Team:Michigan/Results

Results

Switch 1.0 was induced with minimal background with its corresponding DNA trigger. However, when induced with thrombin, with aptamer and trigger preincubated, GFP expression became sporadic and had a loose negative correlation with increasing thrombin concentration. Additionally, the negative controls (switch, trigger, aptamer, and no protein) showed the same induction as the positive controls. Switch 2.0 adapted to thrombin protein showed increasing GFP expression with increasing thrombin and minimal background as desired. However, Switch 2.0 adapted to gliadin had a negative control (switch and no protein) with GFP expression at the same rate as when induced with gliadin. Only Switches 1.0 and 2.0 had been tested at the point of writing. All switches are RNA, transcribed from constitutively expressed DNA, and all reporters are GFP, unless otherwise noted. All in vitro transcription and translation reactions were performed using an E. coli based, cell free expression kit (PURExpress In Vitro Protein Synthesis, New England BioLabs).

Figure 1: Toehold Switches with and without triggers
Switch 1.0 was adapted for thrombin detection with a GFP reporter and induced by the corresponding DNA trigger, showing over 100,000 RFU, while the uninduced switch showed minimal background.

Figure 2: Thrombin Titration (switch 1.0 + trigger + aptamer + thrombin), 9/7/15 results
The positive control, consisting of Switch 1.0 with trigger had high expression. The negative control, consisting of Switch 1.0 with pre-incubated trigger and aptamer had high expression. Switch 1.0 with trigger, aptamer, and 0.2 uM-5.0 uM thrombin had high expression, and 25.0 uM had low expression, the opposite of the intended design. This experiment was repeated several times with the same result.

Figure 3: Positive Control (switch + trigger, no aptamer) with thrombin titration, 9/7/15 results
If Switch 1.0 worked the way it was intended to, thrombin would have no impact on just switch and trigger in solution. After observing that the switch “induced” with 25uM thrombin had the lowest GFP expression, the effect of just thrombin on the Switch 1.0 adapted for thrombin detection was tested. The switch was induced at a constant concentration of trigger, with varying thrombin concentration. Switch 1.0 showed decreasing GFP expression loosely correlated with increasing concentrations of thrombin protein. This suggests that the thrombin may be able to bind to the RNA version of the aptamer that is completely exposed in the toehold region of the switch and Switch 2.0 was designed to take advantage of the this.

Figure 4: Switch 2.0 Induced with thrombin, 9/17/15 results
Switch 2.0 adapted for thrombin detection and induced with 0.2-5uM thrombin had increasing expression with increasing thrombin concentrations. Negative control, consisting of Switch 2.0 with no thrombin, had minimal expression. Blank, consisting of water, had minimal fluorescence. This switch was designed so thrombin directly binds the exposed aptamer/toehold region, causing the hairpin to unfold so the ribosome binding site is exposed and translation can occur.

Figure 5: Switch 2.0 Induced with gliadin, 9/18/15 results
Switch 2.0 adapted to gliadin showed high GFP expression, regardless of gliadin concentration.

Discussion
Two different switch designs were tested with thrombin and Switch 2.0 responded successfully, even at the lowest concentration of thrombin tested.

Switch 1.0 was successfully induced by the DNA trigger; however, while just Switch 1.0 in solution produced minimal background, trigger and aptamer together in solution with Switch 1.0 resulted in maximum measurable GFP expression. This implies that the aptamer and trigger are unbound under the experimental conditions (no added salt, 37 °C), allowing high background. Additionally, fluorescence is low at high concentrations of thrombin, implying that thrombin is inhibiting translation of GFP. One possibility is that thrombin inhibits switch activation by binding to the toehold region, which is the RNA version of the DNA aptamer. This inspired the successful Switch 2.0 design.

Switch 2.0 was successfully induced by thrombin. Switch 2.0 showed an increase in GFP expression as thrombin protein concentration increased and responded to the lowest concentration tested, 0.2uM. The thrombin version of Switch 2.0 shows much potential, and it seems promising that adapting the switch for other proteins could yield similar results.

Altogether, the results of the Aptapaper project are promising. High GFP expression was seen with no optimization of experimental conditions, demonstrating the robustness of this system. While many different switch designs were discussed during planning, the results obtained only show one iteration of each switch. Thus, Switch 2.0 was successful with zero optimization of the design, and it is likely that performance could be further improved. Binding strength of different parts of each switch (toehold to protein, hairpin) could be optimized to adapt Switch 2.0 for a wide range of proteins, while maintaining high reporter expression and low background.

Future Directions Include:
Adapt, test, and determine limit of detection of Switch 2.0 for a variety of proteins
Optimize to increase reporter expression for Switch 2.0 and reduce background
Test specificity using biological sample
Demonstrate on paper using a colorimetric reporter