Mimicry between streptococcal M protein and cardiac myosin is important in the pathogenesis of rheumatic heart disease. M protein-specific human T cell clones derived from rheumatic carditis were cross-reactive with human cardiac myosin, and laminin, a valve protein. Among the 11 CD4+ and CD8+ cross-reactive T cell clones, at least 6 different reactivity patterns were distinguished, suggesting different degrees of cross-reactivity and a very diverse T cell repertoire. The latter was confirmed by a heterogeneous Vβ gene and CDR3 usage. HLA restriction and Th1 cytokine production in response to rM6 protein were preserved when the T cell clones were stimulated by human cardiac myosin or other α-helical proteins, such as tropomyosin and laminin. The cross-reactive human T cell clones proliferated to B2 and B3A, dominant peptide epitopes in the B repeat region of streptococcal M protein. In human cardiac myosin, epitopes were demonstrated in the S2 and light meromyosin regions. In our study, T cell mimicry was defined as recognition of structurally related Ags involved in disease and recognized by the same T cell. Mimicry in our study was related to α-helical coiled coil proteins which have a repetitive seven-aa residue periodicity that maintains α-helical structure and thus creates a high number of degenerate possibilities for recognition by T cells. The study of human T cell clones from rheumatic heart disease revealed potential sites of T cell mimicry between streptococcal M protein and human cardiac myosin and represents some of the most well-defined T cell mimicry in human autoimmune disease.

Molecular mimicry between group A streptococci and heart tissue was first described by Zabriskie (1) and Kaplan (2) and was defined as Ab reactivity directed toward streptococcal and host Ags. Although mimicry between similar or dissimilar molecules has been well established by studies of cross-reactive mAbs in streptococcal sequelae (3, 4, 5, 6), little is known about T cell mimicry in disease. Recognition of host and microbial epitopes by T cells may be an important influence on the development of autoimmune responses in humans (7, 8, 9, 10, 11, 12, 13, 14, 15, 16). Mimicry between host and pathogen may have an important role in maintaining autoimmune T cell responses in the presence of inflammatory cytokines during infection. Cross-reactive T cells from the periphery infiltrate target organs, where they can continue to be activated and proliferate into disease-producing T cells.

Our study of the dual specificity of human T cell clones from rheumatic heart disease provides a unique view of T cell mimicry in a disease where mimicry is a hallmark of the disease. Previously, we investigated human cross-reactive B cell responses against mimicking Ags in rheumatic carditis and found that sera and human mAbs from rheumatic carditis reacted with the major heart autoantigen cardiac myosin (17, 18, 19, 20). The importance of cardiac myosin is related to its ability to produce myocarditis and valvular heart disease in animal models (21, 22, 23). Group A streptococcal M protein, an α-helical coiled coil molecule with high homology to cardiac myosin, induced valvular rheumatic-like heart disease in Lewis rats (24).

In the first study of T cells isolated from rheumatic heart valves, it was shown that cross-reactive epitopes of streptococcal M5 protein and cardiac proteins were recognized by T cell clones from inflamed rheumatic hearts (25). Our study of peripheral T cell clones from rheumatic carditis focuses on their cross-reactive epitope specificity for streptococcal M protein and human cardiac myosin. The T cell clones recognized epitopes in the B repeat region of M protein and in the S2 and light meromyosin (LMM)3 regions of human cardiac myosin. The results demonstrated the presence of recombinant M6 protein-specific peripheral T cell clones from rheumatic carditis which proliferated in response to cardiac myosin as well as other α-helical coiled coil molecules such as tropomyosin and laminin, a protein found in valve tissue. Most importantly, the study demonstrates T cell cross-reactivity between streptococcal M protein and human cardiac myosin and defines the peptide epitope specificity of T cell mimicry in rheumatic heart disease. The hypothesis of mimicry and epitope spreading may explain how cardiac myosin, an intracellular protein in myocardium, can lead to valvular heart disease.

Screening for the HLA haplotype was performed by PCR DNA typing. The HLA profile of the patient was: HLA-A*0101, 02011; B*15011, 3701; C*03031, 0602; DRB1*1301, 15011; DRB3*02021; DRB5*01011; DQA1*0103, 0102; DQB1*0603, 0602; and DPB1*0301, *02012. The likely inherited maternal and paternal HLA class II haplotypes of this patient are: haplotype 1, DRB1*1301-DRB3*02021-DQA1*0103-DQB1*0603; and haplotype 2, DRB1*1501-DRB5*01011-DQA1*0102-DQB1*0602.

Streptococcal recombinant M6 protein (rM6) was purified in the laboratory of Dr. Vincent A. Fischetti (Rockefeller University, New York, NY), and purified human cardiac myosin was prepared in our laboratory according to a procedure described previously (20). Purified tropomyosin from rabbit muscle, skeletal myosin, and laminin were obtained from Sigma-Aldrich. Tetanus toxoid was obtained from Massachusetts Biological Laboratories.

Previously described peptides of the LMM fragment (26) of the human cardiac myosin β-chain rod region and of streptococcal M5 protein (27) were synthesized as 18-mers with 5-aa overlap and were purified by HPLC. Peptides of the S2 fragment of the human cardiac myosin β-chain rod region were synthesized by Genemed Synthesis as 25-mers with 11-aa overlap (28) and were purified by HPLC.

Peripheral blood was obtained from a patient with rheumatic heart disease at the Oklahoma Children’s Heart Center (Oklahoma City, OK). The research protocol was reviewed and approved by the University of Oklahoma Institutional Review Board (Oklahoma City, OK). Rheumatic carditis was identified in the patient with acute rheumatic fever and was based on the revised Jones criteria including an elevated anti-streptolysin O titer of 1250 (29). For production of T cell lines and clones, PBMC were prepared from heparinized venous blood by Ficoll gradient separation. PBMC were plated at 5 × 104 cells/well in U-bottom 96-well plates (Costar) in the presence of 20 μg/ml rM6 at 37°C in 5% CO2. Autologous medium used for cell culture was RPMI 1640 (Invitrogen Life Technologies) supplemented with 10% heat-inactivated autologous human serum and l-glutamine (2 mM), sodium pyruvate (1 mM), nonessential amino acids (0.1 mM), and 10 mM HEPES buffer (Invitrogen Life Technologies). Seven days later, the cultures were restimulated with irradiated (5000 rad) autologous PBMC pulsed with rM6 (105 cells/well) as a source of APCs, and 5 U/ml rIL-2 (Boehringer Mannheim) were added after 48 h. After another 5 days, cultures were tested for specific response to rM6 protein (20 μg/ml) in a [3H]thymidine proliferation assay. [3H]Thymidine (0.5 μCi; Valeant Pharmaceuticals) was added, and 18 h later the T cells were harvested onto membranes (glass fiber filter; Wallac) by a cell harvester (MACH II; Wallac) and subsequently counted in a liquid scintillation counter (Betaplate 1250; Wallac). A T cell line was considered to be reactive to rM6 when the cpm exceeded the medium control cpm by at least three times. Stimulation indices were calculated by dividing the cpm of the test wells by the cpm of medium control wells. All proliferation experiments were performed in duplicate and repeated at least three times. rM6-specific T cell lines were subsequently cloned by limiting dilution at 0.3 cell/well and cultured with 105 irradiated allogeneic PBMC and 2 μg/ml PHA-protein. After 48 h, 5 U/ml rIL-2 were added, and cultures were subsequently fed every 3 days with fresh autologous medium and 5 U/ml rIL-2. After ∼10–12 days, growth-positive wells were tested for specific reactivity to rM6 and other α-helical proteins in the [3H]thymidine proliferation assay. Human T cell clones were selected by dual recognition of rM6 and human cardiac myosin, laminin, or tropomyosin. Clones were sorted using V-β TCR-specific Ab, as described under FACS, to achieve 100% homogeneous T cell populations.

Mouse monoclonal anti-CD3-, anti-CD4-, anti-CD8-, anti-αβ TCR-, anti-γδ TCR-, and anti-Vβ TCR-specific Abs (Vβ1-Vβ2 -Vβ3-Vβ5.1-Vβ5.2-Vβ5.3-Vβ6.7-Vβ7-Vβ8-Vβ9-Vβ11-Vβ12.2-Vβ13.1-Vβ14-Vβ16-Vβ17-Vβ18-Vβ20-Vβ21.3-Vβ22-Vβ23-Vβ24) conjugated with fluorescein or PE (Serotec) were used for flow cytometry analysis. Analysis was performed on a FACScan flow cytometer using CellQuest software (BD Biosciences).

Total RNA was extracted from 2 × 106 cells using the RNeasy mini kit (Qiagen). Five micrograms of total RNA served as template for first-strand cDNA synthesis using random hexamers, and the Superscript Preamplification system (Invitrogen Life Technologies). Subsequently, the cDNA was subject to PCR amplification with a primer specific for one of the 24 Vβ gene families (as determined by flow cytometry as described above) as the forward primer, and a Cβ gene-specific primer as the reverse primer (30). The amplified PCR products were separated in a 1% agarose gel by electrophoresis and stained with ethidium bromide. The visualized PCR products were cut and purified using a QIA Quick Gel extraction kit (Qiagen). Purified PCR products (10 ng/100 bp) were sequenced with the Applied Biosystems Big dye terminator kit using the 377 Applied Biosystems sequencer. Sequence analysis and identity searches were performed using the software package of the Genetics Computer Group of the University of Wisconsin, release 9.1, GenBank database, and the BLAST program (31).

