close

Вход

Забыли?

вход по аккаунту

?

The B cell repertoire of patients with rheumatoid arthritis. II. Increased frequencies of IgG+ and IgA+ B cells specific for mycobacterial heat-shock protein 60 or human type II collagen in synovial fluid and tissue

код для вставкиСкачать
ARTHRITIS & RHEUMATISM
Val. 40, No. 8, August 1997, pp 1409-1419
0 1997, American College of Rheumatology
1409
THE B CELL REPERTOIRE OF PATIENTS WITH
RHEUMATOID ARTHRITIS
11. Increased Frequencies of IgG+ and IgA+ B Cells Specific for
Mycobacterial Heat-Shock Protein 60 or Human Type I1 Collagen in Synovial Fluid and Tissue
ULRIKE RUDOLPHI, RITA RZEPKA, STEPHEN BATSFORD, STEFAN H. E. K A U F M A " ,
KLAUS VON DER MARK, HANS H. PETER, and INGA MELCHERS
Objective. A qualitative and quantitative analysis
of the functional, antigen-specific B cell receptor repertoire of patients with rheumatoid arthritis (RA) in
synovial and peripheral compartments.
Methods. B cells were activated to grow and
differentiate at high efficiency in vitro under limitingdilution conditions. Isotype and specificity of the secreted Ig were tested by enzyme-linked immunosorbent
assay.
Results. In contrast to peripheral B cells, most
synovial B cells had already switched to IgG/IgA in vivo.
The frequencies of B cells specifically recognizing foreign antigens were decreased within the synovial population, whereas the frequencies of B cells specific for
type I1 collagen, mycobacterial heat-shock protein 60
(hsp60), or IgG Fc fragments were significantly increased, revealing a negative correlation in terms of
frequencies.
Conclusion. B cells specific for human type I1
collagen, hsp60, and IgG Fc fragments are produced
and/or expanded locally within the affected joints of RA
patients. Thus, the specific immune system is definitely
involved in the local inflammatory and destructive
processes.
Supported by Deutsche Forschungsgemeinschaft grant Pe
15lilO.
Ulrike Rudolphi, MS, Rita Rzepka, BS, Stephen Batsford,
PhD, Hans H. Peter, MD, Inga Melchers, PhD: Albert-Ludwigs
University, Freiburg, Germany; Stefan H. E. Kaufmann, PhD: University of Ulm, Ulm, Germany; Klaus von der Mark, PhD: University of
Erlangen-Nuremberg, Erlangen, Germany.
Address reprint requests to Inga Melchers, PhD, Klinische
Forschergruppe fur Rheumatologie, Klinikum der Albert-LudwigsUniversitat, Breisacher Strasse 64, D79106 Freiburg, Germany.
Submitted for publication November 13, 1996; accepted in
revised form March 17, 1997.
Rheumatoid arthritis (RA) is immunologically
characterized by infiltrations of lymphocytes into the
synovium, the presence of rheumatoid factors (RFs) in
most patients (seropositive RA), and a well-established
association with certain alleles of the HLA-DRB1 gene
(1-3). This phenomenology has led to the hypothesis
that T lymphocytes, restricted to the associated
HLA-DR alleles, recognize a synovial autoantigen and
induce an autoimmune reaction that encompasses the
activation of autoreactive B cells and the production of
RF (for review, see refs. 4 and 5). During recent years,
attention has focused on the search for appropriate T
cell autoantigens. However, neither attempts to directly
stimulate synovial T cells with (auto) antigens (6-11) nor
a more indirect approach via analysis of T cell receptor
V-gene usage by synovial T cells (12-15) has provided
satisfactory answers to this question.
B cells form a minority of the infiltrating lymphocytes (16). Interestingly, B cells in all stages of differentiation, from mature B cells to plasma cells, have been
observed (16-20). There is some molecular evidence
that B cells are specifically activated and clonally expanded in the synovium (21-23). Organized structures
resembling germinal centers have been described, which
contain follicular dendritic cells and CD4+ T cells and
reveal clonal expansion of B cells exhibiting isotype
switching and somatic hypermutation (19,22). Numerous
plasma cells have been found which produce IgM, IgG,
and IgA (24); some of these also secrete R F or antibodies specific for type I1 collagen (CII) (17,25-27).
To study the B cell repertoire, we used a limitingdilution system, in combination with highly efficient
activation, proliferation, and differentiation of B cells
(28-30). We measured antibodies derived from clones of
RUDOLPH1 ET AL
1410
Table 1. Characteristics of the patients with rheumatoid arthritis*
Patient
Sedage
HLA-DRt
ESR
(mmlhour)
CRP
(mddl)
RA1
RA2
RA3
RA4
RA5
RA6
RA7
RA8
RA9
RAlO
MI60
Fl66
Fl56
Fl59
MI39
F136
Fl32
MI50
F156
Fl67
ND
ND
ND
ND
41w6
517
213
040111302
040110404
417
50
99
68
25
24
25
5
32
ND
54
10.9
ND
1.9
0.8
1.9
1.2
0.5
3.4
ND
5.3
RF
+
+
ND
-
+
+
+
+
+
-
Disease duration
(years)
8
3
1
26
2
2
6
13
3
8
Medication$
Material
studied5
GC
MTX
GC
BT
GC
GC, anti-CD4
BT, GC
BT
GC
GC
SFIST
SFIST
SFIST
SFIST
SF
SF
SF
SF
SF
SF
* ESR = erythrocyte sedimentation rate; CRP = C-reactive protein; R F = rheumatoid factor (seropositive [+I or seronegative [-I); ND = not
determined.
t Given is either the HLA-DR serotype or the DRBl alleles (31).
3 Patients were treated with basic therapy (BT; auranofin, gold sodium thiomalate), glucocorticoids (GC), methotrexate (MTX), or anti-CD4
monoclonal antibody therapy.
