- Email: [email protected]
Immune complexes and antinuclear, antinucleolar, and anticentromere antibodies in scleroderma
I II
I
I I
I I
II
I
Immune complexes and antinuclear, antinucleolar, and anticentromere antibodies in scleroderma* Ze-yi Chen, M.D.,** Gabriel Virella, M.D., Ph.D., Hsiaoho Edward Tung, Ph.D., Sterling K. Ainsworth, Ph.D., Richard M. Silver, M.D., An-Chuan Wang, Ph.D., Mariano F. LaVia, M.D., Hildegard R. Maricq, M.D., and Richard L. Dobson, M.D. Charleston, SC Forty-one patients with various forms of systemic sclerosis (scleroderma) and positive antinuclear antibodies of nucleolar (ten patients), speckled (eleven patients), or centromere pattern (twenty patients) were selected for study of immune complexes by the radioisotope labeled Clq binding and the radioisotope labeled protein A binding methods. The presence of immune complexes was found by the Clq binding assay in sixteen patients (39%) and by a protein A binding assay in eight patients (20%). Overall, 46% of patients (19/41) had immune complexes. A lower incidence of organ involvement and fewer positive results in the screening of serum immune complexes were observed in patients with centromere antibody (35%) than in patients with nucleolar (60%) or speckled pattern (55%). Paiients with immune complexes had higher frequencies of kidney, heart, and muscle involvement and digital ulceration than did patients with no detectable immune complexes, but the differences were not statistically significant. Diffuse skin involvement was not related to the presence of immune complexes. (J AM ACADDERMATOL 11:461-467, 1984.)
Scleroderma (systemic sclerosis, SS) is a connective tissue disease characterized by tightness,
From the Departments of Dermatology (Chen, Dobson), Basic and Clinical Immunology and Microbiology (Virella, Tung, Wang), Pathology (Ainsworth), Medicine (Silver, Maricq), and Laboratory Medicine (LaVia), Medical University of South Carolina. Supported in part by National Institutes of Health Grant AM 31283 from the National Institute of Arthritis, Diabetes, and Digestive and Kidney Diseases. Accepted for publication Feb. 22, 1984. Reprint requests to: Dr. Gabriel Virella, Department of Basic and Clinical Immunology and Microbiology, 171 Ashley Ave., Charleston, SC 29425. *Publication No. 662 from the Department of Basic and Clinical Immunology and Microbiology, Medical University of South Carolina. **A M.U.S.C. Visiting Associate Professor of Dermatology (research) from Shanghai Second Medical College, Shanghai, The People's Republic of China.
thickening, and induration of skin (scleroderma), Raynaud's phenomenon, and systemic involvement affecting a variety of organs. The cause of the disease remains obscure. The presence of vascular abnormalities in the tissues involved 2-4 and the deposition of immune complexes (IC) in the kidneys and skin of some patients suggest that circulating IC may be involved in the pathogenesis of SS. s'6 Although previous reports of IC in SS have produced conflicting results, 7-9 there has been the general suggestion that the presence of IC is related to more severe organ involvement. Antinuclear antibodies (ANA) have been detected by indirect immunofluorescence using tissue culture cells as substrate in almost all patients with SS. t°,11 The use of tissue culture cells as substrate allows easier differentiation of the various nuclear patterns (e.g., centromere, nucleolar, 461
462
Chen et al
and speckled), r2.1~ The association of the nucleolar and speckled patterns with scleroderma has been recognized by several groups, ~4-~7 and the different nuclear patterns appear to be related to different incidences of various clinical features. TM Anticentromere antibody (ACA) has been demonstrated in association with a less severe form of systemic sclerosis, the CREST syndrome (calcinosis, Raynaud's phenomenon, esophageal involvement, sclerodactyly, and telangiectasia). 19-..,2 A significant correlation between the presence of ACA and characteristic capillary patterns has also been observed. 11,2"~ Since previous studies have shown that SS patients with circulating IC may have a greater frequency of visceral involvement, 9 we sought to determine whether the presence of IC was related to different nuclear patterns, a point that had not been previously investigated, to the best of our knowledge. In this study the sera of forty-one patients with various forms of SS, selected on the basis of having speckled, nucleolar, or centromere ANA patterns, were examined for the presence of IC by radioisotope-labeled Clq binding and radioisotope-labeled protein A binding. The correlation of circulating IC with different nuclear patterns and various clinical features was investigated. MATERIAL AND METHODS Selection of patients. Single serum samples from forty-one patients with various forms of SS were included in the study. The patients were selected for study of IC on the basis of the following criteria: (1) patients had been admitted at least once since 1978 to the Medical University of South Carolina Hospital for evaluation of scleroderma and (2) a positive ANA titer with clear nuclear patterns that include centromere, nucleolar, or speckled patterns, as demonstrated by indirect immunofluorescence with human epithelioid cells (HEp-2) as substrate. Only patients with manifestations suggestive of systemic lupus erythematosus, locaiized scleroderma, and Raynaud's phenomenon were excluded from this study. Clinical assessment. Diagnosis of SS was based On the American Rheumatism Association (ARA) criteria. 24 Diffuse scleroderma was characterized by skin invoNement proximal to the elbow and/or the knee joints. Calcinosis and Raynaud's phenomenon were assessed by patient history and clinical examination. Esophageal involvement was evaluated by esophageal
Journal of the American Academyof Dermatology
