The micropillar array substrates were prepared as described previously [12]. Briefly, an array of photoresist posts was made on a silicon wafer by standard photolithography as a male mold. To make a female mold for the array of microposts, poly(dimethylsiloxane) (PDMS; Sylgard 184, Dow-Corning, Midland, MI, USA) was poured over the male mold, cured at 70˚C for 6 h, peeled off, and silanized to aid the release of PDMS from the mold. PDMS was then poured into the female mold, cured at 100˚C for 3 h, and peeled off to obtain the micropillar array substrate. In this study, we designed the PDMS micropillar array substrates with a hexagonal arrangement and with a pillar diameter, length, and center-to-center spacing of 3, 9, and 9 µm, respectively (Figure 1). The Young’s modulus of the

PDMS substrate was ~1.6 kPa [13], which was of the same order of magnitude as extracellular matrices such as elastin (~0.6 MPa [14]) surrounding VSMCs in vivo.

Porcine aortic smooth muscle cell lines (VSMCs; CAP35405, Cell Applications) were used as the test model. VSMCs were cultured in Dulbecco’s Modified Eagle’s Medium (Invitrogen) supplemented with 10% fetal bovine serum (JRH Bioscience), penicillin (100 unit/ml), and streptomycin (100 µg/ml) (Sigma) in a 5% CO₂ incubator at 37˚C. The cells were passaged repeatedly at a 1:4 split ratio when they reached ~80% confluence. Cells at passages 6 - 8 were used for all experiments.

Both micropillar array substrates and the flat substrates made of the same lot of PDMS were placed in 35-mm glass-bottom culture dishes (No. 0, Matsunami, Osaka, Japan). The surfaces of both substrates were exposed to oxygen plasma (5 mA, 10 Pa) for 2 min using a plasma generator (SEDE-P, Meiwafosis, Tokyo, Japan). All surfaces of both the pillar and flat substrates were coated with fibronectin (50 µg/ml, Sigma) to allow cell adhesion. VSMCs were cultured on these two substrates in DMEM supplemented with 10% FBS at 37˚C. The initial cell density was controlled to ~50 cells/mm². Time course images of the cells cultured on the flat and micropillar substrates were obtained for 24 h at 3-min interval using a microscope imaging system with a compact CO₂ incubator (BZ-X700, Keyence, Japan) to analyze cell migration on the substrates. The changes in the number of cells were also measured in the same areas of the substrates during culture; the specimen dishes were set on an inverted microscope, and their XYZ coordinates of the regions of interest (24 sets, 800 µm ´ 800 µm) were recorded. Then the bright field images of the cells at the same location on the substrates were obtained at every observation period during the culture period.

EdU (5-ethynyl-29-deoxyuridine) is a nucleoside analog of thymidine that is incorporated into DNA only during synthesis, allowing visualization of the newly synthesized DNA (Zeng et al. 2016). This method is a less toxic alternative to BrdU incorporation. VSMCs seeded on the micropillar array substrates and the flat substrates were preincubated at 37˚C for 24 h and treated with EdU in a dedicated buffer (Click iT EdU Imaging Kits, Molecular Probes) for 4 h and 12 h, as per the manufacturer’s protocol, to ensure the capture of the majority of proliferating cells. The cells were fixed, permeabilized, blocked for nonspecific protein binding, and stained with Alexa Fluor 488 for 30 min under Cu(I)-catalyzed click reaction conditions, as described by the manufacturer. The cells were washed with PBS and fluorescently stained for their intranuclear DNA, as described in the previous section.

Standard methods to quantify intracellular protein expression, such as Western blotting, require large number of specimen cells (>10⁶ cells) and it cannot be used for the analysis of protein expression in a single cell level. However, it was unable to obtain a sufficient amount of protein for western blot analysis in this study. Thus, in order to investigate the effects of the mechanical deformation of the nucleus on the smooth muscle contractile protein expression, such as α-smooth muscle actin (α-SMA), the fluorescence intensity was measured accurately in a single cell level using the method described previously [15]. Briefly, VSMCs cultured on the micropillar and flat substrates for 3 days were detached with trypsinization and moved to the glass-bottom culture dishes coated with poly-D-lysine. Then, the detached round-shaped cells were fixed immediately with a phosphate buffered saline containing 3.7% formaldehyde for 5 min, permeabilized with PBS containing 0.5% Triton X-100 for 5 min, and rinsed with PBS containing 1% BSA. For α-SMA staining, the fixed and permeabilized cells were incubated for 1 h at 37˚C with α-SMA antibody (clone 1A4 (A2547), Sigma) in PBS containing 1% bovine serum albumin. After washing, the cells were incubated for 1 h at room temperature with Alexa Fluor 488-conjugated secondary antibody (Molecular Probes; 1:200). Actin filaments in the cells were also stained fluorescently by incubation with Alexa Fluor 546-conjugated phalloidin (Molecular Probes, USA) at a concentration of ~200 nM for 1 h at room temperature.

The stained isolated cells were set on the stage of an inverted fluorescent microscope equipped with a confocal system (CSU-X1; Yokogawa, Tokyo, Japan) with a multicolor light emitting diode (LED) fluorescence system (Light Engine Spectra-X; Opto-line, Osaka, Japan) and a digital complementary metal-oxide semiconductor (CMOS) camera (ORCA-Flash4.0 V2; Hamamatsu Photonics, Hamamatsu City, Japan). The fluorescence images slices of their actin filaments and α-SMA were obtained in the same cell using an oil-immersion objective (60´, NA = 1.40) and recorded in a personal computer. These image slices were collected in the range of the thickness of the isolated cells (~25 µm) at a 0.5-µm interval. The reconstructed plane images were obtained with an integration of the slices and their fluorescent intensity was measured as indices of the intracellular expression of actin filaments and α-SMA.

Data were expressed as mean ± SD. Differences were analyzed by the Student’s paired and unpaired t-test. For cell proliferation, data were assessed using ANOVA with a correction for multiple comparisons, followed by a Steel-Dwass test for multiple comparisons of the means between two groups using the statistical analysis program MEPHAS (in Japanese, http://www.gen-info.osaka-u.ac.jp/testdocs/tomocom/). P-values < 0.05 were considered significant for all analyses.