Molecular switch for inducible gene expression

Abstract

Described herein are termination-readthrough-based gene switches that are small (only 9 nucleotides) and do not introduce any foreign protein into the target cell, and that can be used to alter expression levels of transgenes in a cell.

Description

[0001] MOLECULAR SWITCH FOR INDUCIBLE GENE EXPRESSION

[0002] CLAIM OF PRIORITY

[0003] This application claims the benefit of U.S. Provisional Application Serial No. 63 / 547,345, filed on November 4, 2023. The entire contents of the foregoing are incorporated herein by reference.

[0004] TECHNICAL FIELD

[0005] Described herein are termination-readthrough-based gene switches that are small (only 9 nucleotides) and do not introduce any foreign protein into the target cell, and that can be used to alter expression levels of transgenes in a cell.

[0006] BACKGROUND

[0007] Gene switches that can artificially regulate gene expression are in high demand for developing safe and effective gene therapies.

[0008] SUMMARY

[0009] Provided herein are nucleic acid cassettes comprising a coding region for a transgene of interest, wherein the coding region comprises a 9-nucleotide TRSTOP sequence selected from TGACTAGAC, TAGCTAGAC and TAACTAGAC inserted into the coding region of the transgene to minimize expression of the transgene, preferably wherein the TRSTOP sequence is TGACTAGAC. In some embodiments, the TRSTOP sequence is inserted within a 5’ half of the coding region, and wherein expression of the 5’ half does not result in production of a functional protein product. These constructs can be referred to herein as TREAD constructs.

[0010] In some embodiments, the transgene of interest comprises a CRISPR Cas protein. In some embodiments, the CRISPR Cas protein is SaCas9 or a variant thereof. In some embodiments, the CRISPR Cas protein is a base editor.

[0011] Additionally provided herein are nucleic acid cassettes comprising, from 5’ to 3’: a first loxP sequence; a coding region for a transgene of interest, wherein the stop codon of the transgene is replaced with a TRSTOP sequence selected from TGACTAGAC, TAGCTAGAC and TAACTAGAC; a coding region for a Cre recombinase, preferably linked to a nuclear localization sequence (NLS); and a second loxP element in the same direction as the first loxP sequence. In some embodiments, the cassettes comprise a linker sequence between the TRSTOP and the sequence encoding the Cre recombinase / NLS. These constructs can be referred to herein as TREAD-dimmer constructs.

[0012] In some embodiments, the transgene of interest comprises a CRISPR Cas protein, optionally SaCas9 or a variant thereof.

[0013] In some embodiments, the transgene of interest is a suicide protein. In some embodiments, the suicide protein is herpes simplex virus thymidine kinase (HSVTK); cytosine deaminase (CD); carboxypeptidase G2 (CPG2); nitroreductase (NTR); Cytochrome P450 (CYP); purine nucleoside phosphorylase (PNP); horseradish peroxidase (HRP); or carboxylesterase (CE).

[0014] Also provided herein are expression constructs comprising the nucleic acid cassettes described herein and a promoter (and optionally other regulatory sequences) to drive expression of the transgene.

[0015] In some embodiments, the expression construct is a viral vector or a recombinant bacterial or eukaryotic plasmid.

[0016] In some embodiments, the transgene of interest comprises a CRISPR Cas protein, and wherein the construct further comprises sequences encoding one or more guide RNAs (gRNAs), optionally further comprising at least one additional promoter to drive expression of the one or more gRNAs.

[0017] Also provided herein are methods of increasing expression of a transgene in a cell. The methods comprise expressing in the cell a TREAD nucleic acid cassette as described herein; contacting the cell with a chemical inducer of stop codon readthrough, optionally wherein the chemical inducer is selected from the group consisting of gentamicin, G418, and PTC 124, in an amount sufficient to increase readthrough of the TRSTOP, thereby increasing expression of the transgene in the cell.

[0018] Additionally, provided herein are methods of reducing expression of a transgene in a cell. The methods comprise expressing in the cell a TREAD-dimmer nucleic acid cassette as described herein; contacting the cell with a chemical inducer of stop codon readthrough, optionally wherein the chemical inducer is selected from the group consisting of gentamicin, G418, and PTC 124. in an amount sufficient to increase readthrough of the TRSTOP, thereby decreasing expression of the transgene in the cell.

[0019] In some embodiments, the transgene is a suicide protein, and the method further comprises contacting the cell with a nontoxic prodrug that is a substrate for the suicide protein, wherein action of the suicide protein on the nontoxic prodrug results in production of a toxic metabolite that induces cell death. In some embodiments, the suicide protein is herpes simplex virus thymidine kinase (HSVTK) and the nontoxic prodrug is ganciclovir (GCV); the suicide protein is cytosine deaminase (CD) and the nontoxic prodrug is 5-flourouracil (5-FU); the suicide protein is carboxypeptidase G2 (CPG2) and the nontoxic prodrug is nitrogen mustard (NM) or a derivate thereof such as ZD2767P or CMDA (4-[(2-chloroethyl)(2-mesyloxyethyl)amino]benzoyl-L- glutamicacid); the suicide protein is nitroreductase (NTR) and the nontoxic prodrug is CB1954 or an analog thereof; the suicide protein is Cytochrome P450 (CYP) and the nontoxic prodrug is an oxazaphosphorine drug such as cyclophosphamide (CPA) and ifosfomide (IFO); the suicide protein is purine nucleoside phosphorylase (PNP) and the nontoxic prodrug is 6-Methylpurine Deoxyriboside or an analog thereof, optionally fludarabine phosphate (F-araAMP) or 2-fluoro-2-deoxyadenosine (F- dAdo); the suicide gene is horseradish peroxidase (HRP) and the nontoxic prodrug is indole-3 -acetic acid (HRP / IAA); or the suicide protein is carboxylesterase (CE) and the nontoxic prodrug is irinotecan.

[0020] Also provided herein are cells comprising the constructs or expression constructs described herein. In some embodiments, the cell is a mammalian cell. In some embodiments, the cell is in vitro. In some embodiments, the cell is in vivo.

[0021] Provided herein are methods for creating an adjustable termination-read through-based gene expression mechanism comprised of the insertion of a six, nine, or twelve nucleotide DNA sequence (called TRSTOP) wherein (a) the nucleotides include one stop codon and one or two flanking sequences; (b) the stop codon is a leaky stop codon; (c) the leakiness of the leaky stop codon is affected by the flanking sequence(s), and. optionally, (d) the leakiness of the leaky stop codon can be regulated by an aminoglycoside antibiotic.

[0022] In some embodiments, the nine-nucleotide sequence that makes up the leaky stop codons is comprised of, but not necessarily restricted to: TGA, TAG or TAA. In some embodiments, there can be one flanking sequence on one side of the stop codon or two flanking sequences on both sides of the stop codon.

[0023] In some embodiments, one flanking sequence is CTAGAC.

[0024] In some embodiments, the nucleotide sequence of TRSTOP is TGACTAGAC, TAGCTAGAC, or TAACTAGAC.

[0025] In some embodiments, the aminoglycoside antibiotic is Gentamicin or G418 .

[0026] Also provided herein is the use of the TRSTOP mechanism as described above a regulatable gene switch comprised of, but not limited to, the gene switch examples as described herein. For example, the gene switch can be used for RNA therapy, wherein the 9 nt DNA sequence TRSTOP and a chemical inducer such as gentamicin are used, e.g., in an AAV-TRSTOP-Cas9-gRNA (targeting eye disease) plus gentamicin, or for gentamicin-inducible mRNA therapy.

[0027] Additionally provided herein is the use of the TRSTOP mechanism as described herein as a bicistronic design for gene co-expression comprised of, but not limited to, AAV-AADC-TRSTOP-GDNF or AAV-antiPD-Ll-TRSTOP-IL-12. In this embodiment, the design includes Gene #1 plus TRSTOP plus Gene #2 (wherein 5-10% of gene#l level is necessary and sufficient for therapy.

[0028] Further provided herein is the use of the TRSTOP mechanism as described herein as a “Gene Dimmer” comprised of, but not limited to, examples such as LoxP plus Gene #1 plus Cre plus LoxP (or others such as FLP-FRT)(preventing overexpression toxicity).

[0029] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs. Methods and materials are described herein for use in the present invention; other, suitable methods and materials known in the art can also be used. The materials, methods, and examples are illustrative only and not intended to be limiting. All publications, patent applications, patents, sequences, database entries, and other references mentioned herein are incorporated by reference in their entirety. In case of conflict, the present specification, including definitions, will control.

[0030] Other features and advantages of the invention will be apparent from the following detailed description and figures, and from the claims. DESCRIPTION OF DRAWINGS

[0031] FIGs. 1A-C. Termination readthrough efficiencies of the TRSTOP sequences as measured using Cre as reporter. (A) Design of an expression cassette consisting of GFP, a 9-nucleotide TRSTOP sequence, Cre, and a termination signal TAATAATAA. Coding sequences for four fusion proteins of GFP-Cre were designed in such that a TGA-containing TRSTOP " sequence TGACTAGAC and three its derivatives TGGCTAGAC, TAGCTAGAC and TAACTAGAC were used to link the coding sequences for GFP and Cre. The parent sequence TGACTAGAC was derived from the Sindbis virus. DNA plasmids expressing the four GFP-Cre fusion proteins were transfected into HEK293 cells with a second plasmid DI0-H2B-RFP that expresses nuclear RFP in a Cre-dependent manner. No termination readthrough a the TRSTOP sequences would result in expression of GFP only, whereas readthrough would lead to the expression of GFP-Cre fusion proteins. Relative fluorescence intensity of the Cre- induced RFP versus GFP was quantified as a measurement of readthrough efficiencies of TRSTOP sequences. A second plasmid pDIO-H2B-RFP was co-transfected which expresses a Cre-dependent nucleus-bound RFP. Figure adapted from Li & Rice, 1993. (B) Each of the four plasmids pGFP-TRSTOPTGG-Cre, pGFP-TRSTOPTGA-Cre, pGFP-TRSTOP.TAG-Cre, and pGFP-TRSTOP.TAA-Cre was co-transfected with a second plasmid DI0-H2B-RFP that expressed nuclear RFP in a Cre-dependent manner. Representative images of GFP and H2B-RFP-labeled HEK293 cells are shown. (C) Relative fluorescence intensity of RFP versus GFP was quantified as show n in the histogram. All three stop codons (TGA, TAG, and TAA) exhibits termination readthrough. Scale bar in b: 200 pm. **** P < 0.0001, one-way ANOVA with Dunnetf s post-test.