manometry (18/41), cine-esophagogram (14/41), or both methods (9/41). Pulmonary abnormalities were defined as bibasilar reticular fibrosis on the basis of standard chest roentgenography and/or abnormalities indicated by pulmonary function tests (forced vital capacity <80% of predicted capacity or diffusion coefficient <80%). Cardiac involvement was defined as pericarditis, conduction disturbances, or congestive heart failure as indicated by clinical manifestations, electrocardiography, and echocardiography (36/41), in the absence of other causes. Renal involvement was defined as creatinine clearance <75 ml/min or proteinuria >2 gm/24 hr. Joint involvement was assessed by patient history or clinical examination of joint inflammation. Muscle involvement was defined as proximal muscle weakness in the prescence of one or more of the following: increased serum levels of creatine phosphokinase or aldolase, positive muscle biopsy findings, or a positive electromyogram. The patients were divided into three groups on the basis of the following ANA patterns: centromere, nucleolar, and speckled patterns. Patients with anticentromere antibody (ACA). This group included twenty patients; ACA titers varied from 320 to 10,240. Fifteen of twenty patients fulfilled the preliminary ARA diagnostic criteria for SS ,24 Three of these fifteen patients had diffuse scleroderma. Twelve had complete or incomplete CREST syndrome. Of the remaining five patients, two had Raynaud's phenomenon and telangiectasia; one had Raynaud's phenomenon and esophageal involvement; one had Raynaud's phenomenon with pulmonary and cardiac involvement; and the last one had Raynaud's phenomenon, sclerodactyly, arthritis, and cardiac involvement. Patients With ANA of nucleolar pattern. This group consisted of ten patients; ANA titers were from 320 to 163,840. Nine of ten patients fulfilled the preliminary ARA criteria for systemic sclerosis; five of these nine patients had diffuse skin involvement and four had proximal skin involvement. One patient had Raynaud's phenomenon, sclerodactyly, and esophageal and joint involvement, but these manifestations were not enough to fulfill the diagnostic criteria of SS. Patients with ANA of speckled pattern. This group included eleven patients, all of whom fulfilled the preliminary ARA criteria for SS; ANA titers ranged from 40 to 81,920. Seven of these patients had diffuse seleroderma, and the remaining four patients had skin thickening proximal to the metacarpophalangeal or metatarsophalangeal joints. The sera obtained from all patients included in the study were separated into
Volume 11 Number 3 September, 1984
Immune complexes and antibodies in scleroderma
463
T a b l e I. Comparison of clinical characteristics of patients with speckled, nucleolar; or centromere nuclear pattern Nuclear pattern
Patient No. (F:M) Mean age at study (yr) Mean disease duration (yr) Organ system involved Diffuse skin (trunk and/or proximal extremities) Esophagus Lung Kidney Heart Joint Muscle Digital ulceration Raynaud's phenomenon Laboratory abnormalities Anti-DNA antibody Rheumatoid factor Cryoglobulin Erythrocyte sedimentation rate C3 and/or C4 Hypergammaglobulinemia ANA other than ACA ACA Anti-Sin antibodies [mmunodiffusion Hemagglutination Anti-RNP antibodies Immunodiffusion Hemagglutination
11 (9:2) 39.8 - 15.5" 5.6 _-_ 2.7
10 (8:2) 39.5 --- 14.6 6.5 --- 6.8
7 (64%)'~ 9 10 4 8 9 5 6 10
(82%) (91%) (36%) (73%) (82%) (46%) (55%) (91%)
5 i50%) 7 7 3 6 7 2 5 9
(70%) (70%) (30%) (60%) (70%) (20%) (50%) (90%)
20 (15:5) 48.2 --- 11.1 8.2 ± 8.2 3 (15%) 14 (70%) 10 (50%) 0 (0%) 5 (25%) 6 (30%) 0 (0%) 11 (55%) 20 (100%)
3 (27%) 2 (18%) 0/8 (0%) 5 (46%)
0 (0%) 0 (,0%) 0/8 (0%) 5 (50%)
3 (15%) 2/19 (11%)$ 0/19 (0%) 3/19 (16%)
3/9 (33%) 3/9 (33%) 11 (100%)§ 0 (0%)
2 (20%) 2/9 (22%) 10 (100%)§ 0 (0%)
0/18 (0%) 3/20 (15%) 0 (0%) 20 (100%)§
1 (9%) 3 (27%)
0 (0%) 0 (0%)
0 (0%) 1 (5%)
3 (27%) 4 (36%)
1 (10%) 2 (20%)
1 (5%) 1 (5%)
*Mean ± SD. t% = Numberpositive/numberof each pattern, all patientshave beentested. ~Number positive/numbertested. §By patient selection.
aliquots and stored at - 7 0 ° C for determination of ANA, IC, Sm antigens and ribonucleoprotein (RNP) antibodies. Laboratory tests. The following tests were completed at the time of clinical evaluation: (1) doublestranded deoxynucleic acid antibody test performed by the Farr radioimmunoassay (values >20 units were considered positive); (2) C3 and C4 levels quantified by radioimmunodiffusion or immunonephelometry (the patients' levels were defined as normal or abnormal according to the normal ranges established for each assay technic); (3) rheumatoid factor determined by
latex agglutination (agglutinating titers equal to Or exceeding 1:20 were considered positive); (4)cryoprecipitation determined by incubating a serum sample separated from 37° C clotted blood at 4° C for 72 hours; (5) erythroeyte sedimentation rate performed by the Westergren method; and (6) serum protein levels determined by electrophoresis. Indirect immunofluorescence technics. The ANA tests were done with the use of HEp-2 cells as substrate (Immuno-Concepts Inc., Sacramento, CA). The cell line is an epithelial cell line originally propagated from a human laryngeal carcinoma. The substrate was grown
464
Journal of the American Academy of Dermatology
Chen et al
T a b l e II. Frequency of ICs in patients with speckled, nucleolar, or centromere pattern of A N A Nuclear pattern
Speckled (N = 11) IC assays
No.
Clq method Protein A method Both Clq and protein A methods Either Clq or protein A method
5 3 2 6
]
Nucleolar (N = 10)
%
No?
46 27 19 55
5 2 1 6
on the slide and fixed by the manufacturer with absolute ethanol at 0° C for 6 minutes. The ANA tests were performed according to instructions provided by the manufacturer. The sera were screened at 1:10, 1:20, and 1 : 40 to determine the presence of mixed patterns, but the results were considered positive only when nuclear fluorescence staining was clearly observed with discernible nuclear or nucleolar staining patterns at 1 : 40 dilution. The staining patterns of ANA were photographed with Kodak Ektachrome 400 film. Immunodiffusion. For determination of the anti-Sm and anti-RNP antibodies, all sera were analyzed by immunodiffusion in agarose against extractable nuclear antigens of rabbit thymus (Pel-Freeze Biologicals, Rogers, AR). Immunodiffusion was performed as previously reported. 2~ The precipitin lines of anti-Sm and anti-RNP were confirmed with reference sera kindly supplied by the Centers for Disease Control, Atlanta, GA. Hemagglutination. Anti-Sin and anti-RNP antibodies were also detected by the hemagglutination technic. 2G IC assays. t251-1abeled Clq binding assay. A modification of the method of Zubler and Lambert''7 was used. 28 Clq was purchased from DNW Biologics, San Antonio, TX, and was labeled with 1~5I by the peroxidase-glucose oxidase technic,29 with the use of reagents supplied by Bio-Rad Laboratories (Richmond, CA) and 12~I (NaI), 15.5 mCi/mg specific activity, supplied by Amersham Corp. (Arlington Heights, IL). For a comparison of results obtained in different rtins, results are expressed as a percent of the mean of the values obtained from the simultaneously assayed sera Of six control subjects (the mean being considered as 100%). Using the ninety-fifth percentile of forty-four sera from sixteen healthy control subjects assayed with the same a25I-labeled Clq batch as the sera of patients included in the study, we set the lower limit of positivity at 130%. *~5l-labeled protein A binding assay. This assay is
'" % 50 20 10 60
Centromere (N = 20)
No.[ 6 3 2 7
"Total" (N = 41)
%
No.]
