| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
*
Laboratoire d’Immunologie et d’Histocompatibilité, Institut National de la Santé et de la Recherche Médicale, Unit 396, Université Paris VII, Centre G. Hayem, Hôpital Saint-Louis, Paris, France;
Institut National de la Santé et de la Recherche Médicale, Unit 463, Institut de Biologie, Nantes, France;
Service de Rhumatologie A, Hôpital Cochin, Paris, France;
§
Service de Rhumatologie, Hôpital Saint-Antoine, Paris, France; and
¶
Service de Rhumatologie A, Hôpital Lariboisière, Paris, France
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The partner of the HLA-B27/antigenic peptide complex recognition is the
ß TCR heterodimer. The arthritogenic peptide hypothesis supposes a
limited diversity of the T cell repertoire, since both HLA-B27 and the
putative antigenic peptide(s) could be shared by different patients,
especially in the case of bacteria-triggered ReA. T cell repertoire
studies in rheumatoid arthritis (13) revealed more frequent T cell
expansions at the pathogenic site than in PBL and that some were shared
in different joints of the same patient and persisted over time (14, 15). Overall, these data suggest an in situ Ag-driven immune response.
With regard to HLA-B27-related arthritis, lack of a preferential usage
of the TCR ß V regions encoding genes (16) but the occurrence of
multiple T cell expansions (17) have been reported in PBL and SF T
cells from ReA patients. Analysis of cytolytic CD8+
HLA-B27-restricted clones produced in vitro from the synovial fluid of
three ReA patients revealed limited use of BV families (BV13, 14, 17)
and BJ segments (18) as well as some structural constraints in the
ß-chain hypervariable complementarity-determining region 3 (CDR3)
without sequence homology (19). The TCR ß-chain is produced by the
combination of V, D, J, and C gene segments. In addition, this
combinatorial diversity is increased by the nibbling of germline
nucleotides and addition of N and P residues at the V-D-J junction
sites. The crucial nature of the CDR3 region encompassing the V-D-J
junction in the contact with the central part of the peptide in the
MHC-peptide complex was shown by crystallography (20). Two methods
called spectratyping (21) and Immunoscope (22) have been developed to
determine the size of CDR3 regions in transcripts of whole BV families
or in given BV-BJ combinations without any requirement for in vitro
cell culture and are therefore particularly suitable for analysis of
ß T cell clonality in complex clinical situations: tumors,
autoimmune diseases, and graft-vs-host disease (23). We took advantage
of this approach to compare the ß T cell repertoire in paired PBL
and SF samples of seven HLA-B27-positive ReA or ankylosing spondylitis
patients. Some oligoclonal expansions defined by the BV, BJ gene
segments, and the CDR3 size were found in common in different patients.
CDR3 amino acid sequence comparisons revealed strikingly conserved
patterns. Identical sequences were observed in expansions, firstly
using different BV genes in one patient and secondly between two
different ReA patients. One sequence was identical with a previously
reported sequence of a cytolytic clone derived in vitro from a
Salmonella-triggered ReA patient (18). Finally, a
CD8+ SF line from the ReA patient whose SF lymphocytes
expressed these sequence identities recognized several B*2705
expressing EBV-transformed lymphoblastoid cell lines (B-LCL). These
results strongly support the arthritogenic peptide hypothesis and argue
for an Ag-driven skewing of the T cell repertoire common to
patients with ReA triggered by different bacteria.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
PBL from 19 HLA-B27-positive patients fulfilling the European
Spondylarthropathy Study Group diagnostic criteria for SA (1)
were included in the study and compared with samples from 10 healthy
HLA-B27-positive individuals obtained from the Blood Bank facility of
our institution. Among the 19 patients, 12 had classical ankylosing
spondylitis, three had psoriatic arthritis, and four had ReA with a
bacterial triggering agent defined in three cases (Chlamydia
trachomatis, Shigella sonnei, Yersinia enterocolitica 0:3).
