WO2020044039A1

WO2020044039A1 - Modified sgrnas

Info

Publication number: WO2020044039A1
Application number: PCT/GB2019/052403
Authority: WO
Inventors: Antonio GARCIA GUERRA; Andrew J. Turberfield; Matthew J. Wood
Original assignee: Oxford University Innovation Limited
Priority date: 2018-08-29
Filing date: 2019-08-28
Publication date: 2020-03-05
Also published as: GB201814041D0

Abstract

The present invention relates to an inducible sgRNA comprising a crRNA or a crRNA linked to a tracrRNA (crRNA-tracrRNA) and a Blocker Domain. The Blocker Domain is capable of hybridizing to one or more regions of the crRNA or crRNA-tracrRNA, such that the sgRNA is capable of forming at least two structural conformations: one which is capable of binding to its cognate CRISPR enzyme and thus enabling the CRISPR enzyme to bind, or bind and cleave, the targeted nucleic acid sequence, and one which is not. The invention also provides methods of using the inducible sgRNA.

Description

MODIFIED SGRNAS

The recent development of gene editing by biological system CRISP R/Cas9 (and related RNA- directed nucleases) creates the prospect of dramatic developments in medicine including gene repair and cancer therapy. For safe and effective applications, it is necessary to target and control this powerful system. There is a concern about off-target Cas9-driven DNA cleavage, which could result in disruption of essential genes. Off-target cleavage can be promoted by the prolonged expression of Cas9 in constitutive expression-based systems, such as those produced by viral integration. Organisms and/or cells that have been engineered to stably express the CRISPR/Cas9 system experience different types of undesired mutations due to the continued presence of active Cas9 complexes.

Currently, there are 4 main strategies that are being explored to reduce off-target effects:

reducing the length of time over which the active enzyme Cas9 is present in the cell (reducing residence time); using a pair of Cas9 nickases; using dCas9 fused to non-specific nucleases (Fok1); and energy-driven approaches.

Reduced residence time can be accomplished by splitting Cas9 into two inactive domains that need to come together to assemble a functional Cas9. This may be accomplished, for example, by using rapamycin-dimerization domains or light-induced dimerization between Cry2 and CIB1. Alternatively, non-functional Cas9 can be exogenously activated by small molecules, as shown with 4-hydroxytamoxifen-responsive inteins.

The double-nickase or dCas9-Fok1 methods rely on an increase in specificity that results from targeting two distinct sites simultaneously that flank a region of interest. Energy-driven approaches engineer the sgRNA or Cas9 to restrict off-target cleavage by introducing energy penalties for binding to off-target sequences, such as the truncation of 5’- ends of the sgRNA to discourage non-complementary binding or by engineering Cas9 to prevent the stabilisation of non-complementary DNA-RNA hybrids.

A completely different approach is to provide the CRISPR/Cas system with sensing modules. Such‘theranostic’ Cas9 technologies are very promising for precision medicine because they would enable the implementation of cell-specific activation, independently of the delivery system used.

At the moment, there are only a few, limited, examples of such theranostic control methods, none of which have been developed beyond simple proof of principle. For instance, by engineering the Cas9 mRNA to contain a microRNA-binding site (e.g. Hirosawa, M., et al. Cell- type-specific genome editing with a microRNA-responsive CRISPRCas9 switch. Nucleic Acids Res., 2017. 45:e118. doi: 10.1093/nar/gkx309) it is possible to arrange that Cas9 is conditionally expressed in the absence of the specific single microRNA. Whilst using endogenous signals such as the microRNAs is interesting (e.g. Xie, Z., et al. Multi-Input RNAi Based Logic Circuit for Identification of Specific Cancer Cells. Science, 2011. 333:1307- 131 1.

doi: 10.1 126/science.1205527), this system only responds to one microRNA, and cannot prevent leak translation of Cas9 even in the presence of the designed repressor microRNA.

Other attempted approaches rely on hiding away the targeting sequence of the sgRNA within a RNA secondary structure (e.g. Ferry, Q. R. V., et al. Rational design of inducible CRISPR guide RNAs for de novo assembly of transcriptional programs. Nature Communications, 8:14633 (2017)). The targeting strand can be reactivated by the external additional of chemically- modified oligonucleotides or the expression of RNA-motif-dependent nucleases. This approach would not be useful in a therapeutic setting because, to ensure efficient targeted operation, it would be necessary to ensure that the chemically-modified oligonucleotides would reach only the therapeutic site. sgRNAs have previously been modified to include additional sequences, such as sensing loops (e.g. WO2015/168404). Such additional sequences have previously been incorporated into the guide RNA at the 3' end of Stem Loop 3. This is because this loop is not located within the RNA/protein complex when Cas9 binds to the guide RNA. Therefore, extra sequences can be easily added without compromising the binding between Cas9 and the guide RNA. RNA polymerase-driven RNA synthesis (transcription) occurs in the 5’ 3’ direction. If transcription is aborted, it will be at the expense of the 3’-end of the guide RNA. Since transcription is not always perfect, some truncated transcripts will inevitably be produced. Chemical synthesis conventionally occurs in the opposite direction. If an error occurs during chemical synthesis then typically the 5’ end of a truncated product strand is missing. Guide RNAs comprise different functional domains, some of them being essential for their recognition by their cognate enzyme. For example, in the case of the nuclease Cas9, a minimal guide RNA must incorporate the crRNA and Stem Loop 1 of tracrRNA in order to be able to bind to Cas9. To ensure that all functional guide RNAs contain a blocker module, it should be located in the guide RNA before all essential functional domains for a minimal guide RNA are transcribed or synthesized.

The addition of control sequences on the 3' side of the minimal guide RNA domains, such as within Stem Loop 3, is problematic because truncated guide RNAs that lack the control sequences will be functional even if the additional signalling molecules, that the sensors are designed to recognise, are not present. Consequently, a mixture of full guide RNAs and truncated versions that do not contain the extra control sequences will have undesired activity via Cas9/guide RNA complexes. The traditional way to exclude such truncated versions has been by in vitro transcription and PAGE purification of the modified guide RNA. This ensures that only the desired, full-length transcripts are used, but it is time-consuming and restrictive as explained below.

A molecular switch designed to regulate Cas9 activity should be Off” by default to prevent any undesired Cas9 activity. As explained above, relying on the purification of the correct molecular switch is not a desirable method as, if the purification is not perfect, there could be

contamination with truncated products that are capable of inducing Cas9 activity. Another disadvantage is that, if purification of truncated product is necessary, molecular switches could not be encoded in plasmids because subsequent purification is not possible in the cell.

Consequently, the only way to use molecular switches that require purification to remove imperfect products with undesired activity is to synthesize them in vitro and deliver them to target cells with specialised RNA carriers.

Encoding molecular switches or sensors, that are to be used to control Cas9 activity, in plasmids is very desirable for a variety of reasons. Firstly, producing plasmids is cheap and, once they have been introduced in the target cell, they can constitutively express the switch. This could be advantageous because increasing the absolute amount of intracellular molecular switch enhances the probability that it will encounter the Opening Keys, i.e. the specific biomolecular triggers that will, by design, activate the engineered guide RNA and allow it to cut the target genomic sequence. Secondly, all currently available delivery systems for plasmids, from virus to commercial transfection reagents, could be used, eliminating the necessity of developing specialised sgRNA delivery systems. Therefore, for all the above mentioned reasons, it is desirable to develop a switchable sgRNA that can perform in the presence of its truncation products and whose activity can be controlled by the presence or absence of molecular signals in its environment.

The invention provides a sgRNA molecule recognisable by an RNA-driven endonuclease comprising an additional sequence, i.e. a Blocker Domain. The Blocker Domain is present in the sgRNA, close to or within the crRNA or tracrRNA domains, and is positioned such that, if it does not interact with other parts of the sgRNA, it does not significantly affect the stability of binding of the RNA molecule to its cognate CRISPR enzyme. The Blocker Domain is preferably present in a region of the sgRNA that is either joined to the 5’-end of the sgRNA or inserted in a position between the 5’-end of the sgRNA and the 3’-end of Stem Loopl of the tracrRNA domain. Placing the Blocker Domain at one of these positions ensures that all functional sgRNA transcripts, that include both crRNA and tracrRNA domains, also contain a Blocker Domain. Because all functional transcripts will contain a Blocker Domain, truncated products of transcription are not a problem because substantially inactive and therefore the sgRNA of the invention can be encoded in a plasmid and transcribed in situ.

The nucleotide sequence of one or more regions of the Blocker Domain is at least partially complementary to the nucleotide sequences of one or more regions of the crRNA or crRNA- tracrRNA. When substantially bound through base-pairing to this complementary region or regions (its Off state”), the Blocker Domain substantially prevents the binding of the sgRNA to its cognate CRISPR enzyme or the activation of the DNA-binding or nuclease function of the enzyme.

However, in the presence of one or more Opening Keys (which may, for example, be a cellular miRNA or a protein or other biomolecule) which is capable of binding to the Blocker Domain, the sgRNA forms a structure (its On state”) wherein it is capable of binding to its cognate CRISPR enzyme and forming an active sgRNA/CRISPR enzyme complex. In this way, the sgRNA of the invention is inducible and it is capable of activation in response to sensing moieties from the cellular environment or an in vivo or in vitro environment.

The invention therefore provides a programmable molecular switch for a CRISPR enzyme system that allows the enzyme to be activated only when certain conditions are met. Such a “safety device” addresses the challenges mentioned above. The switch produces a CRISPR enzyme system that is (usually)“Off’ by default, which avoids the problems associated with non-specific delivery of the gene-editing system and/or constitutive expression because the enzyme will not be active except in cells in which the pre-programmed conditions, i.e. the presence of the Opening Key or Keys, are met. For instance, if accumulation of exogenously delivered CRISPR enzyme in the liver should happen it will not cause off-target CRISPR activity if the Opening Key or Keys is not present in the liver. Similarly, if the CRISPR enzyme and its sgRNA were expressed constitutively in a wide range of cell types, off-target activity would be minimized because gene editing would only be activated in the presence of the Opening Key or Keys. By restricting the activity of the enzyme to cells in which the Opening Key or Keys is present, the number of CRISPR enzyme complexes capable of producing off-target mutations is drastically reduced.

In one embodiment, the invention provides an inducible sgRNA comprising a crRNA or a crRNA linked to a tracrRNA (crRNA-tracrRNA) and a Blocker Domain, wherein the nucleotide sequences of one or more regions of the Blocker Domain are at least partially complementary to the nucleotide sequences of one or more regions of the crRNA or crRNA-tracrRNA, such that the Blocker Domain is capable of (substantially) hybridizing to those one or more regions of the crRNA or crRNA-tracrRNA, and wherein

(i) the position of the Blocker Domain in the inducible sgRNA and

(ii) the location(s) of the one or more regions of the crRNA or crRNA-tracRNA are selected such that the sgRNA is capable of adopting at least the following two

conformations:

(A) a first conformation wherein the Blocker Domain is hybridized or substantially hybridized to the one or more regions of the crRNA or crRNA-tracrRNA,

and

(B) a second conformation wherein the Blocker Domain is not hybridized or substantially not hybridized to the one or more regions of the crRNA or crRNA-tracrRNA. Preferably, in conformation (A), the sgRNA is incapable of binding to its cognate CRISPR enzyme or otherwise incapable of activating the nucleic-acid-binding or nuclease functionalities of the cognate CRISPR enzyme.

Preferably, in conformation (B), the sgRNA is capable of binding to its cognate CRISPR enzyme and/or of activating the nucleic-acid-binding or nucleic-acid-binding and nuclease functionalities of the cognate CRISPR enzyme.

Preferably, the hybridization of the Blocker Domain to the one or more of the regions of the crRNA or crRNA-tracrRNA is capable of being disrupted by the binding of one or more Opening Keys to the Blocker Domain, thereby promoting the change in conformation of the sgRNA from (A) to (B) (and preferably thus allowing the sgRNA to form a structure wherein it is capable of binding to and activating its cognate CRISPR enzyme).

In another embodiment, the invention provides an inducible sgRNA comprising a crRNA or a crRNA linked to a tracrRNA (crRNA-tracrRNA) and a Blocker Domain, wherein the nucleotide sequences of one or more regions of the Blocker Domain are at least partially complementary to the nucleotide sequences of one or more regions of the crRNA or crRNA-tracrRNA or to an additional domain or domains inserted into the crRNA or crRNA-tracrRNA, such that the Blocker Domain is capable of (substantially) hybridizing to those one or more regions of the crRNA or crRNA-tracrRNA or to the additional domain or domains inserted into the crRNA or crRNA- tracrRNA, and wherein

(i) the position of the Blocker Domain in the inducible sgRNA and

(ii) the location(s) of the one or more regions of the crRNA or crRNA-tracRNA or the additional domain or domains inserted into the crRNA or crRNA-tracrRNA

are selected such that the sgRNA is capable of adopting at least the following two

conformations:

(A) a first conformation wherein the Blocker Domain is hybridized or substantially hybridized to the one or more regions of the crRNA or crRNA-tracrRNA or to the additional domain or domains inserted into the crRNA or crRNA-tracrRNA,

and

(B) a second conformation wherein the Blocker Domain is not hybridized or substantially not hybridized to the one or more regions of the crRNA or crRNA-tracrRNA or to the additional domain or domains inserted into the crRNA or crRNA-tracrRNA . Preferably, in conformation (A), the sgRNA is incapable of binding to its cognate CRISPR enzyme or otherwise incapable of activating the nucleic-acid-binding or nuclease functionalities of the cognate CRISPR enzyme.

Preferably, the hybridization of the Blocker Domain to the one or more of the regions of the crRNA or crRNA-tracrRNA or to the additional domain or domains inserted into the crRNA or crRNA-tracrRNA is capable of being disrupted by the binding of one or more Opening Keys to the Blocker Domain, thereby promoting the change in conformation of the sgRNA from (A) to (B) (and preferably thus allowing the sgRNA to form a structure wherein it is capable of binding to and activating its cognate CRISPR enzyme).

The invention relates to an inducible sgRNA. As used herein, the term“sgRNA” refers to a guide RNA or single guide RNA. The terms guide RNA (gRNA) and sgRNA are used interchangeably herein. Such sgRNAs are also known as CRISPR sgRNAs. CRISPR is an acronym for Clustered, Regularly Interspaced, Short, Palindromic Repeats.

In one embodiment, a sgRNA is a chimeric RNA which is formed from a crRNA and a tracrRNA which are used in the native CRISPR/Cas systems (e.g. Jinek, M. et al. (2012),“A

programmable dual-RNA-guided DNA endonuclease in adaptive bacterial immunity”, Science 337, 816-821). The sgRNA of the invention may also be called a modified sgRNA because it comprises not only a crRNA and (in some embodiments) a tracrRNA, but also a Blocker Domain, and optionally other introduced domains.

In some embodiments of the sgRNA, the crRNA and the tracrRNA are linked through a chemical bond, i.e. they are covalently-linked between contiguous RNA nucleotides.

In other embodiments of the sgRNA, the crRNA and tracrRNA are linked by hybridisation between regions (nucleotide sequences) of the crRNA and tracrRNA, i.e. they are not covalently-bonded. Such embodiments also form part of the invention. The term“crRNA- tracrRNA” therefore also encompasses embodiments wherein a crRNA and tracrRNA are hybridised together. The term sgRNA may also encompass embodiments herein a crRNA is present without a tracrRNA. In particular, the sgRNA may additionally comprise one or more additional domains inserted into the crRNA or crRNA-tracrRNA, wherein the nucleotide sequences of one or more regions of the Blocker Domain are at least partially complementary to the nucleotide sequences of one or more regions of these additional domains. As used herein, the terms“crRNA” and“crRNA-tracrRNA” encompass embodiments comprising such additional domains.

In other embodiments, the sgRNA is a crRNA such as that which is capable of binding to and activating Cpf1 , and derivatives, analogues and variants thereof.

The sgRNA may be made up of ribonucleotides A, G, C and U. Modified ribonucleotides, deoxyribonucleotides, other synthetic bases and synthetic backbone linkages (such as peptide nucleic acid (PNA), phosphorothioate linkage, locked nucleic acid (LNA), etc.) may also be used.

The crRNA comprises a targeting RNA sequence. The targeting sequence has a degree of sequence complementarity with the desired target nucleic acid sequence. Preferably, the degree of sequence complementarity between the targeting RNA sequence and the target nucleic acid sequence is at least 80%, more preferably at least 90%, 95%, 99% or 100%.

Preferably, the targeting RNA sequence is 14-30 nucleotides, more preferably 20-30 nucleotides in length.

The target nucleic acid may be DNA or RNA. Preferably, the target nucleic acid is DNA.

The target may be single- or double-stranded. The target DNA is preferably eukaryotic DNA, e.g. DNA within eukaryotic host cells.

The target DNA may be any DNA within the host cells. The target DNA may, for example, be chromosomal DNA, mitochondrial DNA, plastid DNA, plasmid DNA or vector DNA, as desired.

In some embodiments, the target nucleic acid may be a regulatory element, e.g. an enhancer, promoter or terminator sequence. In other embodiments, the target nucleic acid is an intron or exon in a polypeptide-coding sequence. In some embodiments, the target nucleic acid is a gene associated with cancer (e.g. an anti-apoptosis gene) or a gene associated with drug sensitisation (e.g. Bcl2). In some embodiments (e.g. wherein the cognate CRISPR enzyme is Cas9 or a variant thereof), the crRNA comprises a Repeat Sequence. The crRNA Repeat Sequence normally binds to the Anti-repeat Sequence in the tracrRNA. These Repeat:Anti-repeat Sequences also include the duplex bulge. The Repeat Sequence is, for example, 8-30 nucleotides in length.

