Proof of Concept
Background

          As part of our project, we wanted to create a biosensor that could be used to easily detect binding of small molecules to an upstream riboswitch. To do this, we used RNA Mango (Dolgosheina et al., 2014) as our detection platform. Our aim was to test several small molecule aptamers binding to their ligand and measuring the fluorescent output of RNA Mango. If we were successful in building the ligand, the amount of properly folded RNA would increase, thus increasing the amount of correctly folded RNA Mango and in turn the level of fluorescence.



We produced three different small molecule RNA biosensors to test:

  1. RNA-Glucose1
  2. RNA-Scramble1
  3. RNA-Theophylline1
         The sequences of these biosensors were found in the literature and have shown to bind glucose (Yang et al., 2014)  and theophylline (Rankin et al., 2006), respectively. The RNA-Scramble sequence was to be used as a negative control, as it should not bind glucose.

We initially chose these small molecule biosensors as we thought it would be important to be able to measure the amount of glucose in your blood and respond with insulin production, if necessary. The theophylline biosensor was to act as our positive control as it has been well characterized and binds theophylline tightly in vitro with a sub micromolar affinity.



Results

          The RNA biosensors were produced by T7 in vitro transcription using a Promega kit, phenol:chloroform extracted and finally dissolved in nuclease-free water. RNA concentration was determined by BioDrop and quality assessed using a 15% urea-PAGE (see lab notes). Several tests were completed where the RNA was refolded in the presence of ligand or fluorophore to test the effectiveness of the RNA Mango biosensor as an indicator of ligand binding by the small molecule biosensors.



    1. RNA Refolded in Presence of TOI Fluorophore Only

          All three biosensors were unfolded by heating to 95°C and then snap cooled on ice for 10 min. TOI was added to the RNA to a final concentration of 1.4 uM and the reaction was warmed to 37°C on a heat block. The reaction mixture was then placed in a quartz cuvette and the fluorophore was excited at 510 nm. The fluorescence emission was recorded between 520 and 600 nm. Ligand (glucose or theophylline) was titrated into the reaction mixture and the fluorescence spectra was recorded after each addition (figure 1).

          All three constructs show fluorescence, indicating that the RNA Mango construct is folding correctly and able to bind to the thiazole orange dye. However, as ligand was added to the reaction mixture the fluorescence output decreased for both theophylline and scramble biosensors. Interestingly, a slight fluorescence decrease is also observed for the glucose biosensor until a glucose ligand concentration of 33 µM is reached. A two-fold increase in fluorescence was observed.





    2.RNA Refolded in Presence of Both TOI and Ligand

          Each biosensor was unfolded and refolded by heating to 95°C and then letting the solution cool to room temperature. Fluorophore TOI was added to the reaction mixture prior to the refolding step. To measure the effect of ligand (glucose or theophylline) on RNA refolding, reaction mixtures with and without ligand were prepared. The ligand (final concentration 20 µM) was added to the RNA prior to refolding at room temperature. The reaction mixtures were then excited and the resulting emission spectra recorded (figure 2).

          Again all biosensors are fluorescent indicating that the RNA Mango fluorophore is folding correctly and fluorescing when excited. However, no change in fluorescence is observed for either the glucose (figure 2A) or theophylline (figure 2B) biosensors when their respective ligands are included in the reaction mixture. Confusingly, there is an increase in fluorescence when glucose is added to the glucose scramble biosensor (figure 2C).





    3.RNA Refolded in Presence of Ligand Only

          All three biosensors were unfolded by heating to 95°C. Ligand was then added to the reaction mixture to a final concentration of 20 µM and the RNA was allowed to cool to room temperature. TOI dye was added to the reaction mixture just prior to fluorescence measurements (figure 3).

          Interestingly, the glucose biosensor fluorescence increases when glucose is present, almost a two-fold increase compared with the no glucose mixture (figure 3A). The theophylline biosensor however only appears to have a slight increase in fluorescence when ligand was included during the refolding step (figure 3B). The glucose scramble control biosensor, again shows a fluorescent increase in the presence of ligand (figure 3C).





    Discussion and Next Steps

          Unfortunately, we are not able to ascertain a clear conclusion on whether the RNA Mango biosensor is a good indicator of binding for small molecules to their respective aptamers. It appears that RNA refolding in the presence and absence of both TOI dye and ligand greatly impacts the usefulness of the biosensor.
          To validate the proposed RNA Mango biosensor, additional experiments should be completed using the small molecule biosensors to measure the level of ligand binding. This could potentially be done using SELEX, surface plasmon resonance (SPR), isothermal titration calorimetry (ITC), fluorescence polarization/anisotropy (FA/FP), or capillary electrophoresis (CE) (Ruscito & DeRosa, 2016), which is beyond the scope of our current abilities.

          Additionally, the free energy of the different RNA aptamers could be calculated to determine the general stability of different biosensors using modeling software (Modelling). This would allow us to hypothesize that if the free energy of the RNA structure is lower for the biosensor without ligand, then switching to the ligand-binding conformation would not be thermodynamically favourable.

          Following discussions with diabetes experts, it was decided to design an insulin biosensor instead of a glucose biosensor for our final product. This was suggested as the final sa-RNA construct will be inside cells in order to access the transcription and translation machinery but the glucose that the aptamer was supposed to detect would be extracellular in the bloodstream. Therefore, the glucose biosensor would not work as designed. Using an insulin biosensor would still allow for control of our insulin gene expression and use a feedback loop that would detect intracellular insulin levels. This biosensor will need to be tested in vitro as well.



References

[1] Dolgosheina EV, Jeng SCY, Panchapakesan SSS, Cojocaru R, Chen PSK, Wilson PD, Hawkins N, Wiggins PA, andUnrau PJ. ACS Chemical Biology 2014 9 (10), 2412-2420 DOI: 10.1021/cb500499x

[2] Rankin CJ, Fuller EN, Hamor KH, Gabarra SA, Shields TP. (2006). A Simple Fluorescent Biosensor for Theophylline Based on its RNA Aptamer. Nucleosides, Nucleotides and Nucleic Acids, 25(12), 1407–1424. doi:10.1080/15257770600919084

[3] Ruscito A, DeRosa MC. (2016) Small-Molecule Binding Aptamers: Selection Strategies, Characterization, and Applications. Frontiers in Chemistry DOI=10.3389/fchem.2016.00014   

[4] Yang KA, Barbu M, Halim M, Pallavi P, Kim B, Kolpashchikov DM, Pecic S, Taylor S, Worgall TS, Stojanovic MN. Recognition and sensing of low-epitope targets via ternary complexes with oligonucleotides and synthetic receptors. Nat Chem. 2014 Nov;6(11):1003-8. doi: 10.1038/nchem.2058. Epub 2014 Sep 28. PMID: 25343606; PMCID: PMC4339820.