Characterization of pathogenic auto-antibodies directed against desmoglein 3 and desmocollin 3 in sera of pemphigus patients
Authors
David Ali Rafei-Shamsabadi
Institution
University Marburg
Country
Germany
Year
2013
Journal
Dissertation
Abstract
Pemphigus vulgaris (PV) represents the most frequent clinical type of the pemphigus group of autoimmune bullous skin disorders. There is substantial evidence that blister formation in pemphigus patients is mediated by auto-antibodies (auto-Abs) targeted against certain desmosomal cadherins, namely desmoglein1 (Dsg1) and desmoglein3 (Dsg3). Several pathogenic epitopes of Dsg3 are located at the amino-(NH2)-terminal end of the Dsg3 ectodomain, namely the extracellular domain 1 (EC1). On the other hand a great number of pemphigus patients exhibit auto-Abs directed against the more carboxy-(COOH)-terminal epitopes of Dsg3, e.g. within the EC4- and EC5-domain and these domains may play an essential role in maintaining desmosomal adhesion. Some pemphigus patients exhibit additional or solely auto-Abs against other desmosomal cadherins, especially desmocollin 3 (Dsc3). However, the pathogenic relevance of Dsc3-reactive immunoglobulin G (IgG) has not been directly shown. This study aimed to first establish a method to specifically isolate Dsg3-reactive IgG from PV sera and to further investigate their pathogenic capacity using a keratinocyte based in vitro assay. This method was further applied to sera of four Japanese patients suffering from atypical pemphigus all of them exhibiting a positive IgG reactivity against Dsc3. Sepharose based affinity chromatochraphy columns coated with recombinant baculovirus produced proteins of the extracellular domains of Dsg3 and Dsc3, respectively, were used to specifically isolate auto-Abs from pemphigus sera. Affinity purified IgG fractions were subsequently tested for antigen specificity using enzyme-linked immunosorbent assay (ELISA) and immunoblotting (IB). Reactivity with native Dsg3- and Dsc3-protein, respectively, was proven by immunofluorescence (IF) on cultured human keratinocytes, monkey esophagus and frozen sections of normal human skin. Finally, a keratinocyte based so-called dissociation assay served to investigate the in vitro pathogenicity of the affinity purified IgG fractions. Eight Dsg3-reactive PV patients showed Dsg3-domain-specific auto-antibodies by ELISA. Four of these patients were selected for further investigation based on their antibody profile, i.e. their epitope specificity. Two patients (#1 and #8) exhibited IgG directed against Dsg3EC1 and Dsg3EC4. Patient #2 exclusively expressed auto-Abs directed against Dsg3EC1 whereas patient #6 showed IgG reactivity against Dsg3EC4, only. Serum IgG was then affinity purified using the respective recombinant Dsg3-subdomains. Antigen specificity of the eluted IgG-fractions was subsequently verified by IB and ELISA. Isolated IgG fractions showed a characteristic intercellular staining pattern by IF using cultured human keratinocytes indicating positive reactivity with native Dsg3-protein. Recognition of Dsg3EC1 but not of Dsg3EC4 by the isolated IgG fractions of patients #1 and #8 was Ca2+-dependent. Finally Dsg3-, Dsg3EC1- and Dsg3EC4-specific IgG caused keratinocyte dissociation which was comparable to the positive control, a monoclonal antibody (AK23) directed against the NH2-terminus of Dsg3. These techniques were then applied to the sera of four atypical, i.e. two pemphigus vegetans and two pemphigus herpetiformis patients, in order to isolate Dsc3-specific IgG. All but one of these patients, who showed additional Dsg1 reactivity, exhibited IgG reactivity exclusively against Dsc3 but no other desmosomal cadherin. From all sera IgG fractions were successfully isolated and antigen specificity to Dsc3 was verified. Dsc3-reacitve IgG showed a characteristic intercellular staining pattern by IF on cultured human keratinocytes, monkey esophagus and human skin. Finally all isolated IgG fractions were able to induce loss of keratinocyte adhesion in vitro.Taken together this data strongly suggests a significant acantholytic effect of IgG directed against COOH-terminal epitopes of Dsg3 in addition to the well known pathogenic epitopes at the NH2-terminus of this auto-antigen. Furthermore recognition of COOH-terminal epitopes seems to be Ca2+ independent which goes in line with previous studies of our group. Moreover Dsc3-reactive IgG isolated from patients with atypical pemphigus variants proved to be pathogenic in vitro. For the first time these results directly show the acantholytic effect of Dsc3-reactive IgG and provides evidence for the pathogenic relevance of Dsc3-IgG in pemphigus patients lacking reactivity against other desmosomal cadherins. Further investigations are needed to elucidate the mechanisms by which auto-Abs directed against COOH-terminal epitopes of Dsg3 induce acantholysis. The pathogenic relevance of other epitopes of Dsg3, especially within the Dsg3EC5-domain, needs to be addressed. Finally screening of pemphigus patients’ sera for Dsc3-reactive IgG should provide further knowledge about their correlation with atypical pemphigus variants.