CN118166081A - A method for detecting off-target effects of a gene editing system - Google Patents

Abstract

The present disclosure relates to a method (OliTag-seq) for detecting off-target of a gene editing system, the method comprising: integrating double-stranded oligodeoxynucleotide dsODN into a double-stranded break site of genomic DNA of a cell, amplifying a portion of genomic DNA comprising the integrated dsODN, sequencing the amplified portion of genomic DNA, thereby detecting a double-stranded break in genomic DNA of the cell; wherein the length of the double-stranded oligodeoxynucleotide dsODN is 35 bp-200 bp, and 1-3 GC DNA pairs are respectively contained at the 3 '-end and the 5' -end of the dsODN. Compared with the existing method, the method has higher sensitivity to the identification of the off-target site, and has wide application prospects in the fields of disease diagnosis and risk assessment, clinical research and drug development, transgenic biotechnology and the like.

Description

Off-target detection method for gene editing system

Technical Field

The present disclosure relates to the field of molecular biology, in particular, the disclosure relates to a method for detecting off-target of a sensitive gene editing system (OliTag-seq).

Background

Double strand breaks in genomic DNA of cells, particularly those caused by off-target cleavage of CRISPR/Cas gene editing systems, are a non-negligible problem in research. It is therefore desirable to perform a detection assessment of double strand breaks to understand and reduce the risk in the study. For double strand breaks in genomic DNA generated by off-target effect, the possible off-target sites are predicted and sequenced mainly by computer software in the prior art, and then the top-ranked sites are subjected to targeted amplification and then deep sequencing to verify whether the Cas nuclease is subjected to off-target cleavage. However, the method has great bias based on the number of mismatched bases with sgrnas, and cannot determine whether Cas is cleaved at other sites not predicted by a computer. Other evaluation methods, such as whole genome sequencing of edited single cell clones, can identify off-target sites without bias, but the method is costly, requires comprehensive analysis of the genomes of many single cell clones, and has great workload in practical research and no practical operability. T7 endonuclease I (T7E 1) assays can also be used for detection of off-target sites, but have low sensitivity and cannot detect off-target sites with mutation frequencies less than 1%, and similarly, large-scale assays are also expensive. In addition, chIP-seq has also been used to identify off-target binding sites for catalytically inactive dCas9, but most published research data indicate that the method recognizes only a few of the resulting sites as true Cas9 off-target cleavage sites. Therefore, it is critical to develop an unbiased, highly sensitive detection method that can analyze off-target sites across the genome.

A variety of cell-based methods (GUIDE-seq, iGUIDE, etc.) and in vitro cleavage methods (Digenome-seq, CIRCEL-seq, CHANGE-seq, etc.) have been published to quantify the Cas-induced genome-wide off-target sites. In vitro methods first require obtaining the complete genome of the cell and then cleaving the genome in vitro with Cas9/sgRNA complexes, generally exhibit higher sensitivity than cell-based methods. However, the chromosomal accessibility and epigenetic modification of genomic DNA in cells is complex compared to gDNA isolated in vitro, and thus the identification of off-target sites by in vitro methods requires further verification in cells or organisms, and as a result, it has been shown that there are many off-target sites without evidence that mutations actually occur in cells. In addition, the in vitro method also brings great additional burden to the subsequent verification work.

In cell-based methods, the GUIDE-seq is most commonly used, which is based on the mechanism of NHEJ repair of DNA double strand breaks, introducing a double-stranded oligonucleotide sequence (dsODN) of 34bp in length at the Cas 9-mediated DNA double strand break for labelling. Then extracting gDNA and randomly cutting into short fragments with the length of about 500bp, then carrying out terminal repair, adding A tail at the 3' end, adding a connector containing a molecular tag (UMI), amplifying genome sequences at the two ends of dsODN, and carrying out high-throughput sequencing. In GUIDE-seq, recognition of off-target sites depends largely on dsODN insertion efficiency and unbiased amplification of the marker sites. To protect the dsODN from nuclease degradation in the cell, researchers have performed phosphorothioate bond modifications at the ends of the dsODN. However, the 34bp dsODN ends with a and T bases, which may cause ligation instability at the interface, making dsODN insertion incomplete. Studies have shown that SpCas9 is more prone to NHEJ+1 editing, further impeding insertion of dsODNs at the cleavage site. Thus, trying to increase the integration efficiency of the dsODN in the Cas cleavage site will help to further increase the sensitivity of off-target site detection.

Disclosure of Invention

The technical problems to be solved are as follows:

One aspect of the present disclosure is to provide a new method for detecting off-target in a gene editing system (OliTag-seq) against the problem of insufficient sensitivity of the method caused by low integration efficiency of dsODN in the GUIDE-seq method.

Specifically, the inventors creatively extended the two ends of the dsODN with base G, respectively, and the results indicate that this improvement increases the overall and full length insertion efficiency of the dsODN in the Cas cleavage site. Meanwhile, the inventor uses three primers to perform one round of PCR in the sequencing library preparation process for the first time, so that the time and the experiment cost are greatly saved. In addition, the inventors also verify that chromatin status affects detection of off-target sites, and found that ipscs exhibit higher sensitivity in off-target site analysis. Finally, the inventors further increased off-target detection capability of cells by constitutive Cas expression and HDAC inhibitors. Using the optimized approach, the inventors analyzed the cleavage specificity of Cas9 at several CAR-T cell therapy-related gene loci and found that the optimized approach detected multiple new off-target sites compared to published data. Through the above research, the problems existing in the prior art are solved.

