A Unidirectional Cell Switching Gate by Engineering Grating Length and Bending Angle

Abstract

On a microgrooved substrate, cells migrate along the pattern, and at random positions, reverse their directions. Here, we demonstrate that these reversals can be controlled by introducing discontinuities to the pattern. On "V-shaped grating patterns", mouse osteogenic progenitor MC3T3-E1 cells reversed predominately at the bends and the ends. The patterns were engineered in a way that the combined effects of angle- and length-dependence could be examined in addition to their individual effects. Results show that when the bend was placed closer to one end, migration behaviour of cells depends on their direction of approach. At an obtuse bend (135°), more cells reversed when approaching from the long segment than from the short segment. But at an acute bend (45°), this relationship was reversed. Based on this anisotropic behaviour, the designed patterns effectively allowed cells to move in one direction but blocked migrations in the opposing direction. This study demonstrates that by the strategic placement of bends and ends on grating patterns, we can engineer effective unidirectional switching gates that can control the movement of adherent cells. The knowledge developed in this study could be utilised in future cell sorting or filtering platforms without the need for chemotaxis or microfluidic control.

Fig 1

(a) Basic design of “V-shaped grating patterns” used in this study. (b) Directionality ratio of MC3T3-E1 cells on V-shaped grating patterns where bending angles are 45°, 90°, and 135°, respectively (N = 140, 3 independent experiments for 45° and 90° patterns, 4 independent experiments for 135° pattern). (c) Number of directional reversals per cell observed on these three patterns over 15 h (N = 140, one-way ANOVA with Dunnett’s post-hoc test, *p <0.05, **p <0.01).

Fig 2

(a) Directional reversal of MC3T3-E1 cell at 135° bend. (b) Locations of directional reversals recorded on various V-shaped grating patterns. (c) Speed of cells with and without directional reversals on V-shaped grating patterns (N = 140, 3 independent experiments for 45° and 90° patterns, 4 independent experiments for 135° pattern, one-way ANOVA with Dunnett’s post-hoc test, **p <0.01, ***p <0.001).

Fig 3

(a) Three possible routes taken by MC3T3-E1 cell approaching bend of V-shaped grating pattern. (b) “Expelled” cells at 45° bend. (c) Proportion of cells taking three possible routes on three V-shaped grating patterns (N = 140, 3 independent experiments for 45° and 90° patterns, 4 independent experiments for 135° pattern, chi-square test, *p <0.05, ***p <0.001). (d) Morphology of MC3T3-E1 cells at bends. Red: F-Actin, green: Vinculin.

Fig 4

Histogram of focal adhesion sizes on V-shaped grating patterns with (a) 45°, (b) 90°, and (c) 135° bends. (d) Total focal adhesion area per cell and (e) number of focal adhesions per cell area of cells at bends (N = 15, one-way ANOVA with Dunnett’s post-hoc test, *p <0.05, NS denotes not significant).

Fig 5

(a) Cell speed and number of reversal per cell on 135° V-shaped grating patterns with various segment lengths (N = 157, 5 independent experiments for 50 and 100 m patterns, 4 independent experiments for 200 m pattern, one-way ANOVA with Dunnett’s post-hoc test, *p <0.05, NS denotes not significant). Probability of cells that (b-d) advanced pass 135° bend of V-shaped patterns with various segment lengths from opposite directions, (e) moved in forward and reversed directions on straight gratings, and (f) advanced pass 45° bend of V-shaped pattern from opposite directions.

Fig 6

(a) Schematic diagram of cell touching both grating bend and end. (b) Total focal adhesion area developed at the side near grating bend and end (N = 14, ratio paired t-test, *p <0.05). (c) Total focal adhesion area developed at two ends of polarized cells on straight gratings (N = 16, ratio paired t-test, NS denotes not significant).