Patient PG developed ReA after an enteric infection contracted on
travels in Peru where Salmonella is a frequent cause of ReA
(24). SF lymphocytes were obtained under informed consent from knee
effusions during SF analysis before steroid administration in three
ankylosing spondylitis patients and four ReA patients. Two ReA patients
were studied at the onset of the disease (patients PG and CN), while
the remaining two (patients CF and CL) had acute relapses. The main
clinical features, HLA class I typing, and treatment at the time of
sample collection are shown in Table I
for these seven
patients who where extensively studied.
|
PBL were obtained by density gradient centrifugation (Ficoll/Hypaque) and SF cells were isolated directly by three successive washes in PBS. RNA was extracted from cell pellets frozen in liquid nitrogen by lysis in guanidium thiocyanate buffer, and the cDNA was prepared from 5–10 µg total RNA with AMV reverse transcriptase (cDNA cycle kit, Invitrogen, Leek, The Netherlands) as previously described (25).
Oligonucleotides and CDR3 size analysis
The nomenclature of BV families and the primers used have been previously described (25, 26). Fluorescent primers for BC, BJ, and the BV1-BJ2S3 clonotypic primer (5'-CTGCGTATCTGTGGAATATAGA-3') were labeled at the 5' end with the Fam fluorophore (Applied Biosystems, Foster City, CA). Aliquots of the cDNA synthesis reaction (corresponding to 250 ng total RNA) were amplified in 50-µl reactions with one of the BV-specific oligonucleotides as the 5'-primer and the BC oligonucleotide as the 3'-primer. The final concentration was 0.5 mM for each primer, 0.2 mM dNTP, and 2 mM MgCl2 in Taq polymerase buffer (Promega, Madison, WI) in the presence of 1 U of Taq polymerase (Promega) on a DNA thermal cycler (model 9600, Perkin-Elmer, Norwalk, CT). The PCR cycle profile was denaturation at 94°C for 30 s, annealing at 60°C for 45 s, and primer extension at 72°C for 45 s for 40 cycles and a final polymerization step of 5 min at 72°C. Aliquots from each BV-BC PCR product (2 µl) were copied in six cycle run-off reactions primed with a fluorophore labeled BC, BJ, or clonotypic-specific oligonucleotide. The final concentrations were 0.2 mM dNTP and 3 mM MgCl2 in the presence of 0.2 U of Taq polymerase. The run-off reactions were migrated on 4.25% acrylamide sequencing gels (377A DNA sequencer, Applied Biosystems) for size (Genescan-500 size marker, Perkin-Elmer) and fluorescence intensity determination. The raw data were analyzed with the help of the Immunoscope software (22). The CDR3 region was defined to include residues 95–106 (23). Since the positions of the BV and the BC primers are fixed, the length distribution observed in the PCR fluorescent BV-BC products depends only on the size of the V-D-J junctions. Statistical analysis was performed to determine whether a profile could be considered Gaussian; a profile was not considered Gaussian if one peak was excluded from the 95% confidence interval of peak level intensities. TCRB subfamilies BV10 and BV19 were omitted from this analysis as they are pseudogenes in most individuals (27).
BV and BJ gene usage
A competitive PCR was used as previously described (25) to
quantify the TCR transcripts in each cDNA sample using a deleted (4-bp)
CD3 
chain plasmid. About 3 x 106 copies of cDNA
from each sample were then amplified for 30 cycles with the BV primers
and an internal fluorescent BC primer. BJ usage was defined after
run-off reactions of the unlabeled BV-BC amplification product and is
quantitative, since the fluorescent primers have comparable
amplification efficiencies (23). The fluorescence intensity in each BV
or BJ family was expressed as the percentage of total signals from the
22 BV or 13 BJ subfamilies. Statistical comparisons between samples
from healthy donors and patients or between PBL and SF were made using
the Mann-Whitney test. Clonotypic sequence usage was defined after
run-off reactions with the labeled BJ2S3 and clonotypic primers. The
amplification efficiency of the clonotypic primer was assessed
beforehand in comparison with the BJ primer on a clonal plasmid DNA.
DNA sequencing
BV-BJ PCR products were cloned into pCR2.1 vector (Invitrogen) and transformed into XL1 Blue supercompetent cells (Stratagene, La Jolla, CA). After blue/white screening of recombinant plasmids on X-galactoside/isopropylthiogalactoside indicator plates, plasmids were purified by alkaline lysis followed by phenol/chloroform/iso-amyl alcohol. Inserts were checked by agarose gel electrophoresis after BV-BJ PCR amplification, and both strands were sequenced with the ABI PRISM Dye Primer Cycle sequencing kit (Perkin-Elmer). Products were loaded on 4.25% acrylamide sequencing gels (377A DNA sequencer, Applied Biosystems) and analyzed with the Sequence Navigator software.
