Cloned Human TCR from Patients with Autoimmune Disease Can Respond to Two Structurally Distinct Autoantigens¹

Systemic lupus erythematosus (SLE)⁴s an autoimmune disease characterized by the presence of immune reactivity against a variety of self Ags. Some SLE patients are characterized by B and T cell reactivity against small nuclear ribonucleoprotein (snRNP) polypeptides; particularly by reactivity against the ribonucleoprotein (RNP) and the Smith (Sm) Ag (1, 2). The snRNP polypeptides are part of the spliceosome complex, which is noncovalently linked to uridylate (U)-rich RNA and includes the U1-specific polypeptides 70 kDa, A, and C found in the RNP Ag, and the core or Sm polypeptides B1, B2, D1, D2, D3, E, F, and G (3). These snRNP polypeptides are abundant in all nucleated cells and normally function in the processing of pre-mRNA to mature mRNA (4).

A surprising feature of TCR usage in autoimmunity that has recently been described in a number of diseases is the use of a limited range of TCR in autoantigen recognition (5, 6, 7). We have found that TCR usage is highly restricted in recognition of certain autoantigens in SLE, including Sm and U1-70 kDa (8, 9, 10, 11). The basis for this limited TCR usage in the recognition of complex Ags found in autoimmune diseases is incompletely understood (12).

It is well established that TCR-ligand-MHC interactions are degenerate or plastic. A single TCR interacts with self peptide/self MHC complexes during thymic selection and later has the potential to interact with nonself (foreign) peptide/self MHC complexes. The same TCR can also interact with self peptide/nonself peptide-nonself MHC (i.e., allo) complexes. The full molecular details of such interactions, especially as it related to autoantigen recognition, remain to be elucidated. Although it has been proposed that plasticity of the TCR might contribute to autoimmunity, there have been few studies to examine this theory in animal models or in human disease (13).

To characterize the molecular interactions between lupus autoantigenic peptides, MHC and the TCR, we have generated human T cell lines and clones and have done T cell epitope mapping of the lupus autoantigens, Sm-B, Sm-D, and U1-70 kDa (14, 15, 16). In the course of these studies, we have identified some T cell clones that appeared to exhibit cross-reactivity between the Sm-B and U1-70 kDa Ags. Previously, we were unable to determine the molecular basis of this apparent T cell cross-reactivity. In the studies reported in this work, we have now established an experimental approach that allowed us to define the molecular basis for this cross-reactivity.

In the present study, we examined human T cell clones that were selected using one of two lupus autoantigens, U1-70 kDa or Sm-B, and found that their TCR had substantial sequence homology within the complementarity-determining region 3 (CDR3); from these T cells, we molecularly cloned and expressed individual TCR/A and TCR/B genes in the TCR-negative human cell line J.RT3-T3.5 (17, 18, 19). We then examined the interaction between cells containing transfected TCR genes and U1-70 kDa or Sm-B antigenic peptides. This work revealed that TCR reactive with the lupus autoantigens Sm-B and U1-70 kDa can be highly plastic, in that a single TCR that was selected using a single snRNP autoantigen can also recognize and respond to an unrelated peptide from another autoantigen.

Previous studies have demonstrated that intermolecular spreading contributes to autoimmune T cell reactivity with snRNP self peptides in SLE (20, 21, 22, 23). The study presented in this work implicates a new mechanism of cross-reactivity of T cells from lupus patients against self Ags, in which a single TCR can recognize two structurally distinct lupus-associated autoantigenic peptides. Furthermore, these studies provide evidence for a novel pathway of autoreactivity in lupus, in addition to those pathways previously defined, such as T cell epitope spreading via inter- and intramolecular help (20, 21, 22, 23). Finally, these studies provide new direct experimental evidence for a role of plasticity of the TCR in human autoimmune disease.