9'0
30 15 10 35
16 8 5 19
39 20 12 46
based on the coprecipitation of 12~I-labeled protein A with soluble IC in the presence of 3% (w/v) polyethylene glycol 6000. Basically, we prepared 1/4.0 dilutions of sera in pH 8.4 borate buffer, 0.1 M, and prepared a mixture containing 0.1 ml of the serum dilution, 0.4 ml of borate buffer, 10/xl of a25I-labeled protein A (30 Ci/g specific activity, obtained from I C N Radioisotopics, Irvine, CA), and 0.5 ml of 6% polyethylene glycol 6000 in borate buffer. The mixtures were incubated overnight at 4 ° to 8° C, and the precipitate was washed once with ice-cold 3% polyethylene glycol. After the supernatant was carefully decanted, the radioactivity in the precipitate was measured and the results were expressed as a percentage of the ratio* relative to the sera of normal control subjects included in the same run, considering as i 0 0 % the average of the normal control subjects' ratio* for each run. Values greater than 153% (ninety-fifth percentile value for a series of sera from nineteen different normal donors) were considered abnormal. RESULTS The major clinical features of our forty-one patients with different ANA patterns are presented in T a b l e I. The female/male ratio was similar for each o f the three ANA groups. Patients with A C A t e n d e d to be older and to have a longer disease duration than patients with either a speckled or a n u c l e o l a r pattern. The centromere pattern was associated with less diffuse skin involvement, as well as lower frequencies of renal, cardiac, m u s c l e , and joint abnormalities. R a y n a u d ' s p h e n o m e n o n (present in all but two patients), digital ulceration, and esophageal dysmotility occurred with n e a r l y Clinical and laboratory
manifestations.
*Ratio of countsper minute in polyethyleneglycolprecipitateto the total counts per minute added.
Volume 11 Number 3 September, 1984
Immune complexes and antibodies in scleroderma
465
Table III. Clinical differences between patients with and without circulating immune complexes (CIC) (detected by either method) CIC results by either method Without CIC (N = 22)
With CIC (N = 19) %
No.
%
7
37
8
36
14' 13 5 11 11 5 13 19
74 68 26 58 58 26 68 100
16 14 2 8 11 2 9 20
73 64 9 36 50 9 41 91
3 3/18? 0/18 8/18
16 17 0 44
3 1/21 0/17 5/22
14 5 0 23
3/17 5 / 18
18 28
2/20 3/20
10 15
No.
Organ system involved Diffuse skin (trunk and/or proximal extremities Esophagus Lung Kidney Heart Joint Muscle Digital ulceration Raynaud's phenomenon Laboratory abnormalities Anti-DNA antibody Rheumatoid factor Cryoglobulin Erythrocyte sedimentation rate C3 and/or C4 Hypergammaglobulinemia Anti-Sm antibodies Imnaunodiffusion Hemagglutination Anti-RNP antibodies Immunodiffusion Hemagglutination
I
0 2
0 11
3 4
16 21
1
5
2
9
2
9
3
14
*Numberpositive/numbertested, all patientshavebeentested. tNumber positive/numbertested.
equal frequency in each of the three ANA groups. Lung disease was noted most frequently in the speckled A N A group (91%) but occurred frequently in the nucleolar and centromere group (70% and 50%, respectively). In terms of laboratory abnormalities, the three A N A groups differed in several respects. The group with a centromere pattern always had normal C3 and/or C4 values and had a lower frequency of an elevated erythrocyte sedimentation rate. Antibodies to deoxyribonucleic acid were detected most frequently in the speckled ANA group. As might be expected, the speckled group had the highest frequency of anti-Sm and antiRNP antibodies as well. R e s u l t s o f s o l u b l e IC screening tests. Serum
IC was detected in nineteen of forty-one patients (46%) with scleroderma by either Clq binding assay or protein A binding assay or both (Table II). The Clq binding assay produced a higher frequency of positive results (39%) than the protein A binding assay (20%). The presence of IC was detected in five patients (12%), as determined by both methods. Correlation of IC with ANA patterns. When the presence of serum IC was compared among the patients with different ANA patterns, the results of IC assays in patients with the nucleolar pattern (60%) were similar to those in patients with the speckled pattern (55%). A lower frequency of positive results was found in patients with ACA (25%) than in patients with either the speckled or
466
Chert et al
the nucleolar pattern, but these differences did not reach statistical significance. Correlation of IC determinations with clinical features, When patients with IC detected by at least one of the two methods were compared with patients who had negative test results for IC by both methods, the former group had a higher incidence of kidney, heart, and muscle involvement and digital ulceration (Table III). Rheumatoid factor positivity, increased erythrocyte sedimentation rate, hypergammaglobulinemia, and anti-RNP antibodies were also observed more frequently in patients with IC. These clinical and laboratory differences were not, however, statistically significant. Results of IC screening by the Clq binding assay also correlated with kidney, heart, and muscle involvement, as well as with digital ulceration. No such correlation was seen with the protein A assay (data not presented). DISCUSSION The efiopathogenesis of SS is poorly understood, but immunologic imbalance is believed to play an important roleJ ° Perivascular deposits of immunoglobulins have been demonstrated in the kidneys of patients studied and in the lungs of some patients with SS. a',3'a Immunoglobulin deposits have also been found at the dermoepidermal junction of some patients with scleroderma. 3z-3~ Recently our group has demonstrated immunoglobulin deposits at the dermoepidermal junction of the nailfold and forearm in some patients with SS. ~,:~6 These findings raise the possibility that soluble IC may be formed and deposited perivascularly in different tissues. A pathogenetic role for IC in scleroderma remains speculative. Some previous studies have detected circulating IC in a large proportion of their patients and have suggested a possible role for IC in the pathogenesis of SS; other studies have detected IC only rarelyY -41 The discrepancies among various studies may relate to the particular IC assays employed, as well as to differences in IC levels among different subsets of scleroderma patients studied. In the present study, the presence of IC was detected in 39% of patients by the Clq binding
Journal of the American Academy of Dermatology
assay and in 20% of patients by the protein A binding assay. In 46% of all our patients with various forms of SS, the presence of circulating IC was detected by either one of these two methods. Seibold et al:~7 found IC in 47% of scleroderma patients tested by two or more assays. Our study indicates that the detection of IC may differ a m o n g patients with different nuclear patterns of A N A in scleroderma; patients with ACA had a lower frequency of detectable circulating IC than did patients with ANA of a nucleolar or a speckled pattern. Since these same patients had a l o w e r frequency of skin and visceral organ involvement, one might speculate that organ system involvement seen in patients with a nucleolar or speckled pattern is due in part to the deposition of IC. H o w ever, as previously noted, a point prevalence survey such as ours does not allow such a conclusion.aT It may, however, explain some of the differences in frequencies of IC reported by various investigators. REFERENCES 1. Rodnan GP: Progressive systemic sclerosis (seleroderma), in McCartyDJ, editor: Arthritis and allied conditions, ed. 9. Philadelphia, 1979, Lea & Febiger, pp. 762-809. 2. Campbell PM, LeRoy EC: Pathogenesis of systemic sclerosis: A vascularhypothesis. Semin Arthritis Rheum 4:351-368, 1975. 3. FleischmajerR, Perlish JS, Shaw KV, Pirozzi D/: Skin capillary changes in early systemic scleroderma. Arch Dermatol 112:1553-1557, 1976. 4. Maricq HR, LeRoyEC: Progressive systemic sclerosis: Disorders of the microcirculation. Clin Rheum Dis S: 81-101, 1979. 5, LapenasD, RodnanGP, CavalloT: Immunopathologyof the renal vascularlesion of progressive systemicsclerosis (scleroderma). Am J Pathol 91:243-258, 1978. 6. Chen ZY, Ainsworth SK, Maricq HR, et al: Immunofluorescence studies in scleroderma. Clin Res 31:612A, 1983. 7. O'LoughlinS, Tappeiner G, Jordon RE: Circulatingimmune complexes in systemic scleroderma and generalized morphia. Dermatologica160:25-30, 1980. 8. CunninghamPH, Andrews BS, Davis JS IV: Immune complexes in progressive systemic sclerosis and mixed connective tissue disease. J Rheumatol7:301-308, 1980. 9. SiminovitchK, KleinM, Pruzanski W, et al: CircuIating immunecomplexesin patients with progressive systemic sclerosis. Arthritis Rheum 25:1174-1179, 1982, 10. BernsteinRM, SteigerwaldJC, Tan EM: Association of antinuclear and antinucleolar antibodies in progressive systemic sclerosis. Clin Exp Immunol 48:43-51, 1982. 11. Chen ZY, Silver RM, Ainsworth SK, et al: Association between fluorescentantinuclearantibodies, capillarypat-
Volume 11 Number 3 September, 1984
12.
