DESCRIPTION

The invention relates to the field of molecular biology, in particular genetic engineering. The invention relates to a method for modifying double stranded DNA employing an RNA guided DNA endonuclease to generate two double strand breaks in the dsDNA molecule to be modified, and replacement of the sequence positioned between the double strand breaks with a substitute DNA sequence using the non-homologous end joining (NHEJ) pathway. The invention also relates to corresponding kits and compositions for modifying double stranded DNA molecules.

The invention relates in particular to an in vitro method for modifying a double stranded DNA (dsDNA) molecule in a cell, the method comprising a) introducing into the cell i) an RNA guided DNA endonuclease or a nucleic acid encoding an RNA guided DNA endonuclease, ii) at least one guide RNA, and iii) an exogenous nucleic acid molecule comprising or encoding a DNA substitute sequence, b) generating at least two double strand breaks of the dsDNA molecule to be modified, wherein i) the dsDNA molecule comprises at least two target sequences (target sequences 1 and 2) which are targeted by the at least one guide RNA, and ii) the at least two double strand breaks occur within or adjacent to the at least two target sequences, c) replacing a DNA sequence of the dsDNA molecule to be modified, wherein the replaced sequence is located between the double strand breaks (between target sequences 1 and 2) and is replaced by the DNA substitute sequence of the exogenous nucleic acid molecule by the non- homologous end joining (NHEJ) pathway.

BACKGROUND

With the rapid discovery and implementation of RNA guided nucleases such as SpCas9 or CPF1 , the ability to probe biological questions and engineer biological systems is quickly expanding (Barrangou, R., and Doudna, J.A. (2016). Nat. Biotechnol. 34, 933-941 ). The use of such nucleases was originally shown to form a double strand break that would be rejoined and in some cases mutated by the non-homologous end joining (NHEJ) pathway to create small insertions and deletions (INDELs) that would cause damage to the loci and inhibit gene function, often to create protein knock-outs (Lieber, M.R. (2010). Annu. Rev. Biochem. 79, 181-21 1 ). This non- homologous end joining pathway is able to ligate DNA in all parts of the cell cycle.

Additionally, early studies showed that when the RNA-guided nuclease cuts the DNA in the presence of a“repair template” the cells can undergo homologous recombination using the supplied template DNA (Mali, P., et al. (2013). Science 339, 823-826); Jinek, M.,et al. (2013). Elife 2, e00471 ; Heyer, W.-D., et al. (2010). Annu. Rev. Genet. 44, 1 13-139). This can create a precise insertion or a replacement of sequence. Initially the template was supplied as double stranded DNA, but it was also shown that single stranded DNA with homology arms can also be used to replace or add DNA sequences, using single strand template repair (SSTR) (Chen, F., et al. (201 1 ). Nat. Methods 8, 753-755). However, the replacement of sequences by knock-ins using homologous recombination or SSTR are frequently reported to be of low efficiency, raising the challenge of whether these pathways for DSB repair can be further improved or replaced (Danner, E. et al (2017), Mamm. Genome, 28 262-274). Moreover, homologous recombination and SSTR require enzymatic processes that are relegated to the late S/G2 phases of the cell cycle and are therefore restricted to dividing and cycling cells. This ablates the ability to produce knock-ins in non-dividing or slowly dividing cells.

Replacement of DNA sequences located between two DSBs by a different DNA sequence by means of the NHEJ repair pathway has been described for systems employing Zinc Finger Nucleases (ZFNs) (Orlando S.J. et al. (2010). Nucleic Acids Research, Vol. 38, No. 15;

Weinthal D.M. et al. (2013). Plant Physiology, Vol. 162, pp. 390-400). The use of Transcription Activator-like Effector Nuclease (TALENs) in such methods has also been suggested

(US9005973B2).

However, the described replacement of DNA sequences by means of the NHEJ repair pathway and ZFNs is typically inefficient, exhibiting replacement rates of only around 5 %. Furthermore, such methods depend on the generation of complementary single strand overhangs at the open ends of the DNA molecule to be modified and at the ends of the insert sequence. A further disadvantage of these replacement methods is that it is time-consuming to generate sequence specific ZFN and TALEN enzymes, and the use of known enzymes reduces the flexibility of the user to select suitable sites for generating DSBs. Accordingly, the methods of the prior art with respect to NHEJ-mediated sequence replacement are hampered by poor specificity and flexibility regarding election of insertion sites and require time- and cost-intensive preparation. Furthermore, the methods employing ZFNs are focused primarily on DNA replacement in plant cells, where HDR is inefficient. The application of such methods in mammalian cells is also of low efficiency.

Furthermore, NHEJ repair has been shown to ligate exogenously provided DNA segments in between the ends of a single DSB. For example, methods for homology-independent knock-in of a donor sequence into a single CRISPR/Cas9-generated DSB in zebra fish (Auer et al Genome Research 2013, vol. 24, no. 1 , 31 , 142-153) have been described. However, in approaches such as this, and also as shown for similar approaches in human cells, designated as Non-Homology (NH) Targeting (He X et al. 2016, Nucleic Acids Res 44:e85), the orientation of the inserted fragments is stochastic and cannot be controlled. This stochastic insertion has been recently overcome by a modification in the technique that allows a high degree of preference in the insertion orientation (Suzuki et al. (2016) Nature 540:144-149). This

Homology-Independent Targeted Integration (HITI) method was shown to be effective in primary cells in vitro and in vivo, suggesting that c-NHEJ can be a method for achieving targeted integrations.

Furthermore, exon replacement in rice cells through NHEJ using CRISPR/Cas9 mediated- cleavage of genomic DNA employing two site-specific guide RNAs has been described (Jun Li et al, Nature Plants, vol. 2, no. 10, 2016, 16139). However, this method revealed a low efficiency for correct exon replacement. NHEJ-based methods of the prior art such as those described above were not primarily designed for purposes that aim to replace a DNA sequence in a targeted and controlled manner. Due to the known problem of the uncontrolled and frequent generation of insertions and deletions at the DSB site, these methods are primarily intended for introducing genetic disruptions or insertions and have essentially been disqualified for targeted DNA editing approaches that are geared towards a controlled/pre-designed sequence replacement.

Importantly, previous references to NHEJ for ligating exogenously provided DNA segments between the ends of a DSB generated by a CRISPR/Cas system in vertebrate or mammalian systems are limited to the presence of a single DSB in the DNA molecule to be modified. As such, the methods of the prior art are limited to insertion events, in which DNA segments are introduced, but no sequence segment is removed or replaced from the dsDNA molecule to be modified. The previous technology is therefore seriously limited in the type of modifications possible to the DNA molecule to be modified. The targeted and efficient replacement of a portion of the target dsDNA molecule via CRISPR/Cas systems and the NHEJ mechanism for the purpose of precise genome editing was until now not thought to be possible.

Despite advances in methods for editing DNA molecules in a cell, improvements are needed that would overcome the disadvantages of the prior art.

SUMMARY OF THE INVENTION

In light of the prior art the technical problem underlying the present invention is the provision of alternative and/or improved means and methods for modifying dsDNA in cells. In particular, improved means for modifying the genome of eukaryotic cells, such as e.g. mammalian or vertebrate cells, are needed. More specifically, the provision of means for the replacement of a genomic sequence segment by a new segment harbored on a donor DNA molecule, especially in non-dividing or slowly dividing cells, is one aim of the present invention.

This problem is solved by the features of the independent claims. Preferred embodiments of the present invention are provided by the dependent claims.

The invention relates to a method for modifying double stranded DNA employing an RNA guided DNA endonuclease to generate two double strand breaks in the dsDNA molecule to be modified, and replacement of the sequence positioned between the double strand breaks with a substitute DNA sequence using the non-homologous end joining (NHEJ) pathway. The invention also relates to corresponding kits and compositions.

The invention therefore relates to an in vitro method for modifying a double stranded DNA (dsDNA) molecule in a cell, the method comprising:

introducing into the cell

i. an RNA guided DNA endonuclease or a nucleic acid encoding an RNA guided DNA endonuclease,

ii. at least one guide RNA, and

iii. an exogenous nucleic acid molecule comprising or encoding a DNA substitute sequence, generating at least two double strand breaks of the dsDNA molecule to be modified, wherein

i. the dsDNA molecule comprises at least two target sequences (target sequences 1 and 2) which are targeted by the at least one guide RNA, and ii. the at least two double strand breaks occur within or adjacent to the at least two target sequences,

replacing a DNA sequence of the dsDNA molecule to be modified, wherein the replaced sequence is located between the double strand breaks (between target sequences 1 and 2) and is replaced by the DNA substitute sequence of the exogenous nucleic acid molecule by the non-homologous end joining (NHEJ) pathway.

It was entirely surprising that it is possible to efficiently replace a sequence of a dsDNA molecule that is located between two cut sites or double strand breaks generated by an RNA guided endonuclease by a substitute DNA sequence that is ligated to the dsDNA molecule to be modified by means of the non-homologous end joining pathway in a targeted and controlled manner.

An entirely surprising effect or result of the method of the invention is that the generation of unwanted and/or uncontrolled DNA sequence modifications at the site of DNA sequence replacement can be largely avoided, although the method relies upon the NHEJ pathway. The method of the invention solves the problem of unwanted modifications frequently observed in DNA editing techniques that employ NHEJ repair. In a preferred embodiment, the replacement of a DNA sequence of the dsDNA molecule to be modified located between the at least two double strand breaks by the DNA substitute sequence by NHEJ occurs without the generation of INDELs or sequence modifications at the sites of the at least two DSBs.

It is unexpected for a person skilled in the art that a DNA sequence can be substituted or replaced by an exogenous DNA sequence using an RNA guided endonuclease and the NHEJ mechanism with high precision and efficiency without the generation of INDELs at the ligation sites, in the majority of sequence replacement events. Additionally, the desired, targeted replacement occurs with a high frequency in the method of the present invention. Furthermore, the method of the invention is relatively straightforward and user friendly, as it does not require the laborious design and generation of donor DNA molecules comprising homology sequences or suitable single strand overhangs, as required for the DNA replacement methods of the prior art. The present invention therefore opens a new field for the creation of pre-designed, controlled and targeted DNA replacement strategies using the NHEJ pathway, which is efficient, safe and user friendly.

So far, to the knowledge of the inventors, it was commonly accepted in the field that for targeted and controlled sequence replacement, the homology-directed repair (HDR) pathway has to be employed, which requires that the cells comprising the DNA to be modified must be cycling cells in S or G2 phase of the cell cycle. Furthermore, in HDR-mediated sequence replacement methods, a donor DNA molecule is typically provided that comprises homology arms flanking the DNA sequence to be inserted into the dsDNA molecule, wherein the homology arms share a high sequence identity with the DNA sequence adjacent to the cut sites of the DNA molecule to be modified, which enables physical interaction between the homology arms and the free ends at the double strand breaks.

According to the method of the present invention, two double strand breaks (DSBs) are generated in a dsDNA molecule present in a biological cell. The cut sites or locations of the DSBs are generated through cleavage by an RNA guided endonuclease, which is complexed with a guide RNA. The guide RNA leads the endonuclease to the target site where the RNA guided endonuclease is activated through the interaction of the guide RNA and the target sequence and the presence of a PAM sequence. The activated RNA guided endonuclease generates a DSB at or adjacent to the target sequence. The exact cut site of the RNA guided endonuclease and therefore the exact location of the DSB relative to the location of the target sequence and the PAM depends on the specific CRISPR/Cas system or RNA guided

endonuclease used in the method of the present invention. Multiple CRISPR/Cas systems and corresponding RNA guided endonucleases are known to one skilled in the art and may therefore be selected appropriately, and target sequences may therefore be designed or selected appropriately, for use in the present invention.

It was surprising in the context of the present invention that DNA sequence replacement is possible between two cut sites (DSBs) that have been generated by RNA guided endonucleases in a way that is completely independent of homology arms or homology sequences present on the DNA substitute sequence and the open ends of the cut DNA molecule to be modified.

Furthermore, no single strand sequence overhangs (sticky ends) need to be generated.

Especially, no complementary single strand overhangs on the open ends of the cut DNA molecule or on the DNA substitute sequence need be present, as in known replacement methods employing ZFNs for generating DSBs. A further advantage of the method of the present invention in comparison to ZFN-mediated replacement methods is the dramatically increased flexibility with respect to the choice of the target sites for generating the DSBs. In the context of the present invention, such cut-sites can essentially be freely chosen, whereas in ZFN-mediated methods suitable enzymes for a specific target site have to be engineered.

In some embodiments of the invention one guide RNA is used. In other embodiments of the invention two or more guide RNAs are used.

In embodiments employing only one guide RNA, the specific sequence of the guide RNA recognizes at least two target sequences, which might be identical or highly homologous to each other, leading to the generation of two DSBs via the use of a single guide RNA.

Through the generation of two DSBs in the dsDNA molecule to be modified the DNA sequence located between the two DSBs can be excised and removed from the dsDNA molecule, leaving two open ends of the dsDNA molecule that are not comprised by the excised sequence. These two open ends can be ligated through NHEJ to the two ends of a DNA substitute sequence, which may be a linear or linearized dsDNA molecule, which has been introduced into the cell by means of an external nucleic acid molecule comprising or encoding the DNA substitute sequence and which thereby replaces or“substitutes” the sequence that had initially been located between the sites of the two DSB that are defined by the two target sequences. Through ligation of the exogenous DNA substitute sequence, the target sequences of the dsDNA molecule may preferably be disrupted, preventing excision of the inserted substitute DNA sequence by means of the introduced RNA guided endonuclease and guide RNA. However, it is possible to design and/or select DNA substitute sequences and/or target sequences in such a manner, that the target sequences are only disrupted when the substitute DNA sequence has been ligated to the open ends of the dsDNA molecule in the desired orientation. This represents a great advantage of the present invention, which is based on the surprising observation that replacement of DNA sequences located between two DSBs by the DNA substitute sequence occurs primarily without the generation of INDELs.

Since the cut sites of the RNA guided endonuclease are known, it is possible to design and provide a linear dsDNA substitute sequence with nucleotide sequences at its ends that lead to either reconstitution or disruption of the target sequences upon ligation of the open ends of the dsDNA molecule to be modified and the DNA substitute sequence, depending on the orientation of the DNA substitute sequence after ligation. If the target sequences are reconstituted after ligation, the DNA substitute sequence can be excised again by the RNA guided endonuclease that is complexed with the guide RNA. This process can be repeated until the DNA substitute sequence has been inserted in the desired orientation leading to disruption or disappearance of the target sequence. This principle is further illustrated by Figure 1 1.

