Skewed T-cell receptor BV14 and BV16 expression and shared CDR3 sequence and common sequence motifs in synovial T cells of rheumatoid arthritis


T-lymphocytes play an important role in rheumatoid arthritis (RA). In this study, we evaluated the hypothesis that common T-cell receptor (TCR) structural features may exist among infiltrating T cells of different RA patients, if the TCR repertoire is shaped by interaction with common self or microbial antigens in the context of susceptible HLA genes in RA. Synovial lesion tissue (ST), synovial fluid (SF) and blood specimens from RA patients and controls were analyzed for TCR V gene repertoire by real-time PCR. There was highly skewed BV14 and BV16 usage in synovial T cells of RA as opposed to those of controls, which was accompanied with a trend for correlation between skewed BV16 and DRB1*0405. Immunoscope analysis of the V–D–J region of ST-derived T cells demonstrated oligoclonal and polyclonal expansion of BV14+ and BV16+ T cells. Detailed characterization using specific BV and BJ primers further revealed common clonotypes combining the same BV14/BV16, BJ and CDR3 length. DNA cloning and sequence analysis of the clonotypes confirmed identical CDR3 sequences and common CDR3 sequence motifs among different RA patients. The findings are important in the understanding of BV gene skewing and CDR3 structural characteristics among synovial infiltrating T cells of RA.


Rheumatoid arthritis (RA) is characterized by chronic inflammation of the synovium of the peripheral joints, in which T cells are thought to play an important role in the pathogenesis.1, 2 This is supported by marked infiltration and accumulation of Th1 proinflammatory cells in the synovial membrane of RA in close association with MHC Class II genes, including DR4 (genotypes DRB1*0404 and DRB1*0401) and DQ (DQB1*0302 and DQB1*0301), in Caucasian RA patients.3, 4, 5 Further supporting evidence includes skewing of cytokine environment in favor of T cell-mediated inflammation and clonal expansion of infiltrating T cells in the affected joints.6, 7, 8, 9, 10, 11, 12 However, the antigen specificity of the infiltrating T cells in rheumatoid synovium is unknown. Several self-antigens, including collagen type II, heat-shock proteins and others, are implicated in RA based on T-cell reactivity to these antigens in patients with RA.13, 14, 15, 16 Microbial antigens, such as mycobacterial antigens and staphylococcal superantigens, may also contribute to T-cell activation in RA.17, 18 However, it is unclear whether T-cell responses to any of these antigens are clinically relevant to RA.

In the absence of an eliciting antigen(s) associated with RA, attempts have been made to identify T-cell receptor (TCR) structural features characteristic of infiltrating T cells derived from synovial fluid (SF) or synovial membrane of RA patients. It was hoped that a common TCR structural feature(s) or characteristic clonotypes of T cells associated with rheumatoid synovium may provide better understanding of the mechanism whereby these infiltrating T cells are activated and perpetuated in the synovium and may potentially lead to therapeutic strategies. As the TCR repertoire is shaped by the genetic background of the individual and the response to self- or environmental antigens, antigen-driven stimulation in the context of similar MHC Class II molecules leads to oligoclonal expansion of T cells utilizing common V–D–J segments. On the other hand, T-cell activation induced by superantigen stimulation is characterized by polyclonal expansion of a particular TCR BV gene family with different D–J segments. Therefore, it is important to delineate the BV gene distribution pattern and structural features of the third complementarity-determining region (CDR3) among T cells in the rheumatoid synovium.

A number of studies in Caucasian RA patients have shown that TCR BV usage of T cells derived from SF and, in some cases, synovial membranes of RA patients is variably skewed to certain BV genes, including BV14, BV17 and several others.19, 20, 21, 22, 23 Analysis of CDR3 of overexpressed BV genes has revealed some clonotypes that only exist in rheumatoid synovium but not peripheral T cells, suggesting T-cell clonal expansion in the affected joints in RA.24, 25, 26 However, clonality and TCR BV gene usage of infiltrating T cells in RA SF or membranes are relatively heterogeneous, which complicates BV gene analysis using regular or semiquantitative PCR. This may be attributable to significant variations in the detection of overexpressed BV genes and CDR3 structural features of RA synovial T cells reported in different studies. Furthermore, relatively diverse clonality of infiltrating T cells seen in RA has significantly increased difficulties in the identification of common CDR3 structural features. Multiple CDR3 sizing peaks typically found in synovial TCR transcripts of RA require several hundreds to thousands of measurements for each sample when multiplied by the number of BV and BJ genes (25 BV and 13 BJ genes), to characterize clonotypes of interest.27

This study was undertaken to examine the BV usage pattern of infiltrating T cells derived from synovial material of a cohort of Chinese RA patients whose HLA background differed from that of Caucasian patients to determine potential association of BV gene distribution with HLA. It was attempted to identify common clonotypes and CDR3 structural characteristics in overexpressed BV transcripts of synovial T cells from different RA patients. The analyses were performed with TCR transcripts derived ex vivo from synovial material and blood of patients and did not require in vitro culture that is likely to introduce biases. It was hypothesized that infiltrating T cells in the rheumatoid synovium might be driven and shaped by common self- or microbial antigens in the context of RA-associated DR or DQ molecules, and that these T cells might display common or shared TCR structural features among different individuals. The main approach used in this study was to first determine overexpressed BV gene(s) in rheumatoid synovium by quantitative real-time PCR. The analysis was performed in peripheral blood (PB), SF and synovial lesion tissues (ST) of a group of well-defined RA patients and a control group of patients with osteoarthritis (OA). This was then followed by serial CDR3 length analyses within the regions spanning 5′BV–3′BC (clonality analysis) by immunoscope technique. Individual CDR3 length of multiple peaks within the V–D–J region was further dissected using BV- and BJ-specific primers to identify common clonotypes that used the same BV and BJ genes with similar CDR3 length. These clonotypes were then analyzed by DNA cloning and DNA sequencing.


