T cell receptors recognizing type II collagen in HLADRtransgenic mice characterized by highly restricted V usage.код для вставкиСкачать
ARTHRITIS & RHEUMATISM Vol. 50, No. 6, June 2004, pp 1996–2004 DOI 10.1002/art.20289 © 2004, American College of Rheumatology T Cell Receptors Recognizing Type II Collagen in HLA–DR–Transgenic Mice Characterized by Highly Restricted V␤ Usage Xiaowen He, Edward F. Rosloniec, Linda K. Myers, William Lewis McColgan, III, Marina Gumanovskaya, Andrew H. Kang, and John M. Stuart Objective. To determine the T cell receptor (TCR) structure recognizing type II collagen (CII) in HLA– DR–transgenic mice, and to examine the role of T cells with certain V␤-chains in collagen-induced arthritis (CIA). Methods. T cell hybridomas were established from DR1- and DR4-transgenic mice and selected for their responses to CII and CII peptide containing the T cell determinants. RNA was extracted and reverse transcribed into complementary DNA, which was then amplified using appropriate V␤- and V␣-subfamily–specific primers. The polymerase chain reaction products were purified and directly sequenced. To determine the role of T cells with certain V␤-chains in CIA, V␤-subfamily– specific antibodies were administered and the development and characteristics of arthritis were determined. Results. TCRs of 23 clonally distinct T cell hybridomas that were derived from DR1-transgenic mice and that were reactive to the CII peptide containing the immunodominant determinant were analyzed. These hybridomas predominantly used the TCR V␤14 and V␤8 gene segments (70% and 30%, respectively). The same restriction in V␤ usage was also found in CIIreactive T cell hybridomas from DR4-transgenic mice. There was also restricted use of V␣ genes, although this was less marked than that of V␤. In contrast, the hybridomas expressed a diverse third complementaritydetermining region. Deletion of both V␤14-bearing and V␤8-bearing T cells significantly reduced the incidence and severity of CIA. Conclusion. These data demonstrate that DR1 and DR4 not only bind and present the same CII immunodominant peptide, but also stimulate a highly restricted subset of T cells. Collagen-induced arthritis (CIA) is a tissuespecific autoimmune disease that develops in susceptible animals when they are immunized with type II collagen (CII). It is characterized by inflammation of synovial joints and possesses many characteristics of human rheumatoid arthritis (RA) (1). Susceptibility to CIA is dependent on the development of a strong immune response to CII. In rodents, susceptibility to CIA is linked to the class II genes of the major histocompatibility complex (MHC) (2,3). Although most mouse strains are resistant to CIA, those bearing H-2q and H-2r are highly susceptible. How the MHC regulates the immune response to induce disease is not entirely clear. Although the initiation of CIA is caused by deposition of antibody in joint tissue with the subsequent activation of complement, T cells play a critical role in the full expression of disease (4,5). Mice that have deletions in the area of the genome that codes for T cell receptors (TCRs) often have reduced susceptibility to CIA (6). However, even in mice with massive deletions of TCR genes, there is sufficient plasticity in the immune response that most animals will develop disease, although the overall severity is reduced (7). We have recently shown that mice transgenic for the human immune-response genes DRB1*0101 (DR1) or DRB1*0401 (DR4), both susceptibility markers for Supported by the Office of Research and Development, Medical Research Services, Department of Veteran Affairs, and NIH grants AR-39166 and AR-47379. Xiaowen He, MD, Edward F. Rosloniec, PhD, Linda K. Myers, MD, William Lewis McColgan, III, BS, Marina Gumanovskaya, MD, PhD, Andrew H. Kang, MD, John M. Stuart, MD: Memphis Veterans Affairs Medical Center, and University of Tennessee, Memphis. Address correspondence and reprint requests to Xiaowen He, MD, Research Service 151, VA Medical Center, 1030 Jefferson Avenue, Memphis, TN 38104. E-mail: firstname.lastname@example.org. Submitted for publication September 10, 2003; accepted in revised form February 16, 2004. 1996 RESTRICTION OF V␤ USAGE IN TCRs RECOGNIZING CII human RA, are also highly susceptible to CIA (8,9). The susceptibility of DR1- and DR4-transgenic mice to CIA offers the opportunity to study the role of the human RA-susceptibility markers DR1 and DR4 in the selection of arthritogenic T cells in an easily controlled environment. The development of autoimmune arthritis in both DR1- and DR4-transgenic mice is accompanied by a strong HLA–DR restricted T cell response to human CII. The core of the dominant epitope for the T cell response in both types of transgenic mice is the same and is located at CII263–270. In the present study, we developed a panel of CII-reactive T cell hybridomas from DR1- and DR4transgenic mice immunized with CII, and analyzed the structural characteristics of the TCRs used by hybridomas specific for the dominant determinant. We found that the hybridomas used the TCR V␤14 and TCR V␤8 gene segments almost exclusively for the recognition of the CII249–281 peptide containing the dominant determinant. In contrast, these hybridomas used a number of different