Furthermore, it is possible to design or select the two target sequences 1 and 2 comprised by the dsDNA molecule to be modified in a way that - after generation of the two DSBs and excision of the DNA sequence located between the DSBs - a target sequence and/or target site that is recognized by the at least one guide RNA is generated upon ligation of the open ends of the dsDNA molecule that are not located on the excised DNA sequence. This enables recognition of the newly generated target sequence by the complex of RNA guided endonuclease and guide RNA leading to cleavage at the site of ligation of the two open ends of the dsDNA molecule. This may lead to the regeneration of the open ends of the dsDNA molecule, which are again available for ligation to the DNA substitute sequence. Moreover, upon re-ligation of the excised DNA sequence of the dsDNA molecule, the target sequences may be reconstituted leading to repeated excision of the DNA sequence to be replaced from the dsDNA molecule. Preferably, a further guide RNA may be used that targets the protospacer site formed by the deletion between the two original target sites enabling reopening of the ligated DNA. This allows for more chances to ligate in the replacement sequence. Furthermore, if the target is inserted in the inverted direction, this guide can excise out the incorrectly ligated sequence.

Accordingly, it is possible to design and select target sequences and DNA substitute sequences that favor replacement of the DNA sequence of the dsDNA molecule located between the DSBs by a DNA substitute sequence in a specific and desired orientation over re-ligation of the excised sequence, ligation of the open ends of the dsDNA molecule and ligation of the DNA substitute sequence in an undesired orientation.

In the context of the present invention, in some embodiments, target sequence 1 and target sequence 2 may be the same (meaning identical) or different from each other. Furthermore, in some embodiments, target sequence 1 and target sequence 2 may be recognized by the same and/or different guide RNAs.

The use of two different target sequences may be advantageous in embodiments comprising the replacement of a genomic DNA sequence. However, in case of excision and replacement of repetitive sequences or gene duplications in the genome, two target sequences and/or target sites may be used that are identical or at least recognized by the same guide RNA.

Both the guide RNA and the RNA guided endonuclease may be introduced into the cell, either as a protein and an RNA molecule, respectively, or by introducing a nucleic acid molecule into the cell which encodes the RNA guided endonuclease and/or the guide RNA, allowing expression of the respective molecule inside the cell after introduction of the encoding nucleic acid. A person skilled in the art knows how to configure such nucleic acids with suitable promoters and other required or useful genetic elements.

The method of the present invention represents an improved and simplified method for sequence replacement in a dsDNA molecule located in a cell. Surprisingly, the method can be performed to modify any kind of dsDNA molecule present in a biological cell, such as, without limitation, a plant cell, a prokaryotic cell, a eukaryotic cell or an engineered artificial cell. The dsDNA molecule to be modified can be naturally present in the cell or can be introduced or inserted into the cell.

In some embodiments, the method of the invention is particularly suited to modify genomic DNA in eukaryotic cells, wherein the term genomic DNA comprises, for example, mitochondrial DNA and/or chromosomal DNA.

It was completely unexpected that it is possible to replace a DNA sequence located between two double strand breaks that are introduced into the dsDNA molecule to be modified through provision of an external DNA substitute sequence, such as a linear or linearized dsDNA molecule in a controlled manner. A skilled person would have expected that the two free or open ends of the dsDNA molecule that are not present on the excised DNA fragment (i.e. flanking the excised sequence of the dsDNA to be replaced) would be ligated with each other, leading to a simple deletion of the sequence positioned between the double strand breaks. Since the exogenous nucleic acid molecule and the DNA substitute sequence do not require sequences that are homologous to the target sites of the dsDNA molecule, a skilled person would not have thought that the exogenous DNA sequence will be present at the sites of both DSBs and will be available for ligation to the open ends of the dsDNA molecule through NHEJ, especially not when the open ends of the DNA molecule to be modified that should be ligated to the DNA substitute sequence are generated by two DNA independent DSBs that might be located at hundreds or thousands of base pairs distance from each other. Furthermore, it was surprisingly found that the method of the present invention is more efficient for site-specific sequence replacement compared to known HDR-mediated techniques.

Since NHEJ is also active in the G1 and GO phase of the cell cycle, it is a great advantage of the method of the present invention that sequence replacement can also be performed in nondividing or slowly dividing cells, which are not accessible to sequence replacement by HDR. In the context of the present invention, the dsDNA molecule to be modified may also be referred to as the dsDNA molecule, target molecule, target DNA molecule, target dsDNA or target dsDNA molecule.

The present invention can be applied for engineering and manipulating dsDNA molecules in biological cells. The dsDNA molecules to be modified can be an artificial molecule or a naturally occurring molecule. One application of the present invention is the correction or modification of genetic sequences of a cell. For example, an exon of a gene or a complete gene could be replaced by a DNA substitute sequence provided by an exogenous nucleic acid molecule. It is possible to correct genetic mutations that are associated with diseases by replacing a mutated sequence, for example an exon comprising a pathological mutation, by a corrected exon sequence without the respective mutation. Furthermore, the method can be used to generate cells that carry modified genetic sequences, wherein an endogenous gene or part of a gene is replaced by an artificial genetic sequence, which might have, for example, a therapeutic effect.

The cells and the dsDNA molecule to be modified comprised by the cell, such as for example the genomic dsDNA, can be manipulated by the method of the present invention in vitro or in vivo. In case of in vitro manipulation of cells, which might be used for therapeutic applications, through sequence replacement according to the present invention, the modified cells can be administered subsequently to a subject in need thereof. Furthermore, it is possible by employing suitable delivery vehicles such as viral vectors, such as AAV or non-integrating lentiviruses, to introduce the components of the present invention into a target cell in vivo.

Another application of the method of the present invention is genetic engineering of complex dsDNA molecules, as for example artificial chromosomes. By using the method of the present invention, complex replacement steps of DNA sequences in dsDNA molecules can be performed in a facilitated way as compared to previous approaches, involving HDR-dependent methods or classical molecular cloning strategies.

Importantly, the use of NHEJ for replacing a DNA sequence of the dsDNA molecule to be modified by a DNA substitute sequence is identifiable by analyzing the genetic scar resulting from the ligation of the DNA substitute sequence caused by insertions and deletion at the site of cleavage and ligation. As such, the replacing of a DNA sequence by the DNA substitute sequence by the non-homologous end joining (NHEJ) pathway can be determined after replacement has occurred using methods known to a skilled person, for example by sequencing the relevant region of the DNA molecule that has been modified.

Further, the delivered DNA for integration might have its own unique scarring pattern. The inserted sequence may always have this scarring pattern, except when the DNA is not linearized in the cell by an RNA guided endonuclease such as Cas9.

Preferably, the exogenous nucleic acid molecule is a circular DNA molecule, preferably a plasmid or mini-circle. The use of a circular plasmid DNA and a mini-circle molecule as an exogenous nucleic acid molecule is particularly advantageous since delivery of such molecules to target cells is well established and highly efficient, for example through transfection, lipofection or electroporation. Furthermore, such plasmids can be designed to comprise one or more target sequences that are recognized by the at least one guide RNA and are cleaved inside the cell by the complex of the RNA guided endonuclease and the associated guide RNA.

If the plasmid contains two target sequences, a linear dsDNA molecule serving as the DNA substitute sequence can be excised from the plasmid through generation of two DSBs.

Alternatively, the whole plasmid can be integrated into the dsDNA molecule to replace the sequence present between the cut sites if only one DSB in the plasmid is generated, leading to the provision of a linear dsDNA molecule.

The use of mini-circles as exogenous nucleic acid molecules is particularly advantageous because such molecules may substantially consist of the DNA substitute sequence in circular form, which can be linearized by generating a single DSB at the cut site marked by a target sequence comprised in the mini-circle sequence, which is recognized by the at least one guide RNA. Sequence replacement through provision of a mini-circle derived substitute sequence is particularly efficient compared to other exogenous DNA molecules serving as a source of the DNA substitute sequence.

Through the specific configuration, selection and/or localization of a target sequence comprised by the plasmid or mini-circle it is possible to direct a specific orientation of the DNA substitute sequence upon integration into the dsDNA molecule. A further advantage of plasmids as well as of mini-circles is the stability of these molecules, allowing prolonged storage before use in the method of the present invention.

According to a further embodiment of the method of the present invention, the exogenous nucleic acid molecule is a linear DNA molecule. This embodiment is particular advantageous because a linear dsDNA molecule does not have to be linearized and can serve as a DNA substitute sequence without further processing inside the cell. As such, PCR products or linear segments of synthetic DNA may be employed directly as the exogenous DNA molecule comprising or consisting of the substitute DNA sequence.

In a preferred embodiment, the linear DNA molecule serving as an exogenous nucleic acid molecule comprising the DNA substitute sequence is a PCR product. This is particularly advantageous because PCR fragments can be easily generated by amplifying the sequence of interest. Furthermore, PCR products are very stable and isolation, purification and delivery of PCR product is well established is well established.

According to a further preferred embodiment of the invention, the exogenous nucleic acid molecule is comprised by a viral vector, such as AAV or a non-integrating lentivirus. Delivery of the exogenous nucleic acid molecule through viral vectors is particularly advantageous for manipulation of dsDNA molecules in cells, in which delivery of exogenous nucleic acid molecules such as linear and circular dsDNA is difficult to achieve through other means, such as transfection, electroporation or microinjection. A skilled person is capable of determining the ability of the target cells to be transformed with an exogenous DNA molecule and is capable of selecting the form or structure of the exogenous DNA molecule appropriately. Transformation frequency using various vectors or structures of the exogenous DNA molecule can be measured and the transformation rate assessed accordingly.

Cells that are difficult to transform, in some embodiments, relate to slowly dividing or non-dividing cells. The frequency or rate of cell division can be established by a skilled person using known techniques, such as methods that measure proliferation of cells, for example by an estimation of DNA synthesis or the number of cells in cell culture in S phase. Direct measurement of cell proliferation may involve the incorporation of a labeled nucleoside into genomic DNA. Examples include the tritiated thymidine ([3H]dT) and BrdU (bromodeoxyuridine) methods.

Furthermore, the method of the invention allows to deliver nucleic acid molecules to cells that are inaccessible to other techniques, for example due to their localization. This relates in particular to cells that are present and integrated into complex structures such as organs or organisms or complex three-dimensional tissue culture systems.

Preferably, the DNA substitute sequence has a length of 25 bp to 1 Mega bp, 30 to 800.000, 40 to 600.000 bp, 50 to 500.000 bp, 60 to 400.000 bp, 70 to 300.000 bp, 80 to 200.000 bp, 90 to 100.000 bp, 100 to 90.000 bp, 1 10 to 80.000 bp, 120 to 70.000 bp, 130 to 60.000 bp, 140 to 50.000 bp, 150 to 40.000 bp, 160 to 30.000 bp, 180 to 20.000 bp, 200 to 10.000 bp, 220 to 9.000 bp, 240 to 8.000 bp, 260 to 7.500 bp, 280 to 7.000 bp, 300 to 6.500 bp, 320 to 6.000 bp, 340 to 5.500 bp, 360 to 5.000 bp, 380 to 4.500 bp, 400 to 4.000 bp, 420 to 3.800 bp, 440 to 3.600 bp, 460 to 3.400 bp, 480 to 3.200 bp, 500 to 3.000 bp, 520 to 2.800 bp, 540 to 2.600 bp, 560 to 2.400 bp, 580 to 2.200 bp, 600 to 2.000 bp, 650 to 1 .800 bp, 700 to 1.750 bp, 750 to 1.700 bp, 800 to 1.650 bp, 850 to 1 .650 bp, 900 to 1.600 bp, 950 to 1.550 bp, 1000 to 1 .500 bp, 1.050 to 1.450 bp, 1.100 to 1.400 bp, 1.150 to 1.350 bp, 1.200 to 1 .400 bp or 1.250 to 1.350 bp. It was shown that integration of sequences in the range of 200 to 10.000 bp is particularly efficient.

Furthermore, in preferred embodiments of the present invention the replaced sequence of the dsDNA molecule to be modified has a length of 25 bp to 1 Mega bp, 30 to 800.000, 40 to 600.000 bp, 50 to 500.000 bp, 60 to 400.000 bp, 70 to 300.000 bp, 80 to 200.000 bp, 90 to 100.000 bp, 100 to 90.000 bp, 1 10 to 80.000 bp, 120 to 70.000 bp, 130 to 60.000 bp, 140 to 50.000 bp, 150 to 40.000 bp, 160 to 30.000 bp, 180 to 20.000 bp, 200 to 10.000 bp, 220 to 9.000 bp, 240 to 8.000 bp, 260 to 7.500 bp, 280 to 7.000 bp, 300 to 6.500 bp, 320 to 6.000 bp, 340 to 5.500 bp, 360 to 5.000 bp, 380 to 4.500 bp, 400 to 4.000 bp, 420 to 3.800 bp, 440 to 3.600 bp, 460 to 3.400 bp, 480 to 3.200 bp, 500 to 3.000 bp, 520 to 2.800 bp, 540 to 2.600 bp, 560 to 2.400 bp, 580 to 2.200 bp, 600 to 2.000 bp, 650 to 1 .800 bp, 700 to 1.750 bp, 750 to 1.700 bp, 800 to 1.650 bp, 850 to 1 .650 bp, 900 to 1.600 bp, 950 to 1.550 bp, 1000 to 1 .500 bp, 1.050 to 1.450 bp, 1.100 to 1.400 bp, 1.150 to 1.350 bp, 1.200 to 1 .400 bp or 1.250 to 1.350 bp. It was shown that excision of sequences in the range of 100 to 10.000 bp is particularly efficient.

It is a great advantage of the present invention that neither the length of the DNA substitute sequence nor the length or the replaced DNA sequence is significantly limiting to the feasibility of the present invention.

In a preferred embodiment of the present invention, the exogenous nucleic acid molecule comprises at least one target sequence (target sequence 3) which is targeted by at least one guide RNA, wherein at least one double strand break occurs within or adjacent to target sequence 3 thereby resulting in a DNA substitute sequence.

In preferred embodiments, target sequence 3 may be the same or different from target sequence 1 and/or target sequence 2.

Furthermore, in some embodiments, target sequence 3 may be recognized by the same and/or different guide RNA as target sequence 1 and/or target sequence 2.

By providing a target sequence for the RNA guided endonuclease in the exogenous nucleic acid molecule it is possible that a dsDNA molecule that either is the exogenous nucleic acid molecule or is derived from the exogenous nucleic acid molecule is further processed inside the cell by the RNA guided endonuclease to generate the linear or linearized DNA substitute sequence for replacement of the sequence located between the DSBs. This feature allows generation of the DNA substitute sequence within the cell. Accordingly, the DNA substitute sequence does not have to be provided to the cell in a ready-to-use configuration.

Furthermore, through selection or design of a suitable target sequence and/or target site it is possible to influence the favored orientation of the DNA substitute sequence after integration between the two DSBs sites in the target DNA molecule, as explained above.

According to a further preferred embodiment of the invention, the exogenous nucleic acid molecule comprises at least two target sequences (target sequences 3 and 4) that are targeted by the at least one guide RNA, wherein double strand breaks occur within or adjacent to target sequences 3 and 4 thereby resulting in a DNA substitute sequence.

In preferred embodiments, target sequence 4 may be the same or different from target sequence 1 , target sequence 2 and/or target sequence 3. Furthermore, in some embodiments, target sequence 4 may be recognized by the same and/or different guide RNAs as target sequence 1 , target sequence 2 and/or target sequence 3.

