close

Вход

Забыли?

вход по аккаунту

?

IgA rheumatoid factor synthesis by dissociated synovial cells. Characterization and relationship to IgM rheumatoid factor synthesis

код для вставкиСкачать
1219
IgA RHEUMATOID FACTOR SYNTHESIS BY
DISSOCIATED SYNOVIAL CELLS
Characterization and Relationship to IgM Rheumatoid Factor Synthesis
WILLIAM J . KOOPMAN, RALPH E . SCHROHENLOHER, SYLVIA S. CRAGO,
DAVID M. SPALDING, and JIRl MESTECKY
We examined patterns of IgA rheumatoid factor
(RF) and IgM-RF synthesis by dissociated synovial cells
obtained from 27 patients with seropositive rheumatoid
arthritis. Synthesis of IgA-RF was observed in 19 of 34
synovial cell preparations from these patients and constituted a mean of 16% of the total IgA produced. IgARF expression correlated only weakly with IgM-RF
prodiiction (r = 0.385) and could be dissociated from
prodiiction of IgA-RF (and IgM-RF) exhibited by simultaneously obtained peripheral blood plasma cells. While
wide variations were observed in the ratio of IgARF:IgM-RF produced by synovial B cells in the patient
sample studied, remarkable consistency in the relationship of IgA-RF to IgM-RF synthesis was observed over
time in different joints of the same patient. IgA-RF
synthesized by dissociated synovial cells was predominantly of the IgAl subclass and existed in both monomeric and polymeric forms. Our results are compatible
From the Division of Clinical Immunology and Rheumatology and the Department of Microbiology, University of Alabama
at Birmingham, School of Medicine, and the Birmingham Veterans
Administration Hospital, Birmingham, Alabama.
Supported by USPHS grants AM-03555, AM-14807, Al18745, and AI-10854 and the Veterans Administration Research
Program.
William J. Koopman, MD: Professor of Medicine and
Director, Division of Clinical Immunology and Rheumatology;
Ralph E. Schrohenloher, PhD: Associate Professor of Medicine,
Division of Clinical Immunology and Rheumatology; Sylvia S .
Crago, PhD: Research Instructor, Department of Microbiology;
David M . Spalding, MD: Associate Professor of Medicine, Division
of Clinical Immunology and Rheumatology; Jiri Mestecky, MD:
Professor of Microbiology, Department of Microbiology.
Address reprint requests to William J. Koopman, MD,
Division of Clinical Immunology and Rheumatology, University of
Alabama at Birmingham, Birmingham, AL 35294.
Submitted for publication December 6 , 1984; accepted in
revised form May 28, 1985.
Arthritis and Rheumatism, Vol. 28, No. 11 (November 1985)
with the view that local production of IgA-RF and
IgM-RF are regulated independently of each other.
Since the initial descriptions of rheumatoid factors (RF) in the sera of patients with rheumatoid
arthritis (RA) and other diseases (1-3), considerable
effort has been directed toward determining the pathogenetic significance of RF. Evidence that both IgM-RF
and IgG-RF were capable of activating the complement cascade (4-6) suggested that these autoantibodies might contribute to tissue injury in diseases such as
RA. This view has been supported by considerable
circumstantial evidence, including the occurrence of
R F in immune complexes present in the sera (7),
synovial fluids (8), and synovial fluid phagocytes (9) of
RA patients; local production of R F in rheumatoid
synovium (lo), pleura ( I l ) , muscle (12), and pericardium (13); local consumption of complement in the
joint space of RA patients (14,15); and correlation of
factor 13 and C4 turnover rates with levels of IgM-RF
in the sera of RA patients (16).
,4lthough these lines of evidence supported a
role for R F in tissue injury associated with RA, it
should be stressed that previous studies focused entirely on IgM-RF and IgG-RF. It has become increasingly clear, however, that the R F response in RA is not
restricted to IgM and IgG classes. Indeed, the presence of IgA (17-20) and IgE (21) R F in the sera of RA
patients has been well documented. Moreover, in
the case of IgA-RF, striking variations (up to fiftyfold)
have also been observed in the ratios of IgA-RF:IgMR F concentrations present in individual RA sera (20),
suggesting that expression of these R F isotypes may
be under distinct regulatory controls, as has been
demonstrated in mice (22,23).
1220
KOOPMAN ET AL
In an effort to determine the contributions of
IgA-RF to tissue injury in RA patients, we have
undertaken studies of the properties of IgA-RF synthesized at local sites of tissue inflammation in RA and
the relationship of IgA-RF expression t o IgM-RF
production in these tissues. O u r results indicate that:
1) IgA-RF is frequently synthesized in the synovium of
RA patients; 2) local production of IgA-RF does not
correlate well with peripheral blood B cell synthesis of
IgA-RF; 3) IgA-RF a n d IgM-RF synthesis are only
weakly correlated at local sites of tissue injury; and 4)
IgA-RF produced by synovial cells is predominantly of
t h e IgAl subclass a n d exists in both monomeric and
polymeric forms.