13.
14. 15.
16. 17. 18.
19. 20. 21, 22.
23.
24. 25.
26.
hnmune complexes and antibodies in scleroderma
terns and clinical features in scleroderma spectrum disorders. Am J Med. (In press.) Parker MD, Kerby GP: Combined titre and fluorescent pattern of IgG antinuclear antibodies using cultured cell monolayers in evaluating connective tissue diseases. Ann Rheum Dis 33:465-472, 1974. Pisko E, Gallup K, Turner R, et al: Cardiopulmonary manifestations of progressive systemic sclerosis: Associations with circulating immune complexes and fluorescent antinuclear antibodies. Arthritis Rheum 22:518523, 1979. Ritchie RF: Antinucleolar antibodies: Their frequency and diagnostic association. N Engl J Med 282:11741178, 1970. Burnham TK, Fine G, Neblett TR: The immunofluorescent tumor imprint technique. II. The frequency of antinuclear factors in connective tissue diseases and dermaroses. Ann Intern Med 65:9-19, 1966. Rothfield NF, Rodnan GP: Serum antinuclear antibodies in progressive systemic sclerosis (scleroderma). Arthritis Rheum 11:607-617, 1968. Jordan RE, Deheer D, Schroeter A, Winkelmann RK: Antinuclear antibodies: Their significance in scleroderma. Mayo Clin Proc 46:111-113, 1971. Kleinsmith DM, Heinzerling RH, Burnham TK: Antinuclear antibodies as immunologic markers for a benign subset and different clinical characteristics of scleroderma. Arch Dermatol 118:882-885, 1982. Fritzler MJ, Kinsella TD, Garbutt E: The CREST syndrome: A distinct serologic entity with anticentromere antibodies. Am J Med 69:520-526, 1980. Tuffanelli DL, McKeon F, Kleinsmith DM, et al: Anticentromere and anticentriole antibodies in the scleroderma spectrum. Arch Dermatol 119:560-566, 1983. Chen ZY, Fedrick JA, Pandey JP, et al: Anticentromere antibody and immunoglobulin allotypes in scleroderma. (Submitted for publication.) McCarty GA, Rice JR, Bembe ML, Barada Jr FA: Anticentromere antibody: Clinical correlations and association with favorable prognosis in patients with scleroderma variants. Arthritis Rheum 26:1-7, 1983. Maricq HR, Harper FE, Khan MM, et al: Microvascular abnormalities as possible predictors of disease subsets in Raynaud phenomenon and early connective tissue disease. Clin Exp Rheum 1:195-205, 1983. Masi AT, Rodnan GP: Preliminary criteria for the classification of systemic sclerosis (scleroderma). Bull Rheum Dis 31:I-6, 1981. Northway JD, Tan EM: Differentiation of antinuclear antibodies giving speckled staining patterns in immunofluorescence. Clin Immunol Immunopathol 1:140154, 1972. Nakamura RM, Peebles CL, Tan EM: Microhemagglutination test for detection of antibodies to nuclear Sm and ribonucleoprotein antigens in systemic lupus erythema-
27. 28.
29.
30.
31. 32.
33.
34. 35.
36.
37. 38. 39.
40.
4t.
467
tosus and related diseases, Am J Clin Pathol 70:800-807, 1978. Zubler RH, Lambert PH: 12H-Clq binding test for solubIe immune complexes. Ann Rheum Dis 36:27-30, 1977. Kilpatrick JM, Virella G: Statistical analysis of five immune complex screening assays: Patterns of detection in patients with rheumatoid arthritis, systemic lupus erythematosus, infectious endocarditis, and diabetes mellitus. Diagn Immanol 1:57-63, 1983. Tower BB, Clark BR, Rubin RT: Preparation of 12~lpolypeptide hormones for radioimmunoassay using glucose oxidase with lactoperoxidase. Life Sci 21:959-966, 1977. Jablonska S, Chorzelski TP, Blaszczyk M, Glinski M: Immunologic phenomenon in scieroderma, in Beutner EH, editor: Immunopathology of the skin. New York, 1979, John Wiley & Sons, Inc., pp. 363-380. Gerber MA: Immunohistochemical findings in the renal vascular lesions of progressive systemic sclerosis. Hum Pathol 6:343-347, 1975. Salerni R, Rodnan GP, Leon DF, Shaver JA: Pulmonary hypertension in the CREST syndrome variant of progressive systemic sclerosis (seleroderma). Ann Intern Med 86:394-399, 1977. Winkelmann RK, Carapeto FJ, Jordan RE: Direct immunofiuorescence in the diagnosis of scleroderma syn. dromes. Br J Dermatol 96:231, 1977. Connolly SM, Winkelmann RK: Direct immunofluorescent findings in scleroderma syndromes. Acta Derm Venereol (Stockh) 61:29-36, 1981. Reimer G, Huschka J, Kelter J, et al: Immunofluorescence studies in progressive systemic sclerosis (scleroderma) and mixed connective tissue disease. Br J Dermatol 109:27-36, 1983. Chen ZY, Dobson RL, Ainsworth SK, et al: Immunofluorescence in skin specimens from three different biopsy sites in patients with scleroderma. (Submitted for publication.) Seibold JR, Medsger TA, WinkelsteJn A, et al: Immune complexes in progressive systemic sclerosis. Arthritis Rheum 25:1167-1173, 1982. Yanase K, Imamura S: Detection of circulating immune complexes in some skin diseases by platelet aggregation test. Br J Dermatol 100:227-228, 1979. Solling J, Solling K, Jacobsen KN: Circulating immune complexes in lupus erythematosus, scleroderma and dermatomyositis. Acta Derm Venereol (Stockh) 59: 42•-426, 1979. Dowd PN, Kirby JD, Holborow EJ, et al: Detection of immune complexes in systemic sclerosis and Raynaud's phenomenon. Br J Dermatol 105:179-188, 1981. Hughes P, Cunningham J, Day M, et al: Immune complexes in systemic sclerosis: Detection by Clq binding, K-cell inhibition and Raji cell radioimmunoassays. J Clin Lab Immunol 10:133-138, 1983.