V␣ genes and had diverse third complementaritydetermining regions (CDR3). The requirement for cells with V␤14 and V␤8 in the development of disease was ascertained by depletion of these cells using V␤-subfamily– specific antibodies. The data indicate that V␤14⫹ and V␤8⫹ T cells play a crucial role in the induction of CIA in DR-transgenic mice. MATERIALS AND METHODS Animals and immunization. Construction of chimeric (human/mouse) DRB1*0101 and DRB1*0401 genes and generation of mice with DR1 and DR4 transgenes has been previously described (8–10). The transgenic founders were backcrossed with B10.M mice (H-2f). The B10.M-DR1 (DR1)– transgenic homozygotes were obtained by further intercrossing, while the B10.M-DR4 (DR4)–transgenic mice were maintained as heterozygous, since the homozygotes do not survive (9). All mice were bred and maintained at the Veterans Affairs Medical Center of Memphis (Tennessee) in a specific pathogen–free environment, and sentinel mice were tested routinely for the presence of mouse hepatitis and Sendai viruses. For immunization, CII was purified as previously described (11) and then dissolved in cold 0.01M acetic acid and emulsified at a 1:1 (volume/volume) ratio with Freund’s complete adjuvant (Gibco BRL, Gaithersburg, MD). Mice were injected subcutaneously at the base of the tail with 100 g of CII in a total volume of 0.1 ml. Beginning 19 days after immunization, the mice were observed for the development of arthritis, and each paw was scored for the degree of inflammation on a scale of 0–4. Production of hybridomas. T cell hybridomas were produced by fusion of lymph node cells with TCR ␣⫺/␤⫺ BW5147 thymoma cells (12). Lymph nodes were obtained 1997 from transgenic mice that had been immunized 10 days previously with human CII emulsified with Freund’s complete adjuvant. Cells were isolated and cultured with the ␣1-chain of human CII. After 3 days, interleukin-2 (IL-2) was added so that a final concentration of 10 units/ml was achieved. The cells from DR1-transgenic (homozygous) mice were cultured for an additional 2–3 days, whereas the cells from DR4-transgenic mice were cultured for an additional 5–6 days, since there was a weaker response to CII by T cells from DR4-transgenic mice. Primed T cells (8 ⫻ 105) were fused to 1.5 ⫻ 107 BW5147 cells with 50% polyethylene glycol (Boehringer Mannheim, Indianapolis, IN). The fused cells were cultured in the presence of nonfused BW5147 cells (104 cells/well) as filler cells, in 96-well plates with round bottoms. Hypoxanthine– aminopterin–thymidine (HAT)–containing medium was added after 24 hours. We cultured the fused cells at a limiting dilution concentration, so that clonal lines were obtained. Only fusions with surviving hybridomas were used (i.e., those wells that contained surviving hybridomas after HAT media selection, accounting for ⬍30% of the total number of wells). HATresistant hybrids were expanded and maintained in 24-well plates. The ability of the hybridomas to recognize the human ␣1-chain of CII and the CII249–281 peptide was tested at least twice in separate experiments. Antigen-presentation assay. T cell hybridoma cells (105) were cultured with 4 ⫻ 105 syngeneic spleen cells and with 100 g/ml of purified CII ␣-chains or 25 g/ml of synthetic peptide in a 0.2-ml culture. After 24 hours, 2-fold serial dilutions of 80 l of culture supernatant were made for determination of IL-2 titers. HT-2 cells (4,000) were added to each supernatant dilution, and after 16–20 hours, the viability of the HT-2 cells was evaluated by visual inspection and cleavage of thiazolyl blue (13,14). IL-2 titers were quantified by the reciprocal of the highest 2-fold serial dilution maintaining 90% viability of the HT-2 cells; results were expressed as units of IL-2 per ml of undiluted supernatant, as described previously (15). Polymerase chain reaction (PCR) amplification. Total cellular RNA was extracted from T cell hybridomas by guanidinium thiocyanate–phenol chloroform in a single-step extraction method (RNA Isolation Kit; Stratagene, La Jolla, CA) and reverse transcribed into complementary DNA (cDNA) using oligo(dT) primer (Advantage RT-for-PCR Kit; Clontech, Palo Alto, CA). The resulting cDNA was amplified using appropriate V␤ family–specific 5⬘ primers and a constant-region C␤-1 reverse 3⬘ primer or appropriate V␣ family–specific 5⬘ primers and a constant-region C␣-1 reverse 3⬘ primer (Table 1). PCR reaction mixtures contained 2 l of cDNA, 12.5 pmoles of each primer, 5 l of each dNTP, 5 l of 10⫻ Taq buffer (500 mM KCl, 100 mM Tris HCl, 2.0 mM MgCl2, pH 8.3), 1 unit of Taq DNA polymerase (Promega, Madison, WI), and water to 50 l. The reaction was carried out for 30 cycles using 94°C for denaturation, 50°C for annealing, 72°C for polymerization at a duration of 1 minute for each. The PCR-amplified products were electrophoresed through a 1.2% agarose gel and the DNA visualized by ethidium bromide staining. DNA sequencing. The PCR products were purified using Centricon Centrifugal Filter Devices with YM-100 membranes (Amicon, Bedford, MA) and electrophoresed through a 1.2% agarose gel. Relevant bands were excised from the gel 1998 HE ET AL Table 1. Primers for polymerase chain reaction and sequencing of T cell receptor (TCR) ␤- and ␣-chains TCR