Beta3-integrin mediates satellite cell differentiation in regenerating mouse muscle

Abstract

Skeletal muscle satellite cells can sense various forms of environmental cues and initiate coordinated signaling that activates myogenesis. Although this process involves cellular membrane receptor integrins, the role of integrins in myogenesis is not well defined. Here, we report a regulatory role of β3-integrin, which was previously thought not expressed in muscle, in the initiation of satellite cell differentiation. Undetected in normal muscle, β3-integrin expression in mouse hindlimb muscles is induced dramatically from 1 to 3 d after injury by cardiotoxin. The source of β3-integrin expression is found to be activated satellite cells. Proliferating C2C12 myoblasts also express β3-integrin, which is further up-regulated transiently on differentiation. Knockdown of β3-integrin expression attenuates Rac1 activity, impairs myogenic gene expression, and disrupts focal adhesion formation and actin organization, resulting in impaired myoblast migration and myotube formation. Conversely, overexpression of constitutively active Rac1 rescues myotube formation. In addition, a β3-integrin-neutralizing antibody similarly blocked myotube formation. Comparing with wild-type littermates, myogenic gene expression and muscle regeneration in cardiotoxin-injured β3-integrin-null mice are impaired, as indicated by depressed expression of myogenic markers and morphological disparities. Thus, β3-integrin is a mediator of satellite cell differentiation in regenerating muscle.

Figure 1.

Expression of β3-integrin is induced in activated satellite cells in regenerating muscle. A) β3-Integrin expression is induced in regenerating mouse soleus. Mouse soleus was injured by direct injection of 100 μl of 10 μM cardiotoxin. At indicated times, soleus was excised from euthanized mice. Muscle homogenates were analyzed for β3-integrin and myogenin expression by Western blot analysis. Sample loading quality was monitored by probing p38 MAPK that was relatively constant in cardiotoxin-injured muscle. B) Quiescent satellite cells do not express β3-integrin. Frozen sections of uninjured soleus were subjected to immunofluorescence staining of β3-integrin (β3-ITG). Costaining of Pax7 was performed to identify satellite cells. Nuclei were stained with DAPI. C) β3-Integrin expression by regenerating soleus takes place in activated satellite cells. Frozen sections of soleus injured by cardiotoxin for 3 d were subjected to immunofluorescence staining of β3-integrin. Costaining of MyoD was performed to identify activated satellite cells. Scale bar = 50 μm. D) β3-Integrin expression by proliferating C2C12 myoblasts is up-regulated during differentiation. C2C12 myoblasts were cultured in GM and then switched to DM. Cells were harvested at indicated times, and cell lysates were analyzed for β3-integrin and myogenin expression by Western blot analysis. Sample loading quality was monitored by probing α-tubulin. Densitometry analysis was performed to quantify detected bands. *P < 0.05; ANOVA.

Figure 2.

Myogenic gene expression in C2C12 myoblasts is dependent on β3-integrin. A) Expression of β3-integrin was knocked down by β3-integrin siRNA in C2C12 myoblasts. C2C12 myoblasts were transfected with control or β3-integrin-specific siRNA. After 48 h, β3-integrin in cell lysates was determined by Western blot analysis. B) Knockdown of β3-integrin impairs myogenic gene expression in C2C12 myoblasts. C2C12 myoblasts transfected with control or β3-integrin (β3-ITG) siRNA were plated at low cell density and incubated in DM for 72 h. Immunofluorescence staining of myogenin or MHC was performed to evaluate myogenic gene expression (green). Nuclei were stained with DAPI (blue). Myogenin- or MHC-expressing cells were counted in random fields and presented as a percentage of total cells in a field. Scale bar = 200 μm. *P < 0.05, **P < 0.01; Student's t test.

Figure 3.