I
I I
I I
II
I
Immune complexes and antinuclear, antinucleolar, and anticentromere antibodies in scleroderma* Ze-yi Chen, M.D.,** Gabriel Virella, M.D., Ph.D., Hsiaoho Edward Tung, Ph.D., Sterling K. Ainsworth, Ph.D., Richard M. Silver, M.D., An-Chuan Wang, Ph.D., Mariano F. LaVia, M.D., Hildegard R. Maricq, M.D., and Richard L. Dobson, M.D. Charleston, SC Forty-one patients with various forms of systemic sclerosis (scleroderma) and positive antinuclear antibodies of nucleolar (ten patients), speckled (eleven patients), or centromere pattern (twenty patients) were selected for study of immune complexes by the radioisotope labeled Clq binding and the radioisotope labeled protein A binding methods. The presence of immune complexes was found by the Clq binding assay in sixteen patients (39%) and by a protein A binding assay in eight patients (20%). Overall, 46% of patients (19/41) had immune complexes. A lower incidence of organ involvement and fewer positive results in the screening of serum immune complexes were observed in patients with centromere antibody (35%) than in patients with nucleolar (60%) or speckled pattern (55%). Paiients with immune complexes had higher frequencies of kidney, heart, and muscle involvement and digital ulceration than did patients with no detectable immune complexes, but the differences were not statistically significant. Diffuse skin involvement was not related to the presence of immune complexes. (J AM ACADDERMATOL 11:461-467, 1984.)
Scleroderma (systemic sclerosis, SS) is a connective tissue disease characterized by tightness,
From the Departments of Dermatology (Chen, Dobson), Basic and Clinical Immunology and Microbiology (Virella, Tung, Wang), Pathology (Ainsworth), Medicine (Silver, Maricq), and Laboratory Medicine (LaVia), Medical University of South Carolina. Supported in part by National Institutes of Health Grant AM 31283 from the National Institute of Arthritis, Diabetes, and Digestive and Kidney Diseases. Accepted for publication Feb. 22, 1984. Reprint requests to: Dr. Gabriel Virella, Department of Basic and Clinical Immunology and Microbiology, 171 Ashley Ave., Charleston, SC 29425. *Publication No. 662 from the Department of Basic and Clinical Immunology and Microbiology, Medical University of South Carolina. **A M.U.S.C. Visiting Associate Professor of Dermatology (research) from Shanghai Second Medical College, Shanghai, The People's Republic of China.
thickening, and induration of skin (scleroderma), Raynaud's phenomenon, and systemic involvement affecting a variety of organs. The cause of the disease remains obscure. The presence of vascular abnormalities in the tissues involved 2-4 and the deposition of immune complexes (IC) in the kidneys and skin of some patients suggest that circulating IC may be involved in the pathogenesis of SS. s'6 Although previous reports of IC in SS have produced conflicting results, 7-9 there has been the general suggestion that the presence of IC is related to more severe organ involvement. Antinuclear antibodies (ANA) have been detected by indirect immunofluorescence using tissue culture cells as substrate in almost all patients with SS. t°,11 The use of tissue culture cells as substrate allows easier differentiation of the various nuclear patterns (e.g., centromere, nucleolar, 461
462
Chen et al
and speckled), r2.1~ The association of the nucleolar and speckled patterns with scleroderma has been recognized by several groups, ~4-~7 and the different nuclear patterns appear to be related to different incidences of various clinical features. TM Anticentromere antibody (ACA) has been demonstrated in association with a less severe form of systemic sclerosis, the CREST syndrome (calcinosis, Raynaud's phenomenon, esophageal involvement, sclerodactyly, and telangiectasia). 19-..,2 A significant correlation between the presence of ACA and characteristic capillary patterns has also been observed. 11,2"~ Since previous studies have shown that SS patients with circulating IC may have a greater frequency of visceral involvement, 9 we sought to determine whether the presence of IC was related to different nuclear patterns, a point that had not been previously investigated, to the best of our knowledge. In this study the sera of forty-one patients with various forms of SS, selected on the basis of having speckled, nucleolar, or centromere ANA patterns, were examined for the presence of IC by radioisotope-labeled Clq binding and radioisotope-labeled protein A binding. The correlation of circulating IC with different nuclear patterns and various clinical features was investigated. MATERIAL AND METHODS Selection of patients. Single serum samples from forty-one patients with various forms of SS were included in the study. The patients were selected for study of IC on the basis of the following criteria: (1) patients had been admitted at least once since 1978 to the Medical University of South Carolina Hospital for evaluation of scleroderma and (2) a positive ANA titer with clear nuclear patterns that include centromere, nucleolar, or speckled patterns, as demonstrated by indirect immunofluorescence with human epithelioid cells (HEp-2) as substrate. Only patients with manifestations suggestive of systemic lupus erythematosus, locaiized scleroderma, and Raynaud's phenomenon were excluded from this study. Clinical assessment. Diagnosis of SS was based On the American Rheumatism Association (ARA) criteria. 24 Diffuse scleroderma was characterized by skin invoNement proximal to the elbow and/or the knee joints. Calcinosis and Raynaud's phenomenon were assessed by patient history and clinical examination. Esophageal involvement was evaluated by esophageal
Journal of the American Academyof Dermatology
