WO2021215932A1 - Improved method of gene editing - Google Patents

Abstract

The invention relates to a method for increasing the efficiency of introducing a mutation using CRISPR/Cas technology. More specifically, the method is directed to preventing the occurrence of re-cutting after successful introduction of the desired mutation by introducing an inactive guide RNA. The invention further relates to use of the method in therapy and kits comprising components for carrying out the method.

Description

Title: Improved method of gene editing

Description

CRISPR technology is a simple yet powerful tool for editing genomes. It allows to easily alter DNA sequences and modify gene function. Recent advances in nuclease-assisted gene modification technology are transforming fundamental and clinical science. Paramount to these developments are the unparalleled ease-of-use and high efficiency of engineered RNA-guided nucleases, chief among them Streptococcus pyogenes Cas9 (spCas9, Cas9). The location of a Cas9-induced DNA double-stranded break (DSB) is specified by an approximately 20 nucleotide (nt) sequence (called: spacer) in the single guide RNA (sgRNA, gRNA) that forms a ribonucleoprotein complex with Cas9. If this sequence matches a genomic sequence, the so-called protospacer, that is followed by an “NGG” ‘protospacer adjacent motif (PAM), a DSB is induced 3 base pairs (bp) upstream of the PAM. Subsequent error- prone DSB repair can leave a scar that in many cases cripples the gene product. Alternatively, the DSB can be repaired by homology-directed repair (HDR) when a single- or double-stranded DNA template is provided. In this process, genetic information is transferred from the template to the genome, allowing pre-designed gene modification as subtle as the substitution of a single base pair.

Unfortunately, DSB induction has also been observed in sequences differing by 1 nt or more from the protospacer sequence and this is at least partially due to a tolerance for spacerprotospacer mismatches. The consequence of this “off-target” promiscuity is twofold: first, the induction of genomic DSBs may occur at sequences different from the original target site, potentially leading to unwanted gene disruptions; second, a single base-pair substitution introduced into the protospacer sequence by templated repair may not necessarily render the modified sequence refractory to Cas9 activity. This “on target off-target activity" may cause re-cutting and destruction of the desired modified sequence, often frustrating the generation of single base-pair substitutions in cell lines and laboratory animals. To avoid re-cutting, additional mutations are often deliberately introduced in the repair template in order to disrupt the PAM or to increase the divergence with the gRNA spacer. However, these extra mutations are not always silent, e.g., when they generate a non-synonymous codon or affect regulatory motifs. Promiscuous Cas9 activity, both at distant sites or on target, is undesirable and poses a major hurdle to the clinical maturation of Cas9 technology. Several strategies have been demonstrated to reduce off-target activity, such as engineering a requirement for the targeting of two guide RNAs for DSB induction (‘dual nickase’ or Fokl fusions), truncating the guide RNA or engineering the Cas9 protein (Fu Y. et al. , (2014) Improving CRISPR-Cas nuclease specificity using truncated guide RNAs. Nat. Biotechnol., 32, 279-284). Even though these implementations were successful in reducing overall off-target activity, break induction at some individual off-target sites remained (Tsai S.Q. et al., (2015) GUIDE-seq enables genome-wide profiling of off- target cleavage by CRISPR-Cas nucleases. Nat. Biotechnol., 33, 187-197; Tsai S.Q. et al., (2014) Dimeric CRISPR RNA-guided Fokl nucleases for highly specific genome editing. Nat. Biotechnol., 32, 569-76; Slaymaker I.M. et al., (2016) Rationally engineered Cas9 nucleases with improved specificity. Science, 351 , 84-8; Chen J.S. et al., (2017) Enhanced proofreading governs CRISPR-Cas9 targeting accuracy. Nature, 550, 407-10). Most recently, single base-pair substitution with low level off- target events was accomplished by integrating Cas9-mediated single-strand break induction with reverse transcription of an RNA template (Anzalone A.V. et al., (2019) Search-and-replace genome editing without double-strand breaks or donor DNA. Nature, 576, 149-57). Nonetheless, with ssODN-templated repair of Cas9-induced DSBs being the best studied and most versatile method of introducing base-pair substitutions, a requirement exists for techniques that can reduce off-target Cas9- activity.

The present invention aims to overcome the above drawbacks, among others, by the methods and products as defined in the appended claims.

Summary of the Invention

In a first aspect, the invention relates to a method of modifying a chromosomal sequence in a cell, the method comprising the steps of: a) introducing into the cell a Clustered Regularly Interspersed Short Palindromic Repeats (CRISPR)/CRISPR-associated (Cas) (CRISPR/Cas) endonuclease or a nucleic acid encoding the CRISPR/Cas endonuclease; b) introducing into the cell guide RNA, or a nucleic acid encoding the guide RNA, wherein the guide RNA comprises a DNA-targeting segment and a CRISPR/Cas endonuclease-binding segment, wherein the protein-binding segment comprises a tracrRNA, and wherein the DNA-targeting segment comprises a complementary region that is substantially complementary to a region at a target site in the chromosomal sequence in the cell and that can base pair with the region at the target site, wherein the region at the target site is upstream of a protospacer adjacent motif (PAM) recognized by the CRISPR/Cas endonuclease, and wherein a protein-RNA complex is formed between the guide RNA and the CRISPR/Cas endonuclease, the guide RNA guides the CRISPR/Cas endonuclease to the target site, the CRISPR/Cas endonuclease introduces a single-stranded break or a double-stranded break at the target site, and a cellular DNA repair process repairs the single-stranded break or double- stranded break, thereby introducing a modification at the target site, wherein the method further comprises the step of c) introducing into the cell hide RNA, or a nucleic acid encoding the hide RNA, wherein the hide RNA comprises a DNA-targeting segment and a CRISPR/Cas endonuclease-binding segment, wherein the protein-binding segment comprises a tracrRNA, and wherein the DNA-targeting segment comprises a complementary region that is in part identical to the complementary region of the DNA-targeting segment of the guide RNA and that is substantially complementary to the region at the target site in the chromosomal sequence in the cell wherein the modification has been introduced and that can base pair with the region at the target site wherein the modification has been introduced, and wherein the complementary region in the hide RNA is shorter in length than the complementary region in the guide RNA. In a second aspect, the invention relates to a hide RNA as defined in method according to the invention for use as a medicament.

In a third aspect, the invention relates to a hide RNA as defined in method according to the invention in the treatment of a subject, wherein the treatment comprises the modification of a chromosomal sequence in a cell of the subject.

In a fourth aspect, the invention relates to a cell comprising a guide RNA, a hide RNA, a CRISPR-Cas endonuclease, and preferably a donor polynucleotide as defined herein. In a fifth aspect, the invention relates to a kit of parts comprising a hide RNA as defined in the first aspect of the invention or a nucleic acid encoding a hide RNA as defined in the first aspect of the invention, and a guide RNA as defined in the first aspect of the invention or a nucleic acid encoding a guide RNA as defined in the first aspect of the invention, and optionally a donor polynucleotide as defined herein, optionally further comprising CRISPR/Cas endonuclease or a nucleic acid encoding CRISPR/Cas endonuclease.

Drawings

Figure 1. Re-cutting after CRISPR/Cas9-mediated single-nucleotide substitution.

A. In the Rosa26 locus of mouse Embryonic Stem Cells, the enhanced GFP gene (grey arrow) is integrated, whose activity depends on conversion of an in-frame upstream AAG triplet (rectangle) into ATG. Part of the reporter sequence is shown, with the AAG triplet marked in grey. The bottom sequence shows the PAM (underlined); the upper stand is hybridized to the gRNA spacer sequence (light gray). The 5’ G (marked in grey) of the RNA spacer does not hybridize to the target sequence. Gray triangles indicate the position of the double strand break.

B. Cas9 nuclease delivery system and ssODN repair templates. A puromycin acetyltransferase gene (encoding resistance to puromycin) was inserted into the Cas9/gRNA expression vector from Cong et al. (2013). Two 120 nt ssODN HDR templates were used: ‘ATG’ instructs conversion of the AAG into ATG; ‘ATG+PAM’ is identical to ‘ATG’ but also instructs disruption of the PAM.

C. Percentage of GFP positive cells determined by Flow cytometry as a function of the puromycin concentration used to select for transfected cells. Diamonds represent GFP percentages obtained with the ATG ssODN repair template; Squares represent GFP percentages obtained with the ATG+PAM ssODN repair template.

D. Reconstruction of the target sequence after HDR-templated base-pair substitution. The top part shows that repair with ATG ssODN template still allows gRNA hybridization despite the single T-U mismatch. Repair with the ATG+PAM ssODN template induced PAM disruption, precluding Cas9-mediated re-cutting.

Figure 2 Protection against re-cuttinq by hide RNAs. A. Upper part: Cas9 in complex with a full-length gRNA (represented as a curving line) can still induce a break despite the presence of a spacer-protospacer mismatch (grey rectangle). Lower part: a gRNA with truncated protospacer that has complete homology to the target sequence from the top part stably binds this sequence, but does not induce a break. It hides the target site from Cas9 bound to a full-length gRNA.

B. Hide RNAs (hRNAs) with matching, truncated protospacer sequence protect the modified reporter sequence from re-cutting. hRNAs 1 and 2 use the original PAM (upper part); hRNAs 3 and 4 use an adjacent PAM. The start codon is marked grey. Note that the sequence of the hRNA protospacers allows hybridization to the T of the ATG whereas a T-U mismatch is formed with the protospacer of the gRNA (indicated in light grey).

C. Experimental setup for evaluating the re-cutting prevention capability of hRNAs. A vector expressing a hRNA and a gRNA is combined with an ATG-inducing ssODN.

D. GFP percentages obtained with 6 different conditions: gRNA + re-cutting-sensitive (ATG) or -protected (ATG+PAM) template; gRNA + hRNA 1-4 plus re-cutting sensitive (ATG) template. * p<0.05, ** p<0.01 , *** p<0.001, **** p<0.0001 (Welch one-sided test comparison with ‘ATG ssODN’ condition, pattern indicates for which condition this applies).

E. The protection score was calculated as follows: (Gi - Gsens)/(GPAM - Gsens), with Gi the percentage of GFP positive cells for a sample, Gsens the percentage of GFP positive cells obtained with the gRNA and the re-cutting sensitive template, and GPAM the percentage of GFP positive cells obtained with the gRNA and the re-cutting- protected template. Patterns of each bar as indicated.

Figure 3. Parameters affecting hRNA performance.

