- Email: [email protected]
Immunoglobulin subclass distribution of synovial plasma cells in rheumatoid arthritis determined by use of monoclonal anti-subclass antibodies
CLINICAL
IMMUNOLOGY
AND
IMMUNOPATHOLOGY
35, 346-351 (1!8s)
lmmunoglobulin Subclass Distribution of Synovial Plasma Cells in Rheumatoid Arthritis Determined by Use of Monoclonal Anti-Subclass Anti bodies’ PATRICIA L. HABER,*‘* HIROMI KUBAGAWA,? AND WILLIAM J. KOOPMAN**~ *Division of Clinicul Immunology und Rheumutology Tumor Institute, The Comprehensive Cancer Center, Medicine. University of Alabama in Birmingham Hospital. Birmingham.
und iThe Cellular Immunobiology Unit qf the Departments of Pediutrics. Microbiology ond und Birmingham Veteruns’ Administrution Alabama 35294
The distribution of IgG and IgA subclass plasma cells among dissociated synovial cells from 14 rheumatoid arthritis (RA) synovia was examined by immunofluorescence using mouse monoclonal anti-human subclass antibodies. Of the IgG plasma cells 81 2 9% were IgG,, 4 2 2% IgG,, 14 2 9% IgG,. and 0.9 k 0.6% IgG,. While IgG, predominated in all 14 synovia (which is similar to what is seen in normal tissues), in
INTRODUCTION
A characteristic feature of the inflamed synovium in rheumatoid arthritis (RA) is the presence of numerous plasma cells. Some of these plasma cells produce rheumatoid factor (l-3) but the antigenic specificity of the nonrheumatoid factor immunoglobulin remains largely unknown. IgG is the predominant immunoglobulin class produced in most RA synovia (2-4). Using heterologous antibodies to IgG subclasses or indirect isolation techniques, several investigators have suggested that a relatively increased proportion of IgG, is expressed in rheumatoid synovium (2-4). Since certain antigens have been demonstrated to elicit an immunoglobulin response largely confined to one IgG subclass (5-7), elucidation of the distribution of IgG subclasses in RA synovium could provide clues to basic pathogenic mechanisms. Recently, mouse monoclonal antibodies specific for each of the IgG and IgA subclasses have become available. In the following study these antibodies were used to reexamine the issue of immunoglobulin subclass distribution in RA synovium. ’ A preliminary report of this study was presented at the 1984 meeting of the American Rheumatism Association, Minneapolis, Min., June 2-6, 1984. 2 Present address: Department of Internal Medicine, Rheumatology Division, St. Louis University Medical Center, St. Louis, MO. 63104. 3 To whom correspondence should be sent at: Division of Clinical Immunology and Rheumatology. University of Alabama in Birmingham, University Station, Birmingham, Ala. 35294. 346 0090-1229/85 $1.50 Copyright All rights
8 1985 by Academic Press. Inc of reproduction m any form reserved.
