close

Вход

Забыли?

вход по аккаунту

?

Specificity of antibodies to type II collagen in rheumatoid arthritis.

код для вставкиСкачать
1493
SPECIFICITY OF ANTIBODIES TO TYPE I1
COLLAGEN IN RHEUMATOID ARTHRITIS
KUNIAKI TERATO, YASUNORI SHIMOZURU, KOU KATAYAMA, YOSHIHARU TAKEMITSU,
IZUMI YAMASHITA, MAKOTO MIYATSU, KATSUYUKI FUJII, MITSUSADA SAGARA,
SHIEGETO KOBAYASHI, MAKOTO GOTO, KUSUKI NISHIOKA,
NOBUYUKI MIYASAKA, and YUTAKA NAGAI
To reassess the role of autoantibodies to type I1
collagen in the pathogenesis of diseases, we studied
antibodies from patients with rheumatoid arthritis (RA)
and from patients with relapsing polychondritis for
species specificity and collagen type specificity, using an
improved enzyme-linked immunosorbent assay. Antibodies were found in the sera of 15% of the RA patients
From the Eisai Tsukuba Research Laboratories, Ibaraki;
Asahikawa Medical College, Asahikawa; Jikei University School of
Medicine, Tokyo; Kanazawa University School of Medicine, Kanazawa; Juntendo University Medical School, Tokyo; Institute of
Rheumatology, Tokyo Women’s Medical College: and Medical
Research Institute. Tokyo Medical and Dental University, Tokyo.
Japan.
Supported by a grant from the Japan Health Sciences
Foundation.
Kuniaki Terato, PhD: Senior Investigator, Eisai Tsukuba
Research Laboratories (current address: Department of Medicine,
University of Tennessee and the VA Medical Center. Memphis,
Tennessee); Yasunori Shimozuru, BS: Investigator. Eisai Tsukuba
Research Laboratories; Kou Katayama, MD: Department of Orthopaedic Surgery, Asahikawa Medical College; Yoshiharu Takemitsu.
MD: Professor, Department of Orthopaedic Surgery, Asahikawa
Medical College; Izumi Yamashita, MD: Department of Orthopaedic Surgery, Asahikawa Medical College; Makoto Miyatsu, MD:
Department of Orthopaedic Surgery, Asahikawa Medical College:
Katsuyuki Fujii, MD: Associate Professor, Department of Orthopaedic Surgery, Jikei University School of Medicine; Mitsusada
Sagara, MD: Department of Orthopaedic Surgery, Kanazawa University School of Medicine; Shiegeto Kobayashi. MD: Department
of Internal Medicine, Juntendo University Medical School; Makoto
Goto, MD: Institute of Rheumatology. Tokyo Women’s Medical
College; Kusuki Nishioka, MD: Professor of Medicine. Institute of
Rheumatology, Tokyo Women’s Medical College; Nobuyuki Miyasaka. MD: Professor, Department of Immunology, Medical Research Institute, Tokyo Medical and Dental University; Yutaka
Nagai, PhD: Professor, Department of Tissue Physiology, Medical
Research Institute, Tokyo Medical and Dental University.
Address reprint requests to Kuniaki Terato, PhD, Veterans
1030 Jefferson Avenue, MemAdministration Medical Center (M),
phis, T N 38104.
Submitted for publication August 29, 1989; accepted in
revised form April 30, 1990.
Arthritis and Rheumatism, Vol. 33, No. 10 (October 1990)
and 50% of the relapsing polychondritis patients, as well
as in the cartilage of 69% of the RA patients examined.
Reaction with both homologous and heterologous type I1
collagens was common. Analysis of 19 selected RA sera
revealed that autoantibodies were generally associated
with specific antibodies to some species of heterologous
type I1 collagen. In contrast, antibodies found in 4% of
the non-RA controls were specific for either bovine or
chick type I1 collagen. These findings indicate that
autoantibody formation in RA and relapsing polychondritis may occur as a result of an immune response to
heterologous type I1 collagen. However, since RA and
relapsing polychondritis patient sera differed in their
reactivity with the cyanogen bromide-digested peptides,
it is possible that the clinical manifestation of collagen
autoimmunity might be influenced by the epitope specificity of the antibodies.
The hypothesis that autoimmunity to type I1
collagen is instrumental in the pathogenesis of rheumatoid arthritis (RA) is supported by 3 lines of investigation. First, anti-type I1 collagen antibodies have
been demonstrated in the serum (1-3), synovial fluid
(43). and cartilage (6) of RA patients. Characterization of these autoantibodies has shown that they
consist primarily of complement-fixing IgG isotypes
and are capable of binding to homologous cartilage and
of converting C5 to C5a (7). Second, susceptible
strains of rodents (8-12) and nonhuman primates (1315) immunized with type I1 collagen produce high
titers of autoreactive antibodies and develop an erosive polyarthritis that shares phenotypic features with
RA. Third, the pathogenic potential of purified human
anti-type I1 collagen antibodies has been cogently
TERATO ET AL
1494
demonstrated by their ability to passively transfer
arthritis into naive mice (16).
