Inflammatory variant of epidermolysis bullosa acquisita with IgG autoantibodies against type VII collagen and laminin alpha3.

BACKGROUND: The inflammatory variant of epidermolysis bullosa acquisita (EBA) may clinically closely resemble bullous or cicatricial pemphigoid. Patients with inflammatory EBA have IgG autoantibodies against type VII collagen. Patients with anti-epiligrin cicatricial pemphigoid have IgG autoantibodies against laminin 5. OBSERVATION: We describe a patient with inflammatory EBA exhibiting nonscarring oral and vaginal involvement. Indirect immunofluorescence using skin substrate lacking an epidermal basement membrane molecule, direct immunoelectron microscopy, immunoblot, and immunoprecipitation studies revealed the simultaneous presence of circulating IgG autoantibodies against type VII collagen and laminin alpha3. A final diagnosis of EBA was based on the sublamina densa level of blister formation. CONCLUSION: This case illustrates the clinical and immunological overlap between EBA and anti-epiligrin cicatricial pemphigoid, a unique finding that may have developed as a consequence of epitope spreading.

Bullous pemphigoid and linear IgA dermatosis sera recognize a similar 120-kDa keratinocyte collagenous glycoprotein with antigenic cross-reactivity to BP180.

Circulating IgG from a large subset of bullous pemphigoid (BP) patients reacted on immunoblot with a 120-kDa protein in conditioned keratinocyte culture medium and in keratinocyte cell extracts. A protein with a similar molecular weight was recognized by circulating IgA from a subset of patients with linear IgA dermatosis (LAD). Both affinity-purified 120-kDa-specific BP IgG and 120-kDa-specific LAD IgA bound to the roof of salt-split skin. Both proteins recognized are collagenous glycoproteins. Deglycosylation with N-glycosidase F resulted in an identical reduction in molecular weight for both the BP-IgG-recognized protein and the LAD-IgA-recognized protein. Both proteins were equally susceptible to digestion with type VII collagenase. Furthermore, both proteins were absent from conditioned culture medium of keratinocytes from patients with BP180-deficient general atrophic benign epidermolysis bullosa (GABEB). Immunodepletion studies showed that the 120-kDa LAD antigen could be removed from conditioned culture medium by anti-120-kDa BP IgG. Thus these results indicate that these proteins are either highly related or, most probably, identical. A strong antigenic relationship between the 120-kDa protein and the 180-kDa bullous pemphigoid antigen (BP180) was detected by cross-reaction of affinity-purified anti-120-kDa BP patient antibodies to BP180 and cross-reaction of monoclonal anti-180-kDa antibodies to the 120-kDa protein. Notwithstanding this cross-reactivity, the 120-kDa protein also exhibits unique epitopes demonstrated by the nonreactivity of individual anti-120-kDa BP and LAD patient serum with the 180-kDa antigen.

Revertant mosaicism in epidermolysis bullosa caused by mitotic gene conversion.

Mitotic gene conversion acting as reverse mutation has not been previously demonstrated in human. We report here that the revertant mosaicism of a compound heterozygous proband with an autosomal recessive genodermatosis, generalized atrophic benign epidermolysis bullosa, is caused by mitotic gene conversion of one of the two mutated COL17A1 alleles. Specifically, the maternal allele surrounding the mutation site on COL17A1 (1706delA) showed reversion of the mutation and loss of heterozygosity along a tract of at least 381 bp in revertant keratinocytes derived from clinically unaffected skin patches; the paternal mutation (R1226X) remained present in all cell samples. Revertant mosaicism represents a way of natural gene therapy.

Effects of keratin 14 ablation on the clinical and cellular phenotype in a kindred with recessive epidermolysis bullosa simplex.

We studied a kindred with recessive epidermolysis bullosa simplex in which the affected members lacked expression of the basal cell keratin 14. The patients had severe generalized skin blistering that improved slightly with age. The basal cells of the patients did not express keratin 14 and contained no keratin intermediate filaments. The expression of keratin 5, the obligate copolymer of keratin 14, was slightly reduced. The expression of keratin 15, the alternative basal cell keratin, was increased, suggesting upregulation or stabilization to compensate for the lack of keratin 14. The expression of keratin 16, keratin 17, and keratin 19 in the patient's skin was not different from controls. Immunoelectron microscopy showed a loose network of keratin 5/keratin 15 protofilaments in the basal cells. Keratin 15 filaments did not aggregate into higher order bundles. Sequence analysis of genomic DNA revealed a homozygous mutation in the 3'-acceptor splice site of intron 1 (1840 A-->C) in the affected individuals. This mutation led to the skipping of exon 2 in 24% of the KRT14 transcripts and to the use of a cryptic splice site in 76% of the transcripts. Premature termination codons were generated in all transcripts (codons 175+1 or 175+29), leading to a truncated keratin 14 protein within the helical 1B rod domain. The disorder was associated with circumscribed hyperkeratotic lesions with the histology of epidermolytic hyperkeratosis. The prognosis of keratin 14 ablation is much better in the human than in the mouse.

