Blister fluid for the diagnosis of subepidermal immunobullous diseases: a comparative study of basement membrane zone autoantibodies detected in blister fluid and serum

The subepidermal immunobullous diseases bullous pemphigoid (BP), cicatricial pemphigoid (CP), pemphigoid gestationis (PG) and linear IgA disease (LAD) are characterized by circulating and in vivo deposition of antibodies to antigens in the cutaneous basement membrane zone (BMZ). Indirect immunofluorescence (IMF) of serum is a routine diagnostic test to detect circulating BMZ antibodies in these diseases. We have compared the titres of IgG and IgA and their subclasses, also of IgM and IgE BMZ antibodies in serum and aspirated blister fluid in 35 adult patients with subepidermal immunobullous diseases: BP ( n  = 30), PG ( n  = 2), CP ( n  = 1), and LAD ( n  = 2), by indirect IMF on intact and salt-split skin. The antibody titre in blister fluid was the same or one dilution less than serum in most cases and there was no significant difference between these results ( P  > 0.05). IgG1 and IgG4 were the predominant subclasses in both blister fluid and serum in BP. Indirect IMF of serum and blister fluid was also carried out on trypsinized epidermal cells in a subgroup of patients with BP ( n  = 19). Typical polar fluorescence was obtained in all 14 cases which had positive indirect IMF on intact and split skin. Our findings demonstrate that blister fluid can be used as an alternative to serum for indirect IMF in subepidermal immunobullous diseases. This avoids the need for venesection and has a practical application in children and those with poor venous access.     

表皮下大疱病的鉴别诊断和抗原表达区域性差别的研究

通过间接免疫荧光和盐裂皮损周围皮肤直接免疫荧光（简称盐裂ＤＩＦ），分别研究正常人皮肤、类天疱疮（ＢＰ）及获得性大疱性表皮松解症（ＥＢＡ）抗原表达的区域性差别和表皮下大疱病鉴别诊断。 窝、肘窝、上背、下背、股内侧和下腹部皮肤ＢＰ抗原表达率较高；膝、阳窝、足背、肘、肘窝和下腹部皮肤ＥＢＡ抗原表达率较高。皮肤ＤＩＦ显示２５例表皮下大疱病中１６例（６４％）基底膜带有Ｃ３或ＩｇＧ或伴Ｃ３和ＩｇＡ沉积；盐裂ＤＩＦ表明２５例（１００％）均有ＩｇＧ或伴Ｃ３和ＩｇＡ沉积在表皮侧或真皮侧。结果提示，ＢＰ抗原高表达率与皮损好发部位相一致；ＥＢＡ抗原高表达率一部分与皮损好发部位一致。盐裂ＤＩＦ不仅提高ＤＩＦ阳性率，而且根据免疫反应物沉积部位可以鉴别出ＢＰ与ＥＢＡ以及大疱性系统性红斑狼疮。     

U-serrated immunodeposition pattern differentiates type VII collagen targeting bullous diseases from other subepidermal bullous autoimmune diseases

BACKGROUND: Epidermolysis bullosa acquisita (EBA) can be differentiated from other subepidermal bullous diseases by sophisticated techniques such as immunoelectron microscopy, salt-split skin antigen mapping, fluorescence overlay antigen mapping, immunoblot and enzyme-linked immunosorbent assay. OBJECTIVES: To determine whether the diagnosis can also be made by routine direct immunofluorescence microscopy. METHODS: We studied frozen skin biopsies from 157 patients with various subepidermal immunobullous diseases. RESULTS: We found three distinct 'linear' fluorescence patterns at the basement membrane zone: true linear, n-serrated and u-serrated. The true linear pattern, often seen in conjunction with either the n- or the u-serrated pattern, was found in any subepidermal immunobullous disease with nongranular depositions. In bullous pemphigoid, mucous membrane pemphigoid, antiepiligrin cicatricial pemphigoid, p200 pemphigoid and linear IgA disease the n-serrated pattern was found, corresponding with depositions located in hemidesmosomes, lamina lucida or lamina densa. However, in EBA and bullous systemic lupus erythematosus the u-serrated staining pattern was seen, corresponding with the ultralocalization of type VII collagen in the sublamina densa zone. The diagnosis of EBA with IgG or IgA autoantibodies directed against type VII collagen was confirmed by immunoelectron microscopy, salt-split skin antigen mapping, fluorescence overlay antigen mapping or immunoblotting. CONCLUSIONS: Using this pattern recognition by direct immunofluorescence microscopy we discovered several cases of EBA which would otherwise have been erroneously diagnosed as a form of pemphigoid or linear IgA disease.     

