The impact of signaling pathways on the desmosome ultrastructure in pemphigus

Abstract

Introduction: The autoantibody-driven disease pemphigus vulgaris (PV) impairs desmosome adhesion in the epidermis. In desmosomes, the pemphigus autoantigens desmoglein 1 (Dsg1) and Dsg3 link adjacent cells. Dsgs are clustered by plaque proteins and linked to the keratin cytoskeleton by desmoplakin (Dp). The aim of this study was to identify the impact of several PV-related signaling pathways on desmosome ultrastructure.

Methods: STED microscopy, Dispase-based dissociation assay.

Results: As observed using STED microscopy, pemphigus autoantibodies (PV-IgG) reduced desmosome number, decreased desmosome size, increased plaque distance and thickness and caused loss of adhesion. Decreased desmosome number, increased plaque distance and thickness and loss of adhesion correlate with features found for newly assembled immature desmosomes, observed after Ca²⁺ depletion and repletion. This was paralleled by plaque asymmetry, keratin filament retraction and fragmentation of Dsg1 and Dsg3 immunostaining. Inhibition of each individual signaling pathway investigated here prevented the loss of adhesion and ameliorated keratin retraction. In addition, inhibition of p38MAPK or PLC completely rescued all parameters of desmosomes ultrastructure and increased desmosome number under basal conditions. In contrast, inhibition of MEK1/2 was only partially protective for desmosome size and plaque thickness, whereas inhibition of Src or increase of cAMP decreased desmosome size but increased the desmosome number even in the presence of PV-IgG.

Discussion: Alterations of the desmosomal plaque ultrastructure are closely related to loss of adhesion and regulated differently by signaling pathways involved in pemphigus pathogenesis. This insight may allow identification of novel treatment options targeting specific steps of desmosome turn-over in the future.

Keywords: adhesion; autoantibodies; autoimmune disease; desmosomes; epidermis; pemphigus; skin.

Figure 1

The composition of desmosomes in epidermal keratinocytes in vitro. (A) Representative STED-microscopy images (Scale bar: 500 nm). (B) Quantification results for relative area of colocalized pixels (N=4-5). * indicates statistically significant differences as indicated, # indicates statistical significant results as compared to epidermal keratinocytes in the spinous layer of human epidermis ex vivo as determined in a previous study (49), both in two-way-ANOVA with Sidaks post analysis for multiple comparisons p<0.05.

Figure 2

Impact of the used mediators and Ca²⁺-depletion on cell adhesion in HaCaT cells in vitro, as determined by dispase-based dissociation assays. (A) Representative Images of cells pretreated with mediators for 1 h followed by PV- or C-IgG for 24 h, stained for viability using MTT before sheering. (B) Quantification results for the number of fragments (N=3). (C) Representative Images of cells after depletion of Ca²⁺ with either 2.5 mM EGTA or 5 mM EGTA for 30 min or 1 h and with or without repletion of Ca²⁺ for 24 h. (D) Quantification results for the number of fragments (N=4). (E) Visual representation of the time course of Ca²⁺ depletion and repletion using EGTA. All experiments were performed using the human epidermal keratinocyte cell line: HaCaT. * indicates statistically significant differences as indicated in two-way-ANOVA with Sidaks post analysis for multiple comparisons p<0.05.

Figure 3

Qualitative summary of effects of PV-IgG or anisomycin treatment on the Dp plaque ultrastructure. Representative STED-microscopy images (Scale bars left to right 2 µm, 500 nm, 100 nm or yellow: 250 nm). (A) Control condition. (B) effects of PV-IgG treatment. (C) Effects of Anisomycin treatment. The lines indicate measurements of the example desmosome under control conditions. Red: length and width, purple: plaque thickness, blue: Plaque distance. Green arrows indicate missing halfs of split desmosomes, red arrows indicate linear streaks, blue arrows indicate areas where Dp staining is clustered and red circles indicate double Dp structures in the cytoplasm below the cell border. Green boxes indicate zoomed areas.

Figure 4

Quantification of effects of PV-IgG or anisomycin treatment on the Dp plaque ultrastructure. Quantification results for desmosome number, desmosome size, desmoplakin distance and plaque thickness (N=4-8). (A) inhibition of p38MAPK. (B) after activation of p38MAPK. (C) inhibition of PLC. (D) inhibition of MEK1/2. (E) inhibition of Src. (F) After induction of increased cAMP levels. * indicates statistically significant differences as indicated in two-way-ANOVA with Sidaks post analysis for multiple comparisons p<0.05.

