Modular and Versatile Trans‐Encoded Genetic Switches

Abstract Current bacterial RNA switches suffer from lack of versatile inputs and are difficult to engineer. We present versatile and modular RNA switches that are trans‐encoded and based on tRNA‐mimicking structures (TMSs). These switches provide a high degree of freedom for reengineering and can thus be designed to accept a wide range of inputs, including RNA, small molecules, and proteins. This powerful approach enables control of the translation of protein expression from plasmid and genome DNA.

Growth of the E. coli cells and transcription of the TMS switches. To verify the function of the TMS switches, we used E. coli BL21(DE3) cells. Two separate tubes of the BL21(DE3) cells were used to execute the transformation process. In the first tube of BL21(DE3) cells, the TMS switch plasmid and the reporter plasmid were co-transformed. The anti-repressor RNA plasmid, which contains both the gene for the anti-repressor and TMS switch, was co-transformed with the reporter plasmid in the second tube of the cells. The transformation was carried out by electroporation (using MicroPulser Electroporator from BIO-RAD with 2.5 kV for 5 milliseconds). We spread the transformed cells on antibiotic plates, single colonies were picked and grown in LB media shaking at 200 rpm overnight at 37°C in the presence of antibiotics (ampicillin 50 µg/ml and chloramphenicol 25 µg/ml). The starter culture was then diluted by 200 times in LB medium containing antibiotics and four separate cultures were prepared -the first culture was to express the reporter gene only, the second culture was to express the switch and the reporter gene, the third culture was to express the switch, the reporter and the anti-repressor RNA and the fourth culture was used as a negative control. All the samples were grown at 37°C with 200 rpm to OD600 0.4 -0.6, after which 0.1% arabinose (w/v) was added into the first culture to induce the expression of the reporter gene only. Into the 3 second culture, 1mM IPTG and 0.1 % arabinose (w/v) were added to induce the switch and the reporter.
Into the third culture, 1mM IPTG and 0.1 % arabinose (w/v) were added to induce the switch, the antirepressor and the reporter. In the fourth culture, no inducer was added. After inducing each sample at the log phase, we incubated for six hours and then the output signal from each sample was measured by flow cytometry. To control gene expression by the TMS switch with protein as an input, we used mcherry as an output signal by replacing the GFP in the reporter plasmid with mcherry protein. The mcherry protein was placed under the control of an arabinose promoter. Here the GFP was used as an input signal to control the expression of the mcherry by the TMS switch. The GFP was cloned in the same reporter plasmid under a tet promoter. To study the gene expression with the TMS switch and protein as input, all the experimental procedure was same as mentioned above except the third culture was further divided into five separate subcultures. At the log phase 1 mM IPTG, 0.1% (w/v) arabinose were added into each five sub-cultures. Along with IPTG and arabinose, anhydrotetracycline was also added into those cultures with different concentrations (0.0125-0.2 µM). The flow cytometry measurement was taken after six hours of incubation.  Next, we decided to include two 9 nts sequences of Initial Binding Elements (IBE) into the repressor domain, to provide initial binding sites between the repressor domain and the anti-repressor RNA. To attach the IBE to the repressor sequence, a stem of 8 nts length was incorporated into the structure. Finally, a stem of 4 nts length was chosen to attach the repressor domain to the respective tRNA structure.
After setting up these initial parameters, we used NUPACK software (Supplementary Ref. 5) to design the whole repressor domain and the stem that connects the repressor domain with the tRNA structure. In order to design these parts of the switch, the following algorithm was used: After executing the algorithm, the NUPACK software determined the sequences of the IBE domains, the stem that connects the repressor sequence to the IBE and the stem that connects the repressor domain to the tRNA structure.
The sequence of the IBE domains was the following: The sequence of the stem that connects the IBE to the repressor sequence was the following: The sequence of the stem that connects the repressor domain to the tRNA structure was the following: The free energy of the structural complex was calculated to be -22.18 kcal/mol. The software also allowed to obtain the average number of nucleotides (2.9 nts) in the complex that could be incorrectly paired at the equilibrium relative to the specified secondary structure (the number was evaluated over the Boltzmanweighted ensemble of secondary structures). The normalized ensemble defect was 4.4%.
We also used NUPACK software to design the different repressor domains used in the orthogonality test.
To conduct the orthogonality test, we designed six different TMS switches with the same stem sequence that connects the repressor domains to the tRNA structure. The TMS switches differ in the sequences of  Colour code: