c-MET regulates myoblast motility and myocyte fusion during adult skeletal muscle regeneration

Abstract

Adult muscle stem cells, satellite cells (SCs), endow skeletal muscle with tremendous regenerative capacity. Upon injury, SCs activate, proliferate, and migrate as myoblasts to the injury site where they become myocytes that fuse to form new muscle. How migration is regulated, though, remains largely unknown. Additionally, how migration and fusion, which both require dynamic rearrangement of the cytoskeleton, might be related is not well understood. c-MET, a receptor tyrosine kinase, is required for myogenic precursor cell migration into the limb for muscle development during embryogenesis. Using a genetic system to eliminate c-MET function specifically in adult mouse SCs, we found that c-MET was required for muscle regeneration in response to acute muscle injury. c-MET mutant myoblasts were defective in lamellipodia formation, had shorter ranges of migration, and migrated slower compared to control myoblasts. Surprisingly, c-MET was also required for efficient myocyte fusion, implicating c-MET in dual functions of regulating myoblast migration and myocyte fusion.

Figure 1. c-MET is required for SC mediated muscle regeneration.

A-C) Control and mutant myoblasts were isolated by FACS via YFP fluorescence (cells were sorted only for data in A-C). A) RT-PCR analysis shows deletion of the 80bp exon 16 in mutant cells. B) Western analysis of total c-MET protein shows presence of both the preprocessed and processed forms of c-MET protein in control cells, while mutant cells lack the processed c-MET protein. C) Western analysis with a c-MET phospho Tyr 1234/1235 specific antibody shows c-MET activation is abolished in mutant cells. D) Hematoxylin and eosin staining of 10 dpi muscle sections shows robust muscle regeneration, indicated by cells with centrally localized nuclei, in control tissue, while mutant muscle shows fewer and smaller regenerated fibers (arrows, newly regenerated fibers; black line, boundary between injured and uninjured tissue; scale bar = 50 μm). E) Control and mutant muscle tissue showing IF against sarcomeric MHC in uninjured, 10, and 20 dpi muscle sections (scale bar = 50 μm). F) Quantification of regenerated fiber number per 0.38 mm² area in 20 dpi control and mutant muscle sections (p = 3.36E-6 t test; N = 3 mice; 3 fields per mouse; error bars = SEM). G) Quantification of regenerated fiber diameter in 20 dpi control and mutant muscle sections (p = .0005 t test; N = 3 mice; 3 fields per mouse; error bars = SEM).

Figure 2. Increased density of TCF4+ fibroblasts and fibrosis during regeneration when SCs lack c-MET.

A) Control and mutant muscle tissue showing IF against TCF4 and Laminin in 10 dpi muscle sections (arrows, TCF4+ DAPI+ nuclei external to Laminin enclosed fibers; inset shows enlarged regions; scale bar = 50 μm). B) Quantification of TCF4+ DAPI+ nuclei external to Laminin per 0.15 mm² 10 dpi area (p = 1.19E-6 t test; N = 3 mice; 3 fields per mouse; error bars = SEM). C) Trichrome and hematoxylin staining of 20 dpi muscle sections shows fibrotic tissue (blue stain) throughout the injured area in mutant muscle (arrowheads, newly regenerated fibers; scale bar = 50 μm).

Figure 3. c-MET is not required for SC maintenance or differentiation.

A) X-gal histochemistry applied to uninjured TA muscle from TMX injected control and mutant mice (arrows, X-gal positive cells; scale bar = 50 μm). B - E) Cultured single fibers from the EDL of control and mutant TMX injected mice. Fibers were cultured for B) 24 h and probed by IF for MYOD and β-GAL (arrows, MYOD+ β-GAL+ cell) or D) 72 h and probed by IF for MYOG and β-GAL (arrows, MYOG+ β-GAL+ cell; scale bar = 20 μm). Quantification of C) MYOD+ β-GAL+ per total β-GAL+ cells (p = .46 t test; N = 3 mice, n = 50 cells per mouse; error bars = SEM) and E) MYOG+ β-GAL+ out of total β-GAL+ cells (p = .80 t test; N = 3 mice; n = 100 cells per mouse; error bars = SEM). F) IF for β-GAL and MHC applied to 10 dpi control and mutant TA muscle sections (arrows, MHC+ β-GAL+ cells; scale bar = 50 μm).

