Matrix stiffness-induced myofibroblast differentiation is mediated by intrinsic mechanotransduction

Abstract

The mechanical properties of the extracellular matrix have recently been shown to promote myofibroblast differentiation and lung fibrosis. Mechanisms by which matrix stiffness regulates myofibroblast differentiation are not fully understood. The goal of this study was to determine the intrinsic mechanisms of mechanotransduction in the regulation of matrix stiffness-induced myofibroblast differentiation. A well established polyacrylamide gel system with tunable substrate stiffness was used in this study. Megakaryoblastic leukemia factor-1 (MKL1) nuclear translocation was imaged by confocal immunofluorescent microscopy. The binding of MKL1 to the α-smooth muscle actin (α-SMA) gene promoter was quantified by quantitative chromatin immunoprecipitation assay. Normal human lung fibroblasts responded to matrix stiffening with changes in actin dynamics that favor filamentous actin polymerization. Actin polymerization resulted in nuclear translocation of MKL1, a serum response factor coactivator that plays a central role in regulating the expression of fibrotic genes, including α-SMA, a marker for myofibroblast differentiation. Mouse lung fibroblasts deficient in Mkl1 did not respond to matrix stiffening with increased α-SMA expression, whereas ectopic expression of human MKL1 cDNA restored the ability of Mkl1 null lung fibroblasts to express α-SMA. Furthermore, matrix stiffening promoted production and activation of the small GTPase RhoA, increased Rho kinase (ROCK) activity, and enhanced fibroblast contractility. Inhibition of RhoA/ROCK abrogated stiff matrix-induced actin cytoskeletal reorganization, MKL1 nuclear translocation, and myofibroblast differentiation. This study indicates that actin cytoskeletal remodeling-mediated activation of MKL1 transduces mechanical stimuli from the extracellular matrix to a fibrogenic program that promotes myofibroblast differentiation, suggesting an intrinsic mechanotransduction mechanism.

Figure 1.

Stiff matrix promotes α-smooth muscle actin (α-SMA) expression by normal human lung fibroblasts. (A) Stiffness/elasticity of soft and stiff polyacrylamide (PA) hydrogels were determined by atomic force microscope indentation over multiple locations. Young’s modulus of each individual location (dots) on gels and the mean values of soft and stiff gel stiffness (long horizontal lines) ± SD (short horizontal lines) are shown. (B) Normal human lung fibroblasts (CCL-210) were cultured on collagen-coated soft, stiff, and 2-kPa PA gels for 24 to 72 hours. Levels of α-SMA mRNA were determined by real-time PCR. 18S rRNA was used as reference control. The level of α-SMA mRNA from cells cultured on soft gels for 24 hours was set at 1. (C) Levels of α-SMA protein were determined by immunoblot. Relative levels of α-SMA protein were determined by scanning densitometry of the blots and normalized to glyceraldehyde 3-phosphate dehydrogenase (GAPDH) expression. The level of α-SMA protein from cells cultured on soft gels for 24 hours was set at 1. Results are the means of three separate experiments ± SD. *P < 0.05 for comparisons as indicated.

Figure 2.

Stiff matrix induces changes in actin dynamics in favor of F-actin polymerization and α-SMA–containing stress fiber formation and promotes nuclear translocation of MKL1. CCL-210 cells were cultured on soft and stiff PA gels for 48 hours. (A) Changes in actin dynamics were determined by measurements of F/G-actin content using immunoblot and densitometric analyses. (B) Cells were fixed and stained for α-SMA with FITC-conjugated anti–α-SMA antibody (green) and F-actin with rhodamine-conjugated phalloidin (red). Confocal immunofluorescent images were taken and overlaid to colocalize α-SMA expression and F-actin (yellow). Nuclei were stained with DAPI. Scale bars: 50 μm. (C) Subcellular localization of MKL1 was determined by protein levels of MKL1 in the cytosolic fraction (Cytosol) and the nuclear fraction (nuclear). Relative levels of MKL1 protein were determined by scanning densitometry of the blots and normalized to GAPDH (for cytosolic MKL1) or lamin A/C (for nuclear MKL1). The level of MKL1 protein from cells cultured on soft gels was set at 1. Results are the means of three separate experiments ± SD. *P < 0.05. (D) Immunofluorescent staining and confocal microscopy were performed to visualize subcellular localization of MKL1 (green). Nuclei were stained with DAPI (blue). Scale bars: 50 μm.