All human studies were approved by the University of Missouri Institutional Review Board. T cell clones were generated from two patients, P1 and P2. Patient P1 was classified as having systemic lupus. The HLA-DRB1 genotype of patient P1 was DRB1*0101, *1302, as determined by serologic and molecular typing (15, 16). Patient P2 was classified as having mixed connective tissue disease, and the HLA-DRB1 genotype of patient P2 was DRB1*0101, 0101.

T cell clones used in these studies were generated and characterized, as described previously (14, 15, 16). Full-length fusion proteins for the U1-70 kDa and Sm-B Ags, as well as control maltose-binding protein, were produced, as described previously (14, 15, 16). In brief, cells were obtained by apheresis and then subject to density-gradient centrifugation using Histopaque (Sigma-Aldrich, St. Louis, MO). Patient PBMC were used immediately or cryopreserved for use as APC to restimulate clones. A total of 5 × 10⁶ cells was cultured in RPMI 1640 with 2 mM l-glutamine (complete medium), supplemented with 20 μg/ml gentamicin, 15% AB-positive human serum, and containing fusion protein or pooled peptides. As Ags, U1-70 kDa fusion protein, Sm-B fusion protein, or a series of synthetic peptides 25 residues in length spanning U1-70 kDa or Sm-B were used, as described (14, 15, 16). Cells in a final volume of 5 ml were placed in a 25-cm² flask and incubated in 5% carbon dioxide at 37°C. Cells were restimulated with 5 × 10⁶ autologous APC that had been irradiated with 30 Gy and Ag in fresh medium on day 7. On day 14, T cell blasts were separated by centrifugation on a density gradient (Percoll; Pharmacia, Peapack, NJ) and cloned by limiting dilution in the presence of Ag, fresh irradiated autologous APC, and 10 U/ml IL-2 (R&D Systems, Minneapolis, MN), as described (14, 15, 16). Cells were cloned by limiting dilution, and those that were positive for growth (seeded at the lowest number of cells per well) were selected for expansion. Cell surface phenotypes of clones were determined by flow microfluorometry, as described (14, 15, 16). A series of HLA-homozygous lymphoblastoid cell lines of known HLA genotypes were used in Ag stimulation experiments to help define the restriction element used by individual TCR (14, 15, 16). All CD4-positive clones that we have tested, including those used in this study, have been restricted in Ag presentation by HLA-DR restriction elements (14, 15, 16).

A total of 2 × 10⁴ T cells in complete medium was cultured for 48 h in 96-well flat-bottom tissue culture plates, and then pulsed for 18 h with 1 μCi/well [³H]TdR. Cells were harvested, and [³H]TdR incorporation was detected by liquid scintillation counting.

Peptides were synthesized using N-[9-fluorenly]methoxycarbonyl solid-phase chemistry on an Applied Biosystems (Foster City, CA) model 433A or 432A peptide synthesizer. Peptides were analyzed for purity and sequence fidelity using HPLC and mass spectrometry. The U1-70 kDa (16) and Sm-B peptides (14) were resynthesized and analyzed to confirm their fidelity. The newly synthesized peptides were used to repeat all key experiments as an additional control for the fidelity of the peptide.

mRNA was extracted from cell pellets using an oligo(dT)-cellulose micro affinity column adsorption method (MicroFastTrack 2.0; Invitrogen, San Diego, CA). First-strand synthesis of cDNA was conducted using reverse transcriptase and RNaseH (SuperScript First-Strand Synthesis System for RT-PCR; Life Technologies, Rockville, MD). Aliquots of cDNA were PCR amplified in the presence of primers specific for 1 of 29 AV or 1 of 25 BV regions of the TCR (8, 9, 10). As a positive control, the amplification of a portion of the TCR/AC region was conducted in parallel. As described previously, extensive precautions were taken to ensure that cross-contamination of samples did not occur (8, 9, 10). The amplified DNA was subjected to gel electrophoresis in 3% Nusieve, 1% SeaKem agarose and visualized by staining with ethidium bromide. PCR-amplified fragments were purified using QIAquick PCR purification kit (Qiagen, Valencia, CA) and cloned into pCR-Blunt II-TOPO vector using Zero Blunt TOPO PCR cloning kit with One Shot chemically competent Escherichia coli (Invitrogen).