In one embodiment of the present invention, the exogenous nucleic acid molecule neither comprises nor encodes homology arms targeted to the dsDNA molecule to be modified.

Accordingly, in some embodiments, the DNA substitute sequence does not comprise homology arms targeted to the dsDNA molecule to be modified. In some embodiments, homology arms that are targeted to the dsDNA molecule to be modified are sequences flanking the DNA substitute sequence.

Homology arm sequences are homologous to the sequences of the target dsDNA present at the open ends of the dsDNA molecule, after creation of two double strand breaks, that are not comprised by the DNA sequence to be replaced. Homology arms have a length of at least 30 nucleotides, preferably at least 50 nucleotides, and may have 90 %, preferably 95 %, 97 %, 98 %, 99 % or 100 % sequence identity to the corresponding sequences flanking the open ends of the target dsDNA molecule.

It is a great advantage of the method of the present invention that it enables precise replacement of a DNA sequence located between two DSBs of the dsDNA molecule to be modified by a DNA substitute sequence in the absence of homology sequences or homology arms on the sequence to be inserted. This embodiment further differentiates the method of the present invention from methods known in the art that use HDR-mediated insertion or replacement of sequences, which require the presence of homology sequences.

One advantage of the invention is therefore that it is not necessary to design substitute DNA sequences that comprise homology arms, whereby the use of homology arms typically leads to extra complication in cloning or costs for synthesis of the substitute DNA for integration into the dsDNA to be modified.

However, the use of homology arms in the exogenous nucleic acid, flanking the substitute sequence to be inserted, may in some embodiments also lead to induction of the NHEJ pathway for ligation of the substitute sequence between the two DSBs. For example, the homology arms may be cleaved by any given endonuclease and NHEJ subsequently employed, if for example the homology arm containing exogenous nucleic acid sequence is provided to the cell when cell division is not taking place. Therefore, the exogenous nucleic acid molecule may in some embodiments comprise homology arms, when they are not determinative in inducing the HDR pathway.

According to a preferred embodiment of the present invention, the cell is a non-dividing cell, preferably a cell in the G1 or GO phase. Alternatively, the cell can be a slowly dividing cell. In a further preferred embodiment of the invention, the cell is in G2 or in S phase.

It is particularly advantageous that the method of the invention can be performed in resting cells that are in G1 of the cell cycle for prolonged times, such as, for example, more than one day, preferably more than 1 week, 1 month or one year, or in cells, which have left the cell cycle and are in GO. Known methods for replacing DNA sequences in dsDNA molecules in cells, especially in eukaryotic cells, cannot be performed in G1 and GO phase, since these methods depend on the HDR pathway that is only active in G2 and S phase or the cell cycle.

Preferably, the method of the present invention comprises introducing into the cell at least two guide RNAs (guide RNA 1 and 2), wherein guide RNA 1 targets at least target sequence 1 and guide RNA 2 targets at least target sequence 2.

The use of two guide RNAs with specificities to different target sequences provides more flexibility to the method of the present invention since DNA sequences located between the cut sites associated with two different target sequences can be excised and replaced by a DNA substitute sequence. This is a great advantage when modifying genomic DNA which comprises only few identical target sequences.

In a preferred embodiment of the invention, guide RNA 1 targets at least a target sequence 1 of the dsDNA molecule to be modified and a target sequence 3 of the exogenous nucleic acid molecule, and/or guide RNA 2 targets at least a target sequence 2 of the dsDNA molecule to be modified and a target sequence 4 of the exogenous nucleic acid molecule.

In a further embodiment of the invention, the RNA guided endonuclease generates blunt end double strand breaks and/or the target sequences are configured to generate blunt end double strand breaks. The generation of blunt ended double strand breaks may be preferred in embodiments that aim to achieve integration of the DNA substitute sequence or ligation of open ends of the dsDNA molecules to be modified present after cleavage without the generation of INDELs at the site of ligation.

As shown in the examples, it was surprisingly found that blunt end cleavage allows perfect ligation and integration of DNA substitute sequences into the dsDNA molecule without the generation of INDELs in the vast majority of sequence replacement events. This can enable reconstitution of previously present or generation of newly formed target sequence that can be cleaved by the RNA guided endonuclease in cooperation with at least one of the guide RNAs. This enables the directed and/or favorable integration of the DNA substitute sequence to achieve a desired orientation of the integrated DNA substitute sequence. A skilled person is aware of suitable RNA guided endonucleases that preferably generate blunt ends, such as Cas9 and especially SpCas9. Furthermore, it is known that the target sequence and/or the target site influence the tendency of RNA guided endonucleases to generate blunt or sticky ends.

Accordingly, a skilled person is able to configure the target sequences and/or target sites in a way that favors the generation of blunt or sticky ends, respectively.

It is possible to determine whether an RNA guided endonuclease generates or has generated blunt or sticky ends by sequencing the cut-site after cutting and subsequent ligation of the open ends. If there are base-insertions that match to the protospacer sequence, this indicates that a sticky end cut occurred. This is further illustrated in Figure 7.

Preferred RNA guided DNA endonucleases that generate blunt end double strand breaks cuts and can be used in the context of the present invention comprise, without limitation, all RNA guided endonucleases of class 2 type II CRISPR systems, including, without limitation, Cas9 of Streptococcus pyogenes , Streptococcus thermophiles , Streptococcus pasteurianus,

Staphylococcus aureus, Neisseria cinerea, Campylobacter lari, Corynebacterium diphtheria and Parvibaculum lavamentivorans (Ran, F.A. et al. (2015). Nature 520, 186-191 ; Murugan, K. et al. (2017). Mol. Cell 68, 15-25).

In another preferred embodiment of the invention, the RNA guided endonuclease generating the at least two double strand breaks of the dsDNA molecule to be modified is a nickase.

The use of nickases for the generation of DSBs may be particularly advantageous to increase the specificity of DSB generation. It is highly unlikely, that double strand breaks at undesired sites of the dsDNA molecule to be modified or in the genome of a cell are generated when nickases are used to generate single strand breaks on the (+) and the (-) strand of the dsDNA molecule that are so close to each other that a DSB is generated. In this case the presence of two different guide RNAs may be required for the generation of one DSB. This embodiment may be used in particular in therapeutic application of the method of the present invention, for example, when the cells are present in a patient after modification of the dsDNA molecule.

In a further embodiment of the invention the RNA guided endonuclease generates sticky end double strand breaks. The generation of sticky end DSB may be favorable in embodiments of the invention that aim to generate genetic scars or INDELs at the site of integration of the DNA substitute sequence. Also, it may be advantageous in certain embodiments that the target sequences of the dsDNA target molecule and potentially the exogenous nucleic acid molecules are disrupted after integration of the DNA substitute sequence. This normally is the case for sticky end DSBs, since the shorter strand will be filled up during ligation by NHEJ, leading to the insertion of additional base pairs at the site of ligation. Sticky-end DSB can be generated by RNA guided endonucleases in various ways known to the person skilled in the art. For example, two single-strand breaks on the (+) and the (-) strand of the dsDNA molecule can be generated by using a nickase. If these single-strand breaks are close to each other, for example within a distance of less than 30, preferably less than 20 nt, more preferably less than 10 nt, a DSB with sticky ends will be formed. Alternatively, an RNA guided endonuclease that regularly generates sticky end DSBs may be used, such as Cpf1. Furthermore, the target sequence or target site can be selected in a way that sticky-end DSB formation is favored.

Preferred RNA guided DNA endonucleases that generate sticky end double strand breaks cuts and can be used in the context of the present invention comprise, without limitation, all RNA guided endonucleases of class 2 type V or type V-A CRISPR systems, such as, without limitation, Cpf1 or Cas12a of Acidaminococcus sp. BV3L6 and Lachnospiraceae bacterium ND2006 (Zetsche, B. et al. (2015). Cell 163, 759-771 ; Murugan, K. et al (2017). Cell 68, 15-25).

Preferred RNA guided DNA endonucleases that generate sticky end double strand breaks cuts and can be used in the context of the present invention comprise, without limitation,

SpCas9(D10A) and SpCas9(H840A) (Ran, F.A. et al. (2013). Cell 154, 1380-1389).

In a preferred embodiment of the method of the present invention, ligation of the DNA substitute sequence by the non-homologous end joining (NHEJ) pathway occurs in orientation 1 or orientation 2, wherein

upon ligation in orientation 1 , target sequences 1 and target sequence 2 are restored, and

upon ligation in orientation 2, target sequences 1 and target sequence 2 are disrupted.

It is clear to a skilled person that a DNA substitute sequence, which is a dsDNA molecule, can be inserted into the dsDNA molecule to be modified in two orientations. However, one of the two orientations may be favored over the other, for example when a coding sequence or a genomic DNA molecule, such as an exon or a part of an exon, is replaced by a modified or corrected version of the sequence. In this case, the target sequences and/or target sites of the dsDNA molecule to be modified and the exogenous nucleic acid molecule may be configured in a way that favors integration in the desired orientation.

For example, if the DNA substitute sequence comprises a coding sequence with 5’ to 3’ directionality, for favoring the desired integration orientation, the exogenous nucleic acid molecule comprising the DNA substitute sequence may have the same target sequence located on the 5’-end of the DNA substitute sequence as the dsDNA molecule to be modified comprises at the cleavage site that should be ligated to the 3’-end of the DNA substitute sequence.

Additionally, the exogenous nucleic acid molecule comprising the DNA substitute sequence may have the same target sequence located on the 3’-end of the DNA substitute sequence as the dsDNA molecule to be modified comprises at the cleavage site that should be ligated to the 5’- end of the DNA substitute sequence.

In that way, it is possible that the respective target sequences are restored upon integration of the DNA substitute sequence in the undesired orientation. Restoration of a target sequence means that the same sequence that is recognized by the at least one guide RNA is formed again after ligation of the DSB sites to the DNA substitute sequence. This allows re-excision of the DNA substitute sequence after ligation and integration so that the DNA substitute sequence and the open ends of the dsDNA molecule to be modified are again available for integration.

In contrast, upon integration of the DNA substitute sequence in the desired orientation, the respective target sequences will be disrupted, which prevents re-excision of the DNA substitute sequence. Disruption of the target sequence means that upon ligation of the DSB a new sequence is formed that is different from the previously target sequence and that is not recognized by the at least one guide RNA. As such, a bias for the desired integration orientation of the DNA substitute sequence can be created. This example demonstrates that the specific configuration of the target sequences employed in the method of the present invention can be used to direct the orientation of the integration of the DNA substitute sequence. A person skilled in the art is able to perform and/or design such configurations on the basis of the present description and the examples and figures provided herein.

According to a further embodiment of the invention, at least two guide RNAs (guide RNA 1 and 2) are introduced into the cell, wherein guide RNA 1 targets target sequence 1 and target sequence 3, and guide RNA 2 targets target sequence 2 and target sequence 4.

In a preferred embodiment of the invention, the exogenous nucleic acid molecule is an exogenous dsDNA molecule.

In a further preferred embodiment of the invention,

target sequence 1 is positioned at a 5’-end and target sequence 2 is positioned at a 3’-end of a (preferably coding) sequence located between the double strand breaks of the dsDNA molecule to be modified, and

target sequence 4 is positioned at a 5’ end and target sequence 3 is positioned at a 3’ end of a (preferably coding) DNA substitute sequence located between the double strand breaks within target sequences 3 and 4 of the exogenous nucleic acid or dsDNA molecule.

In a further embodiment of the invention, at least one additional guide RNA targets a sequence of the dsDNA molecule to be modified, wherein said sequence arises when the double strand breaks of the dsDNA molecule to be modified are ligated together, without a) introduction of the DNA substitute sequence and b) without reintroduction of the sequence located originally between the double strand breaks.

This method is particularly advantageous because in case of ligation of the open ends of the dsDNA molecule to be modified, which are not comprised by the excised DNA sequence to be replaced, a target sequence is generated that is recognized by a specific guide RNA, leading to separation of the open ends, which are then again available for the integration of the DNA substitute sequence. An example of this embodiment of the invention is illustrated in Figure 10 and example 3 provided herein.

In another embodiment of the present invention,

a PAM associated with target sequence 1 and a PAM associated with target sequence 2 are comprised by the DNA sequence located between the double strand breaks of the dsDNA molecule to be modified, and/or

a PAM associated with target sequence 3 and a PAM associated with target sequence 4 are not comprised by the DNA substitute sequence.

This configuration of the target sequences 1 to 4 is particularly advantageous, because it was shown that it leads to a very efficient replacement of the DNA sequence located between the DSBs of the dsDNA molecule to be modified and integration of the DNA substitute sequence without the generation of INDELs. Therefore, this embodiment is particularly advantageous for the directed integration of the DNA substitute sequence.

According to a further embodiment of the method or the invention, the DNA sequence to be replaced between the DSBs of the dsDNA molecule to be modified and the DNA substitute sequence comprised by the exogenous dsDNA molecule are both coding DNA sequences and therefore define a (+)-strand and a (-)-strand of the dsDNA molecule to be modified and the exogenous dsDNA molecule, and

target sequences 1 and 2 are located on the (+)-strand and target sequence 3 and 4 are located on the (-)-strand,

target sequence 1 and 3 are located on the (+)-strand and target sequence 2 and 4 are located on the (-)-strand,

target sequences 1 and 4 are located on the (+)-strand and target sequence 2 and 3 are located on the (-)-strand,

target sequences 2 and 3 are located on the (+)-strand and target sequence 1 and 4 are located on the (-)-strand,

target sequences 2 and 4 are located on the (+)-strand and target sequence 1 and 3 are located on the (-)-strand, or

target sequences 3 and 4 are located on the (+)-strand and target sequence 1 and 2 are located on the (-)-strand.

The present invention further relates to a Kit for modifying a dsDNA molecule in a cell, comprising:

an RNA guided DNA endonuclease or a nucleic acid encoding an RNA guided DNA endonuclease,

at least one guide RNA, and

a nucleic acid molecule comprising or encoding a DNA substitute sequence, wherein: the at least one guide RNA is configured for generating at least two double strand breaks of a dsDNA molecule to be modified within or adjacent to a target sequence 1 and a target sequence 2 within the dsDNA molecule, and

the nucleic acid molecule comprising or encoding a DNA substitute sequence is configured for replacement of the DNA sequence of the dsDNA molecule located between the two double strand breaks with the DNA substitute sequence by the NHEJ pathway.

The at least one guide RNA that is configured for generating at least two double strand breaks of a dsDNA molecule to be modified within or adjacent to a target sequence 1 and a target sequence 2 may comprise a target specific sequence that is at least partially, preferably completely complementary to target sequences 1 and 2 and interacts with the RNA guided DNA endonuclease of the kit of the present invention.

A skilled person is therefore capable of configuring the guide RNA to target particular sequences in the dsDNA target. By analysis of the guide RNAs in the kit or composition a skilled person can determine whether and how the guide RNAs are configured for generating two double strand breaks. The presence of two guide RNAs in the kit is clearly indicative of generating two double strand breaks, for example in the context of editing genomic dsDNA sequences.

