To investigate the role of γδ T cells in human autoimmune disease we expressed and characterized a γδ TCR from an autoimmune tissue lesion. The TCR was first identified in a rare form of polymyositis characterized by a monoclonal infiltrate of γδ T cells which invaded and destroyed skeletal muscle fibers. The Vγ1.3-Jγ1-Cγ1/Vδ2-Jδ3 TCR cDNA of the original muscle invasive γδ T cell clone was reconstructed from unrelated cDNA and transfected into the mouse hybridoma BW58αβ. Appropriate anti-human γδ TCR Abs stimulated the TCR transfectants to produce IL-2, thus demonstrating that the human γδ TCR functionally interacted with murine signaling components. The transfected Vγ1.3/Vδ2 TCR recognized a cytosolic protein expressed in cultured human myoblasts and TE671 rhabdomyosarcoma cells. The Ag was recognized in the absence of presenting cells. Using a panel of control γδ TCR transfectants with defined exchanges in different positions of both TCR chains, we showed that the γδ TCR recognized its Ag in a TCR complementarity-determining region 3-dependent way. To our knowledge, this is the first example of a molecularly defined γδ TCR directly derived from an autoimmune tissue lesion. The strategy used in this study may be applicable to other autoimmune diseases.

Some autoimmune diseases are thought to be mediated by T cells, but it must still be established if these T cells actually recognize an autoantigen. The indirect evidence for a pathogenic role of T cells includes the histological demonstration of T cell infiltrates in affected organs, combined with molecular evidence of oligoclonally expanded TCR repertoires. However, a more direct demonstration of T cell autoreactivity would require 1) suggestive morphological evidence, e.g., direct cell-cell contact with target cells plus evidence of cytotoxic effector function; and 2) the identification of both TCR chains from single infiltrating T cells (or members of an expanded clone), followed by the functional expression and characterization of their TCRs.

To our knowledge, these two postulates have not yet been met by any human autoimmune disorder. For example, TCR-α and β chains have been cloned from whole polyclonal infiltrates of multiple sclerosis brain lesions (1), but the T cells from which these sequences were derived were not characterized morphologically, nor was it possible to relate the corresponding TCR-α and β chains in these polyclonal infiltrates. Very recently, single-cell PCR techniques have allowed the amplification of rearranged TCR genes from morphologically defined T cells, but thus far, it has only been possible to identify the TCR-β chains from genomic DNA of single infiltrating T cells (2).

We report on the reconstruction and functional expression of the TCR of a γδ T cell clone first identified in a rare form of polymyositis (3). Our paradigm offers two significant advantages that help meet the above postulates. First, there is strong morphological evidence for a direct pathogenic contribution of the γδ T cells, and second, the putatively autoaggressive γδ T cells are essentially monoclonal, thus allowing the identification of both chains of their TCR.

The form of polymyositis in our patient was characterized by severe mononuclear infiltrates in skeletal muscle tissue. Numerous muscle fibers were surrounded and deeply invaded by CD4CD8 γδ T cells, which exclusively expressed an unusual Vγ1.3 and Vδ2 combination (Fig. 1; Ref. 4). In the present study, we reconstructed the complete sequences of both receptor chains of the muscle invasive γδ T cell clone and functionally expressed them in a mouse lymphoma line lacking intrinsic TCR expression. The Vγ1.3/Vδ2 receptor recognized a muscle-associated protein in a complementarity-determining region (CDR)33-dependent, MHC-nonrestricted way.

FIGURE 1.

Immunohistochemistry of a muscle biopsy of a polymyositis patient (adapted from Ref. 4 ). a, Localization of CD3 Ag with red rhodamine fluorescence and of the Vδ2 Ag with green FITC fluorescence as double exposure. CD3+Vδ2+ double-positive γδ T cells appear in yellow. CD3+Vδ2 single positive αβ T cells appear in red. b, Schematic representation of a. The big cells are muscle fibers; the small, bold-lined, gray cells are CD3+Vδ2+ γδ T cells; and the small, thin-lined, white cells are CD3+Vδ2 αβ T cells. The cells that surround and invade muscle fibers exclusively belong to the CD3+Vδ2+ γδ T cell population, whereas the CD3+ αβ T cells are bystanders. Molecular characterization of the TCR CDR3 region of the muscle invasive T cells revealed that this population was essentially monoclonal (4 ).

FIGURE 1.

Immunohistochemistry of a muscle biopsy of a polymyositis patient (adapted from Ref. 4 ). a, Localization of CD3 Ag with red rhodamine fluorescence and of the Vδ2 Ag with green FITC fluorescence as double exposure. CD3+Vδ2+ double-positive γδ T cells appear in yellow. CD3+Vδ2 single positive αβ T cells appear in red. b, Schematic representation of a. The big cells are muscle fibers; the small, bold-lined, gray cells are CD3+Vδ2+ γδ T cells; and the small, thin-lined, white cells are CD3+Vδ2 αβ T cells. The cells that surround and invade muscle fibers exclusively belong to the CD3+Vδ2+ γδ T cell population, whereas the CD3+ αβ T cells are bystanders. Molecular characterization of the TCR CDR3 region of the muscle invasive T cells revealed that this population was essentially monoclonal (4 ).

Close modal

TE671 human rhabdomyosarcoma and LTK mouse fibroblasts were obtained from American Type Culture Collection (Manassas, VA). The following Abs were used (all anti-TCR Abs were directed against human molecules): anti-pan-γδ, IMMU510; anti-Vγ1.3, 23D12; anti-Vδ2, IMMU389; anti-Vγ9, IMMU360; anti-Vδ1, R9.12; and anti-Vβ9, FIN9 were from Immunotech (Marseille, France); anti-mouse-CD3ε, 145-2C11; and anti-CD56 (neural cell adhesion molecule), MY31 were from BD PharMingen (San Diego, CA). As secondary Ab we used the dichlorotriazinyl-fluoresceine (DTAF)-labeled goat-anti-mouse F(ab′)2 115-016-068 (Dianova, Hamburg, Germany). The myelin basic protein (MBP) peptide MBP139–151 (KGFKGVDAQGTLS) was synthesized by F-moc chemistry and purified by HPLC. Integrity was verified by mass spectrometry. Polymyositis and connective tissue disease-associated Ags, which normally are used for Ab detection in patient sera, were obtained from Varelisa ReCombi kit (Pharmacia & Upjohn, Freiburg, Germany). Titin was kindly donated by S. Labeit (European Molecular Biology Laboratory, Heidelberg, Germany).