SYNOVIAL
PLASMA
CELLS
MATERIALS
IN
RHEUMATOID
ARTHRITIS
347
AND METHODS
Antibodies. Mouse monoclonal antibodies used in these studies included BAM 15 (anti-IgG,, mouse isotype yl~) (Seward Laboratories, London); BAM 14 (antiIgG,, mouse isotype 71~) (Seward); C3-8-34 (anti-IgG,, mouse isotype y2,~) prepared by M. E. Conley (University of Pennsylvania, Philadelphia); 9441-SA (antiIgG,, mouse isotype yl~) (BRL, Gaithersburg, Md.); l-155-1 (anti-IgA,, mouse isotype y3h); and 14-3-26 (anti-IgA,, mouse isotype YORK) prepared by M. E. Conley. Evidence for the subclass specificity of these mouse monoclonal antibodies has been reported previously (8, 9). In addition to these antibodies, fluorochrome-coupled [either fluorescein isothiocyanate (FITC) or tetramethylrhodamine isothiocyanate (RITC)], affinity-purified goat antibodies against mouse immunoglobulin and against human F, y, or (;Ychains were provided by W. Gathings (University of Alabama, Birmingham) (Southern Biotechnology, Birmingham). Synovial cell preparation and immunojluorescent staining. Synovial tissues from 14 patients with classical/definite RA as defined by the American Rheumatism Association (10) were obtained during surgery for either joint replacement or synovectomy. The tissue was digested using Joklik’s medium containing 10% fetal bovine serum (Flow Laboratories) and Dispase (1 mg/ml; Grade II; Boehringer Mannheim Biochemicals, Indianapolis, Ind.) as previously described (11). The mean cell yield was 1.6 + 0.8 x lo6 cells/g tissue. After washing, the cells were cytocentrifuged onto glass slides then fixed with acid-alcohol. Analysis of immunofluorescence for the frequency of IgM, IgG, and IgA plasma cells as well as the distribution of IgG and IgA subclasses was performed according to previously described methods (812). Briefly, for the determination of the frequency of the IgG subclasses among IgG plasma cells and IgA subclasses among IgA plasma cells, the fixed cell preparations were first incubated with mouse monoclonal anti-subclass antibody, then with FITC-coupled goat anti-mouse Ig that had been extensively absorbed with human immunoglobulin. The mouse monoclonal antibodies used were all in ascitic fluid diluted to a titer predetermined to be optimal for immunofluorescence. Generally, 100-200 IgG plasma cells were counted except for two patients in which fewer numbers of plasma cells were present (I.C. and L.M). We considered the possibility that staining with anti-y subclass monoclonal antibodies might be influenced by the presence of plasma cells producing rheumatoid factor (RF). It should be stressed in this regard that the monoclonal antibodies directed against IgG,, IgG,, and IgG, were all yl~. Insignificant staining with the anti-IgG, reagent was consistently observed (see Results). To control for the possibility that RF plasma cells might disproportionately bind the monoclonal antibody directed against IgG, (YORK), we stained three synovia in parallel with two other anti-IgG, monoclonal antibodies (both -yl~). Identical results were obtained with each of the three monoclonal anti-IgG, monoclonal antibodies. RESULTS
Dissociated
synovial
cells from 14 patients
with RA were examined
by im-
348
HABkK.
KUBAGAWA.
.4ND
KOOI’MAN
