Abstract
T cell responses to the immunodominant peptide (residues 83–99) of myelin basic protein are potentially associated with multiple sclerosis (MS). This study was undertaken to examine whether a common sequence motif(s) exists within the TCR complementarity-determining region (CDR)-3 of T cells recognizing the MBP83–99 peptide. Twenty MBP83–99-reactive T cell clones derived from patients with MS were analyzed for CDR3 sequences, which revealed several shared motifs. Some Vβ13.1 T cell clones derived from different patients with MS were found to contain an identical CDR3 motif, Vβ13.1-LGRAGLTY. Oligonucleotides complementary to the shared CDR3 motifs were used as specific probes to detect identical target CDR3 sequences in a large panel of T cell lines reactive to MBP83–99 and unprimed PBMC. The results revealed that, in contrast to other CDR3 motifs examined, the LGRAGLTY motif was common to T cells recognizing the MBP83–99 peptide, as evident by its expression in the majority of MBP83–99-reactive T cell lines (36/44) and PBMC specimens (15/48) obtained from randomly selected MS patients. The motif was also detected in lower expression in some PBMC specimens from healthy individuals, suggesting the presence of low precursor frequency of T cells expressing this motif in healthy individuals. This study provides new evidence indicating that the identified LGRAGLTY motif is preferentially expressed in MBP83–99-reactive T cells. The findings have important implications in monitoring and targeting MBP83–99-reactive T cells in MS.
Multiple sclerosis (MS)3 is an inflammatory demyelinating disease of the CNS. Although the etiology and pathogenesis is unknown, there is an increasing body of evidence suggesting that T cell responses to myelin basic protein (MBP) may play an important role in the disease processes. In particular, T cells recognizing the immunodominant peptide (residues 83–99) of MBP are frequently detected in DR2 patients with MS (1, 2, 3, 4). It was demonstrated that the MBP83–99 peptide had the highest binding affinity to DRB1*1501, which may contribute to its association with the T cell responses to MBP83–99 frequently seen in DR2 patients with MS (5). Although MBP-reactive T cells, including MBP83–99-reactive T cells, were also found in healthy individuals (1, 4, 6), there is evidence that these T cells undergo in vivo activation and clonal expansion in peripheral blood as well as in cerebrospinal fluid in patients with MS, as opposed to healthy individuals (4, 7, 8, 9). These studies lend support for the potential role of MBP-reactive T cells in the pathogenesis of MS and have provided the rationale for development of novel immunotherapy for MS (10, 11, 12).
Although TCR Vα and Vβ gene usage of MBP-reactive T cells has been shown to vary between patients with MS (8, 13, 14), the Vα and Vβ gene rearrangements of MBP-reactive T cells are highly restricted in some MS patients. This finding is consistent with in vivo clonal expansion of MBP-reactive T cells in some patients with MS (8, 13, 14). It is well established that DNA sequence within the complementarity determining region (CDR)-3 encodes hypervariable Vα-Jα junctional region of α-chain and Vβ-Dβ-Jβ junctional region of β-chain, both of which are most critical to Ag recognition (15, 16). It is hypothesized that T cell clones with identical peptide specificity would share conserved CDR3 amino acid sequence motifs that are characteristic of TCR recognition of a given epitope(s). Although identical and related CDR3 sequence motifs were reported previously in MBP-reactive T cell clones, they were largely confined to and specific for given individuals as a result of in vivo clonal expansion (14, 17). The CDR3 motif study has been unrevealing largely because of the limited number of MBP-reactive T cell clones examined in the previous studies and their functional and structural heterogeneity (e.g., different epitope recognition and MHC restriction) (18). However, recent studies indeed suggest that there may be some structural similarities within TCR of T cells that uniformly recognize a single peptide (e.g., MBP83–99) in the context of the same MHC restriction element (17, 19). Oksenberg and colleagues identified a number of repeated motifs in Vβ5.2 T cells isolated from the MS brain plaques of two HLA-identical patients, one of which was identified previously in other human and rat MBP-reactive T cell clones (2, 20).