In some embodiments of the invention wherein the nucleotide sequence of the Blocker Domain is at least partially complementary to one or more regions (or all) of the Anti-Repeat Sequence, it is preferable that the Repeat Sequence is modified (for example, reduced in length compared to the length of the Anti-repeat Sequence) in order to reduce the strength of the binding between the Repeat and Anti-repeat sequences thus allowing introgression of the Blocker Domain. In such embodiments, the length of the Repeat Sequence is preferably 10-15, more preferably 1 1-13 and most preferably about 12 nucleotides.

The crRNA and the tracrRNA, when present, are preferably linked by a linker, such that the crRNA and tracrRNA form a sgRNA. The linker is preferably a short (e.g. 1 -10) sequence of nucleotides.

Preferably, the crRNA and tracrRNA are linked using an RNA tetraloop. Most preferably, the tetraloop has the nucleotide sequence 5’-GAAA-3’. Other tetraloop sequences may also be used.

The tracrRNA comprises an Anti-repeat sequence. In some embodiments, the length of the Repeat sequence is less than the length of the Anti-repeat sequence in order to reduce the strength of the binding between the Repeat and Anti-repeat sequences, thus allowing introgression of the Blocker Domain. In some embodiments, defects such as bulges or base pair mismatches are present in the duplex formed between the Repeat and Anti-repeat sequences in order to achieve the same effect. In some embodiments, the length of the Anti repeat sequence is 8-30 nucleotides, more preferably more than 15, 14 or 13 nucleotides.

The tracrRNA also comprises one or more stem-loop sequences. Preferably, the tracrRNA comprises 1 , 2 or 3 stem-loop sequences; most preferably the tracrRNA comprises 2 or 3 stem- loop sequences. The stem-loop sequences may be linked by short (e.g. 1 -10 nucleotide) linkers.

Preferably, the tracrRNA comprises the genetic elements which are normally found in sgRNAs, i.e. - an Anti-repeat Sequence,

- Stem Loop 1 ,

- a linker sequence,

- Stem Loop 2 and

- Stem Loop 3.

The inducible sgRNA of the invention comprises a Blocker Domain. The Blocker Domain is a stretch or domain of RNA nucleotides which is joined contiguously to the crRNA-tracrRNA or is inserted within the sequence of the crRNA-tracrRNA.

In some embodiments, the Blocker Domain is 5-250 nucleotides in length. For example, it might be 5-10, 10-20, 20-30, 30-40, 40-50 or 50-100 or 100-200 nucleotides in length.

Preferably, the Blocker Domain is 40-50 or 50-100 nucleotides in length.

The Blocker Domain is present in (e.g. inserted in) a region of the crRNA or crRNA-tracrRNA which does not affect the ability of the crRNA or crRNA-tracrRNA to bind to its cognate CRISPR enzyme (when the Blocker Domain is not hybridized to one or more regions of the crRNA or crRNA-tracrRNA). In other words, the presence of the Blocker Domain when it is attached to or inserted in the crRNA or crRNA-tracrRNA must not prevent the sgRNA and CRISPR enzyme from forming an active sgRNA/CRISPR enzyme complex when the Blocker Domain is not hybridized to one or more regions of the crRNA or crRNA-tracrRNA (e.g. when the hybridization of the Blocker Domain to one or more regions of the crRNA or crRNA-tracrRNA has been disrupted by the Opening Key).

Preferably, the Blocker Domain is:

(a) joined to the 5’-end of the crRNA;

(b) inserted into the tetraloop (i.e. between the Repeat and Anti-repeat sequences);

(c) inserted into Stem Loop 1 ;

(d) inserted into Stem Loop 2;

(e) inserted into Stem Loop 3; or

(f) joined to the 3’-end of the tracrRNA.

It is advantageous to ensure that the Blocker Domain is included in all transcribed sgRNAs. Preferably, therefore, the closer the Blocker Domain is to the 5’-end of the sgRNA, the better. If the Blocker Domain is added after all of the essential features of the sgRNA (i.e. those that are required for the sgRNA to bind to its cognate CRISPR enzyme), then there might be functional truncation products that would be transcribed without a Blocker Domain. These truncation products would be unregulated, and, therefore, active.

Preferably, the Blocker Domain is either joined to the 5’ end of the sgRNA or inserted in a position between the 5’ end of the sgRNA and the 3’-end of Stem Loopl .

More preferably, the Blocker Domain is:

(a) joined to the 5’-end of the crRNA;

(b) inserted into the tetraloop (i.e. between the Repeat and Anti-repeat sequences); or

(c) inserted into Stem Loop 1.

If the Blocker Domain is present at the 3’-end of the sgRNA, the sgRNA would have to be synthesised or transcribed, followed by appropriate in vitro purification (e.g. by polyacrylamide gel electrophoresis) in order to ensure that all sgRNAs contain a Blocker Domain. This approach would not allow the encoding of sgRNAs within plasmids to be transcribed within the targeted cells, thus preventing the use of plasmid DNA delivery systems for their

transfection/transduction to cells.

In some embodiments, the Blocker Domain is not joined to the 5’-end or not joined to the 3’-end of the crRNA or crRNA-tracrRNA. In some embodiments, the Blocker Domain does not bind to an immediately-contiguous sequence, i.e. the Blocker Domain sequence and its complementary target sequence are not contiguous. Most preferably, the Blocker Domain is inserted in the tetraloop.

The Blocker Domain is a stretch of RNA nucleotides which is joined contiguously to the 5’- or 3’ end of the crRNA or crRNA-tracrRNA or which is inserted within the sequence of the crRNA or crRNA-tracrRNA. The Blocker Domain may comprise a plurality of regions, i.e. stretches of contiguous RNA nucleotides having different functions.

The Blocker Domain comprises one or more“feet”. As used herein, the term“feet” refers to one or more of the regions of the Blocker Domain which have nucleotide sequences which are at least partially complementary to one or more regions of the crRNA or crRNA-tracrRNA. (The singular term“foot” shall be construed similarly, mutatis mutandis). The degree of complementarity between each foot and the region of the crRNA or crRNA- tracrRNA to which it may be bound may be less than 100%, e.g. preferably 80%-100%, more preferably 90-100%.

The Blocker Domain may, for example, comprise 1 , 2, 3, 4, 5, 6, 7, 8, 9 or 10 feet. Preferably, the Blocker Domain comprises 1 , 2 or 3 feet, most preferably 2 feet. The length of each foot is preferably 3-25, more preferably 8-20 nucleotides.

One or more or all of the feet regions of the Blocker Domain are preferably flanked on one side or both sides (preferably on both sides) by“toeholds”. As used herein, the term“toeholds” refers to regions of the Blocker Domain which do not have sequences which are complementary to the regions of the crRNA or crRNA-tracrRNA which flank the regions which are

complementary to the feet regions.

Preferably, the Blocker Domain toeholds do not bind stably to any other part of the crRNA or crRNA-tracrRNA (e.g. the sequence complementarity between the Blocker Domain toeholds and any other part of the crRNA or crRNA-tracrRNA is less than 50%, more preferably less than 40%, 30% or 20%, or the Blocker Domain toeholds are too short to bind stably to

complementary regions unless an adjacent region of the Blocker Domain is also hybridized).

Preferably, the Blocker Domain comprises 1 , 2, 3, 4, 5 or 6 toeholds. In some particularly preferred embodiments, the Blocker Domain comprises 2, 3 or 4 toeholds, most preferably 3 toeholds. These Blocker Domain toeholds may (independently) consist of 1 -80 nucleotides, more preferably 5-80 nucleotides.

In some preferred embodiments, the Blocker Domain comprises first, second and third toeholds consisting of 5-10, 5-20 and 5-10 nucleotides, respectively.

In one embodiment, the Blocker Domain comprises:

(a) a first toehold,

(b) a foot, and

(c) a second toehold,

in the above order 5’-3’ .

In other preferred embodiments, the Blocker Domain comprises: (a) a first toehold, preferably 5-10 nucleotides;

(b) a first foot;

(c) a second toehold, preferably 10-30 nucleotides;

(d) a second foot; and

(e) a third toehold, preferably 5-10 nucleotides.

Most preferably, the above elements are present in the above order 5’-3’.

In other embodiments, one or more or all of the feet regions of the Blocker Domain are flanked on one side or both sides by sequences which are capable of forming stem-loops (Blocker Domain stem loops).

The Blocker Domain stem loops do not have sequences which are substantially complementary to the regions of the crRNA or crRNA-tracrRNA which flank the regions which are

complementary to the feet regions.

Preferably, the sequences of the Blocker Domain stem loops are not substantially

complementary to any other part of the crRNA or crRNA-tracrRNA (i.e. the sequence complementarity between the Blocker Domain stem loops and any other part of the crRNA or crRNA-tracrRNA is less than 50%, more preferably less than 40%, 30% or 20%).

Preferably, the Blocker Domain comprises 0, 1 , 2, 3, 4, 5 or 6 Blocker Domain stem-loops. In some particularly preferred embodiments, the Blocker Domain comprises 2, 3 or 4 Blocker Domain stem loops, most preferably 3 Blocker Domain stem loops.

These Blocker Domain stem-loops may (independently) have internally substantially complementary regions of 5-80 base pairs in length (i.e. the total number of nucleotides in the Blocker Domain stem loop region is 10-160 nucleotides).

In some preferred embodiments, the Blocker Domain comprises first, second and third Blocker Domain stem-loops having internally complementary regions of 5-10, 5-20 and 5-10 base pairs in length, respectively.

In some preferred embodiments, the Blocker Domain comprises:

(a) a first Blocker Domain stem loop, preferably 5-20 base pairs in total;

(b) a first foot; and (c) a second Blocker Domain stem loop, preferably 10-30 base pairs in total.

In other preferred embodiments, the Blocker Domain comprises:

(a) a first Blocker Domain stem loop, preferably 5-10 base pairs in total;

(b) a first foot;

(c) a second Blocker Domain stem loop, preferably 10-30 base pairs in total;

(d) a second foot; and

(e) a third Blocker Domain stem sloop, preferably 5-10 base pairs in total.

Most preferably, the above elements are present in the above order 5’-3’.

In some embodiments, the Blocker Domain is inserted into the tetraloop; and first and second feet have sequences which are complementary to regions of the Anti-repeat Sequence.

In other embodiments, the Blocker Domain is inserted into the tetraloop; and the first and second feet have sequences which are complementary to regions of Stem loop 1 , Stem loop 2, Stem loop 3 or the guide sequence.

In other embodiments, the Blocker Domain comprises:

(a) a first foot;

(b) an RNA aptamer or an anti-miRNA or anti-ncRNA; and

(c) a second foot;

or an RNA aptamer.

The inducible sgRNA of the invention can be used to“sense” the presence of the ligand of the RNA aptamer in cells.

Preferred examples of RNA aptamers and their ligands include the following:

The inducible sgRNA of the invention can be used to“sense” the presence of ncRNAs, preferably miRNAs, in cells.

The Blocking Domain may therefore comprise the sequence of an anti-Opening Key, i.e. an RNA which has a sequence which is at least partially complementary to that of the Opening Key.

The Blocking Domain may therefore comprise an anti-ncRNA (preferably an anti-miRNA), i.e. an RNA which has a sequence which is at least partially complementary to that of a selected miRNA.

Examples of such miRNAs include immature miRNA (pri-miRNA, premiRNA), and mature miRNA (duplex RNA or ssRNA). Preferred miRNA include Mature mir21 -5p, Mature mir122-5p and Mature mir127-3p.

For example, pre-miRNAs 100, and 141 are expressed in HeLa, MCF7cell lines, respectively. Additionally, miRNA 122-5p is present in hepatocytes. Therefore, these pre-miRNAs can be used to distinguish these cells.

The nucleotide sequences of one or more regions of the Blocker Domain are at least partially complementary to the nucleotide sequences of one or more regions of the crRNA or crRNA- tracrRNA. The Blocker Domain is therefore capable of hybridizing (preferably reversibly hybridizing), under appropriate conditions, to the one or more regions of the crRNA or crRNA- tracrRNA.

The position of the Blocker Domain in the inducible sgRNA and the location(s) of the one or more regions of the crRNA-tracRNA are selected such that the sgRNA is capable of adopting (preferably inducibly adopting) at least the following two conformations (A) and (B): (A) a first conformation wherein when the Blocker Domain is hybridized to the one or more regions of the crRNA-tracrRNA, the sgRNA is incapable of binding to its cognate CRISPR enzyme or incapable of activating its nucleic-acid-binding or nuclease functionalities, and

(B) a second conformation wherein when the Blocker Domain is not hybridized to the one or more regions of the crRNA-tracrRNA, the sgRNA is capable of binding to its cognate CRISPR enzyme and of activating its nucleic-acid-binding or nucleic-acid-binding and nuclease functionalities.

In other words, the Blocker Domain must be capable of preventing the sgRNA and CRISPR enzyme from forming an active sgRNA/CRISPR enzyme complex when the Blocker Domain is hybridized to the one or more regions of the crRNA-tracrRNA (when the hybridization of the Blocker Domain to the one or more regions of the crRNA-tracrRNA has not been disrupted by the Opening Key).

Preferably, the nucleotide sequences of one or more regions of the Blocker Domain (e.g. one or more of the feet of the Blocker Domain) are at least partially complementary to one or more of the following:

(a) an RNA 5’ extension to the crRNA;

(b) the crRNA guide sequence;

(c) the crRNA Repeat Sequence or duplex bulge;

(d) the tetraloop;

(e) an RNA inserted into the tetraloop;

(f) the tracrRNA Anti-repeat Sequence or duplex bulge;

(g) Stem Loop 1 ;

(h) an RNA inserted into Stem Loop 1 ;

(i) Stem Loop 2;

0 an RNA inserted into Stem Loop 2;

(k) Stem Loop 3;

(L) an RNA inserted into Stem Loop 3; or

(m) an RNA 3’ extension to the tracrRNA.

In the cases of embodiments (a), (e), (h) 0, (I) and (m), where RNA extensions or inserts are included, the extensions/inserts must not be of a size or structure such that they would prevent the sgRNA from being capable of binding to its cognate CRISPR enzyme or of activating its nucleic-acid-binding or nuclease functionalities. It will be appreciated therefore that the crRNA or crRNA-tracrRNA may comprise an additional domain or domains inserted into it (e.g. RNA inserts) to which one or more regions of the Blocker Domain (e.g. one or more of the feet of the Blocker Domain) are at least partially complementary.

More preferably, the nucleotide sequences of one or more regions of the Blocker Domain (e.g. one or more of the feet) are at least partially complementary to one or more of the following:

(b) the crRNA guide sequence;

(c) the crRNA repeat sequence or duplex bulge;

(f) the tracrRNA anti-repeat sequence or duplex bulge; or

(h) an RNA inserted into Stem Loop 1.

Most preferably, the nucleotide sequences of one or more regions of the Blocker Domain (e.g. one or more of the feet) are at least partially complementary to

(h) an RNA inserted into Stem Loop 1.

In other preferred embodiments, the Blocker Domain is inserted into the tetraloop and the nucleotide sequences of one or more regions of the Blocker Domain (e.g. one or more of the feet) are at least partially complementary to an RNA inserted into Stem Loop 1 ; or the Blocker Domain is inserted into the Stem Loop 1 and the nucleotide sequences of one or more regions of the Blocker Domain (e.g. one or more of the feet) are at least partially complementary to an RNA inserted into the tetraloop.

Preferably, the inserted RNA is shorter than the regions of the Blocker domain to which the inserted RNA is at least partially complementary.

As used herein, the term“at least partially complementary” means that the nucleotide sequences of the Blocker Domain and its target have at least 50%, 60%, 70%, preferably at least 75%, 80%, 85%, 90%, 95% or 100% sequence complementarity.

The cognate CRISPR enzyme is one which, when complexed with an inducible CRISPR sgRNA of the invention, is capable of binding or binding and cleaving a target nucleic acid (single- stranded or duplex) which comprises a nucleotide sequence which is complementary to that of the target/guide element in the sgRNA. In some embodiments, the CRISPR enzyme has nuclease, preferably endonuclease, activity. In some embodiments, the CRISPR enzyme is nuclease-deficient. Hence it is not necessary for the CRISPR enzyme to have intrinsic nuclease activity.

In some embodiments, the CRISPR enzyme is a Class 2 CRISPR system enzyme, preferably Class 2, Type II or Type V.

Examples of CRISPR enzymes which may be used in this invention include

SpCas9, FnCas9, St1 Cas9, St3Cas9, NmCas9, SaCas9, AsCpfl , LbCpfl , VQR SpCas9, EQR SpCas9, VRER SpCas9, RHA FnCas9 and KKH SaCas9 (see Komor et ai,“CRISPR-Based Technologies for the Manipulation of Eukaryotic Genomes”, Cell (2017),

http://dx.doi.Org/10.1016/j.cell.2016.10.044).

A further example is the expanded PAM SpCas9 variant (xCas9) that can recognize a broad range of PAM sequences including NG, GAA, and GAT (Hu et ai,“Evolved Cas9 variants with broad PAM compatibility and high DNA specificity”, doi: 10.1038/nature26155).