munflourescence for the presence of IgM, IgG and IgA plasma cells (Table I). They were found to contain 0.9 -t I. I% IgM + (mean f I SD). I .6 z 2.0% IgG ’ . and 0.8 + I. 1% IgA’ plasma cells. These synovia were chosen for study because of the presence of adequate numbers of IgG plasma cells for distinguishing subclass distribution. In order to determine the distribution of IgG and IgA subclases, dual-staining studies were performed. Of the IgG plasma cells, 81 i 9% (mean +- I SD) were IgG,, 4 + 2% were IgG?. I4 I 9% were IgG,, and 0.9 ? 0.6% were IgG,. For each of the synovia studied the sum of the frequencies of the four subclasses was within the range of 100 ? 10% making it unlikely that there was a significant distortion of the results due to rheumatoid factor-producing plasma cells binding either the mouse or the goat antibodies. In all the synovia. IgG, was the predominant IgG subclass. Eight synovia had IgG, as the next most frequent subclass. In five of these, IgG, comprised 20% or greater of the IgG plasma cells .This is a higher percentage than has been reported for normal blood. bone marrow. or spleen IgG plasma cells using the same reagents (8). In addition, the frequency of IgG, plasma cells in most of the synovia was lower than found in normal adult tissues. The reduced frequency of IgG2 occurred even in those synovia without increased IgG, making it unlikely that the reduced frequency of IgG, was entirely secondary to the increased IgG,. Nine synovia were examined for IgA subclasses. IgA,, comprising more than 80% of the IgA plasma cells in each synovium, was by far the dominant subclass. This distribution of IgA subclasses is similar to that described for serum IgA and for bone marrow IgA plasma cells but is unlike the pattern seen in mucosal tissue (13). DISCUSSION Plasma cells from RA synovia were examined by immunofluorescence using mouse monoclonal antibodies against IgG and IgA subclasses. For all fourteen synovia, IgG, was the major subclass of IgG plasma cell, which is similar to what has been reported previously for normal tissues (8). The frequency of IgG, plasma cells, however, was lower than previously reported for normal tissues using these same anti-subclass antibodies and, for Yl4 synovia, IgG, was proportionately higher. Among IgA plasma cells, IgA, uniformly predominated with the relative proportions of IgA, and IgAz similar to what is found in serum. The occurrence of a relative increase in IgG, in RA synovium has been reported previously by Munthe and Natvig (2) for synovial plasma cells and by Hoffman et al. (4) for IgG secreted by synovial explant cultures. Our results, however, do not show as pronounced a disproportion as has been reported. Furthermore, the increased IgG, occurred only in a minority of the synovia studied. While differences in patient populations, particularly with regard to drug regimens, cannot be excluded, all three studies examined patients with advanced disease who had been exposed to a variety of anti-inflammatory drugs. Differences between our observations and those of Munthe and Natvig (2) are likely due to two factors. First, we used monoclonal anti-subclass antibodies which may offer finer specificity than previously available heterologous antibodies. Second, rather than using
” Not done. b Too few cells for frequency
1.6 2 0.8
Mean
( h 1 SD)
0.9 1.6 1.6 1.5 1.7 2.0 0.2 2.5 1.6 1.3 1.0 2.5 0.9 3.5
L.W. M.S. ES. G.P. W.S. E.B. R.A. M.A.S. W.M. A.F. R.J. I.C. L.M. C.M.
Patient
0.9 It I.1
3.0 3.0 1.0 0.5 0.5 1.0 0.5 0.2 0.1 0.1 0.1 0.2 N.D.” N.D.
kM
determination.
Cell yield (per g x IOh)
TABLE
1
All IgA
cell (%)
staining
1.6 ” 2.0
1.5 7.5 2.0 0.5 2.0 1.0 1.0 4.0 2.0 0.3 0.2 0.3 0.2 0.5
kG
Plasma frequency
cells seen were
0.8 -+ 1.1
3.5 2.0 0.5 1.0 0.2 (0. I 0.2 2.0 0.3 0.1 0.1
&A
IgA,.
81 2 9
90 87 89 84 86 78 67 75 79 60 76 90 87 92
Gl
SUBCLASS DISTRIBUTIONOFSYNOVIALPLASMACELLSIN
422
5 3 4 10 4 3 4 8 2 4 4
G2
14 i
3 10 3 6 2 21 28 10 20 36 30
G3
IgG subclass distribution (%)
9
RHEUMATOID ARTHRITIS
1 2
0.9 rt 0.6
3
4 (1
(1
Cl
3 2 4 3 2
G4
89 + 5
T N.D. N.D.
84 97 N.D.” N.D. 90 t 90 85 N.D. t 85
Al
823
4 4 N.D. N.D. 10 J,b 10 11 N.D. 1 10 4 N.D. N.D.
A2
IgA subclass distribution (%)
8 =j t;
$
5 0 G
z
E
i?
t
mo
f
f
$
5
i5
2
350
HABEK,
KUBAGAWA.