Figure 4. c-MET does not affect myoblast proliferation.

A) Representative image of β-GAL+ cells in clonal clusters on EDL single fibers cultured for 72 h (scale bar = 20 μm). B) Histogram showing the number of β-GAL+ cells per cluster on control and mutant EDL single fibers cultured for 72 h (N = 3 mice; n = 50 cells per mouse). C) Representative images of 5 dpi TA muscle from control and mutant mice, probed by IF for β-GAL and EdU and stained with DAPI (arrows, β-GAL+ EdU+ cells; scale bar = 50 μm). D) Quantification of β-GAL+ EdU+/β-GAL+ DAPI+ cells in 5 dpi TA muscle sections (p = .36 t test; N = 3 mice; n= 866 for control, 333 for mutant; error bars = SEM). E) Quantification of β-GAL+ DAPI+ cells per 0.15mm² injured TA muscle area (p = .02 t test; N = 3 mice; n = 4 injured muscle fields per mouse; error bars = SEM). F) Quantification of DAPI+ cells per 0.15mm² injured TA muscle area (p = .18 t test; N = 3 mice; n = 4 injured muscle fields per mouse; error bars = SEM).

Figure 5. c-MET plays a role in cell morphology, distance traveled, and velocity during myoblast migration.

A) Sequence of Hoffman modulation contrast (HMC) images showing live myoblasts in migration media. Images of the same cells were captured every 4 minutes for 68 minutes (arrows, lamellipodia in the control cell; arrowheads, membrane protrusions in the mutant cell; scale bar = 20 μm. B) Traces showing migratory paths of 15 myoblasts. Cell positions recorded every 4 minutes for 6 h (Axes = ± 300 μm). C) Average velocities of myoblasts in migration media without (no GF) or with supplement of HGF at 4 ng/ml. Images were taken once every 4 minutes, velocity measurements were recorded over 6 h. (p < .02 t test; N = 3 mice; n = 10 cells per mouse; error bars = SEM).

Figure 6. c-MET contributes to lamellipodia formation and peri-nuclear INTβ1 localization.

A) Representative images of cells stained with phalloidin and DAPI with IF for YFP (arrow, lamellipodia; arrowhead, long actin protrusion; scale bar = 5 μm). B) Representative images of cells with IF for INTβ1 and YFP (arrow, peri-nuclear INTβ1 localization; scale same as in (A)). C) Average percentage of YFP+ cells with phalloidin stained lamellipodia as example shown in (A)(p = .00029 t test; N = 3; n = 120 cells; error bars = SEM). D) Average percentage of YFP+ cells with peri-nuclear INTβ1 as example shown in (B) (p = .013 t test; N = 3; n = 120 cells; error bars = SEM).

Figure 7. Myoblasts require c-MET for efficient fusion.

A) HMC images showing the same fields of low-density, live, myoblasts in differentiation media at T0, 24, and 48 h. (scale bar = 100 μm). B) Representative images of high-density R26R ^YFP control and mutant myoblasts after 3 days in differentiation media. Cells were probed by IF for MHC and YFP and stained with DAPI (scale bar = 100 μm). C) Differentiation index (MHC+ YFP+ nuclei per YFP+ nuclei) and D) Index of unfused cells (MHC+ YFP+ cells with single nucleus per total MHC+ YFP+ nuclei) was quantified in 3-day differentiation cultures. Two independent cell cultures were tested for each condition. Low-density cells were plated at 13,000 cells per cm²; high-density cells were plated at 40,000 cells per cm². At least 110 MHC+ YFP+ cells were counted for each culture and condition.