Identification of the target antigen in chronic bullous disease of childhood and linear IgA disease of adults

Disease-associated autoantibodies to basement membrane proteins have been used to characterize structural components of the epidermal basement membrane such as bullous pemphigoid (BP) antigen and epidermolysis bullosa acquisita (EBA) antigen (type VII collagen). The autoimmune bullous diseases characterized by IgA autoantibodies to the basement membrane zone (BMZ), i.e. linear IgA disease of adults (LAD) and chronic bullous disease of childhood (CBDC) may have circulating antibodies. Previous studies of tissue distribution and ultrastructural binding have suggested that the LAD and CBDC antigens are similar, if not identical, and differ from the target antigens of the other bullous diseases. We present the molecular characterization of the LAD/CBDC antigens by Western blotting of a large series of antisera. Seven of 33 sera (21%) were positive on immunoblotting and bound to the same antigen which has a molecular weight (MW) of 285 kDa. Using both defined polyclonal antisera to BP and LH 7.2 monoclonal antibody to type VII collagen (carboxy terminal) we have shown that the LAD and CBDC antisera both bind to an identical molecular weight protein which clearly differs from both the BP and EBA (type VII collagen) antigens. Although detectable in dermal tissue extracts like EBA, the MW of 285 kDa is heavier than type VII collagen (250 kDa, in our system, using non-collagenous standards). This study confirms the identity of LAD and CBDC antigens to be the same and to differ from previously described basement membrane proteins.     

Defining target antigens in linear IgA disease using skin from subjects with inherited epidermolysis bullosa as a substrate for indirect immunofluorescence microscopy

Identification of target antigens in immunobullous disorders usually involves laborious techniques such as immunoblotting and immunoprecipitation, which do not always provide conclusive data. This is particularly true of linear IgA disease (LAD) in which the target antigen has often proved difficult to identify. As an alternative means of antigen identification in five adult patients with LAD, we performed indirect immunofluorescence (IIF) microscopy on a panel of skin samples taken from subjects with different forms of inherited epidermolysis bullosa (EB). Skin samples were selected that showed a complete absence of immunostaining for a specific basement membrane zone (BMZ) molecule (type VII collagen, laminin 5 or the 180-kDa bullous pemphigoid antigen BP180). In each case, the underlying genetic mutations had been defined and shown to consist of premature termination codons on both alleles of the particular gene, resulting in total ablation of the encoded protein. Two epidermal-binding LAD sera showed BMZ fluorescence on all substrates except BP180-deficient skin, suggesting that the target antigen was BP180, or a closely related molecule. In contrast, two dermal-binding LAD sera were positive on all substrates except the type VII collagen-deficient skin, suggesting that the target antigen was likely to be type VII collagen. One LAD serum sample, which showed combined dermal and epidermal fluorescence on normal salt-split skin, was also positive on all substrates tested, suggesting a target antigen other than type VII collagen, laminin 5 or BP180. The study confirms that LAD is a heterogeneous disorder and illustrates that IIF using a panel of skin samples which lack specific BMZ molecules, taken from subjects with inherited EB, is a relatively simple and useful tool to help identify target antigens in immunobullous disorders.     