In Table I, the sequences of the previously described γ- and δ-chains (Vγ1.3-Jγ1-Cγ1/Vδ2-Jδ3-Cδ; nomenclature used in this study Vγ1.3+Vδ2+) from the muscle invasive TCR clone (4) and control chains are listed (the γ-chain, now designated Vγ1.3, has formerly been designated as Vγ3). Because no original material of the patient was accessible anymore and only a cDNA fragment of the γ-chain was left (received from Dr. G. Pluschke, Swiss Tropical Institute, Basel, Switzerland), we reconstructed the rest of the γ-chain and the full-length sequence of the δ-chain by PCR using lymphocyte cDNA of unrelated donors as template. Missing or inappropriate base pairs were corrected by site-directed mutagenesis. All constructs were verified by full-length sequencing in both directions.

Table I.

Sequences of γ- and δ-chains of γδ-TCR-control transfectantsa

Amino Acid Sequences
TCR-γ chain   
  v | nn | J | c   
 RLQNLIENDSGVYYCATWD-G-YYYKKL—FGSGTTLVVTDKQ..IKT——– Vγ1.3+ 
 IHNVEKQDIATYYCALWEVT-ELGKKIKVFGPGTKLIITDKQ..IKT——— Vγ9*altVJγ 
 RLQNLIENDSGVYYCATWD-GPNYYKKL–FGSGTTLVVTDKQ..IKT——— Vγ1.3*altγCDR3 
 RLQNLIENDSGVYYCATWD-G-YYYKKL–FGSGTTLVVTDKQ..IKTDVTTVDPKY Vγ1.3*altγC 
TCR-δ chain   
  v | nDn | J | c   
 AKNLAVLKILAPSERDEGSYYCACDTVWGTSKSSWDTRQMFFGTGIKLFVEPRSQP Vδ2+ 
 AAKSVALTISALQLEDSAKYFCALGE–QGDWDL–––KLIFGKGTRVTVEPRSQP Vδ1*altVDJδ 
 AKNLAVLKILAPSERDEGSYYCACDTITGGYQEE–YTDKLIFGKGTRVTVEPRSQP Vδ2*altδCDR3J 
 AKNLAVLKILAPSERDEGSYYCACDT––PSSPWDTRQMFFGTGIKLFVEPRSQP Vδ2*altδCDR3 
Amino Acid Sequences
TCR-γ chain   
  v | nn | J | c   
 RLQNLIENDSGVYYCATWD-G-YYYKKL—FGSGTTLVVTDKQ..IKT——– Vγ1.3+ 
 IHNVEKQDIATYYCALWEVT-ELGKKIKVFGPGTKLIITDKQ..IKT——— Vγ9*altVJγ 
 RLQNLIENDSGVYYCATWD-GPNYYKKL–FGSGTTLVVTDKQ..IKT——— Vγ1.3*altγCDR3 
 RLQNLIENDSGVYYCATWD-G-YYYKKL–FGSGTTLVVTDKQ..IKTDVTTVDPKY Vγ1.3*altγC 
TCR-δ chain   
  v | nDn | J | c   
 AKNLAVLKILAPSERDEGSYYCACDTVWGTSKSSWDTRQMFFGTGIKLFVEPRSQP Vδ2+ 
 AAKSVALTISALQLEDSAKYFCALGE–QGDWDL–––KLIFGKGTRVTVEPRSQP Vδ1*altVDJδ 
 AKNLAVLKILAPSERDEGSYYCACDTITGGYQEE–YTDKLIFGKGTRVTVEPRSQP Vδ2*altδCDR3J 
 AKNLAVLKILAPSERDEGSYYCACDT––PSSPWDTRQMFFGTGIKLFVEPRSQP Vδ2*altδCDR3 
a

Amino acid sequences of the V(D)J regions of the transfected TCR γ- and δ-chains. The reconstructed γ- and δ-chains of the muscle invasive TCR are designated Vγ1.3+ and Vδ2+. Unrelated control chains with altered TCR Vγ- and Jγ-regions (Vγ9*altVJγ), V-, D-, J-δ region (Vδ1*altVDJδ), defined amino acid substitutions in the CDR3 regions (Vγ1.3*altγCDR3, Vδ2*altδCDR3J, Vδ2*altδCDR3), or exchange of the Cγ-region (Cγ2 instead of Cγ1, Vγ1.3*altCγ) are compared. The amino acids differing from the parent chains are in bold and underlined.

The γ-chain (Vγ1.3(GV1S3A1N1T)-Jγ1-Cγ1) was reconstructed from two DNA fragments as PCR templates. One was a TCR-γ chain PCR product of the original cDNA covering the V-, J-, and parts of the C-region. The second was unrelated lymphocyte cDNA. Three fragments were derived from those two templates by PCR and subsequently ligated for full-length reconstruction of the γ-chain. The following three primer pairs were used: Vγ1.3-SalI-for (5′-GATGTCGACATGCGGTGGGCCCTAGCGGTGCTTCTAGCTTTCCTGTCTC) and Vγ-Bsi-rev (5′-GCACAGTAATAGACTCCGGAATCATTTTC), Vγ-Bsi-for (5′-GAAAATGATTCCGGAGTCTATTACTGTGC) and Cγ-Aat-rev (5′-GCTTGGGGGAGACGTCTGCATC), and Cγ-Aat-for (5′-GATGCAGACGTCTCCCCCAAGC) and Cγ-BglII-rev (5′-GGAAGATCTTTATGATTTCTCTCCATTGC). The fragments were ligated via BsiMI and AatII restriction sites and inserted into pRSVneo via SalI and BglII.

The δ-chain (DV102S1A1T, Vδ2-Jδ3-Cδ) was reconstructed from two unrelated lymphocyte cDNA using the primer pairs Vδ2-SalI-for (5′-GATGTCGACATGCAGAGGATCTCCTCCC) and nDn-Vδ2-Kpn-rev (5′-GAGGTACCCCAGACAGTGTCACAGGCACAGTAGTAAG), nDn-Vδ2-Kpn-for (5′-TGGGGTACCTCAAAGAGCTCCTGGGACACCCGACAGATG) and Cδ-BamHI-rev (5′-CGCGGATCCTTACAAGAAAAATAACTTGGC). After digestion with SalI, KpnI, and BamHI, the two δ-chain fragments and the expression vector pRSVhygro were ligated.

