Deaminase-based RNA sensors

Abstract

Disclosed herein is an RNA editing tool for use in a system designed to measure RNA and manipulate specific cell types in vivo.The RNA sensor system comprises: a) a single-stranded RNA (ssRNA) sensor that comprises a stop codon and a payload, and optionally further comprises a normalization gene; and b) an adenosine deaminase that acts on RNA (ADAR), where the sensor can bind to target ssRNA to form a double-stranded RNA (dsRNA) that is a substrate for ADAR deaminase; the substrate comprises a mismatch within the stop codon; the mismatch can be edited by ADAR deaminase, and the edit can effectively remove the stop codon, allowing the payload to be translated and expressed.Also disclosed is a method for quantifying ribonucleic acid (RNA) levels using the RNA sensor system.

Description

[Technical Field]

[0001] This application claims the benefit of U.S. Provisional Application No. 63 / 267,177, filed January 26, 2022, and U.S. Provisional Application No. 63 / 210,829, filed June 15, 2021, the entireties of which are incorporated herein by reference.

[0002] STATEMENT REGARDING FEDERALLY SPONSORED RESEARCH OR DEVELOPMENT This invention was made with government support awarded by the National Institutes of Health (NIH) under Grant No. 5DP5OD024583. The U.S. Federal Government has certain rights in this invention. [Background technology]

[0003] In recent years, the ability to precisely and programmably edit nucleic acids has been refined. New technologies enable this precision editing in vivo, opening the possibility of treating patients at the genotype level. However, no viable tools exist for measuring and tracking RNA levels in vivo without genetic manipulation or tagging. Genetic manipulation often requires genome manipulation, which has unpredictable consequences for expression and overall cellular activity, rather than a strictly observable sensor system where changes can be measured. Furthermore, manipulation at the genome level to fully integrate sensors requires transgenic organisms, an approach that is impractical for many experimental designs. Summary of the Invention [Problem to be solved by the invention]

[0004] While recent advances have enabled the determination of numerous specific cell types, the ability to track and manipulate these cells remains lacking. This disclosure relates to RNA editing tools for use in systems designed to measure RNA and manipulate specific cell types in vivo. [Means for solving the problem]

[0005] (Summary of the Invention) The present disclosure is directed to an RNA sensor system. The present disclosure provides an RNA sensor system comprising: a) a single-stranded RNA (ssRNA) sensor comprising a stop codon and a payload, and optionally further comprising a normalization gene; and b) an adenosine deaminase (ADAR) acting on RNA; the sensor can bind to target ssRNA to form double-stranded RNA (dsRNA) that serves as a substrate for the ADAR deaminase; the substrate comprises a mismatch within the stop codon; the mismatch can be edited by the ADAR deaminase, and the edit can effectively remove the stop codon, allowing the payload to be translated and expressed.

[0006] The present disclosure presents an RNA sensor system in which mispairing between the ssRNA sensor and the target ssRNA involves an adenine:cytidine mispairing within the dsRNA duplex.

[0007] The present disclosure provides an RNA sensor system, wherein the mismatch between ssRNA sensor and target ssRNA comprises adenine:cytidine mismatch, and ADAR deaminase edits adenine into inosine in the mismatch of dsRNA duplex.In one embodiment, the RNA system comprises more than one mismatch.

[0008] The sensor strand of the RNA sensor system can include a payload comprising a reporter protein, a transcription factor, an enzyme, a transgene protein, or a therapeutic protein. The present disclosure also provides an RNA sensor system in which the payload comprises a fluorescent reporter. The present disclosure also provides an RNA sensor system in which the payload comprises an EGFP reporter or a luciferase reporter. The present disclosure also provides an RNA sensor system in which the payload comprises a caspase.

[0009] The present disclosure also provides an RNA sensor system in which the ADAR is an endogenous ADAR or an exogenous ADAR.The present disclosure also provides an RNA sensor system in which the ADAR is a modified ADAR.

[0010] The present disclosure also provides RNA sensor systems in which the ADAR comprises a programmable A to I(G) replacement RNA editing (REPAIR) molecule, a Cas13b-ADAR fusion molecule, a Cas13d-ADAR fusion molecule, a Cas7-11-ADAR fusion molecule, and an MS2-ADAR fusion molecule, the deaminase domain of ADAR2, full-length ADAR2, or truncated ADAR2.

[0011] The present disclosure also provides an RNA sensor system comprising a plurality of RNA sensors.

[0012] The disclosure also provides an AND gate comprising: a) an AND gate comprising one or more payloads and an ssRNA sensor comprising multiple stop codons complementary to different target ssRNAs, wherein the ssRNA sensor is capable of binding to the target ssRNA to form a double-stranded RNA (dsRNA) that is a substrate for an ADAR deaminase; the substrate comprises a mismatch within each stop codon; and the mismatch within each stop codon is editable by an ADAR deaminase, the editing effectively removing the stop codon to allow translation and expression of one or more payloads; and / or b) an AND gate comprising multiple independent ssRNAs Also presented is a cellular logic system comprising an OR gate comprising multiple independent ssRNA sensors, each of the sensors comprising a payload and a stop codon complementary to one or more different target RNAs, where each ssRNA sensor is capable of binding to a different target ssRNA to form a double-stranded RNA (dsRNA) that is a substrate for an ADAR deaminase; the substrates contain mismatches within each stop codon; and the mismatches within each stop codon are editable by an ADAR deaminase, where the editing can effectively remove the stop codon to allow translation and expression of one or more payloads.The present disclosure also provides an AND gate comprising: a) one or more payloads and an ssRNA sensor comprising multiple stop codons complementary to different target ssRNAs, wherein the ssRNA sensor is capable of binding to the target ssRNA to form a double-stranded RNA (dsRNA) that is a substrate for an ADAR deaminase; the substrate comprises a mismatch within each stop codon; and the mismatch within each stop codon is editable by an ADAR deaminase, the editing effectively removing the stop codon to allow translation and expression of one or more payloads; or b) a payload Also disclosed is a cellular logic system comprising an OR gate comprising a load and multiple independent ssRNA sensors comprising stop codons complementary to one or more different target RNAs, wherein each ssRNA sensor is capable of binding to the target ssRNA to form a double-stranded RNA (dsRNA) that is a substrate for an ADAR deaminase; the substrates contain mismatches within each stop codon; and the mismatches within each stop codon are editable by the ADAR deaminase, where the editing can effectively remove the stop codons to allow translation and expression of one or more payloads.

[0013] The present disclosure provides a method for detecting or quantifying ribonucleic acid (RNA) levels with an RNA sensor system, the method comprising the steps of: a) providing a single-stranded RNA (ssRNA) sensor comprising a stop codon and a payload, and optionally further comprising a normalization gene; and b) providing an adenosine deaminase acting on RNA (ADAR); the sensor is capable of binding to a target ssRNA of interest to form a double-stranded RNA (dsRNA) that is a substrate for the ADAR deaminase; the substrate comprises a mismatch within the stop codon; and the mismatch is editable by the ADAR deaminase, which editing can effectively remove the stop codon to allow translation and expression of the payload.

[0014] The present disclosure also presents an RNA sensor system in which the mispairing comprises an adenine:cytidine mispairing within a dsRNA duplex.

[0015] The present disclosure provides a method for detecting or quantifying ribonucleic acid (RNA) levels using an RNA sensor system in which mismatches include adenine mismatches with cytidine, and ADAR deaminase edits the adenine to inosine in a dsRNA duplex. The present disclosure provides a method for detecting or quantifying ribonucleic acid (RNA) levels using an RNA sensor system that includes more than one mismatch.

[0016] The present disclosure provides methods for detecting or quantifying ribonucleic acid (RNA) levels with RNA sensor systems that contain payloads translated into reporter proteins, transcription factors, enzymes, transgene proteins, or therapeutic proteins.

[0017] The present disclosure also provides an RNA sensor system, wherein the ADAR is an endogenous ADAR or an exogenous ADAR. In some embodiments, the ADAR is a modified ADAR. In some embodiments, the ADAR is an ADAR that is endogenous to the cell type in which the sensor can be used.

[0018] The present disclosure also provides an RNA sensor system comprising: a) a single-stranded RNA (ssRNA) sensor comprising at least a first stop codon and a payload, and optionally further comprising a normalization gene; and b) an adenosine deaminase acting on RNA (ADAR); the sensor is capable of binding to the target ssRNA to form a double-stranded RNA (dsRNA) that is a substrate for the ADAR deaminase; the substrate comprises a mismatch within the first stop codon; and the mismatch is editable by the ADAR deaminase, which editing can effectively remove the stop codon to allow translation and expression of the payload.

[0019] The present disclosure also provides an RNA sensor system, wherein the single-stranded RNA sensor comprises more than one stop codon.The present disclosure also provides an RNA sensor system, wherein the single-stranded RNA sensor further comprises a second stop codon.The present disclosure also provides an RNA sensor system, wherein the single-stranded RNA sensor further comprises a third stop codon.

[0020] The present disclosure also provides an RNA sensor system in which the mismatch comprises CCA on the target strand and TAG / UAG on the sensor strand. The present disclosure also provides an RNA sensor system in which the sensor strand comprises a stop codon, TAG / UAG, but does not mismatch with the CCA codon on the target strand. The present disclosure also provides an RNA sensor system in which the sensor strand comprises a stop codon that can create a match or mismatch with a codon on the target strand selected from the group consisting of ACA, ACT, ACC, ACG, TCA, TCT, TCC, TCG, GCA, GCT, GCC, GCG, CCA, CCT, CCC, and CCG.

[0021] The present disclosure also provides an RNA sensor system, which comprises: a) a single-stranded RNA (ssRNA) sensor comprising a stop codon and a payload, and optionally further comprising a normalization gene; and b) an adenosine deaminase (ADAR) that acts on RNA; the sensor can bind to target ssRNA to form double-stranded RNA (dsRNA) that is a substrate for ADAR deaminase; the substrate comprises a stop codon that can be edited by ADAR deaminase, and editing can effectively remove the stop codon so as to allow payload translation and expression.In some embodiments, the ssRNA sensor comprises a stop codon TAG / UAG.In some embodiments, the stop codon TAG / UAG forms a dsRNA duplex with target ssRNA at a codon with the formula nCn, where n is any nucleotide, and C is cytidine.

[0022] The present disclosure also provides an RNA sensor system described herein, wherein the ssRNA sensor is 50 nucleotides or more, 100 nucleotides or more, 150 nucleotides or more, 200 nucleotides or more, 250 nucleotides or more, 300 nucleotides or more, or 500 nucleotides or more. In some embodiments, the ssRNA sensor is 51 nucleotides. In some embodiments, the ssRNA sensor is 81 nucleotides. In some embodiments, the ssRNA sensor is 171 nucleotides. In some embodiments, the ssRNA sensor is 225 nucleotides. In some embodiments, the ssRNA sensor is 279 nucleotides. In some embodiments, the ssRNA sensor is longer than 279 nucleotides.

[0023] The present disclosure also provides an RNA sensor system as described herein, wherein the ssRNA sensor is a circular sensor. In some embodiments, the circular sensor is a rolling circle translation sensor. In some embodiments, the circular sensor is a regular circular sensor.

[0024] The present disclosure also provides the RNA sensor system described herein, wherein ssRNA sensor comprises two stop codons.In some embodiments, ssRNA sensor comprises three stop codons.In some embodiments, ssRNA sensor comprises two stop codons, and in this case, only one stop codon is targeted for ADAR editing.In some embodiments, ssRNA sensor comprises three stop codons, and in this case, only one stop codon is targeted for ADAR editing.

[0025] The present disclosure also provides the RNA sensor system described herein, wherein the ssRNA sensor comprises at least one avidity binding region. In some embodiments, the ssRNA sensor comprises at least three avidity binding regions. In some embodiments, the ssRNA sensor comprises at least five avidity binding regions. In some embodiments, the ssRNA sensor comprises at least seven avidity binding regions. In some embodiments, the ssRNA sensor comprises more than seven avidity binding regions. In some embodiments, the avidity binding regions are separated by MS2 hairpin regions.

[0026] The present disclosure also provides the RNA sensor system described herein, wherein the payload comprises Cre recombinase. In some embodiments, the payload comprises a Cas protein. In some embodiments, the payload comprises Cas9. In some embodiments, the payload comprises a transcription factor. In some embodiments, the payload comprises an ADAR payload. In some embodiments, the payload is a reporter for cellular stress response.

[0027] The present disclosure also provides a composition comprising the RNA sensor system described herein and a delivery vehicle.In some embodiments, the composition comprises an RNA sensor system and lipid nanoparticles, wherein the RNA sensor system comprises: a) a single-stranded RNA (ssRNA) sensor comprising a stop codon and a payload, and optionally further comprising a normalization gene; and b) an adenosine deaminase (ADAR) that acts on RNA; the sensor can bind to target ssRNA to form double-stranded RNA (dsRNA) that is the substrate for ADAR deaminase; the substrate comprises a stop codon that can be edited by ADAR deaminase, and editing can effectively remove the stop codon, so as to allow payload translation and expression, and the RNA sensor system is encapsulated in lipid nanoparticles.

[0028] The present disclosure also provides a method for killing a specific cell or a specific cell type, comprising providing a single-stranded RNA (ssRNA) sensor or guide comprising a stop codon and a payload, and optionally further comprising a normalization gene; the payload is a self-dimerizing caspase, and the ssRNA sensor or guide is capable of binding to the target ssRNA to form a double-stranded RNA (dsRNA) that is a substrate for adenosine deaminase acting on RNA (ADAR), and the target ssRNA is enriched for expression in the specific cell or cell type.

[0029] The present disclosure also provides an RNA sensor system, which comprises: a) an RNA sensor comprising a stop codon and a payload, and optionally further comprising a normalization gene; and b) an adenosine deaminase (ADAR) acting on RNA; the sensor can bind to target RNA to form a double-stranded RNA (dsRNA) region that serves as a substrate for ADAR deaminase; the substrate comprises a mismatch within the stop codon; the mismatch can be edited by ADAR deaminase, and the edit can effectively remove the stop codon, allowing payload translation and expression.In some embodiments, the RNA sensor is single-stranded RNA.In some embodiments, the RNA sensor comprises one or more double-stranded RNA (dsRNA) domains.In some embodiments, the target RNA is single-stranded RNA.In some embodiments, the RNA comprises one or more double-stranded RNA (dsRNA) domains.

[0030] The present disclosure also provides an RNA sensor system described herein, wherein the RNA sensor is 50 nucleotides or more, 100 nucleotides or more, 150 nucleotides or more, 200 nucleotides or more, 250 nucleotides or more, 300 nucleotides or more, or 500 nucleotides or more. In some embodiments, the RNA sensor is 51 nucleotides. In some embodiments, the ssRNA sensor is 81 nucleotides. In some embodiments, the ssRNA sensor is 171 nucleotides. In some embodiments, the ssRNA sensor is 225 nucleotides. In some embodiments, the ssRNA sensor is 279 nucleotides. In some embodiments, the ssRNA sensor is longer than 279 nucleotides.

[0031] The present disclosure also provides the RNA sensor system described herein, wherein the ssRNA sensor is a circular sensor.In some embodiments, the circular sensor is a rolling circle translation sensor.In some embodiments, the circular sensor is a conventional circular sensor.In some embodiments, the RNA sensor comprises two stop codons.In some embodiments, the RNA sensor comprises three stop codons.In some embodiments, the RNA sensor comprises two stop codons, and in this case, only one stop codon is targeted for ADAR editing.In some embodiments, the RNA sensor comprises three stop codons, and in this case, only one stop codon is targeted for ADAR editing.

[0032] The present disclosure also provides the RNA sensor system described herein, wherein the RNA sensor comprises at least one avidity binding region. In some embodiments, the RNA sensor comprises at least three avidity binding regions. In some embodiments, the RNA sensor comprises at least five avidity binding regions. In some embodiments, the RNA sensor comprises at least seven avidity binding regions. In some embodiments, the RNA sensor comprises more than seven avidity binding regions. In some embodiments, the avidity binding regions are separated by MS2 hairpin regions.

[0033] In some embodiments, the payload comprises a Cre recombinase. In some embodiments, the payload comprises a Cas protein. In some embodiments, the payload comprises Cas9. In some embodiments, the payload comprises a transcription factor. In some embodiments, the payload comprises an ADAR. In some embodiments, the payload is a reporter for a cellular stress response.

[0034] In some embodiments, the ADAR is selected from the group consisting of ADAR2, ADAR1, ADAR1 p150, ADAR1 p110, ADAR2 R455G, ADAR2 R455G, ADAR2 S486T, ADAR2 T375G E488Q T490A, ADAR2 T375G, ADAR2 T375S, ADAR2 N473D, ADAR2 deaminase domain, ADAR2 T490S, ADAR2 T490A, MCP-ADAR2 deaminase domain, ADAR2 R455E, ADAR2 T375G T490A, ADAR2 E488Q, MCP-ADAR2 deaminase domain E488Q T490A, ADAR2 R510E, ADAR2 R455S, ADAR2 V351L, and derivatives or engineered variants thereof. In some embodiments, the ADAR is endogenously expressed in the target cell in which the RNA sensor system may be used. [Brief explanation of the drawings]