Generalized atrophic benign epidermolysis bullosa. Either 180-kd bullous pemphigoid antigen or laminin-5 deficiency.

BACKGROUND: Generalized atrophic benign epidermolysis bullosa (GABEB) is a form of nonlethal junctional epidermolysis bullosa, clinically characterized by generalized blistering after birth, atrophic healing, and incomplete universal atrophic alopecia with onset in childhood. Recently, we discovered a deficiency of the 180-kd bullous pemphigoid antigen (BP180) and a reduced amount of BP180 messenger RNA in three patients with GABEB. It is not yet clear, however, whether GABEB is invariably caused by BP180 deficiency. RESULTS: We examined 18 patients with nonlethal junctional epidermolysis bullosa from unrelated families; nine of these individuals presented with the clinical characteristics of GABEB. Specimens of clinically normal skin obtained from the patients were stained by immunofluorescence with monoclonal antibodies to BP180 and laminin-5. The BP180 epitopes were not expressed in eight patients, all of whom were sharing the typical clinical features of GABEB. In one of the nine patients with GABEB, the BP180 level was sufficient, but the laminin-5 level was reduced. Among the nine patients with junctional epidermolysis bullosa without atrophic alopecia, laminin-5 level was not expressed in one patient, while in the other patients both antigens were normally expressed. CONCLUSIONS: Not all patients with GABEB are deficient in BP180, since some individuals with GABEB only exhibit reduction of the laminin-5 expression. The BP180 deficiency in the skin invariably seems to result in GABEB. Immunofluorescence analysis using monoclonal antibodies against BP180 (and laminin-5) may allow early subtyping, which is of prognostic significance, in children born with junctional epidermolysis bullosa.

Bullous pemphigoid and epidermolysis bullosa acquisita. Differentiation by fluorescence overlay antigen mapping.

BACKGROUND AND DESIGN: From previous studies, we concluded that the fluorescence overlay antigen mapping (FOAM) technique could be of value to the differential diagnosis of the acquired subepidermal bullous skin disorders, bullous pemphigoid (BP) and epidermolysis bullosa acquisita (EBA). In these diseases, ultrastructural identification of the site of skin-bound IgG deposits at the epidermal basement membrane zone (EBMZ) may be essential to the correct diagnosis. Since ultrastructural studies are more expensive, time-consuming, and less widely available than immunofluorescence, we addressed the question of whether the FOAM technique can reliably identify the site of IgG deposits at the EBMZ, and distinguish BP from EBA. For this purpose, the technique was applied to perilesional skin from seven patients with BP and six with EBA, using computer-aided imaging of red-stained type VII collagen and green-stained IgG, according to previous findings. RESULTS: Digitized multicolor FOAM images of perilesional skin from patients with BP showed nonoverlap band patterns of green-stained lamina lucida IgG deposits (ultrastructurally proven) and red-stained type VII collagen. By contrast, FOAM images of EBA skin typically showed overlap patterns of green-stained sublamina densa IgG deposits and red-stained type VII collagen. These findings were observed also in skin tissue stored in Michel's transport medium or stored frozen for 15 years. CONCLUSIONS: The computer-aided FOAM technique may have great potential in distinguishing between IgG deposits above (BP) and just below (EBA) the lamina densa of the EBMZ in skin tissue. The technique is not as simple as saline-split skin methodology but offers more flexibility, and it certainly is quicker and less expensive than electron microscopy. Furthermore, the use of digitized fluorescence images offers improved possibilities for evaluating the various "linear" patterns of immune reactant deposition at the EBMZ in subepidermal bullous autoimmune skin diseases.

Bullous pemphigoid: serum antibody titre and antigen specificity.