The technical scheme provided by the disclosure is as follows:

a method (OliTag-seq) for detecting off-target in a gene editing system, the method comprising:

Integrating double-stranded oligodeoxynucleotide dsODN into a double-stranded break site of genomic DNA of a cell, amplifying a portion of genomic DNA comprising the integrated dsODN, sequencing the amplified portion of genomic DNA, thereby detecting a double-stranded break in genomic DNA of the cell;

Wherein the length of the double-stranded oligodeoxynucleotide dsODN is 35 bp-200 bp, and 1-3 GC DNA pairs are respectively contained at the 3 '-end and the 5' -end of the dsODN.

In certain embodiments of the present disclosure, the double-stranded oligodeoxynucleotide dsODN described above can be 35bp, 37bp, 39bp, 41bp, 45bp, 50bp, 60bp, 70bp, 90bp, 120bp, 150bp, 180bp, 200bp in length. To achieve better results, in one embodiment of the present disclosure, the double-stranded oligodeoxynucleotide dsODN is 39bp in length.

In the present disclosure, the above GC DNA pairs may be equally or unequally distributed at both ends of the dsODN. Preferably, in certain embodiments of the present disclosure, one end of the dsODN is two GC DNA pairs and the other end of the dsODN is three GC DNA pairs.

To better achieve certain objects of the present disclosure, e.g., save time and test costs for gene amplification, in certain embodiments of the present disclosure, the methods of amplification may include:

Step 1), after repairing the tail ends of random length fragments integrated with the sequence of the dsODN, respectively connecting the two tail ends with a connector, and performing n rounds of PCR amplification, wherein n is more than or equal to 2;

Step 2) performing a first round of PCR amplification using three primers, wherein primer one is a pair of primers complementary to the sequence of the adaptor, primer two is a primer complementary to the sequence of the dsODN along the 3 'end direction, and primer three is a primer complementary to the sequence of the dsODN along the 5' end direction;

Step 3) carrying out the nth round of PCR amplification by using the three primers to the product of the first round of PCR amplification obtained in the step 2).

Preferably, in certain embodiments of the present disclosure, the molar ratio of primer one, primer two and primer three in step 2) is 2:1:1.

Preferably, in certain embodiments of the present disclosure, the primer in step 3) is provided with a molecular barcode.

In the present disclosure, the off-target may be one resulting from constitutive expression of an endogenous engineered nuclease or activation of an exogenous engineered nuclease for gene editing in the cell.

Constitutive Cas9 expression in cells can promote Indel efficiency as compared to electrotransport Cas9 transient plasmid expression, because the Cas protein is ready to interact with sgrnas to form a complex once the editing element is delivered into the cell. Thus, preferably, in certain embodiments of the present disclosure, the off-target is one resulting from constitutive expression of an endogenous engineered nuclease in the cell.

In the present disclosure, the engineered nuclease may be a meganuclease, zinc finger nuclease, transcription activator effector-like nuclease (TALEN), or CRISPR/Cas RNA Guided Nuclease (RGN), which can result in Double Strand Breaks (DSBs) of genomic DNA in a cell. Preferably, in certain embodiments of the present disclosure, the engineered nuclease is an engineered nuclease of a CRISPR/Cas system; more preferably, in certain embodiments of the present disclosure, the engineered nuclease is a Cas9 nuclease.

In the present disclosure, the cell may be any suitable cell. In the detection of Cas9 off-target sites, researchers typically use Cas9/sgRNA expression plasmids and dsODN templates for gene delivery to immortalized cell lines, commonly used are U2OS and HEK293T cells. The inherent DNA repair pathways of these cell lines have been deregulated, making efficient insertion of dsodns somewhat easier than other cell types. However, clinically relevant ipscs have a high degree of similarity to Embryonic Stem Cells (ESCs) in transcriptional profile compared to cancer cell lines, and therefore the inventors believe that assessing the off-target effect of Cas nuclease in ipscs may be more convincing. This is true through verification in the presently disclosed embodiments. Thus, in certain embodiments of the present disclosure, the cells are iPSC cells.

In the presently disclosed embodiments, the cleavage specificity of Cas9 at engineered T cell (CAR-T cell) treatment-related gene sites was also verified, and it was found that the methods of the present disclosure detected multiple new off-target sites compared to published data. Thus, in certain embodiments of the present disclosure, the cell is an engineered T cell; preferably, in other embodiments of the present disclosure, the engineered T-cell is an engineered T-cell obtained by induction.

To achieve better technical results, in certain embodiments of the present disclosure, the cells are contacted with an HDAC inhibitor.

The beneficial effects are that:

The method disclosed by the invention can effectively identify the off-target site induced by the gene editing system, and has higher sensitivity to the identification of the off-target site compared with the existing method. The method has wide application prospect in the fields of disease diagnosis and risk assessment, clinical research and drug development, transgenic biotechnology, environmental monitoring and ecological system protection, research and education, biosafety and biological defense, personalized medicine and the like.

Application scenarios of the technical scheme of the disclosure:

1. Disease diagnosis and risk assessment: the off-target detection technology can be used for detecting adverse reactions possibly caused by the gene editing technology in the treatment process, so that a more personalized and safer treatment scheme is provided for patients, and the method is one of safety evaluation of CRISPR/Cas9 mediated cellular genetic engineering. The WGS method can be used for genome sequencing of different individuals of the existing reference sequence (REFERENCE SEQUENCE) species, comparing the whole genome sequencing data of the edited species with the reference genome, comprehensively detecting single nucleotide mutations (SNPs) and indel mutations (InDels) caused by gene editing, and then detecting possible off-target sites by analyzing the unique mutation on the homologous region of sgRNA.