[0035] [Figure 1] This is a graphical illustration of how an RNA sensor using ADAR technology can generate new gene output. The sensor RNA contains an optional marker protein, a guide RNA region with a stop codon (red octagon), and a downstream payload protein. The sensor's association with the target RNA forms a duplex with an AC mismatch, which serves as a substrate for RNA editing by the ADAR protein (brown). For example, RNA editing can convert the stop codon UAG to UIG, allowing translation of the payload (green protein). [Figure 2A] Figure 2A is a graphical illustration of dual-transcript ADAR sensor design. Luciferase dual-transcript ADAR sensor contains a normalization protein under constitutive expression and a payload protein under the control of the ADAR sensor. The fold change ratio can be calculated by calculating the activation (gluc / cluc) of the normalization sensor, and then normalizing it to the ratio value in the absence of target. The target can be delivered through exogenous transfection under the control of a doxycycline-inducible promoter. The sensor can either recruit endogenous ADAR to sense the target EGFP transcript, or utilize exogenously delivered ADAR to sense the target transcript with enhanced sensitivity. [Figure 2B] FIG. 2B is a graphical illustration of a comparison of the fold increase in activation of luciferase levels in HEK293FT cells transfected with an eGFP plasmid or a control plasmid, as well as a sensor strand that recognizes eGFP, in the presence or absence of a supplemented ADAR. [Figure 2C] FIG. 2C is a graphical illustration comparing luciferase values ​​for the untargeted sensor, the targeted sensor, or the constitutively active plasmid in the presence or absence of supplemented ADAR. [Figure 2D]FIG. 2D is a graphical illustration of next-generation sequencing results quantifying editing of the stop codon UAG in targeted sensors, non-targeted sensors, and constitutively active plasmids. [Figure 3A] Figure 3A is a schematic diagram showing a fluorescent ADAR sensor with a single-transcript design containing a constitutively expressed normalizing fluorescent protein (mCherry) upstream of an ADAR sensor guide controlling a second fluorescent protein (mNeon). Due to dual fluorescence on a single transcript, non-functional eGFP must be used with this format. [Figure 3B] Figure 3B is a series of representative images of HEK293FT cells transfected with targeted or non-targeted sensors, non-functional eGFP, with or without supplemented ADARs, and Figure 3C is quantification of the fold increase in mNeon activation by measuring fluorescence. [Figure 3C] FIG. 3C is a quantification of the fold increase in mNeon activation by measuring fluorescence. [Figure 3D] FIG. 3D is a graphical illustration of next-generation sequencing results quantifying editing of the stop codon UAG in the presence of supplemented ADARs for both non-targeted and targeted sensors. [Figure 4A] Graphical illustration of luciferase expression (A) and fold increase in luciferase expression compared to a negative control (B) in HEK293FT cells. Fold-change ratios can be calculated by calculating the normalized sensor activation (gluc / cluc) and then normalizing to the ratio value in the absence of target. (A) Expression of the Gluc luciferase gene was quantified in HEK293FT cells transfected with two distinct guide strands (Design 2 and Design 4) targeting EGFP or a negative control scrambled sequence (negative control). (B) Increase in EGFP expression compared to the negative control guide strand was quantified. [Figure 4B]Graphical illustration of luciferase expression (A) and fold increase in luciferase expression compared to a negative control (B) in HEK293FT cells. Fold-change ratios can be calculated by calculating the normalized sensor activation (gluc / cluc) and then normalizing to the ratio value in the absence of target. (A) Expression of the Gluc luciferase gene was quantified in HEK293FT cells transfected with two distinct guide strands (Design 2 and Design 4) targeting EGFP or a negative control scrambled sequence (negative control). (B) Increase in EGFP expression compared to the negative control guide strand was quantified. [Figure 5] Figure 2 shows a graphical illustration of the increase in luciferase expression by experimental guide strands 1-4 targeting the EGFP transcript in the presence of endogenous ADAR2 (Figure 3, blue bars), endogenous ADAR2 supplemented with the exogenously supplied deaminase domain of ADAR2 (ADAR2dd; Figure 2, white bars), or a fusion construct expressing the catalytically inactive enzyme Cas13b fused to the deaminase domain of ADAR2 (dPspCas13b-ADAR2dd; Figure 2, red bars). All fold increases are compared to a scrambled guide designed not to target EGFP, used as a negative control. [Figure 6A]Visual representation of the ADAR mutant and catalytic domain mutation screen. (A) From left to right, a schematic diagram of different ADARs tested includes ADAR1p150, ADAR1p110, ADAR2, and MS2 coat protein (MCP)-ADAR fusion protein (MCP-ADAR). Fl = full length. DD = deaminase domain. Catalytic domain mutations are not shown in the schematic but all reside within the deaminase domain. (B) Bar graph showing activation of exogenously transfected sensors in the presence of exogenously transfected iRFP and different ADAR mutants. ADAR mutants selected for target-wide screening are shown in red, along with RNA sequencing data showing conversion of the stop codon TAG to TIG in the presence and absence of target. Error bars indicate standard deviation of n=3 technical replicates. [Figure 6B] Visual representation of the ADAR mutant and catalytic domain mutation screen. (A) From left to right, a schematic diagram of different ADARs tested includes ADAR1p150, ADAR1p110, ADAR2, and MS2 coat protein (MCP)-ADAR fusion protein (MCP-ADAR). Fl = full length. DD = deaminase domain. Catalytic domain mutations are not shown in the schematic but all reside within the deaminase domain. (B) Bar graph showing activation of exogenously transfected sensors in the presence of exogenously transfected iRFP and different ADAR mutants. ADAR mutants selected for target-wide screening are shown in red, along with RNA sequencing data showing conversion of the stop codon TAG to TIG in the presence and absence of target. Error bars indicate standard deviation of n=3 technical replicates. [Figure 7A]7A-7C are graphical representations of ADAR sensor activation by targets eGFP and iRFP. (A) Experiments with ADAR mutants on a 69-nucleotide iRFP sensor. The indicated fold change indicates the fluorescence ratio value (mNeon / mCherry) in the presence of target divided by the ratio value in the absence of target. (B) Non-normalized mNeon / mCherry fluorescence ratio values ​​for the data shown in FIG. 7A. (C) Experiments with ADAR mutants on a 51-nucleotide eGFP sensor. The indicated fold change indicates the fluorescence ratio value (mNeon / mCherry) in the presence of target divided by the ratio value in the absence of target. (D) Non-normalized mNeon / mCherry fluorescence ratio values ​​for the data shown in FIG. 7C. Error bars indicate the standard deviation of n=3 technical replicates. [Figure 7B] 7A-7C are graphical representations of ADAR sensor activation by targets eGFP and iRFP. (A) Experiments with ADAR mutants on a 69-nucleotide iRFP sensor. The indicated fold change indicates the fluorescence ratio value (mNeon / mCherry) in the presence of target divided by the ratio value in the absence of target. (B) Non-normalized mNeon / mCherry fluorescence ratio values ​​for the data shown in FIG. 7A. (C) Experiments with ADAR mutants on a 51-nucleotide eGFP sensor. The indicated fold change indicates the fluorescence ratio value (mNeon / mCherry) in the presence of target divided by the ratio value in the absence of target. (D) Non-normalized mNeon / mCherry fluorescence ratio values ​​for the data shown in FIG. 7C. Error bars indicate the standard deviation of n=3 technical replicates. [Figure 7C]7A-7C are graphical representations of ADAR sensor activation by targets eGFP and iRFP. (A) Experiments with ADAR mutants on a 69-nucleotide iRFP sensor. The indicated fold change indicates the fluorescence ratio value (mNeon / mCherry) in the presence of target divided by the ratio value in the absence of target. (B) Non-normalized mNeon / mCherry fluorescence ratio values ​​for the data shown in FIG. 7A. (C) Experiments with ADAR mutants on a 51-nucleotide eGFP sensor. The indicated fold change indicates the fluorescence ratio value (mNeon / mCherry) in the presence of target divided by the ratio value in the absence of target. (D) Non-normalized mNeon / mCherry fluorescence ratio values ​​for the data shown in FIG. 7C. Error bars indicate the standard deviation of n=3 technical replicates. [Figure 7D] 7A-7C are graphical representations of ADAR sensor activation by targets eGFP and iRFP. (A) Experiments with ADAR mutants on a 69-nucleotide iRFP sensor. The indicated fold change indicates the fluorescence ratio value (mNeon / mCherry) in the presence of target divided by the ratio value in the absence of target. (B) Non-normalized mNeon / mCherry fluorescence ratio values ​​for the data shown in FIG. 7A. (C) Experiments with ADAR mutants on a 51-nucleotide eGFP sensor. The indicated fold change indicates the fluorescence ratio value (mNeon / mCherry) in the presence of target divided by the ratio value in the absence of target. (D) Non-normalized mNeon / mCherry fluorescence ratio values ​​for the data shown in FIG. 7C. Error bars indicate the standard deviation of n=3 technical replicates. [Figure 8A] Graphical depiction of the editing rates at stop codons for the sensor panel within targeted and untargeted groups for (A) exogenous recruitment with MCP-ADAR2dd, (B) exogenous recruitment with ADAR1 p150 isoform, (C) exogenous recruitment with ADAR2, and (D) no exogenous ADAR recruitment. [Figure 8B]Graphical depiction of the editing rates at stop codons for the sensor panel within targeted and untargeted groups for (A) exogenous recruitment with MCP-ADAR2dd, (B) exogenous recruitment with ADAR1 p150 isoform, (C) exogenous recruitment with ADAR2, and (D) no exogenous ADAR recruitment. [Figure 8C] Graphical depiction of the editing rates at stop codons for the sensor panel within targeted and untargeted groups for (A) exogenous recruitment with MCP-ADAR2dd, (B) exogenous recruitment with ADAR1 p150 isoform, (C) exogenous recruitment with ADAR2, and (D) no exogenous ADAR recruitment. [Figure 8D] Graphical depiction of the editing rates at stop codons for the sensor panel within targeted and untargeted groups for (A) exogenous recruitment with MCP-ADAR2dd, (B) exogenous recruitment with ADAR1 p150 isoform, (C) exogenous recruitment with ADAR2, and (D) no exogenous ADAR recruitment. [Figure 9] This heatmap shows the results of an experiment in which HEK293FT cells were transfected with plasmids expressing ADAR p150 and the target transcripts and target-sensing ADAR sensor constructs, as shown on the y-axis and x-axis, respectively. Data shown are fold changes calculated as the fluorescence ratio (mNeon / mCherry) in the target-containing condition divided by the fluorescence ratio in the target-free condition (pUC19). All conditions represent data from n=3 technical replicates. [Figure 10A](A) Selected ADAR variants screened against four different targets combined with their respective RNA sensors. Numbers in the heatmap represent fold-change ratios. All conditions represent data from n=3 technical replicates. (B) Representative images are shown for neuropeptide Y (NPY), a target for ADAR1 p150. Cells were transfected with the NPY sensor, ADAR variant, and target combinations listed around the image. Tenfold and fourfold digitally enhanced image data for HEK293 cells obtained via confocal microscopy. [Figure 10B] (A) Selected ADAR variants screened against four different targets combined with their respective RNA sensors. Numbers in the heatmap represent fold-change ratios. All conditions represent data from n=3 technical replicates. (B) Representative images are shown for neuropeptide Y (NPY), a target for ADAR1 p150. Cells were transfected with the NPY sensor, ADAR variant, and target combinations listed around the image. Tenfold and fourfold digitally enhanced image data for HEK293 cells obtained via confocal microscopy. [Figure 11A]Figure 1 shows graphical illustrations of normalized (A, C, E, G) and non-normalized (B, D, F, H) fluorescence values ​​for target and ADAR sensor combinations when examined in HEK293FT cells. The test targets were iRFP (A, B), eGFP (C, D), neuropeptide Y (E, F), and dCas9 (G, H). (A, C, E, G): Fold change in sensor fluorescence ratio (mNeon / mCherry) representing the target-containing condition normalized to the target-free condition for each target-sensor combination. (B, D, F, H): Non-normalized mNeon / mCherry fluorescence ratio values ​​for each sensor-target combination with different ADAR mutants. Next-generation sequencing data of RNA sensors for UAG to UIG conversion for the targets iRFP and EGFP. Editing % indicates the % of A→I edited reads. All conditions represent data from n=3 technical replicates. [Figure 11B] Figure 1 shows graphical illustrations of normalized (A, C, E, G) and non-normalized (B, D, F, H) fluorescence values ​​for target and ADAR sensor combinations when examined in HEK293FT cells. The test targets were iRFP (A, B), eGFP (C, D), neuropeptide Y (E, F), and dCas9 (G, H). (A, C, E, G): Fold change in sensor fluorescence ratio (mNeon / mCherry) representing the target-containing condition normalized to the target-free condition for each target-sensor combination. (B, D, F, H): Non-normalized mNeon / mCherry fluorescence ratio values ​​for each sensor-target combination with different ADAR mutants. Next-generation sequencing data of RNA sensors for UAG to UIG conversion for the targets iRFP and EGFP. Editing % indicates the % of A→I edited reads. All conditions represent data from n=3 technical replicates. [Figure 11C]Figure 1 shows graphical illustrations of normalized (A, C, E, G) and non-normalized (B, D, F, H) fluorescence values ​​for target and ADAR sensor combinations when examined in HEK293FT cells. The test targets were iRFP (A, B), eGFP (C, D), neuropeptide Y (E, F), and dCas9 (G, H). (A, C, E, G): Fold change in sensor fluorescence ratio (mNeon / mCherry) representing the target-containing condition normalized to the target-free condition for each target-sensor combination. (B, D, F, H): Non-normalized mNeon / mCherry fluorescence ratio values ​​for each sensor-target combination with different ADAR mutants. Next-generation sequencing data of RNA sensors for UAG to UIG conversion for the targets iRFP and EGFP. Editing % indicates the % of A→I edited reads. All conditions represent data from n=3 technical replicates. [Figure 11D] Figure 1 shows graphical illustrations of normalized (A, C, E, G) and non-normalized (B, D, F, H) fluorescence values ​​for target and ADAR sensor combinations when examined in HEK293FT cells. The test targets were iRFP (A, B), eGFP (C, D), neuropeptide Y (E, F), and dCas9 (G, H). (A, C, E, G): Fold change in sensor fluorescence ratio (mNeon / mCherry) representing the target-containing condition normalized to the target-free condition for each target-sensor combination. (B, D, F, H): Non-normalized mNeon / mCherry fluorescence ratio values ​​for each sensor-target combination with different ADAR mutants. Next-generation sequencing data of RNA sensors for UAG to UIG conversion for the targets iRFP and EGFP. Editing % indicates the % of A→I edited reads. All conditions represent data from n=3 technical replicates. [Figure 11E]Figure 1 shows graphical illustrations of normalized (A, C, E, G) and non-normalized (B, D, F, H) fluorescence values ​​for target and ADAR sensor combinations when examined in HEK293FT cells. The test targets were iRFP (A, B), eGFP (C, D), neuropeptide Y (E, F), and dCas9 (G, H). (A, C, E, G): Fold change in sensor fluorescence ratio (mNeon / mCherry) representing the target-containing condition normalized to the target-free condition for each target-sensor combination. (B, D, F, H): Non-normalized mNeon / mCherry fluorescence ratio values ​​for each sensor-target combination with different ADAR mutants. Next-generation sequencing data of RNA sensors for UAG to UIG conversion for the targets iRFP and EGFP. Editing % indicates the % of A→I edited reads. All conditions represent data from n=3 technical replicates. [Figure 11F] Figure 1 shows graphical illustrations of normalized (A, C, E, G) and non-normalized (B, D, F, H) fluorescence values ​​for target and ADAR sensor combinations when examined in HEK293FT cells. The test targets were iRFP (A, B), eGFP (C, D), neuropeptide Y (E, F), and dCas9 (G, H). (A, C, E, G): Fold change in sensor fluorescence ratio (mNeon / mCherry) representing the target-containing condition normalized to the target-free condition for each target-sensor combination. (B, D, F, H): Non-normalized mNeon / mCherry fluorescence ratio values ​​for each sensor-target combination with different ADAR mutants. Next-generation sequencing data of RNA sensors for UAG to UIG conversion for the targets iRFP and EGFP. Editing % indicates the % of A→I edited reads. All conditions represent data from n=3 technical replicates. [Figure 11G]Figure 1 shows graphical illustrations of normalized (A, C, E, G) and non-normalized (B, D, F, H) fluorescence values ​​for target and ADAR sensor combinations when examined in HEK293FT cells. The test targets were iRFP (A, B), eGFP (C, D), neuropeptide Y (E, F), and dCas9 (G, H). (A, C, E, G): Fold change in sensor fluorescence ratio (mNeon / mCherry) representing the target-containing condition normalized to the target-free condition for each target-sensor combination. (B, D, F, H): Non-normalized mNeon / mCherry fluorescence ratio values ​​for each sensor-target combination with different ADAR mutants. Next-generation sequencing data of RNA sensors for UAG to UIG conversion for the targets iRFP and EGFP. Editing % indicates the % of A→I edited reads. All conditions represent data from n=3 technical replicates. [Figure 11H] Figure 1 shows graphical illustrations of normalized (A, C, E, G) and non-normalized (B, D, F, H) fluorescence values ​​for target and ADAR sensor combinations when examined in HEK293FT cells. The test targets were iRFP (A, B), eGFP (C, D), neuropeptide Y (E, F), and dCas9 (G, H). (A, C, E, G): Fold change in sensor fluorescence ratio (mNeon / mCherry) representing the target-containing condition normalized to the target-free condition for each target-sensor combination. (B, D, F, H): Non-normalized mNeon / mCherry fluorescence ratio values ​​for each sensor-target combination with different ADAR mutants. Next-generation sequencing data of RNA sensors for UAG to UIG conversion for the targets iRFP and EGFP. Editing % indicates the % of A→I edited reads. All conditions represent data from n=3 technical replicates. [Figure 12A]Representative images of full-screen 10x images with insets without target (FIG. 12A) or with target (FIG. 12B) for the ADAR p150 image shown in FIG. 10B. Scale bar is 100 μm. [Figure 12B] Representative images of full-screen 10x images with insets without target (FIG. 12A) or with target (FIG. 12B) for the ADAR p150 image shown in FIG. 10B. Scale bar is 100 μm. [Figure 13A] Graphical comparison of normalized luciferase values ​​of the sensor panel in target and non-target groups for (A) exogenous supplementation with MCP-ADAR2dd, (B) exogenous supplementation with ADAR1 p150 isoform, (C) exogenous supplementation with ADAR2, and (D) no exogenous ADAR supplementation. [Figure 13B] Graphical comparison of normalized luciferase values ​​of the sensor panel in target and non-target groups for (A) exogenous supplementation with MCP-ADAR2dd, (B) exogenous supplementation with ADAR1 p150 isoform, (C) exogenous supplementation with ADAR2, and (D) no exogenous ADAR supplementation. [Figure 13C] Graphical comparison of normalized luciferase values ​​of the sensor panel in target and non-target groups for (A) exogenous supplementation with MCP-ADAR2dd, (B) exogenous supplementation with ADAR1 p150 isoform, (C) exogenous supplementation with ADAR2, and (D) no exogenous ADAR supplementation. [Figure 13D] Graphical comparison of normalized luciferase values ​​of the sensor panel in target and non-target groups for (A) exogenous supplementation with MCP-ADAR2dd, (B) exogenous supplementation with ADAR1 p150 isoform, (C) exogenous supplementation with ADAR2, and (D) no exogenous ADAR supplementation. [Figure 14]10 is a heat map of the titration of exogenously supplemented MCP-ADAR2dd sensors versus IL6-targeted sensors against 20 ng of tetracycline-inducible human IL6 transgene transfected into IL6. Fold changes represent the normalized luciferase ratio between the target and no-target groups. [Figure 15] Figure 15 shows a graphical illustration of the fold increase in luciferase expression for guide strand 1 and guide strand 2 targeted to EGFP compared to a scrambled guide strand designed not to specifically target EGFP. Each guide strand was introduced into HEK293T cells. Cells were then probed using Cas13b (Programmable A to I(G) Replacement RNA Editing (REPAIR)), an enzyme fused to the deaminase domain of ADAR2, or its catalytically inactive form (REPAIR K370A). Guide strands were probed in cells containing only endogenous ADAR2 (Figure 15, blue bars), cells containing endogenous ADAR2 in addition to an exogenous catalytically inactive REPAIR molecule (REPAIR K370A; Figure 15, white bars), or cells containing a catalytically active REPAIR molecule (REPAIR; Figure 15, red bars). [Figure 16] This is a heatmap showing the fold increase in luciferase activation (white: lowest fold increase to dark blue: highest fold increase) when examined in HEK293FT cells. The y-axis shows guide strands targeting EGFP with several different designs, varying in length and mismatch. The x-axis shows the exogenous ADAR molecules tested (none = endogenous only; ADAR2 fl = full-length ADAR2; REPAIR = Cas13b, an enzyme fused to the deaminase domain of ADAR2; MS2-ADAR2dd = MS2-binding protein fused to ADAR2dd; dDisCas7-11-ADAR2dd = catalytically inactive Cas7-11 fused to the deaminase domain of ADAR2). [Figure 17A](A) Schematic diagram showing sensors of different lengths screened against the target iRFP transcript. (B) Bar graph showing increasing sensor activation with increasing sensor length. Sensor activation is calculated by dividing the normalized fluorescence (mNeon / mCherry) value in the presence of target by the normalized fluorescence (mNeon / mCherry) value in the absence of target for each sensor. (C) mNeon-positive cell fraction indicates the proportion of cells with expression above a given threshold. All conditions represent data from n=3 technical replicates. [Figure 17B] (A) Schematic diagram showing sensors of different lengths screened against the target iRFP transcript. (B) Bar graph showing increasing sensor activation with increasing sensor length. Sensor activation is calculated by dividing the normalized fluorescence (mNeon / mCherry) value in the presence of target by the normalized fluorescence (mNeon / mCherry) value in the absence of target for each sensor. (C) mNeon-positive cell fraction indicates the proportion of cells with expression above a given threshold. All conditions represent data from n=3 technical replicates. [Figure 17C] (A) Schematic diagram showing sensors of different lengths screened against the target iRFP transcript. (B) Bar graph showing increasing sensor activation with increasing sensor length. Sensor activation is calculated by dividing the normalized fluorescence (mNeon / mCherry) value in the presence of target by the normalized fluorescence (mNeon / mCherry) value in the absence of target for each sensor. (C) mNeon-positive cell fraction indicates the proportion of cells with expression above a given threshold. All conditions represent data from n=3 technical replicates. [Figure 18A](B) Single-cell image analysis similar to fluorescence cytometry for iRFP-targeting ADAR sensors with guide lengths of 69, 249, and 600 nucleotides. Histograms show the population density of mNeon expression across all cells for conditions with (blue) and without (pink) targeted iRFP. The dotted line indicates a constant intensity threshold across all conditions for gating individual cells as mNeon (+) or mNeon (-). Colored boxes indicate the % mNeon-positive cells for conditions with (blue) and without (pink) targeted iRFP. (B) Representative images are shown for (A). Cells were transfected with combinations of targeted iRFP, ADAR p150, and different ADAR sensor guide lengths, as listed around the image. Scale bar: 100 microns. [Figure 18B] (B) Single-cell image analysis similar to fluorescence cytometry for iRFP-targeting ADAR sensors with guide lengths of 69, 249, and 600 nucleotides. Histograms show the population density of mNeon expression across all cells for conditions with (blue) and without (pink) targeted iRFP. The dotted line indicates a constant intensity threshold across all conditions for gating individual cells as mNeon (+) or mNeon (-). Colored boxes indicate the % mNeon-positive cells for conditions with (blue) and without (pink) targeted iRFP. (B) Representative images are shown for (A). Cells were transfected with combinations of targeted iRFP, ADAR p150, and different ADAR sensor guide lengths, as listed around the image. Scale bar: 100 microns. [Figure 19A] Representative image of a full-screen 10x image with inset for the image shown in Figure 18B. Scale bar is 100 μm. [Figure 19B] Representative image of a full-screen 10x image with inset for the image shown in Figure 18B. Scale bar is 100 μm. [Figure 20] 1 is a graphical comparison of different exogenously recruited ADAR variants on an IL6-targeted sensor transiently transfected with a tetracycline-inducible human IL6 transgene. [Figure 21A] 1 is a series of graphical illustrations showing normalized luciferase values ​​of the sensor panel in groups with and without targets for (A) exogenous recruitment with MCP-ADAR2dd(E488Q, T490A), (B) exogenous recruitment with ADAR1 p150 isoform, (C) exogenous recruitment with ADAR2, and (D) no exogenous ADAR recruitment. [Figure 21B] 1 is a series of graphical illustrations showing normalized luciferase values ​​of the sensor panel in groups with and without targets for (A) exogenous recruitment with MCP-ADAR2dd(E488Q, T490A), (B) exogenous recruitment with ADAR1 p150 isoform, (C) exogenous recruitment with ADAR2, and (D) no exogenous ADAR recruitment. [Figure 21C] 1 is a series of graphical illustrations showing normalized luciferase values ​​of the sensor panel in groups with and without targets for (A) exogenous recruitment with MCP-ADAR2dd(E488Q, T490A), (B) exogenous recruitment with ADAR1 p150 isoform, (C) exogenous recruitment with ADAR2, and (D) no exogenous ADAR recruitment. [Figure 21D] 1 is a series of graphical illustrations showing normalized luciferase values ​​of the sensor panel in groups with and without targets for (A) exogenous recruitment with MCP-ADAR2dd(E488Q, T490A), (B) exogenous recruitment with ADAR1 p150 isoform, (C) exogenous recruitment with ADAR2, and (D) no exogenous ADAR recruitment. [Figure 22]1 is a graphical illustration of a comparison between sensors containing a conventional guide and sensors containing a guide containing multiple binding sites for the target human IL6 and an MS2 hairpin loop in HEK293 cells with endogenous ADAR1, exogenously supplemented ADAR1 p150 isoform, full-length ADAR2, or MCP-ADAR2dd (E488Q, T490A). Fold changes are calculated by normalizing luciferase values ​​(Gluc / Cluc) from the target condition to the no-target condition. [Figure 23] 1 is a visual representation of the engineering of the ADAR sensor with the MS2 hairpin loop and avidity region. The addition of the MS2 hairpin loop and avidity enhances the sensitivity and dynamic range of the ADAR sensor. [Figure 24-1] Schematic diagram of the step-by-step creation of a three-avidity ADAR sensor with 5-nucleotide spacing between the avidity guide regions. [Figure 24-2] Schematic diagram of the step-by-step creation of a three-avidity ADAR sensor with 5-nucleotide spacing between the avidity guide regions. [Figure 25A] (A) Schematic of the variation of linker length outside the target region. (B) Graphical representation of the effect of linker length between avidity regions. Linker lengths of 5, 30, and 50 nucleotides between the avidity regions of the MS2 hairpin-connected 5-avidity sensor for IL6 were investigated. [Figure 25B] (A) Schematic of the variation of linker length outside the target region. (B) Graphical representation of the effect of linker length between avidity regions. Linker lengths of 5, 30, and 50 nucleotides between the avidity regions of the MS2 hairpin-connected 5-avidity sensor for IL6 were investigated. [Figure 26A](A) Schematic diagram of the double / single stop codon avidity / MS2 hairpin sensor. (B) Comparison of sensor activation fold between a conventional MS2 hairpin-connected 7-avidity sensor and a double stop codon 7-avidity sensor with an insertion of a 3' downstream stop codon within the most posterior avidity region. [Figure 26B] (A) Schematic diagram of the double / single stop codon avidity / MS2 hairpin sensor. (B) Comparison of sensor activation fold between a conventional MS2 hairpin-connected 7-avidity sensor and a double stop codon 7-avidity sensor with an insertion of a 3' downstream stop codon within the most posterior avidity region. [Figure 27A] (A) Comparison of background versus activation for avidity sensors versus naive ("long") sensors. (B) Scatter plot of fold change versus background luciferase values ​​for avidity sensors versus naive ("long") sensors. [Figure 27B] (A) Comparison of background versus activation for avidity sensors versus naive ("long") sensors. (B) Scatter plot of fold change versus background luciferase values ​​for avidity sensors versus naive ("long") sensors. [Figure 28A] (A) A bar graph showing a comparison of the five-binding site avidity sensor versus the seven-binding site avidity double stop codon sensor across MCP-ADAR2dd (E488Q, T490A) and ADAR1 p150. (B) A bar graph showing a comparison of the activation signal versus the background signal for the seven-binding site avidity single stop codon sensor versus the seven-binding site avidity double stop codon sensor. [Figure 28B](A) A bar graph showing a comparison of the five-binding site avidity sensor versus the seven-binding site avidity double stop codon sensor across MCP-ADAR2dd (E488Q, T490A) and ADAR1 p150. (B) A bar graph showing a comparison of the activation signal versus the background signal for the seven-binding site avidity single stop codon sensor versus the seven-binding site avidity double stop codon sensor. [Figure 29A] Comparison of target mismatch tolerance across all 16 possible mismatches (16 targets containing 5' or 3' nucleotide changes from the canonical CCA) between the naive 51 bp sensor, a 3-avidity sensor design, and a 5-avidity sensor design. (A) Schematic representation of mismatch tolerance. (B) Heatmap showing the log activation fold (blue) and log10 normalized tolerance (red) for different target mismatches compared to the native CCA target for all three sensor designs across the 16 target mismatches. [Figure 29B] Comparison of target mismatch tolerance across all 16 possible mismatches (16 targets containing 5' or 3' nucleotide changes from the canonical CCA) between the naive 51 bp sensor, a 3-avidity sensor design, and a 5-avidity sensor design. (A) Schematic representation of mismatch tolerance. (B) Heatmap showing the log activation fold (blue) and log10 normalized tolerance (red) for different target mismatches compared to the native CCA target for all three sensor designs across the 16 target mismatches. [Figure 30] 1 is a heatmap displaying the normalized preference of sensor designs among each target mismatch combination between the naive 51 bp sensor, the triple-binding site type sensor, and the five-binding site type sensor. [Figure 31A](A) Visual representation of the creation and activation of circular sensors. Conventional circular sensors are created using a twister ribozyme backbone driven by a U6 promoter for in vitro self-circularization. Self-circularization of the sensor-HiBit tag utilizes mammalian cell RtcB ligase. A rolling-circle translation form of the circular sensor is created by deleting the stop codon at the C-terminus of the HiBit protein and inserting a T2A peptide, which allows cyclic readthrough of the ribosome. (B) Sensors of various lengths, between 50 and 120 nucleotides, are compared in terms of fold change in sensor activation upon induction of the transgene target (human IL-6). [Figure 31B] (A) Visual representation of the creation and activation of circular sensors. Conventional circular sensors are created using a twister ribozyme backbone driven by a U6 promoter for in vitro self-circularization. Self-circularization of the sensor-HiBit tag utilizes mammalian cell RtcB ligase. A rolling-circle translation form of the circular sensor is created by deleting the stop codon at the C-terminus of the HiBit protein and inserting a T2A peptide, which allows cyclic readthrough of the ribosome. (B) Sensors of various lengths, between 50 and 120 nucleotides, are compared in terms of fold change in sensor activation upon induction of the transgene target (human IL-6). [Figure 32A] (A) Schematic diagram of assessed RNA modifications. (B) Heatmap comparing different mRNA modifications for synthetic mRNA ADAR sensors detecting IL6 transgene expression in HEK293FT cells supplemented with MCP-ADAR2dd(E488Q, T490A) by transient transfection of the plasmid 24 hours prior to mRNA sensor transfection. (C) Heatmap comparing different mRNA modifications for synthetic mRNA ADAR sensors detecting IL6 transgene expression in HEK293FT cells supplemented with MCP-ADAR2dd(E488Q, T490A) mRNA at the time of sensor transfection. [Figure 32B] (A) Schematic diagram of assessed RNA modifications. (B) Heatmap comparing different mRNA modifications for synthetic mRNA ADAR sensors detecting IL6 transgene expression in HEK293FT cells supplemented with MCP-ADAR2dd(E488Q, T490A) by transient transfection of the plasmid 24 hours prior to mRNA sensor transfection. (C) Heatmap comparing different mRNA modifications for synthetic mRNA ADAR sensors detecting IL6 transgene expression in HEK293FT cells supplemented with MCP-ADAR2dd(E488Q, T490A) mRNA at the time of sensor transfection. [Figure 32C] (A) Schematic diagram of assessed RNA modifications. (B) Heatmap comparing different mRNA modifications for synthetic mRNA ADAR sensors detecting IL6 transgene expression in HEK293FT cells supplemented with MCP-ADAR2dd(E488Q, T490A) by transient transfection of the plasmid 24 hours prior to mRNA sensor transfection. (C) Heatmap comparing different mRNA modifications for synthetic mRNA ADAR sensors detecting IL6 transgene expression in HEK293FT cells supplemented with MCP-ADAR2dd(E488Q, T490A) mRNA at the time of sensor transfection. [Figure 33A]