In some embodiments, the CRISPR enzyme is Cas9 or a Cas9-like polypeptide. In some embodiments, the Cas9 enzyme is derived from S. pneumoniae, S. pyogenes, or S.

thenvophilus Cas9, or a variant thereof. In some embodiments, the CRISPR enzyme is Cpfl

In some embodiments, the CRISPR enzyme possesses nuclease (preferably endonuclease) activity. In such embodiments, the CRISPR enzyme may, for example, be a wild-type Cas9 or Cpf1 , or a variant or derivative thereof which has endonuclease activity.

In other embodiments, the CRISPR enzyme is an endoribonuclease, e.g. C2c2 or Cas13b, or a variant or derivative thereof.

In other embodiments, the aim of the complex is to target functional domain(s) or cargo(es) to the desired target DNA; the aim is not to cleave the target DNA. Consequently, in such embodiments, there is no need for the CRISPR enzyme to possess any endonuclease activity. In such embodiments, it is in fact desirable that the CRISPR enzyme does not have any or any significant endonuclease activity. Preferably, the CRISPR enzyme is a catalytically-inactive or nuclease-deficient enzyme.

Preferably, the CRISPR enzyme is an enzyme which has no or substantially no endonuclease activity. Lack of nuclease activity may be assessed using a Surveyor assay to detect DNA repair events (Pinera et ai, Nature Methods (2013) 10(10):973-976). In such embodiments the CRISPR enzyme is unable to cleave dsDNA but it retains the ability to target and bind the DNA.

In some embodiments, the CRISPR enzyme has no detectable nuclease activity. The CRISPR enzyme may, for example, be one with a diminished nuclease activity or one whose nuclease activity has been inactivated.

The CRISPR enzyme may, for example, have approximately 0% of the nuclease activity of the non-mutated or wild-type Cas9 enzyme; less than 3% or less than 5% of the nuclease activity of the non-mutated or wild-type Cas9 enzyme. The parent or wild-type Cas9 enzyme may, for example, be SpCas9.

Reducing the level of nuclease activity is possible by introducing mutations into the RuvC and HNH nuclease domains of the SpCas9 and orthologs thereof. For example utilising one or more mutations in a residue selected from the group consisting of D10, E762, H840, N854, N863, or D986; and more preferably introducing one or more of the mutations selected from the group consisting DI0A, E762A, H840A, N854A, N863A or D986A. A preferred pair of mutations is DI0A with H840A; more preferred is DI0A with N863A of SpCas9 and orthologs thereof.

In some embodiments, the CRISPR enzyme is dCas9 enzyme. In some embodiments, the CRISPR enzyme is a nuclease-deficient Cpf1 (dCpfl).

In some embodiments, the inducible CRISPR sgRNA/CRISPR enzyme complex comprises one or more functional domains or cargoes which, when juxtaposed to a target nucleic acid (e.g. a target DNA), promote a desired functional activity, e.g. transcriptional activation of an associated gene. In this case, the aim of the complex is to target the functional domain(s) or cargo(es) to the desired target nucleic acid. In some embodiments, the complex may act as a programmable transcription regulator. Upon binding of the inducible CRISPR RNA to the target nucleic acid, the functional domain or cargo is placed in a spatial orientation and proximity to a target nucleic acid sequence that allows the functional domain or cargo to function in a targeted manner.

In some embodiments, one or more functional domains or cargoes are attached, directly or indirectly, to the inducible CRISPR sgRNA. In some embodiments, one or more functional domains or cargoes are attached via stem-loop RNA binding proteins (RBPs) to the CRISPR sgRNA. In other embodiments, one or more functional domains or cargoes are attached, directly or indirectly, to the CRISPR enzyme.

In some embodiments, the CRISPR sgRNA additionally comprises: one or more stem loops to which one or more stem-loop RNA binding proteins (RBPs) are capable of interacting. These stem loops are in addition to those that are formed between substantially complementary regions of the scaffold part of the sgRNA. Preferably, these one or more stem loops are positioned within the non-targeting sequence region of the sgRNA, such that the one or more stem loops do not adversely affect the ability of the non-targeting sequence region of the sgRNA to interact with the cognate CRISPR enzyme (e.g. with Cas9), or the ability of the targeting sequence to hybridize to its target DNA.

Examples of suitable stem-loop binding proteins include MS2, PP7, <3b, F2, GA, fr, JP501 , M12, R17, BZ13, JP34, JP500, KU1 , M1 1 , MX1 , TW18, VK, SP, FI, ID2, NL95, TW19, AP205, 0>Cb5, OCb8r, OCb12r, 0>Cb23r, 7s, PRR1 and com.

The CRISPR sgRNA may therefore additionally comprise one or more stem-loops which are capable of interacting with one or more of the above-mentioned stem-loop binding proteins. Preferred examples of such stem-loop RNA binding proteins include the bacteriophage MS2 coat proteins (MCPs) which bind to MS2 RNA stem loops; and the PP7 RNA-binding coat protein of the bacteriophage Pseudomonas.

Tagging of RNA stem loops with MS2 coat proteins is a technique based upon the natural interaction of the MS2 protein with a stem-loop structure from the phage genome. It has been used for biochemical purification of RNA-protein complexes and partnered to GFP for detection of RNA in living cells (see, for example, Johansson et al., (1997), "RNA recognition by the MS2 phage coat protein", Sem. Virol. 8 (3): 176-185). PP7 RNA-binding coat protein of the bacteriophage Pseudomonas binds a specific RNA sequence and secondary structure. The PP7 RNA-recognition motif is distinct from that of MS2.

The stem-loop RNA binding proteins (RBPs) may themselves be linked to or be capable of interacting with other moieties, e.g. other proteins or polypeptides.

In some embodiment, the stem-loop RNA binding proteins (RBPs) act as adaptor proteins, i.e. intermediaries, which bind both to the stem-loop RNA and to one or more other proteins or polypeptides.

Preferably, the stem-loop RNA binding proteins (RBPs) act as adaptor proteins, i.e.

intermediaries, which bind both to the stem-loop RNA and to one or more functional domains or cargoes.

In some embodiments, the stem-loop RNA binding protein forms a fusion protein with one or more functional domains or cargoes.

In other embodiments, the one or more functional domains or cargoes are attached, directly or indirectly, to the CRISPR enzyme.

In some embodiments, the one or more functional domains or cargoes are attached to the Red domain, the Rec2 domain, the HNH domain, or the PI domain of the SpCas9 protein or any ortholog corresponding to these domains.

In certain embodiments, the one or more functional domains or cargoes are attached to the Red domain at position 553 or 575; the Rec2 domain at any position of 175-306 or

replacement thereof; the HNH domain at any position of 715-901 or replacement thereof; or the PI domain at position 1 153 of the SpCas9 protein; or any ortholog corresponding to these domains.

In other embodiments, the dCas9 forms a fusion protein with one or more functional domains or cargoes.

The functional domain or cargo is generally a heterologous domain, i.e. a domain which is not naturally found in the stem-loop RNA binding protein or dCas9. ln some embodiments of the invention, at least one of the one or more functional domains or cargoes have one or more activities selected from the group consisting of methylase activity, demethylase activity, transcription activation activity, transcription repression activity, transcript release factor activity, histone modification activity, RNA cleavage activity, DNA cleavage activity, nucleic acid binding activity and base-conversion activity.

The functional domain or cargo may be an effector domain (e.g. a domain which is capable of stimulating transcription of an associated target gene). The functional domain or cargo is preferably a polypeptide or part thereof, e.g. a domain of a protein which has the desired activity. In some preferred embodiments, the functional domain has transcription activation activity, i.e. the functional domain acts as a transcriptional activator. Preferably, one or more of the functional domains is a transcriptional activator which binds to or activates a promoter, thus promoting transcription of the cognate gene. Examples of transcription factors include heat- shock transcription factors (e.g. HSF1 , VP16, VP64, p65 and MyoDI, p300).

Transcriptional repression may be achieved by blocking transcriptional initiation (e.g. by targeting the sgRNA to a promoter) or by blocking transcriptional elongation (e.g. by targeting the sgRNA to an exon). It may also be achieved by fusing a repressor domain to the CRISPR enzyme which induced heterochromatization (e.g. the KRAB domain). Examples of transcriptional repressor domains include KRAB domain, a SID domain and a SID4X domain.

As used herein, the term“cargo” includes any moiety that is to be bound to, or localized in the vicinity of, the targeted nucleic acid sequence.

In some embodiments, the function of the cargo is to label the target nucleic acid sequence such that its position may be determined by, e.g., optical or electron microscopy. Examples of such cargoes include: optical emitters such as fluorescent proteins, fluorescent aptamers and synthetic dye molecules; and electron-dense particles used as labels in electron microscopy such as gold nanoparticles.

Under physiological conditions, the inducible sgRNA is capable of adopting conformations (A) and (B). As used herein, the term“physiological conditions” includes conditions which are normally found within cells or in tissue culture media. Examples of physiological conditions include: (a) DMEM, 10% Fetal Bovine Serum, 1 % Pen/Strep and 20 mM HEPES;

(b) 100 mM NaCI, 5 mM MgCI₂, and 0.1 mM EDTA; or

(c) 1x Phosphate Buffer Saline (PBS) supplemented with 2-5mM MgCI₂.

The conformation that the sgRNA adopts (under physiological conditions) will be dependent on the presence or absence of one or more (preferably one) Opening Keys. Preferably, the hybridization of the Blocker Domain to the one or more of the regions of the crRNA-tracrRNA is capable of being disrupted by one or more Opening Keys, thereby promoting the change in conformation of the sgRNA from (A) to (B) (and preferably thus allowing the sgRNA to form a structure wherein the crRNA-tracrRNA it is capable of binding to its cognate CRISPR enzyme and of activating its nucleic acid binding and/or nuclease functionality).

The Blocker Domain may be displaced by one or more’’Opening Keys”, thus preferably turning the sgRNA On” and capable of binding and activating its cognate CRISPR enzyme.

As used herein, the term Opening Key” refers to a moiety which is capable of binding to one or more regions of the Blocker Domain, thus preventing, inhibiting or blocking the hybridization of one or more regions of the Blocker Domain to one or more of the regions of the crRNA- tracRNA. The inducible sgRNA may then adopt a structure such that it is recognisable by the cognate CRISPR enzyme.

In other embodiments, the displacement of the Blocker Domain by one or more Opening Keys turns the sgRNA“Off and renders it incapable of being bound by its cognate CRISPR enzyme.

The Opening Key is not part of the sgRNA. The Opening Key is not attached to the sgRNA. Examples of Opening Keys include nucleic acids and aptamer ligands. In some embodiments, the Opening Key is a single-stranded nucleic acid. The nucleic acid may be DNA or RNA. The nucleic acid of the Opening Key may comprise one or more stem loops and/or duplexes, with or without overhangs.

The nucleic acids of the Opening Key may, for example, be 5-100 nucleotides in length, more preferably 8-80 nucleotides.

In embodiments of the invention wherein the Opening Key is a nucleic acid, this nucleic acid initiates a strand displacement reaction, displacing the RNA Blocking Domain from its hybridization site in the sgRNA. The free sgRNA can now fold into a different secondary structure, preferably making it recognisable by the CRISPR enzyme, and thus turning On” the sgRNA/CRISPR enzyme complex.

The nucleotide sequence of the nucleic acid of the Opening Key is preferably at least partially complementary to that of one or more of the toehold sequences or one or more of the Blocker Domain stem-loop sequences.

In embodiments of the invention wherein the Blocker Domain comprises:

(a) a first toehold,

(b) a foot, and

(c) a second toehold,

in the above mentioned 5’-3’ order, the nucleotide sequence of the Opening Key comprises a first region which is complementary to the sequence of the first toehold and a second region which is complementary to the sequence of the second toehold. The first and second regions of the Opening Key may be contiguous or linked by a short (e.g. 1-10 nucleotide) linker.

In embodiments of the invention wherein the Blocker Domain comprises:

(a) a first toehold, preferably 5-10 nucleotides;

(b) a first foot;

(c) a second toehold, preferably 10-30 nucleotides;

(d) a second foot; and

(e) a third toehold, preferably 5-10 nucleotides,

two or more Opening Keys may be used.

The nucleotide sequence of the first Opening Key comprises a region which is complementary to the sequence of the first toehold and a region which is complementary to the sequence of the first part of the second toehold. These two regions may be contiguous or linked by a short (e.g. 1 -10 nucleotides) linker.

The nucleotide sequence of the second Opening Key comprises a region which is

complementary to the sequence of the second part of the second toehold and a region which is complementary to the sequence of the third toehold. These two regions may be contiguous or linked by a short (e.g. 1 -10 nucleotides) linker. In embodiments of the invention wherein the Blocker Domain comprises:

(a) a first Blocker Domain stem loop, preferably 5-10 nucleotides in total;

(b) a first foot; and

(c) a second Blocker Domain stem loop, preferably 10-30 nucleotides in total,

the nucleotide sequence of the Opening Key comprises a first region which is complementary to the sequence of at least the second strand of the first Blocker Domain stem loop and a second region which is complementary to the sequence of at least the first strand of the second Blocker Domain stem loop. The first and second regions of the Opening Key may be contiguous or linked by a short (e.g. 1 -10 nucleotides) linker.

In embodiments of the invention wherein the Blocker Domain comprises:

(a) a first Blocker Domain stem loop, preferably 5-10 nucleotides in total;

(b) a first foot;

(c) a second Blocker Domain stem loop, preferably 10-30 nucleotides in total;

(d) a second foot; and

(e) a third Blocker Domain stem sloop, preferably 5-10 nucleotides in total,

two or more Opening Keys may be used.

The nucleotide sequence of the first Opening Key has a region which is complementary to the sequence of at least the second strand of the first Blocker Domain stem loop and a region which is complementary to the sequence of at least the first strand of the second Blocker Domain stem loop. These two regions may be contiguous or linked by a short (e.g. 1 -10 nucleotides) linker.

The nucleotide sequence of the second Opening Key comprises a region which is

complementary to the sequence of at least the second strand of the second Blocker Domain stem loop and a region which is complementary to the sequence of at least the first strand of the third Blocker stem loop. These two regions may be contiguous or linked by a short (e.g. 1 -10 nucleotides) linker.

Preferably, the nucleotide sequences of the Opening Keys are not complementary to the foot sequences.

In other embodiments of the invention wherein the Blocker Domain comprises an RNA or DNA aptamer, the Opening Key is the cognate ligand for that aptamer. The ligand may, for example, be a peptide, polypeptide or (non-poly amino acid) small chemical entity; or a non- polynucleotide ligand. Examples of RNA aptamers and their cognate ligands have been given above.

In some embodiments, the Opening Key is:

(a) a non-coding RNA, or

(b) a nucleic acid-protein complex.

The term“non-coding RNA” (ncRNA) refers to RNA which does not encode a peptide, polypeptide or protein sequence (as opposed to a coding RNA which encodes mRNA), but has other cellular functions. The term“non-coding RNA” may also refer to RNA which has a function in a cell other than to code for a peptide, polypeptide or protein. The ncRNA be eukaryotic, prokaryotic, archaeal or viral origin. The ncRNA may be endogenous to the cell in question or exogenously-introduced.

Examples of ncRNA include miRNA, pre-miRNA, pri-miRNA, piRNA, IncRNA, lincRNA, sRNA, siRNA, eRNA, PARs, introns, sisRNA, lariat intronic RNA, linear introns, intron-retained transcripts, exon-intron circular RNA, hairpin RNA (hpRNA), Intron-containing hairpin RNA (ihpRNA), telomerase RNA, IRES, RepA RNA, snoRNA, scaRNA, snRNA, rRNA, tRNA, bifunctional RNA (bifRNA or cncRNA), HOTAIR RNA, HOTTIP RNA, Airn RNA, B2 RNA, circRNA, viral ncRNA, viral small RNA, viral miRNA, the ncRNA coming from Retrotransposons Transposable elements (TEs), Long interspersed nuclear elements (LINEs), Short Interspaced Nuclear Elements (SINEs)(including ALU elements), Long Terminal Repeats (LTR),

Endogenous Retroviruses (ERV), Penelope-like elements (PLE), Dictyostelium intermediate repeat sequence (DIRS). The definitions and functions of such RNAs may be found in Cech and Steitz, Cell 157, 77-94 (2014). Preferably, the non-coding RNA is miRNA, siRNA or IncRNA, most preferably miRNA.

Particularly preferred examples of miRNA include miR-21 , miR-122, miR-16, miR-206, miR-124, miR-147, miR-199, miR-146b, miR-10a, miR-155, miR-191 , miR-92a, miR-100, miR-32, miR- 200c, miR-7, miR-124, miR-223, miR-142, miR-9, miR-128a/b, miR-7, miR-375, miR-200a, miR- 142, miR-144, miR-150, miR-29a, miR-135b, miR-452, miR-490, miR-34c, miR-10b, miR-127, miR-296, miR-470, miR-145, miR-132, miR-212, miR-29b, and miRNA clusters: let-7, miR-290, miR-17/19, miR-106b/25, miR-106a/363, miR-302b/367, miR-15b/16, miR-32, miR-106b/25, miR-144/451 , miR-221/222, miR-543/374a, miR-212/132, miR-143/145, miR-23b/27b, miR- 1/133a, miR-200c/141. Preferred examples of IncRNA include Lncend , LincU, AK028326, AK141205, ES1 , ES2, ES3, MEG3/GTL2, MIRA, TUNA/MEGAMIND, Dum, lincRNA-1592, lincRNA-1552, GAS5, LINC- PINT, NBAT1 , PR-lncRNA-1 , PTENP1 , CDKN2B-AS1 , BCAR4, HOTAIR, MALAT1 , PCAT1 , SCHLAP1 , Xist, NEAT1 , 7SL RNA.