5 Either synovial fluid (SF) was aspirated, or a mixture of SF and synovial tissue (ST) was collected by arthroscopic synovectomy.
antibody-secreting cells (ASC) after activation in vitro,
and deduced the numbers and specificities of their
precursors in vivo (i.e., mature Ig+ B cells with receptors of the same kind [p-ASC]). Herein we present the
results of the analysis of the synovial IgG+ and IgA+ B
cell receptor repertoire, performed to identify autoantigens recognized by local B cells. In addition, we compared
the functional B cell repertoires in 2 compartmentssynovial fluidhynovial tissue (SF/ST) and peripheral blood
(PB)-of individual patients with RA.
PATIENTS AND METHODS
Patients. In this study, we used material from 10
patients with RA. Data on these patients, including clinical
parameters and therapy, are shown in Table 1. The diagnosis
of RA was established according to the American College of
Rheumatology (formerly, the American Rheumatism Association) criteria (31). Arthroscopic synovectomy was performed
as clinically indicated. HLA typing was done by standard
cytotoxicity assays or as described previously (32).
Preparation and characterization of B cells. SF/ST
material was concentrated by centrifugation and incubated
overnight in the presence of collagenase and DNase. Released
cells were harvested the next day. These cells, as well as SF
cells and PB mononuclear cells (PBMC), were separated by
Ficoll-Hypaque gradient centrifugation and depleted of T
lymphocytes by rosetting with sheep red blood cells. The
remaining cell population was enriched for B lymphocytes and
is referred to as B+ cells. B+ cells were characterized by
staining with antibodies to CD3 (OKT3; American Type
Culture Collection, Rockville, MD), CD14 (Leu-M3; Becton
Dickinson, Mountain View, CA), or CD19 (IOB4; Dianova,
Hamburg, Germany) and fluorescein isothiocyanate-labeled
goat anti-mouse Ig (Dianova), followed by flow cytometry
(FACScan; Becton Dickinson).
Limiting dilution (LD) cultures. LD cultures were
performed as described previously (28,30). Briefly, 100 pl of
LD medium (LDM) containing 10% conditioned medium
(CM) and 3 ng/ml of phorbol myristate acetate (PMA) were
placed into round-bottomed microtiter wells. EL-4 B5 cells
(5 X 104/culture) irradiated with 50 Gy (137Cs source; IBL,
Oris Industries, Gif-sur-Yvette, France) were added in 50 pl of
LDM, consisting of RPMI 1640, L-glutamine (2 mM), HEPES
(10 mM), penicillin (100 units/ml), streptomycin (100 pg/ml),
5 X lOP5M 2-mercaptoethanol, and 10% fetal calf serum. B+
cells were added in 2 series of 2-fold dilutions, ranging from
32-0.5 cells/culture and 1,000-60 cells/culture. For each cell
concentration, 224 replicates were set up. Control cultures did
not receive B+ cells.
Cultures were incubated at 37°C for 10 days, and the
supernatants were collected and stored at -70°C. CM was
prepared from bulk cultures of nylon wool-purified T lymphocytes (from several healthy donors) stimulated at 1 X lo6
cells/well in LDM supplemented with PMA (10 ng/ml) and
phytohemagglutinin (5 pg/ml) for 36 hours. Supernatants were
centrifuged, filtered (0.22 pm), and stored in aliquots at
-20°C.
Antigens. As coating antigens in the enzyme-linked
immunosorbent assays (ELISAs), we used human IgG Fc
fragments (Dianova), human CII, and recombinant mycobacterial heat-shock protein 60 (hsp60), both purified as described
previously (28,33,34). Tetanus toxoid (7T)was kindly provided
by K. D. Hungerer (Behringwerke, Marburg, Germany). The
streptococcal proteinase proenzyme (SPP) was purified as
described (35). For coating ELISA plates, antigens were used
at a concentration of 5 pg/ml.
Assays for antibody quantification and specificity.
Antibodies present in tissue culture supernatants were assayed
with ELISAs as described (28), using goat anti-human IgM,
IgA, or IgG labeled with alkaline phosphatase as second
antibodies and paranitrophenyl phosphate as substrate. The
amount of Ig/culture was determined in ELISAs using goat
anti-human Ig as the coating and dilutions of human IgM, IgG,
SYNOVIAL B CELL REPERTOIRE IN FL4
1411
Table 2. Activation of B lymphocytes in vitro
SFL: IgH and IgG
Compartment*
Determination
PB
SFiST
No. of samples tested
10
4
CD19+ cells (%)t
Mean i SD
15.7 t 13.0 12.8 f 9.1
Range
1.0-40
1.0-20
Activation (% CD19+ cells)$
Total Ig
Mean -C SD
88.9 ? 17.6 42.6 2 41.2
Range
60-100
1.2-85
IgM
86.7 t 21.8 4.4 i 5.9
Mean f SD
Range
50-100
0.7-13
IgG
Mean i SD
63.8 t 35.0 10.8 2 18.9
Range
25-100
0.3-39
IgA
Mean f SD
65.3 i 26.2 35.1 t 35.7
Range
25-100
0.3-85
* PB
=
peripheral blood; SF = synovial fluid; ST
=
SF
6
28.0 i 13.5
8-45
PEL: IgH and
Id
32.0 2 21.5
10-71
*%
4.7 2 3.7
2.0-11
17.8 f 18.2
3.0-50
PBL:
9
0
1
2
3
0
1
2
3
I& and IgG
11.0 2 8.1
3.0-20
1
synovial tissue.
t Cells were enriched for B cells by depletion of T cells.
i Activation of B cells is expressed as the percentage of CD19+ cells
placed into culture which were induced to secrete the isotype indicated.
j
0
1
2
M
i 1
Patient
3
EAU 1
Figure 2. Segregation patterns of clones producing different Ig
classes, determined at low concentrations of B+ cells (32-0.5 cellsi
culture). Each culture was tested for the presence of IgM, IgG, and
IgA. The enzyrne-linked immunosorbent assay absorbance units
(EAU) that were obtained are plotted in pairs, with the firstmentioned Ig class on the abscissa (EAUI). Data obtained with cells
from 1 representative experiment are shown. Clones switching in vitro
appear on the diagonal line (peripheral blood lymphocytes [PBL]: IgM
and IgG and PBL IgM and IgA). Clones that already switched in vivo
are detected as IgM-,IgG+ and IgM-,IgA+ (top 2 plots of synovial
fluid lymphocytes [SFL]). IgA- and IgG-producing B cells segregate in
both populations (PBL and SFL).
or IgA as standards (all reagents from Dianova). ELISA
absorbance units (EAU) were read at a Atest/hreference
ratio of
405 nm/490 nm.
