A synthetic Escherichia coli predator-prey ecosystem

Abstract

We have constructed a synthetic ecosystem consisting of two Escherichia coli populations, which communicate bi-directionally through quorum sensing and regulate each other's gene expression and survival via engineered gene circuits. Our synthetic ecosystem resembles canonical predator-prey systems in terms of logic and dynamics. The predator cells kill the prey by inducing expression of a killer protein in the prey, while the prey rescue the predators by eliciting expression of an antidote protein in the predator. Extinction, coexistence and oscillatory dynamics of the predator and prey populations are possible depending on the operating conditions as experimentally validated by long-term culturing of the system in microchemostats. A simple mathematical model is developed to capture these system dynamics. Coherent interplay between experiments and mathematical analysis enables exploration of the dynamics of interacting populations in a predictable manner.

Figure 1

Individual growth behaviors (without interactions) of (A) predator and (B) prey cells in liquid media. For each condition, 6 ml LBK medium containing chloramphenicol and kanamycin was inoculated with a single bacterial colony and was divided into three 2 ml cultures: ‘OFF' cultures contained no inducers, ‘+IPTG' cultures contained 1 mM IPTG and ‘+IPTG+AHL' contained 1 mM IPTG and 100 nM AHL, respectively. After 20 h of incubation (light gray bars), optical densities (ODs) of these cultures were measured with a microplate reader (see Supplementary information). Error bars represent standard deviation of triplicate cultures.

Figure 2

Long-term characterization of system dynamics using the microchemostat. (A) Typical oscillatory dynamics of the system with a period of ∼180 h (IPTG=5 μM, dilution rate=0.1125 h⁻¹). (B) System dynamics for varying IPTG induction levels. The microchemostat dilution rate was 0.1125 h⁻¹. Without induction, prey cells are washed out. At increased IPTG level (IPTG⩾5 μM), oscillatory dynamics were elicited in several reactors. (C) Bifurcation diagram of the predator density versus IPTG level. The inset shows the oscillation period as a function of IPTG. Time courses of the two populations at different IPTG concentrations are illustrated.

Figure 3

Dependence of systems dynamics on dilution rate (D). (A) Experimental dynamics of predator and prey populations at different D in the microchemostat: for the pair of predator (MG1655) and prey (Top10F′). (B) For the pair of predator (Top10F′) and prey (Top10F′). (C) Bifurcation diagram of oscillatory period versus D. The inset is the oscillation period versus D. Qualitatively different systems dynamics (sustained and damped oscillations, and steady state) can be obtained by the variation of D. These experiments were carried out at IPTG=50 μM.