Preferred examples of viral ncRNA include VAI, VAN, l_AT, sRNA1 , sRNA2, EBER1 , EBER2, ebv-sisRNA1-2, v-snoRNA1 , HBZ, HIV-1-encoded antisense RNA. Preferred examples of ERV RNA include HERVH, TROJAN, HCP5.

In other embodiments, the Opening Key is a nucleic acid-protein complex, preferably a ncRNA- protein complex.

The majority of ncRNAs are found in the cell as RNA-protein complexes. Therefore, the Opening Key may be in the form of a ncRNA-protein complex (i.e. comprising a ncRNA and one or more protein molecules).

In some embodiments, the protein part of the RNA-protein complex is a nucleotidyltransferase, a ribonuclease, a RNA-modifying enzyme, a helicase, a GTPase or a protein having a RNA- binding domain. Subclasses within these classes may be found in Jankowsky and Harris (Nature Reviews Molecular Cell Biology, volume 16, pages 533-544 (2015)), the contents of which are explicitly incorporated by reference.

Preferably, the ncRNA in the ncRNA-protein complex is a ncRNA as mentioned above.

The ncRNA in the ncRNA-protein complex may also be a modified synthetic RNA.

In other embodiments, the Opening Key is a DNA molecule. Examples of suitable DNA molecules include promoter DNA sequences, enhancer DNA, Circulating Free DNA or Cell free DNA (cfDNA), circulating tumour DNA (ctDNA), cell-free fetal DNA (cffDNA), DNA with modifications (e.g 5-methylcytosine), DNA with biological relevant structural features (e.g specific minor groove width or specific rotational parameters such as helix and propeller twist, etc.), sequence-specific motifs for DNA binding proteins and DNA binding motifs. In other embodiments, the nucleic acid-protein complex is a DNA-protein complex. Such complexes may act as Opening Keys via complete or partial sequence complementarity between the Blocker Domain in the inducible sgRNA and the nucleic acid bound in the complex.

Preferably the DNA part of the DNA-protein complex is a DNA molecule as mentioned above. The DNA in the DNA-protein complex may be chemically-modified.

When the Opening Key is a nucleic acid-protein complex, it may be desirable to introduce specific modifications in the inducible sgRNA in order to prevent or modify a (undesirable) natural function of the complex.

For example, in embodiments wherein the Opening Key is a miRNA-Argonaute protein complex, it is preferable that the Blocker Domain is protected from Argonaute-mediated cleavage. Within a cell, miRNA is associated with Argonaute protein and hence, within a cell, the miRNA Opening Key will be associated with Argonaute protein when the miRNA binds to the Blocker Domain. If the Blocker Domain sequence is not protected, then the Argonaute protein may cleave the sgRNA (crRNA or crRNA-tracrRNA). Argonaute proteins are well known in the art (e.g. see Swarts et al., Nature Structural & Molecular Biology, volume 21 , pages 743-753 (2014), the contents of which are explicitly incorporated by reference). Argonaute proteins normally belong to the PIWI protein superfamily, defined by the presence of a PIWI (P element-induced wimpy testis) domain. Prokaryotic argonautes (pAgo) and eukaryotic argonautes (eAgo) normally have a conserved number of domains. In addition to PIWI domains, all eAgos feature an N (N- terminal) domain, a PAZ (PIWI-Argonaute-Zwille) domain and a MID (middle) domain, along with two domain linkers, L1 and L2 (see Fig. 1 , Box 2 of Swarts, ibid).

The Blocker Domain may be protected from Argonaute-mediated cleavage by, for example, introducing mismatches into the relevant regions of the Blocker Domain or by chemically- modifying the sgRNA (crRNA or crRNA-tracrRNA) to prevent Argonaute-mediated cleavage of the sgRNA (crRNA or crRNA-tracrRNA). For example, one or more mismatches may be introduced into the region of the Blocker Domain which binds to the central region of the miRNA (e.g. to nucleotides 9-13 of the miRNA).

Modifying the Blocker Domain to introduce mismatches to the miRNA sequence can accomplish different things. It can prevent binding if the seed sequence (nts g2-g8) is not complementary enough to the target. This can also affect the kinetics of binding and disassociation, affect the thermodynamics of binding, and prevent slicing from a catalytically-active Argonaute by mismatching the central region of the miRNA/siRNA molecule (nt t9-t12), despite Argonaute cutting the bond between nucleotides t10 and t11 1 ,2 (see Wee, C. et al. Argonaute Divides Its RNA Guide into Domains with Distinct Functions and RNA-Binding Properties . Cell, Volume 152, Issues 1-2, 17 January 2013, Pages 366; and Salomon, et al., Single-Molecule Imaging Reveals that Argonaute Reshapes the Binding Properties of Its Nucleic Acid Guides. Cell, Volume 162, Issue 1 , 2 July 2015, Pages 84-95). (In this context, in tX-tY,“t” refers to the target RNA molecule to which the guide RNA (miRNA or siRNA) in association with Argonaute binds; and in gX-gy, “g” refers to the guide RNA in association with Argonaute.

In the case of chemically-modified oligos, they are normally used to prevent slicing; however, they can also be used in a similar fashion as mentioned above. To prevent cleavage, a target RNA can contain a phosphorothioate linkage flanked by 2’-0-methyl ribose at positions t10 and t1 1 1 .

To determine the effects of mismatches on kinetics, single molecule experiments, similar to those discussed in Wee et al. and Salomon et al., supra, can be used. These use fluorescent single-molecule microscopy to study the effects on K_on and K_0ff using a test-tube system.

Regarding the mismatches and chemical modifications to prevent slicing, a simple gel electrophoresis assay (polyacrylamide for small nucleic acids targets and agarose for larger ones) would suffice. In Wee et al. and Salomon et al. (supra), radiolabelled RNA targets were used and gels were run at different time points to perform kinetic studies.

Alternatively, a more sophisticated approach without gels could entail a fluorescent method. In a fluorimeter, one can measure at different times points the amount of fluorescence coming from the sample. Depending of the experimental design, one could follow the increase of fluoresce as a proxy of slicing, as a result of using a target molecule containing a Quencher-fluorophore pair (the more target molecules are cut, the more fluorophore molecules will stop being in proximity of the quencher, resulting in an increase of fluorescence). Another possible approach would be to follow the decrease of Forster resonance energy transfer (FRET) when the target contains a FRET pair (FRET requires an pair of acceptor and donor fluorophores in close proximity). If the RNA target is cleaved, the pair would diffuse away from each other, producing a markedly decrease in FRET signal. As discussed above, the Blocker Domain will comprise a RNA sequence which is complementary to or substantially complementary to at least part of the Opening Key.

In other embodiments of the invention wherein the Blocker Domain comprises an anti-miRNA, the Opening Key is the cognate miRNA.

In the presence of one or more Opening Keys, the binding of the one or more Opening Keys to the Blocker Domain and the binding of the resulting complex to the cognate CRISPR enzyme must be more energetically favourable than the binding of the Blocker Domain to the one or more regions of the crRNA or crRNA-tracrRNA.

Factors that will affect this energetic favourability include the following:

(i) the degree and extent of nucleotide sequence complementarity between the Blocker Domain and the one or more regions of the crRNA or crRNA-tracrRNA to which it hybridizes;

(ii) the presence of one or more toehold sequences in the Blocker Domain;

(iii) in embodiments wherein the Opening Key is a nucleic acid, the degree and extent of nucleotide sequence complementarity between the Opening Key and the Blocker Domain;

(iv) in embodiments wherein the Opening Key is a non-polynucleotide ligand, the binding affinity of that ligand for the Blocker Domain; and

(v) the conditions under which the Opening Key is contacted with the sgRNA (e.g. pH, concentrations of sgRNA, cognate enzyme and Opening Key, composition and concentrations of other components of the cytosol or solution, etc.).

Each of the above can readily be tested by the person of average skill in this area of technology, following the guidance given herein (e.g. regarding degrees of sequence complementarity, toe holds, etc.). For example, a 30mer single-stranded polynucleotide Opening Key that has 100% sequence complementarity with the Blocker Domain should readily displace the binding of a Blocker Domain to a region of the crRNA or crRNA-tracrRNA wherein the sequence

complementarity between the Blocker Domain and the region of the crRNA or crRNA-tracrRNA is only 80% over 21 polynucleotides. sgRNA of the invention may readily be produced (e.g. using recombinant DNA technology or solid-phase synthesis) and tested against appropriate control Opening Keys (for example under the physiological conditions defined above) in order to establish the relative energetic favourabilities of different states of the sgRNAs of the invention. In another embodiment, the invention provides an inducible sgRNA comprising a crRNA or a crRNA linked to a tracrRNA (crRNA-tracrRNA), a first Blocker Domain and a second Blocker Domain, wherein the nucleotide sequences of one or more regions of the first Blocker Domain are at least partially complementary to the nucleotide sequences of one or more first regions of the crRNA or crRNA-tracrRNA, such that the first Blocker Domain is capable of (substantially) hybridizing to those one or more first regions of the crRNA or crRNA-tracrRNA, and wherein

(i) the position of the first Blocker Domain in the inducible sgRNA and

(ii) the location(s) of the one or more first regions of the crRNA or crRNA-tracRNA are selected such that the sgRNA is capable of adopting at least the following two

conformations:

(A1) a first conformation wherein the first Blocker Domain is hybridized or substantially hybridized to the one or more first regions of the crRNA or crRNA-tracrRNA, and

(B1) a second conformation wherein the first Blocker Domain is not hybridized or substantially not hybridized to the one or more first regions of the crRNA or crRNA- tracrRNA;

and wherein the nucleotide sequences of one or more regions of the second Blocker Domain are at least partially complementary to the nucleotide sequences of one or more second regions of the crRNA or crRNA-tracrRNA, such that the second Blocker Domain is capable of

(substantially) hybridizing to those one or more second regions of the crRNA or crRNA- tracrRNA, and wherein

(i) the position of the second Blocker Domain in the inducible sgRNA and

(ii) the location(s) of the one or more second regions of the crRNA or crRNA-tracrRNA are selected such that the sgRNA is capable of adopting at least the following two

conformations:

(A2) a first conformation wherein the second Blocker Domain is hybridized or substantially hybridized to the one or more second regions of the crRNA or crRNA- tracrRNA,

and

(B2) a second conformation wherein the second Blocker Domain is not hybridized or substantially not hybridized to the one or more second regions of the crRNA or crRNA- tracrRNA. Preferably, in conformations (A1)(A2), (A1)(B2) and (A2)(B1), the sgRNA is incapable of binding to its cognate CRISPR enzyme or otherwise incapable of activating the nucleic-acid-binding or nuclease functionalities of the cognate CRISPR enzyme.

Preferably, in conformation (B1)(B2), the sgRNA is capable of binding to its cognate CRISPR enzyme and/or of activating the nucleic-acid-binding or nucleic-acid-binding and nuclease functionalities of the cognate CRISPR enzyme.

Preferably, the hybridization of the first Blocker Domain to the one or more of the first regions of the crRNA or crRNA-tracrRNA is capable of being disrupted by the binding of one or more first Opening Keys to the first Blocker Domain, thereby promoting the change in conformation of the sgRNA from (A1) to (B1).

Preferably, the hybridization of the second Blocker Domain to the one or more of the second regions of the crRNA or crRNA-tracrRNA is capable of being disrupted by the binding of one or more second Opening Keys to the second Blocker Domain, thereby promoting the change in conformation of the sgRNA from (A2) to (B2).

This embodiment of the invention provides an AND gate, where only the binding of first and second Opening Keys to first and second Blocker Domains leads to the binding of the sgRNA to its cognate CRISPR enzyme and/or of activating the nucleic-acid-binding or nucleic-acid-binding and nuclease functionalities of the cognate CRISPR enzyme.

In some preferred embodiments, the inducible gRNA comprises a crRNA-tracrRNA (either covalently-attached or hybridized); the Blocker Domain is inserted into the tetraloop; the nucleotide sequences of one or more feet of the Blocker Domain are at least partially complementary to Stem Loop 1 ; the Opening Key is a miRNA, (preferably an Argonaute-miRNA complex wherein the sequence of the Blocker Domain or Stem Loop 1 is modified to prevent Argonaute-mediated cleavage); and the cognate CRISPR enzyme is spCas9 or SaCas9, or derivative thereof which is capable of recognising the inducible gRNA.

In a further preferred embodiment, the inducible gRNA comprises a crRNA; the Blocker Domain is joined to the 5’-end of the crRNA; the nucleotide sequences of one or more feet of the Blocker Domain are at least partially complementary to an RNA 3’-extension to the crRNA; the Opening Key is a miRNA (preferably an Argonaute-miRNA complex wherein the sequence of the Blocker Domain is modified to prevent Argonaute-mediated cleavage); and the cognate CRISPR enzyme is Cpf1 , or derivative thereof which is capable of recognising the inducible gRNA.

In yet a further preferred embodiment, the inducible gRNA comprises a crRNA; the Blocker Domain is joined to the 5’-end of the crRNA; the nucleotide sequence of the Blocker Domain is at least partially complementary to the targeting sequence of the crRNA; the Opening Key is a miRNA (preferably an Argonaute-miRNA complex wherein the sequence of the Blocker Domain is modified to prevent Argonaute-mediated cleavage); and the cognate CRISPR enzyme is Cpf1 , or derivative thereof which is capable of recognising the inducible gRNA.

There are many established algorithms available to assess the sequence complementarity between two nucleic acid sequences. Typically, one sequence acts as a reference sequence, to which test sequences may be compared. The sequence comparison algorithm calculates the percentage sequence complementarity for the test sequence(s) relative to the reference sequence, based on the designated program parameters. Alignment of nucleic acid or amino acid sequences for comparison may be conducted, for example, by computer-implemented algorithms (e.g. GAP, BESTFIT, FASTA or TFASTA), or BLAST and BLAST 2.0 algorithms.

Percentage nucleotide sequence identities may be obtained using the BLAST methods of alignment (Altschul et al. (1997), "Gapped BLAST and PSI-BLAST: a new generation of protein database search programs", Nucleic Acids Res. 25:3389-3402; and

http://www.ncbi.nlm.nih.gov/BLAST). Preferably the standard or default alignment parameters are used.

With regard to nucleotide sequence comparisons, MEGABLAST, discontiguous-megablast, and blastn may be used to accomplish this goal. Preferably the standard or default alignment parameters are used. MEGABLAST is specifically designed to efficiently find long alignments between very similar sequences. Discontiguous MEGABLAST may be used to find nucleotide sequences which are similar, but not identical, to the nucleic acids of the invention.

The BLAST nucleotide algorithm finds similar sequences by breaking the query into short subsequences called words. The program identifies the exact matches to the query words first (word hits). The BLAST program then extends these word hits in multiple steps to generate the final gapped alignments. In some embodiments, the BLAST nucleotide searches can be performed with the BLASTN program, score=100, wordlength=12. One of the important parameters governing the sensitivity of BLAST searches is the word size. The most important reason that blastn is more sensitive than MEGABLAST is that it uses a shorter default word size (1 1). Because of this, blastn is better than MEGABLAST at finding alignments to related nucleotide sequences from other organisms. The word size is adjustable in blastn and can be reduced from the default value to a minimum of 7 to increase search sensitivity.

A more sensitive search can be achieved by using the newly-introduced discontiguous megablast page (www.ncbi.nlm.nih.gov/Web/Newsltr/FallWinter02/blastlab.html). This page uses an algorithm which is similar to that reported by Ma et al. (Bioinformatics. 2002 Mar; 18(3): 440-5). Rather than requiring exact word matches as seeds for alignment extension, discontiguous megablast uses non-contiguous word within a longer window of template. In coding mode, the third base wobbling is taken into consideration by focusing on finding matches at the first and second codon positions while ignoring the mismatches in the third position. Searching in discontiguous MEGABLAST using the same word size is more sensitive and efficient than standard blastn using the same word size. Parameters unique for discontiguous megablast are: word size: 11 or 12; template: 16, 18, or 21 ; template type: coding (0), non coding (1), or both (2).

In some embodiments, the BLASTP 2.5.0+ algorithm may be used (such as that available from the NCBI) using the default parameters. In other embodiments, a BLAST Global Alignment program may be used (such as that available from the NCBI) using a Needleman-Wunsch alignment of two sequences with the gap costs: Existence 1 1 and Extension 1.

In a further embodiment, the invention provides a kit comprising:

(a) an inducible sgRNA of the invention; and

(b) a cognate CRISPR enzyme.

The kit may additionally comprise (c) one or more cognate Opening Keys.

The invention also provides a DNA molecule encoding an inducible sgRNA of the invention.

The DNA molecule may additionally encode a CRISPR enzyme.

The invention also provides a plasmid or vector encoding a DNA molecule of the invention. The plasmid or vector may additionally encode a CRISPR enzyme. The invention also provides a kit comprising:

(a) a DNA molecule encoding an inducible CRISPR RNA of the invention; and

(b) a DNA molecule encoding its cognate CRISPR enzyme.

The inducible sgRNA of the invention is capable of adopting at least two structural forms.

Preferably, the sgRNA of the invention is capable of inducibly- and/or reversibly-switching between at least two structural forms.

In conformation (A), the Blocker Domain is hybridized to one or more of the regions of the crRNA or crRNA-tracrRNA. The sgRNA is preferably then not capable of binding to its cognate CRISPR enzyme or not capable of activating its nucleic-acid-binding or nuclease activity. In conformation (B), the Blocker Domain is not hybridized to one or more of the regions of the crRNA or crRNA-tracrRNA. The sgRNA is preferably then capable of binding to its cognate CRISPR enzyme and of activating its nucleic-acid-binding or nucleic-acid-binding and nuclease activity.