Primers were designed and used to PCR amplify and clone full-length TCR based upon the TCR subtype identified using the initial screening assay (11). The following primers were used for cloning of the TCR: AV12-GGG GTA CCC CTC AGT GAA CCA GGG CAG; AV17-GGG GTA CCA TGG AAA CTC TCC TGG GAG TG; AC-GCT CGA GCA GGC TGT CTT ACA ATC TGG CA; BV20-GGG CTC GAG AAG GTG GTG TGA GGC CAT; BC1-GGG GGA TCC ATG ACG GGT TAG AAG CTC C. The PCR amplification was conducted in a total volume of 100 μl containing DNA template, 1× reaction buffer (Stratagene, La Jolla, CA), 100 μM of each dNTP (Applied Biosystems), 2.5 U of PFU Turbo DNA polymerase (Stratagene), and 0.2 μM each primer. Each reaction cycle included denaturation at 94°C for 1 min, annealing at 55°C for 1 min, and extension at 72°C for 1 min for a total of 35–40 cycles. The amplified DNA was then subjected to gel electrophoresis in1% SeaKem agarose and visualized by staining with ethidium bromide. Gel-purified PCR-amplified DNA fragments were cloned into pCR-Blunt II-TOPO vector (Invitrogen). TOPO vectors carrying full-length TCR fragments and the episomal expression vectors pREP7 and pREP9 (Invitrogen) were restriction digested, and the fragments were separated on a 1% Nusieve agarose gel, in which they were excised and the DNA was eluted using QIAquick gel extraction kit (Qiagen). Full-length TCR/A DNA fragments were ligated to pREP7, and full-length TCR/B DNA fragments were ligated to pREP9. Aliquots of the ligation reactions were transformed into chemically competent TOP 10 cells (Invitrogen), and antibiotic-resistant colonies were selected. Plasmid DNA was extracted using QIAprep Spin Miniprep kit (Qiagen).

The nucleotide sequences of full-length TCR cloned into pREP vectors (Invitrogen) were determined using the appropriate forward (pREP) and reverse (EBV) primers.

The TCR β-chain-deficient Jurkat T cell line J.RT3-T3.5 (12 × 10⁶ cells per transfection) (17) was cotransfected with 20 μg of pREP7 (TCR/A) and/or pREP9 (TCR/B) DNA by electroporation at 250 V, 960 μF using a Gene Pulser apparatus (Bio-Rad, Hercules, CA). Cells were cultured for 2 days in complete medium containing 10% heat-inactivated newborn calf serum, and then resuspended in the same medium to which 1 mg/ml G418 (Life Technologies) and 0.5 mg/ml hygromycin (Invitrogen) had been added. Drug-resistant cell lines were selected after 3–4 wk of growth. Following expansion of each Jurkat transfectant, TCR cell surface expression was analyzed by flow cytometry using FITC-conjugated anti-CD3 and FITC-conjugated anti-TCR αβ framework-specific Abs (Caltag Laboratories, Burlingame, CA), on a FACSort flow cytometer, and the results were analyzed using CellQuest software (BD Biosciences, San Jose, CA).

