Human umbilical vein endothelial cells (HUVECs, Sciencell), passage 2 - 8, were cultured in ECM (Sciencell) with 5% fetal bovine serum (FBS) and 1% penicillin/streptomycin (P/S), and 1% endothelial cell growth supplement (ECGS) at 37˚C in a humidified 5% CO₂ atmosphere. Medium was changed every 2 - 3 days. Subculture cells at 80% - 90% confluence. The cells were seeded on the PDMS substrates for 3 days before proceeding to the following experiments.

To acquire patterned substrates with parallel microgrooves (10 μm in width and 3 μm in depth), we used microfabrication technique to fabricate materials with desired surface topography. Poly (dimethyl siloxane) (PDMS) (Sylgard 184, Dow Corning, Midland, MI) was prepared according to the manufacturer’s instruction. PDMS was spin-coated onto the patterned silicon wafers to achieve desired thickness (~250 μm), degassed under vacuum, and cured at 75˚C for 1.5 h [6] [16]. The PDMS with micropatterned surface were removed from the template, cut into appropriate shapes and thoroughly cleaned by sonication, treated with Plasma Prep (11050Q-AX) to enhance the surface hydrophilicity, and coated with 2% gelatin for 1.5 hour to promote cell attachment. The images of the PDMS surface were collected by using scanning electron microscopy (SEM, JEOL JSM-5600).

The total RNA and microRNA were isolated using TRIzol® plus RNA Purification (No. 12183555, Invitrogen) and miRcute miRNA isolation kit (DP501, TIANGEN) respectively following the manufacture’s instruction. For qRT-PCR, cDNA of mRNA synthesis was performed by using the FastQuant RT kit (with gDNase) (KR106, TIANGEN), the mRNA level was measurement using SuperReal PreMix Plus (SYBR Green) (FP209, TIANGEN). GAPDH was used as a normalization control. The sequences of the primers used for real-time PCR are listed in Table 1. cDNA of microRNA was synthesized by miRcute miRNA First-Strand cDNA Synthesis Kit (KR211, TIANGEN) and the quantitative real-time PCR was performed by miRcute miRNA qPCR Detection Kit (SYBR Green) (FP411, TIANGEN) according to the manufacturer’s instruction. RNU6-2 was used as a normalization control in all miRNA measurements. Primers used were listed as shown in Table 2. The qRT-PCR was run on Applied Biosystems 7900HT Fast Real-Time PCR system (ABI). And the qRT-PCR data were analyzed using the comparative 2^−∆∆CT method.

For immunostaining, the actin-cytoskeleton and cell membrane was stained by antibody of FITC-phalloidin (FITC, AAT Bioquest) and β-catenin (Abcam) respectively, Nuclei were stained with 4',6-diamino-2-phenylindole (DAPI, Beyotime Biotechnology). The microscopic images acquired by Confocal Laser Scanning Microscope TCS SP5II (Leica).

50 ng/ml LPS was added to the ready HUVECs for 5 hours. THP-1 cells (ATCC) were washed with serum-free RPMI 1640 medium and suspended in medium with 4 μM of Calcein AM (1631214; Invitrogen). Cells were kept in an incubator for 30 minutes. The labeling reaction was stopped by adding the cell growth medium, and cells were washed with growth medium twice and resuspended in growth medium. After 5 hours of LPS treatment, HUVECs were washed once with THP-1 cell growth medium, and Calcein AM-loaded THP-1 cells were added to each well. After 1 hour of incubation, non-adherent cells were removed carefully. Adherent cells were gently washed with pre-warmed RPMI 1640 medium 3 times. The number of THP-1 cells per view was quantified from randomly acquired images.

Unless otherwise indicated, data represent the mean of at least 3 independent experiments and error bars represent the standard error of the mean (SEM). For all cases, p-values less than 0.05 were considered statistically significant. In all figures, * represent P < 0.05, ** represent P < 0.01, and *** represent P < 0.001. GraphPad Prism 6.0 software was used for all statistical evaluations.