The localization of the binding site of circulating IgA antibodies in linear IgA disease of adults, chronic bullous disease ofchildhood and childhood cicatricial pemphigoid

Biopsies from suction blisters raised in three normal volunteers were used as substrate in the indirect immunofluorescence technique to determine the binding site of circulating IgA antibodies in serum from three patients with adult linear IgA disease (LAD), nine with chronic bullous disease of childhood (CBDC), three with childhood cicatricial pemphigoid (CCP), and four with bullous pemphigoid (BP). Direct immunofluorescence was done using suction blisters raised in two patients with LAD and one with CCP. The circulating IgA antibodies in LAD and CBDC bound mainly to the roof of the blister but also to the base, and in CCP they bound only to the base of the blister. The circulating IgG antibodies in BP bound to the roof and base of the blister. These results demonstrate that the antigens in the various linear IgA dermatoses are heterogenous and are localized at different sites. The LAD and CBDC antigens are present in the lamina lucida, and the antigen in CCP is associated with the basal lamina.     

大疱性系统性红斑狼疮的皮肤基底膜带相关抗原

免疫印迹和盐裂皮肤间接免疫荧光检测 ５例大疱性系统红斑狼疮（ＢＳＬＥ）血清，对照为 ５例获得性大疤性表皮松解症（ＥＢＡ）、２０例类天疱疮（ＢＰ）、２０例ＳＬＥ和１０例正常人血清。结果表明，３例（３／５）ＢＳＬＥ血清结合盐裂皮肤真皮侧和真皮提取物中２９０ ０００抗原，其中２例ＢＳＬＥ血清也结合表皮提取物中 １６５ ０００抗原，结果与 ＥＢＡ和部分ＢＰ血清相同。ＳＬＥ血清未结合 ２９０ ０００和 １６５ ０００抗原。提示ＢＳＬＥ血清中存在ＥＢＡ和ＢＰ抗体，推测ＥＢＡ和ＢＰ抗原可能是ＢＳＬＥ的皮肤基底膜带相关抗原。     

Linear IgA dermatosis with IgA and IgG autoantibodies to the 180 kDa bullous pemphigoid antigen (BP180): evidence for a distinct subtype

BACKGROUND: Autoantibodies in linear immunoglobulin A (IgA) disease (LAD) are reported to be of IgA class and directed against a 97-120 kDa epidermal antigen. METHODS: We report a 39-year-old woman with clinical features of LAD and with circulating IgA and IgG autoantibodies to the 180 kDa bullous pemphigoid antigen (BP180). RESULTS: Histopathology of lesional skin revealed a subepidermal blister with mixed inflammatory cell infiltrate. Direct immunofluorescence of perilesional skin showed linear deposits of IgA along the dermal-epidermal junction. The antigen specificity of the patient's circulating antibodies was determined by Western blotting and enzyme-linked immunoabsorbent assay (ELISA) using various antigen sources, including cultured human keratinocytes, dermal protein lysates, and purified laminin-5, as well as proteins corresponding to BP180, the 230 kDa bullous pemphigoid antigen (BP230), laminin-5 subunits, and collagen IV alpha1-alpha6 chains. IgA and IgG antibodies in the patient's serum were directed against BP180, and no IgA or IgG reactivity was found against the other skin antigens. CONCLUSIONS: These data provide evidence for the presence of a subtype of LAD with dual IgA and IgG autoimmune response to BP180.     

Anatomical distribution and immunological characteristics of epidermolysis bullosa acquisita antigen and bullous pemphigoid antigen

A Japanese patient with epidermolysis bullosa acquisita (EBA) was autopsied, and direct immunofluorescence (DIF) testing was performed. Using this patient's serum (EBA serum) and three bullous pemphigoid (BP) sera, the anatomical distribution and immunological characteristics of EBA antigen and BP antigen were investigated by indirect immunofluorescence (IIF). EBA antigen showed the same anatomical distribution as BP antigen in DIF and IIF studies; both antigens were limited to the skin, tongue, oesophagus, trachea, cornea and bladder. EBA antigen was located on the dermal side of both NaCl and PBS-separated skin, whereas BP antigen was limited to the epidermal side. Ethanol fixation abrogated the antigenic stability of BP antigen, but not that of EBA antigen. No difference was found when acetone or formalin fixation was used. The separation methods and prefixation in ethanol could be useful techniques applicable to the classification of the bullous disorders which manifest circulating anti-BMZ antibodies.     