family ␤-chain V␤1 V␤2 V␤3 V␤4 V␤5 V␤6 V␤7 V␤8 V␤9 V␤10 V␤11 V␤12 V␤13 V␤14 V␤15 V␤16 V␤17 V␤18 Constant region of ␤-chain C␤-1 C␤-2 ␣-chain V␣1 V␣2 V␣3 V␣4 V␣5 V␣6 V␣7 V␣8 V␣9 V␣10 V␣11 V␣12 V␣13 V␣14 V␣15 V␣17 V␣18 V␣19 V␣20 Constant region of ␣-chain C␣-1 C␣-2 Primer sequence AGCTGCAGGCTTCTCCTC CAGACCCCACAGTGACTTTG CTAGGAATTTTGAATTCAAAAG GCAGGTCCAGTCGACCC TCTGGGGTTGTCCAGTCTCC CCCTCCAAACTATGAACAAG CAGGCCTTGTGGACATGAA CACATGGAGGCTGCAGTCA GCCACTTTTGTGGATACTACGG CCTATTGGTACAAGCAAGACTCT GAGCAGAACCAACAAATGCTG GCAAATCACACAGATGCTGG CCTCTATAACAGTTGCCCTCG TTCTTGGGTGTTAGTGCTCAG GTGGAACTTCCATGAGGATG AGCAGGACACACAGGACC AGGCTCTTTATGTTGCTGGA CTTTGGAGCCAAGTTCA GAAAGCCCATGGAACTGCACT GCTTTTGATGGCTCAAACAAGGA AGACTCCCAGCCCAGTGACTC AATCTCTGACAGTCTGGGAAGGAG GAGTTCAGCAAGAGCAACTCT ACCACCTCCTTCCACTTGC TGTGTGAGCAGAGGCGAGCAGG TTGGGTATTGCGGGTGATGCTAAG AGAAAGTGATTCAGGTCTGGTCAAC GCCACTCTCCATAAGAGCAGC TTGTTAAAGGCACCAAGGGCT AGACTGACATCCACTACAGTC ATTTGCTGGGTGAGAGGAGATCAG TGGAGCCTTGCCAAGACCACC ATGGAAAGAAAGCAGACCCAAG CAAACAGGACACAGGCAAAGGT GTGGACAGAAAACAGAGCCAAAG ACAAGCAAACAGCAAGTGGG CGTCCACGAGGGTGAAAGTGT TGGATGGTTTGAAGGACAGTGG AAATACCCTGACAACAGCCCC GTTTTCGGCACATTGATTTGGG AAGTCGGTGAACAGGCAGAGG and purified further using Geneclean Spin Kits (Bio101, Vista, CA). The protocols recommended by the manufacturers were followed exactly. Thirty to ninety nanograms of purified PCR product was mixed with 3.2 pmoles of C␤-2 or C␣-2 sequencing primers, which are internal to the C␤-1 or C␣-1 used for the initial PCR. The prepared samples were sequenced at a core facility in the Molecular Resource Center of the University of Tennessee Health Science Center (Memphis), using automated analysis. The sequences obtained were compared with all known mouse germline V␤, J␤, and D␤ gene segments for ␤-chains and V␣ and J␣ gene segments for ␣-chains, using the MacVector computer program (Accelrys, Burlington, MA). The database was constructed based on the data in the International ImMunoGeneTics Database (available at the IMGT Web site at http://imgt.cnusc.fr:8104, the GenBank Web site at http://ncbi.nlm.nih.gov, and from published sequence information ). Junctional nucleotides that did not appear to be part of the segments were assigned as N-region additions. Flow cytometric analysis. To verify the usage of the V␤ or V␣ subfamily by T cell hybridomas, the cells were divided into multiple tubes. Both phycoerythrin (PE)–conjugated antimouse V␤ monoclonal antibodies (mAb) and one of the fluorescein isothiocyanate (FITC)–conjugated anti-mouse V␤or V␣-subfamily–specific mAb were added into each tube at the concentration recommended by the supplier (BD PharMingen, San Diego, CA). To determine the overall V␤ expression, spleen cells from nonimmunized animals were isolated and divided into multiple tubes. Peridin chlorophyll protein– conjugated anti-CD4, PE-conjugated anti-CD8, and one of the FITC-conjugated anti-mouse V␤-subfamily–specific mAb were added to each tube (BD PharMingen). To determine the effectiveness of deletion of V␤14⫹ and V␤8⫹ T cells, peripheral blood was collected from each mouse. Mononuclear cells were isolated by density-gradient centrifugation on Lympholyte-M (Accurate Chemicals, Westbury, NY). Both PE-conjugated anti-mouse V␤ mAb and the FITC-conjugated anti-mouse V␤8 or V␤14 mAb were added (BD PharMingen). In all of the cases, the cells were incubated for 30–45 minutes at 4oC in the dark. After washing 3 times, the cells were suspended in 400 l cold phosphate buffered saline (PBS)/bovine serum albumin and analyzed by flow cytometry. Depletion of V␤14ⴙ and V␤8ⴙ T cells in DR1transgenic mice. Specific mAb for V␤14, V␤8, or both were injected peritoneally 3 days before the immunization that induced CIA, at a dosage of 100 g of each antibody per mouse. Anti-V␤14 was obtained from a commercial source (BD PharMingen). Anti-V ␤ 8 was purified by protein A–Sepharose column chromatography (Pharmacia Biotech, Uppsala, Sweden) from supernatant fluids of cultured F23.1 cells. A control group received PBS. The treatments were repeated on the day of immunization. Statistical analysis. Statistical evaluation of differences in the expression of each V␤-chain family by spleen lymphocytes from DR1- and DR4-transgenic mice was performed by using one-way repeated-measures analysis of variance (ANOVA) followed by all pairwise multiple comparison procedures (Tukey’s test). The incidence of CIA was analyzed by using Fisher’s exact test. The difference in the severity of CIA was analyzed by Kruskal-Wallis one-way ANOVA on ranks, followed by multiple comparisons with the control group (Dunnett’s method). Data were judged statistically significant at P values of less than or equal to 0.05. RESULTS Recognition of the CII249–281 peptide by most CII-reactive T cell hybridomas. A total of 74 T cell hybridoma lines reactive to the ␣1-chain of human CII was obtained from DR1 mice. Among them, 16 lines RESTRICTION OF V␤ USAGE IN TCRs RECOGNIZING CII Figure 1. Responsiveness of T cell hybridomas to a peptide representing human type II collagen (hCII) (249–281) (referred to as CII249–281). The hybridomas were cultured with the purified ␣1chain of hCII or CII249–281, or without antigen, for 24 hours in the presence of spleen cells isolated from DR1 mice. The concentration of interleukin-2 (IL-2) in the culture supernatant was determined by HT-2 cell assay. All of the hybridomas secreted ⱖ1,280 units/ml of IL-2 in the presence of the purified ␣ 1-chain of hCII, but secreted ⱕ20 units/ml of IL-2 in the absence of antigen. Hybridomas reactive to CII249–281 secreted ⱖ1,280 units/ml in response to the peptide. Nonresponsive or weakly reactive hybridomas to CII249–281 secreted ⱕ80 units/ml of IL-2 in response to the peptide. There were no hybridomas secreting IL-2 resulting in concentrations between 80 and 1,280 units/ml in response to CII249–281. 