The identification of common CDR3 motifs among MBP-reactive T cells is critical to our understanding of T cell recognition of MBP and its potential association with the disease processes in MS. It is also important in our attempt to develop a useful TCR marker for identifying and monitoring the activity of MBP-reactive T cells in MS. This study was undertaken to address whether a common CDR3 motif(s) exists in MBP-reactive T cells possessing similar characteristics. Twenty independent T cell clones exclusively recognizing MBP83–99 were analyzed for the CDR3 sequences using reverse-transcribed PCR and DNA sequencing technique. The analysis revealed four shared CDR3 motifs in both Vα-Jα and Vβ-Dβ-Jβ regions. One of the motifs (Vβ13.1-LGRAGLTY) was identical among T cell clones derived from two patients with MS, suggesting that the expression of this CDR3 motif was not restricted to a given individual. A panel of oligonucleotides complementary to various shared motifs were used as probes in a combined PCR-hybridization detection system to trace identical sequence in a large panel of T cell lines specific for MBP83–99 and unprimed PBMC specimens. The study confirmed that, in contrast to other CDR3 motifs, the LGRAGLTY motif is preferentially expressed in MBP83–99-reactive T cells in some patients with MS. The findings also suggest that T cells expressing this common CDR3 motif are present but at low precursor frequency in healthy individuals. The study provides new evidence for further investigating common CDR3 motifs in MBP-reactive T cells and their potential value in monitoring in vivo activity and migration patterns of MBP83–99-reactive T cells in different clinical stages of MS.
Materials and Methods
Reagents and peptides
rIL-2 was purchased from Boehringer Mannheim (Indianapolis, IN). Medium used for cell culture was RPMI 1640 supplemented with 10% heat-inactivated FCS and l-glutamine, sodium pyruvate, nonessential amino acids, and 10 mM HEPES buffer (HyClone, Logan, UT). A panel of overlapping peptides of human MBP (17 aa in length) and a panel of 17 peptides analogue to the 83–99 region of MBP were synthesized by the Merrified solid phase method and were purified by HPLC (courtesy of Dr. Stefen Boheme, Neurocrine Biosciences, San Diego, CA). The analogue peptides were synthesized with single alanine substitutions. The purity of all peptides used in this study was greater than 95%.
MS patients and healthy individuals
Forty-eight patients with clinically definite MS confirmed by magnetic resonance imaging were included in this study (21). The patients were characterized as having relapsing-remitting or chronic progressive MS for more than two years. The patients had not taken any immunomodulatory drugs 2 mo before the study. Consent forms were obtained from the patients after explaining the experimental procedures. The protocol was approved by the Institutional Human Subjects Committee at Baylor College of Medicine. The control blood specimens used in this study were obtained from healthy blood donors (Gulf Coast Blood Center, Houston, TX).
Generation of a MBP83–99-reactive T cell clone from PBMC in MS patients
To generate specific T cell lines (4, 6), PBMC were plated out at 200,000 cells/well in U-bottom plates (Costar, Cambridge, MA) in the presence of the 83–99 peptide (10 μg/ml). Seven days later, all cultures were restimulated with irradiated autologous PBMC pulsed with the peptide as a source of APC. After another week, each culture was examined for specific proliferation to the 83–99 peptide in proliferation assays. Briefly, each well was split into four aliquots (∼104 cells per aliquot) and cultured in duplicate in the presence of 105 APC pulsed with the 83–99 peptide or a control peptide, respectively. Cells were cultured for 72 h and pulsed with [3H]thymidine (Amersham, Arlington Heights, IL) at 1 μCi per well during the last 16 h of the culture. Cells were then harvested, and [3H]thymidine incorporation was measured in a Betaplate counter (Wallac, Turku, Finland). A T cell line was considered to be specific for the 83–99 peptide when the cpm were greater than 1500 (in the presence of the peptide) and exceeded the reference cpm (in the absence of the peptide) by at least threefold (4, 10). Some of these T cell lines were used in the detection of the common CDR3 motif.
To establish stable MBP83–99-reactive T cell clones, the resulting T cell lines were cloned by PHA (Sigma, St. Louis, MO) in the presence of autologous PBMC as accessory cells (10). Briefly, T cells were plated out at 0.3 cell per well under limiting dilution condition and cultured with 105 irradiated autologous PBMC and 2 μg/ml PHA. Cultures were fed with fresh medium containing 50 IU/ml rIL-2 every 3 to 4 days. After ∼10–12 days, growth-positive wells became visible and were tested in proliferation assays for specific responses to the 83–99 peptide.
Proliferation assays with peptides of MBP and analogue peptides
Twenty thousand cells of a given T cell clone were cultured with irradiated autologous PBMC (100,000 cells/well) in the presence of each alanine-substituted peptide (20 μg/ml) and the wild-type 83–99 peptide, respectively. Cultures were set up in duplicate for each peptide. In all cases, cell proliferation was measured after 72 h by [3H]thymidine incorporation assays as described above.
