HK40008760A - Method to analyze and optimize gene editing modules and delivery approaches - Google Patents

Description

Methods of analyzing and optimizing gene editing modules and delivery protocols

Technical Field

The present invention is in the field of gene editing to produce modified cell lines or organisms. More precisely, a method for determining a genomic integration event is reported herein.

Background

The technology to engineer biological systems and organisms is essential for basic science, medicine and biotechnology (Ran et al, 2013). In recent years, several genome editing technologies have emerged, including Zinc Finger Nucleases (ZFNs) (Miller et al, 2007; Sander et al, 2011; Wood et al, 2011), transcription activator-like effector nucleases (TALENs) (Hockemeyer et al, 2011; Sanjana et al, 2012; Wood et al, 2011; Zhang et al, 2011), and RNA-guided CRISPR/Cas9 nuclease systems (Cho et al, 2013; Cong et al, 2013; Makarova et al, 2011; Ran et al, 2013). The CRISPR/Cas9 nuclease system is guided by a small RNA containing 20 nucleotides complementary to the target DNA sequence (Fu et al, 2014). In contrast to TALENs and ZFNs, the CRISPR/Cas9 system is said to be easier to design, efficient and well suited for high throughput and multiplex gene editing for a variety of cell types and organisms (Ran et al, 2013). In addition, to produce transgenic animals (Drosophila melanogaster, zebrafish, mouse, rat, and rabbit), gene editing techniques are necessary for genetically-driven tools (genetic motivatools), such as in cell lines, primary cells (somatic and pluripotent stem cells), and fertilized oocytes (McMahon et al, 2012; Urnov et al, 2010). The first application of therapeutic genome editing was CCR5(Gori et al, 2015; Tebas et al, 2014) in autologous CD4T cells of HIV patients entering the clinic. In addition to this application, gene editing has also been successfully applied in a number of diseases at the preclinical level as well as in phase I clinical trials (Cox et al, 2015; Holt et al, 2010; Li et al, 2011; Perez et al, 2008; Tebas et al, 2014; Yin et al, 2014). The mechanisms to achieve genome editing-based therapy are to correct or inactivate harmful mutations, introduce protective mutations, add therapeutic transgenes and destroy viral DNA (Cox et al, 2015). Thus, gene editing protocols, including those with CRISPR/Cas9 technology, can lead to a whole new class of therapeutics directed to different diseases.

For therapeutic applications (and for effective use in research and development), characterization and comparison of gene editing techniques and modules is necessary. This includes analyzing and comparing their efficiency and specificity, as well as optimizing delivery to target cells. Optimal delivery and assessment of specificity are crucial for safe and effective clinical transformation of gene editing technology (Gori et al, 2015). For example, there is an urgent desire to reduce off-target activity and increase specificity in the CRISPR/Cas9 system (Zhang et al, 2015). Off-target mutations may occur frequently, at much higher rates than expected for targeted mutations and insertions. This may induce genomic instability and disruption of the function of normal genes (Cho et al, 2014; Fu et al, 2013; Mali et al, 2013 a; Pattanayak et al, 2013; Zhang et al, 2015). The objective was to optimize the various components of the CRISPR/Cas9 system, promoting a reduction in off-target activity without loss of on-target cleavage efficiency (Zhang et al, 2015).

One premise for developing and optimizing CRISPR/Cas 9-derived applications is to reliably and robustly detect and unambiguously determine heterozygous and homozygous gene inactivation as well as non-specific and targeted integration events. Existing methods, such as determining insertion-induced phenotype (e.g., drug resistance) or lack of phenotype (gene inactivation) or genetic analysis/sequencing protocols, often do not distinguish between homozygous and heterozygous inactivation. In addition, the prior art has little to resolve the genetic makeup of individual cells or has not supported statistical analysis of the stability based on a large number of individual gene editing cells.

Picco, g. et al disclose diphtheria toxin resistance markers (scientific. rep.5(2015)1-11) for selection of stably transduced human cells in vitro and in vivo. Stahl, S. et al disclose that loss of diphtheria amide pre-activates the NF-. kappa.B pathway and the death receptor pathway and causes MCF7 cells to be hypersensitive to tumor necrosis factor, including supporting information (Proc. Natl. Acad. Sci. USA 112(2015)10732-10737+6 p).

US 2016/058889 discloses that CRISPR/Cas 9-mediated gene editing (Myo-editing) is effective in correcting dystrophin gene mutations in mdx mice (duchenne muscular dystrophy (DMD) model). WO 2016/109840 discloses cell lines for efficient editing of genomes using cas/CRISPR systems, methods of producing such cell lines, and methods of producing mutations in the genome of an organism using such cell lines. Pradeep, g.k. et al disclose that modification of elongation factor-2 by diphtheria amide renders mammalian cells resistant to ricin (cell. microbiol.10(2008) 1687-.

WO 2007/143858 discloses Dph2 gene deletion mutants and uses thereof.

Carette, J.E. et al discloses the identification of host factors utilized by pathogens by haploid gene screening in human cells (Science 326(2009) 1231-1235). Roy, v. et al disclose that a dominant negative regime to prevent diphtheria amide formation can confer resistance to pseudomonas exotoxin a and diphtheria toxin (PLOS ONE 5(2010) 1-7).

Brief description of the invention

Herein is reported a robust and simple method to quantify and optimize gene editing protocols based on gene inactivation in combination with toxin and optionally antibiotic selection. This method not only allows to determine gene editing efficiency on a large number of independent cells, i.e. suitable for high throughput methods or applications, but also allows to distinguish between homozygous and heterozygous integration events and to distinguish between site-specific and non-specific gene disruption and/or integration events. The simplicity and robustness of this approach allows analysis and direct comparison of the efficiency and specificity of different gene editing modules to identify appropriate cell cloning and integration events, leading to desirable characteristics for research and development as well as therapeutic applications.

In one aspect, herein is reported a robust method for quantifying CRISPR/Cas-mediated gene changes. Using this method, for example, the efficiency of gene editing events, the difference between site-specific and non-site-specific gene disruptions, and sequence integration events can be determined. Thus, the methods as reported herein can be used to quantify and/or optimize CRISR/Cas mediated gene inactivation and integration events.

One aspect as reported herein is a method for determining the introduction of a nucleic acid into the genome of a mammalian cell, wherein the mammalian cell comprises in one embodiment one or more transcriptionally active alleles of the DPH1, DPH2, DPH4 and/or DPH5 genes, in another embodiment two transcriptionally active alleles of the DPH1, DPH2, DPH4 and/or DPH5 genes, the method comprising the steps of

-transfecting mammalian cells with one or more plasmids comprising the nucleic acid to be introduced and the elements required for gene editing of the DPH gene,

-culturing the transfected cells in the presence of a DPH gene transcriptional susceptible toxin,

-determining that the nucleic acid is introduced into the genome of the mammalian cell if the transfected cell survives in the presence of the toxin.

In one embodiment of the method as reported herein, the DPH gene transcription susceptible toxin is selected from the group consisting of pseudomonas exotoxin, diphtheria toxin and cholix toxin.

In one embodiment of the method as reported herein, the method comprises the steps of:

-transfecting mammalian cells with one or more plasmids comprising the nucleic acid to be introduced, the nucleic acid conferring resistance to a selectable marker, and the elements required for gene editing of the DPH gene,

-culturing the transfected cells in the absence of selective pressure,

-dividing the culture into at least two aliquots, or taking at least two samples from the culture, and

-incubating the first aliquot or sample in the presence of a DPH gene transcriptional susceptible toxin and incubating the second aliquot or sample in the presence of the corresponding selection marker.

The method as reported herein can be used to determine gene editing efficiency. In order to perform this analysis on a statistical basis, a large number of individual cells must be processed and analyzed independently.

In one embodiment of the method as reported herein, the mammalian cells are a plurality of mammalian cells and the method comprises the following step immediately preceding the step of culturing the cells in the presence of a toxin and/or a selectable marker

-placing each of the transfected plurality of cells separately as a single cell.

In one embodiment of the method as reported herein, said plurality of mammalian cells is 1000 to 10,000,000 cells.

This analysis of multiple cells obtained from the same transfection allows the determination of the efficiency and specificity of the gene editing step.

In one embodiment of the method as reported herein, the method is used for determining gene editing efficiency, for determining gene editing specificity, or for determining gene editing efficiency and specificity.

This analysis further allows to distinguish between homozygous and heterozygous gene modifications, as well as to distinguish between site-specific and non- (site) specific gene disruption and integration events, since statistical analysis can be performed on the transfection results.

In one embodiment of the method as reported herein, the method is used for determining homozygous and heterozygous gene modifications, or for determining site-specific and non- (site) specific gene disruption and integration.

Inactivation of the homozygous DPH gene results in toxin resistance. In this case, all alleles of the DPH gene are inactivated in the cell.

In one embodiment of the method as reported herein, the introduction of the nucleic acid is at least a single introduction into the genome of the mammalian cell, if the transfected cell survives in the presence of the selectable marker.

Toxin sensitivity can distinguish between homozygous target gene modifications and heterozygous target gene modifications and no modifications. Thus, if the transfected cell is sensitive to the toxin but not to the selectable marker, the transfection results in either heterozygous or non-specific gene integration.

In one embodiment of the method as reported herein, the introduction of the nucleic acid is the introduction of the nucleic acid into the genome of the mammalian cell, if the transfected cell is not viable in the presence of the toxin but survives in the presence of the selectable marker.

In one embodiment of the method as reported herein, the introduction of the nucleic acid is a knock-out (complete functional inactivation) of the nucleic acid in the genome of the mammalian cell, if the transfected cell survives in the presence of the toxin.

Resistance to the selectable marker indicates integration of the nucleic acid. If this is the case in the absence of toxin resistance, the integration occurs at a different location in the genome than the DPH gene. If the transfected cells are simultaneously resistant to the toxin, then, without being bound by theory, integration of the nucleic acid occurs in the DPH gene, resulting in its inactivation.

In one embodiment of the method as reported herein, DPH gene inactivation events and nucleic acid integration events are quantified by a combination of incubation in the presence of a toxin and a selectable marker and optionally High Resolution Melting (HRM) PCR.

By comparing the frequency of toxin resistance, selectable marker resistance or dual resistance, integration and integration combined inactivation can be distinguished. In the first case, this is non- (site-) specific integration (nucleic acid integration, but the target gene has not been inactivated), while in the second case, without being bound by theory, this is site-specific integration (nucleic acid integration and the target gene has been inactivated). Inactivation of the homozygous gene results in toxin resistance; targeted nucleic acid integration results in toxin and selectable marker resistance; non-targeted integration events only lead to selectable marker resistance.

By comparing the frequency of toxin resistance, selectable marker resistance and dual resistance, the specificity and efficiency of the gene editing process can also be determined. Thus, with the method as reported herein, different gene editing methods can be evaluated and ordered, or similarly, different elements used in the same/one gene editing method can be evaluated and ordered.

Thus, a robust readout of the method as reported herein, i.e. the number of toxin-resistant or selectable marker-resistant colonies, can be used to evaluate the effect of the method variables (e.g. the sequence length of the guide RNA in the CRISPR/Cas gene editing method) on the efficiency and type of gene modification.

In one embodiment of the method as reported herein, the method is for evaluating/comparing the efficiency of different gene editing methods and comprises the following steps

-transfecting a mammalian cell with one or more plasmids comprising the nucleic acid to be introduced, the nucleic acid conferring resistance to a selectable marker, and the elements required for a first gene editing method for editing the DPH gene,

-culturing the transfected cells in the absence of selective pressure,

-dividing the culture into at least two aliquots, or taking at least two samples from the culture,

-incubating a first aliquot or sample in the presence of a toxin dependent on the expression of the DPH gene and incubating a second aliquot or sample in the presence of a corresponding selection marker,

-repeating these steps for all gene editing methods to be tested, and

-ranking different gene editing methods based on the frequency of toxin resistance, selectable marker resistance or dual resistance.

The diphtheria toxin performs ADP-ribosylation of the diphtheria amide on eEF2, thereby inactivating eEF 2. This irreversibly stops protein synthesis and kills the cells. Diphtheria amide is a defined histidine modification produced by diphtheria amide synthesis genes such as diphtheria amide biosynthesis genes 1, 2, 4 and 5(DPH1, DPH2, DPH4 and DPH 5). Inactivation of these genes will stop the synthesis of the toxin target and render the cell resistant to pseudomonas exotoxin a and diphtheria toxin (Stahl et al, 2015).

The frequency of inactivation of all alleles of the target gene can be detected in a fast and robust manner by counting toxin-resistant colonies.

In one embodiment of the method as reported herein, the frequency of inactivation of all alleles of the target gene (in vitro) in the cells is detected by counting toxin-resistant colonies.

Cells in which only one allele of the target gene has been modified can be identified by directly performing HRM-PCR assays on cultured cells. Modification of the CRISPR/Cas target site will alter the melting temperature of the corresponding DPH gene-derived PCR fragment compared to the wild-type gene-derived fragment. This can be reflected by a biphasic melting curve in the HRM spectra.

In one embodiment of the method as reported herein, the (in vitro) inactivation of one allele of the target gene in the cell is detected by HRM-PCR by means of the presence of a biphasic melting curve.

In one embodiment of the method as reported herein, the HRM-PCR is performed directly on the cultured cells.

Since the CRISPR/Cas 9-mediated gene inactivation event is rarely identical on both (all) alleles in a cell, many cells with complete gene inactivation will also show a biphasic HRM curve. These cells can be distinguished from single allele changes by their toxin resistance phenotype.

In one embodiment of the method as reported herein, the frequency of inactivation of all alleles of the target gene by CRISPR/Cas9 is detected by counting the duplex melting curves determined by toxin resistant colonies in combination with HRM-PCR.

Thus, the combination of high throughput HRM-PCR assay and toxin selection (colony counting) assay on cells enables quantification of heterozygous and homozygous DPH gene-specific modification events.

The exemplarily used puromycin-N-acetyl-transferase (PAC), encoded by the integration cassette of the applied CRISPR/Cas9 plasmid, can inactivate puromycin (PM, selectable marker) and thus render the cells resistant to PM. Since this is a general principle, it applies to any antibiotic commonly used for selecting positively transfected cells.

Thus, toxin resistance, e.g., Pseudomonas Exotoxin (PE) and Diphtheria Toxin (DT) resistance, is produced by inactivation of specific and homozygous target genes.

Selectable marker (e.g., PM) resistance is a characteristic of any integration event, regardless of the location of integration.

The frequency of site-specific integration versus non- (site) -specific integration can be resolved by comparing the number of selectable marker (e.g., PM) resistant cells exposed to the target gene-specific guide RNA to the number of cells exposed to the scrambled non-specific guide RNA.

In one embodiment of the method as reported herein, a plasmid encoding a DPH gene-specific CRISPR/Cas9 module is transfected into mammalian cells, followed by HRM-PCR and colony counting assays (to Detect Toxin (DT) resistant cells and selectable marker (PM) resistant cells) on the transfected cells, wherein the method is used to determine the frequency of site-specific integration and non-site-specific integration.

Inactivation of the DPH gene showed absolute dependence on the matching guide RNA sequence, whereas the scrambled guide RNA did not give rise to any DT-resistant colonies. This suggests that CRISPRR/Cas 9/DPH-guided PAC gene integration occurs preferentially at the DPH gene, but without absolute specificity.

In one embodiment of the method as reported herein, the DPH gene is selected from the group consisting of DPH1 gene, DPH2 gene, DPH4 gene and DPH5 gene.

Inactivation of both alleles of any of the DPH1 gene, DPH2 gene, DPH4 gene, and DPH5 gene will confer absolute toxin resistance.

In one embodiment of the method as reported herein, the method comprises the steps of: the number of toxin-resistant colonies, the number of antibiotic-resistant colonies, and the number of toxin-to-antibiotic-resistant colonies were determined, wherein the ratio between the integration event (antibiotic-resistant colony number) and the inactivation event (toxin-resistant colony number) is/reflects the specificity of the method.