[0033] Figures 33A, 33B are graphical illustrations depicting EGFP expression (Figure 33A, 33B) and the fold increase in GFP expression (Figure 33C) in HEK293FT cells. EGFP expression was constitutive or expressed as a gradient using a doxycycline-inducible EGFP construct. HEK203T cells were then exposed to doxycycline at concentrations ranging from 8 ng / mL to 200 ng / mL. [Figure 33B]

[0033] Figures 33A, 33B are graphical illustrations depicting EGFP expression (Figure 33A, 33B) and the fold increase in GFP expression (Figure 33C) in HEK293FT cells. EGFP expression was constitutive or expressed as a gradient using a doxycycline-inducible EGFP construct. HEK203T cells were then exposed to doxycycline at concentrations ranging from 8 ng / mL to 200 ng / mL. [Figure 33C]

[0033] Figures 33A, 33B are graphical illustrations depicting EGFP expression (Figure 33A, 33B) and the fold increase in GFP expression (Figure 33C) in HEK293FT cells. EGFP expression was constitutive or expressed as a gradient using a doxycycline-inducible EGFP construct. HEK203T cells were then exposed to doxycycline at concentrations ranging from 8 ng / mL to 200 ng / mL. [Figure 34A] 1 is a graphical illustration depicting the dose-dependent luciferase activity of guide strand 1 (A) and guide strand 3 (B) as a function of doxycycline dose in HEK293FT cells simultaneously exposed to full-length ADAR2 and a guide strand targeting EGFP under the control of a doxycycline-inducible promoter. [Figure 34B] 1 is a graphical illustration depicting the dose-dependent luciferase activity of guide strand 1 (A) and guide strand 3 (B) as a function of doxycycline dose in HEK293FT cells simultaneously exposed to full-length ADAR2 and a guide strand targeting EGFP under the control of a doxycycline-inducible promoter. [Figure 35A] 1 is a graphical illustration depicting the luciferase activity of guide strand 1 (A) and guide strand 3 (B) as a function of GFP fluorescence in HEK293FT cells simultaneously exposed to full-length ADAR2 and a guide strand targeting EGFP under the control of a doxycycline-inducible promoter. [Figure 35B]1 is a graphical illustration depicting the luciferase activity of guide strand 1 (A) and guide strand 3 (B) as a function of GFP fluorescence in HEK293FT cells simultaneously exposed to full-length ADAR2 and a guide strand targeting EGFP under the control of a doxycycline-inducible promoter. [Figure 36A] (B) A visual representation of the results of combined treatment with tetracycline-inducible IL6 and stable lentiviral integration. Relative expression of IL6 was then quantified using a double stop codon 7 avidity IL6 sensor, and the corresponding luciferase fold change is plotted against the Cq value for IL6 expression detected by quantitative polymerase chain reaction (QCPR). [Figure 36B] (B) A visual representation of the results of combined treatment with tetracycline-inducible IL6 and stable lentiviral integration. Relative expression of IL6 was then quantified using a double stop codon 7 avidity IL6 sensor, and the corresponding luciferase fold change is plotted against the Cq value for IL6 expression detected by quantitative polymerase chain reaction (QCPR). [Figure 37] This scatter plot demonstrates the usefulness of the double stop codon 7 avidity IL6 sensor for quantifying relative IL6 expression due to its large dynamic range. The target expression range was created through a combination of tetracycline-inducible transient overexpression of IL6 and stable lentiviral integration of a tetracycline-IL6 cassette into HEK293FT cells. The fold change in ADAR sensor activity relative to basal conditions is plotted against the change in IL6 gene expression, as determined by quantitative polymerase chain reaction (qPCR). [Figure 38] Scatter plot displaying linear regression of fold change in sensor activation against fold change in gene expression detected by qPCR. [Figure 39] Adenosine editing within the sensor stop codon, UAG, across different IL6 gene expression levels, corresponding to Figure 38. [Figure 40A](A) Schematic representation of an AND gate. (B) Schematic representation of an OR gate. (C) Graphical illustration comparing activation fold change of a naive 51 nucleotide-guided AND gate sensor and a 5 avidity-guided AND gate sensor across all four combinations of targeted IL6 / EGFP induction. [Figure 40B] (A) Schematic representation of an AND gate. (B) Schematic representation of an OR gate. (C) Graphical illustration comparing activation fold change of a naive 51 nucleotide-guided AND gate sensor and a 5 avidity-guided AND gate sensor across all four combinations of targeted IL6 / EGFP induction. [Figure 40C] (A) Schematic representation of an AND gate. (B) Schematic representation of an OR gate. (C) Graphical illustration comparing activation fold change of a naive 51 nucleotide-guided AND gate sensor and a 5 avidity-guided AND gate sensor across all four combinations of targeted IL6 / EGFP induction. [Figure 41A] (A) Graphical illustration of normalized sensor activation of an AND-gated ADAR sensor for EGFP / IL6 transcript input across all four possible target combinations. (B) Graphical illustration of normalized sensor activation of an OR-gated ADAR sensor for EGFP / IL6 transcript input across different target combinations. [Figure 41B] (A) Graphical illustration of normalized sensor activation of an AND-gated ADAR sensor for EGFP / IL6 transcript input across all four possible target combinations. (B) Graphical illustration of normalized sensor activation of an OR-gated ADAR sensor for EGFP / IL6 transcript input across different target combinations. [Figure 42A](A) Schematic diagram of IL6-responsive caspase using an ADAR sensor with a 5-avidity sensor targeting human IL6 transcript. Activation of the sensor expresses FKBP, which causes caspase-9 to self-dimerize. (B) Graphical illustration of the fold change in cell death (apoptosis) in response to ADAR sensor activation by detection of IL6 transcript. The positive control sensor contains a scrambled guide sequence preceding iCaspase without a stop codon in frame. The fold change in cell death is determined by calculating the fold change in cell viability in the target-containing condition compared to the no-target condition. (C) Bar graph comparing the percentage cell viability values ​​of the ADAR sensor and no-stop codon control for IL6-responsive iCaspase in the target-containing and no-target groups. [Figure 42B] (A) Schematic diagram of IL6-responsive caspase using an ADAR sensor with a 5-avidity sensor targeting human IL6 transcript. Activation of the sensor expresses FKBP, which causes caspase-9 to self-dimerize. (B) Graphical illustration of the fold change in cell death (apoptosis) in response to ADAR sensor activation by detection of IL6 transcript. The positive control sensor contains a scrambled guide sequence preceding iCaspase without a stop codon in frame. The fold change in cell death is determined by calculating the fold change in cell viability in the target-containing condition compared to the no-target condition. (C) Bar graph comparing the percentage cell viability values ​​of the ADAR sensor and no-stop codon control for IL6-responsive iCaspase in the target-containing and no-target groups. [Figure 42C](A) Schematic diagram of IL6-responsive caspase using an ADAR sensor with a 5-avidity sensor targeting human IL6 transcript. Activation of the sensor expresses FKBP, which causes caspase-9 to self-dimerize. (B) Graphical illustration of the fold change in cell death (apoptosis) in response to ADAR sensor activation by detection of IL6 transcript. The positive control sensor contains a scrambled guide sequence preceding iCaspase without a stop codon in frame. The fold change in cell death is determined by calculating the fold change in cell viability in the target-containing condition compared to the no-target condition. (C) Bar graph comparing the percentage cell viability values ​​of the ADAR sensor and no-stop codon control for IL6-responsive iCaspase in the target-containing and no-target groups. [Figure 43A] This is a visual representation of an experiment examining the efficiency of ADAR sensors in a heat shock assay. (A) Heat shock of Hela cells at 42°C is used to induce upregulation of HSP40 / HSP70 gene expression. Hela cells are transfected with HSP40 / HSP70-targeted sensors alone or with MCP-ADAR2dd(E488Q, T490A) and then maintained at 42°C or 37°C for 24 hours. (B) qPCR verification of upregulated HSP40 / HSP70 levels after 24 hours of heat shock. (C) Sensor activation is calculated between the 42°C and 37°C groups and normalized to the sensor with a scrambled non-targeted guide to account for changes in protein degradation. [Figure 43B]This is a visual representation of an experiment examining the efficiency of ADAR sensors in a heat shock assay. (A) Heat shock of Hela cells at 42°C is used to induce upregulation of HSP40 / HSP70 gene expression. Hela cells are transfected with HSP40 / HSP70-targeted sensors alone or with MCP-ADAR2dd(E488Q, T490A) and then maintained at 42°C or 37°C for 24 hours. (B) qPCR verification of upregulated HSP40 / HSP70 levels after 24 hours of heat shock. (C) Sensor activation is calculated between the 42°C and 37°C groups and normalized to the sensor with a scrambled non-targeted guide to account for changes in protein degradation. [Figure 43C] This is a visual representation of an experiment examining the efficiency of ADAR sensors in a heat shock assay. (A) Heat shock of Hela cells at 42°C is used to induce upregulation of HSP40 / HSP70 gene expression. Hela cells are transfected with HSP40 / HSP70-targeted sensors alone or with MCP-ADAR2dd(E488Q, T490A) and then maintained at 42°C or 37°C for 24 hours. (B) qPCR verification of upregulated HSP40 / HSP70 levels after 24 hours of heat shock. (C) Sensor activation is calculated between the 42°C and 37°C groups and normalized to the sensor with a scrambled non-targeted guide to account for changes in protein degradation. [Figure 44A]Figure 1 shows a visual representation of the analysis of SERPINA1 in three cell types that differentially express SERPINA1. (A) A bar graph comparing SERPINA1 expression across HEK293FT, HepG2, and Hela cells. (B) A SERPINA1-sensing 5-avidity sensor is transfected into three different cell types (HEK293FT, Hela, and HepG2) with or without exogenous MCP-ADAR2dd (E488Q, T490A). (C) A bar graph comparing the fold change in sensor activation between Hela and HepG2 cells across SERPINA1 sensors targeting different CCA sites. (D) Sensor activation is determined by calculating raw luciferase values ​​from the SERPINA1 sensor normalized to the scrambled non-targeting guide sensor to account for differences in protein production / secretion and background ADAR activity between cell types, followed by normalization to the gluc / cluc ratio in HEK293FT cells. [Figure 44B] Figure 1 shows a visual representation of the analysis of SERPINA1 in three cell types that differentially express SERPINA1. (A) A bar graph comparing SERPINA1 expression across HEK293FT, HepG2, and Hela cells. (B) A SERPINA1-sensing 5-avidity sensor is transfected into three different cell types (HEK293FT, Hela, and HepG2) with or without exogenous MCP-ADAR2dd (E488Q, T490A). (C) A bar graph comparing the fold change in sensor activation between Hela and HepG2 cells across SERPINA1 sensors targeting different CCA sites. (D) Sensor activation is determined by calculating raw luciferase values ​​from the SERPINA1 sensor normalized to the scrambled non-targeting guide sensor to account for differences in protein production / secretion and background ADAR activity between cell types, followed by normalization to the gluc / cluc ratio in HEK293FT cells. [Figure 44C] Figure 1 shows a visual representation of the analysis of SERPINA1 in three cell types that differentially express SERPINA1. (A) A bar graph comparing SERPINA1 expression across HEK293FT, HepG2, and Hela cells. (B) A SERPINA1-sensing 5-avidity sensor is transfected into three different cell types (HEK293FT, Hela, and HepG2) with or without exogenous MCP-ADAR2dd (E488Q, T490A). (C) A bar graph comparing the fold change in sensor activation between Hela and HepG2 cells across SERPINA1 sensors targeting different CCA sites. (D) Sensor activation is determined by calculating raw luciferase values ​​from the SERPINA1 sensor normalized to the scrambled non-targeting guide sensor to account for differences in protein production / secretion and background ADAR activity between cell types, followed by normalization to the gluc / cluc ratio in HEK293FT cells. [Figure 44D]Figure 1 shows a visual representation of the analysis of SERPINA1 in three cell types that differentially express SERPINA1. (A) A bar graph comparing SERPINA1 expression across HEK293FT, HepG2, and Hela cells. (B) A SERPINA1-sensing 5-avidity sensor is transfected into three different cell types (HEK293FT, Hela, and HepG2) with or without exogenous MCP-ADAR2dd (E488Q, T490A). (C) A bar graph comparing the fold change in sensor activation between Hela and HepG2 cells across SERPINA1 sensors targeting different CCA sites. (D) Sensor activation is determined by calculating raw luciferase values ​​from the SERPINA1 sensor normalized to the scrambled non-targeting guide sensor to account for differences in protein production / secretion and background ADAR activity between cell types, followed by normalization to the gluc / cluc ratio in HEK293FT cells. [Figure 45] FIG. 16 is a graphical representation of the normalized fold change in editing rates for 5-avidity sensors for SERPINA1 in different cell types (HEK293, Hela, and HepG2). [Figure 46] 1 is a bar graph comparing the fold change in mRNA SERPINA1 sensor activation targeting different CCA sites on the SERPINA1 transcript in transiently transfected Hepa1-6 cells in which SERPINA1 expression was induced by tetracycline. [Figure 47A](A) Schematic illustration of an in vivo sensing experiment for human SERPINA1 transcripts using a mADAR sensor construct for SERPINA1. SERPINA1-targeted sensor mRNA with Akaluc output is generated in vitro with 25% 5-methylcytosine and 0% pseudouridine. Constitutive Akaluc sensor constructs (without a stop codon) and non-targeted guide sensor constructs (with a stop codon) are synthesized using the same protocol. All mRNAs are packaged in lipid nanoparticles and injected via the tail vein into wild-type or NSG-Piz mice with a human SERPINA1 PiZ mutant cassette. In vivo sensor activation is measured 18 hours after injection. (B) Representative images of sensor activation are shown for various synthetic mRNA ADAR sensor constructs. [Figure 47B] (A) Schematic illustration of an in vivo sensing experiment for human SERPINA1 transcripts using a mADAR sensor construct for SERPINA1. SERPINA1-targeted sensor mRNA with Akaluc output is generated in vitro with 25% 5-methylcytosine and 0% pseudouridine. Constitutive Akaluc sensor constructs (without a stop codon) and non-targeted guide sensor constructs (with a stop codon) are synthesized using the same protocol. All mRNAs are packaged in lipid nanoparticles and injected via the tail vein into wild-type or NSG-Piz mice with a human SERPINA1 PiZ mutant cassette. In vivo sensor activation is measured 18 hours after injection. (B) Representative images of sensor activation are shown for various synthetic mRNA ADAR sensor constructs. [Figure 48A](A) Graphical illustration of Akaluc-generated radiance calculated for liver and compared between wild-type and NSG-PiZ mutant mice. Fold change between NSG-PiZ and WT mice is calculated for each ADAR sensor construct. Significance is determined via a two-tailed t-test, N=2 mice. *: p<0.05. p-values ​​less than 0.05 are indicated by an asterisk for statistical significance. (B) Bar graph comparing Akaluc-generated radiance in NSG-PiZ and WT mice across non-targeted, constitutive, SERPINA1 CCA35-targeted, and SERPINA1 CCA30-targeted sensors. [Figure 48B] (A) Graphical illustration of Akaluc-generated radiance calculated for liver and compared between wild-type and NSG-PiZ mutant mice. Fold change between NSG-PiZ and WT mice is calculated for each ADAR sensor construct. Significance is determined via a two-tailed t-test, N=2 mice. *: p<0.05. p-values ​​less than 0.05 are indicated by an asterisk for statistical significance. (B) Bar graph comparing Akaluc-generated radiance in NSG-PiZ and WT mice across non-targeted, constitutive, SERPINA1 CCA35-targeted, and SERPINA1 CCA30-targeted sensors. [Figure 49A] Graphical illustration of differential gene analysis for 37 tissues using the Human Protein Atlas and GTEX datasets for the minimum number of genes required to classify tissues according to gene (A), as well as the number of protein-coding genes enriched in specific tissues, enhanced in specific tissues, or less specific (B). (C) Heatmap showing the relative transcript abundance of 34 tissue-specific mRNAs across 34 different tissue types. [Figure 49B]Graphical illustration of differential gene analysis for 37 tissues using the Human Protein Atlas and GTEX datasets for the minimum number of genes required to classify tissues according to gene (A), as well as the number of protein-coding genes enriched in specific tissues, enhanced in specific tissues, or less specific (B). (C) Heatmap showing the relative transcript abundance of 34 tissue-specific mRNAs across 34 different tissue types. [Figure 49C] Graphical illustration of differential gene analysis for 37 tissues using the Human Protein Atlas and GTEX datasets for the minimum number of genes required to classify tissues according to gene (A), as well as the number of protein-coding genes enriched in specific tissues, enhanced in specific tissues, or less specific (B). (C) Heatmap showing the relative transcript abundance of 34 tissue-specific mRNAs across 34 different tissue types. [Figure 50A] Graphical illustrations showing the characterization of RADARS safety with respect to immune response and endogenous RNA knockdown. (A, B) Effect of sensor-target duplex formation on the innate antiviral pathway. RADARS sensors were transfected in the presence or absence of complementary target sequences. Total RNA was analyzed using quantitative PCR (qPCR) to determine the relative expression levels of MDA5 (A) and IFN-β (B). (C, D) Effect of sensor-target duplex formation on the abundance of endogenous target transcripts. The relative abundance of NEFM and PPIP transcripts upon transfection of complementary or non-targeting RADARS sensors was assessed by qPCR. Data are presented as mean ± SD (n = 4); non-significant by unpaired, two-tailed Student's t-test: p > 0.05. [Figure 50B]Graphical illustrations showing the characterization of RADARS safety with respect to immune response and endogenous RNA knockdown. (A, B) Effect of sensor-target duplex formation on the innate antiviral pathway. RADARS sensors were transfected in the presence or absence of complementary target sequences. Total RNA was analyzed using quantitative PCR (qPCR) to determine the relative expression levels of MDA5 (A) and IFN-β (B). (C, D) Effect of sensor-target duplex formation on the abundance of endogenous target transcripts. The relative abundance of NEFM and PPIP transcripts upon transfection of complementary or non-targeting RADARS sensors was assessed by qPCR. Data are presented as mean ± SD (n = 4); non-significant by unpaired, two-tailed Student's t-test: p > 0.05. [Figure 50C] Graphical illustrations showing the characterization of RADARS safety with respect to immune response and endogenous RNA knockdown. (A, B) Effect of sensor-target duplex formation on the innate antiviral pathway. RADARS sensors were transfected in the presence or absence of complementary target sequences. Total RNA was analyzed using quantitative PCR (qPCR) to determine the relative expression levels of MDA5 (A) and IFN-β (B). (C, D) Effect of sensor-target duplex formation on the abundance of endogenous target transcripts. The relative abundance of NEFM and PPIP transcripts upon transfection of complementary or non-targeting RADARS sensors was assessed by qPCR. Data are presented as mean ± SD (n = 4); non-significant by unpaired, two-tailed Student's t-test: p > 0.05. [Figure 50D]Graphical illustrations showing the characterization of RADARS safety with respect to immune response and endogenous RNA knockdown. (A, B) Effect of sensor-target duplex formation on the innate antiviral pathway. RADARS sensors were transfected in the presence or absence of complementary target sequences. Total RNA was analyzed using quantitative PCR (qPCR) to determine the relative expression levels of MDA5 (A) and IFN-β (B). (C, D) Effect of sensor-target duplex formation on