2. Clinical study and drug development: in drug development for gene therapy, it is critical to ensure that the drug acts on the intended gene locus without affecting other non-target genes. Off-target detection techniques can help researchers assess the safety and specificity of drugs, thereby more effectively performing drug screening and optimization. Meanwhile, the method can also be used as a monitoring guide before clinical medicines or gene therapy products are marketed.

3. Transgenic biotechnology: in the agricultural and animal breeding fields, off-target detection techniques can ensure that gene editing tools only modify the intended gene and do not cause unnecessary gene changes, thereby ensuring the safety and stability of transgenic organisms.

4. Environmental monitoring and ecosystem protection: before releasing the genetically edited organisms into the environment, it is necessary to ensure that they do not adversely affect the ecosystem. Off-target detection techniques can help researchers assess the impact of these organisms on the environment.

5. Study and education: in the academy, researchers can use off-target detection techniques to evaluate the effectiveness of their gene editing tools to conduct more accurate experiments. Meanwhile, the technology can also be used as a teaching tool to help students better understand the principle and practice of gene editing.

6. Biosafety and biosafety: off-target detection techniques can increase the level of biosafety.

7. Personalized medicine: through off-target detection techniques, doctors can provide more personalized gene editing treatment schemes for patients, thereby improving treatment effects and reducing side effects.

In summary, with the wide application of gene editing technology in various fields, the importance of off-target detection technology will be increasingly highlighted, and the application scenario will be further expanded.

Dominance analysis of off-target assessment technique:

1. The practicability is strong: can reflect the on-target and off-target conditions of gene editing of CRISPR in eukaryotic cells, so as to evaluate the safety and effectiveness.

2. The detection flux is high: the on-target and off-target conditions of a plurality of samples can be detected at one time, and the actual number of sequencing reads is reduced through optimized label design.

3. The accuracy is high: most of the detection results can be verified.

4. The sensitivity is high: can efficiently detect low-frequency off-target sites and detect off-target mutations as low as 0.1%.

5. The experiment difficulty is low: the operation process is simple and easy to implement.

6. The cell adaptability is strong.

Drawings

FIG. 1 is a flow chart and a result chart of the 39bp ODN promoting the identification of off-target sites in the embodiment of the disclosure, wherein A is an off-target analysis flow chart, cells are first electrotransformed with a plasmid expressing Cas9/sgRNA and a dsODN template, genomic DNA is extracted after 3 days and a library is built for 150PE sequencing; b is GUIDE-seq (34 bp, left panel) and ODN sequence comparison panels used in the method of this example (OliTag-seq, 39bp, right panel), P represents 5' phosphorylation, represents phosphorothioate linkages, and extended GC clamps in 39bp ODNs are circled in boxes; c is a result graph comparing full-length insertion efficiency of 34bp ODN and 39bp ODN at six clinically relevant gene loci, and the calculation method of the integration frequency of the full-length ODN is that reads containing the full-length ODN are divided by reads of total editing events, and the value of the 39bp ODN is normalized by the value of the 34bp ODN; d is a graph of the effect of several small molecule compounds on the integration frequency of NHEJ-like dsODN, M3814 as a negative control due to its NHEJ inhibition properties; e is a result graph of comparing the on-target and off-TARGET READS numbers (n=5) of the EMX1, VEGF3 and TRAC loci obtained by sequencing 34bp ODN or 39bp ODN, and the value of the 39bp ODN is normalized by the value of the 34bp ODN; f is a target-off event result graph obtained by comparing 34bp ODN or 39bp ODN identified at EMX1, VEGF3 and TRAC sites, the value of 39bp ODN is normalized by the value of 34bp ODN, the data are shown as mean value +/-standard deviation, and the P value is obtained by unpaired double-tail Student's t-test calculation;

FIG. 2 is a graph of results demonstrating that the insertion efficiency of the 39bp ODN is higher than that of the 34bp ODN in the embodiment of the disclosure, wherein A is a graph of results comparing the overall insertion efficiency and the full-length insertion frequency of the 34bp ODN and the 39bp ODN, and data in the graph are relative values of the overall editing efficiency and are normalized by the value of the 34bp ODN; b is a result graph comparing the insertion condition of 34bp ODN and 39bp dsODN in DSB, the data are shown as mean value +/-standard deviation, and the P value is obtained through unpaired double-tail Student's t-test calculation;

FIG. 3 is a schematic diagram and a result diagram of PCR in the embodiment of the disclosure, wherein A is a schematic diagram of amplifying genome sequences at two ends of an ODN by nested PCR, gDNA is randomly sheared into fragments of 500-700 bp by ultrasonic disruption, then end repair is performed, A tail is added, then a linker containing a molecular tag (UMI) is connected, and finally a library is built by two rounds of PCR; b is a graph of the results of the first round of PCR using the effect of two primers and three primers on sequencing reads; c is a result graph of the influence of two primers and three primers on the number of off-target sites in the first round of PCR, the data are shown as mean value +/-standard deviation, and the P value is obtained through paired double-tail Student's t-test calculation;