A. Three hRNAs using different PAMs (1 , 2 and 3, marked light grey, black, dark grey, respectively) and varying protospacer lengths were designed to protect the activated GFP reporter against re-cutting. Part of the GFP reporter sequence after repair with the ATG only template is shown (T of ATG marked grey) and the PAMs. Part of the spacer sequence of the hRNAs that target PAMs 1 to 3 is shown, with the A that marks their specificity marked in grey.

B. Protection scores of hRNAs 1-3 with different spacer lengths determined at three different puromycin concentrations. If the first nucleotide of the protospacer was not a G, a G was added to the sequence. The hRNA protospacer length does not take a mismatching 5’ G into account.

Figure 4. hRNAs protecting a mutation outside the target site.

A. Top: sequences of the reporter before (WT) and after ssODN-directed insertion of TGG (ATG) or TGG insertion plus PAM disruption (ATG+PAM). The ‘ATG’ template leaves the entire Cas9 target site intact and thus sensitive to re-cutting. Bottom: gRNA and hRNA A and B spacers are shown, interacting with the target sequence after TGG insertion.

B. Hide RNAs can protect a mutation that is outside of the Cas9 target site. The percentages of GFP-expressing cells are scored after puromycin selection at 3 different concentrations for combinations of Cas9 and the gRNA, the ATG, ATG+PAM ssODN template and hRNA A or B.

Figure 5. hRNA protection at endogenous loci.

A. Introduction of 1 nt substitution in the MSH6 DNA MMR gene that results in a motif recognized by the Type II restriction endonuclease Nael. Top: the wild-type sequence with single-letter amino acid code. The PAM is indicated with a line. Middle: one nucleotide change (in grey) changing the encoded amino acid from proline to alanine. In addition, an Nael site is created (indicated with a black line). Bottom: additional, silent mutations have been instructed to disrupt the protospacer. Note that PAM disruption in this case would have disrupted the Nael site.

B. Three different transfections were performed for each mutation: gRNA + template instructing a single nucleotide change (SNS), gRNA + template instructing SNS and additional Cas9 target site disrupting (TSD) mutations, or gRNA + hRNA + template instructing SNS.

C. Experimental workflow. Quantification of the reaction products was performed using the Caliper GX.

D. Hide RNAs can be used to protect mutations generated at endogenous loci. On the left, the gene name and intended codon substitution is shown. For each mutation, 4 sequences are shown: the wildtype genomic CRISPR/Cas9 target sequence (WT), the sequence where the SNS has been introduced (marked with a blue ‘M’), the sequence with the SNS plus additional target site disruption (marked with red TSD’), the hRNA spacer sequence. Dots denote sequence identity; dashes indicate that this sequence is not part of the hRNA protospacer. The scores are calculated based on the percentages of digested DNA. Amounts of cut and uncut DNA were quantified on the labchip caliper GX, and the fraction of DNA that contained a Type II restriction site was calculated as: (cut DNA/all DNA)*100%. The experiment was performed three times. Within each replicate, a ‘max-normalization’ was performed -fraction of DNA was divided by the value of the sample with the largest amount of digested DNA. Max normalizations for all replicates were averaged and shown with SEM. * p<0.05, ** p<0.01 , *** p<0.001 , **** p<0.0001 (Welch one-sided test comparison with ‘M’ (no protection) condition, unless otherwise indicated).

Figure 6. hide RNA protection in zygotes.

A. Sequences of wildtype (WT), oligonucleotide template (ssODN) and guide and hide RNAs for three different mutations.

B. For each mutation two rounds of zygote injections were performed: ssODN plus RNP with guide RNA or ssODN plus RNP with guide RNA and RNP with hRNA. Injected zygotes were cultured to blastocyst stage and analyzed for the presence of the mutant allele by Sanger sequencing.

C. The percentage of HDR-deriving alleles was calculated per embryo by the TIDER algorithm. Horizontal bars indicate the average percentage of mutant allele.

D. Examples of the chromatograms produced by Sanger sequencing blastocyst gDNA from guide-only and guide RNA+hRNA injections to make the D304V mutation. The arrow indicates the position of the base-change achieved upon HDR with the oligonucleotide template and the black line indicates the PAM of the guide and hRNA (the orientation of the chromatogram sequence is complementary to the one depicted in Fig. 6A). **** p<0.00001 (Welch one-sided test).

Figure 7. Protection score of six hideRNAs with unrelated spacer sequence, as in Fig.

The blue and red colors denote data from cells selected with 9.2 and 22.3 pg/mL puromycin, respectively. In all experiments, cells were transfected with the same guideRNA; ATG indicates the use of an ssODN only instructing the ATG-generating mutation; ATG+PAM indicates the use of an ssODN repair template instructing additional PAM disrupting base changes. C1 to C6 indicate hRNAs 1 to 6 with an unrelated spacer sequence. Figure 8. Live cell numbers in the hideRNA experiments from Figure 3, using hideRNA PAM 1 and PAM 2.

The control conditions (guideRNA + ssODN, no hideRNA) are indicated as ‘no HideRNA’. In these conditions, data from the cells transfected with the ssODN that instructs a PAM disrupting mutation is designated with a ‘P’ while data from the transfections with ssODN that does not instruct a PAM mutation is designated with ‘NP’. For each puromycin concentration, the data was normalized. (n=3).

Detailed description of the invention

A portion of this disclosure contains material that is subject to copyright protection (such as, but not limited to, diagrams, device photographs, or any other aspects of this submission for which copyright protection is or may be available in any jurisdiction). The copyright owner has no objection to the facsimile reproduction by anyone of the patent document or patent disclosure, as it appears in the Patent Office patent file or records, but otherwise reserves all copyright rights whatsoever.

Various terms relating to the methods, compositions, uses and other aspects of the present invention are used throughout the specification and claims. Such terms are to be given their ordinary meaning in the art to which the invention pertains, unless otherwise indicated. Other specifically defined terms are to be construed in a manner consistent with the definition provided herein. Although any methods and materials similar or equivalent to those described herein can be used in the practice for testing of the present invention, the preferred materials and methods are described herein.

For purposes of the present invention, the following terms are defined below.

As used herein, the singular form terms “a,” “an,” and “the” include plural referents unless the content clearly dictates otherwise. Thus, for example, reference to “a cell” includes a combination of two or more cells, and the like. For example, a method for administrating a drug includes the administrating of a plurality of molecules (e.g. 10's, 100's, 1000's, 10's of thousands, 100's of thousands, millions, or more molecules).

As used herein, “and/or” refers to a situation wherein one or more of the stated cases may occur, alone or in combination with at least one of the stated cases, up to with all of the stated cases. As used herein, "at least" a particular value means that particular value or more. For example, "at least 2" is understood to be the same as "2 or more" i.e. , 2, 3, 4, 5, 6, 7, 8, 9, 10, 11 , 12, 13, 14, 15, ... , etc. As used herein, the term "at most " a particular value means that particular value or less. For example, "at most 5 " is understood to be the same as "5 or less" i.e., 5, 4, 3, ... .-10, -11 , etc.

As used herein, “comprising” or “to comprise” is construed as being inclusive and open ended, and not exclusive. Specifically, the term and variations thereof mean the specified features, steps or components are included. These terms are not to be interpreted to exclude the presence of other features, steps or components. It also encompasses the more limiting “to consist of”.

As used herein, “conventional techniques” or “methods known to the skilled person” refer to a situation wherein the methods of carrying out the conventional techniques used in methods of the invention will be evident to the skilled worker. The practice of conventional techniques in molecular biology, biochemistry, cell culture, genomics, sequencing, medical treatment, pharmacology, immunology and related fields are well-known to those of skill in the art and are discussed, in various handbooks and literature references.

As used herein, "exemplary" means "serving as an example, instance, or illustration," and should not be construed as excluding other configurations disclosed herein.

It is desirable to reduce the sensitivity of a successfully-modified locus to re-cutting by Cas9 when generating single point mutations by templated DSB repair. The inventors have surprisingly found that a DNA target site can be ‘hidden’ by persistent binding to a non-active Cas9 RNP. One possibility is to use “dead Cas9” with mutated nuclease domains, which can bind a target site without inducing a DSB. Alternatively, binding without cutting can also be achieved by wildtype Cas9 complexed with gRNAs containing a trimmed spacer. The inventors reasoned that Cas9 complexed to a gRNA with a trimmed spacer that perfectly matches the successfully mutated sequence, may confer protection against re-cutting by Cas bound to the full length gRNA with a now imperfectly matching spacer. The inventors demonstrate that the addition of a gRNA with a trimmed spacer sequence, which we will refer to as ‘hide RNA’ in addition to a convention guide RNA still allows initial introduction of a single or double strand break but reduces re-cutting after introducing the desired mutation. This facilitates the recovery of single base-pair substitutions both in cell lines and in zygotes.

Therefore, in a first aspect the invention relates to a method of modifying a chromosomal sequence in a cell, the method comprising the steps of: a) introducing into the cell a Clustered Regularly Interspaced Short Palindromic Repeats (CRISPR)/CRISPR-associated (Cas) (CRISPR/Cas) endonuclease or a nucleic acid encoding the CRISPR/Cas endonuclease; b) introducing into the cell a guide RNA, or a nucleic acid encoding the guide RNA, wherein the guide RNA comprises a DNA-targeting segment and a CRISPR/Cas endonuclease-binding segment, wherein the protein-binding segment comprises a tracrRNA, and wherein the DNA-targeting segment comprises a complementary region that is substantially complementary to a region at a target site in the chromosomal sequence in the cell and that can base pair with the region at the target site, wherein the region at the target site is upstream of a protospacer adjacent motif (PAM) recognized by the CRISPR/Cas endonuclease, and wherein a protein-RNA complex is formed between the guide RNA and the CRISPR/Cas endonuclease, the guide RNA guides the CRISPR/Cas endonuclease to the target site, the CRISPR/Cas endonuclease introduces a single-stranded break or a double-stranded break at the target site, and a cellular DNA repair process repairs the single-stranded break or double- stranded break, thereby introducing a modification at the target site, wherein the method further comprises the step of c) introducing into the cell a hide RNA, or a nucleic acid encoding the hide RNA, wherein the hide RNA comprises a DNA-targeting segment and a CRISPR/Cas endonuclease-binding segment, wherein the protein-binding segment comprises a tracrRNA, and wherein the DNA-targeting segment comprises a complementary region that is in part identical to the complementary region of the DNA-targeting segment of the guide RNA and that is substantially complementary to the region at the target site in the chromosomal sequence in the cell wherein the modification has been introduced and that can base pair with the region at the target site that wherein the modification has been introduced, and wherein the complementary region in the hide RNA is shorter in length than the complementary region in the guide RNA.