the abundance of endogenous target transcripts. The relative abundance of NEFM and PPIP transcripts upon transfection of complementary or non-targeting RADARS sensors was assessed by qPCR. Data are presented as mean ± SD (n = 4); non-significant by unpaired, two-tailed Student's t-test: p > 0.05. [Figure 51A] Graphical illustration showing the decay of RADARS signal corresponding to endogenous target knockdown. (A) Schematic diagram of siRNA-mediated knockdown of endogenous transcripts. (B) qPCR and fluorescent RADARS detected differential expression of siRNA targeting PPIB or NEFM compared with control non-targeting siRNA in HEK293FT cells. RADAR sensor activation was calculated for targeting siRNA and normalized to control siRNA. Data are mean ± SD of technical replicates (n≧3). [Figure 51B] Graphical illustration showing the decay of RADARS signal corresponding to endogenous target knockdown. (A) Schematic diagram of siRNA-mediated knockdown of endogenous transcripts. (B) qPCR and fluorescent RADARS detected differential expression of siRNA targeting PPIB or NEFM compared with control non-targeting siRNA in HEK293FT cells. RADAR sensor activation was calculated for targeting siRNA and normalized to control siRNA. Data are mean ± SD of technical replicates (n≧3). [Figure 52]Graphical comparison of fold activation of engineered guide RNAs for #CCA8 IL6 with a 171-nucleotide guide and four MS2 loops when used with exogenously supplemented ADAR1 p150 or endogenous ADAR. Data are mean ± sem of technical replicates (n=3). [Figure 53A] Figure 53A is a graphical comparison of the fold activation of mRNA RADARS sensors for detecting IL6 transcripts following transfection of ADAR1 p150 with a plasmid. Synthetic mRNA sensors are synthesized with different chemically modified bases at different incorporation levels ranging from 0 to 100%. Data are the mean ± sem of technical replicates (n = 3). [Figure 53B] Figure 53B is a graphical comparison of the fold activation of mRNA RADARS activation in detecting IL6 transcripts that utilize endogenous ADARs when synthesized with different chemically modified bases at different incorporation levels ranging from 0 to 100%. Data are means ± sem of technical replicates (n = 3). [Figure 53C] Figure 53C is a graphical comparison of the induction of interferon-beta responses resulting from transfection of mRNA RADARS. Synthetic mRNA was synthesized with different levels of chemically modified bases, and the interferon response was measured by a plasmid (1 avidity Glo luciferase) reporter assay (Gentili et al., 2015). [Figure 54A]Figure 54A is a visual characterization of the evolution of different RADARS sensor designs when used with exogenous ADAR1p150 supplementation. The inset depicts the scaffolds for the different RADARS designs. Fold RADARS activation is calculated as the ratio of Gaussia luciferase (Gluc) luminescence to Cypridina luciferase (Cluc) constitutive luminescence in the presence of target IL6 compared to the absence of target (Gluc / Cluc) (see "Methods"). The sensor, target (IL6), and ADAR1p150 are co-delivered via transient transfection. Data are mean ± sem of technical replicates (n = 3). [Figure 54B] Figure 54B is a graphical comparison of the Gluc / Cluc ratio between on-target and off-target conditions for #CCA8 IL6-targeting engineered guide RNAs with different lengths and different MS2 hairpin loops while maintaining a 5' peptide length of 0 amino acids. Error bars indicate the standard error of the mean (n=3 technical replicates). [Figure 55A] Figure 55A is a graphical comparison of the Gluc / Cluc ratio between target and no-target conditions for 5-avidity binding site (4 MS2 loops) and 9-avidity binding site (8 MS2 loops) engineered guide RNAs with varying 5' peptide lengths. Error bars indicate standard error of the mean (n=3 technical replicates). [Figure 55B] Figure 55B is a graphical comparison of the Gluc / Cluc ratio between no out-of-frame stop codon and two out-of-frame stop codons for five-avidity binding site-based engineered guide RNAs with 200 amino acid 5' peptide residues. The last column represents constitutive gluc driven under the Efl-alpha promoter. Error bars indicate the standard error of the mean (n=3 technical replicates). [Figure 56A]Figure 56A is a graphical depiction of the fold activation of IL6-, EGFP-, and NPY-targeting RADARS with supplementation with exogenous ADAR1p150. For each transcript, 12 engineered guide RNAs were engineered to target different CCA sites across the transcript. The depicted CCA site numbers follow the convention that #CCAx indicates the number of CCA triplicates counted from the 5' end of the transcript coding region. Each dot represents the average of three technical replicates for an individual sensor. The solid horizontal line represents the average value of all 12 engineered guide RNAs. [Figure 56B] Figure 56B graphically depicts the percentage editing of the target adenosine within the stop codon UAG for a non-targeting engineered guide RNA and 14 IL6-targeting engineered guide RNAs tiling the CCA site on IL6 using the RADARSv2 design with the addition of exogenous ADAR1p150 in the presence and absence of the target IL6 transcript. Error bars indicate the standard error of the mean (n=3 technical replicates). [Figure 57] 1 is a graphical depiction of RADARSv2 with engineered guide RNAs targeting high TPM genes (RPS5), low TPM genes (KRAS), or non-targeting scrambled sequences, used with exogenous ADAR1 p150 supplementation or endogenous ADAR to sense downregulation of their corresponding genes via gene-specific siRNA. Fold activation is calculated by the activation of payloads in the targeted siRNA group compared to the non-targeting siRNA group. Data are the mean ± sem of technical replicates (n = 3). [Figure 58A]Figure 58A is a visual representation of a schematic showing a fluorescent output RADARS construct containing a constitutively expressed normalizing fluorescent protein (mCherry) (top) upstream of a RADARS-driven guide RNA that controls the mNeon fluorescent protein, and an image of fluorescent RADARS showing HEK293FT cells expressing the mNeon payload only in the presence of the target transcript (out-of-frame EGFP). HEK293FT cells are transfected with EGFP-targeted RADARS, ADAR1 p150, and with or without target (out-of-frame EGFP) as indicated (bottom). Scale bar: 100 microns. [Figure 58B] Figure 58B is a visual representation of flow cytometry analysis for fluorescent RADARS showing a histogram of mNeon / mCherry fluorescence for HEK293FT cells transfected as in Figure 58A, with beige and blue distributions indicating target absence and target presence, respectively. [Figure 59A] Figure 59A is a visual representation of the gating strategy used for flow cytometry analysis of fluorescent RADARS in HEK293 cells. Gates are set using a control population transfected with pUC19 plasmid. [Figure 59B] Figure 59B is a visual representation of the gate placed on the population of cells transfected with EGFP-targeted RADARS, ADARp150 and pUC19 plasmids. [Figure 59C] Figure 59C is a visual representation of the gate placed on the population of cells transfected with EGFP-targeted RADARS, ADARp150 and EGFP target (frameshift) plasmids. [Figure 60A]Figure 60A is a visual depiction of the fold activation of RADARS relative to basal conditions (0 ng / mL doxycycline in integrated HEK293FT cells) plotted against the change in IL6 gene expression, as determined by quantitative polymerase chain reaction (qPCR), on a log10-log10 scale. The dashed blue line represents the linear regression results for the data. Data are means ± sem of technical replicates (n=3). [Figure 60B] Figure 60B is a visual depiction of RADARS corresponding raw Cq values ​​and fold activation of the IL6 transgene normalized to the Cq number of the GAPDH gene by subtraction. Error bars indicate the standard error of the mean (n=3 biological replicates). [Figure 60C] Figure 60C is a visual depiction of the effect of the best titration of RADARS-engineered guide RNA sensor for IL6 under target conditions on resulting activation and total protein production (gluc / cluc ratio). For conditions below 40 ng, the remaining plasmid amount was replaced with pUC19 plasmid. Error bars indicate the standard error of the mean (n=3 technical replicates). [Figure 61] Figure 1 shows a visual representation of the results of a validation experiment of siRNA knockdown of 10 endogenous transcripts as measured by qPCR expression. Fold change is calculated by the gene expression of the target transcript in the targeted siRNA group compared to the non-targeted siRNA group (n = 3 biological replicates). [Figure 62A]Figure 62A (top) is a visual depiction of gene expression in transcripts per million (TPM) shown on a logarithmic scale across 10 genes ranging from 10,381 TPM (RSP5) to 13 TPM (KRAS). Bottom: Detection of transcripts by RADARSv2 in cells treated with 100 nM of targeted or non-targeting siRNA pools. Bars represent fold activation (Gluc / Cluc ratio of RADARS in the targeted siRNA group compared to the non-targeting siRNA group) for targeted versus non-targeting RADARS constructs. Significance is determined via unpaired t-tests (*: p<0.5; **: p<0.01; ***: p<0.001; ****: p<0.0001) with Welch's correction assuming separate variances for each group between targeted and non-targeted RADARS. [Figure 62B] Figure 62B is a graphical representation of the editing rate of the stop codon UAG in the best-performing sensor for each gene group shown in Figure 62A. An unpaired one-tailed t-test was performed between the targeting siRNA group and the non-targeting siRNA group (*: p<0.05; **: p<0.01; ***: p<0.001; ****: p<0.0001). Error bars indicate the standard error of the mean (n=3 technical replicates). [Figure 63A] Figure 63A is a graphical depiction of the performance of eight randomly selected engineered guide RNAs and eight randomly selected non-targeting engineered guide RNAs targeting ten endogenous transcripts. The RADARS fold activation is calculated by the Gluc / Cluc ratio in the targeted siRNA group being lower than that in the non-targeting siRNA group. The best-performing targeted sensor for each gene is labeled as a yellow sensor and is discussed in Figure 63. The horizontal lines represent the average value for each group. [Figure 63B]Figure 63B is a visual representation of RADARSv2 targeting RPL41, GAPDH, ACTB, HSP90AA1, PPIB, and KRAS, tracking the expression of these transcripts across a range of siRNA concentrations. The blue and beige curves represent the RADARS activation fold (Gluc / Cluc) ratio relative to 0 nM siRNA and the fold change in expression quantified by qPCR, respectively. The gray curve represents the activation fold of a non-targeting engineered guide RNA (non-complementary to the target transcript). Data are the mean ± sem of technical replicates (n = 3). (R values ​​represent Pearson correlation between qPCR and targeted RADARS; *: p < 0.05; **: p < 0.01; ***: p < 0.001). [Figure 64A] Figure 64A is a visual schematic of the upregulation of HSP70, a heat shock protein family gene, upon heat shock at 42 degrees Celsius. [Figure 64B] Figure 64B is a visual depiction of results from an experiment in which HeLa cells were transfected with four HSP70-targeting engineered guide RNAs and a scrambled non-targeting (NT) engineered guide RNA, all targeting different CCA sites, with supplementation of exogenous ADAR1 p150, followed by 24 hours at 42°C or 37°C. qPCR and RADARSv2 detected differences in HSP70 expression between the 37°C (control) and 42°C (heat shock) groups. Sensor activation was calculated between the 42°C and 37°C groups and normalized to the NT condition. Data are mean ± sem of technical replicates (n = 3). [Figure 65A] Figure 65A is a visual schematic diagram of a two-input AND gate in RADARS. [Figure 65B] Figure 65B is a graphical depiction of normalized sensor activation by AND-gated RADARS for EGFP / IL6 transcript input across all four possible target combinations. Data are means ± sem of technical replicates (n = 3). [Figure 66A] Figure 66A is a visual schematic diagram of a two-input OR logic gate in RADARS. [Figure 66B] Figure 66B is a graphical depiction of sensor activation by OR-gated RADARS for all possible EGFP / IL6 transcript input combinations. Data are means ± sem of technical replicates (n = 3). [Figure 67A] Figure 67A is a schematic representation of SERPINA1-targeting RADARS with an inducible caspase-9 payload. [Figure 67B] Figure 67B is a graphical depiction of cell viability in A549, Hela, and HepG2 cells after transfection of RADARS, which senses SERPINA1 and expresses iCaspase9 in combination with exogenous ADAR1p150. The non-targeting control engineered guide RNA contains a scrambled sequence with a stop codon before the payload. Data are the mean ± sem of technical replicates (n = 3). [Figure 67C] Figure 67C is a graphical depiction of cell viability of HepG2, Hela, and A549 cells 48 hours after transfection of iCaspase9 constructs and SERPINA1 or non-targeted RADARS expressing ADAR1 p150, as determined using an MTS assay and normalized to a control transfected with only a GFP-expressing plasmid. [Figure 68]This is a graphical depiction of the results of an experiment in which HeLa cells were transfected with the best HSP70-targeting RADARS construct and a non-targeting (NT) scrambled RADARS construct for HSP70 transcripts without exogenous ADARs, followed by 24 hours at 42°C or 37°C. qPCR and RADARSv2 detected differences in HSP70 expression between the 37°C (control) and 42°C (heat shock) groups. Sensor activation was calculated between the 42°C and 37°C groups and normalized to the NT condition, which was a sensor with a scrambled non-targeting (NT) engineered guide RNA, to compare changes in protein production. Data are the mean ± sem of technical replicates (n=3). [Figure 69A] Figure 69A is a visual schematic of the double-loxP EGFP Cre reporter and IL6 RADARS-CRE. Right: Fluorescence from HEK293FT 48 hours after transfection of the double-loxP EGFP reporter, ADAR1 p150, and IL6-targeted RADARS with Cre payload, with or without the target IL6. Images are shown for the untargeted and targeted conditions. The white scale bar represents 100 microns. [Figure 69B] Figure 69B is a visual depiction of the results of an experiment in which cells from Figure 69A were harvested for flow cytometry analysis for expression of EGFP. [Figure 70A] Figure 70A is a visual schematic of the SERPINA1-targeting RADARS construct with a Cre payload. [Figure 70B] Figure 70B is a visual depiction of the results of the percentage of GFP+ cells analyzed by flow cytometry for Hela, HepG2, and A549 cells 48 hours after transfection of SERPINA1-targeting RADARS constructs expressing Cre together with exogenous ADAR1 p150. Data are means ± sem of technical replicates (n = 3). [Figure 70C] Figure 70C is a visual depiction of EGFP expression quantified by flow cytometry 48 hours after transfection of IL6-targeted RADARS with a CRE reporter, ADAR1p150, and Cre payload, with or without targeting IL6. The distribution of EGFP signal is analyzed by flow cytometry for HeLa, A549, and HepG2 cells. For all three cell types, GFP-positive cells are defined as the cell population with EGFP intensity in the FITC channel greater than 10. [Figure 71A] Figure 71A is a visual representation of bioluminescence images of sensor activation for various synthetic mRNA RADARS constructs. [Figure 71B] Figure 71B is a graphical comparison of Akaluc-induced luminescence radiance across substrate background, constitutive sensors, and #CCA32 SERPINA1-targeted RADARS in the livers of NSG-PiZ and NSG-WT mice. Data are mean ± sem of technical replicates (n = 3). Significance is determined between radiance of NSG-WT samples and radiance of NSG-PiZ samples via an unpaired, two-tailed t-test with N = 3 mice (non-significant: p > 0.05; **: p < 0.01). [Figure 71C] Figure 71C is a graphical comparison of Akaluc luminescence radiance calculated for the liver and compared between wild-type and NSG-PiZ mutant mice. Fold activation between NSG-PiZ and NSG-WT mice is calculated for each RADARS construct. Significance is tested for each group relative to substrate background via an unpaired, two-tailed t-test with N=3 mice (non-significant: p>0.05; ***: p<0.001). [Figure 72A]Figure 72A is a graphical representation of the transcript expression levels of 10 endogenous genes from HEK293FT cells transfected with targeted or non-targeted (NT) RADARS constructs, as quantified by qPCR, with supplementation of exogenous ADAR1p150. Data shown are normalized to targeted RADARS. Significance between targeted and non-targeted RADARS is determined via an unpaired t-test (non-significant: p>0.05) with a Welch correction assuming independent variances for each group. [Figure 72B] Figure 72B is a graphical representation of endogenous ACTB / PPIB protein expression in HEK293FT cells transfected with ACTB / PPIB-targeted or non-targeted RADARS (with supplementation of ADAR1 p150), as quantified by Western blot. Significance between targeted and non-targeted RADARS is determined by an unpaired t-test (non-significant: p>0.05) with Welch's correction assuming independent variances for each group. [Figure 72C] Figures 72C and 72D are visual depictions of the effect of sensor-target hybridization on protein production, analyzed by Western blot. Protein levels of ACTB (Figure 72C) and PPIB (Figure 72D) are shown in response to hybridization with RADARS, with GAPDH used as a normalization protein control. [Figure 72D] Figures 72C and 72D are visual depictions of the effect of sensor-target hybridization on protein production, analyzed by Western blot. Protein levels of ACTB (Figure 72C) and PPIB (Figure 72D) are shown in response to hybridization with RADARS, with GAPDH used as a normalization protein control. [Figure 73A]Figure 73A is a graphical depiction of the expression levels of interferon beta, OAS1, RIG-1, and MDA5 detected by qPCR upon transfection of RADARS constructs targeting IL6, ACTB, RPS5, or PPIB, non-targeting (NT) RADARS, and high molecular weight poly(I:C). Significance is determined by one-way ANOVA test between the untreated, RADARS, and poly(I:C) groups (non-significant: p>0.05; ****: p<0.0001). [Figure 73B] Figure 73B is a graphical depiction of the fold change in gene expression of four dsRNA-responsive genes (IFNb, OAS1, MDA5, and RIG-1) in response to RADARS in HEK293FT cells as detected by qPCR, using ACTB as a normalization gene. [Figure 73C] Figure 73C is a graphical depiction of the fold change in gene expression of four dsRNA-responsive genes (IFNb, OAS1, MDA5, and RIG-1) in response to RADARS or poly(I:C) in HepG2 cells as detected by qPCR, with GAPDH as the normalizing gene. [Figure 74A] Figure 74A is a visual representation of quantification of RNA editing within the 200 bp hybridization region of RADARS targeting transcripts for engineered guide RNAs against ACTB, PPIB, and non-targeting engineered guide RNAs. A to I(G) conversion is depicted in the heatmap (non-significant: p>0.05). [Figure 74B]Figure 74B shows a scatterplot analysis of transcriptome-wide off-target editing. The scatterplot shows the allele fraction (AF) of A-to-G mutations in (i) PPIB-sensor-transfected versus untransfected HEK293T cells with ADAR1 p150 overexpression (n = 3.17 × 10 6 sites); (ii) the same as (i) but with IL6 sensor (n = 3.07 × 10 6 sites); and (iii) the same as (i) but with non-targeting RADAR sensor without ADAR overexpression (n = 3.14 × 10 6 sites). Sites are colored by FDR-corrected p-value (color bar on the right). For i and iii, experiments were performed in triplicate. [Figure 74C] Figure 74C shows a scatterplot analysis of transcriptome-wide off-target editing by MCP-ADAR2(E488Q) using the dataset from NCBI Geo accession number GSE123905 (Katrekar et al., 2019). The scatterplot shows the allele fraction (AF) of A→G mutations in MCP-ADAR2(E488Q) overexpression (n=2.35×106 sites) compared to non-transfected HEK293T cells. Sites are colored by FDR-corrected p-value (color bar on the right). [Figure 75A] Figure 75A is a graphical representation of the analysis of sequence homology between transcriptome off-targets (n=23 sites) and targeted PPIB homology regions. Corresponding FDR-corrected p-values ​​(Monte Carlo permutation test) for the significance of local alignment between each off-target site and 200 bp surrounding the PPIB homology region targeted by the engineered guide RNA (red curve indicates p=0.05). [Figure 75B] Figure 75B is a visual representation of sequence logo analysis for off-target editing sites with overexpression of ADAR1p150 together with PPIB-targeted RADARS. DETAILED DESCRIPTION OF THE INVENTION

