close

Вход

Забыли?

вход по аккаунту

?

Genomic absence of the gene encoding T cell receptor V 7.2 is linked to the presence of autoantibodies in Sjgren's syndrome

код для вставкиСкачать
ARTHRITIS & RHEUMATISM
Vol. 50, No. 1, January 2004, pp 187–198
DOI 10.1002/art.11429
© 2004, American College of Rheumatology
Genomic Absence of the Gene Encoding T Cell
Receptor V␤7.2 Is Linked to the Presence of
Autoantibodies in Sjögren’s Syndrome
Sanil J. Manavalan,1 Jennifer R. Valiando,1 Westley H. Reeves,2 Frank C. Arnett,3
Antje Necker,4 Ronit Simantov,1 Robert Lyons,2 Minoru Satoh,2 and David N. Posnett1
Objective. It is not yet known whether the absence
of certain T cell receptor V␤ (TCRBV) genes (e.g., due to
genomic deletion) has functional significance. We examined this question in relation to a known 21.6-kb
insertion/deletion-related polymorphism (IDRP) in the
human BV locus.
Methods. New polymerase chain reaction (PCR)
genotyping methods were used. Monoclonal antibodies
to TCRBV gene products were used to confirm the
absence of the relevant proteins. Patients with Sjögren’s
syndrome (SS) or systemic lupus erythematosus (SLE)
were compared with normal controls with regard to
TCR genotypes and serologic profiles.
Results. There are 3 known haplotypes (I, D1, D2)
and 6 possible genotypes related to the 21.6-kb IDRP.
Novel PCR-based methods were used to define these
genotypes. In subjects with deleted/deleted (D/D) genotypes, T cells could not express V␤7.2 TCRs, as assayed
with a new antibody specific for V␤7.2. This was the sole
significant difference between subjects without the insertion and those with either 1 or 2 copies. Surprisingly,
we found that the D/D genotype was associated with
primary SS, but only when pathogenic autoantibodies
were present.
Conclusion. These results suggest that T cells
expressing TCRs with V␤7.2 are protective against a
pathogenic immune response in SS. Thus, genomic
polymorphism of TCR genes (along with the correct
HLA alleles) determines whether T cells can direct a
pathogenic autoimmune response.
In both human and mouse T cell receptor (TCR)
gene loci, major genomic deletions exist. These deletions
result in a contracted TCR repertoire, with certain TCR
V␤ genes missing in some individuals. Whether this is of
any functional significance is unknown. For instance, it is
well appreciated that TCR transgenic mice do not
overtly suffer from immunodeficiency, although they
have a severely contracted TCR repertoire that is sometimes monoclonal, as occurs when these mice are bred
onto a RAGo/o background. In contrast, mice with TCR
haplotypes bearing major genomic deletions have been
reported to be susceptible (1) or resistant (2,3) to
collagen-induced arthritis, and resistant to autoimmune
disease in the NZB ⫻ NZW murine model of systemic
lupus erythematosus (SLE) (4). The mechanism underlying these associations is not fully understood, and
similar associations in humans have not been described.
Primary Sjögren’s syndrome (SS) is an autoimmune disease characterized by keratoconjunctivitis
and sialadenitis, with lymphocytic infiltration of these
exocrine glands with both T and B cells (5). Extraglandular site involvement may cause lymphocytic interstitial
pneumonitis, atrophic gastritis, biliary cirrhosis, vasculitis of the skin (hypergammaglobulinemic purpura), thyroiditis, neuropathy, various cytopenias, and tubulointerstitial nephritis. The autoantibodies characteristic of
SS include antinuclear antibodies termed anti-Ro/SSA
and anti-La/SSB (6). These antibodies are also found in
patients with SLE, although less frequently. Anti-Ro
antibodies are directed to 60-kd and 52-kd proteins
Supported by NIH grant AR-40391, GCRC grant M01R00082, and by funding from the State of Florida to the Center for
Autoimmune Disease.
1
Sanil J. Manavalan, MD, Jennifer R. Valiando, MD, Ronit
Simantov, MD, David N. Posnett, MD: Weill Medical College of
Cornell University, New York, New York; 2Westley H. Reeves, MD,
Robert Lyons, Minoru Satoh, MD, PhD: University of Florida,
Gainesville; 3Frank C. Arnett, MD: University of Texas–Houston
Health Sciences Center; 4Antje Necker, PhD: Immunotech, BeckmanCoulter, Marseilles, France.
Address correspondence and reprint requests to David N.
Posnett, MD, Cornell University Weill Medical College, 1300 York
Avenue, Box 56, Room D601, New York, NY 10021. E-mail:
dposnett@mail.med.cornell.edu.
Submitted for publication January 22, 2003; accepted in
revised form September 11, 2003.
187
188
MANAVALAN ET AL
(Ro 60 and Ro 52, respectively) that associate with the
Y series of small RNA through RNA binding domains
and zinc-finger motifs. Anti-La antibodies are specific
for a 48-kd phosphoprotein that complexes transiently
with nascent RNA polymerase III transcripts. La
antigens associate with Epstein-Barr virus (EBV)–
encoded RNA transcripts.
In the human TCR ␤-chain (TCRB) locus, there
is an insertion/deletion-related polymorphism (IDRP) in
the region of the variable (V) gene segments (TCRBV),
which was originally described by Zhao et al (7). It is a
21.6-kb fragment, containing 3 BV gene segments. The
insertion is flanked by BV13S2 (upstream) and BV6S7
(downstream). These are both polymorphic BV genes.
Kay et al have described an association of SS with the
BV13S2*2 allele (8). BV13S2*1 and BV13S2*2 (the 2
known alleles of BV13S2) differ, however, by only a
single amino acid in the leader region (9). Because the
leader region is not part of the mature TCR ␤-chain
protein, there is no difference in the expressed TCR
␤-chain protein. Therefore, the data reported by Kay et
al suggest that a neighboring, closely linked gene may
actually be relevant, rather than the BV13S2 alleles
themselves.
In this study, we examined the 3 known TCRBV
haplotypes in the region of BV13S2. Six possible genotypes were determined by novel polymerase chain reaction (PCR)–based genotyping methods. Genotypes and
phenotypes were compared using V␤-specific monoclonal antibodies (mAb) to detect expression of specific
TCRs. The specificity of a new antibody to V␤7.2 was
determined. This gene is encoded within the IDRP
insertion, and individuals lacking the insertion on both
chromosomes also lack V␤7.2-positive T cells. Surprisingly, we observed that SS was significantly associated
with the absence of BV7S2, but only when antibodies to
Ro and La were present.
PATIENTS AND METHODS
Human subjects. Patients with SLE or SS fulfilled
accepted diagnostic criteria (5,10). All patients with SS, including those without anti-Ro/La antibodies, had positive results
on lip biopsies, with Daniels’ focus scores of ⬎2. Race, age,
sex, and clinical features were taken into account. Normal
controls were from either Marseilles, France or Houston,
Texas. The former subjects were randomly selected clinic
outpatients and normal volunteers of unselected ethnic background. All of the latter subjects were Caucasian and were
selected as a matched population for a group of patients with
primary SS. These normal adult volunteers were medical
school and hospital personnel from Houston. Ethnicity was
based on both grandparents being Caucasian.
Other study populations included SLE patients from
New York Presbyterian Hospital and the Hospital for Special
Surgery in New York, and patients with SLE or SS from the
University of Florida in Gainesville. All samples were obtained
in conformity with protocols accepted by the respective human
institutional review board committees. Peripheral blood mononuclear cells (PBMCs) were used for flow cytometry or for
DNA isolation in some cases. Otherwise, stored DNA samples
were used for genotyping. Stored serum or plasma was used for
autoantibody tests. Analysis of patient and control groups was
limited to Caucasians, because TCR genotypes in the area of
the IDRP differ considerably in different ethnic groups
(11,12).
