Systematic Comparison and Rational Design of Theophylline Riboswitches for Effective Gene Repression

Abstract

Riboswitches are promising regulatory tools in synthetic biology. To date, 25 theophylline riboswitches have been developed for regulation of gene expression in bacteria. However, no one has systematically evaluated their regulatory effects. To promote efficient selection and application of theophylline riboswitches, we examined 25 theophylline riboswitches in Escherichia coli MG1655 and found that they varied widely in terms of activation/repression ratios and expression levels in the absence of theophylline. Of the 20 riboswitches that activate gene expression, only one exhibited a high activation ratio (63.6-fold) and low expression level without theophylline. Furthermore, none of the five riboswitches that repress gene expression were more than 2.0-fold efficient. To obtain an effective repression system, we rationally designed a novel theophylline riboswitch to control a downstream gene or genes by premature transcription termination. This riboswitch allowed theophylline-dependent downregulation of the TurboRFP reporter in a dose- and time-dependent manner. Its performance profile exceeded those of previously described repressive theophylline riboswitches. We then introduced as the second part a RepA tag (protein degradation tag) coding sequence fused at the 5'-terminal end of the turborfp gene, which further reduced protein level, while not reducing the repressive effect of the riboswitch. By combining two tandem theophylline riboswitches with a RepA tag, we constructed a regulatory cassette that represses the expression of the gene(s) of interest at both the transcriptional and posttranslational levels. This regulatory cassette can be used to repress the expression of any gene of interest and represents a crucial step toward harnessing theophylline riboswitches and expanding the synthetic biology toolbox. IMPORTANCE A variety of gene expression regulation tools with significant regulatory effects are essential for the construction of complex gene circuits in synthetic biology. Riboswitches have received wide attention due to their unique biochemical, structural, and genetic properties. Here, we have not only systematically and precisely characterized the regulatory properties of previously developed theophylline riboswitches but also engineered a novel repressive theophylline riboswitch acting at the transcriptional level. By introducing coding sequences of a tandem riboswitch and a RepA protein degradation tag at the 5' end of the reporter gene, we successfully constructed a simple and effective regulatory cassette for gene regulation. Our work provides useful biological components for the construction of synthetic biology gene circuits.

Keywords: biological parts; posttranslational regulation; protein degradation tag; theophylline riboswitch; transcriptional regulation.

FIG 1

Evaluation of the regulatory efficiencies of various theophylline riboswitches. (A) Design of a plasmid-containing gene circuit to assess the repression efficiencies of theophylline riboswitches. The turborfp gene (shown as a red arrow) is controlled by a constitutive promoter (shown as a black arrow). The coding sequences for theophylline riboswitch and RBS are shown as a blue box and black semicircle, respectively. Control plasmid pWA143 contains gene circuit but without the riboswitch coding sequence. (B) Definition of activation/repression ratio and dynamic range. The activation/repression ratio was calculated by dividing the maximum value by the minimum value. In the case of activation, the minimum value is equal to the expression level in the absence of theophylline; in the case of repression, the maximum value is equal to the expression level in the absence of theophylline. “Dynamic range” refers to the maximum and minimum values of the interval. (C) TurboRFP fluorescence intensity (FI) measured in the absence (light blue) and presence (dark blue) of 2 mM theophylline. Theophylline was added at 2 h, and the relative turborfp mRNA expression levels or FI was measured at 12 h. Numbers above the column represent activation/repression ratios. Data represent the mean ± standard deviation (SD) from three biological replicates.

FIG 2

Evaluation of the regulatory efficiencies of the rationally designed TC-OFF theophylline riboswitches. (A) Design strategy for theophylline-dependent riboswitch R1 to control transcription. The theophylline aptamer (red) was fused to an intrinsic transcription terminator (cyan and yellow). Sequences modified from the previous riboswitch B are marked in yellow. The RBS sequence (black semicircle) and the open reading frame of the reporter gene turborfp are located downstream of this construct. In the absence of theophylline, intrinsic terminator formation is inhibited, resulting in transcription readthrough and turborfp expression. Upon binding of theophylline (black solid circle), an intrinsic terminator is formed and transcription is prematurely stopped, resulting in repression of turborfp expression. (B) Relative expression levels of turboRFP mRNA measured in the absence (white and light blue) and presence (gray and dark blue) of 2 mM theophylline. 16S rRNA was used as an internal control. Data were subjected to one-way analysis of variance (ANOVA) using the Bonferroni test. ns, not significant (P > 0.05). Asterisks indicate significant differences: *, 0.01 < P < 0.05; **, 0.001 < P < 0.01. (C) TurboRFP FI measured in the absence (white and light blue) and presence (gray and dark blue) of 2 mM theophylline. Numbers above the column represent activation/repression ratios. Data represent the mean ± SD from three biological replicates.