Flow cytometric analysis
Direct immunofluorescence was performed in triple labeling analysis. After washing, SF cells or PBL were incubated for 15 min with one of the BV region-specific labeled mAb (Immunotech, Marseille, France) BL37.2-phycoerythrin (BV1), TAMAYA1.2-FITC (BV16), FIN9-phycoerythrin (BV9), or AF23-phycoerythrin (BV23) together with a labeled CD3, CD4, CD8, or CD45RO mAb (Immunotech, Marseille, France). Events (5 x 103–104) gated on CD3 or CD8 expression were analyzed using a FACScan flow cytometer (Becton Dickinson, Mountain View, CA), and results were expressed as percentages of cells staining above the background level.
Selection of a CD8-positive cytolytic T cell line from patient PG and 51Cr release assay
Synovial T cells from patient PG were sorted by means of the CD8 mAb (>95% CD8+ cells) and cultured in medium supplemented with rIL-2 (Genzyme, Cergy, France) at 150 IU/ml, purified PHA-L at 0.5 µg/ml (leukoagglutinin, Sigma, St. Louis, MO) and irradiated (50 Gy) allogeneic feeder cells as described previously (28). Cells were maintained by adding IL-2 (150 IU/ml) twice weekly and were restimulated every 2 wk in the presence of allogeneic feeder cells in which HLA-B27-positive cells had been excluded. Cytotoxicity was measured by a standard 4-h 51Cr release assay, and the percentage of specific lysis was calculated as described previously (28). Targets were B-LCL generated in the laboratory, except ADA (28), HOM-2, COX, MGAR, Sweig, and YAR (homozygous B-LCL from the Tenth International Histocompatibility Workshop), and the T2 and B*2705-transfected T2 cell lines (29). PHA blasts were generated from PBL of individual GC (2 x 106) by three cycles of a 4-day stimulation with PHA-L at 1 µg/ml. In inhibition experiments, targets were preincubated for 1 h with a 1/100 dilution of ascites of W6/32, an HLA class I monomorphic mAb (30) or B1.23.2, an HLA-B- and -C-specific mAb (31). SF1.111, an anti-H-2 Kd IgG2a isotypic control, was provided by Dr. M. Pla (Hôpital Saint-Louis, Paris, France). Blocking experiments were also performed using the CD3-specific mAb OKT3 (Ortho Diagnostics, Raritan, NJ). HLA class I typing was determined by standard microcytotoxicity and HLA-B27 subtyping using DNA methods (32).
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This approach of T cell repertoire analysis was chosen to provide
both a qualitative and a quantitative appraisal of the T cell response
in terms of clonality and of BV or BJ gene segment usage, respectively.
We first compared the TCR BV gene usage in PBL isolated from
age-matched HLA-B27-positive individuals who were either healthy
(n = 10) or were affected by various forms of SA
(n = 19). The overall picture of BV usage was
comparable in the two groups (data not shown). We observed T cell
expansions in different BV families without any correlation with
disease status. Their frequency increased with age in the group of
healthy subjects as well as in the patients as reported previously
(33). In line with the hypothesis that a local stimulus could initiate
the arthritic process, we thereafter chose to focus on the comparison
between the T cell repertoire in PBL and in SF lymphocytes. Sample
pairs from seven different HLA-B27 patients were available for
analysis. Perturbations of the CDR3 size distribution profiles were the
rule in SF T cells with few truly polyclonal families. Oligoclonal
expansions were found in three to six BV families in SF and in zero to
three families in blood. Some were common at both sites, but the
majority were detected in SF lymphocytes only. Major SF expansions and
corresponding BV-BC amplifications from PBL are shown in Fig. 1 for the seven patients studied. Expansions were found
in different BV families in different patients. However, four of seven
patients had expansions in the BV1 or in the BV11 family, and three of
seven had expansions in BV23. Patients PG and CN, both affected by an
acute form of ReA triggered by an enteric infection, shared T cell
expansions in the BV1 and BV23 families. These BV expansions were
defined more precisely in most cases with regard to their BJ usage by
performing run-off reactions of the BV-BC amplification with BJ
fluorescent primers (Table II). At this level of
definition some BV expansions apparently common to different patients
could be distinguished, for instance the BV16 expansions (11 aa) of
patients GM and PG used, respectively, the BJ2S5 and BJ2S7 segments.
