вход по аккаунту


HLA-B27-derived peptides as autoantigens for T lymphocytes in ankylosing spondylitis.

код для вставкиСкачать
Vol 40, No 1 1. November 1997, pp 2047-2054
0 1997, American College of Rheumatology
Objective. To study whether peptides derived from
the HLA-B27 molecule sequence can stimulate peripheral blood T lymphocytes (PBL) from patients with
HLA-B27-associated spondylarthropathies.
Methods. PBL from 55 HLA-B27+ patients with
ankylosing spondylitis (AS), 28 HLA-B27+ patients
with other spondylarthropathies, 7 rheumatoid arthritis
patients, and 30 HLA-B27+ and 22 HLA-B27- healthy
controls were tested in lymphocyte proliferation assays
with 4 synthetic peptides derived from the HLA-B*2705
Results. A 13-mer peptide (B27PA) induced significant proliferative responses in 17 of the 55 AS
patients (stimulation index [SI]2.5-17.5), as well as in
3 of the HLA-B27+ healthy controls (SI 2.5-9.8).
Another 13-mer peptide (B27PC) induced PBL proliferation (SI 2.7-5.5) in 10 AS patients and in some donors
of the control groups. In B27PA-specific T cell lines, an
expansion of cells positive for the y/S T cell receptor
could be demonstrated.
Conclusion. These results indicate that HLAB27-derived peptides can be recognized as autoantigens
by PBL of HLA-B27+ AS patients and B27+ healthy
controls. Recent infections preceding the manifestation
of AS may be involved in this process of anti-self major
histocompatibility complex reactivity.
Since the demonstration of the strong association
between the major histocompatibility complex (MHC)
Dr. MBrker-Hermann’s and Dr. Wildner’s work was supported by the Deutsche Forschungsgcmcinschaft (projects SFB 311
and A12 and projcct Wi 138211.1, respectively).
Elisabcth Marker-Hermann, MD, Karl-Hermann Meyer zum
Biischenfelde, MD, PhD: Johannes-Gutenberg-University, Mainz,
Germany; Gerhild Wildner, PhD: Univcrsity Eye Hospital, Munich,
Address reprint requests to Elisabeth Mirker-Hermann, MD,
Department of Medicinc I , Johannes Gutenberg University of Mainz,
D-55 10I Mainz, Germany.
Submitted for publication January 29, 1997; accepted in
revised form June 19, 1997.
class I molecule HLA-B27 and ankylosing spondylitis
(AS) in 1973 (l), multiple hypotheses have been proposed trying to explain this linkage between an HLA
allele and disease susceptibility in humans. Among these
hypotheses, (i) molecular mimicry between HLA-B27
and enterobacterial antigens (2,3), (ii) presentation of
bacterial and self “arthritogenic” peptides by HLA-B27
to CD8+ T cells (4-6), and (iii) HLA-B27-induced
modification of bacterial invasion (7) have currently
been under investigation.
Animal models of transgenic rats expressing high
copy numbers of HLA-B27 and human p2microglobulin (&m) have confirmed the role of HLAB27 in disease pathogenesis (8). Furthermore, the presence of T cells (9) and of physiologic gut flora (10) has
been demonstrated to be crucial for disease manifestation. Interestingly, the development of spontaneous inflammatory spondylarthropathy-like disease has recently
been reported in HLA-Ba7-transgenic mice lacking
&rn (11). In these mice, the HLA-B27 heavy chain in
the absence of P2m is not expressed on the cell surface,
and the number of circulating CD8+ T lymphocytes is
very low. This animal model suggests that aberrant
assembly, transport, and expression of HLA-B27 may
predispose to spondylarthropathy. The HLA-B27 heavy
chain itself may be retained in the endoplasmic reticulum or the cytoplasm, undergo protein degradation, and
be presented by class 11 MHC molecules to CD4+ T
cells (11). Thus, peptides derived from the HLA-B27 a
chain, probably mimicking epitopes of organ-specific
autoantigens, may become candidate autoantigens in
the development of AS or related spondylarthropathies. That mechanism has previously been described
for autoimmune uveitis, another HLA class I-associated
disease (12).
In this study, we examined several synthetic peptides derived from the HLA-B27 hypervariable chain.
We sought to determine whether peripheral blood lymphocytes (PBL) from AS patients proliferated in re-
Table 1. Amino acid (AA) sequence, position within the HLA molecule, and specificity of HLA-B27- and HLA-B7-derived peptides used lor
T cell proliferation
AA position in
HLA molecule
AA sequence
B27 AA 60-72
B27 AA 109-118
B27 AA 151-165
B27 AA 125-138
B7 AA 125-138
B*2701, B*2702, B*2703, B*2704, B*2705, B*4701
B"2701, B*2702, B"2703, B*2705, B*2707, B13 (B*1301, B*1302), B*4011,
B*4002, B*4003, B*4004, B"4006, B"4701, B*4801
All HLA-B except B*2707, B7, B8, B"4001-4006, B*4101, B*4201, B*4801
B*2707, B7, B8, B*4001-4006, B*4101, B*4201, B*4801
sponse to one of these peptides, and we compared our
findings with PBL proliferation in patients with other
HLA-B27-associated spondylarthropathies and in
HLA-B27+ and HLA-B27- healthy control subjects.