PATIENTS AND METHODS
Patients. Rheumatoid synovial tissue (total of 34
samples from the knee, hip, or wrist) was obtained at the
time of surgery (synovectomy or joint replacement) from 27
patients who fulfilled the American Rheumatism Association
criteria for definite or classic RA (24). All patients were
seropositive, as defined by a serum latex fixation titer (25) of
at least 1 :320. The mean age of the patients was 57 (range
.33-77).
Enzymatic dissociation and culture of rheumatoid
synovial cells. Synovial tissue was placed directly into chilled
Dulbecco's saline in the operating room and was transported
to the laboratory for immediate processing. Freshly obtained
rheumatoid synovial tissue was dissected free of capsular
and cartilaginous tissue and divided into l-2-mm3 fragments.
The fragments were washed exhaustively in chilled minimal
essential medium (MEM), and the wet weight of the tissue
was determined.
A modification of previously described methods for
the preparation of single cell suspensions from murine Peyer's patch (26) was utilized to obtain dissociated synovial
cells. Synovial tissue fragments were placed into Joklik's
modified MEM containing Dispase (1 mg/ml, grade 11;
Boehringer Mannheim Biochemicals, Indianapolis, IN) at
37°C in a beaker or tissue culture flask, with constant
agitation (10 gm of tissue per 50 ml of enzyme solution).
After 30 minutes, medium containing dissociated synovial
cells was removed by gentle aspiration and was placed into
an equivalent volume of RPMI 1640 containing 5% fetal calf
serum (FCS). Fresh prewarmed enzyme solution was immediately added to the partially extracted synovial tissue fragments, and the extraction procedure was repeated until
sequential cell counts indicated insignificant cell yields from
the residual synovial tissue (generally, 4-5 extractions were
performed). Dissociated synovial cells were washed 5-6
times with RPMI 1640 containing 10% FCS, 2 mM
glutamine, and 20 pg/ml gentamycin. Cell viability (trypan
blue exclusion) consistently exceeded 90% prior to culture,
and cell yields varied from 0.4-21 x lo6 cells/gm of tissue
(mean 5.5 x lo6 ceIIs/gm of tissue).
Synovial cells were cultured (1 ml, 1 x lo6 cells/ml)
in complete medium in the absence of mitogens at 37°C in a
5% C 0 2 and air atmosphere for 7 days. Culture supernatants
were harvested by centrifugation at the termination of culture and stored at -20°C until assayed.
The fetal calf serum lot used in these studies was
selected because it exhibited minimal mitogenic activity for
murine B cells and was associated with low levels of in vitro
spontaneous IgM production by peripheral blood mononuclear cells (PBMC) from healthy adult donors.
Preparation and culture of PBMC. Peripheral blood
samples were obtained from 20 of the 27 RA patients undergoing joint surgery. PBMC were prepared and cultured in
complete medium in the absence of mitogens as previously
described (27).
Radioimmunoassay (RIA) of IgA-RF. Culture supernatants were assayed for IgA-RF utilizing the quantitative
solid-phase RIA previously described (20).
RIA of IgM-RF. The procedure utilized for quantitation of IgM-RF was identical to that previously described
(28), except that polystyrene microtiter wells were utilized
as described above for the IgA-RF RIA (20).
RIA of IgA-RF subclasses. In order to determine the
subclass distribution of secreted IgA-RF, an RIA was developed utilizing affinity-purified monoclonal anti-IgA subclass
antibodies (a kind gift from Dr. Mary Ellen Conley,
Children's Hospital, Philadelphia, PA). The production and
characterization of the monoclonal anti-IgA1 and anti-IgA2
antibodies (clones 1-155-1 and 14-3-26, respectively) have
been previously described (29). The goat anti-human IgA
utilized was produced by immunizing a goat with a highly
purified LgAlK protein. Antibodies to the kappa light chain
and contaminating immunoglobulins were removed by sequential absorption over a series of cyanogen bromide-activated Sepharose 4B columns to which Cohn fraction 11,
monoclonal IgGK, and monoclonal IgMK proteins had been
covalently coupled. The anti-IgA antibody was then isolated
by affinity chromatography using a column to which a
purified IgA2 protein had been covalently bound, to ensure
that the eluted anti-IgA had no subclass specificity.
For the RIA, aliquots of an isolated monoclonal
IgAI-RF standard (20,30) (dose range 0.25-25 ng/well) or
test samples were incubated in 2 parallel sets of polystyrene
microtiter wells coated with human IgG or with bovine
serum albumin (BSA) as a control for nonspecific binding.
After 18-20 hours, the wells were washed as previously
described for the IgA-RF assay (20). Ten nanograms of
'2SI-labeled affinity-purified anti-IgA was then added to the
first set of wells, and 10 ng of '2SI-labeled anti-IgA1 was
added to the second set. Standard curves were then constructed for total IgA-RF (the wells receiving '251-labeled
anti-IgA) and IgA 1-RF (the wells receiving '251-labeIed antiIgAl) in each experiment by plotting mean counts per
minute of '251-labeled anti-IgA (or anti-IgAl) bound per
IgG-coated well minus the corresponding BSA control mean
for each input of IgA-RF. Quantitative values for unknowns
were calculated by determining the mean cpm of '2sI-labeled
anti-IgA (or anti-IgAl) bound (IgG well minus BSA well
mean) for each dilution of unknown and then referring to the
appropriate standard curve.
