Tempo-Spatial Change of Cellular Stiffness and Geometry in the Process of Developing Epithelial Cell-Cell Adhesion Measured by Atomic Force Microscopy

For sheet-like tissue, mechanical strength at cell-cell adhesion is important. However, it is unclear how long the mechanical strength is constructed. To clarify temporal change of the strength, we measured elasticity of epithelial cells during cellular migration to cover a vacancy by atomic force microscopy. Immediately after the construction of cell-cell contact, elasticity was low. However, elasticity increased following 90 min. Time-lapse observation of cellular morphology revealed that cells became spindle-shaped after the construction of cell-cell contact and then became round. The changes in elasticity and morphology would be multi-cellular dynamics in the epithelial wound healing process.


Introduction
Mechanical strength at the cell-cell contact point plays an important role in the maintenance of cellular sheet.When cellular sheet is injured, cells which are facing the wound migrate to the injured area and regenerate cell-cell adhesion (i.e.wound closure).In this process, cells do not only make a new contact with a neighboring cell but also regenerate mechanical strength at the new cell-cell adhesion.The regenerated mechanical strength enables the cell sheet to endure external stress such as body fluid and stretching.From the view point of dynamics of specific molecules, the strength of cell-cell adhesion was investigated [1,2].However, mechanical properties at the cell-cell level and their time course are still not fully investigated.Atomic force microscopy (AFM) has been used to measure mechanical property distribution of living cells.Radmacher et al. measured elasticity mapping of a single cell by AFM [3].Besides, elasticity mapping of epithelial cell colony was measured by Hassan et al. [4].Our group modified a commercial AFM to wide-range scanning probe microscopy (WR-SPM) which enabled acquisition of elasticity and topography distribution of thick sample like epithelial tissue over sub-millimeter scan range [5].By using the AFM or WR-SPM, some research groups showed that cell-cell boundaries are stiffer than the center of the cells [4,6].
In the present study, we measured a time course of topography and elasticity distribution of epithelial cells during regenerating cell-cell adhesion using WR-SPM.To mimic wound closure, we made a micro stencil on a cell culture dish, cultured cells on the dish and peeled the stencil to make a vacancy for cell migration.After peeling the stencil, cells migrated and made first contact with their neighbors.In this process, elasticity at the cell-cell contact point increased.Besides, cells changed their morphology from round to spindle to round.We will discuss these changes of elasticity and cell shape in the wound closure process.

Mimicking wound closure and cell culture
To mimic wound closure in vitro, we chose an experimental system which is based on removal of silicone elastomer on cell culture dish and following cell migration into the generated space [7].Base and cure agent of silicone elastomer (SH 9555; Dow Corning Toray, Japan) were mixed in the ratio of 1:10 and put onto the cell culture dish.Successively, the elastomer was shaped into a thin line with the width of about 150 µm using a glass needle controlled by a micromanipulator (MP-285; Sutter Instrument, U.S.A) and incubated overnight at 20 °C to make sure of hardening [Fig.1(a), (b)].To ease removal of hardened elastomer, a cell culture dish was coated with 300 µg/ml collagen (Cellmatrix Type I-C; Nitta Gelatin, Japan) in HCl at 20 °C for 1 h prior to the forming of elastomer.Madin-Darby canine kidney cells (MDCK; RIKEN BioResource Center, Japan), which were suspended in cell culture medium (DMEM; Sigma-Aldrich, Japan) containing 10% fatal bovine replacement (BIST-TEC; Equitech-Bio, U.S.A) and 1% antibiotics (Sigma-Aldrich), were seeded onto the cell culture dish and incubated at 37 °C [Fig.1(c)].After 2 h incubation, culture medium was replaced with HEPES-buffered DMEM (Sigma-Aldrich).After additional 2 h incubation, the silicone elastomer was removed [Fig.1(d)], that is to say, we began AFM measurement 4 h after the cell seeding.To keep the cell medium fresh during the AFM measurement, HEPES-buffered DMEM was substituted every 2 h.

Measurement of cellular elasticity and topography by WR-SPM
We measured temporal changes in the cellular elasticity and topography using WR-SPM.Since the WR-SPM has maximal scan range of 400 µm square in the xy-plane and 23 µm along the z-axis [5], we can easily measure elasticity and topography of epithelial cells getting higher during migration and making cell-cell adhesion.We used a cantilever with a spring constant of 0.01 N/m (MLCT-AUNM, Veeco Instruments, U.S.A).Topography and elasticity distribution were obtained simultaneously by force volume mode described previously [8].In the force volume mode, force-versus-distance curves were obtained, the curves were fitted to Hertz's model [9], and thereby cellular elasticity was evaluated.We took 30 min to acquire a set of topography and elasticity distribution by 64 * 64 points of the force volume mode.To observe more detailed temporal changes in at cell-cell contact regions, we introduced elasticity and height kymography (64 * 1 points of the force volume mode) on a line perpendicular to the linear silicone elastomer.The time interval of the kymography was set as 10 min.To observe high spatial resolution of cellular morphology, we used conventional contact mode which gives us height and deflection images.