[0032] FIGs. 2A-C. Effects of gentamicin (2A), PTC124 (2B), and G418 (2C) on enhancing termination readthrough. (A) Gentamicin increased readthrough efficiencies of TGA, TAG, and TAA in dose-dependent manners, as measured bypercentages of RFP-positive out of GFP-positive cells, with the most marked effects seen on TGA codons. (B) PTC 124 and (C) G418 also modulated TGA codons in a dose-dependent manner. Data: mean ± SEM. * P < 0.05, ** P < 0.01, *** P < 0.001, **** P < 0.0001, one-way ANOVA with Dunnetf s post-test.

[0033] FIGs. 3A-H. Design and testing of the gene sw itch TREAD " (A) TREAD w as comprised of two components: 1) a 9-nucleotide TRSTOP (TGACTAGAC) that could be inserted into the coding sequence of the target gene and limited its expression and 2) a chemical inducer (such as gentamicin, PTC124, G418). For a typical gene therapy vector that consists of a promoter, TRSTOP-incorporated coding sequence of a target gene and a polyadenylation signal (pA), administering an inducer of desired doses could be used to adjust expression levels of the target gene. Top, basic expression construct. Bottom, AAV expression construct. For temporal control of CRISPR-Cas9 gene editing, the 9-nucleotide TRSTOP sequence was inserted into the saCas9 expression cassette at the Ser431 / Gln432 site of the Cas9 protein. (B) TREAD regulated CRISPR gene editing in cell culture, (top) Design of two “all-in- one” AAV constructs consisting of expression cassettes for Cas9 (wildtype or TRSTOP-incorporated) and targeting gRNA. The gRNA was designed to direct Cas9 to cut out the “LSL” site in a co-transfected plasmid pLSL-GFP and activate GFP expression upon successful editing. The 9-nucleotide TRSTOP sequence was inserted into the Cas9 expression cassette at the Ser431 / Gln432 site of the Cas9 protein, (bottom left) Representative images of DAPI and GFP-labeled HEK293 cells after cotransfection of Cas9 plasmid with pLSL-GFP with different chemical treatments.

[0034] (bottom right) Percentage numbers of GFP -positive cells versus DAPI-positive total cells. Scale bar in b: 200 pm. **** p < 0.0001, one-way ANOVA with Dunnetf s posttest. (C) Schematic illustration of an exemplary system for testing TREAD-ON in vivo in a mouse model. EFS, elongation factor la short promote (D) Validation of “Switchable” AAV-CRISPR gene editing in mice was evaluated using two doses of PTC 124. AAV2 carrying an “all-in-one” CRISPR-Cas9 system was injected intravenously into an adult Ail4-RFP reporter mouse at a dose of 2ell vg per mouse. A pair of specific gRNAs was designed in the vector to induce expression of RFP in cells after Cas9-mediated DNA deletion. AAV2 with wild-type Cas9 resulted in robust RFP expression indicative of extensive gene editing. AAV with TREAD-controlled Cas9 (“switchable” Cas9) did not result in gene editing in the absence of PCT124. PTC 124 induced “switchable” Cas9-mediated gene editing in a dose-dependent manner (low dose: 60 mg / kg / day via gavaging for two weeks plus 50 mg / kg IP every two days for two weeks; high dose: 300 mg / kg / day via gavaging for two weeks plus 50 mg / kg IP every two days for two weeks). Images show expression in liver. (E-F) “Switchable” AAV-CRISPR gene editing was evaluated in mouse liver using

[0035] AAV.CPP. 16 for delivery. AAV.CPP. 16 carrying an “all-in-one” CRISPR-Cas9 system was injected intravenously into an adult AH4-RFP reporter mouse at a dose of 5ell vg per mouse. A pair of specific gRNAs was designed in the vector to induce expression of RFP in cells after Cas9-mediated DNA deletion. AAV.CPP. 16 with TREAD-controlled Cas9 (“switchable” Cas9) resulted in little gene editing in the liver with com oil control treatment alone. AAV with wild-type Cas9 resulted in robust RFP expression indicative of extensive gene editing. PTC124 induced “switchable” Cas9-mediated gene editing with a comparable efficiency to wild-type Cas9 (300 mg / kg / day via gavaging for two weeks plus 50 mg / kg IP every two days for two weeks). Scale bar: 200 m. * P < 0.05, ns = not significant, one-way ANOVA with Dunnetf s post-test. The bar graph in (F) shows quantification of the expression seen in liver shown in (E). (G) AAV.CPP. 16 carrying an “all-in-one” CRISPR-Cas9 system was injected intravenously into an adult Ail4-RFP reporter mouse at a dose of 5ell vg per mouse. A pair of specific gRNAs was designed in the vector to induce expression of RFP in cells after Cas9-mediated DNA deletion. AAV.CPP. 16 with TREAD-controlled Cas9 (“switchable” Cas9) did not result in gene editing in the skeletal muscle with com oil control treatment alone. AAV with wild-type Cas9 resulted in robust RFP expression indicative of extensive gene editing. PTC 124 induced “switchable” Cas9-mediated gene editing with a comparable efficiency to wild-type Cas9 (300 mg / kg / day via gavaging for two weeks plus 50 mg / kg IP every two days for two weeks). Scale bar: 200 pm. (H) AAV.CPP. 16 carrying an “all-in- one” CRISPR-Cas9 system was injected intravenously into an adult Ail 4-RFP reporter mouse at a dose of 5el 1 vg per mouse. A pair of specific gRNAs was designed in the vector to induce expression of RFP in cells after Cas9-mediated DNA deletion. AAV.CPP. 16 with TREAD-controlled Cas9 (“switchable” Cas9) did not result in gene editing in the heart with com oil control treatment alone. AAV with wild-type Cas9 resulted in robust RFP expression indicative of extensive gene editing. PTC 124 induced “switchable” Cas9-mediated gene editing with a comparable efficiency to wild-type Cas9 (300 mg / kg / day via gavaging for two weeks plus 50 mg / kg IP every’ two days for two weeks). Scale bar: 200 pm.

[0036] FIGs. 4A-B. Schematic illustration of an exemplary TREAD-dimmer embodiment. (A) schematic illustration of an exemplary transgene expression construct in which the stop codons of a transgene are replaced with a TRSTOP sequence, followed by a Cre recombinase; loxP elements are present at the 5’ and 3’ end of the constructs, flanking the transgene-TRSTOP-Cre cassette. To regulate expression of transgene of interest, the sequence for the gene of interest is linked downstream to a 9-bp TRSTOP sequence (e.g., TGACTAGAC), a sequence for a linker peptide (e.g., 9-glycine), a sequence for a Cre recombinase, and a 3x TAA translation termination sequence (i.e., TAATAATAA), in one open reading frame. The entire reading frame is then flanked by two 34-bp loxP sequences oriented in the same direction. (B) When the gene therapy vector is administered to the target cell at a low genetic dosage, expression level of the transgene is low and no or little expression of Cre is expected due to the low readthrough efficiency of TRSTOP. When the target cell receives high-dosage gene therapy vector, the expression levels of both the transgene and Cre are increased. Increased Cre expression eventually triggers DNA recombination within the gene therapy vector and deletion of the floxed transgene and Cre sequences. Eventually, the genetic dosage of the transgene is reduced and so is the expression product of the transgene. Thus, in cells in which only a low genetic dose of the construct is present, the low likelihood of readthrough limits the expression of the Cre, and provides low expression of the Cre as weak as the transgene (top). However, in cells receiving a high genetic dose (numerous copies of the construct), the higher likelihood of readthrough allow s for expression of Cre, which excises at least some of the floxed constructs, reducing copy number and thus expression of the transgene (bottom).

[0037] FIGs. 5A-B show' that ‘ TREAD-Dimmer” diminishes gene expression at a high genetic dose in vitro. (A) Schematic illustration of a conventional (left) and TREAD-dimmer (right) plasmid. (B) Fluorescent images of GFP expression in cells transfected with the conventional (left) and TREAD-dimmer (right) plasmids. On the left, the conventional plasmid provided variable expression across the cells, with some showing low expression and some showing high expression. On the right, more uniform expression w as seen in the cells transfected w ith the TREAD-Dimmer plasmid. GFP DNA plasmids with conventional gene expression cassette (i.e., CAG- GFP) or TREAD-Dimmer-controlled gene expression cassette (CAG-loxP-GFP- TRSTOP-Cre-loxP) were transfected using polyethylenimine into HEK293 cells cultured in a 24-well plate. A high dose of plasmid DNA w as used (2 pg per w ell). GFP fluorescent images were captured 48 hours after transfection. FIGs. 6A-B show that TREAD-Dimmer reduced AAV-mediated gene expression in the mouse brain. (A) schematic illustration of AAV construct AAV.CPP. 16-GFAP-loxP-GFP-TRSTOP-Cre-loxP. AAV.CPP. 16-GFAP-loxP-GFP- TRSTOP-loxP was injected intravenously into an adult mouse at a dose of 5ell vg per mouse. (B) Mouse brain was processed for GFP imaging three weeks after AAV administration. Without GFP antibody staining to amplify GFP signal, very low level of GFP fluorescence was observed in the mouse brain (left panel). Notable GFP- positive cells were observed after GFP antibody staining (right panel), indicating TREAD-Dimmer-controlled AAV vector w as capable of mediating GFP expression in target cells albeit with a reduced expression level. GFAP is an astrocyte-specific promoter. Compare to (C) which shows expression patterns from a conventional AAV.CPP. 16.