Restricted TCR V gene usage in T cells derived from SF and lesion tissues of patients with RA

A group of well-defined patients with RA and a control group of patients with OA were included in the study as described in the Materials and methods section. Clinical characteristics and HLA DR and DQ genotypes are illustrated in Table 1. In this cohort of Chinese RA patients, genotype DRB1*0405 represented the most dominant DR4 (16/37, 43%) compared to two other DR4 genotypes (DRB1*0401 (8%) and DRB1*0404 (3%)) that are typically associated with Caucasian RA patients.3, 4, 5 In addition, DRB1*09012 (35%) and three DQB1 genotypes (0301, 0303 and 0401) were also frequently expressed in this cohort of RA patients (30–41%).

Table 1 Clinical characteristics and HLA genotypes of patients

We first examined whether T cells derived from RA synovial lesions and SF display restricted TCR BV genes and whether the restricted BV genes are associated with HLA genotypes. To this end, PB and synovial (SF and ST) specimens were first analyzed for TCR BV gene usage by real-time quantitative PCR using 25 specific primers. The real-time PCR method used in this study was sensitive and specific for detection of selective expansion of T cells based on BV expression pattern. As shown in Figure 1, selective expansion of BV2 was readily detected by real-time PCR analysis in four PB mononuclear cell (PBMC) preparations after stimulation with toxic shock syndrome toxin, a superantigen known to activate BV2+T cells. As shown in Figure 2, a series of real-time PCR analyses revealed a highly significant BV skewing for BV14 (mean expression level of 27%), BV16 (mean expression level of 31%) and, to a lesser extent, BV20 (17%) in RA-derived ST specimens. Similar overexpression of BV16 was also observed in SF specimens (28%) obtained from the same RA patients, while BV14 skewing was not significant in paired SF specimens. It was determined that the mean ratio of CD4+ T cells to CD8+ T cells was 1.36±0.74 for PB, 1.41±0.87 for SF and 1.80±1.10 for ST-derived single-cell suspension by flow cytometry. In contrast, BV gene distribution appeared highly heterogeneous in RA-derived PB specimens as well as ST and SF specimens obtained from patients with OA (Figure 2). BV14 and BV16 were not overexpressed in these OA-derived synovial specimens. Furthermore, percentage of Vβ14+ and Vβ16+ cells in single-cell suspensions prepared from fresh ST specimens of 10 randomly selected RA patients was examined. The results revealed that the expression of TCR Vβ14 and Vβ16 was detected predominantly in CD8+ T-cell population of ST specimens of RA patients. Representative flow cytometric profiles are shown in Figure 3.

Figure 1

Optimization of BV-specific primers for real-time PCR analysis and BV gene analysis of peripheral T cells after stimulation with a superantigen. (a) A set of oligonucleotide primers specific for 25 BV family and BC gene were tested with diluted plasmid DNA of each BV gene as templates for PCR amplification efficiency profile by an ABI 7000 Sequence Detection System. Similar slopes of fluorescence intensity (Delta Rn) in function of cycle numbers indicate similar amplification efficiency of TCRBV and TCRBC primers under the PCR conditions as described in Materials and methods. (b) PBMCs were prepared separately from four healthy individuals and cultured in the presence and absence of toxic shock syndrome toxin (TSST-1) at the predetermined concentration of 1 μg/ml for 7 days. Cells were collected and washed by centrifugation, and RNA was extracted for real-time PCR analysis using primers specific for 25 BV genes. The PCR conditions are described in Materials and methods. Results are presented as % expression of each BV gene relative to the total expression of all 25 BV genes. The error bars represent standard deviation.

Figure 2

BV gene distribution of synovial and blood specimens of RA patients and controls by real-time PCR analysis. RNA was extracted from 37 ST specimens paired with 20 PB and SF specimens of the same RA patients as some of SF and PB specimens were not available for analysis. BV gene expression in each transcript was analyzed quantitatively by real-time PCR using specific primers for 25 BV genes. Seven all paired specimens of OA patients were tested as a control. BV gene distribution is presented as mean % expression of each BV gene relative to BC expression on the Y-axis. Student's t-test was used to calculate the statistical difference in BV14 and BV16 expression between the specimen groups. Asterisks represent statistically significant differences (P<0.05). Results are presented as % expression of each BV gene relative to the total expression of all 25 BV genes. The error bars represent standard deviation.