[0036] The present disclosure provides systems and sensors for detecting and quantifying RNA. The present disclosure also provides systems and methods for gene editing. Also disclosed is a system for in vivo imaging of RNA expression. The present disclosure provides a sensor system for converting adenine to inosine, which is recognized as guanosine (G) by the translation machinery. The conversion of adenine to inosine is the result of hydrolytic deamination of adenosine (Cox et al., Science, 358(6366):1019-1027(2017)). Therefore, the adenosine deaminase (ADAR) family of enzymes acting on RNA can convert codons within transcripts so that the translation product is functionally altered. The present disclosure provides, inter alia, editing of transcripts to remove functional alterations, such as stop codons, thereby enabling the expression of a payload.

[0037] definition Unless otherwise stated, the terms and techniques used within this application have the meanings commonly known to those skilled in the art.

[0038] As used herein, the term "about" is understood to modify the specified value. Unless expressly stated otherwise, the term "about" is understood to modify the specified value by ±10%. As used herein, when applied to a range, the term "about" modifies both endpoints of the range. For purposes of example, the range "about 5 to 10" is understood to mean "about 5 to about 10."

[0039] As used herein, the terms "sensor" and "sensor strand" are used interchangeably. As used herein, a "sensor" or "sensor strand" is understood to refer to a single-stranded RNA that includes at least a stop codon, where the RNA strand is capable of hybridizing or forming a duplex with another RNA strand. The terms "sensor" strand and "guide" strand may be used interchangeably throughout this disclosure.

[0040] As used herein, the terms "ADAR," "ADAR enzyme," and "deaminase" are used interchangeably unless expressly stated otherwise. Thus, within the present disclosure, "when the ADAR enzyme is a prokaryotic RNA-editing enzyme" is understood to also mean "when the deaminase is a prokaryotic RNA-editing enzyme."

[0041] As used herein, the term "RNA sensor system" or "sensor system" is understood to mean the minimum components required for (i) hybridization of a single-stranded RNA with a transcript of interest such that the resulting hybridized RNA contains at least one mismatch and at least one stop codon, (ii) recognition of the hybridized RNA as a substrate, and (iii) editing of the single-stranded RNA to remove the stop codon.

[0042] As used herein, the term "cellular logic system" or "logic system" refers to a system composed of multiple individual sensor systems. The cellular logic system or logic system of the present disclosure is a composite system that may be composed of one or more individual RNA sensor systems. When integrated into a larger cellular logic system, an individual RNA sensor system may depend on a separate RNA sensor system within the same cell for activation. Alternatively, multiple individual RNA sensor systems may be integrated into a cellular logic system such that an individual RNA sensor system is not required for another RNA sensor system.

[0043] Unless expressly stated otherwise, the term "payload," as used herein, is generally understood to mean a portion of single-stranded RNA that may hybridize with another single-stranded RNA, may be an invading strand to an already duplexed RNA molecule, or may be translated to express a protein. Thus, when an embodiment of the present disclosure, for purposes of example, states that "the payload comprises a therapeutic protein," it is generally understood that the payload is a fragment or portion of single-stranded RNA that can be translated to express a therapeutic protein.

[0044] As used herein, "cell-specific," "cell type-specific," and "activatable by a specific cell type" are understood by those skilled in the art to mean that activation of the RNA sensor requires the presence of factors that are present at substantially higher levels in the specific cell type than in other cell types. Those skilled in the art will recognize that expression of these factors may occur in other cell types, such that activation of the RNA sensor is possible, but is unlikely.

[0045] As used herein, unless otherwise stated, "avidity region" or "avidity binding region" can be used interchangeably to describe a region on the guide strand that has a degree of complementarity to target a transcript. Avidity refers to the design of multiple binding sites within the guide strand. These binding sites can be separated by linkers. The avidity region can optionally be separated from the main sensor region of the guide strand by one or more secondary structures, including hairpin structures. In some embodiments, the hairpin structure is an MS2 hairpin. The avidity region can optionally contain a stop codon that is not targeted for editing by ADARs. In some embodiments, the avidity region comprises a linker sequence. In some embodiments, the avidity region comprises one or more linker sequences.

[0046] Adenosine deaminase acting on RNA (ADAR) and other RNA-editing enzymes The ADAR enzyme is evolutionarily conserved among animals. Mammals have three known ADAR enzymes: ADAR1, ADAR2, and ADAR3. ADAR1 and ADAR2 are known to be catalytically active. In contrast, ADAR3, despite its substantial similarity to ADAR2, is generally considered catalytically inactive (Savva et al., Genome Biol., 13(12):252 (2012)). The ADAR enzyme contemplated in the present disclosure includes mammalian ADAR enzymes or modified enzymes derived from mammalian ADARs. In some embodiments, the RNA sensor enzyme ADAR is human ADAR1. In some embodiments, the RNA sensor enzyme ADAR is modified human ADAR1. In some embodiments, the RNA sensor enzyme ADAR is human ADAR2. In some embodiments, the RNA sensor enzyme ADAR is modified human ADAR2. In some embodiments, the RNA sensor enzyme ADAR is modified human ADAR3. In some embodiments, the RNA sensor enzyme ADAR is a synthetic enzyme. In some embodiments, the RNA sensor enzyme ADAR is an enzyme non-mammalian ADAR.

[0047] The ADAR enzyme of the present disclosure includes modified enzymes. The ADAR enzyme contemplated in the present disclosure includes enzymes modified to increase the affinity of the enzyme for the sensor strand. In some embodiments, the ADAR is modified to include an additional RNA-binding domain. In some embodiments, the ADAR is modified to eliminate one or more non-catalytic domains. In some embodiments, the ADAR enzyme of the sensor system includes an ADAR2 deaminase domain. In some embodiments, the ADAR consists of an ADAR2 deaminase domain. In some embodiments, the ADAR includes an ADAR2 deaminase domain fused to an MS2-binding protein. In some embodiments, the ADAR consists of an ADAR2 deaminase domain fused to an MS2-binding protein. In some embodiments, the ADAR is fused to a Cas (CRISPR-associated system) protein or a fragment or derivative thereof. In some embodiments, the ADAR is fused to a modified Cas protein. In some embodiments, the modified Cas protein is mutated to lack catalytic activity. In some embodiments, the ADAR is fused to a modified Cas13. In some embodiments, the ADAR is fused to a Cas13b containing a mutation at the amino acid corresponding to K370. In some embodiments, the ADAR is fused to a Cas13b containing the mutation K370A. In some embodiments, the ADAR is fused to a modified Cas13d. In some embodiments, the ADAR is fused to a modified Cas7-11.

[0048] The ADAR enzyme of the present disclosure can be endogenous to the cell that the sensor is delivered to.The ADAR enzyme envisioned in the present disclosure is exogenous, and can be delivered to cell simultaneously with the sensor, or can be delivered to cell separately from the sensor.In some embodiments, exogenous ADAR is delivered separately from the sensor.In some embodiments, exogenous ADAR is delivered simultaneously with the sensor.In some embodiments, exogenous ADAR can be used to supplement endogenous ADAR.In some embodiments, more than one exogenous ADAR is provided to cell.

[0049] Further RNA-editing enzymes In the methods of the present disclosure, additional deaminating enzymes may be used. In some embodiments, the deaminating enzyme may be a modified ADAR enzyme. The modified ADAR enzyme may include an ADAR enzyme modified to have increased cytidine deaminating activity, such as RESCUE (Abudayyeh et al., Science, 365, 382-386 (2019)).

[0050] In some embodiments, the deaminase is a modified enzyme that edits cytidine to uracil. In some embodiments, the deaminase can be a member of the APOBEC (apolipoprotein B mRNA-editing enzyme, catalytic polypeptide like) family of cytidine deaminases. In some embodiments, the deaminase is a modified APOBEC1. In some embodiments, the deaminase is a modified APOBEC2. In some embodiments, the deaminase is a modified APOBEC3. In some embodiments, the deaminase is a modified APOBEC3A. In some embodiments, the deaminase is a modified APOBEC3B. In some embodiments, the deaminase is a modified APOBEC3C. In some embodiments, the deaminase is a modified APOBEC3D. In some embodiments, the deaminase is a modified APOBEC3E. In some embodiments, the deaminase is a modified APOBEC3F. In some embodiments, the deaminase is a modified APOBEC3G. In some embodiments, the deaminase is a modified APOBEC3H.

[0051] In some embodiments, the deaminase may be a prokaryotic RNA-editing enzyme, hi some embodiments, the deaminase is derived from Escherichia coli (E. coli).

[0052] Sensor / Sensor Chain The sensors of the present disclosure contain at least one stop codon. The sensor can be located on the same RNA strand as the payload, the normalization gene, or both the payload and the normalization gene. The sensors of the present disclosure can be designed so that when a single-stranded RNA (ssRNA) sensor strand binds to a target ssRNA strand to create a double-stranded RNA (dsRNA), the duplex contains a mismatch in the region corresponding to the stop codon in the sensor strand. The disclosed sensors can be modified in a number of ways. In some embodiments, the sensor is administered to a cell as a DNA template that can then be transcribed into a single-stranded RNA sensor molecule.

[0053] The present disclosure also provides sensor strands that include more than one stop codon. In some embodiments, the sensor strand includes 1, 2, 3, 4, 5, 6, 7, 8, 9, or 10 stop codons. In some embodiments, the sensor strand includes more than 10 stop codons. In some embodiments, the sensor strand includes 2 stop codons. In some embodiments, the sensor strand includes 3 stop codons. In some embodiments, the sensor strand includes 4 stop codons. In some embodiments, the sensor strand includes 5 stop codons. In some embodiments, the sensor strand includes 6 stop codons. In some embodiments, the sensor strand includes 7 stop codons. In some embodiments, the sensor strand includes 8 stop codons. In some embodiments, the sensor strand includes 9 stop codons. In some embodiments, the sensor strand includes 10 stop codons.

[0054] The present disclosure also provides a sensor / guide strand that contains one or more avidity binding regions.In some embodiments, the sensor strand comprises three avidity regions.In some embodiments, the sensor strand comprises five avidity regions.In some embodiments, the sensor strand comprises seven avidity regions.The avidity region can incorporate a stop codon that is not a target for ADAR editing.The avidity binding region can also contain a stop codon that is a target for ADAR editing.In some embodiments, the avidity binding region comprises a stop codon for the intended ADAR editing.

[0055] Also provided within the present disclosure are sensor / guide strands incorporating one or more MS2 hairpins. In some embodiments, the sensor strand comprises two MS2 hairpins. In some embodiments, the sensor strand comprises three MS2 hairpins.

[0056] In some embodiments, the sensor / guide strand comprises both an avidity region and an MS2 hairpin region.

[0057] In some embodiments, the sensor / guide strand comprises a modified RNA. In some embodiments, the modified RNA comprises 5-methylcytosine. In some embodiments, the modified RNA comprises pseudouridine.

[0058] payload In some embodiments, the payload comprises a reporter transcript. In some embodiments, the payload consists of a reporter transcript. In some embodiments, the reporter transcript is a fluorescent reporter. In some embodiments, the reporter transcript comprises a luciferase transcript. In some embodiments, the reporter transcript comprises a GFP transcript.

[0059] The sensor systems of the present disclosure can be designed to deliver a payload encoding a therapeutic protein, which in some embodiments can be used in conjunction with another therapeutic agent.

[0060] In some embodiments, the payload comprises a transcription factor, in some embodiments, the payload comprises an enzyme, in some embodiments, the payload comprises a transgene protein.

[0061] In some embodiments, the payload comprises a protein for use in editing the genome of a cell. In some embodiments, the payload comprises a Cas protein. In some embodiments, the payload comprises a Cas9 protein.

[0062] In some embodiments, the payload comprises a protein capable of converting one cell type into another cell type.

[0063] In some embodiments, the payload comprises an ADAR. In some embodiments, the payload comprises an ADAR that is capable of inducing a positive feedback loop.

[0064] In some embodiments, the payload comprises a protein capable of killing a specific cell type, hi some embodiments, the payload comprises a protein capable of killing tumor cells, hi some embodiments, the payload comprises an immunomodulating protein.

[0065] logic gates The present disclosure also relates to multiplexed multi-sensor reporter systems. In some embodiments, these multi-sensor reporter systems utilize logic gates. These logic gates may consist of AND gates, OR gates, or AND / OR gates as individual decision points within the same reporter system.

[0066] An AND gate for use in the reporter system of the present disclosure can be implemented by having multiple guide strand-binding compartments on the same ssRNA sensor. In this type of AND gate, each individual guide compartment senses a distinct endogenous transcript within the cell. Activation of the AND gate in this type of gate is equivalent to activation of the entire ssRNA sensor strand; that is, removal of the stop codon and expression of the terminal payload. This type of logic gate requires that each guide compartment interacts with the target sequence and an ADAR or other deaminating molecule. Deamination of the stop codon in each guide compartment allows full expression of the payload. In some embodiments, each guide compartment is further separated by a distinct reporter. In some embodiments, each guide compartment is further separated by a distinct, distinct reporter. In some embodiments, each reporter on the ssRNA sensor for use in an AND gate is a distinct fluorescent reporter.

[0067] In some embodiments, the AND gate can operate sequentially. In this type of AND gate, each of the multiple guide strand binding compartments is placed on a separate ssRNA sensor. Activation of the gate in this type of AND gate involves activation of multiple sensors in a defined sequence. In this type of AND gate, activation of the first sensor results in the expression of an intermediate payload. This intermediate payload allows the expression of a second RNA sensor in a specific cellular environment. In such a system, the cascade of RNA sensor activation can only occur in the context of a specifically determined cellular stimulus upon activation.

[0068] OR gates for use in the reporter systems of the present disclosure can be implemented through the use of multiple independent ssRNA sensors within the same cell, each of which can deliver a payload without activating another sensor.

[0069] Delivery System The present disclosure also provides a system for delivering an ADAR sensor.

[0070] In some embodiments, the ADAR sensor is delivered directly to a cell. In some embodiments, the ADAR sensor is encapsulated in a lipid nanoparticle. In some embodiments, the ADAR sensor is delivered via a viral vector.

[0071] In some embodiments, the ADAR sensor is a circular RNA.

[0072] Methods of the present disclosure The present disclosure also provides methods for using the sensor systems described herein.

[0073] In some embodiments, the RNA sensor delivers a payload that can be optically observed. In some embodiments, the RNA sensor is tracked via an imaging system. The imaging system can be any suitable imaging system that is compatible with the system. In some embodiments, the imaging system uses fluorescent molecules. In some embodiments, the RNA sensor system includes multiple fluorescent molecules. In other embodiments, the imaging system is a non-invasive imaging system. In some embodiments, the RNA sensor system is compatible with a fluorescence-activated cell sorting (FACS) system.

[0074] In some embodiments, the RNA sensor system is tracked via a non-invasive imaging system. In some embodiments, the imaging system tracks Deep Red luciferase.

[0075] The present disclosure also provides a method for quantifying RNA in vivo. The quantification method of the present disclosure can rely on the incorporation of a normalization gene on the sensor strand. The translation of this normalization gene occurs independently of RNA editing. Quantification of the normalization gene provides a reference standard for the total sensor strand delivered to each individual cell. Reference to this normalization gene allows the determination of activated RNA sensor as a percentage of the total sensor delivered.

[0076] The present disclosure also contemplates live cell imaging. In some embodiments, fluorescent reporters are visualized. In some embodiments, multiple fluorescent reporters are tracked. The present disclosure also provides long-term cell lineage tracking. In some embodiments, activation of the RNA sensor system results in a permanent change in the expression of the reporter molecule, so that cells in which the sensor system has already been activated can be identified at a later time point.

[0077] In some embodiments, the RNA sensor system can be used to target specific cell types. In some embodiments, the RNA sensor system is engineered to be activated in specific cell types. In some embodiments, the RNA sensor system targets specific cell types. In some embodiments, the RNA sensor system targets tumor cells. In some embodiments, the RNA sensor system delivers a payload that kills specific cell types. In some embodiments, the RNA sensor system delivers a payload that converts one cell type into another cell type. In some embodiments, the RNA sensor system delivers a payload that edits the genome of a cell.

[0078] In some embodiments, the RNA sensor system is druggable. In some embodiments, the RNA sensor system is drug-sensitive. In some embodiments, the RNA sensor system is activated only in the presence of a single drug or compound. In some embodiments, the RNA sensor system is activated only in the presence of multiple drugs or compounds.