[0038] DETAILED DESCRIPTION

[0039] Genetic mutation is a fundamental cause of disease. According to the National Institute of Health, it directly contributes to more than 7,000 known rare genetic diseases that affect millions of people in the United States, ft is also linked to many common diseases such as cancer, diabetes, heart diseases, and dementia. Gene therapy aims to treat or even cure diseases by using genetic tools and materials. It has demonstrated great promise as a new- therapeutic modality with technical advances in areas including gene delivery’, gene editing and mRNA therapeutics (Dunbar et al., 2018). ‘'On-command” controlling technologies that can turn on or off gene therapy drugs are in high demand; they are particularly desired for management of treatment- associated side effects and fine-tuning of genetic dosing for better clinical outcomes (Madderson et al., 2021). For instance, such gene switch tools could be applied in optimizing expressions of CR1SPR (i.e.. Clustered Regularly Interspaced Short Palindromic Repeats) gene-editing enzymes to when and where they’re needed, thereby reducing immunogenicity’ and risk of off-target editing associated w ith their overexpression. Making CRISPR gene editing “switchable” would substantially improve the safety and feasibility of in vivo gene editing.

[0040] Previous engineered gene switches, such as the “classic” Tet-On and Tet-Off systems (Gossen & Bujard, 1992), typically include tw o components: 1) a responsive DNA “effector” that was incorporated into the target gene of interest and made its function inducible; and 2) a corresponding “inducer” that could be a small chemical, light, heat, electrical signal, or magnetic force and that acted on the DNA effector or its protein products for regulating the target gene (Madderson et al., 2021). There are two major limitations with existing gene switches: first, the DNA effectors are often very large in size (several thousands of nucleotides or base pairs long), which makes it challenging to incorporate them into many gene therapy vectors including the widely used adeno-associated virus (AAV) vectors (Wang et al., 2019). Such size problem would be exacerbated when simultaneously working with a large-sized target gene such as all CRISPR enzymes and a relatively low -capacity delivery vector such as AAV. Second, foreign “sensor” proteins are often generated to mediate the effects of the exogenous inducers in previously reported gene switches, and those sensor proteins have the inherent risk of immunogenicity. Developing a small-sized gene switch without introducing additional immunogenicity would provide a safety control mechanism for cell and gene therapy, especially for in vivo gene therapy.

[0041] Termination readthrough occurs when protein translation does not stop at one of the three stop codons: UGA, UAG, and UAA (or TGA, TAG, and TAA as their corresponding stop codons in DNA). This phenomenon has been observed in both prokaryotic and eukary otic systems. Some viruses such as bacteriophage Q , tobacco mosaic virus, Moloney murine leukemia virus, and Sindbis virus utilize the mechanism of termination readthrough to produce essential viral proteins probably to maximize economy of their small genomes (Li & Rice, 1993; Pelham, 1978; Weiner & Weber, 1971; Yoshinaka et al., 1985). Termination readthrough has also been explored as a therapeutic strategy' to overcome premature termination codons in genetic diseases that arise from non-sense mutations (Keeling et al., 2014). Small- molecular compounds that promote termination readthrough include both aminoglycosides, such as gentamicin and G418, and non-aminoglycosides, such as PTC 124 and amlexanox (Li et al., 2023).

[0042] Described herein are termination-readthrough-based “mini” gene switches that are small (only 9 nucleotides) and do not introduce any foreign protein into the target cell. Using a reliable in vitro assay for studying termination readthrough, all three stop codons were shown to be leaky to varying degrees. This leakiness could be modified by nucleotide contexts and chemical drugs such as gentamicin, G418. and PTC 124 (i.e., ataluren), which have stop-codon-dependent effects on termination readthrough. A gene switch was engineered using a 9-nucleotide, stop codon TGA-based DNA effector and a chemical inducer. As shown herein, these gene switches can be used for controlling CRISPR gene editing in cells and animals, and can be translated into clinical use for safe and efficacious gene therapy.

[0043] Termination Readthrough based Expression Adjusting Device (TREAD)

[0044] The TREAD cassettes described herein include a 9-nucleotide TRSTOP sequence (e.g., TGACTAGAC) as a DNA “effector’ that is inserted into the coding region of a target gene at a place to keep expression of the target gene at a minimal baseline level. Methods using the TREAD cassettes can include two components; the tread cassette (preferably in an expression construct comprising a promoter and / or other regulatory sequences to drive expression of the transgene) and a chemical “inducer” that increases target gene expression (by increasing readthrough of the TRSTOP sequence) when applied (see, e.g., FIG. 3A). The TRSTOP sequence can be inserted into the transgene coding sequence, optionally at a point that creates a nonfunctional gene product (e.g., any transcript produced codes for a non-functional protein), or as a replacement of the endogenous stop codon.

[0045] TREAD has a number of advantages. The small size (9nts) works within the size limitations of AAVs, which have been shown to be a versatile delivery vehicle (e.g., for ocular, neuro, muscular, or global targets). The “All-in-one” design (e.g.. FIG. 3A) facilitates clinical translation, since only one AAV needs to be produced. Adding the TRSTOP sequence would have no impact on AAV manufacturability. In addition, transient expression potentially circumvents issues associated with permanent CRISPR activity, including potential immunogenicity of CRISPR enzymes and off-target effects, as well as resolves issues around AAV genome integration. Thus, the methods and compositions provide better safety profiles in therapeutic DNA editing using small-sized Cas9s and base editors.

[0046] TREAD-Dimmer

[0047] Under some circumstances, overexpression of a transgene can be detrimental; because of the nature of viral delivery methods, inconsistent delivery can result in some cells having numerous copies of a construct that could result in deleterious overexpression. The present methods provide for self-regulated gene expression, referred to herein generally as ’TREAD-Dimmer”. The present methods and compositions can include the use of “TREAD- dimmef ’ cassettes in which the stop codons of a transgene are replaced with a TRSTOP sequence, followed by a Cre recombinase; loxP elements are present at the 5’ and 3’ end of the cassettes, flanking the transgene-TRSTOP-Cre cassette (see FIG. 4A; cassettes with flanking loxP sequences are referred to as “floxed”). In cells in which only a low genetic dose of the cassette (preferably in an expression construct comprising a promoter to drive expression of the transgene) is present, the low likelihood of readthrough limits the expression of the Cre, and provides low expression of the Cre as well as the transgene (see FIG. 4B, top). However, in cells receiving a high genetic dose (numerous copies of expression constructs comprising the TREAD-dimmer cassette), the higher likelihood of readthrough allows for expression of Cre, which excises at least some of the floxed constructs, reducing copy number and thus expression of the transgene (see FIG. 4B, bottom). This holds true with AAV delivery in vitro and in vivo (see FIGs. 5A-B and 6A-B). The level of readthrough can be controlled by administration of one or more doses of a chemical inducer as described herein that increases readthrough.

[0048] TREAD-dimmer is particularly useful for preventing transgene overexpression in those contexts where gene dosage is critical; for preventing overexpression of RNA therapeutics (e.g., AAV-microRNA for HD, PD and AD); and for restricting trans gene expression in the CNS (e.g.. cancer gene therapy for brain tumors using a construct including a floxed HSV-TK1 -TRSTOP-Cre cassette, or for neurotropic factor therapy for neurodegeneration). The methods provide the option of controlling trans gene dosage (e.g., with a chemical inducer), as well as the option of deleting trans genes outside the CNS (e.g., by systemically administering a chemical inducer that doesn’t penetrate the blood-brain barrier.

[0049] Optionally, there is a linker sequence between the various elements, e.g., between TRSTOP and Cre. In some embodiments, the constructs include one or more flexible linkers. The linkers can be used to attach the separate parts of the fusion protein together. In some embodiments, the linker is a peptide linker. Peptide linkers can be from about 2-100, 10-50, or 15-30 amino acids long. In some embodiments, peptide linkers may be at least 2, 4, 5, 6, 10, 15, or at least 20 amino acids long and / or up to 20, 25, 35, 40, 60, 80, 90, or no more than 100 amino acids long. In some embodiments, the linker is a peptide linker comprising one or more glycines and / or serines, e.g., a single or repeating GGGGS (SEQ ID NO: 1), GGGS (SEQ ID NO:2), GS. GGGGGG, GSGGS (SEQ ID NO: 3). GGSG (SEQ ID NO: 4). GGSGG (SEQ ID NO: 5), GSGSG (SEQ ID NO: 6), GSGGG (SEQ ID NO: 7), GGGSG (SEQ ID NO: 8), and / or GSSSG (SEQ ID NO: 9) sequence(s). Other linkers are known in the art. Intact antibodies with desired specificity can also be fused to glycosylation enzy mes, enabling specific targeting of the enzymes. Further, similar protein fusions can be generated using dog / cat / horse / cow equivalent / homologous antibodies or glycosylation enzymes, enabling treatment of non-human animals (e.g., pets and livestock).

[0050] TRSTOP sequences

[0051] The present methods and compositions use transgene coding sequences that include 9-nucleotide TRSTOP sequences TGGCTAGAC, TGACTAGAC, TAGCTAGAC and TAACTAGAC; preferably the TRSTOP sequence is TGACTAGAC. In some embodiments, the sequence comprises 6-nt portion of the above sequence such as TGACTA, or 9-nt sequences NNNTGACTA, wherein N is any amino acid.

[0052] Cre.'Lox Sequences

[0053] Cre-loxP recombination systems include two main components: a Cre recombinase (preferably linked to a nuclear localization sequence) and locus of crossover (x) Pl (loxP) sites. The loxP are DNA sequences consisting of two 13 basepair inverted and palindromic repeats separated by an 8 base-pair core asymmetric sequence spacer region; the canonical loxP sequence is ATAACTTCGTATA- GCATACAT-TATACGAAGTTAT (SEQ ID NO: 10). When two loxP sites flank a segment of DNA in the same direction (referred to as '‘flexed’’), the Cre recombinase excises the segment between the sites, deleting any intervening sequences. See, e.g., Schntitgen et al., Nat Biotechnol. 2003 May;21(5):562-5; Lee and Saito, Gene. 1998 Aug 17;216(l):55-65; Sauer and Henderson, Proc Natl Acad Sci U S A. 1988 Jul;85(14):5166-70; Orban et al.. Proc Natl Acad Sci U S A. 1992 Aug l;89(15):6861-5; Sternberg and Hamilton, J Mol Biol. 1981 Aug 25;150(4):467-86.

