Human umbilical vein endothelial cells (HUVECs) were purchased from Sciencell. The cells were cultured in ECM (Sciencell) consisted of 5% fetal bovine serum (FBS), 1% penicillin/streptomycin, and 1% endothelial growth factors provided by the manufacturer. Cell at passage 4 to 8, were seeded on the PDMS membrane for 3 days before proceeding to the following experiments.

To create patterned membranes with parallel microgrooves (10 μm in width and 3 μm in depth, with 10 μm distance between each groove), we used microfabrication techniques [16]. Poly (dimethyl siloxane) (PDMS) was prepared according to the manufacturer protocol (Sylgard 184, Dow Corning, Midland, MI), spin-coated onto the patterned silicon wafers to desired thickness (~250 μm), degassed under vacuum, and cured at 70˚C for 2 h [17].

The resulting micropatterned membranes were removed from the template, cut to appropriate dimensions and thoroughly washed and sonicated before use. The surface topography of micropatterned PDMS membranes was examined by scanning electron microscopy (SEM) [18].

Antibody against β-catenin and phalloidin-fluorescein conjugate was obtained from Abcam and AAT Bioquest, respectively. DAPI was obtained from Beyotime Biotechnology. The immunohistostaining of HUVECs was examined by Confocal Laser Scanning Microscope TCS SP5II (Leica).

We isolated total RNA and microRNA using TRIzol plus RNA Purification (Invitrogen) and miRcute miRNA isolation kit (TIANGEN) respectively following the manufacture’s instruction.

PureLink® miRNA Isolation Kit (Ambion) was used to purify microRNA which then collected to run miRNA microchip (Oebiotech Co. Ltd). The chip applied in this experiment was Agilent Human miRNA (8*60K) and analyzed by Feature Extraction software (version10.7.1.1) and Genespring software (version13.1) by Agilent Technologies.

For qRT-PCR, first-stand cDNA synthesis and quantitative assessment of microRNA was measured using miRNA first-strand cDNA synthesis kit and miRNA qPCR detection kit (CYBR green) from Tiangen, respectively. RNU6-2 was used as a normalization control in all miRNA measurements. Primers used were listed as shown in Table 1.

SuperReal PreMix Plus (SYBR Green) was used to determine the mRNA level by using the GAPDH as a normalization control. The sequences of primer were shown below: GAPDH (Forward: 5’ GGG AAG GTG AAG GTC GGA GT, Reverse: 5’ GGG GTC ATT GAT GGC AAC A), PTEN (Forward: 5’ TTT TGA AGA CCA TAA CCC AC, Reverse: 5’ TAT CAT TAC ACC AGT TCG TC).

The qPCR was run on 7900HT Fast Real-Time PCR system (Applied Biosystems).

Whole Cell Extraction kit (Chemicon) was use to collect protein from HUVECs. Protein lysates were quantified by BCA Protein assay kit (Novagen). SDS-PAGE and polyvinylidene fluoride membranes were used to analyze the protein sample. Before incubating with primary antibodies, the membranes were blocked in 5% non-fat milk. Primary antibodies included PTEN (ab32199, Abcam), caspase-3 (sc-7148, SANTA CRUZ) and β-Actin (#4970, Cell Signaling). Peroxidase AffiniPure Goat Anti-Rabbit IgG (H+L) (111-035-003, Jackson immunoResearch) was used as Secondary antibody.

To assess whether topography might influence the expression of microRNA, HUVECs were seeded onto flat PDMS membranes (F) or those fabricated with parallel microgrooves of 10 μm width and spacing (G). The SEM images of PDMS membranes with or without 10 μm micro-grooves were shown in Figure 1(A). Phalloidin staining demonstrated that microgroove substrates had a pronounced effect on HUVEC alignment. (Figure 1(B)). In general, cells cultured on microgrooves exhibited a more elongated morphology relative to those cultured on flat surfaces, and the stress fibers was also parallel to the axis of groove alignment. The nuclear elongation was also observed.

In addition, we applied microRNA chips to screen the expression levels of we measured the expression level of a number of miRNAs, however, we observed that microgroove topography significantly increased the expression of these miRNAs (Figure 2(B)). Among these microRNAs, miR-21 was increased significantly after culturing on microgroove topography.

According to related articles, miR-21 was demonstrated to involve in modulating the decrease of apoptosis in different adult tissues, such as HUVECs. We hypothesized that the increase of miR-21 expression caused by grooved materials could lead to the decrease of apoptosis. To test this hypothesis, HUVECs were cultured on grooved and flat PDMS; and then the protein was collected to test the expression of Procaspase-3 and cleaved caspase-3. The activation of caspase-3 was tested to confirm the apoptosis of HUVECs. Caspase family is important in the process of apoptosis, especially Caspase-3. When apoptosis happens in cells, caspase-3 will active and divide into two small sub-units (12KD and 17KD). Procaspase-3 expression levels increased and cleaved caspase-3 (17KD) expression levels decreased in HUVECs cultured on grooved PDMS in

comparison with flat PDMS (Figure 3). However the expressions of cleaved caspase-3 which is 12 KD have no difference.

Some articles showed that miR-21 was involved in expression of the proapoptotic gene PTEN. MiR-21 was illustrated to decrease the apoptosis of HUVECs and vascular smooth muscle cells by downregulation of PTEN.

Overexepression of miR-21 in HUVECs which cultured on grooved materials reduced PTEN protein expression. These results are consistent with negative regulation of PTEN by miR-21 in HUVECs (Figure 4).