[0079] The present disclosure also contemplates the use of the RNA sensor system described herein in an in vitro diagnostic assay. For example, the RNA sensor system may be an RNA sensor system in a diagnostic assay in which the payload comprises a fluorescent protein, a luciferase protein, an antigen, or an epitope. In some embodiments, the diagnostic assay is a lateral flow strip.

[0080] Now that the present technology has been described in detail, the present technology will be more clearly understood by referring to the following examples.The following examples are incorporated for illustrative purposes only and are not considered to limit the embodiments of the present technology.All patents and publications mentioned in this specification are expressly incorporated by reference. [Example]

[0081] [Example 1] Example method Unless otherwise indicated, the following experimental methods were utilized in the examples presented herein.

[0082] Measurement of luciferase activity Culture media containing secreted luciferase was harvested 48 hours after transfection unless otherwise noted. 20 μL of culture media was used to measure luciferase activity using the Targeting Systems Cypridinia luciferase assay kit and the Targeting Systems Gaussia luciferase assay kit (Targeting Systems) on a Biotek Synergy 4 plate reader using the injection protocol. All replicates were performed as biological replicates.

[0083] Transfection for fluorescent sensors The day before transfection, cells were seeded at 10,000 cells into Corning 96-well tissue culture-treated black plates, resulting in approximately 40-50% confluency on the day of transfection. For all fluorescent sensors, HEK293FT cells were transfected with 100 ng of total plasmid DNA using TransIT-LT1 according to the manufacturer's specifications (1 μg DNA:3 μL Trans Reagent). Unless otherwise specified, ADAR sensor, ADAR, and target plasmids were mixed at equal concentrations (33.3 ng per condition); experiments without one or more of the foregoing were substituted with pUC19, thereby maintaining the total DNA concentration at 100 ng.

[0084] Confocal microscopy of fluorescent ADAR sensors 48 hours after transfection, all wells were measured via confocal microscopy under the following settings: For each well, 2 x 2 images at 10x magnification were collected around the center point and stitched together: Images were collected in the 488 nm channel (32.8% magnification, 100 ms exposure), 561 nm channel (35.2% magnification, 100 ms exposure), 640 nm channel (80% magnification, 100 ms exposure), and bright field channel (25 ms exposure).

[0085] Quantification of fluorescent signals from images Images were opened in Matlab and segmented within the mCherry channel via Watershed. The total pixel area and average pixel intensity were calculated for the mNeon (488 nm), mCherry (561 nm), and iRFP (640 nm) channels for each segmented cell and exported to a summary csv file. The csv files were batch-processed in R by the following steps: merging all csv files, merging conditions with small summary areas (conditions with few cells or conditions not transfected with the sensor), subtracting the background fluorescence for each channel from all conditions within that channel, and dividing the summary value for each condition by the area to obtain the average fluorescence intensity. Standard deviations were calculated by comparing the average values ​​for triplicate technical replicates for transfection. For mNeon / mCherry ratio values, the average mNeon fluorescence intensity for a condition was divided by the average mCherry value for that same condition. For fluorescence ratio and fold-change ratio values, error propagation was performed using the formula:

[0086]

number

[0087] Quantification of the percentage of mNeon-positive cells in confocal images We observed some consistent leakage of mNeon within the fluorescent sensor due to very low levels of plasmid contamination or ribosome slippage. Therefore, we gated on detection of mNeon-positive cells 30 arbitrary units above background and determined the percentage of mCherry-positive cells under conditions where mNeon was expressed above this threshold. The mNeon values ​​were plotted as a histogram with kernel density estimation by base 10 logarithm to generate the plot in Figure 2E.

[0088] RNA extraction and next-generation sequencing for ADAR sensors To calculate the editing rate of the ADAR sensor, cells were harvested 48 hours after transfection after imaging. Total RNA was extracted using the RNeasy 96 Kit (Qiagen) with DNase treatment. cDNA was prepared using SuperScript IV reverse transcriptase (Invitrogen) and sensor-specific primers. The sensor guide regions were amplified, indexed, and sequenced on an Illumina MiSeq platform. Reads were demultiplexed and aligned to each sensor, and the A-to-I editing rate was calculated using a proprietary MATLAB pipeline.

[0089] Quantification of protein expression Two days after transfection into HEK293FT cells, the HiBiT tag was quantified in cell lysates using the Nano-Glo HiBiT Lytic Detection System (Promega). To prepare the Nano-Glo HiBit Lytic Reagent, Nano-Glo HiBit Lytic Buffer (Promega) was mixed with Nano-Glo HiBit Lytic Substrate (Promega) and LgBiT Protein (Promega) according to the manufacturer's protocol. The volume of Nano-Glo HiBit Lytic Reagent added was equal to the culture medium present in each well, and the samples were shaken at 600 rpm on an orbital shaker for 3 minutes. After a 10-minute incubation at room temperature, readout was performed using a plate reader (Biotek Synergy Neo 2) with a gain of 125 and an integration time of 2 seconds. The background control was subtracted from the final measurement.

[0090] mRNA synthesis Prior to in vitro transcription, DNA templates were obtained by PCR with a targeting forward primer containing a T7 promoter. Sensor mRNA and MCP-ADAR2dd mRNA were transcribed and poly(A)-tailed using the HiScribe™ T7 ARCA mRNA kit (NEB, E2065S) according to the manufacturer's protocol, supplemented with 50% 5-methyl CTP and pseudo-UTP (Jena Biosciences). The mRNA was then cleaned up using the MEGAclear™ Transcription Clean-Up Kit (Thermo Fisher, AM1908).

[0091] Harvesting of total RNA and quantitative PCR For gene expression experiments in mammalian cells, 48 ​​hours after transfection, cells were harvested and reverse-transcribed to generate cDNA using a previously described modification of the commercially available Cells-to-Ct kit (Thermo Fisher Scientific) (Joung et al., 2017). Transcript expression was then quantified by qPCR using Fast Advanced Master Mix (Thermo Fisher Scientific) and TaqMan qPCR probes (Thermo Fisher Scientific) along with GAPDH control probes (Thermo Fisher Scientific). All qPCR reactions were performed in duplicate technical replicates in 10 μl reactions in a 384-well format and read using a LightCycler 480 Instrument II (Roche). For multiplexed targeting reactions, reads of different targets were performed in separate wells. Expression levels were calculated by subtracting the cycle threshold (Ct) of the housekeeping control (GAPDH) from the target Ct value and normalizing to the total input, resulting in a ΔCt level. Relative transcript abundance was calculated as 2 -ΔCt All replicates were performed as biological replicates.

[0092] Automated production of avidity sensors Avidity sensors were created using the python scripts in the following repository: https: / / github.com / abugoot-lab / ADAR SENSOR. A schematic for the creation of a typical three-avidity guided ADAR sensor with 5 nucleotide spacing between the two MS2 hairpin loops and the guide region (on target) is shown in Figure 24.

[0093] Animal Care and Protocols All experiments were performed on female B6(Cg)-Tyrc-2J / J (Albino B6) mice and female NOD.Cg-Prkdcscid Il2rgtm1Wjl Tg(SERPINA1*E342K)#Slcw / SzJ (NSG-PiZ) mice (Jackson Laboratory) with ad libitum access to food and water. NSG-PiZ mice express mutant human SERPINA1 on an immunodeficient NOD scid gamma background. All mice were housed in individually ventilated cages (IVCs) in a temperature-controlled animal facility (usually operating on a 12:12-h light / dark cycle) and used in accordance with procedures approved by the MIT Committee on Animal Care.

[0094] RADARSv2 Design Stop codons were engineered in the +1 and +2 frames following the engineered guide RNA region to capture translating ribosomes across all frames. These out-of-frame stop codon designs synergized with the extended 5' peptide to significantly reduce background and result in approximately 200-fold activation. We selected this sensor design, termed RADARSv2, incorporating a structured guide, upstream peptide, and out-of-frame stop codon as a unifying architecture for future sensors (Figure 54A).

[0095] [Example 2] Development of luciferase / fluorescent sensor and detection of EGFP transcripts

[0096] Cloning of luciferase sensors Luciferase sensors were cloned by Gibson assembly of PCR products. The sensor backbone was created by cloning Cypridinia luciferase (Cluc) under the CMV promoter and Gaussia luciferase (Gluc) under the EF1-α promoter on a single vector. Expression of both luciferases on a single vector allowed one luciferase to be used as a dosing control for normalization of the knockdown of the other luciferase, controlling for variations due to transfection conditions. The short sensor primers were ordered and then phosphorylated and annealed using T4 polynucleotide kinase. The annealed oligos were ligated into the backbone using T4 DNA ligase (NEB) for 20 minutes at room temperature in a typical 10-μL ligation reaction with 1 μL of T4 DNA ligase, 30 ng of insert, 50 ng of backbone, and 1 μL of 10x ligation buffer. The long avidity sensor region was ordered directly from Integrated DNA Technologies (IDT) as Eblocks. The PCR product was purified by gel extraction (Monarch gel extraction kit, NEB) and assembled into the backbone using the NEB HiFi DNA Assembly master mix kit with 2.5 μL of mastermix, 30 ng of backbone, and 5 ng of insert in a 5-μL reaction. The reaction was incubated at 50°C for 30 minutes in a thermocycler, and 20 μL of competent Stbl3 cells produced with the Mix and Go! competency kit (Zymo) were transformed with 2 μL of the assembled reaction and plated onto agar plates supplemented with the appropriate antibiotic. After overnight growth at 37°C, colonies were picked into Terrific Broth (Thermo Fisher Scientific) and incubated at 37°C with shaking for 24 hours.Cultures were harvested using a QIAprep Spin Miniprep kit (Qiagen) according to the manufacturer's instructions.

[0097] This ADAR sensor for luciferase contains a 51-nucleotide EGFP transcript sensing guide and a Gaussia luciferase (Gluc) payload (Figure 2A). Constitutive Cypridiana luciferase (Cluc) was incorporated on a separate transcript to allow ratiometric control of transfection variability. This dual-reporter, dual-transcript luciferase reporter system targets functional eGFP under the control of a doxycycline-inducible promoter. We tested this ADAR sensor design in the presence or absence of an exogenous ADAR2 deaminase domain with the hyperactive mutation E488Q and the specific mutation T490A fused to the MS2 coat protein (MCP-ADAR2dd(E488Q, T490A)) (Kuttan and Bass, 2012; Cox et al., 2017), along with a scrambled guide control. HEK293FT cells were co-transfected with a plasmid expressing an ADAR sensor and a plasmid expressing EGFP or a control plasmid. We observed that when relying solely on endogenous ADAR, the ADAR sensor resulted in a 5-fold increase in normalized luciferase levels, and when supplemented with exogenous MCP-ADAR2dd (E488Q, T490A), it resulted in a 51-fold activation of the signal (the fold change in luciferase expression in the presence of target / in the absence of target) (Fig. 2B). In addition, we observed that the luciferase signal upon induction of the ADAR sensor supplemented with exogenous ADAR was equivalent to that of a constitutively expressed transcript without an upstream stop codon (approximately 78%, Fig. 2C). Therefore, this high level of protein production upon activation of the ADAR sensor confirms the suitability of the ADAR sensor for applications requiring high absolute payload expression.To confirm that payload expression is dependent on RNA editing, we harvested RNA from cells and quantified editing by next-generation sequencing. We observed that editing of the stop codon UAG was increased approximately 24-fold with the EGFP-targeting sensor, whereas the increase in editing with the non-targeting sensor was negligible (Figure 2D).

[0098] Cloning of fluorescent sensors The parent fluorescent ADAR sensor was cloned in three pieces via Gibson assembly using HindIII- and NotI-cut pcDNA3.1(+) as the backbone. mCherry was amplified from Addgene vector 109427, and T2A mNeon was ordered as gBlock from IDT. All fluorescent ADAR sensors were subcloned into the parent fluorescent plasmid via Golden Gate cloning using the enzyme Esp3I (an isoschizomer of BsmBI). Inserts were ordered as complementary strands with overhangs and annealed in phosphorylated form, or generated via PCR. Golden Gate reactions were assembled from components using the BsmBIv2 golden gate assembly kit from NEB or in a 20 μL reaction containing 25 ng of vector and 2 μL of a 1:200 diluted insert (approximately 5–10 ng). The reaction was thermocycled for 1 hour, alternating between 25°C and 37°C for 5 minutes each, and then 12.5 μL of Zymo Mix and Go Competent Cells was transformed with 0.75 μL of the reaction mix. The transformed cells were diluted 1:1 with SOC medium, and 10 μL was streaked onto a carbenicillin agar plate at 50 μg / mL. After overnight incubation at 37°C, single colonies were picked into 4 mL of LB (Luria broth) supplemented with 50 μg / mL carbenicillin. Plasmids were prepared from the cultures as described above for the luciferase sensor.

[0099] This dual-reporter single-transcript fluorescent sensor contains a single-transcript fluorescent reporter, with mCherry constitutively expressed upstream of the 51-bp eGFP sensor and an mNeon reporter downstream that is activated only upon interaction with the target (Figure 3A). HEK293 cells were transfected with non-functional eGFP under the control of a doxycycline-inducible promoter. In the presence of 1 μg / mL doxycycline, cells were also transfected with the dual-reporter single-transcript targeted or non-targeted sensor. Figure 3B shows representative images from experiments with and without targets, and experiments with and without exogenous ADARs. Figure 3C shows the quantification of the fold change in EGFP fluorescence upon target induction. The fold change ratio indicates the mNeon / mCherry fluorescence value (fluorescence ratio value) in the presence of the target divided by the fluorescence ratio value in the absence of the target for the ADAR mutant. In the presence of exogenous MCP-ADAR2dd(E488Q, T490A), the targeted ADAR sensor exhibited a >21-fold increase in activation. In addition, low background activation was observed in the absence of the targeted ADAR sensor. The TAG→TIG editing rate of the sensor codon TAG was also measured in the presence or absence of the target. In the presence of exogenous ADAR and the targeted ADAR sensor, the stop codon UAG was edited at an editing rate of 9.4%, whereas in the absence of the targeted ADAR sensor, the stop codon UAG was edited at an editing rate of 0.2%, suggesting that target-driven editing drives the expression of the fluorescent payload.

[0100] To further establish proof-of-principle for the biological luciferase sensor, we co-transfected three guide strands into HEK293FT cells with an exogenous EGFP reporter transcript. The ADAR sensor was a dual-transcriptomic system using Gaussia luciferase (Gluc) / Cypridiana luciferase (Cluc) transcripts, allowing ratiometric control for transfection variability. The dual-reporter dual-transcriptomic luciferase reporter system targeted an exogenous eGFP reporter transcript. Designs 2 and 4 were different guides targeting the EGFP transcript. No exogenous ADAR was introduced into the cells. The negative control was a guide strand that did not recognize EGFP. Analysis of EGFP expression after transfection of the three guide strands showed a significant increase in luciferase expression levels in both Designs 2 and 4 (Figure 4A). Both Design 2 and Design 4 guide strands exhibited a significant increase in luciferase signal compared to the negative control scrambled guide (Figure 4B).

[0101] Example 3: Increased transcript expression following administration of exogenous ADAR2 To determine whether the introduction of additional ADAR molecules increases luciferase expression, five guide strands were co-transfected into HEK293FT cells with exogenous EGFP transcripts as a reporter. Guides 1–4 are different guides targeting the EGFP transcript. The negative control was a scrambled control designed not to recognize EGFP. Each guide strand was tested under three experimental conditions. First, each guide was transfected into HEK293FT cells without exogenous ADAR (Figure 5, blue bars). Next, each guide was co-transfected with the deaminase domain of ADAR2 (ADARdd) into HEK293FT cells (Figure 5, white bars). Finally, each guide was co-transfected with dPspCas13b-ADAR2dd, a transcript that overexpresses ADAR2dd (Figure 5, red bars).

[0102] Guide strands 1-4 exhibited between a 1.5-fold and 2-fold increase in luciferase expression when normalized to a negative control without the addition of exogenous ADAR molecules (Figure 5, blue bars). When the deaminase domain of ADAR2 (ADAR2dd) was co-transfected into cells, guides 1, 2, and 4 exhibited similar increases in luciferase expression as the same guides in the presence of endogenous ADAR (Figure 5, white bars). Guide strand 3 exhibited a 3-fold increase in luciferase expression in the presence of additional ADAR2dd molecules (Figure 5, white bars).

[0103] When a vector overexpressing ADAR2dd was introduced simultaneously with guide strands 1-4, luciferase expression increased at least twofold (Figure 5, red bar). Guide strand 3, in particular, exhibited a four-fold increase in luciferase expression when compared to the negative control. This data highlights the possibility of utilizing exogenous / modified ADAR molecules to enhance sensor performance.

[0104] Example 4: ADAR optimization and length screening reduces background and increases ADAR sensor activation in the presence of target During validation of these ADAR sensors, we observed that activation could occur for some guides in the presence of exogenous ADARs, despite the absence of target RNA (Figures 2C and 3B). We attempted to determine whether ADAR activity could be optimized to increase activation and reduce background. To optimize the ADAR sensors and minimize this background, we selected and examined a panel of different ADAR1 / ADAR2 mutants combined with 69-nucleotide guides targeting frameshifted EGFP or iRFP transcripts (Figures 6 and 7). Figure 6A shows, from left to right, a schematic diagram of the different ADARs tested, including the p150 isoform of ADAR1, the p110 isoform of ADAR1, ADAR2, and MS2 coat protein (MCP)-ADAR fusion protein (MCP-ADAR). fl = full length. DD = deaminase domain. The catalytic domain mutations are not shown in the schematic, but all mutations are within the deaminase domain.

[0105] We screened full-length human ADAR isoforms (ADAR1 p110, ADAR1 p150, and ADAR2) (Galipon et al., 2017; Merkle et al., 2019) and their catalytic deaminase domains alongside specific mutants designed to destabilize ADAR-dsRNA interactions and reduce nonspecific editing (Cox et al., 2017; Matthews et al., 2016). Our initial exogenous ADAR selection, MCP-ADAR2dd(E488Q, T490A), performed best against a frameshifted EGFP transcript, but several candidates in our screen also conferred comparable activation upon cotransfection with the target (Figure 6B) and reduced background against the two-target set (Figure 7). We also examined the editing rates of stop codons by various sensors in the following conditions: (Fig. 8A) exogenous recruitment with MCP-ADAR2dd, (Fig. 8B) exogenous recruitment with ADAR1 p150 isoform, (Fig. 8C) exogenous recruitment with ADAR2, and (Fig. 8D) no exogenous ADAR recruitment. The editing rates of these candidate substances were also examined. The editing rates were calculated via RNA sequencing data showing the conversion of the stop codon UAG to UIG in the presence and absence of the ADAR mutant targets selected for further screening (Fig. 8).

[0106] Because guide selection can affect the overall sensitivity of the sensor, we screened the top ADAR candidates against multiple guide sequences and targets in an orthogonal panel (Figure 10). First, we observed activation above background only for properly matched ADAR sensor and target transcripts (Figure 9). ADAR1 p150 generally provided the greatest fold activation for three of the four targets, driven by a low overall background signal in the absence of target, while MCP-ADAR2dd(E488Q, T490A) performed best against the target EGFP due to its generally high absolute signal, but suffered from high background against other targets, which reduced its overall activation (Figures 6, 10, 11, and 12).

[0107] Analysis of the activation of the sensor panel by various sensors relative to background was also performed with (Figure 13A) exogenous recruitment with MCP-ADAR2dd, (Figure 13B) exogenous recruitment with ADAR1 p150 isoform, (Figure 13C) exogenous recruitment with ADAR2, and (Figure 13D) no exogenous ADAR recruitment.

[0108] The optimal exogenous ADAR amount was also investigated (Figure 14). In a titration experiment for tetracycline-inducible IL6, with the target (IL6) amount fixed at 20 ng, HEK293 cells were transfected with varying amounts of MCP-ADAR2dd ranging from 10 ng to 100 ng, as well as varying amounts of the three-site avidity-linked IL6 sensor chain ranging from 10 ng to 100 ng. The fold change represents the normalized luciferase ratio between the target and no-target groups.

[0109] Example 5: Use of the programmable A to I(G) replacement RNA editing (REPAIR) system for biological sensors Next, we assessed the feasibility of the programmable A to I(G) replacement RNA editing (REPAIR) system as a mechanism for triggering these genetic sensors. The REPAIR system consists of a fusion of a catalytically active enzyme, Cas13b, fused to the deaminase domain of ADAR2. A catalytically inactive enzyme, Cas13b, incorporating the K370A mutation was also fused to the deaminase domain of the ADAR2 molecule, forming a fusion protein without Cas13b activity (REPAIR K370A).

[0110] Two guide strands designed to target EGFP, as well as a negative control designed not to target EGFP, were transfected into HEK293FT cells without the addition of exogenous ADAR molecules (Figure 15, blue bars). The guide strands were also co-transfected with the REPAIR molecule (Figure 15, red bars) or the catalytically inactive REPAIR K370A molecule (Figure 15, white bars). Guide 1 and guide 2, designed to be targeted by Cas13b, did not show increased luciferase expression in cells compared to cells relying solely on endogenous ADAR expression. However, when Cas13 activity was turned off (REPAIR K370A), guide 1 showed a four-fold increase in luciferase expression compared to cells relying solely on endogenous ADAR expression. This increase in expression supports the potential of the REPAIR system for use with the genetic sensors disclosed herein.