[0054] In some embodiments, the Cre sequence is modified to include a nuclear localization sequences (NLSs), e.g., at the C- and / or N-terminus of the Cre protein. Exemplary NLSs include SV40 large T antigen 5 NLS (PKKKRRV (SEQ ID NO: 1 1)); PKKKRKV (SEQ ID NO: 12); KRTADGSEFESPKKKRKV (SEQ ID NO: 13); and nucleoplasmin NLS (KRPAATKKAGQAKKKK (SEQ ID NO: 14)). Other NLSs are known in the art; see, e.g., Cokol et al., (2000) EMBO Rep. Nov 15; 1 (5):411- 415; Freitas and Cunha, (2009) Curr Genomics. Dec; 10(8): 550-557; and Leung et al (2003) Journal of Biol Chem. 10 278(43):41947-41953.

[0055] Transgenes

[0056] Transgenes that can be used in the present methods and compositions include any gene of interest, such as mammalian and non-mammalian genes, e.g., non-sindbis viral genes. Preferably, the genes are therapeutic genes.

[0057] In some embodiments, the transgenes are or comprise CRISPR Cas proteins, e.g., nucleases, base editors, prime editors, and so on. In some embodiments, the Cas protein comprises SaCas9, Nme2Cas9, or SpCas9, or a variant or fusion protein thereof. Variants that have altered PAM requirements or activity are known in the art. In some embodiments, the Cas protein fusion comprises a base editor, e.g., Nme2ABE8e, SaKKHABE8e. CjABE8e, or SauriABE8e. See, e.g., Pacesa et al., Cell, Volume 187, Issue 5, 1076 - 1100; Zhou and Yao, Mol Biomed. 2023 Apr 7;4(1): 10; Tyumentseva et al., Int J Mol Sci. 2023 Nov 8;24(22): 16077; and references cited therein.

[0058] In some embodiments, the TREAD-dimmer constructs comprise a transgene encodes a toxin or suicide protein; such constructs, can be used in treating a subject who has been diagnosed with brain cancer. The methods can include systemically administering a non-brain permeant chemical inducer to the subject sufficient to result in deletion of the construct outside the brain, and then administering a nontoxic prodrug that is a substrate for the suicide protein, wherein action of the suicide protein on the nontoxic prodrug results in production of a toxic metabolite that induces cell death. These methods w ould limit that production to the brain, thereby reducing systemic toxicity. Examples include wherein the suicide protein is herpes simplex virus thymidine kinase (HSVTK) and the nontoxic prodrug is ganciclovir (GCV); the suicide protein is cytosine deaminase (CD) and the nontoxic prodrug is 5-flourouracil (5-FU); the suicide protein is carboxypeptidase G2 (CPG2) and the nontoxic prodrug is nitrogen mustard (NM) or a derivate thereof such as ZD2767P or CMDA (4-[(2- chloroethyl)(2-mesyloxyethyl)amino]benzoyl-L-glutamicacid); the suicide protein is nitroreductase (NTR) and the nontoxic prodrug is CB1954 or an analog thereof; the suicide protein is Cytochrome P450 (CYP) and the nontoxic prodrug is an oxazaphosphorine drug such as cyclophosphamide (CPA) and ifosfomide (IFO); the suicide protein is purine nucleoside phosphorylase (PNP) and the nontoxic prodrug is 6-Methylpurine Deoxyriboside or an analog thereof, e.g., fludarabine phosphate (F- araAMP) or 2-fluoro-2-deoxyadenosine (F-dAdo); the suicide gene is horseradish peroxidase (HRP) and the nontoxic prodrug is indole-3-acetic acid (HRP / IAA); or the suicide protein is carboxylesterase (CE) and the nontoxic prodrug is irinotecan. See, e.g., Karjoo et al., Adv Drug Deliv Rev. 2016 Apr 1; 99(Pt A): 113-128.

[0059] Chemical Inducers

[0060] Drugs that can be used to modulate expression of the modified genes comprising TRSTOP sequences described herein include drugs identified as translational readthrough-inducing drugs (TRIDs) (Li et al., Biomolecules. 2023 Jun 14;13(6):988). The compounds identified in Li et al. were divided into two major categories: aminoglycosides and non-aminoglycosides. Exemplified herein are gentamicin, G418, and PTC124 (ataluren); others include NB30, NB54, NB74. NB84, TC007, ELX-02, negamycin, tylosin. Erythromycin, Azithromycin, Escin, Amlexanox, CC-90009, CC-885, SJ6986, Clitocine, H7, 2,6-DAP, SRI-37240, SRI- 41315, 2-Guanidino-Quinazoline, 5-Fluorouridine, RTC13, GJ071, GJ072, RTC204, RTC219, NV2899. NV2909. NV2913, and NV2907.

[0061] PTC 124 is a preferred compound. It has received conditional approval for use in Duchenne muscular dystrophy from the EMA. The approved regimen used oral administration, 3 times daily (6 / 6 / 12 hrs) of a dose of 10 / 10 / 20 mg / kg (40 mg / kg in total). A dose of 80 mg / kg / day (20 / 20 / 40) was tested in DMD patients for 58 weeks, and no serious adverse effects were seen. A single oral dose of 200 mg / kg in healthy volunteers results in transient, low-grade symptoms, but no serious adverse reactions. There was no evidence of affecting normal termination codons even at high doses. Preclinical data (in rats and dogs) showed good tolerability for oral doses through 1500 mg / kg / day for 28 days (Hirawat et al 2007; Welch et al 2007; Du et al 2007).

[0062] Dosage, toxicity and therapeutic efficacy of the compounds can be determined by standard pharmaceutical procedures in cell cultures or experimental animals, e.g., for determining the LD50 (the dose lethal to 50% of the population) and the ED50 (the dose therapeutically effective in 50% of the population). The dose ratio between toxic and therapeutic effects is the therapeutic index and it can be expressed as the ratio LD50 / ED50. Compounds that exhibit high therapeutic indices are preferred. While compounds that exhibit toxic side effects may be used, care should be taken to design a delivery system that targets such compounds to the site of affected tissue in order to minimize potential damage to uninfected cells and, thereby, reduce side effects.

[0063] The data obtained from cell culture assays and animal studies can be used in formulating a range of dosage for use in humans. The dosage of such compounds lies preferably within a range of circulating concentrations that include the ED50 with little or no toxicity. The dosage may vary within this range depending upon the dosage form employed and the route of administration utilized. For any compound used in the method of the invention, the therapeutically effective dose can be estimated initially from cell culture assays. A dose may be formulated in animal models to achieve a circulating plasma concentration range that includes the IC50 (i.e., the concentration of the test compound which achieves a half-maximal inhibition of symptoms) as determined in cell culture. Such information can be used to more accurately determine useful doses in humans. Levels in plasma may be measured, for example, by high performance liquid chromatography.

[0064] In some embodiments, in human subjects, a target dose of 25-100 pg / mL plasma, or about 50 pg / mL plasma, is desired. This can be achieved, e.g., by administering 100 mg / kg twice a day, or 30-50 mg / kg three times a day (Hirawat et al 2007). In some embodiments, the methods include administration for up to two weeks.

[0065] Methods and Compositions for Controlling Transgene Expression

[0066] The methods and compositions provided herein can be used to control expression of transgenes in cells (e.g., isolated cells in vitro) and animals (e.g., transgenic non-human animals, or in human subjects).

[0067] Thus, provided herein are the TREAD and TREAD-dimmer cassettes described herein, as well as expression constructs comprising the TREAD and TREAD-dimmer cassettes described herein, with promoters and / or other regulatory sequences to drive expression of the transgenes.

[0068] Approaches include insertion of the TREAD and TREAD-dimmer cassettes in expression constructs such as viral vectors or recombinant bacterial or eukaryotic plasmids. Viral vectors transfect cells directly; plasmid DNA can be delivered naked or with the help of, for example, cationic liposomes (lipofectamine) or derivatized (e.g., antibody conjugated), polylysine conjugates, gramicidin S, artificial viral envelopes or other such intracellular carriers, as well as electroporation or direct injection of the gene construct or CaPO4 precipitation carried out in vivo.

[0069] A preferred approach for in vivo introduction of nucleic acid into a cell is by use of a viral vector, including recombinant retroviruses, adenovirus, adeno- associated virus, lentivirus, and herpes simplex virus-1, containing nucleic acid, e.g., a cDNA. encoding the TREAD or TREAD-dimmer cassette. Infection of cells with a viral vector has the advantage that a large proportion of the targeted cells can receive the nucleic acid. Additionally, molecules encoded within the viral vector, e.g., by a cDNA contained in the viral vector, are expressed efficiently in cells that have taken up viral vector nucleic acid.

