Desmin Modulates Muscle Cell Adhesion and Migration

Abstract

Cellular adhesion and migration are key functions that are disrupted in numerous diseases. We report that desmin, a type-III muscle-specific intermediate filament, is a novel cell adhesion regulator. Expression of p.R406W mutant desmin, identified in patients with desmin-related myopathy, modified focal adhesion area and expression of adhesion-signaling genes in myogenic C2C12 cells. Satellite cells extracted from desmin-knock-out (DesKO) and desmin-knock-in-p.R405W (DesKI-R405W) mice were less adhesive and migrated faster than those from wild-type mice. Moreover, we observed mislocalized and aggregated vinculin, a key component of cell adhesion, in DesKO and DesKI-R405W muscles. Vinculin expression was also increased in desmin-related myopathy patient muscles. Together, our results establish a novel role for desmin in cell-matrix adhesion, an essential process for strength transmission, satellite cell migration and muscle regeneration. Our study links the patho-physiological mechanisms of desminopathies to adhesion/migration defects, and may lead to new cellular targets for novel therapeutic approaches.

Keywords: desmin; focal adhesion; intermediate filaments; migration; myopathies; vinculin.

FIGURE 1

Human desmin expression and DNA chip analysis in stable cell lines. (A) The amounts of endogenous and human (Myc tagged) desmin in WT and p.R406W expressing clones were compared by Western blot. Actin was used as a loading control. (B) Genes that were upregulated or downregulated in C2C12 cells expressing R406W desmin or WT desmin were compared (right) and clustered using gene ontology enrichment analysis (left). The number of hits and p-values are respectively shown in brackets (C) Examples of up-regulated (right) or down-regulated (left) genes from the clusters determined in (B).

FIGURE 2

Desmin and vinculin morphology in C2C12 cells. (A) Representation of WT or R406W mutant desmin fused to GFP at the N-terminus. (B) GFP-WT desmin or GFP-R406W desmin were electroporated into C2C12 myoblasts. Vinculin was stained in red with Alexa-568. Epifluorescence images were taken 24 h after transfection. The p.R406W expressing cells show intracytoplasmic aggregates of desmin, unlike those expressing wild-type desmin. The cells were observed using a ×60 immersion objective. The scale bar corresponds to 20 μm.

FIGURE 3

Effect of desmin mutation on the total area of focal adhesions in C2C12 cells. (A) Average area of cells electroporated with GFP, GFP-WT desmin or GFP-R406W desmin. (B) Average vinculin patch number per cell counted for the three cell lines and (C) Average total vinculin patch area per cell area for the three cell lines, for the entire cells (“Total”), for the region of the cells located under the nucleus (“Near nucleus”) and for the cell minus this nuclear zone (“External”). Kruskal-Wallis tests were performed. *p < 0.05, **p < 0.01. The measurements were performed with a home-designed ImageJ macro described in the Materials and Methods section and Supplementary Figure S1. These results represent the average of n > 40 cells per condition from three independent experiments.

FIGURE 4

Relative protein expression of desmin, vinculin and vimentin in proliferating and differentiating satellite cells. Relative protein expression of desmin in proliferating (A,B) and differentiating (C,D) satellite cells was quantified by immunodetection and normalized by GAPDH protein expression. Each histogram represents the average of two lines per condition (3 for DesKI) and three independent experiments. The relative protein expression of vinculin was measured in proliferating (E,F) and differentiating (G,H) satellite cells by Western blot and normalized to the expression of GAPDH. The relative protein expression of vimentin was measured in proliferating (I,J) satellite cells by Western blot and normalized to the expression of GAPDH. Non-parametric Kruskal-Wallis tests were performed on the samples with: n.s, not significant; *p < 0.05; **p < 0.01 and ***p < 0.001.

FIGURE 5

Detachment of satellite cells during the application of a shear flow. (A) A syringe-pump delivered a constant flow of medium through tubing on cells seeded on Ibidi slides. Images were acquired every 15 s during the experiment. (B) Satellite cells detached over time. After 24 min, the remaining cells were counted. More WT satellite cells remained adhered to the substrate than DesKO and DesKI-R405W satellite cells. Images were taken on an inverted microscope using a ×10 objective. Scale bar = 200 µm. (C) The number of cells remaining adhered to the substrate were counted to measure the percentage of detachment over time. A significant difference in detachment was measured between the WT satellite cells and the DesKO and DesKI-R405W mutant satellite cells. A non-parametric Friedman test was used: ***p < 0.001. The error bars correspond to the SD. The experiment was repeated 3 times with two independent satellites cells lines extracted from two mice each for WT and KO genotype and three independent satellites cells lines extracted from three mice for KI genotype.

FIGURE 6

DesKO and DesKI-R405W cells move faster on fibronectin and laminin substrates. Cells were seeded on culture wells coated with fibronectin (A,C,E) or laminin (B,D,F). Pictures were taken every 7 min. Cells were tracked by hand using ImageJ software and its manual tracking plugin. Each average represents two different cell extractions for each mice genotype, and experiments were repeated 3 times. with n > 150 cells and the error bars represent the SD. (A,B) The mean speed of each cell line is represented. The distance to the origin on fibronectin (C) and laminin (D) was calculated from the coordinates obtained during the migration follow-up with ImageJ software. WT DesKO and DesKI-R405W cell migration way was calculated with Excel software from a representative sample of eight cells per genotype. Cell persistence on fibronectin (E) and laminin (F) was calculated by their distance from the origin and their total migration distance. A Kruskal-Wallis test was performed, and the error bars correspond to the SD: n.s, not significant; *p < 0.05; ***p < 0.001

FIGURE 7

Relative vinculin and desmin protein expression in DesKO and DesKI-R405W mouse soleus. The amount of vinculin was measured in the soleus of mutant DesKO (A,B) and DesKI-R405W mice (C,D). The amount of desmin was measured in the soleus of mutant DesKI-R405W mice (E,F). Each histogram represents the average relative protein expression of four independent GAPDH-standardized Western blot experiments for 7 WT mice and 7 DesKO mice or 6 WT and 5 DesKI-R405W mice. A Mann-Whitney non-parametric test was performed, with *p < 0.05; **p < 0.01; ***p < 0.001. Error bars represent the SD.

FIGURE 8

Localization of vinculin in the soleus of DesKO and DesKI-R405W mice. Microscopic images of soleus cross sections from DesKO mice at 5 months (A) and 1 year of age (B) and DesKI-R405W 3-month-old mice (C). Vinculin accumulations in and around the fibers are shown by the white arrows and central nuclei by red arrows. Laminin was stained in red, vinculin in green and nuclei in blue. Images were taken with a ×40 objective, and the scale bar corresponds to 30 μm.

FIGURE 9

Vinculin relative protein expression in patient muscle biopsies. The amounts of vinculin, actin, and desmin were measured in muscle biopsies from controls (CT) or patients carrying homozygous mutations in the desmin gene. GAPDH was used as a loading control.

FIGURE 10

Schematic representation of desmin roles in adhesion and migration and the potential consequences on muscle.