manometry (18/41), cine-esophagogram (14/41), or both methods (9/41). Pulmonary abnormalities were defined as bibasilar reticular fibrosis on the basis of standard chest roentgenography and/or abnormalities indicated by pulmonary function tests (forced vital capacity <80% of predicted capacity or diffusion coefficient <80%). Cardiac involvement was defined as pericarditis, conduction disturbances, or congestive heart failure as indicated by clinical manifestations, electrocardiography, and echocardiography (36/41), in the absence of other causes. Renal involvement was defined as creatinine clearance <75 ml/min or proteinuria >2 gm/24 hr. Joint involvement was assessed by patient history or clinical examination of joint inflammation. Muscle involvement was defined as proximal muscle weakness in the prescence of one or more of the following: increased serum levels of creatine phosphokinase or aldolase, positive muscle biopsy findings, or a positive electromyogram. The patients were divided into three groups on the basis of the following ANA patterns: centromere, nucleolar, and speckled patterns. Patients with anticentromere antibody (ACA). This group included twenty patients; ACA titers varied from 320 to 10,240. Fifteen of twenty patients fulfilled the preliminary ARA diagnostic criteria for SS ,24 Three of these fifteen patients had diffuse scleroderma. Twelve had complete or incomplete CREST syndrome. Of the remaining five patients, two had Raynaud's phenomenon and telangiectasia; one had Raynaud's phenomenon and esophageal involvement; one had Raynaud's phenomenon with pulmonary and cardiac involvement; and the last one had Raynaud's phenomenon, sclerodactyly, arthritis, and cardiac involvement. Patients With ANA of nucleolar pattern. This group consisted of ten patients; ANA titers were from 320 to 163,840. Nine of ten patients fulfilled the preliminary ARA criteria for systemic sclerosis; five of these nine patients had diffuse skin involvement and four had proximal skin involvement. One patient had Raynaud's phenomenon, sclerodactyly, and esophageal and joint involvement, but these manifestations were not enough to fulfill the diagnostic criteria of SS. Patients with ANA of speckled pattern. This group included eleven patients, all of whom fulfilled the preliminary ARA criteria for SS; ANA titers ranged from 40 to 81,920. Seven of these patients had diffuse seleroderma, and the remaining four patients had skin thickening proximal to the metacarpophalangeal or metatarsophalangeal joints. The sera obtained from all patients included in the study were separated into
Volume 11 Number 3 September, 1984
Immune complexes and antibodies in scleroderma
463
T a b l e I. Comparison of clinical characteristics of patients with speckled, nucleolar; or centromere nuclear pattern Nuclear pattern
Patient No. (F:M) Mean age at study (yr) Mean disease duration (yr) Organ system involved Diffuse skin (trunk and/or proximal extremities) Esophagus Lung Kidney Heart Joint Muscle Digital ulceration Raynaud's phenomenon Laboratory abnormalities Anti-DNA antibody Rheumatoid factor Cryoglobulin Erythrocyte sedimentation rate C3 and/or C4 Hypergammaglobulinemia ANA other than ACA ACA Anti-Sin antibodies [mmunodiffusion Hemagglutination Anti-RNP antibodies Immunodiffusion Hemagglutination
11 (9:2) 39.8 - 15.5" 5.6 _-_ 2.7
10 (8:2) 39.5 --- 14.6 6.5 --- 6.8
7 (64%)'~ 9 10 4 8 9 5 6 10
(82%) (91%) (36%) (73%) (82%) (46%) (55%) (91%)
5 i50%) 7 7 3 6 7 2 5 9
(70%) (70%) (30%) (60%) (70%) (20%) (50%) (90%)
20 (15:5) 48.2 --- 11.1 8.2 ± 8.2 3 (15%) 14 (70%) 10 (50%) 0 (0%) 5 (25%) 6 (30%) 0 (0%) 11 (55%) 20 (100%)
3 (27%) 2 (18%) 0/8 (0%) 5 (46%)
0 (0%) 0 (,0%) 0/8 (0%) 5 (50%)
3 (15%) 2/19 (11%)$ 0/19 (0%) 3/19 (16%)
3/9 (33%) 3/9 (33%) 11 (100%)§ 0 (0%)
2 (20%) 2/9 (22%) 10 (100%)§ 0 (0%)
0/18 (0%) 3/20 (15%) 0 (0%) 20 (100%)§
1 (9%) 3 (27%)
0 (0%) 0 (0%)
0 (0%) 1 (5%)
3 (27%) 4 (36%)
1 (10%) 2 (20%)
1 (5%) 1 (5%)
*Mean ± SD. t% = Numberpositive/numberof each pattern, all patientshave beentested. ~Number positive/numbertested. §By patient selection.
aliquots and stored at - 7 0 ° C for determination of ANA, IC, Sm antigens and ribonucleoprotein (RNP) antibodies. Laboratory tests. The following tests were completed at the time of clinical evaluation: (1) doublestranded deoxynucleic acid antibody test performed by the Farr radioimmunoassay (values >20 units were considered positive); (2) C3 and C4 levels quantified by radioimmunodiffusion or immunonephelometry (the patients' levels were defined as normal or abnormal according to the normal ranges established for each assay technic); (3) rheumatoid factor determined by
latex agglutination (agglutinating titers equal to Or exceeding 1:20 were considered positive); (4)cryoprecipitation determined by incubating a serum sample separated from 37° C clotted blood at 4° C for 72 hours; (5) erythroeyte sedimentation rate performed by the Westergren method; and (6) serum protein levels determined by electrophoresis. Indirect immunofluorescence technics. The ANA tests were done with the use of HEp-2 cells as substrate (Immuno-Concepts Inc., Sacramento, CA). The cell line is an epithelial cell line originally propagated from a human laryngeal carcinoma. The substrate was grown
464
Journal of the American Academy of Dermatology
Chen et al
T a b l e II. Frequency of ICs in patients with speckled, nucleolar, or centromere pattern of A N A Nuclear pattern
Speckled (N = 11) IC assays
No.
Clq method Protein A method Both Clq and protein A methods Either Clq or protein A method
5 3 2 6
]
Nucleolar (N = 10)
%
No?
46 27 19 55
5 2 1 6
on the slide and fixed by the manufacturer with absolute ethanol at 0° C for 6 minutes. The ANA tests were performed according to instructions provided by the manufacturer. The sera were screened at 1:10, 1:20, and 1 : 40 to determine the presence of mixed patterns, but the results were considered positive only when nuclear fluorescence staining was clearly observed with discernible nuclear or nucleolar staining patterns at 1 : 40 dilution. The staining patterns of ANA were photographed with Kodak Ektachrome 400 film. Immunodiffusion. For determination of the anti-Sm and anti-RNP antibodies, all sera were analyzed by immunodiffusion in agarose against extractable nuclear antigens of rabbit thymus (Pel-Freeze Biologicals, Rogers, AR). Immunodiffusion was performed as previously reported. 2~ The precipitin lines of anti-Sm and anti-RNP were confirmed with reference sera kindly supplied by the Centers for Disease Control, Atlanta, GA. Hemagglutination. Anti-Sin and anti-RNP antibodies were also detected by the hemagglutination technic. 2G IC assays. t251-1abeled Clq binding assay. A modification of the method of Zubler and Lambert''7 was used. 28 Clq was purchased from DNW Biologics, San Antonio, TX, and was labeled with 1~5I by the peroxidase-glucose oxidase technic,29 with the use of reagents supplied by Bio-Rad Laboratories (Richmond, CA) and 12~I (NaI), 15.5 mCi/mg specific activity, supplied by Amersham Corp. (Arlington Heights, IL). For a comparison of results obtained in different rtins, results are expressed as a percent of the mean of the values obtained from the simultaneously assayed sera Of six control subjects (the mean being considered as 100%). Using the ninety-fifth percentile of forty-four sera from sixteen healthy control subjects assayed with the same a25I-labeled Clq batch as the sera of patients included in the study, we set the lower limit of positivity at 130%. *~5l-labeled protein A binding assay. This assay is
'" % 50 20 10 60
Centromere (N = 20)
No.[ 6 3 2 7
"Total" (N = 41)
%
No.]