[0070] Retrovirus vectors and adeno-associated virus vectors can be used as a recombinant gene deliver}7system for the transfer of exogenous genes in vivo, particularly into humans. These vectors provide efficient deliver}' of genes into cells, and the transferred nucleic acids are stably integrated into the chromosomal DNA of the host. The development of specialized cell lines (termed '‘packaging cells”) which produce only replication-defective retroviruses has increased the utility of retroviruses for gene therapy, and defective retroviruses are characterized for use in gene transfer for gene therapy purposes (for a review see Miller, Blood 76:271 (1990)). A replication defective retrovirus can be packaged into virions, which can be used to infect a target cell through the use of a helper virus by standard techniques. Protocols for producing recombinant retroviruses and for infecting cells in vitro or in vivo with such viruses can be found in Ausubel, et al., eds., Current Protocols in Molecular Biology, Greene Publishing Associates, (1989), Sections 9.10-9.14. and other standard laboratory manuals. Examples of suitable retroviruses include pLJ, pZIP, pWE and pEM w hich are known to those skilled in the art. Examples of suitable packaging virus lines for preparing both ecotropic and amphotropic retroviral systems include 'PCrip, 'PCre, 2 and f Am. Retroviruses have been used to introduce a variety of genes into many different cell types, including epithelial cells, in vitro and / or in vivo (see for example Eglitis, et al. (1985) Science 230: 1395-1398; Danos and Mulligan (1988) Proc. Natl. Acad. Sci. USA 85:6460-6464; Wilson et al. (1988) Proc. Natl. Acad. Sci. USA 85:3014-3018; Armentano et al. (1990) Proc. Natl. Acad. Sci. USA 87:6141-6145; Huber et al. (1991) Proc. Natl. Acad. Sci. USA 88:8039-8043; Ferry et al. (1991) Proc. Natl. Acad. Sci. USA 88:8377-8381; Chowdhury et al. (1991) Science 254: 1802-1805; van Beusechem et al. (1992) Proc. Natl. Acad. Sci. USA 89:7640-7644; Kay et al. (1992) Human Gene Therapy 3:641-647; Dai et al. (1992) Proc. Natl. Acad. Sci. USA 89: 10892-10895; Hwu et al. (1993) J. Immunol. 150:4104-4115; U.S. Patent No. 4,868,116; U.S. Patent No. 4,980,286; PCT Application WO 89 / 07136; PCT Application WO 89 / 02468; PCT Application WO 89 / 05345; and PCT Application WO 92 / 07573).

[0071] Another viral gene deliver}' system useful in the present methods utilizes adenovirus-derived vectors. The genome of an adenovirus can be manipulated, such that it encodes and expresses a gene product of interest but is inactivated in terms of its ability to replicate in a normal lytic viral life cycle. See, for example, Berkner et al., BioTechniques 6:616 (1988); Rosenfeld et al., Science 252:431-434 (1991); and Rosenfeld et al., Cell 68:143-155 (1992). Suitable adenoviral vectors derived from the adenovirus strain Ad type 5 dl324 or other strains of adenovirus (e.g., Ad2, Ad3, or Ad7 etc.) are known to those skilled in the art. Recombinant adenoviruses can be advantageous in certain circumstances, in that they are not capable of infecting nondividing cells and can be used to infect a wide variety of cell types, including epithelial cells (Rosenfeld et al., (1992) supra). Furthermore, the virus particle is relatively stable and amenable to purification and concentration, and as above, can be modified so as to affect the spectrum of infectivity. Additionally, introduced adenoviral DNA (and foreign DNA contained therein) is not integrated into the genome of a host cell but remains episomal, thereby avoiding potential problems that can occur as a result of insertional mutagenesis in situ, where introduced DNA becomes integrated into the host genome (e.g., retroviral DNA). Moreover, the carrying capacity of the adenoviral genome for foreign DNA is large (up to 8 kilobases) relative to other gene delivery vectors (Berkner et al., supra; Haj-Ahmand and Graham, J. Virol. 57:267 (1986).

[0072] A preferred viral vector system useful for delivery of nucleic acids in the present methods is the adeno-associated virus (AAV). AAV is a tiny non-enveloped virus having a 25 nm capsid. No disease is known or has been shown to be associated with the wild type virus. AAV has a single-stranded DNA (ssDNA) genome. AAV has been shown to exhibit long-term episomal transgene expression, and AAV has demonstrated excellent transgene expression in the brain, particularly in neurons. Vectors containing as little as 300 base pairs of AAV can be packaged and can integrate. Space for exogenous DNA is limited to about 4.7 kb. An AAV vector such as that described in Tratschin et al., Mol. Cell. Biol. 5:3251-3260 (1985) can be used to introduce DNA into cells. A variety of nucleic acids have been introduced into different cell types using AAV vectors (see for example Hermonat et al., Proc. Natl. Acad. Sci. USA 81:6466-6470 (1984); Tratschin et al., Mol. Cell. Biol. 4:2072-2081 (1985); Wondisford et al., Mol. Endocrinol. 2:32-39 (1988); Tratschin et al., J. Virol. 51:611-619 (1984); and Flotte et al., J. Biol. Chem. 268:3781-3790 (1993). There are numerous alternative AAV variants (over 100 have been cloned), and AAV variants have been identified based on desirable characteristics. In some embodiments, the AAV is AAV1, AAV2, AAV4, AAV5, AAV6, AV6.2, AAV7, AAV8. AAV9. rh.10, rh.39, rh.43 or CSp3. For targeting the CNS, preferred AAV for use in the present methods and compositions include AAV.CPP.16 (described in WO 2020 / 014471), and AAV.CPP.21 (described in WO 2020 / 014471). Other AAV as known in the art (e.g., AAV1, 2, 3, 4, 5. 6, 7, 8 and variants thereof and others as known in the art or described herein) can also be used. AAV.CPP.16 can also be used for targeting CNS, liver, heart / muscle, and / or lung.

[0073] Regulatory Sequences / Promoters

[0074] The constructs can also include one or more sequences that promote expression of a transgene, e.g., one or more promoter sequences; enhancer sequences, e.g., 5’ untranslated region (UTR) or a 3’ UTR; a polyadenylation site; and / or insulator sequences. In some embodiments, the promoter is a brain tissue specific promoter, e.g., a neuron-specific or glia-specific promoter. In certain embodiments, the promoter is a promoter of a gene selected to from: neuronal nuclei (NeuN), glial fibrillary acidic protein (GFAP), MeCP2, adenomatous polyposis coli (APC), ionized calcium-binding adapter molecule 1 (Iba-1), synapsin I (SYN), calcium / calmodulin- dependent protein kinase II, tubulin alpha I, neuron-specific enolase and platelet- derived growth factor beta chain. In some embodiments, the promoter is a pan-cell type promoter, e.g., cytomegalovirus (CMV), beta glucuronidase, (GUSB), ubiquitin C (UBC), or rous sarcoma virus (RSV) promoter. GRP 78 or HMGB2 promoters can also be used.

[0075] In some embodiments, the promoter drives expression in neuronal cells or glial cells. For example, a promoter that drives expression in glial cells can be a GFAP promoter, gfaABCID promoter, gfa2 promoter, ALDH1L1 promoter, SLC1A3 promoter, Gjb6 promoter, Mbp promoter, MAG promoter, CBh promoter, F4 / 80 promoter, CD68 promoter, or CD1 IB promoter. As another example, a promoter that drives expression in neuronal cells can be a neuronal-specific enolase (NSE) promoter, Synapsin promoter, calcium / calmodulin-dependent protein kinase II promoter, tubulin alpha 1 promoter, platelet-derived growth factor beta chain promoter, parvalbumin promoter, GAD67 promoter, or CCK promoter.

[0076] In some embodiments, the promoter is a ubiquitous promoter, optionally major immediate early human cytomegalovirus promoter (MIEhCMV), Chicken |3-Actin Promoter (CBA); Human Cytomegalovirus Immediate / Early Gene Promoter and Enhancer (CMV); Chicken P-Actin / Cytomegalo virus Hybrid Promoter (CAG); Rous Sarcoma Virus Long Terminal Repeat Promoter (RSV); SV40 promoter; EFlalpha promoter.

[0077] Synthetic promoters are also known, see, e.g, Jiittner et al., Nature Neuroscience volume 22, pages 1345- 1356(2019); Morelli et al., J Gen Virol. 1999 Mar;80 ( Pt 3):571 -583; O’Carroll et al.. Front Mol Neurosci. 2020; 13: 618020 (review).

[0078] The expression constructs can also include an enhancer, such as a CMV Enhancer, mDlx enhancer, or AQP4 enhancer. See, e.g., WO2020168279, Nair et al., iScience. 2020 Mar 27; 23(3): 100888, Gruh et al., J Gene Med. 2008 Jan; 10(1): 21- 32, Abe et al., FEBS Lett. 2017 Dec; 591 (23): 3906-3915, and Dimidschstein et al., Nat Neurosci.. 2016 Dec; 19(12): 1743-1949.

[0079] The expression constructs can also include a Woodchuck Hepatitis Virus Posttranscriptional Regulatory Element (WPRE) or alternatives, e.g., as described in PCT / EP2014 / 072852.

[0080] Preferably, the expression constructs includes a polyA signal sequence at the 3 ’ end of the coding sequence for the transgene. Exemplary' polyA signal sequence include human growth hormone (hGH), SV40, bovine growth hormone (bGH). or beta-globin, e.g.. rabbit beta-globin (rbGlob).

[0081] Delivery

[0082] The constructs can be administered in any effective carrier, e.g., any formulation or composition capable of effectively delivering the component gene to cells in vitro or in vivo. In clinical settings, the constructs can be introduced into a subject by any of a number of methods, each of which is familiar in the art. For instance, a pharmaceutical preparation of the constructs can be introduced systemically, e.g., by intravenous injection, and specific transduction of the protein in the target cells will occur predominantly from specificity' of transfection, provided by a gene delivery vehicle, cell-type or tissue-type expression due to transcriptional regulatory sequences controlling expression of the receptor gene, or a combination thereof. In some embodiments, initial delivery of the recombinant gene is more limited, with introduction into the subject being localized. For example, the gene delivery vehicle can be introduced by catheter (see U.S. Patent 5,328,470) or by stereotactic injection (e.g., Chen et al.. PNAS USA 91: 3054-3057 (1994)).

[0083] A pharmaceutical preparation comprising a construct or cassette as described herein can consist essentially of the construct or cassette in an acceptable diluent, or can comprise a slow release matrix in which the gene delivery vehicle is embedded. Alternatively, where the construct or cassette can be produced intact from recombinant cells, e.g., retroviral vectors, provided herein are the one or more cells that produce the construct or cassette.

[0084] EXAMPLES

[0085] The invention is further described in the following examples, which do not limit the scope of the invention described in the claims.

[0086] Materials and Methods

[0087] The following materials and methods were used in the examples below.