FIG. 4 is a graph of the results of a validation of the method (OliTag-seq) showing good reproducibility in the examples of the present disclosure, where A is a graph of the results of the correlation between the first 20 off-target sites Indels (x-axis) of VEGF3 and dsODN insertion efficiency (y-axis); b is a graph of the results of the correlation between the integration frequency of dsODN at the first 20 off-target sites of VEGF3 (x-axis) and the number of OliTag-seq sequencing reads (y-axis); c is OliTag-seq verification result diagram of technical repeatability; d is a visual result diagram of the OliTag-seq off-target site in three biological repeated experiments of the VEGF3 site, mismatched bases of the off-target site are marked by different colors, and OliTag-seq ready numbers corresponding to the positions are displayed on the right side; e is a result graph of overlapping conditions among off-target sites in three biological repeated experiments of VEGF3 sites, pearson linear regression analysis is carried out in graphs A-C, and a correlation coefficient P value is determined by double-tail t distribution;

FIG. 5 is a graph of the results of a validation of the method (OliTag-seq) in the examples of the present disclosure showing higher sensitivity than GUIDE-seq on off-target site recognition, wherein A, B, C, D is a graph of the results of EMX1 (A), VEGF1 (B), VEGF2 (C) or VEGF3 (D) comparisons OliTag-seq with GUIDEseq in U2OS cells, mismatched bases at off-target sites are labeled with different colors, the number of OliTag-seq reads corresponding to each site is shown on the right, and off-target sites for VEGF1, VEGF2 and VEGF3 are truncated due to insufficient space;

FIG. 6 is a graph of the effect of transient and constitutive expression of Cas9 on Off-target recognition in an embodiment of the disclosure, wherein A, B is a graph comparing the effect of transient and constitutive expression of Cas9 on EMX1 Off-target recognition in an iPSC with the visual result, off-index values have been shown at the bottom; C. d is a visual result graph comparing the effect of Cas9 transient expression and constitutive recognition on VEGF3 Off-target conditions in ipscs, off-index values have been shown at the bottom;

FIG. 7 is a graph of the results of analysis of the specificity of Cas9 nuclease in the CAR-T associated treatment site by method (OliTag-seq) in the examples of the present disclosure, wherein A, B, C, D, E is a graph of the results of visualization of OliTag-seq in the CAR-T associated sites (TRAC-CJ (A), TRAC-MH (B), TRBC-CJ (C), TRBC-MH (D) and PDCD1-CJ (E)), the mismatched bases of the off-target site are labeled with different colors, and the corresponding OliTag-seq reads number for each site is shown on the right. On-target sites are marked with black squares and known off-target sites are marked with open diamonds.

Detailed Description

The invention discloses a method (OliTag-seq) for detecting off-target of a gene editing system, and a person skilled in the art can refer to the content of the text and properly improve the technological parameters. It is to be particularly pointed out that all similar substitutes and modifications apparent to those skilled in the art are deemed to be included in the invention and that the relevant person can make modifications and appropriate alterations and combinations of what is described herein to make and use the technology without departing from the spirit and scope of the invention.

In the present invention, unless otherwise indicated, scientific and technical terms used herein have the meanings commonly understood by one of ordinary skill in the art. Throughout the specification and claims, unless explicitly stated otherwise, the term "comprise" or variations thereof such as "comprises" or "comprising", etc. will be understood to include the stated element or component without excluding other elements or components. The terms "such as," "for example," and the like are intended to refer to exemplary embodiments and are not intended to limit the scope of the present disclosure.

Definition:

GUIDE-seq: GUIDE-seq is a molecular biological technology that allows unbiased detection of gene editing off-target events on DNA caused by CRISPR/Cas and other RNA-guided nucleases in living cells. The principle is that a short double-stranded oligonucleotide (dsODN) is used for marking off-target breakage induced by CRISPR/Cas, then high-throughput sequencing is carried out on a genome region where a tag is located, and off-target sites are determined through bioinformatic analysis. See Tsai S Q, zheng Z, nguyen N T, etc .GUIDE-seq enables genome-wide profiling of off-target cleavage by CRISPR-Cas nucleases[J].Nature Biotechnology,2015,33(2):187–197.

In certain embodiments of the present disclosure, the GUIDE-seq method may specifically include:

The cells are contacted with blunt-ended double-stranded oligodeoxynucleotides dsodns. An exogenous engineered nuclease is expressed or activated in the cell, resulting in a Double Strand Break (DSB). Amplifying a portion of genomic DNA comprising the integrated dsODN. Sequencing the amplified portion of genomic DNA, thereby detecting DSBs in genomic DNA of the cell.

The term "double-stranded oligodeoxynucleotide dsODN" is a molecule consisting of two complementary single-stranded oligodeoxynucleotides (ssODN) that form a double-helical structure by base pairing. dsODN can be used as a marker or primer in molecular biology experiments, and can also be used as a homologous recombination template in gene editing. In the present disclosure, the dsODN is orthologous to the genomic DNA of the cell. In certain embodiments of the disclosure, the end of the dsODN may contain two phosphorothioate bond modifications that may insert the sequence of interest at the DSB. Since repair of non-homologous end joining (NHEJ) is inhibited, the occurrence of indels (indels) is reduced. In certain embodiments of the present disclosure, the length of the dsODN may be 15-200 bp, for example ,15bp、16bp、17bp、18bp、19bp、20bp、21bp、22bp、23bp、24bp、25bp、26bp、27bp、28bp、29bp、30bp、31bp、32bp、33bp、34bp、35bp、36bp、37bp、38bp、39bp、40bp、41bp、42bp、43bp、44bp、45bp、46bp、47bp、48bp、49bp、50bp、60bp、70bp、90bp、120bp、150bp、180bp、200bp. preferably, the length of the dsODN may be 30-200 bp; more preferably, in one embodiment of the present disclosure, the dsODN is 39bp in length. In the present disclosure, 1 to 3 GC DNA pairs are contained at each of the 3 'end and the 5' end of the dsODN. For example, 1 GC DNA pair, 2 GC DNA pairs or 3 GC DNA pairs, the positions of which may be randomly distributed. Preferably, in one embodiment of the present disclosure, one end of the dsODN is two GC DNA pairs and the other end of the dsODN is three GC DNA pairs. The dsODN can be attached to genomic DNA using a suitable method.