For analysis of HLA restriction pattern, 104 rM6-specific T cells and 105 irradiated autologous APC were cultured in the absence or presence of dialyzed anti-HLA class I and class II Abs (10 μg/ml) in rM6-containing culture medium for 3 days. Inhibition of cell proliferation was measured by [3H]thymidine incorporation assays. Murine mAbs anti-HLA-DR (L243), anti-HLA-DP (B7/21), and anti-HLA-DQ (TÜ169) against human MHC class II framework determinants as well as anti-HLA class I Ab (G46-2.6) were all purchased from BD Biosciences.

Groups of 104 cells of each rM6-specific T cell clone were cultured with irradiated autologous PBMC (105 cells/well) in the presence of 10-μg/ml quantities of the indicated synthetic peptides or 20-μg/ml quantities of the α-helical proteins for 72 h. Cultures were performed in duplicate for each Ag and control wells (medium alone) to examine fine specificities of the T cell clones. In all cases, cell proliferation was measured by [3H]thymidine incorporation assays as described above and by the IFN-γ ELISA as described below.

For analysis of cytokine profile, 104 cells of each rM6-reactive T cell clone were cultured with irradiated autologous PBMC (105 cells/well) in the presence and absence of Ag for 48 h. The resulting supernatant was used for cytokine detection using the Cyto Sets kit for IFN-γ and IL-4 (Biosource International). With an ELISA reader/spectrophotometer (Dynatech MR 700) set at 450 nm, the cytokine levels were detected in comparison with known amounts of cytokine standards.

ELISPOT plates (Immunospot M200; Cellular Technology) were precoated with human IFN-γ (Endogen) capture Ab (10 μg/ml) in 0.1 M carbonate coating buffer (pH 9.5, 50 μl/well) and placed at 4°C overnight. Ab-coated ELISPOT plates were washed three times with 200 μl/well sterile PBS. Plates were blocked to prevent nonspecific binding of proteins using 200 μl/well sterile PBS containing 10% autologous serum for 1 h at room temperature and then washed again three times with 200 μl/well sterile PBS and 1 time with autologous serum. To the plates precoated with IFN-γ capture Ab, 300 cells/well of the T cell clones and 105 irradiated autologous APCs were transferred. Ags were added to the wells in increasing concentrations, and the cells were incubated for 24 h at 37°C in 5% CO2. Cells were then washed away with eight washes of 100 μl/well ice cold PBS, 0.05% Tween 20. Secondary biotinylated anti-IFN-γ Ab was diluted in PBS at 5 μg/ml with 10% autologous serum, added at 50 μl/well, and incubated overnight at 4°C. Plates were then washed eight times with 200 μl/well PBS-Tween 20, and a 1/2000 dilution of streptavidin-HRP in PBS with 10% autologous serum was added at 100 μl/well for 2 h at room temperature. Plates were washed eight times with 200 μl/well PBS, and spot color was developed by adding 100 μl/well Aminoethylcarbazole substrate diluted 1/30 in 0.1 M acetate buffer (pH 5.0) containing a 1/2000 dilution of 30% H2O2. Plates were observed for spot development for a maximum of 1 h at room temperature and then were washed three times with dH2O (200 μl/well) to stop the reaction. Plates were dried overnight at room temperature. Images of the wells were acquired and saved on compact disc using an automated ImmunoSpot Series 3, and the spots were enumerated on an Immunospot Satellite Analyzer (Cellular Technology) using software specifically designed for ELISPOT. Positive responses were defined as two or more adjacent Ag-stimulated wells giving responses greater than the mean ± 3 SD of unstimulated wells.

Peripheral human T cell clones responsive to group A rM6 were derived from a rheumatic carditis patient and selected for dual recognition of rM6 and human cardiac myosin, laminin, or tropomyosin. Thirteen rM6-responsive CD4+ and four CD8+ cross-reactive T cell clones were produced, comprising 23% of the total number of rM6-reactive T cell clones. For further analysis, we selected 2 non-cross-reactive clones and 11 cross-reactive clones. Three CD8+ and 10 CD4+ rM6-reactive T cell clones were sorted using Vβ TCR-specific Ab to achieve 99–100% homogeneous populations. Table I shows the cross-reactive proliferative responses of these 13 T cell clones to rM6 and several host α-helical proteins including human cardiac myosin and its heavy meromyosin (HMM) subfragment, laminin, and tropomyosin. Seven (54%) rM6-reactive T cell clones recognized human cardiac myosin, but only one clone responded in proliferation assays to the human cardiac myosin HMM subfragment. Several of the T cell clones recognized laminin (8 of 13 clones) and tropomyosin (7 of 13 clones), α-helical proteins that have been shown previously to be recognized by cross-reactive anti-myosin/anti-streptococcal Abs (32, 33). The two non-cross-reactive clones, G4.1 (CD4+) and 4G5-10.6 (CD8+), were rM6 specific but not cross-reactive (Table I). For all cross-reactive T cell clones produced, none responded in proliferation assays above the medium control values to tetanus toxoid, a globular protein and control Ag used throughout our studies. Among the 11 cross-reactive T cell clones, at least 6 different reactivity patterns for recognition of rM6, human cardiac myosin, human cardiac HMM, laminin and tropomyosin were distinguished (Table II), suggesting a very diverse T cell repertoire. The increase in cross-reactivity of the T cell clones was proportionate to the increase in the stimulation index, as shown in Fig. 1.

Table I.

Cross-reactivity of rM6-reactive T cell clones with human cardiac myosin and other α-helical proteinsa

T Cell CloneMedium (cpm)α-Helical Proteins (cpm)
rM6MyosinHMMLamininTropomyosin
CD4+       
 C8-13.1 421 ± 40 15,601 ± 523 5,234 ± 643 715 ± 71 5,488 ± 590 640 ± 16 
 3D4-3.3 370 ± 21 18,754 ± 925 395 ± 207 458 ± 38 5,830 ± 198 2,582 ± 139 
 3E11-1.2 556 ± 17 6,878 ± 405 475 ± 63 495 ± 35 510 ± 72 1,861 ± 142 
 3E11-10.1 488 ± 41 30,567 ± 451 3,029 ± 107 953 ± 57 1,473 ± 62 634 ± 76 
 F3s 777 ± 192 9,880 ± 556 632 ± 24 839 ± 22 465 ± 104 1,736 ± 121 
 F7-3.2 962 ± 73 4,829 ± 241 5,642 ± 431 543 ± 108 3,241 ± 186 1,896 ± 97 
 F7-3.5 894 ± 102 3,325 ± 233 3,462 ± 280 2,751 ± 213 3,700 ± 264 3,990 ± 142 
 G4.1 441 ± 106 5,183 ± 344 285 ± 82 464 ± 110 414 ± 63 189 ± 117 
 G4s 511 ± 104 23,781 ± 1,265 14,520 ± 2970 643 ± 99 16,328 ± 894 8,646 ± 607 
 3G8-1.10 544 ± 76 3,520 ± 164 2,076 ± 150 607 ± 101 569 ± 70 486 ± 92 
CD8+       
 3E11-1.6 216 ± 28 5,375 ± 342 582 ± 75 322 ± 74 1,284 ± 160 3,025 ± 187 
 4G5-10.6 197 ± 66 7,863 ± 275 686 ± 142 326 ± 62 539 ± 127 189 ± 84 
 3G8-3.1 299 ± 32 4,036 ± 320 1,004 ± 58 538 ± 134 1,084 ± 215 378 ± 97 
T Cell CloneMedium (cpm)α-Helical Proteins (cpm)
rM6MyosinHMMLamininTropomyosin
CD4+       
 C8-13.1 421 ± 40 15,601 ± 523 5,234 ± 643 715 ± 71 5,488 ± 590 640 ± 16 
 3D4-3.3 370 ± 21 18,754 ± 925 395 ± 207 458 ± 38 5,830 ± 198 2,582 ± 139 
 3E11-1.2 556 ± 17 6,878 ± 405 475 ± 63 495 ± 35 510 ± 72 1,861 ± 142 
 3E11-10.1 488 ± 41 30,567 ± 451 3,029 ± 107 953 ± 57 1,473 ± 62 634 ± 76 
 F3s 777 ± 192 9,880 ± 556 632 ± 24 839 ± 22 465 ± 104 1,736 ± 121 
 F7-3.2 962 ± 73 4,829 ± 241 5,642 ± 431 543 ± 108 3,241 ± 186 1,896 ± 97 
 F7-3.5 894 ± 102 3,325 ± 233 3,462 ± 280 2,751 ± 213 3,700 ± 264 3,990 ± 142 
 G4.1 441 ± 106 5,183 ± 344 285 ± 82 464 ± 110 414 ± 63 189 ± 117 
 G4s 511 ± 104 23,781 ± 1,265 14,520 ± 2970 643 ± 99 16,328 ± 894 8,646 ± 607 
 3G8-1.10 544 ± 76 3,520 ± 164 2,076 ± 150 607 ± 101 569 ± 70 486 ± 92 
CD8+       
 3E11-1.6 216 ± 28 5,375 ± 342 582 ± 75 322 ± 74 1,284 ± 160 3,025 ± 187 
 4G5-10.6 197 ± 66 7,863 ± 275 686 ± 142 326 ± 62 539 ± 127 189 ± 84 
 3G8-3.1 299 ± 32 4,036 ± 320 1,004 ± 58 538 ± 134 1,084 ± 215 378 ± 97 
a

rM6-reactive T cell clones were generated from the blood of a patient with previous rheumatic heart disease. The reactivity of the clones toward human cardiac myosin and other α-helical proteins is measured via a [3H]thymidine incorporation experiment, expressed in cpm, and compared with control cpm (in the absence of Ags). Bold numbers represent positive stimulations with values ≥3 times the control cpm. Underlined numbers represent cpm that are considered positive but lower than the cutoff value set at 3 times the control cpm.