Figure 3

Analysis of TCR Vβ14 and Vβ16 expression of ST specimens by flow cytometry. Single-cell suspensions were prepared from freshly obtained ST specimens of RA patients. Cells were analyzed for the percentage of TCR Vβ14+ or Vβ16+ cells in T-cell populations gated for CD3+, CD4+ and CD8+ subsets, respectively, by flow cytometry using conjugated monoclonal antibodies. Mean percentage of Vβ14+ and Vβ16+ T cells gated on CD8+ subset ranged from 12.8 to 16.2% and from 8.2 to 14.7%, respectively, in ST specimens tested. The mean percentage of Vβ14+CD8+ and Vβ16+CD8+ T cells was less than 2.2 and 0.8%, respectively, in paired PBMC derived from the same patients. Representative flow cytometric profiles are shown as histograms.

Further analysis indicated a trend for correlation between the overexpression of BV16 but not BV14 with the expression of DRB1*0405 in this group of RA patients. Of 37 RA patients analyzed for both DR and DQ genotypes and BV gene usage in ST specimens, mean % expression of BV16 and BV14 was 29 and 11% in DRB1*0405-positive subjects (n=16), and was 21 and 20%, respectively, in DRB1*0405 negative group (n=21). However, the differences did not reach statistical significance. In contrast, the expression level of both BV14 and BV16 was slightly lower in RA patients of other frequently detected genotypes, including DRB1*09012 and DQB1*0301, 0303 and 0401.

Preferential BJ gene usage and CDR3 length analysis of the overexpressed BV14 and BV16 transcripts of RA lesion-derived T cells

Next, we examined clonality of the overexpressed BV14 and BV16 of T cells derived from selected synovial material (>20% relative expression level in these specimens) by CDR3 length analysis using immunoscope technique. As BV14 and BV16 were not overexpressed in ST specimens of OA, four OA ST specimens that exhibited relatively higher expression of BV 14 and BV 16 (6.2–8.8%) were examined as a control. Additional controls included 12 paired PBMC specimens that had detectable expression of either BV14 (mean 8.1 vs an average of 5.7% for all PBMC specimens) or BV16 (mean 4.7 vs an average of 2.9% for all PBMC specimens) by real-time PCR. As illustrated in Figures 4 and 5, both BV14 and BV16 genes exhibited heterogeneous CDR3 length profile in both SF and ST specimens derived from RA patients when two pairs of 5′BV14–3′BC- and 5′BV16–3′BC-specific primers were used to analyze the sequence regions between BV14/BV16–BJ–3′BC. Some ST specimens displayed highly limited clonality with characteristic clonotypes (eg RA2, RA10, RA17, RA20 and RA34 for BV14 and RA12 and RA19 for BV16), while others showed polyclonal patterns. In contrast, highly diverse patterns of clonality were detected in paired PBMC specimens that had relatively higher BV14 or BV16 expression (Figures 4 and 5). Furthermore, both BV14 and BV16 showed polyclonal patterns in OA-derived ST specimens (Figure 6).

Figure 4

CDR3 length analysis of 5′BV–3′BC region of the overexpressed BV14 transcripts. Transcripts of overexpressed BV14 derived from synovial material and PBMC were analyzed for clonality of the 5′BV–BD–BJ–3′BC region by immunoscope using specific primers for 5′BV14–3′BC. CDR3 length is expressed as peak areas (X-axis). Y-axis represents arbitrary units of fluorescence intensity. Selection of BV14 transcripts for analysis was based on the level of BV expression (>20%) in selected specimens.

Figure 5

CDR3 length analysis of 5′BV–3′BC region of the overexpressed BV16 transcripts. Transcripts of overexpressed BV16 derived from synovial material and PBMC were analyzed for clonality of the 5′BV–BD–BJ–3′BC region by immunoscope using specific primers for 5′BV16–3′BC as described in Figure 3 legend. Selection of BV16 transcripts for analysis was based on the level of BV expression (>20%) in selected specimens. BV16 was not expressed in ST specimens of patients RA2, RA17 and RA32.

Figure 6

CDR3 length analysis of 5′BV–3′BC region of BV14 and BV16 transcripts of OA controls. Transcripts of BV14 and BV16 derived from synovial tissue specimens of four OA patients that exhibited relatively higher expression of BV14 or BV16 were analyzed for clonality of the 5′BV–BD–BJ–3′BC region by immunoscope using specific primers for 5′BV14–3′BC and 5′BV16–3′BC.

CDR3 length profile of BV14 and BV16 examined was further dissected and refined by immunoscope using BV14 or BV16 forward primers and reverse primers for 13 BJ genes. A total of 689 CDR3 length profiles were generated from selected transcripts derived from overexpressed BV14 and/or BV16 genes of ST specimens. The analyses revealed several important findings. First, three to four BJ genes were preferentially used in the context of the overexpressed BV14 and BV16. BJ1S1, BJ2S1 and BJ2S7 were preferentially used with BV16 while BJ1S1, BJ2S1, BJ1S4 and BJ2S7 were associated with BV14. Representative examples are shown in Figure 7. Furthermore, some overexpressed BV14 and BV16 transcripts contained common and dominant clonotypes that had the same structural features of BV and BJ with similar CDR3 length, which were present in various ST specimens of different RA individuals. At least three identical clonotypes with CDR3 length of 21, 18 and 12 bp, respectively, were detected in BV16 transcripts with BJ2S1, BJ2S7 and BJ1S1 combinations. Similar common clonotypes also appeared in BV14 with BJ2S1, BJ2S7, BJ1S4 and BJ1S1 combinations in ST-derived TCR transcripts of independent RA patients. Representative clonotype patterns are illustrated in Figure 8. The findings raised the possibility that some of these common clonotypes may have identical CDR3 sequences or common CDR3 sequence motifs. To address this issue, we selected some of the dominant clonotypes for further analysis.