Although no IgA2-RF standard was available,
semiquantitative results obtained for uptake of '251-labeled
IgA-RF SYNTHESIS
anti-lgA2 were in good agreement with the data obtained for
the IgA-RF and IgAl-RF assays (i.e., relative binding of
'2SI-labeledanti-IgA2 was inversely related to the calculated
proportion of IgAl-RF present). All assays were performed
in duplicate or triplicate.
RIA of IgM and IgA. As previously described, total
IgM and IgA in culture supernatants were quantitated by
solid-phase RIA (3 I ) .
Identification of monomeric and polymeric forms of
IgA-RF. Two synovial cell culture supernatants were concentrated approximately tenfold (Amicon C-25 ultrafiltration
membrane cone) and 0.3 ml was layered on a 4.7-ml linear
12-30% sucrose gradient prepared in 0.1M sodium acetate
buffer, pH 4.1. The gradients were centrifuged at 37,500
revolutions per minute for 15.5 hours at 2°C in an SW-50.1
rotor, and fractions (0.25 ml) were collected by flotation.
Aliquots of each fraction were neutralized in 0.2M Na2HP04
and assayed for monomeric and polymeric IgA-RF as previously described (32).
Briefly, neutralized fractions were diluted in either
0.15J4 NaCl buffered at pH 7.4 by 0.02M sodium phosphate
(PBS) containing 1% BSA (PBS-BSA), or in PBS-BSA with
0.0556 murine ascites containing monoclonal anti-human
a-chain antibody. Duplicate 0.25-ml aliquots of each dilution
were placed in IgG-coated wells and incubated at room
temperature for 16 hours. The wells were then aspirated,
washed 3 times with 0.5% Tween 80 in PBS, and bound IgA
was detected by adding 0.25 ml of '*'I-labeled affinityisolal ed goat anti-human IgA as described previously for the
IgA-IIF RIA (20). Dilutions of isolated monomeric and
polymeric IgA-RF standards (32) (2-250 ng/ml) and controls
of PEG-BSA and the 0.05% murine ascites fluid alone were
included in each assay.
The quantity of monomeric IgA-RF was estimated
from the difference between '251-labeled anti-IgA bound by
wells incubated with the sample diluted in the 0.05% murine
ascites fluid reagent and that bound by wells incubated with
the sample diluted in PBS-BSA alone by comparison to a
curve constructed from the monomeric IgA-RF standard.
The quantity of polymeric IgA-RF was deduced from the
'*'I-liibeled anti-IgA bound by wells incubated with the
sample diluted in PBS-BSA alone by comparison with a
curvt: constructed from the polymeric IgA-RF standard. For
each gradient run, aliquots of isolated monomeric and polymeric: IgA-RF standards were run in parallel gradient tubes
and were assayed simultaneously with the experimental
t ubea .
Immunofluorescence analysis. Fixed cytocentrifuge
preparations of dissociated synovial cells were examined for
the presence of cytoplasmic immunoglobulins (Ig), IgA,
IgAl and IgA2 subclasses, IgM, and RF (staining with
heat-aggregated human IgG [33]).
In initial experiments, slides were incubated with
fluorescein isothiocyanate (F1TC)-labeled polyvalent antihuman Ig (heavy and light chain specific; Behring Diagnostics, Calbiochem, Davis, CA) for 45 minutes at room temperature and washed for 2 hours in PBS at 4°C. Preparations
containing Ig-staining cells were then subsequently analyzed
for the presence of IgA and the distribution of IgAl and IgA2
subclasses. Slides were incubated with unlabeled monoclonal anti-human IgAl (29) or monoclonal anti-IgA2 (29)
1221
followed by tetramethylrhodamine isothiocyanate (TR1TC)labeled, affinity-purified goat anti-mouse IgG (Kirkegaard
and Perry Laboratories, Gaithersburg, MD). After washing,
slides were incubated with FITC-labeled polyvalent rabbit
anti-human IgA (heavy chain specific).
To determine the presence of cytoplasmic RF activity, synovial cell preparations were incubated with TRITClabeled heat-aggregated normal human IgG (33). Synovial
cell samples containing cells which stained with TRITClabeled heat-aggregated human IgG were then examined to
determine the isotype distribution of the RF by co-staining
with FITC-labeled polyvalent goat anti-human IgA or IgM
(heavy chain specific). The subclass distribution of IgA-RF
was determined by co-staining with TRITC-labeled heataggregated IgG and unlabeled monoclonal anti-IgA I or IgA2,
followed by FITC-labeled goat anti-mouse IgG. Controls
consisting of murine myeloma proteins of the relevant
isotype and subclass for the anti-al (IgC3) and anti-a2
(IgG2b) hybridomas did not result in positive staining of
selected synovial samples.
Preparation of immunoglobulins and antisera,
fluorochrome-labeling techniques, and specificity controls
for polyclonal reagents have been previously described (34).