Table II.

Overview of the characteristics of the rM6-reactive T-cell clonesa

T Cell ClonePhenotypeCross-reactivity ProfileFine SpecificityVβ UsageHLA Restriction
4G5-10.6 CD8 rM6 B1A Vβ5.1 rM6 HLA class I 
G4.1 CD4 rM6 B1B2,B3A Vβ17 rM6 HLA-DQ 
3G8-1.10 CD4 rM6, myob B2,B3A Vβ17 rM6 HLA-DR 
3E11-1.2 CD4 rM6,tropo B2,B3A,S2–4 Vβ5.1 rM6, tropo HLA-DR 
F3s CD4 rM6, tropo B2,B3A None detected ND 
C8-13.1 CD4 rM6, myo, lam  Vβ13.1 rM6, lam HLA-DR 
3G8-3.1 CD8 rM6, myo, lam B2,B2B3B None detected rM6 HLA class I 
3E11-10.1 CD4 rM6, myo, lam S2–16 Vβ5.3 rM6, myo, lam HLA-DQ 
3E11-1.6 CD8 rM6, lam, tropo  Vβ21 rM6 HLA class I 
3D4-3.3 CD4 rM6, lam, tropo B2 None detected rM6, lam, tropo HLA-DQ 
F7-3.2 CD4 rM6, myo, lam, tropo  Vβ5.1 rM6, myo, lam HLA-DR 
G4s CD4 rM6, myo, lam, tropo B3A Vβ2 rM6, myo, lam HLA-DR 
F7-3.5 CD4 rM6, myo, lam, HMM, tropo  Vβ2 rM6 HLA-DR 
T Cell ClonePhenotypeCross-reactivity ProfileFine SpecificityVβ UsageHLA Restriction
4G5-10.6 CD8 rM6 B1A Vβ5.1 rM6 HLA class I 
G4.1 CD4 rM6 B1B2,B3A Vβ17 rM6 HLA-DQ 
3G8-1.10 CD4 rM6, myob B2,B3A Vβ17 rM6 HLA-DR 
3E11-1.2 CD4 rM6,tropo B2,B3A,S2–4 Vβ5.1 rM6, tropo HLA-DR 
F3s CD4 rM6, tropo B2,B3A None detected ND 
C8-13.1 CD4 rM6, myo, lam  Vβ13.1 rM6, lam HLA-DR 
3G8-3.1 CD8 rM6, myo, lam B2,B2B3B None detected rM6 HLA class I 
3E11-10.1 CD4 rM6, myo, lam S2–16 Vβ5.3 rM6, myo, lam HLA-DQ 
3E11-1.6 CD8 rM6, lam, tropo  Vβ21 rM6 HLA class I 
3D4-3.3 CD4 rM6, lam, tropo B2 None detected rM6, lam, tropo HLA-DQ 
F7-3.2 CD4 rM6, myo, lam, tropo  Vβ5.1 rM6, myo, lam HLA-DR 
G4s CD4 rM6, myo, lam, tropo B3A Vβ2 rM6, myo, lam HLA-DR 
F7-3.5 CD4 rM6, myo, lam, HMM, tropo  Vβ2 rM6 HLA-DR 
a

The clones are grouped for similar reactivity profiles towards rM6 protein, human cardiac myosin, laminin and tropomyosin. The Vβ gene of F3s and 3D4–3.3 could not be detected via anti-Vβ Abs in flow cytometry or via PCR with Vβ gene primers; the Vβ gene of 3G8–3.1 was not detected with the available anti-Vβ Abs.

b

Lam, laminin; tropo, tropomyosin; myo, human cardiac myosin; ND, not determined.

FIGURE 1.

Cross-reactivity of T cell clones was proportionate to the stimulation index of streptococcal rM6 protein. Cross-reactivity of T cell clones G4s and 3D4-3.3 was measured via [3H]thymidine uptake in cpm and compared with the stimulation index of rM6 protein. With the increase in stimulation index shown for rM6 protein, the clones demonstrated increased cross-reactivity with human cardiac myosin, laminin, or tropomyosin.

FIGURE 1.

Cross-reactivity of T cell clones was proportionate to the stimulation index of streptococcal rM6 protein. Cross-reactivity of T cell clones G4s and 3D4-3.3 was measured via [3H]thymidine uptake in cpm and compared with the stimulation index of rM6 protein. With the increase in stimulation index shown for rM6 protein, the clones demonstrated increased cross-reactivity with human cardiac myosin, laminin, or tropomyosin.

Close modal

Using flow cytometry and a panel of 22 anti-Vβ TCR-specific Abs, we analyzed the Vβ gene usage of the rM6-reactive T cell clones. Table II shows the Vβ gene usages and how they correspond with different cross-reactivity patterns. For three of the clones, we could not determine the Vβ gene usage with the given Abs. One CD8+ and 2 CD4+ rM6-reactive T cell clones used Vβ5.1, but each of these clones expressed a different cross-reactivity pattern (Tables I and II). In addition, two clones that expressed Vβ2 (G4s and F7-3.5) and 2 clones that expressed Vβ17 (G4.1 and 3G8-1.10) each expressed different cross-reactivity patterns. Junctional Vβ sequences of selected rM6-reactive T cell clones were determined by using TCRVβ family-specific PCR followed by direct sequencing (Table III). The lack of specific sequence similarities is in accordance with the broadly diverse T cell repertoire that was observed for the T cell clones.

Table III.

V-(D)-J junctional region sequences of Vβ chains expressed by selected rM6 protein-reactive T cell clonesa

T Cell CloneVβ-JβCDR3 Sequence
3E11-10.1 Vβ5.3-Jβ1.4 YLCASSSGDSTNEKLFFG 
3E11-1.2 Vβ5.1-Jβ1.1 YLCASSDGQGEAFFG 
F7-3.2 Vβ5.1-Jβ2.1 YLCASSSGTSGNNEQFFG 
4G5-10.6 Vβ5.1-Jβ2.3 YLCASSLTYGPYEQYFG 
F7-3.5 Vβ2.3-Jβ1.4 YICSAKRNEKLFFG 
G4s Vβ2.3-Jβ2.5 YICSALDAGGETQYFG 
T Cell CloneVβ-JβCDR3 Sequence
3E11-10.1 Vβ5.3-Jβ1.4 YLCASSSGDSTNEKLFFG 
3E11-1.2 Vβ5.1-Jβ1.1 YLCASSDGQGEAFFG 
F7-3.2 Vβ5.1-Jβ2.1 YLCASSSGTSGNNEQFFG 
4G5-10.6 Vβ5.1-Jβ2.3 YLCASSLTYGPYEQYFG 
F7-3.5 Vβ2.3-Jβ1.4 YICSAKRNEKLFFG 
G4s Vβ2.3-Jβ2.5 YICSALDAGGETQYFG 
a

For each clone, the deduced amino acid sequence of the CDR3 loop is shown. GenBank accession numbers are: AY848703 for the TCR β-chain mRNA of T cell clone 3 E11-10.1, AY848704 for clone 3E11-1.2 TCR β-chain, AY848705 for clone F7-3.2 TCR β-chain, AY848706 for clone F7-3.5 TCR β-chain, AY848707 for clone 4G5-10.6 TCR β-chain, AY848708 for clone G4s TCR β-chain mRNA.

Table II shows that the cross-reactive CD4+ response was MHC II restricted to DR (6 of 9 clones) or DQ (3 of 9 clones), whereas the cross-reactive CD8+ response was MHC I restricted. For 6 cross-reactive T cell clones, the HLA restriction was studied for the different cross-reactive Ags. Fig. 2 shows DR-restricted clone G4s and DQ restricted clone 3D4-3.3. HLA restriction was preserved among the different cross-reactive Ags.

FIGURE 2.

MHC restriction of cross-reactive streptococcal M protein-specific T cell clones. The figure represents a typical blocking experiment using 10 μg/ml monoclonal anti-HLA class I, anti-HLA class II, anti-HLA-DR, anti-DP, and anti-DQ Abs added at the initiation of the cultures. T cell proliferation assays were performed for cross-reactive Ags rM6 protein, human cardiac myosin, laminin, and tropomyosin and measured via [3H]thymidine incorporation assays. T cell clone G4s was DR restricted for the different cross-reactive Ags, whereas T cell clone 3D4-3.3 was DQ restricted.

FIGURE 2.

MHC restriction of cross-reactive streptococcal M protein-specific T cell clones. The figure represents a typical blocking experiment using 10 μg/ml monoclonal anti-HLA class I, anti-HLA class II, anti-HLA-DR, anti-DP, and anti-DQ Abs added at the initiation of the cultures. T cell proliferation assays were performed for cross-reactive Ags rM6 protein, human cardiac myosin, laminin, and tropomyosin and measured via [3H]thymidine incorporation assays. T cell clone G4s was DR restricted for the different cross-reactive Ags, whereas T cell clone 3D4-3.3 was DQ restricted.