In order to increase the cell attachment and biocompatibility of the biomaterial, 2% gelatin was used to coat the PDMS substrates surface to improve the hydrophilicity. After coating with gelatin, the hydrophilicity of PDMS surfaces was determined by water contact angle measurements. The results are shown in (Figure 1) and Table 3, which indicated that gelatin coating for 1 hour can significantly increase the hydrophilicity of the PDMS substrates, and no significant difference was observed between the flat surface and microgrooved surface.

To observe the HUVEC morphology modified by micro-patterned substrates, HUVECs were seeded onto flat poly (dimethyl siloxane) (PDMS) substrates and the micro-patterned one fabricated with a 10 μm parallel grooves in width and 3 μm in depth. SEM characterized the three-dimensional structure of PDMS substrates (Figure 2(a)), HUVECs morphology was demonstrated by FITC-phal- loidin, DAPI, and β-catenin staining (Figure 2(b)). As expected, phalloidin and β-catenin staining indicated that microgroove substrates had a pronounced effect on HUVEC morphology. In general, cells grown in microgroove PDMS substrates exhibited a more elongated morphology followed the guidance of parallel grooves, and the stress fibers were also parallel to the axis of groove alignment. The nuclear elongation was also observed in some cells.

The effect of the microgrooved surface on modifying the morphology of HUVEC shown here was selected from a series designed width based on our preliminary experiments. The width should be fine-tuned according to the size of the cells culture on it. In case of HUVEC, microgrooves with 10 μm in width were suitable to elongate the cells regardless of seeding density.

Previous studies have shown that topographic cues can significantly affect cell gene and microRNA expression levels in human vascular endothelial cells. To test whether our designed microgrooved PDMS substrates can also affect anti-inflammatory microRNAs, HUVECs were cultured on grooved and flat PDMS substrates biomaterial, and then the microRNAs were collected to examine the expression of miR-146a and miR-181b. The results showed that microgrooved PDMS substrates significantly increased the expression of miR-146a (P < 0.05) and miR-181b (P < 0.001) (Figure 3(a)). TRAF6 and importin-α3 were known

as the target genes of miR-146a and miR-181b, respectively. We further tested the mRNA levels of TRAF6 and importin-α3 and found both were significantly suppressed on grooved PDMS substrates (P < 0.01 and P < 0.001, respectively, Figure 3(b)).These results consistently supported our hypothesis that biomaterial substrate may induce specific microRNA expression including those regulating cell inflammatory response.

In response to endothelial activation, adhesion molecules, such as VCAM-1, ICAM-1, and E-selectin, act to initiate, promote, and sustain leukocyte attachment to the vascular endothelium. To assess whether microgrooved surface might suppress the expression of these specific proteins to inhibit the inflammation, the expression levels of VCAM-1, ICAM1, SELE (E-Selectin), and MCP-1, were measured in LPS-stimulated HUVECs by qRT-PCR. The results showed that the level of those inflammatory genes was all significantly decreased on microgrooved PDMS substrates relative to the flat substrates (VCAM-1, P < 0.05; ICAM1, P < 0.001; SELE (E-Selectin), P < 0.001, and MCP-1, P < 0.05, Figure 4(a)), suggesting that the endothelial inflammatory response was suppressed.

To further confirm the functional consequence of the decreased expression of the inflammatory genes, we employed an in vitro cell adhesion assay to assess leukocyte-endothelium interactions. Mononuclear cell line THP-1 was used in this assay. The adhesion of THP-1 cells to HUVECs on microgrooved PDMS substrates was markedly decreased (P < 0.05, Figure 4(b) and Figure 4(c)).

Taken together, these findings suggest that microgrooved PDMS substrates can inhibit the HUVECs inflammatory response by induction of the expression of anti-inflammatory microRNAs.