Conversely, antibodies to different types of
collagen have been associated with RA (17-25), and
antibodies to type I1 collagen have been identified in a
variety of diseases (26-31). Because of this lack of
specificity, the pathogenic significance of collagen
antibodies has been questioned. We have modified
the standard enzyme-linked immunosorbent assay
(ELISA) system to reduce nonspecific background
interference (32) and have used this new system to
determine the prevalence of anticollagen antibodies in
the sera of patients with RA, patients with relapsing
polychondritis, and in non-RA controls and normal
controls. Using this method, we found anti-type I1
collagen antibodies in 15% of RA sera, 50% of RP sera,
and 4% of non-RA control sera. Analysis of antibodies
in these sera showed differences in the specificity of
the collagen antibodies in RA patient sera and relapsing polychondritis patient sera compared with that in
sera from the controls. Further analysis also suggested
that there is a difference in the fine specificity of
collagen autoantibodies in relapsing polychondritis
patients compared with RA patients.
PATIENTS AND METHODS
Patients and controls. Serum samples were obtained
from 202 patients with RA and 26 patients with relapsing
polychondritis. Sera were also obtained from 19 patients
with osteoarthritis (OA), 54 patients with gout, and 19
patients with other, nonrheumatic diseases. Two hundred
age- and sex-matched normal control subjects also provided
serum samples. Sera were stored at -20°C until used.
The RA patients studied were drawn from among
those attending the outpatient clinic of Kanazawa University
School of Medicine and the Institute of Rheumatology,
Tokyo Women’s Medical College. The diagnosis and anatomic staging of RA were established according to the
American Rheumatism Association criteria (33). The RA
patients consisted of 35 men (mean a SD age 52.7 2 13.8)
and 167 women (mean 2 SD age 53.2 f 11.6). Seventeen
patients had possible RA, 10 had probable RA, 83 had
definite RA, 85 had classic RA, and 7 had RA of unknown
type. Ten were classified as having stage 0 disease, 41 as
stage I, 74 as stage II,46 as stage III,21 as stage IV, and 10
as stage unknown.
The normal control subjects consisted of 95 men
(mean ? SD age 41.3 2 12.8) and 105 women (mean 2 SD
age 35.6 t 11.3) who were citizens of Tokyo and received
regular medical care at Tokyo Women’s Medical School.
Sera from patients with relapsing polychondritis
were provided by the Department of Internal Medicine,
Juntendo University Medical School. In addition to these
sera, 13 RA sera with high titers of antibody to type I1
collagen (human, bovine, or chick) were obtained from the
Department of Orthopaedic Surgery, Jikei University School
of Medicine, and were included in this study.
Articular cartilage was obtained from 13 patients
with RA and 5 patients with OA who had undergone arthroplasty of knee joints at the Department of Orthopaedic
Surgery, Asahikawa Medical College. Cartilage samples
were stored at -20°C until used.
Preparation of collagens and peptides. Type I1 collagen from human (HII), monkey (MkII), bovine (BII), rat
(RII), mouse (MsII), and chick (CII) sources was prepared
from sternal or articular cartilage as described elsewhere
(1 1,12,15). Type 1 collagen from human (HI) and bovine (BI)
was purchased from the Collagen Research Center (Cosmo
Bio, Tokyo, Japan). Cyanogen bromide (CB) peptides from
HI1 were purified by a combination of ion-exchange and
molecular-sieve chromatography and renatured into their
original triple-helical conformation by stepwise cooling, as
described previously (34). The purity of the collagens and
CB peptides was determined by electrophoresis using 6%
and 15% sodium dodecyl sulfate-polyacrylamide gels, respectively. The renaturation of the CB peptides was not
assessed because the quantity was insufficient, the antibodies used to test the reactivity were purified by a n i t y
chromatography with native type I1 collagen, and the reactivity of polyclonal antibodies with each renatured and
nonrenatured CB peptide was generally parallel, as has been
shown in a previous paper (12).
Antibody assay and inhibition studies. The detection
of antibodies to collagen was performed by ELISA using
buffered normal rabbit serum (NRS) as a blocking agent to
diminish nonspecific binding of human IgG to the plastic
plates, as described earlier (32). The ELISA plates (Serocluster U plates; Costar, Cambridge, MA) were coated with
100 pl of collagen (5 pglml) or CB peptide (10 CLg/ml)
dissolved in phosphate buffer, pH 7.5. Serum samples were
diluted 1 :50 in buffered NRS and added to the plates. The
second antibody consisted of peroxidase-conjugated rabbit
anti-human IgG (Fc specific; Cooper Biomedical, Cochranville, PA) diluted 1:2,000 with 25% NRS. Inhibition studies
were performed by diluting sera with buffered NRS containing each antigen at a concentration of 50 pglml.
The reactivity of serum with type I and I1 collagens
was more than 80% inhibited by the corresponding antigen
(results not shown). Antibody levels were calculated by
reference to the optical density values of a standard serum
from an arthritic cynomolgus monkey that had been immunized with CII (15,32). Results are given in arbitrary units, as
defined by the reference serum diluted 1 :128,000.