After expansion and analysis by flow microfluorometry, transfected Jurkat cells (10⁵) were cultured in triplicate with homozygous cell lines of known HLA genotypes (2 × 10⁵) that had been irradiated with 70 Gy in complete medium containing 10% AB⁺ serum, 1 ng/ml PMA, and synthetic peptides in 96-well flat-bottom microtiter plates at 37°C. Following culture for 18–24 h, cells and supernatants were collected and used immediately or stored at −80°C for future analysis. Total RNA was isolated from cells using the RNAqueous isolation kit (Ambion, Austin, TX). Synthesis of cDNA from cell pellets was performed in the presence of murine leukemia virus reverse transcriptase and dNTPs. PCR was performed in an ABI Prism 7000 Sequence Detection System (Applied Biosystems). PCR amplifications were performed with 1.0 μl of cDNA sample in the presence of Taqman Universal Master Mix without AmpErase, plus target primers and FAM probe for human IL-2 (Applied Biosystems). As a control for total RNA, eukaryotic 18S ribosomal RNA was used with specific primers, and probe. PCR amplifications were always performed in duplicate, using the following conditions for amplification: 10 min at 95°C, followed by a total of 45 two-temperature cycles (15 s at 95°C and 1 min at 60°C). ELISA for IL-2 was performed on culture supernatants using a commercial assay, as recommended by the manufacturer (R&D Systems).

Accession numbers assigned to novel gene sequences identified in this study were AY232281–232285, AY239351-AY239357, and AY247832-AY247837.

As shown in Fig. 1, there was apparent cross-reactivity observed for some human T cell lines that were selected using the Sm Ag when they were restimulated with the U1-70 kDa Ag. Conversely, similar results were observed for some T cell lines that were selected using the U1-70 kDa Ag (data not shown). Because T cells were cloned by limiting dilution, the possibility that two individual cell lineages were present could not be excluded by these experiments. A large panel of human T cell clones reactive with individual snRNP has been studied and has been the subject of previous publications from our laboratory (8, 9, 10, 11, 12, 14, 15, 16). Previously published work did not describe characterization of clones exhibiting cross-reactivity; characterization of these cell lines is now presented in this study.

TCR genes were amplified by PCR using a series of subtype-specific primers, cloned, and then subjected to DNA sequencing. These results are summarized in Table Iand revealed that there was substantial similarity within the CDR3 region of the TCR, both within groups generated against either U1-70 kDa or Sm-B, as well as between the U1-70 kDa and Sm-B groups. We have previously reported similar findings among another group of T cell clones (11). The findings of structurally similar CDR3 differ, however, from those found for clones generated against the related Sm-D Ag. The structure of the TCR from Sm-D-reactive human T cell clones was not similar to TCR from Sm-B- and U1-70 kDa-reactive T cell clones. Furthermore, proliferative cross-reactivity of Sm-B or U1-70 kDa clones with Sm-D was not observed (data not shown) (8).

As shown in Fig. 2, some apparently clonal T cells were identified that had two functionally rearranged TCR/A genes, such that they could have expressed two functional TCR. This dual TCR expression could explain some cases of dual Ag recognition by the T cell clones. Subsequent analysis, however, demonstrated that dual Ag recognition could occur when a single TCR/A-TCR/B gene combination was transfected into the J.RT3.T3.5 TCR-negative cell line (see below) and, therefore, dual TCR/A gene expression was not required for dual Ag recognition.

To be able to isolate the effects of individual TCR/A and TCR/B genes on Ag recognition, as well as to subsequently be able to experimentally modify and express altered TCR genes, primers were designed so that individual full-length TCR could be cloned using PCR. The full-length TCR/A and TCR/B genes that were cloned and studied are shown in Fig. 3.

Full-length TCR/A and TCR/B genes were stably transfected into the TCR-deficient Jurkat cell line J.RT3-3.5, and the transfected cells were expanded in the presence of selective medium. Cells were then examined by flow cytometry using mAbs directed against TCR αβ framework determinants or CD3. As shown in Fig. 4, there were comparable levels of TCR cell surface expression among the three constructs examined in this study. Similar results were obtained using anti-CD3 mAb (Fig. 4). In that the J.RT3-3.5 used expresses an endogenous TRA8S4 J3 gene (data not shown), we also examined J.RT3-T3.5 cells transfected with the TRBV20S1 BJ1-5 (BV20 only or control transfectant) gene to determine whether it could form an intact TCR that could be detected on the cell surface. As shown in Fig. 4, TRBV20 did pair with the endogenous TRAV8S4 J3 gene product, and an intact TCR αβ protein could be detected on the cell surface of the BV20 only-transfected cells using an anti-TCR framework mAb. This cell line was then used as a control in subsequent experiments, as a TRBV20-positive Jurkat cell line expressing a TCR/A receptor of an irrelevant specificity.

