Myotubes differentiate optimally on substrates with tissue-like stiffness: pathological implications for soft or stiff microenvironments

Abstract

Contractile myocytes provide a test of the hypothesis that cells sense their mechanical as well as molecular microenvironment, altering expression, organization, and/or morphology accordingly. Here, myoblasts were cultured on collagen strips attached to glass or polymer gels of varied elasticity. Subsequent fusion into myotubes occurs independent of substrate flexibility. However, myosin/actin striations emerge later only on gels with stiffness typical of normal muscle (passive Young's modulus, E approximately 12 kPa). On glass and much softer or stiffer gels, including gels emulating stiff dystrophic muscle, cells do not striate. In addition, myotubes grown on top of a compliant bottom layer of glass-attached myotubes (but not softer fibroblasts) will striate, whereas the bottom cells will only assemble stress fibers and vinculin-rich adhesions. Unlike sarcomere formation, adhesion strength increases monotonically versus substrate stiffness with strongest adhesion on glass. These findings have major implications for in vivo introduction of stem cells into diseased or damaged striated muscle of altered mechanical composition.

Figure 1.

Micropatterned collagen on PA gels. (A) Cross-linked PA was first bound to an aminosilanized glass slide or coverslip. Sulfo-SANPAH (x), a heterobifunctional cross-linker, was then applied to the free surface and attached to the gel by exposure to 365 nm light. (B and B′) A micropatterned glass stamp with 20-μm-wide channels was dipped in collagen-I to fill its channels. The “loaded” microstamp was overlaid on the PA gel bearing exposed succinimidyl ester groups and the reaction was allowed to proceed for 12 h at 37°C. (C) This gave a striped pattern of cross-linked collagen as visualized by immunofluorescence (C′). Addition of C2C12 myoblasts showed strong alignment on the patterns after two d (D) on both soft (1 kPa) and stiff (8 kPa) PA gels. Bars, 100 μm.

Figure 2.

Passive stiffness of normal and dystrophic muscle tissue. EDL muscle was dissected from normal (C57) and dystrophic (mdx) mice and probed in buffer by AFM. The inset schematic depicts the AFM probe indenting a thick tissue section on a glass support. Indentations of 1–2 μm were randomly made on ∼100-μm-thick samples at 10 or more locations per left and right EDL samples (n = 387). Force versus indentation curves were fit with a Hertz cone model (Engler et al., 2004a) in order to determine the passive elasticity as a Young's modulus, E. Table I lists the predominant stiffness for both normal C57 muscle and mdx-derived muscle.

Figure 3.

Initial myoblast responses to collagen-coated PA gels. Myoblasts were plated (5 × 10³ cells/cm²) on unpatterned gels to observe initial responses to elasticity. (A) Images of initial cell spreading and elongation on gels. Bars, 20 μm. (B) Spread cell area increases as a function of the substrate elastic modulus as well as time. A hyperbolic fit (Eq. 1) captures the trends at each time point. (C) Cells at 4 h spontaneously elongate (minor axis < major axis) only on stiff gels and glass, but by 24 h the elongated morphology predominates on all substrates. (D) An orientational correlation function (OCF, Eq. 2) evaluated between any two cells on unpatterned PA gels decays exponentially over ∼5–10 cell widths after 24 h For a fully aligned culture, OCF = 1; for a random orientation OCF = 0.5 (indicated by a dashed line); and for anticorrelated interactions OCF = 0. The fit, OCF = B e^{− Ax} + (1 − B), gives a decay constant A and the long range orientation B (B = 0.5 is random). Cells in close proximity thus prefer to align with their nearest neighbors, facilitating formation of myotubes. Error bars are SEM.

Figure 4.

Myotube striation is substrate stiffness dependent, whereas myocyte patterning is not. (A) By 2 d in culture, myoblasts align on the collagen patterns and begin to fuse into nascent myotubes. This occurs regardless of substrate elastic modulus and is sustainable in culture for weeks. Unbranched, isolated myotubes are consistently >90% of cells throughout the lifetime of all micropatterned cultures. Mouse primary cells (C57 derived) exhibit a similar behavior. (B and C) Several weeks after plating myoblasts on collagen-coated PA gels of varied stiffness or rigid glass, cells were stained for myosin (green), actin (not depicted), and the nucleus (blue) in order to assess cytoskeletal expression and organization. All substrates supported fusion into multi-nucleated myotubes, but after 2 or 4 wk only myotubes on gels of intermediate stiffness showed significant myosin striation. (D) On rigid glass, nascent myotubes showed robust arrays of stress fibers (F-actin is red) and tight substrate adhesion from vinculin by 1 wk and, as extrapolated from the PA gel results, myotubes showed no tendency to striate, even by 4 wk. Note the precision of the end-patterning on glass (see Materials and methods). Bars: (A) 100 μm; (B–D) 20 μm.

