MODULATED CAS-INHIBITORS

Abstract

The present invention relates to a polynucleotide encoding a fusion polypeptide comprising an anti-CRISPR (Acr) polypeptide, wherein said fusion polypeptide further comprises a receptor domain changing conformation upon reception of a stimulus. The present invention also relates to a vector comprising the polynucleotide of the present invention, to a bipartite Acr polypeptide comprising a first partial Acr polypeptide comprising amino acids corresponding to amino acids 10 to 62 of SEQ ID NO: 1, and a second partial Acr polypeptide comprising amino acids corresponding to amino acids 67 to 77 of SEQ ID NO: 1, and to a host cell comprising the aforesaid polynucleotide compounds. The present invention also relates to the said compounds for use in medicine, in particular for use in treatment and/or prevention of genetic disease, neurodegenerative disease, cancer, and/or infectious disease. Moreover, the present invention also relates to a kit, methods, and uses related thereto.

Claims

1. A polynucleotide encoding a fusion polypeptide comprising an anti-CRISPR (Acr) polypeptide, wherein said fusion polypeptide further comprises a receptor domain changing conformation upon reception of a stimulus.

2. The polynucleotide of claim 1, wherein said stimulus is light, preferably blue light, or wherein said stimulus is a chemical compound, preferably is rapamycin.

3. The polynucleotide of claim 1, wherein said receptor domain is inserted into the Acr at an insertion site corresponding to one of amino acids 62 to 69 of the AcrIIA4 polypeptide (SEQ ID NO:1) and/or is fused to one of the terminal amino acids of the Acr polypeptide.

4. The polynucleotide of claim 1, wherein said receptor domain is selected from a light-oxygen-or-voltage (LOV) domain, a rapamycin-binding domain, a phytochrome (Phy) domain, a cryptochrome (Cry) domain, a steroid receptor domain, and tetracycline binding domain, preferably is a LOV domain.

5. The polynucleotide of claim 1, to wherein said fusion polypeptide comprises an amino acid sequence at least 70% identical to an amino acid sequence selected from SEQ ID NOs: 78 to 114, preferably to an amino acid sequence selected from SEQ ID NOs: 88 to 107.

6. (canceled)

7. A bipartite anti-CRISPR (Acr) polypeptide comprising a first partial Acr polypeptide comprising amino acids corresponding to amino acids 10 to 62 of SEQ ID NO: 1, and a second partial Acr polypeptide comprising amino acids corresponding to amino acids 67 to 77 of SEQ ID NO: 1.

8. The bipartite Acr polypeptide of claim 7, wherein said first and second partial Acr polypeptide are comprised in the same fusion polypeptide; or wherein said first and second partial Acr polypeptide are separately fused to the components of a receptor/ligand pair.

9. A bipartite anti-CRISPR (Acr) polypeptide comprising a first partial Acr polypeptide comprising amino acids corresponding to amino acids 10 to 62 of SEQ ID NO: 1, and a second partial Acr polypeptide comprising amino acids corresponding to amino acids 67 to 77 of SEQ ID NO: 1, wherein said bipartite Acr polypeptide is encoded by a polynucleotide according to claim 1.

10. (canceled)

11. (canceled)

12. (canceled)

13. A method of providing a host cell comprising a stimulus-modulatable activity of a CRISPR-associated (Cas) nuclease comprising a) introducing into said host cell a Cas nuclease; b) introducing into said host cell a fusion polypeptide comprising an Acr polypeptide and a receptor domain according to claim 9; c) thereby, providing a host cell comprising a stimulus-modulatable activity of a Cas nuclease.

14. (canceled)

15. (canceled)

16. A method for treating genetic disease, neurodegenerative disease, cancer, and/or infectious disease in a subject suffering therefrom, said method comprising a) contacting a host cell of said subject with a Cas nuclease and with a fusion polypeptide comprising an anti-CRISPR (Acr) polypeptide and a receptor domain according to claim 7; b) optionally, providing a stimulus causing the receptor domain to change conformation; and c) thereby, treating genetic disease, neurodegenerative disease, cancer, and/or infectious disease.

17. The method of claim 16, wherein said method comprises contacting at least a fraction of cells of said subject with said stimulus causing the receptor domain to change conformation.

18. The method of claim 16, wherein said method further comprises contacting said host cell with at least one gRNA.

19. The method of claim 18, wherein contacting a host cell with a gRNA is contacting said host cell with a polynucleotide comprising an expressible gene encoding said gRNA.

20. The method of claim 16, wherein contacting a host cell with a Cas nuclease is contacting said host cell with a polynucleotide comprising an expressible gene encoding said Cas nuclease.

21. The method of claim 16, wherein contacting a host cell with a fusion polypeptide comprising an Acr polypeptide and a receptor domain is contacting said host cell with a polynucleotide comprising an expressible gene encoding said fusion polypeptide comprising an Acr polypeptide and a receptor domain.

Description

FIGURE LEGENDS

[0132] FIG. 1: Light-induced disorder enables tight control of AcrIIA4. (A) Schematic of LOV2 insertion library generation. The LOV2 domain was inserted at 41 different positions into AcrIIA4 (indicated in C). Light-induced unfolding of the LOV2 domain should then cause disorder of the AcrIIA4 structure and hence inhibition of AcrIIA4, i.e. release of Cas9 activity (see FIG. 1B). (B) LOV2-AcrIIA4 enables light-dependent genome editing. The LOV2-AcrIIA4 hybrid indicated with * in (C) was co-transfectcd into HEK 293T alongside a Cas9 expression vector and a CFTR locus-targeting guideRNA. Sixteen h post transfection, cells were irradiated with blue light (7 s ON. 7 s OFF, 3 W/m.sup.2) for 32 h or kept in the dark as control. A T7 endonuclease assay was performed to monitor target locus cleavage. (Q Screen of LOV2-AcrIIA4 hybrid constructs. The LOV2 domain was inserted at the indicated positions into AcrIIA4 sequence (SEQ ID NO: 1). primarily into loops (bottom; Sec.Strct., secondary structure, derived from Dong et al., 2017). AcrIIA4-LOV2 hybrid variants were then investigated for their ability to inhibit SpyCas9 in the absence of blue light (bar chart, top). HEK 293T cells were transfected with the constructs in FIG. 6A alongside the indicated AcrIIA4-LOV2 hybrid. Forty-eight h post-transfection, a dual luciferase assay was performed. Firefly photon counts were normalized to Renilla photon counts. One loop comprising amino acids 62 to 69 tolerates the LOV2 insertion. Data represent means±s.e.m., 3 replicates, polypeptides with insertions at positions between amino acids 62 and 69 are those of SEQ ID NOs: 78 to 84 (encoded by SEQ ID NOs: 38 to 44, respectively), insertion between amino acids 85 to 87 or appended to amino acid 87 are those of SEQ ID NOs: 85 to 87 (encoded by SEQ ID NOs: 45 to 47, respectively).

[0133] FIG. 2: Improved light-induced disorder by systematic linker variation and embedding into the target protein. (A) Second generation LOV2-AcrIIA4 hybrids based on the variant indicated with * in FIG. 1C. Variants differ by the presence or absence of short OS linkers (bold) as well as optional deletion of AcrIIA4 residue E66 or Q65/E66; fusion site sequences are those of SEQ ID NOs: 7 to 15; fusion polypeptide sequences are those of SEQ ID NOs: 88 to 96 (encoded by SEQ ID NOs: 48 to 56). (B) Screen of the second generation LOV2-AcrIIA4 hybrid variants (numbers correspond to constructs in A). HEK 293T cells were transfected with the constructs in FIG. 6A alongside the indicated AcrIIA4-LOV2 hybrid. Six h after transfection, cells were irradiated with blue light (5 s ON, 15 s OFF, 2 W/m2) or kept in the dark for 42 h. A dual luciferase assay was performed. Firefly photon counts were normalized to Renilla photon counts. Data represent means from 2 replicates. (C) Third generation LOV2-AcrIIA4 hybrids with elongated linkers. (B, D) Acr, Res. #, AcrIIA4 residue number; fusion site sequences are those of SEQ ID NOs: 7, 15, and 16 to 19, fusion polypeptide #10 to 13 sequences are those of SEQ ID NOs: 97 to 100 (encoded by SEQ ID NOs: 57 to 60). (D) Screen of the third generation L0V2-AcrIIA4 hybrid variants by luciferase assay (numbers correspond to constructs in A and C). Experiment was performed as in B. Data represent means from at least 3 replicates. (E) LOV2-AcrIIA4 enables light-dependent genome editing. Indicated LOV2-AcrIIA4 hybrids from B and D were co-transfected into HEK 2931 alongside a Cas9 expression vector and a CFTR locus-targeting guideRNA. Six h post transfection, cells were irradiated with blue light (5 s ON, 10 s OFF, 3 W/m2) for 42 h or kept in the dark as control. A T7 endonuclease assay was performed to monitor target locus cleavage.