However, the BV1 expansions (10 aa) of patients PG and CN used the same
BJ2S3 segment as did the BV11 expansions (6 aa) of patients CF and CN.
As the experiments were also semiquantitative, we could compare the
percentage of the T cell expansions in the PBL vs SF. In some cases SF
expansions were markedly over-represented compared to blood, for
instance the BV1-BJ2S3 expansion, which could be considered the
dominant expansion in patient PG accounting for about 4.75% of in situ
T cells (Table II).
|
|
Patient PG presented at diagnosis an acute right knee arthritis
and 15 days later developed a controlateral knee effusion. Fluid was
obtained for analysis at both articular sites (SF-R and SF-L). Similar
oligoclonal patterns were observed in both knee effusions. The
similarity of the BV1, BV9, BV16, and BV23 expansions was confirmed by
BJ run-off reactions (Fig. 2). Moreover, the expansions
detected in two right knee samples obtained a few days apart were
identical (data not shown). These data confirm the relevance of the
method and the fixed bias of the T cell response in a given patient
during an acute episode of arthritis. The BV1-BJ2S3 expansion at a
10-aa CDR3 size was detected in blood during the acute phase of
arthritis but no longer in a sample collected 11 mo after recovery
(Fig. 2 and data not shown).
|
). It confirmed the over-representation of the
BV1 family in the SF and not in the blood, notably in the CD8
subset. The frequency of CD8+CD45RO+ memory
cells was markedly increased in the SF, accounting for 40% of the
total CD8+ cell population. About one-third of these cells
(24%) belonged to one of the four expanded BV families, and about 14%
belonged to the single BV1 family.
|
Immunoscope data and cytometric analysis clearly indicated that in
the case of patient PG the dominant ß T cell expansion found in
situ was in the BV1 family and was characterized by the usage of the
BJ2S3 gene segment and a CDR3 size of 10 aa. Since such expansions that
appear to be monoclonal could in fact reflect the occurrence of many
different T cell clones, it was necessary to obtain sequences from this
CDR3 junctional region. Oligoclonal T cell expansions in the BV9, BV16,
and BV23 families were also sequenced, and the results are shown in
Table IV. The sequence SVGLYSTDTQ was present in five of
11 clones and could be considered major in the BV1 expansion. The six
other clones provided sequences at the expected CDR3 size and differed
from the major sequence by a single amino acid at position 97, 98, or
99 considering the last serine in the CASS motif as position 95. One
sequence (SVGVYSTDTQ) has even been reported in a T cell clone
derived from a Salmonella-triggered ReA patient (18). In the
other BV families, the most notable results were obtained in BV23, in
which the dominant sequence SVGLYSTDTQ present in eight of 24 clones
was the same as that in BV1. Two other sequences, SVGLFSTDTQ and
SVGDYSTDTQ, were also common to BV1 and BV23. These sequences had a
characteristic BV-encoded residue at position 79, either leucine
for BV1 or methionine for BV23, showing that they definitely belonged
to different families. The high conservation of the CDR3 within these
two families was also demonstrated in 35 sequences by the use of a
small (proline, glycine, serine) or aliphatic/hydrophobic (valine,
alanine, isoleucine, leucine) residue at position 96 and of an
aromatic/hydrophobic residue (tyrosine or phenylalanine) at position
99, while the CDR3 positions 97 and 98 were more variable. A clonotypic
primer was designed to estimate the representation of the dominant CDR3
sequence shared in the BV1 and BV23 expansions. It amplified about
0.6% and 0.2% of the total SF T cells in BV1 and BV23, respectively,
and was negative in PBL (Fig. 2
). However, this clonotype gave a low
estimate, since different nucleotide sequences could encode the same
CDR3 amino acid sequence, as will be detailed later.