9'0
30 15 10 35
16 8 5 19
39 20 12 46
based on the coprecipitation of 12~I-labeled protein A with soluble IC in the presence of 3% (w/v) polyethylene glycol 6000. Basically, we prepared 1/4.0 dilutions of sera in pH 8.4 borate buffer, 0.1 M, and prepared a mixture containing 0.1 ml of the serum dilution, 0.4 ml of borate buffer, 10/xl of a25I-labeled protein A (30 Ci/g specific activity, obtained from I C N Radioisotopics, Irvine, CA), and 0.5 ml of 6% polyethylene glycol 6000 in borate buffer. The mixtures were incubated overnight at 4 ° to 8° C, and the precipitate was washed once with ice-cold 3% polyethylene glycol. After the supernatant was carefully decanted, the radioactivity in the precipitate was measured and the results were expressed as a percentage of the ratio* relative to the sera of normal control subjects included in the same run, considering as i 0 0 % the average of the normal control subjects' ratio* for each run. Values greater than 153% (ninety-fifth percentile value for a series of sera from nineteen different normal donors) were considered abnormal. RESULTS The major clinical features of our forty-one patients with different ANA patterns are presented in T a b l e I. The female/male ratio was similar for each o f the three ANA groups. Patients with A C A t e n d e d to be older and to have a longer disease duration than patients with either a speckled or a n u c l e o l a r pattern. The centromere pattern was associated with less diffuse skin involvement, as well as lower frequencies of renal, cardiac, m u s c l e , and joint abnormalities. R a y n a u d ' s p h e n o m e n o n (present in all but two patients), digital ulceration, and esophageal dysmotility occurred with n e a r l y Clinical and laboratory
manifestations.
*Ratio of countsper minute in polyethyleneglycolprecipitateto the total counts per minute added.
Volume 11 Number 3 September, 1984
Immune complexes and antibodies in scleroderma
465
Table III. Clinical differences between patients with and without circulating immune complexes (CIC) (detected by either method) CIC results by either method Without CIC (N = 22)
With CIC (N = 19) %
No.
%
7
37
8
36
14' 13 5 11 11 5 13 19
74 68 26 58 58 26 68 100
16 14 2 8 11 2 9 20
73 64 9 36 50 9 41 91
3 3/18? 0/18 8/18
16 17 0 44
3 1/21 0/17 5/22
14 5 0 23
3/17 5 / 18
18 28
2/20 3/20
10 15
No.
Organ system involved Diffuse skin (trunk and/or proximal extremities Esophagus Lung Kidney Heart Joint Muscle Digital ulceration Raynaud's phenomenon Laboratory abnormalities Anti-DNA antibody Rheumatoid factor Cryoglobulin Erythrocyte sedimentation rate C3 and/or C4 Hypergammaglobulinemia Anti-Sm antibodies Imnaunodiffusion Hemagglutination Anti-RNP antibodies Immunodiffusion Hemagglutination
I
0 2
0 11
3 4
16 21
1
5
2
9
2
9
3
14
*Numberpositive/numbertested, all patientshavebeentested. tNumber positive/numbertested.
equal frequency in each of the three ANA groups. Lung disease was noted most frequently in the speckled A N A group (91%) but occurred frequently in the nucleolar and centromere group (70% and 50%, respectively). In terms of laboratory abnormalities, the three A N A groups differed in several respects. The group with a centromere pattern always had normal C3 and/or C4 values and had a lower frequency of an elevated erythrocyte sedimentation rate. Antibodies to deoxyribonucleic acid were detected most frequently in the speckled ANA group. As might be expected, the speckled group had the highest frequency of anti-Sm and antiRNP antibodies as well. R e s u l t s o f s o l u b l e IC screening tests. Serum
IC was detected in nineteen of forty-one patients (46%) with scleroderma by either Clq binding assay or protein A binding assay or both (Table II). The Clq binding assay produced a higher frequency of positive results (39%) than the protein A binding assay (20%). The presence of IC was detected in five patients (12%), as determined by both methods. Correlation of IC with ANA patterns. When the presence of serum IC was compared among the patients with different ANA patterns, the results of IC assays in patients with the nucleolar pattern (60%) were similar to those in patients with the speckled pattern (55%). A lower frequency of positive results was found in patients with ACA (25%) than in patients with either the speckled or
466
Chert et al
the nucleolar pattern, but these differences did not reach statistical significance. Correlation of IC determinations with clinical features, When patients with IC detected by at least one of the two methods were compared with patients who had negative test results for IC by both methods, the former group had a higher incidence of kidney, heart, and muscle involvement and digital ulceration (Table III). Rheumatoid factor positivity, increased erythrocyte sedimentation rate, hypergammaglobulinemia, and anti-RNP antibodies were also observed more frequently in patients with IC. These clinical and laboratory differences were not, however, statistically significant. Results of IC screening by the Clq binding assay also correlated with kidney, heart, and muscle involvement, as well as with digital ulceration. No such correlation was seen with the protein A assay (data not presented). DISCUSSION The efiopathogenesis of SS is poorly understood, but immunologic imbalance is believed to play an important roleJ ° Perivascular deposits of immunoglobulins have been demonstrated in the kidneys of patients studied and in the lungs of some patients with SS. a',3'a Immunoglobulin deposits have also been found at the dermoepidermal junction of some patients with scleroderma. 3z-3~ Recently our group has demonstrated immunoglobulin deposits at the dermoepidermal junction of the nailfold and forearm in some patients with SS. ~,:~6 These findings raise the possibility that soluble IC may be formed and deposited perivascularly in different tissues. A pathogenetic role for IC in scleroderma remains speculative. Some previous studies have detected circulating IC in a large proportion of their patients and have suggested a possible role for IC in the pathogenesis of SS; other studies have detected IC only rarelyY -41 The discrepancies among various studies may relate to the particular IC assays employed, as well as to differences in IC levels among different subsets of scleroderma patients studied. In the present study, the presence of IC was detected in 39% of patients by the Clq binding
Journal of the American Academy of Dermatology
assay and in 20% of patients by the protein A binding assay. In 46% of all our patients with various forms of SS, the presence of circulating IC was detected by either one of these two methods. Seibold et al:~7 found IC in 47% of scleroderma patients tested by two or more assays. Our study indicates that the detection of IC may differ a m o n g patients with different nuclear patterns of A N A in scleroderma; patients with ACA had a lower frequency of detectable circulating IC than did patients with ANA of a nucleolar or a speckled pattern. Since these same patients had a l o w e r frequency of skin and visceral organ involvement, one might speculate that organ system involvement seen in patients with a nucleolar or speckled pattern is due in part to the deposition of IC. H o w ever, as previously noted, a point prevalence survey such as ours does not allow such a conclusion.aT It may, however, explain some of the differences in frequencies of IC reported by various investigators. REFERENCES 1. Rodnan GP: Progressive systemic sclerosis (seleroderma), in McCartyDJ, editor: Arthritis and allied conditions, ed. 9. Philadelphia, 1979, Lea & Febiger, pp. 762-809. 2. Campbell PM, LeRoy EC: Pathogenesis of systemic sclerosis: A vascularhypothesis. Semin Arthritis Rheum 4:351-368, 1975. 3. FleischmajerR, Perlish JS, Shaw KV, Pirozzi D/: Skin capillary changes in early systemic scleroderma. Arch Dermatol 112:1553-1557, 1976. 4. Maricq HR, LeRoyEC: Progressive systemic sclerosis: Disorders of the microcirculation. Clin Rheum Dis S: 81-101, 1979. 5, LapenasD, RodnanGP, CavalloT: Immunopathologyof the renal vascularlesion of progressive systemicsclerosis (scleroderma). Am J Pathol 91:243-258, 1978. 6. Chen ZY, Ainsworth SK, Maricq HR, et al: Immunofluorescence studies in scleroderma. Clin Res 31:612A, 1983. 7. O'LoughlinS, Tappeiner G, Jordon RE: Circulatingimmune complexes in systemic scleroderma and generalized morphia. Dermatologica160:25-30, 1980. 8. CunninghamPH, Andrews BS, Davis JS IV: Immune complexes in progressive systemic sclerosis and mixed connective tissue disease. J Rheumatol7:301-308, 1980. 9. SiminovitchK, KleinM, Pruzanski W, et al: CircuIating immunecomplexesin patients with progressive systemic sclerosis. Arthritis Rheum 25:1174-1179, 1982, 10. BernsteinRM, SteigerwaldJC, Tan EM: Association of antinuclear and antinucleolar antibodies in progressive systemic sclerosis. Clin Exp Immunol 48:43-51, 1982. 11. Chen ZY, Silver RM, Ainsworth SK, et al: Association between fluorescentantinuclearantibodies, capillarypat-
Volume 11 Number 3 September, 1984
12.