To confirm Ag specificity of the muscle invasive γδ TCR, α-, β-, γ-, and δ-control chains were cloned 1) from the human T cell line BBC9 (Vα22+Vβ9+ TCR), which recognizes the MBP peptide 139–151 in the context of HLA-DRB1*0101 and -DRB5*0101 (5); 2) from unrelated cDNA; or 3) were established by mutagenesis from the γ- and δ-chains described above and cloned into pRSVneo or pRSVhygro. The amino acid sequences of all γ- and δ-chains are given in Table I. Amino acids differing from the original γ- and δ-chain are underlined and bold. Transfection and establishment of clones was performed as described below.

The plasmids were transfected into a subclone (6) of the mouse T cell hybridoma BW58αβ which lacks endogenous TCR chains, but expresses all CD3 molecules (7, 8). A total of 1 × 107 cells/ml in RPMI medium were electroporated at 260 V, 960 μF with 30 μg of each plasmid in Gene Pulse cuvettes (Bio-Rad, Hercules, CA). Transfectants were sequentially selected with 1.5 mg/ml G418 and 0.3 mg/ml Hygromycin B (Sigma-Aldrich, St. Louis, MO), and clones were established by picking single colonies. Several independent clones exhibiting TCR surface expression and capable of IL-2 production were established for each transfection.

L cell transfectants expressing either DRA1*0101/DRB1*0101 (DR1) or DRA1*0101/DRB5*0101 (DR2a) were established by transfecting LTK cells with appropriate cDNA clones of the HLA molecules in pRSVneo or pRSVhygro and selection with G418 and Hygromycin B, respectively.

Surface expression of TCR-CD3 complexes was monitored by FACS analysis. G418 and Hygromycin-resistant clones (∼105 cells) were incubated with primary Ab for 30 min at 4°C in FACS buffer (PBS containing 0.5% BSA and 0.1% NaN3). After two washes with FACS buffer, secondary DTAF-labeled Abs were added for 30 min at 4°C. Controls were treated only with DTAF-labeled secondary goat anti-mouse Ab. Cells were analyzed on a FACSort flow cytometer using CellQuest software (BD Biosciences, Heidelberg, Germany). A total of 10,000 cells per sample were analyzed. Viable cells were gated after adding 1 μg/ml propidium iodide. All transfectants showed comparable levels of TCR-CD3 complex expression as judged from the FACS staining efficiency with the anti-CD3 Ab 145-2C11.

The transfectants were stimulated in different ways, namely with 1) immobilized anti-TCR Abs; 2) living target cells; 3) target cell lysates; 4) purified or recombinantly expressed proteins; or 5) nonpeptidic Ags. For activation of transfectants by Abs against TCR or CD3 molecules, activating and control Abs were coated for 16 h at 4°C at 1 μg/ml in PBS onto 96-well flat-bottom plates (Costar, Corning, NY). Target cell lysates (see Preparation of target Ags) were coated at 0.1–1.0 mg total protein/ml for 2 h at 37°C. In some experiments, 50–90% confluent living cells were tested for their Ag presentation capability to γδ TCR transfectants directly. For stimulation of MBP139–153-specific αβ TCR control transfectants, HLA-DRA/B1*0101 or DRA/B5∗0101-transfected LTK cells (1 × 105) were loaded with peptides (0.1–100 mM) for 10 h and washed twice before adding TCR transfectants. To all samples, 50,000 transfectants in 150 μl RPMI with 10% FCS were added per well and incubated for 16 h at 37°C. Supernatants were tested for mouse IL-2 by ELISA (Endogen, Woburn, MA). Autopresentation of the Ag by the transfectants was excluded, because first the transfectants were diluted so that direct cell-cell contacts were unlikely. Second, some of the samples were negative. This would not be expected if the transfected cells presented an endogenous Ag to each other.

All target cells except human myoblasts were maintained in RPMI 1640 supplemented with 10% FCS, 2 mM l-glutamine. Human myoblasts were isolated from normal subjects and patients with inflammatory myopathy, purified for CD56 by magnetic bead separation, and cultured with skeletal muscle growth medium (Promocell, Heidelberg, Germany) as described (9). Myoblasts used in our experiments stained >95% positive for the neural cell adhesion molecule (CD56) by FACS analysis. Spodoptera frugiperda cells were grown in TMN-FH medium, 10% FCS (Invitrogen, San Diego, CA). Adherent cells were detached with trypsin-EDTA (Life Technologies, Karlsruhe, Germany). All cells were washed twice with PBS, and resuspended at 107–1.5 × 107 cells/ml in sterile H2O for osmotic lysis. Alternatively, cells were resuspended in 1 ml PBS and disrupted by one freeze/thaw step (freezing for at least 12 h at −20°C). The lysates were centrifuged first at 1,200 × g for 10 min, then at 10,000 × g for 1 h, and finally at 100,000 × g for 1 h. Both lysis methods resulted in comparable results concerning the yield of total protein (∼1 mg/ml) and content of the TCR-stimulating Ag as measured by T cell activation.

TE671 ultracentrifugation supernatant and control protein (BSA, 10 μg/ml; Sigma-Aldrich) were digested with 50 μg/ml proteinase K (Boehringer-Mannheim, Mannheim, Germany) in 10 mM Tris-HCl/5 mM EDTA (pH 7.4) or Trypsin-EDTA one time in HBSS (pH 7.4; Life Technologies) for 2 h at 37°C. Digestion efficiency was confirmed by SDS-PAGE. For heat denaturation, TE671 ultracentrifugation supernatant or BSA was incubated at elevated temperatures for 4 min each, or for different periods of time at 95°C and immediately cooled on ice. Samples were adsorbed to microtiter plates as described above.

We reconstructed the γδ TCR of the previously described Vγ1.3+Vδ2+ T cell clone (3, 4). The expanded, muscle invasive γδ T cells were originally identified in muscle biopsy sections of a patient with polymyositis (3). Combined immunohistochemical and RT-PCR analysis revealed that the inflammatory γδ T cells were essentially monoclonal and expressed an unusual Vγ1.3+/Vδ2+ TCR (4). As shown in Fig. 1, two populations of inflammatory T cells could be clearly distinguished, the CD3+Vδ2+ muscle invasive γδ T cells (stained yellow in Fig. 1 a), and a bystander population of CD3+Vδ2 αβ T cells (stained red).