AND KOOPMAN
tissue sections. we utilized dissociated synovial cells, perhap:, yielding ;I hccler sampling of the entire tissue. Comparison between our results and those of Hoffman and colleagues (4) i\ difficult since our study analyzed plasma cells rather than secreted immunoglobulin. The possibility exists that the distribution of IgG subclass plasma cells may not strictly parallel the distribution of immunoglobulin found in cell cultures. In this regard, our laboratory previously demonstrated that plasma cell frequency can be dissociated from levels of secretion of IgA subclasses using peripheral blood B cells (14). Previous studies have suggested that IgG, has a higher rate of catabolism than the other IgG subclasses and a slightly higher rate of synthesis has also been suggested (15, 16). This could account in part for the higher proportion of IgG, Hoffman and co-workers (4) found in their synovial explant cultures, though it seems unlikely that IgG, production for any of the synovia we examined would be greater than 90% of the total IgG as these workers reported for several synovia. Studies currently in progress in our laboratory using radioimmunoassays to quantitate the IgG subclasses in cell culture supernatants should help clarify this issue. In our study a relative decrease in IgGl plasma cells was observed more frequently than the increase in IgG,. This is in contrast to the wide variability in the frequency of IgG, plasma cells Munthe and Natvig noted (2). Again, the different findings may relate to previously mentioned differences in techniques. Shakib and Stanworth have reported a decrease in serum IgG, in RA patients which suggests that our finding of a decrease in synovial IgG, may not be a localized phenomenon ( 17). Our findings as well as those of Munthe and Natvig (2) and Hoffman et cl/. (4) suggest that restricted immunoglobulin production exists in the synovia of certain RA patients; however, for the majority of patients, other than the relative decrease in IgGz, the IgG subclass distribution did not differ from that present in normal tissues. Our data therefore do not support the hypothesis of uniform induction of a selective IgG subclass by putative antigen(s) in RA synovium. It is, nonetheless, possible that disease duration or drug therapy may have influenced our findings and that more pronounced distortions in IgG subclass distribution would be observed earlier in the course of the disease. Advances in arthroscopic tissue sampling should permit analysis of synovial plasma cell distribution earlier in the disease. Other approaches, however, such as examination of immunoglobulin variable region diversity, may be more useful for determining the extent of clonal restriction of immunoglobulins produced in RA synovium. ACKNOWLEDGMENTS The authors express their appreciation for the technical assistance of Ms. Jennifer Collins and Teresa Langford and for the assistance of Ms. Marie Mays and Ms. Brenda Gilmore in preparing the manuscript. This work was supported by NIH Awards AM 32073. AI 18745. and AM 03555 and the Research Program of the Veteran’s Administration.
REFERENCES 1. Mellors. R. C., Heimer, R.. Corcos, J.. and Korngold. L.. J.
Exp.
Med.
110, 875. 1959
SYNOVIAL 2. 3. 4. S. 6. 7. 8. 9. 10. It.
CELLS
IN RHEUMATOID
ARTHRITIS
351
Munthe, E., and Natvig, J. B., C/in. Exp. Immrtnol. 12, 55. 1972. Egland. 0.. Lea, T., Saari. G.. Mellbye, 0. J., and Natvig, J. B.. Arthritis Rheum. 25, 1445. 1982. Hoffman, W. L., Goldberg, M. S., and Smiley, J. D., J. Clin. In\,esr. 69, 136, 1982. Yount, W. J., Domer, M. M., Kunkel, H. G., and Kabat, E. A., J. Exp. Med. 127, 633, 1968. Perlmutter, R. M.. Hansburg. D., Briles. D. E., Nicolotti, R. A.. and Davie. J. M.. J. Immwwl. 121, 566. 1978. Slack, J.. Der-Balian, G. P, Nahm, M.. and Davie. J. M.. J. &p. Med. 151, 853, 1980. Mayumi, M..Kuritani. T., Kubagawa, H., and Cooper, M. D.. J. ImmLlno/. 130, 671, 1983. Conley. M. E.. Lawton. A. R., and Cooper. M. D., J. Immunol. 125, 121 I. 1980. Ropes, M. W.. Bennett. G. A.. Cobbs, S.. Jacox. R.. and Jessar. R. A.. Bd/. Rhelrm. DiA. 9, 175. 1958. Koopman, W. J., Miller. R.K., Crago. S. S., Mestecky, J.. 3rd Schrohen]oher. R. E.. Am. N. y. Acad.
12. 13. 14. 15. 16. 17.
PLASMA
Sci. 409,
258, 1983.
Gathings. W. E.. Lawton, A. R.. and Cooper, M. D., Eur. J. Immunol. 7, 804, 1977. Andre, C., Andre. F., and Fargier. C.. Ciin. Exp. Immunol. 33, 327. 1978. Conley, M. E.. and Koopman. W. J., J. Exp. Med. 156, 1615, 1982. Spiegelberg. H. L., Fishkin, B. G., and Grey, H. M.. J. C/in. Inrest. 47, 2323. 1968. Morel], A., Terry. W. D.. and Waldmann, T. A., J. C/in. [nrjest. 49, 673, 1970. Shakib. F.. and Stanworth, D. R.. Ann. Rheum. Djs. 35, 263. 1976.