The DNA substitute sequence that is configured for replacement of the DNA sequence of the dsDNA molecule located between the two double strand breaks with the DNA substitute sequence by the NHEJ pathway may be a linear or linearized dsDNA molecule.

The nucleic acid molecule comprising or encoding a DNA substitute sequence may be configured for employing the NHEJ pathway by comprising a linearizable DNA element that is suitable as the substitute DNA sequence. The configuration of the DNA substitute sequence for employing the NHEJ pathway may in some embodiments relate to the absence of homology arms.

However, in some embodiments, the presence of homology arms may not be determinative for using the HDR pathway, and the presence of homology arms may in some embodiments lead also to the employment of the NHEJ pathway.

In a preferred embodiment of the kit of the invention the nucleic acid molecule comprising or encoding a DNA substitute sequence is configured

to restore target sequence 1 and target sequence 2 upon ligation of the DNA substitute sequence by the NHEJ pathway in orientation 1 , and

to disrupt target sequence 1 and target sequence 2 upon ligation of the DNA substitute sequence by the NHEJ pathway in orientation 2.

Such a nucleic acid molecule may comprise target sequences 3 and 4, wherein target sequence 3 is the same as target sequence 1 , and target sequence 4 is the same as target sequence 2, or target sequence 3 is the same as target sequence 2, and target sequence 4 is the same as target sequence 1. Suitable configurations of the nucleic acid molecule comprising or encoding a DNA substitute sequence have been described herein in the context of the method of the present invention and are obvious to the person skilled in the art in light of the disclosure of the present patent application. In a preferred embodiment of the kit of the invention the RNA guided endonuclease generates blunt end double strand breaks and/or the target sequences are configured to generate blunt end double strand breaks.

In a preferred embodiment of the kit of the invention the RNA guided endonuclease generating the at least two double strand breaks of the dsDNA molecule to be modified is a nickase and/or the RNA guided endonuclease generates sticky end double strand breaks.

Furthermore, the present invention encompasses a composition for use as a medicament, for modifying a double stranded DNA (dsDNA) molecule in a cell, comprising:

an RNA guided DNA endonuclease or a nucleic acid encoding an RNA guided DNA endonuclease,

at least one guide RNA, and

a nucleic acid molecule comprising or encoding a DNA substitute sequence, wherein:

the at least one guide RNA is configured for generating at least two double strand breaks of a genomic dsDNA molecule within or adjacent to a target sequence 1 and a target sequence 2 within the genomic dsDNA molecule, and

the nucleic acid molecule comprising or encoding a DNA substitute sequence is configured for replacement of the DNA sequence of the genomic dsDNA molecule located between the two double strand breaks with the DNA substitute sequence by the NHEJ pathway.

Preferably, the dsDNA molecule of the composition of the present invention is a genomic dsDNA molecule. Furthermore, the cell of the preferentially is a eukaryotic cell.

The composition of the present invention is therefore suitable for the treatment of any given medical condition in which a change in the genomic DNA (including mitochondrial DNA) of a subject is necessary. Preferably, the change in the genomic DNA is to be carried out in somatic cells. This genetic modification may be employed potentially directly in vivo to a subject in need thereof, for example by modifying the genome of a subject, or may be employed in cells removed from a patient before subsequent transplantation, or may be employed to modify the genome or other genetic element of a pathogen inside a host.

In a preferred embodiment of the composition for use as a medicament the nucleic acid molecule comprising or encoding a DNA substitute sequence is configured

to restore target sequence 1 and target sequence 2 upon ligation of the DNA substitute sequence by the NHEJ pathway in orientation 1 , and

to disrupt target sequence 1 and target sequence 2 upon ligation of the DNA substitute sequence by the NHEJ pathway in orientation 2.

In a further embodiment of the composition for use as a medicament of the invention the RNA guided endonuclease generates blunt end double strand breaks and/or the target sequences are configured to generate blunt end double strand breaks. In an alternative embodiment of the composition for use as a medicament the RNA guided endonuclease generating the at least two double strand breaks of the dsDNA molecule to be modified is a nickase and/or the RNA guided endonuclease generates sticky end double strand breaks.

In further embodiments of the invention the method as described herein is used for exon replacement or exchange. Exon replacement allows coding regions in genes, which are later spliced into functional coding mRNA molecules, to be precisely and efficiently modified, for example by removing exons in which undesired gene sequences, such as mutations or other sequences encoding unwanted traits, are present. The present invention enables via the use of two DSBs the replacement of an entire exon, for example by targeting sites for DSB creation in the neighboring introns flanking an exon and providing to the cell a replacement exon in the form of an exogenous nucleic acid molecule for replacing the undesired exon.

All advantages and preferred features of the method of the present invention provided herein also apply to the kit and the composition of the invention, and vice versa.

DETAILED DESCRIPTION OF THE INVENTION

The invention relates to a method for modifying double stranded DNA employing an RNA guided DNA endonuclease to generate two double strand breaks in the dsDNA molecule to be modified, and replacement of the sequence positioned between the double strand breaks with a substitute DNA sequence using the non-homologous end joining (NHEJ) pathway.

In the context of the present invention, the term“modifying” a double stranded DNA refers to any kind of alteration, modification or change of a dsDNA molecule. In particular, modifying relates to deleting, inserting, replacing, substituting or translocating one or one or more nucleotides or pairs of nucleotides or nucleotide sequences from a dsDNA molecule.

The term“replacement” or "substitution", as used herein, is defined in accordance with the pertinent art and refers to the replacement of nucleotides with other nucleotides. The term includes for example the replacement of single nucleotides resulting in point mutations. Said point mutations can lead to an amino acid exchange in the resulting protein product but may also not be reflected an the amino acid level (i.e. silent mutations). Also encompassed by the term "substitution" are mutations resulting in the replacement of multiple nucleotides, such as for example parts of genes, such as parts of exons or introns as well as the replacement of entire genes. The number of nucleotides that replace the originally present nucleotides may be the same or different (i.e. more or less) as compared to the number of nucleotides removed. Preferably, the number of replacement nucleotides corresponds to the number of originally present nucleotides that are substituted.

The term“insertion", in accordance with the present invention, is defined in accordance with the pertinent art and refers to the incorporation of one or more nucleotides into a nucleic acid molecule. Insertion of parts of genes, such as parts of exons or introns as well as insertion of entire genes is also encompassed by the term "insertion". When the number of inserted nucleotides is not dividable by three, the insertion can result in a frameshift mutation within a coding sequence of a gene. Such frameshift mutations will alter the amino acids encoded by a gene following the mutation. In some cases, such a mutation will cause the active translation of the gene to encounter a premature stop codon, resulting in an end to translation and the production of a truncated protein. When the number of inserted nucleotides is instead dividable by three, the resulting insertion is an "in-frame insertion". In this case, the reading frame remains intact after the insertion and translation will most likely run to completion if the inserted nucleotides do not code for a stop codon. However, because of the inserted nucleotides, the finished protein will contain, depending on the size of the insertion, one or multiple new amino acids that may affect the function of the protein.

The term "deletion", as used in accordance with the present invention, is defined in accordance with the pertinent art and refers to the loss of nucleotides or larger parts of genes, such as exons or introns as well as entire genes. As defined with regard to the term "insertion", the deletion of a number of nucleotides that is not evenly dividable by three will lead to a frameshift mutation, causing all of the codons occurring after the deletion to be read incorrectly during translation, potentially producing a severely altered and most likely nonfunctional protein. If a deletion does not result in a frameshift mutation, i.e. because the number of nucleotides deleted is dividable by three, the resulting protein is nonetheless altered as the finished protein will lack, depending on the size of the deletion, one or several amino acids that may affect the function of the protein.

The term double stranded DNA relates to two deoxyribonucleic acid polynucleotide strands that are bound together or hybridizes through pairing of the bases or nucleotides of the two strands through hydrogen bonds resulting in double-stranded DNA. The method of the present invention can be performed using any kind of dsDNA molecule, including, without limitation, genomic dsDNA of any origin or organism, synthetic dsDNA, amplified or isolated dsDNA.

A“cell” in the sense of the present invention refers to, without limitation, any biological cell, which might be derived from any kind of organism, comprising unicellular organisms as well as multicellular organisms, such as for example any kind of plant or animal, including mammals, fish, amphibians, reptiles, birds, molluscs, arthropods, annelids, nematodes, flatworms, cnidarians, ctenophores and sponges. Cells of the present invention further comprise prokaryotic cells, bacteria, eukaryotic cells, blood cells, stem cells, immune cells (such as 13- cells, dendritic cells, granulocytes, innate lymphoid cells (ILCs), megakaryocytes, monocytes, macrophages, myeloid-derived Suppressor Cells (MDSC), natural killer (NK) cells, platelets, red blood cells (RBCs), T-cells or thymocytes), cancer cells, tumor cells and circulating tumor cells.

In some embodiments of the invention, the cell in which the dsDNA is to be modified is a vertebrate cell, more preferably a mammalian cell, such as a human cell. In some

embodiments, the cell is not a rice cell. In some embodiments, the cell is not a plant cell.

The term "introducing into the cell", as used herein, relates to any known method of bringing a protein or a nucleic acid molecule into a cell. Non-limiting examples include microinjection, infection with viral vectors, electroporation, transfection, such as transfection using formulations with cationic lipids. Suitable methods for introducing the components of the present invention into a cell are known to the skilled person. CRISPR is an abbreviation of Clustered Regularly Interspaced Short Palindromic Repeats and is a family of DNA sequences in bacteria. The sequences contain snippets of DNA from viruses that have attacked the bacterium. These snippets are used by the bacterium to detect and destroy DNA from further attacks by similar viruses. These sequences play a key role in a bacterial defense system, and form the basis of a technology known as CRISPR/Cas that effectively and specifically changes genes within organisms.

Sequences of the CRISPR loci are transcribed and processed into CRISPR RNAs (crRNAs) which, together with a trans-activating crRNAs (tracrRNAs), complex with CRISPR-associated (Cas) proteins to dictate specificity of DNA cleavage by Cas nucleases through Watson-Crick base pairing between nucleic acids (Wiedenheft, B et al (2012). Nature 482: 331-338; Horvath, P et al (2010). Science 327: 167-170; Fineran, PC et a. (2012). Virology 434: 202-209).

It was shown that the three components required for the type II CRISPR nuclease system are the Cas9 protein, the mature crRNA and the tracrRNA, which can be reduced to two components by fusion of the crRNA and tracrRNA into a single guide RNA (sgRNA) and that re targeting of the Cas9/sgRNA complex to new sites could be accomplished by altering the sequence of a short portion of the gRNA (Garneau, JE et al (2010). Nature 468: 67-71 ;

Deltcheva, E et al. (201 1 ). Nature 471 : 602-607, Jinek, M et al (2012) Science 337: 816-821 ).

CRISPR-Cas systems are RNA-guided adaptive immune systems of bacteria and archaea that provide sequence-specific resistance against viruses or other invading genetic material. This immune-like response has been divided into two classes on the basis of the architecture of the effector module responsible for target recognition and the cleavage of the invading nucleic acid (Makarova KS et al. Nat Rev Microbiol. 2015 Nov; 13(1 1 ):722-36.). Class 1 comprises multi subunit Cas protein effectors and Class 2 consists of a single large effector protein. Both Class 1 and 2 use CRISPR RNAs (crRNAs) to guide a Cas nuclease component to its target site where it cleaves the invading nucleic acids. Due to their simplicity, Class 2 CRISPR-Cas systems are the most studied and widely applied for genome editing. The most widely used CRISPR-Cas system is CRISPR-Cas9.

It was demonstrated that the CRISPR/Cas9 system could be engineered for efficient genetic modification in mammalian cells. The only sequence limitation of the CRISPR/Cas system appears to derive from the necessity of a protospacer-adjacent motif (PAM) located

immediately 3’ to the target site. The PAM sequence is specific to the species of Cas9. For example, the PAM sequence 5’-NGG-3’ is necessary for binding and cleavage of DNA by the commonly used Cas9 from Streptococcus pyogenes. However, Cas9 variants with novel PAMs have been and may be engineered by directed evolution, thus dramatically expanding the number of potential target sequences. Cas9 complexed with the crRNA and tracrRNA undergoes a conformational change and associates with PAM motifs throughout the genome interrogating the sequence directly upstream to determine sequence complementarity with the gRNA. The formation of a DNA-RNA heteroduplex at a matched target site allows for cleavage of the target DNA by the Cas9-RNA complex. These methods and mechanisms are well known in the art. As known in the art, CRISPR/Cas9 has been exploited to develop potent tools for genome manipulation in animals, plants and microorganisms. The RNA-guided Cas9 endonuclease first recognizes a 2- to 4-base-pair conserved sequence named the protospacer-adjacent motif (PAM), which flanks a target DNA site. Upon binding to the PAM, Cas9 interrogates the flanking DNA sequences for base-pairing complementarity to a guide RNA. If there is complementarity between the first 12 base pairs (the‘seed’ sequence) of the guide RNA and the target DNA strand, RNA strand invasion accompanies local DNA unwinding to form an R-loop. Precise cleavage of each DNA strand by the RuvC and HNH domains of Cas9 generates a blunt double-strand DNA (dsDNA) break (DSB) at a position three base pairs upstream of the 3' edge of the protospacer sequence, measuring from the PAM.

CRISPR/Cas9 genome-editing experiments have been exploiting the host cell machinery to repair the genome precisely at the site of the Cas9-generated DSB. Mutations can arise either by non-homologous end joining (NHEJ) or homology-directed repair (HDR) of DSBs. NHEJ can produce small insertions or deletions (INDELs) at the cleavage site, whereas HDR uses a native (or engineered) DNA template to replace the targeted allele with an alternative sequence by recombination. Additional DNA repair pathways such as single-strand annealing, alternative end joining, microhomology-mediated joining, mismatch and base- and nucleotide-excision repair can also produce genome edits.

Cas9 variants derived from the Streptococcus pyogenes Cas9 (SpCas9) have been generated for use as nickases, dual nickases or Fokl fusion variants. More recently, Cas9 orthologs, and other nucleases derived from class 2 CRISPR-Cas systems including Cpf1 and C2c1 , have been added to the CRISPR toolbox. These ongoing efforts to mine the abundant bacterial and archaeal CRISPR-Cas systems should increase the range of molecular tools available to researchers.

In the context of the present invention, the term“RNA guided DNA endonuclease” refers to DNA endonucleases that interact with at least one RNA-Molecule. In the context of the present invention the terms RNA guided DNA endonuclease and RNA guided endonuclease are used interchangeably. DNA endonucleases are enzymes that cleave the phosphodiester bond within a DNA polynucleotide chain. In case of RNA guided DNA endonuclease the interacting RNA- Molecule may guide the RNA guided DNA endonuclease to the site or location in a DNA where the endonuclease becomes active. In particular, the term RNA guided DNA endonuclease refers to naturally occurring or genetically modified Cas nuclease components or CRISPR-Cas systems, which include, without limitation, multi-subunit Cas protein effectors of class 1 CRISPR- Cas systems as well as single large effector Cas proteins of class 2 systems.

