Inducible systems able to sense and respond to fluctuation of a compound of interest, are some of the most practical tools synthetic biology has to offer. However, such systems are not easy to find and, more often than not, are inadequately characterized. During this year’s competition, our team tried to characterize a major component of such a system, the FR1 promoter BBa_K4624000, the actuator of the malonyl-CoA sensor-actuator system we have implemented in our design (Fig. 1). To learn about the involvement of the FR1 promoter on oPHAelia’s system, see our Design page.
PFR1 is a synthetic promoter containing binding sites for the malonyl-CoA-responsive transcription factor FapR, derived from the Gram-positive bacteria Bacillus subtilis1. In the presence of low malonyl-CoA levels, FapR binds to FapRO sites and inhibits the promoter. On the other hand, the binding of malonyl-CoA to FapR triggers a conformational change to the protein, causing FapR-DNA complex dissociation.

Fig. 1. Schematic representation of the malonyl-CoA-based negative feedback system for the regulation of acc expression. In the presence of low malonyl-CoA levels, the FapR protein is in its unbound state and is able to bind to the PFR1 promoter causing its inhibition. Thus, the lacI is not expressed and the LacI-repressive PT7 promoter is active transcribing the acc. The expression of acc is equivalent to the rise of malonyl-CoA levels, which bind to and restrict the FapR protein, leaving the PFR1 active for lacI transcription. LacI inhibits the PT7, switching the expression of the acc off and consequently, the malonyl-CoA levels drop.

Malonyl-CoA is a central metabolite in microorganisms, serving as a key building block for synthesizing a wide range of crucial compounds2,which can be used as or converted to commodity chemicals, pharmaceuticals and biofuels3,4. The FapR/PFR1 system has the potential to enable the effective utilization of this critical intermediate in future applications, offering solutions to numerous challenges in the field of synthetic biology. So, the straightforward design of this system and the various applications it could have, motivated us to conduct 2 experiments for the characterization PFR1 by itself but also the inhibitory role of the fapR protein on the promoter.
  To adequately characterize the promoter we had to shed light on two aspects: 1. the strength of the synthetic PFR1 , compared to a standard one, such as an Anderson promoter and 2. the expression levels of FapR needed to effectively regulate the promoter. Therefore, we planned to design, build and test 2 constructs for an initial characterization of the system:

For the 1st experiment, a level 1 (alpha) construct with the PFR1 upstream the reporter syfp2 (Fig. 2) was designed for promoter strength evaluation. By comparing it to a standard promoter, we aimed to provide future users with a straightforward way to assess the suitability of this promoter for their specific purposes.

Fig. 2. Schematic representation of the level 1 (alpha) construct for the characterization of the PFR1 (1st experiment).

For the 2nd experiment, a level 2 (omega) construct was designed, containing the aforementioned level 1 (alpha) reporter construct and an additional one, carrying the fapR regulated by the AraC/PBAD system (Fig. 3). So, by titrating arabinose to control the levels of FapR we will be able to give sufficient data to determine the levels of repression the protein exerts on the promoter, and the degree of fapR transcription required to make the system responsive.

Fig. 3. Schematic representation of the level 2 (omega) construct for the evaluation of the inhibition FapR poses on PFR1 (2nd experiment).

Additionally, a positive control construct carrying the Anderson promoter PJ23118 (BBa_J23118) regulating the expression of syfp2 was designed. To better understand our cloning process, we encourage you to check the section “Cloning Method” on our Engineering page.

To begin, we domesticated the sequences of the FR1 promoter, the AraC/PBAD system, and fapR. The domestication involved removing internal restriction sites that were not compatible with the GoldenBraid cloning method and the addition of appropriate 3' and 5' 4-nt overhangs. Once designed, the domesticated sequences were ordered to be synthesized by IDT. The next step was to proceed with the assembly. Detailed instructions for the standard cloning procedures can be found on our Experiments page. The domesticated sequences were inserted into a pUPD2 part domestication vector to create level 0 constructs that we could then combine to assemble our complete constructs. Once the insertion was verified through restriction-digestion reaction and gel electrophoresis (Fig. 4, 5 and 6), we used the resulting level 0 parts along with the level 0 parts of the B0030 RBS (BBa_J428032), syfp2 (BBa_K864100) and B0015 double terminator (BBa_J428092), provided by this year’s iGEM distribution kit, to create the level 1 reporter construct (Fig. 2) and the level 1 construct carrying fapR. Both constructs were successfully assembled, and the inserts were once again confirmed through restriction-digestion reactions (Fig. 7 and 8). Subsequently, we utilized the reporter construct for the PFR1 in the 1st experiment.

Fig. 4. Diagnostic digestion of pUPD2_pFR1 with EcoRI and EcoRV, expected bands (bp): 1305 and 896. Lane 2: pUPD2 (no insert)

Fig. 5. Diagnostic digestion of pUPD2_araC/pBAD with EcoRI and EcoRV, expected bands (bp): 2107, 896 and 466. Lane 2: pUPD2 (no insert).

Fig. 6. Diagnostic digestion of pUPD2_fapR with EcoRV, expected bands (bp): 1477 and 1196. Lane 2: pUPD2 (no insert).

Fig. 7. Diagnostic digestion of pDGB3α1_pFR1-syfp-rrnB T1/T7TE with BsaHI, expected bands (bp): 2327, 2022, 1628, 1291 and 58. Lane 2: pDGB3α1 (no insert).

Fig. 8. Diagnostic digestion of pDGB3α2_araC/pBAD-fapR-rrnB T1/T7TE with EcoRV and BsaHI, expected bands (bp): 2327, 1628, 1379, 1291, 807, 559, 391 and 58. Lane 2: pDGB3α2 (no insert).