|
Some oligoclonal expansions were shared between different patients
(BV1-BJ2S3 in patients PG and CN, BV11-BJ2S3 in patients CF and CN) or
were similar in size or BV usage, for instance the BV16 expansion at 11
aa in patients PG and GM (Fig. 2 and Table II). Although sharing the
same size and BV and BJ usage, the major sequences from the BV11-BJ2S3
peaks were different, being SLVDTQ in patient CF and TTGYTQ in patient
CN (Table V). In marked contrast, we found identical
CDR3 sequences in patients CN (Table V) and PG (Table IV): SVGLFSTDTQ
in the BV1 and BV23 oligoclonal peaks of patient PG and in the BV1 peak
of patient CN and SPGLYSTDTQ shared by BV23 and BV1 peaks of patients
PG and CN, respectively. The most frequent CDR3 sequence in the BV1
expansion of patient CN (SVAHYSTDTQ) differed at positions 97 and
98 only from the main SVGLYSTDTQ sequence found in patient PG.
Comparisons at the nucleotide level of identical sequences in patients
PG and CN excluded the possibility that these shared sequences could be
explained by contamination of samples during PCR amplification, since
multiple silent substitutions were found in different clones encoding
the same CDR3 sequence (Table VI).
|
|
A CD8+ T cell line was established from SF lymphocytes
of patient PG. The BV, BJ, and clonotypic expression was checked by
flow cytometry and immunoscope analysis. The BV1 family was expressed
at a similar level as in SF, and the clonotype frequency in BV1 and
BV23 families was estimated as 1.2% of T cells (data not shown). This
line was tested against a panel of B-LCL, either HLA-B27-positive or
-negative (Fig. 3). Several HLA-B27-positive targets
were lysed reproducibly, with >15% specific lysis at a 12.5:1 E:T
cell ratio. HOM-2, A16, and A19 shared only B*2705 with the
CD8+ effector line. The cytotoxic activity was blocked by
two different HLA class I-specific mAbs, W6/32 and B1.23.2, as well as
by a CD3-specific mAb, therefore confirming the HLA-B27 specificity of
the T cell recognition (Fig. 4A). The
autologous EBV line was not lysed, and this could not be explained by
resistance of the target to cytotoxicity, as it was recognized by
effector cells from another arthritic patient (M. Bonneville and
M. A. Peyrat, unpublished observations). In this panel, several
EBV lines from B*2705 individuals were also not recognized (Fig. 3).
There was no correlation with disease status, as HOM-2, A16, A19, or GC
obtained from healthy individuals and TD or ADA from arthritic patients
were lysed. The T2 and T2-B*2705 cell lines, which are deficient
in peptide transporters and express empty class I molecules at 37°C,
were not lysed (data not shown). The idea that EBV Ags could be
recognized by intrasynovial lymphocytes (28) prompted us to test
comparatively another kind of target than B-LCL. PHA blasts could be
generated from the healthy HLA-B27 individual GC. They were recognized
by the CD8+ line, and lysis was blocked upon incubation
with W6/32 (Fig. 4B), arguing for the recognition of
HLA-B27-presented antigenic peptides from endogenous self proteins and
not from EBV.
