Method for screening site-directed integration site, vector microparticle, site-directed integration method and microparticle

Abstract

Provided are a method for screening a site-directed integration site, a vector microparticle, a site-directed integration method and a microparticle. Provided are a method for screening a site-directed integration site, a vector microparticle, a site-directed integration method and a microparticle. The objective of the present invention lies in that by integrating a marker gene having recognition sequences upstream and downstream on a vector microparticle, protein expression screening and single clone screening are simultaneously performed so as to obtain a single clone vector having good performance by means of one-step screening, and the marker gene is then replaced when an exogenous gene of interest is expressed so as to greatly ensure the overall stability of the microparticle, which ensure the overall expression performance and stability of the expression microparticle while shortening the development of the expression microparticle, thereby solving the technical problem in the prior art that expression vectors are long in development time or poor in performance. The provided method for screening a site-directed integration site can effectively shorten the development period of expression vectors having high expression and overall stability. The provided vector microparticle has good adaptability, good expression and good passaging stability.

Description

Site-directed integration site screening method, carrier microparticle, site-directed integration method and microparticle Technical Field

[0001] The present invention belongs to the field of biotechnology, and more specifically, relates to a method for screening site-specific integration sites, carrier microparticles, a site-specific integration method, and microparticles. Background Art

[0002] Since the US FDA approved Genetech's recombinant insulin for marketing in 1982, recombinant proteins have been successfully used in scientific research, as well as in the diagnosis, treatment, and prevention of various diseases. Currently, recombinant proteins are primarily produced and expressed using exogenous protein expression systems. Recombinant protein production using mammalian cells has become the mainstream technology for the expression and production of biopharmaceuticals. The majority of biopharmaceuticals currently on the market, in clinical trials, and in preclinical studies, including but not limited to peptides and proteins, can be expressed in mammalian cells. Among these, Chinese Hamster Ovary (CHO) cells are the most widely used host cell for recombinant protein production. There are numerous CHO cell lines. The main CHO cell lines currently used for industrial production include CHO-K1, CHO-DG44, CHO-DXB11 (also known as CHO-DUKX), CHO-S, and CHOZN (CHOZN ZFN-Modified CHO Cell Lines), with CHO-MK currently under development.

[0003] A Chinese invention patent (authorization publication number CN100381573C) discloses a method for constructing a cell line containing an amplifiable high transcriptional activity site in the genome. The method uses an integration marker vector to insert a marker into an amplifiable transcriptional activity site in the mammalian genome. Subsequently, the mammalian cells are co-transfected with a dual-plasmid targeting expression vector and a recombinase expression vector. This allows the exogenous target gene to be integrated into the genomic marker site under the action of a site-specific recombination system. Through gene amplification, the gene copy number is rapidly increased, achieving high gene expression. In this patented scheme, a recombinase is used to recognize and cleave a pair of recombination signal sequences, such as attP and attB, and the cell line selected is CHO-dhfr-.

[0004] A Chinese invention patent (authorization publication number CN114107380A) discloses a method for constructing a cell line using a bacteriophage Bxb1 site-specific recombinase system, which can significantly reduce cell line development time. First, the site-specific integration site attP is randomly integrated into the CHO-S cell genome using a random integration method. Antibiotic screening is then performed to obtain a CHO-S.attP recombinant cell model. Using the CHO-S.attp recombinant cell line as the host cell, DNA carrying the attB site and the target protein gene is site-specifically integrated into the CHO-S.attp cell line with the aid of Bxb1 recombinase. Stable, high expression of the target protein is achieved through antibiotic pressure screening.

[0005] Currently, when constructing cell lines, an exogenous target gene is inserted into a specific site. Through gene amplification, the gene copy number is rapidly increased to achieve high gene expression. Existing screening methods for known sites include mother cell line construction, mother cell line stability screening, site-specific integration screening of the exogenous target gene, and target cell line stability screening. Mother cell line construction is one step, while traditional mother cell line construction uses a two-step process of "mother cell line construction + screening" to select mother cell lines with the marker gene inserted into the site. This method is lengthy and labor-intensive.

[0006] At the same time, when screening the mother cell line, a marker gene needs to be used. The target cell line finally integrated cannot remove the marker gene from the genome, which affects the overall stability of the target cell line.

[0007] Summary of the Invention

[0008] In response to the above-mentioned deficiencies or improvement needs of the prior art, the present invention provides a method for screening site-specific integration sites, vector microparticles, a site-specific integration method, and microparticles. The present invention aims to integrate marker genes with upstream and downstream recognition sequences into vector microparticles, thereby simultaneously screening for protein expression levels during monoclonal screening. After obtaining a monoclonal vector with good performance through one-step screening, the marker gene can be replaced when expressing the exogenous target gene, thereby greatly ensuring the overall stability of the microparticles. This shortens the development time of the expression microparticles while ensuring the overall expression performance and stability of the expression microparticles, thereby resolving the technical problems of the prior art expression vectors, such as the long development time or poor performance.

[0009] To achieve the above object, according to one aspect of the present invention, a method for screening site-specific integration sites is provided, comprising the following steps:

[0010] (1) randomly integrating an exogenous gene containing two recognition sequences and a marker gene between the two recognition sequences into an endogenous gene of a vector particle to obtain a vector particle carrying the marker gene;

[0011] (2) amplifying and culturing the vector particles carrying the marker gene obtained in step (1) to form a vector particle pool to be screened;

[0012] (3) performing monoclonal screening on the vector particle pool to be screened obtained in step (2), and retaining monoclonal vector particles whose expression levels of the corresponding marker genes and / or exogenous target genes are higher than a preset threshold value based on the expression levels of the marker genes and / or exogenous target genes during the monoclonal screening, thereby obtaining monoclonal vector particles with high protein expression;

[0013] (4) Subculture the high-protein expression monoclonal vector particles obtained in step (3), determine the subculture stability of the monoclonal vector particles based on the marker gene expression levels of multiple generations of vector particles, obtain high-expression monoclonal vector particles with subculture stability, and use the insertion position of the exogenous gene on the endogenous gene as the site-specific integration site.

[0014] Preferably, in the method for screening site-specific integration sites, the marker gene is of comparable length to the exogenous target gene to be integrated;

[0015] The marker gene has multiple optional lengths, preferably 0.75k, 1k, 3k, and 15k;

[0016] It is preferred to use marker genes of 3k or less and predict the expression levels of exogenous target genes of other lengths through intelligent learning algorithms.

[0017] Preferably, in the method for screening site-directed integration sites, the marker gene is selected from one or more of a neomycin resistance gene, a thymidine kinase gene, a hygromycin phosphotransferase gene, a dihydrofolate reductase gene, a thymidine kinase gene, a glutamine synthetase gene, an asparagine synthetase gene, a tryptophan synthetase gene, a histidinol dehydrogenase gene, an aminoglycoside phosphotransferase gene, a tryptophan synthetase gene, and a fluorescent protein gene; and / or,

[0018] The carrier particles are used to express one or more of amino acids, polypeptides, proteins, nucleotides, nucleic acids, and secretions, and are selected from cells, bacteria, and viruses, and are commonly carrier cells; the carrier cells are eukaryotic cells; optionally, the eukaryotic cells are mammalian cells; optionally, the mammalian cells include Chinese hamster ovary CHO cells and human embryonic kidney HEK293 cells; and / or,

[0019] The recognition sequence is independently selected from one or more of the following sequences: LoxP sequence, LoxPL3 sequence, LoxP 2L sequence, LoxFas sequence, Lox511 sequence, Lox2272 sequence, Lox2372 sequence, Lox5171 sequence, Loxm2 sequence, Lox71 sequence, Lox66 sequence, FRT sequence, Bxb1 attP sequence, Bxb1attB sequence, attP sequence, attB sequence, attP-GA sequence and attB-GA sequence.

[0020] Preferably, in the method for screening site-specific integration sites, step (3) uses image processing technology to screen monoclonal vector particles, and specific schemes include but are not limited to deep learning target detection models YOLO (You Only Look Once) series and SSD (Single Shot MultiBox Detector) series algorithm models.

[0021] Preferably, the method for screening site-directed integration sites, wherein step (3) predicts the expression level of the marker gene and / or the exogenous target gene using an intelligent algorithm based on the cell image, specifically comprises the following steps:

[0022] (3-1) imaging the pool of vector particles to be screened obtained in step (2) to obtain an image of the vector particles to be screened; preferably, high-throughput scanning imaging is used, and when the marker gene is a fluorescent protein, high-throughput fluorescence scanning imaging is preferably used;

[0023] (3-2) performing image processing on the image of the vector particles to be screened obtained in step (3-1) to obtain an image of each vector particle, and using an intelligent algorithm to predict the expression level of the marker gene and / or the exogenous target gene of the vector particle based on the image features; preferably, the image features used include: cell morphological characteristics and / or optical characteristics, including but not limited to roundness, size, and grayscale of specific cell sites;

[0024] Preferably, the expression levels of the marker genes and / or exogenous target genes of the vector particles predicted by the multiple intelligent algorithms are evaluated using the Normalized Discounted Cumulative Gain (NDCG), an evaluation standard of the ranking algorithm, to evaluate the ranking accuracy of the protein expression levels predicted by the multiple intelligent algorithms. According to the rule that the greater the standard Normalized Discounted Cumulative Gain, the better the ranking accuracy of the protein expression levels, the intelligent algorithm with the best ranking accuracy in predicting the protein expression levels is selected;

[0025] For a specific intelligent algorithm for predicting protein expression, the standard normalized loss cumulative gain calculation formula is as follows:

[0026] Wherein, IDCG is the DCG value of the actual ranking of the protein expression of the sample monoclonal vector particles, i represents the predicted ranking number of a monoclonal vector particle in the sample monoclonal vector particles according to the predicted protein expression of the intelligent algorithm, rel(i) represents the actual protein expression of the i-th monoclonal vector particle; k is the sample size of the sample monoclonal vector particles;

[0027] When the predicted protein is a fluorescent protein, rel(i) indicates the fluorescence value; when the predicted protein is a non-fluorescent protein, rel(i) indicates the value of the protein expression level tested by Western blot, ELISA, mass spectrometry, electrophoresis, etc.

[0028] (3-3) The carrier particles whose expression levels of the marker gene and / or the exogenous target gene exceed a preset threshold are acquired according to image positioning and moved to a preset position in the culture medium for retention.

[0029] Preferably, in the method for screening site-specific integration sites, step (3) employs semi-solid culture medium monoclonal vector particle screening.

