Matrix stiffness and architecture drive fibro-adipogenic progenitors' activation into myofibroblasts

Abstract

Fibro-adipogenic progenitors (FAPs) are essential in supporting regeneration in skeletal muscle, but in muscle pathologies FAPs the are main source of excess extracellular matrix (ECM) resulting in fibrosis. Fibrotic ECM has altered mechanical and architectural properties, but the feedback onto FAPs of stiffness or ECM properties is largely unknown. In this study, FAPs' sensitivity to their ECM substrate was assessed using collagen coated polyacrylamide to control substrate stiffness and collagen hydrogels to engineer concentration, crosslinking, fibril size, and alignment. FAPs on substrates of fibrotic stiffnesses had increased myofibroblast activation, depicted by αSMA expression, compared to substrates mimicking healthy muscle, which correlated strongly YAP nuclear localization. Surprisingly, fibrosis associated collagen crosslinking and larger fibril size inhibited myofibroblast activation, which was independent of YAP localization. Additionally, collagen crosslinking and larger fibril diameters were associated with decreased remodeling of the collagenous substrate as measured by second harmonic generation imaging. Inhibition of YAP activity through verteporfin reduced myofibroblast activation on stiff substrates but not substrates with altered architecture. This study is the first to demonstrate that fibrotic muscle stiffness can elicit FAP activation to myofibroblasts through YAP signaling. However, fibrotic collagen architecture actually inhibits myofibroblast activation through a YAP independent mechanism. These data expand knowledge of FAPs sensitivity to ECM and illuminate targets to block FAP's from driving progression of muscle fibrosis.

Figure 1

FAPs’ response to substrate stiffness. (A) Expression of perilipin and αSMA in FAPs after 7 days in adipogenic or FAP media. (B) YAP immunofluorescence of FAPs on polyacrylamide gels on 8 kPa and 25 kPa. Yellow outlines indicate nuclei and red outlines indicate cytoplasm on insets. (C) Quantification of YAP signal intensity in nucleus and cytoplasm. (D, E) EdU signal and quantification of percent proliferating cells after 24 h EdU treatment (F) αSMA signaling as a measure of myofibroblast activation in FAPs. (G) Quantification of αSMA and actin area ratio. (H) Pearson correlation of αSMA expression and YAP nuclear localization fitted with linear regression line. *P < 0.05, **P < 0.01, compared to 8 kPa via one-way ANOVA using repeated measures with Dunnett correction. Scales bars set to 50 μm. Scale bars for insets set to 20 μm. N = 3 mice and n = 3 independent gels.

Figure 2

FAP behavior in response to scaling collagen concentration. (A) YAP immunofluorescence of FAPs on 1.5 and 6.0 mg/ml collagen gels and tissue cultured plastic. Yellow outlines indicate nuclei and red outlines indicate cytoplasm on insets. (B) Quantification of YAP signal intensity as a nuclear to cytoplasmic ratio. (C, D) EdU signal and quantification across collagen concentrations and plastic. (E) αSMA signaling in FAPs after 7 days in FAP media. (F) Quantification of αSMA and actin area ratio as a measure of myofibroblast activation. *P < 0.05, **P < 0.01, ***P ≤ 0.001, ****P < 0.0001, One-way ANOVA with multiple comparisons with Dunnett correction. Data normalized to 3.0 mg/ml. Scales bars set to 50 μm. Scale bars for insets set to 20 μm. N = 3 and n = 3 for 1.5, 4.5, 6.0 mg/ml and plastic. N = 4 and n = 4 for 3.0 mg/ml. N = number of mice, n = number of independent gels.

Figure 3

FAPs’ response to 3 mg/ml crosslinked and non-crosslinked collagen gels. (A) YAP immunofluorescence of FAPs on telocollagen and atelocollagen gels. Yellow outlines indicate nuclei and red outlines indicate cytoplasm on insets. (B) Quantification of YAP signal intensity to find nuclear to cytoplasmic ratio of YAP. (C, D) EdU signal and quantification of percent proliferating cells after 24 h EdU treatment (E) αSMA signaling as a measure of myofibroblast activation in FAPs. (F) Quantification of αSMA and actin area ratio. ***P < 0.001. Data normalized to telecollagen gels. Scales bars set to 50 µm. Scale bars for insets set to 20 µm. N = 3 mice and n = 3 independent gels.

Figure 4

SHG imaging of 3 mg/ml crosslinked and non-crosslinked collagen gels. (A) SHG images of collagen and FAP cells. Red outlines indicate a αSMA− FAP and green outlines indicate a αSMA+ FAP. Scale bars on no cell images set to 10 μm. Scales bars on day 2 and day 7 images set to 50 µm. (B) αSMA and actin area ratio after 2 and 7 days on the collagen gels. (C) Collagen intensity of gels under αSMA+ or αSMA− cells, or under no cells normalized to areas of collagen without any cells. *P < 0.05 using a two-way ANOVA with multiple comparisons using Tukey correction. N = 2 independent gels and n = 3 fields per gel.

Figure 5

Effect of fibril size on FAP behavior. (A) YAP immunofluorescence of FAPs on 3 mg/ml telocollagen gels polymerized at 37 and 22 °C. Yellow outlines indicate nuclei and red outlines indicate cytoplasm on insets. (B) Quantification of YAP signal intensity to find nuclear to cytoplasmic ratio of YAP. (C, D) EdU signal and quantification of percent proliferating cells after 24 h EdU treatment. (E) αSMA signaling in FAPs after 7 days. Scales bars set to 50 μm. (F) Quantification of αSMA and actin area ratio. **P < 0.01. Data normalized to 37 °C. Scales bars set to 50 µm; insets set to 20 µm. N = 1 mouse and n = 3 independent gels.

Figure 6

SHG imaging of 3 mg/ml collagen gels polymerized at different temperatures. (A) SHG images of collagen gels polymerized at 37 °C and 22 °C at day 2 and day 7 after cell seeding. Scale bars on no cell images set to 10 μm. Scales bars on day 2 and day 7 images set to 50 μm. Red outlines indicate a αSMA− FAP and green outlines indicate a αSMA+ FAP. (B) Myofibroblast activation at day 2 and day 7 across the gels. (C) Collagen intensity under no cells, αSMA+, or αSMA− FAPs. Intensity normalized to areas of gels without cells. *P < 0.05 using a two-way ANOVA with multiple comparisons using Tukey correction. N = 2 independent gels and n = 3 fields per gel.

Figure 7

Yap inhibition reduces myofibroblast activation. (A) YAP localization in FAPs cultured on plastic in DMSO or 0.5 µM verteporfin for 2 days. Scale bars are 50 µm, insets are 20 µm. Yellow outlines indicate nuclei and red outlines indicate cytoplasm on insets. (B) Myofibroblast activation in FAPs cultured for 7 days with 0.5 µM verteporfin or 0.1% DMSO on collagen-coated polyacrylamide gels at 25 kPa and 3.0 mg/ml telocollagen gels polymerized at 37 °C. Scale bars set to 50 µm. (C) Quantification of myofibroblast activation of FAPs in DMSO or 0.5 µM verteporfin on collagen-coated polyacrylamide gels. (D) Quantification of myofibroblast activation of FAPs in DMSO or 0.5 µM verteporfin on collagen gels polymerized at different temperatures. *P < 0.05, ****P < 0.0001 between DMSO and verteporfin treatment. #P < 0.05, ####P < 0.0001 between substrates.