Figure 1.
Depiction of the adhesive substratum on which surface features and chemistry could be independently controlled.
Etching times (0–90 min) in KOH determine the size of pyramidal features and hence surface roughness. Subsequent functionalisation with self-assembled monolayers control RGD densities. Monolayer chemistry consists of a base layer of undecylenic acid that couples 1-amino hexa(ethylene glycol) moieties (EO6-X) to the surface. GRGDS peptides were grafted to activated EO6-OH. RGD surface densities therefore were controlled by the ratio of EO6-OH and EO6-CH3.
Figure 2.
AFM images of silicon before and after etching.
(A) Silicon (100) before etching and after 90 min etching in 20% KOH at 50°C (B–C). The following topographical characteristics were measured (D): *root mean square (RMS) roughness, Rq, averaged over the entire surface; †RMS roughness of flat sections of an etched surface that contains no pyramids as indicated by the white square in B; ‡average height of pyramids from base to apex as shown in C. For the latter, 3D surface plots were used with pitch and roll angles set to 0°, thus giving a side view of the surface. Data are means ± standard deviation of 9 independent measurements.
Figure 3.
The number and cell area of endothelial cells is different on flat silicon and etched silicon.
Serum-starved endothelial cells were adhered for 30 min on chemically unmodified silicon that were etched for 0–90 min. (A) The number of adherent cells per square micron is significantly different between flat (0 min) and pyramid-presenting surfaces (10–90 min) as indicated by the asterisks (p<0.05). No difference was detected on etched surfaces (p>0.05). (B) Cell area was similar on all surfaces (p>0.05). Data correspond to a minimum of three independent experiments. Error bars represent standard deviations.
Table 1.
Average RGD densities per surface area and RGD spacing for various RGD∶OMe ratios.
Figure 4.
F-actin staining shows that endothelial cells adhere to flat and to etched surfaces.
Serum-starved endothelial cells were adhered for 30 min on silicon surfaces with various topographies and RGD densities. Characteristics of topographical features as a function of etching times are shown in Fig. 2. RGD densities determined by the RGD∶OMe ratio are listed in Table 1. Cells were stained with phalloidin-Alexa Fluor 555 to visualize F-actin and imaged with an epifluorescence (A) and a confocal (B–E) microscope. (B–E) To view cells and the underlying substratum, images were acquired in fluorescence (B) and in bright field mode (C) at the same focal depth and merged (D–E). Zoomed region indicated in D is shown in E. In D–E, F-actin is pseudo-colored green and substrate red. Closed arrows point to the silicon pyramids; dashed arrows to agglomerates of actin. Scale bars are 20 µm (A to D) and 5 µm (H).
Figure 5.
Quantification of the number of adherent endothelial cells on flat and on etched silicon surfaces.
(A) Number of endothelial cells after 30 min incubation with flat surfaces (A) and surfaces etched for 10 min (B), 30 min (C) and 90 min (D) for different RGD∶OMe ratios. The corresponding RGD densities are listed in Table 1. RGD stands for 100% RGD modification or 6×1011 RGD/mm2. OMe is the anti-fouling modification and contains no RGD. Data correspond to at least three independent surface preparations and >10 images per surface. Error bars represent standard deviations. Significant differences between data points in A–D are indicated (p<0.05; p>0.05).
Figure 6.
Confocal images of GFP-paxillin in adherent endothelial cells after 3 h incubation with surfaces with different topographical features, determined by the etching times, and RGD densities, defined by the RGD∶OMe ratio.
Dashed arrows point to the paxillin agglomerates, closed arrows point to ‘classical’ dash-shaped focal adhesions. Scale bar 20 µm (A), 2 µm (B).
Figure 7.
Quantification of endothelial cell area on flat and etched silicon surfaces.
(A) Cell spreading of adherent endothelial cells after 3 h incubation with flat surfaces (A) and surfaces etched for 10 min (B), 30 min (C) and 90 min (D) for different RGD∶OMe ratios. In A–D, each data point is significantly different from all other data points. Data are derived from at least three independent surface preparations and >10 images per surface. Error bars represent standard deviations.
Figure 8.
Endothelial cell migration over surfaces with different topographies, as defined by the etching time, and RGD surface densities of 6×105 RGD/mm2 (1∶106 RGD∶OMe ratio).
(A) Wound immediately after scratching; (B–E) wound after a further 3 h incubation. (C) Zoomed images of cells some distance from the wound edge as indicated in B. (C–E) Zoomed of images of cells migrating into the wound. Cells were stained with phalloidin-Alexa Fluor 555. Scale bar 100 µm (A, B) and 20 µm (D). (E) Quantification of wound closer after 3 h incubation.