Details of the technical application of CRISPR/Cas systems and suitable RNA guided endonuclease are known to the skilled person and have been described in detail in the literature, as for example by Barrangou R et al. (Nat Biotechnol. 2016 Sep 8;34(9):933-941 ), Maeder ML et al. (Mol Ther. 2016 Mar;24(3):430-46) and Cebrian-Serrano A et al. (Mamm Genome. 2017; 28(7): 247-261 ). The present invention is not limited to the use specific RNA guided

endonucleases and therefore comprises the use of any given RNA guided endonucleases in the sense of the present invention suitable for use in the method described herein. Any RNA guided DNA endonuclease known in the art may be employed in accordance with the present invention. RNA guided DNA endonuclease comprise, without limitation, Cas proteins of class 1 CRISPR-Cas systems, such as Cas3, Cas8a, Cas5, Cas8b, Cas8c, Casl Od, Cse1 , Cse2, Csy1 , Csy2, Csy3, GSU0054, Cas10, Csm2, Cmr5, Csx1 1 , Csx10 and Csf1 ; Cas proteins of class 2 CRISPR-Cas systems, such as Cas9, Csn2, Cas4, Cpf1 , C2c1 , C2c3 and C2c2; corresponding orthologous enzymes/CRISPR effectors from various bacterial and archeal species; engineered CRISPR effectors with for example novel PAM specificities, increased fidelity, such as SpCas9-HF1/eSpCas9, or altered functions, such as nickases. Particularly preferred RNA guided DNA endonuclease of the present invention are

Streptococcus pyogenes Cas9 (SpCas9), Staphylococcus aureus Cas9, Streptococcus thermophilus Cas9, Neisseria meningitidis Cas9 (NmCas9), Francisella novicida Cas9

(FnCas9), Campylobacter jejuni Cas9 (CjCas9), Cas12a (Cpf1 ) and Cas13a (C2C2) (Makarova KS et al. (November 2015). Nature Reviews Microbiology. 13 (1 1 ): 722-36).

The definition and explanations provided herein are mainly focused on the SpCas9 Crispr/Cas system. However, the person skilled in the art is aware of how to use alternative Crispr/Cas systems as well as tools and methods that provide or allow the gain of information on the details of such alternative systems.

The use of Cpf1 in the context of the present invention can be advantageous because it requires only one associated guide RNA, it generates staggered/sticky end cuts and it cuts in non-dividing cells, such as nerve cells.

The RNA guided DNA endonuclease and in particular Cas9 may also be a modified protein, wherein the nuclease function of the protein is altered into a nicking endonuclease function, which only cuts one of the two DNA strands of the dsDNA. In other words, the naturally occurring endonucleases function of cleaving both strands of a double-stranded target DNA, is altered into an endonuclease that cleaves (i.e. nicks) only one of the strands. Such modified RNA guided DNA endonucleases are also calls“nickases” in the context of the present invention. Means and methods of modifying RNA guided DNA endonuclease such as Cas9 accordingly are well known in the art, and include for example the introduction of amino acid replacements into Cas9 that render one of the nuclease domains inactive. More specifically, aspartate can be replaced against alanine at position 10 of the Streptococcus pyogenes Cas9 (SpCas9 D10A; Cong et al. (2013) Science 339:819-823). Further examples are known in the art, for example the H840A replacement in SpCas9 (Mali P et al. Nat Biotechnol. 2013 Sep;

31 (9):833-8; Ran FA et al. Cell. 2013 Sep 12; 154(6): 1380-9).

In accordance with the method of the invention, the RNA guided DNA endonuclease may be introduced as a protein, but alternatively the RNA guided DNA endonuclease may also be introduced in form of a nucleic acid molecule encoding said protein. It will be appreciated that the nucleic acid molecule encodes said RNA guided DNA endonuclease in expressible form such that expression in the cell results in a functional RNA guided DNA endonuclease protein such as Cas9 protein. Means and methods to ensure expression of a functional polypeptide are well known in the art. For example, the coding sequences for the endonuclease may be comprised in a vector, such as for example a plasmid, cosmid, virus, bacteriophage or another vector used conventionally e.g. in genetic engineering. The coding sequences inserted in the vector can e.g. be synthesized by standard methods, or isolated from natural sources. The coding sequences may further be ligated to transcriptional regulatory elements and/or to other amino acid encoding sequences. Such regulatory sequences are well known to those skilled in the art and include, without being limiting, regulatory sequences ensuring the initiation of transcription, internal ribosomal entry sites (IRES) and optionally regulatory elements ensuring termination of transcription and stabilization of the transcript. Non-limiting examples for regulatory elements ensuring the initiation of transcription comprise a translation initiation codon, transcriptional enhancers such as e.g. the SV40-enhancer, insulators and/or promoters, such as for example the cytomegalovirus (CMV) promoter, SV40-promoter, RSV-promoter (Rous sarcome virus), the lacZ promoter, chicken beta-actin promoter, CAG-promoter (a combination of chicken beta- actin promoter and cytomegalovirus immediate-early enhancer), the gai 10 promoter, human elongation factor 1 a-promoter, A0X1 promoter, GAL1 promoter CaM-kinase promoter, the lac, trp or tac promoter, the lacUV5 promoter, the autographa californica multiple nuclear polyhedrosis virus (AcMNPV) polyhedral promoter or a globin intron in mammalian and other animal cells. Non-limiting examples for regulatory elements ensuring transcription termination include the V40-poly-A site, the tk-poly-A site or the SV40, lacZ or AcMNPV polyhedral polyadenylation signals, which are to be included downstream of the nucleic acid sequence of the invention. Additional regulatory elements may include translational enhancers, Kozak sequences and intervening sequences flanked by donor and acceptor sites for RNA splicing. Moreover, elements such as origin of replication, drug resistance gene or regulators (as part of an inducible promoter) may also be included.

Nucleic acid molecules encoding said RNA guided DNA endonuclease include DNA, such as cDNA or genomic DNA, as well as RNA and in particular mRNA. It will be readily appreciated by the skilled person that more than one nucleic acid molecule may encode an RNA guided DNA endonuclease in accordance with the present invention due to the degeneracy of the genetic code. Degeneracy results because a triplet code designates 20 amino acids and a stop codon. Because four bases exist which are utilized to encode genetic information, triplet codons are required to produce at least 21 different codes. The possible e possibilities for bases in triplets give 64 possible codons, meaning that some degeneracy must exist. As result, some amino acids are encoded by more than one triplet, i.e. by up to six. The degeneracy mostly arises from alterations in the third position in a triplet. This means that nucleic acid molecules having different sequences, but still encoding the same RNA guided DNA endonuclease, can be employed in accordance with the present invention.

The nucleic acid molecules used in accordance with the present invention may be of natural as well as of (semi) synthetic origin. Thus, the nucleic acid molecules may, for example, be nucleic acid molecules that have been synthesized according to conventional protocols of organic chemistry. The person skilled in the art is familiar with the preparation and the use of said probes (see, e.g., Sambrook and Russel "Molecular Cloning, A Laboratory Manual", Cold Spring Harbor Laboratory, N.Y. (2001 )). As used herein“nucleic acid” shall mean any nucleic acid molecule, including, without limitation, DNA, RNA and hybrids or modified variants thereof. An“exogenous nucleic acid” or

“exogenous genetic element” relates to any nucleic acid introduced into the cell, which is not a component of the cells“original” or“natural” genome. Exogenous nucleic acids may be integrated or nonintegrated in the genetic material of the target mesenchymal stem cell, or relate to stably transduced nucleic acids.

The nucleic acid molecules used in accordance with the invention may be nucleic acid mimicking molecules known in the art such as synthetic or semi-synthetic derivatives of nucleic acid molecules and mixed polymers. They may contain additional non-natural or derivatised nucleotide bases, as will be readily appreciated by those skilled in the art. Nucleic acid mimicking molecules or nucleic acid derivatives according to the invention include, without being limiting, phosphorothioate nucleic acid, phosphoramidate nucleic acid, morpholino nucleic acid, hexitol nucleic acid (HNA), peptide nucleic acid (PNA) and locked nucleic acid (LNA).

Furthermore, the method of the present invention comprises introducing into the cell at least one guide RNA. In the context of the present invention, a“guide RNA” refers to RNA molecules interacting with RNA guided DNA endonuclease leading to the recognition of the target sequence to be cleaved by the RNA guided DNA endonuclease. According to the present invention, the term“guide RNA” therefore comprises, without limitiation, target sequence specific CRISPR RNAs (crRNA), trans-activating crRNAs (tracrRNA) and chimeric single guide RNAs (sgRNA).

crRNAs differ depending on the RNA guided endonuclease and the CRISPR/Cas system but typically contain a target specific sequence of between 20 to 72 nucleotides in length, flanked by two direct repeats (DR) of a length of between 21 to 46 nucleotides. In the case of S.

pyogenes, the DRs are 36 nucleotides long and the target sequence is 30 nucleotides long. The 3' located DR of the crRNA is complementary to and hybridizes with the corresponding tracr RNA, which in turn binds to the Cas9 protein.

As used herein, the term "trans-activating crRNA (tracr RNA)" refers to a small RNA, that is complementary to and base pairs with a pre-crRNA (3' located DR of the crRNA), thereby forming an RNA dupiex. This pre-crRNA is then cleaved by an RNA-specific ribonuclease, to form a crRNA/tracrRNA hybrid, which subsequently acts as a guide for the endonuclease Cas9, which cleaves the invading nucleic acid.

As described herein, the genes encoding the elements of a CRISPR/Cas system, such as for example Cas9, tracrRNA and crRNA, are typically organized in operon(s). DR sequences functioning together with RNA guided endonuclease such as Cas9 proteins of other bacterial species may be identified by bioinformatic analysis of sequence repeats occurring in the respective Crispr/Cas operons and by experimental binding studies of Cas9 protein and tracrRNA together with putative DR sequence flanked target sequences.

Alternatively, a chimeric single guide RNA sequence comprising such a target sequence specific crRNA and tracrRNA may be employed. Such a chimeric (ch) RNA may be designed by the fusion of a target specific sequence of 20 or more nucleotides (nt) with a part or the entire DR sequence (defined as part of a crRNA) with the entire or part of a tracrRNA, as shown by (Jinek et al. Science 337:816-821 ). Within the chimeric RNA a segment of the DR and the tracrRNA sequence are complementary able to hybridize and to form a hairpin structure.

Moreover, the at least one guide RNA of the present invention may also be encoded by a nucleic acid molecule, which is introduced into the cell. The definitions and preferred embodiments recited above with regard to the nucleic acid molecule encoding the

endonuclease equally apply to the nucleic acid molecule encoding these RNAs. Regulatory elements for expressing RNAs are known to one skilled in the art, for example a U6 promoter.

The present invention relates to the generation of double strand beaks of the dsDNA molecule to be modified, wherein the dsDNA molecule comprises at least two target sequences, which are targeted by the at least one guide RNA.

In accordance with the present invention, a "target sequence" is a nucleotide sequence in the dsDNA molecule that is recognized by the at least one guide RNA that is associated with the RNA guided endonuclease due to the target specific sequence comprised by the guide RNA. The target sequence is at least partially complementary to the target specific sequence of the guide RNA and is associated with a so-called protospacer adjacent motif (PAM). The PAM is a 2-6 base pair DNA sequence located adjacent to the target sequence and can be located either at the 5’-end (for example for the Crispr/Cpf1 system) or at the 3’-end of the target sequence (for example for the Crispr/Cas9 system), depending on the Crispr/Cas system employed. An RNA guided endonuclease, such as Cas9 or Cpf1 , will not successfully bind to and cleave the targeted dsDNA molecule if the recognized target sequence is not associated with a PAM sequence. The formation of a DNA-RNA heteroduplex between the target sequence and the target specific sequence of the guide RNA allows for cleavage of the target DNA by the guide RNA/RNA guided endonuclease complex. Cleavage of the targeted dsDNA molecule occurs within the target sequence or at a site adjacent to the target sequence, depending on the used RNA guided endonuclease and CRISPR/Cas system.

For example, in case of SpCas9 the DNA target sequence of at least 20 nucleotides is located directly upstream/at the 5’-end of an invariant 5’-NGG-3’ PAM. Correct pairing of the guide RNA to the DNA target sequence leads to the generation of a double strand break in the dsDNA molecule (“cleavage” of the dsDNA molecule by SpCas9) 3 base-pairs (bp) upstream of the PAM within the target sequence.

In case of the CRISPR/Cpf1 system, the Cpf1-crRNA complex cleaves target DNA by identification of a target sequence that may be located downstream/at the 3’ end of a protospacer adjacent motif (for example 5 -YTN-3' (where "Y" is a pyrimidine and "N" is any nucleobase) or 5 -TTN-3'). Cpf1 can introduce a sticky end/staggered end DNA double strand breaks. In case of AsCpfl and LbCpfl a double strand break with a 4 nucleotides overhang can be generated, which can occur 19 bp downstream of the PAM on the targeted (+)-strand and 23 bp downstream of the PAM on the (-)-strand.

As illustrated by the above examples, the exact site of the double strand break depends on the Crispr/Cas system or the RNA guided endonuclease employed in the method of the invention and can therefore be determined by the person skilled in the art upon selection of the RNA guided endonuclease. In the context of the present invention, a site of a double strand break “adjacent” to the target sequence is located within 100 nucleotides or base pairs upstream or downstream from the 5’- or 3’-end of the target sequence. Preferably, the double strand break is generated within 90, 80, 70, 60, 50, 40, 30, 20, 10, 5, 4, 3, 2 or 1 nucleotides/base pairs upstream or downstream from the 5’- or 3’-end of the target sequence or within the target sequence.

In the context of the present invention, the target sequence may also be called“protospacer". The term“target site” may refer to a location or sequence in the dsDNA molecule comprising the target sequence and an associated PAM.

In the context of the present invention, the term“double strand break” or“DSB” refers to interruption of both strands of a dsDNA molecule leading to the separation of the parts of the dsDNA molecule that lie upstream and downstream of the side of the double strand break. In contrast, a single strand break refers to the interruption of only one of the two DNA strands and will not lead to a separation of the parts of the dsDNA molecule that lie upstream and downstream of the side of the double strand break.

In the context of the present invention, double strand breaks can occur due to cleavage of both strands by one RNA guided endonuclease or due to two single-strand cuts on both the (+)- and the (-)-strand by nickases.

A double strand break can be generated by cleavage of both strands of the dsDNA at the same/corresponding position on the complementary strands, leading to the formation of blunt ends of the resulting separated ends of the dsDNA molecule, as it is mostly the case for Cas9 mediated cleavage. However, in case of non-canonical cleavage, Cas9 may also induce the formation of double strand breaks with sticky/staggered ends, wherein the strand breaks on the two complementary DNA strands of the dsDNA are located at different positions, leading to the formation of strand-overhangs on the ends of the cleaved dsDNA molecule.