[0134] FIG. 3: Rapamycin control of AcrIIA4. (A) Schematics of rapamycin-inducible AcrIIA4 variants bearing a UniRapR domain inserted into the identified engineering hotspot. Variants differ by the presence or absence of short OS linkers (bold) as well as optional deletion of AcrIIA4 residue E66 or Q65/E66; fusion site sequences are those of SEQ ID NOs: 1 (#1), 10 (#2), 13 (#3), 15 (#4), 21 to 22 (#5 to #6), and 17 (#7); fusion polypeptide sequences are those of SEQ ID NOs; 101 to 107 (encoded by SEQ ID NOs: 61 to 67). (B) Screen of UniRapR-AcrIIA4 hybrids (numbers correspond to constructs in A). HEK 293Y cells were transfected with the constructs in FIG. 6A alongside the indicated UniRapR-AcrIIA4 hybrid in A. Six h post transfection, cells were treated with 10 μM rapamycin for 42 h or DMSO (rapamycin solvent) as control. A dual luciferase assay was performed. Firefly photon counts were normalized to Renilla photon counts and resulting values were normalized to the corresponding positive controls (reporter+gRNA). Data represent means from 2 replicates.

[0135] FIG. 4: N-terminal LOV2 fusion also enables negative AcrIIA4 regulation with light. (A) Schematic of used constructs and partial sequences of the tested LOV2-Jα-AcrIIA4 hybrids. The LOV2-Jα part is underlined. Variant #8 is a fusion of the wild type LOV2 (L404-L546) to wild type AcrIIA4 (without the AcrIIA4 methionine-encoding start codon). All other variants bear mutations or insertions (indicated in bold), or deletions introduced to alter the Jα hydrophobicity or its possible interaction with the AcrIIA4 domain; fusion site sequences are those of SEQ ID NOs: 23 to 28; fusion polypeptide sequences are those of SEQ ID NOs: 108 to 113 (encoded by SEQ ID NOs: 68 to 73). (B) Bar plot showing dCas9-VP64 transactivation assay in HEK 293T cells transfected with the indicated vectors in A alongside a dCas9-VP64 vector (SEQ ID NO: 117), a TetO-dependent luciferase reporter (SEQ ID NO: 118), a TetO-targeting guideRNA as well as a constitutive Renilla expression vector for normalization purposes (refer to FIG. 6C). Cells were irradiated with blue light (3 s ON, 17 s OFF, 2 W/m.sup.2) for 42 h or kept in the dark as control. Subsequently, a dual luciferase assay was performed. Firefly photon counts were normalized to Renilla photon counts. Data represent means±s.e.m., 3 replicates. (C) Bar plot showing Cas9 reporter cleavage assay in HEK 293T cells transfected with the indicated vectors in FIG. 6A alongside the corresponding LOV2-Jα-AcrIIA4 hybrid in A. Cells were irradiated with blue light (3 s ON, 17 s OFF, 2 W/m2) for 42 h or kept in the dark, followed by a dual luciferase assay as in B. Data represent means±s.e.m., 3 replicates.

[0136] FIG. 5: Induced nuclear export inhibits AcrIIA4 function. (A) Schematic of AcrIIA4 construct fused to mCherry-LEXY. LEXY, light-inducible nuclear export system; fusion polypeptide sequence is SEQ ID NO: 114 (encoded by SEQ ID NO: 74). (B) Fluorescence images of HEK 293T cells transfected with the construct in A. mCherry images were taken prior to induction (Preinduction), after 15 min of irradiation with blue light (Post Activation) and after an additional 20 min recovery phase in the absence of blue light (Post Recovery). (C) Line profile of the indicated cell in B, (D) Bar plot showing control of dCas9-VP64-mediated luciferase reporter activation by NLS-AcrIIA4-mCherry-LEXY. HEK 293T cells were transfected with the constructs in FIG. 6C alongside a constitutive Renilla expression vector (for normalization purposes) and optionally a wildtype AcrIIA4 expression vector or the NLS-AcrIIA4-mCherry-LEXY vector in A. The used ratio of dCas9-VP64 and AcrIIA4-mCherry-LEXY-encoding vector during transfection was varied as indicated. Cells were irradiated with blue light (3 s ON, 17 s OFF, 2 W/m.sup.2) for 42 h or kept in the dark, followed by dual-luciferase assay. Firefly luciferase photon counts were normalized to Renilla photon counts, 1, reporter+guideRNA; 2-4, reporter+guide RNA+dCas9-VP64; 2, no AcrIIA4; 3, NLS-AcrIIA4-mCherry-LEXY; 4, wild type AcrIIA4.

[0137] FIG. 6: Targeting AcrIIA4 to the nucleus improves Cas9 inhibition. (A) Schematic of Cas9 reporter cleavage assay. Co-delivery of Cas9, a firefly luciferase reporter and a luciferase-targeting guideRNA results in potent luciferase knockdown. The firefly reporter also bears a constitutive Renilla expression cassette for normalization purposes (SEQ ID NO: 119). AcrIIA4-mediated Cas9 inhibition prevents reporter knockdown. (B) Cas9 reporter cleavage assay in HEK 293T cells co-transfected with the constructs in A alongside an AcrIIA4 expression vector bearing a SV40 NLS at the N-terminus or not; the N-terminal SV40 NLS fusion to AcrIIA4 had the amino acid sequence of SEQ ID NO: 115, encoded by the nucleic acid sequence of SEQ ID NO: 75. Forty-eight h post-transfection, firefly and Renilla luciferase activity were measured by dual-luciferase assay. Firefly luciferase photon counts were normalized to Renilla photon counts. 1, reporter+guideRNA: 2-4 reporter+guideRNA+Cas9; 2, no AcrII4; 3, wild-type AcrIIA4; 4, SV49 NLS-AcrIIA4. Data are means±s.e.m., 3 replicates. (C) Schematic of dCas9-VP64 reporter trans-activation assay. Co-delivery of dCas9-VP64 and a TetO-targeting guideRNA results in potent activation of a firefly luciferase reporter driven from a minimal promoter preceded by TetO repeats. (D) dCas9-VP64 reporter trans-activation assay in HEK 293T cells co-transfected with the constructs in C as well as a constitutive Renilla expression vector (for normalization) and an AcrIIA4 expression vector bearing a cMyc.sup.PIA NLS at the N-terminus or not. Forty-eight h post -transfection, firefly and Renilla luciferase activity were measured by dual-luciferase assay. Firefly luciferase photon counts were normalized to Renilla photon counts. 1, reporter 4 guideRNA; 2-4 reporter+guideRNA+dCas9-VP64; 2, no AcrIIA4; 3, wild-type AcrIIA4; 4, cMyc.sup.PIA NLS-mCherry-AcrIIA4. Data are means±s.e.m., 3 replicates; the N-terminal cMyc.sup.PIA NLS fusion to AcrIIA4 had the amino acid sequence of SEQ ID NO: 116, encoded by the nucleic acid sequence of SEQ ID NO: 76.

[0138] FIG. 7: Schematic and sequences of further Acr-LOV hybrids. The Acr-LOV hybrids can e.g. be encoded by the nucleic acid sequences provided as SEQ ID NOs: 134 to 149.

[0139] FIG. 8: Eight control of luciferase reporter cleavage by different Act-LOV hybrids. HEK 293T cells expressing Cas9, a luciferase reporter (Rep), a reporter-targeting gRNA and the indicated LOV-Acr variant in FIG. 7 were irradiated with 3 W per m pulsatile blue light for 48 h or kept in the dark followed by luciferase assay. Data arc means±s.d.