Elution of antibodies from articular cartilage. Antibodies were eluted from articular cartilage with 0.1M Tris
HCI buffer, pH 7.5, containing 0.15M NaCI, 0.1% bovine
serum albumin (BSA), and 3M potassium thiocyanate
(KSCN-Tris-BSA).Cartilage specimens were sliced into thin
pieces with a razor blade and washed with the same buffer
without KSCN. The antibodies were then eluted with 2
volumes (weightlvolume) of KSCN-Tris-BSA by shaking for
60 minutes at room temperature. The eluates were filtered
through a 0.22 pm membrane to remove cartilage debris and
then dialyzed at 4°C in 0.1M Tris HCl buffer, pH 7.5,
containing O.15M NaCl and 0.05% azide (Tris-azide). The
eluents were diluted 1:2-1: 10 with buffered NRS for ELISA
analysis. The validity of the elution method was verified by
1495
ANTI-TYPE I1 COLLAGEN SPECIFICITY
incubating cartilage samples with serum containing known
quantities of antibodies to type I1 collagen and then quantitating the amount of antibody bound. The cartilage was then
processed as described above, and the amount of antibody
that eluted was quantified. Under these conditions, 92.8 ?
9.1% (mean 2 SD) of antibodies bound to cartilage were
recovered.
Purification of anti-HI1 antibodies. Antibodies to HI1
were purified from sera by affinity chromatography using
HII-coupled Sepharose as described previously (15). Fifty
microliters of each serum sample was loaded on a column
containing 1 ml of HII-Sepharose and allowed to stand for 60
minutes at room temperature. Unbound antibodies were
washed out with 2.5 ml of buffered NRS. After washing with
20 ml of Tris-BSA, anti-HI1 antibodies were eluted from the
column with 2.5 ml of KSCN-Tris-BSA at a flow rate of 2.5
d / 1 0 minutes at room temperature. The eluent was immediately dialyzed against Tris-aide and stored at 4°C.
Table 2. Species specificity of anti-type I1 collagen antibodies in
non-rheumatoid arthritis control subjects*
Serum
Normal
40127
40144
90237
90247
OA patient
2073
HI1
MklI
BII
RII
MsII
CII
-
-
0.207 (100)
1.301 (96)
1.744 (89)
0.484 (83)
-
-
0.311
-
-
-
0.555 (100)
-
-
-
* Serum samples were diluted 150 with buffered normal rabbit
serum (NRS) and assayed in an enzyme-linked immunosorbent
assay on plates coated with type I1 collagen ( 5 pg/ml) from human,
monkey, bovine, rat, mouse, and chick sources (HII, MkII, BII,
RII, Msll, CII, respectively). For the inhibition study, serum was
diluted with buffered NRS containing 50 pg/ml of antigen. Data are
expressed in optical density (OD) units (at 492 nm). - = values
<0.100 OD units. Numbers in parentheses are the % inhibition. OA
= osteoarthritis.
RESULTS
Using our improved ELISA, 3 1 of 202 sera from
patients with RA reacted with some species of type I1
collagen. Antibodies against homologous type 11 collagen were identified in 23 (1 1%) of these 202 sera from
patients with RA; an additional 8 patients did not have
antibodies against HII, but had antibodies reactive
with either CII or BII. Similarly, HI1 antibodies were
found in 11 (42%) of the 26 sera from the patients with
relapsing polychondritis, but in only 1 (0.3%) of the
292 sera from the entire group of control subjects (a
patient with OA) (Table 1).
Antibodies against type I collagen were also
observed in 6 (3%) of the 202 RA sera and in 1 I (3.8%)
Table 1. Incidence of antibodies to homologous and heterologous
type I1 collagen in sera from patients with rheumatic diseases and
from controls*
Group (n)
RA patients (202)
Relapsing polychondritis
patients (26)
OA patients (19)
Gout patients (54)
Other, nonrheumatic
disease patients (19)
Normal controls (200)
Total no. of
sera positive
for type I1
collagen
No. of sera
positive for
HI1
BII
CII
31
13
23
11
21
ND
17
13
3
4
0
1
0
0
I
0
0
4
0
5
0
4
1
1
* Serum samples were diluted 150 with buffered normal rabbit
serum and assayed for reactivity with human, bovine, and chick
type I1 collagen (HII, BII, and CII, respectively) in an enzymelinked immunosorbent assay. Reactivity was confirmed by inhibition studies, and those inhibited >80% by the corresponding antigen
were listed. RA = rheumatoid arthritis; ND = not done; OA =
osteoarthritis.
of the 292 non-RA control sera (1 OA and 10 normal);
however, the antibodies were specific to BI, and
cross-reactivity with HI was not observed in either
group. Interestingly, antibody to BI was associated
with antibody to type I1 collagen in 3 patients with RA
and in 5 non-RA control subjects (1 OA and 4 normal)
(data not shown).
To ascertain the species specificity of the antibodies, we tested comparatively high-titer sera from
17 RA patients (including the 13 sera obtained from
Jikei University), 7 relapsing polychondritis patients,
and 5 controls for reactivity with type I1 collagen from
a variety of sources. Although 4 normal control sera
and 1 sample from a patient with OA showed strong
reactivity with B11, none of the serum samples tested
showed cross-reactivity with other species of type I1
collagens except for sample number 90247, which
demonstrated modest cross-reactivity with MsII (Table 2). In contrast, extensive cross-reactivity with type
11 collagen from most sources was found in the RA and
relapsing polychondritis patient sera tested.
