Force mapping in epithelial cell migration

Abstract

We measure dynamic traction forces exerted by epithelial cells on a substrate. The force sensor is a high-density array of elastomeric microfabricated pillars that support the cells. Traction forces induced by cell migration are deduced from the measurement of the bending of these pillars and are correlated with actin localization by fluorescence microscopy. We use a multiple-particle tracking method to estimate the mechanical activity of cells in real time with a high-spatial resolution (down to 2 microm) imposed by the periodicity of the post array. For these experiments, we use differentiated Madin-Darby canine kidney (MDCK) epithelial cells. Our data provide definite information on mechanical forces exerted by a cellular assembly. The maximum intensity of the forces is localized on the edge of the epithelia. Hepatocyte growth factor promotes cell motility and induces strong scattering activity of MDCK cells. Thus, we compare forces generated by MDCK cells in subconfluent epithelia versus isolated cells after hepatocyte growth factor treatment. Maximal-traction stresses at the edge of a monolayer correspond to higher values than those measured for a single cell and may be due to a collective behavior.

Fig. 1.

Scanning electron micrographs. (A) Closely spaced microfabricated posts after PDMS molding. (B) Individual cells lying on μFSA (1 μm diameter and 2 μm distance center-to-center). (C) A cell monolayer (2 μm diameter and 3 μm distance center-to-center). (C Inset) Magnified view of the area delimited by the black square. Cells spread only on the top of pillars (B and C).

Fig. 2.

Magnitude of forces applied in and at the edge of a monolayer. (A) Transmission image of a monolayer grown onμFSA for 2 days (2μm diameter and 4 μm distance center-to-center). (Inset) Drawing of the adherent cell-cell junctions. (Scale bar: 12 μm.) (B) Histogram of magnitude of forces measured on the whole film. The stack contains 72 images and corresponds to 2 h. (C) Reconstructed image: each post is associated with a gray square localized at the undeflected position of the post in the image. Grayscaled mapping depends on the magnitude of the force from white (for low forces) to black (for high forces).

Fig. 3.

Magnitude and orientation of traction stress along the edge of a monolayer. (A) A growing cell monolayer on μFSA (2 μm diameter, 4 μm center to center) observed by transmission microscopy (60× air objective). The edge of the monolayer is outlined by the white line. The white arrows indicate the resulting force applied on four consecutive posts along the edge (indicated by circles). The magnitude of these traction stresses is not uniform, but the orientation remains centripetal all along the edge. (B) Average traction stress versus distance from the edge. Equidistant posts from the edge are pooled together to calculate for each distance the corresponding average traction stress. Posts used to calculate the last point are further than 16 μm toward the interior of the monolayer of A.

Fig. 4.

Temporal evolution of forces applied by cells on individual posts. The time-lapse sequence analyzed here corresponds to the area boxed in Fig. 2 A. (A) Force cartographies at different times. The edge of the cell monolayer is figured by the purple line; the cells are located in the white part of the image. Center-to-center spacing between pillars is 4 μM. An arrow of such a length corresponds to a 30-nN force. The origin of each arrow is the undeflected position of the corresponding post. Four differently located posts have been labeled with colored circles. (B) Forces applied by cells on these four posts as a function of time. Vertical arrows indicate the time of the image sequence (a–d). The blue curve illustrates the increase (≈10 min) and release (≈30 min) of the force as the leading edge passes over the post. The middle curves (magenta and red) reflect the residual activity further from the edge than the blue one. The green curve corresponds to an uncovered post.

Fig. 5.

Immunofluorescence labeling of actin. Repartition of actin in the interior of a normal monolayer (A) and for a monolayer treated with HGF (B) for 6 h (100× oil objective). White arrows indicate actin-rich fibrillar structures due to dissociation of the epithelium.

Fig. 6.

Forces applied by a single motile cell during migration. (A) Temporal variation of the magnitude of forces exerted on four posts (a–d) during a 30-min-long experiment. Posts positions are shown on B; the colors of the circles drawn around each post correspond to the ones of the curves on A. (B) Cell position toward the chosen posts and corresponding intensity pictures at three different times. The yellow arrow represents the direction of migration. Intensity pictures are presented in grayscale (white, low forces; black, high forces). Cell border is figured by a black line. Times corresponding to the three pictures in B are reported on A by vertical brown dotted lines. Post d, chosen as control, is not covered by the cell. The dimensions of the posts used here are 5.2 μm in height and 1 μm in diameter. (Scale bar: 10 μm.)