Ag-induced IL-2 expression was studied in the TCR-negative Jurkat mutant cell line transfected with full-length TRAV17S1 AJ43 and BV20S1 BJ1-5 genes (J43), AV12S1 AJ29 and BV20S1 BJ1-5 genes (J29), AV17S1 AJ48 and BV20 genes (J48), or BV20 only control (which paired with the endogenous intact Jurkat AV8S4 J3). In kinetic experiments, done to determine the optimum time point to detect IL-2 mRNA by real-time PCR, it was found that the detection of IL-2 mRNA was optimal at 18 h following stimulation with Ag. In select instances, IL-2 mRNA expression and IL-2 protein production were analyzed following stimulation of a transfectant with Ag or anti-CD3 mAb. The real-time PCR and ELISA results parallel each other, but kinetic studies revealed that the optimal detection of IL-2 measured by ELISA was obtained at 24 h after stimulation of the transfectants, while the detection of IL-2 mRNA by real-time PCR was more sensitive and responses peaked earlier (data not shown).

Fig. 5,A illustrates a representative experiment in which real-time PCR was used to detect IL-2 mRNA expression from the various TCR transfectants following stimulation with U1-70 kDa Ag at a concentration of 25 mM. TCR transfectants J43 and J29 showed responses significantly above those for the BV20 only negative control. Transfectant J48 demonstrated lower and more variable levels of IL-2 production in response to Ag stimulation. Results shown in this work were confirmed in additional independent experiments. Transfect J43 was also examined using Ag at limiting concentrations. Fig. 5 B illustrates the dose-dependent response in IL-2 gene expression seen from J43 when stimulated with U1-70 kDa Ag at 5, 2.5, and 1.25 μM concentration. At lower concentrations of U1-70 kDa Ag J43 failed to respond above control levels of IL-2 expression.

The J43, J29, J48, and control transfectant were stimulated with either U1-70 kDa or Sm-B Ag, and IL-2 mRNA expression was assayed (Fig. 6). The data demonstrate that for J43 and J29, a single TCR can respond to either the U1-70 kDa (▪) or Sm-B (□) Ag. Although overall responses from J48 were less robust than for the other transfectants (despite comparable levels of TCR expression as detected using flow cytometry shown in Fig. 4), it did not respond to the Sm-B Ag (□), although it did consistently respond to the U1-70 kDa Ag (▪). The BV20 only control failed to respond to either the U1-70 kDa (▪) or Sm-B Ag (□).

Cloned TCR Ag reactivity was found to be limited to one U1-70 kDa and one Sm-B peptide, and was not a broad or nonspecific response. Fig. 7 illustrates that there was reactivity with the Sm-B peptide spanning the dominant T cell epitope 1 (RMRCILQDGRIFIGT) and the U1-70 kDa peptide spanning the dominant T cell epitope 3 (HMVYSKRSGKPRGYA), but not with other dominant T cell epitopes (Sm-B 2 and 3; U1-70 kDa-1, 2, 4, and 5; Sm-D 1 and 2) nor control peptides outside of the defined T cell epitopes (U1-70 kDa-6 and 7) (14, 15). The transfectant responded only to U1-70 kDa epitope-3 (HMVYSKRSGKPRGYA) and Sm-B epitope-1 (RMRCILQDGRIFIGT) peptides, as demonstrated by increased IL-2 gene expression using real-time PCR. Anti-CD3 mAb was used as a positive control for stimulation of IL-2 gene expression (8). The results shown in Fig. 7 are representative of two independent experiments.