In one embodiment of the method as reported herein, the method is for selecting a guide RNA to achieve CRISPR/Cas 9-directed integration of a nucleic acid, wherein the method comprises the steps of: a plurality of different guide RNAs are provided and the guide RNA with the highest ratio between integration events (antibiotic resistant colony numbers) and inactivation events (toxin resistant colony numbers) is selected.

The frequency of toxin-resistant colonies reflects the inactivation of the target gene-specific homozygous gene. At the same time, the number of antibiotic resistant colonies and the number of toxin-antibiotic dual resistant colonies were evaluated to monitor cassette integration. The ratio between integration events (antibiotic resistance) and inactivation events (toxin resistance) can be used as a 'specific indicator' to identify conditions where specific integration occurs and where there are minimal simultaneous gene inactivation events. Such conditions may be advantageous if targeted integration is desired while not encountering a large number of non-productive target gene lesions.

Low values (e.g., fewer antibiotic-resistant colonies relative to toxin-resistant colonies) reflect ineffective integration relative to concurrent inactivation events. High values (more antibiotic resistant colonies and/or relatively reduced number of toxin resistant colonies) reflect non-specific cleavage at the target gene affected by CRISPR/Cas 9.

Shorter guide RNAs not only improved the integration efficiency (overall higher number of antibiotic resistant colonies), but also improved the ratio between productive gene editing and non-productive gene editing (reduced toxin resistant colonies without insertions).

In one embodiment of the method as reported herein, the gene editing method is selected from CRISPR/Cas, zinc finger nuclease and TALEN.

Gene editing modules and parameters can be optimized by DPH gene modification and thereafter shifted to optimizing gene editing efficiency or specificity of other genes.

Thus, 20 mers may be selected if the goal is for the most efficient inactivation of the gene, and 16-18 mers may be preferred if integration without excessive destructive editing is desired.

Since gene inactivation alters the guide RNA target sequence, the most suitable guide RNA can be determined using the methods reported herein if the aim is to reapply the CRISPR/Cas module to previously treated cells to increase the integration efficiency. Thus, only the unaltered gene can be modified by the original CRISPR/Cas9 component, while the modified gene (without integration) is not susceptible to the first approach used using the same guide RNA.

One aspect as reported herein is a method for identifying/selecting a CRISPR/Cas9 or a ZFN or TALEN (mutant forms or variants thereof) or other gene editing module comprising the following steps

-providing/preparing a plurality of variants of one or more gene-editing modules,

determining the efficiency and/or the highest ratio between integration events (antibiotic resistant colony numbers) and inactivation events (toxin resistant colony numbers) with the method as reported herein, and

-identifying/selecting the variant with the highest efficiency and/or the highest ratio.

One aspect as reported herein is a method for selecting a compound or combination of compounds that alters (enhances or decreases) the efficiency or specificity of a gene editing module/method, said method comprising the following steps

-providing one or more compounds or one or more combinations of compounds,

-determining the efficiency and/or ratio between integration events (antibiotic resistance colony numbers) and inactivation events (toxin resistance colony numbers) optionally in the absence of said compound or combination of compounds, using the method as reported herein,

determining the efficiency and/or ratio between integration events (antibiotic resistant colony numbers) and inactivation events (toxin resistant colony numbers) separately/individually for each of said compounds or combinations of compounds in the presence of said compounds or combinations of compounds using the methods as reported herein,

-identifying/selecting at least one compound or combination of compounds having an efficiency and/or ratio which is different from the efficiency and/or ratio obtained by carrying out the method as reported herein in the absence of said compound or combination of compounds.

One aspect as reported herein is a method for determining the concentration of a compound and its addition time point to enhance the efficiency or specificity and minimize growth inhibition or toxicity of a gene editing method comprising the following steps

-providing one or more compounds or one or more combinations of compounds,

-determining the efficiency and/or ratio between integration events (antibiotic resistance colony numbers) and inactivation events (toxin resistance colony numbers) optionally in the absence of said compound or combination of compounds, using the method as reported herein,

determining the efficiency and/or ratio between integration events (antibiotic resistant colony numbers) and inactivation events (toxin resistant colony numbers) separately/individually for each of said compounds or combinations of compounds in the presence of said compounds or combinations of compounds using the method as reported herein at different concentrations and/or addition time points,

-identifying/selecting for each compound or combination of compounds of said at least one compound or combination of compounds, respectively, a concentration and/or point of addition having a higher efficiency and/or a higher ratio than the efficiency and/or ratio obtained by carrying out the method as reported herein in the absence of said compound or combination of compounds.

In one embodiment of the previously reported method, the compound with the highest efficiency and/or the highest ratio is identified/selected.

Detailed Description

Definition of

It must be noted that, as used herein and in the claims, the singular forms "a," "an," and "the" include plural referents unless the context clearly dictates otherwise. Thus, for example, reference to "a cell" includes a plurality of such cells and equivalents thereof known to those skilled in the art, and so forth. Likewise, the terms "a" (or "an"), "one or more" and "at least one" can be used interchangeably herein.

It should also be noted that the terms "comprising," including, "and" having "are used interchangeably and include the term" consisting of … ….

Procedures and methods for converting amino acid sequences (e.g., polypeptides) into the corresponding nucleic acid sequences encoding such amino acid sequences are well known to those skilled in the art. Thus, a nucleic acid can be characterized by its nucleic acid sequence consisting of individual nucleotides and likewise by the amino acid sequence of the polypeptide encoded thereby.

The term "about" refers to a range of +/-20% of the subsequent numerical value. In one embodiment, the term "about" refers to a range of +/-10% followed by a numerical value. In one embodiment, the term "about" refers to a range of +/-5% followed by a numerical value.

Pseudomonas exotoxin a (pe), Diphtheria Toxin (DT), Cholix Toxin (CT) and related toxins are bacterial proteins that undergo ADP-ribosylation of diphtheria amide residues of eukaryotic translation elongation factor a (eEF 2). ADP was transferred to diphtheria amide of eEF2 using NAD as co-substrate. This inactivates the function of eEF 2. Cells with ADP-ribosylated eEF2 stop their protein synthesis and thus die. PE, DT, CT and related toxins require the presence of diphtheria amide on eEF2 for ADP-ribosylation of eEF2 to occur. EEf2 free of diphtheria amide is unable to cause ADP-ribosylation by these toxins.

The terms "cell," "cell line," and "cell clone" are used interchangeably and refer to a cell into which an exogenous nucleic acid has been introduced, including the progeny of such a cell. "cell" includes "transformants" and "transformed cells," which include the primary transformed cell and progeny derived therefrom, regardless of the number of passages. Progeny may not be identical in nucleic acid content to the parent cell, but may instead contain mutations. Included herein are mutant progeny that have the same function or biological activity as the function or biological activity screened or selected in the originally transformed cell.

An "isolated" nucleic acid refers to a nucleic acid molecule that has been separated from components of its natural environment. An isolated nucleic acid includes a nucleic acid molecule that is contained in a cell that normally contains the nucleic acid molecule, but that is present extrachromosomally or at a chromosomal location that is different from its natural chromosomal location.

A "plasmid" is a nucleic acid that provides all of the elements necessary for the expression of the structural genes it contains in a host cell. In general, expression plasmids include a prokaryotic plasmid amplification unit comprising an origin of replication, such as that used in E.coli, and a selection marker, a eukaryotic selection marker, and one or more expression cassettes for expressing a structural gene of interest, each comprising a promoter, a structural gene, and a transcription terminator comprising a polyadenylation signal. Gene expression is typically placed under the control of a promoter, and such a structural gene is said to be "operably linked" to the promoter. Similarly, a regulatory element is operably linked to a core promoter if the regulatory element modulates the activity of the core promoter. The term "plasmid" includes, for example, shuttle and expression plasmids as well as transfection plasmids.

A "selectable marker" is a nucleic acid that allows for the specific selection or counter-selection of cells carrying the selectable marker in the presence of a corresponding selection agent. Generally, a selectable marker will confer drug resistance or complement a metabolic or catabolic defect in the host cell. The selectable marker may be positive, negative, or bifunctional. Useful positive selection markers are antibiotic resistance genes. Such a selectable marker allows for the positive selection of host cells transformed therewith in the presence of a corresponding selection agent (e.g., an antibiotic). Untransformed cells are not able to grow or survive in culture under selective conditions, i.e., in the presence of a selective agent. Positive selection markers allow selection of cells carrying the marker, while negative selection markers allow selective elimination of cells carrying the marker. Selectable markers for use with eukaryotic cells include, for example: aminoglycoside Phosphotransferase (APH) genes, e.g., hygromycin (hyg) selectable marker, neomycin (neo) selectable marker, and G418 selectable marker; genes of dihydrofolate reductase (DHFR), thymidine kinase (tk), Glutamine Synthetase (GS), asparagine synthetase, tryptophan synthetase (indole as a selective agent), histidinol dehydrogenase (histidinol D as a selective agent); and nucleic acids conferring resistance to puromycin, bleomycin, phleomycin, chloramphenicol, Zeocin, and mycophenolic acid. Further marker genes are described, for example, in WO 92/08796 and WO 94/28143.

Detailed Description

A prerequisite for optimized gene editing is reliable and robust detection and differentiation of heterozygous and homozygous gene inactivation as well as non-specific and targeted integration events. Existing methods, such as determining a phenotype (e.g., resistance) or lack of a phenotype (gene inactivation) or sequencing protocols, due to an insertion often do not distinguish between homozygous and heterozygous inactivation. In addition, the prior art has little to resolve the genetic composition of individual independent cells, or is not based on a large number of independent gene editing cells to allow robust statistical analysis to be performed.

Off-target effects caused by site-specific nucleases can be toxic to cells and difficult to predict and monitor comprehensively. Complex genomes often contain multiple sequences that are identical or highly homologous to the intended DNA target, resulting in off-target activity and cytotoxicity (Gaj, t. et al, Trends biotechnol.31(2013) 397-.

Herein is reported a simple and robust method for characterizing gene editing events. The combination of DPH gene inactivation, toxin treatment and (antibiotic) selectable marker selection allows the determination of gene editing potency on a very large number of independent cells. By adopting the method, homozygous gene inactivation and heterozygous gene inactivation can be distinguished, and site-specific integration and non-site-specific integration can be distinguished. The simplicity and robustness of the methods as reported herein can be used, for example, to optimize gene editing programs and modules and to identify and compare gene editing regulators.

The efficacy and specificity of, for example, CRISPR/Cas 9-mediated or Zinc Finger Nuclease (ZFN) -mediated homozygous and heterozygous gene inactivation and cassette integration events have been quantified by a combination of toxin (e.g., Diphtheria Toxin (DT)) and (antibiotic) selection marker (e.g., Puromycin (PM)) selection and High Resolution Melting (HRM) PCR using the methods as reported herein.

The overall gene inactivation frequency can be detected by HRM-PCR: inactivation of the homozygous DPH1 or DPH2 genes results in toxin resistance (e.g., diphtheria toxin resistance (DTr)), so homozygous and heterozygous DPH modifications can be distinguished by toxin sensitivity. Selectable marker resistance (e.g., puromycin resistance (PMr)) results from the integration of the expression cassette.

Target gene specific colony counts can be 10 per experiment⁴-10⁵Homozygous or heterozygous inactivation and integration events are distinguished in individual cells, and hundreds of independent clones carrying inactivated genes are evaluated in one experiment.

Homozygous inactivation (DTr) has been found to occur at a frequency of about 30-50 times higher than the frequency of targeted cassette integration (DTr & PMr) or non-targeted integration (PMr using scrambled rna (scrna)).

It has been found that inactivation of a hybrid gene without integration occurs more than 100 times more frequently than the frequency of integration.

Preference for gene inactivation over integration is independent of target sequence, gene or chromosomal location.

It has been found that colony counting can resolve a number of variables, including guide rna (grna) length, selection of enzymes for gene editing, or regulators of non-homologous end joining (NHEJ) or Homologous Recombination (HR).

Using the methods as reported herein, it has been found that 20mer grnas are most effective for inactivation, while 16-18mer grnas provide the highest integration potency.

With the approach as reported herein, it has been found that unmodified CRISPR/Cas9 is twice as efficient as ZFN-editing and 5 times more efficient than engineered high fidelity CRISPR/Cas9 in terms of gene inactivation and integration. It has been found that the ratio between inactivation events, non-targeted integration events and targeted integration events is similar for different enzymes.

The methods reported herein are also suitable to resolve the effect of modulation of NHEJ or HR on editing efficiency and specificity.

Therefore, a method for evaluating the above aspects is beneficial.

Diphtheria amide modification

Stahl, s. et al (proc. natl. acad. sci. usa 112(2015)10732-10737) reported that eukaryotic translational elongation factor 2(eEF2) is a highly conserved protein and essential for protein biosynthesis. The diphtheria amide modification at His715 of human eEF2 (or at a corresponding position in other species) is conserved in all eukaryotes and in the archaeal counterpart. It is produced by proteins encoded by seven genes. Proteins encoded by diphtheria amide biosynthetic protein 1(DPH1), DPH2, DPH3 and DPH4(DNAJC24) have 3-amino-3-carboxypropyl (ACP) groups conjugated to eEF 2. This intermediate is converted to diphtheria by the methyl-transferase DPH5, which is subsequently amidated with DPH6 and DPH7 to diphtheria amide.

Diphtheria amide modified eEF2 is the target for ADP ribosylating toxins, including pseudomonas exotoxin a (pe) and Diphtheria Toxin (DT). These bacterial proteins enter the cell and catalyze ADP ribosylation of diphtheria amide using Nicotinamide Adenine Dinucleotide (NAD) as a substrate. This inactivated eEF2 arrests protein synthesis and kills cells.

In Stahl et al, gene-specific Zinc Finger Nucleases (ZFNs) were used to generate DPH gene-inactivated MCF7 cells.

Thus, the plasmid encoding the ZFN was transfected into MCF7 cells. Forty-eight hours after transfection, mutated cells were identified either by phenotypic selection or by genetic analysis in order to be able to achieve ZFN binding, double strand breaks and fault repair. For phenotypic selection, cells were exposed to a lethal dose of PE (100nM) to kill all cells whose eEF2 is a substrate for the toxin. After another 48 hours, dead cells were removed and the cultures were propagated in medium containing toxins. This procedure produced colonies of cells transfected with ZFNs against DPH1, DPH2, DPH4 and DPH 5. Under toxin selection, no colonies were obtained in cells transfected with ZFNs targeting DPH3, DPH6, and DPH7, or transfected with mock (mock).

To identify MCF7 mutants under avirulence selection, individual single cells from each transfection were subjected to High Resolution Melting (HRM) analysis of genes. This technique allows the identification of cells containing two different alleles of the gene to be analyzed, since these cells produce a biphasic or odd-shaped melting curve. Analysis of candidate clones with biphasic HRM curves confirmed the presence of different DPH allele sequences. This method gives clones with one gene copy inactivated and another wild type copy functional for all DPH genes.

In wild-type cells, only the diphtheria amide modified eEF2 could be detected with no evidence of either unmodified eEF2 or modified diphtheria element or ACP. Cells with complete inactivation of DPH1, DPH2, DPH4, and DPH5 genes contained no diphtheria amide modified eEF 2. Thus, these genes are essential for diphtheria amide synthesis and other genes cannot compensate for their inactivation. Complete inactivation of DPH1, DPH2, or DPH4 resulted in cells that were detectable only in unmodified eEF2 and not in other modified forms. Complete inactivation of DPH5 yielded ACP intermediates (antibodies used in previous western blots did not recognize eEF2 with such intermediates). In cells with one inactive copy and one functional copy of DPH1-7, the predominant species of eEF2 is diptheria amide-modified eEF 2.