Data processing and statistical analysis. EAU values
were further processed as described elsewhere (28). All frequencies shown were estimated in experiments with 2 3 F,
values # 1.00, using
minimization; only estimations with P
values > 0.05 were accepted and are shown here. Segregation
analyses were performed with cultures containing a mean of
(1 B cell with the attribute of interest (i.e., F, 2 0.37). In
addition to tl, a second threshold t2 (t2 = 2 X t l ) was used in
the segregation analyses.
2
Figure 1. Dominant Ig isotypes after activation in culture among
rheumatoid arthritis (RA) patients. Black squares indicate the Ig
isotype which was most frequently produced after in vitro activation of
B cells under limiting-dilution conditions (32-0.5 cellsiculture). PB =
peripheral blood; SF = synovial fluid; ST = synovial tissue; M = IgM;
G = IgG; A = I&.
RESULTS
In vitro activation of synovial B cells. Synovial B
cells were obtained from patients who underwent either
aspiration of synovial fluid (6) or arthroscopic synovec-
RUDOLPH1 ET AL
1412
Table 3. Frequencies of antigen-specific p-ASC in synovial and peripheral B cell populations*
Synovial B cells
Peripheral blood B cells
Antigen,
isotype
Sample size
(ndnt)?
Frequencies
(% of Ig-producing cells)
(mean t SD)$
Sample size
(n,,InJ t
Frequencies
(% of Ig-producing cells)
(mean 5 SD)$
414
718
3.73 t 1.85
0.36 F 0.31
414
718
1.75 ? 1.01
1.50 5 1.01
317
415
0.23 +- 0.06
0.58 2 0.44
517
415
3.48 ? 6.03
1.60 ? 1.66
416
415
0.15 ? 0.10
1.10 5 0.91
616
515
2.63 t 2.65
1.02 ? 1.23
111
111
(0.3)
(2.6)
111
111
(0.3)
(0.1)
212
(242)
212
(<0.1,0.2)
* Data from individual experiments are shown in parentheses. p-ASC = precursor antigen-secreting cells;
CII = type I1 collagen; hsp60 = mycobacterial heat-shock protein 60; TT = tetanus toxoid; SPP =
streptococcal proteinase proenyzme.
t Sample size is given as number of experiments with frequency estimates >0.03% (n,) and number of
experiments performed (q).
$ Geometric means and standard deviations were calculated from all experiments with frequency
estimates >0.03% (n,,).
tomy (4) (Table 1). The latter material consisted of
pieces of synovial tissue, but also contained single cells
set free during the removal procedure or before. Free
synovial fluid cells were isolated immediately. SF/ST was
incubated overnight in the presence of enzymes, and
single cells were collected the next day. Peripheral blood
was also obtained at the same time.
These cell populations contained varying numbers of CD19+ B cells, which were enriched by the
depletion of T cells (B+ cells). B+ cells were activated
in limiting dilution cultures as described previously (28).
Replicate cultures (r24/group) with low concentrations
of B cells, ranging from 100 to 0.5 B + cells per culture,
served to estimate the frequencies of CD19-t cells that,
in culture, could be activated to grow and differentiate
into ASC. A summary of these data is given in Table 2.
Whereas most peripheral CD19+ B cells responded to activation in vitro with clonal expansion and
Ig secretion (89%), only about one-third of the synovial
CD19+ B cells could be expanded (mean 36%, range
14.5%).B cells derived from SF did not significantly
differ from B cells derived from SF/ST in terms of
activation and Ig isotype distribution (Table 2). We
therefore subsequently refer to both as synovial B cells.
Synovial B cells preferentially produce IgG or
IgA. Nearly all peripheral B cells were precursors of IgM
ASC and switched to the production of IgG or IgA in
vitro, under the culture conditions used (Table 2 and
Figures 1 and 2). In contrast, in the synovial B cell
population, p-ASC for IgG or IgA were more frequent
than p-ASC for IgM (Table 2 and Figures 1 and 2).
Segregation analyses demonstrated that most synovial
p-ASC for IgG or IgA had already switched in vivo,
before their reactivation in vitro (Figure 2).
In some experiments, we calculated the clone
sizes of p-ASC reached in LD cultures in vitro (not
shown). In general, clone sizes of p-ASC for IgG and
IgA were slightly lower than those of p-ASC for IgM
(28). There were no significant differences between
peripheral and synovial B cells.
In conclusion, the LD system allows activation of
a high proportion of peripheral and synovial B cells,
which will expand and differentiate in vitro into clones
that produce antibodies. The analysis of these antibodies
allows the composition of the original patient B cell
receptor repertoire to be deduced.
Frequencies of antigen-specific B cells in synovial and peripheral lymphocyte populations. The following experiments were performed to determine whether
B cells with defined specificities undergo antigeninduced expansion at the site of inflammation, that is, in
the synovial environment of patients with RA. We tested
supernatants from in vitro-activated synovial and peripheral B cell cultures for antibodies that recognize
SYNOVIAL B CELL REPERTOIRE IN FL4
15
1413
Fi
1gG:TTISSP
v
SFIST
1gG:CII
A IgG:hsp60
z
0 1gA:TTISSP
Y
v
.-5
-a
1gA:CII
* IgA:hsp6O
U
Q
0 1gA:RF
$0
-n
al
0
-m
Q
>
0
c
*
(I)
E
-* 5
0
C
al
3
U
2
LL
0
*.A0
-1
0
1
2
3
Frequency in peripheral B cell population (%)
A
-1
0
2
1
3
Frequency in peripheral B cell population (%)
B
Figure 3. Negative correlation between frequencies estimated in the synovial and the peripheral B cell populations. Frequencies of precursor
antibody-secreting cells (p-ASC) for IgG and IgA specific for several antigens were estimated in both compartments of individual patients and
plotted against each other (A). In addition, the origin of synovial B cells from synovial fluidisynovial tissue (SF/ST) or SF is indicated (B). The
Spearman rank correlation coefficient was calculated using all data (n = 34; r = -0.42; P = 0.0168). TT = tetanus toxoid; SPP = streptococcal
proteinase proenzyme; CII = type I1 collagen; RF = rheumatoid factor.
autologous molecules, which are supposed to function as
autoantigens in the development of arthritis, as well as
control antigens. As autoantigens we chose human CII
and, to test for IgM and IgA RFs, human IgG Fc
fragments. As control antigens, we used IT and a
streptococcal proteinase proenzyme. In addition, antibodies binding to purified recombinant hsp60 of Mycobacterium tuberculosis were analyzed.
