612 ANTINUCLEAR ANTIBODIES IN LOCALIZED SCLERODERA KAZUHIKO TAKEHARA, YASUOKI OROI, YASUHARU NAKABAYASHI, and YASUASA ISHIBASHI When HeLa cells were used as the substrate for detection by the indirect immunofluorescence method, antinuclear antibodies were demonstrated in 16 of 22 (72.7%) sera from patients with localized scleroderma. When mouse kidney sections were used, the positive rate for antinuclear antibodies was 50% (11 of 22). In the 3 subgroups of localized scleroderma, frequencies of antinuclear antibodies on HeLa cells were as follows: morphea, 50% (2 of 4), generalized morphea, 100% (6 of a), linear scleroderma, 67% (8 of 12). Antibodies to centromere, Scl70, nuclear RNP, Sm, and SSB antigens were not detected in any patient with localized scleroderma. The high frequency of antinuclear antibodies in localized scleraderma sera suggests that localized scleroderma is a disease which, though different from diffuse scleroderma, also involves an immunologic abnormality. Scleroderma is a chronic disease of unknown etiology and is divided into two types: localized scleroderma and diffuse scleroderma, or progressive systemic sclerosis (PSS). Tuffanelli and Winkelmann classified the two as follows: localized scleroderma includes From the Departments of Dermatology and Internal edicine and Physical Therapy, Faculty of edicine, University of Tokyo, Tokyo, Japan. Supported in part by Grant in Aid for Cooperative Research No. 56370023 from the inistry of Education, Science and Culture. Kazuhiko Takehara, B: Department of Dermatology; Yasuoki oroi, D: Department of Internal edicine and Physical Therapy: Yasuharu Nakabayashi, D: Department of Dermatology; Yasumasa Ishibashi, D: Department of Dermatology. Address reprint requests to Kazuhiko Takehara. B, Department of Dermatology, Faculty of edicine, University of Tokyo, 731 Hong6, Bunkyoku, Tokyo, 113. Japan. Submitted for publication July 29, 1982; accepted in revised form December 10, 1982. morphea, generalized morphea, or linear scleroderma. Systemic scleroderma includes acrosclerosis or diffuse scleroderma (1). In localized scleroderma, the lesions are usually limited to the skin and to the subcutaneous tissue beneath the cutaneous lesions. The absence of Raynaud s phenomenon, acrosclerosis, and involvement of internal organs differentiates localized scleroderma from diffuse scleroderma, and the prognosis of patients with localized scleroderma is good. orphea is usually characterized by circumscribed, sclerotic plaques with an ivorycolored center and a surrounding violaceous halo. Generalized morphea, a severe form of the local disease, is characterized by widespread skin involvement with multiple indurated plaques, hyperpigmentation, and frequent muscle atrophy. Usually generalized morphea is not associated with internal involvement, but a variety of abnormal immune reactions such as positive band test, positive antinuclear antibody (ANA), positive lupus erythematosus cell phenomenon, and positive rheumatoid factor have been reported in this group of patients (24). Linear scleroderma, a form of the localized disease, appears in a linear bandlike distribution. The lesions are usually unilateral. Frontal or frontoparietal linear scleroderma (en coup de sabre) is characterized by atrophy and a furrow, or depression, that extends below the level of the surrounding skin. Rodnan et al reported also finding some serologic abnormalities such as rheumatoid factor, antinuclear antibody, and antibody to single stranded DNA in linear scleroderma patients (5). Previous studies have shown that antinuclear antibodies have been detected in the sera of 15.0 Arthritis and Rheumatism, Vd. 26, No. 5 (ay 1983)
ANA IN LOCALIZED SCLERODERA 613 Table 1. Frequencies of antinuclear antibodies in localized scleroderma in previous reports Positive/ Reference Substrate* total % positive 6 (1%7) Rat liver 3/20 (2/5)t 15.0 (40.0)t 7 (1%9) Human tumor 1 OD6 38.5 imprint 8 (1975) Rabbit mucosa 20/42 47.6 9 (1975) ND 4/12 33.3 5 (1977) Rat liver 8/30 (7/15)$ 26.7 (46.7)$ 10 (199) ND IU47 (3/9)t 25.5 (33.3)t 11 (1981) ND Ull (1/3)t 18.2 (33.3)t * ND = not described. t Generalized morphea. $ Linear scleroderma. 