The EEF2 of the parent MCF7 and all seven heterozygote-inactivated MCF7 derivatives (DPH1-7) can cause ADP ribosylation by PE. In contrast, eEF2 from cells with completely inactivated DPH1 or DPH2 or DPH4 or DPH5 genes did not accept ADP ribosylation. eEF2 with only diphtheria amide is the substrate for ADP ribosylating toxins, whereas eEF2 without modification (DPH1, DPH2, DPH4) or with partial modification (ACP in DPH5) is not.

Under normal growth conditions, no effect of DPH inactivation on the growth of all heterozygous clones was observed. Furthermore, complete inactivation of DPH1, DPH2, or DPH4 did not result in significant reduction in cell growth or viability. Cells with complete inactivation of DPH1, DPH2, and DPH4 carried only unmodified eEF 2. Thus, the presence of only unmodified eEF2 by itself did not inhibit MCF7 growth. A decrease in the growth rate of all clones with complete inactivation of DPH5 was observed.

Cells with complete inactivation of DPH1, DPH2, DPH4, or DPH5 showed increased sensitivity to TNF. The NF-. kappa.B signal transduction pathway and the death receptor signal transduction pathway (known to be involved in and essential for development) are pre-activated in diphtheria amide deficient cells. However, these cells are viable because the induction of the pathway does not cross the threshold sufficient to induce apoptosis without additional stimulation. When these pre-induced pathways are triggered, pre-sensitization becomes phenotypically significant: all diphtheria amide synthesis deficient cells (not associated with target gene knockdown) were hypersensitive to TNF-induced apoptosis. The results of this study indicate that the presence or absence of diphtheria amide affects the NF-. kappa.B pathway or the death receptor pathway.

The DPH genes have nucleotide sequences described in NM _001383.3(DPH1), NC _000001.11(DPH2), NC _000003.12(DPH3), NM _181706.4(DPH4), BC053857.1(DPH5), NM _080650.3(DPH6), and NC _000009.12(DPH 7). Mutant clones were obtained by isolation of survivor clones after exposure to lethal doses of pseudomonas exotoxin a or by PCR-based HRM analysis.

Thus, in one embodiment of the methods herein for toxin selection, cells are treated with 100nM PE 48 hours after transfection and further propagated to generate toxin-resistant colonies, these colonies are isolated and re-cloned from single cells, and for genetic screening gene-specific PCR fragments are generated and HRM is performed (clones containing mutations are indicated according to a biphasic melting curve).

In one embodiment, growth of the parent and mutant MCF-7 is assessed by: 10,000 cells were seeded in flat bottom 96-well plates and humidified 5% CO at 37 deg.C₂(iii) incubation, cells are exposed to toxin twenty-four hours post inoculation, cell growth is measured, and a cell proliferation assay (e.g., CellTiterGlo 96Aqueous OneSolution cell proliferation assay, Promega, according to the manufacturer's instructions, wherein amplification (DNA replication) is resolved 72 hours post toxin exposure by the BrdU incorporation assay (roche diagnostics, mannheim FRG).

Definition (taken from Gaj, T. et al, Trends Biotechnol.31(2013) 397-405):

CRISPR/Cas (CRISPR-associated) system: clustered regularly spaced short palindromic repeats are loci containing multiple short direct repeats and provide acquired immunity to bacteria and archaea. CRISPR systems rely on crRNA and tracrRNA for sequence-specific silencing of invading foreign DNA. There are three types of CRISPR/Cas systems: in type II systems, Cas9 acts as an RNA-guided DNA endonuclease that cleaves DNA upon recognition of a crRNA-tracrRNA target.

crRNA: CRISPR RNA base-pair with the tracrRNA to form a 2-RNA structure that directs the Cas9 endonuclease to a complementary DNA site for cleavage.

PAM: the prodomain sequence adjacent motif is a short nucleotide motif present on crRNA that is specifically recognized by Cas9 and is required for it to cleave DNA.

tracrRNA: transactivating chimeric RNAs are non-coding RNAs that facilitate crRNA processing, and are required for activating RNA-guided cleavage of Cas 9.

And (3) DSB: the products of the action of ZFNs, TALENs and CRISPR/Cas9, double strand breaks, are forms of DNA damage that occur when both DNA strands are cleaved.

HR: homology-directed repair is a template-dependent pathway for DSB repair. HDR faithfully inserts donor molecules at targeted loci by supplying donor templates containing homology along with site-specific nucleases. This method enables the insertion of single or multiple transgenes, as well as single nucleotide substitutions.

NHEJ: non-homologous end joining is a DSB repair pathway that joins or joins two cleaved ends together. NHEJ does not utilize homologous templates for repair and therefore generally results in the introduction of small insertions and deletions at the site of the fracture, often resulting in a frame shift in the function of the knockout gene.

TALEN: the transcription activator-like effector nuclease is a fusion of the FokI cleavage domain and a DNA binding domain derived from a TALE protein. TALEs contain multiple 33-35 amino acid repeat domains that each recognize a single base pair. Like ZFNs, TALENs induce directed DSBs that activate DNA damage response pathways and make custom alterations possible.

ZFN: zinc finger nucleases are fusions of a non-specific DNA cleavage domain from fokl restriction endonuclease with a zinc finger protein. The ZFN dimers induce targeted DNA DSBs that stimulate DNA damage response pathways. The binding specificity of the designed zinc finger domain directs the ZFN to a specific genomic site.

ZFNick enzyme: zinc finger nickases are ZFNs that contain inactivating mutations in one of the two fokl cleavage domains. The ZFNic enzyme generates only single-stranded DNA breaks and induces HDR without activating the mutagenic NHEJ pathway.

Gene editing method

During the last decade, protocols have evolved that can manipulate virtually any gene in a variety of cell types and organisms. This core technology, often referred to as 'genome editing', is based on the use of engineered nucleases consisting of a sequence-specific DNA-binding domain fused to a non-specific DNA-cleavage module. These chimeric nucleases enable efficient and precise genetic modification by inducing targeted DNA Double Strand Breaks (DSBs) that stimulate cellular DNA repair mechanisms including error-prone non-homologous end joining (NHEJ) and homology-directed repair (HDR). The versatility of this approach is facilitated by the programmability of the DNA binding domain.

The versatility of these approaches stems from the ability to customize the DNA binding domain to recognize virtually any sequence.

Thus, the ability to perform genetic alterations depends mainly on the DNA binding specificity and affinity of the designed proteins (Gaj, T. et al, Trends Biotechnol.31(2013) 397-405).

Two types of gene-specific manipulations can be envisaged: non-specific, non-targeted mutagenesis and targeted gene modification or gene replacement.

Unspecified mutagens were targeted to one gene. The result of site-directed mutagenesis is localized sequence changes.

Targeted gene replacement produces localized sequence changes by homologous recombination between the original gene copy and the exogenous gene copy. In targeted gene replacement, the goal is to replace an existing sequence with a designed sequence in the laboratory. The designed sequence allows for the introduction of more subtle and more extensive changes. Generation of targeted genetic alterations is often referred to as "gene targeting" (Carroll, D., Genetics,188(20111) 773-782).

Zinc Finger Nucleases (ZFNs) and transcription activator-like effector nucleases (TALENs) and CRISPR/Cas represent a powerful tool for redefining the boundaries of biological studies. These chimeric nucleases consist of a programmable, sequence-specific DNA binding module linked to a non-specific DNA cleavage domain. ZFNs and TALENs can achieve a wide range of types of genetic modification by inducing DNA double-strand breaks that stimulate error-prone non-homologous end joining or homology-directed repair at specific genomic locations. RNA-guided DNA endonucleases based on Clustered Regularly Interspaced Short Palindromic Repeats (CRISPR)/Cas rely on crRNA and tracrRNA for sequence-specific modification of DNA. There are three types of CRISPR/Cas systems: in type II systems, Cas9 acts as an RNA-guided DNA endonuclease that cleaves DNA upon recognition of a crRNA-tracrRNA target.

By co-delivering a site-specific nuclease with a donor plasmid carrying a locus-specific homology arm, a single or multiple transgenes can be efficiently integrated into an endogenous locus. In addition to their role in promoting HR, site-specific nucleases also allow rapid generation of cell lines and organisms with null phenotype; NHEJ-mediated nuclease-induced repair of DSBs leads to the introduction of small insertions or deletions at the targeted site, resulting in the knock-out of gene function by frameshift mutations. Site-specific nucleases can also induce deletion of large chromosomal segments. This approach has been shown to support large chromosomal inversions and translocations. Finally, large transgenes (up to 14kb) have been introduced into a variety of endogenous loci by NHEJ-mediated ligation through synchronized nuclease-mediated cleavage of donor DNA to chromosomal targets (Gaj, T. et al, trends Biotechnol.31(2013) 397-405).

NHEJ-mediated nuclease-induced repair of DSBs leads to the efficient introduction of variable length insertion/deletion (indel) mutations starting at the site of the break. Thus, NHEJ-mediated repair of DSBs introduced into gene coding sequences often results in frameshift mutations that can lead to gene functional knockouts.

If a double stranded DNA "donor template" is supplied, nuclease-induced DSB HR can be used to introduce precise nucleotide substitutions or insertions of up to 7.6kb at or near the site of the break. Recent work has also shown that oligonucleotides can be used in conjunction with ZFNs to introduce precise changes, small insertions, and large deletions. ZFNs have been used to introduce NHEJ-mediated or HR-mediated gene changes (Joung, j.k. and Sander, j.d., nat. rev.mol.cell biol.14(2013) 49-55).

Generally, the nuclease-encoded gene is delivered into the cell via plasmid DNA, viral vectors, or in vitro transcribed mRNA. Transfection of plasmid DNA or mRNA by electroporation or cationic lipid-based agents may be toxic and restricted to certain cell types. Viral vectors are also limited in that they are complex, difficult to produce, potentially immunogenic, and involve additional regulatory hurdles. Integrase-deficient lentiviral vectors (IDLV) are an attractive alternative to deliver ZFNs to cell types resistant to transfection. AAV is a promising ZFN delivery vector that has been used to enhance the efficiency of ZFN-mediated HR and to drive ZFN-mediated gene modification in vivo. Although adenoviral vectors can accommodate full-length TALEN genes and deliver into human cells, lentiviral plasmid vectors carrying TALEN sequences tend to rearrange upon transduction (Gaj, t. et al, trends biotechnol.31(2013) 397-.

Zinc Finger Nuclease (ZFN)

Zinc finger nucleases combining the non-specific cleavage domain (N) of FokI endonucleases with Zinc Finger Proteins (ZFPs) provide a universal route to deliver site-specific Double Strand Breaks (DSBs) to the genome.

The modular structure of the Zinc Finger (ZF) motifs and the modular recognition of the ZF domains make it possible to make diverse DNA recognition motifs for the design of artificial DNA binding proteins each ZF motif consists of approximately 30 amino acids and folds into ββ a structures stabilized by the conserved Cys2His2 residue chelating zinc ions, the ZF motif binds to DNA by inserting a α -helix into the major groove of the DNA duplex, each finger binds primarily to triplets in the DNA substrate, relative to the start of the α -helix of each ZF motif, the key zinc finger residues at positions-1, +2, +3, +4, +5 and +6 contribute most of the sequence-specific interactions with the DNA sites.

FokI restriction enzyme, bacterial type IIS restriction endonuclease, recognizes the non-palindromic pentadeoxyribonucleotides 5'-GGATG-3':5'-CATCC-3' in duplex DNA and cleaves 9/13nt downstream of the recognition site. Durai et al suggested that the FokI recognition domain could be replaced with other native DNA binding proteins or other designed DNA binding motifs that recognize longer DNA sequences to generate chimeric nucleases (Durai, S. et al, nucleic. acids Res.33(2005) 5978-5990).

FokI nucleases function as dimers and therefore two zinc finger arrays must be designed for each target site. Early ZFNs used wild-type homodimeric FokI domains, which can form unfavorable dimers of the same monomeric ZFN. The use of strictly heterodimeric FokI domains can reduce the formation of unfavorable homodimeric species and thus have improved specificity (Joung, j.k. and Sander, j.d., nat. rev. mol. cell biol.14(2013) 49-55). Thus, the ZFN target site consists of two zinc finger binding sites separated by a 5-7 bp spacer recognized by the FokI cleavage domain (Gaj, T. et al, Trends Biotechnol.31(2013) 397-405).

Plasmids for use in the single-hybrid genetic selection System on pDB series plasmids, reporter genes, Chloramphenicol Acetyltransferase (CAT) or GFP, located downstream of a lac-derived weak promoter (Pwk). the 9bp target site bound by the ZF is located at a specific distance from the start of transcription.on pA series plasmids, the gene of the ZF is fused to a fragment of the RNA polymerase α -subunit (rpoA [ 1-248 ]) by means of a sequence encoding an amino acid linker.rpoA [ 1-248 ] -ZF fusion binds to the 9bp site in the reporter plasmid and can recruit further RNA polymerase subunits to stimulate transcription of the reporter gene (Durai, S.et al, nucleic acids Res.33(2005) 5978-.

Transcription activator-like effector nucleases (TALEN)

A fusion TALEN of a plant pathogenic Xanthomonas (Xanthomonas) species transcription activator-like (TAL) effector to fokl nuclease, can bind in pairs and cleave DNA. Binding specificity is determined by a customizable array of polymorphic amino acid repeats in TAL effectors.

TAL effectors enter the nucleus, bind to effector-specific sequences in the host gene promoter and activate transcription. Their targeting specificity is determined by the central domain of the 33-35 amino acid repeats in tandem followed by a single truncated repeat of 20 amino acids. The naturally occurring recognition site is uniformly preceded by a T necessary for TAL effector activity (Cermak, T. et al, nucleic. acids res.39(2011) e 82).

TALE specificity is determined by two hypervariable amino acids called repeat-variable diresidues (RVDs). Similar to zinc fingers, modular TALE repeats join together to recognize adjacent DNA sequences (Gaj, t. et al, Trends biotechnol.31(2013) 397-.

TAL effectors can be fused to the catalytic domain of FokI nucleases to generate targeted DNA Double Strand Breaks (DSBs) for genome editing in vivo. Since fokl effects cleavage as a dimer, these TAL effector nucleases (TALENs) act in pairs, binding opposing targets across a spacer over which the fokl domains converge to cause cleavage. In almost all cells, DSBs can be repaired by one of two highly conserved processes: often resulting in small insertions or deletions and can be used for non-homologous end joining (NHEJ) for gene disruption, and Homologous Recombination (HR) for gene insertion or replacement.

Assembly of a TALEN or TAL effector construct comprises two steps: (i) assembling the repeat sequence modules into an intermediate array of 1-10 repeats and (ii) ligating the intermediate array into a backbone to produce the final construct (Cerak, T. et al, Nucl. acids Res.39(2011) e 82).

The TALEN target site consists of two TALE binding sites separated by a spacer sequence of variable length (12-20 bp) (Gaj, T. et al, Trends Biotechnol.31(2013) 397-405).

For common heterodimeric target sites (i.e., sites as would normally occur in the native DNA sequence), paired TALEN constructs are transformed together into the target cell.

One of the paired TALENs targeting the human gene of interest is subcloned into a mammalian expression vector using a suitable restriction endonuclease. These enzymes cleave the entire TALEN pair and place the coding sequence under the control of the promoter. The resulting plasmid was introduced into HEK293T cells by transfection using lipofectamine 2000(Invitrogen) following the manufacturer's protocol. Cells were harvested 72 hours after transfection (Cerak, T. et al, Nucl. acids Res.39(2011) e 82).

