Plasmid Design

We were originally going to be using fUSE5 until Dr. George P. Smith, an advisor, informed us the most current plasmid is f3-55nm, requiring a change in the plasmid model. We have also sequenced the previously unsequenced fMCS via Sanger Sequencing for other teams to reference.

scFv Design

Once the sequence was found, we had to modify the surrounding base pairs to properly align with the sequence for the pIII to allow for integration with the coat protein.

Expression Cassette Design

The cassette went through multiple design iterations to reach the final product.

Design Cycle summary:

• Original Cassette
• Design: TNFa-GSG+T2A-GrzB
• Build & Testing: via Benchling
• Problems:
• Multiple undesirable restriction cut sites.
• Learn/ Solution:
• Used silent mutations to remove the extra cut sites.
• Cassette 2
• Design: eGFP-GSG+P2A-TNFa-GSG+T2A-GrzB
• Build & Testing: via Benchling
• Problems:
• eGFP reduced downstream synthesis of TNFa and GrzB
• IDT could not synthesize the cassette
• Learn/ Solution:
• Added a modified kozak sequence
• Codon optimization
• Cassette 3
• Design: TNFa-GSG+P2A-GrzB-GSG+T2A-eGFP
• Build & Testing: via Benchling
• Problems:
• P2A linker not exportable out of the US
• Learn/ Solution:
• Use E2A linker, which is approved for export and has similar, albeit lower cleaving efficacy
• Final Cassette 4
• Design: TNFa-GSG+T2A-GrzB-GSG+E2A-eGFP
• Build & Testing: via Benchling

The sequence of our expression cassette went through many phases of the engineering cycle before reaching our final version. Our initial design phase consisted of researching and deciding on the genes of choice for expression. Once we had settled on TNFa and Granzyme B, we looked at ways of linking the two genes, and decided to use a GSG-T2A self cleaving peptide, which we had used the year before. We then built up the sequence in Benchling to model it, using BamHI restriction enzymes for ligation into the pAAV plasmid. After the first virtual digest, we ended up with multiple different fragments, resulting from BamHI sequences scattered throughout the cassette. These were then removed by modifying individual base pairs to remove the sites while retaining the correct amino acid sequence. The next virtual digest resulted in the desired fragment, allowing for the ligation into pAAV.

While doing work on the ligation of the pAAV into the f3-55nm-MCS phage vector, it was realized that based off of our insertion into pAAV, that we would get limited if any production of our cassette due to its location after eGFP and without a Kozak sequence. We then designed a Kozak sequence that would fit our requirements, and looked for another 2A self-cleaving peptide to use in addition to T2A, which we found in P2A. We then built up the new sequence, adding eGFP and GSG-P2A before the original cassette, taking care to ensure there were no additional BamHI sites again. We additionally added the Kozak sequence after the BamHI site and before the eGFP. The virtual digest worked properly, allowing for the removal of the old eGFP sequence from pAAV, and reintroducing our cassette in including a new eGFP sequence.

Our next problem arose during the ordering of our cassette from IDT, where we learned that the sequence was too difficult to construct. We hadn’t anticipated this, and decided to do another review of the cassette before ordering again. After some additional reading on the cleavage efficacy, we decided to place eGFP as the last gene in the sequence as it’s only used for observation instead of the primary goal of inducing apoptosis. This also necessitated designing a new Kozak sequence, as the first gene being TNFa required a slight modification based on its amino acid sequence. This was then assembled again in Benchling, modifying as little as possible. Before ordering from IDT, we ran their codon optimization program for expression in humans to increase efficiency, as well as to remove as many construction problems on their side so we could order.

While we didn’t have any problems ordering the sequence initially, we later learned that P2A is illegal for export out of the USA, canceling our order. While frustrating, it did cause us to go back to the drawing board, and find a new 2A self-cleaving peptide, which we created as a new part in GSG-E2A, which is legal for export out of the USA. We then replaced the P2A sequence with the E2A, and after verifying it didn’t introduce any BamHI sites, we successfully ordered and received the final expression cassette.