Figure 7

Representative CDR3 length analysis using BV and BJ primers in BV14 and BV16 transcripts. BV14 and BV16 transcripts were further analyzed for CDR3 length profile by immunoscope using BV14 or BV16 primers and a set of primers specific for 13 individual BJ genes, respectively, to identify dominant clonotype patterns in combinations with various BV and BJ. Transcripts derived from ST specimens of two OA patients were included as a control.

Figure 8

Representative clonotype patterns sharing the same BV and BJ combination with similar CDR3 length. Representative clonotypes that had the same BV and BJ combinations with similar CDR3 length (peak areas, 12, 18 and 21 bp for BV16 and BV14) are shown. Selected TCR transcripts of common clonotypes were cloned using TA cloning kits and the resulting DNA clones were subsequently analyzed for CDR3 sequences using corresponding BV primers.

CDR3 sequence analysis of TCR transcripts containing common clonotypes

It is of great interest to address whether the identified clonotypes sharing the same V–D–J structural features have the same CDR3 sequence or common CDR3 sequence motifs among different patients. This could be achieved effectively by characterizing common clonotypes present in TCR transcripts of independent ST specimens. To this end, we selected some of the dominant clonotypes for analysis. Selected TCR transcripts of common clonotypes were cloned into TA cloning vector and the resulting DNA clones were subsequently analyzed for CDR3 sequences using corresponding BV and BJ primers. Each cluster of the BV and BJ combination had approximately 15 independent DNA clones randomly selected for sequence analysis. A total of 490 DNA clones were successfully sequenced. The results indicated that most of individual DNA clones of the same cluster had identical CDR3 sequences of the selected clonotypes, indicating in vivo clonal expansion of T cells carrying the clonotypes. The majority of the clonotypes displayed independent CDR3 sequences specific for each individual. Interestingly, some of these similar clonotypes found in different individuals had identical CDR3 sequences. A CDR3 sequence (QADGTH) was found in BV16–BJ2S7 transcripts of RA12 and RA16 (Table 2), while another CDR3 sequence (SGGSLF) appeared in BV14–2S7 transcripts of RA7 and RA8 (Table 3). Remarkably, as shown in Tables 4 and 5, these clonotypes exhibited shared/common sequence motifs. Motifs QD, LL and WGG were detected in 6/12 BV16 individuals, and the LS motif was found in 5/14 BV14 individuals.

Table 2 CDR3 sequences of BV16 clonotypes derived from ST specimens of RA
Table 3 CDR3 sequences of BV14 clonotypes derived from ST specimens of RA
Table 4 Primers specific for 25 TCRBV genes and TCRBC gene used in real-time PCR analysis
Table 5 Primers for run-off reactions


The results of the study revealed that infiltrating T cells derived from ST of RA patients displayed marked skewing of BV gene distribution toward BV14 and BV16. The analysis was performed quantitatively in a cohort of Chinese RA patients using real-time PCR, which has significant technical advantages over regular or semiquantitative PCR that was employed in previous studies. The skewed BV14 pattern seen in synovial membrane did not appear in paired SF, while the overexpression of BV16 remained in SF. The findings suggest selective activation and accumulation of BV14 and BV16 T cells in the synovial compartment of RA. It is of interest to note that although overexpression of BV14 gene has been reported in T cells derived from RA SF and membrane previously, the discovery of BV16 overexpression in synovial T cells of RA has never been reported previously. Moreover, in this study, we did not detect skewed BV17 and other BV genes that were described in Caucasian RA patients.19, 20, 21, 22, 23 The observations have raised the possibility that in this study, overexpression of BV16 and lack of skewed BV17 and some other BV genes described in other reports may be characteristically associated with Chinese RA patients, while BV14 skewing is common to both Caucasian and Chinese patients with RA. Such discrepancies in BV gene skewing may be attributable to both genetic background (eg HLA genes) and environmental factors of geographic significance. In this regard, it is important to note that this cohort of Chinese RA patients is preferentially associated with DRB1*0405 (43% patients), which is different from two other genotypes of DR4 (DRB1*0404 and DRB1*0401) closely linked with Caucasian RA patients.3, 4, 5 Our study further revealed a trend toward correlation between the overexpression of BV16 but not BV14 in synovial T cells and DRB1*0405 in RA patients. However, the sample size was too small to allow meaningful statistical analysis. In contrast, no correlation was found between the expression level of BV16 and BV14 and the other frequently used DR and DQ genotypes in these RA patients. The finding supports the notion that HLA genotypes and racial background of the individual may influence BV skewing of synovial infiltrating T cells in RA and provides an explanation for characteristic overexpression of BV16 in this Chinese RA population. Although our study did not compare the repertoire of peripheral naïve and memory T cells in RA patients with that in healthy individuals, other studies indicate that there is repertoire contraction in peripheral naïve T cells in RA patients, when compared to healthy controls, without clonal expansion of peripheral memory T cells.28

