Figure 6. Hysteresis in rel expression.

(A) Filled triangles and squares represent the experimental data on expression of rel with increasing and decreasing (denoted by arrows) concentrations of inducer (tetracycline) respectively. (B) Hysteresis plot obtained through stochastic simulation based on the Gillespie algorithm. At a particular range of concentrations, we observed two distinct alternative steady states of rel expression depending on whether inducer concentration was going up or coming down. Figure 4C shows the GFP distributions in the stationary phase for two sets of experiments, one in which the inducer concentration is increased from a low to a specific value (indicated as “Low” in black) and the other in which the same value is reached by decreasing the inducer concentration from a high value (indicated as “High” in red). The first and third panels in which the specific inducer concentration is respectively low and high correspond to regions of monostability. The distributions with different histories, i.e., initial conditions more or less coincide. The middle panel describes a region of bistability. The distinct distributions for different histories indicate persistent memory (see SI).