Cell culture
Mouse MSCs (Invitrogen, US) within passage 10 were cultured in α-MEM medium (Hyclone, UT) supplemented with 10% (v/v) fetal bovine serum (Gibco, NY) and 1% (v/v) penicillin and streptomycin (Invitrogen, CA) under humidified conditions at 37 °C and 5% CO2. After reaching 70% confluence, the cells were treated with 2% trypsin/EDTA solution (Invitrogen, CA). The suspended cells were seeded on the bottom of 6-well plates at a density of 5×103 cells/cm2. After seeding for 1 h, the culture medium was gently removed and changed every 3 days. All methods were carried out in accordance with relevant guidelines and regulations.
Application of fluid shear stress
In our previous study, we designed a cone-plate flow chamber based on 6-well plates, in which FSS was uniformly distributed on the annular region of the well bottom 34. In the present study, the cone-plate flow chamber was further modified to apply FSS independently to each well (Fig. 1A). The device consisted of a stationary plate underneath a rotating cone (Fig. 1B). The distance between the cone’s tip and plate surface was controlled by placing a silicon membrane (Fig. 1C). Wall FSS exerted on the cells was controlled by specifying the cone’s rotation speed. Cone was fabricated according to hydrodynamic calculation 34,35, in which a uniform wall FSS field was provided on the cell surface. MSCs were cultured on the bottom of a 6-well plate. MSCs were treated with a specific FSS thrice a day for 1 h each time.
Numerical simulation
Using COMSOL Multiphysics software, FSS was simulated by setting the parameters of device and the angular rate of the cone in accordance with the method used in the previous study 34. For the cone-plate flow chamber model, the cone’s generatrix was machined as polyline to establish a uniform wall FSS field on the plate surface. The maximum radius \({R_c}\) of the cone was 15 mm, and its vertical distance \({h_c}\) to the tip was 1.3 mm. The gap \({h_0}\) between the cone’s tip and the plate surface was 0.2 mm. The radius of a well for the 6-well culture plate was 17 mm. A no-slip boundary condition was assumed for all rigid surfaces in the model except for the cone. A free surface boundary condition was used for the upper fluid surface within the well.
Navier–Stokes equations were used to define the flow behavior of viscous fluids. Incompressible viscous fluid was assumed to have a density of 1×103 kg/m3 and a viscosity of 1×10− 3 Pa∙s. An iterative method was used to solve the equations for steady flow, and convergence was identified when the relative tolerance was less than 0.001.
During numerical simulation, Reynolds number was computed as defined by Sdougos et al. 36. According to the calculated Reynolds number, fluid flow in the cone-plate should be assumed as steady laminar flow, and FSS is calculated according to the following equation:
\(\tau {\text{=}}\mu \frac{{\partial \nu }}{{\partial z}}\)
We first calculated the shear strain rate\(\gamma {\text{=}}{{d\nu } \mathord{\left/ {\vphantom {{d\nu } {dz}}} \right. \kern-0pt} {dz}}\)along thedirection close to the plate surface based on numerical simulation results and then obtained FSS by multiplying the shear strain rate with the viscosity coefficient \(\mu\). All rigid surfaces in the model were assumed to be non-slip boundary conditions. Upper flow surface in the well was adopted as an open boundary condition, and the rotating cone was adopted as a sliding boundary condition. MATLAB and Origin software were used for data processing.
Immunofluorescence staining
After incubation in a 6-well plate for 3 days, the cells were fixed with 4% paraformaldehyde for 30 min at room temperature, rinsed with phosphate buffer saline twice, permeabilized with 0.2% Triton-X100 for 10 min, and blocked with 1% bovine serum albumin for 60 min at room temperature. In addition, the cells were labeled with primary antibody against alkaline phosphatase (ALP, Santa Cruz Biotechnology, Inc., USA), type I collagen (COL I, Santa Cruz Biotechnology, Inc., USA), osteocalcin (OCN, Santa Cruz Biotechnology, Inc., USA), or YAP (ABclonal, CN) at a 1:200 dilution overnight at 4 °C. Second antibody labeling was performed in the dark for 60 min at 37 °C. The nuclei were stained with 0.1% Hoechst (Invitrogen, USA) for 10 min. The cells were double-stained with an apoptosis assay kit (One Step TUNEL Apoptosis Assay Kit, Beyotime, CA) and ALP-conjugated TRITC. After fixation and permeabilization, the cells were incubated in the dark with a TUNEL reaction mixture for 60 min at 37 °C.
Image analysis
For mechanical experiments, MSCs were chosen from annular regions with inner and outer diameters of 5 and 12 mm, respectively, where the wall FSS appears uniform. The spreading area and shape of a single cell were obtained from bright-field image. Its nuclear area was acquired from Hoechst-staining image. Bright-field or fluorescence images were analyzed using ImageJ software 27.
Statistical analysis
Statistical data were expressed as mean ± standard error of mean. Three independent experiments were performed with at least 50 cells for each group. One-way analysis of variance (ANOVA) with Turkey’s post hoc test for multiple comparisons was performed for statistical analysis by using Origin software. The osteogenesis ratio of three osteogenic markers was calculated by chi-square test. The mean values of different groups were regarded as significantly different when p < 0.05.