Lamins and nesprin-1 mediate inside-out mechanical coupling in muscle cell precursors through FHOD1

Abstract

LINC complexes are crucial for the response of muscle cell precursors to the rigidity of their environment, but the mechanisms explaining this behaviour are not known. Here we show that pathogenic mutations in LMNA or SYNE-1 responsible for severe muscle dystrophies reduced the ability of human muscle cell precursors to adapt to substrates of different stiffness. Plated on muscle-like stiffness matrix, mutant cells exhibited contractile stress fibre accumulation, increased focal adhesions, and higher traction force than controls. Inhibition of Rho-associated kinase (ROCK) prevented cytoskeletal defects, while inhibiting myosin light chain kinase or phosphorylation of focal adhesion kinase was ineffective. Depletion or inactivation of a ROCK-dependent regulator of actin remodelling, the formin FHOD1, largely rescued morphology in mutant cells. The functional integrity of lamin and nesprin-1 is thus required to modulate the FHOD1 activity and the inside-out mechanical coupling that tunes the cell internal stiffness to match that of its soft, physiological-like environment.

Figure 1

Cell responses to different substrate stiffness. (A) Phalloidin staining of the F-actin of fixed WT, Nespr-1^ΔKASH and LMNA^ΔK32 myoblasts on fibronectin-coated glass and gel substrates of 700 kPa, 20 kPa, 12 kPa and 5 kPa. Nuclei are stained with DAPI. Scale bar: 40 µm. (B) Projected cell area as a function of substrate stiffness. Analysis was performed on glass and gel substrates of 700 kPa, 20 kPa, 12 kPa, and 5 kPa (each n > 50 cells). Values are means ± SEM; ^$p < 0.001 vs corresponding cell line value on glass; *p < 0.001 vs WT value at similar substrate rigidity.

Figure 2

Actin cytoskeleton on soft matrix (12 kPa). (A) Confocal images of WT, Nespr-1^ΔKASH and LMNA^ΔK32 myoblasts on soft matrix (12 kPa) close to physiological muscle stiffness and stained for F-actin (phalloidin, red) and non-muscle myosin 2 A (NM-2A, green). Nuclei are stained with DAPI (blue). Scale bar: 10 µm. (B,C) Zoom-in of actin cytoskeleton at the cell periphery (B) and in the perinuclear regions (C). In C, confocal images are taken at the apical and basal surface of the cell. Scale bar: 5 µm. (D) Supranuclear actin cable number in WT, Nespr-1^ΔKASH and LMNA^ΔK32 myoblasts. Values are means ± SEM; ***p < 0.001 compared with WT. (E) Nuclear thickness in WT, Nespr-1^ΔKASH and LMNA^ΔK32 myoblasts. Values are means ± SEM; **p < 0.01 compared with WT. (F) mRNA expression of MYH9 gene expression coding for NM-2A in WT, Nespr-1^ΔKASH and LMNA^ΔK32 myoblasts. Values are means ± SEM; **p < 0.01, *p < 0.05 compared with WT.

Figure 3

Cell matrix adhesions on soft matrix (12 kPa). (A) Confocal images of WT, Nespr-1^ΔKASH and LMNA^ΔK32 myoblasts on soft matrix (12 kPa) and stained with antibody against vinculin (green). Nuclei are stained with DAPI (blue). Scale bar: 10 µm. (B) Binary images of vinculin staining obtained from A. (C,D) Quantification of vinculin size (C) and number of focal adhesions per cell (D) in WT, Nespr-1^ΔKASH and LMNA^ΔK32 myoblasts obtained from binary images. Values are means ± SEM in at least 12 myoblasts per line; *p < 0.05, **p < 0.01, ***p < 0.001 compared with WT. (E) Histogram of VCL mRNA expression in WT, Nespr-1^ΔKASH and LMNA^ΔK32 myoblasts. Values are means ± SEM; *p < 0.05 compared with WT and expressed in arbitrary units (au). Values are means ± SEM, n = 5 in each cell line from 2 separate experiments. (F,G) Effects of FAK phosphorylation inhibition on cell spreading in WT, Nespr-1^ΔKASH and LMNA^ΔK32 myoblasts. Values are expressed as µm² (F) and as percent of baseline values for each cell line (G). At least 50 cells of each type were measured, ***p < 0.001 compared with WT. (H,I) Traction force microscopy in WT, Nespr-1^ΔKASH and LMNA^ΔK32 myoblasts. (H) Typical images of traction isostresses (Pa) and traction stress vectors (Pa) in WT, Nespr-1^ΔKASH and LMNA^ΔK32 myoblasts. (I) Mean values ± SEM obtained from n ≥ 9 in each cell line; ***p < 0.001 compared with WT.