PCR amplifications and direct sequencing of PCR-amplified DNA products
Total RNA was extracted from 106 cells of each MBP83–99-reactive T cell clone using RNeasy mini kit (Qiagen, Santa Clarita, CA). TCR α- and β-chain genes were amplified and directly sequenced as previously described (8, 17). Briefly, extracted RNA was reverse transcribed into first-strand cDNA using an oligo(dT) primer and Reverse Transcriptase (Life Technologies, Gaithersburg, MD). cDNA was then subject to PCR amplification with a set of primers specific for TCR Vα and Vβ gene families, whose sequences were published previously (8, 17). PCR was performed with 1 μl of cDNA in the following amplification mixture: 5 μl of 10× PCR buffer II (100 mM Tris-HCl (pH 8.3) and 500 mM KCl), 3 μl of 25 mM magnesium chloride, 1 μl of 10 mM dNTP mix, 0.3 μl of Taq polymerase (5 U/μl) (AmpliTaq Gold; Perkin-Elmer, Norwalk, CT), 20 pmol of each 5′Vα or 5′Vβ primer as the forward primer, 20 pmol of 3′Cα or 3′Cβ primer as the reverse primer. The PCR amplification profile used was 1 min at 95°C for denaturation, 20 s at 56°C for annealing, and 40 s at 72°C for extension in a total of 35 cycles. The amplified PCR products were separated in a 1% agarose gel by electrophoresis and stained with ethidium bromide. The visualized PCR products were cut and purified subsequently using a QIAquick gel extraction kit (Qiagen) before sequence analysis. The purified PCR products were directly sequenced with the T7 sequencing kit (Pharmacia, Uppsala, Sweden). Ten nanograms of template was sequenced with 2 pmol of the corresponding V gene primer using the method of dideoxy chain termination (22).
CDR3-specific oligonucleotide probes
Oligonucleotide probes corresponding to unique CDR3 sequences were synthesized (Genosis Biotechnologies, The Woodlands, TX) and subsequently labeled with digoxigenin-dUTP using terminal transferase according to the manufacturer’s instructions (Boehringer Mannheim). The sequences of the probes are underlined in Tables I and II. The specificity of these oligonucleotides was confirmed by cross-examination with original clones from which the sequences were derived and with unrelated clones. The labeled CDR3-specific oligonucleotide probes were used to trace similar/identical sequences in short-term T cell lines specific for MBP83–99 and unprimed PBMC specimens as described below.
Detection of CDR3 sequences in short-term T cell lines and PBMC specimens
cDNA products reverse-transcribed from total RNA of the T cell lines and PBMC specimens were analyzed by PCR with a corresponding 5′Vβ forward primer and a 3′Cβ reverse primer as previously described. The amplified PCR products were electrophoretically separated in a 1% agarose gel and transferred to a positively charged nylon membrane (Boehringer Mannheim) using vacuum blot (Bio-Rad, Hercules, CA) at 5 mm Hg for 90 min. DNA was fixed onto the membrane by 3 min-exposure to UV cross-linking and prehybridized at 68°C for at least 1 h. Poly(A) (0.1 mg/ml) was added to prehybridization solution (5× SSC, 1% blocking solution, 0.1% N-lauroylsarkosine, and 0.02% SDS) to reduce nonspecific binding of the probe to non-target DNA. Hybridization temperature and washing conditions were optimized according to different CDR3-specific probes to ensure a stringent hybridization condition. Hybridization was conducted in a buffer containing 5× SSC, 1% blocking solution, 0.1% N-lauroylsarkosine, 0.02% SDS, and 0.3 pmol/ml digoxigenin-labeled CDR3-specific probe for 6 h. The detection of DNA hybrid products was performed using the Digoxigenin Luminescent Detection Kit according to the manufacturer’s instruction (Boehringer Mannheim). The membrane was then exposed to x-ray film for 15–30 min at room temperature. To ensure the specificity, the original T cell clones from which CDR3-specific probes were derived were used as positive control, and MBP83–99 T cell clones of unrelated CDR3 sequences were used as negative control.
DNA cloning and sequencing of PBMC-derived PCR products
PCR products amplified by the 5′Vβ13.1 primer and the 3′Jβ primer from four PBMC specimens positive for the expression of the LGRAGLTY motif were cloned into the TA cloning vector pCR2.1 (Invitrogen, Carlsbad, CA). Plasmid DNA was prepared from transformed Escherichia coli using QIAprep Kit (Qiagen) and screened by PCR with a M13 primer as a forward primer and a reverse primer specific for the LGRAGLTY sequence. The positive plasmids that showed visible amplification by PCR were selected and sequenced for Vβ-Dβ-Jβ sequences with a Vβ13.1 primer.