47.6% of patients with localized scleroderma (51 1, Table 1). In the present report we describe the frequency and specificity of ANA in localized scleroderma, detected by the techniques of immunofluorescence, immunodiffusion, and other methods. ATERIA AND ETHODS Sera. Sera from 22 patients with localized scleroderma were studied. The sera were stored at 70 C for varying periods prior to use. Immunofluorescence. Indirect immunofluorescence studies of the sera were performed on two different substrates: HeLa cells grown directly on glass slides and 4p sections of frozen mouse kidney. These substrates were fixed in acetone for 10 minutes. The conjugate used was a goat antiserum raised against human immunoglobulin (Eiken Chemical Limited, Tokyo). The fluoresceidprotein ratio of the conjugate was 1.46. Intensity of positive staining was graded from 1 + to. When HeLa cells were used as the substrate, a staining at 1 : 64 dilution was regarded as positive; in the case of mouse kidney sections, a 1 + staining at 1 :4 dilution was regarded as positive. The percentage of positivity in normal control populations was 5% (1/20), both on HeLa cells and on mouse kidney sections. Immunodiffusion. Immunodiffusion was carried out with a modified'ouchterlony technique according to a procedure described previously (12). The source of nuclear antigens was lyophilized rabbit thymus acetone powder purchased from Pel Freez, Rogers, AR. A saline extract of this material contained multiple soluble nuclear antigens, including Sm antigen, nuclear RNP (nrnp), Scl70 antigen, and SSB antigen (13). For optimal results, the saline extract was used at a concentration of 20 mg proteidml. Antinative DNA antibody. Antibody to native DNA was determined by the Farr technique (14), using the test kit of Radio Chemical Center (RCC), UK (14). Normal values are under 10.0 unitdml. Antibistone antibody. Antibody to histone was measured with an indirect immunofluorescence method accord ing to the procedure previously described by Tan et al (IS). Total histone (HLY, Worthington Biochemicals, Freehold, NJ) was used. RESULTS Clinical. Twentytwo patients were examined, Seven were males and I5 were females. Their ages ranged from 649 years, with a mean of 18.5. Four of the patients were diagnosed as having morphea, 6 as having generalized morphea, and 12 as having linear scleroderma (Table 2). Immunofluorescence. When HeLa cells were used as the substrate, 16 of 22 sera (72.7%) produced positive nuclear fluorescence (Table 2). Two of 4 sera (50%) from patients with morphea, all 6 sera (100%) from patients with generalized morphea, and 8 of 12 sera (66.7%) from patients with linear scleroderma were positive (Table 3). When mouse kidney sections were used as the substrate, 11 of 22 sera (50%) exhibited nuclear fluorescence. In the sera from patients with generalized morphea and in some from patients with linear scleroderma, homogeneous patterns with chromosomal staining were often observed on HeLa cells (Figure 1). Speckled patterns were observed in 2 sera from patients with linear scleroderma (Figure 2). Nucleolar patterns were also observed in some cases (Figure 3). None of the sera contained anticentromere antibody. On mouse kidney sections, homogeneous patterns were observed in 4 generalized morphea sera. Speckled patterns were found in 1 morphea serum, 1 generalized morphea serum, and 4 linear scleroderma sera. Nucleolar pattern was not observed, except in 1 generalized morphea serum. Immunodiffusion. No precipitating antibodies to Sm, nrnp, Scl70, or SSB antigens were detected in the 22 patients' sera. We also could not find any other precipitating antibody in the 22 sera. Antinative DNA antibody. Abnormal DNA binding activities (> 10.0 units/ml) were measured only in 4 of the 6 sera from patients with generalized morphea. These binding activities were much lower than those in the sera from patients with active systemic lupus erythematosus (usually higher than 100 unitslml) (Table 2). All other sera showed normal DNA binding activities (< 10 unitdml). Antihistone antibody. Antibody to histone was not detected by the reconstitution method in any of the 11 sera which had produced positive nuclear staining on mouse kidney sections before 0.1N HCI extraction. Nine sera did not produce any kind of nuclear
614 TAKEHARA ET AL Table 2. Serologic studies in patients with localized scleroderma Antinuclear antibody by immunofluorescencet Type of DNA HeLa cells localized ouse binding Patient Ageisex scleroderma* Nucleus Chromosome Nucleolus kidney activity$ 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 6lF I7l 8lF 28l 6/F 22lF 16l 16/ 49lF 18lF 91F 16lF 41lF 1?IF 7/F 101F 61F 201F 151 43l 28l I I/F H2+ HI + H2+ sl+ SI + H2 + H1+ % positive7 54.5 54.5 * = morphea; = generalized morphea; = linear scleroderma. t H = homogeneous; S = speckled: N = nucleolar. $ Normal <10.0 unitslml. Reciprocal endpoint titer. 