[0030] Preferably, in the method for screening site-specific integration sites, in step (4), the passage stability of the monoclonal vector particles is determined using an intelligent algorithm; the specific steps are as follows:

[0031] (4-1) obtaining the marker gene expression levels of the monoclonal vector particles of multiple generations;

[0032] (4-2) evaluating the differences in marker gene expression levels of the multiple generations of monoclonal vector microparticles using an intelligent algorithm based on the marker gene expression levels of the multiple generations of monoclonal vector microparticles obtained in step (4-1);

[0033] (4-3) Judging whether the monoclonal vector microparticles have generation stability based on the differences in marker gene expression levels of the multiple generations of monoclonal vector microparticles obtained in step (4-2) is conducted according to the principle that the smaller the difference in expression levels, the greater the probability of having generation stability.

[0034] According to another aspect of the present invention, there is provided a site-specific integration vector particle obtained by screening using the site-specific integration site screening method, wherein an exogenous gene is inserted into an endogenous gene of the vector particle; the exogenous gene comprises two first recognition sequences and a marker gene expression segment for expressing a marker protein, which is integrated between the two first recognition sequences;

[0035] The expression level of the marker gene of the site-directed integration vector particle is higher than the preset threshold;

[0036] The site-directed integration vector particles have passage stability.

[0037] According to another aspect of the present invention, a system for screening for site-specific integration sites is provided, wherein the system is an electronic device and / or a non-transitory computer-readable storage medium;

[0038] The electronic device includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the steps of the method for screening site-specific integration sites provided by the present invention;

[0039] The non-transitory computer-readable storage medium stores a computer program thereon, and when the computer program is executed by a processor, the steps of the method for screening site-specific integration sites provided by the present invention are implemented.

[0040] According to another aspect of the present invention, a method for site-directed integration of an exogenous target gene is provided, comprising the following steps:

[0041] Obtain the site-directed integration vector particles provided by the present invention, and use the first recognition sequence in the exogenous gene in the site-directed integration vector particles to perform homologous recombination, so as to recombine the exogenous target gene expression fragment into the gene of the site-directed integration vector particles while removing the marker gene expression fragment.

[0042] Preferably, in the method for site-directed integration of the exogenous target gene, the exogenous target gene expression segment is set in a recombinant vector, and the exogenous target gene is site-directedly integrated into the site of the site-directed integration vector particle through the recombinant vector; the recombinant vector is a lentiviral vector, an adenoviral vector, an adeno-associated viral vector, a herpes virus vector, a poxvirus vector, a baculovirus vector, a papillomavirus vector, a papillomatovirus vector, an integrative phage vector, a non-viral vector, a transposon, a transposase, an integrase substrate or a plasmid.

[0043] Preferably, in the method for site-directed integration of the exogenous target gene, the length of the marker gene expression segment of the site-directed integration vector particle is comparable to that of the exogenous target gene expression segment.

[0044] Preferably, the method for site-directed integration of the exogenous target gene, wherein the homologous recombination is performed using the first recognition sequence of the site-directed integration fragment of the site-directed integration vector particle, is specifically as follows:

[0045] Constructing an exogenous target gene recombinant fragment comprising an exogenous target gene; wherein the exogenous gene recombinant fragment contains two second recognition sequences and an exogenous target gene expression fragment; the exogenous target gene expression fragment is between the two second recognition sequences; the exogenous target gene expression fragment includes a promoter for regulating expression of the exogenous target gene and an exogenous target gene sequence;

[0046] The exogenous gene is integrated into the site of the site-directed integration vector particle, and the two first recognition sequences on both sides of the marker gene are respectively changed into a third recognition sequence and a fourth recognition sequence; the third recognition sequence is formed by the reaction of the first recognition sequence and the second recognition sequence, and the fourth recognition sequence is formed by the reaction of the first recognition sequence and the second recognition sequence, and the third recognition sequence is different from the fourth recognition sequence.

[0047] Preferably, in the method for site-directed integration of the exogenous target gene, the third recognition sequence and the fourth recognition sequence will not become the second recognition sequence or the first recognition sequence in the same reaction system.

[0048] Preferably, in the method for site-directed integration of the exogenous target gene, the third recognition sequence and the fourth recognition sequence are generated by a recombinase, and the recombinase is one of Bxb1 integrase, ΦC31 integrase, Cre recombinase or FLP recombinase; and / or

[0049] The second recognition sequence is independently selected from one or more of the following sequences: LoxP sequence, LoxPL3 sequence, LoxP 2L sequence, LoxFas sequence, Lox511 sequence, Lox2272 sequence, Lox2372 sequence, Lox5171 sequence, Loxm2 sequence, Lox71 sequence, Lox66 sequence, FRT sequence, Bxb1 attP sequence, Bxb1attB sequence, attP sequence, attB sequence, attP-GA sequence and attB-GA sequence; and / or,

[0050] The third recognition sequence and the recognition sequence are independently selected from one or more of the following sequences: LoxP sequence, LoxPL3 sequence, LoxP 2L sequence, LoxFas sequence, Lox511 sequence, Lox2272 sequence, Lox2372 sequence, Lox5171 sequence, Loxm2 sequence, Lox71 sequence, Lox66 sequence, FRT sequence, attL sequence and attR sequence.

[0051] Preferably, the method for site-directed integration of the exogenous target gene comprises the following steps:

[0052] S1. Obtaining a plurality of site-directed integration vector particles provided by the present invention, performing homologous recombination using the first recognition sequence in the exogenous gene within the site-directed integration vector particles, recombining the exogenous target gene expression fragment into the gene of the site-directed integration vector particle while removing the marker gene expression fragment, thereby obtaining a plurality of candidate exogenous target gene expression vector particles;

[0053] S2, amplifying and subculturing the multiple candidate exogenous target gene expression vectors obtained in step S1;

[0054] S21. Retaining, based on the expression level of the exogenous target gene, monoclonal exogenous target gene expression vector particles whose expression level of the corresponding exogenous target gene is higher than a preset threshold;

[0055] S22 determines the passage stability of the exogenous target gene expression vector microparticles based on the exogenous target gene expression level of multiple generations of monoclonal exogenous target gene expression vector microparticles;

[0056] Obtain exogenous target gene expression vector microparticles with high protein expression and stable passage.

[0057] Preferably, in the method for site-directed integration of an exogenous target gene, step S21 predicts the expression level of the exogenous target gene by an intelligent learning algorithm;

[0058] The expression level of the exogenous target gene is retained and predicted by an intelligent algorithm based on the cell image; specifically comprising the following steps:

[0059] S211, imaging the monoclonal exogenous target gene expression vector microparticles obtained in step S21 to obtain an image of the candidate monoclonal exogenous target gene expression vector microparticles; preferably, high-throughput scanning imaging is used;

[0060] S212. Predicting the expression level of the exogenous target gene in the monocloned exogenous target gene expression vector microparticles using an intelligent algorithm based on image features; preferably, the image features used include: cell morphological features and / or optical features, including but not limited to roundness, size, and grayscale of specific cell sites;

[0061] Preferably, the expression levels of the exogenous target gene of the monoclonal exogenous target gene expression vector microparticles predicted by multiple intelligent algorithms are evaluated using the Normalized Discounted Cumulative Gain (NDCG), an evaluation standard of the ranking algorithm, to evaluate the ranking accuracy of the protein expression levels predicted by the multiple intelligent algorithms. According to the rule that the larger the standard Normalized Discounted Cumulative Gain, the better the ranking accuracy of the protein expression level, the intelligent algorithm with the best ranking accuracy in predicting the protein expression level is selected;

[0062] For a specific intelligent algorithm for predicting protein expression, the standard normalized loss cumulative gain calculation formula is as follows:

[0063] Wherein, IDCG is the DCG value of the actual ranking of the protein expression of the sample monoclonal vector particles, i represents the predicted ranking number of a monoclonal vector particle in the sample monoclonal vector particles according to the predicted protein expression of the intelligent algorithm, rel(i) represents the actual protein expression of the i-th monoclonal vector particle; k is the sample size of the sample monoclonal vector particles;

[0064] When the predicted protein is a fluorescent protein, rel(i) indicates the fluorescence value; when the predicted protein is a non-fluorescent protein, rel(i) indicates the value of the protein expression level tested by Western blot, ELISA, mass spectrometry analysis, electrophoresis, etc.

[0065] Preferably, in the method for site-directed integration of the exogenous target gene, in step S22, the passage stability of the monoclonal exogenous target gene expression vector microparticles is determined using an intelligent algorithm; the specific steps are as follows:

[0066] S221, obtaining the expression levels of the exogenous target gene of multiple generations of the monoclonal exogenous target gene expression vector microparticles;

[0067] S222, using an intelligent algorithm to evaluate the difference in the expression levels of the exogenous target gene among the multiple generations of monoclonal exogenous target gene expression vector microparticles obtained in step S221;

[0068] S223 determines whether the monoclonal exogenous target gene expression vector microparticles have passage stability based on the difference in exogenous target gene expression levels of the multiple generations of monoclonal exogenous target gene expression vector microparticles obtained in step S222, according to the principle that the smaller the difference in expression levels, the greater the probability of having passage stability.

[0069] According to another aspect of the present invention, there is provided an exogenous target gene expression vector particle obtained by screening the exogenous target gene site-directed integration method, wherein the exogenous target gene is inserted into the endogenous gene of the vector particle; the exogenous gene includes a third recognition sequence, a fourth recognition sequence, and an exogenous target gene expression segment disposed between the third recognition sequence and the fourth recognition sequence for expressing a target protein;

[0070] The target gene expression level of the exogenous target gene expression vector microparticle is higher than the preset threshold value;

[0071] The exogenous target gene expression vector microparticles have passage stability.

[0072] According to another aspect of the present invention, a system for site-directed integration of an exogenous target gene is provided, wherein the system is an electronic device and / or a non-transitory computer-readable storage medium;

[0073] The electronic device includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, the steps of the method for site-directed integration of an exogenous target gene provided by the present invention are implemented.

[0074] The non-transitory computer-readable storage medium stores a computer program, which, when executed by a processor, implements the steps of the method for site-directed integration of an exogenous target gene provided by the present invention.

[0075] In general, the above technical solutions conceived by the present invention can achieve the following beneficial effects compared with the prior art:

[0076] The present invention achieves two effects by constructing an exogenous gene containing two recognition sequences and a marker gene between the two recognition sequences onto the endogenous gene of the vector particle: on the one hand, single clone screening and expression level screening are completed simultaneously, and a high-expression monoclonal vector is obtained in one step; on the other hand, the marker gene is directly replaced when the exogenous target gene is expressed, without the need to load the marker gene in the final protein expression vector, thereby improving the overall stability of the cell. Combining the above two effects, the method for screening site-specific integration sites of the present invention can effectively shorten the development cycle of expression vectors with high expression levels and overall stability.

