 35kDa/pI8.4 proteins, as well as with the streptococcal M5(81-103) peptide. Another CD4 + T cell clone obtained from myocardium-infiltrating T cells also recognized the same 56-53kDa/pI6.76 mitral valve protein. A similar myocardium protein fraction (50-54 kda) was also recognized by peripheral T lymphocytes from RHD patients after being stimulated in vitro with streptococci antigens (51). TCR analysis characterized the CD4 + intralesional T cell clone responsive to the 56-53kDa/pI6.76 mitral valve protein as being V 13 J 2S7, V 2 Fig. (2); however they differ by only one alpha chain (V 3 J 44 or V 7 J 29) Fig. (2) [52]. Interestingly, we found 4.2% of the V 13 family in the periphery and an expansion of V 13 (8.2%) in the heart tissue, with an oligoclonal expansion of V 13 J 2S7 (61.8%) in the heart Fig. (2) [53]. Our group has recently shown that the immunological response and consequent cardiac lesions are exacerbated by in situ cytokine production [55]. In this work we observed that IFN-gamma, TNF-alpha and IL-10 positive cells were consistently predominant in both myocardium and valvular tissue. IL-4 positive cells were also predominant in the myocardium, whereas IL4-positive cells were scarce in the valves. In agreement with these data, in vitro experiments using heart-infiltrating T cell lines stimulated with M5 recombinant protein and the immunodominant M5(81-96) peptide showed that these cells produce preferentially IFNgamma and IL-10. IL-4 production was detected only with T cell lines derived from the myocardium but not from the valve. These results suggest that in the heart there is a predominance of Th1 cytokine production, and that these proinflammatory cytokines may be involved in the maintenance and perpetuation of rheumatic lesions. Possibly, the significantly lower production of IL-4 in the valvular tissue may contribute to the progression of RHD, leading to permanent and more severe valvular damage. Fig. (1). Cross-recognition of streptococcal M5 peptides and heart-tissue proteins by peripheral and intralesional T cell clones derived from severe RHD patients. Intralesional T-cell lines were derived from in vitro culture of surgical fragments of mitral and aortic valves and myocardium from severe RHD patients undergoing valve-replacement surgery, as previously described [48].

5 T Cell Response in Rheumatic Fever Current Protein and Peptide Science, 2007, Vol. 8, No Fig. (2). Distinct heart tissue crossreactive T cell clones recognized the same antigens as peripheral T cell clones. TCR repertoire was analyzed by Immunoscope [54]. TCR VB-13 was represented in both peripheral blood and heart tissue-infiltrating T cells [53]. Antigen-specific isolated T cell clones from mitral valve and myocardium bear the same BV 13 JB 2S7, differing by only one TCR -chain [52]. Both T cell clones recognized the kda/pi 6.76 protein, characterized previously as kda, and the 35 kda/pi 8.4 protein (previously kda) {Fig. (1)}; and mitral valve-derived T cell clones also recognize the M5 (81-103) peptide. The same pattern of recognition was found in the periphery [52]. CONCLUDING REMARKS All the findings contributed by our group and by many other investigators have permitted a better understanding of the pathogenesis of RF/RHD. According to this view, during the acute phase of the disease, susceptible individuals show an exaggerated immune response to group A carbohydrates or to the M protein. Specific streptococcus antibodies deposit in the heart tissue and crossreact with alpha helical proteins such as myosin, leading to heart tissue damage and triggering an inflammatory process in the myocardium. Inflammation increases the expression of adhesion molecules such as VCAM-1 that facilitate cell recruitment and migration to the lesion