Clustered regularly interspaced short palindromic repeats/CRISPR-associated protein 9(CRISPR/Cas9)

The naturally occurring CRISPR/Cas type II system has evolved into a powerful gene editing tool for eukaryotic cells. In particular, the incorporation of crRNA and tracrRNA into a single guide rna (sgrna) was demonstrated, and a way was laid for this development. Cas9 creates a single double-strand break in DNA, an important feature of gene editing tools. The method provides two ways to generate genetic variation using DNA repair pathways in eukaryotic cells. The first approach relies on non-homologous end joining (NHEJ), which joins the ends of the nicks, but proceeds in a manner that often deletes a few bases, which may compromise the gene product, or cause a frameshift that renders it inactive. In a second approach, the damaged allele is repaired by homology-directed repair (HDR) using another piece of DNA with homology to the target. By providing a DNA element that can be inserted by recombination, any type of insertion, deletion or change in sequence can be achieved. The main limitation is the need for a PAM adjacent to the target. The off-target effect of Cas9 interaction with unintended targets is a problem that requires strategies to predict and prevent (Rath, d. et al, biochim.117(2015) 119-128).

In type II CRISPR/Cas systems, a short segment of foreign DNA (referred to as the "spacer") is integrated inside the CRISPR genomic locus and is transcribed and processed into short CRISPR RNA (crRNA). These crrnas anneal to trans-activating crRNA (tracrrna) and direct sequence-specific cleavage and silencing of pathogenic DNA by Cas proteins. Recent studies have shown that target recognition of Cas9 protein requires a "seed" sequence inside the crRNA and a Protospacer Adjacent Motif (PAM) sequence containing a conserved dinucleotide upstream of the crRNA binding region. It has been shown that CRISPR/Cas systems can be conveniently transported into human cells by co-delivering a Cas9 endonuclease expressing plasmid and an essential crRNA component (Gaj, t. et al, trends biotechnol.31(2013) 397-.

High resolution melting method for determining homozygote and heterozygote small amplicon

A high resolution melting method of small amplicons for determining homozygotes and heterozygotes is reported by Liew, M.et al (Clinical Chemistry 50(2004) 1156-. Heteroduplexes are easy to identify because they alter the shape of the melting curve. Adding 15% of known homozygous genotypes to an unknown sample can allow melting curve separation for all three genotypes.

Melting analysis of short PCR products in the presence of a heteroduplex detection dye (LCGreen I) was used to genotype SNPs.

Although heterozygotes can be identified by the presence of a second, low temperature melting transition using standard techniques, genotyping is much easier using high resolution methods.

PCR was performed in a LightCycler using reagents commonly used in clinical laboratories. 10 μ L of the reaction mixture was composed of 10-50ng of genomic DNA, 3mM MgCl₂1XLightCycler FastStart DNA Master hybridization Probe Master mix, 1 XLCGreen I, 0.5. mu.M forward and reverse primers, and 0.01U/. mu.L E.coli (Escherichia coli) uracil N-glycosylase (UNG; Roche). PCR was initiated with a 10 minute hold at 50 ℃ to control contamination by UNG and a 10 minute hold at 95 ℃ to activate the polymerase. A rapid thermal cycle was performed between 85 ℃ and the renaturation temperature at a programmed transition rate of 20 ℃/s.

Immediately after cycling melting analysis was performed on a LightCycler. Twenty samples were analyzed immediately by: first heated to 94 ℃, cooled to 40 ℃, heated again to 65 ℃ (both at 20 ℃/s) and then melted at 0.05 ℃/s with continuous fluorescence acquisition until 85 ℃. LightCycler software was used to calculate the derivative melting curve.

When high resolution melting is used, the amplified sample is heated to 94 ℃ in a LightCycler and rapidly cooled to 40 ℃. The LightCycler capillary was then transferred to the HR-1 high resolution instrument and heated at 0.3 ℃/s. Samples were analyzed between 65 ℃ and 85 ℃ with a turnaround time of 1-2 minutes. The high resolution melting data was analyzed using HR-1 software. In most cases, the fluorescence versus temperature curve is normalized. For direct comparison with LightCycler data, non-normalized derivative curves were used. After data output, all curves were plotted using Microsoft Excel.

Methods reported herein

Gene editing techniques are required to produce tools, cells (cell lines) or organisms for genetic modification. In order to be suitable for routine use, the technique must have high specificity and high efficiency, while at the same time, low levels of non-targeted genetic modification (i.e., side reactions).

Herein is reported a high throughput method that can be used to determine and thereby compare/assess gene editing specificity, gene editing efficiency and/or non-targeted gene editing levels of a gene editing method/gene editing module.

Herein is reported a robust high-throughput method that can be used for quantifying gene editing method/gene editing module (e.g., CRISPR/Cas) mediated gene changes. The method is suitable for distinguishing between the efficiency of a gene editing method/gene editing module, its site-specific and non- (site) specific gene disruption and sequence integration events (e.g., quantitative CRISR/Cas-mediated gene inactivation and integration events).

In one embodiment of the method as reported herein, the gene editing method/module/technology is selected from CRISPR/Cas, zinc finger nuclease and talon nuclease.

In one embodiment of the method as reported herein, the toxin is an enzymatically active toxin or a fragment thereof, including but not limited to diphtheria toxin a chain, non-binding active fragments of diphtheria toxin, exotoxin a chain (from Pseudomonas aeruginosa), ricin a chain, abrin a chain, goldenafil a chain, α -sarcin, Aleurites fordii (Aleurites fordii) protein, carnation protein, phytolacca americana (phytopaca americana) protein (PAPI, PAPII and PAP-5), momordica charantia (momordia charantia) arrestin, curcin, crotin, saponaria officinalis (sapaonaria officinalis) arrestin, gelonin, mitogellin (mitogellin), restrictocin, phenomycin, enomycin and trichothecenes.

Exemplary embodiments

The methods as reported herein are exemplified below using exemplary sequences and targets. This should not be construed as limiting the invention. Evidence is provided herein as only one of many alternatives (i.e., specific embodiments) and as to the usefulness of the methods reported herein.

In one embodiment of the method as reported herein, a robust and simple method of quantifying and optimizing gene editing protocols as reported herein, utilizes a combination of DPH gene inactivation (in preferred embodiments DPH1 or DPH2 gene inactivation) with diphtheria toxin and/or puromycin selection. With this method it is possible to determine not only gene editing efficiency for a large number of independent cells (i.e. at high throughput), but also to distinguish between homozygous and heterozygous genes and to distinguish between site-specific and non- (site-) specific gene disruption and integration events.

CRISPR/Cas mediated events of homozygous and heterozygous DPH gene inactivation and cassette integration were quantified by a combination of Diphtheria Toxin (DT) and Puromycin (PM) selection and High Resolution Melting (HRM) PCR. Inactivation of the DPH gene can be detected by high resolution melting HRM-PCR, and inactivation of the homozygous DPH gene results in toxin resistance. Thus, toxin sensitivity distinguishes between homozygous and heterozygous target gene modifications. PM resistance detects the integration of CRIPR/Cas cassettes. Comparing the frequency of DT-resistant, PM-resistant or double-resistant clones/colonies, it has been observed that the frequency of homozygous gene inactivation (DT-resistance) occurs about 30-50 times higher than targeted cassette integration (DT-resistance as well as PM-resistance) or non-targeted integration events (PM-resistance). The frequency of inactivation of the heterozygous target gene without cassette integration (detectable by HRM-PCR) was even higher (>100 fold) than the frequency of cassette integration. The bias and higher efficiency of gene inactivation relative to cassette integration is independent of CRISPR/Cas9 target sequence, chromosomal location, or target gene. Robust readout results (number of DT-resistant or PM-resistant colonies) of the methods as reported herein can be used to evaluate the effect of CRISPR/Cas9 variables (e.g., sequence length of guide RNA) on gene modification efficiency and modification type. The results presented herein also suggest that CRISPR/Cas 9-derived therapies may be more suitable for gene inactivation protocols rather than applications requiring targeted integration events.

Exemplary results

Diphtheria toxin resistance assay and HRM-PCR quantification and differentiation of homozygous and heterozygous DPH1 gene inactivation

Diphtheria Toxin (DT) performs ADP-ribosylation of the white larynx amide and thus inactivates eukaryotic translation elongation factor 2(eEF 2). This irreversibly stops protein synthesis and kills the cell (Weidle et al). Diphtheria amide is a histidine modification at eEF2 by the enzyme encoded by the diphtheria amide synthesis gene, including DPH 1. Complete inactivation of DPH1 in MCF7 cells can prevent the synthesis of the toxin target diphtheria amide. This renders the cells resistant to DT (Stahl et al). As a result, inactivation of all copies of DPH1 will result in a 'DT resistance' phenotype (DTr). The frequency of this phenotype can be detected in a robust manner by counting toxin-resistant colonies (see figure 3).

Since the presence of one remaining functional DPH1 allele is sufficient for toxin sensitivity, DTr cells carry a functional knock-out of the entire DPH1 allele. Cells with only one allelic modification can be identified by on-cell HRM-PCR assay (fig. 4). Modification of the target site altered the melting temperature of the DPH1-PCR fragment compared to the wild-type fragment, thereby generating a biphasic HRM curve. Since CRISPR-Cas 9-mediated gene inactivation is an independent event and therefore rarely identical on both alleles, most cells with complete gene inactivation will also show a biphasic HRM curve. These cells can be distinguished from single allele changes based on their toxin resistance phenotype.

Thus, in one embodiment, heterozygous and homozygous inactivation events are quantified and distinguished by a combination of (high throughput) HRM-PCR and DTr colony counting assays.

Selectable marker resistance detection and differentiation between specific and non-specific integration events

puromycin-N-acetyl-transferase (PAC), encoded by the integration cassette of the CRISPR/Cas9 donor plasmid used in the current example, can inactivate Puromycin (PM) and render the modified cells resistant to PM (Vara et al). Thus, PAC integration was detected and quantified by the PM resistance assay in the same manner as described for DTr colonies (PM was applied instead of DT as a selection agent, fig. 3B). In contrast to DTr, which is caused only by inactivation of specific and homozygous target genes, PMr appeared independent of the integration position. In one embodiment, the frequency of site-specific integration versus non- (site) -specific integration is resolved by comparing double resistant colonies (DTr + PMr) and PMr colonies.

Comparison of DPH1 inactivation event and Targeted integration event by CRISPR/Cas9

To compare the frequency of target-specific inactivation, integration and non-target integration, a plasmid encoding the DPH 1-specific CRISPR/Cas9 module (see sequence figure 1) was transfected into MCF7 cells. These cells were then subjected to HRM-PCR and colony counting assays to detect DT resistance and PM resistance. The results of these assays are summarized in fig. 5, and the individual data sets are shown in the table below. Fig. 5A shows that complete inactivation of the DPH1 gene (indicating loss of function of all DPH1 alleles) occurred at a frequency of about 6% of all transfected cells (2.5% of all cells were considered as transfection efficiency of 40%, see first table below). DPH1 inactivation showed absolute dependence on the matching gRNA sequence: scrambled control RNA (scRNA) did not produce any DTr colonies. The frequency of HRM-identified clones is shown in FIG. 5B as compared to the appearance of DTr colonies. These analyses revealed that single allele inactivation (HRM-confirmed toxin sensitivity) occurred twice as frequently as double allele inactivation (DTr cells).

Table: colony count and phenotypic frequency of transfected cells

'TF eff.': transfection efficiency was determined by FACS analysis by monitoring the frequency of fluorescent cells after transfection of MCF7 with GFP-reporter plasmid. The relative number of GFP positive cells in% of all cells is listed. One of these assays demonstrated unusually high GFP positivity and unusual FACS patterns. The average transfection efficiency for all assays was 30% -40%. A 'HRM': high resolution melting point PCR positive cells are defined as cells that exhibit a well defined divergent (biphasic) melting curve pattern compared to wild type cells. O.' indicates the frequency of cells that do not carry a functional copy of the DPH1 or DPH2 gene and are therefore resistant to DT. 'int.' denotes the frequency of cells carrying the PAC expression cassette and thus resistant to PM. 'A-D' indicates each sample of independent experiments.

Table: effect of gRNA length on inactivation of the orientation Gene and cassette integration

'TF eff.': transfection efficiency was determined by FACS analysis by monitoring the frequency of fluorescent cells after transfection of MCF7 with GFP-reporter plasmid. The relative number of GFP positive cells in% of all cells is listed. Cells that do not carry a functional copy of the DPH1 gene are resistant to DT. Cells carrying the PAC expression cassette are therefore resistant to PM. 'A-D' indicates each sample of independent (quadruplicate) experiments.

FIG. 6 shows a comparison of DTr and PMr colony frequencies (two different experiments, experiment 1: A-C, experiment 2: D-F): DPH1 biallelic inactivation occurred at 30-50 fold more potent than the PAC expression cassette integration event mediating PM resistance (fig. 6C and 6E, the location or zygosity (zygosity) of PAC integration could not be determined). Scrnas produced 2-fold fewer PMr colonies compared to DPH 1-specific grnas under the same conditions. Thus, Cas9/DPH1-gRNA mediated integration occurs preferentially at the DPH1 gene, but without absolute specificity. Thus, this effect can be verified using the methods as reported herein. Based on preferential integration in the DPH1 gene, many PMr colonies obtained using DPH1 guide RNA were DT resistant (fig. 6A and 6D). Neither of the PMr colonies obtained with scRNA were resistant to DT. Thus, Cas 9-mediated gene inactivation (including gene inactivation on both alleles) occurred highly specific and with much higher frequency than directed PAC integration (fig. 6C and 6E). Thus, this effect can be verified using the methods as reported herein. Targeted integration not only occurs less frequently, but also has less specificity for the location defined by the gRNA.

Quantification of gene editing was applied in an equivalent manner to another target gene, DPH2

The same protocol as outlined above was also used for a different gene, the DPH2 gene. That is, Cas 9-induced DPH2 gene modification was performed. This modification was made to show the general applicability of the method as reported herein.

DPH2 encodes a different enzyme with a different sequence on a different chromosome, however, it is also essential for diphtheria amide synthesis. Lack of DPH2 renders cells resistant to DT in the same manner as lack of DPH1 (Stahl et al).

The results of DPH2 compilation, followed by DT resistance and PM resistance assessment results are shown in fig. 6F: consistent with our observations of DPH1, it was observed that the homozygous DPH2 inactivation event was more frequent than PAC cassette integration, with variations in the same order of magnitude (DPH2 inactivation was about 90-fold higher than PAC cassette overall). Inactivation strictly depends on the presence of cognate grnas, whereas cassette integration is site-biased, but not absolutely specific for the target gene (compare the frequency of DPH2 guidance with scrnas).

Thus, the analytical principles developed to characterize DPH1 modifications may also be applied to analyze DPH2 modifications.

By using two different genes, DPH1 and DPH2, the results provide evidence that the present analysis system can be used for other genes.

Comparison and optimization of Cas9 Gene manipulation modules-gRNA length

As shown above, the method as reported herein can resolve the overall impact of the composition of the genetic modification module.

Figure 8 shows how gene inactivation and the efficacy and specificity of integration of Cas9 grnas of different lengths can be assessed and compared. All grnas used target the same sequence segment within DPH1, but vary in length from 14 to 26 bases (fig. 8D, gRNA details in fig. 1). The number of DTr colonies was recorded to reflect the specific homozygous inactivation of the target gene. At the same time, PMr resistance numbers and DTr-PMr dual resistance numbers were evaluated to monitor cassette integration.

Using the methods as reported herein, it can be demonstrated that gRNA length affects the efficacy of gene inactivation.

20 mers have been shown to confer maximum deactivation efficacy of DPH 1. By shortening the complementary stretch to 18 or 16 bases or extending it to 26 bases, determined using the method as reported herein: significant specific gene inactivation function was retained, although the potency was below 20 mers.