[0140] FIG. 9: Cas9 inhibition is dose-dependent. HEK 293T cells were co-transfected with plasmids encoding (i) Acr-LOV hybrid, (ii) Cas9 and (iii) a luciferase reporter as well as a gRNA targeting the luciferase gene. The vector mass ratio of the transfected Cas9 and Acr-LOV construct was varied between 10:1 and 1:1, as indicated. Six hours post-transfection, cells were irradiated with pulsatile blue light (5 s ON, 10 s OFF; 2.5 W per m.sup.2) for 30 h or kept in the dark as control before assessing luciferase activity. Data are means±s.e.m.

[0141] FIG. 10: Light-dependent editing of endogenous loci in HEK 293T cells. Transgenes were delivered by transient plasmid DNA transfection or AAV transduction. Light-mediated indel mutation of human CCR5 (a), CFFR (b) or EMX1 (c) locus. Cells were co-transfected or transduced with vectors expressing CASANOVA, Cas9 and a locus-specific gRNA, and then exposed to blue light for 70 h or kept in the dark as control. For the transfection samples, the used vector mass ratio of the Acr:Cas9 construct is indicated. Editing frequencies wore evaluated by mismatch-sensitive T7 endonuclease assay. Representative gel images and corresponding quantifications of editing efficiencies are shown. Data arc means±s.e.m. Wt, wild-type. CN, CASANOVA. *P<0.05. **P<0.01, ***P<0.001 by Student's t-test. (g) Note that CCR5 T7 fragments have the same size.

[0142] FIG. 11: Acr-LOV hybrids carrying a C450A LOV2 pseudodark mutation arc still light-responsive. (a) Light-dependent luciferase reporter cleavage mediated by different Acr-LOV hybrid pseudodark mutants. HEK 293T cells were co-transfected with plasmids encoding (i) the indicated Acr-LOV hybrid variant, (ii) Cas9 and (iii) a luciferase reporter as well as a gRNA targeting the luciferase gene. Six hours post-transfection, cells were irradiated with pulsatile blue light for 48 h or kept in the dark as control before assessing luciferase activity. Data arc means±s.d., (b) Quantification of light-mediated indel mutation of the human CCR5 focus by T7 endonuclease assays. HEK 293T cells were co-transfected with constructs expressing Cas9, CCR5 locus-targeting gRNA and the indicated Acr-LOV hybrid variant and exposed to blue light for 470 h or kept in the dark as control. During transfection, the vector mass ratio of Acr-LOV:Cas9 construct was varied as indicated, n.d., not determined. The Acr-LOV hybrids can e.g. be encoded by the nucleic acid sequences provided as SEQ ID NOs:150 to 155.

[0143] FIG. 12: Cas9 inhibition can be modulated via mutations that affect docking of the LOV2 terminal helices, (a) Light-dependent luciferase reporter cleavage mediated by different Acr-LOV hybrid mutants. HEK 293T cells were co-transfected with plasmids encoding (i) the indicated Acr-LOV hybrid variant, (ii) Cas9 and (iii) a luciferase reporter as well as a gRNA targeting the luciferase gene. Six hours post-transfection, cells were irradiated with pulsatile blue light for 48 h or kept in the dark as control before assessing luciferase activity. Data arc means±s.d., (b) Quantification of light-mediated indel mutation of the human CCR5 locus by T7 endonuclease assay. HEK 293T cells were co-transfected with constructs expressing the Cas9, the CCR5 locus-targeting gRNA and the indicated Acr-LOV hybrid variant and exposed to blue light for 70 h or kept in the dark as control. During transfection, the vector mass ratio of Acr-LOV:Cas9 construct was varied as indicated. Data arc moans. The Acr-LOV hybrids can e.g. be encoded by the nucleic acid sequences provided as SEQ ID NOs:156 to 165.

[0144] FIG. 13; In silico docking analysis reveals Acr mutations that improve CASANOVA performance, (a) Light-dependent luciferase reporter cleavage mediated by different Acr-LOV hybrid mutants. HEK 293T cells were co-transfected with vectors encoding (i) the indicated Acr-LOV hybrid variant, (ii) Cas9 and (in) a luciferase reporter as well as a gRNA targeting the luciferase gene. Six hours post-transfection, cells were irradiated with pulsatile blue light for 48 h or kept in the dark as control before assessing the luciferase activity. Data arc means±s.d., (b) Quantification of light-mediated indel mutation of the human CCR5 locus by T7 endonuclease assay. HEK 293T cells were co-transfected with constructs expressing Cas9, the CCR5 locus-targeting gRNA and the indicated Acr-LOV hybrid variant and exposed to blue light for 70 h or kept in the dark as control. During transfection, the vector mass ratio of Acr-LOV:Cas9 construct was varied as indicated. n.d., not determined. The Acr-LOV hybrids can e.g. be encoded by the nucleic acid sequences provided as SEQ ID NOs:166 to 176.

[0145] FIG. 14: Comparison of light-dependent indel mutation by CASANOVA and its corresponding S46D and T16F mutants. HEK 293T cells were co-transfected with constructs expressing Cas9, a gRNA and the indicated CASANOVA variant and exposed to blue light for 70 h or kept in the dark as control. During transfection, the vector mass ratio of Acr-LOV:Cas9 construct was varied as indicated. Editing frequencies were evaluated by mismatch-sensitive T7 endonuclease assay, (a) Indel mutation of CCR5 locus and (b) indel mutation of EMX1 locus, (a-b) Data arc means±s.e.m.

[0146] FIG. 15: Optogenetic control of xCas9. Light-dependent luciferase reporter cleavage mediated by different Acr-LOV hybrid mutants. HEK 293T cells were co-transfected with vectors encoding (i) the indicated Acr-LOV hybrid variant, (ii) xCas9 and (iii) a luciferase reporter as well as a gRNA targeting the luciferase gene. Six hours post-transfection, cells were irradiated with pulsatile blue light for 48 h or kept in the dark as control before assessing luciferase activity. Data are mean±s.d.

[0147] FIG. 16: Optogenetic control of gene expression, (a) Schematics showing concept of light-mediated activation of IL1RN expression using CASANOVA and a dCas9-based acetyltransferase (dCas9-p300) targeted to the IL1RN promoter, (b) Light control of IL1RN expression. HEK 293T cells expressing CASANOVA and a dCas9-p300 fusion targeted to the IL1RN promoter via four gRNAs were exposed to blue light for 44 h or kept in the dark. IL1RN expression was assessed by quantitative RT-PCR. Data arc means±s.e.m.

[0148] FIG. 17: Schematics showing concept of optogenetic control of telomere labeling.

[0149] FIG. 18; Analysis of light-mediated telomere recruitment in fixed samples. U2OS cells expressing dCas9-3xRFP. a tetomere-targeting gRNA and CASANOVA for wild-type or no Acr instead of CASANOVA) were exposed to blue light for 20 h or kept in the dark as control. Telomere labeling in individual nuclei was then quantified by automated image analysis in KNIME. The violin plot shows the distribution, while black bars and grey dots indicate the median and individual data points, respectively. **P<2.2.Math.10.sup.−16 by Wilcoxon rank-sum test.

[0150] The following Examples shall merely illustrate the invention. They shall not be construed, whatsoever, to limit the scope of the invention.

EXAMPLE 1

Cell Culture

[0151] Human embryonic kidney cells with SV40 large T-antigen (HEK 293T) were maintained in phenol red-free Dulbecco's Modified Eagle Medium (DMEM) supplemented with 10% fetal calf serum (Biochrom AG), 2 mM L-glutamine (Invitrogen/Gibco), 100 U/ml penicillin and 100 mg/ml streptomycin (Invitrogen/Gibco). Cells were cultivated at 37° C. and 5% CO2 and passaged when reaching ˜90% confluency. Before usage, the cell line was authenticated and tested tor mycoplasma contamination using the commercial Multiplex Cell Line Authentication and Mycoplasma Test services (Multiplexion).

