Adhesion and migration of ovarian cancer cells on crosslinked laminin fibers nanofabricated by multiphoton excited photochemistry

Abstract

Ovarian cancer is the deadliest gynecological cancer, which may arise in part due to the concurrent invasion and metastasis of high grade tumors. It is thus crucial to gain insight into the adhesion and migration mechanisms in vivo, as this may ultimately lead to new treatment/detection options. To explore this possibility, we have used multiphoton excited photochemistry (MPE) to synthesize models of the ovarian basal lamina consisting of crosslinked laminin nanofibers to quantify the adhesion/migration dynamics. The nanostructured laminin patterns permit the systematic comparison of total migration, directed migration, adhesion, and morphology of "normal" immortalized human ovarian epithelial cells (IOSE) and three lines of varying metastatic potential (OVCA433, SKOV-3.ip1, and HEY-1 cells). We find that the migration of all the cell lines is directed by the crosslinked fibers, and that the contact guidance enhances the total migration rates relative to monolayers. These rates increase with increasing metastatic potential, and the more invasive cells are less rigid and more weakly adhered to the nanofibers. The extent of directed migration also depends on the cell polarity and focal adhesion expression. For the invasive cells, these findings are similar to the integrin-independent ameboid-like migration seen for polar cells in collagen gels. Collectively, the results suggest that contact mediated migration as well as decreased adhesion may be operative in metastasis of ovarian cancer in vivo.

Fig. 1

maging of the crosslinked laminin structures used for the migration and cytoskeletal studies. (a) Immunofluorescence imaging of the crosslinked laminin fibers, scale bar = 20 microns; (b) 10× phase contrast showing OVCA433 cells on the laminin fibers, scale bar = 100 microns; (c) 4× phase contrast showing cells on the pattern (delineated by the dashed box) and those located on the background BSA monolayer, scale bar = 200 microns.

Fig. 2

epresentative migration trajectories for the IOSE (a) and HEY-1 cells (b) over 10 h, where each step size is 10 min. (c) Bar graph of the average migration speeds for the IOSE, OVCA433, SKOV-3.ip1, and HEY-1 cells where the error bars are standard error. The resulting p values are given in Table 1.

Fig. 3

irectional migration data for the four cell lines, plotted in terms of y/x displacement where y corresponds to the laminin fiber axis. All cell lines have ratios greater than one, indicating the importance of contact guidance. The corresponding p values from pairwise t-tests are given in Table 2.

Fig. 4

rajectories of SKOV-3.ip1 cells on (a) fibers, where the box delineates the region of fabricated laminin fibers from the background BSA; and on a self assembled monolayer (b). In addition to the difference in directionality, cells on the fibers have increased total migration (37.8 ± 1.2 vs. 30.7 ± 1.1 μm h⁻¹).

Fig. 5

(a) Cytoskeletal staining for IOSE, OVCA433, SKOV-3.ip1 and HEY-1 cells, where the columns from left to right are phase contrast, focal adhesions (red, anti-vinculin), F-actin (green, phalloidin), and the two color overlap. Field size = 200 microns for all frames. (b) Enlarged version for IOSE cells showing overlap of the focal adhesions with the laminin fibers, where the latter are drawn in based on the phase contrast image.

Fig. 6

ar graph of the average numbers of focal adhesions for the IOSE, OVCA433, SKOV-3.ip1, and HEY-1 cells where the error bars represent the standard error. The resulting p values are given in Table 3.

Fig. 7

reas of spread IOSE, OVCA433, SKOV-3.ip1, and HEY-1 cells, where the error bars represent the standard error. The resulting p values are given in Table 4.