Strains and media

E. coli Top10 was used for all cloning. For characterization, we used E. coli strain AB360 (ref. 14) for strains containing rTetR plasmids and/or T7RNAP plasmids. The strain contains the transcription factor AraC, whereas arabinose-metabolizing genes araBAD are deleted, and lacYA177C that allows for titratable arabinose regulation. SKA684 (MG1655 ΔaraCBAD ΔlacIZYA ΔaraE ΔaraFGH attB::lacYA177C ΔrhaSRT ΔrhaBADM Tn7::tetR kan(FRT)), a kanamycin-resistant SKA703 (ref. 48) variant was used for experiments involving the TetR repressor. Plasmids were transformed using a one-step preparation protocol of competent E. coli for transformation of plasmids in testing strains49.

Autoclaved LB-Miller medium was used for strain propagation. Sterile-filtered M9 medium supplemented with 0.2% casamino acids, 0.4% glucose, 0.001% thiamine, 0.00006% ferric citrate, 0.1 mM calcium chloride, and 1 mM magnesium sulfate was used for all gene expression experiments. Antibiotics (Sigma-Aldrich Chemie GmbH) were used as necessary for plasmid maintenance at concentrations of 100 μg/mL ampicillin, 34 μg/mL chloramphenicol, and 50 μg/mL kanamycin. Arabinose was received from Sigma-Aldrich Chemie GmbH. aTc was received from Chemie Brunschwig AG.

Plasmids and genetic parts

As a reporter plasmid for TetR- or rTetR-regulated expression, we used a pZ-series plasmid30 containing a chloramphenicol resistance gene and a pSC101 origin of replication. The resulting plasmid pAB269 encodes mCherry under control of a Ptac promoter50 followed by a tetO2 operator sequence and an RBS (see genetic parts list in Supplementary Data 1).

rTetR was derived from rtTA3 (pLenti CMV rtTA3, a gift from Eric Campeau, Addgene plasmid # 26429;; RRID:Addgene_26429). We used amino acids 1–207 from rtTA3, under control of ParaB* (ref. 14) and containing an ampicillin resistance gene and a colEI origin of replication to create pAB286.

Growth and light incubation conditions

All experiments were performed in an environmental shaker (INNOVA 40 R, New Brunswick) at 37 °C with shaking at 230 r.p.m. and black, clear bottom 24-well plates (Dunn Labortechnik GmbH, #303008), which was sealed with one layer of pierced polyolefin foil (HJ-BIOANALYTIK GmbH, Art. Nr. 900371) to reduce liquid evaporation, and an additional layer of gas-permeable membrane (BREATHseal™ Greiner Bio-One GmbH) foil to guarantee sterility, as well as a plastic lid (Greiner Bio-One GmbH, Product #: 656171). For experiments, overnight cultures were inoculated in M9 medium and grown overnight to an OD600 of ~4. aTc was added to 50 ng/ml for overnight cultures containing rTetR and grown in light-proof black tubes (Greiner Bio-One) to reduce accumulation of mCherry in overnight cultures. These cultures were diluted 1:20,000 into fresh M9 medium containing the respective inducer concentrations, right before the start of the experiment. This high dilution ensures that the cells are still in logarithmic growth phase after 5 h, at the end of the experiment14 (Supplementary Fig. 10). One ml of inoculated culture was incubated per well of the 24-well plates.

For light induction experiments, we used a 24-well light plate apparatus (LPA)34 equipped with UVA (375 nm, LEDsupply, PART #: L7-0-U5TH15-1), blue (472 nm, Super Bright LEDs Inc., Part #: RL5-B2545), green (525 nm, Super Bright LEDs Inc., Part Number: RL5-G8045), and infrared (740 nm, Marktech Optoelectronics, Part #: MTE1074N1-R) LEDs. The LEDs were calibrated (Photometer: Thorlabs PM100USB with S170C) to the LED with the lowest intensity of the respective wavelengths, resulting in the following calibrated maximal intensities: UVA 4.1 ± 0.5 mW/cm2, blue 4.8 ± 0.3 mW/cm2, green 1.9 ± 0.1 mW/cm2, and infrared 0.70 ± 0.01 mW/cm2. Light intensity correlated linearly with the input number given to the LPA via its program IRIS (Supplementary Fig. 11).

Cells were grown for 5 h before transcription, and translation was stopped with rifampicin and Tc14. A total of 100 µL inhibition solution was aliquoted in covered 96-well U-bottom plates (Thermo Scientific Nunc), precooled on ice, and samples were added in equal volumes (100 µL), resulting in a final inhibitor concentration of 250 μg/mL rifampicin (Sigma-Aldrich Chemie GmbH) and 25 μg/mL Tc (Sigma-Aldrich Chemie GmbH). The inhibition solution contained 500 μg/mL rifampicin and 50 μg/mL Tc in phosphate-buffered saline (Sigma-Aldrich Chemie GmbH, Dulbecco’s phosphate-buffered saline), and was filtered using a 0.2 μm syringe filter (Sartorius). After sample was added, the solution was incubated on ice for at least 30 min. Then mCherry maturation was carried out at 37 °C for 90 min. The samples were kept at 4 °C until measurement through flow cytometry. Due to the slow degradation of aTc with blue, green, and infrared light, we preincubated the M9 medium with aTc without cells. After this incubation, we inoculated the preincubated media and grew the cells for 5 h before inhibition.

