(49f) The Role of Cytoskeleton and Ion Channels in Cell Decision-Making Under Confinement

Introduction: Lymphatic vessels as well as perivascular, perimuscular and perineural spaces provide longitudinal tracks, which facilitate cancer metastasis. The cell decision-making process at intersections of either partially or fully confined microenvironments, has been analyzed in a generic way through paradigms of hydraulic resistance or chemical gradients without any mechanistic explanation. We herein present the first comprehensive analysis on how cross-sectional areas of resulting branches affect tumor cell decision-making, focusing on the role of cytoskeleton and ion channels in the absence of chemical gradients.

Materials and Methods: To examine the effect of channel dimensions on decision-making, we created Ψ-shaped PDMS-based microchannels coated with collagen I. The microchannel array confined MDA-MB-231 breast cancer cells either partially (WxH=10x10µm²) or fully (WxH=3x10µm²) before reaching a 3-way intersection consisting of channels of prescribed dimensions. Furthermore, we created microchannels with a dead end to account for decisions under infinite hydraulic resistance. To quantify cell speed and persistence we used the MTrackJ plugin of FIJI and a custom made MATLAB code. Statistical significance was assessed with non-paired Studentâ??s t-test.

Results and Discussion: Upon decision-making within a bifurcation of an infinite hydraulic resistance (dead end) versus a lower resistance (open end) branch, 20% of MDA-MB-231 cells chose to enter the dead end compared to ~5% of the HL-60 leukocyte-like cells. Hence MDA-MB-231 cells in full confinement donâ??t make decisions solely based on hydraulic resistance. Breast cancer cell decisions in Ψ-shaped asymmetric branches can be predicted in relatively high accuracy by the ratio of the cross-sectional areas that cells encounter within an intersection. Treatment with 125uM colchicine, which disrupts microtubules, lowers significantly the speed and persistence of cells. This, in turn, increases the number of cells which remain undecided (>4hrs to enter a branch). More importantly, in the fully confined regime, the probability of cells continuing on the same base without turning diminishes. We postulate that differences in branch cross-sectional areas reflect differences in cell membrane stretching. Consequently, mechanosensitive and stretch-activated Ca²⁺channels play a significant role in cell-decision making. When MDA-MB-231 cells are treated with GsMTx4 (pharmacological inhibitor of TRPC1, TRPC6 and PIEZO1 channels), the decision-making differences between the partially and fully confined cases are completely abrogated. Ongoing experiments include transfection of MDA-MB-231 cells with a Ca²⁺ FRET probe to determine Ca²⁺ localization at intersections. Moreover, we are elucidating the contributions of mechanosensitive and stretch-activated Ca²⁺channels (TRPC1, TRPV6, TRPM7) as well as Ca²⁺ activated K⁺ channels (SK3, BK_Ca).

Conclusion: Our data provide novel insight into the mechanism of cell decision-making under confinement. Cross-sectional area can be used as a predictor of cell decision, microtubules are key players for maintaining cell directionality, and stretch activated channels help cells sense their microenvironment before making a decision.