[0111] [Example 6] Examination of variation in guide strand characteristics to increase luciferase expression Next, the characteristics of guide strand design were varied to determine which variables could be adjusted to increase the efficiency and expression of gene sensor. Figure 16 shows a heat map showing luciferase expression fold change against the variation of guide / ADAR combination. Exogenously introduced full-length ADAR (column 2) consistently shows the greatest increase in luciferase expression fold. However, this is not the case for the guide strand designed as MS2 agonist, which shows almost consistent expression regardless of introduced ADAR.

[0112] By selecting the low background level from Example 4, we began further optimization of the sensor using the ADAR1 p150 construct. To improve both binding stability and target search time, increasing the guide length (Qu et al., 2019) centered around the premature stop codon was examined against the constitutive target, iRFP. As the guide length increased from 51 nucleotides to 600 nucleotides, sensor activation improved from 2.2-fold to 18.22-fold (Figure 17B). Additionally, in the presence of the target, a significant shift in the distribution of mNeon expression levels per cell was observed for all guide lengths (Figures 17C, 18A), with the substantial mNeon(+) population increasing with increasing guide length. Meanwhile, in the absence of the target, the percentage of mNeon(+) cells remained consistently <5% for all guide lengths. With a guide length of 600 nucleotides, 66.7% mNeon(+) cells were observed in the presence of the target and 1.6% mNeon(+) cells in the absence of the target, suggesting a robust ability to separate cell populations based on target mRNA expression.

[0113] The two best-performing ADARS, ADAR1 p150 and MCP-ADAR2dd(E488Q, T490A), from Example 4, achieved optimal signal through reduced background or increased activation. Because engineering the ADAR enzyme revealed that MCP-ADAR2dd(E488Q, T490A) resulted in the highest activation by the luciferase sensor ( FIG. 20 ), we hypothesized that a guide engineering strategy that reduces background, when coupled to MCP-ADAR2dd(E488Q, T490A), would result in maximal activation of the optimal sensor.

[0114] We designed a novel sensor targeting IL6 mRNA, a virtually unpredicted transcript in HEK293FT cells (Uhlen et al., 2015), allowing us to both supplement IL6 mRNA via exogenous transfection and generate an integrated diagram of IL6 mRNA under the control of a doxycycline-inducible promoter that modulates low-level IL6 expression for sensitivity testing (see Figure 20).

[0115] Due to increased background signal in the absence of target, potentially due to readthrough of a stop codon within the long guide region (Figure 21 (A) MCP-ADAR2dd(E488Q, T490A) exogenous recruitment, (B) exogenous recruitment with ADAR1 p150 isoform, (C) exogenous recruitment with ADAR2, and (D) no exogenous ADAR recruitment), we engineered the guide region to block aberrant translation by introducing an MS2 hairpin loop (Chao et al., 2008), which provides the additional benefit of recruiting the MCP-ADAR2dd(E488Q, T490A) protein to the guide:target duplex (Figure 23). We repeated our search for increasing the guide region with a luciferase sensor targeting IL6 transcripts, and found a significant reduction in the fold change in ADAR sensor activation beyond the 81-nucleotide guide using the MCP-ADAR2dd(E488Q, T490A) construct (Figure 22) when only one binding site was present on the guide strand. However, structural additions and modifications to the guide strand design, such as the addition of an MS2 hairpin loop and further manipulation of the guide binding region (referred to as the "avidity binding region") on the sensor / guide strand, were performed to determine whether they enhanced sensitivity, resulting in a significant increase in ADAR sensor activation (Figure 22). Figure 22 presents the results of an experiment to determine the fold change for guide / sensor strands with different formats. HEK293 cells were transfected (Lipofectamine 3000, Thermo Fisher Scientific) with naive reverse complements (reverse complements of the target IL6) and sensors containing MS2 hairpins with additional avidity regions (x-axis: 51-bp sensor, sensor with a 171-bp continuous binding region, sensor with a 171-bp avidity region separated by two MS2 hairpins, sensor with a 225-bp single continuous binding region, sensor with a 225-bp binding region separated by four MS2 hairpins, sensor with a 279-bp continuous binding region, sensor with a 279-bp binding region separated by six MS2 hairpins, and non-targeted sensors).Separation of the binding domains by an MS2 hairpin significantly increased target expression in all sensors with varying avidity domain lengths. Several types of ADAR proteins were examined, including full-length ADAR2 (ADAR2FL), the p150 isoform of ADAR1, and fusion proteins of the MS2 coat binding protein fused to the deaminase domains of endogenous ADAR1 and human ADAR2. Fold-change expression (y-axis) is calculated by comparing the raw luciferase value under target conditions with that under target-free conditions.

[0116] ADAR sensor activation was highest with a five-site avidity-linked guide, achieving approximately 70-fold higher activation than uninterrupted guide designs and substantially lower background (Figure 27). Avidity-linked guides improved performance for all exogenous ADAR constructs, but supplementation with MCP-ADAR2dd demonstrated the greatest increase in performance. Avidity-linked guides with five or seven binding sites can result in detectable activation relying solely on endogenous ADARs. Figure 23 presents several possible MS2 hairpin / avidity modification formats, and Figure 24 provides design guidelines for avidity sensors and an easy-to-use software program (github.com / abugoot-lab / ADARSENSOR) that automatically generates avidity sensors for input target sequences.

[0117] We investigated whether the best-performing IL6-targeting engineered guide RNA could utilize endogenous ADARs to sense IL6, a synthetic target transfected into cells. We observed that supplementation with exogenous ADAR1p150 improved the performance of RADARSv2, but that endogenous ADARs activated the payload by over 50-fold (Figure 52).

[0118] To further explore the concept of avidity-linked guides, we varied the spacing between binding sites (5, 30, and 50 nucleotides). The length between binding sites represents the number of nucleotides on the guide strand, starting just before the MS2 hairpin and ending at another complementary region. Closely spaced binding sites on the target transcript resulted in the highest degree of activation (Figure 25).

[0119] We also explored whether avidity-linked guide improvements could be combined with orthogonal methods to block translational readthrough, such as additional stop codons. We compared a single stop codon-based sensor with a double stop codon-based sensor with an additional stop codon in the most posterior avidity region (Figure 26). Figure 26B shows the fold change in luciferase payload between the single stop codon-based sensor and the double stop codon-based sensor. The double stop codon-based sensor exhibited a significantly increased fold change compared to the ADAR-dependent single stop codon-based sensor. We found that the additional stop codon increased the activation fold for the seven-site avidity-linked guide beyond the performance of the five-site avidity-linked guide (Figure 28A). This improvement was driven by both a decrease in background activation rates and an increase in stop codon editing rates in the presence of target (Figure 28B, Figure 8).

[0120] Despite the abundance of CCA codons on potential target transcripts, we explored whether incremental manipulation of avidity guide design could improve mismatch tolerance, increasing targeting flexibility. We also examined target mismatch tolerance (Figure 29A) for 16 possible mismatches (derived from the canonical CCA) between the 51-bp naive sensor, the 3-avidity sensor, and the 5-avidity sensor. We designed 16 targets (nCn) covering all nucleotide changes to 5' cytosine or 3' adenosine. By examining guides containing UAG beyond these codon variations, we found that mismatches with guanine or cytosine were generally better tolerated than mismatches with adenosine or uridine (Figure 29B). Furthermore, with the exception of targets ACA and ACU, the five-site avidity-linked guide-ADAR sensor design yielded the best activation fold change (Figure 30).

[0121] The modularity of protein payloads also allows for small payloads, such as HiBit payloads (Schwinn et al., 2018), which allow for in vivo transcript circularization. Circular RNAs offer a platform for enhanced residence time and minimal immunotoxicity (Katrekar et al., 2019), and we hypothesized that ADAR sensors with small payloads could be circularized to exploit these properties (Figure 31A). First, we developed two forms of short circular sensors. The canonical circular sensor is a scaffold of a U6 promoter-driven, double-twister ribozyme system (Litke and Jaffrey, 2019) and circularizes in vitro in the presence of RtcB ligase. The canonical circular sensor also contains a HiBit tag with a stop codon at the C-terminus of the HiBit. We found that the circular ADAR sensor induced target (IL6)-specific expression of HiBit (Figure 31B). To amplify the signal, we enhanced these circular ADAR sensors as endless ADAR sensors by removing the termination codon at the end of the payload and inserting 2A peptides at either end of the HiBit tag, enabling expression via rolling circle translation (RCT) (Abe et al., 2015). These rolling circle translation sensors are similar to conventional circular sensors, except that the termination codon within HiBit is removed and a T2ToA peptide is inserted to enable circular readthrough by the ribosome. HEK293 cells were transfected with these circular sensors (targeting IL6), and sensors of varying lengths were compared for sensor activation. The long sensor consistently increased the fold change in sensor activation. We found that rolling circle sensors can direct protein expression in a target-specific manner with minimal background leakage (FIG. 31B).

[0122] We investigated the effect of mRNA modifications on the fold change in sensor activation. Synthetic mRNAs have emerged as a useful therapeutic modality, but there are no methods for transcript-specifically controlling their payload expression. We explored the application of synthetic mRNA ADAR sensors for transcript-specific expression in a mouse model expressing human SERPINA1 transcripts in mouse hepatocytes. When delivering mRNA, the incorporation of base modifications such as 5'methylcytosine (5mc) and pseudouridine (Ψ) is essential for reducing host immune responses (Kauffman et al., 2016). However, these modifications can interfere with ADAR activity and affect the sensor function of mADARs.

[0123] The incorporation of 5-methylcytosine and pseudouridine was analyzed in HEK293 cells. 24 hours before mRNA transfection, HEK293 cells were supplemented with MCP-ADAR2dd either as a plasmid or directly as mRNA. An IL6 sensor with tetracycline-inducible IL6 was also used. We found that increasing the amount of Ψ reduced activation of the ADAR sensor, whereas 5mc was better tolerated, with 25% incorporation of 5mc resulting in the highest signal activation (Figure 32).

[0124] Using our inducible IL6 system to measure sensor activation, we further assayed the effects of different incorporation levels of a large panel of chemically modified bases on mRNA RADARS activation by transfecting modified IL6-sensing mRNA RADARS with exogenous ADAR1p150 (Figure 53A) or endogenous ADAR (Figure 53B). We found that all tested modifications reduced mRNA RADARS activation, likely due to interference with the ability of ADAR1p150 to edit the modified mRNA. Among modifications, we found that in the case of exogenous ADARp150, 50% incorporation of modified bases, such as 5-methylcytosine or 5-methyluridine, was best tolerated, while 100% incorporation of 5-methylcytosine resulted in the highest activation by endogenous ADAR. To determine whether this level of modification was sufficient to reduce the host immune response, we assayed for induction of interferon beta-related genes by chemically modified mRNA RADARS. We observed that even at a 25% incorporation level of modified bases, we still achieved minimal induction of inflammatory genes (Figure 53C).

[0125] We investigated a sensor called RADARSv1, which contains a 51-nucleotide IL6 transcript-sensing guide in front of a Gaussia luciferase (Gluc) payload, in combination with a constitutive Cypridiana luciferase (Cluc) on a separate transcript to provide ratiometric control for transfection variability (Figure 54A). With this RADARSv1 design and cotransfection with exogenous ADAR1p150, we observed approximately 5-fold activation (Figure 54A), as quantified by an increase in the Gluc / Cluc ratio in the presence of exogenous target IL6 expression (Figure 54B; elsewhere in this paper, the change in the Gluc / Cluc ratio between conditions is defined as the RADARS fold activation). Because ADAR1p150 prefers long double-stranded RNA as a substrate, we titrated the guide region from 51 nucleotides to 279 nucleotides in length, resulting in a slight increase in activation at 81 nucleotides, but with increasing length, activation decreased due to increased background payload expression in the absence of target RNA (Figure 54A, Figure 54B).

[0126] Three strategies were used to prevent the formation of dsRNA in the absence of a target, which is due in part to translational readthrough and self-folding. First, we introduced binding sites interspersed with multiple MS2 hairpin loops into the guide region to create secondary structure to prevent self-folding and enable multivalent binding. We optimized these engineered guides, termed engineered guide RNAs, by varying the number of MS2 loops and binding sites on the guide. RADARS activation was highest with an engineered guide RNA containing five binding sites interspersed with MS2 hairpin loops, which reduced background payload expression in the absence of a target compared to an uninterrupted guide design and achieved approximately 20-fold activation (Figures 54A and 54B).

[0127] Next, we increased the length of the translatable open reading frame (ORF) in front of the engineered guide RNA to promote termination and prevent ribosome re-triggering, which is known to depend on the upstream ORF length. We investigated 5' peptide lengths at ORF lengths of 0, 100, and 200 residues and found that at 200 residues, we could substantially reduce background translational readthrough (Figure 55A) and achieve >100-fold activation.

[0128] Finally, we engineered stop codons in the +1 and +2 frames following the engineered guide RNA region to capture translating ribosomes across all frames. These out-of-frame stop codon designs synergized with the extended 5' peptide to significantly reduce background and result in approximately 200-fold activation. We selected this sensor design, termed RADARSv2, incorporating a structured guide, upstream peptide, and out-of-frame stop codon as a unifying architecture for future sensors (Figures 54A and 55B).

[0129] We assessed our RADARSv2 design across exogenously expressed targets, IL6, EGFP, and neuropeptide Y (NPY), by tiling engineered guide RNAs across 14 CCA sites distributed across the transcripts. We found that while RADARS activation depends on the hybridization site chosen for a given target, the majority of sensors resulted in substantial payload activation in the presence of their targets, with activation less than 1,000-fold, demonstrating the generalizability of the RADARSv2 design (Figure 56A). To confirm that payload expression resulted from RNA editing, we harvested RNA from cells transfected with a panel of 14 IL6-targeting engineered guide RNAs and quantified editing by next-generation sequencing. In the presence of the target transcript, all 14 engineered guide RNAs resulted in greater than 15% editing, with an average of 35.1% ± 11.4%. In the absence of target transcripts, 13 of 14 engineered guide RNAs resulted in minimal editing (0.32% ± 0.34%). We also observed minimal editing by non-targeting sensors, reaffirming that RNA editing by RADARS sensors requires recognition of the target RNA by a specific engineered guide (Figure 56B).

[0130] [Example 7] Quantitative / correlation analysis of gene sensors To determine whether the above gene sensor could be used as a "dose-sensitive" sensor, an inducible EGFP transcript was introduced into HEK293FT cells. This EGFP transcript was placed under the control of a doxycycline-inducible promoter, and the cells were then exposed to 0, 8, 40, or 200 ng / mL of doxycycline to vary the expression of the EGFP transcript. In HEK293FT cells without exogenous ADAR transfection (Figure 6A), none of the guide strands showed significant differences in luciferase activity. However, guide strand 3 showed a trend toward dose sensitivity (Figure 33A, white bar).

[0131] When full-length ADAR2 was co-transfected with the guide strand into cells (Figure 33B), guide strand 3 exhibited a clear dose sensitivity, whereas guide 1 exhibited some dose sensitivity, but only to a lesser extent (Figures 33B-33C, white bars). Cells treated with 200 ng / mL doxycycline exhibited luciferase activity similar to that of cells constitutively expressing EGFP transcripts. As the dose of doxycycline decreased, a corresponding decrease in the luciferase activity identified in the cells was observed. This same trend was observed in cells co-exposed to full-length ADAR2 and guide strand 1 (Figure 33B, blue bars). Figure 34 depicts dose-dependent reporter expression when guide strand 1 (Figure 34A) and guide strand 3 (Figure 34B) targeting EGFP were transfected into cells. The luciferase fold activation also followed this dose-dependent trajectory (Figure 33C). Figure 35 depicts the level of luciferase activity as a function of GFP fluorescence. These results indicate that these gene sensors may be useful as quantitative sensors for transcript levels and not just as "on / off" sensors.

[0132] We performed siRNA perturbation experiments to compare the performance of RADARSv2 between exogenous ADAR1p150 and endogenous ADARs, comparing target expression for the gene with the highest transcript per million (TPM) (RPS5) and the least expressed gene (KRAS). We observed that both sensors detected siRNA knockdown without supplementation with exogenous ADARp150, but the RPS5 sensor benefited more from exogenous ADAR (Figure 57), suggesting that larger expression changes may benefit more from exogenous ADARs.

[0133] To further explore the quantitative value of the ADAR sensor, we measured the luciferase response of a seven-site avidity-linked guide with a double stop codon, which resulted in a wide range of expression levels, using both transfected and virally integrated versions of our inducible IL-6 expression system (Figure 36A). We found that luciferase activation by the ADAR sensor was linearly correlated with the concentration of the target transgene, as confirmed by qPCR (Figures 37 and 38, R2 = 0.96). Thus, RNA editing of the first stop codon within the ADAR sensor guide strongly correlated with gene expression levels (Figure 39), indicating that the ADAR sensor can quantitatively sense transcripts at both the RNA editing and payload levels.

[0134] To enable single-cell measurements with RADARSv2, we engineered a fluorescent payload for microscopy and flow cytometry-based readout (Figure 58A). We designed a fluorescent sensor as a single transcript containing the self-cleaving peptide sequence T2A preceding the mNeon payload, followed by an mCherry normalization control upstream, separated from the best EGFP-targeted engineered guide RNA by the self-cleaving peptide P2A sequence. We transfected HEK293FT cells with EGFP-targeted RADARS with or without a combination of exogenous ADAR1 p150 or a frameshifted, non-fluorescent EGFP target transcript. We observed mNeon fluorescent signal by microscopy in the presence of the target transcript and negligible background in the absence of the target transcript (Figure 58A). Quantification of the fluorescent signal by flow cytometry revealed a shift in the distribution of mNeon / mCherry ratios from 1.00% mNeon / mCherry positive cells in the absence of the target transcript to 56.1% mNeon / mCherry positive cells in the presence of the target transcript, representing a 38-fold increase in the geometric mean ratio (Figure 58B, Figures 59A-59C).

[0135] To further explore the quantitative accuracy of RADARS, we used transfected and virally integrated versions of the tetracycline-inducible IL-6 expression system to generate a wide range of expression levels and measured luciferase responses with the best IL-6-sensing engineered guide RNAs. RADARS-mediated luciferase activation was quantitative and linearly correlated with target transgene concentration, as confirmed by qPCR (Figures 60A and 60B, R2 = 0.95). Furthermore, RADARS activation was invariant to the amount of transfected sensor, and activation rates were robust across large titers of sensor loading, allowing fine-tuning of total sensor output independent of total sensor activation (Figure 60C).

[0136] For validation, we further designed sensors for a panel of 10 different transcripts, with TPMs ranging from approximately 10,000 to approximately 10 in HEK293FT cells. For each transcript, we compared eight different targeting engineered guide RNAs with eight non-targeting engineered guide RNAs. After siRNA transfection, we validated knockdown of these 10 genes by qPCR (Figure 61) and observed a significant reduction in RADARS signal for each transcript compared to the non-targeting sensor control. The robustness of the engineered guide RNAs correlated with expression: for highly expressed genes, the majority of the eight different targeting engineered guide RNAs detected knockdown of the target transcript, whereas for low-expressing genes, fewer engineered guide RNAs successfully detected knockdown (Figure 63A). Although RADARS sensitivity decreased with decreasing TPM, at least one of the eight engineered guide RNAs tested was still able to significantly detect transcript knockdown (Figure 62A). These data suggest that RADARS is sensitive to relative changes in gene expression across a wide range of expression levels. By measuring the editing rate of the stop codon UAG for the best-performing sensor for each of the 10 target transcripts (Figure 62B), we found that while overall editing rates were low for all 10 genes, there was a statistically significant reduction in editing rate when the target was knocked down. Because RADARS is overexpressed relative to endogenous targets with low turnover rates, the editing rate of the stop codon is less sensitive to perturbations in the target copy number.

[0137] We next sought to determine the sensitivity of RADARS by measuring changes in gene expression of endogenous targets. To examine RADARS across a range of endogenous transcript expression levels, we applied the sensor to measure siRNA-mediated transcriptional downregulation (Figure 63). We utilized a commercially available, validated siRNA pool targeting six endogenous genes, divided into highly expressed genes RPL41, GAPDH, and ACTB, and moderately to lowly expressed genes HSP90AA1, PPIB, and KRAS. For each transcript, we first compared eight different engineered guide RNAs for the highest sensitivity to knockdown (Figure 63A). We then titrated the amount of siRNA that resulted in a range of expression levels, confirmed by qPCR, and tracked changes in expression levels using the best engineered guide RNA supplemented with exogenous ADAR1p150. We observed that for all six genes, RADARS tracked transcript levels measured by qPCR with a high degree of Pearson correlation (R>0.86, Figure 61B). We found that for KRAS (Karlsson et al., 2021) expressed at 13 transcripts per million (TPM) in HEK293FT cells, the raw activation fold for RADARS activation deviated from the fold change measured by qPCR, likely due to loss of sensitivity at such low expression levels. However, RADARS responses were still highly correlated with KRAS levels determined by qPCR (R=0.93, Figure 61B).