13.
14. 15.
16. 17. 18.
19. 20. 21, 22.
23.
24. 25.
26.
hnmune complexes and antibodies in scleroderma
terns and clinical features in scleroderma spectrum disorders. Am J Med. (In press.) Parker MD, Kerby GP: Combined titre and fluorescent pattern of IgG antinuclear antibodies using cultured cell monolayers in evaluating connective tissue diseases. Ann Rheum Dis 33:465-472, 1974. Pisko E, Gallup K, Turner R, et al: Cardiopulmonary manifestations of progressive systemic sclerosis: Associations with circulating immune complexes and fluorescent antinuclear antibodies. Arthritis Rheum 22:518523, 1979. Ritchie RF: Antinucleolar antibodies: Their frequency and diagnostic association. N Engl J Med 282:11741178, 1970. Burnham TK, Fine G, Neblett TR: The immunofluorescent tumor imprint technique. II. The frequency of antinuclear factors in connective tissue diseases and dermaroses. Ann Intern Med 65:9-19, 1966. Rothfield NF, Rodnan GP: Serum antinuclear antibodies in progressive systemic sclerosis (scleroderma). Arthritis Rheum 11:607-617, 1968. Jordan RE, Deheer D, Schroeter A, Winkelmann RK: Antinuclear antibodies: Their significance in scleroderma. Mayo Clin Proc 46:111-113, 1971. Kleinsmith DM, Heinzerling RH, Burnham TK: Antinuclear antibodies as immunologic markers for a benign subset and different clinical characteristics of scleroderma. Arch Dermatol 118:882-885, 1982. Fritzler MJ, Kinsella TD, Garbutt E: The CREST syndrome: A distinct serologic entity with anticentromere antibodies. Am J Med 69:520-526, 1980. Tuffanelli DL, McKeon F, Kleinsmith DM, et al: Anticentromere and anticentriole antibodies in the scleroderma spectrum. Arch Dermatol 119:560-566, 1983. Chen ZY, Fedrick JA, Pandey JP, et al: Anticentromere antibody and immunoglobulin allotypes in scleroderma. (Submitted for publication.) McCarty GA, Rice JR, Bembe ML, Barada Jr FA: Anticentromere antibody: Clinical correlations and association with favorable prognosis in patients with scleroderma variants. Arthritis Rheum 26:1-7, 1983. Maricq HR, Harper FE, Khan MM, et al: Microvascular abnormalities as possible predictors of disease subsets in Raynaud phenomenon and early connective tissue disease. Clin Exp Rheum 1:195-205, 1983. Masi AT, Rodnan GP: Preliminary criteria for the classification of systemic sclerosis (scleroderma). Bull Rheum Dis 31:I-6, 1981. Northway JD, Tan EM: Differentiation of antinuclear antibodies giving speckled staining patterns in immunofluorescence. Clin Immunol Immunopathol 1:140154, 1972. Nakamura RM, Peebles CL, Tan EM: Microhemagglutination test for detection of antibodies to nuclear Sm and ribonucleoprotein antigens in systemic lupus erythema-
27. 28.
29.
30.
31. 32.
33.
34. 35.
36.
37. 38. 39.
40.
4t.
467
tosus and related diseases, Am J Clin Pathol 70:800-807, 1978. Zubler RH, Lambert PH: 12H-Clq binding test for solubIe immune complexes. Ann Rheum Dis 36:27-30, 1977. Kilpatrick JM, Virella G: Statistical analysis of five immune complex screening assays: Patterns of detection in patients with rheumatoid arthritis, systemic lupus erythematosus, infectious endocarditis, and diabetes mellitus. Diagn Immanol 1:57-63, 1983. Tower BB, Clark BR, Rubin RT: Preparation of 12~lpolypeptide hormones for radioimmunoassay using glucose oxidase with lactoperoxidase. Life Sci 21:959-966, 1977. Jablonska S, Chorzelski TP, Blaszczyk M, Glinski M: Immunologic phenomenon in scieroderma, in Beutner EH, editor: Immunopathology of the skin. New York, 1979, John Wiley & Sons, Inc., pp. 363-380. Gerber MA: Immunohistochemical findings in the renal vascular lesions of progressive systemic sclerosis. Hum Pathol 6:343-347, 1975. Salerni R, Rodnan GP, Leon DF, Shaver JA: Pulmonary hypertension in the CREST syndrome variant of progressive systemic sclerosis (seleroderma). Ann Intern Med 86:394-399, 1977. Winkelmann RK, Carapeto FJ, Jordan RE: Direct immunofiuorescence in the diagnosis of scleroderma syn. dromes. Br J Dermatol 96:231, 1977. Connolly SM, Winkelmann RK: Direct immunofluorescent findings in scleroderma syndromes. Acta Derm Venereol (Stockh) 61:29-36, 1981. Reimer G, Huschka J, Kelter J, et al: Immunofluorescence studies in progressive systemic sclerosis (scleroderma) and mixed connective tissue disease. Br J Dermatol 109:27-36, 1983. Chen ZY, Dobson RL, Ainsworth SK, et al: Immunofluorescence in skin specimens from three different biopsy sites in patients with scleroderma. (Submitted for publication.) Seibold JR, Medsger TA, WinkelsteJn A, et al: Immune complexes in progressive systemic sclerosis. Arthritis Rheum 25:1167-1173, 1982. Yanase K, Imamura S: Detection of circulating immune complexes in some skin diseases by platelet aggregation test. Br J Dermatol 100:227-228, 1979. Solling J, Solling K, Jacobsen KN: Circulating immune complexes in lupus erythematosus, scleroderma and dermatomyositis. Acta Derm Venereol (Stockh) 59: 42•-426, 1979. Dowd PN, Kirby JD, Holborow EJ, et al: Detection of immune complexes in systemic sclerosis and Raynaud's phenomenon. Br J Dermatol 105:179-188, 1981. Hughes P, Cunningham J, Day M, et al: Immune complexes in systemic sclerosis: Detection by Clq binding, K-cell inhibition and Raji cell radioimmunoassays. J Clin Lab Immunol 10:133-138, 1983.