Patients and controls. PBL were obtained from 55
HLA-B27+ patients who fulfilled thc modified New York
Criteria for definite AS (13) (36 men and 19 women; mean age
40.9 years, age range 18-71). All AS patients met the requirements of having evidence of clinically active disease, which was
defined as inflammatory-type back pain requiring treatment
with nonsteroidal antiinflammatory drugs, morning stiffness
of 245 minutes' duration, and/or active peripheral arthritis.
We also evaluated PBL samples from 28 HLA-B27+
patients with other spondylarthropathies (5 with reactive arthritis, 10 with psoriatic arthritis, and 13 with enteropathic
arthritis in association with Crobn's disease or ulcerative
colitis; 15 men and 13 women, mean age 42.4 years, age range
22-76), from 30 HLA-B27+ healthy blood donors (18 men
and 12 women; mean age 46.2 years, age range 20-78), from 22
HLA-B27- healthy blood donors (10 men and 12 women;
mean age 30.8 years, age range 22-53), and from 7 HLA-B27patients with active rheumatoid arthritis (RA) (3 men and 4
women; mean age 56.8 years, age rangc 30-72 years). None of
the control blood donors had a personal or family history of
spondylarthropathies or evidence of a recent gastrointestinal
or urogenital infection.
Separation of PBL. PBL obtained from heparinized
peripheral blood were isolated by standard centrifugation on a
Ficoll-Hypaque gradient, washed 3 times, and resuspended in
culture medium RH10 ( i t . , RPMI 1640 [Gibco, Karlsruhe,
Germany] containing 2 mmoles of L-glutamine (Sigma,
Deisenhofen, Germany), HEPES (Gibco), 100 unitsiml of
penicillin (Gibco), 100 pg/ml of streptomycin (Gibco), and
10% heat-inactivated human AB serum).
Peptides. Peptide., B27PA (patent pending), B27PB,
B27PC, and B27PD from polymorphic regions of the HLAB27 molecule were originally chosen according to the ChouFasman prediction of protein conformation (14), based on
their B27-specific amino acid sequences for the generation of
monoclonal antibodies (Wildner G: unpublished observations). We selected those regions from the HLA-B27 a chain
that were hydrophilic and/or contained a-helical structures.
The amino acid sequences of these peptides are shown in
Table 1.
B27PD has previously been demonstrated to be uveitogenic in the rat model of experimental autoimmune uveitis
(EAU) (12). Peptide B7PD represents the sequence of some
HLA-B antigens including HLA-B7, which differ from peptide B27PD by a single amino acid exchange, and was used as
a control (Table 1). Peptides B27PA, B27PB, and B27PC were
synthesized on a semiautomatic peptide synthesizer and controlled for purity; peptides B27PD and B7PD were obtained
from Neosystems, Strasbourg, France. All peptides were dissolved in unsupplemented RPMI 1640.
Lymphocyte proliferation assay. PBL were placed in
96-well round-bottomed microtiter plates (Nunc, Roskilde,
Denmark) at 5 X lo4 cells per well, with or without stimulants,
to a final volume of 100 pl of culture medium RH10. The
peptides were used at a final concentration of 10 pgiml. All
tests were performed in triplicate. Cultures were incubated for
6 days at 37°C in a humidified atmosphere of 5% CO, and 95%
air. On day 5, 0.25 pCi of tritiated thymidine (3H-TdR;
Amersham, Braunschweig, Germany) was added to each well.
On day 6, the cells were harvested onto glass filter paper by
standard harvesting procedures. 'H-TdR incorporation was
quantitated by liquid scintillation counting. The results were
expressed as the mean counts per minute incorporated into
triplicate cultures and as stimulation indices (SI; [cpm with
stimulant - cpm without stimulant]/[cpm without stimulant]).
