Plug-and-play protein biosensors using aptamer-regulated in vitro transcription

Abstract

Molecular biosensors that accurately measure protein concentrations without external equipment are critical for solving numerous problems in diagnostics and therapeutics. Modularly transducing the binding of protein antibodies, protein switches or aptamers into a useful output remains challenging. Here, we develop a biosensing platform based on aptamer-regulated transcription in which aptamers integrated into transcription templates serve as inputs to molecular circuits that can be programmed to a produce a variety of responses. We modularly design molecular biosensors using this platform by swapping aptamer domains for specific proteins and downstream domains that encode different RNA transcripts. By coupling aptamer-regulated transcription with diverse transduction circuits, we rapidly construct analog protein biosensors and digital protein biosensors with detection ranges that can be tuned over two orders of magnitude and can exceed the binding affinity of the aptamer. Aptamer-regulated transcription is a straightforward and inexpensive approach for constructing programmable protein biosensors that could have diverse applications in research and biotechnology.

Fig. 1. The ARTIST platform.

a In the absence of an aptamer’s protein ligand, dARTs are transcribed to produce an RNA output (left), but protein binding represses transcription (middle). The input and output domains are decoupled (right), which enables modular design of dARTs by swapping out the aptamer domain or customizing the output sequences to encode different RNA outputs. DNA and RNA are represented with solid and dashed lines, respectively. IFN, Thr, IL6, and TNF indicate DNA aptamers for IFN-γ, thrombin, IL-6, and TNF-α, respectively. b dARTs serve as the protein sensing layer (left) whose outputs can be coupled with downstream circuits to demonstrate versatile functionalities (middle). The RNA output of molecular circuits can react with a DNA reporter complex (right), which produces measurable fluorescence for detection.

Fig. 2. Design of aptamer-regulated transcription templates (dARTs).

a dART templates consist of a double-stranded promoter region (pink boxed region), a single-stranded aptamer domain on the template strand read by T7 RNAP (Apt), and a double-stranded output domain (O1). The single-stranded aptamer domain permits aptamer-ligand binding. b Secondary structure of the RNA transcript predicted by NUPACK. c The aptamer sequences of IFN-O1-dART, which binds IFN-γ and Dummy-O1-dART, whose aptamer domain does not have a tertiary structure or specific protein affinity. d dART transcripts react with an O1 DNA reporter via toehold-mediated strand displacement. e Reacted reporter kinetics from IFN-O1-dART and Dummy-O1-dART transcription with and without IFN-γ and potassium. f Simulated and experimental dose–response curves for 10 nM IFN-O1-dART with 0 to 1000 nM IFN-γ for K_d,apparent = 8 nM (Supplementary Information Note 2). The dashed line is a simulation for K_d,apparent matching experiments. The shaded region shows simulation results for K_d,apparent values spanning 1 to 20 nM (see “Methods”). g Experimentally measured (bold) and simulated (dashed) reacted reporter kinetics for 10 nM of IFN-O1-dART and 0 to 1000 nM of IFN-γ. h Simulated and experimental dose–response curves and i Experimental (bold) and simulated (dashed) reacted reporter kinetics (h) for 1 nM IFN-O1-dART and 0 to 100 nM of IFN-γ using the K_d,apparent determined in (f). For (f) and (h), the reacted reporter concentration for each plot was measured at 120 min. Three technical replicates (blue) are plotted. Error bars indicate the average reacted reporter concentrations of three technical replicates per [IFN-γ] ± one s.d. (N = 3). Supplementary Information Note 2 describes the process for determining K_d,apparent.

Fig. 3. dART dose–response, selectivity, and modularity.

a A series of dARTs with different inputs, achieved by swapping aptamer domain sequences. b Reacted reporter kinetics by IFN-O1-dART, Thr-O1-dART, IL6-O1-dART, and TNF-O1-dART in the presence of 0 to 1000 nM IFN-γ, thrombin, IL-6, and TNF-α, respectively. Shaded regions represent minimum/maximum values of reacted reporter concentration of three replicates (N = 3). c Heat map showing reacted reporter concentrations for 10 nM Thr-O1-dART, IFN-O1-dART, IL6-O1-dART, TNF-O1-dART, and Dummy-O1-dART (a control) each subjected to 100 nM of BSA, thrombin, TNF-α, IL-6, or IFN-γ. d A series of dARTs with different outputs, achieved by swapping template output sequence. e Reacted reporter kinetics for IFN-O1-dART, IFN-O2-dART, and IFN-O3-dART for 0 to 1000 nM of IFN-γ. All experiments were conducted in the presence of 100 mM KCl. Sequences of IFN-O1-dART, Thr-O1-dART, IL6-O1-dART, TNF-O1-dART, IFN-O1-dART, IFN-O2-dART, and IFN-O3-dART are in Supplementary Information Note 1.