Close modal

To study the dose-response reactivity of the cross-reactive T cell clones toward the different cross-reactive Ags, IFN-γ-release was monitored for six cross-reactive T cell clones (G4s, C8-13.1, 3E11-1.2, 3E11-10.1, F7-3.5, and 3G8-1.10) via the ELISPOT while varying the concentration of Ag. Fig. 3 shows a representative dose-response experiment for clone G4s. The evidence demonstrated that streptococcal rM6 protein was substantially more effective in stimulating the cross-reactive T cell clones than human cardiac myosin. This 100-fold difference between rM6 and myosin stimulation was also observed for T cell clones 3G8-1.10, F7-3.5, 3E11-10.1, and C8-13.1. Clone 3E11-1.2 demonstrated a 1000-fold difference between rM6 and tropomyosin stimulation. T cell clone F7-3.5 was analogous to clone G4s and demonstrated a reactivity pattern of rM6 protein≫human cardiac myosin>laminin and tropomyosin. T cell clones 3E11-10.1 and C8-13.1 also had a comparable pattern (rM6≫human cardiac myosin>laminin). For all clones tested, the response to rM6 outweighed the response toward the other cross-reactive Ags by a factor of ∼100 or ∼1000, respectively. The response to human cardiac myosin was 10-fold greater than the other cross-reactive α-helical coiled coil proteins laminin and tropomyosin. In addition and not unexpectedly, there was also a response to skeletal myosin similar to that demonstrated by other α-helical coiled coil proteins (data not shown).

FIGURE 3.

Dose response of cross-reactive T cell clone G4s. A cross-reactive Ag dose-dependent IFN-γ response curve is shown for 1000 cells/well G4s cells at 24 h after stimulation with rM6 protein, human cardiac myosin, laminin, and tropomyosin. The response was greatest for rM6 protein, which was ∼100-fold > human cardiac myosin. The cardiac myosin response was 10-fold > laminin and tropomyosin for T cell clone G4s. Other T cell clones had a similar responsive profile to M protein and cardiac myosin.

FIGURE 3.

Dose response of cross-reactive T cell clone G4s. A cross-reactive Ag dose-dependent IFN-γ response curve is shown for 1000 cells/well G4s cells at 24 h after stimulation with rM6 protein, human cardiac myosin, laminin, and tropomyosin. The response was greatest for rM6 protein, which was ∼100-fold > human cardiac myosin. The cardiac myosin response was 10-fold > laminin and tropomyosin for T cell clone G4s. Other T cell clones had a similar responsive profile to M protein and cardiac myosin.

Close modal

Cross-reactive T cell clones F3s and G4s were further tested for production of IL-2, IFN-γ, IL-4, TNF-α, IL-10, and TGF-β cytokine responses toward the different cross-reactive Ags (Table IV). Both T cell clones produced IFN-γ and TNF-α, and the production of these cytokines was consistent for the different α-helical cross-reactive Ags to which these clones responded in proliferation assays. Levels of IFN-γ and TNF-α cytokine expression correlated very well with degrees of proliferation as measured via [3H]thymidine incorporation. Tetanus toxoid as a control Ag did not induce cytokine production above the medium control from any of our clones.

Table IV.

IL-2, IFN-γ, IL-4, TNFα, IL-10, and TGF-β cytokine responses of 2 cross-reactive T cell clones toward different cross-reactive Agsa

T Cell CloneAg[3H]Thymidine (cpm)SIIL-2 (pg/ml)IFN-γ (pg/ml)IL-4 (pg/ml)TNF-α (pg/ml)IL-10 (pg/ml)TGF-β (pg/ml)
F3s Medium 764        
 rM6 7,880 12.9 58 2,000  5,000 30  
 myo 492        
 Laminin 345   37  350   
 Tropomyosin 1,833 2.27  230  370   
 B2 4,098 5.36  1,000  1,200 38 28 
 B3A 5,450 7.13 21 1,850  1,990  24 
G4s Medium 502        
 rM6 23,001 45.8 72 5,730  8,030   
 HCM 7,640 15.2 23 997  860   
 Laminin 6,023 11.9 18 562  740   
 Tropomyosin 4,297 8.6 17 630  677   
 B2 967     30   
 B3A 16,740 33.3 102 6,028  9,940   
T Cell CloneAg[3H]Thymidine (cpm)SIIL-2 (pg/ml)IFN-γ (pg/ml)IL-4 (pg/ml)TNF-α (pg/ml)IL-10 (pg/ml)TGF-β (pg/ml)
F3s Medium 764        
 rM6 7,880 12.9 58 2,000  5,000 30  
 myo 492        
 Laminin 345   37  350   
 Tropomyosin 1,833 2.27  230  370   
 B2 4,098 5.36  1,000  1,200 38 28 
 B3A 5,450 7.13 21 1,850  1,990  24 
G4s Medium 502        
 rM6 23,001 45.8 72 5,730  8,030   
 HCM 7,640 15.2 23 997  860   
 Laminin 6,023 11.9 18 562  740   
 Tropomyosin 4,297 8.6 17 630  677   
 B2 967     30   
 B3A 16,740 33.3 102 6,028  9,940   
a

Cross-reactive T cell clones F3s and G4s were stimulated with different cross-reactive Ags and IL-2, IFN-γ, IL-4, TNFα, IL-10, and TGF-β cytokine responses were measured for each of the Ags. [3H]Thymidine incorporation was performed simultaneously to verify the proliferation via another method. The stimulation index (SI) of the different Ags corresponded with the quantity of IFN-γ and TNF-α secretions of the T cell clones. Bold numbers highlight the significant stimulations. The myo abbreviation was used for human cardiac myosin. The sequence of streptococcal M5 peptide B2 is TIGTLKKILDETVKDKIA and of peptide B3A is IGTLKKILDETVKDKLAK.

Overlapping peptides of group A streptococcal M5 protein, which share a large amount of homology with streptococcal M6 protein, were reacted with eight of the cross-reactive T cell clones to determine epitope specificity using cytokine (IFN-γ) response assays (Table V). The α-helical coiled coil M protein structure is characteristically divided into three regions based on similarity of amino acid sequence repeats. The A repeat region represents the N-terminal one-third of the molecule that determines serotype specificity and is the most variable among the different M serotypes. The B repeat region of M proteins is the central part and less variable among M types than the A repeat region. The amino acid sequence within the B repeat region is significantly homologous between serotype M5 and M6 proteins (54% identity) (34). The C repeat region amino acid sequence is highly conserved among all M proteins, is ∼80% identical between M5 and M6 (34), and comprises the region of the M protein molecule before insertion into the peptidoglycan wall-membrane region.

Table V.

Streptococcal M protein epitope specificity and IFN-γ expression of rM6 protein cross-reactive T cell clonesa

T Cell CloneMedium (pg/ml)rM6B1AB1BB1B2B2B2B3AB3AB2B3BB3B
Non-cross-reactive T cell clones           
 4G5–10.6 85* 3200 1450 45 72 46 38 42 65 54 
 G4.1 73 5644 47 39 1003 88 26 2602 78 90 
Cross-reactive T cell clones           
 3G8–3.1 53 1756 51 40 39 2239 72 52 2600 48 
 3G8–1.10 77 3607 81 26 52 4532 29 2904 45 92 
 3D4–3.3 74 4976 37 62 85 4673 45 62 25 67 
 3E11–1.2 55 3826 72 23 24 2300 58 4201 65 71 
 F3s 40 4427 67 52 53 2760 64 2510 53 62 
 G4s 39 1742 27 43 39 44 26 1907 41 33 
T Cell CloneMedium (pg/ml)rM6B1AB1BB1B2B2B2B3AB3AB2B3BB3B
Non-cross-reactive T cell clones           
 4G5–10.6 85* 3200 1450 45 72 46 38 42 65 54 
 G4.1 73 5644 47 39 1003 88 26 2602 78 90 
Cross-reactive T cell clones           
 3G8–3.1 53 1756 51 40 39 2239 72 52 2600 48 
 3G8–1.10 77 3607 81 26 52 4532 29 2904 45 92 
 3D4–3.3 74 4976 37 62 85 4673 45 62 25 67 
 3E11–1.2 55 3826 72 23 24 2300 58 4201 65 71 
 F3s 40 4427 67 52 53 2760 64 2510 53 62 
 G4s 39 1742 27 43 39 44 26 1907 41 33 
a

The reactivity of the clones toward streptococcal M5 peptides was measured in the ELISA and expressed as picograms of IFN-γ per milliliter. A and C repeat peptide reactivity was measured but not shown here, because the reactivity was located in the B repeat region of streptococcal M protein. B repeat sequences are B1A = TRQELANKQQESKENEKAL (aa 111–129), B1B = ENEKALNELLEKTVKDKI (aa 124–141), B1B2 = VKDKIAKEQENKETIGTL (aa 137–154), B2 = TIGTLKKILDETVKDKIA (aa 151–167), B2B3A = KDKIAKEQENKETIGTLK (aa 163–180), B3A = IGTLKKILDETVKDKLAK (aa 176–193), B2B3B = DKLAKEQKSKQNIGALKQ (aa 189–206), B3B = GALKQELAKKDEANKISD (aa 202–219). Bold numbers represent positive IFN-γ values ≥3 times the background value (∗).