To characterize the autoantibodies, we purified
them by affinity chromatography on an HII-Sepharose
column and studied their specificity. Of the original 17
RA sera, only 12 had antibody titers high enough for
further characterization, and only 3 of the original
relapsing polychondritis samples were available. In 8
of 12 RA sera and 3 of 3 relapsing polychondritis sera,
antibodies that did not bind to HI1 were recovered in
the effluent. These antibodies reacted little with HII,
but reacted strongly with some of the heterologous
type I1 collagens (Table 31, as did the antibodies
observed in the non-RA control sera. In the remaining
4 RA sera, antibodies were not detected in the effluent.
1496
TERATO ET AL
Table 3. Species-specificand speciescross-reactive antibodies to type
I1 collagen in sera from patients with RA or relapsing polychondritis*
Serum
RA patient
5868
Bound
Unbound
~89i
Bound
Unbound
3976
Bound
Unbound
3 I427
Bound
Unbound
51429
Bound
Unbound
5 1448
Bound
Unbound
31483
Bound
Unbound
KB30
Bound
Unbound
5975
Bound
Unbound
J995
Bound
Unbound
KB88
Bound
Unbound
KB260
Bound
Unbound
Relapsing polychondritis
patient
HI1
MkII
BII
RII
MsII
CII
1.193
0.927
-
0.939
2.316
0.740
-
0.119
0.199
0.732
0.260
1.291
0.122
1.314
0.177
1.019
0.336
0.999
-
0.331
-
0.751
0.159
1.683
0.112
1.598
0.184
1.815
0.210
1.316
-
0.216
0.171
3.342
0.128
1.614
1.512
0.107
1.102
2.829
1.317
0.103
0.497
0.451
1.260
-
-
1.056
0.102
0.281
-
1.940
-
1.864
-
1.792
-
1.886
0.498
0.853
1.914
3.195
3.360
1.987
0.128
2.2796
-
1.000
0.266
1.914
'
-
-
-
-
-
-
-
0.319
0.122
0.227
-
0.206
0.212
0.165
0.110
-
-
0.165
0.130
0.848
0.739
-
0.440
-
0.300
-
-
0.210
1.239
-
1.277
-
1.027
-
1.091
-
0.667
0.640
0.294
0.266
-
0.278
-
0.155
-
-
0.139
1.244
1.038
-
0.803
-
1.045
0.332
0.783
-
-
-
-
-
-
-
-
In contrast, antibodies eluted with 3M KSCN reacted
efficiently with both homologous and heterologous
type I1 collagens (Table 3), indicating that the autoantibodies against type I1 collagen recognize common
antigenic determinants shared on type I1 collagen of
various species. Low reactivity with MsII indicates
that the number of common antigenic determinants
expressed on MsII are fewer than on type I1 collagen
from other species.
Assays of knee cartilage samples from 5 OA
patients showed no detectable antibodies to human
type I1 collagen (data not shown). However, 10 of the
14 cartilage specimens from the 13 RA patients
showed antibodies to HI1 (Table 4). Antibodies were
detectable in serum samples from only 5 of these 9 RA
patients. These findings could be due to selective
accumulation of antibodies in the knee joint cartilage,
whereas their presence in serum is in such low
amounts as to be undetectable, or it could be due to
local production of autoantibodies by immune cells
within the synovial membrane.
To localize antigenic determinants on the HI1
molecule, we used affinity-purified antibodies from 19
sera to determine their reactivity with CB peptide in
the ELISA plates. As shown in Table 5, most purified
antibodies reacted with CB-11, which was present in
14 RA sera (74%); CB-8 was present in 13 (68%),
CB-10 was present in 7 (37%), and CB-9-7 and CB-12
were present in 3 (16%) and 1 (5%), respectively.
These results suggest that the major antigenic determinants on homologous type 11 collagen are located
-
Table 4. Accumulation of anti-type I1 collagen antibodies in the
cartilage of rheumatoid arthritis patients*
JU3
Bound
Unbound
5114
Bound
Unbound
Ju7
Bound
Unbound
0.153
ND
0.223
ND
0.182
ND
0.216
ND
0.157
ND
1.582
1.449
0.118
1.159
1.206
0.701
1.382
0.448
0.413
0.223
0.908
0.115
1.977
0.198
1.848
0.475
1.645
1.657
0.201
0.737
0.139
1.062
0.234
-
-
-
* Serum samples (50 4) were loaded on a human type 11 collagen
(HII)-Sepharose column (1 ml), and nonbound antibodies were washed
out with 2.5 ml of buffered normal rabbit serum. Bound antibodies
were then eluted with 2.5 ml of 3M potassium thiocyanate. The effluent
and eluent were assayed for reactivity with different species of type 11
collagen in an enzyme-linked immunosorbent assay. Effluent from 8 of
the 12 rheumatoid arthritis (RA) patient sera and from the 3 relapsing
polychondritis patient sera contained specific antibodies to some
species of type I1 collagen; no antibodies were detected in the effluent
from the other 4 RA sera. Data are expressed in optical density (OD)
units (at 492 nm). Values represent antibodies bound to HI1 or
antibodies not bound to HII. - = values <0.100 OD units; ND = not
done. See Table 2 for other definitions.