Generation of peptide-specific T cell lines. PBL (3 X
10") were incubated with peptide B27PA at 10 pgiml in 12-well
plates at a volume of 2 ml of RH10 culture medium. The PBL
were derived from AS patients AS1 and AS3 and from an
HLA-B27+ healthy blood donor (Hl), who had shown siguificant proliferative responses to peptide B27PA in the initial
lymphocyte proliferation assay. PBL from another AS paticnt
(AS2), who had no proliferative response to B27PA, were used
as a control. After 3 days of culture, the medium was replaced
by fresh RHlO supplemented with 20 IU/ml of recombinant
interleukin-2 (rIL-2; Boehringer, Tngelheim, Germany). On
day 10, the cultures were washed and restimulated with
autologous irradiated (4,000 rads) PBL as feeder cells, peptide
B27PA at 10 p.giml, and rIL-2-supplemented RH10. For
restimulation, 2 X 10' feeder cells were used per 1 X lo6
responder T cells. On day 20, the resulting T cell lines were
functionally characterized in a lymphocyte proliferation assay
using 1 X 10' T cells as responder and 1 X 10' autologous Ecells (i.e., monocyteiB cell-enriched fractions separated from
autologous PBL by sheep erythrocyte rosetting) as feeder cclls
in a total volume of 150 pl RH10, as described previously (15).
Peptide B27PA (at 10 or 1 pg/ml), B27PB, B27PC, B27PD,
and B7PD (at 10 pgiml) were used as stimulants; phytohemagglutinin (PHA) at 1 pdml (Seromed, Berlin, Germany) was
used as a positive control.
To test the ability of monoclonal antibodies (MAb) to
block peptide-specific T cell proliferation, MAb W6132 (antiHLA class I) at 1:400, DA6.231 (anti-HLA class 11) at 1:400
(both antibodies were a kind gift of Prof. Bernhard Fleischer,
Hamburg, Germany), or TCR61 (specific for a framework
determinant on the 6 chain of the y/S T cell receptor [TCR])
(T Cell Sciences, Cambridge, MA) at 1 pgiml was added to
B27PA-containing cultures. The cells were pulsed with 0.25
$3 of 'H-TdR on day 3 of the cultures, and were harvested
and counted as described above.
Phenotypic analysis (fluorescence-activated cell
sorter) of peptide-stimulated T cell lines. Cell sorter analysis
of the T cell lines as well as of unstimulated PBL was
performed on day 8 and day 20 of culture. Cells (3 X 10"per
test) were labeled first with the unconjugated MAb antiLCD3
(OKT3), antiLCD4 (OKT4), anti-CD8 (OKT8) (hybridoma
supernatants), BMA.031 (specific for a framework determinant on TCRaIP; kind gift of Dr. R. Kurrle, Behringwerke,
Marburg, Germany), TCR61, and then, with fluorescein
isothiocyanate-conjugated polyclonal goat anti-mouse Ig antibody (Medac, Hamburg, Germany) in RPMI 1640110% fetal
calf serum. Samples were washed twice between each step.
Cytofluorometric analysis, counting 6,000 cells per sample, was
performed on a fluorescence-activated cell sorter (FACStar;
Becton Dickinson, Mountain View, CA).
Statistical analysis. The groups of patients and normal
controls were compared for the SIs of peptide-induced PBL
proliferation by Mann-Whitney rank sum test, using the Stat
Exact program.
8 -
6 4 -
Proliferative responses of PBL to HLA-B27derived peptides. As demonstrated in Figure 1, peptide
B27PA induced significant PBL proliferation (SI >2.5)
in 17 of the 55 AS patients terted (SI 2.5-17.5), as well
as in 3 of the 30 HLA-B27+ healthy blood donors (SI
2.5-9.8). The difference between the group of AS patients and the group of HLA-B27+ healthy blood
donors was statistically significant (P < 0.05). In 2 of the
28 patients with other HLA-B27-associated spondylarthropathies as well as in 2 of the 22 B27- healthy
controls, only weak PBL proliferation (SI = 2.5) could
be demonstrated. The B27PA-specific response was
dose dependent, reaching a plateau at a final concentration between 10 pg/ml and SO pl/ml in culture medium
(Table 2).
Since it is known that some T cells, in response to
antigen, may be stimulated to secrete cytokines but do
not undergo proliferation, we also determined the fre-
B27 PB
4 4
627 PD
87 PD
827 PA
other Sp.a.
Figure 1. Proliferative responses of peripheral blood lymphocytes
(PBL) against HLAPB27-derived peptides B27PA (WDRETQICKAKAQ), B27PB (LLRGYHQDAY), B27PC (RVAEQLRAYLEGECV); B27PD (ALNEDLSSWTAAUT), and an HLA-B7derived peptide B7PD (ALNEDLRSWTAADT) in a standard
lymphocyte proliferation assay. Results are given as a stimulation
index. The absolute mean cpm of 'H-thymidine incorporated in PBL
from ankylosing spondylitis (A.S.) patients, patients with other spondylarthropathies (other Sp.a.), rheumatoid arthritis (R.A.) patients,
and healthy donors ranged between 125 and 376 cpm for the medium
controls and between 1,702 cpm (B27PC) and 2,945 cpm (B27PA) for
the maximum peptide responses. P represents significant diffcrcnces in
B27PAinduced PBL proliferation between the group of ankylosing
spondylitis patients and the group of HLA-B27+ hcalthy donors.
quency of interferon-y (1FNy)-secreting T cells among
the PBL of 1 responder (ASI) and 1 nonresponder
(AS2) AS patient by using an enzyme-linked immunospot (ELISpot) assay (modified according to ref. 16).