Furthermore, certain RNA-guided endonucleases regularly generate sticky ends, such as for example Cpf1. It is also possible to influence the tendency of RNA guided endonucleases to generate sticky-ends or blunt ends through selection of certain target sequence.

Additionally, it is possible to intentionally generate sticky ended double strand breaks of a certain configuration by inducing single strand breaks on both the (+)- and the (-)-strand by individual nickases targeting different target sequences on the two strands. By using this approach, it is possible to precisely select the site of a single strand break. Induction of two single strand breaks on both complementary strands within a distance of no more than 50 nucleotides, preferably not more than 40nt, 35 nt, 30 nt or 25, most preferably not more than 20, 19, 18, 17, 16, 15 14, 13 ,12, 1 1 , 10, 9, 8, 7, 6, 5, 4, 3, 2 or 1 nt will lead to a separation of the ends of the dsDNA molecule that are distal and proximal of the corresponding single strand breaks, resulting in the formation of stick ends with overhangs of the corresponding length. Importantly, the single strands sticky end overhangs might not be required for NHEJ repair and it might not be required that complimentary single strand overhangs are present on the open ends of the DNA molecule to be modified generated by a DSB and of the corresponding DNA substitute sequence to be ligated with each other. In preferred embodiments, the DNA double strand breaks generate sticky ends with single strand overhangs of at least 5 nucleotides.

By using this approach of inducing a double strand break, it can be precisely selected what kind of overhang at the resulting separated ends of the dsDNA molecule is generated. For each double strand break, at least two guide RNA molecules may be required.

DNA double-strand breaks (DSBs) are produced intentionally by RNA guided nucleases to achieve genome editing through DSB repair. Two main branches of DSB repair mechanisms exist in cells. These either process DSB ends by strand resection and initiate repair by homology-directed repair (HDR) or template-free annealing (a-NHEJ), or seek for the immediate protection of free ends through relegation by classic-NHEJ (c-NHEJ).

While the c-NHEJ pathway operates throughout all phases of the cell cycle, HDR is restricted to the S and G2 phases, seeking for homologous sequences to repair the resected DSB ends. During mitosis, DSB repair is entirely shut down to guard the chromosomes against the fusion of telomeres. In the G1 (and GO) phase and in resting cells, c-NHEJ repair is dominant as HDR is silenced. All pathways are active and competing in the S/G2 phases. DSB induction in a population of cycling cells leads to a variety of edited alleles, with c-NHEJ as the dominant outcome.

In proliferating cells, HDR is mediated by the homologous recombination (HR) pathway, wherein the native pathway uses the intact homologous sequences of sister chromatids as template for the repair of DSB sites and leads to the reconstitution of the wild-type allele. It has been shown in the art that it is possible to achieve precise sequence modifications at targeted DSBs. Therefore, the HR pathway can be co-opted by providing an artificial DNA repair template containing sequence regions homologous the DSB ends. The sequence between the homologous ends, either an insertion or replacement, is then transferred into the targeted locus during HR, enabling the generation of precisely modified‘knockin’ alleles, e.g., for codon replacements or the insertion of reporter genes.

Large sequence insertions require the use of double-stranded, plasmid-based gene-targeting vectors with homology regions of >500 bp. Shorter sequence modifications can be introduced by using synthetic single-stranded DNA oligodeoxynucleotides (ssODN) as repair templates. Repair using ssODNs is mediated by a poorly defined mechanism designated as single-strand template repair (SST-R).

In contrast to HDR, the c-NHEJ pathway mediates the religation of DSB ends without the involvement of a repair template. Although some fraction of c-NHEJ repair events result in precisely reconstituted wild-type sequences, a fraction of cleaved sequences gain a random insertion or deletion of one or more nucleotides (INDELs). Therefore, DSB repair by the supposedly error-prone c-NHEJ pathway is frequently used to generate INDELs within coding regions, which often will cause a frameshift knockout mutation.

In the context of the present invention, the terms NHEJ and c-NHEJ are used interchangeably.

In further embodiments, the present invention further relates to a-NHEJ. In preferred embodiments, a-NHEJ is not comprised by the present invention. NHEJ- and HDR-mediated DSB repair are well established processes or mechanisms that are known to the person skilled in the art and have been described by Danner et al. (Mamm Genome 28 262-274 -). The initiation of DSB repair is identical for both c-NHEJ and HDR. The ATM (ataxia telangiectasia mutated) protein kinase is a key initial regulator of the DNA damage response and coordinates DSB repair. ATM is activated by the MRN (Mre1 1-Rad50-Nibrin) complex and other factors at DNA breaks. Upon monomerization and autophosphorylation,

ATM phosphorylates Serine 139 of histone H2AX, forming yH2AX. The phosphorylated residue on yH2AX is recognized by MDC1 , which in turn recruits more MRN complexes. These further activate ATM and creates a positive feedback loop driving the expansion of yH2AX chromatin domains into yH2AX foci. MDC1 becomes phosphorylated by ATM at its TQXF repeats and initiates downstream signaling by recruiting the E3 ubiquitin ligase RNF8. RNF8 and its E2 enzyme partner UBC13 polyubiquitinate the H1 linker histone. This further promotes the recruitment of the E3 ubiquitin ligase RNF168 that ubiquitinates histone H2A at Lysine 13 and 15. H2A-K15Ub together with dimethylated Lysine 20 of histone H4 (H4K20me2) are chromatin marks for the recruitment of the checkpoint protein 53BP1. The control of accumulation of 53BP1 determines if the DSB event is repaired by c-NHEJ, or through resection and subsequent HDR.

The classical c-NHEJ pathway initiates with the localization of 53BP1 to a DSB and blocks 5' resectioning. 53BP1 blocks CtIP-based resectioning and recruits Rif 1 , which further blocks resectioning and inhibits BRCA1 accumulation. Unresected ends allow Ku70/80 to bind, further inhibiting resection. Ku proteins form a scaffold and recruit DNA-PKcs (catalytical subunit) to form the complete DNA-PK, which then recruits endprocessing factors (like Artemis) and the XRCC4/XLF/DNA Ligase-IV complex. The XRCC4/XLF factors stabilize and align the DNA fibers and DNA Ligase IV ligates the two strands. Repair of chemically or irradiation-induced DSBs is greatly complicated by the need to excise and repair damaged bases. However, this will be left out of this review as DSBs from Cas9 nucleases form blunt ends with 5'

phosphorylated DNA, the substrate for DNA Ligase IV. The ability to excise damaged bases and then ligate non-complementary strands has resulted in c-NHEJ being often thought of as a mutagenic process. However given a complimentary cut, such as created by restriction enzymes, mutation-free ligation events can be 75% or higher. Error-prone mutations that have previously been attributed to c-NHEJ are often a result of DSB resectioning and annealing through the similarly named but mechanistically distinct a-NHEJ.

Alternative non-homologous end joining pathways (a-)NHEJ encompass microhomology- mediated end joining (MMEJ), single-strand annealing (SSA), and thetamediated end joining. Once thought to only be backup pathways, a-NHEJ events can in some cases occur up to 10% of the frequency of c-NHEJ. These repair events can result in deletions of various sizes, and only sometimes anneal and ligate through microhomologies. However, they always begin with the same resection steps as in homologous recombination, involving the MRE1 1 complex and CtIP. Resection can be <20 bp for microhomology or up to thousands of bps for SSA. The choice between a-NHEJ and HR comes from the inability of RPA to be replaced by Rad51 by Rad52/BRCA2. This limits the ssDNA to proceed through the HR pathway. Importantly, the extensive resection, when repaired by a-NHEJ, results in an increased chromosomal translocation frequency, a major driver of human cancer. The homologous recombination pathway requires the exclusion of 53BP1 and resection in order to be initiated. H2A is de-ubiquitinated upon mitotic entry so 53BP1 is excluded from the chromatin. During the S/G2 phase, BRCA1 excludes Rif 1 from the foci, and recruits CtIP and the MRN complex. This complex initiates a cleavage step which is then further 5 -resected by Exo1. The resection extends 2-4 kb on each side of the DSB. The exposed single-stranded DNA (ssDNA) is quickly bound by RPA for protection. RPA is replaced by Rad51 through the action of BRCA2 and Rad52 to form a nucleofilament competent for homology search. The Rad51 filaments maintain the ssDNA in a B-form which has triplets open for Watson-Crick pairing with complementary triplets in homologous duplex DNA. It should be noted that this review highlights only some of the key factors of the HR pathway and more complete reviews are available.

In contrast to HDR, replacement of a DNA sequence of a dsDNA molecule by a DNA substitute sequence of an exogenous nucleic acid molecule according to the method of the present invention does not require that the exogenous nucleic acid molecule comprises or encodes homology arms targeted to the dsDNA molecule to be modified.

In the context of the present invention, the terms“homology arms” or“homology arms that are targeted to the dsDNA molecule to be modified” refer to regions or sequences of the exogenous nucleic acid molecule that are homologous to the sequences at the two double strand break ends of the dsDNA molecule to be modified that are not located on the DNA sequence to be replaced.

This means that for performing the method of the present invention it is not required that there is a sequence homology between the sequences of the dsDNA molecule to be modified that are adjacent to the at least two target sequences and sequences comprised by the exogenous DNA molecule comprising the DNA substitute sequence.

Homology arms have a length of at least 30 nucleotides, preferably at least 50 nucleotides, and may have 90 %, preferably 95 %, 97 %, 98 %, 99 % or 100 % sequence identity to the corresponding sequences flanking the open ends of the target dsDNA molecule. In preferred embodiments of the present invention, homology arms have a length of at least 30 nucleotides, preferably more than 35 nt, 40 nt, 45 nt, 50 nt, 60 nt, 80 nt, 100 nt or 500 nt.

Homology arms have sufficient sequence identity to ensure specific binding to the target sequence. Methods to evaluate the identity level between two nucleic acid sequences are well known in the art. For example, the sequences can be aligned electronically using suitable computer programs known in the art. Such programs comprise BLAST (Altschul et al. (1990) J. Mol. Biol. 215, 403), variants thereof such as WU-BLAST (Altschul and Gish (1996) Methods Enzymol. 266, 460), FASTA (Pearson and Lipman (1988) Proc. Natl. Acad. Sci. USA 85, 2444) or implementations of the Smith-Waterman algorithm (SSEARCH, Smith and Waterman (1981 ) J. Mol. Biol., 147, 195). These programs, in addition to providing a pairwise sequence alignment, also report the sequence identity level (usually in percent identity) and the probability for the occurrence of the alignment by chance (P-value). In accordance with the present invention it is preferred that BLAST is used to determine the identify level between two nucleic acid sequences. The method of the invention further comprises introducing into the cell an exogenous nucleic acid molecule comprising or encoding a DNA substitute sequence. An“exogenous nucleic acid” or“exogenous genetic element” relates to any nucleic acid introduced into the cell, which is not a component of the cells“original” or“natural” genome. Exogenous nucleic acids may be integrated or not integrated in the genetic material of the target cell, or relate to stably transduced nucleic acids.

In the context of the present invention, the“DNA substitute sequence”, which might also be called a“replacement sequence”, is a DNA sequence that is introduced into the dsDNA molecule to be modified at the location between the at least two double strand breaks leading to replacement of the sequence of the dsDNA molecule that was initially located between the at least two double strand breaks.

The DNA substitute sequence may be introduced into the dsDNA molecule to be modified as a linear or linearizable dsDNA molecule that can be generated from the exogenous nucleic acid molecule. The exogenous nucleic acid molecule comprising or encoding a DNA substitute sequence can be, for example, a dsDNA molecule, such as a linear dsDNA molecule, which can result from a PCR amplification, a DNA plasmid or a DNA mini-circle.

In preferred embodiments, the exogenous nucleic acid molecules consist of the DNA substitute sequence. This is possible, for example, if the exogenous nucleic acid molecule is a linear dsDNA molecule, such as for example a PCR amplification product, or a DNA mini-circle. A mini-circle is a plasmid like circular DNA that has had all other parts except the sequence of interest removed. Thus a single cut can linearize a fragment that is then ready to integrate.

In certain embodiments, the DNA substitute sequence can be introduced into the cell through delivery of a nucleic acid encoding for the DNA substitute sequence, for example by means of viral delivery through Adeno-associated viruses (AAV), retroviruses, non-integrating or integrating lentiviruses.

Genetically modified viruses have been widely applied for the delivery of genes into cells. A viral vector may be employed in the context of the present invention.

Non-viral methods may also be employed, such as alternative strategies that include conventional plasmid and nucleic acid transfer and delivery. Physical methods to introduce vectors and nucleic acid molecules or proteins into cells are known to a skilled person. One example relates to electroporation, which relies on the use of brief, high voltage electric pulses, which create transient pores in the membrane by overcoming its capacitance. One advantage of this method is that it can be utilized for both stable and transient gene expression in most cell types. Furthermore, gold nanoparticles for delivery of PCR-like double strand DNA products attached to the gold nanoparticle can be used. Alternative methods relate to the use of liposomes or protein transduction domains. Appropriate methods are known to a skilled person and are not intended as limiting embodiments of the present invention.

In preferred embodiments of the method of the present invention, the cell is a non-dividing or a slowly dividing cell, preferably a cell in the G1 or GO phase. The cell cycle or cell division cycle is the series of events that take place in a cell leading to its division and duplication of its DNA (DNA replication) to produce two daughter cells. In cells with a nucleus, as in eukaryotes, the cell cycle consists of four distinct phases: G1 phase, S phase (synthesis), G2 phase (collectively known as interphase) and M phase (mitosis). M phase is itself composed of two tightly coupled processes: karyokinesis, in which the cell's chromosomes are divided, and cytokinesis, in which the cell's cytoplasm divides forming two daughter cells. GO is a resting phase where the cell has left the cycle and has stopped dividing.

The word "post-mitotic" is sometimes used to refer to both quiescent and senescent cells. Nonproliferative (non-dividing) cells in multicellular eukaryotes generally enter the quiescent GO state from G1 and may remain quiescent for long periods of time, possibly indefinitely (as is often the case for neurons). This is very common for cells that are fully differentiated. Some cells enter the GO phase semi-permanently and are considered post-mitotic, e.g., some liver, kidney, and stomach cells. Many cells do not enter GO and continue to divide throughout an organism's life, e.g., epithelial cells.

In the context of the present invention, a non-dividing cell is a cell that is in the GO phase where the cell has left the cycle and has stopped dividing. A slowly dividing cell is a cell that has not yet left the cell cycle, but remains in the G1 or GO phase for a prolonged period, such as more than 1 day, 2 days, 1 week, 2 weeks, 1 month, 2 months, 1 year, 2 years, 5 years, 10 year, or 20 years before reentering the cell cycle and transitioning to the S phase. Importantly, HDR is only active in the S and G2 phase of the cell cycle, whereas NHEJ is active in GO, G1 , S, and G2. During the M phase no DNA repair mechanisms are active.