Because the patient had died in the meantime and biopsy and cDNA material were not available, we relied on PCR mutagenesis to reconstruct the cDNA for the full-length γ and δ TCR chains from normal PBMC (see Materials and Methods). The reconstructed γ- and δ-chain protein sequences were completely identical with the previously identified sequences (3, 4).

We transfected the reconstructed cDNA of the γδ TCR into the mouse T hybridoma BW58αβ. As shown by flow cytometry, the transfectants displayed the same staining pattern as the γδ T cells in the original muscle biopsy specimen. The Vγ1.3+/Vδ2+ transfectants stained positive with anti-CD3, anti-pan γδ TCR, anti-Vγ1.3 and -Vδ2 mAbs, but not with Abs to Vγ9 or Vδ1. Fig. 2,a shows clone B5 as a representative of three independent clones of identical staining pattern. Control cells transfected with an αβ TCR (BBC9, clone 5-1) specific for HLA-DR/MBP complexes were stained by the specific anti-Vβ9 Ab, but not by a control Ab to Vδ2 (Fig. 2 a).

FIGURE 2.

Characterization of the γδ TCR transfectants. a, Cell surface expression of the reconstructed γδ heterodimer (Vγ1.3+/Vδ2+) on BW58αβ T cell hybridoma as determined by flow cytometry (left). The γδ transfectants stained positive with mAbs against CD3, pan-γδ TCR, Vγ1.3, and Vδ2, but were not stained by Vγ9- and Vδ1-Abs. Clone B5 is shown as a representative of three independent clones. Control cells transfected with a Vα22+/Vβ9+ TCR (BBC9, clone 5-1) were stained with the anti-Vβ9 Ab, but were not stained by an Ab against Vδ2 (right). Histograms show transfectants stained with the Ab (shaded area), overlaid by control (white area). b, Ab induced activation of Vγ1.3+/Vδ2+ (clone B5) and Vα22+/Vβ9+ (clone BBC9-5-1) TCR transfectants. Abs (see Fig. 2,a) were coated to microtiterplates, transfectants were added, and IL-2 was measured in the supernatants by ELISA (see Material and Methods). ▪, The Vγ1.3+Vδ2+ transfectant carrying the muscle invasive γδ TCR; □, The Vα22+Vβ9+ αβ TCR control transfectant BBC9. Note that the γδ TCR transfectants responded to the same mAbs that yielded positive surface staining by FACS (cf Fig. 2 a). The Vα22+Vβ9+ αβ TCR transfectant (BBC9) responded additionally to its cognate Ag, i.e., HLA-DR-transfected fibroblasts preincubated with MBP peptide 139–151 (□ at the right).

FIGURE 2.

Characterization of the γδ TCR transfectants. a, Cell surface expression of the reconstructed γδ heterodimer (Vγ1.3+/Vδ2+) on BW58αβ T cell hybridoma as determined by flow cytometry (left). The γδ transfectants stained positive with mAbs against CD3, pan-γδ TCR, Vγ1.3, and Vδ2, but were not stained by Vγ9- and Vδ1-Abs. Clone B5 is shown as a representative of three independent clones. Control cells transfected with a Vα22+/Vβ9+ TCR (BBC9, clone 5-1) were stained with the anti-Vβ9 Ab, but were not stained by an Ab against Vδ2 (right). Histograms show transfectants stained with the Ab (shaded area), overlaid by control (white area). b, Ab induced activation of Vγ1.3+/Vδ2+ (clone B5) and Vα22+/Vβ9+ (clone BBC9-5-1) TCR transfectants. Abs (see Fig. 2,a) were coated to microtiterplates, transfectants were added, and IL-2 was measured in the supernatants by ELISA (see Material and Methods). ▪, The Vγ1.3+Vδ2+ transfectant carrying the muscle invasive γδ TCR; □, The Vα22+Vβ9+ αβ TCR control transfectant BBC9. Note that the γδ TCR transfectants responded to the same mAbs that yielded positive surface staining by FACS (cf Fig. 2 a). The Vα22+Vβ9+ αβ TCR transfectant (BBC9) responded additionally to its cognate Ag, i.e., HLA-DR-transfected fibroblasts preincubated with MBP peptide 139–151 (□ at the right).

Close modal

Surface expression of the transfected γ- and δ-chains precisely correlated with the functional response to TCR stimulation with various mAbs immobilized to microtiter plates (Fig. 2,b). The transfectants produced up to 250 ng/ml IL-2 after stimulation with anti-CD3, pan-γδ TCR, anti-Vγ1.3, and anti-Vδ2-Abs, but not with Abs to Vγ9 or Vδ1 (Fig. 2,b). The BBC9 (αβ TCR) control transfectants were activated by anti-Vβ9 Ab but not by Vδ2 Abs. In addition, the BBC9 cells could be stimulated by the same peptide/MHC combination recognized by its parent human T cell line (DRB1*0101/MBP139–151) (Ref. 5 ; Fig. 2 b). These results imply that transfection of human αβ as well as γδ TCR chains into mouse BW58αβ yields functional TCR/CD3 complexes, indicating that the human conserved TCR regions of human γδ as well as the αβ TCR can interact with the murine CD3 molecules for signaling.

Our functional γδ T cell transfectants were tested for reactivity against two types of human muscle cells: myoblasts cultured from normal or inflammatory human muscle, and the human rhabdomyosarcoma cell line TE671 as the source of Ag(s). The myoblasts were lysed by osmotic shock or repeated freeze-thaw cycles, subjected to consecutive centrifugation steps of increasing centrifugal force, and the supernatants were used for stimulation of the γδ TCR transfectants. The γδ TCR transfectants were incubated in the absence of APCs in microtiter plates preadsorbed with the human muscle cell preparations. As read out for stimulation of the TCR transfectants, we determined the amount of IL-2 secreted into the supernatant by ELISA.

