Cell-cell adhesions and cell contractility are upregulated upon desmosome disruption

Abstract

Desmosomes are perturbed in a number of disease states - including genetic disorders, autoimmune and bacterial diseases. Here, we report unexpected changes in other cell-cell adhesion structures upon loss of desmosome function. We found that perturbation of desmosomes by either loss of the core desmosomal protein desmoplakin or treatment with pathogenic anti-desmoglein 3 (Dsg3) antibodies resulted in changes in adherens junctions consistent with increased tension. The total amount of myosin IIA was increased in desmoplakin-null epidermis, and myosin IIA became highly localized to cell contacts in both desmoplakin-null and anti-Dsg3-treated mouse keratinocytes. Inhibition of myosin II activity reversed the changes to adherens junctions seen upon desmosome disruption. The increased cortical myosin IIA promoted epithelial sheet fragility, as myosin IIA-null cells were less susceptible to disruption by anti-Dsg3 antibodies. In addition to the changes in adherens junctions, we found a significant increase in the expression of a number of claudin genes, which encode for transmembrane components of the tight junction that provide barrier function. These data demonstrate that desmosome disruption results in extensive transcriptional and posttranslational changes that alter the activity of other cell adhesion structures.

Figure 1. Changes in adherens junctions in DP-null keratinocytes.

(A-B) WT and DP-null keratinocytes stained for E-cadherin (green) and α-catenin (red) after 24 h in Ca²⁺. (C-F) WT (C,D) and DP-null (E,F) keratinocytes stained with the tension-sensitive anti-α-catenin antibody, α18 (red). The α18 epitope is exposed in WT keratinocytes after microtubules are reorganized to the cell cortex with taxol treatment, 1 hour at 10 µM. (D). DP-null keratinocytes have the α18 epitope exposed at steady state (E). The junctional intensity is not significantly changed upon treatment with taxol (F). (G) Quantitation of junctional/cytoplasmic intensity of α18 in indicated samples. n>100 cells from at least two independent experiments. p values as compared to WT DMSO are <0.0001 for both WT with taxol and DP KO DMSO. There was no significant difference between DP KO DMSO verses taxol. (H) α18 (red) staining of DP null keratinocytes treated with the myosin II inhibitor blebbistatin (25 µM for 1 hour). (I,J) E-cadherin (red) and α-catenin (green) localization in DP cells treated either with DMSO (I) or 25 µM blebbistatin (J) for 1 hour. Scale bar, 10 µm.

Figure 2. Myosin II levels and localization are upregulated in DP-null cells and epidermis.

(A-D) WT (A, C) and DP cKO (B, D) epidermis at E17.5 (A, B) and E18.5 (C, D) were stained for myosin IIA (green). Basement membrane is marked with a dashed line. Scale bar, 10 µm. (E) Myosin II levels in E18.5 WT and DP cKO epidermis were evaluated by Western blot. β-tubulin is the loading control. (F, G) WT and DP-null keratinocytes were stained for myosin IIA (green). (H) Quantitation of cortical/cytoplasmic ratios of myosin IIA in WT and DP null cells. n>60 cells for two independent experiments, p = 0.027.

Figure 3. Pathogenic pemphigus antibodies induce increased contractility in WT keratinocytes.

(A-C) WT keratinocytes were stained for myosin IIA (green). Cortical myosin II staining is observed when cells are treated with taxol to increase tension (B), or when treated with pathogenic pemphigus antibodies (AK23, in C). (D-F) WT keratinocytes were stained for α18 (red). The tension-sensitive epitope of α-catenin is exposed after taxol treatment (E) or after pathogenic pemphigus antibody treatment (F). (G,H) Desmoplakin (red) localization in WT cells treated with normal mouse sera (G) or with AK23 antibodies (H). Scale bars, 10 µm. (I) WT and myosin IIA-null keratinocytes were subjected to cell sheet disruption after treatment with control IgG or pathogenic pemphigus antibody. **, p<.005, n = 4.

Figure 4. Tight junctions are altered upon loss of desmoplakin.

(A-B′) Calcium was added to WT and DP-null keratinocytes, and cells were fixed at various time points and stained for tight junction proteins occludin (red) and ZO-1 (green). Scale bar, 10 µm. (C,D) ZO-1 staining of WT and DP-null keratinocytes at 3 hour after calcium switch. (E) Co-stain for ZO-1 (red) and E-cadherin (green) in DP null keratinocytes 3 hours after calcium switch. (F) RNA was isolated from WT and DP-null keratinocytes, and RT-PCR for several claudins was performed. ***, p<.0005, **, p<.005. n = 3. (G-I) Western blot analysis of total levels of claudin-1 and β-actin in lysates from cultured keratinocytes (G,H) and from isolated epidermis (I). (J,K) Claudin 1 staining of WT and DP-null keratinocytes 24 hours after calcium shift. (L,M) Immunofluorescence analysis of claudin 1 (green) and β4-integrin (red) in wild type (L) and desmoplakin conditional null epidermis (M). Scale bars are 10 µm.