Myocardin-related transcription factors A and B are key regulators of TGF-β1-induced fibroblast to myofibroblast differentiation

Abstract

Myofibroblasts are contractile, smooth muscle-like cells that are characterized by the de novo expression of smooth muscle α-actin (SMαA) and normally function to assist in wound closure, but have been implicated in pathological contractures. Transforming growth factor β-1 (TGF-β1) helps facilitate the differentiation of fibroblasts into myofibroblasts, but the exact mechanism by which this differentiation occurs, in response to TGF-β1, remains unclear. Myocardin-related transcription factors A and B (MRTFs, MRTF-A/B) are transcriptional co-activators that regulate the expression of smooth muscle-specific cytoskeletal proteins, including SMαA, in smooth muscle cells and fibroblasts. In this study, we demonstrate that TGF-β1 mediates myofibroblast differentiation and the expression of a contractile gene program through the actions of the MRTFs. Transient transfection of a constitutively active MRTF-A induced an increase in the expression of SMαA and other smooth muscle-specific cytoskeletal proteins, and an increase in myofibroblast contractility, even in the absence of TGF-β1. MRTF-A/B knockdown, in TGF-β1-differentiated myofibroblasts, resulted in decreased smooth muscle-specific cytoskeletal protein expression levels and reduced contractile force generation, as well as a decrease in focal adhesion size and number. These results provide direct evidence that the MRTFs are mediators of myofibroblast differentiation in response to TGF-β1.

Figure 1. TGF-β1 treatment is sufficient for increased expression of smooth muscle-specific cytoskeletal proteins

(A) REF-52 cells were plated in supplemented DMEM and stimulated with 0.25 ng/mL TGF-β1 for 96 hours. Total RNA was collected and analyzed for levels of smooth muscle-specific cytoskeletal protein gene expression by real-time PCR analysis. mRNA levels are relative to the non-stimulated control expression levels. Error bars represent mean ± AAD, **p<0.01, *p<0.05, or n.s.-not significant by Student’s t-test; n=3. (B) Representative Western blot. Total protein was extracted from REF-52 cells, after 96 hours of TGF-β1 treatment, and analyzed using the indicated antibodies. The quantification of each protein level is shown relative to control levels and normalized to β-tubulin. Error bars represent mean ± AAD.

Figure 2. MRTF-A is sufficient for myofibroblast formation in REF-52 cells independent of TGF-β1 treatment

(A) REF-52 cells were cultured in supplemented DMEM for 24 hours and then transfected with FLAG-tagged-CA-MRTF-A or empty vector control for 48 hours. Total RNA was extracted and analyzed for expression of smooth muscle-specific cytoskeletal proteins using real-time PCR. mRNA levels are relative to the TIMP2 empty vector control message levels. Error bars represent mean ± AAD, ***p<0.001, **p<0.01, *p<0.05, or n.s.-not significant by Student’s t-test; n=3–7. (B) Representative Western blot. Total protein was extracted from REF-52 cells after treatment described above, and analyzed using the indicated antibodies. The quantification of each protein level is shown relative to control levels and normalized to β-tubulin. Error bars represent mean ± AAD.

Figure 3. MRTF-A/B knockdown reduces smooth muscle-specific cytoskeletal protein expression in myofibroblasts

(A) TGF-β1-induced myofibroblasts were transfected with MRTF-A/B or NT siRNA for 48 hours. mRNA levels were analyzed by real-time PCR and normalized to NT siRNA; n= 5. (B) Representative Western blot. Total protein was extracted from REF-52 cells after treatment described above and analyzed using the indicated antibodies. Quantification of protein level is shown relative to control level and normalized to β-tubulin. (C) Myofibroblasts were co-transfected with MRTF-A/B siRNA or NT siRNA, CA-MRTF-A or empty vector control, plus a SMαA promoter expressing luciferase and a Renilla luciferase control, then evaluated at 48 hours post-transfection for luciferase expression; n=3. Error bars represent mean ± AAD, ***p<0.001, **p<0.01, *p<0.05, or n.s.-not significant by Student’s t-test.

Figure 4. MRTFs contribute to myofibroblast morphology and contractile function

(A) TGF-β1-induced myofibroblasts were transfected with MRTF-A/B siRNAs or NT siRNA, and then stained with an anti-vinculin antibody at 72 hours posttransfection; Bar= 40 μm. (B) Focal adhesion quantification was performed for 10 cells per treatment. (C) Myofibroblasts were seeded on a deformable substrate (5 kPa stiffness), and transfected with MRTF-A/B siRNAs or NT siRNA for 72 hours. Bar= 40 μm. White arrows point to examples of wrinkles. (D) Quantification of percent cells wrinkling substrate after treatment with MRTF-A/B siRNA or NT siRNA. (E) Quantification of percent cells wrinkling substrate after transfection with FLAG-tagged-CA-MRTF-A, compared to empty vector, after 72 hours. Error bars represent mean ± SEM., ***p<0.001, **p<0.01, or *p<0.05 by Student’s t-test; n=3.

Figure 5. Proposed model of myofibroblast differentiation from fibroblasts in response to TGF-β1

The fibroblast when exposed to TGF-β1 increases assembly of F-actin inducing the MRTF-A/B to translocate into the nucleus, bind to SRF, and promote expression of a contractile gene program, consisting of CArG-containing genes such as SMαA, SMγA, SM22-α, h1-calponin, and vinculin. The expression of these genes causes changes to the cytoskeleton, resulting in a positive feed-back loop of MRTF activity and smooth muscle-specific cytoskeletal protein (SMCP) gene expression, and ultimately, the differentiation of the myofibroblast and the development of contractile function (figure modified from Tomasek and coworkers (Tomasek et al., 2002)).