Whereas intact muscle cells did not stimulate the transfectants, the supernatants triggered IL-2 production (Fig. 3,a). The ultracentrifuge supernatants yielded strong signals, indicating that the γδ TCR recognized a soluble Ag, which was enriched in the cytosol. Further characterization of the putative target Ag(s) indicated that it was a protein, or at least associated with a protein, because it was sensitive to proteinase K and trypsin digestion, as well as to heat denaturation (Fig. 3,b). Although the human muscle cell line TE671 yielded the strongest stimulatory signal, small amounts of the same (or a cross-reactive) Ag seemed to be present in other cell lines, including S. frugiperda insect cells. However, the transfectants could not be stimulated by control proteins such as BSA (Fig. 3,a), nonspecific Abs (Fig. 2,b), or denatured proteins (Fig. 3 b), indicating that the response was specific.

FIGURE 3.

A protein-Ag expressed in muscle cells is recognized by the γδ TCR. a, Subcellular fractions of human myoblasts and TE671 human rhabdomyosarcoma cells stimulate the γδ TCR transfectants. Intact myoblasts and TE671 cells did not activate transfectants, but lysed cells or centrifugation supernatants after lysis coated to microtiter plates induced IL-2 production. This indicates that the Ag is not presented at the cell surface. Strong stimulation was observed with ultracentrifuged supernatants, showing that the Ag is present in the cytosol. BSA (1 mg/ml) served as negative control. b, Proteinase digestion and heat denaturation of the TE671 lysates abolishes γδ TCR activation. Ultracentrifuged supernatant of TE671 was digested with proteinase K or trypsin for 2 h, or were incubated at elevated temperatures (37, 50, 60, 70, 80, and 90°C) for 4 min or at 95°C for increasing times (0–10 min). Both protease digestions and both heat inactivation experiments abrogated bioactivity of the antigenic fraction, suggesting that a protein is recognized by the γδ TCR via a conformational epitope.

FIGURE 3.

A protein-Ag expressed in muscle cells is recognized by the γδ TCR. a, Subcellular fractions of human myoblasts and TE671 human rhabdomyosarcoma cells stimulate the γδ TCR transfectants. Intact myoblasts and TE671 cells did not activate transfectants, but lysed cells or centrifugation supernatants after lysis coated to microtiter plates induced IL-2 production. This indicates that the Ag is not presented at the cell surface. Strong stimulation was observed with ultracentrifuged supernatants, showing that the Ag is present in the cytosol. BSA (1 mg/ml) served as negative control. b, Proteinase digestion and heat denaturation of the TE671 lysates abolishes γδ TCR activation. Ultracentrifuged supernatant of TE671 was digested with proteinase K or trypsin for 2 h, or were incubated at elevated temperatures (37, 50, 60, 70, 80, and 90°C) for 4 min or at 95°C for increasing times (0–10 min). Both protease digestions and both heat inactivation experiments abrogated bioactivity of the antigenic fraction, suggesting that a protein is recognized by the γδ TCR via a conformational epitope.

Close modal

Thus far, obvious candidate Ags, including muscle structural proteins (myosin, titin), and known γδ T cell Ags (10), such as MHC and CD1 molecules, various heat-shock proteins (hsp) (hsp70, hsp73, hsp90, hsp65, hsp38, GroEl), and nonpeptidic alkylphosphates (monoethylphosphate, isopentenylpyrophosphate) all failed to stimulate our Vγ1.3+/Vδ2+ TCR transfectants (negative data not shown). Other known targets of γδ T cells, such as intact EBV-transformed B cells and Daudi lymphoma cells, also did not stimulate our transfectants (data not shown).

MHC-nonrestricted Ag recognition was observed previously with other γδ T cells (reviewed in Refs. 11, 12, 13). For further evidence that our Vγ1.3+Vδ2+ transfectants were indeed stimulated by a TCR-dependent, Ag-specific recognition mechanism, a set of control transfectants was constructed that had either altered γ or δ TCR chains, or single amino acid exchanges in the CDR3 regions of either chain. In Table I, the mutated TCR chains are marked by an asterisk and the type of alteration is indicated as a superscript. The set of altered TCR molecules includes untransfected 58αβ recipient cells, an αβ TCR (BBC9), two transfectants with either an altered γ- or δ-chain (Vγ9*altVJγVδ2+, Vγ1.3+Vδ1*altVDJδ), three transfectants with defined amino acid exchanges in CDR3 regions of either the γ- or δ-chain (Vγ1.3*altγCDR3Vδ2+, Vγ1.3+Vδ2*altδCDR3, Vγ1.3+Vδ2*altδCDR3J), and a transfectant in which only the constant region of the γ-chain Cγ1 is changed to Cγ2, but the V- and J-regions are unchanged (Vγ1.3*altγCVδ2+). All transfectants expressed a functional γδ TCR heterodimer on the cell surface, as determined by FACS analysis after staining with appropriate Abs (data not shown) and by IL-2 production after stimulation with immobilized chain-specific Abs. The levels of IL-2 production were comparable among different clones of the different transfections (Fig. 4 a).

FIGURE 4.

Recognition of the γδ TCR Ag is CDR3-specific as shown by TCR mutagenesis. a, Activation of different TCR transfectants by immobilized Abs. Representative clones of different wild-type and mutagenized γδ TCR transfectants were activated with specific or with control Abs immobilized to microtiter wells, and secreted IL-2 was measured in the supernatant. The amino acid sequences of each chain are listed in Table I. The Vγ1.3+Vδ2+ TCR transfectant (clone B5) is compared with untransfected BW58αβ recipient cells and to mutagenized γδ TCR transfectants, with an altered VJ region of the γ-chain, but correct δ-chain (Vγ9*altVJγVδ2+, clone A3), with the correct γ-chain, but altered V(D)J region of the δ-chain (Vγ1.3+Vδ1*altVDJδ, clone C3), with a two amino acid exchange in the CDR3 sequence of the γ-chain (Vγ1.3*altγCDR3Vδ2+, clone A1), with a four amino acid exchange in the CDR3 sequence of the δ-chain (Vγ1.3+Vδ2*altδCDR3, clone B5), with a complete exchange of the nDn and the J region in the δ-chain (Vγ1.3+Vδ2*altδCDR3J, clone A1) and with identical γ- and δ-chains V(D)J regions, but altered conserved region (Vγ1.3*altγCVδ2+, clone B2). The transfectants produced IL-2 to comparable levels upon stimulation with specific Abs, but were not activated by nonspectific Abs. b, Activation of different TCR transfectants by TE671 ultracentrifuge supernatant as measured by IL-2 ELISA. The same clones with mutagenized TCR as in Fig. 4 a are shown, together with cells transfected with an αβ TCR (Vα22+/Vβ9+, clone BBC9-5-1). None of the control transfectants could be activated by the TE671 supernatant except the Vγ1.3*altγCVδ2+ transfectant, in which only the conserved region Cγ1 is replaced by Cγ2. Even modest amino acid exchanges in the CDR3 regions (2 aa in Vγ1.3*altγCDR3 or 4 aa in Vδ2*altδCDR3) abrogated recognition completely, providing evidence that recognition of the TE671 Ag by the γδ TCR is conveyed via the CDR3 region.