DNA preparations. DNA isolation was performed by
several methods. For the 7.9-kb PCR, DNA was isolated
using the QIAamp DNA Blood Mini Kit (Qiagen, Valencia,
CA) according to the manufacturer’s instructions. Otherwise,
DNA was isolated by incubating cells in lysis buffer (Tris
HCl, 10 mM; EDTA, 1 mM; sodium dodecyl sulfate [SDS],
0.0001%; Triton X-100, 0.001%; proteinase K, 600 ␮g/ml) for
1 hour at 56°C followed by 15 minutes at 95°C, or by similar
protocols (13).
TCR nomenclature. We used the nomenclature described by Arden et al (9), with a single exception: BV6S7 (11),
as used herein, is also called BV6S4 by Arden et al (9). For
BV7S2 within the IDRP and BV7S3 upstream of the IDRP, we
used the nomenclature described by Arden, which coincides
with GenBank nomenclature (accession no. L36092). The
same genes are named in reverse (BV7S3 and BV7S2, respectively) by Zhao et al (7). TCRBV gene products (proteins) are
referred to as V␤7.2, and so forth.
Flow cytometry with TCR mAb. PBMCs from a FicollHypaque gradient were washed in phosphate buffered saline
(PBS) and suspended in staining buffer (RPMI 1640, 2% fetal
calf serum, 0.05% sodium azide). Murine mAb to TCR V␤
epitopes were added at predetermined optimal concentrations.
The cells were stained at room temperature for 40 minutes on
a shaker, then washed again 3 times and stained next with goat
anti-mouse Ig–fluorescein isothiocyanate, followed by antiCD3–phycoerythrin, both at optimal dilutions, for 40 minutes
on a shaker, at room temperature in the dark. Cells were then
washed 3 times with staining buffer and suspended in 0.5 ml of
the same buffer for flow analysis on a Coulter Epics XL
(Beckman Coulter, Miami, FL). The mAb have been described
elsewhere (14,15), except for mAb Zizou (V␤7.2).
PCR-based TCR genotyping. Genotyping for the
IDRP is based on 2 PCR assays. The first assay detects the
presence of the insert, with primers straddling the upstream
margin of the insertion (Figure 1). The primers used are
5⬘-AGGTGAACTTGGAGATGCA-3⬘ (forward) and 5⬘CTGTGGAGGCATGTACACTATGT-3⬘ (reverse). The conditions are as follows: 10 mM Tris HCl, 50 mM KCl, 0.1%
Triton X-100, 1.5 mM MgCl, 200 ␮M of each dNTP, and 1 unit
of Taq polymerase (Promega, Madison, WI) at 94°C for 1
minute, 55°C for 1 minute, 72°C for 1 minute, for 36 cycles,
followed by a 10-minute extension at 72°C. The result is
amplification of a 169-bp fragment.
The second assay is a nested PCR with primers in the
BV6S7 gene (Figure 1). The outer primers are 5⬘GTCTCCAGTCCCCCAGTAACAA-3⬘ (forward) and 5⬘GCAGGAAGGAGGCGACTGTGCCA-3⬘ (reverse). The result is a 311-bp PCR fragment. The conditions are as described
above, except for an annealing temperature of 65°C. The inner
ABSENCE OF TCRBV7S2 ASSOCIATED WITH PRESENCE OF AUTOANTIBODIES IN SS
189
Figure 1. Map of the 3 insertion/deletion-related polymorphism haplotypes (inserted [I], deleted
2 [D2], and deleted 1 [D1]) showing the allelic variants of the T cell receptor BV genes present on
each haplotype. BV genes located outside the insert are shown as open boxes, and those located
within the insert are shown as filled boxes. The official nomenclature for BV genes was used:
13S2*1a means BV13S2, allele *1, copy a (there is also a copy b within the genomic insert, and an
allele *2 on the D1 haplotype). Large shaded arrows indicate genomic duplication; margins of the
insert are indicated by a vertical line at the end of each arrow. Primers (arrowheads) that detect the
insert are shown for haplotype I; nested primers for BV6S7 are on the far right. Primers for a large
7.9-kb polymerase chain reaction that detects deleted haplotypes are shown for D2.
primers are 5⬘-GGTCACAGAGAAGGGAAAGG-3⬘ (forward) and 5⬘-CGGCCGAGTCCTCCTGCTG-3⬘ (reverse),
yielding a 233-bp fragment. The conditions are as described
above, except for an annealing temperature of 60°C, and use of
2.0 mM of MgCl2. The 233-bp fragment is then digested with
10 units of Bam HI (17 ␮l of PCR product, 2 ␮l of 10⫻ buffer,
1 ␮l of Bam HI). If the BV6S7*1 allele is present, there is a
Bam HI site (11), and the 233-bp PCR product is cut into
184-bp and 49-bp fragments. The BV6S7*2 allele does not
have a Bam HI site.
The PCR to distinguish between inserted/inserted (I/I)
and inserted/deleted 2 (I/D2) genotypes (see below) used a
forward primer (5⬘-GTCTACAACTTTAAAGAACAGACTG-3⬘) paired with the reverse primer of the above PCR
5⬘-GCAGGAAGGAGGCGACTGTGCCA-3⬘) to yield a
7.9-kb product. This PCR used the Expand Long Template
PCR System Kit (Roche Laboratories, Nutley, NJ). Each 50-␮l
reaction contained 400 ng of purified genomic DNA, 350 nM
of each dNTP, 400–600 ng of each primer, 20 mM of Tris HCl,
100 of mM KCl, 1 mM of dithiothreitol, 0.1 mM of EDTA,
0.5% Tween 20 (volume/volume), 0.5% Nonidet P40 (v/v),
1.75 mM of MgCl, 2.5 units of Taq, plus 2.5 units of Pwo
polymerase. The cycling parameters were 92°C for 2 minutes,
92°C for 10 seconds, 65°C for 30 seconds, and an 8-minute
extension at 68°C, repeated for 10 cycles; this was followed
by 20 cycles of 92°C for 10 seconds, 65°C for 30 seconds,
and an 8-minute extension at 68°C (with an increase of the
extension time by 20 seconds with each cycle), then a final
extension of 68°C for 7 minutes. This PCR was best performed
using a Perkin-Elmer 9600 thermocycler (Perkin-Elmer,
Emeryville, CA).
Anticardiolipin antibody enzyme-linked immunosorbent assay (ELISA). The ELISA for anticardiolipin antibodies
was performed with cardiolipin antigen (C1649; Sigma, St.
Louis, MO) dissolved in ethanol and coated onto roundbottomed wells by evaporation. After blocking with 2% bovine
serum albumin, human serum serially diluted in PBS containing 10% adult bovine serum (C1507; Sigma) was incubated at
4°C for 2–3 hours, followed by washing and further incubations, first with alkaline phosphatase–conjugated goat antihuman IgG antibody, and second with p-nitrophenylphosphate
disodium substrate. Optical density (OD) was read at 405 ␭
(wavelength).
Antinuclear antibodies by immunoprecipitation. K562
cells (human erythroleukemia; American Type Culture Collection, Rockville, MD) were maintained in RPMI 1640 supplemented with 10% fetal bovine serum, L-glutamine, and
penicillin/streptomycin. Autoantibodies were analyzed by immunoprecipitation of 35S-radiolabeled K562 cell extract, using
8 ␮l of human serum per sample (16). Sera containing anti–Ro
60/SSA, anti-La/SSB, anti–nuclear RNP/Sm, and other autoantibodies were identified by comparison with human reference sera, as previously described (17). Anti-Sm antibodies
were detected by immunoprecipitation of the 200- and 205-kd
components of the U5 small nuclear RNP, as previously
described (18).
Anti–Ro 52 and other antibodies by immunoblotting.