The term "amplification" refers to the process of replicating a target sequence with reasonable fidelity using nucleases and the like. Amplification may be accomplished by natural or recombinant DNA polymerase, e.g., taq DNA polymerase, T7 DNA polymerase, etc. Amplification reactions refer to any chemical reaction, including enzymatic reactions, that results in an increase in the copy of the template nucleic acid sequence or transcription of the template nucleic acid. In the present disclosure, the amplification refers to Polymerase Chain Reaction (PCR). Any suitable PCR reaction can achieve the objects of the present disclosure, e.g., nested PCR. Nested PCR is a variant Polymerase Chain Reaction (PCR) that uses two pairs (rather than one pair) of PCR primers to amplify the entire fragment. The first pair of PCR primers amplified fragments similar to those of conventional PCR. The second pair of primers is called nested primers (because they are inside the first PCR amplified fragment) bound inside the first PCR product, so that the second PCR amplified fragment is shorter than the first one. The advantage of nested PCR is that if the first amplification yields a wrong fragment, the second time primer pairing can be performed on the wrong fragment and the probability of amplification is very low. Thus, amplification by nested PCR is very specific. In certain embodiments of the present disclosure, the inventors utilized improved PCR methods to achieve better results. This method involves a PCR amplification reaction using three primers. Compared with the traditional two-primer tube-separating PCR method, the first round of PCR adopts a three-primer mixing mode, so that the library construction process is greatly simplified, and the sensitivity of off-target site identification is further improved. To further achieve a better result, in certain embodiments of the present disclosure, the number of PCR amplification reactions may be at least two times per round. In certain embodiments of the present disclosure, the primer may be provided with a molecular Barcode (Barcode) in the nth round of PCR reactions. The term "molecular Barcode" (Barcode) as used in this disclosure refers to a tag consisting of a set of base sequences arranged in a certain regular order, and is used to represent certain information. Which is unique and identifiable.

The term "constitutive expression" refers to stable expression without the induction of other factors, without the time, place, or environment of gene expression being affected.

The term "CRISPR/Cas system" is one of the ways in which DNA Double Strand Breaks (DSBs) are caused in the present disclosure. One or more components of the CRISPR/Cas system may be used as nucleotide binding components in the system. The nucleotide binding molecule may be a Cas protein, a fragment thereof, or a mutant form thereof. Cas proteins may have reduced or no nuclease activity, have (i) programmable binding to a nucleic acid of a target DNA, and (ii) the ability to nick a target DNA sequence on one strand. Cas/Cas protein/Cas nucleases in this disclosure include CRISPR CAS protein and variants thereof, as well as equivalents of Cas9 such as Cas12a(Cpf1)、Cas12e(CasX)、Cas12b1(C2c1)、Cas12b2、Cas12c(C2c3)、C2c4、C2c8、C2c5、C2c10、C2c9Cas13a(C2c2)、Cas13d、Cas13c(C2c7)、Cas13b(C2c6) and Cas13 b. In one embodiment of the present disclosure, preferably is a Cas9 nuclease.

The term "iPSC" or "induced pluripotent stem cell" is used interchangeably and refers to a pluripotent stem cell that is artificially derived (e.g., induced or by complete reversal) from a non-pluripotent stem cell (typically an adult cell), e.g., by inducing forced expression of one or more genes.

The term "engineered T cell" refers to a T cell that has been genetically modified by human intervention, such as by recombinant DNA methods or viral transduction methods. These T cells include T helper cells, cytotoxic T cells (CTL T cells), natural killer T cells, regulatory T cells, memory T cells, or γδ T cells.

The term "HDAC" refers to histone deacetylases (histone deacetylase, HDAC) which are a class of proteases that play an important role in structural modification of chromosomes and regulation of gene expression. In general, acetylation of histones facilitates dissociation of DNA from the histone octamers, relaxing the nucleosome structure, allowing specific binding of various transcription factors and co-transcription factors to DNA binding sites, activating transcription of genes. Whereas deacetylation of histones plays an opposite role. The HDAC described in this disclosure may be any human HDAC isoform, including HDAC1, HDAC2, HDAC3, HDAC4, HDAC5, HDAC6, HDAC7, HDAC8, HDAC9, HDAC10, and HDAC11. An example of an "HDAC inhibitor" may be ,Mocetinostat(MGCD0103)、BRD73954、LMK-235、MC2590、Vorinostat、Ebselen oxide、HDAC-IN-52、HDAC-IN-47、HDAC6-IN-14、HDAC6-IN-13、FNDR-20123、AES-135、CRA-026440、TMP195、TMP269、MC1742、CAY10603(BML-281)、Vorinostat-d5(SAHA-d5)、Romidepsin(FK 228)、Quisinostat(JNJ-26481585),, etc.

Examples:

In order to enable those skilled in the art to better understand the technical solution of the present invention, the present invention will be further described in detail with reference to specific embodiments.

Example 1: optimization of the original GUIDE-seq method by increasing the labelling efficiency of dsODN at double strand break site

The flow chart of the method of the embodiment is shown in A of FIG. 1.