Figure 3.

Stiff matrix promotes the formation of nuclear MKL1-SRF transactivator complex and the binding of the complex to α-SMA promoter, resulting in enhanced α-SMA promoter activity. (A) CCL-210 cells were cultured on soft and stiff PA gels for 48 hours. Nuclear proteins were extracted and subjected to immunoprecipitation with anti-SRF antibody (αSRF). Immunoprecipitates were blotted with anti-MKL1 antibody (αMKL1). Relative levels of MKL1 protein were determined by scanning densitometry of the blots and normalized to lamin A/C in the input. The level of MKL1 from cells on soft gels was set at 1. (B) Chromatin isolated from cells cultured on soft and stiff PA gels was immunoprecipitated with anti-SRF antibody. DNA in immunoprecipitated chromatin was purified and subjected to real-time PCR to quantify the amounts of α-SMA promoter fragment. Amplification of the same α-SMA promoter fragment using DNA extracted from preimmunoprecipitated chromatin as templates were used as reference control. (C) Schematic drawing of a 765 nt wild-type (WT) rat proximal α-SMA promoter fragment and three mutated promoter fragments harboring CArG box 1 mutations (CArG-b1 mut), CArG box2 mutations (CArG-b2 mut), and CArG box1 and box2 mutations (CArG-b1and2 mut). WT and mutated promoter reporters were transfected into CCL-210 cells. Transfected cells were cultured on soft and stiff PA gels for 24 hours. Promoter activity was determined by luciferase assay. Results are the means of three separate experiments ± SD; each experiment was performed in triplicate. *P < 0.05.

Figure 4.

Mkl1 deficiency renders mouse lung fibroblasts resistant to stiff matrix–induced myofibroblast differentiation. (A) Cell lysates from Mkl1^+/+ and Mkl1^−/− lung fibroblasts were analyzed for Mkl1 expression with immunoblot. GAPDH was used as loading control. (B) Mkl1^+/+ and Mkl1^−/− lung fibroblasts were cultured on stiff and soft PA gels for 48 hours. Levels of α-SMA mRNA were determined by real-time PCR. 18S rRNA was used as reference control. The level of α-SMA mRNA from cells cultured on soft gels was set at 1. (C) Levels of α-SMA protein were determined by immunoblot. Relative levels of α-SMA protein were determined by scanning densitometry of the blots and normalized to GAPDH expression. The level of α-SMA protein from cells cultured on soft gels was set at 1. (D) Mkl1^−/− lung fibroblasts were transfected with green fluorescent protein (GFP)-human MKL1 cDNA expression vector or GFP vector alone (control). Expression of GFP-MKL1 fusion protein was determined by immunoblot with anti-GFP antibody. Transfected cells were cultured on stiff and soft PA gels for 48 hours. α-SMA protein levels were determined as described in C. Results are the means of at least three separate experiments ± SD. (E) Mkl1^−/− lung fibroblasts transfected with GFP-MKL1 and empty vectors were cultured on soft and stiff PA gels. α-SMA stress fiber formation was visualized by immunofluorescent staining and confocal microscopy. GFP-MKL1 translocated into nuclear under a stiff matrix condition. Scale bars: 20 μm. *P < 0.05 for comparisons as indicated.

Figure 5.