7l Total positive = 16/22 (72.7%) 2+ 2+ 1+ I.o S165 8.0 I.o H1,0245 29.0 H1,0245 14.0 H 1,0245 7.0 N1.0241 11.0 S165 9.0 H1.0245 18.0 9.0 6.0 5.0 S16P 7.0 S165 3.0 S2565 6.0 S165 2.0 6.0 3.0 50 18.2 stainings after O.1N HCI extraction or after the addition of purified histones to the acidextracted mouse kidney sections. However, 2 generalized morphea sera which produced homogeneous patterns before 0.1 N HCI extraction produced clumpy patterns after 0.1N HCI extraction; these patterns remained after the addition of purified histones to the acidiextracted mouse kidney sections. These sera showed elevated DNA binding activities. DISCUSSION It has been demonstrated that in an indirect immunofluorescence test for ANA, the sensitivity varies with the type of substrate used, and cultured Table 3. Antinuclear antibodies in clinical types of localized scleroderma Substrate Type of localized HeLa ouse scleroderma cells* kidnev* orphea 214 (50) 114 (25) Generalized morphea 616 ( 100) 616 (100) Linear scleroderma 8/12 (66.7) 4/12 (33.3) Total 16/22 (72.7) 11122 (50) * No. positiveltotal (% positive). Figure 1. Homogeneous patterns with chromosomal staining, produced by generalized morphea sera on HeLa cells.
ANA IN LOCALIZED SCLERODERA 615 Figure 2. Combination of speckled and nucleolar patterns, produced by linear scleroderma sera on HeLa cells. cell substrates generally show greater sensitivity than tissue sections (13,14,16). In this study cultured HeLa cells were used in addition to mouse kidney sections. The frequency of ANA on HeLa cells in the present study, 72.7%, was much higher than the frequencies reported in previous studies (51 1, Table 1). According to the previous studies, the frequencies of ANA in generalized morphea were higher than those in other types of localized scleroderma; in our study all the sera from patients with generalized morphea produced positive staining. In other types of localized scleroderma, namely in morphea and linear scleroderma, 10 of 16 sera (62.5%) produced positive staining on HeLa cells. This frequency was also higher than those in the previous reports, while the frequency of ANA on mouse kidney sections in the present study, 5 of 16 (31.25%), was almost equal to the frequencies in the previous reports. Therefore it seems that the high frequency of ANA on HeLa cells in the present study was mainly due to the sensitivities of the substrates used, and not to the selection of patients. The high frequency of ANA in the sera of patients with localized scleroderma suggests that localized scleroderma is a disease which involves an immunologic abnormality. In generalized morphea, the titers of ANA on mouse kidney sections were higher than those in other types of localized scleroderma. Also, elevated DNA binding activities were present only in generalized morphea. Therefore it seems ties, such as systemic lupus erythematousus or progressive systemic sclerosis. However, none of our patients with generalized morphea had skin rash, Raynaud s phenomenon, arthralgia, acrosclerosis, pulmonary fibrosis, or esophageal involvement. Only 1, patient 6, had Sjogren s syndrome. She complained of dry mouth and dry eyes, and the disease was confirmed by Schirmer s test, Rose Bengal test, and sialography. The prevalence of association with clinical or subclinical Sjogren s SYRdrome in localized scleroderma patients is unknown, and we are now investigating this question. Recently Tan et al(l7) reported that ANA were demonstrated in the sera of 96% of patients with progressive systemic sclerosis. In progressive systemic sclerosis, 3 types of antibodies appeared to be highly specific: antibody to Scl70 antigen, antibody to centromere, and antinucleolar antibody (17,18). Antibodies to nrnp were also occasionally detected. In the localized scleroderma patients in the present study, antibody to Scl70 antigen, centromere, nrnp, Sm antigen, and SSB antigen were not detected. Antinucleolar antibodies were detected in 9 sera on HeLa cells, but only in 1 serum on mouse kidney sections. Therefore, except for antinucleolar antibody, ANA in localized scleroderma appear to be different from those in progressive systemic sclerosis. Analysis and comparison of nucleolar antigens detected in sera from localized scleroderma patients and from systemic scleroderma patients would be another interesting point to be explored in the future. In the previous reports, specificity of ANA in localized scleroderma was not well discussed, and only Nishikawa et a1 (1 1) likely that generalized morphea may be to Figure 3. Nucleolar patterns, produced by generalized morphea diseases with more distinct immunologic abnormali sera on HeLa cells.