2 antigens have been identified as possible targets for autoantibody depositions in bullous pemphigoid: a 230-kD protein (BP230) and a 180-kD protein (BP180). We studied the relationship of these 2 antigens with the immunofluorescence determined serum antibody titre: 2 groups of bullous pemphigoid patients were selected on the basis of immunoblot-determined antibody specificity. One group (13 patients) had antibody specificity for BP230 and not for BP180, while the other group (9 patients) had antibody specificity for BP180 and not for BP230. The immunofluorescence titres of the circulating antibodies determined on monkey oesophagus substrate displayed, for the BP230-specific group, a mean of 1:1102. The maximal observed titre was 1:5120. The mean titre in the BP180-specific group was only 1:29, with a highest titre of only 1:160. This result suggests that in routine indirect immunofluorescence of bullous pemphigoid sera, the contribution of the BP180-specific antibodies to the total anti-epidermal basement membrane zone antibody titre is relatively much lower than that of the BP230-specific antibodies. Thus, at high dilutions, only the BP230-specific antibodies contribute to the overall indirect immunfluorescence titre.

Fluorescence overlay antigen mapping of the epidermal basement membrane zone: III. Topographic staining and effective resolution.

In this third study on the fluorescence overlay antigen mapping (FOAM) technique, we have addressed the question of which differences of antigen distributions close to the resolving power of the light microscope can be distinguished. An answer to this question should provide clues to future applications of the technique aiming at the topographic differentiation of IgG deposits displayed at the epidermal basement membrane zone (EBMZ) in certain bullous skin disorders. For the present purpose we have developed a topographic staining model in human skin, using structural EBMZ antigens as topographic reference markers. The distribution of these markers relative to one another is visualized in FOAM images obtained by selective double immunofluorescence tracing and videomicroscopic overlay imaging. The theoretical resolution limit of the technique is discussed and suggests an effective lower limit of some 60-65 nm. Although this limit is not reached under present conditions, our results show that it is possible to distinguish topographic differences of antigen distributions with an upper resolution limit of 200 +/- 50 nm. Furthermore, our findings indicate that collagen Type VII and beta 4 integrin are the most suitable molecules to serve as topographic reference markers in future applications of the technique aiming at the differentiation of bullous pemphigoid (BP) and epidermolysis bullosa acquisita (EBA). Preliminary results on this topic are most promising indeed.

Fluorescence overlay antigen mapping of the epidermal basement membrane zone: II. Color fidelity.

In this second report on the fluorescence overlay antigen mapping (FOAM) technique, we highlight some of the errors that may influence faithful color rendition of slide preparations using triple antigen immunofluorescence staining. Reliable interpretation of multicolor fluorescence images requires that the observer can unambiguously assign each color in these images to the presence of a specific combination of the labeled antigens. This is possible only when the image fidelity meets certain standards. The present study concentrates on color fidelity which is easily undermined by spectral matching errors, image contrast errors, and exposure time errors. Evaluation of these errors, using the photomicrographic overlay variant of FOAM, showed the potential unreliability of the simultaneous use of multiple fluorophores for immunofluorescence microscopy. The procedures described here may serve as a solid starting point in formulating technical conditions that allow reliable color rendition in multicolor immunofluorescence microscopy. Furthermore, these procedures can be adapted to studies other than the analysis of basement membrane zone antigens, to which they have been first applied.

180-kD bullous pemphigoid antigen (BP180) is deficient in generalized atrophic benign epidermolysis bullosa.

Generalized atrophic benign epidermolysis bullosa (GABEB) is a form of nonlethal junctional epidermolysis bullosa characterized by universal alopecia and atrophy of the skin. We report a deficiency of the 180-kD bullous pemphigoid antigen in three patients with GABEB from unrelated families. We screened specimens of clinically normal skin from nine junctional epidermolysis bullosa patients (3 GABEB, 4 lethal, 1 cicatricial, 1 pretibial) by immunofluorescence using monoclonal antibodies to the 180-kD and 230-kD bullous pemphigoid antigens (BP180 and BP230). In the skin of the three GABEB patients there was no reactivity with antibodies to BP180, whereas staining for BP230 was normal. In the skin of the other six, non-GABEB patients, included in this study the expression of BP180 and BP230 was normal. Immunoblot analysis of cultured keratinocytes from one of the GABEB patients also failed to detect BP180 antigen, whereas BP230 was present in normal amounts. The deficient expression of BP180 is reflected in the RNA message, as in Northern blot analysis a reduced amount of BP180 transcripts, although of normal length, were detected. Interestingly, in another GABEB patient there were not-involved areas of skin, in which blistering could not be induced by rubbing. Biopsy material from these areas showed interrupted staining for BP180. There was no staining for BP180 in areas of clinically normal but involved skin of this patient. In conclusion, this study reveals that the BP180 antigen is deficient and the BP180 mRNA is reduced in generalized atrophic benign epidermolysis bullosa.