To compare peripheral and synovial B cell populations, all frequencies were calculated as percentages
of activated B cells secreting antibodies of a given
specificity and isotype, regardless of differences in composition and activation potential. The results are sum-
marized in Table 3 and plotted individually in Figures
3A and B.
We detected p-ASC for IgM RFs in all RA
samples tested. The frequencies in the peripheral cell
population (3.73%) were slightly higher than those that
have been described previously (1.48%) (28). We observed p-ASC for IgA RFs in 88% of the samples. The
synovial cell populations contained relatively more pASC for IgA RFs (-4-fold higher) and fewer p-ASC for
IgM RFs (47%).
We found p-ASC for IgG anti-CII in 3 of 7
PBMC samples (43%) as well as in 5 of 7 SF/ST samples
(71%); p-ASC for IgA anti-CII were detected in 80% of
RUDOLPH1 ET AL
1414
I
0
2
0
0
m
h
W
6
o
o
s'
EAU 1
7
Figure 4. Specificity of individual synovial B cell clones. Segregation
analyses of a representative experiment are shown (patient RA8).
Individual culture supernatants (n = 150) were tested for IgA antibody
binding to 3 different antigens (a-c) and for IgG antibody binding to
2 different antigens (d). EAU = enzyme-linked immunosorbent assay
absorbance units.
PBMC and SF/ST samples. Mean frequencies were
higher in the synovial populations (-15-fold for IgG and
-3-fold for IgA).
Precursor ASC for IgG or IgA anti-hsp60 were
detected in most PBL and in all synovial lymphocyte
samples. The frequencies of p-ASC for IgG were similar
(0.15% versus 0.09%) and the frequencies of p-ASC for
IgA were higher (1.1% versus 0.27%) than observed
previously (28). In the patients studied here, the mean
frequencies of p-ASC for IgA anti-hsp60 were comparable in the PB and the synovial populations; p-ASC for
IgG anti-hsp60, however, were increased in the synovial
population (- 18-fold higher).
Precursor ASC reactive with the control antigens
TT and SPP were detected in 7 of 8 samples tested,
which were obtained from 4 patients with RA. Frequencies were either similar in both compartments (1 case,
IgG anti-7T) or higher in the PB population (3 cases,
IgA anti-TT 26-fold and IgG anti-SPP >22-fold and
11-fold).
Synovial IgG- or IgA-producing B cells are antigen specific. Using a LD culture system, it is possible to
determine the specificity of individual B cell clones by
testing their supernatants for binding to 2 or more
different antigens, followed by segregation analyses
(28,36). These assays have to be performed at B cell
concentrations that provide a good probability for
clonality of the B cells with the highest frequency. We
performed segregation analyses of this type whenever
possible. In all cases, we detected only IgA-producing B
cell clones specific for either IgG Fc fragments, hsp60, or
CII and IgG-producing B cell clones specific for either
hsp60 or CII. A representative experiment is depicted in
Figure 4.
Frequencies of p-ASC for different isotypes with
the same specificity do not correlate. We detected
p-ASC for IgG and for IgA with specificity for all
antigens tested. Individual patients showed different
patterns of response, that is, in some cases, we found
similar frequencies of p-ASC for IgG and IgA antibodies
(e.g., in patient RA8, synovial B cells specific for CII),
and in other cases, there was dominance of only 1
isotype (e.g., in patient RA10, synovial B cells contained
>10 times more p-ASC for IgG anti-hsp60 than for IgA
anti-hsp60). As shown in Table 4, we did not detect any
correlation between p-ASC for IgG and for IgA with the
same specificity. This was also true for PBMC, most of
which switch in vitro (compare with Figure 1). In
addition we showed, by segregation analyses, that individual PB-derived clones only secreted 1isotype with the
specificity studied, e.g., only IgM-RF or only IgA-RF
(not shown). As discussed previously (29), we conclude
that we could only detect specific p-ASC for IgG and
IgA that have switched in vivo.
Frequencies of p-ASC in the synovial and the
peripheral compartment are negatively correlated. The
mean frequencies of p-ASC for anti-CII (IgG and IgA)
and anti-hsp60 (IgG) were increased in the synovial
compartment of RA patients, but individual frequencies
Table 4. Lack of correlation between the frequencies of p-ASC for
different Ig isotypes with the same specificity*
Specificity, isotypes compared
CII,
hsp60,
RF,
IgG vs. IgA IgG vs. IgA IgM vs. IgA
Spearman rank correlation
analysis
n
r
P
10
0.1737
0.6023
8
0.1667
0.6591
10
0.0420
0.8944
~
~
~
~~~~
* All pairs of individual frequencies of precursor antigen-secreting
cells (p-ASC) for different Ig isotypes with the same specificity were
analyzed. Shown are the number of pairs (n), the correlation coefficient (r), and the significance (P).No correlation was observed. RF =
rheumatoid factor; see Table 3 for other definitions.
SYNOVIAL B CELL REPERTOIRE IN RA
1415
msynovial B cells
0 peripheral B cells
0.1
1
10
100
Fraction of the total repertoire defined (X)
Figure 5. B cell repertoire defined by a limited set of antigens. Shown is the percentage of the total
repertoire, as calculated by adding the frequencies for individual rheumatoid arthritis (RA) patients.