The below provided examples and figures convincingly demonstrate that re-cutting of the modified chromosomal sequence can be prevented by the described method using an hide RNA as described herein, thereby increasing efficiency of introducing the desired modification in the chromosomal sequence.

Without wishing to be bound by theory, it is believed that the hide RNA competes with the guide RNA for binding of the Cas complex and for binding to the chromosomal sequence. Because the hide RNA targeting segment includes a sequence that can base pairwith the modification introduced in the chromosomal sequence, it is theorized that after modification of the chromosomal sequence the hide RNA preferably binds that target site on the chromosomal sequence. However, due to the trimmed spacer sequence of the hide RNA, the CRISPR/Cas-hide RNA complex is unable to induce a new strand break, or at least (much) less efficiently able to do so compared to the CRISPR/Cas9-guide RNA complex.

The term “chromosomal sequence” when used herein may refer to a chromosome or chromosomal DNA present in a cell, but may also include plasmids, bacterial artificial chromosomes or vectors and the like that may be present in the cell. Preferably the term refers to a chromosome or nuclear DNA.

The term “modified” or “modifying” or “modification” when referring to the chromosomal sequence refers to a change in the nucleotide sequence introduced by a DNA repair mechanism. The term may refer to an insertion of one or more, e.g. 1 , 2, 3, 4, 5, 6, 7, 8, 9, 10 or more nucleotides in the chromosomal sequence. The term may refer to a deletion of one or more, e.g. 1 , 2, 3, 4, 5, 6, 7, 8, 9, 10 or more nucleotides in the chromosomal sequence. Alternatively the term may refer to a substitution of one or more, e.g. 1, 2, 3, 4, 5, 6, 7, 8, 9, 10 or more nucleotides in the chromosomal sequence. CRISPR (clustered regularly interspaced short palindromic repeats) is a family of DNA sequences found in the genomes of prokaryotic organisms such as bacteria and archaea. These sequences are derived from DNA fragments of bacteriophages that had previously infected the prokaryote. Cas9 (or "CRISPR-associated protein 9") is an enzyme that uses CRISPR sequences as a guide to recognize and cleave specific strands of DNA that are complementary to the CRISPR sequence. Cas9 enzymes together with CRISPR sequences form the basis of a technology known as CRISPR- Cas9 that can be used to edit genes within organisms. Native Cas9 requires a guide RNA composed of two disparate RNAs that associate - the CRISPR RNA (crRNA), and the trans-activating crRNA (tracrRNA). Type II CRISPR-Cas systems require a tracrRNA which plays a role in the maturation of crRNA. The tracrRNA is partially complementary to and base pairs with a pre-crRNA forming an RNA duplex. This is cleaved by RNase III, an RNA-specific ribonuclease, to form a crRNA/tracrRNA hybrid. This hybrid acts as a guide for the endonuclease Cas9, which cleaves the invading nucleic acid.

It is generally known in the field that different endonucleases can be used in CRISPR technology. Therefore, when used herein, the term CRISPR/Cas endonuclease refers to any suitable endonuclease, non-limiting examples are Cas9 and Cas12a (formerly known as Cpf1), as well as mutation and modifications thereof. Non limiting examples are CAS9_D10A, CAS9_H820A, CAS9_H839A, and Cpf1_R1226.

The CRISPR/Cas endonuclease may be provided by methods generally known in the field. For example the CRISPR/Cas endonuclease may directly be introduced in the cell by electroporation or pronuclear injection (e.g. pronuclear injection of zygotes), but other suitable methods are known. Alternatively a nucleic acid encoding the CRISPR/Cas endonuclease may be introduced in the cell resulting in the transcription and subsequent expression of the CRISPR/Cas endonuclease in the cell.

In the method of the invention a guide RNA (also referred to as a single guide RNA, sgRNA or gRNA) is introduced in the cell. The guide RNA may be introduced directly in the cell by methods known in the field, such as lipofection, but other suitable methods are known. Alternatively, a nucleic acid encoding the guide RNA may be introduced in the cell.

Suitable methods to introduce a nucleic acid encoding the guide RNA and/or CRISPR/Cas in the cell are known to the skilled person, non-limiting examples being transfection or electroporation with a plasmid or using a viral vector, but other suitable methods are known. Generally a plasmid or viral vector comprises an element that drives expression of the desired sequence such as a promoter region. Suitable viral vectors are known to the person skilled in the art.

The guide RNA comprises a DNA-targeting segment and a CRISPR/Cas endonuclease-binding segment. The guide RNA may therefore be a single RNA molecule comprising both segments, or the guide RNA may be two separate RNA molecules, each comprising one of the elements. When the guide RNA is two separate RNA molecules it is desirable that the RNA molecules comprise regions that allow association of the RNA molecule comprising the DNA-targeting segment and the RNA molecule comprising the CRISPR/Cas endonuclease-binding segment. For example such regions allowing association may be complementary regions or regions that allow secondary structures to be formed.

The CRISPR/Cas endonuclease-binding segment in the guide RNA comprises a tracrRNA which is involved in binding of the CRISPR/Cas endonuclease. The DNA- targeting segment comprises a complementary region that is substantially complementary to a region (also referred to as crRNA) at a target site in the chromosomal sequence in the cell and that can base pair with the region at the target site. Therefore the crRNA serves to localize the target site (by base pairing) in the chromosomal sequence and allowing the formation of a complex of the guide RNA (comprising the crRNA and the tracrRNA) with CRISPR/Cas and the chromosomal sequence at the target site of the chromosomal sequence.

The target site is chosen such that it is upstream of a protospacer adjacent motif (PAM). The PAM is recognized by the CRISPR/Cas endonuclease. A protospacer adjacent motif (PAM) is a 2 to 6 base pair DNA sequence immediately following the DNA sequence targeted by a CRISPR/Cas. The canonical PAM is the sequence 5'- NGG-3', where "N" is any nucleobase. Some versions of Cas9 have been engineered to recognize 5'-YG-3' (where "Y" is a pyrimidine). The Cpf1 nuclease of Francisella novicida recognizes the PAM 5'-TTTN-3' or 5'-YTN-3'. It is however speculated that nucleases in different species may have different PAM sequences. Therefore the invention is not limited to specific PAM sequences, instead the PAM sequence is defined by the endonuclease used, e.g. Cas9 defines the PAM sequence as NGG.

It is understood that the CRISPR/Cas endonuclease may introduce a double stranded break or a single stranded break (referred to as nickase). The method of the invention can be used both with endonucleases that introduce double stranded breaks and with endonucleases that introduce single stranded breaks. The introduction of a single strand break or a double strand break is here collectively referred to as the introduction of a “strand break”.

By introducing in the cell a CRISPR/Cas endonuclease and a guide RNA as described in steps a) and b) of the method, in the cell a protein-RNA complex can form comprising the CRISPR/Cas endonuclease and the guide RNA; this allows the guide RNA to guide the complex to the target site. The guide RNA allows guiding to the target site because it is capable of base pairing with a region in the target site. The guide RNA is designed such that it can base pair with a region upstream of the PAM. Typically the complementary region of the guide RNA, meaning the part that can base pair with the target site, has a length of 16, 17, 18, 19, 20, 21 or 22 nucleotides, preferably between 17 and 20 nucleotides. The complementary region may be fully complementary to the target site, or it may allow for 1 , 2, 3 or 4 mismatches. /pet

Assembly of the CRISPR/Cas endonuclease - guide RNA complex at the target site enables the CRISPR/Cas endonuclease to introduce a single-stranded break or a double-stranded break at the target site. Introduction of a break in the chromosomal sequence of a cell will induce a cellular repair process which will repair the single- stranded break or double-stranded break. It is known that several mechanisms can be used by the cell, such as non-homologous end joining (NHEJ), and homology directed repair (HDR). Both processes may result in the introduction of a modification in the target site. Depending on the repair mechanism that is used, deletions, insertions or substitutions can be introduced at the site of the break. The modifications introduced by NHEJ are predictable to some extend as described in van Overbeek et al. , 2016, Molecular Cell 63, 633-646 August 18. Alternatively HDR can be used by introducing in the cell a template which is largely identical to the target site but may include a desired modification. The HDR process will use the template for the repair mechanism and thereby also introduce the modification in the chromosomal sequence. The method of the invention is envisioned to work with both the NHEJ and the HDR repair mechanisms.

When used herein, a “hide RNA” is an RNA molecule comprising a DNA-targeting segment and a CRISPR/Cas endonuclease-binding segment, wherein the protein binding segment comprises a tracrRNA, and wherein the DNA-targeting segment comprises a complementary region that is in part identical to the complementary region of the DNA-targeting segment of the guide RNA and that is substantially complementary to the region at the target site in the chromosomal sequence in the cell wherein the modification has been introduced and that can base pair with the region at the target site wherein the modification has been introduced, and wherein the complementary region in the hide RNA is shorter in length than the complementary region in the guide RNA.

Therefore, the hide RNA and the guide RNA have the following properties: 1) When introduced in the cell together with the guide RNA the hide RNA and the guide RNA both are capable of binding the CRISPR/Cas endonuclease. In case the amount of the CRISPR/Cas endonuclease is limited in the cell, it may be assumed that the hide RNA and the guide RNA compete for binding to the CRISPR/Cas endonuclease.

2) The guide RNA and the hide RNA both bind to a region at the target site. The guide RNA and the hide RNA can base pair with a part of the target site in the chromosomal sequence that at least partially overlap. The region at the target site may fully overlap, meaning the region where the hide RNA can base pair with the chromosomal sequence is fully comprised within the region where the guide RNA can base pair with the chromosomal sequence. This may for example be the case when the modification is introduced within the region where the guide RNA can base pair with the chromosomal sequence.

Alternatively, the region where the hide RNA can base pair with the chromosomal sequence partially overlaps with the region where the guide RNA can base pair with the chromosomal sequence. This may for example be the case when the modification is introduced outside the region where the guide RNA can base pair with the chromosomal sequence. The region where the guide RNA and the hide RNA can base pair with the chromosomal sequence may overlap with 1 or more, e.g. 2, 3, 4, 5, 6, 7, 8, 9, 10, 11 , 12, 13, 14, 15, 16 or more nucleotides, preferably with 3 or more, more preferably with 6 or more, most preferably with 9 or more nucleotides.

3) The hide RNA is able to base pair with the region in the target site with the modification. For example, when NHEJ is used and the double strand break is predicted to introduce an insertion by NHEJ, the hide RNA comprises a DNA- targeting segment comprising a complementary region which includes the base (or bases) which are complementary to the base(s) inserted by the modification. Or when for example HDR is used and a base pair substitution is introduced in the chromosomal sequence by the template, the hide RNA comprises a DNA-targeting segment comprising a complementary region which includes the base (or bases) which are complementary to the substituted base(s).

