Four patients with the clinical picture of epidermolysis bullosa acquisita were investigated. Biopsies were taken from the involved and uninvolved areas of the skin and the immunohistochemical and microscopic changes were studied. Direct immunofluorescence showed deposition of IgG and C3/4 in a linear or norched pattern along the epidermal basement membrane in both the involved and the uninvolved skin. In addition IgA (3/4), IgM (1/4), C4 (3/4) and properdin (3/4) could be detected. Indirect immunofluorescence revealed the presence of circulating antibodies against inter alia the epithelial basement membrane zone in one patient. Routine electron microscopy showed that the blister was situated in the dermis leaving the basal lamina in the roof of the blister. With immunoelectron microscopy using peroxidase-labelled antibody the in vivo deposition of IgG was observed just beneath the basal lamina in the dermis of both the perilesional and the uninvolved skin. These observations show that epidermolysis bullosa acquisita is a distinct entity, in which autoimmune mechanisms might possibly play a role.

Direct immunofluorescence investigations were performed on skin biopsies from five patients with granuloma eosinophilicum faciale (GEF; facial granuloma). An extensive and brilliant granular picture was observed along the basement membrane (BM) of the epidermis and the hair follicles, in the walls of the vessels, in the cellular infiltrates and on the connective tissue fibers. These granules were positively stained by antisera directed against IgG, complement (C3/4) and, although less consistently, against IgA and IgM. IgE was found in one case along the BM, but anti- IgD was negative. Concomitantly heavy fibrillar deposits of fibrin were present in the walls of the vessels and in the cellular infiltrates, together with granular depositions along the BM. The analysis of the complement factors showed that C1q, C4, C3, C3c, C3d and C5 were present in the same pattern as C3/4. These results indicate that GEF can be considered as a chronic form of leukoclastic vasculitis mediated by an Arthus-like mechanism, maintained by an unidentified, persistent antigen or by locally produced Ig aggregates.

In this study we compared horseradish peroxidase (HRP)-labeled rabbit antihuman immunoglobulin G (IgG) conjugates, prepared by a one-step and a two-step method. Glutaraldehyde was used as a cross-linking agent. Two methods were used for removing unconjugated HRP: Sephadex G-200 gel chromatography and ammonium sulfate precipitation. The conjugates were characterized immunologically, immunochemically and enzymatically. The immunohistoenzymic properties of the conjugates were tested on unfixed cryostat sections of the skin of patients with chronic discoid bupus erythematosus. The influence of the presence of unconjugated HRP and unconjugated IgG was studied. Optimal results were obtained with conjugates prepared by a two-step method. Removing unconjugated HRP improved the immunohistoenzymic properties of the conjugates. Conjugated and unconjugated IgG could be separated by Sephadex G-200 gel chromatography.

Summary Using the ELISA and passive haemagglutination test we investigated the humoral immune response induced by long‐term administration of benzylpenicillin in patients. After the second week of the course in twenty‐four out of sixty‐one patients, synthesis of benzylpenicilloyl (BPO) specific IgG, IgA, IgM and/or IgE could be demonstrated. BPO‐specific IgE was always found in combination with BPO‐specific IgG and/or IgM. No difference in the humoral immune response was found between patients who developed clinical hypersensitivity reactions and those who did not. In a follow‐up study the disappearance of BPO‐specific immunoglobulins of all classes was observed. Results of the epicutaneous tests showed that a long‐term course of benzylpenicillin did not evoke delayed hypersensitivity. However intracutaneous tests with benzylpenicilloyl‐polylysine or benzylpenicillin showed positive immediate type reactions. From the results it can be concluded that a BPO‐specific IgE response is not necessarily associated with the development of a clinical hypersensitivity reaction.

No abstract