Antibody level
Patient
Functional Disease
ESR
Cartilage Serum
class,
duration
(md
(units/ (units/
stage
(years) RF hour) CRP
gm)
ml)
Al
111, IV
A5
A6
A7
A8
A9
Ill, Ill
Ill, 111
IV. Ill
Ill. 111
111, 1V
14
7
10
13
21
10
I+
I+
I+
2+
I+
2+
100
115
A10
All
A13
Ill, IV
IV, 1V
Ill. 11
25
10
7
I+
2t
82
122
70
25
51
60
102
4t
5+
2t
*
3+
2+
2+
1+
38
2,500
7
300
3
0
348
13,500
4
0
321 (R)
900
328 (L)
410
3,250
0
18
9
0
* Cartilage extracts and sera were assayed for antibodies to human
type 11 collagen by enzyme-linked immunosorbent assay. Rheumatoid factor (RF) and C-reactive protein (CRP) values were scored on
a scale of - to 2+ and - to 5 + , respectively. Cartilage was obtained
from the knee (values are per gram wet weight). ESR = erythrocyte
sedimentation rate (Westergren); R = right; L = left.
1497
ANTI-TYPE I1 COLLAGEN SPECIFICITY
Table 5. Localization of antigenic determinants on type 11 collagen recognized by autoantibodies in rheumatoid arthritis and relapsing polychondritis patient sera*
Patient
HII
CB-12
Rheumatoid
arthritis
Al
5713
A9
A7
J975
51427
KB30
5868
5891
5733
51448
TJ6
A10
CB-IO CB-9-7
0.255
0.101
(0.807)
0.732 0.348
(2.141) (1.493)
0.132
0.283
0.670 0.551
0.202 0.176
0.628
0.303
0.100
0.152
0.510
0.374
0.151
0.088 0.101
0.203 0.530 0.258
0.203
(2.563) (2.075)
1.149
0.833
1.143
0.242
0.258
0.677
0.360 0.299
0.619
0.274 0.218
(0.183)
-
0.975
( I . 179)
0.339
0.183
(2.670)
1. I27
(5.566)
1.563
2.248
0.495
1.723
2.406
0.700
1.517
1.192
0.415
(3.521)
2.357
1.915
1.175
3.668
2.589
1.095
(0.531)
51483
KB260
KR147
5976
J995
TJ32
A5t
Relapsing
polGhondritis
0.781
Ju I
1.671
Ju4
TJ7
1.801
~ ~ - 1 (3-8
1
0.411
0.178
-
-
-
0.284
-
0.193
0.647
1.311
* Antibodies from the sera of patients with rheumatoid arthritis
(RA) or relapsing polychondritis were purified by affinity chromatography on human type 11 collagen (HI1)-Sepharose or were
extracted from RA cartilage (in parentheses) and tested for reactivity
with renatured cyanogen bromide (CB) peptide fragments of HI1 in an
enzyme-linked immunosorbent assay, at 4°C. Data are expressed in
optical density (OD) units (at 492 nm). - = values <Q.100 OD units.
t Serum antibody titer was 0.174 and was therefore not considered
reactive with CB peptide.
within the central helical portion of the type I1 collagen
molecule. In addition, antibodies extracted from cartilage (Al, AS, A7, A9, and AlO) gave CB peptide
recognition patterns comparable with those of the
serum antibodies (Table 5). In some instances, reactivity to a specific CB peptide was detected in cartilage
eluates, but not in the corresponding sera, e.g., A1 and
A10. These disparate results reflect the lower concentration of specific antibodies in the serum samples, as
indicated by the OD values for HI1 in Table 5 . These
results indicate that the autoantibodies in RA sera bind
specifically to the cartilage. Thus, it was not possible
to identify a specific RA-reactive epitope. However,
purified antibodies from 3 patients with relapsing polychondritis all reacted with CB-9-7 (Table 5).
DISCUSSION
A number of investigators, using a variety of
methods, such as passive hemagglutination (1,4,1720), radioimmunoassay (5,6,21-24), immunofluorescence (2,3), and ELISA (7,25), have demonstrated that
antibodies that react with various types and sources of
collagen are present in serum and synovial fluid of RA
patients. Despite these findings, there is controversy
regarding the incidence and specificity of anticollagen
antibodies found in RA patients. These discrepancies
may be attributed to the differences in the sensitivities
of the assays and to nonspecific reactions which
complicate many assays, as we (32) and Beard and
coworkers (3,35) have reported. For this reason, we
recently developed an improved ELISA using buffered
normal rabbit serum as a blocking agent to diminish
the nonspecific binding of human IgG to the plastic
microtiter plates (32). Based on this method, we
demonstrated that specific antibodies to type I1 collagen, which have no cross-reactivity with type I collagen, were present in 15% of the serum samples from
RA patients.
Of particular interest are the anticollagen antibodies found in the articular cartilage of RA patients,
which are most likely to contribute to the pathogenesis
of joint injury. We eluted such antibodies from the
cartilage samples of 69% (9 of 13) of RA patients
studied. The detection of cartilage-bound antibodies
was possible even in many patients who did not have
detectable antibody in their sera. These results confirm and extend the observations made by Jasin (6),
who showed that antibodies against native and denatured type I1 collagen are sequestered in 60% of RA
and OA cartilage samples. He observed that most of
these antibodies reacted with heterologous type I1
collagen (CII and BII) and some with type I collagen
(BI). Our data differ, in that we did not observe any
cartilage-bound antibodies which reacted with type I
collagen, and all of our OA cartilage samples were
negative for anti-type I1 collagen antibodies. These
discrepancies may be due to differences in the specificity and sensitivity of assay systems used, differences in the antibody isolation techniques (collagenase
digestion versus KSCN extraction), and differences in
the patient populations.