Because the hide RNA can base pair specifically with the modified chromosomal sequence while the guide RNA can base pair specifically with the unmodified chromosomal sequence it is hypothesized that after the modification is introduced by introducing a strand break and subsequent DNA repair mechanism, the hide RNA can preferentially base pair with the target site.

4) The complementary region in the hide RNA is shorter in length than the complementary region in the guide RNA. The complementary region in the hide RNA may for example by 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11 ,12 or more nucleotides shorter than the complementary region in the guide RNA. Preferably the complementary region in the hide RNA is at least 4 to 10 nucleotides the complementary region in the guide RNA. The typical length of the complementary region in the guide RNA is between 17 and 24 nucleotides. The minimal length of a hide RNA complementary region should be 6 nucleotides. Therefore, the length of the hide RNA complementary region may be 6, 7, 8, 9, 10, 11 , 12, 13, 14, 15, 16, 17, 18, 19, 20, 21 , 22 or 23 nucleotides long, depending on the size of the guide RNA. Preferably the complementary region of the hide RNA has a length of between, and including, 6 - 16 nucleotides, more preferably between, and including, 8 - 15 nucleotides. In a preferred embodiment the length of the complementary region of the hide RNA is between and including 10 - 15 nucleotides.

In a further embodiment of the invention the hide RNA according to invention is 10, 11 , 12, 13, 14, 15 or 16 nucleotides long. The hide RNA according to the invention comprises a DNA-targeting segment comprising a complementary region which includes the base (or bases) which are complementary to the base(s) inserted by the modification. In a preferred embodiment said complementary region of the hide RNA has a length of 10, 11 , 12, 13, 14, 15 or 16 nucleotides.

Therefore, when CRISPR/Cas endonuclease, guide RNA and hide RNA are introduced in the cell (or nucleic acids encoding these components), the CRISPR/Cas endonuclease together with the guide RNA will induce a single strand or double strand break at the target region in the chromosomal sequence in the cell. The target region is defined by the complementary region in the guide RNA. By introducing a strand break in the chromosomal sequence, a DNA repair mechanism is induced, such as NHEJ or HDR. These repair mechanisms can be used to introduce a desired modification in the chromosomal sequence, by the specific design of the guide RNA complementary sequence and by optionally including a template polynucleotide (a donor polynucleotide; to enable HDR). To prevent the CRISPR/Cas endonuclease from re-cutting and potentially introducing unwanted modifications, or removing or further modifying the introduced desired modification, it is surprisingly found by the inventors that a hide RNA as defined herein can be used. The hide RNA competes with the guide RNA for available CRISPR/Cas endonuclease and for binding on the chromosomal sequence. The hide RNA binds to the region of the target site comprising the modification, and the binding site on the chromosomal sequence of the hide RNA at least partially overlaps with the binding site of the guide RNA. Because the complementary region of the hide RNA is shorter than the complementary region of the guide RNA, the hide RNA itself is not able to support induction of a strand break, or at least is much less efficient in doing so. The result is that the instances of re-cutting by the CRISPR/Cas endonuclease after successful modification of the chromosomal sequence is drastically reduced, resulting in an overall increased efficiency of modification of the chromosomal sequence by CRISPR/Cas endonuclease.

In an embodiment the method further comprises the step of d) introducing into the cell a donor polynucleotide comprising a donor sequence comprising the modification to be introduced at the target site, preferably wherein the donor sequence is flanked by upstream and downstream sequences that are identical to upstream and downstream sequences, respectively, of the target site in the chromosomal sequence in the cell. Although the method is envisioned to work both with NHEJ or HDR mechanisms, it may be particularly beneficial to introduce a donor polynucleotide comprising a donor sequence comprising the modification to be introduced to enable the HDR mechanism to introduce the modification in the chromosomal sequence. In such case the HDR mechanism uses the donor polynucleotide as a template to repair the chromosomal sequence at the break site. Because the template comprises a modification, the HDR mechanism will incorporate the modification in the chromosomal sequence.

The upstream and downstream sequences of the donor polynucleotide may be individually 10, 11 , 12, 13, 14, 15, 16, 17, 18, 19, 20, 25, 30, 35, 40 or more nucleotides long. Preferably the upstream and downstream sequences of the donor polynucleotide are at least 25 nucleotides long.

It is understood that the donor polynucleotide may be RNA, single stranded DNA or double stranded DNA. It is further understood that the donor polynucleotide sequence may comprise upstream and downstream sequences that are identical to upstream and downstream sequences in the positive strand or the negative strand of the chromosomal sequence, regardless of the strand to which the guide RNA and/or hide RNA can base pair.

In an embodiment the cellular DNA repair process repairs the single-stranded break or double stranded break such that the chromosomal sequence in the cell is modified by deletion of at least one nucleotide, an insertion of at least one nucleotide, a substitution of at least one nucleotide, or a combination thereof.

In an embodiment the DNA repair process is a non-homologous end-joining repair process, or, when a donor polynucleotide is used, wherein the DNA repair process is a homology-directed repair.

In an embodiment steps a), b), c), and optionally d) are performed in any order, simultaneously, sequentially, or any combination thereof. Therefore, the CRISPR/Cas endonuclease, the guide RNA and the hide RNA, and optionally the donor polynucleotide may be introduced into the cell simultaneously, or in any particular order. It is however preferred that the hide RNA is present in the cell when the strand break is introduced to prevent re-cutting of the modified chromosomal sequence prior to introduction of the hide RNA. Therefore, it is preferred that at least one of steps a) or b) are performed simultaneously with step c) or at least one of steps a) or b) are performed after step c). More preferably, when the CRISPR/Cas endonuclease is activated, all the components (guide RNA, hide RNA and optionally donor polynucleotide) are present in the cell. Most preferably all components (CIRSPR/Cas, guide RNA, hide RNA and optionally donor polynucleotide) are introduced at the same time, meaning step a,), b), c), and optionally d) are performed simultaneously.

In an embodiment the CRISPR/Cas endonuclease, the guide RNA and the hide RNA, and optionally the donor polynucleotide are encoded by a single nucleic acid encoding and introduced simultaneously in the cell, where the individual components can be transcribed from the nucleic acid encoding each of the components. In an embodiment the complementary region of the hide RNA is at least 1 , 2, 3, 4, 5, 6, 7, 8, 9 , 10, 11 , 12 or more nucleotides shorter than the complementary region of the guide RNA. Preferably the complementary region of the hide RNA is at least 3 nucleotides shorter than the complementary region of the guide RNA and at most 13 nucleotides shorter than the complementary region of the guide RNA. More preferably, the complementary region of the hide RNA is at least 4 to 10 nucleotides shorter than the complementary region of the guide RNA.

In a preferred embodiment the length of the hide RNA complementary region is chosen such that it is still capable of targeting CRISPR/Cas to the target region in the chromosomal sequence, but is catalytically inactive, meaning not able to induce a single strand break or a double strand break. Depending on several factors like sequence context and the length of the guide RNA complementary region, the limits of the length of the hide RNA complementary region where these conditions apply may vary. An easy test whether the hide RNA is catalytically inactive is by introducing the hide RNA together with CRISPR/Cas in a cell (but without guide RNA). If a strand break is observed the hide RNA is catalytically active and thus not suitable; if no strand breaks are observed the hide RNA is catalytically inactive. Alternatively the minimum length of the hide RNA complementary region that still allows binding to the target region can be determined, e.g. by performing DNA pull down assays.

In an embodiment the complementary region of the hide RNA has a length of between, and including, 6 - 16 nucleotides, preferably between, and including, 8 - 15 nucleotides, more preferably between and including 10 - 15 nucleotides.

In an embodiment the guide RNA comprises at least one nucleotide in the complementary region that is not complementary to the cognate nucleotide in the region at a target site. It is hypothesized that the hide RNA functions at least in part by competing with the guide RNA for binding to the target site, it is particularly useful to design the guide RNA such that it comprises 1 or more, e.g. 1 , 2, 3, 4 or 5, mismatches in the complementary region when base pairing with the target site (prior to the introduction of the modification). By designing the hide RNA such that it preferable is fully complementary (no mismatches) between the complementary region and the target site after introduction of the modification, the competing for chromosomal sequence binding can be shifted more in favour to the hide RNA, thereby reducing re-cutting even more.

In an embodiment the donor polynucleotide comprising the donor sequence comprises a further modification to be introduced, wherein the modification is to be introduced in the region at the target site that is substantially complementary to the complementary region in the guide RNA. As an alternative to introducing deliberate mismatches in the guide RNA to shift binding preference to hide RNA but may reduce efficiency of introducing the modification, the donor polynucleotide may be used to introduce one or more modifications in the region at the target site that is substantially complementary to the complementary region in the guide RNA, thereby resulting in one or more mismatches between the guide RNA complementary region and the target site after introduction of the modification. These modifications can be taken into account in the design of the hide RNA, thus designing a complementary region in the hide RNA with preferably fully complementary sequence to the target site after the modification has been introduced.

Because the desired modification will generally be introduced in the coding sequence of a gene, it is preferred that the further modification is a nucleotide substitution, more preferably a silent nucleotide substitution, in order not to introduce a frame shift or unintended further modification of the protein encoded by the gene.

In an embodiment the donor polynucleotide is single stranded DNA, double stranded DNA or RNA.

In an embodiment more than one different hide RNAs are introduced in the cell.

In an embodiment the DNA-targeting segment and the CRISPR/Cas endonuclease binding segment are arranged in the 5' to 3' direction or in the 3’ to 5’ direction.

In an embodiment the guide RNA and/or the hide RNA is a single molecule or is comprised of two molecules. It is understood that the crRNA and the tracrRNA may be two separate molecules that form a complex with each other and CRISPR/Cas, or may be single molecule, generally referred to a sgRNA (single guide RNA). In an embodiment the CRISPR/Cas endonuclease in introduced into the cells in the form of a ribonucleoprotein complex, wherein the ribonucleoprotein complex comprises the CRISPR/Cas endonuclease and the guide RNA.

In an embodiment the CRISPR-Cas endonuclease introduces a double-stranded break, and wherein the CRISP-Cas endonuclease is selected from the group consisting of Staphylococcus aureus Cas9, Streptococcus pyogenes Cas9, Neisseria meningitides Cas9, Streptococcus thermophiles Cas9, preferably Streptococcus pyogenes Cas9.

In an embodiment the CRISPR-Cas endonuclease introduces a single stranded break and is selected from the group consisting of CAS9_D10A, CAS9_H820A, CAS9_H839A, and Cpf1_R1226. Alternatively, any CRISPR/Cas endonuclease that comprises a mutation that renders one of the two catalytic domains inactive may be used.