[0088] Plasmids

[0089] DNA plasmids used for testing the “TREAD” gene switch in vitro were modified from a previously published AAV vector backbone by replacing all DNA elements between the two ITR (i.e., inverted terminal repeat) sequences. For plasmids pGFP-TRSTOP.TGG-Cre, pGFP-TRSTOP.TGA-Cre, pGFP-TRSTOP.TAG-Cre, and pGFP-TRSTOP.TAA-Cre, the stop codons for the green fluorescent protein (GFP) cDNAs were replaced with the 9-nucleotide TRSTOP sequences TGGCTAGAC. TGACTAGAC, TAGCTAGAC and TAACTAGAC, respectively. Each TRSTOP sequence was followed by the coding sequences for a 9-glycine linker, a Cre recombinase, and a 3x TAA translation termination sequence (i.e., TAATAATAA). The entire open reading frame for each GFP-TRSTOP-Cre ‘‘fusion protein"’ was regulated by a CAG ubiquitous promoter and the Kozak sequence upstream and a SV40 poly-adenylation (pA) signal sequence downstream. To measure the level of Cre expression, a second plasmid pDIO-H2B-RFP was co-transfected to visualize Cre expression by expressing Cre-dependent nuclear-bound RFP (DIO: Double-foxed Inverted Opening reading frame: H2B: a nuclear localization signal).

[0090] The “all-in-one” CRISPR-Cas9 plasmid pAAV-Cas9-gRNA was modified from a plasmid purchased from Addgene (AAV:ITR-EFlaCore-SaCas9-dual-U6- sgRNA-backbone-ITR; Addgene #207878) by inserting two guide RNA sequences (GCCTCTAGAGTCGCAGATCCTC (SEQ ID NO: 15) and TACGAAGTTATATTAAGGGTT (SEQ ID NO: 16) into the two gRNA scaffolds using two cloning sites (BbsI and Aarl). Both gRNA sequences were designed to target the loxPSTOPloxP (“LSL”) sequence in the plasmid pLSL-GFP (Addgene #51269) and thus direct Cas9 to cut out the LSL sequence that otherwise blocks the expression of GFP. The plasmid aAAV-TRSTOP-Cas9-gRNA (i.e. “switchable” Cas9) was generated from pAAV-Cas9-gRNA by inserting the TRSTOP sequence TGACTAGAC into the Ser431 / Gln432site of the SaCas9 protein (1052 amino acids in total).

[0091] For testing of “TREAD-Dimmer” technology, the plasmid pGFP- TRSTOP.TGA-Cre was used to generate the plasmid pCAG-loxP-GFP-TRSTOP-Cre- loxP. The two 34-bp loxP sequences were oriented in the same direction to achieve excision of the sequence between them after Cre recombination. pCAG-loxP-GFP- TRSTOP-Cre-loxP was mutated to generate pGFAP-loxP-GFP-TRSTOP-Cre-loxP byreplacing the CAG promoter with the GFAP promoter.

[0092] Transgenic Animals

[0093] Ail4 mice, which are homozygous for Rosa-CAG-LSL-tdTomato-WPRE (Madisen et al., Nat Neurosci. 2010 Jan;13(l): 133-40) were obtained from Jackson Labs.

[0094] Chemical drugs

[0095] Gentamicin (# SC-203334) and G418 (# SC-29065) were purchased from Santa Cruz Biotech, dissolved in saline to obtain 100 mg / rnL and 50 mg / rnL stock solutions respectively. Both solutions were sterilized using 0.2 pm cell culture-grade filters, aliquoted into 1.5 ml tubes for storage at -20 °C. PTC124 (# S5059L >99%) was purchased from www.shyuanye.com. For in vitro cell culture experiment, a stock solution of 5 mg / ml PTC124 was made in DMSO. For mouse administration. PTC124 was first dissolved in DMSO and then in com oil (9x in volume) to reach a final working concentration of 10 mg / ml.

[0096] DNA transfection in vitro

[0097] Chemical transfection was performed in HEK293 cells 24 hours after being seeded in a 24-well cell culture plate with a density of 300,000 cells per well. For each well, 0.05 or 0.1 pg of plasmid, diluted in 0.1 ml DMEM solution, was mixed well with the polyethylenimine (PEI) solution at a DNA / PEI ratio of 1 :3 in mass. The DNA / PEI solution was left undisturbed for 15 min to allow formation of DNA / PEI nanoparticles and was then added into the culture medium, which was changed 6-8 hours later with fresh medium to remove excessive PEI. For co-transfection of the plasmids pGFP-TRSTOP-Cre with pDIO-H2B-RFP, treated cells were fixed 72 hours after transfection using cold 4% paraformaldehyde (PFA) for 30 min. Drug treatment was performed by adding pre-determined amount of drug into the fresh culture medium immediately after transfection. All fixed cells were stained with DAPI (0.05%; 30 min), a fluorescent DNA stain that labels cell nuclei.

[0098] For testing CRISPR gene editing in vitro, the plasmid pAAV-Cas9-gRNA, or pAAV-TRSTOP-Cas9-gRNA. was co-transfected with the plasmid pLSL-GFP in HEK293 cells. Treated cells were fixed with 4% PFA 72 hours after transfection. Expression of GFP would suggest deletion of the “LSL” sequence in the plasmid pLSL-GFP, indicating the success of gene editing by the CRISPR-Cas9 machine delivered by the plasmid pAAV-Cas9-gRNA or pAAV-TRSTOP-Cas9-gRNA.

[0099] AAV production

[0100] Recombinant AAVs were packaged using standard three-plasmid cotransfection protocol. Briefly, plasmids pRC, pHelper, and pAAV carry ing the transgene expression cassette were co-transfected into HEK 293T cells using polyethylenimine. rAAVs vectors were collected from the serum-free medium and the cells. AAV particles in the medium were concentrated using a PEG-precipitation method with 8% PEG-8000 (wt / vol). Cell pellets containing viral particles were resuspended and lysed through sonication. Combined viral vectors from PEG- precipitation and cell lysates were treated with DNase and RNase at 37 °C for 30 mins and then purified by iodixanol gradient (15%, 25%, 40% and 60%) with ultracentrifugation (VTi 50 rotor, 48,000 r.p.m, 18°C, Ih). rAAVs were then concentrated using Millipore Amicon filter unit (UFC910008, 100K MWCO) and formulated in Dulbecco's phosphate buffered saline (PBS) containing 0.001% Pluronic F-68 (Gibco).

[0101] CRISPR gene editing in vivo - Mouse study

[0102] Adult mice (over 6 weeks of age) were anesthetized with 2.5% isoflurane. rAAV vectors diluted in 200 pL sterile saline were administered through tail vein using sharp 31-G insulin syringes. Three weeks later, animals were transcardially perfused with cold phosphate buffered saline (PBS) followed by 4% paraformaldehyde (PF A). Tissues were post-fixed in 4% PFA overnight, and then immersed in 30% sucrose solutions for two days prior to embedding and snapfreezing in OCT. Typically, 80 pm thick brain sections were cut for imaging of native fluorescence, 40 pm thick brain sections for immunohistochemistry'.

[0103] PTC 124 administration was initiated one week after AAV injection. 10 mg / ml PTC 124 in DMSO / com oil was administered twice daily for two weeks by gav aging with a daily dose of 300 mg / kg per mouse for most experiments. For testing the effect of low-dose PTC 124, a daily dose of 60 mg / kg per mouse was used. In addition to gavaging, 50 mg / kg PTC 124 was given intraperitoneally in each animal every' two days to help maintain plasma concentration. On average, 325 mg / kg / day (high dose) or 85 mg / kg / day (low-dose) PTC124 was administered in each mouse for a total treatment period of two weeks.

[0104] Image and data analysis

[0105] Fluorescent images of fixed cells were examined and captured using an inverted Nikon fluorescence microscope equipped with a digital camera. All images were analyzed using the publicly available ImageJ software. Data analysis was performed using Microsoft Excel and GraphPad Prism. Student’s / -test and one-way ANOVA with Dunnett’s post-test were used for data comparison. P values less than 0.05 were considered statistically significant.

[0106] Example 1. Verifying the phenomenon of “termination readthrough”.

[0107] Since termination readthrough in protein translation is a relatively low- frequency phenomenon (Keeling et al., 2014), a reliable and sensitive assay was developed to study it. An expression cassette was designed consisting of a coding sequence for GFP, a stop-codon-containing effector sequence TRSTOP and a coding sequence for RFP in one open reading frame. A triple-stop-codon sequence (i.e., 3xTAA) was used downstream of the RFP sequence to signal “hard stop’7of protein translation. If no termination readthrough occurred for TRSTOP, protein translation would be terminated at TRSTOP and only GFP would be expressed; otherwise, protein translation would go beyond TRSTOP, and a GFP-RFP fusion protein would be expressed. Thus, at least in theory’, no detection of RFP signal in all GFP-labeled cells would indicate no termination readthrough; a higher percentage of RFP-positive cells out of GFP-labeled cells would suggest a higher readthrough efficiency.

[0108] To engineer TRSTOP, a 9-nucleotide sequence UGACUAGAC in the RNA genome of the Sindbis virus was evaluated, where the sequence of the first stop codon was varied. Readthrough of the stop codon UGA is utilized by the virus for the synthesis of an essential RNA-dependent RNA polymerase (Li & Rice, 1993). The 6 nucleotides downstream of UGA were included because of published studies suggesting potential impact of flanking sequence on readthrough efficiency of the stop codon (Cassan & Rousset, 2001; Schilff et al., 2021). In addition to plasmid pGFP-TRSTOP.TGA-RFP, three other plasmids were also constructed by replacing TGA with the stop codon TAG, stop codon TAA, and a regular codon TGG that encodes for try ptophan (pGFP-TRSTOP.TAG-RFP, pGFP-TRSTOP.TAA-RFP, and pGFP-TRSTOP.TGG-RFP respectively).