In the presence of the Opening Key, however, the Opening Key is bound to one or more regions of the Blocker Domain (conformation B). Hence the Blocker Domain is then not hybridized to one or more of the regions of the crRNA or crRNA-tracrRNA. The inducible sgRNA may then preferably form a structure which is capable of binding to its cognate CRISPR enzyme to form an active sgRNA/CRISPR enzyme complex.

In some embodiments, therefore, the sgRNA/CRISPR enzyme complex is inducible. It can preferably be turned“on” at a desired time or in a desired environment in order to target a sgRNA/CRISPR enzyme complex to a desired target DNA. Such complexes may be used to reduce off-target effects by limiting the duration of the activity of the complex or by achieving tissue- or cell-type-specific DNA editing in model organisms or in human cells, organs or patients.

In a further embodiment, therefore, the invention provides a method of activating a

sgRNA/CRISPR enzyme complex, the method comprising the step:

contacting a composition comprising an inducible sgRNA of the invention or DNA molecule coding therefore with its cognate CRISPR-enzyme and cognate Opening Key, such that the Opening Key binds to the Blocker Domain of the sgRNA, thereby preventing the Blocker Domain from hybridizing to one or more regions of the crRNA or crRNA-tracrRNA, and thereby allowing the inducible sgRNA to form an active CRISPR enzyme complex.

The invention also provides a method of inducibly-targeting a sgRNA/CRISPR enzyme complex to a desired DNA target, the method comprising the steps:

(a) contacting a composition or cell comprising the desired DNA target with

(i) an inducible sgRNA of the invention or DNA molecule coding therefore, wherein the targeting sequence of the sgRNA has sequence complementarity with the desired DNA target sequence; and

(ii) a cognate CRISPR-enzyme; and

(b) introducing one or more Opening Keys into the composition or cell,

or activating one or more Opening Keys within the composition or cell,

such that the Opening Key(s) binds to the Blocker Domain of the sgRNA, thereby preventing the Blocker Domain from hybridizing to one or more regions of the crRNA or crRNA-tracrRNA, and thereby allowing the inducible sgRNA to form an active CRISPR enzyme complex which targets the desired DNA target.

The invention also provides a method of selectively targeting a sgRNA/CRISPR enzyme complex to a desired DNA target in a composition or cell, the method comprising the step:

(a) contacting a composition or cell comprising a desired DNA target with

(ii) a cognate CRISPR-enzyme; and

wherein if the composition or cell is one which comprises an Opening Key which is capable of binding to the Blocker Domain of the sgRNA,

the Blocker Domain will be prevented from hybridizing to one or more regions of the crRNA or crRNA-tracrRNA, and thereby allowing the inducible sgRNA to form an active CRISPR enzyme complex which targets the desired DNA target.

The invention also provides a method of determining the presence or absence of an

endogenous moiety in a composition or cell, the method comprising the steps:

(a) contacting a composition or cell with

(i) an inducible sgRNA of the invention or DNA molecule coding therefore, wherein the Blocking Domain of the sgRNA comprises a binding site for the

endogenous moiety, such that the binding of the endogenous moiety to the Blocking Domain prevents the Blocker Domain from hybridizing to one or more regions of the crRNA or crRNA-tracrRNA, and thereby allowing the inducible sgRNA to form an active CRISPR enzyme complex; and

(ii) a cognate CRISPR-enzyme;

and determining the presence or absence of an active sgRNA/CRISPR enzyme in the composition or cell,

wherein the presence of an active sgRNA/CRISPR enzyme in the composition or cell is indicative of the presence of the endogenous moiety within the composition or cell, and wherein the absence of an active sgRNA/CRISPR enzyme in the composition or cell is indicative of the absence of the endogenous moiety within the composition or cell.

The presence or absence of the active sgRNA/CRISPR enzyme in the composition or cell may be determined by any suitable means. For example, the cleavage of a reporter gene may be detected, wherein the reporter gene is in the composition or cell and the targeting sequence of the sgRNA has substantial or complete sequence complementarity with the desired DNA target sequence.

The endogenous moiety may be an Opening Key, as defined herein, preferably an endogenous ncRNA. In some preferred embodiments, the endogenous moiety is a miRNA and the Blocker Domain comprises an anti-miRNA RNA sequence. In other embodiments, the endogenous moiety is a part of a virus genome and the Blocker Domain comprises and nucleotide sequence which is complementary to that part of the viral genome. For example, the virus may be a pox virus or HIV.

The invention may also be applied to the identification of cancer cells (e.g. breast cancer cells) on the basis of the miRNA profile that those cells express. Upon the identification of a cancer cell, the activated sgRNA/CRISPR enzyme complex may then be programmed to target a particular gene (e.g. an oncogene) and/or to cleave that gene.

Therefore, in yet other embodiments, the invention provides a method of identifying potential cancer cells, the method comprising the steps:

(a) contacting a cell with

(i) an inducible sgRNA of the invention or DNA molecule coding therefore, wherein the Blocking Domain of the sgRNA comprises a binding site for an Opening Key (preferably a miRNA), whose expression is associated with cancer cells, such that the binding of the Opening Key (preferably miRNA) to the Blocking Domain prevents the Blocker Domain from hybridizing to one or more regions of the crRNA or crRNA- tracrRNA, and

thereby allowing the inducible sgRNA to form an active CRISPR enzyme complex with a cognate CRISPR enzyme; and

(ii) a cognate CRISPR-enzyme;

and determining the presence or absence of an active sgRNA/CRISPR enzyme in the cell, wherein the presence of an active sgRNA/CRISPR enzyme in the cell is indicative of the cell being a cancer cell.

The invention also provides a method of destroying a cancer cell, the method comprising the steps:

(a) contacting a cell with

(i) an inducible sgRNA of the invention or DNA molecule coding therefore,

wherein the Blocking Domain of the sgRNA comprises a binding site for an Opening Key (preferably miRNA) whose expression is associated with cancer cells, such that the binding of the Opening Key (preferably miRNA) to the Blocking Domain prevents the Blocker Domain from hybridizing to one or more regions of the crRNA or crRNA-tracrRNA, and thereby allowing the inducible sgRNA to form an active CRISPR enzyme complex with a cognate CRISPR enzyme; and

(ii) a cognate CRISPR-enzyme;

wherein if the cell expresses the Opening key (preferably miRNA) whose expression is associated with cancer cells, the Opening Key will bind to the Blocker Domain of the sgRNA, thereby preventing the Blocker Domain from hybridizing to one or more regions of the crRNA- tracrRNA, and thereby allowing the inducible sgRNA to form an active CRISPR enzyme complex, wherein the activation of the sgRNA/CRISPR complex leads to the death of the cell.

Cell death may be instigated by, for example, the cleavage of one or more essential genes. In such a case, the targeting RNA of the sgRNA will have sequence complementarity with an essential gene.

In some embodiments, one or more methods of the invention are carried out in vitro or ex vivo. In yet a further embodiment, there is provided a kit comprising:

(a) a crRNA or crRNA-tracrRNA, and

(b) a Molecular Switch comprising a Blocker Domain,

wherein the nucleotide sequences of one or more regions of the Blocker Domain are at least partially complementary to the nucleotide sequences of one or more regions of the cRNA or crRNA-tracrRNA or to an additional domain or domains inserted into the crRNA or crRNA- tracrRNA, such that the Blocker Domain is capable of hybridizing to those one or more regions of the crRNA or crRNA-tracrRNA or additional domain or domains inserted into the crRNA or crRNA-tracrRNA, wherein:

the location(s) of the one or more regions of the crRNA or crRNA-tracRNA or domains inserted into the crRNA or crRNA-tracrRNA are selected such that the crRNA or crRNA-tracrRNA is capable of adopting at least the following two conformations:

(A) a first conformation wherein the Blocker Domain is substantially hybridized to the one or more regions of the crRNA or crRNA-tracrRNA or domains inserted into the crRNA or crRNA-tracrRNA, and

(B) a second conformation wherein the Blocker Domain is substantially not hybridized to the one or more regions of the crRNA or crRNA-tracrRNA or domains inserted into the crRNA or crRNA-tracrRNA,

wherein in conformation (A), the crRNA or crRNA-tracrRNA is incapable of binding to its cognate CRISPR enzyme or otherwise incapable of activating the nucleic-acid-binding or nuclease functionalities of the enzyme; and in conformation (B), the crRNA or crRNA-tracrRNA is capable of binding to its cognate CRISPR enzyme and of activating its nucleic-acid-binding or nucleic-acid-binding and nuclease functionalities,

wherein the change in conformation from (A) to (B) is inducible by the binding of one or more Opening Keys to the Blocker Domain, and wherein the Opening Key is:

(a) a non-coding RNA, or

(b) the nucleic acid part of a protein/nucleic acid complex.

The invention also provides a method of unblocking a blocked crRNA or crRNA-tracrRNA, the method comprising contacting a crRNA or crRNA-tracrRNA which is blocked with a

Molecular Switch comprising a Blocker Domain, with an Opening Key.

As used herein, the term“Molecular Switch” refers to a single-stranded nucleic acid molecule (preferably a single-stranded RNA molecule) having a nucleotide sequence which comprises that of a Blocker Domain. The Molecular Switch is preferably 5-200 or 10-100 nucleotides in length, more preferably about15 nucleotides in length.

The disclosure of each reference set forth herein is specifically incorporated herein by reference in its entirety.

BRIEF DESCRIPTION OF THE FIGURES

All Cas9 digestion reactions presented were performed as described in the manufacturer’s website https://international.neb.com/protocols/2014/05/O1/in-vitro-diaestion-of-dna-with-cas9- nuciease-s-pyogenes-m0388

Figure 1 shows an annotated example of a standard sgRNA with the main elements identified.

Figure 2. Proof-of-concept that Cas9 activity can be controlled by reversibly modifying the secondary structure of the tracrRNA. A) Diagram showing the corresponding DNA blocker/Anti- blocker system. A minimal tracrRNA was targeted by a DNA blocker that bound to it, preventing its recognition by Cas9. In the presence of an Anti-blocker strand, which was complementary to the blocker and its toehold, the blocker was prevented from binding to the tracrRNA, which became available for hybridization to the crRNA and loading to Cas9, activating the

endonuclease. B) Cas9 reactions were prepared to demonstrate the blocker system. Lanes l and 2 show uncleaved plasmid DNA and cleaved plasmid DNA respectively. Lane 3 and 4 show Cas9 reactions missing one essential molecule for Cas9 cleavage (crRNA and tracrRNA respectively). Lane 5 shows Cas9-driven cleavage of plasmid DNA in the presence of all required elements for the RNA-driven nuclease. In lane 6, it can be seen that Cas9 cleavage is significantly reduced in the presence of the DNA blocker molecule, as evidenced by the supercoiled plasmid DNA band that indicates uncleaved target. Lane 7 shows the rescue of Cas9 activity by the addition of the antiblocker molecule, which prevents the blocker from binding to tracrRNA, resulting in Cas9-driven cleavage of the target plasmid DNA. Cas9 reactions were run in a 1 % agarose gel in 0.5*TBE buffer at 80V for 1 h 30’.

Figure 3 A) Screening for better blocking architectures and locations. Different DNA blockers target different regions of the tracrRNA. Blocker 1 : against Stem Loop 1 ; Blocker 5: against duplex bulge; Blocker 6: against Stem Loop 1 + duplex bulge. All version 2 blockers include extra nucleotides (green sequence in the diagram, bottom) that do not hybridize to the tracrRNA but are designed to add bulk to the blocker-tracrRNA complex to promote steric hindrance of the Cas9 - tracrRNA interaction. Cas9 reactions were incubated for 1 h. Blocker 6.2 generates the largest supercoiled plasmid signal, suggesting that this blocking structure is the most efficient at impeding the Cas9 - tracrRNA interaction B) Diagram showing the DNA blocker/Anti-blocker system with added internal sequence for enhanced steric hindrance.

Figure 4 A) DNA blocker adapted from Blocker 6.2 (cf. Fig. 3) to be released in the presence of pre-miRNA 122. Blockers with different strengths of interaction with the tracrRNA were tested. The numbers annotated in the‘blocker’ row indicate the lengths in nucleotides of each of the 2 feet that bind to the tracrRNA. The two gels shown are a time course that sample the same reactions at different time points, and both gels have the same lanes (top gel: reactions took place for 3h 30’, bottom gel: 24h). Lanes 4 and 6 show a strong supercoiled plasmid signal, consistent with blocking; whereas lanes 5 and 7, which contain pre-miRNA 122, show a significant decrease of supercoiled plasmid and an enrichment of linear DNA after 24h. This data suggest that these blockers can control Cas9 activity, and activate it when Opening Key premiRNA122 is present. B) Diagram showing the release mechanism for the DNA blocker with premiRNA122.

Figure 5 A) RNA blockers with different feet lengths were released by opening key pre- microRNA122 using linearised plasmid as target. The length of each of the 2 feet is indicated in the row labelled“RNA Bl.”. It can be seen by the pattern of plasmid fragments that there is no perfect blocking, as there is a significant low MW band that correlates with the cleavage produced by Cas9. There is no significant difference between samples in the presence of the blocker with and without the opening key pre-microRNA122. The mechanism is the same as in Figure 4, with the exception that the blocker is made out of RNA. White horizontal lines mark that some agarose without bands has been removed to reduce the size of the image. These data indicate RNA blockers with this particular design, in which the blocker is a separate oligonucleotide, work less well than blockers with an equivalent base sequence made of DNA.

B) Diagram showing the full tracrRNA-pre-miRNA gating mechanism.

Figure 6: Blocking efficiency of a tracrRNA-3’ Blocker domain fusion (a non-optimal

configuration) produced by in vitro transcription. This gel shows a time course, 1 h30’ and 24 h. It can be seen that this blocker could not elicit perfect blocking, as there is a double digestion band generated by Cas9 even though the opening key was absent. Two different tracrRNA-3’ Blocker domain fusion variants were tested, one with added bases between feet to increase steric hindrance (“sgRNA OStL”) and one with three internal loops (“sgRNA 3StL”). There is no difference in blocking efficiency between the two, and neither of them achieved perfect blocking. Left side of the gel 1 h30’, right side 24 h. The lanes of the gel have been re-arranged for clarity, as noted by the white vertical line. The white horizontal lines mark the removal of excess agarose without bands to reduce the size of the image.

Figure 7: Potential effects of truncation products of in vitro transcription of the modified sgRNA on blocking efficiency. A) This diagram show that if T7 RNA polymerase aborts during transcription events this results in transcripts of different lengths. B) Consequences for transcription of a tracrRNA+3’ Blocker domain fusion. Because transcription occurs from the 5’ to 3’ of the transcripts, it is very likely that the 3’ blocker domain is eliminated from the transcript, allowing unregulated Cas9 activity.

Figure 8: Presence of truncation products in the in vitro transcribed tracr+RNA-3’ blocker fusion. In this denaturing polyacrylamide gel, the left side shows results for a fusion variant that does not contain internal loops, and the right side, an intra-loop-containing fusion. In both samples the highest, brightest band is the intended full-length RNA molecule. Faster (lower) bands are truncation products. The lanes of the gel have been juxtaposed for clarity, as indicated by the white vertical line.

Figure 9: Blocking efficiency of a PAGE-purified sgRNA comprising a 3’ blocker domain, after a pre-incubation of the sgRNA fusion for 10 minutes at 37°C or a temperature ramp, heating the sample and cooling it rapidly. PAGE purification removes truncated transcription products, including sgRNAs produced without effective blocker domains. This gel shows a time course, i.e. reaction products after 1 h 30’ and 24 h. It can be seen that the blocker effectively prevents Cas9 activity at both time points. Two different modalities were tested: one with added bases between feet to increase steric hindrance and one with 3 internal loops (cf. Figure 6). There was not much difference in performance between the two. The improved performance of these blockers compared to those assessed in Figure 6 confirms the hypothesis that, in the case that the blocker domain is appended at the 3’ end of the sgRNA, truncated transcription products lacking effective blocker domains compromise performance. Left side of the gel 1 h 30’, right side 24 h. The white horizontal lines mark the removal of excess agarose without bands to reduce the size of the image.

Figure 10: Blocker domain inserted in the tetraloop. A) Designed operation of a sgRNA with a blocker domain in the tetraloop region. I) WT sgRNA interacting with Cas9 for comparison. II) Modified sgRNA with the blocker domain inserted in the tetraloop region. The two feet of the blocker domain are complementary to the tracrRNA anti repeat sequence and compete against the repeat sequence for binding to this domain. The blocker domain is designed such that it binds more stably to the anti-repeat sequence than the repeat sequence does. As a result, the blocker domain binds preferentially, forming a secondary structure that is not recognisable by Cas9. Ill) Binding of the opening keys (in this case microRNA) to the toeholds flanking the blocker’s feet result in the removal of the blocker from the anti-repeat sequence, which allows the repeat sequence to bind. This conformational change allows Cas9 to recognise the sgRNA and load it, enabling cleavage of the target duplex DNA, as shown in IV). The two opening keys may be different microRNAs or two copies of the same microRNA (as in the data shown in part B).