Semiquantitative measurement of RNA containing the LGRAGLTY motif by RT-PCR and DNA hybridization
First strand cDNA from selected PBMC specimens was amplified by PCR with the 5′CDR3 primer specific for motif LGRAGLTY and a 3′Cβ primer. TCR Cβ was amplified with 5′Cβ and 3′Cβ primers in the same reaction as an internal control. To optimize PCR amplification condition, PCR was performed at different cycles in pilot experiments. A total of 30 cycles was used to ensure that Cβ amplification did not reach a plateau level. The PCR products were then hybridized with a digoxigenin-labeled Cβ-specific probe using the same protocol as described above. The chemiluminescent intensities of hybridized products were visualized on Kodak films and quantified using a Gel Doc 1000 scanning densitometer (Bio-Rad). The level of the LGRAGLTY gene expression was analyzed relative to that of Cβ gene expression (reference gene expression) in each specimen. The results were presented as the ratio of the specific gene expression [(expression of the LGRAGLTY motif/expression of Cβ) × 100%].
Genomic HLA-DR2b typing
HLA-DR2b (DRB1*1501) was identified by genomic typing of the second exon of DRB1*1501 gene that encodes the β-chain of HLA-DR2b. The PCR method used for DR2b typing was described elsewhere with some modification (23). Briefly, cDNA products prepared from PBMC were amplified for the DRB1*1501 allele by specific oligonucleotide primers. The forward primer was 5′-TTCCTGTGGCAGCCTAAGAGG, and the reverse primer was 3′-ACCACCGAATTCTGCACTGTGAAGCTCTCCA, respectively. The primers amplified specifically the DRB1*1501 allele. The conditions used for PCR amplification were as follows: 1 min at 95°C for denaturation, 20 s at 65°C for annealing, and 40 s at 72°C for extension in a total of 30 cycles.
Results
TCR Vβ-Dβ-Jβ sequence and repeated sequence motifs shared among MBP83–99-specific T cell clones derived from different patients with MS
A panel of 20 CD4+ independent T cell clones was generated from seven patients with MS. All T cell clones exclusively recognized the 83–99 peptide of MBP in the context of HLA-DR2, as determined by using mouse fibroblast cells (L cells) transfected with DRB1*1501 as APCs. The T cell clones were analyzed for TCR V gene rearrangements by RT-PCR using Vα- and Vβ-specific primers and were subsequently sequenced for the Vα-Jα and Vβ-Dβ-Jβ junctional regions. As shown in Table I⇓ and Table II⇓, although the Vα and Vβ rearrangements varied between individual MBP83–99 T cell clones, many of these independent T cell clones derived from a given individual expressed the same Vα- and Vβ-chains with identical Vα-Jα and Vβ-Dβ-Jβ junctional region sequences. This finding is consistent with in vivo clonal expansion of MBP83–99-specific T cells in given patients with MS, as reported previously (7, 8). Interestingly, four shared motifs consisting of at least three identical amino acids were found within Vα-Jα and Vβ-Dβ-Jβ junctional regions of these T cell clones derived from different patients with MS (Tables I and II). These include the QDR motif shared by clones Vβ9:MS7-E3.1 and Vβ7:MS37-D9.3, the FGN motif shared by Vα9:MS7-D2.2, Vα17:MS27-C3.1, Vα17:MS27-D7.16, and Vα17:MS27-F3.4 and the S-GGSN motif shared by two clones Vα17:MS7-E2.6 and Vα22:MS7-E3.1 derived from patient MS-7. Remarkably, as indicated in Table II⇓, an independent T cell clone (clone MS7-E2.6) derived from one patient (MS7) expressed the same TCR Vα17 and Vβ13.1 genes in 3/4 T cell clones (MS27-C3.1, MS27-D7.16, and MS27-F3.4) obtained from another patient (MS27). Vβ13.1 of these T cell clones shared an identical sequence (LGRAGLTY) within the Vβ-Dβ-Jβ junctional region while their Vα17 chains had two distinct Vα-Jα junctional region sequences.