TRITC-labeled heat-aggregated IgG did not bind to fixed
normal human PBMC, PBMC stimulated with pokeweed
mitogen, or synovial cell preparations which lacked Igpositive cells. TRITC-labeled anti-mouse IgG did not stain
synovial cells, as was indicated by failure to stain without
having first preincubated the cells with monoclonal anti-IgA 1
or IgA2 reagents. The specificity of monoclonal reagents was
demonstrated by their capacity to selectively block the
immunoprecipitation of radiolabeled myeloma proteins of
relevant subclass specificity.
RESULTS
Production of IgM-RF and IgA-RF by dissociated synovial cells. Synovial cells obtained by enzymatic dissociation of synovium taken from 27 patients
with se:ropositive RA (34 samples) were examined for
their capacity to elaborate IgM-RF and IgA-RF in
vitro. As shown in Figure 1, appreciable amounts of
IgM-RF and lesser quantities of IgA-RF were synthesized. IgM-RF synthesis was observed with 30 of 34
samples (23 of 27 patients). Lack of IgM-RF production was confined to synovia with low cell yields and
was presumably indicative of inactive or "burned out"
disease.
Detectable levels of IgA-RF production occurred in 19 of 34 synovial cell preparations (12 of 27
patients), and IgA-RF production was not observed in
the absence of IgM-RF secretion. De novo synthesis
of R F was documented by the demonstration that
cycloheximide ( W 3 M ) consistently inhibited elaboration of both IgM-RF and IgA-RF (>80% suppression)
by the synovial cell cultures. IgM-RF constituted a
1222
-cn
n
I
Q)
0
6
r
a
a
c
Y
a
KOOPMAN ET AL
ij401
1000
800
600
400
200
80
6o
t
- 800
- 600
- 400
. ' . .
- 200
- 100
-80
-60
-40
.
=
8
6
8
'
0
C
-20
-10
-a
2:
a
a
gJ
IgM RF (ng/lO@ceIls)
Figure 2. Relationship between IgA rheumatoid factor (RF) and
IgM RF synthesis by dissociated synovial cells. Individual points
represent the quantity of IgA RF and IgM R F synthesized by each
of the 35 synovial cell preparations analyzed.
- 6
- 4
20
(10
'.$
LAasid
Figure 1. Synthesis of IgM rheumatoid factor (RF) (left panel) and
1gA RF (right panel) by dissociated synovial cells obtained from
patients with seropositive rheumatoid arthritis. Lines connect points
obtained for the same synovial cell culture supernatant. 0 = values
beneath the level of detection in the RF assay.
that 2 of the 3 synovial cell preparations which elaborated the highest levels of IgM-RF (7,840 ng/106 cells
and 3,000 ng/106cells, respectively) did not synthesize
detectable quantities of IgA-RF.
Since 5 of the 27 patients in this study underwent multiple joint reconstructive surgery, we inves-
Table 1. Comparable ratios of IgA-RF:IgM-RF, synthesized over
time in different involved joints of individual rheumatoid arthritis
patients*
Patient,
sample
no.
significant fraction of the total IgM produced by RA
synovial cells (mean 2 1 SD, 43 & 26%), while IgA-RF
constituted a lesser portion of the total IgA (16 ?
13%).
Relationship between IgM-RF and IgA-RF synthesis by dissociated synovial cells. We previously reported considerable variation in the ratio of IgM-RF to
IgA-RF levels present in individual RA sera (20). It
was therefore of interest to examine the relationship
between IgA-RF and IgM-RF synthesis by RA synovial cell cultures. As shown in Figure 2, there was a
weak correlation between absolute levels of IgM-RF
and IgA-RF synthesized by dissociated synovial cells
(r = 0.385; P < 0.05). It is noteworthy in this regard
SL
1
2
3
LW
1
2
3
AC
I
2
MS
1
2
IgA-RF
(np/lOb cells)
IgM-RF
(ng/106 cells)
IgA-RF:IgM-RF
45
74
72
605
1.800
792
0.07
75
144
I0
208
252
0.36
0.57
0.66
540
792
2,232
0.35
228
120
1,666
540
0.14
0.22
8
12
30
24
0.27
0.50
106
5
,ooo
0.04
0.09
0.11
GM
1
2
* RF = rheumatoid factor.
IgA-RF SYNTHESIS
1223
Table 2. Determination of subclass distribution of IgA rheumatoid factor (RF) plasma cells in
dissociated synovial cell preparations, by immunofluorescence
Patient
SL#
MS
RL
AC
SL#
% cells binding
heat-aggregated
human IgGt
% Ig-positive
cells*
10.8
5.1
3.7
23.0
5.2
1.9
1.o
1.7
3.6
4.3
Heat-aggregated IgG-binding cells
Bb IgAS
% IgAlE)
5% IgA2T
5.3
6.1
19.8
41.1
13.6
5.3
6.1
11.4
21.5
1.7
0.0
0.0
5.1
0.0
2.8
* Cells exhibiting cytoplasmic staining with fluorescein isothiocyanate (FITC)-labeled anti-human Ig.
t Cells exhibiting cytoplasmic staining with tetramethylrhodamine isothiocyanate (TR1TC)-labeled
heat-aggregated human IgG.
$ Cells staining with TRITC-labeled heat-aggregated human IgG and with FITC-labeled anti-human
IgA.