HLA-DR expression is induced on keratinocytes in delayed hypersensitivity but not in allergen induced late-phase reactions.

In view of increasing evidence suggesting an active immunoregulatory role of the skin keratinocytes and the observation that the differentiation of allergen specific T lymphocytes is critical in the development of allergy, we evaluated epidermal expression of HLA-DR antigen in skin reactions induced with an atopen (house dust mite) and with an non-atopic antigen (Hemocyanin). Two groups of patients with house dust mite (Dermatophagoides pteronyssinus [Der p]) allergy were compared, one group was skin tested with Der p, the other group was immunized and subsequently skin tested with Helix pomatia Hemocyanin (HPH). Biopsy specimens taken at 48 h after the HPH (n = 11) and Der p (n = 11) tests were analysed immunohistologically. Reactions in both groups were comparable in size. Immunohistological analysis showed domination by CD4+ lymphocytes. Expression of HLA-DR antigen by epidermal keratinocytes was observed in six out of 11 of the HPH induced reactions, but in none of the Der p induced reactions. Eosinophils were spotted only throughout the Der p induced reactions, showing a good correlation with the number of CD4 positive lymphocytes. The lack of HLA-DR expression by keratinocytes during the allergen-induced reaction, compared with the Hemocyanin induced reaction can be the result of a difference in cytokine profile of the lymphocytes dominating the dermal infiltrate. On the other hand evidence exists that defective HLA-DR expression by keratinocytes enhances antigen induced lymphocyte activation, and may thus contribute to the development of allergen-specific T-lymphocytes.

Fluorescence overlay antigen mapping of the epidermal basement membrane zone: I. Geometric errors.

To identify in tissue sections the relative positions of antigen distributions close to the resolving power of the microscope, we have developed the fluorescence overlay antigen mapping (FOAM) procedure. As this technique makes high demands on the geometric fidelity of the overlay image, it is essential to recognize geometric errors resulting from optical imperfections. This applies in particular to the image shift difference (ISD) that may routinely occur during fluorescence overlay. We describe here procedures for assessment and mechanical correction of the ISD in tissue sections. Furthermore, we describe an alignment verification test to assess the accuracy of the ISD correction procedure, using collagen Type VII as the geometric verification marker. These procedures should enable reliable evaluation of relative antigen distributions in tissue sections using photomicrographic multicolor fluorescence overlay. Further details of the FOAM technique, such as color fidelity and its utility for diagnostic and research purposes, will be published separately.

Expression of integrin alpha 6 beta 4 in junctional epidermolysis bullosa.

The integrin alpha 6 beta 4 is a member of the integrin family of adhesion receptors. The integrin alpha 6 beta 4 is preferentially expressed in stratified squamous epithelia, where it is localized in hemidesmosomes. A reduced number of rudimentary hemidesmosomes is often found in skin from patients with junctional epidermolysis bullosa (JEB). In this study we have investigated the expression of alpha 6 beta 4 in skin specimens of patients with junctional (one non-lethal, two lethal) and dystrophic (two) epidermolysis bullosa, using immunofluorescent (IF) staining with five different monoclonal antibodies against the alpha 6 and beta 4 subunits. The intensity of IF staining of the integrin alpha 6 beta 4 and bullous pemphigoid antigen (BPA) was unreduced along the epidermal basement membrane zone (EBMZ) of all EB patients, compared to that in skin of healthy human controls. However, in the skin of two patients with lethal (Herlitz) JEB, who did not express GB3, IF staining of integrin alpha 6 beta 4 and BPA showed a "stitchy" discontinuous linear pattern along the EBMZ with interruptions at the borders of adjoining basal keratinocytes. The same results were obtained by immunoelectron microscopy. They corresponded with freeze-induced partial cell detachment from the basement membrane at the ultimate baso-lateral edge of the basal keratinocytes in lethal JEB skin. The basal lamellipodia at that location almost completely lacked tonofilaments and hemidesmosomes. Furthermore, in JEB there was a split between the intra- and extracellular epitopes of the integrin alpha 6 beta 4 receptor, whereas the integrin remains intact in salt-split skin. This suggests that the defect is in alpha 6 beta 4 itself or perhaps its ligand.