Fig. 4. Analog biosensors with steady-state outputs.

a Schematic of coupled RNA transcription of dARTs by T7 RNAP and RNA degradation by RNase H. b The concentration of reacted reporter over time for an experiment with 10 nM IFN-O1-dART, 0 to 1000 nM of IFN-γ, 2 U µL⁻¹ of T7 RNAP, and 2 × 10⁻³ U µL⁻¹ of RNase H. Shaded regions enclose the minimum and maximum values for independent trials (N = 2; see “Methods”). c Steady-state analog outputs of IFN-O1-dART, Thr-O1-dART, IL6-O1-dART, and TNF-O1-dART with 5 to 1000 nM of their corresponding proteins after 240 min. Two individual technical replicates are plotted (N = 2).

Fig. 5. Comparator circuits for digital biosensing.

a The comparator circuit. When [IFN-γ] is high, IFN-O1’-dART’s transcription rate is low. Ref-O1-RNA is thus produced in excess, and it reacts with the O1 DNA reporter. When [IFN-γ] is low, IFN-O1’-RNA’s transcription rate is higher than Ref-O1-RNA’s, and Ref-O1-RNA is sequestered by IFN-O1’-RNA before it can react with the reporter. b Illustrations of desired responses of the digital biosensor. Left: when [IFN-γ] is high, reacted reporter rises concentration should increase rapidly until all reporter is reacted, When [IFN-γ] is low it should rise slowly or not at all. Right: the concentration of reacted reporter should therefore either be fully ON or very low (OFF) at the end of the reaction. c Simulated kinetics of IFN-C-50-O1. d Experimentally measured kinetics of IFN-C-50-O1. e Dose–response curve of IFN-C-50-O1 after 240 min of reaction. Dashed lines represent simulations, whereas the points represent experimental values. f The dose–response curves of IFN-C-30-O1, IFN-C-50-O1, and IFN-C-100-O1. g The comparator is designed to be K+-insensitive. IFN-O1’-dART and Ref-O1-dART both form G-quadruplexes, and their transcription rates should change similarly with [K+] concentrations so the output of the comparator should be relatively insensitive to [K +]. h Dose–response curves of IFN-C-50-O1 in (g); responses were measured after 240 min. The line plots are added to aid visualization of the trends in the reacted reporter concentrations for each [K +] concentration. i A [K +]-sensitive comparator. At high [K +], the aptamer domain on the IFN-O1’-dART forms a G-quadruplex, which reduces its transcription rate. If the IFN-O1’-dART’s output is compared to an RNA produced at a rate insensitive to [K +] (from Dummy-O1-dART) output should change with [K +]. j Dose–response curves of IFN-C-50-O1 with Dummy-O1-dART for different [K +]; responses were measured after 240 min. The line plots are added to aid visualization of the trends for each [K+] concentration. For (e), (f), (h), and (j), three individual technical replicates are plotted. Lines connect the average reacted reporter concentrations of three technical replicates per [IFN-γ]. Shaded regions indicate ± one s.d. (N = 3). See Supplementary Information Note 6.2, 6.3, and 6.4 for simulation equations and parameters.

Fig. 6. Sensitive protein detection with amplified comparators.

a Circuit diagram for an amplified IFN-C1’-dART/Ref-C1-dART comparator. The Ref-C1-RNA output, which is high when IFN-γ exceeds a threshold concentration, activates transcription of a genelet that transcribes O4, which is detected by a reporter. b The reactions through which Ref-C1-RNA coactivates G1O4 to produce the O4 output. c Reacted reporter kinetics of IFN-AC-50-O4, which consists of 25 nM Ref-C1-dART, 50 nM IFN-C1’-dART, 100 nM G1O4:B1, 200 nM A1, and 2000 nM O4 DNA reporter, for 0 nM or 50 nM IFN-γ. d Reacted reporter kinetics of the IFN-AC-1-O4, which consists of 0.5 nM of Ref-C1-dART, 1 nM IFN-C1’-dART, 20 nM of G1O4:B1, and 100 nM of A1, and 250 nM of O4 DNA reporter for IFN-γ inputs from 0 to 4 nM. 4 U µL⁻¹ T7 RNAP and 4 × 10⁻³ U µL⁻¹ RNase H were used in IFN-AC-1-O4. e Comparison of the dose–response curves of IFN-AC-1-O4, IFN-C-30-O1, IFN-C-50-O1, and IFN-C-100-O1. Three individual technical replicates are plotted. Lines connect the average reacted reporter concentrations of three technical replicates per [IFN-γ]. Shaded regions indicate the average value of three technical replicates ± one s.d. (N = 3).