It is important to note that that the expression of TCR Vβ14 and Vβ16 was detected predominantly in CD8+ T-cell population of ST specimens of RA patients. Clonal expansion of CD8+ T cells in the rheumatoid synovium may suggest recruitment and local expansion of CD8+ T-cell subsets, which is potentially influenced by MHC and/or by encounter with self- or environmental antigen(s). The role of CD8+ T cells in cartilage damage and arthritic inflammation was examined in a recent study by Cohen and Bodmer.29 The authors demonstrated that chondrocytes normally resistant to MHC-mediated cytolysis could be rendered to lysis by CD8+ T cells through antigen recognition. This process required the presence of inflammation that induced upregulation of MHC Class I molecules in target chondrocytes.30 Furthermore, CD8+ T cells could play an important role in the amplification and perpetuation of rheumatoid inflammation through the induction of a proinflammatory cytokine cascade, involving TNF-α and γ-interferon.31

A key aspect of this study is related to the identification of potential identical CDR3 sequences and common CDR3 sequence motifs among infiltrating T cells representing overexpressed BV14 and BV16 populations in the rheumatoid synovium. If synovial T cells of BV14 and BV16 are driven by some common autoantigen(s) associated with RA in the context of similar HLA background, the TCR repertoire may be shaped during the course of the disease to develop T-cell populations of identical CDR3 or common CDR3 structural features among different RA patients. As the V–D–J region pattern of the overexpressed BV14 and BV16 is relatively diverse, such an attempt is a highly difficult task. Our strategy was to first identify similar clonotypes grouped according to common and dominant V–D–J sequence patterns in overexpressed BV14 and BV16. Transcripts containing these common clonotypes were subsequently cloned and analyzed for CDR3 sequences. The findings described here have confirmed that these common clonotypes have identical V–D–J sequences in synovial lesions derived from different patients with RA. It is remarkable that two identical CDR3 sequences are detected in RA lesions between different patients. The observation is in agreement with a previous report that a CDR3 sequence was found in synovial tissues of two RA patients.25 The results are reminiscent of similar findings among T cells recognizing myelin basic protein (MBP), a candidate autoantigen for multiple sclerosis (MS). Oksenberg and co-workers demonstrated that TCR derived from autopsy material of brain lesions of an MS patient shared an identical V–D–J sequence with that of a well-characterized MBP-specific T-cell clone generated from a different MS patient, confirming the presence of MBP-reactive T cells in MS brain lesions.32 More recently, we reported a common CDR3 sequence among MBP-reactive T cells that recognize the immunodominant peptide of MBP in a significant portion of MS patients.33 Furthermore, our sequence analysis of overexpressed BV14+ and BV16+ has revealed multiple CDR3 sequence motifs that are shared or common among synovial T cells of different RA patients. Again, the findings support the notion that CDR3 of overexpressed BV14 and BV16 is not random and may result from T-cell responses to a common, yet unidentified, autoantigen(s) potentially associated with RA. This is consistent with the observation described here that some lesion tissues showed highly restricted oligoclonal patterns by immunoscope analysis of the V–D–J region. However, it should be noted that sequence motifs of BV genes, with the same specificity and restriction, are highly complex, frequently involving nonadjacent residues and amino acids at each position that have chemically conserved side chains. Furthermore, as the clonotype analysis by immunoscope was limited to selected specimens of >20% expression level of the skewed BV gene(s) and DNA cloning/sequencing was performed selectively in representative clonotypes, further investigation using sequence-specific primers (SSP) corresponding to the identified CDR3 sequence may help to evaluate the true frequency of the CDR3 structural features in synovial material of a large population of Chinese RA patients.

Materials and methods

Patients and specimens

A group of 37 patients with RA as determined according to the American Rheumatism Association criteria were included in this study. Seven patients with OA served as a control group. One of the inclusion criteria was that selected patients had not received steroids or other immunosuppressive treatments 2 months prior to the inclusion. Patients who were on symptomatic treatments were not excluded. PBMCs were prepared from heparinized blood specimens by Ficoll–Hypaque gradient separation. Cells of SF were collected from patients of both groups by centrifugation and subsequent washes. STs were obtained by knee synovectomy via arthroscopy from both RA and OA control patients during surgical procedures unrelated to this study. The tissue specimens were cut into small pieces and immediately processed for RNA extraction. The study protocol was approved by the institutional human subjects review board.