Table V shows that the cross-reactive T cell clones responded to epitopes in the B repeat region of streptococcal M protein. In some cases, the response toward the B repeat peptide was stronger than the response toward streptococcal intact rM6 protein (e.g., T cell clone 3G8-3.1). M protein peptides B2 (aa 151–167) and B3A (aa 176–193) (Table V), which differ from each other only by 3 aa, were the dominant peptides recognized. B2 as well as B3A were each recognized by five T cell clones. However, three T cell clones recognized both B2 and B3A. T cell clone F3s recognized B2 and B3A equally, whereas clone 3G8-1.10 responded better to peptide B2 than to B3A and clone 3E11-1.2 recognized B3A better than B2. Other B repeat epitopes recognized were B1A (aa 111–129) (non-cross-reactive T cell clone 4G5-10.6), B1B2 (aa 137–154) (non-cross-reactive T cell clone G4.1) and B2B3B (aa 189–206) (clone 3G8-3.1). Comparison of M5 peptide sequences with M6 protein B repeat sequences revealed that M5 peptide B2 was 100% identical with M6 protein whereas B1B2 was 94% identical. Less identity was observed with B3A at 88% identity and B2B3B at 83% identity with M6 protein; however, the amino acid substitutions between M5 and M6 in this region are highly conserved.

No significant cross-reactive responses were detected in these clones for the A and C repeat regions (data not shown). However, 6 of 11 of the original rM6-reactive T cell lines recognized the C repeat region at the closely related C1A and C2C3 peptide sequences (data not shown). The positive responses of T cell lines to the C1A and C2C3 M protein peptides indicated that, as expected, there was a T cell response against the C repeat region, but this response was not apparently cross-reactive with human cardiac myosin because we did not observe it after the dual selection for rM6 and human cardiac myosin.

The specificity of the peripheral human T cell clones for peptide epitopes in human cardiac myosin was determined for the S2 and LMM regions via the ELISPOT. T cell clone 3E11-1.2 recognized S2–4 (aa 43–67) and to a lesser degree LMM 35 (aa 1740–1758) in addition to the B repeat peptides B2 and B3A. B2 and B3A have 50% homology with S2–4 and 71% homology with LMM 35. T cell clone 3E11-10.1 recognized S2–16 (aa 201–225) (Table VI). Four other T cell clones were studied, but only weak responses were obtained. The weaker responses of the T cell clones for the human cardiac myosin peptides were in accordance with the lower sensitivity of the clones for human cardiac myosin than for streptococcal M protein. No response was observed in ELISPOTs against the control Ag tetanus toxoid.

Table VI.

ELISPOT responses of the rM6-reactive T cell clones toward streptococcal M5 protein B repeat and human cardiac myosin peptidesa

T Cell CloneMedium (no. of spots/well)rM6MyosinB2B3AS2–4S2–16LMM-35
3E11–1.2 4 ± 1.3 314 ± 6.4 7 ± 1.9 289 ± 28.4 322 ± 89.6 37 ± 9.4 3 ± 2.8 15 ± 11.2 
3E11–10.1 6 ± 2.0 348 ± 12.4 21 ± 4.8 5 ± 3.6 2 ± 0.8 1 ± 1.4 28 ± 10.4 7 ± 2.9 
T Cell CloneMedium (no. of spots/well)rM6MyosinB2B3AS2–4S2–16LMM-35
3E11–1.2 4 ± 1.3 314 ± 6.4 7 ± 1.9 289 ± 28.4 322 ± 89.6 37 ± 9.4 3 ± 2.8 15 ± 11.2 
3E11–10.1 6 ± 2.0 348 ± 12.4 21 ± 4.8 5 ± 3.6 2 ± 0.8 1 ± 1.4 28 ± 10.4 7 ± 2.9 
a

Cross-reactive T cell clones were plated at 300 cells/well. The fine specificity of the clones toward streptococcal M5 protein and human cardiac myosin peptides was measured and expressed as number of IFN-γ-producing spots per well. Control responses were measured toward rM6 protein, human cardiac myosin, and medium. Positive responses in bold were at least 3 times the medium control.

Molecular mimicry has been established as recognition of homologous structures or amino acid sequences and has been primarily based on Ab cross-reactivity (3, 4, 5). Molecular mimicry by Ab is a well-known concept (6, 35, 36), and Ab is known to recognize highly diverse Ags with similar chemical structures (4, 6, 37, 38). However, few studies have addressed T cell mimicry and degeneracy in disease (9, 10). Recognition of Ag by the T cell is different from that of Ab because T cells recognize peptide Ag fragments by the TCR and HLA molecules (39, 40). Because only a few contact residues are required for recognition simultaneously by the TCR and HLA molecule and there are a limited number of positions that 20 aa can assume in proteins, it is a natural phenomenon that unrelated protein Ags are recognized by the same T cell (41). The result is T cell degeneracy which is defined as the ability of T cells to cross-react with unrelated Ags lacking sequence homology due to the presence of similar contact residues for TCR and HLA molecules (9, 10). The importance of degeneracy may be to compensate for the inability of the TCR to undergo somatic mutation against foreign Ags. Although mimicry and degeneracy can be associated with disease, cross-reactive T cells are part of the normal immune repertoire (10). In our study, T cell mimicry was defined as recognition of structurally related Ags involved in the disease process and recognized by the same T cell. The mimicry in our studies involved α-helical coiled coil proteins that have a repetitive 7-aa residue periodicity that maintains an α-helical structure and creates a high number of degenerate possibilities. Reaction of Ag with an Ab or T cell does not necessarily mean that it is the cause of disease. In the case of skeletal myosin, it does not cause disease in animals, whereas cardiac myosin produces fulminant myocarditis and valvulitis in rats and mice (21, 22). Cross-reactive T cells entering the valve in rheumatic carditis lead to chronic inflammation and potentially to epitope spreading and recognition of epitopes within the valve perpetuating heart disease. Epitope spreading has been defined in animal models as recognition of epitopes distinct from and non-cross-reactive with the disease-inducing epitope (42).

In rheumatic heart disease, it is well established that T cells infiltrate heart valves (25, 43, 44, 45, 46), and valve damage appears to be T lymphocyte dependent. In support of cross-reactive T cells in valve lesions, T cells from valves of rheumatic fever patients have been shown to proliferate in response to peptides of streptococcal M5 protein and heart tissue Ags (25). The T cells in the valve infiltrate are believed to be the effectors of tissue damage (45) and would originate from T cells in peripheral blood after their contact with the infectious pathogen (46). It is our hypothesis that the T cells that infiltrate the valve are reactive with the streptococcal M protein and its cardiac myosin-like peptide sequences. The study of how mimicry between M protein and cardiac myosin at the T cell level may produce valvular heart disease is important to fully understand the pathogenesis of rheumatic heart disease.

Our study of cross-reactive peripheral T cell clones from rheumatic heart disease provides an in depth investigation and detailed analysis of the cross-reactivity observed in T cells captured from human disease. Using a dual selection method, we were able to identify and study T cell clones that were cross-reactive with rM6 and human cardiac myosin. The T cell clones all recognized epitopes in the B repeat region of the streptococcal M protein but not the C repeat region, which was recognized by a large proportion of the original T cell lines. The recognition of the C repeat region by a large number of T cell lines is in accordance with the fact that all M protein serotypes share a large region of homology in the C repeat region of the molecule (34, 47, 48). M protein serotypes M5 and M6 share homology within the B repeat region (49, 50), which is the basis for the sharing of epitopes in the B repeat region of M5 and M6 proteins. Consequently, M6-protein-specific T cell clones proliferated to B repeat peptides of M5 protein. In contrast, the A repeat region is the most distinct among M protein serotypes which explains why M5 peptides from the A repeat serotype-specific region were not recognized by the rM6-reactive T cell clones. The B repeat region is not part of any group A streptococcal vaccine that is currently under development.

The recognition of M5 peptides B2 and B3A by T cells from rheumatic heart disease is consistent with previous reports (25, 46), although it remains to be determined which of the class II HLA molecules in this patient are presenting streptococcal peptides B2 and B3A to T cell clones cross-reactive for human myosin. A review of the literature fails to identify a strong association between class II HLA and rheumatic heart disease, but previous studies do indicate that a number of class II HLA alleles found in this patient are not associated with rheumatic heart disease. The haplotype DRB1*1301-DRB3*02021-DQA1*0103-DQB1*0603 is reportedly not associated with rheumatic heart disease (51), whereas alleles DQA*0102 and DQB1*0602 on the second haplotype are also reported as not associated with rheumatic heart disease (52, 53). This process of elimination leaves DRB1*1501 and DRB5*01011 as the most probable candidates for presenting peptides B2 and B3A to cross-reactive T cell clones. A review of peptide ligands and motifs bound by DRB1*1501 and DRB5*01011 (http://HLAligand.ouhsc.edu) confirms that the C-terminal 11 aa in B3A would bind DRB5*01011 and that the N-terminal 11 aa of B3A and B2 (starting with Ile) would bind DRB1*1501. In the future, we intend to embark upon epitope-mapping studies to test the hypothesis that DRB1*1501 and DRB5*01011 present streptococcal peptides B2 and B3A to cross-reactive T cell clones.