[0077] The carrier microparticles provided by the present invention have good adaptability to different exogenous target genes and good expression levels and passage stability.

[0078] The present invention covers the entire process of cell line selection, from the construction of the plasmid of the target protein to the final determination of the protein stable and high-yielding cell line. The entire process is standardized, repeatable and highly operable.

[0079] The preferred solution uses artificial intelligence algorithms to evaluate and predict the performance of monoclonal expression vectors, which increases the probability of screening monoclonal expression vectors with good performance in a single culture, and further shortens the development cycle of expression vectors with targeted integration sites overall.

[0080] In a preferred approach, the present invention uses high-throughput microscopic scanning of cell imaging data to screen for monoclonal cell lines that highly express the target protein. A robotic arm semi-automatically aspirates and transfers the selected cell lines, and finally, an image recognition algorithm is used to select stable expressing cell lines. Compared to other techniques for screening monoclonal cell lines, this method offers the advantages of higher accuracy and reduced time and effort. BRIEF DESCRIPTION OF THE DRAWINGS

[0081] FIG1 is a schematic structural diagram of a recombinant plasmid comprising RMCE constructed in Example 1 of the present invention;

[0082] FIG2 is a schematic diagram of the structure of the marker gene expression fragment contained in the recombinant plasmid provided in Example 1 of the present invention;

[0083] FIG3 is a flow chart of the method for screening fixed integration sites provided in Example 1 of the present invention.

[0084] FIG4 is a diagram showing the detection effect of the target detection algorithm provided in Example 1 of the present invention;

[0085] FIG5 is an operation flow chart provided by Example 2 of the present invention;

[0086] Figure 6 is a recombinant plasmid of target genes 1, 2, 4, and 5 provided in Example 2 of the present invention;

[0087] FIG7 is a recombinant plasmid of target gene 3 provided in Example 2 of the present invention;

[0088] FIG8 is a graph showing the expression level comparison of the expression vector of Example 2 on day 14;

[0089] FIG9 is a graph showing the expression levels of the expression vectors in Example 2 over 7 days;

[0090] Figure 10 is a photograph of electrophoresis of GBB003-2D8 cells into which target gene 1 was integrated in Example 2;

[0091] Figure 11 is a photograph of electrophoresis of GBB003-1G2 cells into which target gene 2 was integrated in Example 2;

[0092] Figure 12 is a photograph of electrophoresis of GBB003-1D6 cells into which target gene 3 was integrated in Example 2;

[0093] 13 is a graph showing the results of the cell viability stability test using Y5E6 and Y6F10 as microparticles expressing exogenous target genes in Example 2;

[0094] FIG14 is a graph showing the results of a stability test of the proliferation rate of microparticles using Y5E6 and Y6F10 as exogenous target gene expression vectors in Example 2;

[0095] FIG15 is a graph showing the glycoform results of the product quality attributes of Y5E6 cells and Y6F10 cells in Example 2;

[0096] FIG16 is a graph showing the charge heterogeneity comparison of proteins expressed in Y5E6 cells and Y6F10 cells in Example 2;

[0097] 17 is a diagram showing the results of sodium dodecyl sulfate-polyacrylamide gel electrophoresis of Y5E6 cells and Y6F10 cells in Example 2;

[0098] Figure 18 is a graph showing the concentration of antibodies expressing target gene 5 in GBB003 cells integrated with target gene 5 in Example 2;

[0099] Figure 19 is a sodium dodecyl sulfate-polyacrylamide gel electrophoresis result of 5GBB003 cells integrated with the target gene in Example 2;

[0100] FIG20 is a graph showing the growth stability test results of GBB-01 cells after 90 days of passage provided in Example 3;

[0101] Figure 21 is a schematic diagram of the structure of the target gene 6 fragment cloned into the exogenous target gene 6 expression fragment in Example 3;

[0102] Figure 22 is a schematic diagram of the Bxb1 integrase plasmid structure provided in Example 3;

[0103] FIG23 is a graph showing the expression levels of the integrated exogenous target gene 6 expression vector particles GBB-01-H8 cells provided in Example 3;

[0104] Figure 24 is a graph showing the expression levels of GBB-01-H8 cells in batches provided in Example 3;

[0105] FIG25 is a graph showing the batch yield of GBB-01-H8 cells provided in Example 3;

[0106] Figure 26 is a comparison of the sugar profiles of GBB-01-H8 cells provided in Example 3;

[0107] FIG27 is a graph showing the comparison of charge heterogeneity of GBB-01-H8 cells provided in Example 3;

[0108] Figure 28 is a diagram of the electrophoresis results of the GBB-01-H8 cell genome provided in Example 3;

[0109] FIG29 is a graph showing the comparative results of the expression levels of the integrated exogenous target gene 3GBBc001 cells provided in Example 4;

[0110] FIG30 is a graph showing the comparative results of the expression levels of the integrated exogenous target gene 1GBBc001 in cells provided in Example 4.

[0111] FIG31 is a graph showing the comparative expression of the integrated exogenous target gene 3GBBc001-14 cells provided in Example 4;

[0112] FIG32 is a graph showing the results of sodium dodecyl sulfate-polyacrylamide gel electrophoresis of GBBc001-14 cells provided in Example 4;

[0113] FIG33 is a graph showing the results of sodium dodecyl sulfate-polyacrylamide gel electrophoresis of GBBc001-3 cells provided in Example 4;

[0114] FIG34 is a schematic diagram of the structure of the fixed integration site screening system provided in Example 5. DETAILED DESCRIPTION

[0115] In order to make the objectives, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below with reference to the following embodiments. It should be understood that the specific embodiments described herein are merely for the purpose of explaining the present invention and are not intended to limit the present invention. In addition, the technical features involved in the various embodiments of the present invention described below may be combined with each other as long as they do not conflict with each other.

[0116] In this article, the preset threshold refers to: a fixed value, or a ranking range value that is greater than a certain ranking range value in an existing numerical range according to the value sorting from large to small, or a ranking value that is greater than a certain ranking value in an existing numerical range according to the value sorting from large to small.

[0117] Herein, the recognition sequence includes but is not limited to homology arm sequences, other sequences that can be recognized, or other homologous recombination sequences.

[0118] Herein, “monoclonal vector particles wherein the expression level of the corresponding marker gene and / or exogenous target gene is retained above a preset threshold value based on the expression level of the marker gene and / or the exogenous target gene during monoclonal screening” refers to “monoclonal vector particles wherein the expression level of the marker gene is retained above a preset threshold value based on the expression level of the marker gene during monoclonal screening”, or “monoclonal vector particles wherein the expression level of the exogenous target gene is retained above a preset threshold value based on the expression level of the exogenous target gene during monoclonal screening”.

[0119] The method for screening site-specific integration sites provided by the present invention comprises the following steps:

[0120] (1) randomly integrating an exogenous gene containing two recognition sequences and a marker gene between the two recognition sequences into an endogenous gene of a vector particle to obtain a vector particle carrying the marker gene;

[0121] The recognition sequence includes but is not limited to the homology arm sequence, other sequences that can be recognized or other homologous recombination sequences. In this embodiment, the recognition sequence refers to the homology arm sequence.

[0122] Since the protein expression level and the stability of the vector microparticles integrated with the exogenous target gene are strongly correlated with the size of the exogenous target gene and its integration site, in order to ensure that the expression level and stability of the exogenous target gene of the constructed vector microparticles integrated with the exogenous target gene meet the screening expectations, the marker gene is of a length comparable to the exogenous target gene to be integrated. In the absence of a clear exogenous target gene, multiple marker genes with representative lengths can be selected for screening. Typical marker gene lengths are 0.75k, 1k, 3k, and 15k. Preferably, a smaller marker gene (3k or less), such as a 0.75k marker gene, can be used to predict the expression level of exogenous target genes of other lengths through an intelligent learning algorithm, thereby significantly reducing costs.

[0123] The marker gene is selected from one or more of a neomycin resistance gene, a thymidine kinase gene, a hygromycin phosphotransferase gene, a dihydrofolate reductase gene, a thymidine kinase gene, a glutamine synthetase gene, an asparagine synthetase gene, a tryptophan synthetase gene, a histidinol dehydrogenase gene, an aminoglycoside phosphotransferase gene, a tryptophan synthetase gene and a fluorescent protein gene.

[0124] The carrier particles are used to express one or more of amino acids, polypeptides, proteins, nucleotides, nucleic acids, and secretions, and are selected from cells, bacteria, and viruses, most commonly carrier cells; the carrier cells are eukaryotic cells; optionally, the eukaryotic cells are mammalian cells; optionally, the mammalian cells include Chinese hamster ovary CHO cells and human embryonic kidney HEK293 cells.

[0125] (2) amplifying and culturing the vector particles carrying the marker gene obtained in step (1) to form a vector particle pool to be screened;

[0126] (3) performing monoclonal screening on the pool of vector particles to be screened obtained in step (2), and retaining monoclonal vector particles whose expression levels of the corresponding marker genes and / or exogenous target genes are higher than a preset threshold value according to the expression levels of the marker genes and / or exogenous target genes during the monoclonal screening, thereby obtaining monoclonal vector particles with high protein expression levels; preferably, using image processing technology to perform monoclonal vector particles, and specific schemes include but are not limited to deep learning target detection models YOLO (You Only Look Once) series and SSD (Single Shot MultiBox Detector) series algorithm models.

