A versatile cis-acting inverter module for synthetic translational switches

Abstract

Artificial genetic switches have been designed and tuned individually in living cells. A method to directly invert an existing OFF switch to an ON switch should be highly convenient to construct complex circuits from well-characterized modules, but developing such a technique has remained a challenge. Here we present a cis-acting RNA module to invert the function of a synthetic translational OFF switch to an ON switch in mammalian cells. This inversion maintains the property of the parental switch in response to a particular input signal. In addition, we demonstrate simultaneous and specific expression control of both the OFF and ON switches. The module fits the criteria of universality and expands the versatility of mRNA-based information processing systems developed for artificially controlling mammalian cellular behaviour.

Figure 1. Design of the switch-inverting module.

(a) Construction of an OFF switch mRNA sequence. An RNA sensory motif that binds to an input molecule is inserted into the 5′-UTR of an mRNA sequence encoding an output gene. The binding of the input molecule to the mRNA inhibits the expression of the output (input=1, output=0), which does not occur in the absence of the input molecule (input=0, output=1). (b) Switch-inverting module for the OFF switch mRNA sequence. The module consists of a bait ORF (N-terminal 152 amino acids of Renilla luciferase) followed by an intron and an IRES. The two PTCs (2 × PTCs) of the bait ORF are located more than 500 nucleotides (nt) upstream of the splice site of the intron to enhance the efficiency of mRNA degradation. The module should be inserted between the RNA motif and the first codon of the OFF switch to generate the ON switch. (c) Behaviour of the inverted ON switch in the absence of the target molecule (input=0). The termination of the translation of the bait ORF triggers mRNA degradation, and the output gene is not sufficiently expressed from the destabilized mRNA (output=0). (d) Behaviour of the ON switch in the presence of the input (input=1). The translation of the bait ORF is blocked in the same manner as in the original OFF switch, and mRNA degradation is therefore not triggered. As a result, the IRES-driven output protein is stably synthesized from the ON switch (output=1).

Figure 2. Behaviour of an inverted ON switch generated via insertion of the module into an OFF switch.

(a) Construction of the INPUT and OUTPUT plasmids transfected into HeLa cells in this study. The INPUT plasmids express either L7Ae or the MS2 coat protein (MS2) as a cognate or noncognate RNA-binding protein, respectively. DsRed-Express is also synthesized from the INPUT plasmid, driven by the IRES. An OUTPUT plasmid expresses ON-switch mRNAs containing K-turn (Kt) or defective K-turn (dKt) as either an active or defective sensory motif, respectively. EGFP driven by the IRES is synthesized from the OUTPUT plasmid. (b–e) Images of cells captured via fluorescence microscopy. DsRed-Express synthesized together with L7Ae (c,e) or MS2 (b,d) is shown in red. EGFP outputs from an ON switch with either an active (b,c; ON-Kt) or a defective (d,e; ON-dKt) sensor are shown in green. These two fluorescent signals are merged and shown in the right most pictures. Scale bars, 200 μm. (f) The mean intensity of EGFP fluorescence of transfected cells. HeLa cells shown in b–e were analysed using a flow cytometer, FACSAria. Cells expressing only DsRed-Express were used as the negative control (DsRed-Express). Bars and error bars represent the mean and s.d., respectively, of three triplicate experiments (n=9). (g) The levels of ON-switch mRNAs, as determined by quantitative RT–PCR. To eliminate the noise from untransfected cells, the results are normalized by employing the mRNA of neomycin-resistant gene transcribed from the OUTPUT plasmid in cells. Bars and error bars represent the mean and s.d., respectively, of three triplicate experiments (n=9). n.a., not analysed. All experiments in b–g were carried out 24 h after the transfection of 100 ng of the indicated OUTPUT plasmid and 20 ng of the INPUT plasmid together with 480 ng of the noncognate INPUT plasmid.

Figure 3. Dependency of inverted switches on PTC and NMD factors.