This sensitive ELISpot technique revealed specific IFNy
Table 2. Dose-response curve for peptide B27PA*
Final peptide
0.1 pgiml
1.0 pgiml
5.0 pgiml
7.5 pgiml
10.0 pgiml
20.0 pgiml
50.0 pgiml
165 % 20
218 t 25
565 I 4 7
1,576 % 70
2,492 i 72
2,718 t 76
2,695 t 88
2,733 t 84
165 i- 20
162 I 2 4
172 t 22
181 t 22
168 t 26
170 I 2 0
175 I 2 4
160 I 2 6
*Peripheral blood lymphocytes (PBL; 1 X 10’ in 100 pI of KH10
culture medium) from a responder subject (patient AS1) were incubated with the synthetic peptide B27PA and an irrelevant peptide
(B7PD) at different concentrations for 6 days, pulsed with 0.25 pCi of
tritiated thymidine on day 5, and harvested onto glass filter paper on
day 6. Values are the mean 2 SD cpm of triplicate cultures.
production by B27PA-stimulated T cells in 8 of 10’ PBL
from donor AS1 and in 1 of 10’ PBL from donor AS2
(data not shown).
Furthermore, peptide B27PC induced weak proliferative responses (SI 2.5-5.5) in 10 of the 55 AS
patients, 7 of the 28 patients with other spondylarthropathies, 7 of the 30 B27+ healthy controls, but also in 5 of
the 22 B27- healthy controls (Figure 1). The results of
PBL proliferation in response to peptides B27PB,
B27PD, and B7PD are also demonstrated in Figure 1.
PBL from the RA patients could not be stimulated by
any of the HLA-B27-derived peptides tested.
Interestingly, 8 months after we obtained the
blood sample from donor H1, the 30-year-old male
HLA-B27i healthy blood donor who had demonstrated
the highest proliferative PBL response to peptide
B27PA (SI = 9.8) and, similarly, to peptide B7PD (SI =
8.9) (among the group of HLA-B27-t healthy subjects),
he developed inflammatory-type low back pain and
morning stiffness of 1 hour’s duration. Unfortunately,
donor H1 was not willing to undergo a medical examination by a rheumatologist.
Functional characterization of B27PA-specific T
cell lines. We generated 3 different B27PA-specific T
cell lines by incubating and restimulating with peptide
B27PA PBL from patients AS1 and AS3 and from an
HLA-B27+ healthy blood donor (Hl), as described in
Patients and Methods. AS1 and AS3 were AS patients
whose PBL had proliferated vigorously in response to
peptide B27PA (AS1 SI = 17.5;AS3 SI = 15.4); the PBL
of blood donor H1 were stimulated by B27PA with an SI
of 9.8. The control T cell line B27PA-AS2 was derived
from PBL which did not proliferate in response to
B27PA (SI = 1).The T cell lines were analyzed on day
20 of culture. As demonstrated in Table 3, lines B27PAAS1, B27PA-AS3, and B27PA-H1 specifically proliferated in response to peptide B27PA, whereas the control
line B27PA-AS2 did not. This specific T cell proliferation could not be inhibited by MAb anti-MHC class I
(w6/32) or class I1 (DA6.231). MAb antibody TCRG1
Table 3. Proliferative responses of T cell lines B27PA-AS1, B27PA-AS3, B27PA-H1, and B27PA-AS2*
T cell line
T + medium
T + E- + medium
T + E- + B27PA (10 pgiml)
+ w6132
+ DA6.231
T + E- + B27PA (1 pgiml)
+ w6132
+ DA6.231
+ TCR8l
T + E- + B27PB (10 pgiml)
T + E- 7 B27PC (10 pgiml)
T + E- + B27PD (10 pgiml)
T + E- + B7PD (10 pgiml)
T + E- + PHA (1 pgiml)
72 t 5
89 t 9
1,548 t 34
1,576 t 55
1,513 I 1 0 2
855 t 85
912 t 90
925 I 8 2
896 t 77
411 I 4 2
86 I 1 1
94 % 14
101 t 16
77 t 18
3,689 i 177
3,650 t 158
75 I 8
96 % 11
1,120 I 4 8
1,105 % 41
1,077 t 83
635 I 6 4
725 % 48
730 t 50
695 I 5 5
304 t 29
80 2 9
92 t 11
95 t 10
8 2 % 10
4,248 % 124
4,268 t 135
68 I 8
85 t 10
1,368 2 58
1,379 t 67
1,322 t 117
893 i 82
816 t 55
827 I 6 9
803 t 53
410 I 4 0
79 I 18
90 % 18
94 I 2 1
81 t 16
3,480 t 145
3,538 I 170
86 t 9
114 ? 16
134 t 18
126 I 2 2
110 t 24
119 t 19
125 i 12
110% I 1
4,068 % 170
4,103 t 182
* The T cell lines B27PA-AS1, B27PA-AS3, B27PA-H1, and B27PA-AS2 were generated in vitro by stimulation of peripheral blood lymphocyies
(PBL) from a healthy donor (HI) and donors with ankylosing spondylitis (AS1, AS2, and AS3) with peptide B27PA. On day 20,l X lo4 T cell5 were
stimulated with peptide B27PA and control peptides at 10 pgiml or 1 pgiml, or with phytohemagglutinin (PHA) at 1 pg/ml in the presence of 1 X
lo4 E- (i.e., monocyteiB cell-enriched fractions separated from autologous PBL by sheep erythrocyte rosetting). Blocking experiments were
performed by adding monoclonal antibodies w6132 (anti-HLA class I; at 1:400), DA6.231 (anti-HLA class 11; at 1:400), or TCR81 (anti-$8 T cell
receptor; at 1 pgiml) to the cultures. Values are the mean t SD cpm of triplicate cultures. NT = not tested.