TCR Vα gene sequence specific for MBP83-89 peptidea
TCR Vβ gene sequence specific for MBP83-99 peptidea
A complete panel of alanine-substituted analogue peptides of the 83–99 region of MBP were used to further examine whether the two representative MBP83–99 T cell clones (MS7-E2.6 and MS27-C3.1) had a similar recognition pattern. The results revealed that the two clones exhibited distinct recognition patterns characteristic of individual T cell clones (Fig. 1⇓). It appeared that Val87 and the three C-terminal amino acids at Arg97, Thr98, and Pro99 were critical to peptide recognition of clone MS7-E2.6 but were not required by clone MS27-C3.1. Since both T cell clones shared the same Vβ-chain with identical Vβ-Dβ-Jβ sequences, the finding suggests that the structural difference in recognition of the 83–99 peptide is attributable to the distinct Vα-Jα junctional sequences. Conversely, two independent clones, MS7-E2.6 and MS7-E3.1, had distinct Vα-chains (Vα17 and Vα22) with a similar Vα-Jα junctional sequence (SPGGSN vs SRGGSN) while their Vβ rearrangements were completely unrelated. They also exhibited a distinct recognition pattern toward alanine-substituted peptides (Fig. 1⇓). All these T cell clones sharing similar motifs were found to have distinct recognition patterns toward alanine-substituted peptides (data not shown). Taken together, these results suggest that identical Vβ-Dβ-Jβ sequence and repeated Vα-Jαand Vβ-Dβ-Jβ sequence motifs exist among DR2-restricted MBP83–99 T cells derived from different patients with MS.
Reactivity patterns of three MBP83–99 T cell clones to analogue peptides with single alanine substitutions. Three MBP83–99 T cell clones that exhibited identical Vβ13.1 rearrangements (for MS7-E2.6 and MS27-C3.1) and a similar Vα-Jα junctional sequence (for MS7-E2.6 and MS7-E3.1), respectively, were examined for reactivity to a panel of alanine-substituted peptides in [3H]thymidine incorporation assays. A mouse fibroblast cell line expressing DRB1*1501 was used as a source of APCs. The proliferative responses of the clones to each analogue peptide were measured after 72 h, and the results are presented as cpm incorporated. The shaded boxes represent >60% inhibition in the proliferation of the T cell clones in response to various analogue peptides.
CDR3-specific oligonucleotides were highly specific and sensitive in detecting target DNA sequences in original T cell clones as well as in PBMC containing original MBP83–99 T cells
A set of 22 oligonucleotides, including 13 Vβ-Dβ-Jβ sequences and 9 Vα-Jα sequences, were synthesized according to the identified CDR3 DNA sequences of independent MBP83–99 T cell clones. The DNA sequences of these oligonucleotides are shown as underlined in Tables I and II. The specificity of the CDR3-specific oligonucleotides was examined in RT-PCR using the oligonucleotides as forward primers and 3′Cα or 3′Cβ as a reverse primer. The CDR3-specific oligonucleotides were able to amplify the DNA sequences in the original T cell clones from which the CDR3-specific oligonucleotides were derived. Fig. 2⇓ illustrates the specificity of the TCR Vβ CDR3-specific oligonucleotides in detecting target DNA sequences in original MBP83–99-reactive T cell clones. No cross amplification with unrelated MBP83–99 T cell clones was observed, with the exception of the two clones MS7-E2.6 and MS27-C3.1, which shared an identical Vβ-Dβ-Jβ junctional sequence (Fig. 2⇓). Experiments with a set of nine Vα CDR3-specific oligonucleotides revealed the same specificity (data not shown). However, no cross-amplification was seen between clone MS-E2.6 and clone MS27-C3.1 (the same Vα17 but distinct Vα-Jα sequences).
Cross-examination of the specificity of CDR3-specific oligonucleotides with original and unrelated T cell clones. A set of oligonucleotides specific for TCR Vβ-Dβ-Jβ regions were examined for their specificity in detecting known target DNA sequences present in original MBP83–99 T cell clones. PCR reactions were performed using CDR3-specific oligonucleotides as the forward primers and a 3′-Cβ primer as the reverse primer. Solid boxes represent positive detection of DNA sequences present in original T cell clones or T cell clone(s) sharing the same CDR3 sequences. All oligonucleotides were also examined for their binding to DNA products of randomly selected T cell clones that had unrelated CDR3 sequences (shaded boxes).
To detect target DNA sequences in unprimed PBMC specimens, we employed a more sensitive detection system that combined RT-PCR with DNA hybridization using the CDR3-specific oligonucleotides as probes. The CDR3-specific oligonucleotides were labeled with digoxigenin to hybridize, under stringent hybridization conditions, with PCR products amplified from cDNA of PBMC specimens using a corresponding 5′Vα/Vβ forward primer and a 3′Cα/Cβ reverse primer. The pilot experiments confirmed highly specific and sensitive detection of the expected target DNA sequences in original PBMC specimens from which the MBP83–99-reactive T cell clones were derived. As expected, there was no cross-hybridization between unrelated CDR3-specific probes. Thus, the experiments described above demonstrated that the combined PCR-DNA hybridization detection system in which CDR3 region sequence served as fingerprint was highly specific and sensitive in tracing MBP83–99-reactive T cells present in PBMC by probing target DNA sequences.