(see text). For synovial and peripheral B cell populations, the number of different test systems used is
shown in parentheses after the patient number.
showed a broad distribution (Table 3 ) . We therefore
evaluated whether patients with the highest synovial
frequencies also present with the highest peripheral
frequencies simultaneously. However, when individual
pairs of frequencies of p-ASC were plotted against each
other, we detected a significant negative correlation
between the frequencies estimated in both compartments (Figure 3). This local increase strongly suggests
that B cells with these specificities are trapped and
expand specifically within the synovial environment.
How much of the B cell repertoire can be determined using a limited set of antigens? Using only a small
panel of antigens in our assay systems, we did not expect
to be able to determine the specificity of the majority of
p-ASC in either the periphery or the local repertoire.
For synovial B cells, we can estimate the size of the
repertoire by simple summation of the frequencies obtained for p-ASC for IgG, IgA, and IgM for the antigens
tested, since we saw no cross-reactivityand the IgG- and
IgA-secreting cells had already switched in vivo. For
peripheral B cells, such estimates are not as easy,
because the dominant p-ASC for IgM switch in culture
to IgG or IgA production, and because many p-ASC for
IgM produce antibodies which cross-react, for example,
with hsp60 and Fc fragments (28). A calculation performed in the same way as for synovial B cells may lead
to overestimation, although the peripheral IgG and IgA
clones analyzed in this study were specific and most
probably derived from B cells switched in vivo (not
shown). We show both sets of data in Figure 5. Interestingly, the percentages of specificities that could be
defined with our assays in most cases were higher in the
synovial repertoire, and in 3 cases close to or even higher
than 10% of the total repertoire.
DISCUSSION
It has been known for more than 20 years that
antibody-secreting cells-plasma cells-exist within the
synovial tissue (for review, see refs. 16 and 37). The
numbers of plasma cells reported vary between <1%
and a maximum of -20% of the total mononuclear cell
infiltrate (24,38). More than 50% of these ASC have
been shown to produce IgG, 20-25% IgA, and 515%
IgM (20,24,39). Within SF, B cells are less frequent, and
within the total B cell population, fewer ASC are
detected. Synovial B cells at stages of differentiation
earlier than ASC (i.e., p-ASC) differ from PB B cells in
terms of their distribution of surface Ig receptors. Youinou et a1 (40) reported that among the K+/A+ cells,
IgG+ and IgA+ cells comprised 54% and 64% in SF
and ST, respectively, compared with 21% in PB. In the
1416
same populations, the percentages of IgM+ ,IgD+ cells
only reached 1.3% and 0.3% in SF and ST, respectively,
instead of the 52.6% found in PB. Accordingly, the
fraction of IgM+,IgD- cells increased from 26% in PB
to 44.7% in SF and 35.7% in ST (40). Synovial B cells
spontaneously produce Ig and respond only poorly to
further activation with mitogens (e.g., pokeweed mitogen) (41). Consequently, the synovial B cell population
is generally regarded as a population of B cells at a
higher stage of differentiation than the PB B cells
(37,40,41).
We present here the first representative comparison of synovial and peripheral populations of mature B
cells in terms of the specificities and frequencies of their
antigen receptors. It was previously shown that, with the
culture system used, we can activate nearly all peripheral
B cells, drive them through proliferation and differentiation stages, and thus perform a representative analysis
of B cell specificities (28,29). The sensitivity of the assay
system allows detection of specific antibodies from individual B cell clones, but not from individual ASC. In this
type of analysis, we will therefore miss all B cells which
reached the stage of ASC in vivo or which, for other
reasons, did not divide or secrete Ig in vitro, but we can
identify all B cells which have expanded in vitro to the
appropriate clonal size and differentiated into ASC. We
refer to this latter cell population as p-ASC.
Consistent with the notion that a large fraction of
synovial B cells is already in a more advanced state of
differentiation, including Ig secretion (37,40,41), we
could only activate a subpopulation of all CD19+ synovial B cells (Table 2). This set of the B cell population
varied from patient to patient, without any obvious
correlation with the parameters studied (age, duration
of disease, or medication). The distribution of Ig isotypes detected in culture supernatants also reflected
variations in the stage of differentiation between synovial and peripheral B cells. Most peripheral B cell clones
produced IgM; in contrast, most synovial clones produced IgG or IgA (Figure 1). Most of the IgG and IgA
secreted in vitro by peripheral B cell clones was derived
from IgM+ p-ASC that had switched isotype in vitro,
whereas the majority of synovial B cells had already
switched their isotype in vivo (Figure 2). Thus, the Ig
isotype distribution of mature synovial B cells, as observed in our cultures, closely resembled the isotype
distribution of synovial plasma cells (20,37). We conclude that the synovial B cell populations mainly consist
of plasma cells and B cells activated in vivo, most
probably by antigen.
With our culture system, we can detect the recep-
RUDOLPH1 ET AL
tors of naive B cells and of activated B cells which are
still able to expand in vitro. Nearly all peripheral B cells,
but only one-third of synovial B cells, fall into this
category. We suggest that the low efficiency of in vitro
activation reflects the high degree of in vivo activation.
Moreover, it appears likely that the same antigens may
be recognized by B cells in these various stages of in vivo
activation. For comparison of B cell specificities, we
included all experiments in which more than a 10% level
of activation was achieved (this excluded only patient
RA3).
TT and SPP are 2 foreign antigens that are
recognized by B cells in most individuals. This reactivity
arises due to vaccination or to natural infection, respectively. We used these antigens to test the repertoires of
synovial and peripheral B cells from 4 patients for
antibody specificities not related to arthritis. The frequencies of p-ASC for IgG anti-TT were the same in
both compartments (1 case), whereas the frequencies of
p-ASC for IgA anti-TI and IgG anti-SPP were decreased in the synovial compartment (3 cases), being
below the limit of detection in 1 case (Table 3). These
results indicate that the higher numbers of IgG+ and
IgA+ B cells within the synovium do not simply represent trapping of B cells with these isotypes, but depend
on their receptor specificity.