In an embodiment the CRISPR-CAS endonuclease is Streptococcus pyogenes Cas9, the complementary region of the guide RNA has a length of between, and including 17 - 20 nucleotides, and the complementary region of the hide RNA is at least 4 - 10 nucleotides shorter than the complementary region of the guide RNA and/or has a length of between and including 8 - 15 nucleotides. In certain embodiments it can be decided that the length of the complementary region of the hide RNA is 16 nucleotides. In a preferred embodiment the complementary region of the hide RNA is between and including 10 - 15 nucleotides, more preferably wherein the complementary region of the hide RNA has a length of 10, 11 , 12, 13, 14, 15 or 16 nucleotides.

In an embodiment the nucleic acid encoding the CRISPR/Cas endonuclease is codon optimized for expression in the cell.

In an embodiment the CRISPR/Cas endonuclease is linked to at least one nuclear localization signal (NLS). In an embodiment the method does not comprise a process for modifying the germ line genetic identity of a human being and/or wherein the method does not comprise a method for treatment of the human or animal body by surgery or therapy.

In a second aspect the invention relates to a hide RNA as defined in any one of the previous claims for use as a medicament.

In a third aspect the invention relates to a hide RNA as defined in any one of the previous claims in the treatment of a subject, wherein the treatment comprises the modification of a chromosomal sequence in a cell of the subject.

In a fourth aspect the invention relates to a cell comprising a guide RNA, a hide RNA, a CRISPR-Cas endonuclease, and preferably a donor polynucleotide as defined in any one of the previous claims.

In a fifth aspect the invention relates to a kit of parts comprising a hide RNA as defined in any one of the previous claims or a nucleic acid encoding a hide RNA as defined in any one of the previous claims, and a guide RNA as defined in any one of the previous claims or a nucleic acid encoding a guide RNA as defined in any one of the previous claims, and optionally further comprising donor polynucleotide as defined in any one of the previous claims.

In an embodiment the kit of parts according to the invention further comprises a CRISPR/Cas endonuclease or a nucleic acid encoding a CRISPR/Cas endonuclease.

Examples Materials and methods mESC GFP reporter cell line, tissue culture, transfection and flow cytometry Mouse embryonic stem cells were cultured, transfected and analyzed by flow cytometry according to Harmsen et al. (2018, DNA mismatch repair and oligonucleotide end-protection promote base-pair substitution distal from a CRISPR/Cas9-induced DNA break. Nuc. Acids Res., 46, 2945-55). Briefly, IB10 mouse embryonic stem cells, maintained on MEF feeder layers in LIF-containing medium were transferred to 0.1% gelatin coated plates one day prior to transfection and cultured in Buffalo Rat Liver (BRL) cell conditioned medium. On the day of transfection, guide RNA or guide+hRNA expression vectors were transfected along with oligonucleotide repair templates with Transit-LT1 (Mirus) transfection reagent, according to manufacturer’s instructions. The day after transfection, cells were reseeded in BRL medium containing indicated amounts of puromycin and grown for 2 d, after which the medium was exchanged for medium without puromycin. After 2-3 d, cells were analyzed by flow cytometry using Cyan ADP flow cytometric analyzer (Beckman Coulter).

Cas9, guide RNA and hRNA expression vectors

For plasmid-based delivery of Cas9 and guide RNAs we used ‘px330.PGKpur’, a polycistronic system based on the popular px330 vector modified by us to contain a puromycin marker. Guide RNAs were inserted as previously described. Briefly; the vector was opened with the Bbsl or Bpil restriction enzyme and the protospacer, a short dsDNA with 5’ overhangs on both sides (5’-CACC on the 5’ end and CAAA-5’ on the 3’ end) was introduced by ligation. To generate a vector expressing puromycin, Cas9, a guide RNA and a hide RNA, we first introduced the respective guide RNA protospacer. Then, we generated an additional cassette for hide RNA expression by first amplifying the empty guide RNA expression cassette from px330 using one perfectly matching primer [gRNA-rev (5’-AACGGGTACCTCTAGAGCC)] and one primer with a one-bp sequence alteration [gRNA-Fw (5’ CTTTTTACGGTACCTGGCCTTTTGC)] resulting in the U6-promoter + single-guide RNA containing fragment flanked by Kpnl sites. Then, we purify the resultant PCR product with the Giaquick PCR purification kit (Giagen) and digested it with Kpnl-HF (New England Biolabs). This expression cassette was then ligated into Kpnl-linearized px330.PGKpur that already contained a guide RNA protospacer (in case of the experiments in Fig. 2-4, this was the spacer targeting GFP, GAAGCTCGATGCATAGGCCT, in case of the experiment in Fig. 5, these were the respective guides targeting the MMR genes). The resultant vector we refer to as the ‘gRNA + empty vector’. Finally, into this vector, the protospacer of the hide RNA was cloned, in the same way as the guide RNA protospacers were cloned to yield a vector expressing a guide RNA, hide RNA, Cas9 and puromycin.

293FT cell line, culture and transfection 293FT cells (Life Technologies) were cultured in optiMEM (Life technologies) with 8% FCS (Life Technologies). One day before transfection, 1.25E4 cells were seeded in a flat bottom 96-well. To transfect the 293FT cells, a mix of 50 ng gRNA or guide RNA + hRNA expressing vector and 150 ng ssODN HDR repair template DNA was prepared in 6.25 pi optiMEM medium (Life Technologies) in a 96-well plate, briefly mixed by pipetting. To this, 6.25 mI optiMEM containing 0.5 mI Transit-LT1 (Mirus) was added, and this was briefly mixed by pipetting. Complexes were formed for 15 to 30 min, and then transferred to 293FT cells of which the medium was gently refreshed 5 to 60 min before. Cells were then incubated for 16-20 h in 37 °C and 5% CO2, upon which the medium was refreshed with optiMEM containing 8% FCS and 7.2 pg/ml puromycin. Cells were incubated for 2 d, and then the medium was aspirated. Cells were washed with PBS (Life Technologies) and optiMEM with 8% FCS was added, followed by incubation for 2 d. Cells were detached by pipetting, and 1/10 of the cell suspension was transferred to a new flat bottom 96-well plate, and incubated for 2 or 4 d, after which the medium was aspirated and cell layers were frozen at -20 °C.

293FT DNA isolation, PCR amplification, digestion and Caliper GX analysis Genomic DNA was isolated from the frozen cell layers by addition of 50 mI of a solution containing 10 mM Tris pH 8.0, 100 mM NaCI, 0.5% SDS, 10 mM EDTA and 50 mI ml Proteinase K, followed by incubation for 2 h at 55 °C, and 45 min at 85 °C. Genomic DNA was precipitated by addition of 125 mI 100% Ethanol, followed by 30 min centrifugation at 4 °C at 4600 RCF. Precipitated DNA was washed with 70% Ethanol, air-dried and eluted overnight in 50 mI Tris/EDTA buffer (10 mM/0.5 mM).

The targets of gene modification were amplified by PCR with Taq polymerase (MRC Holland). For each locus and each sample, multiple PCR reactions were pooled and concentrated by precipitation (briefly: 0.1 volume of 3 M Sodium Acetate pH 5.0 and 0.7 volume of Isopropanol were added, followed by centrifugation at 10,000 RCF at 4 °C for 30 min, washing with 70% Ethanol, air-drying and DNA pellet elution in TE).

1 mI of purified PCR product was digested with 0.1 mI enzyme in a total volume of 2 mI in a PTC-100 thermocycler (MJ research) in PCR tubes for 3 h, otherwise according to manufacturer’s instructions. Each reaction was then supplemented with 18 mI of MO H2O and briefly mixed, after which 15 mI of sample was analyzed on the Labchip GX (Caliper Life Sciences) to record the amounts of DNA at the cut and uncut positions. Zygote injections

Blastocyst injections were performed according to Pritchard et al. (2017) Direct Generation of Conditional Alleles Using CRISPR/Cas9 in Mouse Zygotes. Meth in Mol Biol, 1642, 21-35, with some modifications. Guide RNAs and hRNAs were complexed with Cas9 protein (Alt-R, IDT) in separate reactions for 5 min at room temperature. 75 pi injection mix was prepared containing 1.875 pg guide RNA/15 pg Cas9 and 1.875 pg hRNA/15 pg Cas9 (if included) complex and 0.4 mM of HDR oligonucleotide. The injection mix was injected into the pronucleus of freshly isolated zygotes from the FVB/N strain. Zygotes were cultured in KSOM medium (Millipore) in a tissue culture dish at 5% CO2 and 37 °C until blastocyst state. Blastocyst were lysed in 5 mI of Direct Lysis tail buffer with 0.3125 mg/ml proteinase K for 30 min at 56 °C, followed by inactivation at 80 °C for 45 min. Then, the loci targeted by Cas9 nucleases were amplified using the entire blastocyst lysate as a template with Phusion flash High fidelity polymerase (NEB, F-548S), according to manufacturer’s instructions. A21V and R691W samples: and 1 mI of PCR product was sequenced. D304V samples: the entire volume of each blastocyst PCR was loaded on 2% agarose gel and different bands were separately isolated. Then, DNA was extracted from gel using QiaQuick gel purification kit, according to manufacturer’s instructions. The purified DNA was sequenced. None of the chromatograms deriving from a band migrating separately from the target site product could be algorithmically deconvoluted. The orientation of the sequencing primer was such that indels produced by re-cutting occurred between the sequencing primer and the HDR event in case of A21V, R691 W and D304V.

TIDER analysis

Algorithmic deconvolution of chromatograms using TIDER (Brinkman E.K., et al. (2018) Easy quantification of template-directed CRISPR/Cas9 editing, Nuc Acids Res, 46, e58) was performed using the Deskgen TIDER portal (http://TIDER.deskgen.com, accessed from July 30th 2019 through August 3d 2019). The experimental HDR control samples (‘Reference chromatogram’) were constructed according to instructions using PCR products from control blastocysts. Chromatograms from unmodified blastocysts were used as a wildtype control, in accordance with instructions accompanying TIDER. The TIDER result page was saved with firefox 68.0.2 (Mozilla Corporation) and the resultant HTML files containing the table with frequencies of the wildtype, various indels and the HDR sequence was transferred into an excel sheet using a custom VBA macro.

Svnthego ICE analysis

Synthego ICE analysis made use of the ICE build publically available on “https://github.com/synthego-open/ice” (version January 23, 2018, accessed June 3, 2019), according to the accompanying instructions. Around 30% of the chromatograms were of a quality that did not meet ICE standards and were excluded from analysis. Relevant ICE output was subsequently parsed into an Excel sheet using a custom Micrsoft Excel VBA macro in order to collect the fractions of wildtype, knock-in and indel alleles from each blastocyst. Only chromatograms with an R2 of 0.8 or higher (excluding around 1/3 of all analyzed chromatograms) were used.

