Figure 1.
Schematic drawing of photoactivatable nanopatterned substrate.
(A) The preparation of a nanopatterned substrate with periodically arrayed gold nanoparticles by block co-polymer nanolithography based on spin-coating. (B) Chemical structures of the cRGD ligand, photocleavable PEG12K and PEG2K-silane. The symbols used in C are also shown. (C) Photoactivation of the nanopatterned substrate. Before irradiation, the cRGD ligand is buried underneath the photocleavable PEG12K, and it is therefore not accessible to the cells. Near-UV irradiation releases the PEG12K, makes the cRGD ligand available in the nanoscopic geometry, and changes the surface from non-cell-adhesive to cell-adhesive. The intervening glass regions are passivated by PEG2K-silane. (D) Photoactivation of the homogenous substrate. An entirely gold-coated substrate was functionalized with a mixture of cRGD ligand and photocleavable PEG12K. The photoswitching strategy is the same as on the photoactivatable nanopatterned substrates, but this surface exposes cRGD ligand in a random geometry.
Figure 2.
Functionalization and photoirradiation keep the nanoarrayed gold particles intact.
(A–C) Scanning electron micrographs of (A) the as-prepared nanopatterned surface, (B) the chemically functionalized surface, and (C) the photo-irradiated surface. The scale bars represent 100 nm.
Figure 3.
Photo-induced cell adhesion on the photoactivatable nanopatterned surface.
(A) Dose-dependent increase of surface cell adhesiveness. The number of cells attached to a given area 3 hours after cell seeding is plotted against the irradiation energy. The error bars represent the standard deviations of the data from 3 different experiments. (B) Comparison of cell adhesiveness of the photoirradiated nanopatterned substrate to a similar surface where gold nanoparticles were functionalized with cRGD alone. The number of cells attached to a given area is plotted. The former surface was irradiated at 10 J before cell seeding. In either case, the glass regions were passivated with PEG2K-silane. The error bars represent the standard deviations of the data from 7 regions. (C, D) SEM images of a single HeLa cell migrating on the photoactivated nanopatterned surface. The image shown in D is a magnified image of the region indicated in C. (E, F) Photopatterning of HeLa cells in a stripe pattern on the nanopatterned surfaces, where gold nanoparticles were functionalized with (E) cRGD and photocleavable PEG12K or (F) photocleavable PEG12K alone. (G, H) Photoinduced cell migration. (G) HeLa cells were initially confined to a circular spot (indicated by the dotted line), and cell migration was induced by irradiating a square region (indicated by the solid line). A cellular image after 21 hours is shown in H. In all experiments, HeLa cells (4×105) were allowed to attach to the surface in serum-free medium for 1 hour, and the unattached cells were removed when the medium was changed to normal serum-containing medium. For C-H, the surfaces were irradiated with near-UV light (λ = 365 nm, 10 J) in PBS before cell seeding.
Figure 4.
Cell migration behavior before and after release from geometrical confinement.
(A) Schematic representations of the procedure for inducing cell migration on the photoactivatable homogenous and nanopatterned surfaces. The entire surface is initially non-cell-adhesive, and a 150-μm circular cell-adhesive spot is generated by the first irradiation. Cells are allowed to attach to the spot for 1 hour, and the unattached cells are removed by a medium exchange. The cells are further cultured for 8 hours to make a confluent circular cell cluster (B, F). The migration of the cells is induced by secondary irradiation through another photomask, which allows the selective irradiation of the surrounding regions (not over the patterned cells). The blue and yellow colors represent non-cell-adhesive and cell-adhesive surfaces, respectively. (B-I) Phase contrast images of the cells (B, F) before and (C, G) 3 hours, (D, H) 6 hours, and (E, I) 9 hours after the secondary irradiation on (B-E) the homogenous and (F-I) the nanopatterned gold substrates picked up from the supplemental movies (Movie S1, S2). The cells in the periphery of the images are nonspecifically attached cells or those have migrated from neighboring circles. The scale bar represents 100 μm.
Figure 5.
Immunofluorescence study of N-cadherin, cytoskeleton, and focal adhesion.
HeLa cells were initially confined within a 150-μm circular spot on (A, C, E, G) the homogenous and (B, D, F, H) the nanopatterned surfaces for 9 hours and their migration was induced. The cells were fixed (A, B) before and (C-H) at 3 hours after the confinement release and stained for (A-D) N-cadherin, (E, F) actin, and (G, H) vinculin. The scale bars represent 20 μm.
Figure 6.
(A, G) Circular clusters with a diameter of 150-μm formed on (A) the homogenous and (G) the nanopatterned substrates before the induction of migration. One representative cell which was originally located at the periphery of the cluster is chosen for each substrate and outlined with a red line. (B-F, H-L) The changes in the cellular shape shown in A and G at given time periods after migration induction. (M, N) The trajectories (black) of the cellular centroids of the cells indicated in A and G. The trajectories for 2 or 3 other cells in A and G are also shown, some of which are rotated to some extent so as not to overlap with other trajectories for convenience. (O, P) The averages of migration rate (O) and directional persistence (P) quantified from the centroid trajectories shown in M and N. The error bars represent the standard deviation for (M) 3 and (N) 4 cells. *Significant difference by a two-tailed Student's t-test (p<0.05). The scale bars represent 20 μm. The original images are taken from Movie S1 and S2.
Figure 7.
Phosphorylation of specific tyrosine residues on FAK in migrating cells.
Immunofluorescence images for (A, D) pY397, (B, E) pY861, and (C, F) total FAK in cells migrating on (A-C) the homogenous and (D-F) nanopatterned surfaces 3 hours after the release from the initial confinement in the 150-μm circle. (G-I) The average fluorescence intensity of 8–9 representative cells shown in A-F. (G) pY397, (H) pY861, and (I) total FAK. The error bars represent the standard deviations. The scale bars represent 20 μm. *Significant difference by a two-tailed Student's t-test (p<0.01).