EXAMPLE 2

Plasmid Construction

[0152] Constructs were generated using classical restriction enzyme cloning. Oligonucleotides were obtained from Sigma-Aldrich or Integrated DNA Technologies (IDT) and codon-optimized DMA sequences were purchased as gBlocks from IDT. Restriction enzymes were purchased from New England Biolabs and Thermo Fisher. PCR amplification of DNA fragments was performed using primers at a concentration of 0.5 μM with Phusion® High-Fidelity DNA Polymerase, by Thermo Fisher, or Q5® High-Fidelity DNA Polymerase, by New England Biolabs, according to the manufacturer's recommendations. Gels for electrophoresis were prepared with 1% agarose in 0.5× TAB. QIAquick Gel Extraction Kit (QIAGEH) was used to isolate DNA from prepared gel fragments. QIAquick Nucleotide Removal Kit or QIAquick PCR Purification Kit (both QIAGEN) were used for purification of DNA fragments without gelelectrophoresis from enzymatic digests or PCR reactions. QIAprep Spin Miniprep Kit and QIAQEN Midi Kit (both QIAGEN) were used for isolation of plasmid DNA for subsequent cloning, sequencing or transfection. Plasmids and ligation products were transformed into chemically competent E. coli TOP 10 and plated without recovery in liquid culture. Bacteria were cultivated on LB-agar plates or in LB-liquid cultures with 100 μg/ml ampicillin at 37° C. Sequences of new plasmids were validated through Sanger sequencing using the services of GATC.

EXAMPLE 3

Luciferase Assay

[0153] HEK 293T cells were seeded into black, clear-bottom 96-well plates (Corning) at a density of ˜12.500 cells per well Twenty-four hours after seeding, cells were transfected according to the manufacturer's instructions using Lipofectamine® 2000 and Opti-MEM® reduced serum medium (Thermo Fisher).

[0154] For the dCas9-VP64 trans-activation assays, 50 ng/well of each, a plasmid encoding dCas9-VP64_GFP (Addgene plasmid #61422, kind gift from Feng Zhang), a tet-inducible firefly luciferase reporter (Addgene plasmid #64127, kind gift from Moritoshi Sato) and a TetO-targeting guide RNA vector (Addgene plasmid #64161, kind gift from Moritoshi Sato) were co-transfected alongside 1 ng/well pRL-TK (TK-driven Renilla expression vector for normalization; Promega) as well as 1-50 ng/well of the AcrIIA4 constructs and, optionally, stuffer DMA (pcDNA3.1(−) (Invitrogen)). For the reporter cleavage assay, a plasmid encoding an H1 promoter-driven guideRNA as well as firefly luciferase and Renilla luciferase was co-transfected alongside Cas9 expression vector pSpCas9(BB)-2A-GFP (PX453; Addgene plasmid #48138, kind gift from Feng Zhang) as weft as the corresponding AcrIIA4 variant (50 ng/well each).

[0155] To assess the effects of NLS fusion to AcrIIA4, 40 ng/well of the reporter construct, 40 ng/well of plasmid encoding Cas9 and 120 ng/well of the AcrIIA4 constructs or stuffer DNA were co-transfected. To investigate the different AsLOV2 insertion sites in AcrIIA4, typically 50 ng/well of each vector encoding the luciferase cleavage reporter, Cas9 and the AcrIIA4-LOV2 hybrid were co-transfected.

[0156] Six h post-transfection, medium was replaced and the cells were illuminated with 460 nm pulsatile blue light (light intensity 2-3 W/m.sup.2 as measured with a LI-COR LI-250A Light Meter, pulsatile illumination regime is indicated in figure legends) or kept in the dark under otherwise identical conditions for 32-42 h (indicated in figure legends). A custom-made LED device composed of six high-power LEDs (type CREE XP-E D5-15; LED-TECH.DE) empowered by a Switching Mode Power Supply (Manson, model: HCS-3102) served as light source. A Raspberry Pi with a custom-made python script was used for light intensity and pulsing control. Subsequently, the luciferase activity was measured using the Dual-Glo luciferase assay kit (Promega) according to the manufacturer's protocol. In brief, the cells were lysed using the provided lysis buffer and tire activity of firefly and Renilla luciferase was quantified using a GLOMAX 96 Microplate Luminometer (Promega) with automated injectors (delay time 2 s, integration time 10 s). The relative luciferase activity was calculated by dividing the firefly luciferase photon counts by the Renilla luciferase photon counts. In some eases, for reasons of comparison, the average relative luciferase activities were then normalized to the corresponding light/dark values of the reporter maximum control (cleavage assay=reporter control transactivation assay=dCas9-VP64 control) (indicated with “normalized” on the figure axis).

EXAMPLE 4

T7 Assay

[0157] HEK 293T cells were seeded into transparent 96-well plates (Coming). The next day, cells were transfected with equal amounts of Cas9 expression vector, CFTR guideRNA (sequence 5-GAATGGTGCCAGGCATAATCC-3′, SEQ ID NO: 29) expression vector as well as vector encoding wild-type AcrIIA4 or AcrIIA4-LOV2 hybrid (carrying AsLOV2 inserted between N64 and Y67) using Lipotectamine 2000. Sixteen It post-transfection, cells were irradiated with blue light (7 s ON, 7 s OFF, 3 W/m.sup.2) for 32 h or kept in the dark as control. Subsequently, cells were washed with PBS and lysed using the DirectPCR lysis reagent (PcqLab). A fragment spanning the edited part of the CFTR locus was PCR-amplified using Q5 polymerase and the following primers: CFTR_fw: 5′-GCACATAGAACAGCACTCGAC-3′, SEQ ID NO: 30; CFTR_re: 5′-GATCCATICACAGTAGCTTACCC-3′, SEQ ID NO: 31). The PCR reaction was run on an agarose gel and the amplicon was purified using a QIAGEN Gel Extraction Kit. Two hundred ng of PCR diluted in 19.5 μl 1× NEB Buffer 2 were then healed up to 95° C. followed by re-annealing in a thermocycler. Next, 0.5 μl of T7 Endonuclease (NEB) were added, and the reaction was incubated at 37° C. for 15 min and then stopped by adding EDTA-containing gel loading dye. Lastly, T7 reactions were analyzed on a 2% agarose gel stained with GelRed (Biotium, Inc.).

EXAMPLE 5

Control of AcrIIA4 by Light-Induced Disorder

[0158] We investigated whether AcrIIA4 generally tolerates insertion of a receptor into a surface-exposed region, ideally a loop. As receptor, we chose the LOV2 blue light sensor from Avena sativa phototropin-1 (residues L404-L546 of Genbank Acc No: AAC05083.1, SEQ ID NO: 34). 41 different AcrIIA4 positions were chosen for insertion the AsLOV2 domain (As phototropin-1 residues L104-LS46) without any additional linkers (FIG. 1A and C, bottom). A comparison of insertion sites with the recently reported AcrIIA4 (secondary) structure (Dong et al, 2017) is included in (FIG. 1C, bottom).

[0159] Next, we tested the AcrIIA4-LOV2 hybrids for their ability to inhibit Cas9 catalytic activity using a luciferase reporter cleavage assay (FIG. 1C). One loop spanning from G62 to D69 tolerated A&LOV2 insertions. Cas9 inhibition was highest for the AcrIIA4 variant bearing AsLOV2 between residues E66 and Y67 (FIG. 1C; indicated with *). This is particularly surprising, as residue Y67 as well as neighboring residues D69 and H70 mediate important interactions with the PAM-binding residues of Cas9 and are thus of high importance for Cas9 inhibition by AcrIIA4 (Dong et al., 2017).

[0160] It thus appeared plausible that light-induced unfolding of the LOV2 terminal helices, in particular the AsLOV2-Jα directly preceding residues Y67/D69/E70, could disturb the Cas9 binding of AcrIIA4, thereby releasing Cas9 activity. To prove that light-induced unfolding of the LOV2 terminal helices interferes with Cas9 binding of AcrIIA4. HEK. 293T cells were co-transfected with a Cas9 construct, a vector expressing a CFTR locus-targeting guideRNA as well as a vector encoding the AcrIIA4 variant with the LOV2 domain inserted between residues E66/Y67 or wild-type AcrIIA4 as control. Note, that the ratio of Cas9:AcrIIA4 construct was 1:4.Sixteen h post transfection, cells were exposed to blue light for 32 h or kept in the dark, followed by T7 assay to measure target locus editing. As anticipated, the AcrIIA4-LOV2 hybrid inhibited Cas9 in a light-dependent manner, indicated by the increased editing of the CFTR locus in the light as compared to the dark control sample (FIG. 1B).