1999 were tested for CD4 and CD8 expression by flow cytometry. All 16 T cell hybridoma lines were CD4 positive and none of them expressed CD8. Each of the 74 hybridomas was tested for responsiveness to the CII249–281 peptide containing the dominant determinant. Sixty-two (84%) of the 74 hybridoma lines were found to be reactive (Figure 1). Because the transgenic mice express endogenous H-2f, it was necessary to establish that the observed responses were DR restricted. Twenty CII249–281–reactive hybridomas were tested for their responsiveness to the ␣1-chain of CII and CII249–281 presented by spleen cells from nontransgenic B10.M mice (H-2f). None of the hybridomas was responsive, indicating that the CII reactivity of all of the hybridomas was DR1 restricted (results not shown). V␤ usage by CII-reactive T cells in transgenic mice. The usage of the TCR V␤-chain and V␣-chain specific for human CII249–281 by 23 hybridomas in DR1transgenic mice was analyzed by PCR. All of the 23 hybridomas expressed unique V␤-chain and V␣-chain transcripts. The V␤-chain usage was highly restricted. V␤14 was identified in 70% (16 of 23) and V␤8 in 30% (7 of 23) of the hybridomas. No other V␤-chain family was identified (Table 2 and Figure 2). The V␤ usage of the hybridoma lines was also analyzed by immunofluorescence and flow cytometry using antibodies specific for Table 2. Third complementarity-determining region of T cell receptor ␤-chains expressed by hybridomas reactive to the type II collagen (249–281) peptide in DR1-transgenic mice Hybridoma 13.0 21.0 E70 E143 E166 E168 E187 2.18 5.1 14.0 16.0 19.3 24.0 D1 D34 D41 D48 E113 E156 E174 E201 E204 E205 V␤ V␤8 V␤8 V␤8 V␤8 V␤8 V␤8 V␤8 V␤14 V␤14 V␤14 V␤14 V␤14 V␤14 V␤14 V␤14 V␤14 V␤14 V␤14 V␤14 V␤14 V␤14 V␤14 V␤14 . . .Cys . . .Cys . . .Cys . . .Cys . . .Cys . . .Cys . . .Cys . . .Cys . . .Cys . . .Cys . . .Cys . . .Cys . . .Cys . . .Cys . . .Cys . . .Cys . . .Cys . . .Cys . . .Cys . . .Cys . . .Cys . . .Cys . . .Cys N–D–N Ala Ala Ala Ala Ala Ala Ala Ala Ala Ala Ala Ala Ala Ala Ala Ala Ala Ala Ala Ala Ala Ala Ala Ser Ser Ser Ser Asp Ser Ser Asp Ser Ser Asp Ser Ser Ser Ser Trp Ser Trp Ser Trp Ser Trp Ser Trp Trp Ser Trp Ser Trp Ser Trp Ser Trp Trp Trp Trp Trp Trp Ser Ser Ser Ser Arg Arg Thr Asn Arg Val Phe Gly Gly Ala Trp Thr Gly Gly Gly Ala Arg Gly Ser Gln Val Trp Gly Gly Lys Lys Ser Gln Gly Val Ile Leu Asp Gly Val Leu Asn Ser Ala Leu Ala Leu Gly Gly Ser Leu Gly Ser Ala Thr Arg Gly Thr Gly Leu Arg Gly Gly Gly Pro Gln Gly Leu Met Gly Gly Thr Lys Leu Gly Tyr Ser Arg Thr Gly Glu Leu Gly Thr Gly Gly Tyr Thr Pro Gly Leu Gly Gly Arg Gly Arg Leu Gly Tyr Gly Thr Leu Leu Met Gly Gly Pro Gly Leu Gly Glu J␤ Ser Gly Asn Thr Leu. . . Ser Tyr Glu Gln Tyr. . . Ser Tyr Glu Gln Tyr. . . Ser Gln Asn Thr Leu. . . Ala Pro Leu. . . Asn Thr Leu. . . Ser Asn Glu Arg Leu. . . Ser Gln Asn Thr Leu. . . Asn Ser Asp Tyr Thr. . . Gln Asp Thr Gln. . . Asn Ser Asp Tyr Thr. . . Asn Tyr Ala Glu Gln. . . Ala Glu Thr Leu. . . Tyr Ala Glu Gln. . . Val Phe. . . Tyr Ala Glu Gly. . . Ala Glu Gln. . . Glu Arg Leu. . . Glu Gln Tyr. . . Tyr Ala Glu Gln. . . Tyr Asn Ser Pro. . . Thr Glu Val Phe. . . Asp Thr Gln. . . J␤1.3 J␤2.7 J␤2.7 J␤2.4 J␤1.5 J␤2.4 J␤1.4 J␤2.4 J␤1.2 J␤2.5 J␤1.2 J␤2.1 J␤2.3 J␤2.1 J␤1.1 J␤2.1 J␤2.1 J␤1.4 J␤2.7 J␤2.1 J␤1.6 J␤1.1 J␤2.5 2000 Figure 2. Restricted V␤ usage of T cell hybridomas reactive to a peptide representing human type II collagen (249–281) in DR1transgenic and DR4-transgenic mice. The V␤ phenotype was determined by polymerase chain reaction and flow cytometry as described in Materials and Methods. V␤ usage was confirmed by sequencing and all of the hybridomas expressed unique functional V␤-chains, as evidenced by unique third complementarity-determining region sequences. mouse T cell V␤ subfamilies. The results were fully consistent with the PCR amplification data (results not shown). Both DR1 and DR4 alleles present the same CII peptide and are able to induce DR-restricted T cell responses to CII. To address the question of whether T cells reactive to the human CII249–281, which contains the dominant determinant, share similar restriction in V␤ usage when presented by DR4 molecules, we established CII249–281–reactive T cell hybridomas from DR4transgenic mice. Several hybridomas shared the same TCR CDR3, probably as a result of prolonged in vitro stimulation in the presence of IL-2. The duplicates were excluded from further analysis. We analyzed 11 hybridomas that had unique nucleotide sequences in the TCR CDR3, which was an indication that they were established from different T cells. Among them, 8 (73%) used V␤14 and 3 (27%) used V␤8 (Figure 2). These data indicate a striking similarity in the TCR V␤-chain usage between the