Linear IgA Bullous Dermatosis with Circulating IgG Autoantibodies to the 230 kD Epidermal Antigen

The patient was a 54-year-old woman with wide-spread bullous lesions on her trunk and oral mucosa. Histologic examination revealed a subepidermal blister with infiltration of neutrophils and eosinophils. Direct immunofluorescence showed an exclusively IgA deposition at the basement membrane zone (BMZ). Indirect immunofluorescence showed that the blister fluid, but not the serum, contained IgG antibodies against the BMZ antigen on the epidermal side of salt-split skin. Using immunoblot analysis with normal human epidermal extracts, both serum and blister fluid reacted with the 230 kD epidermal antigen. Using colloidal gold and direct immunoelectron microscopy, IgA deposition was detected in the lamina lucida. Clinically, the skin lesions responded well to dapsone. We diagnosed this case as linear IgA bullous dermatosis (LABD) with IgG class circulating autoantibodies against the epidermal 230 kD antigen. These antibodies were considered to be secondary to the damage to the epidermal basal keratinocyte in this case.     

Respective contribution of neutrophil elastase and matrix metalloproteinase 9 in the degradation of BP180 (type XVII collagen) in human bullous pemphigoid.

Bullous pemphigoid is a blistering disorder associated with autoantibodies directed against two components of hemidesmosomes, BP180 and BP230. Autoantibodies to the extracellular collagenous domain of BP180 are thought to play a key role in the pathogenesis of the disease. In a murine model of bullous pemphigoid, neutrophil elastase and 92 kDa gelatinase (matrix metalloproteinase 9) have been implicated in subepidermal blister formation via proteolytic degradation of BP180. In this study we sought to elucidate the contribution of these two enzymes to subepidermal blister formation by assessing the expression, localization, and activity of the two proteases in lesional skin, serum samples, and blister fluids obtained from 17 patients with bullous pemphigoid. The results indicate that (i) neutrophil elastase is found in skin biopsy specimens from bullous pemphigoid lesions and is recovered as active enzyme in blister fluids, and (ii) although proform of matrix metalloproteinase 9 is present in lesional skin, it is present only as proenzyme in blister fluids, which also contain high levels of tissue inhibitor of metalloproteinase-1. Next, the capacity of matrix metalloproteinase 9 and neutrophil elastase to degrade a recombinant protein corresponding to the extracellular collagenous domain of the BP180 was studied. Our data illustrate that (i) recombinant matrix metalloproteinase 9, neutrophil elastase, and blister fluid from bullous pemphigoid patients are all able to hydrolyze recombinant BP180; (ii) the pattern of recombinant BP180 proteolysis with blister fluid was similar to that obtained with neutrophil elastase; and (iii) recombinant BP180 degradation by blister fluid could be inhibited by chloromethylketone, a specific elastase inhibitor, but not by batimastat, a wide spectrum matrix metalloproteinase inhibitor. Our results confirm the importance of neutrophil elastase but not matrix metalloproteinase 9 in the direct cleavage of BP180 autoantigen and subepidermal blister formation in human bullous pemphigoid.     