[0127] A preferred solution is to predict the expression levels of the marker gene and / or the exogenous target gene using an intelligent algorithm based on the cell image; specifically comprising the following steps:

[0128] (3-1) imaging the pool of vector particles to be screened obtained in step (2) to obtain an image of the vector particles to be screened; preferably, high-throughput scanning imaging is used, and when the marker gene is a fluorescent protein, high-throughput fluorescence scanning imaging is preferably used;

[0129] (3-2) performing image processing on the image of the vector particles to be screened obtained in step (3-1) to obtain an image of each vector particle, and using an intelligent algorithm to predict the expression level of the marker gene and / or the exogenous target gene of the vector particle based on the image features; preferably, the image features used include: cell morphological characteristics and / or optical characteristics, including but not limited to roundness, size, and grayscale of specific cell sites;

[0130] Preferably, the expression levels of the marker genes and / or exogenous target genes of the vector particles predicted by the multiple intelligent algorithms are evaluated using the Normalized Discounted Cumulative Gain (NDCG), an evaluation standard of the ranking algorithm, to evaluate the ranking accuracy of the protein expression levels predicted by the multiple intelligent algorithms. According to the rule that the greater the standard Normalized Discounted Cumulative Gain, the better the ranking accuracy of the protein expression levels, the intelligent algorithm with the best ranking accuracy in predicting the protein expression levels is selected;

[0131] For a specific intelligent algorithm for predicting protein expression, the standard normalized loss cumulative gain calculation formula is as follows:

[0132] Wherein, IDCG is the DCG value of the actual ranking of the protein expression of the sample monoclonal vector particles, i represents the predicted ranking number of a monoclonal vector particle in the sample monoclonal vector particles according to the predicted protein expression of the intelligent algorithm, rel(i) represents the actual protein expression of the i-th monoclonal vector particle; k is the sample size of the sample monoclonal vector particles;

[0133] When the predicted protein is a fluorescent protein, rel(i) indicates the fluorescence value; when the predicted protein is a non-fluorescent protein, rel(i) indicates the value of the protein expression level tested by Western blot, ELISA, mass spectrometry, electrophoresis, etc.

[0134] (3-3) acquiring carrier particles whose expression levels of marker genes and / or exogenous target genes exceed a preset threshold value according to image positioning and moving them to a preset position in the culture medium for retention;

[0135] Since there are differences in the sequence and other properties of marker genes and exogenous target genes, an intelligent algorithm is used to learn and train the differences between the two, and then the expression level of the exogenous target gene is predicted. This can improve the adaptability of the expression of the exogenous target gene at the selected integration site to a certain extent, thereby shortening the overall development cycle and reducing development costs.

[0136] Monoclonal cell vector particle screening generally uses limiting dilution, flow cytometry, and semi-solid culture medium screening. The limiting dilution method requires a large number of experiments, and flow cytometry uses pressure to operate the vector particles, which has a low survival rate and a high risk of contamination. The present invention preferably uses semi-solid culture medium to screen monoclonal vector particles, which can achieve mechanical automation.

[0137] The preset threshold value of the expression level of the marker gene and / or the exogenous target gene can be determined according to the experimental requirements of the protein expression level. For example, if the goal is to require the protein expression level to exceed a certain determined value, the determined value can be used as the preset threshold value; or it can be determined according to the relative protein expression level, for example, the expression level of the marker gene and / or the exogenous target gene is the first fixed proportion or the first fixed ranking in the descending order of expression level.

[0138] (4) Subculture the high-protein expression monoclonal vector particles obtained in step (3), determine the subculture stability of the monoclonal vector particles based on the marker gene expression levels of multiple generations of vector particles, obtain high-expression monoclonal vector particles with subculture stability, and use the insertion position of the exogenous gene on the endogenous gene as the site-specific integration site.

[0139] In the preferred solution, the passage stability of monoclonal vector particles is determined using an intelligent algorithm; the specific steps are as follows:

[0140] (4-1) obtaining the marker gene expression levels of the monoclonal vector particles of multiple generations;

[0141] (4-2) evaluating the differences in marker gene expression levels of the multiple generations of monoclonal vector microparticles using an intelligent algorithm based on the marker gene expression levels of the multiple generations of monoclonal vector microparticles obtained in step (4-1);

[0142] (4-3) Judging whether the monoclonal vector microparticles have generation stability based on the differences in marker gene expression levels of the multiple generations of monoclonal vector microparticles obtained in step (4-2) is conducted according to the principle that the smaller the difference in expression levels, the greater the probability of having generation stability.

[0143] The site-directed integration vector particles provided by the present invention have an exogenous gene inserted into the endogenous gene of the vector particle; the exogenous gene includes two first recognition sequences and a marker gene expression segment for expressing a marker protein, which is arranged between the two first recognition sequences;

[0144] The expression level of the marker gene of the site-directed integration vector particle is higher than the preset threshold;

[0145] The site-directed integration vector particles have passage stability.

[0146] In this embodiment, the first recognition sequence refers to the first homology arm sequence.

[0147] The marker gene is selected from one or more of a neomycin resistance gene, a thymidine kinase gene, a hygromycin phosphotransferase gene, a dihydrofolate reductase gene, a thymidine kinase gene, a glutamine synthetase gene, an asparagine synthetase gene, a tryptophan synthetase gene, a histidinol dehydrogenase gene, an aminoglycoside phosphotransferase gene, a tryptophan synthetase gene and a fluorescent protein gene.

[0148] The exogenous gene containing two recognition sequences and the marker gene between the two recognition sequences contains a marker gene expression fragment, and the marker gene expression fragment has a promoter, which is a CMV promoter, SV40 promoter, RSV promoter, β-globin promoter, UBC promoter, EF1a promoter, ubiquitin promoter, β-actin promoter, PGK1 promoter, Rosa26 promoter, HSP70 promoter, GAPDH promoter, Eif4A1 promoter, Egr1 promoter, FerH promoter, SM22α promoter or Endothelin-1 promoter.

[0149] The first recognition sequence is independently selected from one or more of the following sequences: LoxP sequence, LoxPL3 sequence, LoxP 2L sequence, LoxFas sequence, Lox511 sequence, Lox2272 sequence, Lox2372 sequence, Lox5171 sequence, Loxm2 sequence, Lox71 sequence, Lox66 sequence, FRT sequence, Bxb1 attP sequence, Bxb1 attB sequence, attP sequence, attB sequence, attP-GA sequence and attB-GA sequence.

[0150] The present invention provides a system for screening for site-specific integration sites, wherein the system is an electronic device and / or a non-transient computer-readable storage medium;

[0151] The electronic device includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the method for screening site-specific integration sites when executing the program;

[0152] The non-transitory computer-readable storage medium stores a computer program thereon, and when the computer program is executed by a processor, the steps of the method for screening site-specific integration sites are implemented.

[0153] The method for site-directed integration of an exogenous target gene provided by the present invention comprises the following steps:

[0154] Obtain the site-specific integration vector particles provided by the present invention, and use the first recognition sequence in the exogenous gene in the site-specific integration vector particles to perform homologous recombination, recombining the expression fragment into the gene of the site-specific integration vector particles while removing the exogenous target gene fragment of the marker gene.

[0155] The preferred solution is to screen all site-specific integration vectors for the exogenous target gene. The specific method is as follows:

[0156] S1. Obtaining a plurality of site-directed integration vector particles provided by the present invention, performing homologous recombination using the first recognition sequence in the exogenous gene within the site-directed integration vector particles, recombining the exogenous target gene expression fragment into the gene of the site-directed integration vector particle while removing the marker gene expression fragment, thereby obtaining a plurality of candidate exogenous target gene expression vector particles;

[0157] S2, amplifying and subculturing the multiple candidate exogenous target gene expression vectors obtained in step S1;

[0158] S21. Retaining, based on the expression level of the exogenous target gene, monoclonal exogenous target gene expression vector particles whose expression level of the corresponding exogenous target gene is higher than a preset threshold; preferably, the expression level of the exogenous target gene is retained by performing intelligent algorithm prediction based on cell images; specifically comprising the following steps:

[0159] S211, imaging the monoclonal exogenous target gene expression vector microparticles obtained in step S21 to obtain an image of the candidate monoclonal exogenous target gene expression vector microparticles; preferably, high-throughput scanning imaging is used;

[0160] S212. Predicting the expression level of the exogenous target gene in the monocloned exogenous target gene expression vector microparticles using an intelligent algorithm based on image features; preferably, the image features used include: cell morphological features and / or optical features, including but not limited to roundness, size, and grayscale of specific cell sites;

[0161] Preferably, the expression levels of the exogenous target gene of the monoclonal exogenous target gene expression vector microparticles predicted by multiple intelligent algorithms are evaluated using the Normalized Discounted Cumulative Gain (NDCG), an evaluation standard of the ranking algorithm, to evaluate the ranking accuracy of the protein expression levels predicted by the multiple intelligent algorithms. According to the rule that the larger the standard Normalized Discounted Cumulative Gain, the better the ranking accuracy of the protein expression level, the intelligent algorithm with the best ranking accuracy in predicting the protein expression level is selected;

[0162] For a specific intelligent algorithm for predicting protein expression, the standard normalized loss cumulative gain calculation formula is as follows:

[0163] Wherein, IDCG is the DCG value of the actual ranking of the protein expression of the sample monoclonal vector particles, i represents the predicted ranking number of a monoclonal vector particle in the sample monoclonal vector particles according to the predicted protein expression of the intelligent algorithm, rel(i) represents the actual protein expression of the i-th monoclonal vector particle; k is the sample size of the sample monoclonal vector particles;

[0164] When the predicted protein is a fluorescent protein, rel(i) indicates the fluorescence value; when the predicted protein is a non-fluorescent protein, rel(i) indicates the value of the protein expression level tested by Western blot, ELISA, mass spectrometry analysis, electrophoresis, etc.

[0165] S22 determines the passage stability of the exogenous target gene expression vector based on the exogenous target gene expression level of multiple generations of monoclonal exogenous target gene expression vector particles;

[0166] The passage stability of the monoclonal exogenous target gene expression vector microparticles is determined using an intelligent algorithm; the specific steps are as follows:

[0167] S221, obtaining the expression levels of the exogenous target gene of multiple generations of the monoclonal exogenous target gene expression vector microparticles;

[0168] S222, using an intelligent algorithm to evaluate the difference in the expression levels of the exogenous target gene among the multiple generations of monoclonal exogenous target gene expression vector microparticles obtained in step S221;

[0169] S223. Judging whether the monoclonal exogenous target gene expression vector microparticles have passage stability based on the difference in exogenous target gene expression levels of the multiple generations of monoclonal exogenous target gene expression vector microparticles obtained in step S222, according to the principle that the smaller the difference in expression levels, the greater the probability of passage stability.

[0170] Obtain exogenous target gene expression vector microparticles with high protein expression and stable passage.

[0171] The exogenous target gene expression segment is arranged in a recombinant vector, and the exogenous target gene is site-specifically integrated into the site of the site-specific integration vector particle through the recombinant vector; the recombinant vector is a lentiviral vector, an adenoviral vector, an adeno-associated viral vector, a herpes virus vector, a poxvirus vector, a baculovirus vector, a papillomavirus vector, a papillomatovirus vector, an integrating phage vector, a non-viral vector, a transposon, a transposase, an integrase substrate or a plasmid.

[0172] The length of the marker gene expression segment of the site-directed integration vector particle is comparable to that of the exogenous target gene expression segment.