616 TAKEHARA ET AL mentioned that antibody to ScI70 antigen was not detected, in 9 sera from patients with localized scleroderma. Homogeneous patterns with chromosomal staining were observed in 10 sera on HeLa cells. Usually several types of ANA (antibodies to native DNA, histone, and Scl70 antigen, for example), produce homogeneous patterns with chromosomal staining. Antibodies to histone and Scl70 antigen were not detected in any of the sera we studied. Antibody to native DNA, i.e., elevated DNA binding activity, was detected in 4 of 6 sera from patients with generalized morphea, but in much lower levels than those in sera from patients with active systemic lupus erythematosus. The slight elevatiorls of DNA binding in these patients may partially explain the homogeneous stainings on HeLa cells and on mouse kidney sections. Antibodies detected in the sera from the patients with localized scleroderma, especially morphea and linear scleroderma, seemed to be against unknown nuclear antigens different from those in progressive systemic sclerosis. Heterogeneity of ANA found in localized scleroderma patients might suggest a heterogeneous immunologic background comparable with that in the diffuse type of scleroderma, and might be useful in classification and understanding of localized scleroderma. The chemical nature, apart from DNA, of antigens stained on cultured cell nuclei by the sera from the patients studied remains unknown, and requires further investigation. REFERENCES 1. Tuffanelli DL, Winkelmann RK: Systemic scleroderma. Arch Dermatol 84:359371, 1955 2. Tanioku K, Fukushiro S: Scleroderma. especially involving generalized morphea. Jpn Clin 31:4451, 1973 (in Japanese) 3. itchell AJ. Rush LJ, Diaz LA: Circumscribed scleroderma with immunologic evidence of systemic lupus erythematosus. Arch Dermatol 1 16:6973, 1980 4. King DF, Dore RK, Gilbert DJ, Gurevitch AW, Hirose F: Generalized morphea with peripheral eosinophilia. fascitis and myositis. Int J Dermatol 19:149153, 1980 5. Rodnan GP, Lipinski E, Rabin BS, Reichlin : Eosinophilia and serologic abnormalities in linear localized scleroderma (abstract). Arthritis Rheum 20: 133, 1977 6. Rowell NR, Beck JS: The diagnostic value of an antinuclear antibody test in clinical dermatology. Arch Dermato1 %:290295, 1967 7. Burnham TK, Neblett TR, Fine G, Paula B: The immunofluorescent tumor imprint technique. Arch Dermatol 99:611616. 1969 8. Barthelmes H: Das Verhalten der antinuklearen Faktoren bei Sklerodermie. Dermatol onatsschr 161:53& 544, 1975 9. Scarola JA, Shulman LE: Serologic abnormalities and their significance in localized scleroderma (abstract). Arthritis Rheum 18526, 1975 10. Jablonska S: Immunopathology of the Skin. Second edition. New York, Wiley edical Publications, 1979, p 369 I I. Nishikawa T, Kurihara S, Sugiura A, Shimizu H, Tojo T: Localized scleroderma and PSS specific antibody to Og antigen. Jpn J Dermatol 91:859860, 1981 (in Japanese) 12. Northway JD, Tan E: Differentiation of antinuclear antibodies giving speckled staining patterns in immunofluoiescence. Clin Immunol Immunopathol 1 : 140154, 1972 13. Kurata N, Tan E: Identification of antibodies to nuclear acidic atltigens by counterimmunoelectrophoresis. Arthritis Rheum 19576580, 1976 14. Holian J, Griffiths ID, Glass DN, aini RN, Scott JT: Human antidna antibody: reference standards for diagnosis and management of systemic lupus erythematosus. Ann Rheum Dis 34~438443, 1975 15. Tan E, Robinson J, Robitaille P: Studies on antibodies to histone by immunofluorescence. Scand J Immunol 5:811818, 1976 16. Hohan N, Eckert HC, Stewart J: Evaluation of cellular substrates for antinuclear antibody determination. J Clin icrobiol 5:4245, 1975 17. Tan E, Rodnan GP, Garcia I, oroi Y, Fritzler J, Peebles C: Diversity of antinuclear antibodies in progressive systemic sclerosis: anticentromere antibody and its relationship to CREST syndrome. Arthritis Rheum 23:617625, 1980 18. oroi Y, Peebles C, Fritzler J, Steigerwald J, Tan E: Autoantibody to centromere (Kinetochore) in scleroderma sera. Proc Natl Acad Sci USA 77:1627 1631, 1980