Results

Re-cutting after successful base-pair substitution

As a reporter for single base-pair substitution, we used mouse Embryonic Stem Cells (mESCs) carrying a single-copy sequence that encodes green fluorescent protein (GFP) but lacks a start codon. However, an in frame AAG triplet is present, which allows for GFP expression upon conversion into ATG (Fig. 1A).

To study single base-pair substitution by single-stranded oligodeoxyribonucleotide (ssODN)-templated homology-directed repair (HDR) of a DSB, we designed a guide RNA to direct SpCas9 cleavage close to the AAG triplet (Fig. 1A). This gRNA was expressed from a vector that also encoded SpCas9 and puromycin acetyltransferase (the latter allowing enrichment of vector-transfected cells by puromycin selection) (Fig. 1 B). We designed a 120 nt ssODN to template repair and direct the conversion of AAG to ATG. This conversion alters the protospacer by 1 bp (Fig. 1 B). We transfected the vector and the ssODN repair template into the mESC reporter line, applying different concentrations of puromycin (Fig. 1 B). Quantification of the fraction of GFP-positive cells by flow cytometry showed an increase in HDR efficiency with increased puromycin concentrations up to 3.6 pg/ml (Fig. 1C). However, HDR efficiency decreased when further increasing puromycin concentration.

Cas9 can induce DSBs at target sequences that differ slightly from the protospacer sequence, and this is stimulated by elevated Cas9 concentrations. In our system, high concentrations of puromycin select for high Cas9 levels, which may cause re-cutting of the ATG containing sequence despite the imperfect match with the guide RNA (Fig. 1 D). While successful conversion of AAG into ATG by HDR will activate GFP, subsequent re-cutting and NHEJ-mediated repair may disrupt this sequence and shift the ATG out of frame or remove it entirely, resulting in loss of GFP-expressing cells. To prevent this, the experiment was repeated with an ssODN template instructing an additional, PAM disrupting mutation (Fig. 1 B). We now observed increased nucleotide substitution levels that did not decline at increased puromycin concentrations (Fig. 1C). This experiment demonstrates that a Cas9 target site of which the protospacer had been modified by HDR could still be cleaved by Cas9, and repaired in an error prone fashion. However, this can be prevented by the introduction of an additional Cas9 target site disrupting (TSD) mutation.

Hide RNAs

The introduction of a PAM mutation is highly effective in preventing re-cutting, but is often undesirable, e.g., when it causes a non-synonymous codon substitution. We therefore sought alternative ways to protect the successfully modified target site from promiscuous Cas9 activity. We hypothesized that a guide RNA with a truncated spacer sequence that perfectly matches the modified target sequence may promote tight binding of inactive Cas9 to the mutated sequence, thus conferring effective protection against re-cutting (Fig. 2A). To investigate if this strategy indeed prevents re-cutting and disruption of the repaired GFP reporter, we designed 4 truncated gRNAs that we call “hide RNAs” (hRNAs). The hRNAs consisted of truncated spacer sequences that were on the same strand as the original full-length gRNA, but included the AAG>ATG mutation (Fig. 2B). The protospacers of hRNAs 1 and 2 (length 12 nt), were associated with the same PAM as the gRNA, whereas hRNAs 3 (10 nt) and 4 (15 nt) were associated with a different PAM (Fig. 2B).

The four hRNAs were cloned separately into the vector from Fig. 1 B adjacent to the gRNA expression cassette (Fig. 2C). Together with an HDR template instructing the ATG-creating nucleotide change but not the PAM-disrupting nucleotide change, we introduced these vectors into mESCs and varied Cas9/gRNA/hRNA levels by selecting with 3 different puromycin concentrations (Fig. 2C). We then compared the number of GFP-positive cells to that obtained with the experimental conditions from Fig. 1 B (only gRNA + ATG or ATG+PAM ssODN). Strikingly, we observed that the presence of hRNAs strongly improved the recovery of GFP-positive cells with the ATG ssODN, reaching levels as high as obtained with the ATG+PAM ssODN (Fig. 2D). We calculated a “protection score” for each hRNA, 0 being no protection and 1 being as much protection as a PAM disrupting mutation (Fig. 2E). While we found protection scores around 0.5 for the different hRNAs at 1.8 pg/ml puromycin, protection was maximal for the highest puromycin concentration tested. This demonstrates that the introduced ATG mutation could effectively be protected against re-cutting by any of the 4 hRNAs, obviating the requirement for an additional PAM-disrupting mutation.

The four hRNAs tested seemed to perform equally well, despite being associated with two different PAMs and having different protospacer lengths. To identify parameters for best hRNA protection, we designed additional hRNAs for the PAM used by the original gRNA sequence, and two other PAMs on the same strand (Fig. 3A) and varied the length of the spacer. The protection scores (Fig. 3B) of these hRNAs were determined in the same way as in Fig. 2E. At intermediate puromycin concentration (3.6 pg/ml), most hRNAs had protection scores around 0.6. At the highest puromycin concentration (10.8 pg/ml), we observed protection scores approaching 1 for hRNAs associated with the two PAMs closest to the mutation. Spacer length seemed less critical (8-15 nucleotides were equally effective). The protection score of hRNAs associated with the PAM most distal from the mutation was not increased at this puromycin concentration. Interestingly, a sharp cut-off was seen for hRNAs with spacer lengths >15 nt, showing no protection (16 nt) or even negative protection scores (17 nt). Possibly this reflects a DSB inducing capability of these hRNAs or increased binding to the non-modified sequence causing competition with the gRNA. Hide RNA-bound Cas9 is unavailable for binding gRNA. Thus, competition between hRNA and gRNA binding to Cas9 may reduce Cas9:gRNA re-cutting activity, irrespective of the protospacer sequence of the hRNA. To determine if this was the case, we tested the protection efficiency of 6 hRNAs with protospacer sequences unrelated to the original target site. Two hRNAs showed approximately 20% of the full protection obtained with a PAM-disrupting mutation, albeit at only one of the puromycin concentrations used (Figure 7). This indicated that occasionally competition for Cas9 binding between hRNA and gRNA may lower re-cutting incidence, although the contribution of this mechanism is modest. These results suggest that the requirement for effective shRNAs are relaxed: different spacer lengths and different PAMs perform equally well against re-cutting. In many cases, only the PAM that is part of the original target site will be available and this offers good protection. Protection seems most efficient at high nuclease concentrations.

Protection of a mutation outside the original target site

We next investigated whether hRNAs could also protect mutations that do not modify the original Cas9 target site. We designed an ssODN repair template to insert a TGG triplet that creates an in-frame ATG upstream of the gRNA target site thus leaving the PAM and protospacer sequence intact (Fig. 4A). We designed a similar oligonucleotide template with additional PAM disrupting mutations (ATG+PAM). We hypothesized that HDR templated by the ‘ATG’ only ssODN would result in GFP expression, but subsequent re-cutting in the downstream unaltered Cas9 target site may result in end- joining-mediated disruption of GFP expression. Indeed, compared to the unprotected ATG template, the ‘ATG+PAM’ template increased the recovery of GFP-positive cells approximately 2 fold (Fig. 4B). We designed two hRNAs (A and B) using different PAMs (Fig. 4A) and examined if these were able to prevent re-cutting. Strikingly, we observed protection by hRNA A was as efficient as that obtained with the ATG+PAM template (Fig. 4B). hRNA B, which utilized a PAM that was 3 bp further away did not protect the modified allele from re-cutting. These results demonstrate that hRNAs can be utilized to protect mutations that are created outside of the Cas9 target site.

Hide RNA effectivity at endogenous genes

We next assessed the generalizability of hRNAs for the protection of small mutations introduced with CRISPR/Cas9. We selected 15 naturally occurring variants in the human DNA MMR genes MSH2, MSH6 and MLH1 that can be generated by 1 nt changes that also create Type II restriction enzyme sites. For each mutation, guide RNAs were designed to introduce a DSB close to the intended mutation. We designed ‘unprotected templates’: 90 nucleotide long ssODN templates identical to the target sequence except for the desired single nucleotide substitution (SNS). We also designed templates with a target site disrupting mutation (TSD), which were identical to the SNS templates but included additional PAM or protospacer disrupting mutations (exemplified in Fig. 5A). These additional mutations did not disrupt the restriction enzyme site, and where within 5 bp of the DSB, as we previously found that longer distance could negatively affect introduction efficiency. TSDs were synonymous in 8 cases, non-synonymous in 5 cases and affecting an intronic sequence in 2 cases. In addition, we designed hRNAs for protection of each SNS. These hRNAs had the same PAM sequence as the gRNA and their protospacer was the shortest 10-15 nucleotide sequence preceding the PAM that started with a guanosine. If no such sequence existed, a 12 or 14 nucleotide sequence was selected and provided with a guanosine on the 5’ side. Cassettes expressing these hRNAs were cloned into their respective gRNA vectors to yield 14 vectors that expressed a gRNA and a hRNA. We then transfected 293FT cells with gRNA expressing vectors and ssODN templates with the intended SNS only or with an additional TSD. In addition, we transfected the 293FT cells with the gRNA and hRNA polycistronic vector and the SNS-only template (Fig. 5B). After transfection, we selected for uptake of the vector with 7.2 mI/ml puromycin. PCR fragments containing the target locus were incubated with the appropriate restriction enzyme, and the fraction of DNA that contained the modification was quantified (Fig. 5C). Digestion of the target locus, indicating HDR events had taken place ranged from 1 to 12% (the percentage of digested DNA of the total DNA). For each locus, we calculated a max-normalized HDR efficiency for the three different conditions, which is shown in Fig. 5D.

In only two cases (MSH6 V878A and MSH2 H639Q), the recovery of HDR events did not benefit from TSD or hRNA protection. The remaining cases benefited from TSD (6/14 cases) and/or from hRNA protection (8/14 cases). Somewhat remarkably, only 2 cases overlapped, i.e. , benefited from both, TSD and hRNA protection (MLH1 K52E and G67V). Thus, in 4 cases (MSH2 E188Q, MLH1 Q346R, MSH2 T33P and MSH6 G39E) hRNAs did not reach the improvement achieved by TSD. Conversely, in 7 cases (MLH1 K52E, N338S, F99I, G67V and L555P, and MSH6 P623A, and MSH2 D38A) hRNAs offered the best protection, yielding 2-10-fold improvement over the unprotected situation. Importantly, while in two cases target site disruption seemed to reduce efficiency with respect to the unprotected situation (MSH6 V878A: p=0.07; MLH1 F99I; p=0.11), the addition of a hRNA either gave the same efficiency (6/14 cases) or improved the efficiency 2-10-fold (8/14 cases).