[0173] The method of performing homologous recombination using the first recognition sequence of the site-directed integration fragment of the site-directed integration vector particle is specifically as follows:

[0174] Constructing an exogenous target gene recombinant fragment comprising an exogenous target gene; wherein the exogenous gene recombinant fragment contains two second recognition sequences and an exogenous target gene expression fragment; the exogenous target gene expression fragment is between the two second recognition sequences; the exogenous target gene expression fragment includes a promoter for regulating expression of the exogenous target gene and an exogenous target gene sequence;

[0175] The exogenous gene is integrated into the site of the site-directed integration vector particle, and the two first recognition sequences on both sides of the marker gene are respectively changed into a third recognition sequence and a fourth recognition sequence; the third recognition sequence is formed by the reaction of the first recognition sequence and the second recognition sequence, and the fourth recognition sequence is formed by the reaction of the first recognition sequence and the second recognition sequence, and the third recognition sequence is different from the fourth recognition sequence.

[0176] In a preferred embodiment, the third recognition sequence and the fourth recognition sequence will not become the second recognition sequence or the first recognition sequence in the same reaction system.

[0177] The third recognition sequence and the fourth recognition sequence are generated by a recombinase, and the recombinase is one of Bxb1 integrase, ΦC31 integrase, Cre recombinase or FLP recombinase; and / or

[0178] The second recognition sequence is independently selected from one or more of the following sequences: LoxP sequence, LoxPL3 sequence, LoxP 2L sequence, LoxFas sequence, Lox511 sequence, Lox2272 sequence, Lox2372 sequence, Lox5171 sequence, Loxm2 sequence, Lox71 sequence, Lox66 sequence, FRT sequence, Bxb1 attP sequence, Bxb1attB sequence, attP sequence, attB sequence, attP-GA sequence and attB-GA sequence; and / or,

[0179] The third recognition sequence and the recognition sequence are independently selected from one or more of the following sequences: LoxP sequence, LoxPL3 sequence, LoxP 2L sequence, LoxFas sequence, Lox511 sequence, Lox2272 sequence, Lox2372 sequence, Lox5171 sequence, Loxm2 sequence, Lox71 sequence, Lox66 sequence, FRT sequence, attL and attR sequence.

[0180] In this embodiment, the first recognition sequence refers to the first homology arm sequence, the second recognition sequence refers to the second homology arm sequence, the third recognition sequence refers to the third homology arm sequence, and the fourth recognition sequence refers to the fourth homology arm sequence; the third recognition sequence is formed by the reaction of the first homology arm located on the left side of the marker gene in the marker gene expression fragment and the second homology arm located on the left side of the exogenous target gene in the exogenous target gene expression fragment, and the fourth recognition sequence is formed by the reaction of the first homology arm located on the right side of the marker gene in the marker gene expression fragment and the second homology arm located on the right side of the exogenous target gene in the exogenous target gene expression fragment.

[0181] The exogenous target gene expression vector microparticles provided by the present invention are obtained by screening according to the above-mentioned method for site-directed integration of the exogenous target gene, wherein the exogenous target gene is inserted into the endogenous gene of the vector microparticle; the exogenous gene includes a third recognition sequence, a fourth recognition sequence, and an exogenous target gene expression segment arranged between the third recognition sequence and the fourth recognition sequence for expressing the target protein;

[0182] The target gene expression level of the exogenous target gene expression vector microparticle is higher than the preset threshold value;

[0183] The exogenous target gene expression vector microparticles have passage stability.

[0184] The exogenous target gene expression fragment has a promoter, and the promoter is CMV promoter, SV40 promoter, RSV promoter, β-globin promoter, UBC promoter, EF1a promoter, ubiquitin promoter, β-actin promoter, PGK1 promoter, Rosa26 promoter, HSP70 promoter, GAPDH promoter, Eif4A1 promoter, Egr1 promoter, FerH promoter, SM22α promoter or Endothelin-1 promoter.

[0185] The third recognition sequence and the fourth recognition sequence are generated by a recombinase, and the recombinase is one of Bxb1 integrase, ΦC31 integrase, Cre recombinase or FLP recombinase.

[0186] The third recognition sequence and the recognition sequence are independently selected from one or more of the following sequences: LoxP sequence, LoxPL3 sequence, LoxP 2L sequence, LoxFas sequence, Lox511 sequence, Lox2272 sequence, Lox2372 sequence, Lox5171 sequence, Loxm2 sequence, Lox71 sequence, Lox66 sequence, FRT sequence, attL and attR sequence.

[0187] In this embodiment, the first recognition sequence refers to the attP sequence, the second recognition sequence refers to the attB sequence, the third recognition sequence refers to the attL sequence, and the fourth recognition sequence refers to the attR sequence.

[0188] The present invention provides a system for site-directed integration of an exogenous target gene, wherein the system is an electronic device and / or a non-transient computer-readable storage medium;

[0189] The electronic device includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the steps of the method for site-directed integration of an exogenous target gene are implemented;

[0190] The non-transitory computer-readable storage medium stores a computer program thereon, and when the computer program is executed by a processor, the steps of the method for site-directed integration of an exogenous target gene are implemented as described above.

[0191] The following are examples:

[0192] Example 1: Method for screening site-specific integration sites, preparation of site-specific integration vector microparticles, method for site-specific integration of exogenous target genes, preparation of exogenous target gene expression vector microparticles

[0193] The method for screening site-specific integration sites provided in Example 1, as shown in FIG3 , includes the following steps:

[0194] (1) Randomly integrating a marker gene with homology arms at both ends into the endogenous gene of the vector particle to obtain a vector particle carrying the marker gene;

[0195] In Example 1, the green fluorescent protein gene (EGFP) was used as a screening marker gene, and attP sequences were used as homology arms to construct a recombinant plasmid containing RMCE (recombinase-mediated cassette exchange), as shown in Figure 1 below. The marker gene expression fragment contained in the recombinant plasmid used for random integration is shown in Figure 2.

[0196] After linearizing the constructed plasmid, purify and recover the linearized DNA. Transfer 3×10⁶ CHO-K1 cells to a 1.5mL centrifuge tube and centrifuge at 1000 rpm for 5 minutes at room temperature. Discard the supernatant. Resuspend the cells in 100μL of R Buffer (specific for cell electroporation). Electroporation conditions: 1575V, 10ms, 3 cycles for 3×10⁶ cells. Use 15μg of plasmid. A fluorescent protein gene with attP sequences at both ends is randomly integrated into the endogenous gene on the vector particle. The constructed plasmid integrated into the endogenous gene is the exogenous gene, which integrates the marker gene expression fragment shown in Figure 2 into the endogenous gene. The vector particle is a cell.

[0197] After two days, use G418 as the pressurizing reagent at a concentration of 800 μg / ml. Plate cells in a 96-well plate at a density of 10,000 cells / well, with at least 200 μl of culture medium per well. Check the plate ten days later. Once a large number of fluorescent cell clusters are observed, cells can be enriched. Enrichment methods include, but are not limited to, manual enrichment. Follow the one-well-to-one-well principle during enrichment; do not concentrate cells from multiple wells into one well.

[0198] (2) amplifying and culturing the vector particles carrying the marker gene obtained in step (1) to form a vector particle pool to be screened; specific steps:

[0199] Observe the growth of enriched cells at all times. Expand the cells once the coverage reaches 50% or higher. Two expansion methods are available: 96-well plate, 24-well plate, 6-well plate, and 96-well plate. After expansion to shake flasks, passage the cells two to three times to form a stable cell pool, which can then be used for monoclonal screening.

[0200] (3) performing monoclonal screening on the pool of vector particles to be screened obtained in step (2), and retaining monoclonal vector particles whose expression levels of the marker gene and / or the exogenous target gene are higher than a preset threshold value based on the expression levels of the marker gene and / or the exogenous target gene during the monoclonal screening, thereby obtaining monoclonal vector particles with high protein expression;

[0201] The formation of monoclonal cells and the screening of stable fluorescence are achieved through artificial intelligence, as shown in Figure 3 below:

[0202] The steps are as described by ah:

[0203] (3-1) imaging the pool of vector particles to be screened obtained in step (2) to obtain an image of the vector particles to be screened; preferably, high-throughput scanning imaging is used, and when the marker gene is a fluorescent protein, high-throughput fluorescence scanning imaging is preferably used;

[0204] a) The cells in the stable cell pool are diluted to a certain concentration and then inoculated into a culture dish containing semi-solid culture medium. During the inoculation process, the cells are ensured to be approximately evenly distributed in all positions of the small dish. The culture dish is left to stand for about half an hour to allow the cells to settle to the bottom of the culture dish.

[0205] b) The culture dish is transferred to the microscope stage, and the microscope performs high-throughput scanning on the cells in the culture dish;

[0206] (3-2) performing image processing on the image of the vector particle to be screened obtained in step (3-1) to obtain an image of each vector particle, and using an intelligent algorithm to predict the expression level of the marker gene and / or the exogenous target gene of the vector particle based on the image features;

[0207] c) The microscope-scanned image is uploaded to a server for artificial intelligence image analysis. The analysis process includes monoclonal cell line detection, protein expression level prediction of monoclonal cell lines, protein expression level ranking, and coding and localization of screened high-protein-expressing cell lines. The protein expression level prediction of the monoclonal cell line can be based on fluorescence or not. The following step c is protein expression level prediction not based on fluorescence.

[0208] In step c, the monoclonal cell line target detection is based on image processing technology. In this embodiment, the YOLOv8 target detection algorithm is used, and the actual detection effect reaches mAP 94.2%, as shown in FIG4 .

[0209] The target detection model for monoclonal cell lines in step c is consistent with commonly used deep learning target detection models, so it is not described in detail. Specifically, the cell images must be annotated. To improve the algorithm's prediction accuracy, the bounding boxes of all monoclonal and adherent cell lines are annotated. These bounding boxes serve as the ground truth for the model output loss calculation. After algorithm training, the model learns to extract the bounding box information of monoclonal cell lines. In the actual application scenario, in step c, the trained model can predict the bounding boxes of monoclonal cell lines in the image.

