Periostin Is Required for the Maintenance of Muscle Fibers during Muscle Regeneration

Abstract

Skeletal muscle regeneration is a well-organized process that requires remodeling of the extracellular matrix (ECM). In this study, we revealed the protective role of periostin, a matricellular protein that binds to several ECM proteins during muscle regeneration. In intact muscle, periostin was localized at the neuromuscular junction, muscle spindle, and myotendinous junction, which are connection sites between muscle fibers and nerves or tendons. During muscle regeneration, periostin exhibited robustly increased expression and localization at the interstitial space. Periostin-null mice showed decreased muscle weight due to the loss of muscle fibers during repeated muscle regeneration. Cultured muscle progenitor cells from periostin-null mice showed no deficiencies in their proliferation, differentiation, and the expression of Pax7, MyoD, and myogenin, suggesting that the loss of muscle fibers in periostin-null mice was not due to the impaired function of muscle stem/progenitor cells. Periostin-null mice displayed a decreased number of CD31-positive blood vessels during muscle regeneration, suggesting that the decreased nutritional supply from blood vessels was the cause of muscle fiber loss in periostin-null mice. These results highlight the novel role of periostin in maintaining muscle mass during muscle regeneration.

Keywords: extracellular matrix; periostin; skeletal muscle regeneration.

Figure 1

Periostin is localized at the neuromuscular junction, muscle spindle, and myotendinous junction in skeletal muscle. (a) Representative image of periostin localization (green) with α-BTX (red) using transverse section. Bar: 16 μm (left). (b) Representative image of periostin localization (yellow) with α-BTX (red) and neurofilament H (NF-H) positive-motor neuron (green) using longitudinal section (right). Bar: 25 μm. (c) Representative image of periostin localization (red) at the muscle spindle. Bag fibers and nuclei were stained by anti-S46 (green) and TOTO-3 (blue), respectively. Bar: 20 μm. (d) Upper: Representative image of periostin localization (red) at the myotendinous junction (MTJ). The serial section was stained with anti-periostin antibody (left) and H&E (right). The representative periostin-expressing MTJs were indicated by arrows. Lower: Longitudinal section was stained with anti-periostin (green) and MF-20-positive muscle fibers (red). Bar: 50 μm. (e) The locomotive function was analyzed by foot paint analysis. (f) Maximum running speed was analyzed using a treadmill. Mouse tibialis anterior muscles were used for histological analysis. n = 4. Statistical analysis was performed using Student’s t-test. Error bars indicate SEM.

Figure 2

Expression of periostin is increased during muscle regeneration. (a) Expression of periostin, Tgfβ1, PDGFRα, Acvr1, Bmp2, and Bmp4 during muscle regeneration by quantitative PCR. (b) Expression of periostin isoforms during muscle regeneration. (c) Immunohistochemical analysis of periostin during muscle regeneration. Bar: 75 μm. n = 3. Error bars indicate SEM.

Figure 3

Periostin is expressed in the interstitial space during muscle regeneration. (a) Immunostaining of periostin (green) with MyHC (MF20, red) 5 days after the induction of muscle regeneration. Bar: 75 μm. (b) Immunostaining of periostin (green) and PDGFRα (red) 5 days after the induction of muscle regeneration. Bar: 75 μm. (c) Immunohistochemical images of periostin (green) with CD31 (red) 7 days after the induction of muscle regeneration. Arrows indicated the CD31 signals surrounded by periostin. Bar: 75 μm. n = 3. Error bars indicate SEM.

Figure 4

Loss of periostin causes a decrease in muscle fiber number during repeated muscle regeneration. (a) Muscle weight 7 days after the induction of muscle regeneration. n = 4–10. (b) Muscle weight 7 days after repeated induction of muscle regeneration. n = 14. (c) CSA of regenerating muscle fiber after repeated muscle regeneration. n = 6. (d) The average of CSA. n = 6. (e) The number of muscle fibers after repeated muscle regeneration. n = 6. (f) Immunostaining of CD31 after repeated muscle regeneration. Bar: 75 μm. (g) The number of CD31-positive blood vessels. n = 3. (h) Muscle weight of 12-week-old mdx and mdx/periostin-null mice. n = 3–6. (i) Representative H&E staining of mdx and mdx/periostin-null mice. (j) Plasma CK levels in mdx and mdx/periostin-null mice. n = 3–6. * p < 0.05 and *** p < 0.001 by Student’s t-test. Error bars indicate SEM.

Figure 5

Loss of periostin delayed muscle regeneration in mice with a DBA/2 background. (a) Muscle weight 7 days after the induction of muscle regeneration. n = 7–8. (b) Number of embryonic MyHC-positive fibers 14 days after the induction of muscle regeneration. n = 4. * p < 0.05 by Student’s t-test. Error bars indicate SEM.

Figure 6

Proliferation and differentiation of muscle progenitor cells are not impaired in periostin-null mice. (a) Time-course changes in the number of cultured muscle progenitor cells. n = 4. (b) Immunocytochemical analysis of Pax7 and MyoD in muscle progenitor cells from periostin-null mice. Bar: 200 μm. (c) Ratio of Pax7+/MyoD-, Pax7-/MyoD+, and Pax7+/MyoD+ cells. n = 4. (d) Immunocytochemical analysis of myogenin in muscle progenitor cells from periostin-null mice. Bar: 200 μm. (e) Quantitative analysis of myogenin-positive cells. n = 4. (f) Immunocytochemical analysis of MF-20 in differentiated muscle progenitor cells from periostin-null mice. Bar: 200 μm. (g) Quantitative analysis for the fusion index. n = 5. Error bars indicate SEM.