Hide RNAs increase the fraction of HDR in mouse zygotes Subtle gene modification in mice can nowadays efficiently be achieved by injecting zygotes with Cas9 ribonucleoprotein complexes targeting the genomic DNA along with oligonucleotides templating the site-specific modification. Also here, TSDs are often added to improve the recovery of gene-edited animals, but this may have undesired effects. We therefore investigated whether hRNAs could be used to increase the frequency of ssODN-templated single-nucleotide substitution in murine zygotes. We designed guide RNAs, oligonucleotides and hRNAs to generate three different point mutations in the murine Mlh 1 gene, representing variants of uncertain significance found in suspected Lynch Syndrome patients (Fig. 6A). Guide RNA and hRNA Cas9 ribonucleoprotein (RNP) complexes were assembled in separate reactions. For every point mutation, we performed two rounds of mouse zygote injections: guide RNA RNPs with oligonucleotide and guide RNA plus hRNA RNPs with oligonucleotide (Fig. 6B). Injected zygotes were cultured until blastocyst stage and the fraction of HDR modification among the recovered alleles was determined by Sanger sequencing. It was apparent that the majority of chromatograms consisted of multiple reads. To quantify the number of reads deriving from HDR, we algorithmically deconvoluted chromatograms with TIDER (Fig. 6C) and ICE (not shown). Both algorithms determine fractions of wildtype, HDR-derived and indel-containing sequences. There was considerable overlap between the group of blastocysts that could be successfully deconvoluted by either algorithm, but the number of deconvolutions with an R2 exceeding 0.8 was higher for the TIDER algorithm. The percentage of HDR alleles could be determined for 109 blastocysts out of 239 by both algorithms with an R2 of at least 0.8 for each individual chromatogram, whereas the agreement between the two algorithms had an R2 of 0.91. However, in some cases, TIDER detected HDR alleles up to 20% that were not seen with ICE.

The percentage of HDR alleles per blastocyst found by TIDER is shown in Fig. 6C. In most cases the percentage of HDR alleles was lower than 50%, indicating that HDR events or re-cutting events disrupting HDR alleles took place at or beyond the 2-cell stage. Both, TIDER and ICE have a detection limit around 5%. When we only consider HDR percentages above 5% as reliably indicating successful base substitution, according to both, TIDER and ICE, the D304V substitution strongly benefited from the co-injection of a hRNA, with the presence of HDR alleles above 5% increasing from 0 (0 out of 7) to ±45% of the blastocysts (Fig. 6C). The benefit for R691 W and A21V was modest, if present, with TIDER suggesting a slight benefit for A21V (2/22 to 4/21) (Fig. 6C) and ICE suggesting a slight benefit for R691W (1/19 to 5/29). These results demonstrate that hRNAs not only can work with plasmid-based delivery of CRISPR/Cas9 components, but also with RNP complexes injected into zygotes, albeit not with 100% efficiency. The addition of a hRNA may thus be useful for the generation of transgenic mice carrying only a single base-pair substitution.

Discussion

We developed a novel tool for repression of promiscuous Cas9 activity, called hide RNA (or hRNA). hRNAs are gRNAs with a trimmed spacer that is complementary to the site to be protected. We explored and exploited the hRNA concept to prevent destruction of subtle base-pair substitutions introduced by CRISPR/Cas9 technology.

With Cas9, we induced a DSB and exploited ssODN-templated HDR to substitute a single bp. However, even when located in the protospacer that constituted the original target site, such minor modifications do usually not prevent targeting by the gRNA:Cas9 complex, which may lead to re-cutting and destruction of the modified allele. We reasoned that the modified allele may resist re-cutting when it is shielded by an inactive Cas9 RNP complex rendering it inaccessible to catalytically active Cas9 RNP. An inactive complex could be formed by Cas9 and a gRNA with a trimmed spacer, and because of perfect matching could bind the modified site with greater affinity than the mismatched gRNA:Cas9 complex. Using a disabled GFP gene as reporter for HDR-mediated gene editing, we found that four different hRNAs expressed from a polycistronic vector that also encoded Cas9 and the gRNA efficiently protected the intended single nucleotide substitution against re-cutting. We discovered that protection was conferred by hRNAs with protospacer lengths of 8-15 nucleotides that were associated with the same PAM as the gRNA, or with PAMs more downstream.

This strategy also worked to protect nucleotide changes introduced outside the original target site. However, protection was only conferred by a hRNA using a proximal PAM, but not by another hRNA that relied on a more distal PAM. Achieving protection in this situation may be more challenging as the hRNA faces stronger competition from the gRNA, which still has a perfect match with the target sequence. To improve the efficacy of hide RNA protection, the activity of the gRNA may be attenuated by introducing mismatches in its spacer sequence.

To investigate its general applicability, we studied the effect of hRNAs on the introduction of 14 SNSs in three endogenous genes in a human cell line. Each SNS was introduced within the protospacer sequence. We compared the protection efficiency of the hRNA to that conferred by the introduction of additional, target site disrupting point mutations (TSD) (protospacer- or PAM-disrupting mutations). We observed protection by the hRNAs in 8/14 cases and by TSD in 7/14 cases. Importantly, the introduction of TSD mutations occasionally lowered the introduction efficiency compared to the situation without protection, whereas addition of a hRNA never lowered the efficiency.

We used a polycistronic vector system for delivering Cas9, the gRNA and the hRNA to the cell. This ensures a 100% co-transfection efficiency of the gRNA and the hRNA, but the amount of Cas9 protein is the same as that expressed with one guide RNA. Because of this, two additional processes in our experimental system could enact protection against re-cutting in a spacer-independent way. The first is a ‘sponge’ effect, whereby hRNA binding decreases the concentration of Cas9 available for guide RNA binding, thus lowering the concentration of catalytically active Cas9 complex. We have addressed this possibility (Fig. 7) by expressing 6 hRNAs with a nonsense protospacer, and indeed observed some protection. However, the level of protection was only 20% of that of spacer-matched hRNA-protection, and was seen with only two out of six hRNAs and at one puromycin concentration. The other process that could decrease re-cutting activity is related to cytotoxic or growth-retarding effects incited by the binding of large numbers of inactive Cas9 complexes to the DNA. Upon expression of some, but not all hRNAs, we noticed decreased numbers of living cells (Fig. 8). If progressively higher cellular hRNA expression leads to decreasing cell viability, the average cellular concentration of catalytically active RNP is decreased in the surviving cell population, with an according reduction in re-cutting propensity. However, we did not observe a correlation between protection efficiency and cell numbers, indicating this effect did not significantly contribute to hRNA protection. These results suggest that hRNA activity is largely spacer dependent and effectuated by physical protection of the modified target site.

Studies of Cas9’s structure and biochemistry are uncovering the mechanism that governs its interaction with DNA in the presence and absence of spacerprotospacer mismatches. Upon binding to a gRNA, conformational changes activate Cas9 and enable it to scan DNA for PAM sequences. When a PAM is encountered, Cas9 locally melts the DNA allowing spacer-protospacer hybridization of sequences closest to the PAM. If this hybridization is successful, the full length of the spacer is annealed, followed by nuclease domain activation and DSB induction. Target binding has a higher tolerance for mismatches than cleavage. Activation of the nuclease activity requires a repositioning of the Histidine/Asparagine (HNH)- containing domain that enacts a conformational change in the RuvC domain. The switch of the HNH domain is slower when Cas9 is complexed with a mismatching guide RNA, or when complexed with a gRNA with a 17 nt spacer. Binding without break induction has been observed for gRNAs that have a protospacer even shorter than this, both in vitro and in vivo. The two DNA strands bound by Cas9 are distinguished as the target (gRNA bound) and the non-target (PAM containing) strands. In vitro studies have demonstrated that the binding of Cas9 protects 24 nt of target and non target DNA from DNAse I, including residues composing the PAM. These findings support our working model in which the hRNA-bound Cas9 complex is able to bind DNA, but unable to activate the nuclease domains. This inactive complex hides the PAM residues from gRNA-bound Cas9, and so interferes with the first step of Cas9’s cleavage mechanism. Based on transcription stimulating ability as readout for Cas9 binding strength, hRNAs are expected to exhibit the same binding strength as guide RNAs on the same target DNA. However, the original target sequence has the highest affinity for catalytically active Cas9/gRNA complex, while the modified sequence preferentially binds hRNA-bound inactive Cas9, which likely augments protection from re-cutting.

Investigating protection efficiency across different loci revealed that hRNA- mediated protection was at least as efficient as TSD-mediated protection, but the two types of protection did not always overlap and in some cases, hRNA protection did not work. The exact parameters that govern efficient hRNA binding protection are at this moment unclear. It seems that the length of the hRNA spacer is of little influence (Fig. 3). Likely, hRNA protection efficacy depends on the relative binding strengths of the hRNA and the gRNA. This balance is probably sequence dependent and influenced by the specific spacer: protospacer mismatches resulting from successful gene editing or deliberately designed in the gRNA. The increasingly better methods available for prediction of gRNA binding efficacy should also be useful for predicting hRNA binding efficacy, so that hRNA protection efficiency in different gene editing situations may be optimized.

The polycistronic vector system ensures co-transfection of hRNA and guide RNA, but it has been noted that plasmid-based delivery systems of Cas9 can result in log-fold differences in cellular RNP concentrations as a consequence of low and high plasmid copy numbers, the latter being associated with increased off-target activity. This is one of the main reasons for the increasing tendency in the field to deliver CRISPR/Cas9 nucleases directly as an in vitro assembled RNP complex. This has several advantages, such as a more straightforward workflow, a more homogeneously distributed transfection efficiency and a complete elimination of the risk of integration of vector-derived sequences. In fact, direct injection of RNP complexes and DNA templates into zygotes is currently the most popular route towards the generation of genetically-modified mice. We demonstrate here that also in this application the addition of hRNA-containing RNP can facilitate the recovery of genetically modified embryos.