B) Blocking efficiencies of blockers with the Blocker Domain in the tetraloop region and with different repeat sequence lengths designed to hybridize to the anti-repeat tracrRNA domain. Gels show results after 24 h for Blockers with a) 26-nt, b) 12-nt, and c) both 1 1 -nt and 10-nt repeat regions. Only gel b shows a lane with significant blocking, and release-capacity in the presence of its Opening Key, mature microRNA 122 (lanes 5 and 6 from gel b). The white horizontal lines mark the removal of excess agarose without bands to reduce the size of the image.

Figure 1 1 : Effect of a 37°C incubation of the sgRNA fusion on blocking efficiency. Products of Cas9 digestion with sgRNAs incorporating blockers with 10- and 12-nt anti-repeat regions with or without a 10 minute pre-reaction incubation at 37°C were compared. This experiment was performed twice with newly transcribed blockers in order to assure that the results obtained are reproducible. Each gel represents a different time point, a) 2hours, b) 5 hours, and c) 17 h. All lanes were the same in all three gels. The white horizontal line indicates that the gels have been edited: agarose without bands has been removed to reduce the size of the image.

Figure 12: Logical control of Cas9 using mature microRNA as an Opening Key. In this experiment, 1-input (input mature microRNA 21 or microRNA 122) and 2-input (mature microRNA 21 and 122) logic gates were tested. Numbers in the key indicate the length of each foot in nucleotides. These gels are a time course, in which each represents a Cas9 reaction time point: a) 2 hours, b) 5 hours and c) 21.5 hours. All samples with blockers were incubated for 10 minutes at 37°C. White lines show where the gel has been edited for clarity: horizontal lines mark the removal of agarose with no bands to reduce the size of the image while keeping the plasmid bands and fragments present; vertical lines represent the rearrangement of lanes for clarity.

Figure 13: A) sgRNA containing a 2-part blocker domain. 2-Part blocker sgRNA were designed in order to eliminate competition between the blockers and the repeat sequence for binding to the anti-repeat sequence, and to simplify the blocker design. The diagram shows a WT sgRNA and the different features found in the molecule: guide sequence, repeat sequence, tetraloop, Stem loop 1 , linker sequence, Stem loop 2, and Stem loop 3. In order to ensure that all sgRNA contain a functional 2-part blocker, Part 1 is embedded in the tetraloop region, and Part 2 in the loop from Stem Loop 1. Part 1 and 2 will partially hybridize, changing the secondary structure of the sgRNA and rendering it unrecognisable by Cas9. The unpaired bases in Part 1 will act as the toehold for the opening key, which will initiate the displacement of Part 2 upon binding. The sgRNA changes conformation into a secondary structure recognisable by Cas9, once the opening key displaces Part 2.

B) Gel demonstrating that 2-part blockers can perform logical computation using microRNAs as inputs. With a single-input or Forward gate against hsa-miR 21 -5p, after 24h in the absence of opening key (miR 21-5p), Cas9 cannot digest the target plasmid. Only when miR21 -5p is added is Cas9 digestion activated. 2-input AND gates can be designed by introducing 2 consecutive sequences complementary to two different miRNAs (separated by a spacer) at the tetraloop region, and 2 sequences, in stem loop 1 , that are complementary to those introduced in the tetraloop. This AND gate against hsa-miR-21-5p AND hsa-miR-127-3p can only digest plasmid in the presence of both opening keys. Each reaction was incubated at 37 C for 24 h. Ctrl refers to undigested plasmid, +Ctrl to double-digested plasmid with Ndel and Cas9 with a WT sgRNA. - and + refer to the absence or the presence, respectively, of the specific microRNA opening key during the plasmid digestion reaction. Agarose gel electrophoresis was performed in a 2% gel in 1 X TAE, and run for 150 minutes at 80 V.

Figure 14: RNA aptamer-controlled sgRNA using 2-part blockers.

A) A RNA aptamer-controlled sgRNA can be designed by introducing a RNA aptamer in the tetraloop region (Part 1), and a sequence complementary to it in the loop from stem loop 1 (Part 2). Part 2 will hybridize to Part 1 , changing the secondary structure of the sgRNA and rendering it unrecognisable by Cas9. In the presence of the aptamer’s ligand, Part 1 will bind to the ligand, releasing Part 2. This results in a change of conformation that allows Cas9 to recognise the sgRNA. B) Each gel shows the performance of a different sgRNA controlled by RNA aptamers to:

Adenosine Monophosphate (AMP), Streptavidin, and Danofloxacin. Aptamer-controlled sgRNAs were left for 24 h in the absence (-) or the presence (+) of their respective ligands (AMP at a final concentration of 10 mM, Streptavidin at 5 mM , and Danofloxacin at 100 mM).

Figure 15: 2 Blocker performance inside cells using endogenous microRNAs as opening keys.

A) Stoplight^*, spCas9⁺ 293T cells were transfected with Lipofectamine RNAiMAX (Thermo) and Forward hsa-miR-21-5p and Forward hsa-miR122-5p gates at a final concentration of 5 nM., and left in Dulbecco’s Modified Eagle’s Medium (DMEM, Gibco) supplemented with 10% Fetal Bovine Serum (FBS, Gibco) for 48h. The medium was then exchanged to DMEM 10%, 1 % Penicillin/Streptomycin (Sigma) and cells were imaged 72 after transfection. Cells expressing eGFP, indicating Cas9 cleavage of the Stoplight reporter construct, are mostly found in the population transfected with Forward has-miR21 -5p, as this cell type expresses miR21-5p but not miR122-5p (miR122 is a hepatocyte-specific microRNA).

B) Both gates, Forward hsa-miR-21-5p and Forward hsa-miR122-5p, were tested in in vitro digestion assays, using plasmid containing the Stoplight linker region, to assess their blocking and release efficiency. Plasmid digestion reactions were incubated at 37C for 24h in the absence (-) or presence (+) of their respective opening key. Samples were loaded in a 1 % agarose gel in 1 X TAE, and run for 100 minutes at 80 V.

Figure 16: Blocking efficiency of 2-part blockers inside cells.

A) Stoplight, spCas9⁺ 293T cells were transfected with Lipofectamine RNAiMAX (Thermo) and either tightly- or loosely-blocked Forward hsa-miR-100-5p gates at a final concentration of 5 nM., and left in Dulbecco’s Modified Eagle’s Medium (DMEM, Gibco) supplemented with 10% Fetal Bovine Serum (FBS, Gibco) for 48h. The medium was then exchanged to DMEM 10%, 1 % Penicillin/Streptomycin (Sigma) and cells were imaged 72 after transfection. The opening key, miR100-5p, is expressed at very low levels in this cell line. The tightly blocked gates result in very few eGFP⁺ cells; whereas, the loosely blocked gate result in a large number of eGFP⁺ cells (false positives).

B) Both gates, tightly-blocked Forward hsa-miR-100-5p and loosely-blocked Forward hsa- miR100-5p, were tested in in vitro digestion assays, using plasmid containing the Stoplight linker region, to assess their blocking and release efficiency. Plasmid digestion reactions were incubated at 37C for 24h in the absence (-) or presence (+) of the opening key (hsa-miR 100- 5p). It can be seen that loosely blocked gates can digest plasmid in the absence of the opening key, which is consistent with the difference in reporter activity observed between cells transfected with the tightly and loosely blocked gates in part A). Samples were loaded in a 1 % agarose gel in 1 X TAE, and run for 100 minutes at 80 V.

Figure 17 Cpf1 (AsCpfl) is also susceptible to control via its modified sgRNA. A time course experiment, 2, 5, and 24h, was performed, in which a blocker module with different lengths (8,9, 10) were attached to the crRNA. No Cpf1 activity was observed after 24 hours unless the opening key miRNA 21 was added. This data confirms that the molecular switch system can be implemented to control other RNA-directed endonucleases

Figure 18: sgRNA containing a 2-part blocker domain for Staphylococcus aureus Cas9

(SaCas9).

Figure 19: miRNA 10a-5p -dependent activation of 2-part blockers inside Stoplight-·-, spCas9+ 293T cells.

Figure 20: miRNA 10b-5p -dependent activation of 2-part blockers inside Stoplight-·-, spCas9+ 293T cells.

EXAMPLES

The present invention is further illustrated by the following Examples, in which parts and percentages are by weight and degrees are Celsius, unless otherwise stated. It should be understood that these Examples, while indicating preferred embodiments of the invention, are given by way of illustration only. From the above discussion and these Examples, one skilled in the art can ascertain the essential characteristics of this invention, and without departing from the spirit and scope thereof, can make various changes and modifications of the invention to adapt it to various usages and conditions. Thus, various modifications of the invention in addition to those shown and described herein will be apparent to those skilled in the art from the foregoing description. Such modifications are also intended to fall within the scope of the appended claims.

Example 1 : Proof-of-concept: manipulation of crRNA and tracrRNA affects the activity of Cas9

Example 1 demonstrates that the activity of Cas9 can be affected by interfering with its required RNA strands, crRNA and tracrRNA. Cas9 requires the hybridization of crRNA to tracrRNA through the Repeat : Antirepeat regions. The resulting molecule is recognised by Cas9 and docked within the protein. Then, Ca9s is directed towards DNA sequences that are complementary to the crRNA guide sequence.

Some structural features of tracrRNA are essential in order for Cas9 to recognise crRNA bound to tracrRNA, or any of their variants (e.g crRNA fused to tracrRNA through a tetraloop sequence to form a single guide RNA or sgRNA). Nishimasu et al Cell 2014 showed that stripping the WT tracrRNA of some of its features yields a minimal tracrRNA with the essential features for Cas9 recognition: Anti-repeat sequence and Stem loop 1. These features are depicted in a diagram in Figure 2A.

Our control system is based in the sequence-dependent interaction of a nucleic acid strand, named Blocker, with tracrRNA (or another element of sgRNA) that interferes with its recognition by Cas9. The binding of the blocker can be reversed by a strand displacement reaction, which requires an antiblocker molecule which has a greater affinity for the blocker. The blocker is partially complementary to the target sequence to which is binding within the tracrRNA, leaving some blocker bases unpaired (a“toehold”). On the other hand, the antiblocker molecule is fully complementary to the blocker, and can use the toehold as an invasion site to initiate the displacement of the blocker and stabilise the resulting complex. Figure 2A shows how the blocker strand can bind reversibly to the tracrRNA, preventing Cas9 from recognising its cognate RNA molecule.

To demonstrate that Cas9 activity can be controlled as described above, a dual Cas9 system (crRNA and minimal tracrRNA molecules) was challenged with a DNA blocker strand capable of binding to Stem Loop 1 , and the ability of Cas9 to cleave a plasmid was measured after incubating the reactions for 1 h. In Lane 5 from figure 2B, it is shown that Cas9 is able to cleave the target, as the main band visible corresponds to linearized plasmid, indicative of enzymatic cleavage. In lane 6, it is shown that Cas9 cleavage is greatly reduced in the presence of the DNA blocker strand. Additionally, Cas9 activity can be recovered when a DNA antiblocker strand is added to blocked Cas9, as seen in lane 7, resulting in cleavage levels similar to unchallenged Cas9 (lane 5).

These results confirm that Cas9 activity can be controlled by manipulating the essential RNA molecules via strand invasion and strand displacement reactions. Example 2: Screening optimal blocking locations

In Example 2, we screened for blocking targets within a minimal tracrRNA that result in high blocking efficiency. Moreover, we also tested whether large, bulky DNA blockers could improve blocking efficiency by sterically impeding the interaction of Cas9 with tracrRNA.

Nishimasu et.al. showed that abolishing the duplex bulge formed in the repeat-antirepeat duplex (crRNA-tracrRNA) or abolishing Stem Loop 1 resulted in negligible Cas9 activity. We developed blockers 1 , 5, and 6 which targeted the anti-repeat region (preventing duplex bulge formation), Stem Loop 1 , or both, respectively. Versions 1.2, 5.2, and 6.2 were bulkier DNA blockers, which comprised an unpaired loop sequence flanked by two feet which targeted the same regions within the tracrRNA, as depicted in Figure 3B.

Results presented in Figure 3A show that the optimal region to target is the duplex bulge by means of a blocker binding to the anti-repeat (blocker 1), and that bulky blockers result in enhanced blocking. Lanes 4-6 show the effects of blockers 1 , 5, and 6 respectively. The intensity of the supercoiled plasmid DNA band (uncut) indicates the blocking efficiency of each blocker. Lane 4 (blocker 1) contains the most intense supercoiled plasmid DNA band (indicating effective blocking) among the 3 blockers. The performance of all blockers (1 , 5, and 6) was improved by adding extra bases to add bulk to promote steric hindrance, as seen in lanes 7-9 (blockers 1.2, 5.2, and 6.2).

Therefore, targeting the anti-repeat region to prevent the formation of the duplex bulge is a preferred strategy when using blockers complementary to naturally occurring sequences within tracrRNA. Furthermore, increasing the steric hindrance of blockers can improve blocking efficiency.

Example 3: Gating tests with best performers

In Example 3, we studied the performance of DNA blockers designed to respond to endogenous cellular signals, in this case premature microRNAs (pre-miRNA).

MicroRNAs (miRNA) are non-coding RNA molecules that control genetic expression in eukaryotic cells. Different cell types have different microRNA expression profiles, and these profiles also change due to disease. Therefore, miRNAs can be used to define a particular cell type or tissue (e.g. miRNA 122 is liver specific). Pre-miRNA are mismatched RNA hairpins which are shuttled from the nucleus to the cytosol to continue their maturation. In order to use pre-miRNA as antiblockers, the DNA blockers need to open the pre-miRNA hairpin to hybridize to the bases in the loop. To meet this requirement, the DNA blocker consists of a toehold and a loop complementary to the pre-miRNA, and 2 feet that bind to the anti-repeat region of the sgRNA, as depicted in Figure 4B. The blocker is comprising, from 5’ to 3’: a foot, the loop, the second foot, and the toehold. The toehold allows the docking of the pre-miRNA through its unpaired loop and binds to one arm of the pre-miRNA stem, leaving the other stem arm unpaired and available for binding to the DNA blocker’s loop, promoting the melting of the blocker off the tracrRNA. (The blocker is therefore itself a hairpin: as described, the‘toehold’ is partially complementary to the‘loop’; this secondary structure is disrupted in both the blocker- tracrRNA and blocker-pre-miRNA complexes.)

Figure 4A shows a gel in which we tested the ability of these DNA blockers to prevent Cas9 cleavage, and their response to the opening key or antiblocker, pre-microRNA 122. The gel is split in 2, representing reactions that were incubated for 3h30’(top) and 24h(bottom). Each blocker had a different foot length, as seen in the numbers listed on the gel legend (6, 7, 8, 9), which had a substantial impact in the blocker’s ability to melt off the tracrRNA. Lanes 4-9 show the reactions that used blockers with foot lengths 6, 7, and 8 nucleotides, where it can be seen that pre-miRNA 122 promotes blocker displacement, and the unblocking efficiency is greater for shorter foot lengths. Cas9 digestion is only detected when pre-miRNA 122 is added: the supercoiled plasmid band starts to fade, and the linearized plasmid band enriches

proportionally. Lanes 10 and 11 show the performance of the DNA blocker with foot length 9 nt, which is not displaced from tracrRNA by pre-miRNA 122.

These results validated that DNA blockers can be designed to respond to cellular endogenous signals (e.g. pre-miRNA 122). They also highlighted the importance of foot length for successful blocking and release, and the slow kinetics of RNA hairpin-triggered release of the blocker strand.

Example 4: sgRNA transcription or chemical synthesis result in heterogeneous populations containing truncated products.

In the previous examples, we showed the performance of an intermolecular system comprising separate blocker and sgRNA molecules. For such a system, the concentrations of the interacting molecules are important parameters affecting blocking efficiency. We therefore decided to include the blocker molecule as a fusion within the sgRNA. This results in local concentrations of the interacting blocker and target sgRNA domains that are several orders of magnitude higher than in the intermolecular system.

Introducing the blocker within the sgRNA without affecting Cas9 recognition is only possible in some regions of the RNA molecule. Additionally, the location of the blocker is crucial to its performance because of the RNA synthesis methods available: transcription and chemical synthesis. Transcription naturally occurs from 5’ to 3’. However, it is common during

transcription that the RNA polymerase aborts. As a result, a variety of 3’ truncated transcripts are created. On the other hands, chemical synthesis of RNA oligomers typically occurs in the 3’- 5’ direction, and the truncated products generated in the process may lose portions of the 5’ end of the sgRNA. In the particular case of the molecular switch being built, this is problematic.

Figure 7 shows why the location of the blocker domain within the sgRNA is important. The most obvious location for the blocker domain is at the 3’ end of the sgRNA. However, transcription truncation products have shorter 3’ ends relative to the full length product, resulting in the loss of the blocker domain. This may be overcome by purification of full length sgRNA, but this is not possible in applications in which the sgRNA is transcribed within a cell. Therefore, blocker location is vital for the performance of the blocker domain-containing sgRNA or molecular switches.

Figure 6 shows the inability of in vitro transcribed sgRNA containing blocking domains at the 3’ end to block Cas9 activity. sgRNAs were transcribed with a 3’ linker region followed by a blocker domain. The blocker domain comprised, from 5’ to 3’: a toehold, a foot, a loop (OStL refers to an unpaired RNA loop, and 3Stl_ to 3 internal stem loops flanked by the feet), a second foot, and a second toehold. Assisted by the linker region, the blocker domain is designed to bind the anti repeat region and prevent its recognition by Cas9. However, the sgRNA was produced through in vitro transcription without size purification: as a result, truncated sgRNA missing their 3’ blocking domains were present in solution. Consequently, none of the blockers in Figure 6 show efficient blocking. These Cas9 reactions were plasmid double digestions, allowing the detection of Cas9 activity by the appearance of a faster band in the gel as a result of cleavage of a fragment from the plasmid. Fast bands in lanes 4-6 from the 1 h30 and 24h portions of the gel (representing the time that each reaction was incubated) indicate Cas9 cleavage and imperfect blocking. This gel confirmed that the location and the method of production have a major impact on the performance of the modified sgRNA (molecular switches).