IgG from 3 ␮l of human anti–Ro 52 reference serum was
190
MANAVALAN ET AL
crosslinked to protein A–Sepharose beads using dimethylpimelimidate (19). Proteins immunoprecipitated from K562
cell extract (2 ⫻ 107 cell equivalents) were fractionated by
SDS–polyacrylamide gel electrophoresis and transferred to
nitrocellulose. The membrane was probed with anti–Ro 52
reference serum (positive control), normal human serum (negative control), or test sera at a 1:400 dilution. Blots were
developed with 1:2,000 alkaline phosphatase–labeled goat antihuman IgG (␥- and light chain–specific; Southern Biotechnology, Birmingham, AL) and developed using the Western-Star
chemiluminescent system (Tropix, Bedford, MA). For immunoblot detection of anti–Ro 60/SSA and anti-La/SSB antibodies,
an analogous approach was used, substituting anti–Ro 60 and
anti-La reference sera, respectively.
Antinuclear antibodies, anti–single-stranded DNA
(anti-ssDNA), and antichromatin ELISAs. Antinuclear antibodies were detected using HEp-2 substrate, as previously
described (20). ELISAs for antichromatin and anti-ssDNA
antibodies were performed using sera diluted 1:500 (21,22).
IgM and IgG antichromatin antibodies were considered
positive when sample absorbance was more than 3 SD above
the mean OD in blank wells. Anti-ssDNA antibodies were
considered positive when sample absorbance was more than
3 SD above the mean OD using sera obtained from healthy
volunteers.
Clinical data. Platelet counts and levels of lupus
anticoagulant, factor V Leiden, prothrombin mutation, and
homocysteine were obtained from the clinical laboratory measurements.
Statistical analysis. Statistical analysis was done by
chi-square (with Yates’ correction when necessary), by 2-tailed
Fisher’s exact test, or by analysis of variance.
RESULTS
IDRP genotypes determined by PCR. The human
TCRB locus contains an IDRP in the BV region (7) (see
Figure 1). It is a 21.6-kb fragment, characterized by the
presence of 3 V gene segments: BV9S2P is a pseudogene, BV7S2 is a unique V gene, and BV13S2*1b is an
exact copy of the BV13S2*1a gene located outside of the
IDRP. There are 2 deleted haplotypes, both lacking the
21.6-kb fragment. These 2 deleted haplotypes differ in
terms of the alleles of the flanking TCR BV13S2 and
BV6S7 genes. Thus, the deleted 1 haplotype (D1) has
BV6S7*2 and BV13S2*2 alleles, and the deleted 2 (D2)
haplotype has BV6S7*1 and BV13S2*1 alleles. The
inserted haplotype (I) is similar to the D2 haplotype in
that it also contains BV6S7*1 and BV13S2*1. The
inserted haplotype provides the total BV locus with 63
BV gene segments, of which 52 are functional, while
both deleted haplotypes have 60 BV gene segments, of
which 50 are functional.
There are 2 additional haplotypes (23), which are
not shown in Figure 1. One haplotype is thought to arise
from 2 head-to-tail insertions of the 21.6-kb fragment
shown in Figure 1. The other is similar to the D
haplotypes shown in Figure 1 but has an additional
deletion located at least 15 kb further downstream.
These other haplotypes occur with gene frequencies of
0.005 and 0.008, respectively and thus are very rare.
All of these TCRBV haplotypes were originally
described using pulsed-field gel electrophoresis of Sfi I
DNA digests (7,23). This technique is impractical when
examining clinical samples obtained from a large number of patients and when only small amounts of DNA
are available. Therefore, we developed simple PCRbased assays to type individuals for the IDRP and the 3
haplotypes shown in Figure 1.
Several PCRs were utilized (Figure 2), with DNA
from test individuals with known haplotypes. The first
PCR detects the presence of the insertion. Genotypes
that completely lack the insertion on both chromosomes
are negative in this PCR (Figure 2A). The second PCR
is a nested PCR using primers in the region of the
BV6S7 gene. The external primers yield a 311-bp product, and the internal primers yield a 233-bp product
(Figure 2B). The latter is then digested with the restriction enzyme Bam HI (Figure 2C) at a site that is present
in BV6S7*1 but not in BV6S7*2 (11). This allows
differentiation between D2 and D1 haplotypes (Figure
2C). However, these PCRs do not allow differentiation
between the I/I and I/D2 genotypes (Figure 2C). Both
genotypes are positive for the insertion PCR, and both
have the BV6S7*1 allele. These 2 genotypes can be
distinguished by Sfi I, Bgl II, or Bam HI restriction
fragment length polymorphisms (RFLPs) (7), or by a
special PCR involving a 7.90-kb amplification fragment. The latter approach was successful for samples
with large amounts of highly purified DNA (Figure
2D) and may prove useful under select circumstances.
However, this assay was not practical for most samples
obtained for the population studies performed here.
Therefore, results are presented for the I/I and I/D2
genotypes combined.
IDRP genotypes in normal controls. First, we typed
a cohort of 55 subjects with mixed ethnic backgrounds, who
lived in Marseilles. The genotypes were represented as
follows: for I/I and I/D2, 33.9%; for I/D1, 41.1%; for
D1/D2, 7.1%; for D2/D2, 0%; and for D1/D1, 17.9%. Thus,
the D2 haplotype appears to be relatively rare. Overall,
these percentages fit well with the initial findings on the
distribution of BV6S7 alleles, as performed in populations
of Western European descent (12).
A second group of 41 normal controls were from
Texas. All of these subjects were Caucasian. Overall, the
distribution of genotypes in this group was similar to
that in the French group (Table 1). A total of 85.4%
ABSENCE OF TCRBV7S2 ASSOCIATED WITH PRESENCE OF AUTOANTIBODIES IN SS
191
Figure 2. Polymerase chain reactions (PCRs) used to distinguish the 3 haplotypes. A, The 169-bp product seen
only in DNA samples from individuals with the I haplotype in at least 1 chromosome. B, PCR products of the first
(lane 1) and the second (lanes 2–7) nested PCRs used to determine BV6S7 alleles. Both PCRs were positive in all
haplotypes. C, Bam HI digestion of the second nested PCR product (233 bp) for all 6 possible genotypes. Only the
BV6S7*1 allele was digested yielding 2 bands (184 bp and 49 bp). Thus, the I/I, I/D2, and D2/D2 genotypes showed
2 bands, the I/D1 and D1/D2 genotypes showed an additional undigested fragment because the D1 haplotype does not
carry BV6S7*1, and the D1/D1 genotype showed only the undigested band. D, The 7.9-kb PCR product used to
distinguish between the D2 and I haplotypes. MM ⫽ molecular markers (see Figure 1 for other definitions).
of subjects had at least 1 chromosome with the insert (I⫹),
and 14.6% lacked the insert on both chromosomes (D/D).