[0161] Noticeable CFTR editing was also observed in the dark sample, suggesting that the AcrIIA4-LOV2 hybrid did not fully block Cas9. Next, it was tested whether inserting short, flexible linkers at the LOV2-AcrIIA4 junction sites could improve Cas9 inhibition in the dark. Using our lead construct (LOV2 inserted between AcrIIA4 residues E66/YC7; FIG. 2A, construct #1) as scaffold, we generated a set of 8 additional AcrIIA4-LOV2 hybrids bearing short G or GS linkers at the LOV termini and optionally an E66 single or Q65/E66 double deletion on AcrIIA4 (FIG. 2A, constructs 42-9). The newly generated variants were transfected into HEK 293T cells together with a Cas9 expression vector as well as a corresponding luciferase cleavage reporter. Cells were irradiated with blue light for 42 h or kept in the dark, followed by luciferase assay (FIG. 2B). Several new variants (#4, #7, #8 and #9) outperformed the parental variant (#1) and showed a potent Cas9 inhibition in the dark as well as strong light-induced release of Cas9 catalytic activity (FIGS. 2A and B). Variant #9 (FIGS. 2A and B) showed a ˜2-fold increase in Cas9 inhibition in the dark compared to the parental construct (#1) and a strong (9-fold) increase in Cas9-mediated reporter knockdown upon irradiation.

[0162] Notably, this candidate bears the AcrIIA4 Q65/E66 double deletion and LOV2-flanking GS linkers.

[0163] Based on this knowledge, further constructs bearing either the Q65/E66 double or E66 single deletion with elongated GS linkers were designed (FIG. 2C) and evaluated for light-dependent Cas9 inhibition using a luciferase assay as before (FIG. 2D). This identified LOV-AcrIIA4 hybrids that arc even more potent at inhibiting Cas9 in the dark and strongly releasing this inhibition upon irradiation (FIG. 2D. compare construct 9 to constructs 10-13).

[0164] A subset of these newly created LOV2-AcrIIA4 hybrids (construct 7, 9, 11) was investigated for their ability to control Cas9 editing of the endogenous CPTR locus by T7 assay and compared to the parent variant (construct #1). In an assay using a ratio of transfected Cas9:AcrIIA4 construct of 1:1, the parent LOV2-AcrIIA4 hybrid gave no significant Cas9 inhibition in the dark (FIG. 2E, construct 1). In contrast, the new variants showed moderate (construct 7) to highly potent (constructs 9 and 11) Cas9 inhibition in the dark and a strong release of Cas9 activity upon activation (FIG. 2E). indicating successful light control of Cas9 editing of an endogenous locus using our engineered LOV2-AcrIIA4 hybrids.

Example 6

Rapamycin Control of Cas9 Activity

[0165] To evaluate whether chemical triggers could be used for systemic Cas9 control, e.g. in animals, we replaced the light input by a clinically approved drug. To this end, we employed the UniRapR receptor, a previously reported FRB-iFKBP fusion (Dagliyan et al., 2013), whose conformation is stabilized upon rapamycin binding. Using the LOV2-AcrIIA4 hybrids as blueprint (FIG. 2), we inserted the UniRapR domain between AcrIIA4 residues N64 and Y67 (FIG. 3A). We again introduced optional GS-linkers at the AcrllA4-UniRapR interfaces as well as Q65/E66 deletions on AcrIIA4 (FIG. 3A), both of which has been beneficial in the context of the LOV2-AcrIIA4 hybrids engineered before. The generated constructs were then investigated for their ability to inhibit Cas9 mediated luciferase reporter cleavage in HEK 293T cells in the presence or absence of 10 μM rapamycin (FIG. 3B). Remarkably, in particular engineered UniRapR-AcrIIA4 hybrids comprising linkers showed a prominent regulation of Cas9 catalytic activity upon rapamycin treatment (FIG. 3B, constructs #4-7). Thus, rapamycin treatment activates AcrIIA4 which in turn results in Cas9 inhibition, while AcrIIA4 is disordered (fully inactive) in the absence of the trigger (FIG. 3B).

[0166] Of note, the lead candidate obtained from this initial, small UniRapR-AcrIIA4 hybrid screen (construct 7, FIGS. 3A and B) not only shares the GSG-thinking linkers, but also the beneficial Q65/E66 double deletion with the lead candidate of our LOV2-AcrIIA4 screen (compare construct 11 m FIG. 2C-E and construct 7 in FIG. 3). It can thus be concluded that the AcrIIA4 engineering hotspot identified in this work can be targeted by different receptors to control Cas9 inhibition with diverse triggers.

Example 7

N-terminal photoreceptor fusion also enables AcrIIA4 light control

[0167] An allosteric AcrIIA4 light-switch was also created by connecting the rigid C-terminal LOV2 Jα helix and the N-terminal AcrIIA4 helix. To this end, we generated different LOV2-AcrIIA4 fusions, in which the LOV2-Jα-AcrIIA4 interface was optionally modified as follows. We either (i) inserted previously described mutations into the Jα helix (FIG. 4A; #41, #51, #53-41, #69-41, #71-51), that could improve helix docking and photoswitching (Niopek et al, 2016), (ii) introduced short, N-terminal AcrIIA4 deletions (FIG. 4A; #53-41) or (iii) inserted short linkers between the C-terminal LOV2 and the N-terminal AcrIIA4 helices (FIG. 4A: #69-41, #71-51). The resulting LOV2-AcrIIA4 hybrids were then tested using either a dCas9-VP64 luciferase reporter trans-activation assay (FIG. 4B) or a Cas9 luciferase reporter cleavage assay (FIG. 4C). As hypothesized, the constructs showed noticeable light-dependent dCas9-VP64 and Cas9 inhibition (FIGS. 4B and C), although the dynamic range of regulation was modest (2- to 3-fold change in reporter activity upon irradiation).

Example 8

Light-Induced AcrIIA4 Cytoplasmic Sequestration Releases Cas9 Activity

[0168] To test whether sequestering AcrIIA4 away from the nucleus could be a possible mode for allosteric Acr control we employed a light-inducible nuclear export system (LEXY) previously reported (Niopek et al., 2016), which mediates the nuclear export of proteins fused to the LEXY domain (an AsLOV2-Nuclear Export Signal-Hybrid) in response to blue light. We generated a CMV promoter-driven construct expressing an NLS-tagged AcrIIA4 fused to mCherry (for visualization purposes) and LEXY (FIG. 5A), and transfected it into HEK 293T cells. Upon blue-light irradiation, we observed a fast and strong decrease in nuclear mCherry fluorescence, indicating successful sequestration of the AcrIIA4 fusion proteins into the cytosol (FIGS. 5B and C). The nuclear mChcrry fluorescence fully recovered within 20 min when blue-light irradiation was stopped, showing that AcrIIA4 nuclear export was reversible (FIGS. 5B and Q.

[0169] Finally, we tested whether cytosolic AcrIIA4 sequestration had an impact on Cas9 activity. To this end, we co-transfected HEK 293T cells with different amounts of the NLS-AcrIIA4-mCherry-LEXY fusion-encoding vector, dCas9-VP64 and a corresponding luciferase reporter (see FIG. 6C). Cells were then exposed to blue light for 48 h or kept in the dark, followed by luciferase assay.

[0170] As expected, nuclear export of AcrIIA4 caused a noticeable increase in dCas9-VP64-mediated luciferase reporter induction, indicating successful light control of AcrIIA4 (FIG. 5D). As moreover expected, the efficiency of dCas9-VP64 inhibition and release upon irradiation was dependent on the used NLS-AcrIIA4-mCherry-LE XY dose (FIG. 5C). Of note, the dynamic range of regulation was modest and concentration-dependent, indicating that cytosolic sequestration was not 100% efficient or that AcrIIA4-bound dCas9-VP64 was partially co-exported.