205 1
Table 4. Phenotypic analysis (by FACS) of T cell subpopulations from B27PA-stimulated T cell lines
B27PA-AS1, B27PA-AS2, B27PA-AS3, and B27PA-H1. compared with rIL-2-stimulatcd lines and freshly
isolated PBL from donors ASI, AS2, AS3, and Hl“;
Donor AS1
Freshly isolated PBL
B27PA stimulation day 8
B27PA stimulation day 20
rIL-2 stimulation day 20
Donor AS3
Freshly isolated PBL
B27PA stimulation day 8
B27PA stimulation day 20
rlL-2 stimulation day 20
Donor H1
Freshly isolated PBL
B27PA stirnulation day 8
B27PA stimulation day 20
rIL-2 stimulation day 20
Donor AS2
Freshly isolated PBL
B27PA stimulation day 8
B27PA stimulation day 20
rIL-2 stimulation day 20
TCRaiP (%)
CD4 (5%)
CD8 (%)
TCRyi8 (56)
71 .0
* In the initial proliferation assay (see Figure l), peptide B27PA induced pcripheral blood lymphocyte
(PBL) proliferation in subjects AS1 (stimulation index [SI] 17.5), AS3 (S1 15.4), and H1 (SI Y.8), but not
in subject AS2 (SI I). Values are thc percentages of monoclonal antibodyydefined positive-staining cells
relative to the percentage of total CD3+ T cells. FACS = fluorcscence-activated cell sorter; rIL-2
recombinant interlcukin-2; TCR = T cell receptor.
(anti-y/6 TCR) was able to block the B27PA-specific
proliferative response by more than 50% in the cultures
that were stimulated with 1 pgml of B27PA, whereas it
did not affect PHA-induced proliferation.
Flow cytometric analysis of B27PA-specific T cell
lines. Phenotyping of the B27PA-specific T cell lines
B27PA-AS1, B27PA-AS3, and B27PA-H1 (performed
on day 8 and on day 20 of the culture) revealed a marked
expansion of $6 TCR+ lymphocytes in the B27PAstimulated lines as compared with the unstimulated PBL
(day 0) or the PBL lines stimulated with rIL-2 only
(Table 4). On day 20, $6 TCR+ lymphocytes comprised
25.5% (patient ASl), 30.4% (patient AS3), and 32.7%
(donor H1) of the total CD3+ cell population. In
contrast, the B27PA-stimulated control line B27PA-AS2
contained only 7.8% of $6 TCR+ cells.
These results show that a synthetic peptide from
the HLA-B27 sequence (WDRETQICKAKAQ) can be
recognized as an autoantigen by PBL from HLA-B27+
AS patients and, to a lesser extent, by PBL from
HLA-B27+ healthy controls. Similar results have recently been demonstrated in several patients with auto-
immune uveitis, in whom another peptide from the
sequence of HLA-B27, B27PD, which shares certain
amino acid homologies with an uveitogenic peptide from
retinal soluble antigen, induced proliferative PBL reFponses (12). The same peptide triggered EAU in Lewis
rats (12).
In HLA-B27- control subjects, the HLA-B27derived peptides induced no or only little proliferation,
although these peptides were foreign to the PBL donors.
The lack of proliferation to HLA-B27 determinants by
the control PBL may be explained by the fact that this
kind of reactivity would represent “indirect allorecognition,” which is characterized by host MHC-restricted T
cell alloresponses and is known to be invariably directed
to either a single or a few dominant determinants on the
foreign MHC antigen (17). Thus, the peptides tested
here seem not to represent the particular determinants
on the HLA-B27 molecule which would be potentially
capable of eliciting alloreactive T cell responses in
HLA-B27- donors.