(a) A switch-inverting module without PTC (ONn). Schematic illustrations of the modules and EGFP fluorescence from switches inverted by the modules are shown in left and right, respectively. The level of EGFP was determined using a flow cytometer, BD Accuri (see also Supplementary Fig. S2), 24 h after transfection of 100 ng of the indicated OUTPUT plasmid with 20 ng of cognate and 480 ng of noncognate INPUT plasmids. Bars and error bars represent the mean and s.d., respectively, of three replicates. (b) siRNA-induced knockdown of NMD factors: SMG1, UPF1 and UPF2. siRNAs to SMG1, UPF1, UPF2 or a non-silencing negative control siRNA (siRNA-n.c.) were transfected into HeLa cells. Forty-eight hours after the transfection, total protein was extracted and subjected to western blotting analysis. Glyceraldehyde 3-phosphate dehydrogenase (GAPDH) was also analysed as an internal control of the lysates. w/o siRNA indicates cells treated with only transfection reagent. The left three lanes were twofold serial dilutions of the lysate from untreated HeLa cells (mock). (c) Behaviour of inverted switches in siRNA-treated cells. Forty-eight hours after the transfection of siRNAs, cells were transfected with the plasmids. Plasmid transfection and analytical procedure were the same as in a. Bars and error bars represent the mean and s.d., respectively, of three replicates. In a and c, cells were transfected with 100 ng of the indicated OUTPUT plasmid with 20 ng of cognate and 480 ng of noncognate INPUT plasmids.

Figure 4. Correlation of the sensitivity and reactivity between the inverted ON switches and parental OFF switches.

(a) Behaviour of the switches as a function of the amount of input. The x axis shows the ratio of the INPUT plasmids expressing cognate RNA-binding proteins to the OUTPUT plasmids for the indicated switches (100 ng). The total plasmid content was adjusted up to 600 ng using noncognate plasmids. Twenty-four hours after transfection, individual cells were analysed using FACSAria. Geometrical mean values of the ratio between EGFP and DsRed-Express signals in a cell were shown. ON-Fr15 and OFF-Fr15 are switches responsive to Bacillus ribosomal protein S15 instead of L7Ae. Data points and error bars represent the mean and s.d., respectively, of three replicates. See Supplementary Fig. S4 (ON-Kt and OFF-Kt) and Supplementary Fig. S5 (ON-Fr15 and OFF-Fr15) for plots produced from this analysis. (b,c) Western blotting analysis of input proteins is shown. Total protein from the transfected cells shown in a was extracted 24 h after transfection and subjected to western blotting analysis. Cognate (L7Ae (b) and S15 (c)) and noncognate (MS2CP (b) and L7KK (c)) input proteins were detected using anti-myc antibody. Glyceraldehyde 3-phosphate dehydrogenase (GAPDH) was also analysed as an internal control of the lysates. The right three lanes were fivefold serial dilutions of the lysate prepared in the same condition as the ratio of INPUT plasmid/OUTPUT plasmid equal to 5. Representative result from the three independent experiments was shown. (d) Behaviour of the switches as a function of the affinity between the input protein and the corresponding sensory motif in the switch. See Supplementary Fig. S8 for plots generated from this analysis.

Figure 5. Induction of cell death using an inverted switch.

(a) Schematic illustration of the experiment. Together with the INPUT plasmids (500 ng), we transfected two plasmids: one expressing anti-apoptotic gene, Bcl-xL (10 ng) and ON-Kt-B (or dKt, 100 ng), which outputs apoptotic gene, Bim-EL instead of EGFP. (b) Induction of annexin V-positive cells. For transfection, 20 ng of cognate and 480 ng of noncognate INPUT plasmids were mixed. The left panel shows histograms of Pacific Blue-labelled anti-annexin V using FACSAria. A black line indicates a threshold used in this analysis. The right panel shows a ratio of annexin V-positive cells to the transfected cells. Bars and error bars represent the mean and s.d., respectively, of two replicates.

Figure 6. Simultaneous and symmetrical control of two outputs in a cell by using ON and OFF switches.

(a) The behaviour of ON and OFF switches responsive to a single input (L7Ae). Two reporter proteins, EGFP and enhanced cyan fluorescent protein (ECFP), were produced from ON and OFF switches, respectively. Two OUTPUT plasmids expressing either an ON or an OFF switch (100 ng each) were transfected into HeLa cells together with 20 ng of cognate and 480 ng of noncognate INPUT plasmids. The fluorescent signals were normalized by those of the switches containing Kt in the absence of L7Ae. (b) Control of two switches under two input proteins. Kt or dKt in ON switches were replaced with Fr15 or dFr15 (defective Fr15). Two OUTPUT plasmids were transfected together with L7Ae- (20 ng) and/or S15- (480 ng) expressing plasmids. In this experiment, a viral nucleocapside protein was used as a noncognate RNA-binding protein. The fluorescent signals were normalized by those detected in the absence of both L7Ae and S15. Transfected cells were analysed using FACSAria. Bars and error bars represent the mean and s.d., respectively, of three replicates.