FIGURE 4.

Recognition of the γδ TCR Ag is CDR3-specific as shown by TCR mutagenesis. a, Activation of different TCR transfectants by immobilized Abs. Representative clones of different wild-type and mutagenized γδ TCR transfectants were activated with specific or with control Abs immobilized to microtiter wells, and secreted IL-2 was measured in the supernatant. The amino acid sequences of each chain are listed in Table I. The Vγ1.3+Vδ2+ TCR transfectant (clone B5) is compared with untransfected BW58αβ recipient cells and to mutagenized γδ TCR transfectants, with an altered VJ region of the γ-chain, but correct δ-chain (Vγ9*altVJγVδ2+, clone A3), with the correct γ-chain, but altered V(D)J region of the δ-chain (Vγ1.3+Vδ1*altVDJδ, clone C3), with a two amino acid exchange in the CDR3 sequence of the γ-chain (Vγ1.3*altγCDR3Vδ2+, clone A1), with a four amino acid exchange in the CDR3 sequence of the δ-chain (Vγ1.3+Vδ2*altδCDR3, clone B5), with a complete exchange of the nDn and the J region in the δ-chain (Vγ1.3+Vδ2*altδCDR3J, clone A1) and with identical γ- and δ-chains V(D)J regions, but altered conserved region (Vγ1.3*altγCVδ2+, clone B2). The transfectants produced IL-2 to comparable levels upon stimulation with specific Abs, but were not activated by nonspectific Abs. b, Activation of different TCR transfectants by TE671 ultracentrifuge supernatant as measured by IL-2 ELISA. The same clones with mutagenized TCR as in Fig. 4 a are shown, together with cells transfected with an αβ TCR (Vα22+/Vβ9+, clone BBC9-5-1). None of the control transfectants could be activated by the TE671 supernatant except the Vγ1.3*altγCVδ2+ transfectant, in which only the conserved region Cγ1 is replaced by Cγ2. Even modest amino acid exchanges in the CDR3 regions (2 aa in Vγ1.3*altγCDR3 or 4 aa in Vδ2*altδCDR3) abrogated recognition completely, providing evidence that recognition of the TE671 Ag by the γδ TCR is conveyed via the CDR3 region.

Close modal

To demonstrate that Ag recognition was CDR3-specific, TE671 ultracentrifuge supernatant was coated to microtiter wells, γδ TCR and control transfectants were added, and IL-2 was measured in the supernatant (Fig. 4,b). The only control transfectant that responded to stimulation with muscle cell extract was Vγ1.3*altγCVδ2+, i.e., the one with altered Cγ-region (Fig. 4 b). This provides clear evidence that the MHC-nonrestricted Ag recognition by our γδ transfectants is nevertheless TCR CDR3-dependent and therefore highly unlikely to be mediated by a superantigen effect.

Strong evidence suggests that human polymyositis, one of the most prevalent inflammatory muscle diseases, is an autoimmune disorder, which is at least partially mediated by autoaggressive T cells attacking the body’s own skeletal muscle tissue (reviewed in Refs. 14, 15, 16, 17). There are animal models of myositis following autoimmunization against muscle components or conditional transgenic up-regulation of MHC class I in muscle fibers (18), and the human disease responds to immunosuppressive treatment (reviewed in Refs. 14, 15, 16, 17). The histological picture of the inflammatory muscle lesion in polymyositis also suggests a pathogenic role for T cells. Typically, individual T cells attach to muscle fibers and then deeply penetrate into these target cells. In most cases of polymyositis, these “autoinvasive” T cells are uniformly CD8+ and express a restricted αβ TCR repertoire (19, 20). The muscle-attached CD8+ T cells direct perforin vesicles toward the muscle fiber membrane, suggesting a perforin-dependent mechanism of muscle fiber injury (21). All in all, these features strongly suggest, but do not prove, the autoaggressive function of the invading T cells. To prove their autoimmune nature, it is especially important to demonstrate T cell recognition of a muscle autoantigen.

To demonstrate that autoinvasive T cells in polymyositis can indeed recognize muscle-associated Ag(s), we used a specific paradigm. Several years ago, we discovered an unusual form of γδ T cell-mediated polymyositis (3), in which CD4CD8CD3+ γδ T cells surrounded and invaded the muscle fibers in exactly the same way as CD8+ αβ T cells in the more commonly occurring form of polymyositis (Fig. 1). This finding strongly suggested the myocytotoxic nature of these γδ T cells. Furthermore, invading γδ T cells have been shown to be essentially monoclonal, thus allowing us to identify both chains of their receptor by RT-PCR (4). We reconstructed this unusual Vγ1.3/Vδ2 TCR rearrangement essentially from the published sequences, because no biological material was available from the demised patient. We functionally expressed this human TCR in a mouse hybridoma lacking endogenous TCR, and thus, demonstrated that the human γδ TCR recognizes a muscle-associated protein in a CDR3-dependent, MHC-nonrestricted way. The target Ag was found in normal human myoblasts from an unrelated donor and in the human rhabdomyosarcoma cell line TE671, suggesting that the pathogenic γδ TCR recognizes a muscle-associated autoantigen and not a viral, bacterial, or superantigen. Thus, in this particular case, not only do we have strong (but indirect) histological evidence for the pathogenic contribution of the γδ T cells (Fig. 1), but we have demonstrated that the reconstructed TCR recognizes a muscle-associated Ag. However, it seems that the Ag is not only expressed in human muscle but also in other cells, even from other species. Therefore, we assume that the unknown Ag has been widely conserved during evolution. Alternatively, our γδ TCR recognizes different Ag(s), perhaps in a form of molecular mimicry (reviewed in Refs. 22 and 23). Indeed, some degree of degeneracy has been well-established for αβ TCR-mediated Ag recognition (24, 25, 26), but it is presently unknown whether Ag recognition by γδ TCR has a similar degree of degeneracy. Our TCR transfectants obviously recognize an intracellular Ag in vitro (Fig. 3,a). In contrast, the histological pictures (Fig. 1) strongly suggest that in the patient’s muscle, the γδ T cells likely recognize an Ag on the cell surface. One could speculate that the Ag is normally intracellular, perhaps expressed in other tissues as well. In our patient’s muscle, an unknown initial trigger might have induced an up-regulation and surface expression of the Ag. The nature of the initiating event, leading to local up-regulation of the autoantigen, remains unknown. It could be an inflammatory stimulus, perhaps some form of local infection, or some form of paraneoplastic reaction. In this regard, it may be noteworthy that the patient died from cancer several years after his polymyositis was diagnosed and successfully treated.

