all plasmids confer kanamycin resistance, and all constructions and characterizations were done in E. coli TOP10. Three composite parts were used: P_BLrep-RFP (original part), P_BLrep-RFP-DAS and P_BLrep-RFP-YbaQ. All three were tested simultaneously in each experiment. Engineered cells were inoculated in 5 mL of LB with kanamycin overnight. The next day, 5 uL of cell culture was refreshed in 5 mL of fresh LB with kanamycin until OD₆₀₀ reaches 0.05, upon which 1 ml of cell culture was transferred to each well of a 12-well plate. Cells were then incubated in 37°C at 120 rpm under different light conditions depending on the experimental setup. Fluorescence of the cells was measured at 1-hour intervals using BioTek Synergy H1 microplate reader.

Cells were characterised in 12-well-plates in this experiment, and each sample is measured in triplicates. An example of a setup is shown below.

In this experiment, one setup was incubated under persistent blue light for 8 hours, while another setup was incubated in the dark (covered with black cloth) for the same duration.

Our data suggests that using RFP solely as a reporter is unable to reflect the expected induction and repression activities of the promoter. As seen from Figures 3a and b, fluorescence decreased over the period of 0 to 4 hours in the dark, and increased over the period of 4 to 8 hours under blue light. This is completely opposite of what we expected. We speculated that this phenomenon is due to the mismatch in the rate of increase in OD₆₀₀ and the degradation rate of RFP. Hence, to alter the degradation rate of RFP, we attached 2 different degradation tags to it, YbaQ and DAS. Degradation tags are also referred to as degrons. They are either part of coding regions, or are added to them to direct their degradation by proteases.

Experimental results show that RFP attached to YbaQ could reflect the expected induction and repression activities of P_BLrep. As shown in Figure 3a, RFP/OD increased steadily with time in the dark, where repression is absent, while in Figure 3b, we can see that RFP/OD decreased exponentially with time under blue light. The high intial RFP/OD at 0 hour in Figure 3a can be explained by the low initial OD₆₀₀ (0.05) and the presence of RFP in the system prior to the commencement of the experiment.

Figures 4a and b represent the comparison of RFP/OD of persistent light on and off conditions for P_BLrep-RFP and P_BLrep-RFP-YbaQ. There is a clear increasing trend for the fold change for P_BLrep-RFP-YbaQ, while the fold change for P_BLrep-RFP increased over the period of 0 to 4 hours, then dropped for the subsequent 4 hours.

This characterisation experiment tells us that the degradation rate of RFP-YbaQ is required for this promoter to work as intended.

To verify that the degradation rate of RFP-YbaQ is indeed the most optimal for P_BLrep to work, we carried out further characterisation experiments with different blue light on-off patterns.

As observed from the Figures 5 and 6, even under different light on-off patterns, P_BLrep-RFP-YbaQ still performed the best, showing clearer oscillation patterns with the absence and presence of light, while P_BLrep with just RFP alone or with RFP-DAS showed relatively poor responses with the rate at which light is being switched on and off.

We also modelled the activity of the promoter and our model shows that for P_BLrep to reflect better oscillation patterns, we should incubate the cell cultures in the dark for 45 minutes, and under blue light for the remaining duration of the experiment.

Our wetlab experimental data verified the recommendations made by our model. A dip is no longer observed at the start of the graph, and the change in RFP/OD could reflect our light on-off pattern.