Statistical analysis of cellular elasticity and topography
To show temporal changes in elasticity and topography around the new cell-cell contact point, the following statistical analysis was performed.First, we defined a minimum height point along a line perpendicular to the linear vacancy (the area generated by the removal of the linear elastomer) as cell-cell contact point.For the 15~25 lines which are perpendicular to the vacancy, the cell-cell contact points were defined.The cell-cell contact points were set as the origin.Topographic and elasticity data as a function of distance from the origin were averaged.The mean and the standard deviation of the set of data were plotted in graphs.The significant differences in elasticity between cell-cell contact points (72 data points from 3 cells) and center of cells (179 data points from 5 cells; distances from the contact points were greater than 10 µm) were determined by Student's t-test.

Results and discussion
To measure the change of elasticity during maturation of cell-cell adhesion, we obtained time-lapse images of topography and elasticity in wound closure process.Cells migrated into the vacancy and contacted each other [Figs. 2 and 3].We performed the elasticity kymograph on a line perpendicular to the linear vacancy [Fig.2].We observed increasing elasticity at the cell-cell contact points (Typical data is shown in Fig. 2(b)).We also observed the increase in elasticity at the cell-cell contact points by 64 * 64 force volume mode [Fig.3(b)] and the statistical analysis [Fig.4].Elasticity around the contact point was low at 60 min, however, it significantly increased (P < 0.01) for 150 min and then stabilized at 240 min.Finally, elasticity at the contact points was higher than center of the cells.Elasticity at the center of the cells didn't increase through the measurement [Fig.4(c)].Besides, we observed changes in the height gap between the cell-cell contact point and the highest point.The height gap decreased, in other words, the epithelial sheet got gradually smoother within 120 min.Next, we acquired time-lapse deflection images by contact mode [Fig.5].After cells were contacted, cells around the contact region were spindled.Spindled cells changed into rounded cells following 120 min.The same tendency was observed in the topographic image shown in Fig. 2

(a).
There would be a clear correlation between changes in elasticity and dynamics of molecular component at the cell-cell contact point.Complex of adhesion molecules interacting with actin cytoskeleton [10] is thought to have an important role in strengthening cell-cell adhesion [2].It was reported actin fibers interacting with cell-cell adhesion complexes in MDCK cell were formed within 120 min [11,12].This time-scale is close to our results.Additionally, elasticity at the center of the cells didn't increase over time.This observation indicates that reconstruction of cytoskeleton by cell adhesion and spreading will not affect the increase of elasticity at the cell-cell contact points.Therefore, cytoskeletal connection at cell-cell adhesion is suggested to contribute to the increasing elasticity at the adhesion point.To investigate a more detailed relationship between molecular dynamics and mechanical properties, combination of fluorescent microscopy (e.g.observation of green fluorescent protein-tagged protein in living cells) and WR-SPM is preferable.

Conclusion
We succeeded to measure a time course of topography and stiffness distribution of epithelial cells in the wound healing mimicking condition.Changes of elasticity distribution were observed after cells were contacted.Cells became spindle-shaped and then round.In order to study more detailed mechanisms of wound healing from the view point of cellular mechanics and molecular component, we are planning simultaneous measurement of fluorescent microscopy and WR-SPM.

Acknowledgment
This work is supported by Exploratory Research (23651099) to K. K., and by Grant-in-Aid for Young Scientists (B) (23770167) and Grant-in-Aid for Scientific  Research on Innovative Areas (24106502) to T. M. from the Japan Society for the Promotion of Science.

Fig. 1 .
Fig. 1.Schematics of sample preparation and AFM measurement.(a) Silicone elastomer on glass was shaped by a glass needle.(b) Silicone elastomer was shaped into a thin line.(c) MDCK cells were seeded on glass.(d) Silicone elastomer was peeled off.(e) Side view-image of measurement.(f) Over view-image of measurement.White arrow indicates direction of cell migration.

Fig. 2 .
Fig. 2. (a) An example of the time course of line-scanned topography and elasticity (kymograph).Thin dashed line indicates analyzed cell-cell contact points.The vacancy was closed 40 min after the beginning of the observation.Bar = 20 µm.(b) Time course of elasticity at the cell-cell contact was analyzed from (a).

Fig. 3 .
Fig. 3. Time-lapse (a) topographic and (b) elasticity images of epithelial cells in the process of wound closure obtained by 64 * 64 force volume mode.The area framed by thick dashed line indicates vacancy.Cells migrated into the vacancy for 60 min.Thin dashed line indicates cell-cell contact points (The definition is described in section 2.3).All images consist of 35 * 35 pixel data which is a part of 64 * 64 pixel data.Bar = 10 µm.

Fig. 4 .
Fig. 4. (a) Averaged elasticity (upper) and height (lower) distribution plotted as a function of the distance from the cell-cell contact point.Gray arrows indicate the gap between contact point and peak height.(b) Averaged elasticity around cell-cell contact points.Error bars denote standard deviation from 72 data.Statistical significance was analyzed with the unpaired t-test (* P < 0.01).(c) Averaged elasticity at center of the cells.Error bars denote standard deviation from 179 data.

Fig. 5 .
Fig. 5. Time-lapse deflection images of epithelial cells in wound closure process.Dashed line indicates cell-cell contact points.Identical cells facing contact line were marked as "#" or "*".All images consist of 128 × 128 pixel data.Bar = 20 µm.