[0210] Since the selected sites will be used to insert exogenous genes of different lengths and to express exogenous genes, and the length of the exogenous gene inserted in the site is strongly correlated with the amount of protein expression, several groups of fluorescent protein marker genes of different deoxyribonucleic acid lengths will be selected here. In this embodiment, four exogenous gene expression fragments of approximately 0.75k, 1K, 3K, and 15K lengths are used. The exogenous genes include genes that can express fluorescent proteins and / or genes that can express other proteins. By testing the expression of fluorescent proteins and other proteins, images of cells that can express four exogenous genes of similar lengths of 0.75k, 1K, 3K, or 15K are selected for artificial intelligence image learning. The case where the exogenous gene length is 0.75K corresponds to the case where the exogenous gene can express fluorescent protein in a specific site in the target cell line and the length of the inserted exogenous gene expression fragment is about 0.75k. The case where the inserted exogenous gene length is 1K corresponds to the case where the target gene is double-stranded DNA in a specific site in the target cell line and the length of the exogenous gene expression fragment is about 1k. The case where the inserted exogenous gene length is 3K corresponds to the case where the target gene is a monoclonal antibody at a specific site in the target cell line and the length of the exogenous gene expression fragment is about 3K to 7K. The case where the inserted gene length is 15K corresponds to the case where the target gene is used to express a multivalent antibody or a multi-segment insertion structure at a specific site in the target cell line and the length of the exogenous gene expression fragment is about 15K.

[0211] Step c provides four groups of expression prediction models of different lengths, each corresponding to a set of fluorescence value labels. Artificial intelligence image learning is performed using historical images of CHO cells with exogenous target genes inserted at four different lengths, generating four prediction models for high expression prediction analysis. During site application, the exogenous target gene expression fragments are replaced with exogenous target gene expression fragments of similar lengths. During the construction of CHO cells as historical images, exogenous genes of different lengths can be used as marker gene expression fragments to replace the exogenous genes at high expression sites. The prediction model is then used to determine whether the site is still a high expression site under the protein expression conditions of the inserted genes of different lengths. If all inserted genes of different lengths are predicted to be high expression sites after being replaced as marker genes, the site can be determined to be a protein expression site suitable for exogenous target gene expression fragments of different lengths. In this process, the marker gene can use other proteins that can produce fluorescence, or other proteins, using the fluorescence value after expression or the protein expression of the actual marker gene as the label. In step c, historical images of CHO cells in which the length of the deoxyribonucleotides of the inserted exogenous gene is the same or similar to the length of the DNA sequence of the inserted exogenous gene of the green fluorescent protein gene (EGFP) are used to perform high expression prediction analysis using a prediction model generated by artificial intelligence image learning.

[0212] Step c predicts the fluorescent protein expression level based on the monoclonal cell image. In this embodiment, the SqueezeNet deep learning network and the MSE Loss function are used, and the expression level of the exogenous gene or marker gene in the monoclonal cell image is used as the label of the regression algorithm.

[0213] The expression levels of the exogenous target gene of the monoclonal exogenous target gene expression vector particles predicted by multiple intelligent algorithms are evaluated using the normalized discounted cumulative gain (NDCG) evaluation standard of the ranking algorithm to evaluate the ranking accuracy of the protein expression levels predicted by the multiple intelligent algorithms. According to the rule that the larger the standard normalized discounted cumulative gain, the better the ranking accuracy of the protein expression level, the intelligent algorithm with the best ranking accuracy of the protein expression level prediction is selected;

[0214] For a specific intelligent algorithm for predicting protein expression, the standard normalized loss cumulative gain calculation formula is as follows:

[0215] Wherein, IDCG is the DCG value of the actual ranking of the protein expression of the sample monoclonal vector particles, i represents the predicted ranking number of a monoclonal vector particle in the sample monoclonal vector particles according to the predicted protein expression of the intelligent algorithm, rel(i) represents the actual protein expression of the i-th monoclonal vector particle; k is the sample size of the sample monoclonal vector particles;

[0216] When the predicted protein is a fluorescent protein, rel(i) indicates the fluorescence value; when the predicted protein is a non-fluorescent protein, rel(i) indicates the value of the protein expression level tested by Western blot, ELISA, mass spectrometry analysis, electrophoresis, etc.

[0217] Specifically, in this example, the cell images predicted expression levels, and the predicted expression levels were ranked according to the predicted expression levels. Comparison with the actual fluorescence value ranking showed that NDCG = 0.89. The expression level prediction model in this invention is described in the Chinese patent document: "Cell Screening Method and Apparatus Based on Convolutional Neural Networks."

[0218] (3-3) The carrier particles whose expression levels of the marker gene and / or the exogenous target gene exceed a preset threshold are acquired according to image positioning and moved to a preset position in the culture medium for retention; the preset threshold refers to the first fixed ratio of the expression levels of the marker gene and / or the exogenous target gene in descending order of expression levels.

[0219] d) The coding and location information of the screened protein high-expressing cell lines are returned to the robot control software;

[0220] The cell codes and positions returned in step d, in this embodiment, return the top 100 cells in terms of predicted expression levels, with the codes ranging from 1 to 100. The position information includes the coordinates in the plane coordinate system relative to the center of the microscope (the depth of the culture medium is not considered because all selected cells settle to the bottom of the dish. In addition, during the photography process, cells that have not settled to the bottom of the dish will be out of focus, resulting in blurred cells, and cells with blurred images will also be excluded as candidate cells).

[0221] e) The robot control software automatically operates (or manually assists) the robotic arm to aspirate the selected monoclonal cell lines and transfer them to the designated well plate. This process is repeated until all high-expressing cell lines are transferred.

[0222] Step e: The robot arm picks up the target monoclonal cell line and transfers it to the well plate.

[0223] Patent Chinese patent literature: "Cell manipulation task processing method, device, equipment and medium based on robot" with publication number CN113821287B, "Cell liquid collection control method, device, equipment and storage medium based on robot" with publication number CN113821287B, "Cell manipulation robot control method, device, equipment and storage medium" with publication number CN113771030A, "Cell manipulation robot control method, device, equipment and storage medium" with publication number CN113733087B are all described in detail.

[0224] f) After the cell clones in the well plate have grown to a certain number, they are transferred to a larger well plate for culture, and finally expanded to a shake flask for culture. The cell expression level is then tested to confirm that the cell line selected by the artificial intelligence is a high-yield cell line;

[0225] In this embodiment, step f is to transfer the cells to a 96-well plate for amplification and then transfer them to a shake flask for culture.

[0226] (4) Subculture the high-protein expression monoclonal vector particles obtained in step (3), determine the subculture stability of the monoclonal vector particles based on the marker gene expression levels of multiple generations of vector particles, obtain high-expression monoclonal vector particles with subculture stability, and use the insertion position of the exogenous gene on the endogenous gene as the site-specific integration site.

[0227] (4-1) obtaining the marker gene expression levels of the monoclonal vector particles of multiple generations;

[0228] g) The selected high-yield cell lines are subcultured. Before each generation of cell lines, a diluted portion of the sample is placed in a culture dish and images are taken using a microscope. Images of the cell lines are taken continuously for several generations. Currently, the expression of the selected cells can be predicted by the model learned from the cell morphology using photography, and the expression of the selected cells can also be predicted by fluorescent labeling.

[0229] Step g: Photographing Monoclonal Cells In this example, a Thermo Fisher Scientific M7000 microscope was used to photograph the cells.

[0230] (4-2) evaluating the differences in marker gene expression levels of the multiple generations of monoclonal vector microparticles using an intelligent algorithm based on the marker gene expression levels of the multiple generations of monoclonal vector microparticles obtained in step (4-1);

[0231] (4-3) Determining whether the monoclonal vector microparticles have generation stability based on the differences in marker gene expression levels of the multiple generations of monoclonal vector microparticles obtained in step (4-2).

[0232] h) Inputting the collected images of several generations of the cell line into an artificial intelligence algorithm to predict the cell line's passage stability based on fluorescence intensity and / or protein expression. Cell lines with high protein expression and stable passage stability are selected as the final candidate cell lines. The cell line's passage stability prediction can be based on fluorescence or cell morphology rather than fluorescence. The following step h involves predicting protein expression based on cell morphology rather than fluorescence. The final candidate cell line is the site-directed integration vector particle.

[0233] Step h predicts the cell line's passage stability, including but not limited to using traditional image recognition algorithms to extract and predict image features using histograms, automatically extracting and predicting image features using deep learning neural networks, and other image-based prediction techniques. In this embodiment, a deep learning-based method was used to automatically extract image features of stable and unstable cell lines and predict their stability, achieving an accuracy rate of 84% for predictions across different cell lines. This is described in detail in the Chinese patent document "Cell Line Stability Prediction Method, Apparatus, Computer Device, and Storage Medium."

[0234] (5) After the stable cell lines screened by AI were expanded and cultured into shake flasks, the growth curve was monitored for 14 days using CHO-K1 cells as a control. The inoculation density was 5×105 / ml, the inoculation system was 30ml, and the cells were counted on the day of inoculation, the 3rd day of culture, the 5th day of culture, the 7th day of culture, the 9th day of culture, the 11th day of culture, the 13th day of culture, and the 14th day of culture. Monoclones with a growth rate lower than that of CHO-K1 were screened and eliminated, and a site-specific integration vector with high expression and higher stability than the CHO-K1 cell line was obtained.

[0235] (6) The site-specific integration vector with high expression and higher passage stability than the CHO-K1 cell line was subjected to stable passage, with passage every 3 or 4 days. The cell density of the 4-day passage was 3×105 / ml, and the cell density of the 3-day passage was 5×105 / ml. The fluorescence value of the cell line was regularly monitored during the passage. The above-mentioned high-fluorescence cell line was subjected to a passage stability study for about 90 days using the passage medium. Cells with small fluorescence changes and stable cell line growth were confirmed to be stable high-fluorescence cells; the copy number and integration site of the stable high-fluorescence cells were verified. The stable high-fluorescence cells here are site-specific integration vectors with high passage stability and high expression.

[0236] The above-mentioned site-directed integration vector is used as an example to integrate the exogenous target gene and screen out the exogenous target gene expression vector particles with high target gene expression level and high passage stability. The method for site-directed integration of the exogenous target gene provided by the present invention comprises the following steps:

[0237] Obtain the site-directed integration vector particles provided by the present invention, and use the first homologous arm of the exogenous gene in the site-directed integration vector particles to perform homologous recombination, thereby recombining the exogenous target gene expression fragment into the gene of the site-directed integration vector particles while removing the marker gene expression fragment.

[0238] The specific steps are:

[0239] (7) Stable high-fluorescence cells are electroporated using a plasmid carrying an exogenous target gene and co-transfected with Bxb1 integrase into stable high-fluorescence cells. The plasmid carrying the exogenous target gene is an exogenous gene comprising two attB sequences and a target gene sequence between the two attB sequences. The exogenous gene comprises an exogenous target gene expression fragment that can be replaced with a marker gene expression fragment inserted into an endogenous gene, and the exogenous target gene expression fragment comprises two attB sequences and a target gene sequence between the two attB sequences. Under the action of Bxb1 integrase, the attP on the left side of the marker gene EGFP forms attL with the attB on the left side of the target gene, and the attP on the right side of the marker gene EGFP forms attR with the attB on the right side of the target gene. At the same time, the exogenous target gene expression fragment completes the replacement of the marker gene expression fragment.

