Molecular-scale substrate anisotropy, crowding and division drive collective behaviours in cell monolayers

Abstract

The ability of cells to reorganize in response to external stimuli is important in areas ranging from morphogenesis to tissue engineering. While nematic order is common in biological tissues, it typically only extends to small regions of cells interacting via steric repulsion. On isotropic substrates, elongated cells can co-align due to steric effects, forming ordered but randomly oriented finite-size domains. However, we have discovered that flat substrates with nematic order can induce global nematic alignment of dense, spindle-like cells, thereby influencing cell organization and collective motion and driving alignment on the scale of the entire tissue. Remarkably, single cells are not sensitive to the substrate's anisotropy. Rather, the emergence of global nematic order is a collective phenomenon that requires both steric effects and molecular-scale anisotropy of the substrate. To quantify the rich set of behaviours afforded by this system, we analyse velocity, positional and orientational correlations for several thousand cells over days. The establishment of global order is facilitated by enhanced cell division along the substrate's nematic axis, and associated extensile stresses that restructure the cells' actomyosin networks. Our work provides a new understanding of the dynamics of cellular remodelling and organization among weakly interacting cells.

Keywords: active nematics; anisotropy; cell migration.

Figure 1.

Experimental approach. (a) Chemical structures of the crosslinker (RM82), monoacrylate (RM23) and a photoinitiator. The full chemical details and substrate characterization data are provided in the Methods section. (b) Schematics of the cell (not to scale) placed on a substrate with nematic or isotropic molecular structure. Cells are labelled in both the cytoplasm (CellTracker, red) and nucleus (Hoechst, blue) channels. (c,d) A high-throughput automated microscope stage acquires 10 s of images at every time point, which are stitched together (dotted rectangles) to allow reconstruction of 10³ cell trajectories over the time span of days. The scale bar is $500\hspace{0.17em}\mu \mathrm{m}$. The dotted rectangle represents one field of view (FOV) captured by a 20× objective. (e) A close-up image of a single cell, with nucleus orientation θ_i labelled in grey, referring to the angle between the orientation of the cell nucleus and the horizontal direction or the direction of liquid crystal alignment. The velocity vector (with x- and y-components) is shown in yellow (red and blue), and the angle β denotes the velocity orientation with respect to the substrate. The scale bar is $100\hspace{0.17em}\mu \mathrm{m}$. The substrate alignment direction is parallel to ${\hat{\mathbf{e}}}_x$ unless otherwise specified.

Figure 2.

Snapshots of cell configurations on (a) an isotropic substrate with approximately 2200 cells and nematic substrates with (b) approximately 2200 cells and (c) approximately 4200 cells. The images are obtained by merging the nucleus and cytoplasm channels. The colour scale denotes the nucleus orientation. Panels (d–f) show the corresponding polar histograms of the nucleus orientations θ_i with respect to the direction of nematic order of the substrate ${\hat{\mathbf{e}}}_x$. The scale bars are $500\hspace{0.17em}\mu \mathrm{m}$.

Figure 3.

(a) Evolution of the cell-substrate order parameter S_cs with cell density ρ. Each colour corresponds to a different density obtained at different times after seeding. Solid lines serve as guides to the eye. Shaded regions denote the typical uncertainty in estimating S_cs (electronic supplementary material, figure S9). Errors in ρ (estimated to be approx. $5-10\mathrm{\%}$) are not shown. A red asterisk in the legend denotes an experiment with a different initial seeding density. (b) At the lowest densities, cells are isolated and their orientations are random. (c) At low and moderate densities, cells display spontaneous swirling motion. (d) At intermediate to high densities, lanes of cells of coherent velocities form. (e) These lanes widen, and eventually, (f) structural arrest takes place due to jamming. Panels (b–f) are taken on nematic substrates. The scale bars are $250\hspace{0.17em}\mu \mathrm{m}$ in (b,f) and $500\hspace{0.17em}\mu \mathrm{m}$ in (c–e).

Figure 4.

Analysis of spatial correlations of cell–cell orientation. The cell–cell orientational correlation function is plotted as a function of the separation between cell pairs for (a) isotropic and (c) nematic substrates. The insets show the corresponding semilog and log-log plots in (a,c), respectively. The correlation length ξ_θθ extracted either by fitting equation (3.4) or equation (3.5) is shown in (b) where red and black symbols denote data obtained using isotropic or nematic substrates, respectively. The black vertical line indicates the cell density at the onset of order, ρ_c. Long-range order is observed for cells on nematic substrates for ρ > ρ_c, as demonstrated by the decreasing power law exponent γ in (d) and increasing plateau value $C_{\theta \theta }^{\mathrm{\infty }}$ in (e).

Figure 5.

Asymmetric dynamics of cell motion. Spatial maps of the cell velocity field, colour-coded by the instantaneous direction of each velocity vector, for cells moving on (a) isotropic and (c) nematic substrates. For clarity, the velocities are plotted at 20% density in the full field of view (FOV), whereas the data of 100% density is plotted in the zoomed-in figure. The angular distribution of velocity with respect to ${\hat{\mathbf{e}}}_x$ is shown for (b) isotropic and (d) nematic substrates.

Figure 6.

Angular distribution of the cell division axis. (a) The time sequence of images captures a cell-division event. Here, ${\theta }_i^{\mathrm{div}}$ denotes the cell–cell orientation at the first time point after the division occurs. (b–e) The polar histograms denote the distribution of ${\theta }_i^{\mathrm{div}}$ for a nematic substrate in (b–d) and isotropic substrate in (e). n is the number of observations.

Figure 7.

Panels (a,b) show the (a) trefoil and (b) comet defects. Local director fields are overlaid on the cytoplasm channel. Tracking $+\frac{1}{2}$ defects on a (c) isotropic, and on (d,e) nematic substrate, when the comet is (d) perpendicular or (e) parallel to the alignment direction. Red double-sided arrows denote the director field of the substrate. Left and top panels: orientation overlaid on the nucleus channel, which provides information about the location and orientation of the cell nuclei. Yellow dotted lines denote the approximate outline of the comet and the defect. Right and bottom panels: precise local orientation was determined from the orientation of cell nuclei. The black dots and corresponding arrows denote the location and direction of movement of the defects. Dashed lines are added to guide the detection of defect movement. The scale bars are $200\hspace{0.17em}\mu \mathrm{m}$ in (a,b), and $100\hspace{0.17em}\mu \mathrm{m}$ in (c,e).