Assaying stem cell mechanobiology on microfabricated elastomeric substrates with geometrically modulated rigidity

Abstract

We describe the use of a microfabricated cell culture substrate, consisting of a uniform array of closely spaced, vertical, elastomeric microposts, to study the effects of substrate rigidity on cell function. Elastomeric micropost substrates are micromolded from silicon masters comprised of microposts of different heights to yield substrates of different rigidities. The tips of the elastomeric microposts are functionalized with extracellular matrix through microcontact printing to promote cell adhesion. These substrates, therefore, present the same topographical cues to adherent cells while varying substrate rigidity only through manipulation of micropost height. This protocol describes how to fabricate the silicon micropost array masters (~2 weeks to complete) and elastomeric substrates (3 d), as well as how to perform cell culture experiments (1-14 d), immunofluorescence imaging (2 d), traction force analysis (2 d) and stem cell differentiation assays (1 d) on these substrates in order to examine the effect of substrate rigidity on stem cell morphology, traction force generation, focal adhesion organization and differentiation.

Figure 1 |

Flow diagram of the different sections of the protocol. The required and optional sections are color-coded white and gray, respectively, with arrows indicating the order in which the sections are performed. Note that some sections require key equipment, such as a plasma cleaner, critical point drier or UVO cleaner, in order to be executed. If these machines are not available, limited quantities of either non-functionalized or functionalized micropost array substrates can be ordered through our online service (www.seas.upenn.edu/~chenlab/micropostform.html). These respective shortcuts are indicated by boxes with dashed borders on the flow chart.

Figure 2 |

Characterization of micropost array masters and substrates, (a) Scanning electron micrographs of silicon micropost array masters of four different heights. From left to right, the heights of the microposts are 2.3, 5, 8.3 and 12.9 μm. The scale bar is 20 μm. (b) Finite element model (FEM) simulations of the deflection of PDMS microposts in response to an applied force at the tip. (c) The nominal spring constant (K), as computed from FEM analysis (bars) and Eq. 2 (curve), is plotted for PDMS microposts of different heights L. Reprinted with permission .

Figure 3 |

Replica molding of a micropost array master, (a) A PDMS negative mold is peeled from the silicon master and (b) can be cut to different sizes to cast arrays of different area, (c) The negative molds are then fluorosilanized in a vacuum desiccator, (d) Silanized negative molds are coated with a thin layer of uncured PDMS (middle mold), sandwiched against a coverslip (right mold) and cured, (e) A cured substrate is released by clamping the substrate while using tweezers to peel the mold, (f) Substrate quality can be quickly determined by looking at how the array diffracts light. A flawless substrate (left) diffracts light into many colors while a flawed substrate (right) has opaque regions.

Figure 4 |

Microcontact printing on micropost array substrates, (a) PDMS stamps are cut to the size of the micropost array and then coated with fibronectin solution (right stamp). Micropost array substrates can be left unmounted or mounted to a Petri dish, (b) A dried, coated stamp is aligned and laid on top of the micropost array, (c) Gentle pressure is applied with tweezers to ensure good contact between the stamp and the micropost tips, (d) The substrate and stamp are placed in ethanol before the stamp is peeled, to minimize any damage that peeling forces may impart.

Figure 5 |

Basic imaging of cells on micropost array substrates, (a) A low-magnification, bright field image of round, but adherent, cells after only 30 minutes of spreading on the micropost array. Scale bar is 150 μm. (b) A low-magnification, bright field image of cells after overnight incubation on the micropost array. Scale bar is 150 μm. The inset is magnified in (c) to show a spread cell, (d) A high-magnification, fluorescent image of the tips of the microposts under a cell. The base positions of the same microposts are shown in (e) and the GFP-expressing cell is shown in (f). These three images were acquired with a 40x oil-immersion objective. Scale bars are 20 μm.

Figure 6 |

General algorithm for analyzing traction forces from the micropost array substrate. Key steps for analyzing a representative cell are illustrated. Detailed instructions for a custom-written MATLAB program can be supplied upon request.

Figure 7 |

Representative images of cells on microposts, (a) A live cell, expressing LifeAct-GFP to visualize F-actin, on microposts with K = 7.22 nN/μm. (b) A fixed cell, stained for F-actin (green) and vinculin (cyan). K = 3.78 nN/μm. (c) A fixed cell constrained to a 30 μm x 30 μm micropattern and stained for F-actin (green) and vinculin (cyan). K = 18.19 nN/μm. (d) A fixed cell constrained to a 75 μm x 75 μm micropattern and stained for F-actin (green) and vinculin (cyan). K = 15.75 nN/μm. Scale bars are 20 μm.

Figure 8 |

Analysis of stem cell differentiation on micropost array substrates, (a) Micrographs of hMSCs on rigid and soft microposts that have been stained for alkaline phosphatase activity (blue) and lipid droplet formation (red) to indicate osteogenic and adipogenic markers, respectively. L and K indicate the micropost height and spring constant for the substrates shown in the corresponding micrographs. Scale bar is 300 μm. (b) Quantification of percentage of differentiating cells on rigid and soft microposts as well as coverglass as a control. * indicates statistical significance with P < 0.05. n.s. indicates statistical insignificance with P > 0.05. Reprinted with permission .