|
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Evidence is now accumulating for the occurrence of a limited heterogeneity in the T cell response at the pathogenic site of autoimmune diseases, much of which has been provided by animal models, e.g. experimental autoimmune encephalitis in rodents (36) or the NOD mouse model of spontaneous autoimmune diabetes (37). T cell repertoire studies have also been conducted in human autoimmune diseases with the aim of defining public T cell responses and potential targets for immunotherapy. In the case of rheumatoid arthritis, TCR ß-chain CDR3 motifs shared by different patients could be defined (13), and sequence identity between different patients has been occasionally found (15, 38). Other examples of identical CDR3 sequences have been reported in brain lesions in multiple sclerosis (39), in CD4+ T cell expansions in coeliac disease patients (40), and in a pair of homozygous twins with Crohn’s disease (41). HLA matching for the disease susceptibility allele and a recent disease onset seem to be crucial points in these observations, as was the case in HLA-DR2 multiple sclerosis patients (42). The importance of homogeneous HLA-typed patients to pinpoint such TCR skewing is not surprising if the structural data concerning MHC peptide presentation to T cells are taken into account (20). In that respect, HLA-B27-associated arthritis could be a privileged example of public ß-chain CDR3 sequences, especially in bacteria-triggered cases of ReA. We did not observe overall differences in the frequency of BV transcripts in SF lymphocytes compared with PBL. This is expected in view of the complexity of the T cell response and is in agreement with a previous report (16). A higher degree of resolution was needed, which was achieved using this CDR3 size analysis approach. In these conditions, we were able to select step by step the most relevant T cell populations from the picture of the global T cell repertoire ex vivo. More generally speaking, the conservation of ß-chain junctional regions has been shown during HLA-B27-restricted cytotoxic responses against the influenza A virus nucleoprotein 383–391 peptide (43) and to a lesser extent in anti-HLA-B27 alloresponses in which multiple antigenic peptides could be recognized (44). The comparison of TCRBV heterogeneity in various class I-restricted responses suggested that a highly restricted pattern could reflect a selection of the response toward peptides that were cross-reactive with self Ags (45). In line with this hypothesis, our data would support the idea of molecular mimicry between bacterial and endogenous HLA-B27 ligands. The fact that the dominant clonotype SVGLYSTDTQ is homologous to the sequence SVGTSGTDTQ of an anti-HLA-B27 alloreactive clone (clone 18DLH) reported by Barber et al. (46) also using the same BV1 and BJ2S3 gene segments is consistent with this assumption. The ß-chain CDR3 mainly contacts the COOH-terminal part of the antigenic peptide (20). The conserved aromatic residue at position 99 in the public clonotype could be critical for peptide contact, similarly to the key residue at position 98 in another TCR/MHC/peptide complex (47).
One aim of this work was to take advantage of the T cell repertoire data to select the best candidates for further functional analysis and possibly the definition of the peptide(s) recognized. In the present study it was crucial to focus on the T cell populations expanded in the SF of patient PG, which expressed T cell clonotypes shared by other ReA patients. The major TCRBV1 expansion was no longer detected in blood after recovery. This observation is in contrast with the persistence of T cell expansions observed in other autoimmune diseases or even in healthy individuals, especially in the CD8+ subset (33). This could be explained by the acute course of the disease in this patient compared with more chronic diseases in other patients in this study and in the literature and could provide an opportunity to study more closely the disease triggering. A CD8+ T cell line was established that expressed a clonotypic frequency similar to the initial SF lymphocytes. This frequency of about 1% is expected for autoantigen-specific T cells at the pathogenic site of autoimmune diseases (48). This CD8+ line had cytotoxic activity toward HLA-B27 molecules expressed in B-LCL without any prior in vitro stimulation. There was no correlation with disease status and the patient’s own B-LCL was not lysed. The lysis of different targets sharing B*2705 only with the effector cell, and the blockade by HLA class I and CD3-specific mAbs argue for a classical B27-restricted CD8+ T cell response. The fact that the transporter-deficient T2 cell line transfected with B*2705 was not lysed also suggests recognition of HLA-B27/peptide complexes. Although the triggering organism was unknown for patient PG, because of both the CDR3 sequence identity with a Salmonella-induced ReA patient (18) and of epidemiological data (24), we tested the cytotoxic activity of the CD8+ line against Salmonella typhimurium-infected target cells. Infection did not induce specific lysis of PG B-LCL (data not shown). Our data therefore indicate that a cytotoxic T cell response triggered during the bacterial infection could be directed toward HLA-B27 molecules bound to endogenous peptides and supports the concept of molecular mimicry at the T cell level. The lack of cytolytic activity on some B*2705 B-LCL, including the patient’s own line, could be simply due to inadequate avidity of the effector CD8+ line (49). Otherwise, the different pattern of recognition among B*2705 B-LCL is reminiscent of a previous report in which the inability of HLA-B*2702 viral peptide presentation was attributed to a defect in Ag processing (50). Differences in endogenous peptide presentation could be explained by many factors, such as genetic differences in the source protein of the peptide as in the case of minor histocompatibility Ags (51) or in genes involved in the processing machinery and peptide presentation, the description of which is ever increasing (52). Notably, genetic susceptibility markers in addition to HLA-B27 have been recently defined in the MHC region (53). The final explanation of this observation could be provided by direct characterization of the peptides recognized among acid-eluted peptides from these targets.