We believe that the differences observed in the
specificity of anti-type I1 collagen antibodies found in
the RA patients, as compared with their specificity in
the non-RA controls, are the important determinants
of disease. Most of the autoantibodies in RA sera
cross-reacted with all of the heterologous type I1
collagens tested. It would appear that most RA pa-
1498
tients have antibodies similar to those that are generated in animals immunized with heterologous type I1
collagen. Antibodies produced in arthritis-susceptible
animals react not only with the collagen used for
immunization, but also with common antigenic determinants found on homologous type I1 collagens (1012,15,36,37).
However, another possibility derivable from
our data is that some of the RA patients were sensitized by autologous type I1 collagen itself. Specific
antibodies to HI1 were observed in 3 of the 202 serum
samples from the RA patients and in 1 of the 19
serum samples from the OA patients (Table 1). Four of
12 RA sera that were passed through an HI1 column
did not contain species-specific antibodies (Table 3).
Autologous type I1 collagen is itself immunogenic and
is capable of inducing arthritis in mice (36), rats (I),
and cynomolgus monkeys (unpublished observations).
Antibodies found in the sera of the control
subjects tend to be species specific, and they resemble
the antibodies produced by guinea pigs that are resistant to collagen-induced arthritis (CIA) (38). Such
antibodies do not react with homologous type I1
collagen. Species-specific antibody formation, which
may be accompanied by suppression of autoreactive
antibody formation, has been observed not only in
CIA-resistant male cynomolgus monkeys, but also in
CIA-susceptible female monkeys with late-stage arthritis (15). However, the levels of antibody found in 3
of the 5 control sera (Table 2) were relatively low
compared with the selected RA sera that were studied
in detail. Therefore, we cannot exclude the possibility
that these control sera contain some cross-reactive
antibodies to homologous type I1 collagen.
In this study, we examined antibodies to native
type I1 collagen, because of their ability to induce
arthritis by binding to cartilage and activating the
complement cascade (7). We would like to point out
that the generation of antibodies to native type I1
collagen is not dependent on immunization with native
collagen. Arthritis and generation of antibodies to
native type I1 collagen have been observed in DBN1
mice immunized with denatured CII and with chick
CB-11 peptide (12). Furthermore, severe arthritis can
be induced in female cynomolgus monkeys by immunization with chick CB-11, CB-10, and CB-12 (unpublished observations). The anti-type I1 collagen antibodies produced by both of these species definitely
cross-react with native homologous type I1 collagen.
However, in both species, CB-8 is not arthritogenic,
nor do the antibodies elicited cross-react with homologous type I1 collagen (ref. 12 and unpublished obser-
TERATO ET AL
vations). These observations indicate that the helical
conformation of type I1 collagen, though important, is
not absolute for the induction of autoreactive antibodies and arthritis.
It is unlikely that RA patients and normal
controls were sensitized to native type I1 collagens.
Nevertheless, it is quite apparent that sera from a
number of RA and relapsing polychondritis patients
and from a few of the normal controls contained
antibodies to heterologous native type I1 collagen.
Their presence is not, uncommonly associated with
antibodies to BI, which is a major component of
gelatin in the human diet. These findings suggest that
some anticollagen antibodies might have been stimulated by denatured collagen and degraded collagen
products that were absorbed from the gastrointestinal
tract. This hypothesis is supported by Tarkowski and
coworkers’ findings of IgG and IgA anti-type I1 collagen antibody-secreting cells in the synovium of RA
patients, as determined by ELISPOT assay using rat
type I and type I1 collagen (39). Likewise, IgA has
been shown to be the predominant isotype of antibody
produced by peripheral B cells upon oral stimulation
with bacterial cell wall antigen (40). Hence, it is of
utmost importance to test the present hypothesis that
denatured type I1 collagens in the diet play a role in the
stimulation of autoantibody formation. In this respect,
we have detected IgA anti-type I1 collagen antibodies
in RA and non-RA sera by using a biotin-avidin
ELISA. The incidence of IgA anti-type I1 collagen
antibody-positive sera was found to be twice as high as
that of IgG antibody-positive sera (unpublished observations).
It was of interest to find the autoantigenic
determinants on type I1 collagen in RA and relapsing
polychondritis patients and to compare them with the
arthritogenic epitopes in experimental arthritis induced by type I1 collagen. Although not all CB peptides were studied, because some are quite small (2-33
amino acid residues) and were not isolated in sufficient
quantity to allow testing, the CB peptides used in this
experiment covered more than 95% of the type I1
collagen molecule. In RA patients, various patterns of
peptide reactivity emerged, as we observed previously
in individual arthritic monkeys immunized with CII
(15). All of the peptides tested were reactive in these
RA patients. In addition, it was rare that only a single
peptide was recognized. CB-11 and CB-8 were more
frequently recognized, but some sera failed to recognize either or both of these 2 peptides. In contrast, all
3 of the relapsing polychrondritis patients’ sera that
were tested recognized CB-9-7. Thus, it would appear
ANTI-TYPE I1 COLLAGEN SPECIFICITY
that while no single antigenic specificity is associated
with RA, relapsing polychrondritis may be associated
with an epitope on CB-9-7. However, the number of
relapsing polychrondritis sera available for testing was
limited, and more patient sera will be required to
confirm this finding.