In principle, a similar strategy should be useful for the characterization of autoreactive T cells in αβ T cell-mediated polymyositis, and in other diseases, such as multiple sclerosis. However, the T cell infiltrates in other diseases are usually oligoclonal rather than strictly monoclonal as in our case of polymyositis (27, 28, 29). Thus, it will be necessary to identify the TCR-α and β chains from individual T cells by single-cell manipulation and PCR techniques. Recently, this was partly achieved in multiple sclerosis brain lesions, but only for the TCR-β chain, not for the α-chain (2). Functional expression of candidate αβ TCRs is possible in the BW58αβ hybridoma, as shown by several groups and with our control transfectants (Fig. 2,a). However, in contrast to our γδ transfectants, which seem to recognize their Ag like Abs, αβ TCR transfectants need to be stimulated with Ag presented by appropriate MHC molecules (Fig. 2 b). MHC-nonrestricted yet CDR3-dependent Ag recognition was previously described for other γδ T cells (30, 31). This “Ab-like” behavior is consistent with the CDR3-length distribution of γ- and δ-chains (reviewed in Ref. 11). Recently, the crystal structure of a Vγ9/Vδ2 TCR was determined (32) and again, the overall conformation of the complete Vγ9/Vδ2 TCR looks more Ab-like than “TCR-like.”

To our knowledge, the Vγ1.3+/Vδ2+ TCR investigated in this study is the only example of a molecularly defined human γδ TCR that was directly derived from a morphologically characterized tissue lesion. Under physiological conditions, γδ T cells are thought to function as effectors of host defenses and modulators of innate and adaptive immune responses, which possibly function as important regulatory cells (reviewed in Refs. 13 and 33). They may be involved in certain immunopathological conditions (reviewed in Refs. 13 and 34, 35, 36). However, except for circumstantial evidence, little is known about the role of γδ T cells in human (auto)immune diseases. One can speculate in our case that the Vγ1.3+/Vδ2+ T cells normally recognize a stress or tumor-related Ag and may have physiological functions in tissue repair or tumor defense, perhaps in the context of the “immunological homunculus” (37, 38).

We thank Dr. E. O. Long (National Institute of Allergy and Infectious Diseases, National Institutes of Health, Bethesda, MD) for providing the pRSV expression vectors. We thank Dr. Monika Bradl for stimulating discussions and Judy Benson for helpful comments on the manuscript.

1

This work was supported by the Hermann and Lilly Schilling Stiftung, Deutsche Forschungsgemeinschaft Grants SFB 571/A1, Wi 1722, 1-1, and SFB 217/C16.

3

Abbreviations used in this paper: CDR, complementarity-determining region; MBP, myelin basic protein; DTAF, dichlorotriazinyl-fluoresceine; hsp, heat shock protein.

