Linker

Introduction

In order to design a functional sensor we needed to be able to link the detector to the reporter. We considered a direct linkage between the reporter and the detector which we decided to pursue for the beta-lactamase as a reporter. We have also built and submitted a construct with the a direct linkage between the TALEs and the ferritin. However, due to the massive size of the finished product (~1300 kDa) we needed another system, a system which allows in vitro assembly of the reporter and the detector. This is why we looked into the use of E and K coiled-coils in our system. The addition of the coiled-coils gives us flexibility in our system to disassemble the specifically engineered detector from the reporter in order to switch in another detector for a different application. In other words, the coiled-coils add re-usability to the system making it a great platform technology. In addition, ferritin serves as a great scaffolding protein which allows us to stick the different detectors on it. We can also scale up the reporter response by adding reporters on ferritin, depending on the requirements of the system where the sensor is applied. Therefore, the ability to scale up and down allows us to incorporate informed design in our platform technology application.

What are E/K Coils?

E/K coils are synthetic coiled-coil domains designed specifically to bind to each other with high affinity and specificity (Litowski and Hodges, 2002) (Figure 1). They are composed of a heptad repeat that forms a coil structures that are able to interact with each other. These coils are able to interact with each other in an anti-parallel fashion that makes them useful for applications such as peptide capture, protein purification and in biosensors. For our project we decided to make use of the IAAL E3/K3 coils (BBa_K1189010, BBa_K1189011) due to the balance they offer between affinity and specificity (Table 1, Figure 2). These parts were already in the registry, however the DNA was never received, so we built, sequenced and re-submitted them.

How do These Coils Work?

These E3/K3 coils are able to form heterodimers due to the hydrophobic residues contained within the heptad repeat. In our case these are isoleucine and leucine residues. Designated by empty arrows in the helical wheel diagram below (Figure 3) these residues form the core of the binding domain of the coils. In order to prevent the homodimerization of these coils charged residues are included in the design. The electrostatic interactions between glutamic acid and lysine residues prevent an E-coil from binding with an E-coil for example. We selected the use of E3/K3 coiled-coils over other synthetic E/K-coils as the isoleucine residue present shows a significant increase in the heterodimer over valine found in other coils. The alpha-helical propensity of the residues outside of the core interacting residues is also increased by utilizing an alanine residue instead of the serine residue found in ISAL and VSAL E/K-coils. This selection maximizes the stability and specificity of the coils used in our system.

What did we construct?

The first construct we created involved attaching the previously mentioned E-coil biobrick to the fusion of heavy and light chains of ferritin. (BBa_K1189018). This construct would assemble to form a 12-subunit nanoparticle with E-coils on the outside that would each bind a protein containing the respective K-coil. This means we can simply attach the K-coil to a protein of choice, such as a TALE in order to bind it to the ferritin scaffold (BBa_K1189018) Therefore we constructed the respective K-coil-TALE fusion constructs, BBa_K1189029 and BBa_K1189030. The ideal result of this system is that BBa_K1189029 and BBa_K1189030 will be bound to ferritin, resulting in a protein complex containing 12 TALE proteins bound to the ferritin scaffold through E/K-coils (Figure 8). This system then allows the proteins to properly fold while at the same time allowing the ferritin nanoparticle to assemble without interference.

So do These Coils Actually Bind?

After putting in gratuitous effort to build parts containing these coils and successfully purifying these proteins it was necessary to determine if the coils could actually bind to each other. For this purpose we prepared a preliminary qualitative blot assay that we could use as a foundation to gather more quantitative data in other assays in the future. In the case of the coils we were interested to see if the K-coil fused to TALE proteins (BBa_K1189029, BBa_K1189030) could bind to the E-coil found on one of our Prussian blue ferritin constructs (BBa_K1189018). To complete this task we placed the TALE on the membrane, washed and blocked the membrane. The ferritin protein with the complimentary coil was then added to the membrane. If this coil successfully binds to the other coil then the ferritin will not be washed off during the next wash step. We can then see if Prussian blue ferritin is bound by adding a TMB substrate solution that will cause a colour change. To this extent we saw a blue ring in this trial indicating a positive result. This suggests that our coils are actually binding in an in vitro system.

Another interesting element of this assay is why we used two variants of the TALE K-coil negative control. A blue ring on our TALE negative control confirmed our fear that during the second protein application and wash step that some of the ferritin with coil proteins would drift over and bind to the TALE K-coils on the nitrocellulose. This did not occur for our separate negative control (Figure 8). Another element of our assay was testing the binding of TALEs to DNA. This did not yield any results and will be investigated in the future.

Ferritin Scaffold

The ferritin scaffold we chose to use, BBa_K1189018, allows us the ability to modulate our system according to what we desire. We determined the ferritin-coiled-coil system could be used for more than the detection of our E.coli sequence. It can use this system to detect other sequences of DNA by exploiting the ability of the TALE system to be easily engineered and modified, along with the interchangeability of the coiled-coil system. Thus by attaching K-coils to TALEs that target alternative sequences, we can use this system as a platform for any variety of DNA sequences. In our system, we utilized two TALE proteins, TALE A and TALE B (BBa_K1189029, BBa_K1189030). These two parts were both created with the K-coil, allowing us to use the same ferritin scaffold (BBa_K1189018) to anchor our proteins. This proved to be incredbily useful in binding the TALE proteins to nitrocellulose, a process that typically requires chemical modifications and specialized proteins in order to ensure the active site of these proteins are functional. Instead, we could interchange TALE A for TALE B on the ferritin scaffold due to our coiled-coil system. The protein complex could then simply be blotted onto the nitrocellulose, ensuring at minimum, the upwards-facing TALEs will be able to bind DNA. Furthermore, we can expand this system towards proteins and other molecules as well by replacing the TALEs with antibodies or other binding proteins. In addition to the modularity of our system, ferritin has been found to be highly accepting of protein fusions, while also being found to stabilize proteins as well. This means the attachment of proteins to ferritin is not a large worry when considering use of it in a fusion system. Our ferritin scaffold complex proves to be an easily modifiable system that also stablises fused proteins.

Another aspect of the ferritin nanoparticle that can be exploited is its composition, being composed of 24 subunits, as well as its potential as a novel reporter. Ferritin is typically composed of two chains, heavy and light. In our system, each one of these subunits can be bound to a reporter, effectively amplifying our catalytic activity by 24 due to the increased number of bound reporter proteins. But this activity can be scaled down through previously reported heavy-light chain fusions (Huh and Kim, 2003). Rather than 24 subunits the nanoparticle would be reduced to 12 subunits, as in our system, and thus the number of reporter proteins would decrease accordingly. We can scale this down even further due to our ability to transform the iron core of the protein into a catalytic active substance called Prussian Blue (Zhang et al., 2013). Thus, instead of multiple bound reporter proteins we can further scale down the output of our system to one with essentially one reporter. The ability of our system to scale up and down according to the need of detection means this system is applicable in areas ranging from low sensitivity to high sensitivity.