6 Cells staining with TRITC-labeled heat-aggregated human IgCi and with FITC-labeled anti-mouse Ig
(after preincubation of cells with monoclonal anti-IgAl).
ll Cells staining with TRITC-labeled heat-aggregated human IgG and with FITC-labeled anti-mouse Ig
(after preincubation of cells with monoclonal anti-IgA2).
# Performed on 2 separate occasions with different synovial cell tissue.
tigated whether patterns of IgA-RF and IgM-RF production were comparable in the synovium of different
joints from the same patient. Synovial specimens were
obtained at separate times, over a period of 18 months,
from these 5 patients (3 samples from 2 patients; 2
samples from each of the other 3). As shown in Table
1, there was a remarkably consistent relationship
between IgA-RF and IgM-RF synthesis in each of the
5 patients serially investigated (less than a 3.5-fold
variation in the ratio of IgA-RF:IgM-RF produced in
the same patient) compared with a 2,000-fold variation
observed within the cross-sectional sample of 27 patients studied (range of IgA-RF:IgM-RF, 0.0005-1.O;
mean ? 1 SD, 0.22 ? 0.22).
Characterization of IgA-RF secreted by dissociated synovial cells. Studies were undertaken to determine the subclass distribution and physical characteristics of IgA-RF synthesized by RA synovial cells.
Two independent approaches were used to analyze the
LL-
a
h
-
0.4
1
I
O3 60.3
Figure 3. Proportion of IgA rheumatoid factor (RF) present as IgAl
RF. Culture supernatants from 7 synovial cell preparations were
assayed for IgA R F and IgAl RF. Individual points represent the
ratio of IgAl R F to total IgA R F measured for each supernatant.
subclass distribution of IgA-RF produced by dissociated synovial cells. Immunofluorescence studies of
fixed cytocentrifuge preparations of synovial cells
verified the presence of plasma cells containing IgAR F in 6 of 9 samples tested. There was good agreement
between the immunofluorescence data and the radio-
A
;
FRACTION NUMBER
, s
2
4
(I
8 1 0 1 2
2
4
6
(I
1
10 12 14
FRACTION NUMBER
Figure 4. Density gradient ultracentrifugation of concentrated
synovial cell culture supernatants from rheumatoid arthritis patients
MS (A, upper graph) and LW (B, upper graph), demonstrating the
presence of both monomeric and polymeric IgA rheumatoid factor
(RF). Fractions (0.25 ml) were collected by flotation, and the
concentrations of monomeric IgA R F (0-0)
and polymeric IgA
R F (0--0)
were determined by solid-phase radioimmunoassay.
Monomeric and polymeric IgA R F controls, isolated from patient
PS (29,30), were centrifuged on parallel gradients and assayed
simultaneously. The results are shown in the bottom graphs.
KOOPMAN ET AL
1224
Table 3. Comparison of rheumatoid factor (RF) production by dissociated synovial cells and
simultaneously obtained peripheral blood mononuclear cell (PBMC) cultures*
~~
Source
IgM-RF
(&lo6
cells)
IgM
(ng/106 cells)
14 2 14
276
PBMC
(n = 20)
Synovial cells 940 2 1,827 3,079
(n = 20)
IgM-RF:IgM
?
656
0.14
2
6,962 0.40 2 0.27t
5
O.llt
IgA-RF
(ng/lOh
cells)
626
129 2 235
IgA
(ng/iO"
cells)
* 329
947 * 2,076
298
IgA-RF:IgA
0.04 2 0.06$
0.19
* 0.14$
* Values are mean
tP
2 SD.
by Wilcoxon signed rank test.
0.005 by Wilcoxon signed rank test.
= 0.002
$P =
immunoassay results obtained for the 9 specimens, as
only the 6 samples with IgA-RF-containing plasma
cells secreted detectable IgA-RF.
For each of the 5 synovial samples examined
with monoclonal anti-IgA subclass antibodies, IgAl
represented the predominant subclass of plasma cells
containing IgA-RF (Table 2). That IgAl was indeed
the predominant subclass of IgA-RF secreted by
synovial cells was verified by RIA of selected synovial
cell culture supernatants. As shown in Figure 3 , IgAlR F accounted for 84 t 20% of the total IgA-RF
synthesized by the synovial cell preparations tested.
In view of evidence that IgA-RF can exist in
both polymeric and monomeric forms (32), it was of
interest to determine the size characteristics of IgAR F secreted by synovial plasma cells. Synovial cell
culture supernatants from 2 patients were concentrated by ultrafiltration, fractionated by sucrose density ultracentrifugation at acid pH, and fractions were
assayed for polymeric and monomeric IgA-RF. As
shown in Figure 4, monomeric and polymeric forms of
IgA-RF were detected in both synovial cell culture
supernatants.
Comparison between patterns of RF production
by dissociated synovial cells and paired PBMC cultures.