Received January 7.1985; accepted January 22, 1985
IMMUNOLOGY
AND
IMMUNOPATHOLOGY
35, 346-351 (1!8s)
lmmunoglobulin Subclass Distribution of Synovial Plasma Cells in Rheumatoid Arthritis Determined by Use of Monoclonal Anti-Subclass Anti bodies’ PATRICIA L. HABER,*‘* HIROMI KUBAGAWA,? AND WILLIAM J. KOOPMAN**~ *Division of Clinicul Immunology und Rheumutology Tumor Institute, The Comprehensive Cancer Center, Medicine. University of Alabama in Birmingham Hospital. Birmingham.
und iThe Cellular Immunobiology Unit qf the Departments of Pediutrics. Microbiology ond und Birmingham Veteruns’ Administrution Alabama 35294
The distribution of IgG and IgA subclass plasma cells among dissociated synovial cells from 14 rheumatoid arthritis (RA) synovia was examined by immunofluorescence using mouse monoclonal anti-human subclass antibodies. Of the IgG plasma cells 81 2 9% were IgG,, 4 2 2% IgG,, 14 2 9% IgG,. and 0.9 k 0.6% IgG,. While IgG, predominated in all 14 synovia (which is similar to what is seen in normal tissues), in
INTRODUCTION
A characteristic feature of the inflamed synovium in rheumatoid arthritis (RA) is the presence of numerous plasma cells. Some of these plasma cells produce rheumatoid factor (l-3) but the antigenic specificity of the nonrheumatoid factor immunoglobulin remains largely unknown. IgG is the predominant immunoglobulin class produced in most RA synovia (2-4). Using heterologous antibodies to IgG subclasses or indirect isolation techniques, several investigators have suggested that a relatively increased proportion of IgG, is expressed in rheumatoid synovium (2-4). Since certain antigens have been demonstrated to elicit an immunoglobulin response largely confined to one IgG subclass (5-7), elucidation of the distribution of IgG subclasses in RA synovium could provide clues to basic pathogenic mechanisms. Recently, mouse monoclonal antibodies specific for each of the IgG and IgA subclasses have become available. In the following study these antibodies were used to reexamine the issue of immunoglobulin subclass distribution in RA synovium. ’ A preliminary report of this study was presented at the 1984 meeting of the American Rheumatism Association, Minneapolis, Min., June 2-6, 1984. 2 Present address: Department of Internal Medicine, Rheumatology Division, St. Louis University Medical Center, St. Louis, MO. 63104. 3 To whom correspondence should be sent at: Division of Clinical Immunology and Rheumatology. University of Alabama in Birmingham, University Station, Birmingham, Ala. 35294. 346 0090-1229/85 $1.50 Copyright All rights
8 1985 by Academic Press. Inc of reproduction m any form reserved.
SYNOVIAL
PLASMA
CELLS
MATERIALS
IN
RHEUMATOID
ARTHRITIS
347
AND METHODS
Antibodies. Mouse monoclonal antibodies used in these studies included BAM 15 (anti-IgG,, mouse isotype yl~) (Seward Laboratories, London); BAM 14 (antiIgG,, mouse isotype 71~) (Seward); C3-8-34 (anti-IgG,, mouse isotype y2,~) prepared by M. E. Conley (University of Pennsylvania, Philadelphia); 9441-SA (antiIgG,, mouse isotype yl~) (BRL, Gaithersburg, Md.); l-155-1 (anti-IgA,, mouse isotype y3h); and 14-3-26 (anti-IgA,, mouse isotype YORK) prepared by M. E. Conley. Evidence for the subclass specificity of these mouse monoclonal antibodies has been reported previously (8, 9). In addition to these antibodies, fluorochrome-coupled [either fluorescein isothiocyanate (FITC) or tetramethylrhodamine isothiocyanate (RITC)], affinity-purified goat antibodies against mouse immunoglobulin and against human F, y, or (;Ychains were provided by W. Gathings (University of Alabama, Birmingham) (Southern Biotechnology, Birmingham). Synovial cell preparation and immunojluorescent staining. Synovial tissues from 14 patients with classical/definite RA as defined by the American Rheumatism Association (10) were obtained during surgery for either joint replacement or synovectomy. The tissue was digested using Joklik’s medium containing 10% fetal bovine serum (Flow Laboratories) and Dispase (1 mg/ml; Grade II; Boehringer Mannheim Biochemicals, Indianapolis, Ind.) as previously described (11). The mean cell yield was 1.6 + 0.8 x lo6 cells/g tissue. After washing, the cells were cytocentrifuged onto glass slides then fixed with acid-alcohol. Analysis of immunofluorescence for the frequency of IgM, IgG, and IgA plasma cells as well as the distribution of IgG and IgA subclasses was performed according to previously described methods (812). Briefly, for the determination of the frequency of the IgG subclasses among IgG plasma cells and IgA subclasses among IgA plasma cells, the fixed cell preparations were first incubated with mouse monoclonal anti-subclass antibody, then with FITC-coupled goat anti-mouse Ig that had been extensively absorbed with human immunoglobulin. The mouse monoclonal antibodies used were all in ascitic fluid diluted to a titer predetermined to be optimal for immunofluorescence. Generally, 100-200 IgG plasma cells were counted except for two patients in which fewer numbers of plasma cells were present (I.C. and L.M). We considered the possibility that staining with anti-y subclass monoclonal antibodies might be influenced by the presence of plasma cells producing rheumatoid factor (RF). It should be stressed in this regard that the monoclonal antibodies directed against IgG,, IgG,, and IgG, were all yl~. Insignificant staining with the anti-IgG, reagent was consistently observed (see Results). To control for the possibility that RF plasma cells might disproportionately bind the monoclonal antibody directed against IgG, (YORK), we stained three synovia in parallel with two other anti-IgG, monoclonal antibodies (both -yl~). Identical results were obtained with each of the three monoclonal anti-IgG, monoclonal antibodies. RESULTS
Dissociated
synovial
cells from 14 patients
with RA were examined
by im-
348
HABkK.
KUBAGAWA.
.4ND
KOOI’MAN
munflourescence for the presence of IgM, IgG and IgA plasma cells (Table I). They were found to contain 0.9 -t I. I% IgM + (mean f I SD). I .6 z 2.0% IgG ’ . and 0.8 + I. 1% IgA’ plasma cells. These synovia were chosen for study because of the presence of adequate numbers of IgG plasma cells for distinguishing subclass distribution. In order to determine the distribution of IgG and IgA subclases, dual-staining studies were performed. Of the IgG plasma cells, 81 i 9% (mean +- I SD) were IgG,, 4 + 2% were IgG?. I4 I 9% were IgG,, and 0.9 ? 0.6% were IgG,. For each of the synovia studied the sum of the frequencies of the four subclasses was within the range of 100 ? 