1
Oksenberg, J. R., M. A. Panzara, A. B. Begovich, D. Mitchell, H. A. Erlich, R. S. Murray, R. Shimonkevitz, M. Sherritt, J. Rothbard, C. C. Bernard, L. Steinman.
1993
. Selection for T-cell receptor Vβ-Dβ-Jβ gene rearrangements with specificity for a myelin basic protein peptide in brain lesions of multiple sclerosis.
Nature
362
:
68
2
Babbe, H., A. Roers, A. Waisman, H. Lassmann, N. Goebels, R. Hohlfeld, M. Friese, R. Schroder, M. Deckert, S. Schmidt, et al
2000
. Clonal expansions of CD8+ T cells dominate the T cell infiltrate in active multiple sclerosis lesions as shown by micromanipulation and single cell polymerase chain reaction.
J. Exp. Med.
192
:
393
3
Hohlfeld, R., A. G. Engel, K. Li, M. C. Harper.
1991
. Polymoysitis mediated by T lymphocytes that express the γδ receptor.
N. Engl. J. Med.
324
:
877
4
Pluschke, G., D. Rüegg, R. Hohlfeld, A. G. Engel.
1992
. Autoaggressive myocytotoxic T lymphocytes expressing an unusual γδ T cell receptor.
J. Exp. Med.
176
:
1785
5
Giegerich, G., M. Pette, E. Meinl, J. T. Epplen, H. Wekerle, A. Hinkkanen.
1992
. Diversity of T cell receptor α and β chain genes expressed by human T cells specific for similar myelin basic protein peptide/major histocompatibility complexes.
Eur. J. Immunol.
22
:
753
6
von Bonin, A., S. Plaga, H. Ruh, S. Hebbelmann, U. Pflugfelder, S. Martin, H. U. Weltzien.
1996
. Analysis of major histocompatibility complex class I-restricted hapten recognition by mutation of the V-J joining of T cell receptor α chains.
Eur. J. Immunol.
26
:
179
7
Letourneur, F., B. Malissen.
1989
. Derivation of a T cell hybridoma variant deprived of functional T cell receptor α and β chain transcripts reveals a nonfunctional α-mRNA of BW5147 origin.
Eur. J. Immunol.
19
:
2269
8
Blank, U., B. Boitel, D. Mege, M. Ermonval, O. Acuto.
1993
. Analysis of tetanus toxin peptide/DR recognition by human T cell receptors reconstituted into a murine T cell hybridoma.
Eur. J. Immunol.
23
:
3057
9
Wiendl, H., L. Behrens, S. Maier, M. A. Johnson, E. H. Weiss, R. Hohlfeld.
2000
. Muscle fibers in inflammatory myopathies and cultured myoblasts express the nonclassical major histocompatibility antigen HLA-G.
Ann. Neurol.
48
:
679
10
Steele, C. R., D. E. Oppenheim, A. C. Hayday.
2000
. γδ T cells: non-classical ligands for non-classical cells.
Curr. Biol.
10
:
R282
11
Chien, Y. H., R. Jores, M. P. Crowley.
1996
. Recognition by γδ cells.
Annu. Rev. Immunol.
14
:
511
12
Morita, C. T., R. A. Mariuzza, M. B. Brenner.
2000
. Antigen recognition by human γδ T cells: pattern recognition by the adaptive immune system.
Springer Semin. Immunopathol.
22
:
191
13
Born, W., C. Cady, J. Jones-Carson, A. Mukasa, M. Lahn, R. O’Brien.
1999
. Immunoregulatory functions of γ/δ T cells.
Adv. Immunol.
71
:
77
14
Hohlfeld, R., A. G. Engel.
1994
. The immunobiology of muscle.
Immunol. Today
15
:
269
15
Dalakas, M. C..
1995
. Immunopathogenesis of inflammatory myopathies.
Ann. Neurol.
37
: (Suppl. 1):
S74
16
Karpati, G., S. Carpenter.
1993
. Pathology of the inflammatory myopathies. F. L. Mastaglia, ed. In
Bailliêre’s Clinical Neurology: Inflammatory Myopathies
Vol. 2
:
527
Bailliere Tindall, London.
17
Plotz, P. H..
1994
. Reverse immunology: the lessons from myositis.
Clin. Immunol. Immunopathol.
72
:
204
18
Nagaraju, K., N. Raben, L. Loeffler, T. Parker, P. J. Rochon, E. Lee, C. Danning, R. Wada, C. Thompson, G. Bahtiyar, et al
2000
. Conditional up-regulation of MHC class I in skeletal muscle leads to self-sustaining autoimmune myositis and myositis-specific autoantibodies.
Proc. Natl. Acad. Sci. USA
97
:
9209
19
Bender, A., N. Ernst, A. Iglesias, K. Dornmair, H. Wekerle, R. Hohlfeld.
1995
. T cell receptor repertoire in polymyositis: clonal expansion of autoaggressive CD8+ T cells.
J. Exp. Med.
181
:
1863
20
Mantegazza, R., F. Andreetta, P. Bernasconi, F. Baggi, J. R. Oksenberg, O. Simoncini, M. Mora, F. Cornelio, L. Steinman.
1993
. Analysis of T cell receptor repertoire of muscle-infiltrating T lymphocytes in polymyositis: restricted Vα/β rearrangements may indicate antigen-driven selection.
J. Clin. Invest.
91
:
2880
21
Goebels, N., D. Michaelis, M. Engelhardt, S. Huber, A. Bender, D. Pongratz, M. A. Johnson, H. Wekerle, J. Tschopp, D. Jenne, R. Hohlfeld.
1996
. Differential expression of perforin in muscle-infiltrating T cells in polymyositis and dermatomyositis.
J. Clin. Invest.
97
:
2905
22
Marrack, P., J. Kappler, B. L. Kotzin.
2001
. Autoimmune disease: why and where it occurs.
Nat. Med.
7
:
899
23
Benoist, C., D. Mathis.
2001
. Autoimmunity provoked by infection: how good is the case for T cell epitope mimicry?.
Nat. Immunol.
2
:
797
24
Evavold, B. D., J. Sloan-Lancaster, K. J. Wilson, J. B. Rothbard, P. M. Allen.
1995
. Specific T cell recognition of minimally homologous peptides: evidence for endogenous ligands.
Immunity
2
:
655
25
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
26
Hemmer, B., B. T. Fleckenstein, M. Vergelli, G. Jung, H. McFarland, R. Martin, K. H. Wiesmüller.
1997
. Identification of high potency microbial and self ligands for a human autoreactive class II-restricted T cell clone.
J. Exp. Med.
185
:
1651
27
Hafler, D. A., M. G. Saadeh, V. K. Kuchroo, E. Milford, L. Steinman.
1996
. TCR usage in human and experimental demyelinating disease.
Immunol. Today
17
:
152
28
Steinman, L..
2001
. Myelin-specific CD8 T cells in the pathogenesis of experimental allergic encephalitis and multiple sclerosis.
J. Exp. Med.
194
:
F27
29
Martin, R., C. S. Sturzebecher, H. F. McFarland.
2001
. Immunotherapy of multiple sclerosis: where are we: where should we go?.
Nat. Immunol.
2
:
785
30
Bukowski, J. F., C. T. Morita, Y. Tanaka, B. R. Bloom, M. B. Brenner, H. Band.
1995
. Vγ2Vδ2 TCR-dependent recognition of non-peptide antigens and Daudi cells analyzed by TCR gene transfer.
J. Immunol.
154
:
998
31
Bukowski, J. F., C. T. Morita, H. Band, M. B. Brenner.
1998
. Crucial role of TCRγ chain junctional region in prenyl pyrophosphate antigen recognition by γδ T cells.
J. Immunol.
161
:
286
32
Allison, T. J., C. C. Winter, J. J. Fournie, M. Bonneville, D. N. Garboczi.
2001
. Structure of a human γδ T-cell antigen receptor.
Nature
411
:
820
33
Hayday, A. C..
2000
. γδ cells: a right time and a right place for a conserved third way of protection.
Annu. Rev. Immunol.
18
:
975
34
Fisch, P., A. Moris, H. G. Rammensee, R. Handgretinger.
2000
. Inhibitory MHC class I receptors on γδ T cells in tumour immunity and autoimmunity.
Immunol. Today
21
:
187
35
Kabelitz, D., D. Wesch, T. Hinz.
1999
. γδ T-cells, their T-cell receptor usage and role in human diseases.
Springer Semin. Immunopathol.
21
:
55
36
Hayday, A., L. Geng.
1997
. γδ cells regulate autoimmunity.
Curr. Opin. Immunol.
9
:
884
37
Cohen, I. R..
1992
. The cognitive paradigm and the immunological homunculus.
Immunol. Today
13
:
490
38
Cohen, I. R..
2000
.
Tending Adam’s Garden: Evolving the Cognitive Immune Self
Academic, London.