REFERENCES
1. Andriopoulos NA, Mestecky J , Miller EJ, Bradley EL:
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
Antibodies to native and denatured collagens in sera of
patients with rheumatoid arthritis. Arthritis Rheum 19:
613-617, 1976
Greenbury CL, Skingle J: Anti-cartilage antibody. J Clin
Pathol 32:826-831, 1979
Beard HK, Ryvar R, Skingle J, Greenbury CL: Anticollagen antibodies in sera from rheumatoid arthritis patients. J Clin Pathol 33: 1077-1081, 1980
Andriopoulos NA, Mestecky J, Miller EJ, Bennett JC:
Antibodies to human native and denatured collagen in
synovial fluid of patients with rheumatoid arthritis. Clin
Immunol Immunopathol6:209-212, 1976
Clague RB, Moore LJ: IgG and IgM antibody to native
type I1 collagen in rheumatoid arthritis serum and synovial fluid: evidence for the presence of collagenanticollagen immune complexes in synovial fluid. Arthritis Rheum 27:137&1377, 1984
Jasin HE: Autoantibody specificities of immune complexes sequestered in articular cartilage of patients with
rheumatoid arthritis and osteoarthritis. Arthritis Rheum
28:241-248, 1985
Watson WC, Cremer MA, Wooley PH, Townes AS:
Assessment of the potential pathogenicity of type I1
collagen autoantibodies in patients with rheumatoid arthritis: evidence of restricted IgG3 subclass expression
and activation of complement C5 to C5a. Arthritis
Rheum 29:13161321, 1986
Trentham DE, Townes AS, Kang AH: Autoimmunity to
type I1 collagen: an experimental model of arthritis. J
Exp Med 146:857-868, 1977
Courtenay JS, Dallman MJ, Dayan AD, Martin A,
Mosedale B: Immunization against heterologous type I1
collagen induced arthritis in mice. Nature 283:666-668,
1980
Morgan K , Clague RB, Shaw MJ, Firth SA, Twose TM,
Hold PJL: Native type I1 collagen-induced arthritis in
the rat: the effect of complement depletion by cobra
venom factor. Arthritis Rheum 24:1356-1362, 1981
Terato K, Hashida R, Miyamoto K, Morimoto T, Kato
Y, Kobayashi S, Tajima T, Otake S, Hori H , Nagai Y:
Histological, immunological and biochemical studies on
type I1 collagen-induced arthritis in rats. Biomed Res
3:495-505, 1982
Terato K, Hasty KA. Cremer MA, Stuart JM, Townes
AS, Kang AH: Collagen-induced arthritis in mice: local-
1499
ization of an arthritogenic determinant to a fragment of
type I1 collagen molecule. J Exp Med 162:637-646, 1985
13. Cathcart ES, Hayes KC, Gonnerman WA, Lazzari AA,
Franzblau C: Experimental arthritis in a nonhuman
primate. 1. Induction by bovine type I1 collagen. Lab
Invest 54:26-31, 1986
14. Yo0 TJ, Kim S-Y, Stuart JM, Floyd RA, Olson GA,
Cremer MA, Kang AH: Induction of arthritis in monkeys by immunization with type I1 collagen. J Exp Med
1681777-782, 1988
15. Terato K , Arai H, Shimozuru Y, Fukuda T, Tanaka H,
Watanabe H, Nagai Y, Fujimoto K, Okubo F, Cho F,
Honjo S, Cremer MA: Sex-linked differences in susceptibility of cynomolgus monkeys to type I1 collageninduced arthritis: evidence that epitope-specific immune
suppression is involved in the regulation of type I1
collagen autoantibody formation. Arthritis Rheum 32:
748-758, 1989
16. Wooley PH, Luthra HS, Singh SK, Huse AR, Stuart
JM, David CS: Passive transfer of arthritis to mice by
injection of human anti-type I1 collagen antibody. Mayo
Clin Proc 59:737-743, 1984
17. Michaeli D, Fudenberg HH: The incidence and antigenic
specificity of antibodies against denatured human collagen in rheumatoid arthritis. Clin Immunol Immunopathol 2:153-159, 1974
18. Cracchiolo A 111, Michaeli D, Goldberg LS, Fudenberg
HH: The occurrence of antibodies to collagen in synovial fluids. Clin lmmunol Immunopathol3:567-574, 1975
19. Andriopoulos NA, Mestecky J, Wright GP, Miller EJ:
Characterization of antibodies to the native human collagens and to their components alpha chains in the sera
and the joint fluids of patients with rheumatoid arthritis.
