Part:BBa_K4932004

Main Purpose

The modular biosensor part, based on the protein designed by the Baker Lab. This composite part is made of two parts: key and cage. The key and cage do not have to be co-expressed on the same vector or even in the same organism, but it is crucial to mix the protein they yield before reading the signal. This composite part allows swapping out the small binding region on the LucCage in order to change what the system is targeting. This has been shown by the Baker lab to work for recognising and providing a positive signal for COVID spike, troponin proteins (for detecting heart attacks), HER2 and more. This system was recently named in the literature by the baker lab (Alfredo Quijano-Rubio et al, 2021). The system is incredibly valuable for iGEM teams focussing on diagnostics because all you need to do is swap out the target region in order to target a new protein.

Mechanism behind the part

There are two parts, the cage and the key. Both have a section of split luciferase coded in to them. When the cage binds to the protein of interest, a change in conformation occurs which allows a ‘lever’ to open. This means the cage is in an open conformation and the key can bind. When the key binds, the luciferase is reconstituted and a positive, illuminescent signal is given.

Our input

We used this concept to create a diagnosing tool for E. coli. We decided to focus our project on a naturally occurring family of proteins, bacteriocins. Bacteriocins are produced by many types of bacteria and they bind to the same species in order to compete against them. They have co-evolved meaning they target extremely conserved regions. Specifically we used colicins. These are specific to E. coli and can be used to test for E. coli Note that we had to reverse the sequences because colicins are encoded in reverse. The sequence published here and on benching are NOT reversed. This is because the normal sequences are more useful.

Future use in iGEM

Other iGEM teams can add a protein of their choice to this system in order to create a very specific biosensor. They can either use a protein found in nature (easier) or they can design their own using machine learning methods (rosetta, alpha fold, protein MPNN, RF diffusion) If any iGEM team is interested in this please see our guide on exactly how to achieve this in our contributions page!

If you are designing your own protein, currently (2023) it only really works well with small, alpha-helical targets.

References: Quijano-Rubio, A., Yeh, HW., Park, J. et al. De novo design of modular and tunable protein biosensors. Nature 591, 482–487 (2021). https://doi.org/10.1038/s41586-021-03258-z