Reducing grnas to less than 16 bases (14 mers) reduced DPH1 inactivation function below detectable levels. Integration efficacy (assessed by counting PMr events) was also affected by gRNA length. Guide RNAs smaller than 16mer (14mer) produced only a few PMr colonies, not exceeding the background level of the scrambled controls. Directed integration of 16mer, 18mer, 20mer, 22mer, 24mer and 26mer was observed, with the best insertion efficacy achieved with the 16-18 mer.

Thus using the methods as reported herein it was demonstrated that no efficacy gain was achieved using larger oligonucleotides, indeed, the larger size significantly reduced the specific insertion events.

Using the reported methods, the ratio between integration event (PMr) and inactivation event (DTr) can be calculated as a 'specific indicator' to identify conditions under which specific integration occurs with minimal gene inactivation events.

Such conditions may be advantageous if targeted integration is desired and not too much ineffective target gene damage is caused. Low numbers (e.g., PMr low relative to DTr colonies) reflect that integration is inefficient relative to concurrent inactivation events. High values (more PMr and/or relatively reduced number of DTr colonies) reflect more efficient integration at Cas9 targeted genes.

Fig. 8E shows the calculated specificity coefficients depending on the gRNA length. The highest insertion/inactivation values were found for the 16-18mer and a significant decrease was found for guide RNAs containing 20 or more bases using the methods as reported herein. This suggests that 20 mers are quite efficient for targeted gene inactivation (consistent with previously published observations, see, e.g., Cho et al, 2013; Bauer et al; Jinek et al, 2012 and 2013; Mali et al, 2013). However, shorter guide RNAs appear to be more efficient for integration. Shorter guide RNAs not only improved the integration efficiency (overall higher number of PMr colonies), but also improved the ratio between efficient (productive) and inefficient (non-productive) gene editing (fewer DTr colonies without insertions).

Effect and specificity of different Gene editing protocols-enzymes

Different variants of RNA-guided Cas9 and ZFN-mediated gene editing gene inactivation and integration events, potency and specificity were compared using the methods as reported herein.

Thus, gRNA length and composition remained constant (DPH 120 mer), with three different editing enzymes applied:

(i) 'SpCAS 9' refers specifically to Cas9 nuclease from Streptococcus pyogenes (Streptococcus pyogenes), which can be considered as the current standard application (see, e.g., Ran et al, 2013, Fu et al, 2014);

(ii) SpCas9-HF1 is an engineered variant of SpCas9 with reduced non-specific DNA binding, reduced off-target activity and thus higher fidelity and specificity (kleintiver et al);

(iii) ZFN-mediated editing modules that recognize target sequences through engineered zinc finger-mediated protein-nucleic acid interactions (Miller et al, 2011; Urnov et al).

In the same manner as the gRNA assay, DTr colonies were recorded to reflect targeted gene inactivation and PMr colonies were recorded to monitor cassette integration (fig. 8F; table below).

Table: colony count and phenotypic frequency of MCF-7 cells transfected with different Gene editing entities

MCF-7 cells were transfected with plasmids encoding different genome editing systems (SpCas9, SpCas9, ZFN). The SpCas9 construct was as described previously. According to Kleinstein et al (Kleinstein et al, 2016), SpCas9-HF contains N497A/R661A/Q695A/Q926A substitutions. At the same time, grnas were replaced with scrnas in parallel to resolve nonspecific activity. DPH1 specific ZFN from Sigma AldrichAnd (4) obtaining. Transfection 3x10⁶The total amount of plasmid DNA (editing entity and donor) of the initial cell pool of individual cells was as described in previous experiments. To quantify the transfection efficiency (TF eff. (%)), GFP reporter plasmids were additionally transfected. GFP positive cells were counted by FACS 24 hours after transfection. A defined number of cells (number of cells seeded) were seeded and treated with DT, PM or DT + PM 72 hours thereafter.

Comparing the total efficiency of gene inactivation and cassette integration as determined using the methods as reported herein, it was found that:

the highest values for gene inactivation and cassette integration were determined for CRISPR/SpCas 9.

Compared to CRISPR/SpCas9, CRISPR/SpCas9-HF attenuated the targeted gene inactivation event to less than 20% of the number of DTr colonies. PMr (integration) resistant colonies and DT-PM double resistant colonies (integration plus targeting gene inactivation) frequency were also reduced.

Under the same conditions, the ZFN module reduced the number of DTr colonies to less than 60% of the events observed with CRISPR/SpCas 9. The potency of ZFN directed inactivation is thus reduced by a factor of about 2 compared to SpCAS9 and about 2-3 times higher than engineered SpCAS9-HF 1. PMr colony frequency did not differ significantly between CRISPR/SpCas9 and ZFNs. The dual resistant colonies (cassette integration with simultaneous gene inactivation) of ZFNs were more or less (30%) reduced compared to CRISPR/SpCas 9.

The ratio of DTr (target gene inactivation) colonies/DT + PM double resistant (target site integration) colonies was calculated and the overall efficacy was taken from the formula: CRISPR/SpCas9, CRISPR/Cas9-HF, and ZFN produced the same level per homozygous gene inactivation event (about 4X 10)^-3) Directed integration events.

Thus, by using the methods as reported herein, it can be shown that the overall efficacy of the editing system is variable, but the 'specificity' of cassette integration above the background of target gene inactivation does not appear to be significantly different for all protocols.

Effect of DNA repair Modulator on Gene editing potency and specificity

The methods as reported herein (including colony assays to determine DTr and PMr cells at DPH gene editing) can also be used to resolve the effect of compounds that modulate DNA repair.

Activators of Homologous Recombination (HR) and inhibitors of non-homologous end joining (NHEJ) have been described to modulate gene editing events and increase integration efficiency (Song et al; Ma et al).

To demonstrate the use of the method as reported herein for determining the effect of a DNA repair modulator on editing potency and specificity, the CRISPR/SpCas9/DPH1gRNA (20mer) editing assay according to the method as reported herein was supplemented with such compounds and the effect of the compounds was quantified.

The DNA ligase IV inhibitor SCR7 was applied 4 hours before transfection of the gene editing module ('early addition') or 18 hours after transfection ('late addition'), and exposure continued until 72 hours after transfection. In the same manner, RAD51 modulator RS-1 (compound 1 stimulating RAD 51) was added to stimulate HR. Both compounds were applied at doses that did not affect the growth or viability of MCF7 cells: scr7 was 1. mu.M and RS-1 was 8. mu.M, and when the two were combined, 1. mu.M + 8. mu.M (SCR7+ RS 1). Addition of RS-1 was determined to increase PMr colony numbers by approximately 2-fold compared to untreated controls (no compound) without affecting colony numbers obtained with scrnas (see table below).

Table: phenotype of MCF-7 exposed to SCR7 and/or RS-1 during Gene editing

Cells transfected with plasmids for SpCas 9-mediated DPH1 editing were seeded in defined numbers (number of seeded cells). DT/PM selection was initiated 72 hours after transfection. Values (w, x, y, z) represent colonies obtained in quadruplicate independent experiments. The effect of the point in time of addition of the compounds (RS-1, SCR7 and RS-1+ SCR7) was. The significant differences of PM resistant versus DT resistant colonies for the treated versus no compound samples are indicated as p <0.05, p <0.01, p < 0.001.

To quantify the effect on overall integration efficacy, the percentage of PMr colonies (gene integration) relative to DTr colonies (gene inactivation) was determined using the method as reported herein (see also figure 10).

Based on the results of the method as reported herein, it has been found that addition of RS-1 at an early time point results in significantly higher integration efficiency, however when added late (18 hours after initiation of editing) integration efficiency is not affected. Therefore, selection of an appropriate (early) point in time for effective (productive) editing for RS 1-mediated HR stimulation is important. This confirms HR as a driver for orientation cassette integration.

Based on the results of the methods as reported herein, it has also been found that, to a similar extent, early application of SCR7 significantly increased the relative number of integrations (fig. 10 and table above). This confirms the previous observation of enhanced efficient gene editing when SCR7 was administered (see Ma et al).

Based on the results of the methods as reported herein, it has further been found that the ratio of PMr relative to DTr is 8.1% when the two compounds are combined compared to 6.5% (SCR7 only) or 6.9% (RS-1 only). However, this difference/amplification was not significant (p ═ 0.39vs. RS-1 alone), which was consistent with previous observations (Pinder et al; Gutscher et al).

Identification of conditions or Components Or Condition-component combinations that enable improved Gene editing Using high throughput techniques

The method as reported herein is based on an editing-based inactivation of cellular genes conferring sensitivity to lethal eEF2 inactivating toxins such as DT or pseudomonas exotoxin a or Cholix toxins. The readout of these assays is the 'colony count' of cells resistant to this ADP ribosylating toxin.

The death/survival readout is very robust and the cell-based colony counting protocol is high throughput compatible. Thus, the analysis principle can be adapted to highly parallel assay settings and to run on an automated (robotic) platform. Automated high throughput platforms for identifying phenotypes in cell-based assays are available in a variety of formats and are well known in the art.

The high throughput/automated application of the DPH editing-based analytical principles can measure the impact of a large number of different conditions or parameters or additives/regulators on gene editing, in particular on efficiency and specificity. The conditions or parameters or additives/modulators may be resolved in a collection or library (exemplary and not limiting) of synthetic or natural compounds. This allows identification of compounds and conditions that enhance gene editing efficiency or specificity. Such compounds and conditions may be derived from or resemble, in particular, entities and classes of compounds that directly or indirectly affect DNA recombination and repair, nucleic acid binding, protein-nucleic acid interaction, or protein interaction with proteins involved in nucleic acid rearrangement, metabolism, synthesis, or repair, in structure or function.

Looking at

Genome editing has emerged as the most important technology for scientific applications. However, its full potential is still limited by the efficacy and specificity problems of the editing schemes currently used. The method as reported herein enables a simple and robust quantification and comparison of the efficacy and specificity of gene inactivation and insertion by editing. It allows the exact determination of heterozygous and homozygous gene inactivation and non-specific and targeted integration events based on a large number of independent cells. The knowledge and parameters generated by this method can not only improve the science of gene editing, but can also be of particular importance in developing and optimizing gene editing.

The core principle of the method as reported herein includes editing-based inactivation of cellular genes conferring sensitivity to lethal eEF2 inactivating toxins such as DT. The toxin target is the diphtheria amide placed on eEF2 by a protein encoded by the DPH gene (e.g., DPH1 or DPH 2). Complete loss of function of either of these genes (both alleles) leads to 'absolute' toxin resistance as this can prevent diphtheria amide synthesis. Thus, toxin resistance specifically indicates homozygous functional inactivation of both target genes. Cells that inactivate only one target allele remain toxin sensitive. Integration events can be assessed separately by the genes conferring resistance on the integration cassette. The toxin resistance caused by inactivation of the DPH gene is 'absolute', without any background and independent of the amount of toxin applied. This is a key factor in the robustness and simplicity of the method: the frequency of homozygous target gene inactivation events can be assessed by counting toxin-resistant colonies, insertion events can be assessed by PM-resistant colonies, and dual events can be assessed by counting colonies resistant to both. Thus, the method enables simple and robust individual evaluation of each editing event on a large number of independent cells. In addition (and in contrast to many existing tools, see, e.g., Fu et al, 2014, Doench et al; Maruyama et al; Shalem et al), homozygous and heterozygous gene inactivation can be distinguished from integration events. Thus, a simple colony count reflects the efficacy and ratio between efficient gene editing (integration) and destructive gene editing (inactivation without integration).

This method can give "universal" results as evidenced by the results of comparing editing events (colony frequency) on two different DPH genes. DPH1 and DPH2 encode different enzymes, which are independently essential for white larynx amide synthesis. The results show comparable potency, specificity and disruption/integration ratio on both genes. Comparable reads of different target sequences on different genes of different chromosomes demonstrate that the rules, dependencies and parameters determined by applying this method can be used to optimize the editing of other genes.

The DPH gene inactivation is used as a robust method for representing, distinguishing and optimizing gene editing efficiency, and is not limited by a CRISPR/Cas module. It can be applied in the same way with the same readout parameters to other gene editing principles, such as zinc finger nucleases or TALENS (Carroll 2011; Christian et al, 2010; Gaj et al, 2013; Miller et al, 2011; Urnov et al, 2010). The similar overall gene inactivation and integration frequencies observed on different DPH target genes (located on different chromosomes) also indicate that the 'rules', dependencies and optimization parameters derived from this approach have general applicability (i.e. can be used for other genes). Thus, gene editing modules and parameters can be optimized by DPH gene modification, and thereafter transferred to optimize gene editing efficiency or specificity of other genes.

With the method as reported herein, the previous observations ((Cho et al, 2013; Jinek et al, 2012; Jinek et al, 2013; Mali et al, 2013 b): the 20mer guide RNAs were effective for CRISPR/Cas-mediated gene targeting the highest overall gene inactivation and integration frequency was observed with the 20mer, with many other approaches (Doench et al, 2014; fu et al, 2014; maruyama et al, 2015; shalem et al, 2014) contrary, in the method as reported herein, not only the overall frequency can be evaluated, but also homozygous and heterozygous gene inactivation (and non- (site) -specific integration) events can be distinguished for a large number of independent cells in a fast, simple and robust way (see fig. 9).

These analyses led to meaningful observations (although the highest overall inactivation frequency was achieved with the 20 mer): the ` best ` ratio between efficient and destructive editing was observed with 16-18mer guide RNAs. Thus, if the aim is to inactivate genes most efficiently, a 20mer can be selected that still can be considered the current gRNA 'standard' (Haeussler et al, 2016; Liao et al, 2015; Muller et al, 2016). While on the other hand, the results presented herein show that for an editing event, a 16-18mer may be preferred if integration without too much destructive editing is desired.

It is significant to note that Fu et al (Fu et al, 2014) have evaluated grnas shorter than 20mer in gene inactivation experiments and observed inactivation potency comparable to 20mer with reduced off-target effects. Their analysis is based on single-allele GFP gene inactivation (only one target gene per cell) and thus cannot resolve or distinguish between homozygous and heterozygous inactivation events in diploid cells. Fu et al have not been able to quantify and compare insertion events. The data (based on the large number of cells and on the inactivation of the human gene encoded by the normal chromosome) show that the 20mer induced gene inactivation more efficiently than the shorter guide. However, a shorter guide is a more efficient insertion of the guide.

Optimal integration efficiency with low null gene disruption may be particularly important for 'repeat protocols', i.e., when aiming to reapply CRISPR/Cas module to previously treated cells to increase integration efficiency. Because gene inactivation alters the guide RNA target sequence, only the unaltered gene can be modified by the original CRISPR/Cas component, while the modified gene (without integration) is not susceptible to repeated protocols.

The exact determination of heterozygous and homozygous gene inactivation as well as non-specific and targeted integration events is of particular importance for the development and optimization of CRISPR/Cas derived therapies. The results presented herein show that gene inactivation occurs at a much higher (>100 fold) frequency than targeted integration. Even homozygous inactivation (both alleles affected) occurs at a frequency 30-50 times higher than targeted integration. Gene inactivation shows 'absolute' dependence on guide RNA specificity, while integration shows preferential integration at the target site, but not to the extent that it can be termed 'specific' integration (a 2-fold increase in the frequency of targeted versus non-targeted gene integration). Thus, for applications requiring explicit integration events, currently available CRISPR/Cas-derived gene editing techniques must be specifically evaluated.

The cited documents are:

bauer, D.E. et al 2015, Journal of visual experiments JoVE (95).

Brooks, S.C. et al, 1973, The Journal of biological chemistry 248(17) 6251-3.

Carroll,D.,2011,Genetics 188(4):773-82.

Cho, S.W. et al, 2013, Nature biotechnology 31(3):230-2.

Cho, S.W. et al, 2014, Genome research 24(1):132-41.

Christian, M. et al, 2010, Genetics 186(2):757-61.

Cong, L. et al, 2013, Science 339(6121) 819-23.

Cox, D.B. et al 2015, Nature media 21(2) 121-31.