The concept that plasticity of the TCR contributes to autoimmunity has previously been proposed, but few studies have experimentally studied this possibility in human disease. Recent studies in multiple sclerosis have supported this concept, as structural analysis of TCR reactive with myelin basic protein suggests that plasticity of the TCR in autoantigen recognition may be important in autoimmune disease (11). Previous molecular studies have not, however, examined this question in systemic autoimmunity.

The purpose of the present study was to attempt to characterize the molecular basis of the apparent T cell cross-reactivity observed between human T cell lines selected using the U1-70 kDa or Sm-B lupus-associated autoantigens. We observed that there appeared to be cross-reactivity of some human T cell clones selected against Sm-B when stimulated with U1-70 kDa and vice versa (Fig. 1 and data not shown). In these experiments, in which cells were cloned by limiting dilution, we could not exclude the possibility that two or more cell lineages were present. However, when we analyzed TCR repertoire usage and CDR3 sequences of a large series of T cell clones specific for U1-70 kDa or Sm-B, we observed that TCR usage was highly restricted and that homologous (in some cases identical) TCR CDR3 were present among clones generated against either U1-70 kDa or Sm-B (Table I). We next considered the possibility that this apparent cross-reactivity could represent cross-pairing of specific TCR/A with specific TCR/B genes, in which multiple TCR/A or TCR/B genes were functionally rearranged and expressed in a single cell. When we examined this possibility, we found that several clones did in fact express dual TCR/A genes (Fig. 2), and in these cases we could not formally exclude the possibility that this dual expression accounted for Ag cross-reactivity.

In attempt to further characterize the molecular basis of this apparent T cell cross-reactivity and to have the ability to experimentally manipulate the interaction of the TCR with Ag, we established in our laboratory an experimental approach that allows for molecularly cloning and selective expression of individual human TCR/A-TCR/B gene pairs in a TCR-negative human cell line J.RT3-3.5, as had originally been described by Brawley and Concannon (18), and by Brenner and colleagues (19). We anticipated that this system would allow for accurate determination of the contribution of individual TCR genes in Ag recognition. We were able to clone three full-length TCR/A genes and one TCR/B gene from human T cell clones reactive with these autoantigens (Figs. 2 and 3). We successfully expressed these TCR genes pairs on the cell surface of the mutant Jurkat cell line (Fig. 4). Using real-time PCR, we were then able to quantitatively assess functional receptor-ligand interactions in this model system by measuring IL-2 gene expression in response to antigenic stimulation (Figs. 5–7).

Using transfectants expressing cloned TCR genes, we found that a single TCR/A-TCR/B gene pair could respond to Ag stimulation (Figs. 5–7) in a dose-dependent manner (Fig. 5,B). Furthermore, we observed that Ag recognition occurred in the context of the same HLA-DR restriction elements as had been used by the original T cell clone (Fig. 5, and data not shown). Finally, we observed that the two lupus autoantigenic peptides, U1-70 kDa and Sm-B, could be recognized by a single transfected TCR/A-TCR/B gene pair (Fig. 6). This occurred despite very limited sequence homology between the two peptides (Sm-B RMRCILQDGRIFIGT and U1-70 kDa HMVYSKRSGKPRGYA). Furthermore, this reactivity was restricted to a single U1-70 kDa peptide and a single Sm-B peptide and did not extend to other T cell epitopes, peptides with closely related sequences, or other control peptides (Fig. 7).

Before the experiments presented in this work, various alternative explanations could potentially have explained the apparent TCR cross-reactivity first observed using human T cell clones. The transfection experiments, however, unequivocally demonstrate that the cross-reactivity observed between U1-70 kDa and Sm-B could occur solely due to the interaction of a single expressed TCR/A-TCR/B gene pair with the two autoantigenic peptides.

