Engineering

Our team set out to create a diagnostic system for medically relevant proteins of interest in blood. To create the this, we wanted to build on previous work done in our lab under Dr. Salis. The backbone of our project originated from work done by Grace Vezeau, a Salis Lab graduate. Her paper validated the De Novo DNA Riboswitch Calculator for predicting effective riboswitch sequences that either express or repress a reporter, as shown through fluorescent characterization^{1}.

The other significant piece to our concept stemmed from Keith Pardee’s work with genetic circuits in cell-free systems. Pardee explores how to make low cost diagnostics more practical through the use of glucometers. The experiment demonstrated the effectiveness of the trehalase enzyme in a glucose-based reporter system. Trehalase was shown to cleave trehaslose and produce glucose in a cell-free system, which was then measured with a glucometer^[2]. This work was crucial to the structure of our idea of connecting our modular riboswitches with trehalase expression.

Based on the research we conducted, our team wanted to create a diagnostic system not through the measurement of mRFP fluorescent protein but through the detection of glucose through a glucometer in a cell-free system. We projected that our two biggests constraints would be the amount of myTXTL cell-free expression kits we had and the time we had to complete all the tests.

With constraints in mind, we designed our final cell-free system to include the following:

In theory, the protein of interest would bind to the aptamer region on the modified PFTV1 plasmid, which would create a conformational change, exposing the ribosome binding site (RBS). Exposing the RBS would allow the synthesis of trehalase. Finally, the produced trehalase would cleave the trehalose in the system and produce glucose molecules. The glucose would then be detectable through a instant glucometer reading.

However, we knew that to get to our final cell-free system design, we would have to create a multistep process to avoid changing more than one variable at a time. We decided for the first round of testing that we would measure the expression of mRFP1 fluorescent protein in a cell-free system with the same protocol followed in Grace Vezeau’s paper.

After creating a framework, our team needed to select the proteins of interest we would be testing. Our team utilized this previous research to set the parameters we would look for when picking our own proteins to detect. We researched proteins with previously tested RNA aptamers, optimally 70 base pairs or less, found in the blood and related to medical conditions. Using that criteria, our team decided to design riboswitches for six different proteins: basic fibroblast growth factor (bFGF), vascular endothelial growth factor (VEGF), bovine thrombin, thyroxine and two previously used proteins in Vezeau’s paper, Interleukin-32γ (IL-32γ), and monomeric C-reactive protein (mCRP). The selection of the last two proteins would act as a control group because no variables were changed from her paper.

We utilized the De Novo DNA Riboswitch Calculator to build our riboswitches. This calculator is a physics based model that uses statistical thermodynamics to predict DNA sequences and their functionality as a riboswitch that activates gene expression. From the list of DNA sequences given by the De Novo Riboswitch Calculator for each protein of interest our team picked five different sequences with varying activation ratios. Once we picked our riboswitches, we ordered all 30 off of IDT as oligonucleotides and PCR assemblies, depending on their size. Our next step was to insert them into our pFTV1 vector and clone them in E. coli.

Step 1:
Once we received our oligos and gBlocks, we performed double digestions with XbaI and SacI and ligated the various riboswitches into our vector.

In the first round of testing our cell-free system included:

During this first test, the fluorescence of mRFP was analyzed in the TECAN Spark plate reader.

In this round of testing, we learned that experiments will rarely go expected to plan. Our plasmids did not produce strong fluorescent signals, and we were unsure as to why. These results forced us to go back to our original design and pinpoint a weak spot.

Additionally, due to limited supply of myTXTL reactions, we did not run the samples in triplicate. With the given timeline and resources available, our team needed to reduce our remaining testing and focus on the most promising riboswitches, which were for bFGF (designs 2 and 3) and mCRP (designs 1-5).

Unsure of what to do, we brainstormed what might be causing the problem. As we lacked the materials to test all 30 riboswitches with J23100 in triplicate, we looked to another approach. Our team hypothesized that the J23100 promoter in our plasmid was not strong enough and that a different promoter like T7 would exhibit stronger fluorescence. The goal of our system was also to work quickly, so we believed the T7 promoter would help this.

To test that theory, we began the cloning process again to cut out J23100 and ligate in T7. All 30 riboswitches were cloned with T7 promoter.

In the second round of testing we ran the samples in triplicate with the proper controls. However, even when testing the most promising riboswitches in triplicate, the results showed no significant fluorescence. Interestingly, our consitutive no-riboswitch control with J23100 expressed fluorescence (as expected), but our no-ribowitch control with T7 failed to fluoresce.

Our team is taking these results into consideration as we continue with the testing. Once we see a strong fluorescence with the pFTV1_mRFP plasmid with T7, we will cut our the mRFP1 and add in Tre37A.

As we are still engaged in the Learn phase of this cycle, we will step back and look more closely at each part of our testing. The T7 positive control’s failure to fluoresce indicates that something is inherently wrong with our testing. We will return to the Design phase for both the J23100 and T7 riboswitches, reassess our protocols, and rety our tests with new improvements.