CII-specific T cells of DR4transgenic mice and those of DR1-transgenic mice. HE ET AL V␣ usage by CII-reactive T cells in DR1transgenic mice. The usage of the TCR V␣-chain was analyzed in hybridomas derived from DR1-transgenic mice. The usage of V␣ was also found to be restricted, although less so than that of V␤; only 7 V␣ families were used. Fifteen of the 23 hybridomas (65%) used V␣1 or V␣2. Since only a few anti-V␣ antibodies are available and they often do not react to all of the members of the family, we could not confirm the V␣ protein expression for all of the hybridomas. However, we analyzed hybridomas using antibodies to V␣2, V␣8.3, V␣11.1, and V␣11.2. The results were consistent with the PCR analyses (results not shown). There was no apparent skewing of the V␣ usage between V␤8-bearing and V␤14bearing T cells. V ␤ 8-bearing T cell hybridomas expressed V␣1 in 2 (29%) of 7, V␣2 in 3 (43%) of 7, and the other V␣-chains in 2 (29%) of 7 hybridomas. V␤14bearing T cell hybridomas expressed V␣1 in 4 (25%) of 16, V␣2 in 6 (38%) of 16, and the other V␣-chains in 6 (38%) of 16 hybridomas. Lack of restriction of overall TCR expression in transgenic mice. To determine if the presence of the transgene caused preferential expression of V␤8 and V␤14 regardless of the specific immune response to CII, we analyzed CD4⫹ T cells isolated from the spleens of B10.M DR1– and DR4–transgenic mice for their expression of each V␤-chain family, using multicolor flow cytometry. If the DR transgene preferentially selected any particular V␤-chain, the percentage of lymphocytes expressing that V␤ family would be overrepresented in comparison with the levels in the nontransgenic B10.M mice. We found that the repertoires of both the transgenic and nontransgenic mice were very similar. Specifically, V␤8 was the most commonly expressed ␤-chain, with a mean ⫾ SD level of 28.9 ⫾ 0.3% of CD4⫹ lymphocytes in the nontransgenic B10.M mice, 26.3 ⫾ 0.1% of CD4⫹ lymphocytes in the DR1-transgenic mice, and 29.7 ⫾ 0.8% of CD4⫹ lymphocytes in the DR4transgenic mice. V␤14 was expressed in 8.1 ⫾ 0.5% of CD4⫹ lymphocytes in the nontransgenic mice, 8.6 ⫾ 0.6% of CD4⫹ lymphocytes in the DR1-transgenic mice, and 10.3 ⫾ 0.5% of CD4⫹ lymphocytes in the DR4transgenic mice. These data indicate that there were no significant changes in V␤8 or V␤14 expression by T cells from the transgenic mice. However, there were significant changes in the expression of other V␤-chains, including a reduction in cells expressing V␤11 and V␤12 and an increase in V␤6 and V␤10 in both strains of transgenic mice as compared with the B10.M mice (P ⬍ 0.01) (Figure 3). RESTRICTION OF V␤ USAGE IN TCRs RECOGNIZING CII Figure 3. T cell receptor V␤ expression by splenic CD4⫹ T cells in wild-type B10.M, DR1-transgenic, and DR4-transgenic mice. Spleen cells were isolated from naive mice. The V␤ expression was determined by flow cytometry using peridin chlorophyll protein–conjugated antimouse CD4 and fluorescein isothiocyanate–conjugated anti-mouse V␤ subfamily–specific monoclonal antibodies as described in Materials and Methods. ⴱ ⫽ P ⬍ 0.05 and ⴱⴱ ⫽ P ⬍ 0.01 in comparison with wild-type B10.M mice. We then addressed the question of whether presentation of antigens other than CII by the DR molecules preferentially selects T cells expressing V␤14 or V␤8, by immunizing B10.M mice (the background for the transgenics) and DR1 mice with Freund’s complete adjuvant only. The percentages of V␤14⫹ and V␤8⫹ cells from draining lymph nodes were analyzed by flow cytometry 10 days after immunization. There were no significant differences in the V␤14⫹ T cell expression (mean ⫾ SD 9.8 ⫾ 0.8% versus 10.4 ⫾ 1.0%) and V␤8⫹ T cell expression (31.5 ⫾ 1.5% versus 29.5 ⫾ 2.1%) between the nontransgenic B10.M mice and DR1transgenic mice, respectively, indicating that the multiple epitopes from Freund’s complete adjuvant presented by DR1 did not preferentially select V␤14⫹ and V␤8⫹ T cells. It seems, therefore, that although the DR transgenes affect T cell selection on the basis of V␤ gene expression, the V␤14⫹ and V␤8⫹ T cells might not be preferentially selected in either DR1- or DR4-transgenic mice. Usage of CDR3 by CII-reactive T cells in DR1transgenic mice. The PCRs carried out to identify the V␣ and V␤ usage generated products that included the entire CDR3. To determine the structure of this region, 2001 PCR products from T cell hybridomas of DR1transgenic mice were purified and sequenced directly using a C-region nested primer. Each of the 23 hybridomas had unique CDR3 structures in both the V␤- and the V␣-chain, confirming that unique TCRs had been isolated. A common motif in the CDR3 of the V␤-chain of the TCR for T cells from DR1-transgenic mice could not be established, indicating that there is considerable plasticity in the CDR3, and that T cells utilize a multitude of different functional structures to interact with this single peptide presented by DR1 (Table 2). Analysis of the CDR3 of the V␣-chain for these T cells revealed similar characteristics (results not shown). Reduction in the incidence and severity of CIA in DR1-transgenic mice by deletion of V␤14ⴙ and V␤8ⴙ T cells. Since V␤14-bearing lymphocytes represent only a small percentage of the total, it was of interest to determine if they were