In conclusion, these data provide evidence for shared T cell responses in bacteria-triggered ReA and favor the arthritogenic peptide hypothesis by showing bias in the T cell repertoire as well as HLA-B27-directed T cell responses during the course of the immune process. They also provide tools for further defining the nature and the origin of such peptide(s) before further attempts of therapy.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. A. Toubert, Centre G. Hayem, Hôpital Saint-Louis, 1 av. C. Vellefaux, 75475 Paris Cedex 10, France. E-mail address:
3 Abbreviations used in this paper: SA, spondyloarthropathies; ReA, reactive arthritis; SF, synovial fluid; hsp, heat shock protein; CDR3, complementarity-determining-region 3; aa, amino acid; B-LCL, Epstein-Barr virus-transformed B lymphoblastoid cell line.
Received for publication September 25, 1998. Accepted for publication December 22, 1998.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
and Vß gene transcripts in synovial fluid T cells of HLA-B27 positive reactive arthritis patients. Clin. Rheumatol. 15:91.
ß T-cell repertoire: characteristics of a polyclonal and naive but completely formed repertoire. Blood 91:340. and ß chains. Eur. J. Immunol. 25:2479.[Medline]
This article has been cited by other articles:
|
|
 |
|
  E. May, M. L. Dorris, N. Satumtira, I. Iqbal, M. I. Rehman, E. Lightfoot, and J. D. Taurog CD8{alpha}{beta} T Cells Are Not Essential to the Pathogenesis of Arthritis or Colitis in HLA-B27 Transgenic Rats J. Immunol., January 15, 2003; 170(2): 1099 - 1105. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. J. Bartolome Pacheco, V. M. Martinez-Taboada, R. Blanco, V. Rodriguez-Valverde, J. I. Valle, and M. Lopez-Hoyos Reactive arthritis after BCG immunotherapy: T cell analysis in peripheral blood and synovial fluid Rheumatology, October 1, 2002; 41(10): 1119 - 1125. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. Dulphy, C. Rabian, C. Douay, O. Flinois, S. Laoussadi, J. Kuipers, R. Tamouza, D. Charron, and A. Toubert Functional modulation of expanded CD8+ synovial fluid T cells by NK cell receptor expression in HLA-B27-associated reactive arthritis Int. Immunol., May 1, 2002; 14(5): 471 - 479. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. R. Wedderburn, A. Patel, H. Varsani, and P. Woo Divergence in the degree of clonal expansions in inflammatory T cell subpopulations mirrors HLA-associated risk alleles in genetically and clinically distinct subtypes of childhood arthritis Int. Immunol., December 1, 2001; 13(12): 1541 - 1550. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. J. Costello, R. J. Winchester, S. A. Curran, K. S. Peterson, D. J. Kane, B. Bresnihan, and O. M. FitzGerald Psoriatic Arthritis Joint Fluids Are Characterized by CD8 and CD4 T Cell Clonal Expansions that Appear Antigen Driven J. Immunol., February 15, 2001; 166(4): 2878 - 2886. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. Guilherme, N. Dulphy, C. Douay, V. Coelho, E. Cunha-Neto, S. E. Oshiro, R. V Assis, A. C. Tanaka, P. M. Alberto Pomerantzeff, D. Charron, et al. Molecular evidence for antigen-driven immune responses in cardiac lesions of rheumatic heart disease patients Int. Immunol., July 1, 2000; 12(7): 1063 - 1074. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Fournel, M. Aguerre-Girr, X. Huc, F. Lenfant, A. Alam, A. Toubert, A. Bensussan, and P. Le Bouteiller Cutting Edge: Soluble HLA-G1 Triggers CD95/CD95 Ligand-Mediated Apoptosis in Activated CD8+ Cells by Interacting with CD8 J. Immunol., June 15, 2000; 164(12): 6100 - 6104. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. R. Wedderburn Tracking T cells in arthritis Rheumatology, May 1, 2000; 39(5): 458 - 462. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||