10% making it unlikely that there was a significant distortion of the results due to rheumatoid factor-producing plasma cells binding either the mouse or the goat antibodies. In all the synovia. IgG, was the predominant IgG subclass. Eight synovia had IgG, as the next most frequent subclass. In five of these, IgG, comprised 20% or greater of the IgG plasma cells .This is a higher percentage than has been reported for normal blood. bone marrow. or spleen IgG plasma cells using the same reagents (8). In addition, the frequency of IgG, plasma cells in most of the synovia was lower than found in normal adult tissues. The reduced frequency of IgG2 occurred even in those synovia without increased IgG, making it unlikely that the reduced frequency of IgG, was entirely secondary to the increased IgG,. Nine synovia were examined for IgA subclasses. IgA,, comprising more than 80% of the IgA plasma cells in each synovium, was by far the dominant subclass. This distribution of IgA subclasses is similar to that described for serum IgA and for bone marrow IgA plasma cells but is unlike the pattern seen in mucosal tissue (13). DISCUSSION Plasma cells from RA synovia were examined by immunofluorescence using mouse monoclonal antibodies against IgG and IgA subclasses. For all fourteen synovia, IgG, was the major subclass of IgG plasma cell, which is similar to what has been reported previously for normal tissues (8). The frequency of IgG, plasma cells, however, was lower than previously reported for normal tissues using these same anti-subclass antibodies and, for Yl4 synovia, IgG, was proportionately higher. Among IgA plasma cells, IgA, uniformly predominated with the relative proportions of IgA, and IgAz similar to what is found in serum. The occurrence of a relative increase in IgG, in RA synovium has been reported previously by Munthe and Natvig (2) for synovial plasma cells and by Hoffman et al. (4) for IgG secreted by synovial explant cultures. Our results, however, do not show as pronounced a disproportion as has been reported. Furthermore, the increased IgG, occurred only in a minority of the synovia studied. While differences in patient populations, particularly with regard to drug regimens, cannot be excluded, all three studies examined patients with advanced disease who had been exposed to a variety of anti-inflammatory drugs. Differences between our observations and those of Munthe and Natvig (2) are likely due to two factors. First, we used monoclonal anti-subclass antibodies which may offer finer specificity than previously available heterologous antibodies. Second, rather than using
” Not done. b Too few cells for frequency
1.6 2 0.8
Mean
( h 1 SD)
0.9 1.6 1.6 1.5 1.7 2.0 0.2 2.5 1.6 1.3 1.0 2.5 0.9 3.5
L.W. M.S. ES. G.P. W.S. E.B. R.A. M.A.S. W.M. A.F. R.J. I.C. L.M. C.M.
Patient
0.9 It I.1
3.0 3.0 1.0 0.5 0.5 1.0 0.5 0.2 0.1 0.1 0.1 0.2 N.D.” N.D.
kM
determination.
Cell yield (per g x IOh)
TABLE
1
All IgA
cell (%)
staining
1.6 ” 2.0
1.5 7.5 2.0 0.5 2.0 1.0 1.0 4.0 2.0 0.3 0.2 0.3 0.2 0.5
kG
Plasma frequency
cells seen were
0.8 -+ 1.1
3.5 2.0 0.5 1.0 0.2 (0. I 0.2 2.0 0.3 0.1 0.1
&A
IgA,.
81 2 9
90 87 89 84 86 78 67 75 79 60 76 90 87 92
Gl
SUBCLASS DISTRIBUTIONOFSYNOVIALPLASMACELLSIN
422
5 3 4 10 4 3 4 8 2 4 4
G2
14 i
3 10 3 6 2 21 28 10 20 36 30
G3
IgG subclass distribution (%)
9
RHEUMATOID ARTHRITIS
1 2
0.9 rt 0.6
3
4 (1
(1
Cl
3 2 4 3 2
G4
89 + 5
T N.D. N.D.
84 97 N.D.” N.D. 90 t 90 85 N.D. t 85
Al
823
4 4 N.D. N.D. 10 J,b 10 11 N.D. 1 10 4 N.D. N.D.
A2
IgA subclass distribution (%)
8 =j t;
$
5 0 G
z
E
i?
t
mo
f
f
$
5
i5
2
350
HABEK,
KUBAGAWA.