The detection of a common CDR3 motif in unprimed PBMC specimens from different patients with MS and healthy individuals
Next, we examined whether target DNA sequences corresponding to the identified CDR3 motifs could be detected in PBMC specimens randomly selected from a group of patients with MS and healthy individuals. For this purpose, the above-mentioned detection system that combined RT-PCR and DNA hybridization using CDR3-specific oligonucleotide probes was employed. As shown in Fig. 3⇓, the results indicate that only one oligonucleotide probe corresponding to the Vβ-Dβ-Jβ sequence (LGRAGLTY) shared by clone MS7-E2.6 and clone MS27-C3.1 detected target DNA sequence(s) in the majority of PBMC specimens (15/48, 31%) obtained from different patients with MS. However, two CDR3-specific probes corresponding to their Vα17 sequences (MS7-E2.6 and MS27-C3.1) failed to detect target DNA sequences in the same PBMC specimens. This finding suggests that Vβ13.1 T cells expressing LGRAGLTY motif probably pair with different Vα-chains in PMBC specimens examined. No target sequences were detected by the remaining CDR3-specific probes, including those that exhibited the shared CDR3 motifs in the PBMC specimens, with the exception of autologous PBMC specimens containing original MBP83–99-reactive T cells from which the sequences of the probes were derived (Fig. 3⇓). Furthermore, the expression of the LGRAGLTY motif did not closely correlate with the expression of DRB1*1501 in the PBMC specimens examined (χ2 = 1.71, p > 0.05), even though 53% of the PBMC specimens that exhibited the LGRAGLTY motif expressed the DRB1*1501 molecule. All experiments described above were repeated with reproducible results.
Detection of target DNA sequence complementary to motif Vβ13.1-LGRAGLTY in randomly selected PBMC specimens derived from patients with MS. cDNA prepared from PBMC specimens from randomly selected MS patients (n = 48) were first amplified in RT-PCR using a 5′Vβ13.1-specific primer and a 3′Cβ primer. The amplified PCR products were then hybridized subsequently with a digoxigenin-labeled oligonucleotide probe specific for the LGRAGLTY motif. The original MBP83–99 clone (MS7-E2.6) and an unrelated T cell clone (MS32-B9.8) were used as positive and negative controls, respectively. The MS code used in this figure corresponds to that shown in Tables I and II. MS-7 and MS-27 were the original PBMC specimens from which clone MS7-E2.6 and clone MS27-C3.1 were derived. Asterisks indicate positive expression of DRB1*1501.
The CDR3-specific oligonucleotide probes were then used under the same experimental conditions to detect target DNA sequences in a panel of PBMC specimens derived from healthy individuals. As shown in Fig. 4⇓, in contrast to other CDR3-specific probes, the LGRAGLTY probe detected the target sequence in 5/20 control PBMC specimens. Additional experiments to quantify the specific expression of the LGRAGLTY motif using semiquantitative analysis revealed lower expression of the motif in control PBMC specimens than those from patients with MS (Fig. 5⇓).
Detection of the Vβ13.1-LGRAGLTY motif in randomly selected PBMC specimens derived from normal subjects. PBMC specimens obtained from 20 normal subjects (ns) were analyzed under the same condition as described in the Fig. 3⇑ legend. The original clone (MS7-E2.6) and an unrelated T cell clone (MS32-B9.8) were used as positive and negative controls, respectively. Asterisks indicate positive expression of DRB1*1501.
Semiquantitative comparison of the expression of the LGRAGLTY motif in PBMC specimens derived from MS patients and normal subjects. The specific expression of motif Vβ13.1-LGRAGLTY was analyzed by semiquantitative PCR relative to the Cβ expression in each cDNA derived from PBMC of MS patients and normal individuals. PBMC specimens analyzed corresponded to those expressing the LGRAGLTY motif in Fig. 3⇑ and Fig. 4⇑. The chemiluminescent intensities of hybridized products were quantified using a densitometer (see Materials and Methods). The percentage of relative expression of the motif is calculated as (expression of the LGRAGLTY motif/expression of Cβ) × 100% and is presented in the histogram. NS, normal subjects. MS, patients with MS.