The presence of RF in blood is a hallmark of
patients with RA (42). ASC-containing RF reside in the
synovial tissue. Their numbers, however, vary considerably between individual patients and between studies, in
which various techniques and IgG preparations have
been used (16,26,37-39,42). Precursor ASC have not yet
been studied. We used human IgG Fc fragments as an
autoantigen and the ELISA technique to detect RF. Of
all the synovial Ig-producing clones, 1.8% and 1.5%
secreted IgM-RF and IgA-RF, respectively, corresponding to 12.5% of all IgM- and 7.5% of all IgA-producing
clones (Table 3). In comparison to the periphery, the
total synovial cell population contained a smaller fraction of p-ASC for IgM-RF, but an increased frequency
of p-ASC for IgA-RF (4-fold higher; Table 3). For
technical reasons, we did not assay IgG-RF. All the
IgA-RF-secreting synovial B cell clones analyzed were
monospecific for human Fc fragments and did not
cross-react with hsp60 or CII (Figure 4). These data are
in agreement with the reported specificities of some
monoclonal RF from synovial material (16,23,37,44).
The data also complement our earlier observations regarding the peripheral repertoires of p-ASC for
IgM-RF in RA patients and healthy blood donors
(13,28). In these latter studies, we distinguished mono-
SYNOVIAL B CELL REPERTOIRE IN RA
specific RF from multireactive RF (cross-reacting with
mycobacterial hsp60 and/or human type V collagen and
other antigens), and showed a reversed ratio between
both populations of p-ASC in RA patients, leading to an
unusual dominance of the monospecific cells.
Mycobacterial hsp60 belongs to a family of heatshock proteins (HSP) with remarkable features, including their broad distribution within all living organisms,
the conservation of their primary structure, their immunodominance, and the existence of animal models of
arthritis established with this antigen (for review, see
refs. 45-47). The mycobacterial hsp60 used here shows
-50% sequence homology with its human counterpart.
At present, it is unclear whether the immune response of
RA patients to hsp60 differs from that of the healthy
population, and consequently, whether HSP are involved in the generation or perpetuation of this disease.
Our data show that a significant fraction of the peripheral B cell pool recognizes hsp60, with a predominance of TgA+ over IgG+ cells (28) (Table 3). Within
the synovial population, p-ASC for IgG or IgA antihsp60 were readily detected. Compared with the PB, the
frequency of p-ASC for IgG anti-hsp60 was increased
M-fold, such that IgG+ cells specific for hsp60 outnumbered IgA+ cells. Precursor ASC for hsp60 were specific
in that they did not cross-react with the other antigens
studied (IgG Fc fragment, CII). There are no data
available about synovial ASC with specificity for HSP;
however, some IgG+ hybridomas were recently established from synovial tissue which were specific for recombinant hsp60 from Yersinia enterocolitica in ELISA
(48). It will be important to study the fine specificity of
these cells further, especially to test if they can react with
the human HSP cognates: cross-reactivity between microbial and human HSP could provide an explanation
for the triggering of arthritis with bacteria.
One of the most widely discussed target autoantigens in human RA is CII (for review, see ref. 5).
Serum antibodies as well as ASC to this target have been
described in RA patients, but again, not without discrepancies concerning their quantities and qualities
(5,25,26). In our study, human CII was used to detect
antibodies by an ELISA, which works reliably with tissue
culture supernatants. Precursor ASC for IgG anti-CII
exhibited the highest frequency within the synovial B cell
population (Table 3). Again, these cells were specific, as
were the p-ASC for IgA anti-CII, which were detected in
slightly lower numbers (Figure 4). Both cell frequencies
were increased compared with the peripheral population
(15-fold and 3-fold, respectively). The numbers we estimated for synovial p-ASC were along the same order of
1417
magnitude as those published for synovial ASC (25,26).
There was no indication of an association between the
presence of HLA-DR4 and the frequency of p-ASC for
anti-CII (25). The highest frequency was observed in
patient RA7 (14.1% p-ASC for IgG anti-CII), who was
typed DR4-.
Are synovial B cells activated and/or expanded
within the synovium by mechanisms that involve local
autoantigens? Our data suggest that they are. In RA
patients, the frequencies of p-ASC for IgG and IgA
anti-CII and for IgA-RF were increased within the
synovial compartment; in contrast, the frequencies for
cells recognizing TT and SPP were decreased. All synovial IgG+ and IgA+ B cell clones analyzed were specific
within the panel of antigens tested. Moreover, frequencies estimated within the synovial compartment were
negatively correlated with those estimated in the PB
(Figure 4). We feel that this latter finding is of particular
importance, because it suggests that B cells are most
frequently detectable within the environment in which
they were generated, before distribution (and dilution)
via the bloodstream. As we have shown elsewhere,
p-ASC for IgG anti-TT are not present in detectable
amounts in the PB of subjects before vaccination, but
suddenly appear at high frequencies 9 days afterwards
(29). The present results suggest that TT-specific IgG+
p-ASC are not generated within the synovium; however,
IgG+ and IgA+ p-ASC for anti-CII are, as well as
p-ASC for IgA-RF and IgG anti-hsp60.
Although in most RA patients, the definable
fraction of the B cell repertoire was greater in the
synovial compartment than in the PB, the specificity of
more than 90% of these cells remained unidentified. We
expect a fraction of the IgG+ cells to be p-ASC for RF.
However, the possibility remains that many of the
synovial B cells are activated by relevant but as-yetunknown antigens, among which one may propose other
local autoantigens involved in the generation or perpetuation of RA.
ACKNOWLEDGMENTS
We appreciate the support of our clinical colleagues
Drs. L. Rabenseifner, E. M. Lemmel, and B. Lang, who
allowed us to study their patients. We thank Dr. H. Eibel for
critical reading of the manuscript and helpful suggestions. We
are especially grateful to all donors of blood and samples of
synovial fluid and tissue.
REFERENCES
1. Ollier W, Thomson W: Population genetics of rheumatoid arthritis. Rheum Dis Clin North A m 18:741-759, 1992
1418
2. Gregersen PK, Silver J, Winchester RJ: The shared epitope
hypothesis: an approach to understanding the molecular genetics
of susceptibility to rheumatoid arthritis. Arthritis Rheum 30:12051213, 1987
3. Evans TI, Han J, Singh R, Moxley G: The genotypic distribution of
shared-epitope DRBl alleles suggests a recessive mode of inheritance of the rheumatoid arthritis disease-susceptibility gene.