Example 9

AcrIIA4 Nuclear Targeting Improves Cas9 Inhibition

[0171] We aimed at independently verifying the results by Rauch et al. (2017) suggesting that AcrIIA4 efficiently inhibits target DNA binding of the catalytically active SpyCas9 as well as the catalytically impaired dCas9 mutant. Therefore, we co-transfected HEK 293T cells with vectors expressing Cas9, a firefly luciferase reporter also encoding a firefly luciferase targeting guideRNA as well as two different AcrIIA4 vectors differing by the presence or absence of an additional N-terminal SV40 NLS (nuclear localization signal). In the absence of AcrIIA4, Cas9 caused a prominent firefly luciferase knockdown, which was strongly reduced upon co-expression of AcrIIA4 (FIG. 6B). Remarkably, the AcrIIA4 construct bearing the additional NLS resulted in a more potent Cas9 inhibition, as reflected by a full recovery of the luciferase reporter signal (FIG. 6B). This boost in AcrIIA4 activity due to nuclear targeting could be further confirmed using a cMyc.sup.PIA-NLS-tagged AcrIIA4 variant in combination with a dCas9-VP64 trans-activator and corresponding reporter test system (FIGS. 2C and D).

[0172] These experiments verify the reported, high potency of AcrIIA4 at inhibiting SpyCas9 or dCas9. They further show that targeting AcrIIA4 to the nucleus improves Cas9 inhibition, likely by increasing the AcrIIA4:Cas9 ratio in the relevant cellular compartment. Therefore, we included an N-terminal NLS into all AcrIIA4 constructs used in the Examples above.

Example 10

Materials and Methods for Examples 11 to 13

Computational Design of Improved Acr-LOV Mutants

[0173] Interface design was performed for the interface residues in AcrIIA4 using the RosettaScripts application (Fleishman et al (2011). In silico saturation mutagenesis was performed for residues in close spatial proximity (residue set 1:16, 18, 33 and set 2:19, 28, 45). Designs with interaction energies (ddGs) within the same range (+2.5 rosetta energy units) or lower than that of the wild-type complex were manually inspected and the best mutations were selected for experimental characterization. Table 1 presents the metrics of the mutants experimentally characterized.

TABLE-US-00001 TABLE 1 CASANOVA mutants selected for experimental characterization. ddG indicates the predicted change in free energy upon binding to the Cas9/gRNA complex. The dHbond_gain_overall shows the number of additionally formed buried hydrogen bonds of the designs compared to the wild-type (baseline). Construct Rosetta score ddG dHbond_gain_overall baseline −1434.482 −124.294 0 T16Y −1412.166 −124.173 1 T16F −1432.37 −125.793 0 K18Q −1435.027 −125.65 3 T22H −1432.85 −125.192 0 T28E −1436.272 −124.224 0 T28N −1433.502 −125.348 1 T28Q −1440.055 −125.657 0 E45K −1438.617 −124.158 1 S46D −1396.953 −122.129 1 N64K −1435.808 −125.231 0 N64R −1409.696 −124.297 1

[0174] General Methods and Cloning

[0175] Plasmids were created using classical restriction enzyme cloning, Golden-gate cloning (Chen et al. (2013)) or Gibson assembly (New England Biolabs). Oligonucleotides were obtained from IDT or Sigma Aldrich. Synthetic, double-stranded DNA fragments were obtained from IDT. The CMV promoter-driven SpyCas9 expression vector was obtained by PCR-amplifying the SpyCas9 gene from vector pSpCas9(BB)-2A-GFP (kind git) from Feng Zhang (Addgene plasmid #48138)) followed by ligation into pcDNA3.1.sup.(.) (ThermoFisher) via XhoI/HindIII. AAV vectors encoding SpyCas9 or a U6 promoter-driven, improved gRNA scaffold (F+E Chen et al. (2013)) and RSV promoter-driven GFP (Senis et al. (2014)) were employed for gRNA expression. Annealed oligonucleotides corresponding to the target site sequence were cloned into the gRNA AAV vector via BbsI using Golden-gate cloning. The luciferase reporter for measuring SpyCas9 activity (luciferase cleavage reporter) was developed by cloning an H1-driven expression cassette encoding a firefly luciferase-targeting gRNA into pAAVpsi2Borner et al. (2013). The resulting vector co-encodes an SV40 promoter-driven Renilla luciferase gene and a TK promotor-driven Firefly luciferase gene. The AcrIIA4 coding sequence was obtained as human codon-optimized, synthetic DNA fragment from IDT and cloned into pcDNA3.1.sup.(.) via NheI/NotI. Acr-LOV hybrids were created by linearizing the Acr-encoding vector by PCR followed by insert ion of a human codon-optimized Avena saliva LOV2-encoding fragment (IDT) via blunt-end ligation or Golden-gate cloning. GS linkers were optionally appended to the LOV-encoding DNA fragment via PCR prior to ligation. Mutations were introduced into the Acr part of the Acr-LOV hybrids by site-directed mutagenesis using 5′ phosphorylated primers. Mutations were inserted into the LOV part of the Acr-LOV hybrids by PCR-amplifying the LOV2 domain with primers introducing the mutations into the N- and C-terminal helix and cloning the altered LOV fragment back into a PCR-linearized, patent Acr-LOV hybrid vector using Golden-gate cloning. Note that wild-type Acr as well as all Acr-LOV hybrids bear an N-terminal SV40 nuclear localization signal, which we added to target the Cas9 inhibitor to the nucleus. The xCas9 cDNA was created by Gibson assembly on basis of the reported SpyCas9 mutations (Hu et al. (2018)) using synthetic, double-stranded DNA fragments cloned into pcDNA3.1.sup.(.). The dCas9-p300 construct was a kind gift from Charles Gersbach (Addgene plasmid #61357). pEJS477-pHAGH-TO-SpydCas9_3XmCherry-sgRNA/Telomere-All-in-one was a gift from Erik Sontheimer (Addgene plasmid #85717). Based on this vector, constructs co-expressing dCas9_3XmCherry and CASANOVA or wild-type AcrIIA4 via a P2A peptide were created by cloning a P2A-CASANOVA or P2A-AcrIIA4 cDNA (IDT) behind the SpyCas9-3XmCherry coding sequence.

[0176] In all cloning procedures, PCRs were performed using Q5 Hot Start High-Fidelity DNA Polymerase (New England Biolabs) or Phusion Flash High-Fidelity polymerase (ThermoFisher). Agarose gel electrophoresis was used to analyze PCR products. Bands of the expected size were cut out and DNA extracted using a QIAquick Gel Extraction Kit (Qiagen). Ligations were performed using T4 DNA ligase (New England Biolabs) and optionally heat-inactivated at 70° C. for 45 min before transformation. Chemically-competent Top 10 cells (ThermoFisher) were used for DNA vector amplification. Plasmid DNA was purified using the QIAamp DNA Mini, Plasmid Plus Midi or Plasmid Maxi Kit (all from Qiagen).

[0177] Cell Culture, Transient Transfection and AAV Lysate Production

[0178] Cells lines w ere cultured at 5% CO.sub.2 and 37° C. in a humidified incubator and passaged when reaching 70 to 90% confluency (every two to four days). HEK 293T (human embryonic kidney) and U2OS (human osteosarcoma; kindly provided by Karsten Rippe, German Cancer Research Center (DKFZ), Heidelberg) were maintained in phenol red-free Dulbecco's Modified Eagle Medium (DMEM; ThermoFisher/GIBCO) supplemented with 10% (v/v) fetal calf scrum (Biochrom AG), 2 mM L-glutamine and 100 U per ml penicillin/100 μg per ml streptomycin (both ThermoFisher/GIBCO). The U2OS medium was additionally supplemented with 1 mM sodium pyruvate (GIBCO). Cell lines were authenticated and tested for mycoplasma contamination prior to use via a commercial service (Multiplexion). Transient transfections were performed with JetPrime (Polyplus transfection) or Turbofect (ThermoFisher) according to the manufacturer's protocols. Details are listed in the corresponding experimental sections below. For production of AAV-containing cell lysates, low-passage HEK 293T cells were seeded into 6-well plates (CytoOne) at a density of 350,000 cells per well. The following day, cells were triple-transfected with (i) the AAV vector plasmid, (ii) an AAV helper plasmid carrying AAV serotype 2 rep and cap genes, and (iii) an adenoviral plasmid providing helper functions for AAV production, using 1.33 μg of each construct and 8 μl of Turbofect reagent per well. The AAV vector plasmid encoded either (1) Cas9 driven from an engineered, short CMV promoter (Senis et al. (2014)), (2) a U6 promoter-driven gRNA (Senis et al. (2014)) (based on the improved F+E scaffold: Chen et al. (2013)) and a RSV promoter-driven GFP marker, or (3) a CMV promoter-driven CASANOVA variant. Seventy-two hours post-transfection, cells were collected in 300 μl PBS and subsequently subjected to five freeze-thaw cycles by alternating between snap freezing in liquid nitrogen and 37° C. Finally, the cell debris was removed by centrifugation at ˜18,000 g and the AAV-containing supernatant was stored at −20° C. until use.