Recognition of off-target sites is largely dependent on the labeling of the dsODN at the double strand break. In the initial GUIDE-seq experimental procedure, the investigator extended the 34bp dsODN to 39bp using blunt-ended, 5' -phosphorylated, end-AT-base-rich 34bp dsODN (see left panel of FIG. 1B), with GC DNA clamps AT both ends (right panel of FIG. 1B). We compared these two dsODNs In six gene loci (EEF 1, BCL11a, ALB-In13, COL1A1, ITG2B and GAPDH 1), and showed that the overall insertion efficiency of 39bp ODN was improved by 1.4-fold (see A of FIG. 2). Considering that adding modifications to the ODN tag would decrease the binding force between the primer and ODN, we also used seqkit command line design algorithm to quantify the integration frequency of the full length ODN, and found that the full length insertion efficiency of 39bp ODN was 1.3-2 times higher than 34bp (see C of fig. 1 and a of fig. 2). Interestingly, however, we also observed that both forms of dsODN could integrate at Cas 9-induced DSBs in tandem repeats (see B of fig. 2) at varying proportions of 0-10%, these tandem ODN repeat insertions resulted in 60-100 bp of non-target PCR products, but we could remove these bands by bead purification prior to the second round of PCR.

Since dsODN is integrated into DSB in a manner similar to NHEJ, we also tested several small molecules that have been reported to increase their insertion efficiency. M3814 was used as a negative control as it can significantly reduce NHEJ editing efficiency. Of the five small molecule compounds tested, B02 showed the most pronounced effect, increasing the dsODN integration efficiency by a factor of 1.8 (see D of fig. 1). Mirin and VE-822 increased dsODN insertion efficiency by 1.5 and 1.3 fold, respectively, while RS1 and cyclosporin H (CsH) had no significant effect on dsODN insertion. Consistent with our published data, M3814 reduced the dsODN insertion frequency by a factor of about 3. We used B02 to promote integration of dsODN.

We then investigated whether the 39bp dsODN would affect the sequencing reads and the recognition of off-target events. We electrotransformed K562 cells with Cas9 expressing plasmid, EMX1, VEGF3 and TRAC targeting sgRNA plasmid, 34 or 39bp ODN template. Since the sequence of 34bp ODN is also contained in 39bp ODN, we can selectively amplify the genomic fragment at both ends of the cleavage site marked by the ODN sequence using the same primers. The results showed that 39bp ODN slightly increased the number of sequencing reads, but there was no significant difference from 34bp dsODN (see E of FIG. 1), but 39bp ODN identified significantly higher off-target events than 34bp ODN (see F of FIG. 1). These results indicate that 39bp ODN can produce higher insertion efficiency and more stable full-length integration, ultimately increasing sensitivity for off-target site recognition.

Example 2: the first round of PCR is carried out by adopting a three-primer mixing mode, so that the library establishment flow is simplified

In the traditional GUIDE-seq library construction process, researchers mostly perform nested PCR to specifically amplify genomic sequences at both ends of ODN, and the adaptor-ligated starting DNA template needs to be divided into two tubes to amplify sequences upstream and downstream of the ODN integration site, respectively. In this example, we performed the first round of PCR using a three primer mix, where one primer was complementary to the linker region and the other two were complementary to the ODN in a 2:1:1 ratio (see FIG. 3A). Three primer mix PCR is expected to yield three products, including upstream and downstream sequences of the ODN marker locus, and products comprising both full length ODN and upstream and downstream genomic sequences, with adaptors at both ends. In the second round of PCR, we used a primer with a barcode (barcode) to enrich for sequences flanking the ODN integration site. Hybrid PCR showed no significant difference in total sequencing reads compared to the traditional PCR method, but unexpectedly identified more off-target sites (see B of fig. 3 and C of fig. 3). Therefore, compared with the traditional two-primer tube-splitting PCR method, the first round of PCR adopts a three-primer mixing mode, so that the library construction process is greatly simplified, and the sensitivity of off-target site identification is improved.

The specific experimental method comprises the following steps:

1. Joint annealing

OliTag-53Phos-Adap-R (primer):

/5Phos/GGCGTGGGTGC*C*A/3Phos/

OliTag-UMI-Adap-F (primer):

AAGGAGGTGACCCTTGGAGAGCTGCCCAATCTGNNNNNNVHGCTCCC TCGCC*T

TABLE 1 Joint annealing procedure

And (3) annealing: after one second at 95℃for one second at 60℃and finally slowly (about-0.05℃per second) cooling to 15 ℃. The annealed linker was stored at-20 ℃.

Alternative procedure: joint annealing procedure: after one second at 95℃for one second at 60℃and finally slowly (approximately-2℃per minute) cooling to 4 ℃.

2. DNA disruption

The genomic DNA after electrotransformation was extracted after 72 hours using Gentra Puregene Blood Kit (Qiagen) and the concentration was measured with Qubit fluorimetry (Invitrogen). 1 microgram of genomic DNA fragments 500-700bp were fragmented in a 130 microliter TE buffer system using Covaris S220, following the procedure:

·Peak power:105Watts

·Duty factor:5％

·Cycles/Burst:200

·Treatment time:80seconds

run the gel with 2% agarose gel for identification.

Alternative procedure: DNA was broken and eluted with 15. Mu.l of 1 XTE buffer after 1 Xpurification using AMPure XP SPRI magnetic beads.