[0109] As expected, when HEK293 cells were transfected with plasmid pGFP- TRSTOP.TGG-RFP (0.1 pg per 300,000 cells), all GFP-positive cells were also RFP- positive, indicating expression of GFP -Try -Leu- Asp-RFP fusion protein. When cells were transfected with the other three plasmids with the stop codons TGA, TAG, and TAA, 1 1.4 ± 0.7% (Mean ± SEM) of GFP-positive cells were also RFP-positive for the TGA plasmid and few RFP-labeled cells were detected for either the TAG plasmid (0.4 ± 0.1%, mean ± SEM) or the TAA plasmid (0%). Such observation on “leakiness” of the stop codon TGA in the TRSTOP sequence was in consistence with previous report (Li & Rice, 1993). In addition, such results suggested that the stop codons TAG and TAA were less leaky, as compared with TGA.

[0110] One caveat in using RFP as a reporter gene was that the fluorescence- microscope-based detection method might not be sensitive enough to monitor subthreshold RFP expression. Thus, RFP was replaced with a more sensitive reporter system using Cre recombinase and co-transfected plasmid pDIO-H2B-RFP to express Cre-dependent nuclear-bound RFP (DIO: Double-foxed Inverted Open reading frame; H2B: a nuclear localization signal). When readthrough occurs, Cre is expressed, triggering removal of the DIO sequence and expression of the RFP (FIG. 1 A). We used this more sensitive system to re-examine the leakiness of all three stop codons. All four Cre plasmids (0.1 jag per 300,000 cells) were co-transfected with a second plasmid pDIO-H2B-RFP (0.5 pg per 300,000 cells) that would express nucleus-bound RFP in a Cre-dependent manner (FIG. 1 A). Once again, baseline readthrough of the stop codon TGA was observed (FIG. IB). Interestingly, effects of readthrough with comparable efficiencies were also observed for the stop codons TAG and TAA (FIG. 1C).

[0111] Together, these data suggest that while baseline readthrough efficiency for the stop codon TGA in TRSTOP may be higher than the other two stop codons, all three stop codons are subject to termination readthrough.

[0112] Example 2. Modulation of readthrough efficiency by gentamicin and G418

[0113] Next, we identified small molecule compounds that could modulate the readthrough efficiencies of TRSTOP. The Cre system described above was chosen as the reporter gene given its apparently higher sensitivity than RFP. Several compounds were tested including gentamicin, which is a clinically approved antibiotic drug, and G418, which is a widely reported readthrough-inducing experimental compound (Bukowy-Bieryllo et al., 2016; Li et al., 2023; Omachi et al., 2022), as well as PTC124 (ataluren; Li et al., 2023). Despite their publication in other applications, neither compound was characterized for their effects on the TRSTOP effectors designed in this study.

[0114] We found that gentamicin with concentrations ranging from 100 to 2,000 jag / ml increased readthrough efficiency of TGA in a dose-dependent manner (FIG. 2A, left panel). A maximum percentage of RFP+out of GFP+cells reached 22.8 ± 0.6% with the treatment of 2,000 pg / ml gentamicin, as compared with the percentage ratio without gentamicin (5.9 ± 0.5%, mean ± SEM; P < 0.0001). Gentamicin also increased readthrough efficiency of TAG in TRSTOP albeit only at the highest concentration tested (10.2 ± 0.2% with 2,000 pg / ml gentamicin vs. 5.4 ± 0.4% without gentamicin, mean ± SEM; P < 0.0001; FIG. 2A, middle panel). For TAA in TRSTOP, the maximal readthrough efficiency was observed with 100 pg / ml gentamicin (FIG. 2A, right panel). Gentamicin of higher concentrations tested appears to have diminished effects on TAA, alluding to potential dose-limiting side effect of gentamicin.

[0115] Testing with gentamicin indicates that the stop codon TGA is most amenable to drug modulation and thus could be nominated for constructing a gene switch controlled by a chemical inducer. To confirm that the response of TGA is not drugspecific, PTC124 (FIG. 2B) and G418 (FIG. 2C) were also tested, and enhancement of readthrough was observed.

[0116] Example 3. Design and testing of a two-component mini gene switch: “TREAD”

[0117] Next, we constructed a gene switch, referred to herein as ' TREAD (or Termination Readthrough based Expression Adjusting Device), that included two components: the 9-nucleotide TRSTOP sequence TGACTAGAC as a DNA “effector,” which is inserted into the coding region of a target gene at a place that would keep expression of the target gene at a minimal baseline level, and a chemical “inducer” that would increase target gene expression when applied (FIG. 3A, top).

[0118] To test TREAD, an AAV -based construct with a TRSTOP inserted into SaCas9 (pAAV-TRSTOP-Cre) was designed (FIG. 3 A, bottom). The TRSTOP sequence was inserted into the Cas9 expression cassette at the Ser431 / Gln432 site of the Cas9 protein. The gRNA in the TREAD construct targeted the LSL cassette to specifically cut it out and allow expression of the RFP reporter protein. When the TRSTOP-Cas9 AAV was co-transfected with a RFP reporter construct comprising a RFP interrupted by a loxP-stop (33 polyA signal)-loxP cassette (LSL) in HEK293. expression of functional Cas9 (and thus of the RFP report protein) is determined by the amount of readthrough of the TRSTOP. As shown in FIG. 3B, addition of gentamicin increased expression of the functional Cas9.

[0119] The ability of this system to control expression of a target gene in vivo was also evaluated; a schematic of these experiments is shown in FIG. 3C. In these experiments, AAV2 or AAV.CPP. 16 vectors were administered to Ail4 mice expressing an RFP reporter construct comprising a RFP interrupted by a LSL sequence. The vectors included a TREAD cassette with a TRSTOPped-Cas9 and gRNAs targeting LSL (FIG. 3C). As shown in FIG. 3D. the TRSTOPped-Cas9

[0120] (second row) showed significantly reduced expression as compared to the ‘switchless’ Cas9 (top row), and expression of the TRSTOPped-Cas9 increased in liver in a dose- dependent manner in the presence of increasing doses of PTC124. FIGs. 3E-H show that other AAV including AAV.CPP. 16 provided similar results, with PTC124 increasing expression in liver, muscle, and heart.

[0121] These results suggest that TREAD can be used as a switch in “turning on” gene expression both in vitro and in vivo.

[0122] Example 4. “TREAD-Dimmer” for “self-regulated” gene therapy

[0123] Expression of a self-regulating TREAD construct (“TREAD-dimmer”), which uses a negative feedback loop to reduce transgene dosing, was evaluated in HEK293 cells using conventional (FIG. 5A. left) and TREAD-dimmer (FIG. 5A right) plasmids. The cells were transfected with 2 pg plasmid DNA per well in a 24 well plate and incubated for 48 hours before imaging. Fluorescent images of GFP expression in cells transfected with the conventional (FIG. 5B, left) and TREAD- dimmer (FIG. 5B, right) plasmids. The conventional plasmid provided variable expression across the cells, with some showing low expression and some showing high expression, while more uniform expression was seen in the cells transfected with the TREAD-Dimmer plasmid.

[0124] Similar results were seen in vivo,' a TREAD-dimmer construct (AAV.CPP.16- GFAP-loxP-GFP-TRSTOP-Cre-loxP, FIG. 6A) was delivered intravenously into mice at a dose of 5E1 Ivg per mouse. Imaging with staining for GFP showed low but more uniform expression (FIG. 6B), in contrast with experiments in which a conventional, “non-regulated” AAV.CPP.16 construct was delivered, showing areas of overexpression (FIG. 6C).

[0125] References

[0126] Bukowy-Bieryllo, Z., Dabrowski, M., Witt, M., & Zietkiewicz, E. (2016). Aminoglycoside-stimulated readthrough of premature termination codons in selected genes involved in primary ciliary dyskinesia. RNA Biol, 13(10), 1041-1050. doi: 10. 1080 / 15476286.2016. 1219832

[0127] Cassan, M., & Rousset, J. P. (2001). UAG readthrough in mammalian cells: effect of upstream and dow nstream stop codon contexts reveal different signals. BMC Mol Biol, 2, 3. doi: 10. 1186 / 1471-2199-2-3 Dunbar, C. E., High, K. A., Joung, J. K., Kohn, D. B., Ozawa, K., & Sadelain, M. (2018). Gene therapy comes of age. Science. 359(6372). doi: 10. 1126 / science.aan4672

[0128] Fussenegger, M., Morris, R. P., Fux, C., Rimann, M., von Stockar, B., Thompson, C. J., & Bailey, J. E. (2000). Streptogramin-based gene regulation systems for mammalian cells. Nat Biotechnol, 18(11), 1203-1208. doi: 10. 1038 / 81208

[0129] Gitzinger, M., Kemmer, C., El-Baba, M. D.. Weber. W., & Fussenegger, M. (2009). Controlling transgene expression in subcutaneous implants using a skin lotion containing the apple metabolite phloretin. Proc Natl Acad Sci U S A, 106(26), 10638- 10643. doi : 10. 1073 / pnas.0901501106

[0130] Gitzinger, M., Kemmer, C., Fluri, D. A.. El-Baba, M. D., Weber, W., & Fussenegger, M. (2012). The food additive vanillic acid controls transgene expression in mammalian cells and mice. Nucleic Acids Res, 40(5), e37. doi: 10. 1093 / nar / gkrl251

[0131] Gossen, M., & Bujard, H. (1992). Tight control of gene expression in mammalian cells by tetracycline-responsive promoters. Proc Natl Acad Sci U S A, 89(12), 5547-5551. doi: 10.1073 / pnas.89.12.5547

[0132] Kaminski, R., Bella, R., Yin, C., Otte, J., Ferrante, P., Gendelman, H. E., . . . Khalili. K. (2016). Excision of HIV- 1 DNA by gene editing: a proof-of-concept in vivo study. Gene Ther. 23(8-9), 690-695. doi: 10.1038 / gt.2016.41

