Experiments & Protocols

Toehold experiments

In order to show that toehold switches could be developed to be used in our diagnostic test, we first had to show that the toehold can be expressed cell-free, and that it responds to the correct trigger RNA. To do this we have tested two toehold designs; Green_FET1, which was designed and tested by Green et al., and our own toehold design, code-named Zeus. Before testing these designs in the lab, which is expensive due to the high cost of commercial cell-free kits, we decided to test the toeholds in silico using the online software package NUPACK.The use of the control toehold, Green_FET1, which has previously been reported to work as expected experimentally, allows for comparison of in silico between our toehold (Zeus), and the data of a function toehold. This can then be used to allow for an informed decision as to whether the toehold will work practically.

In silico testing:

The use of in silico testing in synthetic biology, and indeed science as a whole, is something which is not to be overlooked. While practical experimentation is still required and an important aspect of any project, the correct use of this kind of testing and modelling can allow for better informed experiments and more efficient testing, which can be extremely important when testing may take long time, or, as in this project's case, become expensive if many experiments are run.

As our diagnostic test would be carried out in the field, where the temperature can not be easily controlled, it was important to determine the integrity of the toehold structures at different temperatures. In order to do this the analysis function of NUPACK was used to perform a melt between 0C and 70C on both toeholds. Briefly, the Green_FET1 toehold structure was shown to be correct from 25C, and the Zeus toehold was shown to be correct from 15C. A full discussion of the results from the NUPACK melt can be found on the results page.

As well as being able to analyse the toehold's structures at different temperatures, the formation of toehold-trigger complex formations can be determined using free energies. As can be seen, as temperature increases, the ΔG of complex formation becomes less negative, meaning that toehold-trigger binding becomes less. Again, full discussion of this can be found on the results page.

The main aim of this testing was to determine the most suitable toehold design to test in the lab, which therefore decreases resources and time spent testing toeholds which would not have been/had very little function.

As well as using purely NUPACK to advise practical experimentation, two main models of the toehold/diagnostic system was created using MatLab. These are a simulation of the system, and a numerical model which could be used to determine optimal conditions for the test. Further explanation and discussion of these models can be found here.

Toehold ordering and construction:

Once our toehold (Zeus) had been designed and passed in silico testing, it and the control toehold (Green_FET1) needed to be ordered. In the paper where Green_FET1 had been experimentally validated, the transcription of it was under the control of a T7 promoter, however we wished to express our toehold under the control of a constitutive promoter, namely J23100. As we wished to use Green_FET1 not only as a control, but also as a comparison in functionality to Zeus, we decided to order Green_FET1 under the control of both a T7 and J23100 promoter.

The toeholds were constructed as gBlocks containing overhangs for the standard pSB1C3 plasmid so as to allow construction of the toehold into the plasmid. The gBlocks were ordered from IDT (Integrated DNA Technology), and constructed using HiFi assembly from NEB (New England Biolabs). The constructs were then transformed, grown in liquid cultures, minipreped, and sent for sequencing. The DNA concentration was also determined by Qubit for use in later toehold testing.

Protocols used:

Links to lab diary:

Toehold dynamic range:

Once the toeholds had been tested in silico, DNA encoding for these could be ordered up as gBlocks and constructed into pSB1C3 plasmids to practically validate in the lab. In order to begin characterising these