site [38]. The development and progression of lesions are mediated by the infiltration of T cells and macrophages [48, 49, 50, 51, 56]. In the heart, T cells recognize self antigens such as myosin. Myosin is an intracellular protein, and is hence sequestered from the immune system. However, normal cardiac cell turnover may expose the epitopes of the protein to the immune system, consequently sensitizing host T cells [57, 58]. These T cells may then be recalled by subsequent exposure to crossreactive streptococcal M protein epitopes. Molecular mimicry between myosin, highly expressed in the myocardium, and lamimin, highly expressed in the valves, may be a plausible hypothesis to explain how a response against myosin would induces valvulitis [57]. Cells infiltrating the heart produce predominately inflammatory cytokines that mediate the lesions. Differently from the myocardium, in which the lesion is reversible, valves show low numbers of cells producing regulatory cytokine IL-4, thus leading to an imbalance in the immune response and consequently exacerbating inflammation and perpetuating valve lesions [55]. ACKNOWLEDGEMENTS This work was supported by grants from the Fundação de Amparo a Pesquisa do Estado de São Paulo (FAPESP) and Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq). REFERENCES [1] Carapetis, J.R. (2004) World Health Organization, [2] Carapetis, J.R., McDonal, M. and Wilson, N.J. (2005) Lancet, 366(9480), [3] Meira, Z.M.A. (1995) Arch. Brasilian Cardiol., 65, [4] Cheadle, W. (1889) Lancet, 371, [5] Guilherme, L., Weidebach, W., Kiss, M.H., Snitcowsky, R. and Kalil, J. (1991) Circulation, 83,

6 44 Current Protein and Peptide Science, 2007, Vol. 8, No. 1 Guilherme et al. [6] Ozkan, M., Carin, M., Sonmez, G., Senocak, M., Ozdemir, M. and Yakut, C. (1993) Circulation, 87, [7] Weidebach, W., Goldberg, A.C., Chiarella, J., Guilherme, L., Snitcowsky, R., Pileggi, F. and Kalil, J. (1994) Hum. Immunol., 40, [8] Guédez, Y., Kotby, A., El-Demellaway, M., Galal, A., Thomson, G., Zaher, S., Kassem, S. and Kotb, M. (1999) Circulation, 99, [9] Stanevecchia, V., Eglite, J., Sochevs, A., Gardovska, D., Zavadska, D. and Shantere,R. (2003) Arth. Res. & Therapy, 5, [10] Ayoub, E.M. (1984) Circulation, 69, [11] Anastasiou-Nana, M., Anderson, J.L., Carquist, J.F. and Nana, J.N. (1986) Am. Heart J., 112, [12] Rajapakse, N.A., Halim, K., Al-Orainey, L., Al-Nozha, M. and Al- Aska, A.K. (1987) Br. Heart J., 58, [13] Bhat, M.S., Wani, B.A., Koul, P.A., Bisati, S.D., Khan, M.A. and Shah, S.U. (1997) Indian J. Med. Res., 105, [14] Monplaisir, N., Valette, I. and Bach, J.F. (1986) Tissue Antigens, 28(4), [15] Maharaj, B., Hammond, M.G., Appadoo, B., Leary, W.P. and Pudifin, D.J. (1987) Circulation, 76, [16] Reddy, K.S., Narula, J., Bathia, R., Shailendri, K., Koicha, M., Taneja, V., Jhingan, B., Pothineni, R.B., Malaviya, A.N. and Mehra, N.K. (1990) Indian J. Pediatr., 57(5), [17] Taneja, V., Mehra, N.K., Reddy, K.S., Narula, J., Tandon, R., Vaidva, M.C. and Bhatia, M.L. (1989) Circulation, 80, [18] Jhinghan, B., Mehra, N.K., Reddy, K.S., Taneja, V., Vaidya, M.C. and Bhatia, M.L. (1986) Tissue Antigens, 27, [19] Koyanagi, T., Koga, Y., Nishi, H., Toshima, H., Sasazuki, T., Imaizumi, T. and Kimura, A. (1996) J. Mol. Cell Cardiol., 28, [20] Hernandez-Pacheco, G., Aguilar-Garcia, J., Flores-Dominguez, C., Rodriguez-Perez, J.M., Perez-Hernandez, N., Alvarez-Leon, E., Reyes, P.A. and Vargas-Alarcon, G. (2003) Inter. J. Cardiol., 92, [21] Chou, H.T., Chen, C.H. and Tsai, F.J. (2004) Am. Heart J., 148(1), [22] Berdeli, A., Celik, H.A. Ozyurek, R. and Aydin, H.H. (2004) Clin. Biochem., 37(10), [23] Hernandez-Pacheco, G., Flores-Dominguez, C., Rodriguez-Perez, J.M., Perez-Hernandez, N., Fragoso, J.M., Saul, A., Alvarez-Leon, E., Granados, J., Reyes, P.A. and