In order to define the relationship between IgA-RF and
IgM-RF synthesis by dissociated synovial cells and
PBMC, we obtained paired samples of synovium and
peripheral blood from 20 patients. The resulting cell
preparations were cultured under identical conditions,
and the culture supernatants were then assayed simultaneously for R F and immunoglobulin content. Levels
of spontaneous IgM, IgA, IgA-RF, and IgM-RF production by rheumatoid synovial cells exceeded that
observed in paired PBMC cultures (Table 3 ) . Moreover, both IgA-RF and IgM-RF represented a substantially greater fraction of total IgA and IgM synthesized
in the synovium than in paired PBMC cultures, indicating preferential expression of these 2 R F isotypes
by synovial plasma cells. Indeed, production of IgAR F (and of IgM-RF) was frequently observed in dissociated synovial cell cultures in the absence of detectable levels of spontaneous R F production by
PBMC; Table 4 illustrates this for 2 representative
patients.
DISCUSSION
Our results indicate that IgA-RF is frequently
synthesized in the synovium of patients with RA and
that levels of IgA-RF synthesis are only weakly correlated with IgM-RF production. These data are in
agreement with those of Egeland and colleagues ( 3 3 ,
who observed IgA-RF plaque-forming cells (PFC) in
Table 4. Dissociation of spontaneous IgA rheumatoid factor (RF) and IgM-RF synthesis in paired
cultures of peripheral blood mononuclear cells and synovial cells
~
Patient,
cell source
LW
Blood
Synovium
GP
Blood
Synovium
-~
IgA-RF
IgA
IgM-RF
IgM
(ng/106 cells) (ng/106 cells) IgA-RF:IgA (ng/106 cells) (ng/lOh cells) IgM-RF:IgM
<4
75
222
450
(0.018
0.167
<lo
208
270
540
<0.037
0.385
<4
55
234
360
<0.017
0.153
25
900
630
1,860
0.039
0.484
IgA-IRF SYNTHESIS
the synovia of 7 of 11 seropositive RA patients studied. Moreover, those workers also did not observe a
close relationship between numbers of IgM-RF and
IgA-RF PFC. We found that IgA-RF constituted a
significant fraction of the total IgA synthesized in the
synovia of the patients in this study. Of interest was
the evidence that local production of IgA-RF (and
IgM-KF) frequently occurred in the absence of demonstrable synthesis of IgA-RF (or IgM-RF) by simultaneously obtained peripheral blood B cells. Furthermore, both IgA-RF and IgM-RF constituted a significantly higher proportion of total isotype synthesis in
the synovium compared with peripheral blood.
These results clearly indicate that R F B cells
are preferentially activated in rheumatoid synovium
and are in agreement with the results of earlier work
by Sniiley and colleagues (10). Taken in conjunction
with our previous observations that local synthesis of
R F in inflamed rheumatoid pleura ( l l ) , muscle (12),
and pericardium (13) can be dissociated from R F
synthesis by blood B cells, these results suggest that
RA can be considered a polycentric disease in which
immune events at local sites of tissue inflammation are
autonlamously regulated.
Multiple synovial samples from different joints
were obtained over a period of several months from 5
RA patients and were compared with regard to patterns of IgA-RF and IgM-RF production. We found
that the relationship of IgA-RF to IgM-RF synthesis in
the same patient was much less variable (<3.5-fold)
than that observed in the entire cross-sectional sample
examined (-2,000-fold). Our results suggest that distinct regulatory pathways govern expression of IgAR F arid IgM-RF, as previously demonstrated in mice
(22,23).
Earlier studies in which IgA-RF was identified
in the saliva of patients with RA as well as in patients
with Sjogren’s syndrome (SS) (17-19), in addition to
evidence that cells secreting IgA-RF in 129/SV mice
are pirimarily located in lymph nodes draining the
intestinal tract (36), suggested a mucosal origin for
IgA-RF. Additional observations that serum IgA-RF
from i i series of patients with S S was predominantly
polymeric were advanced as evidence that these RFs
might be of mucosal origin (37). In view of these
findings, it was of interest to examine the properties of
IgA-RF synthesized by synovial B cells. Our results
clearly demonstrate that IgA-RF is predominantly of
the IgAl subclass, similar to the distribution of IgA
subclames observed in the serum (38-41) but different
from secretions in which approximately equal amounts
1225
of IgAl and lgA2 occur (39,40). Fractionation of
synovial cell culture supernatants from 2 RA patients
indicated that synovial B cells are capable of elaborating both polymeric and monomeric forms of IgA-RF.
Further studies are required to define the relationship
between synthesis of these 2 forms of IgA-RF. Nonetheless, our results demonstrate that synthesis of
polymeric IgA-RF can occur at nonmucosal sites, a
finding in agreement with a recent report establishing
the presence of polymeric IgA-RF in the sera of
patients with subacute bacterial endocarditis (42).
The significance of IgA-RF in the pathogenesis
of tissue injury in RA remains speculative. The prevalence of IgA-RF and the marked variability in its
relationship to IgM-RF levels, however, justify more
intensive efforts to define the biologic properties of
both the monomeric and polymeric forms of IgA-RF.
Observations that IgA is a weak activator of the
complement system (43) and that polymeric IgA may
facilitate antigen clearance through the biliary tract
(44) are of considerable interest in this regard.
In summary, local synthesis of IgA-RF frequently occurs in the synovia of RA patients and is
only weakly correlated with IgM-RF production. Patterns of local production of both IgA-RF and IgM-RF
in the synovium bear little relationship to those exhibited by peripheral blood B cells. Locally synthesized
IgA-RF is predominantly of the IgAl subclass and
exists in both polymeric and monomeric forms. Additional studies are required to determine the role(s) of
IgA-RF in the pathogenesis of tissue injury in RA.
