Corticosteroid inhibits differentiation of palmar fibromatosis-derived stem cells (FSCs) through downregulation of transforming growth factor-β1 (TGF-β1)

Abstract

Treatment for musculoskeletal fibromatosis remains challenging. Surgical excision for fibromatosis is the standard therapy but recurrence remains high. Corticosteroids, an anti-fibrogenic compound, have been used to treat early stage palmar fibromatosis, but the mechanism is unknown. We investigated the inhibitory mechanism effect of corticosteroids in the murine model of fibromatosis nodule as well as in cultured FSCs. Quantitative reverse transcription/polymerase chain reaction (PCR) analysis and immunofluorescence (IF) staining for markers of myofibroblasts (α-smooth muscle actin and type III collagen) were used to examine the effect of dexamethasone on myofibroblasic differentiation of FSCs both in vitro and in vivo. Transforming growth factor-β1 (TGF-β1) signaling and its downstream targets were examined using western blot analysis. TGF-β1 expression in FSCs before and after dexamethasone treatment was compared. In addition, inhibition of TGF-β1 expression was examined using RNA interference (RNAi) on FSCs, both in vitro and in vivo. Treating FSCs with dexamethasone inhibited FSCs' myofibroblastic differentiation in vitro. Treating FSCs with dexamethasone before or after implantation further inhibited formation of fibromatosis nodules. Dexamethasone suppressed expression of TGF-β1 and pSmad2/3 by FSCs in vitro. TGF-β1 knockdown FSCs showed reducing myofibroblastic differentiation both in vitro and in vivo. Finally, addition of TGF-β1 abolished dexamethasone-mediated inhibition of myofibroblastic differentiation. Dexamethasone inhibits the myofibroblastic differentiated potential of FSCs both in vitro and in vivo through inhibition of TGF-β1 expression in FSCs. TGF-β1 plays a key role in myofibroblastic differentiation.

Fig 1. Dexamethasone inhibits in vitro myofibroblastic differentiation of FSCs.

(A) FSCs seeded at 2000 cells/well in 96-well plates were treated with dexamethasone at the indicated concentrations for 7 days. The IC50-value was measured. (B) mRNA expression of α-SMA, Col3A1 and Col1A3 was analyzed by quantitative RT-PCR after treatment with dexamethasone 0, 0.2 and 2 uM for 14 days. (C) Immunofluorescence staining for α-SMA (red), types III (red) and I collagen (green) after treated with either dexamethasone 2uM or 0 uM (control) for 14 days. Bars = 50 μm. (D) The percentages of stained areas. (E) The percentages of myofibroblasts. Data are shown as mean ± SD (n = 3). Statistical significance is presented as **, p<0.01 compared with other groups. All experiments were repeated with FSCs isolated from three different donors.

Fig 2. Dexamethasone inhibited FSCs formation of fibromatosis nodule in murine model.

(A–C) FSCs treated with 200 nM dexamethasone (Dex) for 3 days were then delivered with Matrigel, followed by transplantation beneath the dorsal skin of nude mice. (A) Macroscopic views of the transplants after 14 days in vivo. Scale = 1 mm. H&E staining and immunofluorescence staining for α-SMA, types III and type I collagen were performed. Bars = 50 μm. (B) The percentages of stained areas. (C) The percentages of myofibroblasts. (D–F) FSCs were delivered in Matrigel and transplanted under beneath the dorsal skin of nude mice. After 7 days, dexamethasone (2 mg/kg/day) dissolved in saline was injected subcutaneously daily for 1 week, and the control group received daily subcutaneous injections of 40 ml of saline alone for 1 week (D) Macroscopic views of the implants at 14 days of implantation in vivo. Scale = 1 mm. H&E staining and immunofluorescence staining for α-SMA, type III and type I collagen. Bars = 50 μm. (E) The percentages of stained areas. (F) The percentages of myofibroblasts. Data are shown as mean ± SD (n = 3). **, p<0.01 denotes statistical significance. All experiments were repeated with FSCs isolated from three different donors.

Fig 3. Inhibition of TGF-β1 signaling, Smad family and down-regulation of TGF-β1 in in vitro dexamethasone-treated FSCs.

FSCs were treated with 0 μM, 0.2 μM, and 2 μM dexamethasone (Dex) for 3 days, followed by (A) and (B) Western blotting analysis of Smad family and Sp1. Immunoblotting of ß-actin & Smad2/3 was performed to show equal protein loading. (C) TGF-β1 protein levels, as measured by enzyme-linked immunosorbent assay (ELISA), in the conditioned media (D) quantitative RT-PCR analysis for mRNA expression of TGF-β1, where GAPDH was used as normalization control. (E) TGF-β1 protein levels between FSCs and BMSCs, as measured by enzyme-linked immunosorbent assay (ELISA), in the conditioned media. Data are shown as mean ± SD (n = 3). **, p<0.01 denotes statistical significance. All experiments were repeated with FSCs isolated from three different donors.

Fig 4. Inhibition of fibromatosis nodule formation through TGF-β1 knockdown of FSCs.

Validation of knockdown efficiency by (A) quantitative RT-PCR analysis and (B) Western blotting for mRNA and protein expression of TGF-β1, respectively after transfection with a lentiviral vector carrying RNAi targeting TGF-β1 or non-targeting RNAi (CTR) for 2 days. (C–E) Transfection with a lentiviral vector carrying RNAi targeting TGF-β1 gene or non-targeting RNAi (CTR) for 14 days. (C) Immunofluorescence staining for α-SMA, type III and I collagen. FSCs were cultured for 14 days. Bars = 50 um. (D) The percentages of stained areas. (E) The percentages of myofibroblasts. (F–H) FSCs were transfected with a lentiviral vector carrying RNAi targeting TGF-β1 and non-targeting RNAi (CTR) were delivered with Matrigel and implanted beneath the dorsal skin of nude mice. Macroscopic views of the implants at 14 days of implantation in vivo. Scale = 1 mm. H&E staining and immunofluorescence staining for α-SMA, type III and type I collagen. Bars = 50 um. (G) The percentages of stained areas. (H) The percentage of myofibroblasts. Data are shown as mean ± SD (n = 3). **, p<0.01 denotes statistical significance. All experiments were repeated with FSCs isolated from three different donors.

Fig 5. Treatment of TGF-β1 abolished TGF-β1 knockdown-mediated inhibition of myofibroblastic differentiation.

(A) Quantitative RT-PCR analysis for mRNA expression of α-SMA, type III and type I collagen genes. FSCs transfected with a lentiviral vector carrying RNAi targeting TGF-β1 or non-targeting RNAi (CTR) were cultured with either 10 ng/ml TGF-β1 or saline for 14 days. (B) Immunofluorescence staining for α-SMA, type III and I collagen. FSCs transfected with a lentiviral vector carrying RNAi targeting TGF-β1 or non-targeting RNAi (CTR) were cultured with either 10 ng/ml TGF-β1 or saline for 14 days. Bars = 50 um. (C) The percentages of stained areas. (D) The percentages of myofibroblasts. Data are shown as mean ± SD (n = 3). **, p<0.01 denotes statistical significance. All experiments were repeated with FSCs isolated from three different donors.