Title: Self-organisation of bidirectional microtubule networks is robust: a mathematical framework and validation in Drosophila.
You and I are but a collection of trillions of cells. Why are we not falling apart? Consider this question on an inter-cellular level: to hold the cells together, the adhesion protein E-cadherin has to be delivered to cell boundaries. This delivery is done along the microtubule cytoskeleton via stochastically moving molecular motors. An outstanding question is how, given a broad range of possible dynamic behaviours of individual microtubules, functional microtubule networks are established and maintained in living cells. Here we analyse microtubule self-organisation in epithelia via mathematical modelling and genetic manipulations of Drosophila embryos. Our stochastic simulations determine that self-organisation of the microtubule network is robust in a wide parameter range: microtubule alignment is not affected by the details of their dynamic instability or interactions. We confirm this using genetic manipulations of Drosophila embryonic epithelial cells and develop a minimal model which suggests that the origin of robustness is the generic separation of scales in microtubule dynamics: polymerisation and depolymerisation predominating over catastrophe and rescue. We demonstrate that the microtubule network self-organisation depends only on cell geometry and the distribution of microtubule minus-ends.