Focal adhesions, F-actin organization, myoblast migration, and myotube formation are dependent on β3-integrin. A) Normal focal adhesions and F-actin organization are dependent on β3-integrin. C2C12 myoblasts transfected with control or β3-integrin (β3-ITG)-specific siRNA were cultured in GM or switched to DM for 24 h. Focal adhesion was evaluated by immunofluorescence staining of phosphorylated tyrosine (p-Tyrosine, red) and F-actin (phalloidin, green). B) β3-Integrin deficiency blocks differentiation-induced increase in focal adhesion cluster. Focal adhesion cluster area per cell was measured by using ImageJ software. *P < 0.05; Student's t test. C) Myoblast migration is dependent on β3-integrin. C2C12 myoblasts transfected with control or β3-integrin-specific siRNA were plated on top of the transwells (2.5×10⁵ cells/well). After incubation for 5 h in DM at 37°C, myoblasts that migrated to the lower face of membrane were stained with crystal violet and counted. *P < 0.001; Student's t test. D) Myotube formation is dependent on β3-integrin. C2C12 myoblasts transfected with control or β3-integrin-specific siRNA were induced to differentiate by incubation in DM for 96 h. Immunofluorescence staining was performed with an antibody against MHC (MF-20, green). Nuclei were stained with DAPI (blue). Myotube formation was quantified by using the fusion index analysis, which was derived by counting the number of myotubes having ≥5 nuclei against the total number of nuclei in the field; ≥ 2000 nuclei/field were counted from randomly selected fields. Scale bars = 100 μm (A); 200 μm (D). *P < 0.01 vs. control siRNA-transfected cells; Student's t test.

Figure 4.

β3-Integrin-mediated signaling is critical to myotube formation. A mouse β3-integrin (β3-ITG)-neutralizing antibody or preimmune IgG (control) was included at 1 or 3 μg/ml in DM during differentiation. After 72 h incubation in DM, myotube formation was analyzed by immunofluorescence staining of MHC. Fusion index was calculated by counting the number of myotubes having ≥3 nuclei against the total number of nuclei in the field. *P < 0.05, **P < 0.01 vs. control antibody; ANOVA.

Figure 5.

Rac1 GTPase mediates β3-integrin effect on myoblast differentiation. A) Rac1 GTPase activity in differentiating myoblasts is dependent on β3-integrin. C2C12 myoblasts were transfected with control or β3-integrin (β3-ITG)-specific siRNA. After incubation in DM for indicated times, cells were harvested. Rac1-GTP (active form of Rac1) was pulled down from cell lysate using PAK1 PBD agarose beads (Cell Biolabs). Pulled-down Rac1-GTP and total Rac1 in cell lysate were analyzed by Western blot analysis. Densitometry data were analyzed by ANOVA. B) Overexpression of constitutively active Rac1 (Rac1 L61) in C2C12 myoblasts. At 24 h after β-3-integrin-specific siRNA transfection, C2C12 myoblasts were transduced with adenovirus encoding Rac1 L61 or GFP. After 24 h, Rac1 L61 expression was verified by Western blot analysis of Rac1-GTP and total Rac1. C) Active Rac1 rescues myotube formation by β3-integrin-deficient myoblasts. C2C12 myoblasts that had been transfected with β3-integrin-specific siRNA and transduced with adenovirus encoding Rac1 L61 or GFP were switched to DM and differentiated for 96 h. Immunofluorescence staining of MHC was performed to assess myotube formation. Fusion index was calculated as described in Fig. 3. Scale bar = 200 μm. *P < 0.05.

Figure 6.

Myogenesis and muscle regeneration are impaired in β3-integrin^−/− mice. A) Initiation of myogenesis is impaired in β3-integrin-null soleus. Soleus was harvested from β3-integrin^−/− (β3-ITG^−/−) mice and wild-type (WT) littermates before (d 0) and after injury by cardiotoxin injection (d 3). Muscle homogenates were analyzed by Western blot. *P < 0.05 vs. WT d 3; Student's t test. B) Myogenic differentiation is impaired in β3-integrin-null soleus. Frozen sections prepared from mouse soleus injured by cardiotoxin for 3 d were analyzed by immunofluorescence staining of eMHC. Nuclei were stained with DAPI. C) Muscle regeneration is impaired in β3-integrin-null soleus. Paraffin-embedded muscle sections prepared from mouse soleus injured by cardiotoxin for 5 d were subjected to hematoxylin and eosin staining. Scale bars = 100 μm.