Use of TCR V␤ elements dependent on IDRP
genotype. In 53 of 55 French subjects, peripheral blood
lymphocytes (PBLs) were collected and analyzed with 4
TCR V␤–specific mAb. The percentages of positive cells
per total T cells for each genotype are summarized in
Figure 3. The specificity of some of the mAb has
previously been described. Thus, the V␤6.7*1 mAb
reacts strongly with TCRs composed of BV6S7*1 alleles
(12,24). Subjects with 2 copies of the BV6S7*1 allele
generally have twice as many V␤6.7*1-positive T cells as
Table 1. Association of the IDRP haplotypes with Sjögren’s syndrome (SS)*
Genotype
Group (n)
I/I & I/D2
I/D1
D1/D2
D2/D2
D1/D1
Total I⫹
Total D/D
P†
Control (41)
SS (55)
Ro⫹ (36)
La⫺ (16)
La⫹ (20)
Any antibody (36)
No antibody (19)
SLE (57)
Ro⫹ (21)
La⫹ (9)
Lupus RNP (12)
Any antibody (35)
No antibody (20)
31.7
12.7
13.9
12.5
15.0
13.9
10.5
19.3
4.8
0.0
8.3
14.3
30.0
53.7
45.5
33.3
37.5
30.0
33.3
68.4
50.9
66.7
66.7
50.0
54.3
50.0
9.8
25.5
30.6
25.0
35.0
30.6
15.8
15.8
9.5
11.1
25.0
17.1
15.0
0.0
1.8
2.8
0.0
5.0
2.8
0.0
5.3
0.0
0.0
0.0
0.0
0.0
4.9
14.5
19.4
25.0
15.0
19.4
5.3
8.8
19.0
22.2
16.7
14.3
5.0
85.4
58.2
47.2
50.0
45.0
47.2
78.9
70.2
71.4
66.7
58.3
68.6
80.0
14.6
41.8
52.8
50.0
55.0
52.8
21.1
29.8
28.6
33.3
41.7
31.4
20.0
0.006
0.0005
0.009
0.0019
0.0004
NS
NS
NS
NS
NS
NS
NS
* Except where indicated otherwise, values are the percentage distribution of insertion/deletion-related polymorphism (IDRP) genotypes. D/D
represents the 3 genotypes that share the complete absence of the insert with BV7S2; I⫹ represents genotypes that contain the insert. If correlations
between genotypes in the different groups are considered and a Bonferroni correction is applied (online at http://home.clara.net/sisa/
bonlp.htm#Corr), the alpha level should be ⬍0.010206218 (SS), or ⬍0.007300832 (SLE). NS ⫽ not significant.
† By Fisher’s exact test, comparing the I⫹ and D/D distribution with that in the control group.
192
MANAVALAN ET AL
Figure 3. T cell receptor V␤ staining as a function of insertion/deletion-related polymorphism
genotypes. The differences in percentages of V␤13.2 cells among the different groups were not
statistically significant, and the elevation of V␤7.1 in the D1/D1 genotype is significant only when
compared with the I/I & I/D2 group (P ⬍ 0.01). The differences in V␤7.2 and V␤6.7 percentages
are highly significant. Bars show the mean and SEM. n ⫽ number of subjects representing each
haplotype (see Figure 1 for other definitions).
those who are heterozygous, and the percentage in
subjects with 2 copies of BV6S7*2 is very low (12,24).
The data in Figure 3 confirm these findings. The V␤13.2
mAb is also known to stain variable percentages of cells
(25,26). In the presence of 2 D haplotypes (D1/D1 or
D1/D2), the percentages of V␤13.2-positive T cells are
smaller than in individuals who have 1 or 2 haplotypes
bearing the insert (Figure 3). This confirms results of
prior studies (25,26).
The V␤7.1 mAb has previously been described.
All subjects would be expected to have 2 functional
copies of the BV7S1 gene, and percentages of positive
T cells should not vary between the genotypic groups.
However, the data showed an increase of V␤7.1-positive
cells in D1/D1 subjects. At present, this result remains
unexplained, because allelic variants of BV7S1 have
not been described (9). BV alleles that differ either in
the promoter region or in the recombination signal
sequences (27) can affect V␤ percentages in the PBLs,
and such potential variants have not been rigorously
excluded.
The specificity of the V␤7.2 mAb was unknown
prior to this study. A transfected cell line expressing
BV7S2 was used to immunize mice for production of the
hybridoma. Variation in the percentage of positive cells
among randomly selected subjects was noted. Figure 3
clearly shows that subjects lacking the insertion on both
chromosomes rarely express the V␤7.2 determinant recognized by the mAb. It was unclear whether subjects
with the I/I genotype have higher percentages of V␤7.2 T
cells than do subjects with the I/D genotype, as would be
expected. This is because the I/I and I/D2 genotypes
were indistinguishable, and the contribution of subjects
with the I/D2 genotype within this combined group was
not known. However, in the I/D1 group, in which all
subjects were heterozygous for the IDRP, there were
slightly lower percentages of V␤7.2-positive cells than in
the I/I and I/D2 group. In sum, the main specificity of
this new mAb is for a V␤7.2 epitope. This mAb can
therefore be used to identify subjects whose T cells are
unable to use the BV7S2 gene because it is not present
in their genome.
Association between SS and D/D genotypes. Kay
et al have described an association between homozygosity of the BV13S2*2 allele and a subset of patients with
SS (8). This subset of patients had a more aggressive
disease, with polyclonal hypergammaglobulinemia, extraglandular disease involvement, anti-La antibodies,
and HLA–DR3. The mature proteins encoded by
BV13S2*2 and BV13S2*1 are exactly the same. Thus,
ABSENCE OF TCRBV7S2 ASSOCIATED WITH PRESENCE OF AUTOANTIBODIES IN SS
use of BV13S2*2 by autoimmune or regulatory T cells
cannot explain the observed association. The BV13S2*2
allele is found exclusively on a deleted haplotype (Figure
1). Therefore, it is possible that the observed association
was actually with the deleted haplotype and with the
absence of genes contained within the deleted genomic
fragment.
Indeed, TCR genotypes obtained in patients with
SS showed a significant association between primary SS
and the complete absence of the IDRP-defined genomic
segment (D/D) (Table 1). D1/D2 subjects are heterozygous for BV13S2 and BV6S7 alleles, but these individuals have the IDRP deletion on both chromosomes. This
genotype was also significantly increased in SS (Table 1).
Thus, SS is associated with the IDRP deletion and not
specifically with the BV13S2*2 allele.
SS is also known to be associated with certain
HLA–DQ alleles (13,28). The presence of alleles encoding Q34 on the DQA chain and L26 on the DQB chain
is associated with more aggressive disease and higher
titers of autoantibodies against Ro/La (13). There was a
gene-dose effect, in that greater numbers of the implicated DQ alleles gave the more striking association (12).
In our study, 50 patients with SS were typed for
HLA–DQ. Anti-Ro/La antibodies were significantly
(P ⬍ 0.025) more frequent as the number of DQ alleles
encoding Q34/L26 increased. This appeared to hold true
regardless of TCR D/D, I/D, and I/I genotypes, but the
numbers of patients were insufficient for statistical analysis within each of these genotypic subsets.
The IDRP insertion contains only 3 known
TCRBV genes (Figure 1) (7). The entire sequence of
this genomic segment is known (GenBank accession no.
L36092), and there are no unexpected genes within it. Of
the 3 TCRBV genes, BV9S2P is a pseudogene.
BV13S2*1b represents an identical copy of BV13S2*1a.
Thus, lack of BV13S2*1b is unlikely to be of any major
consequence. Therefore, individuals who lack the IDRP
insertion on both chromosomes are distinguished primarily by their inability to produce TCRs using BV7S2.
It follows that T cells expressing BV7S2 may alter the
nature of immune responses in SS: lack of these T cells
would perhaps result in exaggerated disease-related T
cell responses, and the outcome might be worsened
clinical manifestations.
It is conceivable that lack of BV7S2 is related to
production of specific autoantibodies. Alternatively, lack
of BV7S2 could be related to severity of disease (which
is independently associated with high levels of autoantibodies). To examine this issue, a multiplex family with
SS was genotyped (Figure 4). A mother with SS (I-2)
had 5 daughters, one of whom (II-4) also had SS. This
193
Figure 4. Analysis of a multiplex family with autoimmune disease.
HLA haplotypes are indicated above each symbol. The haplotypes
were as follows: for a, DRB1*1301, DQB1*0603, DQA1*0103; for b,
DRB1*0101, DQB1*0501, DQA1*0101; for c, DRB1*0301,
DQB1*0201, DQA1*0501; for d, DRB1*1301, DQB1*0603,
DQA1*0103; e and f are unrelated haplotypes. The T cell receptor
(TCR) haplotypes were as follows (shading indicates a deleted haplotype): for I-1, D1/D2; for I-2, I/D2; for II-1, I/D1; for II-2, I/D1; for
II-3, D1/D2; for II-4, D1/D2; for II-5, D1/D2; for III-1, D1/D1. ANA ⫽
antinuclear antibody; HT ⫽ autoimmune hypothyroid disease; SS ⫽
Sjögren’s syndrome; Ro ⫽ anti-Ro antibody; La ⫽ anti-La antibody;
Raynaud’s ⫽ Raynaud’s syndrome; CVI ⫽ common variable immunodeficiency; IgA Def ⫽ IgA deficiency; IDDM ⫽ insulin-dependent
diabetes mellitus (see Figure 1 for other definitions).
individual has a sister (II-3) who is both HLA identical
(haplotypes a and c) and TCR genotype identical (D1/
D2), yet these 2 sisters are not congruent for SS.