Having the level 1 (alpha) reporter construct (Fig. 2) ready, we set out to perform the 1st experiment. The detailed protocol can be found on our Experiments page. In short, the isolated plasmid carrying the reporter construct was used to transform E. coli BL21 (DE3) chemically competent cells, following standard protocol. Single colonies were picked to inoculate 5 ml of LB medium, with the appropriate antibiotic, and cultured O/N at 37°C and 210 rpm. Single colonies were also picked for the positive control (syfp2 under the regulation of the Anderson J23118 promoter), and the negative control (non-transformed E. coli BL21 DE3 cells). The O/N cultures were used to prepare the final dilutions into M9 minimal medium in order to reach the same starting OD600 value. Measurements were taken after 6 hours of incubation (37°C and 210 rpm) at 511 nm (excitation) and 529 (emission).

Results

The results we got from the 1st experiment are depicted in Fig. 9.

Fig. 9. Normalized fluorescence intensity of the pFR1-syfp-rrnB T1/T7TE construct, compared to a positive control (pJ23118-syfp-rrnB T1/T7TE) and a negative control (non-transformed E. coli cells) after 6h incubation.

It is evident that the PFR1 exhibits a fluorescence output similar to that of the constitutive PJ23118 , in absence of the FapR regulatory protein. So we can conclude that these two promoters have a comparable intensity.

  Now that we had tested the functionality of the level 1 (alpha) reporter construct and provided an estimate for the promoter's strength we could start the process of building the level 2 (omega) construct (Fig. 3) to assess the relation of PFR1 with its regulatory protein FapR. To achieve that, we combined the construct from the 1st experiment (Fig. 2), with the level 1 (alpha) construct carrying the fapR under the control of the AraC/PBAD system. Using this construct, we can employ various arabinose concentrations to induce varying expression levels of fapR and observe how PFR1 responds to this regulation. Gathering this type of information is essential for integrating both components into any regulatory circuit, similar to what we have integrated in oPHAelia’s design (see Design page).
  The construct was assembled following our standard protocol for digestion-ligation reaction. After a few failed attempts, we managed to successfully build the level 2 construct which was again confirmed with restriction-digestion reaction and gel electrophoresis (Fig. 10). Similarly to the 1st experiment, we started again by transforming E. coli BL21 (DE3) chemically competent cells with the isolated plasmid carrying the level 2 construct. The next day, for the preparation of liquid cultures, single colonies were picked and inoculated in LB medium, with the appropriate antibiotic, and finally the cultures were incubated O/N at 37°C and 210 rpm. The next morning, final dilutions x5 were prepared in M9 minimal medium for the level 2 construct as well as for the same positive and negative controls used in the 1st experiment. Lastly, addition of L-arabinose followed for the preparation of 4 different final concentrations (0.01, 0.1, 1 and 10 mM), creating various levels of PBAD induction. Measurements at wavelengths of 511 nm (excitation) and 529 nm (emission) were taken at 3h and 6h timepoints.

Fig. 10. Diagnostic digestion of pDGB3ω1_pFR1-syfp-rrnB T1/T7TE + araC/pBAD-fapR-rrnB T1/T7TE with EcoRI and BsaHI, expected bands (bp): 2325, 1898, 1628, 1304, 1291, 758, 470 and 58. Lane 2: pDGB3ω1 (no insert).

Results

The results we got are shown in Fig. 11.

Fig. 11. Normalized fluorescence intensity for the level 2 construct (pDGB3ω1_pFR1-syfp-rrnB T1/T7TE + araC/pBAD-fapR-rrnB T1/T7TE) in different concentrations of L-arabinose, after 3h and 6h incubation.

As we can see from the results (Fig. 11), there is a significant decrease of fluorescence intensity at increasing concentrations of L-arabinose. This fact confirms the existence of the FapR-mediated PFR1 repression, since higher concentration of L-arabinose implies a higher activation of the PBAD , through the AraC regulatory protein, and ultimately a higher expression of fapR.
  For our system however, with the results obtained from the two experiments we cannot say which constitutive Anderson promoter would be ideal for the regulation of fapR expression. So, additional data and experiments are required (see “Future experiments” section below) for the integration of the PFR1 promoter and FapR regulator into the negative feedback system of oPHAelia (Fig. 1), which due to limited time were not performed.

  To obtain optimal sensor-actuator behavior in the malonyl-CoA-based negative feedback system (Fig. 1), the selection of a constitutive promoter with the proper strength for fapR expression is critical5 . Therefore, it would be necessary to run a 3rd experiment to tune the fapR expression levels by titrating with various amounts of arabinose and examine the responses of the sensor-actuator system. For this experiment, we would need to create a construct that would consist of two composite parts. The first part would be the one we tested in the second experiment, which includes the PFR1-driven reporter module and the PBAD-driven FapR repressor module. The second part would involve a PT7-driven malonyl-CoA source module, where the acc gene would be under the control of a LacI-repressive T7 promoter (as depicted in Fig. 12). So, by titrating L-arabinose (0.01, 0.1, 1 and 10 mM) to control the amount of FapR and IPTG to manipulate cellular malonyl-CoA levels, we could evaluate the desired malonyl-CoA response by measuring the fluorescence intensity, indicating that the amount of FapR produced under this concentration of arabinose would be optimal for regulation. Using these data, we could select a constitutive promoter with strength corresponding to that of the induced PBAD at the optimal concentration of arabinose we had observed.

Fig. 12. Schematic representation of the construct needed to determine the appropriate constitutive promoter for the regulation of fadR expression, consisting of the PFR1-driven reporter module, the PBAD-driven FapR repressor module and the PT7–driven malonyl-CoA source pathway ( accABCD ).