Team:OUC-China/Demonstrate
1. Introduction
This year, our project focuses on a standardized design principle for design of modular riboswitch, which can be easily applied by synthetic biologists and future teams. We researched several existing riboswitches, and found that the application of riboswitches was limited by the disadvantages like its context dependent performance and the limited selection range in dynamics. Clearly, the lack of design principles based on engineering strategy for highly modular riboswitch devices is the root cause of the above problems.
We developed the design principles based on a newly emerging engineering strategy for modular riboswitch known as ‘Ribo-attenuator’. By decoupling the subunits of the ‘Ribo-attenuator’ and finely running experiments and modeling on each part like the ‘Stabilizer’, ‘Tuner’, and asRNA, we got a complete design principle and reliable design calculation tools to develop a set of various modular riboswitches - RiboLegos.
The RiboLego consists of the original riboswitch, rationally designed Stabilizer and Tuner. Stabilizer can protect the structure of riboswitch from interference while Tuner can avoid the N-terminal fused non-sense peptide on desired protein. We confirmed the powerful ability of our design rules on different riboswitches including three kinetic switches: Adda riboswitch, Btub riboswitch, cobalamin biosensor, and one thermodynamic switch: FourU riboswitch. What's more, three different kinds of GOI is used (sfGFP, YFP, and mRFP1), and the good results showed the high universality of our design principles.
We believe that we have met gold medal requirement, and we have proved that our system could work under real world conditions.
2. Proof the function of ‘Ribo-attenuator’
First, we successfully demonstrated that the ‘Stabilizer’ restored the normal function of different riboswitches while Tuner tackled the problem of inclusion body generated by the Stabilizer. By fluorescence microscopy, we can clearly observe that the ‘Stabilizer’ + ‘Tuner’ is capable of making GOI express normally without inclusion body formation (Figure 1 and 2).
Figure 1: The fluorescence images show E. coli with different circuits under 2-AP induction to activate the Adda riboswitch. Left: The strain with circuit including Adda riboswitch and the reporter gene (sfgfp). Middle: The strain with circuit including Adda riboswitch and the ‘Stabilizer’. Right: The strain with circuit including Adda riboswitch, the ‘Stabilizer’ and the ‘Tuner’.
Figure 2: The fluorescence images show E. coli under VB12 induction. Left: The strain with circuit including Btub riboswitch and the reporter gene (sfgfp). Middle: The strain with circuit including Btub riboswitch and the ‘Stabilizer’. Right: The strain with circuit including Btub riboswitch, the ‘Stabilizer’ and the ‘Tuner’.
3. Proof of the function of a modular riboswitch toolkit
We made a modular riboswitch toolkit with various riboswitches which were all rationally designed. The riboswitches could provide candidates with different dynamic behaviors. The core strategy to achieve this is employing the ‘Tuner’ as the engineering target. The function of the ‘Tuner’ was successfully modeled by a thermodynamic approach.
By using a series of constructs with various model-based designed ‘Tuners’, we then diversified the response behaviors of modular riboswitches. Five different Tuners were introduced at the downstream of the Adda riboswitch and Stabilizer. Tuners were able to shift the system’s output range for responding to 2-aminopurine in a manner that correlated with the structural strength of the Tuners (Figure 3).
Figure 3: The results of modular Adda riboswitches under different concentrations of 2-AP by microplate reader. Five different Tuners are introduced at the downstream of the Adda riboswitch and Stabilizer respectively.
We collaborated with four teams which helped us prove the results of Tuner A through experiments finished by their research teams. The results show that our system is feasible and stable(Figure 4)! You can get more data on the collaboration page.
Figure 4: The iGEMers helped us to verify the repeatability of the modular Adda riboswitch.The effect of modular riboswitch with different concentrations of 2-AP. Fluorescence/RFU values of the groups of 2-AP (ligand) added with different concentrations (0, 50, 150 μM/ml) are gained after 8 hours' induction.
As we have showed in the section 2, we achieved above engineering achievements on another riboswitch to demonstrate the universal applicability of our design principle. The adenosylcobalamin-triggered translation-repressing Btub riboswitch was employed. We were able to show that the rationally designed ‘Tuners’ could also provide the similar engineering effect on the Btub riboswitch successfully (Figure 5).
In order to reduce the metabolic burden of cells, we also created Tuner S containing SsrA degradation tag, which could degrade the non-sense peptide translated from the Stabilizer.
Figure 5: The fluorescence intensity of sfGFP collected by microplate reader during the entire cultivation period. By using three different Tuners, we could tune the response of the Btub riboswitch. Error bars represent standard deviation of four biological replicates.
We also tested our system working by replacing the CDS of sfGFP with that of YFP which were introduced at the downstream of the Adda riboswitch, Stabilizer and Tuner A. The result shows that, with a new reporter, the modular Adda riboswitch still works well to sense its ligand (Figure 6).
Figure 6: The results by microplate reader. The emission of YFP was measured at a wavelength of 527nm when excited at 514nm. Error bars represent standard deviation of three biological replicates. Data was collected when steady state is reached (at least two consecutive subsequent data points do not increase fluorescence).