critical for the development of arthritis. To address this issue, experiments were carried out to study the role of V␤14⫹ T cells, as well as V␤8⫹ T cells, in DR1-transgenic mice in the development of CIA. In these experiments, V␤14⫹ T cells alone, V␤8⫹ T cells alone, or both were selectively depleted by intraperitoneal injection of purified mAb specific for these V␤ subfamilies. The antibodies were injected 3 days before immunization for induction of CIA, and were repeated on the day of immunization. Eight days after the first injection, V␤14⫹ and V␤8⫹ T cells in the peripheral blood were examined by flow cytometry. In the control groups, V␤14⫹ and V␤8⫹ T cells accounted for 8% and 27%, respectively, of the total V␤⫹ T cells. In the group treated by anti-V␤14, the expression of V␤14⫹ T cells was reduced to 0.3% of the total V␤⫹ T cells, and V␤8⫹ T cells were essentially unchanged (28%). In the group treated by anti-V␤8, the expression of V␤8⫹ T cells was reduced to 0.7% of the total V␤⫹ T cells, and V␤14⫹ T cells were essentially unchanged (9%). In the group treated by a combination of anti-V␤14 and anti-V␤8, the levels of both V␤14⫹ and V␤8⫹ T cells were reduced to ⬍0.5% of the total V␤⫹ T cells. These data indicate that the depletion of V␤14bearing and V␤8-bearing T cells by the antibodies was effective. By using anti–V␤-PE antibodies that stain all ␣/␤ T cells, we observed that the deletion of the V␤8⫹ and/or V␤14⫹ T cells led to a corresponding decrease in the total ␣/␤ T cell population, suggesting that deletion of these particular V␤ TCRs might not be compensated for in the short term. The development and characteristics of CIA in the mice were examined beginning at 19 days after 2002 HE ET AL DISCUSSION Figure 4. Reduction in the incidence and severity of collagen-induced arthritis in DR1-transgenic mice by depletion of V␤14-bearing and V␤8-bearing T cells. The mice (n ⫽ 40) were immunized by subcutaneous injection at the base of the tail with 100 g/ml of bovine type II collagen (CII) in Freund’s complete adjuvant at day 0. Among them, 10 mice were treated with anti-V␤14, 10 with anti-V␤8, 10 mice with a mixture of anti-V␤14 and anti-V␤8, and 10 controls with phosphate buffered saline. The injections were given intraperitoneally at day ⫺3 and day 0. Beginning at 19 days after immunization, mice were monitored for the incidence (A) and severity (B) of arthritis. ⴱ ⫽ P ⬍ 0.05. The reduction in the incidence of arthritis by anti-V␤14 was close to, but did not reach, statistical significance (P ⫽ 0.057). The effects of anti-V␤14 were tested in 2 separate experiments, with similar results. immunization. In comparison with the controls, the mice treated by anti-V␤14 or anti-V␤8 mAb exhibited delayed disease onset and showed a reduction in the incidence and severity of CIA. The combination of both anti-V␤14 and anti-V␤8 had the strongest suppressive effect (P ⬍ 0.05) (Figure 4). These data establish the importance of these cells in the pathogenesis of the disease, and suggest that the response could not compensate by using other TCR families. Analysis of patients with RA has established a linkage between HLA–DR and the incidence and severity of disease. It is known that susceptibility to RA is at least partially conferred by a sequence in the third hypervariable region of the ␤1-chain of the DR molecule. Patients with differing but related DR haplotypes from widely separated population groups all share the same amino acid sequence at residues 70–74 of the DR4␤1 molecule (17). Thus, DRB1*0101, DRB1*0401, DRB1*0404, and DRB1*0405 genotypes are all associated with RA and possess this “shared epitope.” However, the precise role of the shared epitope in the development of RA is unknown. The shared epitope could either confer binding specificity for a particular antigenic peptide, shape the development of the T cell repertoire, act as an antigenic structure itself, or act through some other as-yet-undiscovered mechanism. Fugger and coworkers found that mice transgenic for DR4 would develop an immune response to CII (18). We have shown that mice transgenic for either DR1 or DR4 are susceptible to CIA. Immunization of the transgenic mice with human CII induced arthritis with high incidence (8,9). In addition, the same immunodominant epitope was recognized by both strains. In contrast, mice transgenic for DRB1*1502 are resistant to CIA (19). Diab and coworkers found that only the DR alleles associated with RA susceptibility bind a CII peptide, CII256–271, containing the immunodominant epitope for DR1- and DR4-transgenic mice (20). These data suggest that DR molecules with the shared epitope may preferentially predispose to the development of CII autoimmunity by selection of a particular antigenic peptide. To address the question of whether the restricted usage of V␤-chain families was related to a generalized inability of DR1 and DR4 to select other mouse T cells, we analyzed the repertoire of preimmune mice. These animals showed a highly diverse TCR repertoire comparable with that seen in the wild-type B10.M mouse, but with some small differences. The major effect of the transgenes was selection against V␤5, V␤11, and V␤12. However, these TCRs are present at relatively low levels even in nontransgenic animals. The significance of this negative selection bias is not clear. In DR1 mice, there was a relative positive selection of V␤3, V␤6, and V␤10, whereas in DR4 mice, there was positive selection of only V␤6 and V␤10. Thus, the highly restricted usage of V␤8 and V␤14 is not likely to be due to the inability of other mouse TCRs to interact with the transgenes. Other investigators have studied the TCR reper- RESTRICTION OF V␤ USAGE IN TCRs RECOGNIZING CII toire in H-2q mice. It was shown by Osman and coworkers, who characterized 13 clonally distinct T cell hybridomas specific for bovine CII in DBA/1 mice (H-2q), that the TCRs of the hybridomas utilized restricted V␤- as well as V␣-chain subfamilies (21). However, CII-reactive T cells in DR1-transgenic mice use V␤14, V␤8, V␣2, and V␣1, whereas in contrast, DBA/1 mice use V␤8, V␤1, V␤6, V␣11, V␣8, and V␣22. It is interesting that the core of the dominant determinant of CII associated with DR1 is CII263–270 (8), whereas that with H-2q is CII260–267 (22). These epitopes overlap significantly, but the data indicate that the MHC–peptide complexes apparently interact with very different TCR structures. The differences include not only those in the N–D–N and N regions, but also those in the V␤- and V␣-chain, the C terminals of which are an integral part of the CDR3. Diab and coworkers found that *0401 and *0402 bound the same peptide, but closer examination of the binding properties indicated that they were binding the peptide in different registers. Thus, the core determinant for *0401 was 263–270, but was 256–268 for *0402 (20). How the binding of these different, but overlapping, determinants relates to susceptibility to arthritis has not been experimentally tested. Restricted usage of the V␤- and V␣-chain has also been found in T cells involved in experimental allergic encephalomyelitis (EAE). Urban et al analyzed TCR genes of 33 clonally distinct Th cells specific for a nonapeptide of myelin basic protein (MBP) in mice. These T cells used only V␤8.2 and V␤13 and 2 V␣-chains (23). In Lewis rats with EAE, Gold et al studied 15 T cell clones and hybridomas specific for the 21-mer encephalitogenic fragment MBP (68–88). All of them used V␤8.2 (24). It is interesting to note that most studies of TCR usage in autoimmune disease in mice have identified V␤8-gene family members for a substantial portion of the response. The basis for selective use of V␤8 in a wide range of responses is not entirely clear, but in most mouse strains, cells expressing V␤8 account for a high percentage of cells in the preimmune state. The highly restricted usage of the V␤-chain of T cells reactive to CII249–281 allowed us to test the effects of deletion of V␤14 and/or V␤8 on the induction of CIA. As expected, administration of the antibodies against these V␤-chains significantly reduced the incidence and severity of CIA, indicating the importance of the V␤14⫹ and V␤8⫹ CII-specific T cells in the pathogenesis of CIA in the DR-transgenic mice. These results are consistent with those of a previous study in which selective deletion of V␤8⫹ T cells by mAb was successful in preventing CIA in H-2q mice (25). In contrast, it has 2003 previously been shown that B10.Q mice with the V␤a haplotype, deficient in V␤5, V␤8, V␤9, V␤11, V␤12, and V␤13, showed no difference in arthritis susceptibility, onset, or severity when compared with wild-type B10.Q mice, although B10.Q-V␤c mice, which lack V␤6, V␤15, V␤17, and V␤19 families in addition to the V␤a deletion, were somewhat resistant to CIA, indicating that the plasticity was not unlimited (26). It seems, therefore, that at least in mice, there is a substantial ability to compensate for the congenital deficiency of some V␤chains, including V␤8. Although RA remains a disease of unknown etiology, CII, the main component of articular cartilage, has been regarded as a potential autoantigen. Both B and T cells reactive with CII have been identified in the inflamed joints of RA patients (27,28). However, the TCR structures reacting with CII have not been characterized. To explore the structural requirements of TCR cells that react with CII, we generated T cell hybridomas from DR1- and DR4-transgenic mice. Our experiments demonstrated that the V␤ and V␣ usage of T cells recognizing the dominant determinant on CII presented by DR1 and DR4 molecules is highly restricted. In fact, it is restricted to the same V␤ families for both of these DR molecules. In contrast, T cell hybridomas utilize relatively diverse functional structures in the CDR3 to recognize the CII peptide presented by DR1 molecules. Extrapolating to the in vivo situation, it can be expected that TCRs recognizing CII in patients with either DR1 or DR4 might share similar characteristics. However, because of differences in the V␤ structures between humans and mice, it is not possible to accurately predict which human V␤ families may be involved (29). REFERENCES 1. Trentham DE, Townes AS, Kang AH. Autoimmunity to type II collagen: an experimental model of arthritis. J Exp Med 1977;146: 857–68. 2. Wooley PH, Luthra HS, Stuart JM, David CS. Type II collageninduced arthritis in mice. I. Major histocompatibility complex (I region) linkage and antibody correlates. J Exp Med 1981;154: 688–700. 