Phenotyping and analysis of TCR Vβ expression by flow cytometry

Synovial tissues were processed immediately after surgical removal and cut into small pieces in complete RPMI 1640 medium (Gibco BRL, Life Technologies, NY, USA) containing 10% fetal bovine serum (FBS) and 1% penicillin–streptomycin-L-glutamine. Grinding of small pieces of tissue specimens was performed using a microtissue grinder with Teflon Resin pestle. Single-cell suspension was obtained after filtering through nylon mesh of 400 μm. Analysis of TCR Vβ14 or Vβ16 expression of PB, SF and ST was carried out by double staining using pairs of conjugated mouse anti-human antibodies, CD3-PerCP-Cy5.5, CD4-FITC, CD8-FITC (BD Biosciences, San Jose, CA, USA) and Vβ14-PE or Vβ16-PE (Beckman Coulter Immunotech, Marseille, France). Percentage of positive staining was determined by flow cytometry (FACS Caliber, BD Bioscience, San Diego, CA, USA).

RNA extraction

Total RNA was extracted from experimental materials (PB, SF and ST samples) using the TRIZOL RNA isolation kit (Gibco BRL, Carlsbad, CA, USA). Tissue (ST samples) (50–100 mg) were homogenized by DEPC-treated mortar and pestle in 1 ml of TRIZOL reagent. Cells from PB and SF were directly lysed in 1 ml of TRIZOL reagent. A measure of 0.2 ml of chloroform was added in 1 ml of TRIZOL and mixed vigorously. The preparation was centrifuged and mixed with isopropyl alcohol to precipitate RNA according to the manufacturer's protocol. Total RNA was treated with DNase digestion kit (RNase-free DNase Set, Qiagen GmbH, Hilden, Germany) to remove contaminating genomic DNA.

HLA genotyping

PBMC specimens obtained from all patients were analyzed for HLA DR and DQ genotypes. Briefly, genomic DNA was extracted from EDTA-treated blood of patients and HLA-DRB1 and HLA-DQB1 alleles were determined by PCR with SSP34 using the high-resolution SSP UniTray (PEL-FREEZE Clinical System, Brown Deer, WI, USA). The primer sets amplifying the alleles were described by the international nomenclature committee of WHO ( The panel of HLA-DRB1 alleles and HLA-DQB1 allele were analyzed according to the manufacturer's protocol.

Determination of BV gene usage in synovial and blood T cells by real-time PCR analysis

In all, 25 TCRBV and TCRBC1 (accession number M12887) gene segments were cloned by using the TA Cloning@ kit (Invitrogen, San Diego, CA, USA) and One Shot@ TOP10 Escherichia coli-competent cells (Invitrogen, San Diego, CA, USA) according the manufacturer's protocol. The oligonucleotide sequences of the BV-specific primers are shown in Table 4. cDNA was synthesized from RNA using random primers and Superscript II (Invitrogen, Carlsbad, CA, USA) in a 20-μl reaction. TCR BV gene expression was analyzed by real-time quantitative PCR. An internal reference control for BV–BC amplification and a nontemplate control containing no cDNA were added to each reaction. Real-time PCR was performed in 96-well optical PCR plates on an ABI 7000 Sequence Detection System (Applied Biosystems, Foster City, CA, USA). Briefly, an aliquot of cDNA sample (0.7 μl) was mixed with 25 pairs of BV-specific primers and one pair of BC primers (0.1 mM in final solution), respectively, together with SYBR Green PCR Master Mix (Applied Biosystems, Foster City, CA, USA) to a final reaction volume of 50 μl. The reaction was performed at 50°C for 2 min and at 95°C for 10 min as hot start activation, which was followed by 40 cycles of reaction at 95°C for 15 s and at 60°C for 1 min. The expression of individual BV genes were calculated based on signal intensity of the PCR reactions according to the following formula:

where Ct refers to threshold cycle.

Immunoscope analysis

PCR reactions were performed with 1 μl of cDNA sample derived from ST specimens in the following amplification mixture: 5 μl 10 × PCR buffer (100 mM Tris-HCl, pH 8.3 and 500 mM KCl), 3 μl 25 mM magnesium chloride, 1 μl of 10 mM dNTP mix, 0.5 μl of Taq polymerase (5 U/μl) (Invitrogen, Carlsbad, CA, USA) and 20 pmol of primers (BV14 or BV16 forward primer and BC primer). The PCR amplification profile used was 30 s at 94°C for denaturation, 30 s at 57°C for annealing and 30 s at 72°C for extension in a total of 40 cycles. Immunoscope analysis was performed with a modified protocol.27, 35 A measure of 2 μl of BV14–BC or BV16–BC PCR products were used as templates and run-off reactions were performed with a single internal fluorescent label for each of the 6FAM (expand)-labeled BC or BJ primers (Table 5). The reaction profile consisted of 30 s at 94°C denaturation temperature and 15 cycles at 94°C for 45 s; at 55°C for 45 s, at 72°C for 1 min, followed by 72°C for 5 min as an extension step. The resulting PCR products were then denatured in formamide and analyzed on an Applied Biosystems 3100 Prism using GeneScan 3.7 software (Perkin-Elmer, Boston, MA, USA). Labeled products were analyzed separately as one-color electrofluorographs. The relative intensity of signal (RIS) corresponding to CDR3 length was expressed as the area under the experimental peak divided by the area under the control peak found within a Gaussian distribution. Based on the distribution of the signal intensity, specific BJ primers were selected for sequence analysis of CDR3.