[0240] Two days after electroporation, the plate was checked to record the electroporation effect. Hygromycin was used to pressurize the cells at a concentration of 250 μg / ml. The cells were then divided into 96-well plates with 6,000-10,000 cells per well.

[0241] The 96-well plate was placed in a CO2 constant temperature incubator (37°C, 80% humidity) for 1-2 weeks. The proportion of non-fluorescent cells in the minipool was counted, and the non-fluorescent cells were manually enriched and monocloned by limiting dilution.

[0242] (8) The 96-well plate of monoclonal cells was placed in a CO2 constant temperature incubator (37°C, 80% humidity) and cultured for 14 days. During this period, monoclonal cells were photographed on the day of inoculation, culture day 1, culture day 2, culture day 3, culture day 7, and culture day 14 to confirm the monoclonal cells and fluorescence status.

[0243] The formation of monoclonal cells and the screening of stable, high-expressing cell lines can also be achieved through artificial intelligence. The specific steps of artificial intelligence implementation are as described in the ah process in step 5 above, with the difference that in this step c, historical images of CHO cells containing marker gene expression fragments with a length similar to the DNA sequence length of the exogenous target gene expression fragment inserted into the target cell line are used to perform high expression prediction analysis using a prediction model generated by artificial intelligence image learning, thereby obtaining exogenous target gene expression vector particles.

[0244] (9) After 14 days, the confirmed non-fluorescent and normally growing monoclones were expanded to 24-well plates and cultured in a constant temperature incubator (37°C, 80% humidity) for further expansion and expansion to the shake flask stage. About 1 month later, the cell line can enter the feeding stage through fed-batch culture (Advance + 1% Glutamax, i.e., fed-batch culture medium, for the first three days). On the 3rd and 5th day of culture, 3% (v / v) Cell Boost 7a and 0.3% (v / v) Cell Boost 7b (cytiva) were added simultaneously. On the 7th, 9th, 11th and 13th day of culture, 5% (v / v) Cell Boost 7a and 0.5% (v / v) Cell Boost 7b (cytiva) were added simultaneously every day. On the 3rd, 5th, 7th, 9th, 11th and 13th day of culture, sugar was supplemented according to the sugar consumption of the cells and the expression level was evaluated.

[0245] (10) Monoclonal cells with high expression were subcultured for approximately 90 days. During this period, 5 × 106 cells were retained to extract the genome for site replacement verification and residual gene detection. Subsequently, batch cultures of the 1st, 13th, and 26th passages were cultured for 7 days to confirm the stability of expression levels and QP. “QP stability” refers to the stability of the daily production of individual cells.

[0246] Steps 1-6 in the above process complete the "mother cell line construction + screening" process, completing the mother cell line construction and screening in a one-step process and obtaining site-directed integration vector microparticles. Steps 7 and 8 in the above process complete the "target cell line construction + first screening" process, achieving site-directed integration of the exogenous target gene, and completing the initial screening of the target cell line. Steps 9 and 10 are then used to perform a second and third screening of the target cell line that has been initially screened, selecting exogenous target gene expression vector microparticles with higher expression levels and better stability. The exogenous gene containing the exogenous target gene expression fragment within the exogenous target gene expression vector microparticles is inserted at the same position as the exogenous gene containing the marker gene expression fragment within the site-directed integration vector microparticles. When the same site-directed integration vector microparticles are used, the exogenous gene containing the exogenous target gene expression fragment within the exogenous target gene expression vector microparticles obtained using the above-described exogenous target gene integration method is inserted at the same position, thereby achieving site-directed integration of the target gene.

[0247] The existing two-step screening method of "mother cell line construction + screening" takes 12 months. By using steps 1-6 in the above process to complete the one-step screening of "mother cell line construction + screening", it only takes 6 months, which is 50% shorter. The mother cell line is highly stable and has high expression levels. The main reason for the significant time reduction is the use of the one-step "mother cell line construction + screening" method and the efficient and rapid screening of monoclonal cell lines through artificial intelligence.

[0248] Using existing technology, it takes 12 months to complete the "construction of target cell line + screening". By adopting steps 7-10 in the above process to complete the "construction of target cell line + screening", it only takes 6 months, which shortens the time by 50%. The target cell line is highly stable and has a high expression level of the exogenous target gene. The main reason for the significant shortening of the time is that the mother cell line has high stability and high expression level, and the exogenous target gene is inserted into a specific site in the mother cell line by site-directed integration, thereby shortening the "construction of target cell line + screening" process, and the prepared target cell line has high stability and high expression level of the exogenous target gene.

[0249] Furthermore, when predicting the expression level, images of cells with exogenous genes inserted inside are used for learning. When predicting mother cells, images of historical cells with exogenous genes similar in length to the DNA sequence of exogenous genes inserted into the mother cells are used for learning. This is used for early prediction of expression levels, which is then used for screening mother cell lines. When predicting target cell lines, images of cells with exogenous genes similar in length to the DNA sequence of exogenous genes inserted into the mother cells are used for learning. This is used for early prediction of expression levels, which is then used for screening product cell lines, thereby improving the chances of obtaining sites that can highly express exogenous target genes. This arrangement can reduce the possibility that the exogenous target gene cannot be highly expressed due to a large difference in the length of the exogenous target gene expression fragment containing the exogenous target gene that needs to be site-specifically integrated into a specific site and the length of the marker gene expression fragment inserted when screening mother cells.

[0250] Example 2: Screening of the S6C7 site and preparation of site-specific integration vector particles and exogenous target gene expression vector particles

[0251] The site-directed integration vector microparticles and exogenous target gene expression vector microparticles prepared in this example according to the method described in Example 1 are different from those in Example 1 in that:

[0252] The specific steps of differentiation are:

[0253] (1) The operation flow of Example 2 is shown in FIG5 . Specific consumables include: Neon Resuspension Buffer R (ThermoFisher), a resuspension buffer for cell electroporation; E1 Buffer (ThermoFisher), an electroporation buffer; a recovery medium consisting of 80% (v / v) EX-CELL CHO Cloning Medium (Sigma-Aldrich) and 20% (v / v) EX-CELL Advanced CHO Fed-batch Medium (Sigma-Aldrich) supplemented with 1% GlutaMAX (ThermoFisher); and an expansion medium and a subculture medium consisting of EX-CELL Advanced CHO Fed-batch. CHO medium was supplemented with 1% GlutaMAX (ThermoFisher); pressurized medium 1 was passage medium supplemented with G418 at a final concentration of 800 μg / ml; pressurized medium 2 was passage medium supplemented with hygromycin at a final concentration of 250 μg / ml; the conditioned medium was the supernatant obtained by sterile filtration after inoculating the passage medium with CHO-K1 for 1 day; the cloning medium was 75% (v / v) EX-CELL CHO Cloning Medium, 20% (v / v) conditioned medium, and 5% (v / v) ClonaCell-CHO ACF Supplement supplemented with 1% GlutaMAX; the basal medium in the fed-batch medium was EX-CELL Advanced CHO Fed-batch Medium supplemented with 1% GlutaMAX (ThermoFisher), and the feed medium was Cell Boost 7a / 7b (HyClone).

[0254] (2) The site-directed integration vector particles obtained in step 6 were named GBB003 mother cells.

[0255] (3) The target genes in step 8 are target gene 1 (monoclonal antibody), target gene 2 (monoclonal antibody), target gene 3 (fusion protein), target gene 4 (bispecific antibody; IgG-SCFV), and target gene 5 (bispecific antibody; KiH-Crossmab). Target genes 1 to 5 are cloned into different plasmids using enzyme digestion and ligation to construct recombinant plasmids containing RMCE. The recombinant plasmids for target gene 1, target gene 2, target gene 4, and target gene 5 are shown in FIG6 , and the recombinant plasmid for target gene 3 is shown in FIG7 .

[0256] The exogenous fragment inserted into the endogenous gene in GBB003 mother cells was 7545 bp long, the marker gene expression fragment was 3815 bp long, and the remaining fragment was 3730 bp long. The exogenous target gene expression fragment corresponding to target gene 1 was 6817 bp long, the exogenous target gene expression fragment corresponding to target gene 2 was 6866 bp long, the exogenous target gene expression fragment corresponding to target gene 3 was 2548 bp long, the exogenous target gene expression fragment corresponding to target gene 4 was 7688 bp long, and the exogenous target gene expression fragment corresponding to target gene 5 was 12256 bp long. The stability of the site and / or the editable length of the gene fragment were verified by inserting gene fragments of different lengths.

[0257] The exogenous target gene 1 has a light chain sequence shown in SEQ ID No. 2; the exogenous target gene 1 has a heavy chain sequence shown in SEQ ID No. 3;

[0258] The exogenous target gene 2 has a light chain sequence shown in SEQ ID No. 4; the exogenous target gene 2 has a heavy chain sequence shown in SEQ ID No. 5;

[0259] The exogenous target gene 3 has the sequence shown in SEQ ID No. 24;

[0260] The exogenous target gene 4 has a light chain sequence shown in SEQ ID No. 25; the exogenous target gene 4 has a heavy chain sequence shown in SEQ ID No. 26;

[0261] The exogenous target gene 5 has the target gene 5 light chain 1 sequence shown in SEQ ID No.27, the target gene 5 heavy chain 1 sequence shown in SEQ ID No.28, the target gene 5 light chain 2 sequence shown in SEQ ID No.29 and the target gene 5 heavy chain 2 sequence shown in SEQ ID No.30.

[0262] (4) In step 9, the expression levels of the site-directed integrated product minipool cell pool (GBB003-minipool), the exogenous target gene 1 expression vector particles (denoted as GBB003-2D8 cells), the randomly integrated minipool cell pool (random minipool), and the single-copy cell line (random monoclonal) cultured under the same conditions are compared as shown in Figure 8.

[0263] (5) In step 10, the 1st, 13th, and 26th generations were batch cultured for 7 days, and sugar was supplemented on days 3, 5, and 7 according to the sugar consumption of the cells. The expression levels are shown in FIG9 , which shows that the monoclonal cells are stable and highly productive.