[0179] Blue Light Setup

[0180] For blue light illumination of samples in the cell culture incubator, a custom-made blue light setup comprising six blue light, high-power LEDs (type OREL XP-E D5-15; emission peak ˜460 nm; emission angle ˜130°; LED-TECH.DE) empowered by a Switching Mode Power Supply (Manson, model: HCS-3102) was used. A Raspberry Pi running a custom Python script was used to control the power supply. Samples were irradiated from below, i.e., through the transparent bottom of the culture dishes or well plates by positioning them on an acrylic glass table installed in the incubator, with the LEDs being located underneath the table. A pulsatile illumination regime (5 s ON, 10 s OFF) was used for sample irradiation. Light intensity was ˜3 W per m.sup.2 as measured with a LI-COR LI-250A light meter, unless indicated otherwise below.

[0181] Luciferase Reporter Assays

[0182] HEK 293T were seeded into black, clear-bottom 96-well plates (Corning) at a density of ˜12,500 cells per well. The following day, cells were co-transfected with 33 ng of a Cas9 or xCas9 expression vector, 33 ng of a construct co-expressing Firefly and Renilla luciferase as well as an H1 promoter-driven gRNA targeting the Firefly luciferase cDNA, and, in most cases, 33 ng of the CMV promoter-driven Acr-LOV hybrid using 0.2 μl JetPrime (amounts are per well). For the titration experiment in FIG. 9, either 3, 10 or 33 ng of Acr-LOV hybrid was co-transfected with 30, 23 or 0 ng of an irrelevant DNA to vary the vector mass ratio of Cas9 and Acr-LOV construct between 10:1 and 1:1. For the experiment shown in FIG. 12. 8.25 ng of Acr-LOV constructs and 24.75 ng of an irrelevant DNA was used. Six hours post-transfection, the medium was exchanged and cells were exposed to blue light for 48 h or kept in the dark as control For the titration experiment shown in FIG. 9. irradiation time was 30 h and light intensity ˜2.5 W per m.sup.2. Subsequently, a Dual-Glo Luciferase Assay System (Promega) was applied to quantify luciferase activity. In brief, cells were harvested into the supplied lysis buffer and Firefly and Renilla luciferase activities were measured using a GLOMAX™ Discover or GLOMAX™ 96 microplate luminometer (both Promega). Integration time was 10 s, and delay time between automated substrate injection and measurement was 2 s. Firefly photon counts were normalized to Renilla photon counts and resulting values were further normalized to the positive control.

[0183] T7 Endonuclease Assay

[0184] Cells were seeded into black, clear-bottom 96-well plates (Coming) at a density of 12,500 cells per well for transfection based experiments or 3,500 cells per well for AAV transduction-based experiments. Transfections were performed with JetPrime using 0.3 μl of JetPrime reagent and 200 ng total DNA per well comprising either 33 ng of each, the gRNA, Cas9 and CASANOVA expression vectors as well as 100 ng of an irrelevant DNA (1:1 ratio Cas9:CASANOVA) or 33 ng gRNA, 33 ng Cas9 and 133 ng CASANOVA expression vector (1:4 ratio Cas9:CASANOVA). For AAV-based experiments, cells were co-transduced with 7 μl of each, the Cas9, gRNA and CASANOVA AAV lysates on two subsequent days when targeting the CCR5 locus. For CFTR and EMX1, 33 μl of each AAV lysate was used. Following transfection or transduction, cells were irradiated with blue light for 70 h or kept in the dark as control. Cells were washed with PBS and harvested in DirectPCR Lysis Reagent (Peqlab) supplemented with Proteinase K (Sigma). The genomic CRISPR/Cas9 target locus was PCR-amplified with primers flanking the target site using Q5 Hot Stan High-Fidelity DNA Polymerase (New England Bio labs).

[0185] Quantitative RT-PCR

[0186] HER 293T cells were seeded into transparent 6-well plates (CytoOne) at 250,000 cells per well. The next day, cells were co-transfected with (i) 750 ng IL1RN gRNA construct mix (Hilton et al. (2015) (187.5 ng per vector), (ii) 500 ng of a construct encoding dCas9-p300-P2A-CASANOVA (or an irrelevant DNA as control), and (iii) 250, 500 or 750 ng CASANOVA-encoding vector and 500, 250 or 0 ng of irrelevant stuffer DNA, respectively, using 6 μl JetPrime reagent (all amounts arc per well). The medium was replaced 4 h post-transfection and cells were irradiated with blue light pulses for 44 h or kept in the dark as control, before lysing cells using the QIAzol Lysis Reagent (Qiagen) according to the manufacturer's instructions. Reverse transcription was performed with the RevertAid First Strand cDNA Synthesis Kit (Thermo Fisher) and equal amounts of input RNA for each experiment. Real-time PCR reactions were set up using 2 μl cDNA mix (25 ng per μl). 1.4 μl of each 10 μM primer, respectively, 10 μl PowerSYB® Green PCR Master Mix (Thermo Fisher) and 5.2 μl water. A StepOne Plus real-time PCR system (Applied Biosystems) was employed with the following cycling conditions: 95° C./10 min initial denaturation followed by 45 cycles of (95° C./15-s-58° C./60 s). Fold-changes in IL1RN levels were then calculated using the ΔΔCt method (Livak et al., 2001).

[0187] Telomere Labeling Experiments

[0188] U2OS cells were seeded into 4-compartment CELLview cell culture dishes (Greiner Bio-One) at a density of 30,000 cells per compartment. The next day, cells were transfected with vectors encoding (i) a CMV promoter-driven dCas9-3xRFP-P2A-CASANOVA and a U6 promoter-driven telomere-targeting gRNA, (ii) a telomere-targeting gRNA and GFP transfection marker, and (iii) a CMV promoter-driven CASANOVA in a ratio of 20:6:3 using 362.5 ng total DNA and 1.5 μl JetPrime for transfection (per compartment).

[0189] In the positive control samples, vector i was replaced by a vector encoding dCas9-3xRFP (without the P2A-CASANOVA) and a U6 promoter-driven telomere-targeting RNA, and vector iii was replaced by an irrelevant DNA. In the negative control samples, the CASANOVA in vectors i and iii was replaced by wild-type AcrIIA4. Four hours post-transfection, the medium was changed and cells were either irradiated with blue light pulses for 20 h or kept in the dark followed by fixation of samples with 4% PFA. SlowFade™ Diamond Antifade Mountant with DAPI (Invitrogen) was added and imaging was performed using the aforementioned microscopy setup and the following excitation/detection settings: 405 nm (1% laser powerV410-490 nm for DAPI, 488 nm (2% laser power)/493-578 nm for GFP, or 552 nm (1% laser power)/578-789 nm for RFP. RFP fluorescent spots (i.e., labeled telomeres) were then detected and quantified using a fully automated image analysis pipeline as follows. The ImageJ2 (beta) Integration in KNIME Version 3.5.2 (KNIME AG) was used to create an automated image processing and analysis pipeline employed for quantification of labeled telomeres. Analysts of all images was performed using the identical workflow configuration, apart from the configuration of data input and output nodes. In brief, raw image stacks (Jif files) were imported into KNIME followed by splitting the three fluorescence channels (DAPI, nuclear marker; GFP, a transfection marker co-encoded on foe gRNA vector; RFP corresponding to dCas9-3XmCherry). Nuclei were segmented based on foe DAPI signal. GFP-negative nuclear segments (i.e., negative for the telomere-targeting gRNA construct) were excluded from the analysis. Furthermore, nuclear segments with a mean RFP signal higher than 170 (as images were 8 bit, this corresponds to ⅔ of the maximum) were also excluded from the analysis, as the very high RFP background fluorescence impaired reliable spot detection. The Spot Detection node was employed to identify and segment fluorescent spots in the RFP channel. All spots lying outside of the nuclear segments were excluded and random fluorescence fluctuations were filtered out by selecting for spots with an average fluorescence at least 1.7-fold higher than the RFP background fluorescence in the corresponding nuclear segment. The workflow output comprises a CSV table listing the nuclear segments and corresponding spots detected in each image. Subsequent data visualization and statistical analysis was performed in R version 3.3.2.

