Subepidermal autoimmune bullous disease affecting predominantly mucocutaneous junctions and the palms with autoantibodies to BP230 and laminin γ1

Dear Editor, Subepidermal autoimmune bullous diseases are believed to be caused by autoantibodies against proteins located at the dermoepidermal junction.1 2 These include bullous pemphigoid (BP)180, BP230, integrin subunit β4, laminin 332, laminin γ1 and type VII collagen. The clinical features of these diseases are usually related to specific autoantibodies. Autoantibodies against more than one antigen are sometimes detected, possibly as a result of intermolecular epitope spreading.3 It can be challenging to determine which autoantibodies are responsible for primary blister formation. Here we report a case of subepidermal autoimmune bullous disease exhibiting unique clinical features with autoantibodies against BP230 and laminin γ1, without any other known autoantibodies. A woman aged in her 60s presented with small, tense blisters with pruritus on her lips. Three days later, multiple small, tense blisters without erythema were observed at the mucocutaneous junctions of the lips (Fig. 1a), eyes (Fig. 1b) and periumbilical (Fig. 1c), perianal and genital areas, and large blisters were observed on her palms (Fig. 1d). There were tense blisters with slight erythema at the scratched sites on her trunk and arms. She had no history of psoriasis or neurological disorders. Clinical appearance and pathological findings. Multiple blisters without erythema were detected at mucocutaneous junctions, such as the lips (a), eyes (b) and periumbilical region (c). Large blisters had formed on both palms (d). (e, f) A skin biopsy of a small blister on the upper arm showed a subepidermal blister with infiltration of eosinophils and neutrophils. A skin biopsy revealed a subepidermal bulla with infiltration of eosinophils and neutrophils (Fig. 1e,f). Direct immunofluorescence showed IgG and C3 deposits along the basement membrane zone (BMZ) (Fig. 2a), and IgA was negative. Indirect immunofluorescence of 1 mol L−1 NaCl‐split skin detected circulating anti‐BMZ IgG antibodies that reacted with both the epidermal and dermal sides (Fig. 2b). Enzyme‐linked immunosorbent assay (ELISA) was negative for IgG antibodies to the BP180 NC16a domain, but positive for BP230 (index 61·1; normal range < 9). Immunoblotting studies detected IgG antibodies to the dermal 200‐kDa protein (Fig. 2c) and laminin γ1 (Fig. 2d). There was no positive reactivity to epidermal extracts, recombinant proteins of BP180 NC16a or C‐terminal domain, purified human laminin‐332, 120‐kDa linear IgA dermatosis‐1, or the hemidesmosomal fraction from DJM‐1 cells. Immunofluorescence and immunoblotting studies. (a) Direct immunofluorescence revealed IgG deposits along the dermoepidermal junction. (b) Indirect immunofluorescence using 1 mol L−1 NaCl‐split healthy human skin showed the reactivity of the patient's IgG antibodies to both the epidermal and dermal sides of the basement membrane zone. (c) Immunoblotting of dermal extracts detected IgG antibodies to the 200‐kDa protein. (d) Western blotting of laminin 211/221 detected antilaminin γ1 IgG antibodies. EBA, epidermolysis bullosa acquisita, mAb, monoclonal antibody. Based on these results, we diagnosed the patient as having subepidermal autoimmune bullous disease with circulating antibodies against BP230 and laminin γ1. Oral prednisolone at a dose of 1 mg kg−1 per day was effective and the lesions healed without scar formation. However, the disease relapsed when the prednisolone was tapered to 0·5 mg kg−1 per day. After treatment with intravenous methylprednisolone (1000 mg per day) pulse therapy for three consecutive days followed by oral prednisolone 1 mg kg−1 per day, doxycycline 100 mg per day and nicotinamide 1500 mg per day, the lesions quickly resolved and the oral prednisolone was subsequently tapered. There has been no relapse after 4 months. Our case demonstrates the unique localization of tense blisters on the mucocutaneous junctions and palms, with the first reported coexistence of circulating antibodies against two distinctive antigens, BP230 and laminin γ1. The pathogenicity of neither has been clearly demonstrated.4 In addition, there have been no reports of pemphigoid diseases with a similar distribution to that of our case, as BP and anti‐p200/laminin γ1 pemphigoid usually presents tense blisters on the whole body, including the mucous membranes.5 This case can be classified as uncommon‐type localized BP, possibly in relation to anti‐BP230 antibodies, as localized BP commonly affects the limbs.6 Anti‐BP230 antibodies might be associated with the focal involvement in BP regardless of their pathogenicity.7 Despite the intracellular localization of BP230,8 there have been two reported cases of BP reactive only with BP230, although the clinical features were somewhat different from those in our case: one presented with blisters exclusively on the dorsal side of the feet while the other developed blisters over the whole body.9 We hesitate to conclude that this case was typical anti‐p200/laminin γ1 pemphigoid with incidental production of anti‐BP230 antibodies during the course. The distribution of the affected sites was unique from the beginning of disease, and there were no preceding lesions generally distributed, as is common in cases of anti‐p200/laminin γ1 pemphigoid. In addition, the histology of our case showed eosinophil‐dominant infiltration in the dermis, whereas neutrophils are dominant in anti‐p200/laminin γ1 pemphigoid.10 The mechanism of the production of autoantibodies and the unique blister distribution in the present case remain to be addressed. Anti‐BP230 antibodies may be an epiphenomenon of the tissue damage caused by antilaminin γ1 antibodies, or vice versa. In our case, ELISA detected autoantibodies against BP230 while immunoblotting studies did not, probably because these autoantibodies could react only with certain conformations of the nonreduced proteins. It is unlikely that the autoantibodies reacted with the unique conformation of BP230–laminin γ1 complexes, because evidence of their interaction is missing. We could not rule out the possibility that other autoantibodies that could not be detected by current techniques might be pathogenic in our case. In summary, we report a rare case of subepidermal autoimmune bullous disease with autoantibodies to both BP230 and laminin γ1. Our case indicates that a unique autoantibody profile might contribute to the atypical clinical features. Funding sources: none. Conflicts of interest: none declared.