Doench, J.G. et al 2014, Nature biotechnology 32(12) 1262-7.

Fu, Y, et al, 2013, Nature biotechnology 31(9):822-6.

Fu, Y, et al, 2014, Nature biotechnology 32(3):279-84.

Gaj, T, et al, 2013, Trends in biotechnology 31(7), 397-405.

Gori, J.L. et al 2015, Human gene therapy 26(7) 443-51.

Gutschner T et al, Cell Reports 14(2016) 1555-1566.

Haeussler, M. et al, 2016, Genome biology 17(1):148.

Hockemeyer, D.et al, 2011, Nature biotechnology 29(8):731-4.

Holt, N. et al, 2010, Nature biotechnology 28(8):839-47.

Jinek, m. et al, 2012, Science 337(6096):816-21.

Jinek, m. et al, 2013, ehife 2: e00471.

Kleinstimer, B.P. et al, Nature 529(2016)490-495.

Li, H. et al, 2011, Nature 475(7355):217-21.

Liao, h.k. et al 2015, Nature Communications 6.

Ma Y et al, RNA biol.13(2016)605-12.

Makarova, K.S. et al, 2011, Nature reviews. microbiology 9(6):467-77.

Mali, P.et al, 2013, Nature biotechnology 31(9):833-8.

Mali, P. et al, 2013, Science 339(6121):823-6.

Maruyama, T. et al 2015, Nature biotechnology 33(5):538-42.

Mayer, K. et al 2015, International J.of molecular sciences 16(11) 27497-507.

McMahon, M.A. et al, 2012, Nature methods 9(1):28-31.

Miller, J.C. et al, 2007, Nature biotechnology 25(7):778-785.

Miller, J.C. et al, 2011, Nature biotechnology 29(2):143-8.

Muller, M. et al, 2016, Journal am. society of Gene Therapy 24(3):636-44.

Pattanayak, V. et al, 2013, Nature biotechnology 31(9):839-43.

Perez, E.E. et al, 2008, Nature biotechnology 26(7):808-16.

Pinder, J.et al, Nucl. acids Res.43(2015) 9379-9392.

Ran, F.A. et al, 2013, Nature protocols 8(11):2281-308.

Sander, J.D. et al, 2011, Nature methods 8(1):67-U94.

Sanjana, N.E. et al, 2012, Nature protocols 7(1):171-92.

Shalem, O. et al 2014, Science 343(6166) 84-7.

Song, J, et al, Nature Comm.2016-01-01.

Stahl, S. et al 2015, Proc. Natl. Acad. Sci USA 112(34) 10732-7.

Tebas, P.et al 2014, The New England Journal of medicine 370(10) 901-10.

Urnov, F.D. et al, 2010, Genetics 11(9):636-46.

Vara, J.A. et al, 1986, Nucleic acids research 14(11):4617-24.

Weidle, U.H. et al, Cancer Gen.prot.11(2014)25-38.

Wood, A.J. et al, 2011, Science 333(6040): 307-.

Yin, H.et al, 2014, Nature biotechnology 32(6) 551-3.

Zhang, F. et al, 2011, Nature biotechnology 29(2):149-53.

Zhang, G. et al, 1996, Biochemical and biological research communications227(3):707-11.

Zhang, x.h. et al 2015, Nucleic acids 4: e264.

Brief Description of Drawings

Fig. 1 composition of gRNA: the sequence, size, and exact genomic locus of grnas targeting DPH1(a) and DPH2 (B); exact chromosomal location is indicated in parentheses; LHR: left homologous region of integration cassette on donor plasmid, RHR: the right homologous region of the integration cassette on the donor plasmid. The corresponding mRNA sequence is RefSeq: NM _001383(DPH1) and RefSeq: NM-001039589, NM-001384 (DPH 2).

Figure 2 MCF7 cells and MCF7 containing either heterozygous (wt/k.o.) or homozygous (k.o./k.o.) inactivated DPH1 gene were exposed to different concentrations of diphtheria toxin for 48 hours. Cell viability was assessed by viability assays as previously described (Mayer et al, 2015; Stahl et al, 2015).

Figure 3 MCF7 cells were transfected with plasmids encoding the CRISPR/Cas9 module for inactivation of the Diphtheria Toxin (DT) sensitivity gene, DPH1, and for integration of the PAC expression cassette conferring Puromycin (PM) resistance. (A & B) 48 hours after transfection, cells were exposed to DT, at a concentration lethal to cells containing DPH1 function. This toxin selection resulted in a survivor colony (upper colony in a) in which all copies of the DPH1 gene were inactivated. Colonies that retained a functional DPH1 gene were killed by the toxin (colony debris in a). Colonies can be visualized by staining and their number quantified as colony counts as described in (B). Note that DT resistant colonies appeared only when the cells were treated with DPH1 gene specific guide RNA, without any non-specific background in the cells exposed to the control guide. (C) 48 hours after transfection, cells were exposed to PM at a concentration lethal to cells not carrying the PAC expression cassette. Thus, PM selection results in a survivor colony carrying at least one PAC expression cassette. Colonies can be visualized by staining and their number quantified as colony counts as described in (C). Note that when cells were treated with guide RNA for specific integration into the DPH1 gene, PM resistant colonies appeared in higher numbers and showed background in cells exposed to the control guide, indicating an unspecific integration event.

FIG. 4 MCF7wt cells, MCF7wtko cells with one DPH1 wild-type allele and one inactivated allele, and MCF7koko cells with both DPH1 genes inactivated were subjected to HRM PCR. PCR fragments that received HRM covered the CRISPR/Cas target region in the DPH1 gene. Note that cells carrying the modification in the DPH1 gene can be distinguished from wild-type cells by differences in their melting curves (indicating the presence of an altered allele). Homozygous and heterozygous modifications may differ from unmodified wild-type cells. This method does not distinguish between cells modified on only one allele (heterozygous, toxin-sensitive) and cells with both alleles inactivated (homozygous, toxin-resistant). Toxin-sensitive cells have a biphasic melting curve indicating that the cell has one wild-type allele (conferring toxin sensitivity) and one mutant allele (aberrant melting point).

Figure 5 MCF7 cells transfected with DPH 1-specific CRISPR/Cas module received DT selection, or HRM-PCR followed by DT selection. (A) DT selection only produced resistant colonies with matching DPH1 guide RNA, no colonies appeared in untreated cells or in cells receiving scrambled guide RNA. (B) HRM-PCR revealed the frequency of cells that were modified at one or both alleles of their DPH1 gene. Subsequent DT sensitivity assays showed that the frequency of single allele hits (toxin sensitive and HRM-positive) was twice that of biallelic inactivation (HRM positive and toxin resistant).

Figure 6 MCF7 cells transfected with DPH 1-specific CRISPR/Cas module received PM selection and/or received DT selection. The transfected control showed neither DTr nor PMr colonies. (A) PM selection generated resistant colonies 2 times more frequently when compared to scrambled guide RNA using DPH1 guide RNA. No colonies appeared in the untreated cells. (B) The combination of PM selection and DT selection allows quantification of the frequency of PAC cassette integration in cells with complete inactivation of the DPH1 gene. DPH1 gives rise to clones with dual PM-DT resistance to guide RNA. The scrambled guide RNA gave rise to only PM resistant clones, but not DT resistant clones. (C) Comparison of DT resistance (inactivation of both DPH1 genes) colony frequency and PM resistance (PAC integration at DPH1 gene or at another site) frequency. (D) Mean + SEM (n ═ 4, p <0.001) is shown. PM selection produced resistant colonies 2-fold more frequently when compared to scrambled guide RNA using DPH1 gRNA. The PM selection and DT selection combinations revealed the frequency of PAC cassette integration in cells with both DPH1 alleles inactivated. DPH1gRNA produced clones with dual PM-DT resistance. ScRNA produced only PMr colonies, but not DTr colonies. (E) Mean + SEM (n ═ 4, p <0.001) is shown. Frequency comparison of DTr (both DPH1 genes inactivated) colonies and PMr colonies (PAC integrated at DPH1 or at another site). (F) Mean + SEM (n ═ 4, p <0.001) is shown. MCF7 cells transfected with DPH 2-specific grnas received PM selection and/or received DT selection.

Figure 7 MCF7 cells transfected with DPH 2-specific CRISPR/Cas module received PM selection and/or received DT selection. In the same way as observed for DPH1, the frequency of DT resistant colonies was much higher than PM resistant colonies. Therefore, biallelic inactivation is significantly more efficient than PAC cassette integration.

FIG. 8 optimized Gene editing: effect of gRNA length and editing enzymes on potency and specificity. (A) MCF7 cells transfected with a DPH 1-specific CRISPR/Cas module received PM selection and DT selection using guide rnas (grnas) of different lengths. The absolute gene inactivation frequency and cassette integration frequency can be determined by reading the results, as well as the determination of guide RNA independent non-specific integration events. (B) The conditions under which desired cassette integration occurs and non-integration associated gene inactivation events are reduced are determined by a normalized ratio between DT resistance frequency and PM resistance frequency. Furthermore, it was noted that the maximum efficiency of destructive gene editing was achieved with the 20mer, whereas the maximum integration efficiency was observed with the 16-18 mer. (C) The greatest efficiency of efficient integration (cells resistant to PM and DT) at the target gene was observed with the 18mer guide. (D) Mean + SEM (n ═ 4, Λ/Φ/Ψ Ψ P <0.01, Λ Λ Λ/Φ/Ψ Ψ P <0.001) is shown. MCF-7 cells transfected with the DPH 1-specific Cas9 module received PM selection and DT selection using different lengths of grnas. gRNA length affects gene inactivation and integration frequency. All scRNA/gRNA values were compared to the maximum value of the corresponding selection group. (E) Mean + SEM (n ═ 4, Λ/Φ/Ψ Ψ P <0.01, Λ Λ Λ/Φ/Ψ Ψ P <0.001) is shown. MCF-7 cells transfected with the DPH 1-specific Cas9 module received PM selection and DT selection using different enzymes. The total number of DTr, PMr or DTr + PMr colonies in the DPH1 edits were close with the 20mer gRNA (CRISPR/Cas9) and the designed ZFN. (F) Mean + SEM (n ═ 4, Λ/Φ/Ψ Ψ P <0.01, Λ Λ Λ/Φ/Ψ Ψ P <0.001) is shown. MCF-7 cells transfected with the DPH 1-specific Cas9 module received PM selection and DT selection using different enzymes. The ratio of total integration events to total inactivation events (PMr/DTr) and the ratio of site-specific integration events/total target gene inactivation events (DTr + PMr)/DTr).

Figure 9 frequency of CRISPR/Cas-mediated specific and non-specific gene editing events observed using a 20mer guide RNA targeting the human DPH1 gene.

FIG. 10 Effect of DNA repair modulators on Gene editing. MCF-7 cells were transfected with plasmids encoding 20mer gRNA, SpCas9, and PAC. HR modulator RS-1 (8. mu.M) and NHEJ-modulating SCR7 (1. mu.M) were added 4 hours before transfection or 18 hours after transfection. DT and/or PM selection was initiated 72 hours after transfection. The percentage of PMr colonies (integrated) relative to DTr colonies (sheared) is shown. Mean + SEM (n ═ 4) is shown. Φ, Λ p ═ 0.05; Φ, Λ Λ p ═ 0.01; Φ Φ Λ Λ Λ p ═ 0.001, relative to untreated (no added agent) controls.

The following examples, figures and sequences are provided to aid the understanding of the present invention, the scope of which is set forth in the appended claims. It will be appreciated that modifications may be made to the method described without departing from the spirit of the invention.

Examples

Example 1

Culturing MCF7 cells and transfecting plasmids encoding gene editing entities

MCF7 cells (Brooks et al, 1973) were originally obtained from ATCC (and subsequently propagated in their own cell banks) and maintained in RPMI 1641 medium supplemented with 10% FCS, 2mM L-glutamine and penicillin/streptomycin at 37 ℃ and 85% humidity. To transfect the plasmid carrying the gene editing module, 3,000,000 cells were seeded in 10cm diameter petri dishes and humidified 5% CO at 37 deg.C₂Culturing in medium. 24 hours after inoculation, cells were transfected with 20. mu.g of total DNA using JetPEI (Polyplus) according to the manufacturer's protocol, but with an N/P ratio of 6: 1. After 24 hours, the transfection efficiency of cells transfected with the modified eGFP expression plasmid was determined by flow cytometry (FACS Calibur, BD biosciences) (Zhang et al, 1996).

Plasmids encoding CRISPR/Cas9 gene editing entities directed against Dph1(gRNA target sequence: CAGGGCGGCCGAGACGGCCC (SEQ ID NO:01), from reference sequence: NM-001383) and Dph2(gRNA target sequence: GATGTTTAGCAGCCCTGCCG (SEQ ID NO:02), from reference sequences NM-001039589, NM-001384), and a scrambled control (gRNA sequence: GCACTACCAGAGCTAACTCA (SEQ ID NO:03)) were obtained from OriGene (i.e., knockout and integration systems). The system comprises one plasmid expressing about 100nt gRNA under the control of the U6 promoter and Cas9 nuclease under the control of the CMV promoter, and a donor plasmid with a low promoter GFP/PM expression cassette flanked by homology arms of the target gene (Dph1 or Dph 2). Other Dph1gRNA sequence variants (OriGene) of different sizes are

(see FIG. 1A).

Example 2

Identification and quantification of CRISPR/Cas 9-mediated homozygous knockouts of DPH1 and DPH2 genes

MCF7 cells that inactivated all chromosomal copies of DPH1 or DPH2 were resistant to diphtheria toxin (Stahl et al, 2015). Thus, the appearance and frequency of toxin-resistant cells/colonies when CRISPR/Cas-induced gene inactivation provides a measure of the efficiency of inactivation of all gene copies. Thus, MCF/cells were transfected as described in example 1 with (i) GFP expression plasmids as transfection controls, (ii) CRISPR/Cas9Dph1 or Dph2 knockout/integration systems and (iii) knockout/integration entities containing scrambled grnas to determine Cas9 independent integration frequency. After determining the transfection efficiency, cells were seeded in 6-well plates. To quantify homozygous knockout events by DT, 20000 cells were seeded and to quantify integration events by PM or both 40000 cells were seeded. 3 days after seeding the cells, RPMI medium was replaced with RPMI medium containing DT or PM or both toxins. The exchange of medium is repeated every 2-3 days until all dead cells are removed from the plate and actively growing colonies can be detected by microscopic examination. After this time (between day 12 and 14 after the start of toxin exposure), cells were washed 3 times with PBS and then stained with ice-cold methylene blue (0.2% in 50% EtOH). The methylene blue solution was then removed and the plates carefully washed under running water. Finally, the cell colonies were counted under the microscope with a 5mm grid foil. The results of these analyses are summarized in the table below.

Table: colony count and phenotypic frequency of transfected MCF-7 cells

'TF eff.': transfection efficiency was determined by FACS analysis by monitoring the frequency of fluorescent cells after transfection of MCF-7 with GFP-reporter plasmid. The relative number of GFP positive cells in% of all cells is listed. The average transfection efficiency for all assays was between 30% and 40%. A 'HRM': high resolution melting point PCR positive cells are defined as cells that exhibit a well defined divergent (biphasic) melting curve pattern compared to wild type cells. O.' indicates the frequency of cells that do not carry a functional copy of the DPH1 or DPH2 gene and are therefore resistant to DT. 'int.' denotes the frequency of cells carrying the PAC expression cassette and thus resistant to PM. 'A-D' indicates each sample of independent (quadruplicate) experiments.