DNA cloning and sequencing analysis of BV14 and BV16 transcripts

PCR products amplified by either BV14 or BV16 forward primer and BC primer from ST samples were used as templates for second run PCR with specific unlabeled BJ primers (Table 5). The second run PCR products were cloned into the TA cloning vector pCR2.1 (Invitrogen, Carlsbad, CA, USA). In all, 15 colonies were picked from each sample for colony PCR using BV14 or BV16 forward primer and a corresponding BJ primer. The positive plasmids that showed visible amplification by PCR were selected. Plasmid DNA was prepared from these samples using QIAPrep mini plasmid kit (Qiagen, Valencia, CA, USA) and V–D–J region was sequenced with either BV14 or BV16 forward primer to determine the sequence of the CDR3 region.


  1. 1

    Lee DM, Weinblatt ME . Rheumatoid arthritis. Lancet 2001; 358: 903–911.

  2. 2

    Lipsky PE, Davis LS . The central involvement of T cells in rheumatoid arthritis. Immunologist 1998; 6: 121–129.

  3. 3

    Kerlan-Candon S, Combe B, Vincent R, Clot J, Pinet V, Eliaou JF . HLA-DRB1 gene transcripts in rheumatoid arthritis. Clin Exp Immunol 2001; 124: 142–149.

  4. 4

    MacGregor A, Ollier W, Thomson W, Jawaheer D, Silman A . HLA-DRB1*0401/0404 genotype and rheumatoid arthritis: increased association in men, young age at onset, and disease severity. J Rheumatol 1995; 22: 1032–1036.

  5. 5

    Fries JF, Wolfe F, Apple R et al. HLA-DRB1 genotype associations in 793 white patients from a rheumatoid arthritis inception cohort: frequency, severity, and treatment bias. Arthritis Rheum 2002; 46: 2320–2329.

  6. 6

    Dolhain RJ, van der Heiden AN, ter Haar NT, Breedveld FC, Miltenburg AM . Shift towards T lymphocytes with a T helper 1 cytokine profile in the joints of patients with rheumatoid arthritis. Arthritis Rheum 1996; 12: 1961–1969.

  7. 7

    Berner B, Akca D, Jung T, Muller GA, Reuss-Borst MA . Analysis of Th1 and Th2 cytokines expressing CD4+ and CD8+ T cells in rheumatoid arthritis by flow cytometry. J Rheumatol 2000; 27: 1128–1135.

  8. 8

    Davis LS, Cush JJ, Schultz-Koops H, Lipsky PE . Rheumatoid synovial CD4+ T cells exhibit a reduced capacity to differentiate into IL-4 poducing T helper-2 effector cells. Arthritis Res 2001; 3: 54–64.

  9. 9

    Goronzy JJ, Bartz-Bazzanella P, Hu W, Jendro MC, Walser-Kuntz DR, Weyand CM . Dominant clonotypes in the repertoire of peripheral CD4+ T cells in rheumatoid arthritis. J Clin Invest 1994; 94: 2068–2076.

  10. 10

    Struyk L, Kurnick JT, Hawes GE et al. T-cell receptor V-gene usage in synovial fluid lymphocytes of patients with chronic arthritis. Hum Immunol 1993; 37: 237–251.

  11. 11

    Gonzalez-Quintial R, Baccala R, Pope RM, Theofilopoulos AN . Identification of clonally expanded T cells in rheumatoid arthritis using a sequence enrichment nuclease assay. J Clin Invest 1996; 97: 1335–1343.

  12. 12

    Alam A, Lambert N, Lule H et al. Persistence of dominant T cell clones in synovial tissues during rheumatoid arthritis. J Immunol 1996; 156: 3480–3485.

  13. 13

    Londei M, Savill CM, Verhoef A et al. Persistence of collagen type II-specific T-cell clones in the synovial membrane of a patient with rheumatoid arthritis. Proc Natl Acad Sci USA 1989; 86: 636–640.

  14. 14

    Pope RM, Pahlavani MA, LaCour E, Sambol S, Desai BV . Antigen specificity of rheumatoid synovial fluid lymphocytes. Arthritis Rheum 1989; 32: 1371–1380.

  15. 15

    Devereux D, O'Hehir RE, McGuire J, van Schooten WC, Lamb JR . HLA-DR4Dw4-restricted T cell recognition of self antigen(s) in the rheumatoid synovial compartment. Int Immunol 1991; 3: 635–640.

  16. 16

    Res PC, Struijk L, Leow A, Daha MR, van den Elsen PC, Breedveld FC . Inflamed joints of patients with rheumatoid arthritis contain T cells that display in vitro proliferation to antigens present in autologous synovial fluid. Functional analysis on the basis of synovial-fluid-reactive T cell clones and lines. Hum Immunol 1994; 40: 291–298.

  17. 17

    Paliard X, West SF, Lafferty JA et al. Evidence for the effects of a superantigen in rheumatoid arthritis. Science 1991; 253: 325–329.

  18. 18

    Holoshitz J, Klajman A, Drucker I et al. T lymphocytes of rheumatoid arthritis patients show augmented reactivity to a fraction of mycobacteria cross-reactive with cartilage. Lancet 1986; 2: 305–309.