ACKNOWLEDGMENTS
We wish to express our appreciation for the skillful
technical assistance of Cindy Moore and Jennifer Donoghue.
We are grateful to Dr. Mary Ellen Conley for generously
providing monoclonal anti-IgA1 and anti-IgA2 antibodies
and to the Hybridoma Core Facility of the Multipurpose
Arthritis Center for providing the monoclonal antiisotype
reagents used in this study. We also thank Marie Mays and
Brenda Ciilmore for expert processing of the manuscript.
REFERENCES
1 . Waaler E: On the occurrence of a factor in human serum
activating the specific agglutination of sheep blood corpuscles. Acta Pathol Microbiol Scand 17: 172-188, 1940
2 . Rose HM, Ragan C, Pearce E, Lipman MO: Differential
agglutination of normal and sensitized sheep erythrocytes by sera of patients with rheumatoid arthritis. Proc
SOCExp Biol Med 68:l-6, 1948
3 . Koopman WJ, Schrohenloher RE: Rheumatoid factor,
KOOPMAN ET AL
Rheumatoid Arthritis. Edited by PD Utsinger, NJ
Zvaifler, GE Ehrlich. Philadelphia, JB Lippincott, 1985,
pp 217-241
4. Zvaifler NJ, Schur P: Reactions of aggregated mercaptoethanol treated gamma globulin with rheumatoid factor-precipitin and complement fixation studies. Arthritis Rheum 11:523-536, 1968
5 . Sabhanval UK, Vaughan JH, Fong S, Bennett PH,
Carson DA, Curd JG: Activation of the classical pathway of complement by rheumatoid factors: assessment
by radioimmunoassay for C4. Arthritis Rheum 25: 161167, 1982
6. Brown PB, Nardella FA, Mannik M: Human complement activation by self-associated IgG rheumatoid factors. Arthritis Rheum 25: 1101-1 107, 1982
7. Franklin EC, Holman H, Miiller-Eberhard HJ, Kunkel
HG: An unusual protein component of high molecular
weight in the serum of certain patients with rheumatoid
arthritis. J Exp Med 105:425-438, 1957
8. Winchester RJ, Agnello V, Kunkel HG: Gamma globulin complexes in synovial fluids of patients with rheumatoid arthritis: partial characterization and relationship to
lowered complement levels. Clin Exp Immunol 6:689706, 1970
9. Vaughan JH, Jacox RF, Noell P: Relation of intracytoplasmic inclusions in joint fluid leukocytes to a n t i - 6
globulins. Arthritis Rheum 11: 135-144, 1968
10. Smiley JD, Sachs C, Ziff M: In vitro synthesis of
immunoglobulin by rheumatoid synovial membrane. J
Clin Invest 47:624-632, 1968
11. Halla JT, Koopman WJ, Schrohenloher RE, Darby WL,
Heck LW: Local synthesis of IgM and IgM rheumatoid
factor in rheumatoid pleuritis. J Rheumatol 10:204-209,
1
am
I/"J
12. Halla JT, Koopman WJ, Fallahi S, Oh SJ, Gay RE,
Schrohenloher RE: Rheumatoid myositis: clinical and
histologic features and possible pathogenesis. Arthritis
Rheum 27:737-743, 1984
13. Spalding DM, Haber P, Schrohenloher RE, Koopman
WJ: Production of immunoglobulin and rheumatoid factor by lymphoid cells in rheumatoid pericardium. Arthritis Rheum 28: 1071-1074, 1985
14. Pekin TJ, Zvaifler NJ: Hemolytic complement in
synovial fluid. J Clin Invest 43:1372-1382, 1964
15. Ruddy S, Austen KF: The complement system in rheumatoid synovitis. I. An analysis of complement component activities in rheumatoid synovial fluids. Arthritis
Rheum 13:713-723, 1970
16. Kaplan RA, Curd JG, Deheer DH, Carson DA, Pangburn MK, Miiller-Eberhard HJ, Vaughan JH: Metabolism of C4 and factor B in rheumatoid arthritis: relation
to rheumatoid factor. Arthritis Rheum 23:911-920, 1980
17. Heimer R, Levin FM: On the distribution of rheumatoid
factors among the immunoglobulins. Immunochemistry
3:l-10, 1966
18. Abraham GN, Santucci E, Jacox RF: Evidence for a
common synthetic arigin for secretory and serum IgA.