Flow cytometry measurement

Cell fluorescence was characterized using a Cytoflex S flow cytometer (Beckman Coulter) equipped with CytExpert 2.1.092 software. mCherry fluorescence was measured with a 561 nm laser and 610/20 nm bandpass filter, and following gain settings: forward scatter 100, side scatter 100, mCherry gain 1500 when mCherry was expressed from aTc-regulated promoters, and 300 gain when mCherry was expressed with Opto-T7RNAP due to the difference in expression levels. Thresholds of 2500 FSC-H and 1000 SSC-H were used for all samples. The flow cytometer was calibrated before each experiment with QC beads (CytoFLEX Daily QC Fluorospheres, Beckman Coulter) to ensure comparable fluorescence values across experiments from different days. At least 15,000 events were recorded in a two-dimensional forward and side scatter gate, which was drawn by eye and corresponded to the experimentally determined size of the testing strain at logarithmic growth and was kept constant for analysis of all experiments, and used for calculations of the mean and CV using the CytExpert software (Supplementary Fig. 12).

Cell growth measurement using flow cytometry

Cell growth was determined as the ratio of cells per defined volume of counting particles (2 μm AccuCount Blank Particles). For this, 79 μL of 50 µg/ml Tc and 500 μg/mL rifampicin (Sigma-Aldrich Chemie GmbH) in phosphate-buffered saline for transcription and translation inhibition, and 21 μL of AccuCount Blank Particles (Spherotech) for counting reference was used per sample. The cell solution was added in equal volumes (100 µL cell culture to 100 µL of inhibition and counting solution). Absolute cell counts per particle counts were determined by flow cytometry. For this, samples were measured for 2 min at 10 µL/min. The gating of the AccuCount particles and cells is shown in Supplementary Fig. 12 bottom. The reported data are from three biological replicates pooled from experiments performed on the same day. The doubling time of E. coli was measured to be 36.9 ± 1.2 min in our setup (Supplementary Fig. 13). We further determined that an inoculum of 1:20,000 of overnight culture into fresh media leads to an OD600 of 0.034 ± 0.003 after 5 h incubation, therefore ensuring logarithmic growth throughout the experiments. Further, a calibration of optical density of a cell culture (OD600) to our cell growth measurement using flow cytometry (counts/bead) was performed, which shows a linear correlation until an OD600 of 0.2 (Supplementary Fig. 14).

Mathematical modeling

The TetR aTc dose–response (Fig. 1c) was fitted to a Hill-like equation of the following form:

$$f_{{mathrm{TetR}}}left( x right) = a + r_{{mathrm{max}}}frac{{x^n}}{{k_{mathrm{m}} + x^n}},$$


where (f_{{mathrm{TetR}}}left( x right)) describes the gene expression controlled by TetR as a function of aTc concentration, x represents aTc concentration, a corresponds to the basal promoter activity in fully repressed conditions, (a+rmax) is the maximal promoter expression, km is aTc’s dissociation constant for TetR, and n is the Hill coefficient for TetR. This dose–response was used subsequently to obtain aTc concentration estimates from the fluorescence readouts.

The measured aTc degradation responses to light intensity and light duration (Fig. 2b) were fitted to exponential decay Eq. (2)

$$fleft( u right) = e^{ – cu},$$


where u represents the input (light intensity or light duration), and c is the decay factor. The data were normalized to have a maximal value of one prior to the fitting. Thus, the output of such a function lies in the range of [0,1], and represents the fraction of aTc degraded due to the applied input.

The fitted mathematical models for light duration and intensity-dependent aTc degradation, with TetR as sensor, were used to predict the response of rTetR to a set of inputs. As the values outputted by the fitted mathematical models correspond to the fraction of aTc degraded and not to aTc concentrations, these were multiplied by the aTc initial concentration to make the predictions.

The aTc dose–response curve of rTetR (Fig. 1f) was fitted to a Hill equation of the following form:

$$f_{{mathrm{rTetR}}}left( x right) = r_{{mathrm{max}}} – frac{{left( {r_{{mathrm{max}}} – a} right)x^n}}{{k_{mathrm{m}} + x^n}},$$


where (f_{{mathrm{rTetR}}}left( x right)) describes the gene expression controlled by rTetR as a function of aTc concentration, x represents aTc concentration, rmax corresponds to the maximal promoter expression, a is the promoter activity in fully repressed conditions, km is aTc’s dissociation constant for rTetR, and n is the Hill coefficient for rTetR.

The aTc dose–response of the bandpass circuit (Fig. 4a) was empirically modeled combining the dose responses of aTc to TetR Eq. (1) and rTetR Eq. (3) as follows:

$$f_{{mathrm{prp}}}left( x right) = frac{x}{{x + k}},$$


$$f_{{mathrm{BP}}}left( x right) = f_{{mathrm{TetR}}}(x)(1 – f_{{mathrm{prp}}}(x)) + f_{{mathrm{rTetR}}}(x)f_{{mathrm{prp}}}(x),$$


where (f_{{mathrm{prp}}}left( x right)) describes the proportionality factor that determines how much (f_{{mathrm{TetR}}}(x)) and (f_{{mathrm{rTetR}}}(x))is considered as a function of aTc concentration, x represents aTc concentration, and k defines the concentration of aTc, where (f_{{mathrm{TetR}}}) and (f_{{mathrm{rTetR}}}) are considered in equal proportions; x represents aTc concentration. (f_{{mathrm{BP}}}left( x right)) describes the gene expression controlled by the TetR–rTetR bandpass depending on the aTc concentration. We chose this empirical model over a mechanistic model as it required to fit only one parameter k to the bandpass circuit experimental data (Fig. 4a), while parameters of (f_{{mathrm{TetR}}}(x)) and (f_{{mathrm{rTetR}}}(x)) were not modified.

All data were fitted using a nonlinear least squares optimizer (MATLAB R2015a, MathWorks) and all parameter values are shown in Supplementary Table 1.

Reporting summary

Further information on research design is available in the Nature Research Reporting Summary linked to this article.

Source link

Leave a Reply

Your email address will not be published. Required fields are marked *