  19. 19

    Zagon G, Tumang JR, Li Y, Friedman SM, Crow MK . Increased frequency of V beta 17-positive T cells in patients with rheumatoid arthritis. Arthritis Rheum 1994; 37: 1431–1440.

  20. 20

    Alam A, Lule J, Coppin H et al. T-cell receptor variable region of the beta-chain gene use in peripheral blood and multiple synovial membranes during rheumatoid arthritis. Hum Immunol 1995; 42: 331–339.

  21. 21

    VanderBorght A, Geusens P, Vandevyver C, Raus J, Stinissen P . Skewed T-cell receptor variable gene usage in the synovium of early and chronic rheumatoid arthritis patients and persistence of clonally expanded T cells in a chronic patient. Rheumatology 2000; 39: 1189–1201.

  22. 22

    Jenkins RN, Nikaein A, Zimmermann A, Meek K, Lipsky PE . T cell receptor V beta gene bias in rheumatoid arthritis. J Clin Invest 1993; 92: 2688–2701.

  23. 23

    Williams WV, Fang Q, Demarco D, VonFeldt J, Zurier RB, Weiner DB . Restricted heterogeneity of T cell receptor transcripts in rheumatoid synovium. J Clin Invest 1992; 90: 326–333.

  24. 24

    Li Y, Sun GR, Tumang JR, Crow MK, Friedman SM . CDR3 sequence motifs shared by oligoclonal rheumatoid arthritis synovial T cells. Evidence for an antigen-driven response. J Clin Invest 1994; 94: 2525–2531.

  25. 25

    Mima T, Ohshima S, Sasi M, Nishioka K, Shimizu M, Murata N . Dominant and shared T cell receptor beta chain variable regions of T cells inducing synovial hyperplasia in rheumatoid arthritis. Biochem Biophys Res Commun 1999; 263: 172–180.

  26. 26

    Davey MP, Burgoine GA, Woody CN . TCRB clonotypes are present in CD4+ T cell populations prepared directly from rheumatoid arthritis. Hum Immunol 1997; 55: 11–21.

  27. 27

    Even J, Lim A, Puisieux I, Ferradini L, Dietrich PY, Toubert A . T-cell repertoires in healthy and diseased human tissues analysed by T-cell receptor beta-chain CDR3 size determination: evidence for oligoclonal expansions in tumours and inflammatory diseases. Res Immunol 1995; 146: 65–80.

  28. 28

    Wagner UG, Koetz K, Weyand CM, Goronzy JJ . Perturbation of the T cell repertoire in rheumatoid arthritis. Proc Natl Acad Sci USA 1998; 95: 14447–14452.

  29. 29

    Cohen SE, Bodmer HC . Cytotoxic T lymphocytes recognize and lyse chondrocytes under inflammatory, but not non-inflammatory conditions. Immunology 2003; 109: 8–14.

  30. 30

    Moreland LW, Heck Jr LW, Koopman WJ et al. V beta 17 T cell receptor peptide vaccination in rheumatoid arthritis: results of phase I dose escalation study. J Rheumatol 1996; 23: 1353–1362.

  31. 31

    Feldmann M, Brennan FM, Maini RN . Role of cytokines in rheumatoid arthritis. Annu Rev Immunol 1996; 14: 397–440.

  32. 32

    Hong J, Zang YC, Tejada-Simon MV et al. A common TCR V-D-J sequence in V beta 13.1 T cells recognizing an immunodominant peptide of myelin basic protein in multiple sclerosis. J Immunol 1999; 163: 3530–3538.

  33. 33

    Moreland LW, Morgan EE, Adamson III TC et al. T cell receptor peptide vaccination in rheumatoid arthritis: a placebo-controlled trial using a combination of Vbeta3, Vbeta14, and Vbeta17 peptides. Arthritis Rheum 1998; 41: 1919–1929.

  34. 34

    Tang J, Myracle AD, Allen S, Karita E, Musonda R, Fultz PN . Novel alleles at the lymphotoxin alpha (LTalpha) locus mark extended HLA haplotypes in native Africans. Hum Immunol 2001; 62: 269–278.

  35. 35

    Oksenberg JR, Panzara MA, Begovich A et al. Selection for T-cell receptor V beta-D beta-J beta gene rearrangements with specificity for a myelin basic protein peptide in brain lesions of multiple sclerosis. Nature 1993; 362: 68–70.

Download references

Author information

Correspondence to J Z Zhang.

Additional information

The work was supported by grants from Chinese Academy of Sciences (Bai-ren Program and Grant KSCX2-SW-212), Chinese National Natural Science Foundation (Grant 30028020) and National Projects 863 (Grant 2002AA216121) and Research Grant 2002CCCD2000 of the Chinese Ministry of Science and Technology and by grants from MAXX Genetech.

Supplementary Information accompanies the paper on Genes and Immunity website (

Supplementary information

Supplementary Data 1 (XLS 30 kb)

Supplementary Data 2 (XLS 142 kb)

Rights and permissions

Reprints and Permissions

About this article


  • complementarity-determining region
  • human leukocyte antigens
  • rheumatoid arthritis
  • synovial T cells
  • T-cell receptor

Further reading