Influence of neurite curvature on axonal polarization.

<p>(A) DW4 set of patterns of increasing curvature along directions L2–L4; Scale bars, 20 µm. (B) Partial and complete unhookings observed on fixed cells (microtubules: green, F-actin: red). White arrows point to partial unhooking, characterized by a displaced neuritic shaft still attached to the substrate by a large lamellipodium. The yellow arrow indicates a complete unhooking characterized by a high density of MTs crossing the pattern arch and remaining entities strictly following the curved adhesive line. Scale bar, 10 µm. (C) Time-lapse experiment (indicated in minutes, beginning 30 hours after plating) of a neurite developing on a DW4 pattern showing partial unhooking (white arrow). The black arrowhead points to the neurite tip and the green dashed line marks the position of the adhesive pattern. Scale bars, 20 µm. Refer to <a href="http://www.plosone.org/article/info:doi/10.1371/journal.pone.0033623#pone.0033623.s007" target="_blank">Movie S1</a> for the original time-lapse sequence. (D) Physical modeling of a curved neurite (in red) as an elastic wire under tension Τ. Curvature is reflected by the angle γ (see text), and Fγ = 2 T sin γ (black arrow) is the force experienced by the elastic wire. (E) Percentages of neurons displaying unhookings when grown over DW0 (0%), DW4-1 (4.7%), DW4-2 (7.0%), DW4-3 (10.6%), and DW4 (22.7%) patterns. (<i>n</i> = 117, 129, 128, 132, and 132, respectively). (F) Preferential axonal specification along the straight direction L1 were plotted from stage 3 neurons plated over DW0 (24.4%), DW4-1 (32.6%), DW4-1 (34.7%), DW4-3 (45.5%), and DW4 (52.3%) patterns. (<i>n</i> = 115, 285, 225, 330, and 216, respectively). ***, significantly different from random, <i>p</i><0.001.