4. Selecting the appropriate length of Stabilizer
Guided by math modeling, we determined that the Stabilizer length of Adda and Btub was 150bp. We proved the effectiveness of our software by our wet experiments. We tested different versions of the Stabilizer with various lengths. The result shows that the 9bp and 21bp versions worked as unavailable Stabilizers, but the 81bp and 129bp versions worked as good Stabilizers for Adda. The results showed that the suitable length of Stabilizer is predictable (Figure 7).
Figure 7: The fluorescence intensity of sfGFP by microplate reader during the entire cultivation period. By using four different Stabilizers, we could prove that our software was effective. 9bp and 21bp was too short to stabilize the structure of Adda riboswitch, leading the failure of response to ligand.
5. Improvement- resurrecting defective part by our riboswitch design method
Using our design principle of modular riboswitch, we were successfully able to improve the function of a cobalamin biosensor created by Paris_Bettencourt team in 2015. They used a riboswitch whose ligand is vitamin B12 to express mRFP1. They inserted the first 30bp of the natural downstream gene of the cobalamin biosensor as a ‘Stabilizer’. By confocal microscopy, no fluorescence was observed. This may because the length of Stabilizer is not long enough or a ‘Tuner’ needs to be employed to insulate the translated peptide from the ‘Stabilizer’ with the mRFP1. By introducing rationally designed ‘Stabilizer’ and Tuner A, we constructed an improved cobalamin riboswitch, which can restore his function and express mRFP1 normally (Figure 8 and 9).
Figure 8: The results by confocal microscopy, which indicates that our principle can improve cobalamin biosensor successfully. It's obvious that the modular cobalamin riboswitch can express mRFP1.
Figure 9: The fluorescence intensity of mRFP1 by microplate reader during the entire cultivation period. We measured part BBa_K1678007 designed by Paris_Bettencourt in 2015 and the improved circuit designed by us. As shown that, by introducing Tuner A, modular cobalamin biosensor was capable of expressing mRFP1 normally in response to different concentrations of VB12.
6. RiboLegos design on the thermodynamic switch
Riboswitches can further be classified into thermodynamic and kinetic switches. We explored whether our design principles could be applied to thermodynamic riboswitches. Using Four U, whose temperature threshold is 37℃, we can successfully express sfGFP in 37℃ and 42℃. In this circuit, the first 81bp of mRFP1 was selected as Stabilizer because Four U can control the expression of mRFP1 normally in previous research. And the Tuner A was used. The result shows that the RiboLegos design principle can be applied to the thermodynamic riboswitches perfectly(Figure 10)!
Figure 10: The results of microplate reader show the working effect of the modular Four U element in different temperatures.
7. Control the on-off state by antisense RNA – a strategy for real application scenarios
According to above results, we have demonstrated that the RiboLegos are able to overcome many of the issues preventing widespread usage of riboswitches. However, one problem was how to shut down the function of riboswitch under ligand existing condition. Because in many real application scenarios, it is inconvenient to replace the culture medium.
After constructing modular riboswitches, we have successfully designed antisense RNA to solve above problem and control the on-off state of the riboswitch in real time. The good results demonstrated our effective approach(Figure 11).
Figure 11: The heat map generated from microplate reader data reflecting the change of fluorescence intensities with and without IPTG. Using our IPTG inducible antisense RNA, we could control the on-off state of Adda and Btub riboswitch.
8. Electrophoresis result of expression vector
After plasmid construction, we proved them by gel electrophoresis. The results are shown below(Figure 12).
Figure 12: The results of gel electrophoresis. From left to right respectively: plasmids about modular Adda riboswitches containing five kinds of Tuners and modular Btub riboswitches containing Tuner A and E.
9. Summary
We believe that we have met this medal requirement because we have successfully demonstrated that our design principle could expand riboswitch function. Our system could work under realistic conditions. Please see our other pages for more inspiration and results. Additionally, see our medal requirements for information on how we met our gold medal requirements.
REFERENCE
[1] Folliard T , Mertins B , Steel H , et al. Ribo-attenuators: novel elements for reliable and modular riboswitch engineering[J]. Scientific Reports, 2017, 7(1):4599.
[2] Caron M P , Bastet L , Lussier A , et al. Dual-acting riboswitch control of translation initiation and mRNA decay[J]. Proceedings of the National Academy of Sciences, 2012, 109(50):E3444-E3453.
[3] Bandyra K , Said N , Pfeiffer V , et al. The Seed Region of a Small RNA Drives the Controlled Destruction of the Target mRNA by the Endoribonuclease RNase E[J]. Molecular Cell, 2012, 47(6):943-953.
[4] Serganov A , Yuan Y R , Pikovskaya O , et al. Structural Basis for Discriminative Regulation of Gene Expression by Adenine- and Guanine-Sensing mRNAs[J]. Chemistry & Biology (Cambridge), 2004, 11(12):1729-1741.
[5] Na D , Yoo S M , Chung H , et al. Metabolic engineering of Escherichia coli using synthetic small regulatory RNAs[J]. Nature Biotechnology, 2013, 31(2):170-174.
[6] Ross K , Samantha H , Kate Y , et al. Rationalizing context-dependent performance of dynamic RNA regulatory devices[J]. ACS Synthetic Biology, 2018:acssynbio.8b00041-.
