Fig 1.
Diagram of EC growing states and AFM probing.
Endothelial cells (green and orange) were cultured at NCF (A), SCF (B); and CF (C) status. For acquisition of Mgrt cells, a scratch in a straight line on EC monolayer was created using a cell scraper (D). After continued culture for two more days, individual Mgrt cells (indicated by arrows in E) from the leading edge of the scratch appeared in the scratched area. Cells at different status were probed by either tipped cantilever for EC stiffness measurement and topography, or by tipless cantilever (black triangle in A-C & E) coupled with aMB231 cell (pink in A-C & E) for measuring cancer cell-EC adhesion.
Fig 2.
Representative pictures (A, C) and diagram (B) of the approach for quantitating cell stiffness.
EC stiffness was measured by indenting a HBME cell on the cell cortex (arrow in A) with the tip at the end of the cantilever (A and B). The corresponding force curve included approach (red in C) and retraction (blue in C) curves that show cantilever deflection. Cell stiffness (Young’s modulus) was converted from the approach curve. Scale bar in A = 10 μm.
Fig 3.
Detection of cancer cell adhesion to ECs at four different growth states.
To compare adhesions at different EC growth states, a MB231 cell (Pink with purple) attached to the cantilever (black triangle) was brought into contact with non-confluent (NCF) EC (A), sub-confluent (SCF) EC (B), confluent (CF) EC (C) and migrating (Mgrt) EC (D) over the EC nucleus for three different cell contact times (0.5, 10 and 60 sec).
Fig 4.
EC cortical stiffness at four different growth states.
Box diagrams showing changes in cortical stiffness (Young’s moduli) of HBMEC-60 (left panel) and HPMEC (right panel) endothelial cells at four different growth/confluency states, NCF (non-confluent), SCF (sub-confluent), CF (confluent) and Mgrt (migrating). Whiskers, 5%– 95%. P–Ordinary One-Way ANOVA.
Fig 5.
Representative topography imaging (60 μm x 60 μm) of ECs at different growing states.
Cantilever deflection images (A, E, I and M for HBMEC-60; C, G, K and O for HPMEC) and cell height images (B, F, J and N for HBMEC-60; D, H, L and P for HPMEC) on ECs were performed to visualize the EC cytoskeletal structure underneath cell membrane. Both HBMEC-60 (E and I) and HPMEC (G and K) cells at SCF state revealed denser, thicker and more aligned stress fiber networks, compared to their respective NCF (A and C) states. The Mgrt (M) cell in HBMEC-60 showed less stress fibers than the NCF (A) cells, while in HPMEC, the Mgrt (O) cells retained more fibers inside the cell. There were no remarkable changes seen in cell height as a function of confluency in either HBMEC-60 (B, F, J and N) or HPMEC (D, H, L and P). Gray vertical scale bars (μm) are for cell height. Black scale bars = 10 μm.
Fig 6.
Representative images of EC cortical actin cytoskeleton at different growing states visualized by Phalloidin staining and confocal microscopy.
HBMEC-60 (A, C, E and G) and HPMEC (B, D, F and H) cells were cultured in chamber slides until reaching desired growth states (NCF, SCF, CF, Mgrt). Cells were fixed in 2% paraformaldehyde and permeabilized with 0.1% Triton X-100 as described in Materials and Methods. Cellular actin was visualized using Alexa Fluor 647 conjugated Phalloidin (pseudo green) and nuclei counterstained with Propidium Iodide (red). The Z-stacks (42–62 planes per stack) were acquired with 0.2 μm step size at 60x magnification and Max Intensity images were generated from each stack. Both HBMEC-60 and HPMEC cells at SCF (C and D) and CF (E and F) states revealed denser, thicker and more aligned stress fiber networks, compared to their respective NCF (A and B) and Mgrt (G and H) states. Scale bar in H = 40 mm.
Fig 7.
Total MB231/HBMEC-60 and MB231/HPMEC adhesion forces in different EC growing states for three cell-cell contact times.
Box diagrams showing changes in total adhesion forces between MB231 human breast carcinoma cells and HBMEC-60 (top panel A) and HPMEC (bottom panel B) endothelial cells at four different growth/confluency states, NCF (non-confluent), SCF (sub-confluent), CF (confluent) and Mgrt (migrating) and three (0.5 sec, 10 sec, 60 sec) cell-cell contact times. Whiskers, 5%– 95%. P–Ordinary One-Way ANOVA (ns–not significant).
Fig 8.
Distributions of rupture forces from MB231/HBME-60 and MB231/HPMEC adhesion measurements in four EC growing states.
For all three contact periods, the number of rupture forces were the greatest in NCF HBMEC-60 cells (1097, 3293 and 4669 counts for 0.5, 10 and 60 sec, respectively), reduced in the SCF cells, and lowest (2.59, 2.58 and 3.35 times less than those in NCF cells) in CF HBMEC-60 cells. Counts of rupture forces in Mgrt HBMEC-60 were slightly lower than those in the NCF cells, but higher than in the SCF cells. In HPMEC, the NCF cells for three contact times demonstrated the largest frequency counts (1423, 3108 and 4519 for 0.5, 10 and 60 sec, respectively) in four growing states. Mgrt cells showed very close counts to NCF cells. In each panel, Frequency Counts are plotted on Y axis and Rupture Forces (pN) on X axis.
Fig 9.
Typical AFM detachment curves of MB231 cells interacting with HBMEC-60 (A, C, E and G) and HPMEC (B, D, F and H) cells at NCF (A and B), SCF (C and D), CF (E and F) and Mgrt (G and H) states for 0.5 (green), 10 (red) and 60 (blue) sec cell contact times.
For both HBMEC-60 and HPMEC note significant increase in overall adhesion force between MB231 and ECs in NCF and Mgrt states compared to SCF and CF cells most noticeable for 10 and 60 sec cell contact times and accompanied by obvious increase in the number of both jumps and tethers (indicated with arrows in G) suggesting the involvement of both, receptors that are tightly associated (jumps) and weakly or not associated (tethers) with the cytoskeleton. Scale for Y and X axes are shown in H.
Fig 10.
Regression analysis of the relationship between EC elasticity and total strength of adhesion with MB231 cells at NCF, SCF, CF and Mgrt states for three (0.5, 10 and 60 sec) cell contact times.
For both HBMEC-60 (A) and HPMEC (B) there is a very strong negative correlation between elasticity and total strength of adhesion with MB231 at all three cell contact times with correlation coefficient (R) ranging from -0.76 to -0.97 for HBMEC-60 and from -0.83 to -0.93 for HPMEC.