The ability to manipulate any genomic sequence by gene editing has created diverse

opportunities to treating many different diseases and disorders (For a review see Maeder et al, Molecular Therapy, 24:3, 2016, 430-446). Of relevance to the present invention is the

opportunity of correcting deficient gene sequences in vivo. The present invention enables gene therapy for correcting pathogenic gene sequences by DNA sequence replacement according to the method described herein. For example, hematologic disorders associated with genetic defects could be treated by gene correction in hematopoetic stem cells. Of further relevance is the treatment of liver disease, by liver-targeted gene editing and the treatment of muscle disease by gene correction in muscle stem cells. Respiratory disorders may be treated, for example, cystic fibrosis, which is caused by mutations to the CFTR chloride channel. Gene editing according to the present invention may be employed to repair the CFTR mutations in patient lung cells. Antimicrobials are another potential therapeutic target of the present invention, i.e. by targeting the genomes of other organisms, for example bacterial genes could be replaced by the method described herein.

The present invention encompasses administration of the composition of the present invention to a subject. As used herein, "administration" or "administering" shall include, without limitation, introducing the composition by oral administration. Such administering can also be performed, for example, once, a plurality of times, and/or over one or more extended periods. A single administration is preferred, but repeated administrations over time (e.g., hourly, daily, weekly, monthly, quarterly, half-yearly or yearly) may be necessary in some instances. Such administering is also preferably performed using an admixture and a pharmaceutically acceptable carrier. Pharmaceutically acceptable carriers are well known to those skilled in the art.

Administration may also occur locally, for example by injection at the site where the cells comprising the dsDNA to be modified are located, for example by endoscopic or microinvasive means.

The composition described herein may comprise different types of carriers depending on whether it is to be administered in solid, liquid or aerosol form, and whether it need to be sterile for such routes of administration as injection. The composition of the present invention can be

administered intravenously, intradermally, intraarterially, intraperitoneally, intralesionally, intracranially, intraarticularly, intraprostaticaly, intrapleurally, intratracheally, intranasally, intravitreally, intravaginally, intrarectally, topically, intratumorally, intramuscularly,

intraperitoneally, subcutaneously, subconjunctival, intravesicularlly, mucosally, intrapericardially, intraumbilically, intraocularally, orally, topically, locally, inhalation (e.g., aerosol inhalation), injection, infusion, continuous infusion, localized perfusion bathing target cells directly, via a catheter, via a lavage, in cremes, in lipid compositions (e.g., liposomes), nanoparticles, or by other method or any combination of the forgoing as would be known to one of ordinary skill in the art (see, for example, Remington's Pharmaceutical Sciences, 18th Ed. Mack Printing Company, 1990, incorporated herein by reference).

Additionally, such compositions can comprise pharmaceutically acceptable carriers that can be aqueous or non-aqueous solutions, suspensions, and emulsions, most preferably aqueous solutions or solid formulations of various types known in the art. Aqueous carriers include water, alcoholic/aqueous solutions, emulsions and suspensions, including saline and buffered media. Parenteral vehicles include sodium chloride solution, Ringer's dextrose, dextrose and sodium chloride, lactated Ringer's and fixed oils. Intravenous vehicles include fluid and nutrient replenishers, electrolyte replenishers such as Ringer's dextrose, those based on Ringer's dextrose, and the like. Fluids used commonly for i.v. administration are found, for example, in Remington: The Science and Practice of Pharmacy, 20th Ed., p. 808, Lippincott Williams S- Wilkins (2000). Preservatives and other additives may also be present, such as, for example, antimicrobials, antioxidants, chelating agents, inert gases, and the like.

SEQUENCES

The invention is further described by the following sequences. These are not intended to limit the scope of the invention, but represent preferred embodiments of aspects of the invention provided for greater illustration of the invention described herein.

FIGURES

The invention is further described by the following figures. These are not intended to limit the scope of the invention, but represent preferred embodiments of aspects of the invention provided for greater illustration of the invention described herein.

Brief description of the figures:

Figure 1 : Outline of use of RNA-guided nucleases for the replacement of sequences in cells using the non-homologous end joining pathway, as described in the present invention

Figure 2: Replacement of Sequence by fluorophore reporter in Reporter 1 cell line in Example 1.

Figure 3: Sequencing of 5’ end PCR1 of the replaced sequence in Reporter 1.

Figure 4: Sequencing of the 3’ end PCR 2 of the replaced sequence in Reporter 1.

Figure 5: Gel Showing Replacement of Sequence and larger Replacement

Figure 6: Figure showing the Florescence Readout for Reporter 1 containing Cells and Wild Type Cells Figure 7: Showing the staggered cleavage and filling in caused by Cas9

Figure 8: Replacement of Sequence by fluorophore reporter in Reporter 2 cell line in Example 2: Figure 9: Sequencing of 5’ end and 3’ end showing the replacement of sequence in Reporter 2. Figure 10: Native Gene or Exon Replacement.

Figure 1 1 : Schematic representation of a strategy for directed insertion of a DNA substitute sequence (sequence to add).

Figure 12: Schematic representation of exon replacement therapy for X-linked chronic granulomatous disease (X-CGD) and a-sarcoglycanopathy.

Figure 13: Exon replacement in the Polb gene in human NHDF cells.

Figure 14: Transfection control.

Figure 15: Analysis of Polb gene exon replacement.

Figure 16: Sequence analyis of engineered Polb alleles.

Detailed description of the figures:

Fig. 1. A) This overview underlies the concept of the invention. It utilizes a RNA-guided nuclease, in our case CRISPR/Cas9, to create two double strand breaks flanking the sequence to be replaced. The cleavage sites in this figure are“Site A” and“Site B”. The light and dark grey arrow symbolize this Cas9 recognition sequence and the triangle symbolizes the sequence on the PAM side of the break. At the same time as the Cas9/sgRNA, a sequence to be inserted in place of the original sequence is delivered into the cell. In part A this is exemplified by plasmid delivery. The new“replacement sequence” on the plasmid needs to be linearized and excised from the plasmid backbone; this excision is in this case done through the use of the same guide-RNA recognition sequences, cleaving out the replacement sequence. Finally the linear piece of replacement DNA is ligated in between the two sites of double strand breaks using the classic non-homologous end joining Pathway (c-NHEJ). It is important to note that religating the original sequence back into the genome reforms the original Cas9 recognition site which then can be cleaved out. The recognition site in the delivered replacement sequence has the recognition sequence pointing in the opposite direction so the CRISPR/Cas9 site is ablated when the replacement sequence is ligated in (notice the two light and dark grey rectangles together or two light and dark grey triangles indicating the loss of the functional recognition site). B) The use of c-NHEJ to replace sequence is a completely not enzymatic pathway for creating such a change in sequence. Due to the profoundly different enzymatic machinery, the method of the invention functions in all parts of the cell cycle, in particular G0/G1 , besides S/G2 phase. In contrast, Homology Directed Repair (HDR) functions only in late S/G2 phase of the cell cycle- limiting its use to rapidly dividing cells. C) The delivery of the replacement sequence could be in multiple forms. It could be delivered in a plasmid such as in part of this figure. It could also be delivered by a virus such as AAV or nonintegrating lentivirus. Further it could be delivered by a MiniCircle which is a plasmid like circular DNA that has had all other parts except the sequence of interest removed. Thus a single cut would linearize a fragment that is then ready to integrate. Fig. 2. This is an overview of the genetic‘Reporter 1’ that is used in Example 1 to show a replacement of sequence using the c-NHEJ pathway. The reporter is integrated into both copies of the AAVS1 Locus. A CAG promoter drives a BFP-pA sequence resulting in all cells being blue. This BFP sequence is flanked by a Cas9 recognition site“Site A”. If Cas9 cleaves both sides and excises out the BFP-pA, the following Venus-pA that had not been coding, now is moved behind the CAG promoter and turns the cells green - indicating a deletion. The replacement sequence, in this case an mCherry coding sequence, is delivered by plasmid (Delivery option 1 ) or by MiniCircle (Delivery option 2). When the replacement sequence is co-delivered with Cas9 and the sgRNA the mCherry sequence replaces the BFP-pA sequence, turning the cells Red.

Fig. 3. A) This schematic explains the PCR fragment shown in (B). PCR 1 is between the genomic sequence near the CAG reporter and the 5’ end of the mCherry replacement sequence. This PCR sequence shows that the mCherry sequence was integrated into the target of interest. B) The figure at the top of B shows the original uncleaved genomic cut site on the left and the uncleaved mCherry sequence on the right; both are then cleaved by Cas9. The lines then indicate how the cleaved sequences are ligated together to form the final product (CAG- mCherry) as shown in the bottom of portion A. In this case 70% of the targets are Insertion- Deletion Mutation (INDEL) free. 2 of the INDELs were created by sticky end cleavage explained in Figure 7. If the formation of these INDELs were problematic for a replacement (such as in a coding region), reversal of the direction of the cut site chosen (5’ protospacer - PAM 3’) can ameliorate this situation (explained in Figure 7). In this case and in many possible uses of c- NHEJ for replacement of sequence, these small INDELs would not be problematic.

Fig. 4. A) This schematic explains the PCR fragment shown in B. PCR 2 is between the 3’ end of the mCherry replacement sequence and the 5’ end of the genomic DNA. This PCR sequence shows that the mCherry sequence was integrated into the target of interest. B) The figure at the top of B shows the original uncleaved genomic cut site on the right and the uncleaved mCherry sequence on the left; both are then cleaved by Cas9. The lines then indicate how the cleaved sequences are ligated together to form the final product (CAG- mCherry) as shown in the bottom of part A). In this case 90% of the targets are Insertion-Deletion Mutation (INDEL) free. The one clone with an INDEL is a deletion formed during the c-NHEJ end processing. The lack of “insertion” INDELs is due to the portions of the Cas9 cut sequences that are being ligated. The mechanism is more clearly explained in Figure 7. For design purposes it should be noted that the two protospacer containing sides may results in more perfect ligations compared to the ligation of the PAM containing sides (due to the sticky cleavage of Cas9).

Fig. 5. A) This figure highlights that when using a plasmid delivery option it could be possible to cleave only one side of the mCherry and then integrate the entire plasmid (Option 2) instead of the desired replacement sequence (Option 1 ). The Figure highlights the PCR band in the gel in B is the 5’ region of the integrate. If the integration of mCherry was the small portion cleaved out of the backbone the size is 1.4kb. If the entire plasmid is integrated it is nearly 4kb. B) PCR 2 of single colonies of mCherry+ cells sorted after plasmid transfection. About 50% of the integrates integrated the entire plasmid and about 50% integrated the replacement sequence. While MiniCircle delivery avoids this backbone integration problem, this highlights that larger sequences (up to 4 kb in this case) can be used. Other data (not included) show 6 kb sequences inserted. Fig. 6. A) This figure shows an example transfection of the Reporter 1 containing HeLa cells. This shows that initially all cells are blue. However after transfection with Cas9/gRNA and the replacement sequence (delivered either through plasmid or MiniCircle) the cells undergo a number of outcomes as noted in Figure 2. Notably both the plasmid and MiniCircle delivery resulted in cells that took up and integrated the mCherry sequence. The plasmid based delivery was about 4% mCherry ⁺ cells, while the MiniCircle delivery showed -30% mCherry ⁺ cells.

Further genomic characterization of cells that underwent plasmid based or MC transfection is shown in Figures 3-5. B) The same transfection was done in WildType HeLa cells. These cells are identical to the reporter except they did not have the reporter placed in AAVS1. The aim of this experiment was to show that if any of the mCherry ⁺ cells in part A were due to off target integration that somehow integrated into a reading frame, then the WT cells should also undergo a similar off target integration. As none of the WT HeLa cells became mCherry ⁺ , it points to the mCherry ⁺ cells in part A originating from on target integration. This is further shown by the genotyping in Figures 3-5.

Fig. 7. This figure explains how NHEJ after Cas9 cutting can religate perfectly the restriction site or create some inserted nucleotides depending on how the Cas9 cleaves the genome. It is commonly accepted that Cas9 can blunt end cleave the genome. When this happens the double strand break can religate and reform the original sequence with no mutations. In this case the genome could be cut again if Cas9 is still around. However it is less well known that Cas9 can cut to create 5’ stick ends. In this case the sticky end overhangs are filled in by the NHEJ machinery (in grey) and then the sequence can be religated. When this happens it appears that there has been an insertion, and this destroys the PAM site upon ligation. This figure does not discuss deletions that can form during NHEJ which is a well known phenomena, nor insertions that can also be randomly created during NHEJ, but rather focuses on insertions formed due non- canonical cleavage of the genome by Cas9 that creates sticky ends.

Fig. 8. This figure shows an overview of the sequence put into the AAVS1 loci of HEK293 cells. There is a defective Venus in the loci so the cells are colorless. The sequence has two unique Cas9 recognition sites“Site C and Site D”. A plasmid is transfected that contains Cas9, sgRNAs for Site C and D, and a turboGFP sequence to replace the excised defective Venus sequence. After the turboGFP sequence and the defective Venus are cleaved out of the plasmid and genome respectively, the turboGFP can be ligated in using the c-NHEJ pathway. This results in green cells that can be sorted and genotyped.

Fig. 9. A) Shows a schematic of the transfected replacement of a sequence in HEK293 Cells. A dysfunctional fluorescent sequence was replaced by a functional turboGFP sequence. The cells were sorted and single cell colonies expanded. The gDNA was extracted from each single colony and PCR at the interface of the replacement sequence was done. B) In the 5’ end there was an unusual and highly atypical resection pattern. C) At the 3’ end there was insertions and deletions. However both shows that the turboGFP was correctly integrated into the genome, replacing the sequence in the correct location.

Fig. 10. The gene Lamin A is targeted for replacement of Exon 2 with with a fluorophore (GFP) as described in example 3. Fig. 11. Directed integration of a DNA substitute sequence can be achieved by using three different guide RNAs. Guide RNA 1 (dark grey) recognizes a target side upstream of the DNA sequence to be replaced and downstream of the DNA substitute sequence (sequence to add). Guide RNA 2 (light grey) recognizes a target side downstream of the DNA sequence to be replaced and upstream of the DNA substitute sequence (sequence to add). Upon cleavage and removal of the DNA sequence to be replaced the DNA substitute sequence can either integrate in the desired orientation (Option A, inserted correct direction), in the undesired orientation (Option

B, inserted reverse direction) or the open ends of the DNA molecule to be modified can be ligated with each other without insertion of the DNA substitute sequence (Option C, deletion is formed). If option A occurs, the resulting DNA molecule cannot be cut again, whereas in case of option B or

C, a target side that is recognized by Guide RNA 3 (dark grey rectangle and light grey arrow head) is formed and can be cut again (recleavage) to recover the open ends for another integration event.

Fig. 12. In this schematic representation of a treatment strategy aiming to replace a bad exon carrying a disease-causing mutation, for example in X-CGD or a-sarcoglycanopathy, directed integration of a DNA substitute sequence comprising the corrected/non-pathological exon sequence can be achieved by employing the directed integration strategy describe in Fig. 1 1.