To support the results obtained in Figure 6, we run a Denaturing Polyacrylamide Gel (PAGE) (Figure 8) with samples of the 3’ end blocker-containing sgRNAs produced by in vitro transcription. Denaturing PAGE is necessary to break all the secondary structure within the sgRNAs allowing effective separation of sgRNAs according to molecular weight. Low-molecular- weight bands, corresponding to truncated transcription products of both sgRNAs, are evident.

Example 5: PAGE purification and effect of temperature

In order to obtain full-length molecular switches, one approach is to purify them using denaturing PAGE and excision of the desired band.

Figure 9 shows that the use of PAGE-purified molecular switches showed improved blocking compared to unpurified switches (Figure 6). In this set of double digestion reactions, there are almost no fast bands produced in the presence of 3’ end blocker domain-containing sgRNAs. The 3StL blockers performed better than OStL with which some Cas9 cleavage could be observed.

Additionally, we tested the effect of different temperature histories of the sgRNA on blocking efficiency (10 minutes at 37 C or a temperature ramp). We hypothesized that appropriate folding of the RNA is important in determining blocking efficiency, and that thermal history affects folding. Figure 9 shows that there is no significant difference between the two methods of preparation in this case, but we will return to this point in other Examples.

Example 6: Insertion of Blocker Domain into the tetraloop

In order to avoid the issues associated with incomplete transcription of 3’-blocker fusions, we explored alternative blocker domain insertion sites. We concluded that the tetraloop region (the 4-nucleotide loop used to fuse the crRNA to the tracrRNA) is a preferred location.

As explained before, sgRNAs have a set of essential features: crRNA, anti-repeat sequence, and stem loop 1. In order to ensure that each sgRNA contains an effective blocker module, its location within the sgRNA must be such that the blocker module is synthesized before all essential features. That way, if a sgRNA molecule is missing the blocker module, it will be missing other essential features, resulting in a non-functional molecule. Consequently, we decided to introduce the blocker module in the tetraloop region which lies between the crRNA’s repeat sequence and the tracrRNA’s anti-repeat sequence. Additionally, it can be seen in the crystal structure of Cas9 complexed to its cognate sgRNA that the tetraloop region protrudes from the protein, far from the protein and its target DNA. This is a desirable trait for the insertion site because the disruption of the blocker domain to Cas9 would be minimal.

Figure 10 A shows the designed operation of a sgRNA with a blocker domain in the tetraloop region and Figure 10 B shows the results of initial experiments with this architecture. Repeat sequences with different lengths were tested: only the sgRNA with 12 nt repeat sequence showed a blocking improvement when compared to the positive control, as indicated by a dimmer fragment band. We therefore concluded that a preferred length for the repeat sequence domain of a blocker inserted in the tetraloop is 12 nt.

Example 7: Effect of incubation conditions on blocking efficiency

After determining the optimal repeat sequence length, we explored the effect of temperature in the molecular switch blocking.

Figure 1 1 shows the effect of incubating the molecular switches (sgRNAs with the blocker domain embedded in the tetraloop region) in their in vitro digestion buffer (provided by the manufacturer) at 37C. Switches incubated at 37C for 10 minutes performed better than the same sgRNAs that were added to the in vitro digestion reaction at the same time as all other reaction components. The differences are slight and only noticeable after 17h incubation, which suggests that the effect of temperature history on the folding and efficiency of the modified sgRNA may be modest.

Additionally, Figure 1 1 compares again (cf. Fig. 10) the performance of molecular switches with repeat sequences of 10 vs 12 nt. It confirms the effect seen in Fig. 10: the molecular switch with 12 nt repeat sequence shows better blocking, represented by the very faint fragment bands in the absence of the opening key (e.g. miR122).

We concluded that incubating the molecular switches at 37C slightly improves their blocking performance. Example 8: Use of miRNA as logic gates

We studied the ability of molecular switches with tetraloop blocker domains to perform logic computations using opening keys as input signals. This is accomplished by designing the toehold sequences to respond to only one opening key (single input or Forward logic) or to several (for instance 2-input AND logic).

Single-input or Forward gates are obtained when the toehold sequences flanking both feet respond to the same input; AND gates are produced when different toeholds respond to different inputs. In Figure 12, we demonstrate miR21 and mir122 Forward gates, and a miR21 - miR122 AND gate.

Forward gates respond as intended. Forward miR21 showed perfect blocking after 24h, and only induces Cas9 activity if the opening key, miR21 , is present. Forward miR122 is leakier than Forward miR21 suggesting that the sequence design of this gate is not optimal and that the designed inactive configuration of the sgRNA is destabilized by unanticipated secondary structure.

Although blocking was not complete, the miR21 -miR122 AND gate showed the designed logical behaviour: maximal Cas9 activity was obtained only when both opening keys were present.

We conclude that it is possible to design molecular switches that require two opening keys, but that interaction between toehold sequences, that are constrained by the sequence of an oligonucleotide opening key, and other domains of the sgRNA may compromise blocking.

Example 9: 2-Part blocker domains, and logic computations

As described in the previous Examples, the performance of the blocker domain can be affected by the sequences of the toeholds which, in the case of nucleic acid opening keys, are constrained by the base sequence of the opening key. Additionally, the competition between the repeat sequence and the blocker for the anti-repeat region affects the blocking efficiency. To address these challenges, we updated the blocker domain design.

This new iteration of the blocker comprises 2 parts: Part 1 consists of a sequence (equivalent to the foot and toehold from the previous design) which binds to the opening key and is located in the tetraloop region. Part 2 consists of a sequence which binds to Part 1 in competition with the opening key (if the opening key is a nucleic acid, Part 2 may be identical in sequence to the opening key) and is located in the loop of Stem Loop 1. Part 2 is designed such that the opening key binds more stably to Part 1 than does Part 2. Here, Part 2 is shorter than Part 1 , which results in partial hybridization to Part 1. This leaves several bases unpaired in Part 1 which can act as a toehold for the binding of the opening key. Figure 13 A shows the design. In the absence of the opening key, Part 1 hybridizes to Part 2, resulting in a secondary structure that is not recognisable by Cas9. Alternatively, when the opening key binds to Part 1 , displacing Part 2, it allows the sgRNA to adopt its native secondary structure.

Figure 13 B. shows the performance of the 2-part molecular switches and their abilities to perform logical computations. In the case of single input or forward gates, it can be seen that there is no plasmid digestion in the absence of the opening key (miR21 -5p): the blocking efficiency is superior to previous Examples. Only in the presence of miR21-5p is the plasmid digested by Cas9. 2 -input gates, or AND gates, also perform as intended. In the absence of both keys (miR21-5p and miR127-3p) or in the presence of only one key there is no plasmid digestion. The AND gate requires both opening keys to be present to change the conformation of the sgRNA and produce Cas9-driven plasmid digestion. The design of the AND gate is essentially the same as for the Forward gate, except that it has 2 consecutive Part 1 blocker domains in the tetraloop region (separated by a small spacer region), and 2 consecutive Part 2 blocker domains within the Stem Loop 1 (separated by a small spacer sequence).

We conclude that the blocker domain can be split into 2 parts and introduced in 2 separate regions within the sgRNA. This results in improved blocking efficiency.

Example 10: Use of RNA aptamers as parts of the 2-part blocker domain design

Using the 2-part blocker approach, it is possible to control sgRNAs using RNA aptamers, resulting in Aptamer-controlled CRISPR/Cas9 systems.

Figure 14 A shows how the 2-part blocker approach can be adapted. Part 1 comprises an RNA aptamer, which is included in the tetraloop region to minimise disruption of sgRNA loading into Cas9. Part 2 is partially complementary to the aptamer, such that when it hybridizes to the aptamer it prevents recognition of the sgRNA by Cas9. In the presence of the aptamer’s ligand, the ligand and Part 2 compete for binding to Part 1. If Part 2 is displaced by competitive binding of the ligand, the resulting conformation of the sgRNA is recognisable by Cas9. Figure 14 B shows the performance of 2-Part molecular switches containing 3 different aptamers against Adenosine Monophosphate (AMP), Streptavidin, and Danofloxacin. For each of the three RNA aptamer-controlled molecular switches there is a significant difference in the fraction of digested plasmid between the reactions in the absence (-) or presence (+) of the ligand. Of the 3 switches, the best performance is achieved by the Danofloxacin-triggered system. This is most likely to correlate with the effect of the (typically) complex secondary structure of the aptamer on the designed interactions between Part 1 and Part 2 and the aptamer and Part 1.

We conclude that the 2-part molecular switch architecture can be harnessed to design aptamer- controlled molecular switches.

Example 11 : Performance of 2-part molecular switches in cells

We tested activation of the 2-part molecular switches within cells using endogenous miRNA as the opening key. We used 293T cells stably expressing a Stoplight reporter construct and spCas9. The Stoplight construct is a gain-of-function reporter that comprises a constitutively expressed red fluorescent protein (e.g. mCherry), followed by a linker region and 2 out-of-frame green fluorescent proteins (e.g. eGFP). If Cas9 is directed to the linker region and produces a double strand break, DNA repair mechanisms will be activated; Non-Homologous End Joining (NHEJ) is the most probable repair mechanism. NHEJ produces insertion and deletion (indel) mutations, which can affect the frame in which genes are read. If the frame shift generated in response to Cas9 cleavage is +1 or +2, one of the normally out-of-frame eGFP will be rescued and the cell will become eGFP⁺. The number of eGFP⁺ cells therefore indicates the number of Cas9-induced double strand breaks. Since the Stoplight cells also express Cas9, we only have to transfect the molecular switches; this may be achieved using commercially available transfection reagents. If the opening key is already present inside the cell, the molecular switch will be activated, resulting in Cas9 digestion of the Stoplight linker and eGFP production. On the other hand, if the opening key is not expressed by the cell, the molecular switch will remain blocked, and the cells will not express eGFP.

We produced a Forward gate against miR21-5p, a microRNA expressed in many tissues and highly expressed in 293T cells, and another Forward gate against miR122-5p, a hepatocyte- specific microRNA. These molecular switches were used to transfect 293T cells incorporating the Stoplight system. Since 293T cells are kidney cells, only the forward miR21 -5p gate should produce eGFP⁺ cells. Figure 15 shows the result of transfecting the Forward miR21 and the Forward miR 122 into Stoplight spCas9⁺ 293T cells. It can be seen that eGFP⁺ cells are only present in the sample that was transfected with the Forward miR 21 gate. The Forward miR 122 gate produces only a very low background (2 cells in the image shown). Figure 15 also shows a gel in which the blocking and release efficiencies of both gates, Forward miR 21 and miR 122, are compared using as target a plasmid containing the Stoplight linker region. It is clear that both gates only produce plasmid digestion in the presence of their respective opening keys, which supports the cell data described above.

We conclude that molecular switches can perform as designed in the crowded cellular cytoplasm using endogenous signals, such as microRNAs, as opening keys. Activation is specific to the designed opening key. The engineered sgRNA can be loaded onto Cas9 inside cells, a context very different from the test tube plasmid digestion reactions showed before.

Example 12: Effect of blocking efficiency within cells

In this Example, we tested the effect of blocking efficiency in the production of eGFP⁺cells.

Two Forward miR 100-5p gates were developed, one that blocks tightly and a second one that is leaky. The blocking efficiency of both gates was tested. Gel data presented in Figure 16 B allows the performance of the tight and leaky Forward miR 100-5p gates to be compared. After 24 hours, the leaky gate shows some plasmid digestion in the absence of the opening key whereas the tight gate does not.

Both gates were transfected into 293T Stoplight spCas9⁺ cells. miR100-5p is expressed at a very low level in this cell type: the number of eGFP⁺ cells created using the tight gate is therefore expected to be low. Figure 16 A shows the results. The tightly blocked Forward miR 100 gate produces few eGFP⁺ cells, whereas the leaky gate produces a larger number, as expected. These results indicated that the blocking efficiency of each gate must be validated, as leaky gates are capable of activating Cas9-induced cleavage of the target DNA even in the absence of the opening key.

We conclude that our forward gates can be recognised by Cas9 and activate it to produce double strand breaks, and that our blocking mechanism can prevent Cas9 activity in the absence of the designed opening key. The efficiency of the blocking mechanism should be tested in each case.

Example 13: Use of Cpf1 as a CRISPR enzyme

Our technology is based on RNA engineering rather than protein engineering. It may be used to control RNA-directed endonucleases other than Cas9. We built a molecular switch for Cpf1 to test the universality of the molecular switch technology.

The ideal location to introduce a blocker domain into the cognate RNA of a RNA-directed endonuclease is in a position that ensures that the blocker domain is synthesized (transcribed) before all essential domains. Cpf1 uses a small RNA that contains the targeting sequence (crRNA) and other structural elements (referred to as handle). The polarity of the molecule is inverted relative to that of spCas9: the targeting sequence is found at the 3’ end.

A Blocker Domain was added to the 5’-end of a Cpf1 crRNA (retaining the 5’- handle). The Blocker Domain comprised two feet with nucleotide sequences complementary to the targeting strand of the crRNA (rather than the anti-repeat sequence used in the design of sgRNAs for Cas9). Each foot was flanked by 2 toeholds complementary to miRNA21.

In Figure 17, it can be seen that the molecular switch concept is applicable across RNA-driven endonucleases. Figure 17 shows logical control of AsCpfl with a 1 input (mature microRNA21) gate. The gels show that AsCpfl is also susceptible to control via its sgRNA. A time course experiment, 2, 5, and 24h, was performed, in which a blocker module with different lengths (8,

9, 10) was attached to the crRNA. No Cpf1 activity was observed after 24 hours unless the Opening Key miRNA 21 was added. This data confirms that the molecular switch system can be implemented to control other RNA-directed endonucleases.

Example 14: sgRNA containing a 2-part blocker domain for Staphylococcus aureus Cas9 (SaCas9)

The same designed principles for spCas9 can be used to designed switchable modified sgRNAs for SaCas9. The interest behind using SaCas9 is its small size (useful for in vivo applications) and its different PAM requirements.

The gel in Figure 18 shows how 2-part blockers can perform logical computation using microRNAs as inputs for SaCas9. A single input or Forward gate against hsa-miR 10a-5p showed that, after 24h in the absence of Opening Key (miR 10a-5p), SaCas9 had reduced activity, shown by partial digestion of the target plasmid. Only when miR 10a-5p was added, SaCas9 digestion was fully activated. The reaction was incubated at 37 C for 24 h.“No MiR” and“MiR” refer to the absence or the presence, respectively, of the specific microRNA Opening Key during the plasmid digestion reaction. AGE was performed in a 1 % gel in 1 X TAE, and run for 45 minutes at 1 10V.

Example 15: miRNA 10a-5p -dependent activation of 2-part blockers inside Stoplight+, spCas9+ 293T cells

sgRNA containing Blocker Domains which are responsive to miRNAs require some additional design considerations to work. sgRNAs need to include modifications in the sequence to prevent Argonaute-miRNA complex-mediated cleavage: chemical backbone modifications; and/or mismatches compared to the miRNA sequence. Without these modifications, the sgRNAs cannot be activated using miRNAs.

Stoplight-·-, spCas9+ 293T cells were transfected with Lipofectamine RNAiMAX (Thermo) with Forward hsa-miR-100-5p and Modified Mismatched (MM) has-miR-10a-5p gates at a final concentration of 50 nM and left in Dulbecco’s Modified Eagle’s Medium (DMEM, Gibco) supplemented with 10% Fetal Bovine Serum (FBS, Gibco) for 72h. Cells were imaged 72 after transfection. Only eGFP+ cells were observed when the sgRNA included the design considerations to prevent Argonaute-miRNA complex-mediated cleavage. Scale bar is 400 pm.

Example 16: miRNA 10a-5p -dependent activation of 2-part blockers inside Stoplight+, spCas9+ 293T cells

Features of a model sgRNA molecule

A

Target

3' CCTTTAATCCACGCGAACCGACC 5' Target DNA ( SEQ ID NO: 1)

I I I I I I I I I I I I I I I I I I I I Repeat

21 26 29 32

5 GGAAAUUAGGUGCGCUUGGCGUUUUAGA—GCUAG (SEQ ID NO: 2)

1 10 20 A

Sequences for Figures

Table A.l: Insert for pCRISPR and target sequences for Cas9 and AsCpfl

Table A.2: Sequences from Figure 2B

Table A.3: Sequences from Figure 3

Table A.4: Sequences from Figure 4

Table A.5: Sequences from Figure 5

Table A.6: Sequences from Figures 6, 8, and 9

Table A.7: Sequences from Figure 10

Table A.8: Sequences from Figure 11

Table A.9: Sequences from Figure 12

Table A.14: Sequences from Figure 14

Claims

1 . An inducible sgRNA comprising a crRNA or a crRNA linked to a tracrRNA (crRNA- tracrRNA) and a Blocker Domain, wherein the nucleotide sequences of one or more regions of the Blocker Domain are at least partially complementary to the nucleotide sequences of one or more regions of the crRNA or crRNA-tracrRNA or to an additional domain or domains inserted into the crRNA or crRNA-tracrRNA such that the Blocker Domain is capable of hybridizing to those one or more regions of the crRNA or crRNA-tracrRNA or additional domain or domains inserted into the crRNA or crRNA-tracrRNA, and wherein:

(i) the position of the Blocker Domain in the inducible sgRNA, and

(ii) the location(s) of the one or more regions of the crRNA or crRNA-tracRNA or domains inserted into the crRNA or crRNA-tracrRNA

conformations:

wherein in conformation (A), the sgRNA is incapable of binding to its cognate CRISPR enzyme or otherwise incapable of activating the nucleic-acid-binding or nuclease functionalities of the enzyme; and in conformation (B), the sgRNA is capable of binding to its cognate CRISPR enzyme and of activating its nucleic-acid-binding or nucleic-acid-binding and nuclease functionalities,

(a) a non-coding RNA, or

(b) a nucleic acid-protein complex.

