Construction of a Genetic Toggle Switch in E. coli
Gardner, T.S., Cantor, C.R., and Collins, J.J. 2000. Construction of a genetic toggle switch in Escherichia coli. Nature 403: 339-342.
The earliest example of a synthetic cellular memory network is the genetic toggle switch in E. coli. This paper was published in 2000 by the lab of J. J. Collins at Boston University. Their system uses the mutual repression design that was discussed in the biological designs section of this wiki paper.
Specific Biological Design
Figure 1: Biological design of a genetic toggle switch
Figure 1 on the right shows the specific configuration of the two plasmid constructs that were used, each of which confers cellular memory. Each of the plasmids consist of two promoter/repressor pairs, one of which is the Ptrc promoter and the lacI repressor (inhibited by IPTG). P1 and R1 represent the second promoter/repressor pair that was used in conjunction with Ptrc and lacI. Plasmids with the PLs1con promoter and the temperature sensitive lambda repressor as their second promoter/repressor pair were referred to as pTAK constructs. Plasmids with the PLtetO-1 promoter and the tetR repressor (inhibited by aTC) were referred to as pIKE constructs. Multiple types of each plasmid design were constructed by modifying the ribosomal binding site strength of the repressor genes. The constructs were capable of toggling between “on” and “off” stable steady states of GFP production through exposure to a repressor-inhibitors (IPTG, 42 degrees C, or aTc). In this way, the plasmids were able to “remember” the most recent inhibitor that was in their environment
The mathematical model for this paper will not be described in detail because most of the tools for modeling memory networks have already been explained in the math modeling section. Briefly, two differential equations were created that described the rate of synthesis of each repressor as a function of the concentration of the opposite repressor, cooperativity of repression, and repressor dilution/decay. To calculate the steady states in the system, both of these equations were set equal to 0. This created two nullclines, the intersection of which represented a steady state and determined the concentration of each repressor in that steady state. This model demonstrates the need to balance promoter strength with repressor strength by tuning the strength of ribosomal binding sites in order to construct a function memory circuit. It also shows that repressor cooperativity is necessary for bistability, as was shown in the mathematical modeling section.
Measuring the Toggle Point
Figure 2: Measuring the IPTG toggle-point concentration.
The toggle-point of the plasmids was first investigated to determine the amount of input that must be detected by a cell for a toggle to occur. This was looked at specifically for IPTG. Figure 2 shows that a step-like toggle occurs at an IPTG concentration between 10-4 and 10-5 M. Based on this data, 100% of the cells should “remember” an IPTG concentration >10-4 M, while a concentration below this will cause none or a portion of the cells to toggle. Similar measurements were done for aTc concentration and lambda repressor temperature sensitivity.
Demonstration of the Toggle Switch
The pTAK and pIKE constructs were then tested for the ability to toggle between the on and off states. In Figure 3, below, the top row of boxes a and b represent the experimental results for the different plasmid types. The bottom two rows of boxes a and b are controls, which behave as expected. All of the cells were set to the on state though 6 hours of exposure to IPTG. All but one (pIKE105) of the experimental constructs remained in the on state after removal of IPTG. The promoter and repressor strength were likely not well-balanced in the pIKE105 construct that didn’t retain memory. The cells were then exposed to a second input (42 degrees C or aTc), causing them to toggle into the off state and remain off even when the input was removed. In this way, it was demonstrated that both toggle switch constructs were capable of "remembering" their most recent environmental condition and modifying GFP production based on this memory.
Figure 3: Experimental results for a) pTAK and b) pIKE plasmids.
The work that was accomplished in this paper represents the first synthetic cellular memory network to ever be constructed. The principles it presented laid the groundwork for more complex studies in cellular memory and across synthetic biology. The cell chassis that was used (E. coli) is relatively simple and allowed for accurate mathematical modeling that demonstrated the need for cooperativity and the advantage of tuning system components through modifications in ribosomal binding site strength. As a result of these advances, memory networks have been developed in multiple eukaryotic systems. Two of these eukaryotic memory systems will be discussed next.