AND KOOPMAN
tissue sections. we utilized dissociated synovial cells, perhap:, yielding ;I hccler sampling of the entire tissue. Comparison between our results and those of Hoffman and colleagues (4) i\ difficult since our study analyzed plasma cells rather than secreted immunoglobulin. The possibility exists that the distribution of IgG subclass plasma cells may not strictly parallel the distribution of immunoglobulin found in cell cultures. In this regard, our laboratory previously demonstrated that plasma cell frequency can be dissociated from levels of secretion of IgA subclasses using peripheral blood B cells (14). Previous studies have suggested that IgG, has a higher rate of catabolism than the other IgG subclasses and a slightly higher rate of synthesis has also been suggested (15, 16). This could account in part for the higher proportion of IgG, Hoffman and co-workers (4) found in their synovial explant cultures, though it seems unlikely that IgG, production for any of the synovia we examined would be greater than 90% of the total IgG as these workers reported for several synovia. Studies currently in progress in our laboratory using radioimmunoassays to quantitate the IgG subclasses in cell culture supernatants should help clarify this issue. In our study a relative decrease in IgGl plasma cells was observed more frequently than the increase in IgG,. This is in contrast to the wide variability in the frequency of IgG, plasma cells Munthe and Natvig noted (2). Again, the different findings may relate to previously mentioned differences in techniques. Shakib and Stanworth have reported a decrease in serum IgG, in RA patients which suggests that our finding of a decrease in synovial IgG, may not be a localized phenomenon ( 17). Our findings as well as those of Munthe and Natvig (2) and Hoffman et cl/. (4) suggest that restricted immunoglobulin production exists in the synovia of certain RA patients; however, for the majority of patients, other than the relative decrease in IgGz, the IgG subclass distribution did not differ from that present in normal tissues. Our data therefore do not support the hypothesis of uniform induction of a selective IgG subclass by putative antigen(s) in RA synovium. It is, nonetheless, possible that disease duration or drug therapy may have influenced our findings and that more pronounced distortions in IgG subclass distribution would be observed earlier in the course of the disease. Advances in arthroscopic tissue sampling should permit analysis of synovial plasma cell distribution earlier in the disease. Other approaches, however, such as examination of immunoglobulin variable region diversity, may be more useful for determining the extent of clonal restriction of immunoglobulins produced in RA synovium. ACKNOWLEDGMENTS The authors express their appreciation for the technical assistance of Ms. Jennifer Collins and Teresa Langford and for the assistance of Ms. Marie Mays and Ms. Brenda Gilmore in preparing the manuscript. This work was supported by NIH Awards AM 32073. AI 18745. and AM 03555 and the Research Program of the Veteran’s Administration.
REFERENCES 1. Mellors. R. C., Heimer, R.. Corcos, J.. and Korngold. L.. J.
Exp.
Med.
110, 875. 1959
SYNOVIAL 2. 3. 4. S. 6. 7. 8. 9. 10. It.
CELLS
IN RHEUMATOID
ARTHRITIS
351
Munthe, E., and Natvig, J. B., C/in. Exp. Immrtnol. 12, 55. 1972. Egland. 0.. Lea, T., Saari. G.. Mellbye, 0. J., and Natvig, J. B.. Arthritis Rheum. 25, 1445. 1982. Hoffman, W. L., Goldberg, M. S., and Smiley, J. D., J. Clin. In\,esr. 69, 136, 1982. Yount, W. J., Domer, M. M., Kunkel, H. G., and Kabat, E. A., J. Exp. Med. 127, 633, 1968. Perlmutter, R. M.. Hansburg. D., Briles. D. E., Nicolotti, R. A.. and Davie. J. M.. J. Immwwl. 121, 566. 1978. Slack, J.. Der-Balian, G. P, Nahm, M.. and Davie. J. M.. J. &p. Med. 151, 853, 1980. Mayumi, M..Kuritani. T., Kubagawa, H., and Cooper, M. D.. J. ImmLlno/. 130, 671, 1983. Conley. M. E.. Lawton. A. R., and Cooper. M. D., J. Immunol. 125, 121 I. 1980. Ropes, M. W.. Bennett. G. A.. Cobbs, S.. Jacox. R.. and Jessar. R. A.. Bd/. Rhelrm. DiA. 9, 175. 1958. Koopman, W. J., Miller. R.K., Crago. S. S., Mestecky, J.. 3rd Schrohen]oher. R. E.. Am. N. y. Acad.
12. 13. 14. 15. 16. 17.
PLASMA
Sci. 409,
258, 1983.
Gathings. W. E.. Lawton, A. R.. and Cooper, M. D., Eur. J. Immunol. 7, 804, 1977. Andre, C., Andre. F., and Fargier. C.. Ciin. Exp. Immunol. 33, 327. 1978. Conley, M. E.. and Koopman. W. J., J. Exp. Med. 156, 1615, 1982. Spiegelberg. H. L., Fishkin, B. G., and Grey, H. M.. J. C/in. Inrest. 47, 2323. 1968. Morel], A., Terry. W. D.. and Waldmann, T. A., J. C/in. [nrjest. 49, 673, 1970. Shakib. F.. and Stanworth, D. R.. Ann. Rheum. Djs. 35, 263. 1976.
Received January 7.1985; accepted January 22, 1985