Example 3

Detection of CRISPR/Cas 9-mediated DPH1 and DPH2 Gene heterozygous knockout

Cells modified with only one allele of dph1 or dph2 were rendered toxin-free. To identify and quantify these events, High Resolution Melting (HRM) PCR was applied in a similar manner as described by Stahl et al (Stahl et al 2015): 24 hours after transfection, single cells were deposited in 96-well plates by FACS (FACSAria, BD biosciences) and cultured until confluent. Cells were washed with PBS and lysed by adding 40 μ L cell lysis buffer per well (Roche). After incubation for 15 minutes at 750 rpm on a plate shaker (Titramax 1000, Heidolph), the cell lysates were of PCR grade H₂Diluting O at 1: 5. mu.L of cell lysate was mixed with High Resolution Melt (HRM) master mix (Roche) and with primers covering the gRNA target sequences. PCR and HRM were performed in LC480II (Roche) according to the manufacturer's protocol. Clones with edited target genes were identified by altered melting curves compared to MCF7-WT cells.

With cell viability assays, cells with a biphasic melting curve may still possess one wild-type allele or may have both alleles inactivated. Thus, such clones were expanded (toxin-free or puromycin-free selection) and subjected to cell viability analysis to distinguish heterozygous (toxin-sensitive) and homozygous (resistant) knockout cells. These assays were performed in flat-bottomed 96-well plates containing 10,000 cells at 37 ℃ in humidified 5% CO₂Under the condition of the reaction. 24 hours after inoculation, cells were exposed to toxin for 72 hours. Metabolic activity of viable cells was performed according to the manufacturer's instructionsEvaluation of luminescent viability assay (Promega). The results of these analyses are summarized in the table above.

Example 4

Cassette integration event

The CRISPR/Cas module for targeted integration contains a Puromycin (PM) resistance gene expression cassette without a promoter to avoid transient expression. Thus, by determining the sensitivity of the cell to PM, integration of the recombination sequence into the genome is detected. PM (quantitative elimination of wild-type MCF7 cells) was applied at a concentration of 500 ng/. mu.L to select MCF7 cells stably incorporating the expression cassette. Using the same procedure as described above for DT selection, PM resistant colonies were selected and visualized and quantified in the same manner as described above for toxin resistant colonies (between 12 and 14 days after selection began). The number of PM resistant colonies obtained with the non-targeting leader sequence (scrambled gRNA, figure 1) reflects the non-specific integration background. The increase in PM resistant but toxin sensitive colonies obtained with the target gene gRNA (above non-specific background), reflected in integration into only one target allele, with the other dph1 or dph2 allele remaining unaffected. The PM-resistant colonies, which are also DT-resistant, have the expression cassette integrated into at least one target allele, while the other allele is inactivated or inactivated and accompanied by another integration event. The results of these analyses are summarized in table 1.

Example 5

Combined discrimination and quantification of homozygous and heterozygous dph1 gene inactivation events for diphtheria toxin resistance assay and HRM-PCR

Diphtheria Toxin (DT) performs ADP-ribosylation of diphtheria amide on eukaryotic translational elongation factor 2(eEF2) and thus inactivates eEF 2. This irreversibly stops protein synthesis and kills the cells. Diphtheria amide is a defined histidine placed on eEF2 by means of multiple diphtheria amide synthesis genes, including diphtheria amide biosynthesis 1 gene ═ DPH 1. Complete inactivation of all DPH1 alleles (e.g., by gene editing) in MCF7 cells prevented synthesis of the toxin target diphtheria amide. This renders the cells resistant to Pseudomonas exotoxin A (PE) and DT (Stahl et al). As a result, all functional copies of inactivated DPH1 gave rise to the 'DT-resistant' phenotype (fig. 3). The frequency of this phenotype can be detected in a rapid and robust manner by counting toxin-resistant colonies as described in examples 2-4 (results in the above table and figure 3).

Since the presence of only one remaining functional DPH1 allele is sufficient for toxin sensitivity (fig. 1 and fig. 2), the resistance phenotype specifically defines cells with a full DPH1 allele knock-out. Cells modified by only one allele can be identified by the HRM-PCR assay (described in example 3) performed directly on cultured cells. The modification to the CRISPR/Cas target site changes the melting temperature of the dph 1-gene derived PCR fragment compared to the wild type fragment. This is reflected by a biphasic melting curve in the HRM spectra (exemplarily shown in fig. 4). Since CRISPR-Cas 9-mediated gene inactivation events are rarely the same on both alleles, many cells with complete gene inactivation will also show biphasic HRM curves. These cells can be distinguished from single allele changes by their toxin resistance phenotype.

Thus, the combination of high throughput HRM-PCR assay and toxin selection (colony counting) assay on cells enables quantification of heterozygous and homozygous DPH1 gene-specific modification events.

Example 6

Puromycin resistance assay detects and distinguishes the frequency of gene-specific and non-specific integration events

puromycin-N-acetyl-transferase (PAC), encoded by the integration cassette of the CRISPR/Cas9 plasmid applied, inactivates PM and thus makes cells resistant to PM (Vara et al, 1986). Thus, the integration of PAC expression cassettes can be detected and quantified by the PM resistance assay in the same manner as described above for the DT resistance colonies (PM applied instead of DT as selection agent, fig. 3). In contrast to DT resistance, which is only due to inactivation of specific and homozygous target genes, PM resistance marks any integration event, regardless of the integration position. The frequency of site-specific integration and non-specific integration can be resolved by comparing the number of PM-resistant cells exposed to the target gene-specific guide RNA with the number of cells exposed to the scrambled non-specific guide (see fig. 3).

Example 7

Comparison of DPH1 Gene inactivation event and Targeted integration event by CRISPR-Cas9

To compare the frequency of target-specific inactivation, integration and non-target integration, a plasmid encoding the DPH 1-specific CRISPR-Cas9 module (example 1, fig. 1) was transfected into MCF7 cells. These cells were then subjected to HRM-PCR and colony counting assays to detect DT resistance and PM resistance. The results of these assays are summarized in fig. 5, and various data sets are available in table 1 above.

Fig. 5A shows complete inactivation of the DPH1 gene, indicating that loss of function of the entire DPH1 allele occurs at a frequency of about 6% of the entire transfected cells (2.5% of the entire cells are considered as 40% transfection efficiency, table 1). Inactivation of the DPH1 gene showed an absolute dependence on the matching guide RNA sequence: the out-of-order guide did not generate any DT-dependent colonies. Comparison of HRM 'hits' on DPH1 gene with the appearance of DT resistant colonies is shown in fig. 5B. These analyses revealed that single allele inactivation (toxin sensitive HRM hits) occurred twice as frequently as double allele inactivation (DT resistant cells).

FIG. 6 shows the frequency of PM resistant colonies compared to DT resistant and PM resistant colonies. These results indicate that DPH1 biallelic inactivation occurs at 30-50 fold higher efficiency than integration of the PAC expression cassette mediating PM resistance (fig. 6A, without distinguishing the location or zygotic nature of PAC integration). The scrambled guide (under the same conditions) produced 2-fold fewer PM resistant colonies than did the application of DPH 1-specific guide RNA. This means that CRISPR/Cas9/DPH 1-guide mediated PAC gene integration occurs preferentially at the DPH1 gene, but without absolute specificity. Based on preferential integration at the DPH1 target site, many of the PM-resistant colonies obtained using DPH1 guidance were toxin-resistant (fig. 6B). None of the PM resistant colonies obtained using the shuffling guide were DT resistant.

These results show that CRISPR/Cas 9-mediated inactivation of the target gene (even at both alleles) occurs highly specific and at a much higher frequency than targeted integration (fig. 6C). Targeted integration not only occurs less frequently, but also is less specific for locations defined by the guide RNA.

Example 8

The applicability and results of the analysis/quantification protocol for CRISPR/Cas9 gene editing can be transferred to other target genes

The results presented herein (including the observation of large differences between targeted gene inactivation and integration) are a general feature of CRISPR-Cas 9-mediated gene editing or a special feature of the DPH1 gene? To address this issue, the same experimental protocol was applied for CRISPR-Cas 9-induced DPH2 gene modification. DPH2 is located on a different chromosome than DPH1, has a different sequence and encodes a different enzyme. However, DPH2 is also essential for leucamide synthesis and DPH2 deficiency renders cells resistant to DT in the same way as DPH1 deficiency (Stahl et al). Thus, the detection and quantification principles reported and developed herein to identify and differentiate DPH1 regulation may also be applicable to the analysis of CRISPR/Cas9 mediated DPH2 alterations.

Results of DPH2 gene editing are listed in fig. 7, followed by DT resistance and PM resistance assessment results: consistent with the observation of DPH1, the frequency of homozygous DPH2 gene inactivation events was observed to be about 40-fold higher than PAC expression cassette integration. Inactivation of DPH2 is strictly dependent on the presence of cognate guide RNAs, whereas cassette integration is site-preferred, but not absolutely specific for the target gene (compare the frequency of DPH2 guidance with scrambled guide RNAs). The striking similarity of the results compiled for the DPH1 gene and the DPH2 gene (even though both are different genes on different chromosomes) clearly demonstrates that the knowledge and rules obtained with this analysis system can be transferred to other suitable genes and provide a reasonable basis for: the methods reported herein have broad applicability.

Example 9

The method as reported herein is applied: comparing and optimizing gene manipulation modules

The results of the DPH1 and DPH2 gene editing experiments are highly comparable, even though both are different genes on different chromosomes. It can therefore be reasonably concluded that: the 'rules' and knowledge obtained with this analysis system can be generalized to other genes. Thus, the method as reported herein can be used in an analysis to resolve the overall impact of the composition of the genetically modified modules.

Thus, the methods as reported herein have been used to compare and optimize gene manipulation modules. Figure 8 shows how the efficiency and specificity of gene inactivation and integration of CRISPR/Cas guide RNAs of different lengths were assessed and compared for their efficiency and specificity of gene inactivation and/or integration. The applied guide RNAs all target the same sequence segment within DPH1, but vary in length from 14 to 26 bases (guide RNA is detailed in fig. 1). The frequency of DT resistant colonies was recorded to reflect inactivation of the target gene specific homozygous gene. At the same time, the number of PM resistant colonies and DT-PM dual resistant colonies were evaluated to monitor cassette integration.

As expected, the guide RNA length affected the gene inactivation efficiency, with the 20mer conferring the greatest efficiency of DPH1 gene inactivation. Shortening the complementary stretch to 18 or 16 bases or extending it to 26 bases retained significant specific gene inactivation function, but at a lower efficiency than the 20 mer. Reducing guide RNA fragments to less than 16 bases (14 mers) reduces DPH1 inactivation function below detectable levels.

Integration efficiency (assessed by counting PM-resistant events, comparison and see fig. 5B) was also affected by guide RNA length. The less than 16mer guide (14mer) produced only a few PM resistant colonies, not exceeding the background level of the scrambled control. Significantly above background alignment was observed for the 16mer, 18mer, 20mer, 22mer, 24mer and 26mer, with the best insertion efficiency achieved with the 16-18 mer. Efficiency gains were not achieved with larger oligo-nucleic acids, and in fact, the larger size (including 20 mers) significantly reduced specific insertion events.

It has been found that the ratio between the integration event (PM-r) and the inactivation event (DT-r) can be calculated as a 'specific indicator' to identify conditions where specific integration occurs with minimal gene inactivation events. Such conditions may be advantageous if targeted integration is desired without causing too much damage to the inefficient target gene.

Low values (e.g., fewer PM resistant colonies relative to DT-r colonies) reflect ineffective integration relative to concurrent inactivation events. High values (more PM resistant colonies and/or relatively reduced number of DT-r colonies) reflect more efficient integration at CRISPR/Cas affected target genes.

FIG. 8B shows the calculated specificity coefficients depending on the length of the guide RNA. It has been observed that for 16-18 mers the 'insertion/inactivation' value is highest and that for guide RNAs containing 20 or more bases there is a significant decrease. This indicates that the 20mer is quite efficient for inactivation of the orientation gene (in agreement with previous observations, the guide RNA is detailed), whereas the shorter guide appears to be more efficient for specific integration. The shorter guide not only improved the integration efficiency (overall higher number of PM resistant colonies), but also improved the ratio between efficient and inefficient gene editing (reduction of DT-r colonies without insertions).

Example 10

Comparing efficiency and specificity of different Gene editing protocols

To compare the gene inactivation and integration events, efficiency and specificity of RNA-guided different variants of Cas9 and ZFN-mediated gene editing, gRNA length and composition remained constant (20mer, targeting DPH 1). As variables, three different editing enzymes were applied:

(i) 'SpCAS 9' refers specifically to Cas9 nuclease from Streptococcus pyogenes (Streptococcus pyogenes), which can be considered as the current standard application (Ran et al, 2013, Fu, Sander et al, 2014);

(ii) SpCas9-HF1 is a non-specific DNA-binding reduced, off-target activity reduced and therefore SpCas9 engineered variant with higher accuracy and specificity was proposed (Kleinstiver, Pattanayak et al, 2016);

(iii) ZFN-mediated gene editing modules that recognize target sequences via engineered zinc finger motifs (i.e., protein-nucleic acid interactions, as opposed to nucleic acid hybridization) (Miller, Holmes et al, 2007; Urnov, Rebar et al, 2010).

The frequency of DT-resistant colonies was recorded to reflect target gene-specific homozygous gene inactivation in the same manner as described for gRNA length analysis in the previous examples. Both PM resistant colonies and DT-PM dual resistant colonies were evaluated to assess monitoring cassette integration.

Comparing the overall efficacy of gene inactivation and cassette integration, the highest inactivation and integration values were observed using CRISPR/SpCas 9. For CRISPR/SpCas9-HF, targeted gene inactivation events were fewer than CRISPR/SpCas9, less than 20% of the number of DT resistant colonies. The number of PM resistant colonies (integration) and the number of DT-PM dual resistant colonies (cassette integration and targeted simultaneous gene inactivation) also decreased strongly. Under the same conditions, application of the ZFN module resulted in a reduction in the number of DT resistant colonies (homozygous gene disruption) to less than 60% of the events observed using CRISPR/SpCas 9. Thus, the targeted gene inactivation efficiency was about 2-fold lower compared to SpCAS9 and increased by about 2-3-fold compared to engineered SpCAS9-HF 1. The number of PM-resistant colonies (integrations) did not differ significantly between CRISPR/SpCas9 and ZFNs. Compared to CRISPR/SpCas9, DT-PM double resistant colonies (cassette integration with concomitant inactivation) of ZFNs were more or less (30%) reduced.

Calculation of DT-resistant (target gene inactivation) colonies/DT + PM dual-resistant (target site integration) colony ratios (overall efficiency from disclosure) showed that CRISPR/SpCas9 and CRISPR/Cas9-HF and ZFN systems produced the same level (about 4x 10)^-3) Targeted integration event/one homozygous/complete gene inactivation event. Thus, although the overall efficiency of gene modification varies with existing editing systems, it is above the background of target gene inactivationThe cassette integration 'specificity' of (a) was determined to be not significantly different for all three protocols evaluated.

Thus, the method as reported herein may thus be used to evaluate other editing systems or additions supporting editing events to identify improved editing modules and/or conditions.

Table: colony count and phenotypic frequency of MCF-7 cells transfected with different Gene editing entities

MCF-7 cells were transfected with plasmids encoding different genome editing systems (SpCas9, SpCas9, ZFN). The SpCas9 construct was used as described above. According to Kleinstitver, Pattanayak et al, 2016, SpCas9-HF, comprises N497A/R661A/Q695A/Q926A substitutions. At the same time, grnas were replaced with scrnas in parallel to resolve nonspecific activity. DPH1 specific ZFN from Sigma AldrichAnd (4) obtaining. For transfection of 3X10⁶The total amount of plasmid DNA (editing entity and donor) of the initial cell pool of individual cells was as described for the previous experiments. To quantify the transfection efficiency (TF eff. (%)), GFP reporter plasmids were additionally transfected. GFP positive cells were counted by FACS 24 hours after transfection. A defined number of cells (number of cells seeded) were seeded and treated thereafter with DT, PM or DT + PM for 72 hours.