Figure 5.

Optimum substrate modulus for myotube differentiation. Myoblasts that have fused to myotubes on collagen-coated PA substrates exhibit a striation tendency that is strongly dependent on the substrate's elastic modulus. The percentage of total cells (n > 15 for each sample) exhibiting actin-myosin striation on the collagen-coated PA gels showed an optimum gel modulus of E* = 12 kPa based on simple Gaussian fits with common means and SDs (∼2.5–3 kPa) for 2 and 4 wk. Circles represent differentiation of cells grown on IPN-patterned glass. Error bars are ± SEM.

Figure 6.

Myotube underlayers provide optimal stiffness for myotube striation. Myoblasts were plated in two sequential layers on patterned glass. (A) An initial, bottom layer of cells was added and is schematically represented as nascent myotubes with square ends “pinned” by stress fibers. Subsequent addition of cells generates an top myotube that progressively differentiates. Within 1 wk after plating (B) and even more prominent at 4 wk (C), the top layer of cells striate both F-actin and myosin although clear striation of F-actin seems slower. Merged images of the top myotube at 4 wk demonstrate actin-myosin colocalization. Bars, 10 μm. (D) Striated myotubes in the top layer, after 1 and 4 wk, respectively constituted 68% (n = 34) and 85% of the total myotube population (repeated in duplicate). Immunostaining of vinculin in cells on top of cells shows diffuse labeling and no profound focal adhesions (not depicted).

Figure 7.

Fibroblast underlayers are too soft for myotube striation. (A) Fibroblasts were plated first on micropatterned glass with subsequent addition of myoblasts. The latter generates an top myotube (B), where differentiation is limited to fusion after 1–2 wk as there is no indication of striation. Myosin expression is nonetheless more prominent in the nascent myotubes than in the bottom fibroblasts that appear to contain similar amounts of F-actin.

Figure 8.

Substrate compliance influences myotube adhesion. Myotubes were mechanically peeled from their matrix after 4 wk in culture. (A) By applying a tension to a single myotube, the receptor-ligand bonds mediating adhesion were forcibly disrupted, and the myotube was peeled from the surface. (B) Cell peeling was accomplished by aspirating myotubes into micropipettes. The wall shear stress imposed by the flowing fluid along the length of the myotube pulled the cell into the micropipette. (C) Mean peeling velocity is plotted as a function of the imposed tension, as calculated across the width of a cell and for aspiration of at least three cells at each substrate modulus (Griffin et al., 2004). Although each myotube peels at a different rate due to adhesive differences, a common resting tension could be extrapolated by simple log fits for results at each substrate stiffness. (D) Comparing the zero-velocity tension for myotubes cultured for 4 wk on various substrates to their state of differentiation, i.e., striation per Fig. 4, demonstrates a disconnect. Adhesive strength simply increases with substrate modulus, but striation is minimal on very stiff and rigid substrates, showing instead an optimum at intermediate cell-like compliances. All error bars are SEM.

Figure 9.

Differentiation is controlled by substrate compliance and adhesive tension. Myofibrillogenesis begins with membrane-nucleated premyofibrils and generates mature myofibrils in a contractile force-dependent manner. On substrates of tissue-like stiffness, the adhesion is moderate (based on peeling tension), which allows premyofibrils to couple optimally to contractile forces, f*, fostering striation and myofibril maturation. On rigid or soft substrates, differentiation is delayed, if not stopped. Cells on rigid substrates (e.g., glass) produce large isometric forces from well-formed stress fibers and numerous vinculin-enriched focal adhesions, all of which limit actin reorganization into striations. Cell adhesion and contractile forces on soft substrates are extremely weak leading to poorly spread cells that lack striations or even stress fibers.