3. Griffiths MM, Eichwald EJ, Martin JH, Smith CB, DeWitt CW. Immunogenetic control of experimental type II collagen–induced arthritis. I. Susceptibility and resistance among inbred strains of rats. Arthritis Rheum 1981;24:781–9. 4. Watson WC, Townes AS. Genetic susceptibility to murine collagen II autoimmune arthritis: proposed relationship to the IgG2 autoantibody subclass response, complement C5, major histocompatibility complex (MHC) and non-MHC loci. J Exp Med 1985; 162:1878–91. 5. Seki N, Sudo Y, Yoshioka T, Sugihara S, Fujitsu T, Sakuma S. Type II collagen-induced murine arthritis. I. Induction and per- 2004 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. petuation of arthritis require synergy between humoral and cellmediated immunity. J Immunol 1988;140:1477–84. Banerjee S, Haqqi TM, Luthra HS, Stuart JM, David CS. Possible role of V␤ T cell receptor genes in susceptibility to collageninduced arthritis in mice. J Exp Med 1988;167:832–9. Griffiths MM, Cole BC, Ito J, Harper DS, Anderson GD, Cannon GW. T-cell receptors and collagen induced arthritis in H-2r mice. Autoimmunity 1993;14:221–9. Rosloniec EF, Brand DD, Myers LK, Whittington KB, Gumanovskaya M, Zaller DM. An HLA-DR1 transgene confers susceptibility to collagen-induced arthritis elicited with human type II collagen. J Exp Med 1997;185:1113–22. Rosloniec EF, Brand DD, Myers LK, Esaki Y, Whittington KB, Zaller DM. Induction of autoimmune arthritis in HLA-DR4 (DRB1*0401) transgenic mice by immunization with human and bovine type II collagen. J Immunol 1998;160:2573–8. Woods A, Chen HY, Trumbauer ME, Sirotina A, Cummings R, Zaller DM. Human major histocompatibility complex class IIrestricted T cell responses in transgenic mice. J Exp Med 1994; 180:173–81. Stuart JM, Cremer MA, Townes AS, Kang AH. Type II collageninduced arthritis in rats: passive transfer with serum and evidence that IgG anticollagen antibodies can cause arthritis. J Exp Med 1982;155:1–16. White J, Blackman M, Bill J, Kappler J, Marrack P, Gold DP. Two better cell lines for making hybridomas expressing specific T cell receptors. J Immunol 1989;143:1822–5. Mosmann T. Rapid colorimetric assay for cellular growth and survival: application to proliferation and cytotoxicity assays. J Immunol Methods 1983;65:55–63. Denizot F, Lang R. Rapid colorimetric assay for cell growth and survival: modifications to the tetrazolium dye procedure giving improved sensitivity and reliability. J Immunol Methods 1986;89: 271–7. Kappler JW, Skidmore B, White J, Marrack P. Antigen-inducible, H-2-restricted, interleukin-2-producing T cell hybridomas: lack of independent antigen and H-2 recognition. J Exp Med 1981;153: 1198–214. Arden B, Clark SP, Kabelitz D, Mak TW. Mouse T-cell receptor variable gene segment families. Immunogenetics 1995;42:501–30. Gregersen PK, Silver J, Winchester RJ. The shared epitope hypothesis: an approach to understanding the molecular genetics of susceptibility to rheumatoid arthritis. Arthritis Rheum 1987;30: 1205–13. HE ET AL 18. Fugger L, Rothbard JB, Sonderstrup-McDevitt G. Specificity of an HLA-DRB1*0401-restricted T cell response to type II collagen. Eur J Immunol 1996;26:928–33. 19. Gonzalez-Gay MA, Nabozny GH, Bull MJ, Zanelli E, Douhan J III, Griffiths MM. Protective role of major histocompatibility complex class II Ebd transgene on collagen-induced arthritis. J Exp Med 1994;180:1559–64. 20. Diab BY, Lambert NC, L’Faqihi FE, Loubet-Lescoulie P, de Preval C, Coppin H. Human collagen II peptide 256-271 preferentially binds to HLA-DR molecules associated with susceptibility to rheumatoid arthritis. Immunogenetics 1999;49:36–44. 21. Osman GE, Toda M, Kanagawa O, Hood LE. Characterization of the T cell repertoire causing collagen arthritis in mice. J Exp Med 1993;177:387–95. 22. Brand DD, Myers LK, Terato K, Whittington KB, Stuart JM, Kang AH. Characterization of the T cell determinants in the induction of autoimmune arthritis by bovine ␣1(II)-CB11 in H-2q mice. J Immunol 1994;152:3088–97. 23. Urban JL, Kumar V, Kono DH, Gomez C, Horvath SJ, Clayton J. Restricted use of T cell receptor V genes in murine autoimmune encephalomyelitis raises possibilities for antibody therapy. Cell 1988;54:577–92. 24. Gold DP, Offner H, Sun D, Wiley S, Vandenbark AA, Wilson DB. Analysis of T cell receptor ␤ chains in Lewis rats with experimental allergic encephalomyelitis: conserved complementarity determining region 3. J Exp Med 1991;174:1467–76. 25. Moder KG, Luthra HS, Kubo R, Griffiths M, David CS. Prevention of collagen induced arthritis in mice by treatment with an antibody directed against the T cell receptor ␣␤ framework. Autoimmunity 1992;11:219–24. 26. Nabozny GH, Bull MJ, Hanson J, Griffiths MM, Luthra HS, David CS. Collagen-induced arthritis in T cell receptor V␤ congenic B10.Q mice. J Exp Med 1994;180:517–24. 27. He X, Kang AH, Stuart JM. Accumulation of T cells reactive to type II collagen in synovial fluid of patients with rheumatoid arthritis. J Rheumatol 2000;27:589–93. 28. He X, Kang AH, Stuart JM. Anti-human type II collagen CD19⫹ B cells are present in patients with rheumatoid arthritis and healthy individuals. J Rheumatol 2001;28:2168–75. 29. Clark SP, Arden B, Kabelitz D, Mak TW. Comparison of human and mouse T-cell receptor variable gene segment subfamilies. Immunogenetics 1995;42:531–40.