When we investigated the degeneracy of the cross-reactive T cell responses toward different α-helical proteins such as human cardiac myosin, laminin, tropomyosin, and streptococcal M protein, we observed a mosaic of different responses with at least six distinct α-helical protein patterns of response demonstrating different degrees of cross-reactivity. Our results indicate that multiple Ags are involved in autoimmunity and cross-reactivity as a result of mimicry and perhaps subsequent epitope spreading. Previously, TCR recognition has been shown to be highly degenerate, and investigation of the requirements for TCR recognition of MHC-peptide complexes has revealed highly degenerate peptide binding motifs (54). A recent study demonstrated how conformational changes during TCR engagement may allow T cell cross-reactivity (55). Although the mimicry and cross-reactivity observed indicated that the degeneracy involved α-helical molecules, the globular protein tetanus toxoid did not induce a T cell response from any of the clones. In addition, certain epitopes within the α-helical coiled coil structures were specific because most of the 107 peptides tested against the T cell clones did not induce a response. If the reaction with host tissue α-helical coiled coil molecules was high affinity, one would expect thymic deletion of such clones. It is therefore not surprising that reactivity of the T cell clones with the host α-helical molecules is less than with streptococcal M protein. It is possible that the cross-reactive T cells isolated from rheumatic valves expand upon recognition of cross-reactive Ag in the valve of which laminin is only one example. Fae et al. (56) do demonstrate oligoclonal expanded T cells in rheumatic valves. Although it is possible that our cross-reactive T cell clones may also react with other nonrelated Ags such as reported by Wucherpfennig (7), the evidence of multiple cross-reactivities with α-helical proteins within the heart/valve may indicate a pathologically meaningful molecular mimicry. High density expression of myosin-like α-helical coiled coil Ags on professional and nonprofessional APCs in the valve may result in ongoing activation of the potentially pathogenic T cells, IFN-γ production, and subsequent scarring of the heart valve. Experiments by Riberdy et al. (57), Anderton et al. (58), and Alexander-Miller et al. (59) demonstrated that the immune response can adapt to the level of Ag in the environment. High local concentrations of IFN-γ may be required for up-regulation of MHC class II molecules on nearby APCs in the valve (43, 44, 60). Receptor tuning (61), a process by which the activation threshold of a T cell can be altered, may be another way of generating a cross-reactive TCR with higher affinity for the self Ag cardiac myosin. Hemmer et al. (62) postulated that the MHC strongly influences the TCR/Ag recognition and that low affinity self Ags that give a full spectrum stimulation of a T cell clone, analogous to the high affinity infectious Ag, may be able to do so because the TCR affinity for the MHC is close to the threshold of activation. Certain low affinity peptides may be able to increase the TCR/Ag-MHC affinity enough to achieve full activation (62).

When we further determined the fine specificity of the T cell clones for peptides of the LMM and S2 subfragment of human cardiac myosin, potential sites of mimicry were revealed in human cardiac myosin peptides S2–4, S2–16, and LMM 35 and streptococcal M protein peptides B2 and B3A. Clone 3E11-1.2 recognized the M protein peptides B2 and B3A and cardiac myosin peptides S2–4 and LMM 35 (Table VI). It is probable that the cross-reactive peptides B2, B3A, S2–4, and LMM 35 share homology at the contact residues and trigger the TCR of clone 3E11-1.2. Cross-reactive clone 3E11-1.10 recognized the human cardiac myosin epitope S2–16, an epitope shown to produce severe myocarditis in the Lewis rat (28).

Dose responses toward the different cross-reactive Ags were studied for seven cross-reactive T cell clones and showed an approximate 100-fold difference in sensitivity between streptococcal M6 protein and human cardiac myosin and a 10-fold greater sensitivity for human cardiac myosin than either tropomyosin or laminin, a protein present in valve tissue. The cross-reactive T cell profile observed for clones with high specificity for streptococcal M6 protein can be explained by a degenerate response toward α-helical structural domains that have repetitive sequence patterns shared by proteins such as myosin, laminin, and tropomyosin. Likewise, cross-reactive antigenic peptides may have homologous amino acids at certain positions to preserve a full spectrum T cell response. α-Helical proteins preserve their α-helical structure and seven amino acid residue periodicity which depends on conserved amino acid substitutions. Effects on the heart by T cells cross-reactive to streptococcal M protein and cardiac myosin may be due initially to the fact that cross-reactive Abs likely target the valve endothelium and up-regulate VCAM-1 (45). Entrance of T cells into the avascular valve lead to their cross-recognition of valvular proteins such as laminin through mimicry which may eventually lead to epitope spreading. We would have preferred to compare the T cell clones from PBL with those in the valve of the same patient but it was not possible. Studies by Fae et al. (56) and Guilherme (63, 64) show recognition of streptococcal M protein and cardiac proteins by valve T cell clones, demonstrating their importance in disease and presence in the target organ. However, there are no studies yet investigating the level and specificities of T cells against myosin in disease vs nondisease. These answers are best obtained from surveys of T cells present in the blood of nondisease vs disease. We are planning ELISPOT studies in animals and humans to delineate the T cell repertoire and dominant myosin epitopes. Normal individuals have been shown to harbor autoreactive T cells in their blood (65, 66). One might expect that regulatory T cells would be important in controlling the autoreactive T cells in normal individuals. Single clones, as described in our study from an individual with rheumatic heart disease, can only address the mimicry hypothesis and cannot adequately answer questions about the T cell repertoire in patients with and without disease.

Although the T cell clones in our paper were selected on the basis of their reactivity with streptococcal M protein, we believe that myosin-reactive clones may enter the heart because when we produced T cell lines from hearts of Lewis rats developing cardiac myosin-induced myocarditis, these lines were found to respond to streptococcal M protein peptides in proliferation assays, suggesting that T cells against myosin taken from inflamed hearts reacted with streptococcal M protein. These animal studies suggest that myosin reciprocally induces myosin-reactive T cells in the heart which may cross-react with M protein (21). Although pathogenic T cells may initially be cross-reactive, chronic inflammation in the target organ may lead to the eventual epitope spreading to proteins present in the valve (42).

Several group A streptococcal proteins have been reported to be superantigens, raising the possibility that autoreactive T cells could be driven or amplified by superantigens. Vβ1, -2, -3, -5.1, -8, and -14 have been related to group A streptococcal superantigens (67, 68). Although several of these Vβ genes were expressed, a heterogeneous Vβ gene usage was observed. Guilherme et al. (63, 64) examined the relative frequency of TCR Vβ families in PBMC from six patients with severe rheumatic heart disease and in six heart-infiltrating T cell lines and clones derived from four of these patients. Although several oligoclonal T cell expansions were found, no shared Vβ usages were observed among the patients. Also, Abbott et al. (69) observed no difference in the Vβ repertoire of peripheral blood T cells in 9 children with acute rheumatic fever vs 34 controls. However, studies by Figueroa et al. (70, 71) indicate that a superantigen-driven inflammatory process may be involved in some instances. Our data suggest that a superantigen-driven response was not observed in our cross-reactive T cell repertoire.

Although we cannot be certain that our clones are pathogenic, we do know that the streptococcal M6 protein-responsive T cell clones produced the Th1 cytokine IFN-γ upon stimulation with rM6 protein, and the Th1 response was preserved upon stimulation with human cardiac myosin. These data are supported by induction of IFN-γ by the cross-reactive peptide epitopes responsible for the mimicry in both M protein and human cardiac myosin. Previous studies have demonstrated IFN-γ in valve tissues in rheumatic carditis (72) as well as by peripheral T cell clones isolated from rheumatic patients and stimulated with heart and streptococcal Ags (25, 46, 72). Effector T cells which mediate disease in experimental models of autoimmune disease typically have, with a few exceptions, a Th1 proinflammatory phenotype. The streptococcal M6-specific T cell clones in our study demonstrated a Th1 phenotype, and Th1 cytokines were secreted upon stimulation with the host autoantigen human cardiac myosin. Th1 clonotypes would be expected to generate valve scarring in rheumatic heart disease.

In summary, we believe that streptococcal M protein/human cardiac myosin cross-reactive T cells play an important role in the pathogenesis of rheumatic heart disease. In streptococcal disease, chronic and enhanced stimulation of the immune system occurs through repetitive pharyngitis in children, leading to mimicry that develops in the periphery. After activation of valvular endothelium, mimicking T cells migrate into the avascular valve through the endocardium and are stimulated by local valve Ags. In the valve, the T cells secrete IFN-γ, leading to Th1-mediated granuloma formation in the form of Aschoff nodules, whereas T cells continue to infiltrate the valve through neovascularized regions even years after acute disease (70).

We thank the blood donor and his family involved in this study. We are grateful to Dr. Mark Hemric for human cardiac myosin preparation, to Dr. Caroline Thompson for nucleotide sequence analysis, and to Dr. David Sidebottom for HLA determination.

The authors have no financial conflict of interest.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1

This work was supported by National Institutes of Health Grants HL56267 and HL35280 from the National Heart, Lung, and Blood Institute. M.W.C. is the recipient of a National Heart, Lung, and Blood Institute Merit Award.

3

Abbreviations used in this paper: HMM, heavy meromyosin; LMM, light meromyosin; rM6, recombinant streptococcal M6 protein.