The LGRAGLTY motif is preferentially expressed in short-term MBP83–99-reactive T cell lines in some patients with MS
Short-term MBP-reactive T cell lines are generated by limited passages in culture, thus representing relatively preserved T cell repertoire to MBP in a given individual. Therefore, it is of interest to examine the percentage of short-term MBP83–99 T cell lines that exhibited the LGRAGLTY in patients with MS. As showed in Fig. 6⇓, the experiments revealed that the LGRAGLTY motif was expressed in the majority of short-term MBP83–99-reactive T cell lines (36/44, 82%) derived from five patients with MS. In some patients with MS (MS26 and MS36), all T cell lines expressed the LGRAGLTY motif. These results confirmed that the LGRAGLTY motif is preferentially expressed in T cells recognizing MBP83–99 in some MS patients.
Detection of the Vβ13.1-LGRAGLTY motif in short-term MBP83–99 T cell lines derived from patients with MS. A panel of 44 independent short-term MBP83–99 T cell lines was generated from five patients with MS using a synthetic 83–99 peptide of MBP. All these T cell lines were confirmed for their specific reactivity to MBP83–99 peptide as determined by stimulation index of at least 5 (cpm in the presence of the peptide/cpm in the absence of the peptide). cDNA products were amplified using a 5′Vβ13.1-specific primer and a 3′Cβ primer in PCR. The amplified PCR products were hybridized subsequently with a digoxigenin-labeled oligonucleotide probe specific for the Vβ13.1-LGRAGLTY motif. cDNA products derived from the original MBP83–99 clone (MS7-E2.6) and an unrelated T cell clone (MS32-B9.8) were used as positive and negative controls, respectively.
Verification of the LGRAGLTY motif in PBMC specimens by recombinant DNA cloning and sequencing
Next, we conducted a series of experiments to verify whether the DNA products amplified from PBMC contained the LGRAGLTY sequence using recombinant DNA cloning and sequencing techniques. The experiments were performed with randomly selected PBMC specimens expressing the LGRAGLTY motif using the experimental strategy illustrated in Fig. 7⇓. PCR products were obtained by amplification of cDNA using Vβ13.1-Jβ primers. The amplified PCR products were then ligated into pCR2 vector and transformed into E. coli. Plasmid DNA was prepared and screened for positive insertions by PCR using the CDR3-specific primer. The plasmids that contained Vβ-Dβ-Jβ inserts were analyzed for DNA sequence. The results confirmed that all PBMC specimens obtained from four MS patients and one healthy individual contained the LGRAGLTY sequence in the Vβ-Dβ-Jβ junctional region.
Experimental procedure of cloning and sequencing of PBMC-derived PCR products. cDNA derived from PBMC specimens were amplified by the 5′Vβ13.1 primer, and the 3′Jβ primer from four PBMC specimens positive for the expression of the LGRAGLTY motif was ligated into the TA cloning vector pCR2.1 and transformed into E. coli. Plasmid DNA was screened by PCR with an M13 primer and the LGRAGLTY-specific primer. The positive plasmids that showed visible amplification by PCR were sequenced for Vβ-Dβ-Jβ sequences with a Vβ13.1 primer.
Discussion
In this study, a PCR detection system combined with DNA hybridization in which CDR3 sequences were used as fingerprint provided a powerful tool in tracing MBP83–99-specific T cells by detecting similar or identical CDR3 sequences. The high specificity and sensitivity of the PCR-based detection system allowed the identification of specific CDR3 motifs in short-term T cell lines and unprimed peripheral blood T cells. The present study demonstrated for the first time that a subset of Vβ13.1 T cells that uniformly recognize the immunodominant 83–99 peptide of MBP express a common Vβ-Dβ-Jβ sequence. These T cells are present in more than 30% of patients with MS. The conclusion is made based on step-wise experiments described herein. The common CDR3 motif was identified in independent MBP83–99 T cell clones derived from different individuals and in short-term MBP83–99-reactive T cell lines as well as in unprimed PBMC specimens. The presence of this common CDR3 motif in PBMC specimens was further confirmed by recombinant DNA cloning and sequencing.