Because II-3 has a positive antinuclear antibody test, it is
possible that the 2 sisters share specific autoantibodies,
excluding anti-Ro or anti-La. One may conclude that the
D/D genotype does not necessarily predict a particular
disease manifestation or the presence of anti-Ro/La,
even when HLA genes are identical. Other polymorphic
genes not shared by the 2 sisters and/or environmental
factors must play a role.
To examine whether the absence of the insert is
associated with production of specific autoantibodies,
we tested for different autoantibodies in SS patients with
known genotypes (Figure 5). In the group of SS patients
with the D/D genotype, antibodies to Ro and La were
more frequently detected. The P value was most significant for anti-La antibodies assessed by immunoprecipitation (P ⬍ 0.006). Autoantibodies against other target
antigens (ssDNA, chromatin) were not dependent on
the TCR genotypes.
SLE patients may also have autoantibodies to the
Ro/La complex. The frequency of such autoantibodies is
high in SS (⬃60–70%) but is lower in SLE (⬃30%) (6).
If the D/D genotype were strictly associated with the
presence of antibodies to Ro/La, we reasoned that SLE
194
MANAVALAN ET AL
Figure 5. Autoantibodies in patients with Sjögren’s syndrome. Sera from 9 D/D (deleted/deleted) patients (filled
bars) and 20 I/D (inserted/deleted) or I/I (inserted/inserted) patients (shaded bars) were compared. Statistical
analysis was by analysis of variance. Data for antibodies to single-stranded DNA (ssDNA) and chromatin were
expressed as quantitative enzyme-linked immunoassay (ELISA) units. Antinuclear antibody (ANA) was obtained
with serial dilutions of serum; numbers on the Y axis represent average inverse [inv] titers. Data from
immunoprecipitation (IP) and immunoblotting (IB) experiments were scored as positive (pos) or negative for the
indicated antibody. Other-IP ⫽ other unidentified bands observed by immunoprecipitation.
patients with such antibodies might also be predominantly of the D/D genotype. Figure 6 presents data on 56
Caucasian SLE patients with known TCR genotypes.
The presence of autoantibodies to Ro/La and other
specificities was not significantly different in patients
with genotypes that lacked the insert (D/D) versus those
with 1 or 2 copies of the insert (I/I or I/D). Autoantibodies detected by immunoprecipitation, lupus anticoagulants, and clinical manifestations of thrombotic
events including fetal loss (data not shown) occurred
more frequently in patients with the D/D genotype, but
this difference did not reach statistical significance.
Anti-La antibodies occurred in 5 of 19 patients with the
D/D genotype versus 6 of 37 patients with the I/D or I/I
genotype. To confirm these data, SLE sera were also
tested by ELISA for the presence of La and Ro 52
antibodies (data not shown). These assays were more
sensitive than the immunoprecipitation and immunoblotting assays. However, they too failed to reveal a
significant association between the D/D genotype and
the presence of anti-Ro/La antibodies in SLE patients.
In sum, the D/D genotype and the lack of BV7S2
appeared to be associated with a subset of SS patients
with high titers of Ro/La antibodies, especially those
with anti-La antibodies.
DISCUSSION
Here we describe an association between an
IDRP of the TCRBV locus and the major subset of
primary SS patients with anti-Ro/La autoantibodies. The
implication of our results is that the gene encoding
V␤7.2 may protect against the development of severe SS
or against the generation of pathogenic autoantibodies,
in particular anti-La antibodies, in SS patients. The
occurrence of SS per se was not related to TCR genotypes. Only those patients with autoantibodies tended to
have a TCR D/D genotype (Table 1). However, in SLE
patients with high titers of anti-Ro/La antibodies, the
Figure 6. Features of systemic lupus erythematosus as a function of T cell receptor genotype. Two
groups of patients, one with the D/D (deleted/deleted) genotype (n ⫽ 19) and one with the I/I
(inserted/inserted) or I/D (inserted/deleted) genotype (n ⫽ 37), were compared. Left, Quantitative
titers of antibodies to cardiolipin (aCL). Bars show the mean and SEM. Right, Percent of patients
positive for lupus anticoagulant (AC), Ro 60, La 48, Sm, RNP, ribosomal P (RibP), and other
unidentified immunoprecipitation bands.
ABSENCE OF TCRBV7S2 ASSOCIATED WITH PRESENCE OF AUTOANTIBODIES IN SS
same association with the D/D genotype was not manifest.
Several lines of evidence suggest that T cells may
be important in a disease such as SS. First, both activated B and activated oligoclonal T cells are found in the
inflamed exocrine glandular tissues of patients with SS.
Second, in a mouse model of SS, adoptive transfer of
lymphocytes from a diseased animal into a SCID recipient resulted in disease that could be prevented by
anti-CD4 and anti–TCR antibody treatment (29). Third,
T cells, like B cells, may recognize the antigens Ro
(30,31) and La (32–34). T cells with some of these
specificities are found in the salivary glands of patients
with SS (30). Fourth, SS is associated with certain HLA
class II alleles (13,35), and this association crosses racial
barriers. The DQA1 and DQB1 alleles found in SS
invariably encoded a Q34 residue (DQA1) and an L26
residue (DQB1), regardless of the genetic background
of the patients (13,35). Both Q34 and L26 are situated in
the floor of the class II major histocompatibility
complex–binding groove, suggesting a role in presenting
autoantigens to CD4 T cells in SS. This HLA–DQ
association is primarily with the anti-Ro/La autoantibody response and does not apply to the minority of
patients without anti-Ro antibody (35). Moreover, these
DQ alleles appear to control diversification of the
anti-Ro/La response (i.e., intermolecular epitope
spreading) (36).
The simplest hypothetical model to explain the
findings presented herein is that V␤7.2-positive T cells
are involved in down-modulation of disease manifestations in SS. One possibility is that such T cells may
participate in the regulation of epitope spreading.
Epitope spreading is a process by which autoantibody
specificities to autoantigen complexes spread from an
initial target epitope to other epitopes. If epitope
spreading is limited to the same polypeptide, it is
referred to as intramolecular spreading. When epitope
spreading involves noncovalently associated polypeptides, it is called intermolecular spreading. Epitope
spreading is known to occur with autoantibodies to the
Ro/La antigen complex (34,37). It is currently thought
that Ro 60, Ro 52, La, and other proteins are transiently
associated in an immunogenic RNA/protein particle
(6,38,39). This allows efficient intermolecular epitope
spreading, as demonstrated in mice immunized with a
conserved Ro 60 epitope, which then develop autoantibodies not only to Ro 60 but also to Ro 52 and La
(38,40,41).
When autoantibodies to La are observed, they
generally do not occur in the absence of anti-Ro antibodies. This is thought to be attributable to regulated
195
progression of intermolecular epitope spreading, in
which autoantibodies to Ro 60 are the first to occur,
and autoantibodies to La 48 are the last to be observed.
The TCR D/D relationship described herein was most
tightly associated with the presence of anti-La antibodies (Table 1 and Figure 5). However, TCR D/D
genotypes were not observed in SLE patients with these
antibodies. Thus, if V␤7.2 T cells somehow inhibit epitope spreading and thus formation of anti-La antibodies,
this process would have to be applicable only to patients
with SS.