Overexpression of constitutively active MKL1 (caMKL1) and promotion of actin polymerization enable human lung fibroblasts to differentiate into myofibroblasts on soft matrix. (A) GFP-caMKL1 expression vector or GFP vector (control) was transfected into CCL-210 cells. GFP-caMKL1 expression in the nuclear fraction and the cytosolic fraction was determined by immunoblot with anti-GFP antibody (αGFP). GAPDH and lamin A/C were used as loading control. (B) Subcellular localization of GFP and GFP-caMKL1 in transfected cells were imaged with confocal immunofluorescent microscopy. Nuclei were stained by DAPI. Scale bars: 50 μm. (C) Lung fibroblasts expressing GFP-caMKL1 or GFP alone were cultured on soft PA gels for 48 hours. Levels of α-SMA protein were determined by immunoblot. Relative levels of α-SMA protein were determined by scanning densitometry of the blots and normalized to GAPDH expression. The level of α-SMA protein from control GFP-expressing cells was set at 1. Results are the means of three separate experiments ± SD. α-SMA stress fibers were stained using anti–α-SMA antibody followed by rhodamine-conjugated secondary antibody and imaged by confocal microscopy. (D) CCL-210 cells were cultured on soft PA gels in the presence of 200 nM jasplakinolide (Jas) or an equal volume of PBS for 24 hours. Subcellular localization of MKL1 was determined by immunofluorescent analysis followed by confocal microscopy. Nuclei were stained by DAPI. (E) Cells were treated as described in D. Levels of α-SMA protein and α-SMA stress fiber formation were determined as described in C. *P < 0.05. Scale bars: 50 μm.

Figure 6.

Stiff matrix promotes RhoA production and activation and increases ROCK activity. CCL-210 cells were cultured on soft and stiff PA gels for 48 hours. (A) Levels of RhoA mRNA and protein (arrow) were determined by real-time PCR and immunoblot, respectively. 18S rRNA was used as reference control (for real-time PCR). GAPDH was used as loading control (for immunoblot). (B) RhoA activity was determined by Rhotekin pull-down assay. Levels of Rhotekin-binding active RhoA were determined by immunoblot with anti-RhoA antibody (αRhoA). GAPDH in the input was used as loading control. (C) Subcellular localization of RhoA was determined by RhoA levels in the cytosolic faction (Cytosol) (arrow) and the membrane fraction (Mem) (arrow). GAPDH was used as loading control. (D) RhoA–ROCK interaction. Cell lysates were immunoprecipitated with anti-RhoA antibody (αRhoA). Immunoprecipitated proteins were blotted with anti-ROKα/ROCK II antibody (αROCK). GAPDH in the input was used as loading control. (E) ROCK activity was determined by incubation of cell lysates with ROCK-specific substrates and MYPT1 followed by colorimetric immunoassay. The level of ROCK activity from cells cultured on soft gels was set at 1. Results are the means of three separate experiments ± SD. (F) Levels of phosphorylated 20-kD myosin light chain (pMLC₂₀) and total 20-kD myosin light chain (MLC₂₀) were determined by immunoblot analysis. GAPDH was used as loading control. *P < 0.05 for comparisons as indicated.

Figure 7.

RhoA/ROCK inhibition blocks stiff matrix–induced actin cytoskeletal reorganization, MKL1 nuclear translocation, and lung myofibroblast differentiation. CCL-210 cells were cultured on stiff PA gels (A–D) or compliant and stiff PA gels (E and F) in the presence of 10 μM Y-27632 or an equal volume of PBS (vehicle control) for 48 hours. (A) Changes in actin dynamics were determined by immunoblot and densitometric analyses as described previously. (B) α-SMA–containing stress fibers were determined by costaining α-SMA (green) and F-actin (red). Images were taken under a confocal microscope. Nuclei were stained with DAPI. Scale bars: 50 μm. (C and D) Subcellular localization of MKL1 were determined by cell fractionation followed by immunoblot analysis. Confocal immunofluorescent microscopy was used to visualize MKL1 (green) subcellular localization. Nuclei were stained with DAPI (blue). Scale bars: 50 μm. (E) Levels of pMLC₂₀ and total MLC₂₀ were determined by immunoblot analysis. GAPDH was used as loading control. (F) Levels of α-SMA mRNA were determined by real-time PCR. 18S rRNA was used as reference control. (G) Levels α-SMA protein were determined by immunoblot. GAPDH was used as loading control. Relative levels of α-SMA mRNA and protein were determined by scanning densitometry and normalized to 18S rRNA or GAPDH expression. The levels of α-SMA mRNA or protein from cells cultured on soft gels in the presence of PBS were set at 1. Results are the means of three separate experiments ± SD. *P < 0.05.