[0133] Keeling, K. M., Xue, X., Gunn, G, & Bedwell, D. M. (2014). Therapeutics based on stop codon readthrough. Annu Rev Genomics Hum Genet, 15, 371-394. doi: 10. 1146 / annurev-genom-091212-153527

[0134] Li, G., & Rice, C. M. (1993). The signal for translational readthrough of a UGA codon in Sindbis virus RNA involves a single cytidine residue immediately downstream of the termination codon. J Virol, 67(8), 5062-5067. doi: 10. 1128 / JVI.67.8.5062-5067. 1993

[0135] Li, S„ Li. J., Shi, W.. Nie, Z„ Zhang, S„ Ma, F„ . . . Xie, X. (2023). Pharmaceuticals Promoting Premature Termination Codon Readthrough: Progress in Development. Biomolecules, 13(6). doi: 10.3390 / bioml 3060988

[0136] Madderson, O., Teixeira, A. P., & Fussenegger, M. (2021). Emerging mammalian gene switches for controlling implantable cell therapies. Curr Opin Chem Biol, 64, 98-105. doi: 10.1016 / j.cbpa.2021.05.012 Madigan, V., Zhang, F., & Dahlman, J. E. (2023). Drug delivery' systems for CRISPR-based genome editors. Nat Rev Drug Discov, 22(11), 875-894. doi : 10. 1038 / s41573-023-00762-x

[0137] Maeder, M. L., Stefanidakis, M., Wilson, C. J., Baral, R., Barrera, L. A., Bounoutas, G. S., . . . Jiang, H. (2019). Development of a gene-editing approach to restore vision loss in Leber congenital amaurosis type 10. Nat Med. 25(2), 229-233. doi: 10. 1038 / S41591-018-0327-9

[0138] Omachi, K., Kai, H., Roberge, M., & Miner, J. H. (2022). NanoLuc reporters identify COL4A5 nonsense mutations susceptible to drug-induced stop codon readthrough. iScience, 25(3), 103891. doi:10.1016 / j.isci.2022.103891

[0139] Pelham, H. R. (1978). Leaky UAG termination codon in tobacco mosaic virus RNA. Nature, 272(5652), 469-471. doi: 10.1038 / 272469a0

[0140] Randjelovic, P., Veljkovic, S., Stojiljkovic, N., Sokolovic, D., & Ilic, I. (2017). Gentamicin nephrotoxicity in animals: Current knowledge and future perspectives. Exclij, 16, 388-399. doi: 10.17179 / excli2017-165

[0141] Schilff, M., Sargsyan, Y., Holhuis, J., & Thoms, S. (2021). Stop Codon Context-Specific Induction of Translational Readthrough. Biomolecules, 11(7). doi: 10.3390 / bioml 1071006

[0142] Wang, D., Tai, P. W. L., & Gao, G. (2019). Adeno-associated virus vector as a platform for gene therapy delivery. Nat Rev Drug Discov, 18(5), 358-378. doi : 10. 1038 / s41573-019-0012-9

[0143] Wang, H., Ye, H., Xie, M., Daoud El-Baba, M., & Fussenegger, M. (2015). Cosmetics-triggered percutaneous remote control of transgene expression in mice. Nucleic Acids Res. 43(14), e91. doi: 10. 1093 / nar / gkv326

[0144] Weber, W., Fux, C., Daoud-el Baba, M., Keller, B., Weber, C. C., Kramer, B. P., . . . Fussenegger, M. (2002). Macrolide-based transgene control in mammalian cells and mice. Nat Biotechnol, 20(9), 901-907. doi:10. 1038 / nbt731

[0145] Weiner, A. M., & Weber, K. (1971). Natural read-through at the UGA termination signal of Q-beta coat protein cistron. Nat New Biol, 234(50). 206-209. doi: 10. 1038 / newbio234206a0

[0146] Yoshinaka, Y., Katoh, I., Copeland, T. D., & Oroszlan, S. (1985). Murine leukemia virus protease is encoded by the gag-pol gene and is synthesized through suppression of an amber termination codon. Proc Natl Acad Sci U S A, 82(6), 1618- 1622. doi: 10. 1073 / pnas.82.6. 1618

[0147] Zetsche, B., Volz, S. E., & Zhang, F. (2015). A split-Cas9 architecture for inducible genome editing and transcription modulation. Nat Biotechnol, 33(2), 139- 142. doi: 10.1038 / nbt.3149

[0148] Zhao. C., Zhao, Y.. Zhang. J., Lu, J.. Chen, L., Zhang. Y., . . . Wang, Y. (2018). HIT-Cas9: A CRISPR / Cas9 Genome-Editing Device under Tight and Effective Drug Control. Mol Ther Nucleic Acids, 13, 208-219. doi: 10. 1016 / j.omtn.2018.08.022 OTHER EMBODIMENTS

[0149] It is to be understood that while the invention has been described in conjunction with the detailed description thereof, the foregoing description is intended to illustrate and not limit the scope of the invention, which is defined by the scope of the appended claims. Other aspects, advantages, and modifications are within the scope of the following claims.

Claims

WHAT IS CLAIMED IS:

1. A nucleic acid cassette comprising a coding region for a transgene of interest, wherein the coding region comprises a 9-nucleotide TRSTOP sequence selected from TGACTAGAC. TAGCTAGAC and TAACTAGAC inserted into the coding region of the transgene to minimize expression of the transgene, preferably wherein the TRSTOP sequence is TGACTAGAC.

2. The nucleic acid cassette of claim 1, wherein the TRSTOP sequence is inserted within a 5’ half of the coding region, and wherein expression of the 5' half does not result in production of a functional protein product.

3. The nucleic acid cassette of claim 1, wherein the transgene of interest comprises a CRISPR Cas protein.

4. The nucleic acid cassette of claim 4, wherein the CRISPR Cas protein is SaCas9 or a variant thereof.

5. The nucleic acid cassette of claim 4, wherein the CRISPR Cas protein is a base editor.

6. A nucleic acid cassette comprising, from 5’ to 3’: a first loxP sequence; a coding region for a transgene of interest, wherein the stop codon of the transgene is replaced with a TRSTOP sequence selected from TGACTAGAC, TAGCTAGAC and TAACTAGAC; a coding region for a Cre recombinase, preferably linked to a nuclear localization sequence; and a second loxP element in the same direction as the first loxP sequence.

7. The nucleic acid cassette of claim 6, wherein the transgene of interest comprises a CRISPR Cas protein, optionally SaCas9 or a variant thereof.

8. The nucleic acid cassette of claim 6. wherein the transgene of interest is a suicide protein.

9. The nucleic acid cassete of claim 8, wherein the suicide protein is herpes simplex virus thymidine kinase (HSVTK); cytosine deaminase (CD); carboxypeptidase G2 (CPG2); nitroreductase (NTR); Cytochrome P450 (CYP); purine nucleoside phosphorylase (PNP); horseradish peroxidase (HRP); or carboxylesterase (CE).

10. An expression construct comprising the nucleic acid cassete of any one of claims 1-9 and a promoter to drive expression of the transgene.

11. The expression construct of claim 10, which is a viral vector or a recombinant bacterial or eukaryotic plasmid.

12. The expression construct of claim 11, wherein the transgene of interest comprises a CRISPR Cas protein, and wherein the construct further comprises sequences encoding one or more guide RNAs (gRNAs), optionally further comprising at least one additional promoter to drive expression of the one or more gRNAs.

13. A method of increasing expression of a transgene in a cell, the method comprising: expressing in the cell the nucleic acid cassete of any one of claims 1-5; contacting the cell with a chemical inducer of stop codon readthrough, optionally wherein the chemical inducer is selected from the group consisting of gentamicin, G418, and PTC124, in an amount sufficient to increase readthrough of the TRSTOP. thereby increasing expression of the transgene in the cell.

14. A method of reducing expression of a transgene in a cell, the method comprising: expressing in the cell the nucleic acid cassete of any one of claims 6-9; contacting the cell with a chemical inducer of stop codon readthrough, optionally wherein the chemical inducer is selected from the group consisting of gentamicin, G418, and PTC124, in an amount sufficient to increase readthrough of the TRSTOP, thereby decreasing expression of the transgene in the cell.

15. The method of claim 14, wherein the transgene is a suicide protein, and the method further comprises contacting the cell with a nontoxic prodrug that is a substrate for the suicide protein, wherein action of the suicide protein on the nontoxic prodrug results in production of a toxic metabolite that induces cell death.

16. The method of claim 15, wherein the suicide protein is herpes simplex virus thymidine kinase (HSVTK) and the nontoxic prodrug is ganciclovir (GCV); the suicide protein is cytosine deaminase (CD) and the nontoxic prodrug is 5- flourouracil (5-FU); the suicide protein is carboxypeptidase G2 (CPG2) and the nontoxic prodrug is nitrogen mustard (NM) or a derivate thereof such as ZD2767P or CMDA (4-[(2-chloroethyl)(2-mesyloxyethyl)amino]benzoyl-L-glutamicacid); the suicide protein is nitroreductase (NTR) and the nontoxic prodrug is CB1954 or an analog thereof; the suicide protein is Cytochrome P450 (CYP) and the nontoxic prodrug is an oxazaphosphorine drug such as cyclophosphamide (CPA) and ifosfomide (IFO); the suicide protein is purine nucleoside phosphorylase (PNP) and the nontoxic prodrug is 6-Methylpurine Deoxyriboside or an analog thereof, optionally fludarabine phosphate (F-araAMP) or 2-fluoro-2-deoxyadenosine (F- dAdo); the suicide gene is horseradish peroxidase (HRP) and the nontoxic prodrug is indole-3 -acetic acid (HRP / IAA); or the suicide protein is carboxy lesterase (CE) and the nontoxic prodrug is irinotecan.

17. The method of claim 13 or 14, wherein the cell is in vitro.

18. The method of claim 13 or 14, wherein the cell is in vivo.

19. A cell comprising the expression construct of any of claims 10-12.

20. The cell of claim 19, wherein the cell is a mammalian cell.

21. The cell of claim 19, which is in vitro.

22. The cell of claim 19, which is in vivo.

Images