Results

Protein Design

With the use of protein modeling systems, a visual representation our antibody switch was made. We made two possible models, each one with a unique linker designs. The linkers were modeled and designed for flexibility and strength, in order to pull apart the antibody domains when the AraC domains bind to arabinose. In the absence of arabinose, the FITC-E2 domains will reinteract and be able to bind to fluorescein.

SHORT LINKER:

LONG LINKER:

Plasmid Design

The created protein model was then designed for a vector system. Vectors were created using a pET52a vector system and specific double stranded DNA fragments (ordered from both IDT and Twist DNA). The completed vector, around roughly 7000 base pairs, was made up of nine fragments. The fragments all contained designed BsaI sites at each end, allowing for successful ligation of DNA fragments and incorporation into the pET52a plasmid via Golden Gate assembly.

Figure 1: Antibody Switch Plasmid. The places labeled as linker regions were designed to have BsaI sites between the end of the ScFv region and the AraC region so that we could insert various linker designs to determine which design worked the best to enable our switch. This plasmid construct was split into nine DNA fragments with BsaI sites unique BsaI sites between each fragment.

A control vector was also designed, which includes the fluorescein antibody ScFv domain attached to a FAB domain. The goal behind the control was to determine if attaching a second domain to the ScFv domain would inhibit the antibody from correctly binding to fluorescein, therefore the designed linkers were not used as a part of the control design.

Figure 2: Control Plasmid. The FITC-E2 + FAB domain region consists of the fluorescein binding antibody domains connected to a similar FAB domain. The designed linkers were not applied to the control because we wanted to determine if the general design of attaching a second domain to the ScFv domain would still allow the FITC-E2 domain to bind to fluorescein.

Experimental Results

Golden Gate assembly was used to assemble the DNA fragments for both the experimental and control plasmids. Using PCR primers to check the overlapping regions between the DNA fragments, we were able to determine whether or not the fragments went correctly together in relation to one another. From multiple PCR checks, we were able to determine that the more fragments you had the more mismatching that could occur between fragments. So, once we correctly assembled the three pieces that made up the plasmid backbone, we performed a PCR that amplified the backbone region and then gel extracted to backbone fragment. From here, we performed all future Golden Gates with one backbone piece, which cut down our number of DNA fragments for each reaction.

Figure 3: PCR of the plasmid backbone. The backbone was just over 5000 base pairs so this gel determined that the backbone was successfully amplified. Two trials were done to determine if the addition of DMSO would help amplify the backbone. From the two trials, we found that the addition of DMSO did help the amplification.

All assembled vectors were transformed with electrical based techniques into Neb 10 Beta cells. Given that the control plasmid had fewer DNA fragments, we began with the assembly of that design first. Successful vector formation and transformation occurred with the pET52a plasmid and designed DNA fragments for the control, as seen in the figures below.

Figure 4: Successful colony growth after transformation of the control plasmid into Neb 10 Beta cells.

Figure 5: Colony PCR on Control - determined that the plasmid backbone went into the control DNA fragment (2 piece Golden Gate assembly). We designed a 300 base PCR product that spans one of overlaps between inserts. Therefore, by obtaining the expected 300 base pair results, we knew that the control fragment and the plasmid backbone were assembled correctly in relation to each other.

Protein Identification

Having successfully assembled and transformed our control plasmid, we then transformed the plasmid into chemically competent BL21 cells, as seen in the figure below.

Figure 6: Successful colony growth after transformation of the control plasmid into BL21 cells.

Next we will be performing protein purification on these colonies and testing the control design.

Analysis

One of the unfortunate problems that we encountered with our experiment is that when we got to purification of our control protein, we discovered that our protein was incorrectly folding inside of the cytoplasm of the cell. We needed to add signal tags onto our protein so that it would sent into the periplasm of the cell, and therefore fold correctly. This mistake set our progress back because we had to order new DNA fragments with the included signal tags. Despite this mistake, we were still able to correctly assemble the control plasmid again with the signal tags included.

Future Steps

Our initial future steps would be to continue with Golden Gate experiments to successful assemble our full switch design with the two possible linker options. Once correctly assembled, we then would transform into BL21 cells before moving onto protein purification. Once purified, we will test our protein with both arabinose and fluorescein to determine if our switch model is feasible.

Our big picture future steps would potentially include humanizing the AraC domain and testing out more linker ideas.