Previously, it has been shown that recognition of multiple components of the spliceosomal complex is common in lupus, including recognition of U1-70 kDa and Sm-B by both Abs and T cell in the same patient (8). This has been proposed to be the result of intermolecular spreading of immune recognition to different components of the same macromolecular complex (20, 21, 22, 23). Greidinger and Hoffman (24) found that the first anti-RNP Abs detected were those against U1-70 kDa and Sm-B, and that these Ab responses appeared to be linked among 5882 serum samples from 3668 patients. Similarly, Deshukh et al. (25, 26) found that in A/J mice immunized with snRNP Ags, the anti-U1-70 kDa and Sm-B Ab responses were linked and that there was evidence of the presence of T cells that were cross-reactive with more than one snRNP autoantigen.

The current study is consistent with the work cited above and provides evidence for an alternative mechanism to intermolecular spreading (20, 21, 22, 23, 24, 25, 26). The novel mechanism reported in this work may be of importance in the immune pathogenesis of SLE. We find that a single TCR can recognize multiple lupus autoantigens and that this may be the basis for autoantigen cross-reactivity in some patients. Furthermore, it is interesting that despite functional mimicry, the two cross-reactive autoantigenic peptides described in this study (U1-70 kDa HMVYSKRSGKPRGYA and Sm-B RMRCILQDGRIFIGT) lack clear sequence homology (15, 16). It will be of interest in future studies to more completely define the requirements for such shape mimicry.

There have been previous reports that a single TCR can be activated by more than one peptide. Among the best characterized, the murine TCR 2C has been shown to recognize three antigenic peptides in the context of different class I MHC molecules (27, 28). The structural basis of this cross-reactivity has been examined using a variety of approaches, including x-ray crystallography (27). It was found in the case of 2C that the interface between peptide and MHC had poor shape complementarity with the TCR and that large conformation changes in the three TCR CDR loops were induced upon binding, thus allowing for the TCR to bind to a variety of different peptides. In related work, Lang et al. (13) reported that a single TCR isolated from a patient with multiple sclerosis could recognize both an HLA-DRB1*1501-restricted myelin basic protein peptide and an EBV peptide in the context of the structurally similar and genetically linked HLA-DRB5*0101 allele.

It is of interest that dual functionally rearranged TCR/A genes were detected among some clones (Fig. 2). Although the expression of dual TCR/A genes could have potentially accounted for dual Ag recognition in these specific cases, the transfection experiments demonstrate that this was not required (Figs. 5–7). Control experiments also excluded the possibility that expression of the endogenous Jurkat TCR/A gene contributes to lupus autoantigen recognition. Although the endogenous Jurkat TCR/A gene can pair with the transfected TCR/BV20 and be detected on the cell surface at levels comparable to the TCR/A-TCR/B dual transfectants (Fig. 4), it failed to respond to stimulation with Ag (Figs. 5 and 6).

It will be of considerable interest in future experiments to examine potential differences in TCR affinity for individual peptides among TCR reactive with dual lupus autoantigen. This has not been addressed yet, but will be a subject of future research (28, 29). However, the functional relevance of the findings presented in this study is supported by the in vitro studies on T cell clones in which a single clone responded to both Sm and U1-70 kDa Ags (Fig. 1). Finally, it is important to acknowledge that SLE remains a complex syndrome, and disease pathogenesis is of course likely to extend beyond the pathogenic mechanism reported in this work. Genetic factors, abnormalities in programmed cell death, as well as functional defects in T and B cells all appear to contribute to disease pathogenesis (30, 31, 32).

In summary, using T cell clones derived from SLE patients combined with molecular cloning and expression of lupus autoantigen-reactive TCR/A and TCR/B genes, we found that there was substantial plasticity or degeneracy of the TCR in that two structurally dissimilar lupus-associated autoantigenic peptides could stimulate a single TCR. These studies provide new evidence supporting a role of plasticity of the TCR in the development of human systemic autoimmunity.

We gratefully acknowledge the excellent technical assistance of Craig Bailey, Mark Foecking, and Shannon Primm in this work.