lmmunochemistry 13:709-712, 1976
20. Menzel J, Steffen C, Kolarz G, Eberl R, Frank 0,
Thrumb N: Demonstration of antibodies to collagen and
collagen-anticollagen complexes in rheumatoid synovial
fluid. Ann Rheum Dis 35:446450, 1976
21. Menzel J, Steffen C, Kolarz G, Kojer M, Smolen J:
Demonstration of anticollagen antibodies in rheumatoid
arthritis synovial fluids by ''C-radioimmunoassay. Arthritis Rheum 21:243-248, 1978
22. Clague RB, Shaw JM, Holt PJL: Incidence of serum
antibodies to native type I and type I1 collagens in
patients with inflammatory arthritis. Ann Rheum Dis
39:201-206, 1980
23. Stuart JM, Huffstutter EH, Townes AS, Kang AH:
Incidence and specificity of antibodies to types I, 11,111,
IV, and V collagen in rheumatoid arthritis and other
rheumatic diseases as measured by 'Z'I-radioimmunoassay. Arthritis Rheum 262332440, 1983
24. Rowley M, Tait B, Mackay IR, Cunningham T, Phillips
B: Collagen antibodies in rheumatoid arthritis: significance of antibodies to denatured collagen and their
association with HLA-DR4. Arthritis Rheum 29: 174184, 1986
1500
25. Morgan K, Clague RB, Collins I, Ayad S , Phinn SD,
Holt PJ: Incidence of antibodies to native and denatured
cartilage collagens (types 11, IX, and XI) and to type I
collagen in rheumatoid arthritis. Ann Rheum Dis 4692907, 1987
26. Foidart JM, Abe S, Martin GR, Zizix RM, Barnett EV,
Lawley TJ, Katz SI: Antibodies to type I1 collagen in
relapsing polychondritis. N Engl J Med 299:1203-1207,
1978
27. Ebringer R, Rook G, Swana GT, Bottazzo GF, Doniach
D: Autoantibodies to cartilage and type I1 collagen in
relapsing polychondritis and other rheumatic diseases.
Ann Rheum Dis 40:473-479, 1981
28. Yo0 TJ, Stuart JM, Kang AH, Townes AS, Tomoda K,
Dixit S: Type I1 collagen autoimmunity in otosclerosis
and Meniere’s disease. Science 217:1153-1155, 1982
29. Mackel AM, DeLustro F, Harper FE, LeRoy EC:
Antibodies to collagen in scleroderma. Arthritis Rheum
25~522-531,1982
30. Gay S , Ward WQ, Gay RE, Miller EJ: Autoantibodies to
basement membrane collagen: epidermolysis bullosa
simplex versus bullous pemphigoid. J Cutan Pathol 7:
315-317, 1980
31. Suou T, Hirayama C: Antibodies to denatured bovine
collagens in sera of patients with liver disease. Clin Exp
Immunol39:119-124, 1980
32. Fuji K, Tsuji M, Murota K, Terato K, Shimozuru Y,
Nagai Y: An improved enzyme-linked immunosorbent
assay of anti-collagen antibodies in human serum. J
Immunol Methods 124:63-70, 1989
33. Medsger TA Jr, Masi AT: Epidemiology of the rheumatic diseases, Arthritis and Allied Conditions. Tenth
edition. Edited by DJ McCarty. Philadelphia, Lea &
Febiger, 1985
TERATO ET AL
34. Terato K, Cremer MA, Hasty KA, Kang AH, Townes
AS: Physicochemical and immunological studies of the
renatured cxl(I1) chains and isolated cyanogen bromide
peptides of type I1 collagen. Coll Relat Res 5:469-480,
1985
35. Beard HK, Lea DJ, Ryvar R: Anomalous reactions in
the haemagglutination assay for anti-collagen antibodies: studies on patients with rheumatoid arthritis or
chronic low back pain. J Immunol Methods 31:119-128,
1979
36. Holmdahl R, Jansson L , Larsson E, Rubin K, Klareskog
L: Homologous type I1 collagen induces chronic and
progressive arthritis in mice. Arthritis Rheum 29:106113, 1986
37. Kerwar SS,Englert ME, McReynolds RA, Landes MJ,
Lloyd JM, Oronsky AL, Wilson FJ: Type I1 collageninduced arthritis: studies with purified anticollagen
immunoglobulins. Arthritis Rheum 26:1120-1 131, 1983
38. Hernandez AD, Cremer MA, Townes AS, Stuart JM,
Kang AH: The immune response of guinea-pigs to type
I1 collagen: poor cross-reactivity with homologous type
I1 collagen accounts for resistance to collagen-induced
arthritis. Immunology 63:619-624, 1988
39. Tarkowski A, Klareskog L, Carlsten H, Herberts P,
Koopman WJ: Secretion of antibodies to types I and I1
collagen by synovial tissue cells in patients with rheumatoid arthritis. Arthritis Rheum 32:1087-1092, 1989
40. Czerkinsky C, Prince SJ, Michalek SM, Jackson S,
Russel MW, Moldoveanu Z, McGhee J, Mestecky J:
IgA antibody-producing cells in peripheral blood after
antigen ingestion: evidence for a common mucosal immune system in humans. Proc Natl Acad Sci USA
84:2449-2453, 1987
Документ
Категория
Без категории
Просмотров
2
Размер файла
793 Кб
Теги
antibodies, arthritis, typed, specificity, collagen, rheumatoid
1/--страниц
Пожаловаться на содержимое документа