High-resolution mapping of the period landscape reveals polymorphism in cell cycle frequency tuning

Many biological oscillators exhibit widely tunable frequency in adapting to environmental changes. Although theoretical studies have proposed positive feedback as a mechanism underlying an oscillator's large tunability, there have been no experiments to test it. Here, applying droplet microfluidics, we created a population of synthetic cells, each containing a cell-cycle oscillator and varying concentrations of cyclin B mRNAs for speed-tuning and positive-feedback inhibitors for modulating network interactions, allowing a continuous mapping of the cell-cycle period landscape in response to network perturbation. We found that although the cell cycle's high tunability to cyclin B can reduce with Wee1 inhibition, the reduction is not as great as theoretically predicted, and another positive-feedback regulator, PP2A, may provide additional machinery to ensure the robustness of cell cycle period tunability. Remarkably, we discovered polymorphic responses of cell cycles to the PP2A inhibition. Droplet cells display a monomodal distribution of oscillations peaking at either low or high PP2A activity or a bimodal distribution with both low and high PP2A peaks. We explain such polymorphism by a model of two interlinked bistable switches of Cdk1 and PP2A where cell cycles exhibit two different oscillatory modes in the absence or presence of PP2A bistability.