Arthritis Rheum 38:1754-1761, 1995
4. Panayi GS: The immunopathogenesis of rheumatoid arthritis. Br J
Rheumatol 32:4-14, 1993
5. Klareskog L, Ronnelid J, Holm G: Immunopathogenesis and
immunotherapy in rheumatoid arthritis: an area in transition.
J Intern Med 238:191-206, 1995
6. Res PC, Struijk L, Leow A, Daha MR, Breedveld FC: Inflamed
joints of patients with rheumatoid arthritis contain T cells that
display in vitro proliferation to antigens present in autologous
synovial fluid: functional analysis on the basis of synovial-fluidreactive T-cell clones and lines. Hum Immunol 40:291-298, 1994
7. Londei M, Savill CM, Verhoeff A, Brennan F, Leech ZA, Duance
V, Maini RN, Feldman M: Persistence of collagen type I1 specific
T cell clones in the synovial membrane of a patient with rheumatoid arthritis. Proc Natl Acad Sci U S A 86:636-640, 1989
8. Van Schooten WCA, Devereux D, Ho CH, Quan J, Aguilar BA,
Rust CJJ: Joint-derived T cells in rheumatoid arthritis react with
self-immunoglobulin heavy chains or immunoglobulin-binding
proteins that copurify with immunoglobulin. Eur J Immunol
24:93-98, 1994
9. Devereux DD, O’Hehir RE, McGuire J, van Schooten WCA,
Lamb JR: HLA-DR4/Dw4-restricted T cell recognition of self
antigen(s) in the rheumatoid synovial compartment. Int Immunol
7~635-640, 1991
10. Quayle AJ, Wilson KB, Li SG, Kjeldsen-Kragh J, Oftung F,
Shimick R, Sioud M, Farre 0, Capra JD, Natvig JB: Peptide
recognition, T cell receptor usage and HLA restriction elements of
human heat-shock protein (hsp) 60 and mycobacterial 65-kDa
hsp-reactive T cell clones from rheumatoid synovial fluid. Eur
J Immunol 22:1315-1322, 1992
11. JooB-Rudiger J, Daubenberger C, Lang B, Peter HH, Melchers I:
In rheumatoid arthritis a high proportion of synovial T cell clones
is autoreactive (abstract). Immunobiology 183:183, 1991
12. Struyk L, Hawes GE, Chatila MK, Breedveld FC, Kurnick JT, van
den Elsen PJ: T cell receptors in rheumatoid arthritis. Arthritis
Rheum 38:577-589, 1995
13. Melchers I, Peter HH, Eibel H: The T and B cell repertoire of
patients with rheumatoid arthritis. Scand J Rheumatol24:153-162,
1994
14. Pluschke G, Ricken G, Taube H, Kroninger S, Melchers I, Peter
HH, Eichmann K, Krawinkel U: Biased T cell receptor VLY
region
repertoire in the synovial fluid of rheumatoid arthritis patients.
Eur J Immunol21:2749-2754, 1991
15. Pluschke G, Ginter A, Taube H, Melchers I, Peter HH, Krawinkel
U: Analysis of T cell receptor V p regions expressed by rheumatoid
synovial T lymphocytes. Immunobiology 188:330-339, 1993
16. Natvig JB, Randen I, Thompson KM, Farre 0, Munthe E: The B
cell system in the rheumatoid inflammation: new insights into the
pathogenesis of rheumatoid arthritis using synovial B cell hybridoma clones. Springer Semin Immunopathol 11:301-313, 1989
17. Munthe E, Natvig JB: Immunoglobulin classes, subclasses and
complexes of IgG rheumatoid factor in rheumatoid plasma cells.
Clin Exp Immunol 1255-70, 1972
18. Petersen J, Heilmann C, Bjerrum OJ, Ingemann-Hansen T,
Halkjaer-Kristensen J: IgG rheumatoid factor-secreting lymphocytes in rheumatoid arthritis: evaluation of a hemolytic plaqueforming cell technique. Scand J Immunol 17:471-478, 1983
19. Randen I, Mellbye OJ, Farre 0, Natvig J: The identification of
germinal centers and follicular dendritic cell networks in rheumatoid synovial tissue. Scand J Immunol 41:481-486, 1995
RUDOLPH1 ET AL,
20. Youinou PY, Morrow JW, Lettin AWF, Lydyard PM, Roitt IM:
Specificity of plasma cells in the rheumatoid synovium. I. Immunoglobulin class of antiglobulin-producing cells. Scand J Immunol
20:307-315, 1984
21. Gause A, Gundlach K, Zdichavsky M, Jacobs G, Koch B, Hopf T,
Pfreundschuh M: The B lymphocyte in rheumatoid arthritis:
analysis of rearranged V K genes from B cells infiltrating the
synovial membrane. Eur J Immunol25:2775-2782, 1995
22. Schroder AE, Greiner A, Seyfert C, Berek C: Differentiation of B
cells in the nonlymphoid tissue of the synovial membrane of
patients with rheumatoid arthritis. Proc Natl Acad Sci U S A
93:221-225, 1996
23. Randen I, Brown D, Thompson KM, Hughes Jones N, Pascual V,
Victor K, Capra JD, Farre 0, Natvig JB: Clonally related IgM
rheumatoid factors undergo affinity maturation in the rheumatoid
synovial tissue. J Immunol 148:3296-3301, 1992
24. Egeland T, Lea T, Mellbye OJ, Pahle JA, Ottesen T, Natvig JB:
Quantitation of cells secreting immunoglobulins after elution from
rheumatoid synovial tissue. Scand J Immunol 16:413-419, 1982
25. Ronnelid J, Lysholm J, Engstrom-Laurent A, Klareskog L, Heyman B: Local anti-type I1 collagen antibody production in rheumatoid arthritis synovial fluid: evidence for an HLA-DR4restricted IgG response. Arthritis Rheum 37:1023-1029, 1994
26. Tarkowski A, Klareskog L, Carlsten H, Herberts P, Koopman WJ:
Secretion of antibodies to types I and 11 collagen by synovial tissue