FIG 3

Evaluation of the regulatory efficiencies of tandem TC-OFF theophylline riboswitches at different theophylline concentrations. (A) Schematic of plasmids containing the engineered gene circuit controlled by a tandem theophylline riboswitch. Coding sequences of the promoter (black arrow), riboswitches (blue box), linker (brown box), RBS (black semicircle), and turboRFP (red arrow) are shown. R1 represents one riboswitch, R2 represents two riboswitches in tandem, and R3 represents three riboswitches in tandem. (B) TurboRFP FI measured in each strain harboring a plasmid with different tandem riboswitches (R1, R2, and R3) in the absence and presence of 2 mM theophylline; (C) TurboRFP FI of each strain grown in LB medium supplemented with 0, 0.5, 1.0, and 2 mM theophylline. FI was measured 6 h after theophylline addition. Numbers above the columns represent activation/repression ratios. Data represent the mean ± SD from three biological replicates.

FIG 4

Comparison of experimental measurements with model predictions. (A) TurboRFP FI measured from 2 to 24 h in the pWA140 (R2)-containing strain at various theophylline concentrations from 0 to 2.0 mM. Data represent mean ± SD of three biological replicates. (B) Mathematic model of R2 based on all data measured at different theophylline concentrations and growth times. The x axis represents growth time, the y axis represents the theophylline concentration, and the z axis indicates the TurboRFP FI. Data represent the mean ± SD from three biological replicates.

FIG 5

Evaluation of the regulatory efficiencies of R2 in different strains and culture conditions. (A) Regulation of TurboRFP expression by R2 in different E. coli strains; (B) regulation of TurboRFP expression by R2 in different bacterial species; (C) regulation of TurboRFP expression by R2 in the E. coli JM101/pWA140 strain grown at different temperatures; (D) regulation of TurboRFP expression by R2 in the E. coli JM101/pWA140 strain grown in different media. Theophylline (2 mM) was added at 2 h, and FIs were measured at 24 h. Numbers above the columns represent activation/repression ratios. Data represent the mean ± SD from three biological replicates.

FIG 6

Regulation of TurboRFP expression by the RR system. (A) Schematic of plasmids containing the coding sequences of R2, RBS, or the RepA tag upstream of turborfp. pWA143 was used as a control plasmid without regulatory sequences upstream of turborfp. pWA140, pWA144, and pWA146 represent the plasmids containing the tandem riboswitch coding sequence, the RepA tag coding sequences, and both of these parts, respectively. (B) Regulation of TurboRFP expression by the protein degradation RepA tag at different growth phases. The numbers above the columns represent the TurboRFP FI of the “No riboswitch, no tag” treatment divided by that of the “RepA-tag” treatment. (C) Regulation of TurboRFP expression by R2 at 0 or 2.0 mM theophylline from 2 to 12 h. The numbers above the columns represent the TurboRFP FI at 0 mM theophylline divided by that at 2 mM theophylline. (D) Regulation of TurboRFP expression by the RR system at 0 and 2.0 mM theophylline from 2 to 12 h. Numbers above the column represent repression ratios. All data above represent the mean ± SD from three biological replicates.

FIG 7

Regulation of LacZ expression by R2 or the RR system located in the chromosome. (A) Schematic diagram of strains containing the coding sequences of R2 (R2-lacZ) or RR system upstream of lacZ (RR-lacZ); (B) regulation of LacZ expression by R2 or the RR system at 0 or 2.0 mM theophylline at 4, 8, and 12 h. All data above represent the mean ± SD from three biological replicates.