Prevalence of pemphigus and pemphigoid autoantibodies in the general population

Abstract

Background: Mucocutaneous blistering is characteristic of autoimmune bullous dermatoses (AIBD). Blisters are caused by autoantibodies directed against structural components of the skin. Hence, detection of specific autoantibodies has become a hallmark for AIBD diagnosis. Studies on prevalence of AIBD autoantibodies in healthy individuals yielded contradictory results.

Methods: To clarify this, samples from 7063 blood donors were tested for presence of anti-BP180-NC16A, anti-BP230 and anti-Dsg1/3 IgG by indirect immunofluorescence (IF) microscopy using a biochip.

Results: Cumulative prevalence of these autoantibodies was 0.9 % (CI: 0.7-1.1 %), with anti-BP180-NC16A IgG being most prevalent. Validation of IF findings using ELISA confirmed presence of autoantibodies in 7/15 (anti-Dsg1), 6/7 (anti-Dsg3), 35/37 (anti-BP180-NC16A) and 2/3 (anti-BP230) cases. Moreover, in 16 samples, anti-BP180-NC16A autoantibody concentrations exceeded the cut-off for the diagnosis of bullous pemphigoid. Interestingly, these anti-BP180-NC16A autoantibodies from healthy individuals formed immune complexes with recombinant antigen and dose-dependently activated neutrophils in vitro. However, fine-epitope mapping within NC16A showed a different binding pattern of anti-BP180-NC16A autoantibodies from healthy individuals compared to bullous pemphigoid patients, while IgG subclasses were identical.

Conclusions: Collectively, we here report a low prevalence of AIBD autoantibodies in a large cohort of healthy individuals. Furthermore, functional analysis shows differences between autoantibodies from healthy donors and AIBD patients.

Fig. 1

ELISA values of indirect IF microscopy-positive and indirect IF microscopy-negative samples. Percentage of indirect IF microscopy (IIFM)-positive (green) and indirect IF-microscopy negative (red) samples according to the measured ELISA values. Anti-BP180-NC16A-, anti-BP230-, anti-Dsg1-and anti-Dsg3-positive samples have significantly higher ELISA values than antibody-negative samples by the Wilcoxon rank-sum test with continuity correction. All indirect IF microscopy-positive samples [n: indirect IF microscopy-negative samples] were included: 15 [27] for Dsg1, 7 [13] for Dsg3, 37 [37] for BP180-NC16A, 3 [9] for BP230. *p < 0.05, ***p < 0.001 (comparing indirect IF microscopy-negative with indirect IF microscopy-positive samples). Abbreviations: IIFM: indirect IF microscopy

Fig. 2

Anti-BP180-NC16A autoantibodies from healthy individuals induce ROS release from neutrophils. (a) ROS release of immune complex-activated PMN in relation to cells incubated with antigen only. Owing to the non-parametric distribution, the data are presented as the median (line), 25/75-percentile (boxes), 5/95-percentiles (error bars) and outliers (dots). *p < 0.05 (ANOVA on ranks followed by Dunn’s method for multiple comparisons versus control (NC16A-negative samples). (b) Concentrations of anti-BP180-NC16A IgG correlate with the amount of ROS release from PMN (r = 0.786, p < 0.001, Pearson product-moment correlation)

Fig. 3

Anti-BP180-NC16A autoantibodies from healthy individuals bind to different epitopes within the NC16A domain. Samples from the indicated groups were tested for their IgG reactivity against NC16A2-3 (RSILPYGDSMDRIEKDRLQGMAPAAGADL). As reported previously [29, 30], high reactivity with this epitope was observed in BP patients. By contrast, only a minority of samples from healthy individuals with IgG reactivity against NC16A showed reactivity with this epitope. Abbreviations: NHS: normal human sera

Fig. 4

Anti-Dsg3 autoantibodies detected in healthy blood donors do not lead to Dsg3 internalization in vitro. HaCaT cells were incubated with IgG from healthy individuals (NH-IgG), IgG from a pemphigus vulgaris patient (PV-IgG) or with Dsg3-positive plasma samples from healthy blood donors (samples #1382, #3167 and #9607), as well as with a Dsg3-negative sample (sample #4025). For all experimental conditions (n = 3), the IgG concentration was 1 mg/ml (with the exception of the diluted PV-IgG samples). While NH-IgG and 4025 did not induce Dsg3 internalization, PV-IgG at 1:1 and 1:100 dilutions led to Dsg3 internalization. At low dilutions, PV-IgG and all tested Dsg3-positive samples from healthy blood donors did not show any effects on Dsg3 internalization