The mechanisms that lead to a break in tolerance
against MHC-derived self peptides are far from being
understood. In the animal model described by Khare et
a1 (1l), the HLA-B27 heavy chain aberrantly accumu-
latcs in the endoplasmic reticulum due to the P2m defect
of these HLA-B27-transgenic mice. This may lead to an
overexpression of otherwise latent self epitopes, which
causes presentation of HLA-B27-derived peptides by
class I1 MHC molecules that become targets of autoreactive CD4+ T cells. Also in the HLA-B27-transgenic
rat model of spondylarthropathy, disease susceptibilityis
dependent on a high number of copies of HLA-B27
genes (18), and transfer of spondylarthropathy to lowcopy B27-transgenic rats is successful only with bone
marrow cells that express high numbers of B27 molecules (9), indicating a role for overexpression or accumulation of HLA-B27 gene products. Although in human disease, there seem to be no differences in the
levels of HLA-B27 expression in PBL from AS patients
as compared with HLA-B27+ healthy control blood
donors (19), differential B27 expression in bacteriainfected (7) and/or inflamed tissues has to be
Second, defective B27 expression during ontogenesis or early childhood could lead to a positive
selection of self-reactive T cells (1 1) or could be due to
mutations in the Pzm molecule (20). The findings that
naturally processed self peptides extracted from human
class I1 molecules comprise predominantly peptides
from MHC class I molecules expressed on the same cell
(21) support this possible form of anti-MHC class I
autoimmunity (22). Recent infections preceding the
manifestation of the spondylarthropathies might be involved in the induction of this process. A cross-reacting
bacterial peptide could awaken an autoimmune response to an HLA-B27 peptide not normally noticed by
CD4+ T cells (22). Alternatively, self-reactive T cells
could be directly primed and optimally costimulated in
response to joint-derived dendritic cells bearing endogenous antigen (23). Both hypotheses of T cell-mediated
pathogenesis in the spondylarthropathies discussed here
are based on the presentation of B27-derived peptides
by class I1 MHC molecules and on autoreactive CD4+ T
lymphocytes as effector cells (22).
Another interesting mechanism by which peptides corresponding to sequences of HLA class I molecules could interact with cellular immune responses has
been discussed by Nossner et a1 (24). That group identified peptides with sequences corresponding to the
relatively conserved region of the a1 helix of the HLA
class I molecules that overlaps the public epitope Bw4/
Bw6, which exhibited inhibitory effects on T lymphocytes. It was shown that only inhibitory peptides bound
to 2 members of the heat-shock protein (HSP) 70 family
in a sequence-specific manner, suggesting that small
molecular weight HLA-derived peptides may modulate
T cell responses by directly interacting with HSPs. The
functional relevance of such peptides was demonstrated
by their beneficial effects on the survival of allogenic
organ transplants in mice and rats (24) and are in
accordance with previous findings on the inhibitory or
immunomodulatory properties of soluble class I molecules in transplantation immunology (25,26). However, a
role for immunomodulating HLA-derived peptides in
physiologic or pathogenic conditions other than alloreactivity can not be excluded. In the case of HLA-B27associated spondylarthropathy, we can only speculate
whether HLA-B27-derived peptides stimulate or downregulate disease-relevant T lymphocytes.
The HLA-B27-derived 13-mer peptide B27PA,
identified in this study to induce T cell proliferation in
31% of the AS patients and in 10% of the HLA-B27+
healthy controls tested, shows anchor positions at AS
residues 1,4,6,7,and 9, which have been predicted to be
allele-specific motifs for binding to class I1 MHC alleles
DRB*0401, "0404, and "0405 (27). Therefore, surprisingly, our data from FACScan analysis and from MAbblocking experiments of B27PA-specific T cell lines
indicated that not alp TCR+ CD4+, but rather, ylS
TCR+ T cells, were the major population stimulated
and expanded after in vitro incubation with the HLAB27-derived peptide. In this context, it is worth mentioning that for several human autoimmune disorders
including RA (28), self-reactive y/S TCR+ cells have
been suspected to contribute to the pathogenesis of
disease, although the relative number of ylS TCR+ cells
is not increased in the peripheral blood or in the synovial
fluids and membranes of RA patients (29). In the
spondylarthropathies, synovial fluid ylS TCR+ cells that
proliferated in response to Klebsiella pneumoniae (30) or
showed cytotoxicity against either self targets, Daudi cell
lines, or enterobacteria-infected B cell lines (31) have
been isolated by our group. With regard to the pathogenesis of the spondylarthropathies, we therefore speculated that ylS TCR+ cells might play a role in the early
defense mechanisms directed against cells infected by
arthritogenic bacteria or against autologous inflamed
"stressed" tissues (31).