寻常性银屑病并发成人型线状IgA大疱性皮病

报告l例寻常性银屑病并发成人型线状IgA大疱性皮病.患者男,36岁.因全身红色斑疹伴白色鳞屑反复发生20年.躯干、双上肢出现环状排列的水疱10d伴瘙痒就诊.皮损组织病理检查:表皮下水疱,疱内、真皮浅层和真皮乳头见中性粒细胞、嗜酸性粒细胞浸润;皮损周围皮肤直接免疫荧光显示基膜带Iga、IgG呈带状沉积;取患者血清行BP180NC16A(大疱性类天疱疮18 000抗原的近膜片段)-ELISA检查显示阴性;以盐裂正常人皮肤为底物,取患者血清行间接免疫荧光检查显示IgA、IgG呈带状沉积在真皮侧.诊断为寻常性银屑病并发成人型线状TgA大疱性皮病.     

Assessment of skin basement membrane zone antibodies in the urine of patients with acquired subepidermal immunobullous diseases

BACKGROUND: In bullous pemphigoid (BP), cicatricial pemphigoid (CP) and linear IgA disease (LAD), autoantibodies to the basement membrane zone (BMZ) are found in skin and mucosa, blood and blister fluid. OBJECTIVES: To assess whether BMZ antibodies might also be detected in urine. METHODS: Urine and serum samples from 62 patients (32 with BP, 17 with CP and 13 with LAD) were analysed for antibody isotypes and subclasses by indirect immunofluorescence, and urine and serum samples from 40 patients (25 with BP, eight with CP and seven with LAD) were screened for target antigens using immunoblotting. RESULTS: Fourteen of 32 patients with BP had detectable levels of IgG BMZ autoantibodies in their urine, and all 32 had positive sera. Of these 14 BP patients, 13 had epidermal-binding serum autoantibodies at a titre > 1 : 160, and one had dermal-binding serum antibodies at a titre of 1 : 40. BMZ autoantibodies were not detected in the urine of the CP or LAD patients, but the corresponding sera were of low titre or negative. IgG subclasses (IgG1-4) were less frequently detected in urine than in serum. IgG4 was the predominant subgroup found (10 urine samples and all 14 sera), followed by IgG1 (two urine samples and 12 sera); IgG2 was detected in a single urine sample and three sera, and IgG3 was not detected. Eight of 25 BP and one of eight CP urine samples were positive on immunoblotting, and bound BP230 and/or BP180 with IgA and/or IgG autoantibodies. IgA autoantibodies were not detected in the urine of the seven LAD patients. The corresponding sera were often more positive, with 21 of 25 BP, five of eight CP and six of seven LAD sera immunoblotting the major BP antigens. CONCLUSIONS: The detection of IgG autoantibodies from urine samples using indirect immunofluorescence correlated with a high titre of IgG autoantibodies in the serum. IgG and IgA autoantibodies in the urine were detected by immunoblotting, although less frequently than in serum. The finding of BMZ antibodies in the urine of many BP patients may have clinical relevance, and may have a restricted application in the diagnosis of immunobullous disease.     

Distribution of the antigen in adult linear IgA disease and chronic bullous dermatosis of childhood suggests that it is a single and unique antigen

The tissue distribution and species expression of the antigens in adult linear IgA disease (LAD) and chronic bullous dermatosis of childhood (CBDC) was studied and compared with those of bullous pemphigoid (BP) and epidermolysis bullosa acquisita (EBA) antigens. Positive sera from five adult LAD patients, 15 CBDC patients, three BP patients and two EBA patients were studied, using skin and mucosal tissues of humans and animals as substrates. The tissue distribution and phylogenetic distribution of adult LAD antigen and CBDC antigen were the same, suggesting that they are identical diseases. The antigens were both demonstrable only in stratified squamous epithelia and amnion of mammalian species. They were absent from transitional and columnar epithelium. The EBA antigen was not expressed in pig tissues or human placenta, unlike adult LAD and CBDC antigens. Expression of adult LAD, CBDC and BP antigens was observed in all tissues except bladder, which lacked LAD and CBDC antigens. These results suggest that the antigens in adult LAD and CBDC are the same and differ from EP and EBA antigens.     

