Epidermal nuclear CIq deposits in a patient with amyopathic dermatomyositis

ARTICLE

Various immunofluorescent (IF) patterns have been found in the skin lesions of autoimmune diseases in direct IF studies. Immunoglobulin (Ig) and complement components deposit at the dermoepidermal junction (DEJ) of patients with lupus erythematosus [1], SS-A/Ro dust-like deposits are present on the epidermal cells in patients with subacute cutaneous lupus erythematosus (SCLE) [2] and epidermal nuclear staining of Ig is characteristically observed in patients with mixed connective tissue disease (MCTD) [3]. These IF staining patterns are mostly disease-specific, and are helpful for differential diagnosis. In contrast, no specific IF staining patterns have been reported in patients with dermatomyositis or scleroderma [4]. Recently, we encountered a case of amyopathic dermatomyositis, in which C1q deposits on the epidermal nucleus were found as well as Ig deposits at the DEJ.

Case report

A 76-year-old Japanese woman had a 2-month history of general malaise, fever and body weight loss. One month before the onset of these symptoms she had noticed erythematous lesions on the dorsal surface of her hands and fingers, and had complained of dyspnea. Interstitial pneumonia was diagnosed by a physician. She subsequently visited our clinic in May 1995.

On admission, infiltrated erythematous lesions and hyperkeratotic erythematous lesions were observed on the dorsal sides of her fingers and hands, and on her elbows (Fig. 1). Hyperkeratotic erythematous lesions were also apparent on both knees. She complained of slight itching on the erythematous lesions. There was no evidence of eruptions on her face or other regions. Abnormal results of blood analysis included 50 mm/hr for the erythrocyte sedimentation rate and 567 IU/L for LDH (normal range 101-193). The CPK, aldolase, GOT and GPT values were within normal limits throughout repeated examinations. Serological tests were negative for nuclear components, Jo-1, DNA, RNP, Topo-1, Ro, La and centromere. Respiratory function tests revealed a pattern characteristic of interstitial pneumonia, which was also confirmed by chest X-ray and CT scan. Electronic muscle examination showed no specific findings. Neither chest X-ray non CT scan showed pathological changes of muscle involvement.

Light microscopy was used to examine the skin lesions and muscle tissues from the dorsal site of the thigh. The infiltrated erythematous lesion showed hyperkeratosis, epidermal atrophy, liquefaction changes, vasodilatation and perivascular mononuclear cell infiltration. The hyperkeratotic erythematous lesion revealed hyperkeratosis, hypergranulosis, acanthosis and perivascular mononuclear cell infiltration. Muscle specimens showed no particular findings.

Direct IF examination of infiltrated erythematous lesions revealed C1q deposits on the epidermal nucleus (Fig. 2) and fibrinogen at the DEJ, and examination of hyperkeratotic erythematous lesions showed linear deposits of Ig G and Ig A at the DEJ but not in the nuclei of epidermal cells (Fig. 2). When FITC-labeled anti-human C1q antiserum (MBL, Nagoya, Japan) was absorbed with various concentrations of purified C1q which was isolated by the method of Yonemasu and Stroud [5] according to our previous report [6], no specific pattern of staining was detected in the examined specimens. Similar findings were obtained when FITC-labeled anti-human C1q antiserum was purchased from Dako, Kyoto, Japan.

Three courses of prednisolone pulse therapy, oral prednisolone (30 mg/day) and cyclophosphamide (50 mg/day) were administered, but these treatments were not effective. The patient died in November 1995 of severe interstitial pneumonia. During the clinical course, myopathic findings were absent, based on the clinical evaluations including chest X-ray and CT and various serological examinations.

Direct IF findings in cases of dermatomyositis

Twenty-six cases of dermatomyositis were collected from the IF units of our department, and examined by a direct IF method. Of the 26 cases, IgG deposits at the DEJ were found in 1 case, IgM deposits on blood vessels in 1 case, IgM and/or C3 on colloid bodies were observed in 2 cases. Nuclear staining was detected only in the present case. In 19 cases with MCTD, positive findings were observed in 12 cases. Nuclear IgG deposits were found in 9 cases, nuclear C1q deposits were found in 1 case, IgM deposits at the DEJ were found in 6 cases, IgG at the DEJ was found in 1 case and complement components at the DEJ were detected in 3 cases. The results in lupus erythematosus patients have been described in a previous report [7], which is similar to those reported elsewhere [1].

Discussion

The deposition of immunoreactants in the epidermal nucleus has been reported in several autoimmune diseases. Approximately 1% of examined specimens showed this staining pattern based on the recent report of Burrows et al. [8]. Although it had been controversial whether or not epidermal cell nuclear staining actually occurs in vivo, recent findings suggest the possibility of in vivo binding such as the penetration of antibodies into live human mononuclear cells through Fc receptors [9], the in vivo binding of anti-RNP and anti-DNA antibodies to a cell suspension of live keratinocytes [10] and the establishment of a spontaneous mouse model which show in vivo epidermal nuclear deposits of IgG [11].