[0138] Next, we explored whether RADARS could sense the upregulation of endogenous transcripts using a cellular heat shock model, which results in the upregulation of heat shock family genes. We designed RADARSv2 engineered guide RNAs targeting HSP70, a dynamic heat shock response protein, and co-transfected them with exogenous ADAR1p150 into HeLa cells before exposing the cells to heat shock at 42°C (Figure 64A). RADARS showed strong agreement with qPCR, with the best HSP70-targeting engineered guide RNA resulting in a 5.9-fold activation of HSP70 transcript expression levels compared to a 7.2-fold increase measured by qPCR in response to heat shock (Figure 64B). These results suggest that RADARS is sensitive to the upregulation of endogenous transcripts and can detect relative gene expression changes with high fidelity.

[0139] [Example 8] Logic gate We also attempted to determine whether the disclosed ADAR sensors could be multiplexed into logic systems, which could include AND and OR gates. These AND / OR methods are shown schematically in Figure 40A (AND) and Figure 40B (OR). The AND gate can fully deliver the payload only if both target strands are present. However, the OR gate can deliver the payload in the presence of one target strand, but not both. To create a rudimentary AND gate, we connected two 51-nucleotide single guides, targeting EGFP and IL6, respectively, in tandem with an MS2 hairpin loop. However, this design performed poorly due to a combination of low signal and background readthrough (Figure 40A).

[0140] To improve the AND gate signal, we used the RADARSv2 design and found that the resulting AND gate sensor behaved target-specifically, requiring both targets to reach full activation, with negligible leakage in single-target conditions (Figure 65A, Figure 65B). The AND chain exhibited 36-fold activation in the presence of both target transcripts, while activation in the presence of only one target RNA was only 1.3-1.5-fold (Figure 65B).

[0141] To engineer an OR logic gate, we co-transfected two five-binding site avidity sensors targeting EGFP and IL6. These sensors responded to the target EGFP or IL6 transcripts in a manner consistent with the OR gate (Figure 41B). The OR logic sensor exhibited a significant increase in fold change in the presence of each gene individually, but not in the absence of both genes. Overall, these results suggest that the modularity of ADAR sensors allows logical operations to be performed on mRNAs in living cells.

[0142] To improve the OR logic gate (Figure 66A), we co-transfected two engineered guide RNAs, RADARSv2 (upstream ORF, out-of-frame stop codon), targeting EGFP and IL6 transcripts, and found that the sensor responded to the target EGFP or IL6 transcript in a manner consistent with the OR gate (Figure 66B).

[0143] Example 9: Use of ADAR sensors to induce apoptosis in target cells To determine whether the disclosed ADAR sensor can be used with payloads other than reporters, we determined whether the ADAR sensor can induce apoptotic cell death in target cell populations. To apply the ADAR sensor to cell state-specific killing, we engineered a payload with therapeutically relevant iCaspase-9 (Straathof et al., 2005) (Figure 42A). The payload, iCaspase, was incorporated into a sensor chain capable of targeting human IL6. Mammalian cells were transfected with the ADAR sensor for caspases, target, and MCP-ADAR2dd. 24 hours after transfection, cells were split 1:5 into fresh medium, and samples with drug were supplemented with 10 nM AP20187 (Sigma-Aldrich). After 24 hours of further growth, cells were assayed for viability using the CellTiter-Glo Luminescent Cell Viability Assay (Promega). The control caspase was a sensor strand with a scrambled sensor region (i.e., the sensor strand did not specifically target IL6) and a caspase without an intervening stop codon. Cell death was measured using the CellTiter-Glo Assay (Promega) as the fold change in luminescence values ​​of cell lysates from the target group over the non-target group. We found that fusion of an IL6 sensor using a double stop codon 7 avidity guide in front of the caspase selectively killed IL-6-expressing cells with minimal toxicity in the absence of IL-6 induction (Figures 42B and 42C). IL-6-responsive caspases exhibited significantly enhanced induction of apoptotic cell death, indicating that ADAR sensors can be used to induce cell death in target cell groups. See Figure 42B. The percent cell survival in cells treated with IL6-responsive caspases and caspases without stop codons was also analyzed in cells with and without the target transcript (Figure 42C).

[0144] Next, we combined engineered guide RNAs with the payload iCaspase-9 to use highly specific SERPINA1-targeting engineered guide RNAs for cell-specific killing (Figure 67A) (Straathof et al., 2005). We cotransfected A549, HeLa, and HepG2 cells with RADARS against SERPINA1-iCaspase9 together with ADARp150 and assayed cell viability 48 hours after transfection. We found that SERPINA1-targeting RADARS-iCaspase selectively killed HepG2 cells with minimal toxicity in other cell types, while the non-targeting negative control showed no differential death (Figure 67B, Figure 67C).

[0145] Example 10: Use of ADAR sensors to track cell state and cell type To determine whether the developed ADAR sensor can be used to track cellular states, we first examined the heat shock response of HeLa cells. Two sets of HeLa cells were transfected with ADAR sensors designed with guides targeting heat shock family genes, including HSP70 and HSP40. HSP70 and HSP40 can be upregulated in an in vitro heat shock model (Figures 43A and 43B). ADAR sensors with 5- or 7-site avidity-binding guide designs detected upregulation of both HSP70 and HSP40 in cells exposed to heat shock (Figure 43A). HeLa cells (ATCC CCL-2) were transfected with ADAR sensors for HSP40 or HSP70. 24 hours after transfection, a portion of the cells were shifted to 42°C (5% CO2) for 24 hours. At the end of 24 hours of heat shock, the medium was collected and subjected to luciferase measurement.To control for non-specific changes in translation as a result of heat shock, we transfected scrambled non-targeted guide.By normalizing to non-targeted guide, we found that the activation of ADAR sensor in response to heat shock was less than 3-fold (Figure 43C).

[0146] We repeated the heat shock experiment with the RADARSv2 design, delivering only the RADARSv2 sensor without supplemental ADARs and engineered sensors (Figure 64). We found that the best HSP70 sensor (CCA42) exploits endogenous ADARs within HeLa cells to track HSP70 upregulation upon heat shock (Figure 68), thereby demonstrating the feasibility of a single-component RADARSv2 system deployed with endogenous ADARs.

[0147] Cell-type differences represent major variations in gene expression within tissues. Thus, we sought to determine whether ADAR sensors could accurately track cell-type differences. First, to identify marker transcripts for significant cell-type differences, we performed differential gene analysis between HEK293, HeLa, and HepG2 cells (Figure 44A). We selected SERPINA1, a hepatic serine protease inhibitor with a therapeutically relevant pathogenic variant (Boelle et al., 2019), as a marker expressed only in HepG2 cells but not in other cell lines (Figure 44B). We designed a panel of ADAR sensors with guides targeting SERPINA1 and examined their ability to distinguish HepG2 from HeLa cells. We found that the CCA30 guide design produced the greatest activation fold change between HepG2 and HeLa cells (Figure 44C). We transfected three different cell types with a SERPINA1 (CCA30)-targeted sensor along with a non-targeted scrambled sensor designed to control for differences in background ADAR editing, transfection variability, and protein production and secretion between the three cell types. Each cell type was transfected with the CCA30 SERPINA1 sensor, which contains five avidity regions connected by MS2 hairpins, with or without MCP-ADAR2dd. The fold change contributing to protein production / secretion and background ADAR kinetic differences between cell types (Figure 44D) was calculated via raw luciferase values ​​from the SERPINA1 sensor normalized to those from the scrambled non-targeted sensor, followed by normalization to the ratio within HEK cells for comparison between cell types.

[0148] The normalized fold change in editing rates in all three cell types in the presence of endogenous ADARs, supplemented ADARs, and controls was also analyzed (Figure 45). Various CCA sites on the SERPINA1 transcript were also used as targets.

[0149] To model liver-specific cell targeting in vitro, we expressed human SERPINA1 transcripts in Hepa-1-6 cells, synthesized the top CCA SERPINA1 sensors as mRNA in vitro, and transfected Hepa-1-6 cells with the mRNA sensors alone. We found that both the CCA30- and CCA35-targeting SERPINA1 sensors were able to recruit endogenous ADARs to sense the induction of SERPINA1 transcripts in Hepa-1-6 cells (Figure 46).

[0150] To assess RADARSv2 for cell type discrimination, we first leveraged the modular nature of RADARS to design a system for permanent genetic labeling of cell populations. We designed a double-loxP system for conditional / permanent labeling of cells with EGFP upon Cre expression and examined this reporter in HEK293FT cells in combination with ADAR1p150 and an IL6-targeted engineered guide RNA with a Cre payload. Upon IL6 induction, we observed significant production of EGFP protein with minimal signal in the absence of target RNA (Figures 69A and 69B).

[0151] Next, using the RADARSv2 design, we identified SERPINA1 as a differentially expressed marker gene in the liver-derived cell line, HepG2, compared with two non-liver cell lines, A549 and HeLa (Karlsson et al., 2021), which do not express SERPINA1. Using a SERPINA1-targeting engineered guide RNA to selectively activate Cre in HepG2 cells (Figure 70A), we co-transfected this sensor into HepG2, HeLa, and A549 cells along with ADAR1p150 and a Cre loxP reporter, and assessed activation relative to a non-targeting RADARS construct. While non-targeting engineered guide RNAs showed no reporter activation in either cell type, targeted engineered guide RNAs showed significant activation of the EGFP reporter only in HepG2 cells (Figure 70B, Figure 70C). These results establish that the RADARS system can distinguish cell types based on specific markers and that engineered guide RNAs and payloads can be combined in a modular manner for cell-type-specific expression of diverse transgenes.

[0152] Example 11: Use of ADAR sensors in vivo Next, to determine whether ADAR sensors can be used in vivo, we investigated SERPINA1 sensors in mice. We synthesized ADAR sensors targeting the CCA30 and CCA35 sites of human SERPINA1 in an Akaluciferase (Akaluc)-expressing construct (Yeh et al., 2019), which allows for easy noninvasive bioluminescence imaging to confirm cell-specific ADAR sensor activation (Figure 47). Prior to bioluminescence imaging, 8- to 10-week-old Albino B6 and NSG-PiZ mice were anesthetized with 3% isoflurane and injected with 5 μg of synthetic mRNA via retroorbital injection using in vivo-jet RNA transfection reagent (Polyplus). Eighteen hours after injection, mice were re-anesthetized with 3% isoflurane and immediately administered 100 μl of 15 mM AkaLumine-HCl (Sigma-Aldrich) for imaging. Ventral bioluminescence images were obtained using an IVIS Spectrum In Vivo Imaging System (PerkinElmer). The following conditions were used for image collection: exposure time = 60 seconds, binning = medium: 4, field of view = 12.5 × 12.5 cm, and f / stop = 1. Bioluminescence images were analyzed using Living Image 4.3 software (PerkinElmer). Since albino B6 mice do not express human SERPINA1, they represent a negative control (no site for binding to CCA30 or CCA35). To determine whether endogenous ADAR alone can edit the administered ADAR sensor, mice were not administered with an additional enzyme, ADAR. The SERPINA1-sensing mRNA RADARS design resulted in significant activation of Akaluc expression in NSG-PiZ mice compared to NSG-WT mice (p=0.007, N=3 mice, one-way ANOVA), and we observed no significant differences between the two strains under substrate-only background luciferase and constitutive Akaluc mRNA RADARS conditions (Figures 71A-71C).

[0153] In addition to the ADAR sensor targeting the CCA30 and CCA35 sites of SERPINA1, we also designed Akaluc payloads with constitutive ADAR sensors or scrambled non-targeting guides lacking a stop codon and expressing Akaluciferase (see Figure 47B). Three sensor systems were investigated. The SERPINA1-sensing ADAR sensor design significantly stimulated Akaluc expression in NSG-PiZ mice compared to wild-type mice (p = 0.04, N = 2 mice, unpaired two-tailed t-test), while the control guide did not result in a significant difference between the two strains (Figure 47B, Figure 48). This activation confirms that the ADAR sensor can be delivered as synthetic mRNA to sense cellular conditions in vivo via endogenous ADARs.

[0154] Further considerations Analysis of public tissue gene expression data (GTEx Consortium, 2013) shows that 34 of 37 tissues are differentiated by the sensor with a 3-fold sensitivity involving a single gene (Figure 49A), and an additional 3 tissues are classified by a combination of genes, supporting the straightforward application of both the sensitivity and logic input of the ADAR sensor (Figure 49B).

[0155] [Example 12] Examination of off-target effects and disruption of conventional cellular processes Because the RADARSv2 mechanism involves the formation of a long hybridization region between the sensor-driven guide RNA and the target transcript, we explored whether this duplex perturbs target transcript levels via Dicer knockdown or transcript stabilization. We compared target expression between the top-targeted engineered guide RNA condition and the non-targeting engineered guide RNA sensor condition for each endogenous transcript knocked down via siRNA (Figure 63A) and found no significant changes in target transcript expression (Figure 72A). In addition, to confirm that the resulting engineered guide RNA-target hybridization did not interfere with endogenous translation, we co-transfected ADAR1p150 into HEK293FT cells with ACTB-targeting or PPIB-targeting engineered guide RNA and quantified target protein levels by Western blot. Similar to mRNA levels, we observed that ACTB and PPIB protein levels were unchanged under targeted engineered guide RNA conditions compared to non-targeted engineered guide RNA conditions, confirming the lack of appreciable effects on RADARS-mediated target expression (Figure 72B, Figures 72C-72D).

[0156] Because intracellular dsRNA formation can activate immune response pathways, we next examined RADARS-induced upregulation of key endogenous innate immune signaling pathways involved in dsRNA responses (IFNB1, MDA5, OAS1, and RIG-1) by qPCR using both ACTB and GAPDH as normalization genes (Figures 73A and 73B). To compare with positive controls, we examined RADARS constructs targeting ACTB, PPIB, RPS5, and an exogenously introduced IL6 transgene together with high-molecular-weight poly(I:C), which acts as an analog dsRNA and activates these four pathways. We found that the RADARS constructs did not significantly upregulate any of the four dsRNA-responsive transcripts, whereas poly(I:C) caused significant activation of all four pathways (Figure 73A). To generalize our findings across cell lines, we also examined the same set of RADARS constructs in the HepG2 human hepatocellular carcinoma cell line and observed that the RADARS sensor did not induce a dsRNA response (Figure 73C).

[0157] We then explored whether overexpression of ADAR1p150 could lead to off-target editing within the transcriptome, as observed with ADAR-based therapeutic RNA editing (Cox et al., 2017; Qu et al., 2019; Reautschnig et al., 2022). First, we profiled the regions surrounding the hybridization duplexes of PPIB and ACTB transcripts with RADARS-driven guide RNAs, but found no significant off-target editing due to sensor hybridization or overexpression of ADAR1p150 (Figure 74A). Next, to unbiasedly examine potential off-targets, we performed polyA mRNA sequencing of cells expressing the PPIB sensor and ADAR1p150. We found that overexpression of ADAR1p150 in combination with PPIB-targeted RADARS resulted in only 23 detectable sites in the transcriptome, all of which showed less than 10% editing (Figure 74B). Furthermore, in the absence of ADARp150 overexpression, we found no significant site editing with non-targeting engineered guide RNAs. In contrast, when we analyzed published RNA-seq data from overexpression of the MCP-ADAR2(E488Q) deaminase domain, we detected >10,000 sites with significant A→I RNA editing, highlighting the impact of deaminase construct selection on off-target profiles (Figure 74C).

[0158] We found no significant homology between the engineered guide RNAs and the sequences surrounding the off-target editing sites (Figure 75A). Performing the same sequencing and analysis on a different engineered guide RNA targeting the exogenous target IL6 in the presence of ADARp150 overexpression revealed a low significant off-target editing rate (all 42 sites showed less than 10% editing, Figure 74B). Importantly, 22 of the 23 PPIB off-target sites were shared between the two different engineered guide RNA samples. Finally, we found that the sequence motif of the bases surrounding the editing sites closely resembled the preferred substrate of ADAR1 (Eggington et al., 2011) (Figure 75B). Overall, these observations imply that RADARS, when used with ADARp150 overexpression, results in relatively little nonspecific RNA editing within the transcriptome.

Claims

1. a) a synthetic single-stranded RNA (ssRNA) construct comprising: (i) a sensor domain comprising a first hybridization region and an ADAR-editable stop codon; and (ii) a payload; and b) Adenosine deaminase acting on RNA (ADAR deaminase) An RNA sensor system comprising: the second ssRNA is an endogenous ssRNA target in a human cell; the first hybridization region of the synthetic ssRNA construct is capable of hybridizing to the endogenous ssRNA target, forming a bimolecular double-stranded RNA (dsRNA) duplex with the synthetic ssRNA construct and the second ssRNA, the second ssRNA containing a mismatch within the stop codon of the sensor domain of the synthetic ssRNA construct; the dsRNA duplex is a substrate for the ADAR deaminase; the mismatch is editable by the ADAR deaminase, and the editing can effectively remove the editable stop codon so as to allow translation of the payload from the synthetic single-stranded RNA construct. RNA sensor system.

2. 2. The RNA sensor system of claim 1, wherein the mispairing within the stop codon of the sensor domain of the synthetic ssRNA construct comprises an adenosine-cytosine mispairing, including an adenosine in the sensor domain of the synthetic ssRNA construct opposite a cytosine in the endogenous ssRNA target in the dsRNA duplex.

3. The RNA sensor system of claim 2, wherein an adenosine in the sensor domain of the synthetic ssRNA construct at a mismatch within the stop codon is edited to an inosine by ADAR deaminase.

4. The RNA sensor system of claim 1 , comprising two or more mismatches within the bimolecular dsRNA duplex.

5. The RNA sensor system of claim 1 , wherein the payload comprises a reporter protein, a transcription factor, an enzyme, a transgene protein, or a therapeutic protein.

6. The RNA sensor system of claim 5 , wherein the payload comprises a therapeutic protein.

7. The RNA sensor system of claim 1 , wherein the payload comprises a fluorescent reporter.

8. The RNA sensor system of claim 7 , wherein the payload comprises an eGFP reporter or a luciferase reporter.

9. The RNA sensor system of claim 1 , wherein the payload comprises a caspase.

10. The RNA sensor system according to claim 1 , wherein the ADAR is an endogenous ADAR or an exogenous ADAR.

11. 2. The RNA sensor system of claim 1, wherein the ADAR deaminase comprises a programmable A to I(G) replacement RNA editing (REPAIR) molecule, a Cas13b-ADAR fusion molecule, a Cas13d-ADAR fusion molecule, a Cas7-11-ADAR fusion molecule, and an MS2-ADAR fusion molecule, the deaminase domain of ADAR2, full-length ADAR2, or a truncated ADAR2.

12. The RNA sensor system of claim 1 , wherein the sensor domain of the synthetic ssRNA construct further comprises a normalization gene.

13. The RNA sensor system of claim 1 , wherein the synthetic ssRNA construct is a circular RNA.

14. a) an AND gate, or b) OR gate a) an AND gate; (i) a synthetic single-stranded RNA (ssRNA) sensor construct, and (ii) Adenosine deaminase acting on RNA (ADAR deaminase) Including, the (i) synthetic single-stranded RNA (ssRNA) sensor construct comprises: (1) one or more payloads; and (2) a sensor domain comprising a plurality of ADAR-editable stop codons, at least a first hybridization region, and a second hybridization region; the first hybridization region of the sensor domain of the synthetic ssRNA sensor construct is capable of hybridizing to a first endogenous ssRNA target, the first endogenous ssRNA containing one of a plurality of ADAR-editable stop codons, and forming an RNA complex between the first hybridization region and the first endogenous ssRNA, the first endogenous ssRNA being a substrate for the ADAR deaminase; the second hybridization region of the sensor domain of the synthetic ssRNA sensor construct is capable of hybridizing to a second endogenous ssRNA target, the second hybridization region comprising one of the plurality of ADAR-editable stop codons, and forming an RNA complex of the second hybridization region with the second endogenous ssRNA target, the RNA complex being a substrate for the ADAR deaminase; the substrate contains a mismatch within the ADAR-editable stop codon, and the editing can effectively remove the stop codon to allow translation of one or more payloads; The OR gate in b) is (i) multiple independent synthetic ssRNA sensor constructs, and (ii) Adenosine deaminase acting on RNA (ADAR deaminase) Including, each of the (i) independent synthetic ssRNA sensor constructs comprises (1) one or more payloads, (2) an ADAR-editable stop codon, and (3) at least one first hybridization region; the first hybridization region of each synthetic ssRNA sensor construct is capable of hybridizing to at least one endogenous ssRNA target, forming a double-stranded RNA (dsRNA) duplex between the synthetic ssRNA sensor construct and the endogenous ssRNA target that is a substrate for ADAR deaminase, the dsRNA duplex containing a mismatch within at least one ADAR-editable stop codon of the independent synthetic ssRNA sensor construct; the substrate contains a mismatch within the ADAR-editable stop codon, and the editing can effectively remove the stop codon to allow translation of one or more payloads; Cellular logic system.

Images