Fig. 4. Experimental demonstration of RPA.
a Illustration of the experimental setup. All four controllers of Fig. 3 were tested in a two-plasmid system. Plasmid 1 encodes an IntC segment incorporated in an activator driven by a constitutive promoter. Plasmid 2 encodes IntN (resp. IntC) in the closed-loop (resp. open-loop) setting, which is fused to the fluorophore mVenus via a P2A-T2A linker and is driven by the activator expressed from Plasmid 1. The fluorescent protein mVenus is used as a proxy reporter for its own mRNA which is the regulated output expected to exhibit RPA in closed loop. In open loop, there is no interaction between the two IntC segments; whereas, in closed loop, the inteins pair can perform the intein-splicing reaction to produce (possibly) functional products. The setpoint is tuned by changing the constitutive production rate μ1 via the transfected copy numbers of Plasmid 1. The reference, or undisturbed output level, is obtained by fixing the copy numbers of the transfected Plasmid 2 across all setpoints; whereas, the disturbed output level is obtained by repeating the same experiment for all setpoints but with a higher copy number of Plasmid 2. b Steady-state errors. For each controller, we show a simplified schematic (top left) and two bar graphs. The bottom bar graphs show the normalized fluorescence of the proxy reporter with (red) and without (green) disturbance and for the closed-loop (left) and open-loop (right) settings. The disturbed and reference triplicate measurements were normalized to the mean fluorescence of the reference data for each setpoint. The x-axis follows a log2-scale and shows the amount of Plasmid 1 transfected within every well. The red horizontal lines give the normalized output averaged over all the setpoints, and the numbers above the lines indicate the averaged error of the disturbed output relative to the reference. The top bar graphs show the non-normalized data over a selected subset of setpoints (pointed out by the dashed lines) that match in absolute fluorescence between the open- and the closed-loop circuits. For all the data, the HEK293T cells were measured using flow cytometry 48 h after transfection, and the normalized data are shown as mean + SD for n = 3 technical replicates.