Example 11

AcrIIA4-LOV2 Variants

[0190] To further improve AcrIIA4-LOV2 hybrids, we aimed at embedding the LOV2 domain further into the C-terminal AcrIIA4 loop. Therefore, we stepwise deleted AcrIIA4 residues that directly precede the insertion site, but do not mediate critical contacts with Cas91 (FIG. 7, variants 2-5). Furthermore, GS-linkers of variable length were optionally included at the Acr-LOV boundaries (FIG. 7. variants 5-16). Indeed, several of the so-obtained Acr-LOV hybrids showed a potent Cas9 inhibition in the dark and almost full recovery of Cas9 activity in the light condition (FIG. 8). As expected for a competitive inhibitor, the degree of Cas9 inhibition depended on the dose of transfected Acr-LOV hybrid (FIG. 9). The most potent hybrid inhibitor (Acr-LOV variant 4) carried a three amino acid deletion (delta N64/Q65/E66) preceding the LOV domain insertion site and no GS-linkers (FIGS. 7 and 8). In the following, we will refer to Acr-LOV variant 4 or further optimized mutants thereof (see below) as CASANOVA (for CRISPR/Cas9 activity switching via a novel optogenetic variant of AcrIIA4).

[0191] Using transient transfection or Adeno-associated virus (AAV)-mediated transduction, we co-expressed CASANOVA alongside Cas9 and gRNAs targeting the CCR5, CFTR or EMX1 locus in HEK 293T cells, indel mutations were strongly light-dependent (up to ˜24-fold regulation) for all target loci (FIG. 10). However, in the transfected samples, we observed significant background editing in the dark, suggesting Cas9 inhibition to be imperfect, at least under heterogeneous expression conditions (FIG. 10). Further, mutations known to improve docking of the terminal helices against the LOV core in the dark were introduced into the L0V2 domain of CASANOVA. As expected, Cas9 background activity was reduced in several mutants, albeit at the cost of a low er dynamic range in most eases (FIGS. 11 and 12). Performance was further tuned by introducing mutations in the AcrIIA4 part of CASANOVA. These mutations were screened computationally and selected by manual inspection and several structural metrics aiming to enhance Cas9 binding affinity (FIG. 13). Two mutants (CASAMOVAT16F and CASANOVAS46D) obtained in this manner showed enhanced Cas9 inhibition in the dark without compromising light activation noticeably (FIG. 14). CASANOVA and several of its optimized mutants also conferred strong light regulation on xCas9, a recently developed PAM-relaxed, highly specific SpyCas9 derivative2 (FIG. 15).

Example 12

Transcriptional Activation

[0192] Next, we investigated whether CASANOVA would enable light-mediated regulation of d(ead)Cas9-effector fusions. To this end, we employed a previously reported dCas9 variant fused to the p300 histone acetyltransferase core domain3 and targeted the Interleukin 1 receptor antagonist (IL1RN) promoter, known to be strongly activated upon induced H3K27 acetylation, in HEK 293T cells via four different gRNAs (FIG. 16a). We titrated the transfected CASANOVA dose and incubated cells in the dark or light for 44 h before assessing IL1RN expression by quantitative RT-PCR. IL1RN transcript levels were increased up to 10-fold in the illuminated samples, indicating successful control of the dCas9-p300 epigenetic modifier (FIG. 16b).

Example 13

Labelling of DNA

[0193] Next to conditional CRISPR/Cas9-mediated cellular perturbations, we assessed CaSANOVA's potential for studying the kinetics of Cas9 DNA targeting in living cells. To this end, we performed a CRISPR labeling experiment, in which a dCas9-3xRFP fusion, a telomere-targeting gRNA and CASANOVA were co-expressed in U2OS cells (FIG. 17). Transfected cells were incubated either in the light or dark for 20 h and samples were fixed before microscopy analysis. We also included control samples expressing wild-type AcrIIA4 instead of CASANOVA or no inhibitor at all, and analyzed telomere labeling in a fully unbiased manner using an automated image analysis workflow implemented in KNIME. The CASANOVA samples showed strong telomere labeling similar to the positive control in the light (˜80% of nuclei labeling positive; ˜40% showed more than 15 dots in the nucleus), while labeling was highly diminished and comparable to the negative control in the dark (˜60% of cells labeling negative; <5% cells showed more than 15 dots in the nucleus) (FIG. 18). The no inhibitor and wild-type AcrIIA4 controls showed light-independent, strong or weak telomere labeling, respectively (FIG. 18).

[0194] Confining CRISPR/Cas9 activity in time and space is a key prerequisite for informative genome perturbation experiments. Here, we showed that anti-CRISPR proteins can be substantially modified to render Cas9 inhibition dependent on an exogenous trigger, thereby providing a blueprint for the engineering of conditional Cas9 inhibitors. CASANOVA is not only a highly important add-on to the CRISPR toolbox, but conceptually advances our ability to confer light regulation on non-enzymatic proteins.

[0195] Non-standard literature cited:

[0196] Barrangou et al. (2007), Science 315, 1709-1712.

[0197] Bhaya et al. (2011), Annu Rev Genet 45, 273-297.

[0198] Borner et al. (2013), Nucleic Acids Res 41, e199.

[0199] Brinkman (2014), Nucleic Acids Res 42, e168.

[0200] Chen et al. (2013), Cell 155, 1479-1491.

[0201] Cong et al. (2013), Science 339, 819-823.

[0202] Daglivan et al. (2013), PNAS 110: 6800

[0203] Dai et al. (2016), Mol Ther Nucleic Acids 5, e349.

[0204] de Solis et al. (2016), Front Mol Neurosci 9, 70.

[0205] Dong et al. (2017), Nature 546, 436-439.

[0206] Fleishman et al. (2011), PloS one 6, e20161.

[0207] Engler et al. (2008), PloS one 3, e3647.

[0208] Hilton et al. (2015), Nat Biotechnol 33, 510-517.

[0209] Hu, et al. (2018) Nature 556, 57-63.

[0210] Hynes et al. (2017), Nat Microbiol 2: 1374.

[0211] Jain et al (2016), Angew Chem Int Ed Engl 55, 12440-12444.

[0212] Jinek et al. (2012), Science 337, 816-821.

[0213] Jinek et al. (2013), Elife 2, e00471.

[0214] Livak et al. (2001), Methods 25, 402-408.

[0215] Maji, et al. (2017), Nat Chem Biol 13, 9-11.

[0216] Mali et al. (2013), Science 339, 823-826.

[0217] Nihongaki et al. (2015a). Nat Biotechnol 33, 755-760.

[0218] Nihongaki et al. (2015b). Chemistry & Biology 22, 169-174.

[0219] Niopek et al. (2016), Nat common 7, 10624.

[0220] Oakes et al. (2016), Nat Biotechnol 34, 646-651.

[0221] Pawluk et al. (2016), Cell 167: 1829

[0222] Platt et al. (2014), Cell 159, 440-455.

[0223] Polstein and Gersbach (2015), Nat Chem Biol 11, 198-200.

[0224] Rauch et al. (2017), Cell 168, 150.

[0225] Schmidt and Grimm (2015), Biotechnol J 10, 258-272.

[0226] Senis et al. (2014), Biotechnol J 9, 1402-1412.

[0227] Terns and Terns (2011), Curr Opin Microbiol 14, 321-327.

[0228] Wang et al. (2013), Cell 153, 910-918.

[0229] Wiedenheft et al. (2012), Nature 482, 331-338.

[0230] Xue et al. (2016), Gene Ther 23, 557-559.

[0231] Yin et al. (2014). Nat Biotcchnol 32, 551-553.

[0232] Zetsche et al. (2015), Nat Biotechnol 33, 139-142.