1
Zabriskie, J. B..
1967
. Mimetic relationships between group A streptococci and mammalian tissues.
Adv. Immunol.
7
:
147
.-188.
2
Kaplan, H. A..
1963
. Anatomy of the cerebrovascular system.
Clin. Neurosurg.
9
:
47
.-55.
3
Cunningham, M. W..
2000
. Pathogenesis of group A streptococcal infections.
Clin. Microbiol. Rev.
13
:
470
.-511.
4
Shikhman, A. R., M. W. Cunningham.
1994
. Immunological mimicry between N-acetyl-β-d-glucosamine and cytokeratin peptides: evidence for a microbially driven anti-keratin antibody response.
J. Immunol.
152
:
4375
.-4387.
5
Galvin, J. E., M. E. Hemric, K. Ward, M. W. Cunningham.
2000
. Cytotoxic mAb from rheumatic carditis recognizes heart valves and laminin.
J. Clin. Invest.
106
:
217
.-224.
6
Kirvan, C. A., S. E. Swedo, J. S. Heuser, M. W. Cunningham.
2003
. Mimicry and autoantibody-mediated neuronal cell signaling in Sydenham chorea.
Nat. Med.
9
:
914
.-920.
7
Wucherpfennig, K. W., J. L. Strominger.
1995
. Molecular mimicry in T cell-mediated autoimmunity: viral peptides activate human T cell clones specific for myelin basic protein.
Cell
80
:
695
.-705.
8
Fujinami, R. S., M. B. Oldstone.
1985
. Amino acid homology between the encephalitogenic site of myelin basic protein and virus: mechanism for autoimmunity.
Science
230
:
1043
.-1045.
9
Hemmer, B., B. T. Fleckenstein, M. Vergelli, G. Jung, H. McFarland, R. Martin, K. H. Wiesmuller.
1997
. Identification of high potency microbial and self ligands for a human autoreactive class II-restricted T cell clone.
J. Exp. Med.
185
:
1651
.-1659.
10
Gran, B., B. Hemmer, M. Vergelli, H. F. McFarland, R. Martin.
1999
. Molecular mimicry and multiple sclerosis: degenerate T-cell recognition and the induction of autoimmunity.
Ann. Neurol.
45
:
559
.-567.
11
Croxford, J. L., J. K. Olson, S. D. Miller.
2002
. Epitope spreading and molecular mimicry as triggers of autoimmunity in the Theiler’s virus-induced demyelinating disease model of multiple sclerosis.
Autoimmun. Rev.
1
:
251
.-260.
12
Tejada-Simon, M. V., Y. C. Zang, J. Hong, V. M. Rivera, J. Z. Zhang.
2003
. Cross-reactivity with myelin basic protein and human herpesvirus-6 in multiple sclerosis.
Ann. Neurol.
53
:
189
.-197.
13
Zhao, Z. S., F. Granucci, L. Yeh, P. A. Schaffer, H. Cantor.
1998
. Molecular mimicry by herpes simplex virus-type 1: autoimmune disease after viral infection.
Science
279
:
1344
.-1347.
14
Panoutsakopoulou, V., M. E. Sanchirico, K. M. Huster, M. Jansson, F. Granucci, D. J. Shim, K. W. Wucherpfennig, H. Cantor.
2001
. Analysis of the relationship between viral infection and autoimmune disease.
Immunity
15
:
137
.-147.
15
Bachmaier, K., N. Neu, L. M. de la Maza, S. Pal, A. Hessel, J. M. Penninger.
1999
. Chlamydia infections and heart disease linked through antigenic mimicry.
Science
283
:
1335
.-1339.
16
Croxford, J. L., H. A. Anger, S. D. Miller.
2005
. Viral delivery of an epitope from Haemophilus influenzae induces central nervous system autoimmune disease by molecular mimicry.
J. Immunol.
174
:
907
.-917.
17
Cunningham, M. W., J. M. McCormack, L. R. Talaber, J. B. Harley, E. M. Ayoub, R. S. Muneer, L. T. Chun, D. V. Reddy.
1988
. Human monoclonal antibodies reactive with antigens of the group A Streptococcus and human heart.
J. Immunol.
141
:
2760
.-2766.
18
Cunningham, M. W., J. M. McCormack, P. G. Fenderson, M. K. Ho, E. H. Beachey, J. B. Dale.
1989
. Human and murine antibodies cross-reactive with streptococcal M protein and myosin recognize the sequence GLN-LYS-SER-LYS-GLN in M protein.
J. Immunol.
143
:
2677
.-2683.
19
Cunningham, M. W., H. C. Meissner, J. Heuser, D. Y. M. Leung.
1999
. Anti-human cardiac myosin autoantibodies in Kawasaki syndrome.
J. Immunol.
163
:
1060
.-1065.
20
Galvin, J. E., M. E. Hemric, K. Ward, M. W. Cunningham.
2000
. Cytotoxic monoclonal antibody from rheumatic carditis reacts with human endothelium: implications in rheumatic heart disease.
J. Clin. Invest.
106
:
217
.-224.
21
Galvin, J. E., M. E. Hemric, S. D. Kosanke, S. M. Factor, A. Quinn, M. W. Cunningham.
2002
. Induction of myocarditis and valvulitis in Lewis rats by different epitopes of cardiac myosin and its implications in rheumatic carditis.
Am. J. Pathol.
160
:
297
.-306.
22
Neu, N., N. R. Rose, K. W. Beisel, A. Herskowitz, G. Gurri-Glass, S. W. Craig.
1987
. Cardiac myosin induces myocarditis in genetically predisposed mice.
J. Immunol.
139
:
3630
.-3636.
23
Smith, S. C., P. M. Allen.
1991
. Myosin-induced acute myocarditis is a T cell-mediated disease.
J. Immunol.
147
:
2141
.-2147.
24
Quinn, A., S. Kosanke, V. A. Fischetti, S. M. Factor, M. W. Cunningham.
2001
. Induction of autoimmune valvular heart disease by recombinant streptococcal M protein.
Infect. Immun.
69
:
4072
.-4078.
25
Guilherme, L., E. Cunha-Neto, V. Coelho, R. Snitcowsky, P. M. Pomerantzeff, R. V. Assis, F. Pedra, J. Neumann, A. Goldberg, M. E. Patarroyo.
1995
. Human heart-infiltrating T-cell clones from rheumatic heart disease patients recognize both streptococcal and cardiac proteins.
Circulation
92
:
415
.-420.
26
Quinn, A., E. E. Adderson, P. G. Shackelford, W. L. Carroll, M. W. Cunningham.
1995
. Autoantibody germ-line gene segment encodes VH and VL regions of a human anti-streptococcal monoclonal antibody recognizing streptococcal M protein and human cardiac myosin epitopes.
J. Immunol.
154
:
4203
.-4212.
27
Cunningham, M. W., S. M. Antone, M. Smart, R. Liu, S. Kosanke.
1997
. Molecular analysis of human cardiac myosin-cross-reactive B- and T-cell epitopes of the group A streptococcal M5 protein.
Infect. Immun.
65
:
3913
.-3923.
28
Li, Y., J. S. Heuser, S. D. Kosanke, M. Hemric, M. W. Cunningham.
2004
. Cryptic epitope identified in rat and human cardiac myosin S2 region induces myocarditis in the Lewis rat.
J. Immunol.
172
:
3225
.-3234.
29
Special Writing Group of the Committee on Rheumatic Fever, Endocarditis, and Kawasaki Disease of the Council on Cardiovascular Disease in the Young of the American Heart Association.
1992
. Guidelines for the diagnosis of rheumatic fever.Jones Criteria, 1992 update.
JAMA
268
:
2069
.-2073.
30
Vandevyver, C., N. Mertens, P. van den Elsen, R. Medaer, J. Raus, J. Zhang.
1995
. Clonal expansion of myelin basic protein-reactive T cells in patients with multiple sclerosis: restricted T cell receptor V gene rearrangements and CDR3 sequence.
Eur. J. Immunol.
25
:
958
.-968.
31
Altschul, S. F., W. Gish, W. Miller, E. W. Myers, D. J. Lipman.
1990
. Basic local alignment search tool.
J. Mol. Biol.
215
:
403
.-410.
32
Fenderson, P. G., V. A. Fischetti, M. W. Cunningham.
1989
. Tropomyosin shares immunologic epitopes with group A streptococcal M proteins.
J. Immunol.
142
:
2475
.-2481.
33
Antone, S. M., E. E. Adderson, N. M. J. Mertens, M. W. Cunningham.
1997
. Molecular analysis of V gene sequences encoding cytotoxic anti-streptococcal/anti-myosin monoclonal antibody 36.2.2 that recognizes the heart cell surface protein laminin.
J. Immunol.
159
:
5422
.-5430.
34
Fischetti, V. A..
1989
. Streptococcal M protein: molecular design and biological behavior.
Clin. Microbiol. Rev.
2
:
285
.-314.
35
Oldstone, M. B..
1998
. Molecular mimicry and immune-mediated diseases.
FASEB J.
12
:
1255
.-1265.
36
Barnett, L. A., R. S. Fujinami.
1992
. Molecular mimicry: a mechanism for autoimmune injury.
FASEB J.
6
:
840
.-844.
37
Shikhman, A. R., N. S. Greenspan, M. W. Cunningham.
1993
. A subset of mouse monoclonal antibodies cross-reactive with cytoskeletal proteins and group A streptococcal M proteins recognizes N-acetyl-β-d-glucosamine.
J. Immunol.
151
:
3902
.-3913.
38
Shikhman, A. R., N. S. Greenspan, M. W. Cunningham.
1994
. Cytokeratin peptide SFGSGFGGGY mimics N-acetyl-β-d-glucosamine in reaction with antibodies and lectins, and induces in vivo anti-carbohydrate antibody response.
