Regulation of skeletal muscle regeneration by CCR2-activating chemokines is directly related to macrophage recruitment

Abstract

Muscle regeneration requires CC chemokine receptor 2 (CCR2) expression on bone marrow-derived cells; macrophages are a prominent CCR2-expressing cell in this process. CCR2-/- mice have severe impairments in angiogenesis, macrophage recruitment, and skeletal muscle regeneration following cardiotoxin (CTX)-induced injury. However, multiple chemokines activate CCR2, including monocyte chemotactic proteins (MCP)-1, -3, and -5. We hypothesized that MCP-1 is the chemokine ligand that mediates the impairments present in CCR2-/- mice. We examined muscle regeneration, capillary density, and cellular recruitment in MCP-1-/- and CCR2-/- mice following injury. Muscle regeneration and adipocyte accumulation, but not capillary density, were significantly impaired in MCP-1-/- compared with wild-type (WT) mice; however, muscle regeneration and adipocyte accumulation impairments were not as severe as observed in CCR2-/- mice. Although tissue levels of MCP-5 were elevated in MCP-1-/- mice compared with WT, the administration of MCP-5 neutralizing antibody did not alter muscle regeneration in MCP-1-/- mice. While neutrophil accumulation after injury was similar in all three mouse strains, macrophage recruitment was highest in WT mice, intermediate in MCP-1-/- mice, and severely impaired in CCR2-/- mice. In conclusion, while the absence of MCP-1 resulted in impaired macrophage recruitment and muscle regeneration, MCP-1-/- mice exhibit an intermediate phenotype compared with CCR2-/- mice. Intermediate macrophage recruitment in MCP-1-/- mice was associated with similar capillary density to WT, suggesting that fewer macrophages may be needed to restore angiogenesis vs. muscle regeneration. Finally, other chemokines, in addition to MCP-1 and MCP-5, may activate CCR2-dependent regenerative processes resulting in an intermediate phenotype in MCP-1-/- mice.

Fig. 1.

Inflammation, myofiber necrosis, and tissue regeneration in tibialis anterior (TA) muscle following cardiotoxin (CTX)-induced injury. Images were derived from TA muscle of wild-type (WT; A, D, G, J, and M), monocyte chemotactic protein-1−/− (MCP-1−/−; B, E, H, K, and N), or CC chemokine receptor 2−/− (CCR2−/−; C, F, I, L, and O) mice at the indicated times post-CTX injections. Thus, specimens were derived at 3 (A–C), 7 (D–F), 14 (G–I), 21 (J–L), or 28 (M–O) days after injury. *Necrotic muscle fibers. Arrows, mitotic figures. Paraffin sections (3–4 μm), hematoxylin and eosin stain. Scale bars = 40 μm (A–F), 100 μm (G–I), and 50 μm (J–O).

Fig. 2.

Impaired muscle regeneration, myofiber cross-sectional area, and fat area in MCP-1−/− and CCR2−/− mice. Measurements were performed in the TA muscle at baseline (no injury) and after CTX-induced injury, n = 8–15 mice/strain/time point. A: cross-sectional area (μm²) of muscle fibers in WT, MCP-1−/−, and CCR2−/− mice of mature myofibers at baseline and after injury (regenerated myofibers). B: fat area (%) at baseline was not detectable (ND) in all 3 mouse strains but increased after injury. Data are means ± SE. *Significant difference (P < 0.001) compared with baseline for each mouse strain; †significant difference (P ≤ 0.008) between WT and MCP-1−/− mice at corresponding time points; #significant difference (P ≤ 0.008) between WT and CCR2−/− mice at corresponding time points; §significant difference (P ≤ 0.02) between MCP-1−/− and CCR2−/− mice at corresponding time points.

Fig. 3.

Myofiber size distribution following CTX-induced injury in WT, MCP-1−/−, and CCR2−/− mice. The distribution of myofibers in 500- μm² increments was calculated from the same images used to assess cross-sectional area, n = 8–15 mice/strain/time point. Data are means ± SE in %. The regenerated myofiber distribution was significantly decreased at all postinjury time points for MCP-1−/− (P ≤ 0.004) and CCR2−/− (P < 0.001), and from days 7–21 for WT (P ≤ 0.001) mice compared with baseline. While baseline fiber distribution was similar in all 3 mouse strains, all postinjury time points were significantly different (P ≤ 0.004) between all 3 mouse strains.

Fig. 4.

Delayed increase in CCR2−/− with similar capillary density in MCP-1−/− and WT mice. Measurements were performed in the TA muscle at baseline (no injury) and after CTX-induced injury, n = 7–16 mice/strain/time point. Data were not available at the day 7 time point for MCP-1−/− and CCR2−/− mice due to the presence of extensive necrosis. Capillaries/muscle fiber (A) and capillaries per millimeter to the second power (B) are shown. Data are means ± SE. *Significant difference (P ≤ 0.008) compared with baseline for each mouse strain; †significant difference (P = 0.05) between WT and MCP-1−/− mice at corresponding time points; #significant difference (P < 0.001) between WT and CCR2−/− mice at corresponding time points; §significant difference (P < 0.001) between MCP-1−/− and CCR2−/− mice at corresponding time points.

Fig. 5.

Elevated MCP-5 in injured muscle of MCP-1−/− mice but similar muscle regeneration with MCP-5 blocking antibody. A: measurement of tissue MCP-5 levels in the anterior compartment of WT and MCP-1−/− mice (5–9 mice/strain/time point) at baseline and after CTX-induced injury. *Significant difference compared with baseline (P ≤ 0.02) for each mouse strain; †significant difference (P = 0.03) between WT and MCP-1−/− mice at corresponding time points. Histomorphometric measurements performed in the TA muscle of MCP-1−/− mice after CTX-induced injury and while receiving MCP-5 neutralizing antibody, control antibody, or no antibody, n = 6–21 mice/strain/antibody treatment/time point. B: cross-sectional area (μm²) of regenerating myofibers. C: fat area (%) after injury. Data are means ± SE.

Fig. 6.

Flow cytometry analysis of cell populations present in baseline and injured muscles of WT, MCP-1−/−, and CCR2−/− mice. Measurements of absolute cell numbers performed in the TA muscle at baseline (no injury) and after CTX-induced injury, n = 4–11 mice/strain/time point. Total cells isolated (A), CD45+ immune cells (B), neutrophils (CD45+/CD11b+/Ly6G+ cells) (C), macrophages (CD45+/CD11b+/Ly6G- cells) (D), and CD45+/Sca-1+ cells (E). Data are means ± SE. *Significant difference compared with baseline (P ≤ 0.05) for each mouse strain, †significant difference (P ≤ 0.03) between WT and MCP-1−/− mice at corresponding time points, #significant difference (P ≤ 0.002) between WT and CCR2−/− mice at corresponding time points, §significant difference (P ≤ 0.04) between MCP-1−/− and CCR2−/− mice at corresponding time points.

Fig. 7.

Mac3+ cells in TA muscle following CTX-induced injury. Immunolocalization of the monocyte/macrophage marker (Mac3) in TA muscle from WT (A, D, G, J, and M), MCP-1−/− (B, E, H, K, and N), or CCR2−/− (C, F, I, L, and O) mice at 3, 7, 14, or 21 days post-CTX injection. *Necrotic muscle fibers. Mac3+ cells in formalin-fixed, paraffin-embedded tissue were identified by the red-brown reaction product of horseradish peroxidase and diaminobenzidine+H₂O₂. Control sections treated in parallel with an isotype-controlled monoclonal rat antibody provided no signal (hematoxylin counterstain). Scale bars = 40 μm.