In addition to protecting templated mutations created by HDR, the hide RNA concept has more applications. It is becoming increasingly appreciated that template- free repair can produce desirable outcomes. However, these outcomes may be subject to re-cutting too, as protospacers with small indels compared to the spacer can be targeted. It is envisaged that hide RNAs can be applied in these protocols to increase the frequency of a desirable template-free repair event. In fact, we envisage that hide RNAs may be applied in any protocol aimed at creating a specific mutation, including base-editing and prime-editing. Base-editors consist of a nuclease-dead Cas9 fused to a mutagenic protein. When provided with a guide RNA, the base-editor will install a mutation at a certain position in the protospacer targeted by the guide RNA. However, while the desired mutation is produced with high frequency, other bases in the vicinity may be mutated simultaneously or in subsequent rounds. Hide RNAs could offer a potential solution to avoid the latter, provided they are delivered as in vitro assembled hide RNA:Cas9 RNPs. Yet another hide RNA application is suppression of genome wide off-target activity. The problem of Cas9 off-target activity has been broadly recognized as it can cause undesired gene disruptions with unexpected effects. Therefore, different approaches to increase the stringency of target site recognition have been developed. Guided by Cas9 crystal structures, variants of Cas9 have been engineered that exhibited decreased off-target activity but with limited effect on on- target activity. Biochemical studies revealed a domain involved in proofreading of the RNA:DNA heteroduplex, and modification of this domain was able to greatly increase Cas9 accuracy. Furthermore, a Cas9 variant has been developed through phage- assisted continuous evolution with relaxed PAM sequence requirements and low off- target activity. Guide RNA modification can also reduce off-target activity. These include: the attachment of two guanosines to the 5’ end of the gRNA, including DNA nucleotides in the guide RNA, or changing its structure. We speculate that hide RNAs could be developed into gatekeepers against particularly problematic off-target activity in state-of-the-art Cas9 protocols. As such, it will be appreciated by the skilled person that the specific embodiments and applications as disclosed herein are similarly suitable for protection of Cas9 target sites and of Cas9 non-target sites (i.e. outside of the target site of Cas9) against modification of an on-target and/or an off-target site. Although the rapidly developing gene editing field has created impressive feats of engineering that improved the recovery of single base-pair substitutions and decreased off-target DSB induction, ssODN-templated repair of Cas9-induced breaks remains a versatile and efficient method. The hide RNA approach we disclose here, makes use of ‘classic’ CRISPR/Cas9 components and can easily be implemented in the currently mostly-used and optimized protocols.

Having now fully described this invention, it will be appreciated by those skilled in the art that the same can be performed within a wide range of equivalent parameters, concentrations, and conditions without departing from the spirit and scope of the invention and without undue experimentation.

While this invention has been described in connection with specific embodiments thereof, it will be understood that it is capable of further modifications. This application is intended to cover any variations, uses, or adaptations of the inventions following, in general, the principles of the invention and including such departures from the present disclosure as come within known or customary practice within the art to which the invention pertains and as may be applied to the essential features hereinbefore set forth as follows in the scope of the appended claims.

All references cited herein, including journal articles or abstracts, published or corresponding patent applications, patents, or any other references, are entirely incorporated by reference herein, including all data, tables, figures, and text presented in the cited references. Additionally, the entire contents of the references cited within the references cited herein are also entirely incorporated by references.

Reference to known method steps, conventional methods steps, known methods or conventional methods is not in any way an admission that any aspect, description or embodiment of the present invention is disclosed, taught or suggested in the relevant art.

The foregoing description of the specific embodiments will so fully reveal the general nature of the invention that others can, by applying knowledge within the skill of the art (including the contents of the references cited herein), readily modify and/or adapt for various applications such specific embodiments, without undue experimentation, without departing from the general concept of the present invention. Therefore, such adaptations and modifications are intended to be within the meaning and range of equivalents of the disclosed embodiments, based on the teaching and guidance presented herein.

It is to be understood that the phraseology or terminology herein is for the purpose of description and not of limitation, such that the terminology or phraseology of the present specification is to be interpreted by the skilled artisan in light of the teachings and guidance presented herein, in combination with the knowledge of one of ordinary skill in the art.

Claims

1. A method of modifying a chromosomal sequence in a cell, the method comprising the steps of: a) introducing into the cell a Clustered Regularly Interspersed Short Palindromic Repeats (CRISPR)/CRISPR-associated (Cas) (CRISPR/Cas) endonuclease or a nucleic acid encoding the CRISPR/Cas endonuclease; b) introducing into the cell guide RNA, or a nucleic acid encoding the guide RNA, wherein the guide RNA comprises a DNA-targeting segment and a CRISPR/Cas endonuclease-binding segment, wherein the CRISPR/Cas endonuclease-binding segment comprises a tracrRNA, and wherein the DNA- targeting segment comprises a complementary region that is substantially complementary to a region at a target site in the chromosomal sequence in the cell and that can base pair with the region at the target site, wherein the region at the target site is upstream of a protospacer adjacent motif (PAM) recognized by the CRISPR/Cas endonuclease, and wherein a protein-RNA complex is formed between the guide RNA and the CRISPR/Cas endonuclease, the guide RNA guides the CRISPR/Cas endonuclease to the target site, the CRISPR/Cas endonuclease introduces a single-stranded break or a double-stranded break at the target site, and a cellular DNA repair process repairs the single-stranded break or double- stranded break, thereby introducing a modification at the target site, wherein the method further comprises the step of c) introducing into the cell hide RNA, or a nucleic acid encoding the hide RNA, wherein the hide RNA comprises a DNA-targeting segment and a CRISPR/Cas endonuclease-binding segment, wherein the CRISPR/Cas endonuclease-binding segment comprises a tracrRNA, and wherein the DNA- targeting segment comprises a complementary region that is in part identical to the complementary region of the DNA-targeting segment of the guide RNA and that is substantially complementary to the region at the target site in the chromosomal sequence in the cell wherein the modification has been introduced and that can base pair with the region at the target site that wherein the modification has been introduced, and wherein the complementary region in the hide RNA is shorter in length than the complementary region in the guide RNA.

2. The method of claim 1 wherein the method further comprises the step of d) introducing into the cell a donor polynucleotide comprising a donor sequence comprising the modification to be introduced at the target site, preferably wherein the donor sequence is flanked by upstream and downstream sequences that are identical to upstream and downstream sequences, respectively, of the target site in the chromosomal sequence in the cell.

3. The method according to any one of the previous claims wherein the cellular DNA repair process repairs the single-stranded break or double stranded break such that the chromosomal sequence in the cell is modified by deletion of at least one nucleotide, an insertion of at least one nucleotide, a substitution of at least one nucleotide, or a combination thereof.

4. The method according to any one of the previous claims wherein the DNA repair process is a non-homologous end-joining repair process, or, when a donor polynucleotide is used, wherein the DNA repair process is a homology-directed repair.

5. The method according to any one of the previous claims wherein steps a), b), c), and d) are performed in any order, simultaneously, sequentially, or any combination thereof.

6. The method according to any one of the previous claims wherein the complementary region of the hide RNA is at least 1 , 2, 3, 4, 5, 6, 7, 8, 9, 10, 11 , 12 or more nucleotides shorter than the complementary region of the guide RNA.

7. The method according to any one of the previous claims wherein the complementary region of the hide RNA has a length of between, and including, 6 - 16 nucleotides, preferably between, and including, 8 - 15 nucleotides.

8. The method according to any one of the previous claims wherein the guide RNA comprises at least one nucleotide in the complementary region that is not complementary to the cognate nucleotide in the region at a target site

9. The method according to any one of the previous claims wherein the donor polynucleotide comprising the donor sequence comprises a further modification to be introduced, wherein the modification is to be introduced in the region at the target site that is substantially complementary to the complementary region in the guide RNA.

10. The method according to any one of the previous claims wherein the donor polynucleotide is single stranded DNA, double stranded DNA or RNA.

11. The method according to any one of the previous claims wherein more than one different hide RNAs are introduced in the cell.

12. The method according to any one of the previous claims wherein the DNA-targeting segment and the CRISPR/Cas endonuclease-binding segment are arranged in the 5' to 3' direction or in the 3’ to 5’ direction.

13. The method according to any one of the previous claims wherein the guide RNA and/or the hide RNA is a single molecule or is comprised of two molecules.

14. The method according to any one of the previous claims wherein the CRISPR/Cas endonuclease in introduced into the cells in the form of a ribonucleoprotein complex, wherein the ribonucleoprotein complex comprises the CRISPR/Cas endonuclease and the guide RNA.

15. The method according to any one of the previous claims wherein the CRISPR/Cas endonuclease introduces a double-stranded break, and wherein the CRISPR/Cas endonuclease is selected from the group consisting of Staphylococcus aureus Cas9, Streptococcus pyogenes Cas9, Neisseria meningitides Cas9, Streptococcus thermophiles Cas9, preferably Streptococcus pyogenes Cas9.

16. The method according to any one of the previous claims wherein the CRISPR/Cas endonuclease introduces a single stranded break and is selected from the group consisting of CAS9_D10A, CAS9_H820A, CAS9_H839A, and Cpf1_R1226.

17. The method according to any one of the previous claims, wherein the CRISPR/Cas endonuclease is Streptococcus pyogenes Cas9, the complementary region of the guide RNA has a length of between, and including 17 - 20 nucleotides, and the complementary region of the hide RNA has a length of between and including 8 - 15 nucleotides.

18. The method according to any one of the previous claims wherein the nucleic acid encoding the CRISPR/Cas endonuclease is codon optimized for expression in the cell.

19. The method according to any one of the previous claims wherein the CRISPR/Cas endonuclease is linked to at least one nuclear localization signal (NLS).

20. The method according to any one of the previous claims wherein the method does not comprise a process for modifying the germ line genetic identity of a human being and/or wherein the method does not comprise a method for treatment of the human or animal body by surgery or therapy.

21. A hide RNA as defined in any one of the previous claims for use as a medicament.

22. A hide RNA as defined in any one of the previous claims in the treatment of a subject, wherein the treatment comprises the modification of a chromosomal sequence in a cell of the subject.

23. A cell comprising comprising a hide RNA as defined in any one of the previous claims or a nucleic acid encoding a hide RNA as defined in any one of the previous claims, and a guide RNA as defined in any one of the previous claims or a nucleic acid encoding a guide RNA as defined in any one of the previous claims, and a CRISPR/Cas endonuclease or a nucleic acid encoding a CRISPR/Cas endonuclease, and optionally a donor polynucleotide as defined in claim 2.

24. A kit of parts comprising a hide RNA as defined in any one of the previous claims or a nucleic acid encoding a hide RNA as defined in any one of the previous claims, and a guide RNA as defined in any one of the previous claims or a nucleic acid encoding a guide RNA as defined in any one of the previous claims, and optionally a donor polynucleotide as defined in claim 2.

25. Kit of parts according to claim 24, wherein the kit further comprises a

CRISPR/Cas endonuclease or a nucleic acid encoding a CRISPR/Cas endonuclease.