J. Immunol.
153
:
5593
.-5606.
39
Zinkernagel, R. M., P. C. Doherty.
1974
. Restriction of in vitro T cell-mediated cytotoxicity in lymphocytic choriomeningitis within a syngeneic or semiallogeneic system.
Nature
248
:
701
.-702.
40
Germain, R. N..
1994
. MHC-dependent antigen processing and peptide presentation: providing ligands for T lymphocyte activation.
Cell
76
:
287
.-299.
41
Wucherpfennig, K. W., J. L. Strominger.
1995
. Molecular mimicry in T cell-mediated autoimmunity: viral peptides activate human T cell clones specific for myelin basic protein.
Cell
80
:
695
.-705.
42
McMahon, E. J., S. L. Bailey, C. V. Castenada, H. Waldner, S. D. Miller.
2005
. Epitope spreading initiates in the CNS in two mouse models of multiple sclerosis.
Nat. Med.
11
:
335
.-339.
43
Kemeny, E., T. Grieve, R. Marcus, P. Sareli, J. B. Zabriskie.
1989
. Identification of mononuclear cells and T cell subsets in rheumatic valvulitis.
Clin. Immunol. Immunopathol.
52
:
225
.-237.
44
Raizada, V., R. C. Williams, Jr, P. Chopra, N. Gopinath, K. Prakash, K. B. Sharma, K. M. Cherian, S. Panday, R. Arora, M. Nigam, J. B. Zabriskie, G. Husby.
1983
. Tissue distribution of lymphocytes in rheumatic heart valves as defined by monoclonal anti-T cell antibodies.
Am. J. Med.
74
:
90
.-96.
45
Roberts, S., S. Kosanke, S. Terrence Dunn, D. Jankelow, C. M. Duran, M. W. Cunningham.
2001
. Pathogenic mechanisms in rheumatic carditis: focus on valvular endothelium.
J. Infect. Dis.
183
:
507
.-511.
46
Guilherme, L., S. E. Oshiro, K. C. Fae, E. Cunha-Neto, G. Renesto, A. C. Goldberg, A. C. Tanaka, P. M. Pomerantzeff, M. H. Kiss, et al
2001
. T-cell reactivity against streptococcal antigens in the periphery mirrors reactivity of heart-infiltrating T lymphocytes in rheumatic heart disease patients.
Infect. Immun.
69
:
5345
.-5351.
47
Fischetti, V. A..
1991
. Streptococcal M protein.
Sci. Am.
264
:
58
.-65.
48
Bessen, D., K. F. Jones, V. A. Fischetti.
1989
. Evidence for two distinct classes of streptococcal M protein and their relationship to rheumatic fever.
J. Exp. Med.
169
:
269
.-283.
49
Jones, K. F., B. N. Manjula, K. H. Johnston, S. K. Hollingshead, J. R. Scott, V. A. Fischetti.
1985
. Location of variable and conserved epitopes among the multiple serotypes of streptococcal M protein.
J. Exp. Med.
161
:
623
.-628.
50
Jones, K. F., S. A. Khan, B. W. Erickson, S. K. Hollingshead, J. R. Scott, V. A. Fischetti.
1986
. Immunochemical localization and amino acid sequences of cross-reactive epitopes within the group A streptococcal M6 protein.
J. Exp. Med.
164
:
1226
.-1238.
51
Guedez, Y., A. Kotby, M. El-Demellawy, A. Galal, G. Thomson, S. Zaher, S. Kassem, M. Kotb.
1999
. HLA class II associations with rheumatic heart disease are more evident and consistent among clinically homogeneous patients.
Circulation
99
:
2784
.-2790.
52
Gu, J., B. Yu, J. Zhou.
1997
. HLA-DQA1 genes involved in genetic susceptibilty to rheumatic fever and rheumatic heart disease in southern Hans.
Zhonghua Nei Ke Za Zhi
36
:
308
.-311.
53
Stanevicha, V., J. Eglite, A. Sochnevs, D. Gardovska, D. Zavadska, R. Shantere.
2003
. HLA class II associations with rheumatic heart disease among clinically homogeneous patients in children in Latvia.
Arthritis Res. Ther.
5
:
R340
.-R346.
54
Hammer, J., P. Valsasnini, K. Tolba, D. Bolin, J. Higelin, B. Takacs, F. Sinigaglia.
1993
. Promiscuous and allele-specific anchors in HLA-DR-binding peptides.
Cell
74
:
197
.-203.
55
Lee, J. K., G. Stewart-Jones, T. Dong, K. Harlos, K. Di Gleria, L. Dorrell, D. C. Douek, P. A. van der Merwe, E. Y. Jones, A. J. McMichael.
2004
. T cell cross-reactivity and conformational changes during TCR engagement.
J. Exp. Med.
200
:
1455
.-1466.
56
Fae, K. C., S. E. Oshiro, A. Toubert, D. Charron, J. Kalil, L. Guilherme.
2005
. How an autoimmune reaction triggered by molecular mimicry between streptococcal M protein and cardiac tissue proteins leads to heart lesions in rheumatic heart disease.
J. Autoimmun.
24
:
101
.-109.
57
Riberdy, J. M., A. Zirkel, S. Surman, J. L. Hurwitz, P. C. Doherty.
2001
. Cutting edge: culture with high doses of viral peptide induces previously unprimed CD8+ T cells to produce cytokine.
J. Immunol.
167
:
2437
.-2440.
58
Anderton, S. M., C. G. Radu, P. A. Lowrey, E. S. Ward, D. C. Wraith.
2001
. Negative selection during the peripheral immune response to antigen.
J. Exp. Med.
193
:
1
.-11.
59
Alexander-Miller, M. A., G. R. Leggatt, J. A. Berzofsky.
1996
. Selective expansion of high- or low-avidity cytotoxic T lymphocytes and efficacy for adoptive immunotherapy.
Proc. Natl. Acad. Sci. USA
93
:
4102
.-4107.
60
Steeg, P. S., R. N. Moore, H. M. Johnson, J. J. Oppenheim.
1982
. Regulation of murine macrophage Ia antigen expression by a lymphokine with immune interferon activity.
J. Exp. Med.
156
:
1780
.-1793.
61
Grossman, Z., W. E. Paul.
2001
. Autoreactivity, dynamic tuning and selectivity.
Curr. Opin. Immunol.
13
:
687
.-698.
62
Hemmer, B., M. Vergelli, C. Pinilla, R. Houghten, R. Martin.
1998
. Probing degeneracy in T-cell recognition using peptide combinatorial libraries.
Immunol. Today
19
:
163
.-168.
63
Guilherme, L., N. Dulphy, C. Douay, V. Coelho, E. Cunha-Neto, S. E. Oshiro, R. V. Assis, A. C. Tanaka, P. M. Pomerantzeff, D. Charron, A. Toubert, J. Kalil.
2000
. Molecular evidence for antigen-driven immune responses in cardiac lesions of rheumatic heart disease patients.
Int. Immunol.
12
:
1063
.-1074.
64
Guilherme, L., E. Cunha-Neto, A. C. Tanaka, N. Dulphy, A. Toubert, J. Kalil.
2001
. Heart-directed autoimmunity: the case of rheumatic fever.
J. Autoimmun.
16
:
363
.-367.
65
Tejada-Simon, M. V., J. Hong, V. M. Rivera, J. Z. Zhang.
2001
. Reactivity pattern and cytokine profile of T cells primed by myelin peptides in multiple sclerosis and healthy individuals.
Eur. J. Immunol.
31
:
907
.-917.
66
Oliveira, E. M., A. Bar-Or, A. I. Waliszewska, G. Cai, D. E. Anderson, J. I. Krieger, D. A. Hafler.
2003
. CTLA-4 dysregulation in the activation of myelin basic protein reactive T cells may distinguish patients with multiple sclerosis from healthy controls.
J. Autoimmun.
20
:
71
.-81.
67
Watanabe-Ohnishi, R., J. Aelion, L. LeGros, M. A. Tomai, E. V. Sokurenko, D. Newton, J. Takahara, S. Irino, S. Rashed, M. Kotb.
1994
. Characterization of unique human TCR Vβ specificities for a family of streptococcal superantigens represented by rheumatogenic serotypes of M protein.
J. Immunol.
152
:
2066
.-2073.
68
Tomai, M. A., P. M. Schlievert, M. Kotb.
1992
. Distinct T-cell receptor Vβ gene usage by human T lymphocytes stimulated with the streptococcal pyrogenic exotoxins and pep M5 protein.
Infect. Immun.
60
:
701
.-705.
69
Abbott, W. G., M. A. Skinner, L. Voss, D. Lennon, P. L. Tan, J. D. Fraser, I. J. Simpson, R. Ameratunga, A. Geursen.
1996
. Repertoire of transcribed peripheral blood T-cell receptor β chain variable-region genes in acute rheumatic fever.
Infect. Immun.
64
:
2842
.-2845.
70
Figueroa, F., M. Gonzalez, F. Carrion, C. Lobos, F. Turner, N. Lasagna, F. Valdes.
2002
. Restriction in the usage of variable β regions in T-cells infiltrating valvular tissue from rheumatic heart disease patients.
J. Autoimmun.
19
:
233
.-240.
71
Carrión, F., M. Fernandez, M. Iruretagoyena, L. E. Coelho Andrade, M. Odete-Hilario, F. Figueroa.
2003
. Selective depletion of Vβ2+CD8+ T cells in peripheral blood from rheumatic heart disease patients.
J. Autoimmun.
20
:
183
.-190.
72
Guilherme, L., P. Cury, L. M. F. Demarchi, V. Coelho, L. Abel, A. P. Lopez, S. E. Oshiro, S. Aliotti, E. Cunha-Neto, P. M. A. Pomerantzeff, A. C. Tanaka, J. Kalil.
2004
. Rheumatic heart disease: proinflammatory cytokines play a role in the progression and maintenance of valvular lesions.
Am. J. Pathol.
165
:
1583
.-1591.