Figure 13: For targeting of POLB exon 5 in NHDF cells, a Cas9 plasmid containing two guide RNAs targeting sequences upstream and downstream of exon 5, as well as an mCherry containing mini-circle were nucleofected into NHDF cells. The cells were analysed by flow cytometry and sequencing after a week.

Figure 14: FACS Analysis of Control nucleofection for estimation of transfection efficiency. A CAG-Venus plasmid was nucleofected into NHDF cells . Right panel: 53% of the cells were positive for GFP as compared to the negative control (left panel). The number of highly positive GFP cells (inserted box) reached 23.9 % of the cells, i.e. about half of the transfected cell population.

Figure 15: FACS analysis one week after nucleofection of the Cas9 plasmid and the POLB- Minicircle containing the Cherry reporter gene into NHDF cells. 14% of the fibroblasts are mCherry positive. The mCherry MC alone contains no promoter.

Figure 16: Sanger sequencing of mCherry sequence replacement targeted to POLB exon 5 in NHDF cells. The two regions highlighted are at the ligated interface of genomic and exogenous DNA.

EXAMPLES

The invention is further described by the following examples. These are not intended to limit the scope of the invention, but represent preferred embodiments of aspects of the invention provided for greater illustration of the invention described herein.

Materials and methods of the examples

Plasmid Construction: For SEQ ID NO 1 : oligonucleotides matching the protospacer of SEQ ID N02: named Rosa26 were ordered from IDT Biologika. The oligonucleotides were hybridized and ligated into a backbone vector (Addgene #62988; www.addgene.org) containing the rest of the sgRNA as well as CAG-Cas9-2A-puro.

For SEQ ID NO 6 (plasmid mCherry): The mCherry coding sequence was amplified from a plasmid. Both primers contained the Target Site A + PAM sequence (SEQ ID N02) and each contained either Pad or Pvul for ligation into a backbone vector.

For SEQ ID NO 7 (MC mCherry): The mCherry coding sequence was amplified from SEQ ID N06. Primer included restriction sites for cloning into the MiniCircle SBI MC-Easy production system. MiniCircles were produced as per the manufactures instructions.

Cell Transfection. Sorting, Expansion

Reporter HeLa cells containing the CAG- BFP-pA-Venus-pA sequence in the AAVSI loci (Figure 2, SEQ ID N03) where transfected with the plasmid (SEQ ID N01 ). This plasmid drives Cas9 production, the guide for the targeting sequences flanking the BFP-pA, and contains a puromycin resistance gene. This was co-transfected with the replacement mCherry sequence on either the full plasmid (SSEQ ID N06) or the minimal MiniCircle plasmid (SEQ ID N07). Plasmids were delivered by reverse transfection using Lipofectamine 3000. After 24 hours selection began for puromycin resistance; selection was carried out for 48 hours. The cells were expanded for 10 days and then FACS sorted. mCherry+ cells were collected both as a mixed population and also mCherry+ cells were plated for single cell colonies. Single Cell colonies were expanded and gDNA extracted for genotyping. Cells were cultured in DM EM/ 10% FBS/ 1 % Pen/Strep.

Reporter HEK293 cells containing the nonfunctional Venus in the AAVS1 loci (Figure.8 and SEQ ID N010) were transected with a plasmid containing Cas9, sgRNAs for two cut sites, and a turboGFP sequence that could be linearized and excised (SEQ ID N01 1 ) using Lipofectamine 2000. These cells were cultured 10 days and FACS sorted for GFP+ cells. Single clonal colonies were made and expanded. The gDNA was extracted from such colonies.

Genotyping Cells:

For cells that were a mixed population of mCherry+ or single cell derived clones, the interface was sequenced at the 5’ and 3’ ends of the insert. The 5’ junction (PCR 1 on Figure 2) the primers 5’- TTGTCCCAAATCTGTGCGGA- 3’ and 5’-CAAGTAGTCGGGGATGTCGG-3’ were used. The 3’ junction of the mCherry insert with the 5’ region of genomic DNA (PCR 2 noted on Figure 2) primers 5 -CCGACTACTTGAAGCTGTCCTT-3’ and 5’- CCCATATGTCCTTCCGAGTGAG-3’. Sequencing of this PCR product was done with primer 5’-GACTACACCATCGTGGAACAGT-3’. Both PCRs reactions were done with Hurculase 35x cycles, 98C- 30s, 58C-30s, 72C- 45s. For the mixed population mCherry cell PCR, the fragments were ligated into the CloneJET PCR Cloning kit. Mini-preps were made and sent for sequencing (Figures 3 and 4) (SEQ ID N08 and SEQ ID N09). For HEK293 Cells with Reporter 2 that were GFP+ 10 days after transfection, single cell colonies were made. These colonies were expanded and gDNA extracted. The 5’ end PCR was done with primers 5’-CGTAGGTGTAGCGGTAGTTAAC-3’ and 5’- TTCGGCTTCTGGCGTGTGACC-3’. The 3’ end PCR was done with 5’- GTTAACTACCGCTACACCTACG-3’ and 5’-GGTACAGCATCTCGGTGTTGG-3’. The PCR was done 35x cycles, 98C- 30s, 55C-30s, 72C- 30s.

Methods for NHDF Polb Knock-in:

The human POLB exon 5 genomic region, the expression vector for Cas9 and guide RNAs, as well as the minicircle with the cherry reporter sequences are shown in the sequence section below. For a given experiment 1 pg of a plasmid containing Cas9 and guide RNAs were co-nucleofected with 3 pg of the POLB-Minicircle containing the replacement mCherry sequence (Lonza program DS-150) into 500.000 NHDF cells. Upon correct integration of the minicircle, the transcript contains a 2A-mCherry-pA sequence. NHDF cells (human juvenile foreskin fibroblasts, purchased from Promocell, Heidelberg, Germany) were cultured in

DMEM/10%FCS/1 % Pen/Strep. Expression of mCherry was measured by FACS after 8 days. An overview is shown in Figure 1. Sanger sequencing shown in Figure 3: the genomic DNA of the entire population of Polb Targeted NHDF cells was extracted with QuickExtract and amplified by primers POLBP3/POLBP4 or POLBP5/POLBP6 to amplify the 5’ or 3’ end respectively. The bands were gel extracted, purified, and Topo-cloned. The clones were Sanger sequenced using primer POLBP1 or POLBP2 to obtain sequence.

Primers for Togo Cloning/ sanaer sequencing of POLB exon 5:

PolbPI : 5’CCATACCCGGCCATCTTTTAGA 3’

PolbP2: 5’ GCTTGAGGGCTTGTTCCAAATT 3’

PolbP3: 5’CCATACCCGGCCATCTTTTAGA 3’

PolbP4: 5’ ACTCCTTGATGATGGCCATGTT 3’

PolbP5: 5’ ATCAAGTTGGACATCACCTCCC‘3

PolbP6: 5’ GCTTGAGGGCTTGTTCCAAATT 3’

Results of the Examples

Example 1: Replacement of sequence in Reporter 1 Cell line: replacing a Blue Fluorescent Protein (BFP) with a Red Fluorescent Protein (mCherry) sequence using Non-Homologous End Joining after two genomic cleavages by CRISPR/Cas9.

To test if it is possible to cut out a piece of the genome and insert in another linear fragment in its place all at once, a florescent reporter was used (Figures 1 and 2). Reporter 1 in HeLa cells contains the CAG-BFP-pA-Venus-pA sequence inserted into the AAVSI loci (SEC ID N03). This sequence has a Cas9“target site A” inserted on both sides of the BFP so that when Cas9+Target Site A sgRNA are introduced the reporter can excise out this sequence. Normally this would require two unique guides, but a single one was used for simplicity in this reporter. These Reporter 1 containing HeLa cells where transfected with the plasmid (SEQ ID N01 ). This plasmid drives (i) Cas9 production, (ii) the guide for the targeting Target Site A flanking the BFP-pA, (iii) a Puromycin resistance gene. A second plasmid was added in addition to the Cas9 containing plasmid. This second plasmid contained the mCherry sequence required for sequence replacement in either a plasmid or MiniCircle form. If the plasmid was used, the Cas9 was designed to cut two places on the plasmid, freeing a linear piece of mCherry coding DNA. If the MiniCircle was used, a single cut site linearized the small plasmid and it was then ready for integration (Figure 2).

These reporter cells were selected by Puromycin to enrich for plasmid-transfected cells and analyzed by flow cytometry The cells could be red (mCherry), blue (BFP), green (Venus), or colorless. If they were red, this was due to the integration of the mCherry coding sequence into the reporter. Green was due to a deletion. BFP was due to a maintenance in the native gene, and colorless was due to some unspecific damage to the loci. Using the plasmid based mCherry delivery, about 5% of the cells were red (Figure 6a). Using the MiniCircle delivery, about 30% of the cells became red. Further when wild type HeLa cells (without the Reporter 1 construct) were transfected with these plasmids, no cells became red (Figure 6b). Noting that the red cells were due to on target integration.

To check that the mCherry was correctly inserted the cells were genotyped. First, the gDNA was extracted out of the transfected cells. Genotyping PCRs of the 5’ end integration site (Figure 3, SEQ ID N08) show that the mCherry was correctly inserted. PCRs of the 3’ end show that the integrate mCherry is correctly inserted. It would be possible that the cells turn red on insertion of the mCherry even if the original genomic DNA was not deleted. However sequencing in Figure 4 (SEQ ID NO 12) shows that the region behind the mCherry show that the mCherry replaced the original sequence and did not simply insert.

In comparison to the MiniCircle delivery, the plasmid based delivery requires two cuts to excise the sequence and prepare it for integration into the genomic region. However if it is not completely cut out of the backbone, then the entire sequence can be integrated into the region of interest. While this points to the utility of MiniCircle delivery, it also shows that larger sequences can be effectively integrated into the site. Figure 5 shows that genotyping of the single cell colonies (single cells expanded to allow for easy genotyping) derived from plasmid based replacement show—1/2 of the colonies that were mCherry+ had the entire 3kb plasmid integrated.

Example 2: Replacement of a nonfunctional Venus fluorophore encoding sequence in Reporter 2: replacing the sequence with turboGFP using Non-Homologous End Joining after two genomic cleavages by CRISPR/Cas9

A second reporter cell line containing a nonfunctional venus existed in HEK293 cells as shown in Figure 8 (SEQ ID NO10). This reporter was targeted as a second example of sequence

replacement. The targeting of the sequence used two unique guide sites“Target Site B” and“Target Site C” (Figure (SEQ ID N012 and SEQ ID N013). The transfected vector (SEQ ID N011 ) contained the Cas9, two guides, and the turboGFP to replace the excised genomic sequence. The turboGFP on the transfected plasmid was flanked by Target Site B and Target Site C. After the HEK293 cells with Reporter 2 were transfected with the plasmid, the cells were cultured for 10 days and then FACS sorted. GFP cells were taken and single colonies grown. Genotyping of the insert was done and shown in Figure 9. While the 5’ end shows an unusual deletion pattern, both ends show correct replacement of sequence in the turboGFP in the colonies.

Example 3: Using Replacement of Sequence on Native Gene

The gene Lamin A is targeted for replacement of Exon 2 with with a fluorophore (GFP) (Figure 10). -200 nucleotides upstream and downstream of Exon 2, Cas9 Pam/Protospacers sequences are identified (cut site 1 and 2). These two sequences have the same directionality with regard to the protospacer/Pam and so a new sequence is formed when the two sites are cut and the region in- between is excised. A third guide RNA targeting this newly formed cut-site (cut site 3) is to be used. Synthetic guides that target sites 1 , 2, and 3 are complexed with recombinant Cas9 protein, and then are nucleofected along with the MiniCircle plasmid into the host HeLa cells. This minicricle sequence has a PAM/Protospacer sequence that matches the cut site 1 , upstream of the exon. Further it also contains the intronic region flanking Exon 2 so that it can be spliced into the mature mRNA correctly. Within the coding region, the exon 2 sequence is replaced with a in-frame 2A-GFP sequence so that correctly inserted sequence causes the cells to become blue. Cells are cultured for 3-5 days to recover and then FACS sorted for GFP+ cells. These cells are taken for single colony sorting.

Further, pooled populations of GFP+ cells are analyzed through PCR through the entire genomic region and sequencing with long read PacBio sequencing to quantify successful replacement and the mutation frequency.

Example 4: Using Replacement of Sequence for Therapeutic Applications in Vivo

For the use of the replacement of sequence for gene therapy patients with Limb Girdle Muscular Dystrophy 2D. A mutation in the Sarcoglycan A (SCGA) gene in Exon 7 results in the slow deterioration of the muscles throughout the body. A mouse model with the patient mutation is generated. A C>A mutation in exon 7 results in Ser>Stop (ochre). Treatment of this using replacement of sequence utilizes three Cas9 gRNAs. Guide 1 is 200 nt downstream of Exon 7.

Guide 2 is ~200nt upstream of Exon 7. Guide 3 is formed from the deletion of the genetic region between guide 1 and 2. These three guides are driven by a U6 promoter and encapsulated in an AAV Virus. Cas9 driven by a truncated promoter is also encapsulated in an AAV virus. A third AAV virus contains the replacement sequence. The replacement sequence is the same Exon 7 sequence without the disease causing mutation. The replacement sequence in the AAV is flannel by cut site 1 and 2 in such a way that correct orientation of the integration destroys the Cas9 cut sites, but reverse integration causes reformed cut sites 1 and 2. Injection into the musculature of the mouse is performed with the three viruses. After 3 weeks the muscle is sectioned and stained for correct SGCA sequence and protein production.

Example 5: Replacement of the POLB gene exon 5 in primary human Fibroblasts

In this example it is demonstrated that an exon of an endogenous target gene, specifically the exon 5 comprising 59 basepairs of the Polymerase-b (POLB) gene, can be replaced in primary human fibroblasts by an artificial exon consisting of a splice acceptor element and a fluorescent Cherry reporter coding sequence (Figure 13). Sequence replacement was achieved by transfection of Normal Human Dermal Fibroblasts (NHDF) cells with a minicircle plasmid (carrying a splice acceptor and cherry reporter gene) together with an expression vector for Cas9 and two POLB specific guide RNAs that target sequences up- and downstream of exon 5 at a distance of 407 basepairs (Figure 13). The efficiency of transfection was controlled by using a GFP expression plasmid and FACS analysis. As shown in Figure 14, 53.9% of the cells were transfected, 23.9% showed high expression levels.

Eight days after transfection, the population was analyzed for Cherry expression by flow cytometry to detect successful replacement events. This analysis showed that 14.1 % of the transfected cell population were Cherry positive cells, indicating successful exon replacement at the POLB locus (Figure 15). Since the transfection reached only 53.9 % of the cells it can be estimated that exon replacement occurred within the transfected cell population at a frequency of approximately 26%. Under the assumption that exon replacement may occur only in the fraction of highly transfected cells (23.9%) the efficiency of exon replacement would be extrapolated to occur in 59% of the cells.

PCR amplification followed by Sanger sequencing proved the successful replacement of POLB exon 5 with the reporter gene together with minimal end resections (Figure 16).