Unravel principles of pemphigus vulgaris pathogenesis

Abstract

Pemphigus vulgaris (PV) is an autoimmune disease affecting the mucosa and the skin. The autoantibodies which are produced in this disease, target desmoglein (Dsg) 3 and Dsg1. These two are adhesion molecules providing intercellular connection between epidermal, hair follicle and mucosal keratinocytes. Once the autoantibodies have bound to these structures, the keratinocytes separate from each other, leading to the formation of blisters in the skin and erosions in the mucosa. The exact molecular mechanism of the pathogenesis is not fully understood yet. This project has the aim to investigate and identify the clinically relevant communication code in the affected tissues. The human skin organ culture (HSOC) model for PV was used to uncover the early events in basal keratinocytes which were incubated with either the single chain variable fragment PX43 (directed against Dsg3/1) or the murine antibody AK23 (directed against Dsg3). The skin samples were prepared 5, 10, and 24 hours after injection for the quantification of split formation, proliferation, and the basal keratinocytes were isolated for bulk RNA-sequencing. Similarly, further analysis of the proteome was conducted 24 hours after injection and a qualitative staining for the binding of PX43 to Dsg3/1 was done. As another quality control, the distribution of Dsg3/1 in the human skin was visualized by immunofluorescence. As controls, human and murine immunoglobulin G were used, respectively. As it is evident from hematoxylin and eosin-stained sections, PX43 induced a split in the skin which was increasing with prolonged incubation time. However, AK23 was not able to induce split formation. Further analysis of the proliferation after incubation with PX43 and AK23 shows that the binding of them does not lead to a significant increase in proliferation at the site of basal and suprabasal keratinocytes. The analysis of the transcriptome used a multi-variate linear model including nuisance factors and reveals prominent upregulation of IFNγ and TNF𝜶 pathways in the basal keratinocytes. This coincides with the upregulation of NFkB and JAK-STAT-mediated circuits and increased keratinocyte detachment. In contrast, basal keratinocytes from AK23-injected HSOC models do not show significant differentially regulated genes. These two observations lead to the conclusion that the inflammatory response induced in the basal keratinocytes, does not originate from the binding of PX43 to Dsg3/1 but is a secondary response to the increasing cell detachment. Analysis of the proteome of basal keratinocytes after 24 hours confirmed the described transcriptomic results showing differentially regulated proteins and pathways of inflammation. All-together, the cell-wide and long-lasting changes on the transcriptomic and proteomic level originate from the loss of cell-cell adhesion and mechanical stress but are not caused by the binding of PX43 or AK23 to their targets.