A Novel Stochastic Simulation Approach Enables Exploration Of Mechanisms To Regulate Polarization Dynamics

Cells polarize in the direction of chemical gradients to orient growth or migration. When budding yeast cells are exposed to a source of mating pheromone, polarity factors assemble into highly dynamic clusters at the plasma membrane, sometimes but not always oriented up the pheromone gradient. The clusters assemble, disassemble, and move between different regions of the cell membrane before eventually forming a more concentrated cluster stably oriented towards the pheromone source. Pathways to generate and stabilize polarity factor clusters have been identified, but clustering dynamics are not well understood. Molecular fluctuations are likely to contribute to clustering dynamics. To gain insight into how they do so, we performed efficient stochastic simulations within the reaction-diffusion master equation framework. While less accurate than particle-based simulations, this approach has been improved by recently derived scale-dependent mesoscopic association rates. A remaining problem is that such mesoscopic rates perform poorly at the high local molecular concentrations typical of polarity systems. We therefore derived and tested novel mesoscopic rates that improve the accuracy of such simulations. We observed that although some existing polarity models failed to produce highly dynamic signaling clusters, a single additional reaction, proposed in another model, enabled dynamic clustering behavior. We investigate which reactions affect cluster dynamics, and validate our predictions using a more realistic particle-based model in a 3D geometry. Overall, our novel stochastic simulation approach allowed us to efficiently explore possible mechanisms by which cells may control polarization dynamics important for gradient detection.