3. Purification of disrupted DNA

After 30 minutes of magnetic bead room temperature incubation, using 1.5 x magnetic beads and crushed genomic DNA room temperature for 2 minutes, placing on a magnetic rack for 3-5 minutes, discarding supernatant, washing twice with 200 microliter of DNA washing liquid, finally adding 52 microliter of nuclease-free water to de-resuspend the magnetic beads, placing on the magnetic rack after 2 minutes of room temperature incubation, transferring the clarified supernatant to a new tube. Identification was performed using 1. Mu.l of DNA running gel.

Alternative procedure: the end repair of the Taq enzyme system comprises the following specific procedures: 15 minutes at 12 ℃; 15 minutes at 37 ℃; 15 minutes at 72 ℃.

4. DNA end repair and addition of A tail

TABLE 2 repair of Ends and addition of A tail

After thoroughly mixing, the mixture was reacted at 20℃for 30 minutes and at 65℃for 30 minutes.

5. Joint for connecting pipe

TABLE 3 grafting reactions

After thoroughly mixing, the mixture was reacted at 20℃for 15 minutes. 1.25 Xmagnetic bead purification, elution with 17. Mu.l nuclease free water. Identification was performed using 1. Mu.l of DNA running gel.

Alternative procedure: and (3) adding a joint: t4 polymerase system, the specific procedure is: 16℃for 30 minutes and 22℃for 30 minutes.

6. PCR amplification

One round of PCR:

TABLE 4 round of PCR procedure

PCR program ：98℃ 1min;8cycles of[98℃ for 10s,68℃(-1℃/cycle)for 15s,72℃ for 30s];22cycles of[98℃ for 10s,60℃ for 15s,72℃ for 30s].Ramp from98℃ to 60℃:-1℃/sec.

The first round of PCR products were purified using 1.8Xmagnetic beads and used for the second round of PCR.

Two rounds of PCR:

TABLE 5 two round PCR procedure

PCR program ：98℃ 1min;8cycles of[98℃ for 10s,68℃(-1℃/cycle)for 15s,72℃ for 15s];18cycles of[98℃ for 10s,60℃ for 15s,72℃ for 15s].Ramp from98℃ to 60℃:-1℃/sec.

Alternative procedure: PCR program ：95℃5min;15cycles of[95℃ for 30s,70℃(-1℃/cycle)for 2min,72℃ for 30s];10cycles of[95℃ for 30s,55℃ for 1min,72℃ for 30s].

TABLE 6OliTag-seq primer

7. Illumina sequencing

All products were mixed and submitted to 150bp modified-end Illumina sequencing (Novogene)

Example 3: the optimized method (OliTag-seq) of the embodiment can effectively identify off-target sites induced by CRISPR-Cas9

To verify whether the optimized OliTag-seq method is able to effectively recognize the cleavage site induced by Cas9, we used the VEGF3 site, which has been widely studied for off-target conditions, and electrotransformed K562 using Cas9 and sgRNA expression plasmids, and extracted gDNA 72 hours after electrotransformation for off-target analysis. We then amplified the first 20 off-target sites specifically using primers with barcode for Illumina high-throughput sequencing to verify cleavage at these sites, we obtained the total editing frequency and corresponding dsODN insertion efficiency by CRISPResso2 analysis. We found that the Indel efficiency varied from 0 to 80% among these 20 off-target sites, with the corresponding dsODN insertion efficiency ranging from 0 to 26%, while some sites showed off-target in OliTag-seq but no Indel evidence in Illumina sequencing, possibly due to systematic error limitations of the Illumina sequencing technique itself. We observed a strong positive correlation between dsODN insertion efficiency and Indel (r2=0.85, see a of fig. 4), and more importantly, the number of sequencing reads of OliTag-seq also showed a clear positive correlation with dsODN integration frequency (r2=0.69, see B of fig. 4). Thus, the efficiency of Cas9 cleavage corresponding to each site of OliTag-seq can be reflected indirectly by the number of sequencing reads at that site. We also investigated the technical and biological reproducibility of OliTag-seq and found that the number of sequencing reads from both technical replicates correlated well (R2=0.88, see FIG. 4C). Furthermore, off-target sites with higher sequencing reads occur almost in each independent biological repeat, but off-target sites with reads below 10 are less reproducible (see D in fig. 4 and E in fig. 4). Therefore, in later experiments, to improve data reliability, we generally employed off-target sites that occur in at least two biological replicates.

Example 4: the method of this example (OliTag-seq) shows a higher sensitivity to off-target site recognition than GUIDE-seq

To directly compare OliTag-seq and GUIDE-seq, we evaluated the specificity of four sgrnas targeting EMX1, VEGF2 and VEGF3 that have been studied before in U2OS cells. OliTagseq results show that the off-target conditions of different sgrnas in the genome are different, and the off-target sites are different from 15 to 209 (see a of fig. 5 to D of fig. 5). For VEGF1, oliTag-seq detected all off-target sites that were previously present in GUIDE-seq. While for other sgrnas, the OliTag-seq also detected the vast majority of off-target sites previously identified by the GUIDE-seq method. Importantly, for all sgrnas, the OliTag-seq detected more off-target sites than the GUIDE-seq, while we also observed that almost all overlapping off-target sites in both methods had a higher number of sequencing reads, while those found only by OliTag-seq were much lower. We next further investigated whether the new off-target site identified by OliTag-seq was indeed mutated in U2OS cells. We selected four representative off-target sites found only in OliTag-seq in VEGF1, and targeted amplification sequencing showed dsODN integration at all four new off-target sites with an insertion efficiency of 0.05% -3.96%. Taken together, oliTag-seq is superior to the original GUIDE-seq approach in identifying CRISPR-Cas9 off-target sites.