Epitope spreading can modulate disease expression in different ways. In experimental adjuvant-induced
arthritis of the Lewis rat, a transient disease induced by
immunization with adjuvant containing mycobacterial
antigens, there is T cell epitope spreading to the
carboxy-terminal peptides of mycobacterial hsp65. The
T cell clones with these specificities are protective when
administered prior to initial immunization. When they
arise as a consequence of epitope spreading, they are
thought to down-modulate disease expression, resulting
in the transient nature of this disease (42). However,
epitope spreading is more commonly associated with
progression of disease (e.g., in experimental autoimmune encephalomyelitis [EAE], a model for human
multiple sclerosis) (43–46).
An alternative possibility is that V␤7.2 T cells
promote antiidiotypic responses, which can mask antiLa antibodies (47). Whether these responses differ in
patients with SS and those with SLE is not known.
Prior studies have examined the TCR repertoire
in samples from SS patients. Several reports have described elevated percentages of V␤2 and V␤13 CD4⫹ T
cells in either blood or salivary gland biopsy samples
(48–53), but the results are quite variable and may
depend on disease activity (54) and the tissue site
analyzed (55). V␤7 cells were usually not prominent in
these studies, in which PCR primers were used to
specifically detect the BV7 transcripts (a method that
should detect all 3 BV7 genes but does not distinguish
between the 3 different BV7 genes). In one study (52)
using samples from Japanese patients with primary SS or
with human T lymphotropic virus type I–associated SS,
BV7 transcripts were elevated and clonally restricted. A
conserved third complementarity-determining region
motif (QDXG) was noted among these BV7 transcripts.
These BV7 transcripts were present in 12 of 13 patients
but not in controls, and in both blood and salivary tissue.
In summary, available data do not clearly indicate
whether V␤7.2 T cells participate in the autoimmune
response or whether they might recognize peptide antigens from the Ro/La complex.
196
Another hypothesis to explain the present findings is that V␤7.2 T cells have a T regulatory (Treg or T
suppressor) function. For example, they might downmodulate the pathogenic helper T cells that result in T
cell help for autoantibody formation. There is some
evidence for the presence of such T cells in the murine
EAE system. For example, deletion of V␤14 and V␤3,
used by such regulatory T cells, led to increased disease
severity (56). But classic Treg cells are not thought to
have restricted TCRBV gene usage, nor do they inhibit
a specific autoimmune response (57).
Finally, another hypothesis is that V␤7.2 T cells
alter the production of key cytokines. T cells may direct
an autoimmune response in different ways, depending
on the cytokine profile produced. In EAE, pathogenic
myelin basic protein–specific Th1 CD4 effector cells
could be deviated to Th2 cells by immune regulatory T
cells, resulting in protection from disease (58). In a
mouse model of SS, there is also a regulatory population
of CD4 cells that prevents disease by producing
interleukin-4 (IL-4), IL-10, and transforming growth
factor ␤ (59). In SLE and SS, interferon-␥ (IFN␥)
appears to be one of the key pathogenic cytokines,
possibly acting by stimulating antigen-presenting cells to
secrete pathogenic cytokines (IL-6, tumor necrosis factor ␣) or by stimulating autoantigen presentation (60–
63). T cells reactive with Ro/La can also produce IFN␥
(32). Perhaps such a cytokine response differs if the
relevant T cells use BV7S2 as opposed to other BV
genes.
Polymorphism in the human TCR loci was discovered soon after the TCR genes were identified
(12,64). However, disease associations with TCR alleles
have not been forthcoming (with few exceptions), in
contrast to the many known disease associations with
major histocompatibility complex polymorphisms. A
TCR B association with multiple sclerosis was demonstrated by careful sibpair analysis (65). The mechanism
by which TCR genes influence multiple sclerosis susceptibility remains unclear (66). Another report described a
null allele of BV6S1 associated with a subset of patients
with juvenile rheumatoid arthritis (67,68). How the lack
of V␤6.1-expressing T cells affects susceptibility to this
disease is still a mystery.
Earlier studies, including studies on SLE, used
RFLPs for Bgl II and Kpn I polymorphic sites in the
vicinity of the TCR BC locus encoding the D␤, J␤, and
C␤ segments. There is no linkage between these markers
and the IDRP studied herein, which is located ⬃600 kb
upstream. An early report by Frank et al described an
association between Ro⫹ SLE, but not Ro⫺ SLE, with
a combination of 2 RFLPs (69). Although the authors
MANAVALAN ET AL
could not explain the mechanism underlying this association, it is possible that polymorphisms of the BC
locus are associated with different D␤ or J␤ usage and
thus affect the expressed TCR repertoire. A followup
report by the same authors (70) showed that the TCR
BC polymorphism predicted high-titer (⬎1:106) anti-Ro
antibody when patients also had particular HLA–
DQA1/DQB1 alleles (DQA1*01, DQB1*0201). These
alleles encode the Q34 and L26 residues mentioned
above (35). Anti-Ro antibodies directed to a particular
Ro 60 epitope, a 15-mer termed A480, were most
strongly associated with the TCR Bgl II 9.8-kb allele
(71).
Next came the report (8) linking severe cases of
SS (with high titers of anti-Ro/La) with TCR BV13S2*2
as opposed to BV13S2*1. BV13S2*2 is present only on
the D1 haplotype (Figure 1). Thus, our data are compatible with this earlier finding. They also illuminate a
potential mechanism. The BV13S2 alleles do not differ
in the expressed amino acid sequence, and the relevant
difference is the presence/absence of the 26-kb genomic
segment, and of BV7S2 in particular. Thus, in general,
there are hints throughout the literature that a combination of HLA class II alleles and the right kind of TCR
is required for T cells to direct a pathogenic autoimmune response. The current findings open up new
possibilities for investigating the mechanisms by which T
cells exert this control in SS.
ACKNOWLEDGMENT
We thank Corinne Leget at Immunotech for her
technical expertise.
REFERENCES
1. Ortman RA, Holderbaum D, Qu XM, Banerjee S, Haqqi TM.
BUB/BnJ (H-2q) is a TCR deletion mutant mouse strain (TCR V
beta a, KJ16-) that is susceptible to type II collagen-induced
arthritis. J Immunol 1994;152:4175–82.
2. Nabozny GH, Bull MJ, Hanson J, Griffiths MM, Luthra HS, David
CS. Collagen-induced arthritis in T cell receptor V beta congenic
B10.Q mice. J Exp Med 1994;180:517–24.
3. Osman GE, Hannibal MC, Anderson JP, Lasky SR, Ladiges WC,
Hood L. FVB/N (H2(q)) mouse is resistant to arthritis induction
and exhibits a genomic deletion of T-cell receptor V beta gene
segments. Immunogenetics 1999;49:851–9.
4. Noonan DJ, McConahey PJ, Cardenas GJ. Correlations of autoimmunity with H-2 and T cell receptor beta chain genotypes in
(NZB ⫻ NZW) F2 mice. Eur J Immunol 1990;20:1105–10.
5. Fox RI. Sjögren’s syndrome. In: Kelley WN, Harris ED, Ruddy S,
Sledge CG, editors. Textbook of rheumatology. Philadelphia:
Saunders; 1997. p. 955.
6. Chan EK, Andrade LE. Antinuclear antibodies in Sjögren’s syndrome. Rheum Dis Clin North Am 1992;18:551—70.
7. Zhao TM, Whitaker SE, Robinson MA. A genetically determined
insertion/deletion related polymorphism in human T cell receptor
ABSENCE OF TCRBV7S2 ASSOCIATED WITH PRESENCE OF AUTOANTIBODIES IN SS
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
beta chain (TCRB) includes functional variable gene segments.
J Exp Med 1994;180:1405–14.
Kay RA, Hutchings CJ, Ollier WE. A subset of Sjögren’s syndrome
associates with the TCRBV13S2 locus but not the TCRBV2S1
locus. Hum Immunol 1995;42:328–30.