Vargas-Alarcon, G. (2003) J. Autoimmun., 21(1), [24] Berdeli, A., Celik, H.A., Ozyurek, R., Dogrusoz, B. and Aydin, H.H. (2005) J. Mol. Med., 83(7), [25] Cunningham, M.W. (2000) Clin. Microbiol. Rev., 13(3), [26] Calveti, P.A. (1945) Proc. Soc. Exp. Biol. Med., 60, [27] Kaplan, M.H. (1963) J. Immunol., 90, [28] Kaplan, M.H., Bolande, R., Rakita, R. and BlaiR, J. (1964) N. Engl. J. M., 271, [29] Dale, J.B. and Beachey, E.H. (1982) Trans. Assoc. Am. Physicians, 95, [30] Baird, R.W., Bronze, M.S., Kraus, W., Hill, H.R., Veasey, L.G., Dale, J.B. (1991) J. Immunol., 146 (9), [31] Cunningham, M.W., Hall, N. K., Krisher, K. K., and Spanier A. M. (1985) J. Immunol., 136, [32] Cunningham, M.W. and Swerlick, R.A. (1986) J. Exp. Med., 164, [33] Krisher, K. and Cunningham, M. W. (1985) Science, 227, [34] Dale, J.B. and Beachey, E. H. (1985) J. Exp. Med., 162, [35] Cunningham, M.W., McCormack, J.M., Fenderson, P.G., Ho, M.K., Beachey, E. H. and Dale J. B. (1989) J. Immunol., 143, [36] Goldstein, I., Halpern, B. and Robert L. (1967) Nature, 213, [37] Galvin, J.E., Hemric, M.E., Ward, K. and Cunningham, M.W. (2000) J. Clin. Invest., 106 (2), [38] Roberts, S., Kosanke, S., Terrence, Dunn, S., Jankelow, D., Duran, C.M. and Cunningham, M.W. (2001) J. Infect. Dis., 183(3), [39] Read, S.E., Zabriskie, J. B., Fischetti, V.A., Utermohlen, V. and Falk R. (1974) J.Clin. Invest, 54, [40] Read, S.E., Reid, H. F., Fischetti, V. A., Poon-King, T., Ramkissoon, R., McDowell, M. and Zabriskie, J. B. (1986) J. Clin. Immunol., 6, [41] Hutto, J. H. and Ayoub, E. M. (1980) in Streptococcal diseases and the immune response (Read, S.E. and Zabriskie, J.B, Eds.). pp , Academic Press, New York, N.Y. [42] Dale, J.B. and Beachey, E.H. (1987) J. Exp. Med., 166, [43] Guilherme, L., Oshiro, S.E., Faé, K.C., Cunha-Neto, E., Renesto, G., Goldberg, A.C., Tanaka, A.C., Pomerantzeff, P.M., Kiss, M.H., Silva, C., Guzman, F., Patarroyo, M.E., Southwood, S., Sette, A. and Kalil, J. (2001) Infect. Immun., 69, [44] Manjula, B.N., Trus, B.L. and Fischetti, V.A. (1985) Proc. Natl. Acad. Sci., USA, 82, [45] Phillips, J.G.N., Flicker, P.F., Cohen, C., Manjula, B.N. and Fischetti, V.A. (1981) Proc. Nat. Acad. Sci., USA, 78, [46] Robinson, J.H., Atherton, M.C., Goodacre, J.A., Pinkney, M., Weightman, H. and Kehoe, M.A. (1991) Infect. Immun., 59(12), [47] Cunningham, M.W., Antone, S.M., Smart, M., Liu, R. and Kosanke, S. (1997) Infect. Immun., 65(9), [48] Guilherme, L., Cunha-Neto, E., Coelho, V., Snitcowsky, R., Pillegi, F. and Kalil, J. (1995) Circulation, 92, [49] Raizada, V., Williams, R.C.J., Chopra, P., Gopinath, N., Praskash, K., Sharma, K.B., Cherian, K.M., Panday, S., Arora, R., Nigam, M., Zabriskie, J.B. and Husby, G. (1983) Am. J. Med., 74, [50] Kemeny, E., Grieve, T., Marcus, R., Sareli, P. and Zabriskie, J.B. (1989) Clin. Immunol. Immunopathol., 52, [51] El-Demellawy, M.R., El-Ridi, N.I., Guirguis, M.A. Kotby, A.A., Kotb, M. (1997). Infect. Immun., 65(6 ), [52] Faé, K., Kalil, J., Toubert, A. and Guilherme, L. (2004) Mol. Immunol., 40, [53] Guilherme, L., Dulphy, N., Douay, C., Coelho, V., Cunha-Neto, E., Oshiro, S.E., Assis, R.V., Tanaka, A.C., Pomerantzeff, P.M.A., Charron, D., Toubert, A. and Kalil, J. (2000) Int. Immunol., 12, [54] Pannetier, C., Even, J. and Kourilsky, P. (1995) Immunol. Today, 16, [55] Guilherme, L., Cury, P., Demarchi, L.M., Coelho, V., Abel, L., Lopez, A.P., Oshiro, S.E., Aliotti, S., Cunha-Neto, E., Pomerantzeff, P.M.A., Tanaka, A.C. and Kalil, J. (2004) Am. J. Pathol., 165(5), [56] Guilherme, L. and Kalil J. (2004) Int. Arch. Allergy. Immunol., 134(1), [57] Cunningham, M.W. (2004) Mol. Immunol., 40, [58] Smith, S.C. and Allen, P.M. (1992) Proc. Natl. Acad. Sci. USA, 89, Received: September 15, 2005 Revised: June 28, 2006 Accepted: July 11, 2006