IgG epidermal cell nuclear staining has been demonstrated in the skin lesions of most patients with MCTD and in one-third to one half of the patients with SLE [3, 8, 12]. In SCLE, epidermal IgG deposition is frequently observed in both lesional and non-lesional skin [2] and specific anti-SS-A/Ro binding has been reported to occur in the nucleus as well as the cytoplasm of suction-blister skin specimens [13]. Only complement deposition of the epidermal nucleus is extremely rare [8].

In cases of dermatomyositis, epidermal nuclear staining of IgG is rarely observed, but an exceptionally high incidence of positive Ig staining has been reported by Chen et al. [14] who found positive nuclear deposits in 2 out of 9 cases. The biological meaning of epidermal C1q deposition is still obscure in the present case. Since immunoglobulins were not in the lesional development, we speculate that only the deposition of C1q reflected the activation of the classical pathway of the complement system. It is notable however, that C1q deposition was found in active erythematous lesions but not in inactively hyperkeratotic lesions, which suggest the close association with the development of erythema at least in this patient. The collection of more similar patients will be needed for better understanding of direct in vivo interaction between nucleus and immunoreactants in dermatomyositis.

CONCLUSION

Acknowledgments

This work was in part supported by Japanese Ministry of Education, Science, Sports and Culture.

REFERENCES

1. Dahl MV, Gilliam JN. Direct immunofluorescence in lupus erythematosus. In : Beutner EH, Chorzelski TP, Kumar V, eds. Immunopathology of the skin, 3rd ed. New York : John Wiley & Sons, 1987: 499-518.

2. David-Bajar KM, Bennion SD, DeSpain JD, Golitz LE, Lee LA. The clinical, histologic, and immunofluorescent distinctions between subacute cutaneous lupus erythematosus and discoid lupus erythematosus. J Invest Dermatol 1992; 99: 251-7.

3. Gilliam JN, Prystowsky SD. Mixed connective tissue disease syndrome. Arch Dermatol 1977; 113: 583-7.

4. Targoff IN. Dermatomyositis and polymyositis. Curr Probl Dermatol 1991; 3: 134-80.

5. Yonemasu K, Stroud RM. C1q rapid purification method for preparation of monospecific antisera and for biological studies. J Immunol 1971; 106: 304-13.

6. Furukawa F, Ozaki M, Imamura S, Yoshida H, Pinrat A, Hamashima Y. Associations of circulating immune complexes, clinical activity, and bacterial index in Japanese patients with leprosy. Arch Dermatol Res 1982; 274: 185-8.

7. Furukawa F, Kashihara S, Danno K, Oowatari F, Imamura S. Statistics of immunofluorescent findings of various skin diseases. Acta Dermatol-Kyoto 1991; 86: 415-22 (in Japanese).

8. Burrows NP, Bhogal BS, Jones RR, Black MM. Clinicopathological significance of cutaneous epidermal nuclear staining by direct immunofluorescence. J Cutan Pathol 1993; 20: 159-62.

9. Alarcon-Segovia D, Ruiz-Arguelles A, Fishbein E. Antibody to nuclear ribonucleoprotein penetrates live human mononuclear cells through Fc receptors. Nature 1978; 271: 67-9.

10. Galoppin L, Saurat JH. In vitro study of the binding of antiribonucleoprotein antibodies to the nucleus of isolated living keratinocytes. J Invest Dermatol 1981; 76: 264-7.

11. Furukawa F, Kanauchi H, Wakita H, Tokura Y, Tachibana T, Hori-
guchi Y, Imamura S, Ozaki S, Takigawa M. Spontaneous autoimmune skin lesions of MRL/n mice: autoimmune disease-prone genetic background in relation to Fas-defect MRL/lpr. J Invest Dermatol 1996; 107: 95-100.

12. Velthuis PJ, Kater L, van der Tweel I, Meyling FG, Derksen RHWM, Hene RJ, van Geutselaar JA, de la Faille HB. In vivo antinuclear antibody of the skin: diagnostic significance and association with selective antinuclear antibodies. Ann Rheum Dis 1990; 49: 163-7.

13. Furukawa F, Kashihara-Sawami M, Lyons MB, Norris DA. Binding of antibodies to the extractable nuclear antigens SS-A/Ro and SS-B/La is induced on the surface of human keratinocytes by ultraviolet light (UVL): implications for the pathogenesis of photosensitive cutaneous lupus. J Invest Dermatol 1990; 94: 77-85.

14. Chen ZI, Dobson RL, Ainsworth SK, Silver RM, Maricq HR. Epidermal nuclear immmunofluorescence: serological correlations supporting an in vivo reaction. Br J Dermatol 1985; 112: 15-22.