cells in patients with rheumatoid arthritis. Arthritis Rheum 32:
1087-1092, 1989
27. Vaughan JH, Chihara T, Moore TL, Robbins DL, Tanimoto K,
Johnson JS, McMillan R: Rheumatoid factor-producing cells
detected by direct haemolytic plaque assay. J Clin Invest 58:933941, 1976
28. Rudolphi U, Hohlbaum A, Lang B, Peter HH, Melchers I: The B
cell repertoire of patients with rheumatoid arthritis: frequencies
and specificities of peripheral blood B cells reacting with human
IgG, human collagens, a mycobacterial heat shock protein and
other antigens. Clin Exp Immunol 92:404-411, 1993
29. Schwab J, Lukowsky A, Volk H-D, Peter HH, Melchers I:
Precursor frequencies for DNA-specific B lymphocytes in patients
with systemic lupus erythematosus (SLE). Clin Exp Immunol
96:450-457, 1994
30. Wen L, Hanvanich M, Werner-Favre C, Brouwers N, Perrin LH,
Zubler RH: Limiting dilution assay for human B cells based on
their activation by mutant EL-4 thymoma cells: total and antimalaria responder B cell frequencies. Eur J Immunol 17:887-892,
1987
31. Arnett FC, Edworthy SM, Bloch DA, McShane DJ, Fries JF,
Cooper NS, Healey LA, Kaplan SR, Liang MH, Luthra HS,
Medsger TA Jr, Mitchell DM, Neustadt DH, Pinals RS, Schaller
JG, Sharp JT, Wilder RL, Hunder GG: The American Rheumatism Association 1987 revised criteria for the classification of
rheumatoid arthritis. Arthritis Rheum 31:315-324, 1988
32. Becking A, Pluschke G, Krawinkel U, Melchers I, Peter HH, Lang
B: HLA-DRBl* gene sequences in HLA-DR4 positive and negative patients with rheumatoid arthritis. Eur J Immunogen 20:8389, 1993
33. Yan T, Burkhardt H, Ritter T, Broker B, Mann KH, Bertling WM,
von der Mark K, Emmrich F: Specificity and T cell receptor p
chain usage of a human collagen type 11-reactive T cell clone
derived from a healthy individual. Eur J Immunol 225-56, 1992
34. Koga T, Wand-Wurttenberger A, DeBruyn J, Munk ME, Schoel
B, Kaufmann SHE: T cells against a bacterial heat shock protein
recognize stressed macrophages. Science 245:1112-1115, 1989
35. Vogt A, Batsford S, Rodriguez-Iturbe B, Garcia R: Cationic
antigens in poststreptococcal glomerulonephritis. Clin Nephrol
20:271-279, 1983
36. Lefkovits I, Waldmann H: Limiting dilution analysis of cells in the
immune system. Cambridge, Cambridge University Press, 1979
SYNOVIAL B CELL REPERTOIRE IN RA
37. Natvig JB, Farre 0, Randen I, Steinitz M, Thompson K, Waalen
K B lymphocytes, B cell clones and rheumatoid factor antibodies
in rheumatoid inflammation. Scand J Rheumatol76:217-227,1988
38. Koopman WJ, Schrohenloher RE, Crago SS, Spalding DM,
Mestecky J: IgA rheumatoid factor synthesis by dissociated synovial cells: characterization and relationship to IgM rheumatoid
factor synthesis. Arthritis Rheum 28:1219-1227, 1985
39. Otten HG, Daha MR, Dolhain RJEM, de Rooy HH, Breedfeld
F C Rheumatoid factor production by mononuclear cells derived
from different sites of patients with rheumatoid arthritis. Clin Exp
Immunol 94:236-240, 1993
40. Youinou PY, Irving WL, Shipley M, Hayes J, Lydyard PM:
Evidence for B cell activation in patients with active rheumatoid
arthritis. Clin Exp Immunol 55:91-98, 1984
41. Chattopadhyay C, Chattopadhyay H, Natvig JB, Michaelsen TE,
Mellbye OJ: Lack of suppressor cell activity in rheumatoid synovial lymphocytes. Scand J Immunol 10:309-316, 1979
42. Vaughan JH: Pathogenic concepts and origins of rheumatoid
factor in rheumatoid arthritis. Arthritis Rheum 36:l-6, 1993
43. Egeland T, Lea T, Saari G, Mellbye OJ, Natvig JB: Quantitation of
1419
44.
45.
46.
47.
48.
cells secreting rheumatoid factor of IgG, IgA, and IgM class after
elution from rheumatoid synovial tissue. Arthritis Rheum 25:
1445-1450, 1982
McCulloch J, Lydyard PM, Rook GA: Rheumatoid arthritis: how
well do the theories fit the evidence? Clin Exp Immunol 92:l-6,
1993
Cohen IR: Autoimmunity to chaperonins in the pathogenesis of
arthritis and diabetes. Annu Rev Immunol 9:567-589, 1991
Robbins DL, Kenny TP, Snyder LL, Ermel RW, Larrick JW:
Allotypic dependency of the specificity and avidity of human
monoclonal IgM rheumatoid factors derived from rheumatoid
synovial cells. Arthritis Rheum 36:389-393, 1993
Morimoto RI, Tissikres A, Georgopoulos C, editors: The Biology
of Heat Shock Proteins and Molecular Chaperones. Cold Spring
Harbor, NY,Cold Spring Harbor Laboratory Press, 1994
Krenn V, Vollmers HP, von Landenberg P, Schmauer B, Rupp M,
Roggenkamp A, Miiller-Hermelink H K Immortalized Blymphocytes from rheumatoid synovial tissue show specificity for
bacterial HSP60. Virchows Arch 427:511-518, 1996
Документ
Категория
Без категории
Просмотров
8
Размер файла
989 Кб
Теги
frequencies, patients, typed, shock, tissue, human, fluid, cells, iga, increase, heat, repertoire, specific, protein, mycobacterium, arthritis, igg, synovial, collagen, rheumatoid
1/--страниц
Пожаловаться на содержимое документа