Evaluation of an Avidin-Biotin-Peroxidase Method with a Monoclonal Antibody to Type IV Collagen in the Differential Diagnosis of Bullous Pemphigoid and Epidermolysis Bullosa Acquisita

There are reports in which an immunohistochemical technique with a monoclonal antibody to type IV collagen has been employed for differentiating between bullous pemphigoid (BP) and epidermolysis bullosa acquisita (EBA). The aim of this study was to determine whether this method could be used routinely. Biopsies (paraffin-embedded lesional skin containing a blister) from currently diagnosed patients with clinical features suggesting BP or EBA were examined by an avidin-biotin-peroxidase (ABC) technique. Sera were tested by indirect immunofluorescence on salt-split skin (IF) and immunoblotting (IB). In all cases which exhibited clear type IV collagen staining, the results of the ABC technique agreed with results of both IF and IB. In one confirmed EBA case, it was impossible to unequivocally localize type IV collagen, because it stained very faintly. Taking into consideration the results of our study, data indicating that the level of blistering might not coincide with the localization of immunoreactants in EBA cases and the possibility of an enzymatic destruction of lamina densa, we conclude that the ABC method is unsuitable for differentiation between BP and EBA.     

Acquired bullous diseases of childhood: re-evaluation of diagnosis by indirect immunofluorescence examination on 1 M NaCl split skin and immunoblotting

Summary Acquired autoimmune bullous diseases of childhood are rare, and can be difficult to distinguish clinically. We have studied 12 children, with an initial diagnosis of bullous pemphigoid (BP) in eight patients, cicatricial pemphigoid (CP) in one, chronic bullous disease of childhood (CBDC) in one, and epidermolysis bullosa acquisita (EBA) in two.
All patients had positive indirect immunofluorescence (HF) of the BMZ with IgG. Using 1 M NaCl split skin, six patients showed epidermal binding of IgG, with additional IgA in three cases, and in five patients IgG antibodies bound a dermal protein. Immunoblotting studies revealed an antibody to type VII collagen (EBA antigen) in three patients who had a dermal pattern on IIF. Six sera reacted with an epidermal protein of 180 and/or 220 kDa, characteristic of BP and CP. One of the three IgA-positive sera detected 220-and 180-kDa epidermal proteins using anti-IgA antibody. Following these studies the diagnosis was changed in three of the children. The diagnosis of CBDC was changed to either BP or EBA because of the presence of circulating IgG autoantibodies. In two children with an initial diagnosis of BP the diagnosis was changed to EBA.
We conclude that the clinical picture in bullous disorders of childhood shows considerable overlap, and is often misleading. Additional circulating IgA autoantibodies seem to be more common in BP than has been recognized previously. Indirect immunofluorescence investigation on 1 M NaCl split skin may be helpful in differentiating between BP and EBA, but does not replace immunoblotting studies. EBA is apparently more common in children than in adults. No difference was found between the children with BP and EBA with regard to the duration of disease. The long-term outlook is good, although the course may be protracted.     

Psoriasis vulgaris coexistent with epidermolysis bullosa acquisita

Summary Autoimmune bullous diseases, such as bullous pemphigoid or pemphigus vulgaris. occasionally develop in psoriatic patients. In addition, a novel subepidermal bullous disease with autoantibodies against a lower lamina lucida antigen of 200kDa has recently been reported in association with psoriasis. We describe here a patient with psoriasis vulgaris who developed epidermolysis bullosa acquisita (EBA). Direct immunofluorescence revealed linear deposition of IgCl and C3 at the basement membrane zone. The patient's serum bound to the dermal side of salt-split normal human skin. However, immnumohlol analysis demonstrated that the patient's serum reacted with an EBA antigen of 290 kDa. EBA should be included in the list of autoimmune diseases associated with psoriasis vulgaris.     