Figure 5

Quantitative summary of effects of PV-IgG or anisomycin treatment on the Dp plaque ultrastructure and evaluation of plaque symmetry and keratin retraction. (A) Quantification results for desmosome size for the subset after treatment with mediators and C-IgG compared to vehicle control for c-IgG, PV-IgG and anisomycin (N=4-8). (B) Quantification results for desmosome size for the subset after treatment with mediators and PV-IgG compared to the vehicle control for c-IgG, PV-IgG and anisomycin (controls are the same as for a) N=4-8). (C) Quantification results for desmosome number for the subset after treatment with mediators and C-IgG compared to vehicle controls for C-IgG, PV-IgG and anisomycin (N=4-8). (D) Quantification results for desmosome number for the subset after treatment with mediators and PV-IgG compared to the vehicle control for c-IgG, PV-IgG and anisomycin (controls are the same as for C) N=4-8). (E) Quantification results for plaque asymmetry after treatment with mediators and C-IgG compared to PV-IgG, anisomycin and control conditions (N=4-8). (F) Quantification results for keratin distance (keratin retraction) after treatment with mediators and C-IgG or PV-IgG (N=4-8). (G) Representative STED-microscopy images for delaminated cells showing more desmosomes along the cell borders after PP2 or F/R treatment, compared to cells in the monolayer showing still more desmosomes than under control conditions (Red boxes indicate zoomed in area. Scale bar top: 2 µm; bottom: 10 µm). * indicates statistically significant differences as indicated in two-way-ANOVA with Sidaks post analysis for multiple comparisons p<0.05.

Figure 6

Quantification of effects of Ca²⁺ depletion and repletion on the Dp plaque ultrastructure. Quantification results for (A) desmosome number. (B) desmosome size. (C) desmoplakin distance. (D) plaque thickness (N=4). (E) Representative STED-microscopy images showing alterations in the Dp plaque ulstrastructure (Scale bar: 500 nm). * indicates statistically significant differences as indicated in two-way-ANOVA with Sidaks post analysis for multiple comparisons p<0.05.

Figure 7

Quantification of effects of PV-IgG or anisomycin on Dsg1 and Dsg3 staining distribution and intensity. Quantification results for fragmentation of Dsg1 and Dsg3 (N=3-5). (A) Inhibition of p38MAPK. (B) activation of p38MAPK (C) inhibition of PLC (D) inhibition of MEK1/2 (E) inhibition of Src (F) After induction of increased cAMP levels. (G) Quantification results for Dsg1 staining intensity along the cell borders after treatment with mediators and C-IgG or PV-IgG (N=3-5). (G) Quantification results for Dsg1 staining intensity along the cell borders after treatment with mediators and C-IgG or PV-IgG (N=3-5). (H) Quantification results for Dsg3 staining intensity along the cell borders after treatment with mediators and C-IgG or PV-IgG (N=4-5). (I) Quantification results for Dsg1 staining intensity in the cytoplasm after treatment with mediators and C-IgG or PV-IgG (N=3-5). (J) Quantification results Dsg3 staining intensity in the cytoplasm after treatment with mediators and C-IgG or PV-IgG (N=4-5). * indicates statistically significant differences as indicated in two-way-ANOVA with Sidaks post analysis for multiple comparisons p<0.05.

Figure 8

Qualitative display of the impact of mediators on the distribution and association of Dsg3 with Dp and keratin filaments. Representative STED-microscopy images after 1 h pretreatment with mediators followed by 24 h of treatment with C-IgG or PV-IgG for: (A) Costaining of Dsg3 and Dp. (B) Costaining of Dsg3 and pan-cytokeratin (Ker). (White arrows indicate Dsg3 associated with keratin filaments. White boxes indicate zoomed in area. Scale bar left to right: 2 µm, 500 nm).

Figure 9

Qualitative display of the impact of mediators on the distribution and association of Dsg1 with Dp and keratin filaments. Representative STED-microscopy images after 1 h pretreatment with mediators followed by 24 h of treatment with C-IgG or PV-IgG for: (A) Costaining of Dsg1 and Dp. (B) Costaining of Dsg1 and pan-cytokeratin (Ker). (White arrows indicate Dsg1 associated with keratin filaments. White boxes indicate zoomed in area. Scale bar left to right: 2 µm, 500 nm).