Another important observation has consequences for the role of HLA-B27 peptidc-specific y/6
TCR+ cells in the animal model of EAU. Adoptively
transferred yiS TCR+ cells from rats orally tolerized
with the uveitogenic autoantigenic retinal soluble antigen peptide or the cross-reactive peptide B27PD could
mediate the suppression of uveitis in an antigen-specific
manner (32). Given these findings, it is conceivable that
also in the human disease of AS, y/S TCR+ cells with
specificity for a peptide from the disease-associated
HLA-B antigen B27 are involved in the immunomodulation or even down-regulation of inflammation. The
observation in this study of peripheral T cell reactivity
with peptide B27PA in HLA-B27+ healthy blood donors is not inconsistent with the latter possibility. We do
not know, however, whether there is cross-reactivity
between the B27PA peptide and autoantigens inside the
joint or any bacterial antigens on the level of $8 TCR+
Since the peptides used in this study do not cover
the entire HLA-B27 heavy chain and were not selected
for their putative HLA class I1 binding motifs, we may
have missed additional important epitopes that are
presented by HLA-B27 itself or that have the ability to
stimulate CD4+ T cells in a HLA class IT-restricted
manner. We therefore plan to investigate more peptides
from the HLA-B27 sequence which encompass
HLA-DR binding motifs, namely, so-called “promiscuous” HLA-DR motifs. In addition, such peptides will be
tested for their ability to induce secretion of either
Thl-type (IFNy, IL-2) or Th2-type (IL-4, IL-10) cytokines by responsive T cells, and thus modulate the
immune response.
The findings described here may have relevance
for further investigations in animal models and for
treatment studies such as oral tolerization trials in
patients with AS.
The authors wish to thank Ms Ilona Kumb for excellent
technical assistance and the members of the German Spondylitis Association (Deutsche Vereinigung Morbus Bechtercw
e.V.) for their support.
1. Brewerton DA, Nicholls A, Oates JK, Caffery M, Walters D,
James DCO: Reiter’s disease and HLA-A27. Lancet ii:996-998,
2. Schwiinmbeck P, Yu DTY, Oldstonc MBA: Autoantibodies to
HLA-B27 in the sera of HLA-B27 patients with ankylosing
spondylitis and Reiter’s syndrome: molecular mimicry with Klebsiclla pneumoniae as a potential mechanism of autoimmune
disease. J Exp Med 166:173-181, 1987
3. Chcn J-H, Kono DH, Yong Z , Park MS, Oldstonc MBA, Yu DTY:
A Yersinia pseudotuberculosis protein which cross-reacts with
HLA-B27. J Immunol 139:3003-3011, 1987
4. Benjamin R, Parham P: Guilt by association: HLA-B27 and
ankylosing spondylitis. Immunol Today 11:137-142, 1990
5. Hermann E, Yu DTY, Meyer zum Biischenfelde K-H, Fleischer B:
HLA-B27-restricted CD8 T cclls derived from synovial fluids of
patients with reactive arthritis and ankylosing spondylitis. Lancet
3421646-650, 1993
6. Scofield RH, Kurien B, Gross T, Warren WL, Harley JB: HLAB27 binding of peptide from its own sequence and similar peptides
from bacteria: implications for spondyloarthropathies. Lancet
345:1542-1544, 1995
7. Kapasi K, Inman RD: HLA-B27 exprcssion modulates gramnegative bacterial invasion into transfected L cells. J Immunol
148:3554-3560, 1992
8. Hammer RE, Maika SD, Richardson JA, Tang J-P, Taurog JD:
Spontaneous inflammatory disease in transgenic rats expressing
HLA-B27 and human &m: an animal model of HLA-B27associated human disorders. Cell 63:1099-1112, 1990
9. Breban M, Hammer RE, Richardson JA, Taurog JD: Transfer of
the inflammatory disease of HLA-B27 transgenic rats by bone
marrow engraftment. J Exp Med 178:1607-1616, 1993
10. Taurog JD, Richardson JA, Croft JT, Simmons WA, Zhou M,
Fernandez-Sueiro JL, Balish E, Hammer RE: The germfree state
prevents devclopment of gut and joint inflammatory disease in
HLA-B27 transgenic rats. J Exp Med 180:2359-2364, 1994
11. Kharc SD, Luthra HS, David CS: Spontaneous inflammatory
arthritis in HLA-B27 transgcnic mice lacking P,-microglobulin: a
model of human spondylarthropathies. J Exp Med 182:1153-1158,
12. Wildner G. Thurau SR: Cross-reactivity between an HLA-B27derived peptide and a retinal autoantigen peptide: a clue to major
histocompatibility complex association with autoimmune disease.