Arden B, Clark SP, Kabelitz D, Mak TW. Human T-cell receptor
variable gene segment families. Immunogenetics 1995;42:455–500.
Tan EM, Cohen AS, Fries JF, Masi AT, McShane DJ, Rothfield
NF, et al. The 1982 revised criteria for the classification of systemic
lupus erythematosus. Arthritis Rheum 1982;25:1271–7.
Li Y, Szabo P, Robinson MA, Dong B, Posnett DN. Allelic
variations in the human T cell receptor V␤6.7 gene products. J Exp
Med 1990;171:221–30.
Posnett DN, Wang CY, Friedman S. Inherited polymorphism of
the human T-cell antigen receptor detected by a monoclonal
antibody. Proc Natl Acad Sci U S A 1986;83:7888–92.
Reveille JD, Macleod MJ, Whittington K, Arnett FC. Specific
amino acid residues in the second hypervariable region of HLADQA1 and DQB1 chain genes promote the Ro (SS-A)/La (SS-B)
autoantibody responses. J Immunol 1991;146:3871–6.
Travaglio-Encinoza A, Chaouni I, DerSimonian H, Jorgensen C,
Simony-Lafontaine J, Romagne F, et al. T cell receptor distribution in rheumatoid synovial follicles. J Rheumatol 1995;22:394–9.
Posnett DN, Romagne F, Necker A, Kotzin BL, Sekaly R-P.
Summary: first human TCR monoclonal antibody workshop. Immunologist 1996;4:5–8.
Satoh M, Langdon JJ, Hamilton KJ, Richards HB, Panka D,
Eisenberg RA, et al. Distinctive immune response patterns of
human and murine autoimmune sera to U1 small nuclear ribonucleoprotein C protein. J Clin Invest 1996;97:2619–26.
Reeves WH, Satoh M, McCauliffe DP. Autoantibody testing by
non-FITC methods. In: Rose NR, Hamilton RG, Detrick B,
editors. Manual of clinical laboratory immunology. Washington,
DC: American Society of Microbiology Press; 2002. p. 933.
Wang J, Satoh M, Kabir F, Shaw M, Domingo MA, Mansoor R, et
al. Increased prevalence of autoantibodies to ku antigen in African
American versus white patients with systemic lupus erythematosus. Arthritis Rheum 2001;44:2367–70.
Schneider C, Newman RA, Sutherland DR, Asser U, Greaves MF.
A one-step purification of membrane proteins using a high efficiency immunomatrix. J Biol Chem 1982;257:10766–9.
Bradwell AR. Immunofluorescent antinuclear antibody tests. In:
Rose NR, Hamilton RG, Detrick B, editors. Manual of clinical
laboratory immunology. Washington, DC: American Society of
Microbiology Press; 2002. p. 922.
Satoh M, Kumar A, Kanwar YS, Reeves WH. Anti-nuclear
antibody production and immune-complex glomerulonephritis in
BALB/c mice treated with pristane. Proc Natl Acad Sci U S A
1995;92:10934–8.
Satoh M, Weintraub JP, Yoshida H, Shaheen VM, Richards HB,
Shaw M, et al. Fas and Fas ligand mutations inhibit autoantibody
production in pristane-induced lupus. J Immunol 2000;165:
1036–43.
Seboun E, Houghton L, Hatem CJ Jr, Lincoln R, Hauser SL.
Unusual organization of the human T-cell receptor ␤-chain gene
complex is linked to recombination hotspots. Proc Natl Acad Sci
U S A 1993;90:5026–9.
Liao L, Marinescu A, Molano A, Ciurli C, Sekaly R-P, Fraser JD,
et al. TCR binding differs for a bacterial superantigen (SEE) and
a viral superantigen (Mtv-9). J Exp Med 1996;184:1471–82.
Choi Y, Kotzin B, Lafferty J, White J, Pigeon M, Kubo R, et al. A
method for production of antibodies to human T-cell receptor
␤-chain variable regions. Proc Natl Acad Sci U S A 1991;88:
8357–61.
Akolkar PN, Gulwani-Akolkar B, Robinson MA, Silver J. The
influence of non-HLA genes on the human T-cell receptor repertoire. Scand J Immunol 1995;42:248–56.
Posnett DN, Vissinga CS, Pambuccian C, Wei S, Robinson MA,
Kostyu D, et al. Level of human TCRBV3S1 (V␤3) expression
28.
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.
39.
40.
41.
42.
43.
44.
45.
197
correlates with allelic polymorphism in the spacer region of the
recombination signal sequence. J Exp Med 1994;179:1707–11.
Harley JB, Reichlin M, Arnett FC, Alexander EL, Bias WB,
Provost TT. Gene interaction at HLA-DQ enhances autoantibody
production in primary Sjögren’s syndrome. Science 1986;232:
1145–7.
Hayashi Y, Haneji N, Hamano H, Yanagi K. Transfer of Sjögren’s
syndrome-like autoimmune lesions into SCID mice and prevention of lesions by anti-CD4 and anti-T cell receptor antibody
treatment. Eur J Immunol 1994;24:2826–31.
Namekawa T, Kuroda K, Kato T, Yamamoto K, Murata H,
Sakamaki T, et al. Identification of Ro(SSA) 52 kDa reactive T
cells in labial salivary glands from patients with Sjögren’s syndrome. J Rheumatol 1995;22:2092–9.
Sumida T, Namekawa T, Maeda T, Nishioka K. New T-cell
epitope of Ro/SS-A 52 kDa protein in labial salivary glands from
patients with Sjögren’s syndrome [letter]. Lancet 1996;348:1667.
Halse AK, Wahren M, Jonsson R. Peripheral blood in Sjögren’s
syndrome does not contain increased levels of T lymphocytes
reactive with the recombinant Ro/SS-A 52 kD and La/SS-B 48 kD
autoantigens. Autoimmunity 1996;23:25–34.
Reynolds P, Gordon TP, Purcell AW, Jackson DC, McCluskey J.
Hierarchical self-tolerance to T cell determinants within the
ubiquitous nuclear self-antigen La (SS-B) permits induction of
systemic autoimmunity in normal mice. J Exp Med 1996;184:
1857–70.
Farris AD, Brown L, Reynolds P, Harley JB, James JA, Scofield
RH, et al. Induction of autoimmunity by multivalent immunodominant and subdominant T cell determinants of La (SS-B). J Immunol 1999;162:3079–87.
Reveille JD, Arnett FC. The immunogenetics of Sjögren’s syndrome. Rheum Dis Clin North Am 2000;18:539–50.
Rischmueller M, Lester S, Chen Z, Champion G, Van Den Berg
R, Beer R, et al. HLA class II phenotype controls diversification of
the autoantibody response in primary Sjögren’s syndrome (pSS).
Clin Exp Immunol 1998;111:365–71.
Deshmukh US, Lewis JE, Gaskin F, Kannapell CC, Waters ST,
Lou YH, et al. Immune responses to Ro60 and its peptides in
mice. I. The nature of the immunogen and endogenous autoantigen determine the specificities of the induced autoantibodies.
J Exp Med 1999;189:531–40.
McCluskey J, Farris AD, Keech CL, Purcell AW, Rischmueller M,
Kinoshita G, et al. Determinant spreading: lessons from animal
models and human disease. Immunol Rev 1998;164:209–29.
Scofield RH, Farris AD, Horsfall AC, Harley JB. Fine specificity
of the autoimmune response to the Ro/SSA and La/SSB ribonucleoproteins. Arthritis Rheum 1999;42:199–209.
Scofield RH, Kaufman KM, Baber U, James JA, Harley JB,
Kurien BT. Immunization of mice with human 60-kd Ro peptides
results in epitope spreading if the peptides are highly homologous
between human and mouse. Arthritis Rheum 1999;42:1017–24.