Refractory bullous pemphigoid leaving numerous milia during recovery

Recovery with milia may occur in bullous pemphigoid (BP), mucous membrane pemphigoid (MMP) and epidermolysis bullosa acquisita (EBA). Scarring commonly occurs in MMP and EBA. Here, we report a 62‐year‐old man patient with BP, who was left with numerous milia during recovery. The patient had immunoglobulin (Ig)G autoantibodies to the recombinant protein of the BP180‐NC16a domain and the soluble 120‐kDa ectodomain of BP180 (linear IgA bullous dermatosis [LAD]‐1). There are cases of BP with IgG autoantibodies to LAD‐1 and/or the recombinant protein of BP180 C‐terminal domain. Extensive milia formation during recovery may be associated with immunological predisposition and/or improper interaction between hemidesmosomes and the extracellular matrix components.     

Acquired subepidermal bullous diseases associated with psoriasis: a clinical, immunopathological and immunogenetic study

Summary Clinical, immunopathological and immunogenetic studies of four patients with a subepidermal bullous disease associated with psoriasis were carried out to determine the true nature of the blistering disease and to investigate further the relationship between psoriasis and acquired subepidermal bullous diseases. Autoantibodies in all four patients bound to the epidermal side of salt-split skin by indirect immunofluorescence and detected the major bullous pemphigoid (BP) antigens by immunoblotting. One had additional IgA autoantibodies binding an epidermal polypeptide of 270/280 kDa and another had circulating IgA autoantibodies which detected both BP and epidermolysis bullosa acquisita (EBA) antigens. All patients had active psoriasis at the onset of the bullous disease. Three patients were being treated with tar when blisters developed: one patient also received UVB radiation and experienced a relapse after exposure to sunlight. HLA phenotypes in three patients were determined. All the patients responded well to methotrexate. These findings demonstrate that BP is the subepidermal bullous disease most associated with psoriasis. Changes at the basement membrane zone in psoriasis may be responsible for the heterogeneous antibody response and may trigger the bullous disease, as may antipsoriatic treatment including tar and UV radiation. However, common immunogenetic mechanisms may play a crucial part in this disease association.     

A comparison of the expression of known basement membrane components with the linear IgA disease antigens using the novel substrate cylindroma

Linear IgA disease (LAD) is characterized by IgA basement membrane zone autoantibodies. Immunofluorescence, immunoblotting and immunoelectron microscopy studies have established the complexity and heterogeneity of the target antigens. We have studied the expression of the LAD antigens by cylindroma, a benign epithelial tumour that secretes abundant basement membrane, using 57 LAD sera categorized by indirect immunofluorescence on intact and salt-split skin. The expression of known components of hemidesmosomes, the anchoring filaments, extracellular matrix and the anchoring fibrils could be differentiated and were compared with the expression of the LAD antigens. The results showed that the LAD sera bound to the cylindroma tumour in two distinctive patterns. Thirty-three sera were positive on cylindroma. Twenty-seven sera bound to the basement membrane around the tumour clusters and the islets within the clusters in a thin linear band that was occasionally discontinuous. This was similar to the pattern observed with antibodies to the hemidesmosome components, the alpha6beta4 integrin and the bullous pemphigoid antigens BP230 and BP180. This pattern was observed with sera that bound to the epidermal (11) and the dermal (3) aspects of split skin or were negative (11), and with one serum which bound only to intact skin. Seven sera, all binding to the dermal aspect of split skin, bound the tumour cluster basement membrane in a thick band that was identical in appearance to that seen with antibodies to collagen VII; however, binding to the islets was either identical to collagen VII or similar but differentiated with double staining. Some sera were negative on cylindroma. Using fluorescence overlay antigen mapping we demonstrated colocalization of some epidermal-associated LAD target antigens with known hemidesmosome proteins, and colocalization of some dermal-associated LAD target antigens with anchoring fibril components. The results using cylindroma as substrate suggest that LAD autoantibodies may react with three or more target antigens. We propose from the results of this study that the autoantibodies in LAD target multiple antigens associated with hemidesmosomes and anchoring fibrils.