2. An inducible sgRNA as claimed in claim 1 , wherein the Opening Key is a miRNA.

3. An inducible sgRNA as claimed in claim 1 , wherein the Opening Key is a non-coding RNA-protein complex.

4. An inducible sgRNA as claimed in any one of the preceding claims, wherein the sgRNA comprises a Repeat Sequence and an Anti-repeat Sequence, and wherein the length of the Repeat Sequence is less than the length of the Anti-repeat Sequence.

5. An inducible sgRNA as claimed in any one of the preceding claims, wherein the sgRNA comprises a Repeat Sequence and an Anti-repeat Sequence, and wherein the duplex formed by hybridization of the Repeat Sequence to the Anti-repeat Sequence is destabilized by the presence of defects such as mismatched base pairs or bulges.

6. An inducible sgRNA as claimed in any one of the preceding claims, wherein sgRNA comprises a crRNA linked via a tetraloop to a tracrRNA.

7. An inducible sgRNA as claimed in any one of the preceding claims, wherein the Blocker Domain is 5-250 nucleotides in length, preferably 5-10, 10-20, 20-30, 30-40, 40-50 or 50-100 or 100-200 nucleotides in length, and more preferably 40-50 or 50-100 nucleotides in length.

8. An inducible sgRNA as claimed in any one of the preceding claims, wherein the Blocker Domain is:

(a) joined to the 5’-end of the crRNA;

(b) inserted into the tetraloop (between the Repeat and Anti-repeat sequences);

(c) inserted into Stem Loop 1 ;

(d) inserted into Stem Loop 2;

(e) inserted into Stem Loop 3; or

(f) joined to the 3’-end of the tracrRNA,

preferably, wherein the Blocker Domain is

(a) joined to the 5’-end of the crRNA;

(b) inserted into the tetraloop (between the Repeat and Anti-repeat sequences); or

(c) inserted into Stem Loop 1 ;

and most preferably wherein the Blocker Domain is inserted into the tetraloop.

9. An inducible sgRNA as claimed in any one of the preceding claims, wherein the Blocker Domain comprises one or more feet (preferably 1 , 2 or 3 feet), wherein the nucleotide sequence of each foot is at least partially complementary to one or more regions of the crRNA or crRNA- tracrRNA.

10. An inducible sgRNA as claimed in claim 9, wherein each foot is flanked on one or both sides by toeholds.

1 1. An inducible sgRNA as claimed in claim 10, wherein the Blocker Domain comprises:

(a) a first toehold, preferably 5-10 nucleotides;

(b) a first foot;

(c) a second toehold, preferably 10-30 nucleotides;

(d) a second foot; and

(e) a third toehold, preferably 5-10 nucleotides.

12. An inducible sgRNA as claimed in any one of claims 1 to 9, wherein the Blocker Domain comprises:

(a) a first Blocker Domain stem loop, preferably 5-10 nucleotides in total;

(b) a first foot;

(c) a second Blocker Domain stem loop, preferably 10-30 nucleotides in total;

(d) a second foot; and

(e) a third Blocker Domain stem loop, preferably 5-10 nucleotides in total.

13. An inducible sgRNA as claimed in any one of claims 1 to 9, wherein the Blocker Domain comprises:

(a) a first foot;

(b) an RNA aptamer or an anti-miRNA or anti-ncRNA; and

(c) a second foot;

or a RNA aptamer.

14. An inducible sgRNA as claimed in any one of claims 9 to 13, wherein the

nucleotide sequences of one or more of the feet of the Blocker Domain are at least partially complementary to one or more of the following:

(a) a RNA 5’ extension to the crRNA;

(b) the crRNA guide sequence;

(c) the crRNA Repeat Sequence or duplex bulge;

(d) the tetraloop;

(e) a RNA sequence inserted into the tetraloop;

(f) the tracrRNA Anti-repeat Sequence or duplex bulge;

(g) Stem Loop 1 ; (h) a RNA sequence inserted into Stem Loop 1.

(i) Stem Loop 2;

0 a RNA sequence inserted into Stem Loop 2;

(k) Stem Loop 3;

(L) a RNA sequence inserted into Stem Loop 3; or

(m) a RNA sequence 3’ extension to the tracrRNA,

preferably to one or more of the following:

(b) the crRNA guide sequence;

(c) the crRNA repeat sequence or duplex bulge;

(f) the tracrRNA anti-repeat sequence or duplex bulge; or

(h) a RNA sequence inserted into Stem Loop 1.

15. An inducible sgRNA as claimed in any one of claims 1 to 14, wherein the cognate CRISPR enzyme is a Class 2, Type II or V CRISPR enzyme, preferably Cas9 or Cpf1 , or a variant or derivative thereof.

16. An inducible sgRNA as claimed in any one of the preceding claims, wherein the Opening Key is a ncRNA, preferably wherein the ncRNA is selected from the group consisting of miRNA, pre-miRNA, pri-miRNA, piRNA, IncRNA, lincRNA, sRNA, siRNA, eRNA, PARs, introns, sisRNA, lariat intronic RNA, linear introns, intron-retained transcripts, exon-intron circular RNA, hairpin RNA (hpRNA), Intron-containing hairpin RNA (ihpRNA), telomerase RNA, IRES, RepA RNA, snoRNA, scaRNA, snRNA, rRNA, tRNA, bifunctional RNA (bifRNA or cncRNA), HOTAIR RNA, HOTTIP RNA, Airn RNA, B2 RNA, circRNA, viral ncRNA, viral small RNA, viral miRNA, and ncRNAs from Retrotransposons Transposable elements (TEs), Long interspersed nuclear elements (LINEs), Short Interspaced Nuclear Elements (SINEs)(including ALU elements), Long Terminal Repeats (LTR), Endogenous Retroviruses (ERV), Penelope-like elements (PLE), or Dictyostelium intermediate repeat sequence (DIRS).

17. An inducible sgRNA as claimed in any one of claims 1 to 16, wherein:

(1) (a) the inducible sgRNA comprises a crRNA-tracrRNA (either covalently-attached or hybridized),

(b) the Blocker Domain is inserted into the tetraloop,

(c) the nucleotide sequences of one or more feet of the Blocker Domain are at least partially complementary to Stem Loop 1 , (d) the Opening Key is a miRNA (preferably an Argonaute-miRNA complex wherein the sequence of the Blocker Domain or Stem Loop 1 is modified to prevent Argonaute- mediated cleavage), and

(e) the cognate CRISPR enzyme is spCas9 or SaCas9, or derivative thereof which is capable of recognising the inducible gRNA;

or

(2) (a) the inducible gRNA comprises a crRNA,

(b) the Blocker Domain is joined to the 5’-end of the crRNA,

(c) the nucleotide sequences of one or more feet of the Blocker Domain are at least partially complementary to an RNA 3’-extension to the crRNA;

(d) the Opening Key is a miRNA (preferably an Argonaute-miRNA complex wherein the sequence of the Blocker Domain is modified to prevent Argonaute-mediated cleavage), and

(e) the cognate CRISPR enzyme is Cpf1 , or derivative thereof which is capable of recognising the inducible gRNA;

or

(3) (a) the inducible gRNA comprises a crRNA,

(b) the Blocker Domain is joined to the 5’-end of the crRNA,

(c) the nucleotide sequence of the Blocker Domain is at least partially complementary to the targeting sequence of the crRNA,

(e) the cognate CRISPR enzyme is Cpf1 , or derivative thereof which is capable of recognising the inducible gRNA.

18. A kit comprising:

(a) an inducible sgRNA as claimed in any one of claims 1 to 17; and

(b) its cognate CRISPR enzyme; and optionally

(c) one or more cognate Opening Keys.

19. A DNA molecule encoding an inducible sgRNA as claimed in any one of claims 1 to 17.

20. A plasmid or vector encoding a DNA molecule as claimed in claim 19.

21. A method of activating a sgRNA/CRISPR enzyme complex, the method comprising the step:

contacting a composition comprising an inducible sgRNA as claimed in any one of claims 1 to 17 or a DNA molecule as claimed in claim 19 with its cognate CRISPR-enzyme or a DNA molecule coding therefore and one or more cognate Opening Keys, such that the Opening Key(s) binds to the Blocker Domain of the sgRNA, thereby preventing the Blocker Domain from hybridizing to one or more regions of the crRNA or crRNA-tracrRNA or to an additional domain or domains inserted into the crRNA or crRNA-tracrRNA, and thereby allowing the inducible sgRNA to form an active CRISPR enzyme complex.

22. A method of selectively targeting a sgRNA/CRISPR enzyme complex to a desired DNA target in a composition or cell, the method comprising the step:

(a) contacting a composition or cell comprising a desired DNA target with

(i) an inducible sgRNA as claimed in any one of claims 1 to 17 or DNA molecule as claimed in claim 19, wherein the targeting sequence of the sgRNA has sequence identity complementarity with the desired DNA target sequence; and

(ii) a cognate CRISPR-enzyme; and

wherein if the composition or cell is one which comprises an Opening Key which is capable of binding to the Blocker Domain of the sgRNA, the Blocker Domain will be prevented from hybridizing to one or more regions of the crRNA or crRNA-tracrRNA or to an additional domain or domains inserted into the crRNA or crRNA-tracrRNA, and thereby allowing the inducible sgRNA to form an active CRISPR enzyme complex which targets the desired DNA target.

23. A method of inducibly targeting a sgRNA/CRISPR enzyme complex to a desired DNA target, the method comprising the steps:

(a) contacting a composition or cell comprising the desired DNA target with

(i) an inducible sgRNA as claimed in any one of claims 1 to 17 or a DNA molecule as claimed in claim 19, wherein the targeting sequence of the sgRNA has complementary sequence identity with the desired DNA target sequence; and

(ii) a cognate CRISPR-enzyme; and

(b) introducing one or more Opening Keys into the composition or cell,

or activating one or more Opening Keys within the composition or cell, such that the Opening Key(s) binds to the Blocker Domain of the sgRNA, thereby preventing the Blocker Domain from hybridizing to one or more regions of the crRNA or crRNA-tracrRNA or to an additional domain or domains inserted into the crRNA or crRNA-tracrRNA, and thereby allowing the inducible sgRNA to form an active CRISPR enzyme complex which targets the desired DNA target.

24. A method of determining the presence or absence of an endogenous moiety in a composition or cell, the method comprising the steps:

(a) contacting a composition or cell with

(i) an inducible sgRNA as claimed in any one of claims 1 to 17 or a DNA molecule as claimed in claim 19, wherein the Blocking Domain of the sgRNA comprises a binding site for the endogenous moiety, such that the binding of the endogenous moiety to the Blocking Domain prevents the Blocker Domain from hybridizing to one or more regions of the crRNA or crRNA- tracrRNA or to an additional domain or domains inserted into the crRNA or crRNA-tracrRNA, and

thereby allowing the inducible sgRNA to form an active CRISPR enzyme complex; and

(ii) a cognate CRISPR-enzyme or a DNA molecule coding therefore;

wherein the presence of an active sgRNA/CRISPR enzyme in the composition or cell is indicative of the presence of the endogenous moiety within the composition or cell, and wherein the absence of an active sgRNA/CRISPR enzyme in the composition or cell is indicative of the absence of the endogenous moiety within the composition or cell, preferably wherein the endogenous moiety is an Opening Key.

25. A method of identifying potential cancer cells, the method comprising the steps;

(a) contacting a cell with

(i) an inducible sgRNA as claimed in any one of claims 1 to 17 or a DNA molecule as claimed in claim 19, wherein the Blocking Domain of the sgRNA comprises a binding site for an Opening Key (preferably an miRNA) whose expression is associated with cancer cells, such that the binding of the Opening Key (preferably miRNA) to the Blocking Domain prevents the Blocker Domain from hybridizing to one or more regions of the crRNA or crRNA-tracrRNA or to an additional domain or domains inserted into the crRNA or crRNA-tracrRNA, and thereby allowing the inducible sgRNA to form an active CRISPR enzyme complex with a cognate CRISPR enzyme; and (ii) a cognate CRISPR-enzyme or a DNA molecule coding therefore;

26. A method of destroying a cancer cell, the method comprising the steps:

(a) contacting a cell with

(i) an inducible sgRNA as claimed in any one of claims 1 to 17 or a DNA molecule as claimed in claim 19, wherein the Blocking Domain of the sgRNA comprises a binding site for an Opening Key (preferably an miRNA) whose expression is associated with cancer cells, such that the binding of the Opening Key (preferably miRNA) to the Blocking Domain prevents the Blocker Domain from hybridizing to one or more regions of the crRNA or crRNA-tracrRNA or to an additional domain or domains inserted into the crRNA or crRNA-tracrRNA, and

(ii) a cognate CRISPR-enzyme or a DNA molecule coding therefore;

wherein if the cell expresses the Opening Key (preferably miRNA) whose expression is associated with cancer cells, the Opening Key will bind to the Blocker Domain of the sgRNA, thereby preventing the Blocker Domain from hybridizing to one or more regions of the crRNA- tracrRNA or to an additional domain or domains inserted into the crRNA or crRNA-tracrRNA, and thereby allowing the inducible sgRNA to form an active CRISPR enzyme complex, wherein the activation of the sgRNA/CRISPR complex leads to the death of the cell.

27. An inducible sgRNA comprising a crRNA or a crRNA linked to a tracrRNA (crRNA- tracrRNA), a first Blocker Domain and a second Blocker Domain, wherein the nucleotide sequences of one or more regions of the first Blocker Domain are at least partially

complementary to the nucleotide sequences of one or more first regions of the crRNA or crRNA- tracrRNA or to an additional domain or domains inserted into the crRNA or crRNA-tracrRNA, such that the first Blocker Domain is capable of (substantially) hybridizing to those one or more first regions of the crRNA or crRNA-tracrRNA or to the additional domain or domains inserted into the crRNA or crRNA-tracrRNA, and wherein

(i) the position of the first Blocker Domain in the inducible sgRNA and

(ii) the location(s) of the one or more first regions of the crRNA or crRNA-tracRNA or the additional domain or domains inserted into the crRNA or crRNA-tracrRNA are selected such that the sgRNA is capable of adopting at least the following two

conformations:

(A1) a first conformation wherein the first Blocker Domain is hybridized or substantially hybridized to the one or more first regions of the crRNA or crRNA-tracrRNA or the additional domain or domains inserted into the crRNA or crRNA-tracrRNA,

and

(B1 ) a second conformation wherein the first Blocker Domain is not hybridized or substantially not hybridized to the one or more first regions of the crRNA or crRNA- tracrRNA or the additional domain or domains inserted into the crRNA or crRNA- tracrRNA;

and wherein the nucleotide sequences of one or more regions of the second Blocker Domain are at least partially complementary to the nucleotide sequences of one or more second regions of the crRNA or crRNA-tracrRNA or the additional domain or domains inserted into the crRNA or crRNA-tracrRNA, such that the second Blocker Domain is capable of (substantially) hybridizing to those one or more second regions of the crRNA or crRNA-tracrRNA or to the additional domain or domains inserted into the crRNA or crRNA-tracrRNA, and wherein

(i) the position of the second Blocker Domain in the inducible sgRNA and

(ii) the location(s) of the one or more second regions of the crRNA or crRNA-tracRNA or the additional domain or domains inserted into the crRNA or crRNA-tracrRNA

conformations:

(A2) a first conformation wherein the second Blocker Domain is hybridized or substantially hybridized to the one or more second regions of the crRNA or crRNA- tracrRNA or to the additional domain or domains inserted into the crRNA or crRNA- tracrRNA,

and

(B2) a second conformation wherein the second Blocker Domain is not hybridized or substantially not hybridized to the one or more second regions of the crRNA or crRNA- tracrRNA or to the additional domain or domains inserted into the crRNA or crRNA- tracrRNA.

28. A kit comprising:

(a) a crRNA or crRNA-tracrRNA, and

(b) a Molecular Switch comprising a Blocker Domain, wherein the nucleotide sequences of one or more regions of the Blocker Domain are at least partially complementary to the nucleotide sequences of one or more regions of the cRNA or crRNA-tracrRNA or to an additional domain or domains inserted into the crRNA or crRNA- tracrRNA, such that the Blocker Domain is capable of hybridizing to those one or more regions of the crRNA or crRNA-tracrRNA or additional domain or domains inserted into the crRNA or crRNA-tracrRNA, wherein:

(a) a non-coding RNA (preferably a miRNA), or

(b) the nucleic acid part of a protein/nucleic acid complex.

29. A method of unblocking a blocked crRNA or crRNA-tracrRNA, the method comprising contacting a crRNA or crRNA-tracrRNA as defined in claim 28 which is blocked with a Molecular Switch comprising a Blocker Domain as defined in claim 28, with an Opening Key as defined in claim 28.