Elsevier

Immunology Letters

Volume 122, Issue 2, 21 February 2009, Pages 208-213
Immunology Letters

High-dose pemphigus antibodies against linear epitopes of desmoglein 3 (Dsg3) can induce acantholysis and depletion of Dsg3 from keratinocytes

https://doi.org/10.1016/j.imlet.2009.01.002Get rights and content

Abstract

We have previously demonstrated that serum autoantibodies of patients with pemphigus vulgaris (PV) may affect desmoglein 3 (Dsg3)-mediated adhesion by decreasing its half-life and inducing Dsg3 cleavage. Here we sought to gain more insights into the role of Dsg3-targetting IgG in acantholysis. To do so, alterations of keratinocyte morphology and cell–cell adhesion strength were investigated in the presence of PV serum, PV IgG, and IgG purified from PV patients’ sera against linear epitopes of Dsg3 (anti-Dsg3-L IgG). Changes in Dsg3 protein levels were assessed by Western blotting. Results showed that both PV serum and PV IgG were able to induce acantholysis and decrease the total amount of Dsg3 in cell lysates. Polyclonal anti-Dsg3-L IgG displayed Dsg3-depleting activity solely when used at 1 μg/ml, i.e. under non-physiologic conditions. Furthermore, cell–cell detachment induced by PV IgG and anti-Dsg3-L IgG seemed to precede the loss of Dsg3 from keratinocytes, suggesting that depletion/degradation of Dsg3 represents a late event in acantholysis. Collectively, the data presented here demonstrate that PV IgG recognizing non-conformational epitopes of Dsg3 are pathogenic when administered on doses largely exceeding those found in PV sera.

Introduction

Pemphigus vulgaris (PV) is a potentially fatal autoimmune blistering disease affecting stratified squamous epithelia [1]. Autoantibodies in PV recognize receptor-like molecules regulating epidermal cohesion. Among the array of keratinocyte self antigens immunoprecipitated by circulating PV IgG [2], [3], the best characterized is desmoglein (Dsg) 3, a desmosomal glycoprotein which is critically involved in ensuring calcium-dependent intercellular adhesion among keratinocytes [4]. Indeed, PV associates with the presence of circulating IgG against linear and conformational epitopes of Dsg3 and, less frequently, Dsg1 [5], [6].

In PV, blisters develop as a consequence of the loss of cell–cell adhesion, or acantholysis. Although PV IgG against non-desmoglein antigens as well as serum factors other than IgG can exert pathogenic effects on keratinocytes [7], [8], [9], the key role of Dsg3-targeting antibodies in PV pathophysiology is widely accepted. Binding of IgG to Dsg3 is thought to affect Dsg3 function by steric hindrance (reviewed in ref. [9]) or by inducing desmosomal signalling [7]. Alternatively, or additionally, binding of autoantibodies to Dsg3 could modulate its own synthesis, leading to the formation of aberrant desmosomes lacking Dsg3 [10], [11], [12], [13]. In this regard, we previously reported that PV sera can decrease the cell content of Dsg3 by reducing its half-life and perturbing the subsequent incorporation of Dsg3 into desmosomes [11]. Consistently, it has been suggested that PV IgG may work through the depletion of the Triton X-100 soluble pool of Dsg3, thus depriving cells from free Dsg3 before its assembly into desmosomes [10]. This phenomenon could depend on the down-regulation of Dsg3 at the transcriptional level [13]. A recent study from Kitajima's lab reported the presumptive depletion of Dsg3 from desmosomes to depend on the pathogenicity of anti-Dsg3 IgG against conformational epitopes of Dsg3 [12]. This finding would be in agreement with previous papers demonstrating that the dominant autoimmune epitopes in PV are found in the N-terminal adhesive surfaces of Dsg3. By using domain-swapped Dsg3 recombinants, Futei et al. showed indeed that the majority of autoimmune IgG in PV sera reacted with epitopes formed by aa 1–161 of the NH2-terminus of Dsg3 [14]. IgG against the amino-terminal adhesive interface of Dsg3 were subsequently found to be pathogenic [15].

In general, the role of IgG against conformational sequences of the extracellular domain of Dsg3 in PV has received great attention. In marked contrast, participation of Dsg3 linear epitopes in PV remains largely unknown. The recent finding that IgG titres against a small stretch of the NH2-terminus of Dsg3 are associated with active PV [16] is relevant to this regard. Indeed, IgG in the PV sera detect non-conformational epitopes of Dsg3 in addition to the previously identified conformation-dependent ones, and they may relate to clinical phenotype of PV [17]. However, the pathophysiological significance of these abovementioned anti-Dsg3 IgG still need to be clarified.

The present studies were performed in the attempt to investigate the presumptive pathogenicity of anti-Dsg3 antibodies purified from patients’ sera against linear epitopes of Dsg3 (anti-Dsg3-L IgG). Results also shed light on the role of PV sera and PV IgG in depleting Dsg3 from keratinocytes.

Section snippets

Serum samples and IgG fractions

Sera of PV patients and healthy donors have been previously characterized [11]. For this study, three different PV sera exhibiting similar effects were used. Anti-Dsg3-L antibodies from PV sera were purified in accordance with the procedure for immunoaffinity antibody purification [18]. In brief, HaCaT detergent extracts were immunoprecipitated with the 5H10 antibody (Santa Cruz Biotecnology, Santa Cruz, CA) and the resulting antibody–antigen complexes were denatured and resolved by 8%

Effects of PV serum, PV IgG, and anti-Dsg3-L IgG on cell morphology

We established a cell culture system by using human HaCaT keratinocytes. The ability of whole PV serum to induce acantholysis has been widely demonstrated [24], [25]. Consistently, keratinocytes cultured in 2 ml whole PV serum, but not those exposed to normal human (Nh) sera, shrunk (Fig. 1a–d) and detached from neighbouring cells within 24 h, as revealed by phase contrast microscopy. Some acantholytic keratinocytes assumed a fibroblastoid appearance, and partially lost anchorage with substrate (

Discussion

In the present study we report that pemphigus serum and PV IgG induce depletion of Dsg3 in keratinocytes, whereas the role of anti-Dsg3 IgG against linear epitopes of Dsg3 appears to be strictly dose-dependent. In contrast to both PV serum and PV IgG, indeed, only very high-concentration anti-Dsg3-L – corresponding to the amount potentially found in 1 ml of seventy-fold concentrated PV sera – can exert pathogenic activity in vitro. Our data also suggest that depletion of Dsg3 may be a late event

Acknowledgement

This study was supported in part by Ministero dell’Istruzione, dell’Universita‘e della Ricerca, Programmi di Ricerca Scientifica di Rilevante Interesse Nazionale (MIUR-PRIN 2007).

References (30)

Cited by (5)

View full text