Example 5: constitutive Cas9 expression further improves off-target recognition capability of iPSC

Constitutive Cas9 expression in cells can promote Indel efficiency compared to electrotransport Cas9 transient plasmid expression, because Cas9 proteins are ready to interact with sgrnas to form complexes once the editing element is delivered into the cell. Furthermore, sustained expression of Cas9 also creates more potential off-target cleavage in sites that are difficult to edit. We compared both transient and constitutive Cas9 expression systems, the results indicate that efficient cleavage is shown in both EMX1 and VEGF3, but the constitutive Cas9 system appears to be able to induce more off-target cleavage events in both sites (see a of fig. 6 and C of fig. 6). To more accurately compare the off-target effects of the two systems, we calculated the off-index values for both, showing that the off-index of the constitutive Cas9 system is almost 2 times that of the transient Cas9 system (see B of fig. 6 and D of fig. 6). Thus, it is suggested to use a constitutive Cas9 expression system in performing off-target site recognition.

Example 6: analysis of Cas9 specificity in CAR-T related sites Using the method (OliTag-seq) of the present example

Off-target analysis is necessary for preclinical studies of CRISPR-Cas9 gene editing techniques, especially for patient-derived Cas9 engineered T cells. Using the optimized OliTag-seq method, we analyzed the specificity of Cas9 nucleases at CAR-T treatment-related sites (TRAC-CJ, TRAC-MH, TRBC-CJ, TRBC-MH, and PDCD 1-CJ), where off-target events for TRAC-CJ, TRBC-CJ, and PDCD1-CJ sites have been reported by the iGUIDE method. To fairly compare the results of the two methods, we followed the iGUIDE data filtering criteria, allowing no more than six mismatched bases in the target sequence or PAM sequence.

We observed that the number of off-target sites of these 5 sgRNAs detected by OliTag-seq varies from 4 to 2 (see FIG. 7A-7E). In contrast to the TRAC-CJ, TRBC-CJ and PDCD1-CJ sgRNA results reported earlier by iGUIDE, only 6 of these off-target sites were detected by OliTag-seq and off-target sites with a number of reads below 10 in a number of iGUIDE were not identified by the OliTag-seq method. However, oliTag-seq also detected a different number of new off-target sites.

Of the five investigated sgrnas, we detected more off-target mutations in TRAC-MH than at the other sites (see B of fig. 7). Sgrnas targeting TRBC-MH are most specific, since the least off-target events were identified (see D of fig. 7), but since the number of off-target events reads at this site is comparable to on-target, negative effects are expected. For PDCD1-CJ sgRNA, the abundance of off-target mutations was minimal, with the number of on-TARGET READS being 100-fold greater than the number of off-TARGET READS (see E in FIG. 7).

The foregoing is merely a preferred embodiment of the present invention and it should be noted that modifications and adaptations to those skilled in the art may be made without departing from the principles of the present invention, which are intended to be comprehended within the scope of the present invention.

Claims (10)

1. A method (OliTag-seq) for detecting off-target in a gene editing system, the method comprising:

Integrating double-stranded oligodeoxynucleotide dsODN into a double-stranded break site of genomic DNA of a cell, amplifying a portion of genomic DNA comprising the integrated dsODN, sequencing the amplified portion of genomic DNA, thereby detecting a double-stranded break in genomic DNA of the cell;

Wherein the length of the double-stranded oligodeoxynucleotide dsODN is 35 bp-200 bp, and 1-3 GC DNA pairs are respectively contained at the 3 '-end and the 5' -end of the dsODN;

Preferably, the double-stranded oligodeoxynucleotide dsODN is 39bp in length.

2. The method of claim 1, wherein one end of the dsODN is two GC DNA pairs and the other end of the dsODN is three GC DNA pairs.

3. The detection method according to claim 1 or 2, wherein the amplification method comprises:

Step 1), after repairing the tail ends of random length fragments integrated with the sequence of the dsODN, respectively connecting the two tail ends with a connector, and performing n rounds of PCR amplification, wherein n is more than or equal to 2;

Step 2) performing a first round of PCR amplification using three primers, wherein primer one is a pair of primers complementary to the sequence of the adaptor, primer two is a primer complementary to the sequence of the dsODN along the 3 'end direction, and primer three is a primer complementary to the sequence of the dsODN along the 5' end direction;

Step 3) carrying out the nth round of PCR amplification by using the three primers to the product of the first round of PCR amplification obtained in the step 2).

4. The method according to claim 3, wherein the molar ratio of primer one, primer two and primer three in step 2) is 2:1:1.

5. The method according to claim 3, wherein the primer in step 3) has a molecular barcode.

6. The method of claim 1, wherein the off-target is due to off-target by constitutive expression of an endogenous engineered nuclease in the cell or activation of an exogenous engineered nuclease for gene editing;

preferably, the off-target is due to gene editing by constitutively expressing an endogenous engineered nuclease in the cell.

7. The detection method of claim 6, wherein the engineered nuclease is an engineered nuclease of a CRISPR/Cas system;

Preferably, the engineered nuclease is a Cas9 nuclease.

8. The method of claim 1, 6 or 7, wherein the cells are iPSC cells.

9. The method of claim 1, 6 or 7, wherein the cell is an engineered cell;

Preferably, the engineered cell is an engineered T cell;

more preferably, the engineered T cell is an engineered T cell obtained by induction.

10. The method of claim 1, wherein the cells are contacted with an HDAC inhibitor.