Team:Hamburg/Results
The aim of this project was to enable a transformation of three plasmids into one bacterium with only one antibiotic resistance instead of three. The idea is based on the paper “Complex cellular logic computation using ribocomputing devices” by Green et al. (Nature, 2017). In our project we focused on an AND logic, which means that two or more so called trigger need to be present at a time to enable gene expression. The strategy is based on toehold switches that are forming hairpin structures, which hide the ribosome binding site and the start codon. Those so called gates can be opened by the triggers due to a higher binding affinity.
To read more about the details visit our Description and Design pages.
We characterized the activity of the promoters we used in our project, the two constitutive promoters from the Anderson Family, BBa_J23100 and BBa_J23102. We measured the expression of GFP downstream of the promoters.
The parts were cloned into E. coli DH5α competent cells. The fluorescence measurements were performed using flow cytometry and plate reader measurements. The results of the flow cytometry measurement are shown in figure 1.
The results confirm the different strengths of the constitutive promoters. BBa_J23100 expresses a stronger fluorescence signal than BBa_J23102. As a negative control we used non-competent cells to measure background emission.
The data used to calibrate the plate reader measurements are shown in figure 2. The Excel file used for the calibration can be viewed here.
The results of our measurement of GFP on the plate reader are shown in figure 3.
The corresponding Excel file can be found here.
The results confirm the previous results from the FACS measurement. The promoter BBa_J23100 expresses a stronger fluorescence signal when compared to BBa_J23102.
We calculated the strengths of these promoters relative to the control: BBa_J23100: 2.28 BBa_J23102: 1.84 The relative strength of BBa_J23102 when compared to BBa_J23100 is 0.8. In comparison, the values in the registry is 0.856. We tested the gate plasmid with our two trigger plasmids by performing simultaneous transformations with these plasmids. After transformation we selected the bacteria by plating on LB-agar plates with chloramphenicol. We compared the results with positive and negative controls, using the empty pSB1C3 backbone and pSB1A3 respectively.
The result of the transformation with all three plasmids after selection with chloramphenicol is shown in figure 4. Colonies grew on chloramphenicol plates after transformation with all three plasmids. This shows that our RNA based logic circuit allows the expression of the antibiotic resistance. No colonies grew on the plates with the negative control. Colonies also grew on the plates with the positive control.
To further test the trigger and the gate, we performed transformations with each plasmid individually. The results are shown in figure 5.
Unfortunately colonies also grew on the plates with just the individual transformation of the gate plasmid, though the number of colonies was much lower when compared to the transformation with all three plasmids. We sequenced the gate plasmid to confirm correct assembly. The results showed no mutation in the gate sequence. We concluded that the gate is showing signs of leakage, expressing the antibiotic resistance even in the absence of triggers. This basal expression was rather strong due to the strong promoter we used for our experiments. The data clearly shows that the gate is leaking, but when compared to the triple transformation (red) and the control (green) the number of colonies when transformed with only the gate plasmid was significantly lower. Additionally the number of colonies when transformed with all three plasmids was larger than when transformed with the positive control.
For cells transformed with three plasmids an impaired growth rate can be observed.
We could not show that the transformation of our three plasmids was less harmful to the bacteria than the transformation with three different antibiotics, instead it shows comparable levels of stress.
Several colonies grew on the plate with bacteria transformed with the modified chloramphenicol acetyl transferase. The negative control showed no cell growth while on the plate with the positive control a cell turf has grown. This confirms that the activity of the chloramphenicol acetyl transferase was not significantly impaired by the additional amino acids.Results
Promoter characterization
Additionally, we sequenced the plasmids used in our measurement. The reported sequence matched the Anderson promoter BBa_J23100. The reported sequence for the promoter BBa_J23102 showed three point mutations, which could have impacted the strength of the promoter. The following mutations occurred: 6A>G, 22G>A, 27T>A. This explains the decrease in the relative promoter strength of BBa_J23100 compared to BBa_J23102. Gate and Trigger
To measure the gates leakage, we observed cells transformed with all three plasmids on plates with different chloramphenicol concentrations and compared them to cells transformed with only the gate plasmid. As a positive control we used the pSB1C3 backbone. The Excel file used for the calibration can be viewed here. The number of colonies per plate after twelve hours of incubation is shown in figure 6.
We also used a plate reader to measure the growth rate of cells that were transformed with three different plasmids (pSB1A3, pSB1C3, pSB1K3 or BBa_K2970003, BBa_K2970004, BBa_K2970006) and compared it to cells without any plasmids. The results are shown in figure 7.
Due to the linker between the gate and our gene of interest (chloramphenicol acetyl transferase) additional bases were attached to the gene which might affect the functionality of the protein as it contained an additional 19 amino acids. To test the genes activity we inserted the sequence containing the additional bases into pSB1A3, transformed it into bacteria and plated the cells on chloramphenicol plates. The results of this experiment are shown in figure 8.