Adv Exp Med Biol45:233-240, 1974
19. Dunne JV, Carson DA, Spiegelberg HL, Alspaugh MA,
Vaughan JH: IgA rheumatoid factor in the sera and
saliva of patients with rheumatoid arthritis and Sjogren's
syndrome. Ann Rheum Dis 38: 161-170, 1979
20. Koopman WJ, Schrohenloher RE, Solomon A: A quantitative assay for IgA rheumatoid factor. J Immunol
Methods 50:89-96, 1982
21. Zuraw BL, O'Hair CH, Vaughan JH, Mathison DA,
Curd JG, Katz DH: Immunoglobulin E rheumatoid
factor in the serum of patients with rheumatoid arthritis,
asthma, and other diseases. J Clin Invest 68: 161G.1614,
1981
22. Van Snick JL, Masson PL: Incidence and specificities of
IgA and IgM anti-IgG autoantibodies in various mouse
strains and colonies. J Exp Med 152:45-52, 1980
23. Van h i c k JL: A gene linked to the IgH-C locus controls
the production of rheumatoid factor in the mouse. J Exp
Med 153:736742, 1981
24. Ropes MW, Bennett GA, Cobb S, Jacox R, Jessar RA:
1958 revision of diagnostic
criteria for rheumatoid arthritis. Bull Rheum Dis 9:175-176, 1958
25. Singer JM, Plotz CM: The latex fixation test. I. Application to the serological diagnosis of rheumatoid arthritis. Am J Med 21:888-892, 1956
26. Frangakis MF, Koopman WJ, Kiyono H , Michalek SM,
McGhee JR: An enzymatic method for production of
murine Peyer's patch cells enriched for macrophages. J
Immunol Methods 48:3344, 1982
27. Koopman WJ, Schrohenloher RE: In vitro synthesis of
IgM rheumatoid factor synthesis by lymphocytes from
healthy adults. J Immunol 125:934-939, 1980
28. Koopman WJ, Schrohenloher RE: A sensitive radioimmunoassay for quantitation of IgM rheumatoid factor.
Arthritis Rheum 23:302-308, 1980
29. Conley ME, Kearney JF, Lawton AR, Cooper MD:
Differentiation of human B cells expressing the IgA
subclasses as demonstrated by monoclonal hybridoma
antibodies. J Immunol 125:2311-2316, 1980
30. Kosaka M, Solomon A: Hyperviscosity syndrome associated with an idiopathic monoclonal IgA rheumatoid
factor. Am J Med 69:145-154, 1980
31. Koopman WJ, Miller RK, Crago SS, Mestecky J,
Schrohenloher RE: IgA rheumatoid factor: evidence for
independent expression at local sites of tissue inflammation. Ann NY Acad Sci 409:258-272, 1983
32. Schrohenloher RE, Koopman WJ, Moldoveanu Z,
Solomon A: Activity of rheumatoid factors of different
molecular sizes: comparison of autologous monomeric
and polymeric monoclonal IgA rheumatoid factors. J
Immunol 134: 1469-1474, 1985
33. Mellors RC, Heimer R, Corcos J, Korngold L: Cellular
IgA-RF SYNTHESIS
origin of rheumatoid factor. J Exp Med 110:875-886,
1227
39. Grey HM, Abel CA, Yount WJ, Kunkel HG: A subclass
1959
of human A globulins (yA2) which lacks the disulfide
bonds linking heavy and light chains. J Exp Med
34. Moro I, Umemura S, Crago S S , Mestecky J: Im-
35.
36.
37.
38.
munohistochemical distribution of immunoglobulins,
lactoferrin, and lysozyme in human minor salivary
glands. J Oral Pathol 13:97-104, 1984
Egeland T, Lea T, Saari G, Mellbye OJ, Natvig JB:
Quantitation of cells secreting rheumatoid factor of IgG,
IgA, and IgM class after elution from rheumatoid
synovial tissue. Arthritis Rheum 25: 1445-1450, 1982
Van Snick JL: The production of anti-IgG2a autoantilbody in the 12YlSV mouse: onset in lymph nodes
draining the intestinal tract and prevention by neonatal
thiymectomy. J Immunol 126:815-818, 1981
Elkon KB, Delacroix DL, Gharavi AE, Vaerman JP,
Hughes GRV: Immunoglobulin A and polymeric IgA
rheumatoid factors in systemic sicca syndrome: partial
characterization. J Immunol 129:576581, 1982
Vaerman JP, Heremans JF, Laurel1 CB: Distribution of
chain subclasses in normal and pathological IgAglobulins. Immunology 14:425-432, 1968
128: 1223-1236, 1968
40. Delacroix DL, Dive C, Rambaud JC, Vaerman JP: IgA
41.
42.
43.
44.
subclasses in various secretions and in serum. Immunology 47~383-385, 1982
Conley ME, Koopman WJ: Serum IgAl and IgA2 in
normal adults and patients with systemic lupus erythematosus and hepatic disease. Clin Immunol Immunopathol 16:39&397, 1983
Elkon KB, Inman RD, Culhane L , Christian CL: Induction of polymeric IgA rheumatoid factor in infective
endocarditis. Am J Med 75:785-78Y, 1983
Pfaffenbach G, Lamm ME, Gigli I: Activation of the
guinea pig alternative complement pathway by mouse
IgA immune complexes. J Exp Med 155:231-247, 1982
Brown TA, Russell MD, Mestecky J: Hepatobiliary
transport of IgA immune complexes: molecular and
cellular aspects. J Immunol 128:2183-2186, 1982
Документ
Категория
Без категории
Просмотров
0
Размер файла
819 Кб
Теги
factors, iga, synthesis, igm, characterization, dissociated, synovial, relationships, rheumatoid, cells
1/--страниц
Пожаловаться на содержимое документа