Figure 4

Effects of ROCK and MLCK inhibition on actin cytoskeleton in myoblasts on soft matrix (12 kPa). (A,F) Confocal images of WT, Nespr-1^ΔKASH and LMNA^ΔK32 myoblasts on soft matrix and stained for F-actin (phalloidin, red) and NM-2A (green) after treatment with the ROCK inhibitor Y-27632 (A), and the MLCK inhibitor ML7 (F). Nuclei are stained with DAPI (blue). Scale bar: 10 µm. Images 1 and 2 show higher magnification of the perinuclear and periphery zones respectively. Zoom-in scale bar: 5 µm. (B,C) Supranuclear actin cable number in WT, Nespr-1^ΔKASH and LMNA^ΔK32 myoblasts after treatment with Y-27632 or ML7. Values are expressed as absolute numbers (B) and as percent of baseline values for each cell line (C). (D,E) Nucleus thickness in WT, Nespr-1^ΔKASH and LMNA^ΔK32 myoblasts after treatment with Y-27632 or ML7. Values are expressed as absolute numbers (B) and as percent of baseline values for each cell line (C). In (B–E), values are means ± SEM; ***p < 0.001 compared with WT (B,D) or compared to values before treatment (C,E). Only significant difference is figured.

Figure 5

FHOD1 expression in myoblasts on soft matrix (12 kPa). (A) Confocal images of WT, Nespr-1^ΔKASH and LMNA^ΔK32 myoblasts on soft matrix and stained for F-actin (phalloidin, red) and FHOD1 (green). Nuclei are stained with DAPI (blue). (B) Representative western-blot of FHOD1 in WT, Nespr-1^ΔKASH and LMNA^ΔK32 myoblasts cultured on soft substrates. Histogram represents FHOD1 quantifications obtained in WT, Nespr-1^ΔKASH and LMNA^ΔK32 myoblasts and normalized by GAPDH. Values are means ± SEM, n = 6 in WT, Nespr-1^ΔKASH and LMNA^ΔK32 myoblasts; *p < 0.05. Only significant difference is figured; au: arbitrary units. (C) Histogram represent mRNA concentrations of FHOD1, DIAPH1 and DIAPH3 normalized to β2 microglobulin expression and expressed in arbitrary units (au). Values are means ± SEM, n = 5 in WT, Nespr-1^ΔKASH and LMNA^ΔK32 myoblasts, ***p < 0.001 compared with WT. Only significant difference is figured.

Figure 6

siRNA against FHOD1 reduced actin contractility and nucleus thickness in myoblasts on soft matrix (12 kPa). (A) Confocal images of WT, Nespr-1^ΔKASH and LMNA^ΔK32 myoblasts on soft matrix and stained for F-actin (phalloidin, red) and NM-2A (green) after siRNA against FHOD1. Scale bar: 10 µm (B,C) Zoom-in of actin cytoskeleton at cell periphery (B) and in the perinuclear regions (C). Scale bar: 10 µm. (D,E) Supranuclear actin cable number in WT, Nespr-1^ΔKASH and LMNA^ΔK32 myoblasts after siRNA against FHOD1. Values are expressed as absolute numbers (D) and as percent of baseline values for each cell line (E). Values are means ± SEM, n ≥ 19 in WT, Nespr-1^ΔKASH and LMNA^ΔK32 myoblasts, ***p < 0.001 compared to values before siRNA against FHOD1. Only significant difference is figured. (F,G) Nuclear thickness in WT, Nespr-1^ΔKASH and LMNA^ΔK32 myoblasts after siRNA against FHOD1. Values are means ± SEM. n ≥ 19 in WT, Nespr-1^ΔKASH and LMNA^ΔK32 myoblasts, ***p < 0.001 compared to values before siRNA against FHOD1. Only significant difference is figured. (H,I) Cell spreading area in WT, Nespr-1^ΔKASH and LMNA^ΔK32 myoblasts after siRNA against FHOD1. Values are expressed as absolute values (H) and as percent of baseline values for each cell line (I). Values are means ± SEM, n ≥ 50 cells in each line; ***p < 0.001 compared to values before siRNA against FHOD1. Only significant difference is figured.

Figure 7

Impaired ability to sustain high external force in Nespr-1^ΔKASH and LMNA^ΔK32 cells. (A) Confocal images of WT, Nespr-1^ΔKASH and LMNA^ΔK32 myoblasts on fibronectin-coated rigid matrix (glass) and stained for F-actin (phalloidin, red) and non-muscle myosin IIA (NM-2A, green). Nuclei are stained with DAPI (blue). Scale bar: 10 µm. (B,C) Zoom-in of actin cytoskeleton in the perinuclear regions. Confocal images were taken at the apical (B) and basal (C) surface of the cell. (D,E) Supranuclear actin cable number on hard surface in WT, Nespr-1^ΔKASH and LMNA^ΔK32 myoblasts. Values expressed as absolute values (D) and as percent changes versus values obtained at 12 kPa (E). Values are means ± SEM. n ≥ 20 in WT, Nespr-1^ΔKASH and LMNA^ΔK32 myoblasts, ***p < 0.01 vs WT. Only significant difference is figured (F–H) Cell matrix adhesions on hard surface. (F) Confocal images of WT, Nespr-1^ΔKASH and LMNA^ΔK32 myoblasts on hard matrix (glass) and stained with antibody against vinculin (green). Nuclei are stained with DAPI (blue). Scale bar: 10 µm. (G,H) Histograms of vinculin area in WT, Nespr-1^ΔKASH and LMNA^ΔK32 myoblasts, expressed as absolute values (G) and as percent changes versus values obtained at 12 kPa (H). At least 50 cells of each type were analysed. *p < 0.05 and ***p < 0.001 compared with corresponding value obtained on 12 kPa substrate. Only significant difference is figured.