Scofield RH, Henry WE, Kurien BT, James JA, Harley JB.
Immunization with short peptides from the sequence of the
systemic lupus erythematosus-associated 60-kDa Ro autoantigen
results in anti-Ro ribonucleoprotein autoimmunity. J Immunol
1996;156:4059–66.
Moudgil KD. Diversification of response to hsp65 during the
course of autoimmune arthritis is regulatory rather than pathogenic. Immunol Rev 1998;164:175–84.
McRae BL, Vanderlugt CL, Dal Canto MC, Miller SD. Functional
evidence for epitope spreading in the relapsing pathology of
experimental autoimmune encephalomyelitis. J Exp Med 1995;
182:75–85.
Vanderlugt CL, Miller SD. Epitope spreading in immune-mediated diseases: implications for immunotherapy. Nat Rev Immunol
2002;2:85–95.
McFarland HI, Lobito AA, Johnson MM, Nyswaner JT, Frank JA,
Palardy GR, et al. Determinant spreading associated with demy-
198
46.
47.
48.
49.
50.
51.
52.
53.
54.
55.
56.
MANAVALAN ET AL
elination in a nonhuman primate model of multiple sclerosis.
J Immunol 1999;162:2384–90.
Yu M, Johnson JM, Tuohy VK. A predictable sequential determinant spreading cascade invariably accompanies progression of
experimental autoimmune encephalomyelitis: a basis for peptidespecific therapy after onset of clinical disease. J Exp Med 1996;
183:1777–88.
Routsias JG, Touloupi E, Dotsika E, Moulia A, Tsikaris V,
Sakarellos C, et al. Unmasking the anti-La/SSB response in sera
from patients with Sjögren’s syndrome by specific blocking of
anti-idiotypic antibodies to La/SSB antigenic determinants. Mol
Med 2002;8:293–305.
Sumida T, Yonaha F, Maeda T, Tanabe E, Koike T, Tomioka H,
et al. T cell receptor repertoire of infiltrating T cells in lips of
Sjögren’s syndrome patients. J Clin Invest 1992;89:681–5. Erratum
in: J Clin Invest 1992;90:293.
Pivetta B, De Vita S, Ferraccioli G, De Re V, Gloghini A,
Marzotto A, et al. T cell receptor repertoire in B cell lymphoproliferative lesions in primary Sjögren’s syndrome. J Rheumatol
1999;26:1101–9.
Yonaha F, Sumida T, Maeda T, Tomioka H, Koike T, Yoshida S.
Restricted junctional usage of T cell receptor V␤2 and V␤13
genes, which are overrepresented on infiltrating T cells in the lips
of patients with Sjögren’s syndrome. Arthritis Rheum 1992;35:
1362–7.
Ohyama Y, Nakamura S, Matsuzaki G, Shinohara M, Hiroki A,
Fujimura T, et al. Cytokine messenger RNA expression in the
labial salivary glands of patients with Sjögren’s syndrome. Arthritis
Rheum 1996;39:1376–84.
Karim SN, Murphy EA, Sturrock RD, Goudie RB. T-cell clonality
in synovial fluid from rheumatoid joints before and after culture in
interleukin-2. Br J Rheumatol 1995;34:232–5.
Dwyer E, Itescu S, Winchester R. Characterization of the primary
structure of T cell receptor ␤-chains infiltrating the salivary gland
in the sicca syndrome of HIV-1 infection: evidence of antigendriven clonal selection suggested by restricted combinations of V␤
and J␤ gene segment usage and shared somatically encoded amino
acid residues. J Clin Invest 1993;92:495–502.
Legras F, Martin T, Knapp AM, Pasquali JL. Infiltrating T cells
from patients with primary Sjögren’s syndrome express restricted
or unrestricted T cell receptor V beta regions depending on the
stage of the disease. Eur J Immunol 1994;24:181–5.
Sumida T, Matsumoto I, Maeda T, Nishioka K. T-cell receptor in
Sjögren’s syndrome. Br J Rheumatol 1997;36:622–9.
Kumar V, Stellrecht K, Sercarz E. Inactivation of T cell receptor
peptide-specific CD4 regulatory T cells induces chronic experimental autoimmune encephalomyelitis (EAE). J Exp Med 1996;
184:1609–17.
57. Shevach EM. Regulatory T cells in autoimmmunity. Annu Rev
Immunol 2000;18:423–49.
58. Kumar V, Sercarz E. Induction or protection from experimental
autoimmune encephalomyelitis depends on the cytokine secretion
profile of TCR peptide-specific regulatory CD4 T cells. J Immunol
1998;161:6585–91.
59. Saegusa K, Ishimaru N, Yanagi K, Haneji N, Nishino M, Azuma
M, et al. Autoantigen-specific CD4⫹CD28low T cell subset prevents autoimmune exocrinopathy in murine Sjögren’s syndrome.
J Immunol 2000;165:2251–7.
60. Lawson BR, Prud’homme GJ, Chang Y, Gardner HA, Kuan J,
Kono DH, et al. Treatment of murine lupus with cDNA encoding
IFN-␥R/Fc. J Clin Invest 2000;106:207–15.
61. Reininger L, Santiago ML, Takahashi S, Fossati L, Izui S. T helper
cell subsets in the pathogenesis of systemic lupus erythematosus.
Ann Med Interne (Paris) 1996;147:467–71.
62. Peng SL, Moslehi J, Craft J. Roles of interferon-␥ and interleukin-4 in murine lupus. J Clin Invest 1997;99:1936–46.
63. Kirou KA, Crow MK. New pieces to the SLE cytokine puzzle. Clin
Immunol 1999;91:1–5.
64. Robinson MA, Kindt TJ. Segregation of polymorphic T-cell
receptor genes in human families. Proc Natl Acad Sci U S A
1985;82:3804–8.
65. Seboun E, Robinson MA, Doolittle TH, Ciulla TA, Kindt TJ,
Hauser SL. A susceptibility locus for multiple sclerosis is linked
to the T cell receptor beta chain complex. Cell 1989;57:1095–100.
66. Wei S, Charmley P, Birchfield RI, Concannon P. Human T-cell
receptor V␤ gene polymorphism and multiple sclerosis. Am J
Hum Genet 1995;56:963–9.
67. Maksymowych WP, Gabriel CA, Luyrink L, Melin-Aldana H,
Elma M, Giannini EH, et al. Polymorphism in a T-cell receptor
variable gene is associated with susceptibility to a juvenile rheumatoid arthritis subset. Immunogenetics 1992;35:257–62.
68. Luyrink L, Gabriel CA, Thompson SD, Grom AA, Maksymowych
WP, Choi E, et al. Reduced expression of a human V␤ 6.1 T-cell
receptor allele. Proc Natl Acad Sci U S A 1993;90:4369–73.
69. Frank MB, McArthur R, Harley JB, Fujisaku A. Anti-Ro (SSA)
autoantibodies are associated with T cell receptor beta genes in
systemic lupus erythematosus patients. J Clin Invest 1990;85:33–9.
70. Scofield RH, Frank MB, Neas BR, Horowitz RM, Hardgrave KL,
Fujisaku A, et al. Cooperative association of T cell beta receptor
and HLA-DQ alleles in the production of anti-Ro in systemic
lupus erythematosus. Clin Immunol Immunopathol 1994;72:
335–41.
71. Scofield RH, Dickey WD, Hardgrave KL, Neas BR, Horowitz
RM, McArthur RA, et al. Immunogenetics of epitopes of the
carboxyl terminus of the human 60-kD Ro autoantigen. Clin Exp
Immunol 1995;99:256–61. Erratum in: Clin Exp Immunol 1995;
100:377.
Документ
Категория
Без категории
Просмотров
3
Размер файла
367 Кб
Теги
presence, absence, autoantibodies, syndrome, sjgren, genes, receptov, genomics, cells, linked, encoding
1/--страниц
Пожаловаться на содержимое документа