The findings described in this study raised several important issues pertinent to our understanding of the potential role and structural characteristics of T cells recognizing the immunodominant 83–99 peptide of MBP in MS. First, it is remarkable that the identified common CDR3 motif (LGRAGLTY) is specific for MBP83–99-reactive T cells and is not restricted to a given individual with MS. In contrast, other shared CDR3 motifs identified in this study were not detected in PBMC specimens derived from different individuals. The findings indicate that, although common CDR3 motifs are rarely detectable in T cell clones, they do exist among T cells possessing uniform characteristics in different individuals. Second, MBP83–99-reactive T cells commonly expressing the LGRAGLTY motif carry distinct Vα-chains with unrelated Vα-Jα sequences. This finding suggests that, although these MBP-reactive T cells displayed the conserved Vβ-Dβ-Jβ sequence, they are structurally different from each other and most likely originate from different clonal lineages. They also appear to exhibit different recognition patterns toward the immunodominant region of MBP, as demonstrated in their reactivity to alanine-substituted peptides of MBP83–99. The results are consistent with considerable heterogeneity in the recognition patterns of T cell clones specific for the MBP-83–99 peptides, even though they display limited TCR V gene rearrangements (17, 24). Furthermore, it is not surprising that MBP83–99 T cells expressing the common CDR3 motif are also present in some healthy individuals. Interestingly, our study demonstrated that these T cells are present at much low precursor frequency in healthy individuals, as evident by quantitatively low expression of the LGRAGLTY motif in PBMC specimens derived from healthy individuals as compared with that from patients with MS. This finding is in agreement with the previous observation that some clonal populations of MBP-reactive T cells undergo in vivo activation and expansion in patients with MS, as opposed to healthy individuals (7, 25, 26). Thus, the results suggest that the MBP83–99-reactive T cells expressing the LGRAGLTY motif may undergo in vivo activation and expansion in at least some patients with MS.
These MBP83–99 T cells expressing the LGRAGLTY motif may represent a significant fraction of MBP83–99 T cells found in some patients with MS. This possibility is supported by the observation that T cells expressing this CDR3 motif seem to represent the majority of short-term MBP83–99 T cell lines in some patients with MS and are frequently found in unprimed PBMC specimens. The reason why this common CDR3 motif was not detected in MBP-reactive T cell clones in previous studies may be explained by several possibilities: 1) only a very limited set of long-term MBP83–99 T cell clones expressing Vβ13.1 have been studied so far (18); and 2) the majority of MBP-reactive T cell lines do not survive repeated stimulation cycles and cloning procedure because of various inhibitory mechanisms (e.g., Ag-induced apoptosis) associated with cell culture and Ag stimulation (27, 28, 29). Thus, the current protocol used by most of the investigators dictates that the majority of MBP-reactive T cells would perish during the process in generating T cell clones and never reach the clonal stage that is necessary for TCR analysis. It is most likely that long-term MBP-reactive T cell clones available for TCR analysis represent a small fraction of MBP-reactive T cells. The TCR repertoire of long-term T cell clones that have survived repeated passages in culture is highly skewed. Therefore, it remains unclear whether the LGRAGLTY sequence described in this study represents the predominant motif in MS patients and whether there are more common CDR3 motifs among MBP-reactive T cells that are unidentified.
It is also tempting to consider potential clinical implications of the findings. The identified common CDR3 motif may be used as a specific marker in a quantitative PCR detection system to trace a subset of MBP83–99 T cells in the blood and cerebrospinal fluid in a large group of MS patients for the purpose of monitoring their activity in vivo. This method will be superior to conventional cell culture-based assays for the reasons discussed above. This is consistent with a recent study where the frequency of MBP-reactive T cells was found to be surprisingly high in patients with MS when direct ex vivo analysis was employed to quantify MBP-reactive T cells (29). Furthermore, synthetic peptides corresponding to the TCR V gene elements have been shown to induce antiidiotypic T cell responses to MBP-reactive T cells in patients with MS (30). Therefore, a TCR peptide containing a common CDR3 sequence may be of great potential in eliciting antiidiotypic T cells to suppress a specific subset of MBP-reactive T cells in a group of patients whose T cells bear the common CDR3 motif. Immunization with a common CDR3 peptide would be advantageous over CDR2 peptides or individual-dependent CDR3 peptides as a potential treatment procedure in patients with MS (11). At this time, however, these potential applications seem to be limited by the fact that the LGRAGLTY motif identified in this study is present only in a subset of Vβ13.1 MBP83–99 T cells. Further investigations using the same approach may lead to discovery of additional common CDR3 motifs in MBP-reactive T cells, making these applications possible.
Footnotes
-
↵1 This work was supported in part by research grants from the National Institutes of Health, the National Multiple Sclerosis Society, the Methodist Hospital Foundation, and the Richardson Foundation. Y.C.Q.Z. is recipient of an advanced postdoctoral fellowship award from the National Multiple Sclerosis Society.
-
↵2 Address correspondence and reprint requests to Dr. Jingwu Zhang, Department of Neurology, Baylor College of Medicine, 6501 Fannin Street, NB302, Houston, TX 77030. E-mail address: jzang{at}bcm.tmc.edu
-
↵3 Abbreviations used in this paper: MS, multiple sclerosis; MBP, myelin basic protein; CDR, complementarity-determining region.
- Received April 27, 1999.
- Accepted July 7, 1999.
- Copyright © 1999 by The American Association of Immunologists
References
In this issue