Eur J Immunol 24:2579-2585, 1994
13. Van der Linden S, Valkenburg HA, Cats A: Evaluation of
diagnostic criteria for ankylosing spondylitis: a proposal for modification of the New York critcria. Arthritis Rheum 27:361-368,
14. Chou PY, Fasman GD: Prediction of protein conformation.
Biocherniatry 13:222-245, 1974
15. Hermann E, Fleischcr B, Mayet W-J, Poralla T, Meyer zum
Biischcnfelde K-H: Response of synovial fluid T cell clones to
Yersinia enterocolitica antigens in patients with reactive Ycrsinia
arthritis. Clin Exp Immunol 75365-370, 1989
16. Klinman DM, Nutman TB: ELISpot assay to detect cytokinesecreting murine and human cells. In, Current Protocols in Immunology. Seventh edition. Edited by JE Coligan, AM Kruisbeek,
DH Margulies, EM Shevach, W Strober. Brooklyn, NY, Greene
Publishing Associates, 1994
17. Benichou G, Tam RC, Loares LRB, Fedoseyeva EV: Indirect
T-cell allorecognition: perspectives for peptide-based therapy in
transplantation. Immunol Toddy 18:67-71, 1997
18. Taurog JD, Maika SD, Simmons WA, Breban M, Hammer RE:
Susceptibility to inflammatory disease in HLA-B27 transgenic rat
lines correlates with the levcl of B27 expression. J Immunol
150:4168-4178, 1993
19. Creamer P, Edmonds J , Sullivan J, Matthews S: Measurement of
HLA class I expression in ankylosing spondylitis. Ann Rheum Dis
51:1138-1142, 1992
20. Fukazawa T, Hermann E, Edidin M, Wcn J, Huang F, Kcllner H,
Floege J, Farahmandian D, Williams MK, Yu DTY: ‘The effect of
mutant P,-microglobulins on the conformation of HLA-B27 dctected by antibody and by cytotoxic T lymphocytes. J Immunol
153:3543-3550, 1994
21. Chicz RM, Urban RG, Gorga JC, Vignali DAA, Lane WS,
Strominger JL: Specificity and promiscuity among naturally processed peptides bound to HLA-DR alleles. J Exp Med 178:27-47,
22. Sieper J, Braun J: Pathogenesis of spondylarthropathies: persistent
bacterial antigen, autoimmunity, or both? Arthritis Rheum 38:
1547-1554, 1995
23. Thomas R, Lipsky PE: Could endogenous self-pcptides prcsentcd
by dendritic cells initiate rheumatoid arthritis? Immunol Today
17559-564, 1996
Nossner E, Goldberg JE, Naftzger C, Lyu SC, Clayberger C,
Krensky AM: HLA-derived peptides which inhibit T cell function
bind to members of the heat-shock protein 70 family. J Exp Med
183:339-348, 1996
Zavazava N, Hausmann R, Muller-Ruchholtz W: Inhibition of
anti-HLA-B7 alloreactive CTL by affinity-purified soluble HLA.
Transplantation 512338-842, 1991
Buelow R, Burlingham WJ, Clayberger C: Immunomodulation by
soluble HLA class I. Transplantation 59549-654, 1995
Rammensee H-G, Friede T, Stevanovic S: MHC ligands and
peptide motifs: first listing. Immunogenetics 41:178-228, 1995
Soderstrom J, Halapi E, Nilsson E, Gronberg A, van Embden J,
Klareskog L, Kiessling R: Synovial cclls responding to a 65-kDa
mycobacterial heat shock protein have a high proportion of a
TcRyG subtype uncommon in peripheral blood. Scand J Immunol
32:503-515, 1990
29. Smith M. Broker B. Moretta L. Ciccone B. Grossi CE, Edwards
JCW, Yiiksel F, Colaco B, Worman C, Mackenzie L, Kinne K,
Weseloh G, Gluckert K, Lydyard PM: Ty6 cells and their subsets
in blood and synovial tissue from rheumatoid arthritis patients.
Scand J Immunol 32585-593, 1990
30. Hermann E, Suck6 B, Drostc U, Meyer zum Buschenfelde K-H:
Klebsiella pneumoniae-reactive T cells in blood and synovial fluid
of patients with ankylosing spondylitis: comparison with HLAB27+ healthy control subjects in a limiting dilution study and
determination of the specificity of synovial fluid T cell clones.
Arthritis Rheum 38:1277-1282, 1995
31. Hermann E, Ackermann B, Duchmann R, Meyer zum Buschenfelde K-H: Synovial fluid MHC-unrestricted y6 T lymphocytes
contribute to antibacterial and anti-self cytotoxicity in the spondylarthropathics. Clin Exp Rheumatol 13:187-191, 1995
32. Wildner G, Hiinig T, Thurau S: Orally induced, peptide-specific
y/S TCR+ cells suppress experimcntal autoimmune uveitis. Eur
J Immunol 26:2140-2148, 1996
Без категории
Размер файла
769 Кб
hla, autoantigen, ankylosis, b27, derived, spondylitis, lymphocytes, peptide
Пожаловаться на содержимое документа