Example 11

Effect of DNA repair Modulator on Gene editing efficiency and specificity

The methods as reported herein (including colony assays to determine DT-resistant cells and PM-resistant cells after DPH gene editing) can also be used to analyze the effect of compounds that modulate DNA repair mechanisms. For example, inhibitors of the non-homologous end joining (NHEJ) process have been described to modulate gene editing events (Ma, Y et al, 2016). In the same way, activators of Homologous Recombination (HR) can increase the efficiency of integration of the targeting cassette (Song, J et al, 2016).

To evaluate and quantify the effect of DNA repair modulators on gene editing efficiency and specificity, CRISPR/SpCAS9 and 20mer gRNA targeting DPH1 were combined with compounds that modulate the DNA repair mechanism. The inhibitor Scr7 was applied 4 hours before transfection of the gene editing module or 18 hours after transfection until 72 hours after transfection to inhibit DNA ligase IV. In the same manner, RAD51 modulator RS-1 (compound 1 stimulating RAD 51) was added to stimulate HR. Both compounds were applied at doses that did not affect the growth or viability of MCF7 cells: scr7 was 1. mu.M and RS-1 was 8. mu.M, and when the two were combined, 1. mu.M + 8. mu.M (Scr7+ RS 1). The frequency of DT resistant colonies, PM resistant colonies and dual resistant colonies was then recorded to reflect the efficiency and specificity of target gene inactivation events and cassette integration events under these different conditions.

It can be shown that the modulation of NHEJ by adding Scr7 from 4 hours before transfection up to 72 hours after transfection did not have a detectable effect on the number of DT resistant colonies (reflecting the frequency of inactivation of the target gene). It can also be shown that the co-application of SCR7 and RS-1 4 hours before transfection and SCR7 18 to 72 hours after transfection resulted in a small (still significant) reduction in the number of DT resistant colonies compared to control transfection. Thus, it can be shown that DNA ligase IV mediated NHEJ can affect CRISPR/SpCAS 9-induced inactivation of target genes to some extent. Although its effect on the overall number of DT-resistant colonies was limited, it can be shown that the frequency of PM resistant colonies was significantly stimulated (approximately 2-fold, table below) by SCR7 application 4 hours before transfection up to 72 hours after transfection. This suggests that SCR7 increases cassette integration events, confirming the enhancement of previous observations of efficient gene editing when SCR7 was administered (Ma Y et al, 2016). Thus, these results provide evidence: methods as reported herein can be used to determine such effects.

It can be shown that the modulation of recombination by addition of HR stimulator RS-1 from 4 hours before transfection up to 72 hours after transfection increased the number of PM resistant colonies by approximately 2-fold, while not affecting the number of colonies obtained with the disorder guide (table below). It can also be shown that RS1 does not affect the efficiency of cassette integration when RS1 is applied 18 to 72 hours after editing is initiated. Thus, these results provide evidence: methods as reported herein can be used to determine such effects.

It can be shown that the combination of ligase iv (nhej) inhibition and HR stimulation in gene editing (by adding Scr7 and RS-1 from 4 hours before transfection up to 72 hours after transfection) reduced the appearance of DT resistant cells and further increased the frequency of PM resistant cells (table below). It can be shown that this protocol also resulted in the highest PM resistance colony/DT resistance colony ratios compared to all other protocols. Thus, these results provide evidence: methods as reported herein can be used to determine such effects.

It can be concluded from these experiments that the methods as reported herein can be used to determine compounds, combinations of compounds, the effect of compound application time and the underlying mechanisms affecting the outcome of gene editing protocols. Thus, the methods as reported herein are therefore useful for identifying compounds or conditions that can enhance the desired gene editing effect and reduce unwanted 'side effects'.

Table: colony count and phenotypic frequency of MCF-7 cells exposed to SCR7 and/or RS-1 during Gene editing

As described previously, MCF-7 cells were transfected with plasmids for SpCas 9-mediated editing. Equal numbers of cells (number of cells seeded) were inoculated and treated with DT or PM 72 hours after initiation of editing.

(A) Scr7 (1. mu.M final concentration), RS-1 (8. mu.M final concentration) or Scr7+ RS 1(1 uM +8uM, respectively) was added 4 hours before transfection.

(B) Effect of RS1 and Scr7 application time (4 hours vs. 18 hours post transfection) on compound-mediated SpCas9-gRNA mediated gene editing. (. about) difference was significant, p < 0.008.

A

B

Example 12

Identification and quantification of ZFN-mediated DPH1 gene editing

MCF7 cells that inactivated all chromosomal copies of DPH1 were DT resistant. Thus, the appearance and frequency of DTr colonies when ZFN-induced gene inactivation and/or cassette integration provides a measure of the efficacy of inactivating all gene copies. The ZFN recognition sequence (CAGGTGATGGCGGCGCTGGTCGTATCCGGGGCAGCGGAGCAG, cleavage site, SEQ ID NO:10) was from NM-001383.3 (DPH1-wt) and was obtained from Sigma Aldrich. The PAC integration cassette for this location was obtained from OriGene. MCF7 cells were transfected with (i) a GFP expression plasmid, (ii) a plasmid encoding a ZFN targeting DPH1, and (iii) a PAC integration cassette targeting DPH1 as described above. After determining the transfection efficiency, cells were seeded in 6-well plates. To quantify homozygous knockout events (DTr), 20,000 cells were seeded, and to quantify integration events (PMr) or dual resistance, 40,000 cells were seeded. 3 days after inoculation, RPMI was exchanged for RPMI medium containing DT or PM or both. The medium was changed every 2-3 days. Between day 12 and 14 after the start of toxin exposure, cells were washed 3 times with PBS and then stained with ice-cold methylene blue (0.2% in 50% EtOH), then washed under running water and the number of colonies counted microscopically with 5mm grid foil.

Example 13

Quantifying the effect of HR modulators and NHEJ modulators on CRISPR/Cas 9-mediated editing

Compound 1(RS-1) stimulating RAD51 was applied during gene editing to modulate Homologous Recombination (HR). RS-1(SigmaAldrich) was dissolved in DMSO to generate a 10mg/ml stock solution that was diluted in RPMI medium immediately prior to application to the cells. Viability (Promega CTG) assay identified that the final RS-1 concentration of 8. mu.M was the dose that did not cause growth inhibitory or toxic effects on MCF7 (viability: 1. mu.M-100%; 3.7. mu.M-100%; 11. mu.M-97%; 33. mu.M-61%).

DNA ligase IV inhibitor SCR7 was applied during gene editing to regulate non-homologous end joining (NHEJ). SCR7(Sigma Aldrich) was dissolved in DMSO to produce 10mg/ml of stock solution, which was diluted in RPMI medium immediately prior to application to the cells. Viability (Promega CTG) assay identified a final concentration of 1. mu.M as a dose that did not cause growth inhibitory or toxic effects on MCF7 (viability: 0.37. mu.M-100%; 1.1. mu.M-100%; 3.3. mu.M-97%; 10. mu.M-88%).

In the 'early exposure' setting, 4 hours prior to transfection of the gene editing module, SCR7 (1. mu.M final concentration) or RS-1 (8. mu.M final concentration) or SCR7+ RS-1 (1. mu.M + 8. mu.M final concentration) was added to MCF7 cells. For 'late exposure', SCR7(8 μ M final concentration) or RS-1(1 μ M final concentration) was added 18 hours after transfection to MCF7 cells. In both cases, cells were exposed to the modulator up to 96 hours post-transfection, i.e., 'early exposure' consisted of 100 hours of treatment and 'late exposure' consisted of 78 hours of treatment.

The system on which the effects of DNA repair modulators were determined consisted of a CRISPR/SpCas9 module with a DPH 120 mer gRNA, which was transfected into MCF7 cells as described in the previous examples and subjected to subsequent DT selection and PM selection. The frequency of DTr colonies, PMr colonies and dual resistant colonies were recorded to reflect gene inactivation events and cassette integration events.

Example 14

Statistical results

Unpaired, two-tailed Student's t-test was performed to make a single comparison between the two treatments. Multiple comparisons were analyzed statistically by one-way anova followed by Tukey true significance difference (HDS) post hoc tests. Significant differences are defined by p-value < 0.05. The significance levels determined by the student's t test in the figures are indicated by one, two or three asterisks corresponding to p <0.05, p <0.01 and p < 0.001. Likewise, the level of significance determined by Tukey's HDS test is indicated by Λ, Φ, or Ψ.

Claims (23)

1. Method for determining the introduction of a nucleic acid into the genome of a mammalian cell, wherein the mammalian cell comprises one or two transcriptionally active alleles of a DPH1, DPH2, DPH4 and/or DPH5 gene, the method comprising the steps of

-transfecting mammalian cells with one or more plasmids comprising the nucleic acid to be introduced and the elements required for gene editing of the DPH gene,

-culturing the transfected cells in the presence of a DPH gene transcription sensitive toxin,

the nucleic acid is determined to be introduced into the genome of the mammalian cell if the transfected cell survives in the presence of the toxin.

2. The method of claim 1, wherein the DPH gene transcription sensitive toxin is selected from the group consisting of Pseudomonas exotoxin and diphtheria toxin.

3. The method according to any one of claims 1 to 2, wherein the method comprises the steps of:

-transfecting mammalian cells with one or more plasmids comprising the nucleic acid to be introduced, the nucleic acid conferring resistance to a selectable marker, and the elements required for gene editing of the DPH gene,

-culturing the transfected cells in the absence of selective pressure,

-dividing the culture into at least two aliquots, or taking at least two samples from the culture, and

-incubating the first aliquot or sample in the presence of a DPH gene transcription sensitive toxin and incubating the second aliquot or sample in the presence of the corresponding selection marker.

4. The method according to any one of claims 1 to 3, wherein the mammalian cells are a plurality of mammalian cells, and the method comprises the following step immediately preceding the step of culturing the cells in the presence of a toxin and/or a selectable marker

-placing cells of the transfected plurality of cells in a single cell format.

5. The method of claim 4, wherein said plurality of mammalian cells is 1000 to 10,000,000 cells.

6. The method according to any one of claims 1 to 5, wherein the method is for determining gene editing efficiency, or for determining gene editing specificity, or for determining gene editing efficiency and specificity.

7. The method according to any one of claims 1 to 6, wherein the method is used for determining homozygous and heterozygous gene modifications, or for determining site-specific and non-specific gene disruption and integration.

8. The method according to any one of claims 1 to 7, wherein the introduction of the nucleic acid is to a nucleic acid homozygous in the genome of the mammalian cell if the transfected cell survives in the presence of the toxin.

9. The method according to any one of claims 1 to 8, wherein the introduction of the nucleic acid is a hybrid nucleic acid introduction into the genome of the mammalian cell, if the transfected cell is not viable in the presence of the toxin but survives in the presence of the selectable marker.

10. The method of any one of claims 1 to 9, wherein DPH gene inactivation events and nucleic acid integration events are quantified by a combination of toxin and selectable marker selection and optionally High Resolution Melting (HRM) PCR.

11. The method according to any one of claims 1 to 10, wherein the method is for assessing different gene editing methods and comprises the steps of

-transfecting a mammalian cell with one or more plasmids comprising the nucleic acid to be introduced, the nucleic acid conferring resistance to a selectable marker, and the elements required for the first genetic method for editing the DPH gene,

-culturing the transfected cells in the absence of selective pressure,

-dividing the culture into at least two aliquots, or taking at least two samples from the culture,

-incubating a first aliquot or sample in the presence of a DPH gene transcription sensitive toxin and incubating a second aliquot or sample in the presence of a corresponding selection marker,

-repeating these steps for all gene editing methods to be examined, and

-ranking different gene editing methods based on the frequency of toxin resistance, selectable marker resistance or dual resistance.

12. The method according to any one of claims 1 to 11, wherein the frequency of inactivation of all alleles of a target gene is detected by counting toxin-resistant colonies.

13. The method according to any one of claims 1 to 12, wherein inactivation of one allele of the target gene is detected by HRM-PCR as a function of the presence of a biphasic melting curve.

14. The method of claim 13, wherein the HRM-PCR is performed directly on the cultured cells.

15. The method according to any one of claims 1 to 14, wherein the frequency of inactivation of all alleles of the target gene by CRISPR/Cas9 is detected by enumerating toxin-resistant colonies in combination with a HRM-PCR determined duplex melting curve.

16. The method of any one of claims 1 to 15, wherein the DPH gene is selected from the group consisting of a DPH1 gene, a DPH2 gene, a DPH4 gene, and a DPH5 gene.

17. The method according to any one of claims 1 to 16, comprising the steps of: the number of toxin-resistant colonies, the number of antibiotic-resistant colonies and the number of toxin-to-antibiotic-resistant colonies were determined, wherein the ratio between the integration event (antibiotic-resistant colony number) and the inactivation event (toxin-resistant colony number) reflects the specificity of the method.

18. The method according to any one of claims 1 to 17, wherein the method is for selecting a guide RNA for CRISPR/Cas9 directed integration of a nucleic acid, wherein the method comprises the steps of: a plurality of different guide RNAs are provided and the guide RNA with the highest ratio between integration events (antibiotic resistant colony numbers) and inactivation events (toxin resistant colony numbers) is selected.

19. The method of any one of claims 1 to 18, wherein the gene editing method is selected from CRISPR/Cas, zinc finger nucleases, and TALENs.

20. Method for identifying/selecting (mutant forms or variants of) CRISPR/Cas9 or ZFNs or TALENs or other gene editing modules comprising the following steps

-providing/preparing a plurality of variants of one or more gene-editing modules,

-determining the efficiency of and/or the highest ratio between integrating events (antibiotic resistant colony numbers) and inactivating events (toxin resistant colony numbers) with the method according to any one of claims 1 to 19, and

-identifying/selecting the variant with the highest efficiency and/or the highest ratio.

21. A method for selecting a compound or combination of compounds that alters (enhances or decreases) the efficiency or specificity of a gene editing module/method comprising the steps of

-providing one or more compounds or one or more combinations of compounds,

-determining the efficiency of and/or the ratio between integrating events (antibiotic resistant colony numbers) and inactivating events (toxin resistant colony numbers) with the method according to any one of claims 1 to 19, optionally in the absence of said compound or combination of compounds,

-determining the efficiency of and/or the ratio between integrating events (antibiotic resistant colony numbers) and inactivating events (toxin resistant colony numbers) separately/individually for each of said compounds or combinations of compounds, using the method according to any one of claims 1 to 19, in the presence of said compound or combination of compounds,

-identifying/selecting at least one compound or combination of compounds having an efficiency and/or ratio which is different from the efficiency and/or ratio when carrying out the method as reported herein in the absence of said compound or combination of compounds.

22. A method for determining the concentration of a compound and the time point of its addition to enhance the efficiency or specificity of a gene editing process and minimize growth inhibition or toxicity comprising the steps of

-providing one or more compounds or one or more combinations of compounds,

-determining the efficiency of and/or the ratio between integrating events (antibiotic resistant colony numbers) and inactivating events (toxin resistant colony numbers) with the method according to any one of claims 1 to 19, optionally in the absence of said compound or combination of compounds,

-determining the efficiency of and/or the ratio between integrating events (antibiotic resistant colony numbers) and inactivating events (toxin resistant colony numbers) separately/individually for each of said compounds or combinations of compounds in the presence of said compounds or combinations of compounds with the method according to any one of claims 1 to 19 at different concentrations and/or addition time points,

-identifying/selecting for each compound or combination of compounds of said at least one compound or combination of compounds a concentration and/or a point in time of addition, respectively, wherein said concentration and/or point in time of addition has a higher efficiency and/or a higher ratio compared to the efficiency and/or ratio when carrying out the method as reported herein in the absence of said compound or combination of compounds.

23. The method according to any one of claims 21 to 22, wherein the compounds with the highest efficiency and/or highest ratio are identified/selected.