[0264] Next-generation whole-genome sequencing of the parent cell line, GBB003, which highly expresses both the fluorescent gene and the target gene, revealed a single-copy integration site in the CHO-K1 cell line: NW_003616785.1:83044. This integration site, designated S6C7, represents the insertion site of an exogenous gene into an endogenous gene.

[0265] The sequence fragment after integration of the target gene 1 is shown in SEQ ID No. 6; the breakpoint upstream primer F1 (SEQ ID No. 7): AGACCAGCCTCAGATGTCACAC and the target gene downstream primer R1 (SEQ ID No. 8): AGGCACACAACGGAGGCGGT on the integration plasmid were used to amplify the GBB003-2D8 cell genome. The target gene 1 downstream primer R1 was located in the light chain of the target gene 1, and a fragment with a theoretical size of 3403 bp was obtained, as shown in Figure 10 (corresponding bands in the GBB003-2D8 cell lane).

[0266] The sequence fragment after integration of target gene 1 includes attL, the nucleotide sequence of which is shown in SEQ ID No. 9: SEQ ID No. 9: ATGATCCTGACGACGGAGACCGCGGTGGTTGACCAGACAAACC. In the sequence fragment after integration of target gene 1, the portion before attL is an exogenous sequence, and the portion after attL is an endogenous sequence. The exogenous sequence includes a portion of the light chain reading frame of target gene 1, namely, the sequence amplified from SEQ ID No. 8. Target gene 1 can encode a monoclonal antibody.

[0267] The amplified bands were recovered by gel excision and purified, and multiple pairs of primers were designed for DNA sequencing. The sequence after splicing of the sequencing results was shown as SEQ ID No. 6, which was consistent with the theoretical sequence, thus confirming that the target gene 1 was correctly integrated into the corresponding position of the marker gene expression fragment.

[0268] Furthermore, it is explained that the vector particle CHO-K1 cell has an S6C7 site, which is located within the fragment SEQ ID No. 1 shown in SEQ ID No. 1. Specifically, the annotation information of the integration site S6C7 on CHO-K1 is: NW_003616785.1:83044.

[0269] Similarly, the sequence fragment after site-directed integration of the exogenous target gene 2 into GBB003 mother cells is shown as SEQ ID No.10.

[0270] The above-mentioned breakpoint upstream primer F1 (SEQ ID No. 7): AGACCAGCCTCAGATGTCACAC and the target gene 2 downstream primer R2 (SEQ ID No. 11): CCTTAGAATCCTGCTCGGTGA on the exogenous target gene 2 insertion fragment were used to amplify the genome of the cell line in which the exogenous target gene 2 was site-specifically integrated into the GBB003 mother cell, and a fragment that met the theoretical size (3486 bp) was obtained, as shown in Figure 11 (corresponding band in lane GBB003-1G2).

[0271] The amplified bands were excised and recovered, purified, and multiple pairs of primers were designed for DNA sequencing. The sequence after splicing the sequencing results was shown in SEQ ID No. 10, which was consistent with the theoretical sequence (sequence SEQ ID No. 10), thereby confirming that the exogenous target gene 2 was correctly integrated into the corresponding position of the marker gene expression fragment. Among them, the sequence fragment after the integration of the exogenous target gene 2 contains attL. In the sequence fragment after the integration of the exogenous target gene 2, the part before attL is the exogenous sequence, and the part after attL is the endogenous sequence. The exogenous sequence contains the target gene 2.

[0272] Similarly, the sequence fragment after the site-directed integration of the exogenous target gene 3 in the GBB003 mother cell is shown in SEQ ID No. 31; using the above-mentioned breakpoint upstream primer F1 (SEQ ID No. 7): AGACCAGCCTCAGATGTCACAC and the target gene 3 downstream primer R2 (SEQ ID No. 32): CTCGACGCCATCCACGTACCAG on the exogenous target gene insertion fragment; amplifying the genome of the GBB003 mother cell site-directed integration of the exogenous target gene 3 cell line, a fragment that meets the theoretical size (2713 bp) is obtained, as shown in Figure 12 (the corresponding band in lane GBB003-1D6). This confirms that the exogenous target gene 3 is correctly integrated into the corresponding position of the marker gene expression fragment. Among them, the sequence fragment after the integration of the exogenous target gene 3 contains attL. In the sequence fragment after the integration of the exogenous target gene 3, the part before attL is the exogenous sequence, and the part after attL is the endogenous sequence. The exogenous sequence contains the target gene 3.

[0273] Similarly, after the target gene 4 was integrated into the GBB003 mother cells, two cell lines, Y5E6 and Y6F10, were selected through AI prediction as the exogenous target gene expression vector particles for 90-day subculture. The stability of the subcultured cell viability is shown in Figure 13, and the stability of the cell proliferation rate is shown in Figure 14.

[0274] The glycoforms of the integrated Y5E6 and Y6F10 cell products are shown in Figure 15 , the charge heterogeneity comparison results are shown in Figure 16 , and the sodium dodecyl sulfate-polyacrylamide gel electrophoresis results are shown in Figure 17 . These results demonstrate that GBB003 mother cells can integrate a 7688 bp exogenous target gene expression fragment and stably express the target gene within the 7688 bp exogenous target gene expression fragment.

[0275] Similarly, the concentration of the target gene 5 expression antibody in a six-well batch culture of GBB003 mother cells after integration of target gene 5 is shown in Figure 18, and the results of sodium dodecyl sulfate-polyacrylamide gel electrophoresis are shown in Figure 19. These results demonstrate that the GBB003 mother cell line can integrate a 12256 bp exogenous target gene expression fragment and stably express the target gene within the 12256 bp exogenous target gene expression fragment.

[0276] Combined with the expression and passage stability of target gene 1, target gene 2, target gene 3, target gene 4, or target gene 5 integrated into the GBB003 mother cell line, it was found that the site-specific integration vector particles screened by the method of this example can be used to integrate exogenous target gene expression fragments of different lengths.

[0277] Example 3: Screening of S4F6 sites and preparation of site-specific integration vector particles and exogenous target gene expression vector particles

[0278] This example screens the S4F6 site and prepares site-specific integration vector particles and exogenous target gene expression vector particles according to the protocol of Example 2. The difference from Example 2 is that:

[0279] The specific steps of differentiation are:

[0280] (1) The consumables in the specific embodiment include: Neon Resuspension Buffer R (ThermoFisher), electroporation solution E (ThermoFisher), recovery medium EX-CELL Advanced CHO Fed-batch Medium (Sigma-Aldrich), expansion medium EX-CELL Advanced CHO Fed-batch Medium plus 200 μg / ml hygromycin (ThermoFisher) and 1% GlutaMAX (ThermoFisher), shake flask medium EX-CELL Advanced CHO Fed-batch Medium plus 1% GlutaMAX, basal medium in the fed-batch medium used 100% EX-CELL Advanced CHO Fed-batch Medium, and feed medium 4% Cell Boost 7a / 7b (HyClone).

[0281] (2) In step 6, the high-fluorescence cell line was subjected to a 90-day passage stability study. The cells were passaged every 3 or 4 days, with a cell density of 3 × 10 5 / ml at 4 days and 5 × 10 5 / ml at 3 days. The fluorescence value of the cell line was monitored during the passage. Cells with minimal fluorescence changes and stable cell growth were identified as stable high-fluorescence cells. This cell line was named GBB-01 cells. GBB-01 cells were the parent cell line and the preferred site-directed integration vector particles. Figure 19 shows that GBB-01 cells maintained stable growth after 90 days of passage, and Table 1 shows that protein expression in GBB-01 cells was stable.

[0282] Table 1

[0283] (3) In step 7, the target gene 6 recombinant plasmid contains an exogenous target gene 6 expression fragment having the target gene 6 between attB. The target gene 6 fragment is cloned into the exogenous target gene 6 expression fragment (as shown in FIG21 ) for electroporation verification, and co-transfected with the Bxb1 integrase plasmid (as shown in FIG22 ) into the mother cell line, wherein the prokaryotic resistance gene is used for prokaryotic plasmid construction and screening, homology arm 1 (attP in the marker gene expression fragment in the mother cell GBB-01) and homology arm 2 (attB in the exogenous target gene expression fragment) are used to produce recombination replacement with the screened mother cell line, promoter 1 is used to drive the expression of the hygromycin gene after recombination with the mother cell, promoter 3 is used to drive the expression of the target gene, and is inserted with BamH I and Xho I enzymes. The Bxb1 integrase plasmid structure is shown in FIG22 , wherein the prokaryotic resistance gene is used for prokaryotic plasmid construction and screening, promoter 1 is used to drive the expression of the Bxb1 recombinase gene, and promoter 2 is used to drive the expression of the prokaryotic resistance gene.

[0284] The length of the exogenous gene fragment inserted into the endogenous gene in GBB-01 mother cells was 7545 bp, the length of the marker gene expression fragment in the exogenous gene was 3815 bp, and the length of the remaining exogenous gene fragment was 3730 bp. The length of the exogenous target gene expression fragment containing target gene 6 was 2227 bp.

[0285] The target gene 6 fragment has the sequence shown in SEQ ID No.13.

[0286] (4) In step 9, the non-fluorescent cells in the monoclonal cells (i.e., the expression vector particles of the exogenous target gene 6) were further expanded (using expansion medium), expanded to the shake flask stage culture (using shake flask medium), and the expression level was evaluated by fed-batch culture (using fed-batch medium). The expression level comparison results of the expression vector particles of the exogenous target gene 6 with site-specific integration (denoted as GBB-01-H8 cells) and the cell line with three copies of the randomly integrated target gene 6 cultured under the same conditions are shown in Figure 23, where "targeted integration single copy" refers to GBB-01-H8 cells.

[0287] (5) In step 10, GBB-01-H8 cells were passaged for approximately 90 days for stability assessment, followed by 7 days of batch culture (cultivation with basal medium only). The results showed that GBB-01-H8 cells exhibited passage stability and high expression levels, both in terms of expression level ( Figure 24 ) and specific productivity ( Figure 25 ).

[0288] The quality attributes of GBB-01-H8 cells were studied and compared with the expression products of cell lines obtained by random integration methods in the following aspects:

[0289] According to the sugar spectrum comparison results in Figure 26 and the charge heterogeneity comparison results in Figure 27, the results show that the quality attributes of the targeted integration cell product (i.e., GBB-01-H8 cells) and the control cell product (i.e., the cell line expression product obtained by the random integration method) are similar, with no significant differences.

[0290] Next-generation whole-genome sequencing of GBB-01 cells, corresponding to GBB-01-H8 cells that highly express both the fluorescent protein EGFP and target gene 6, revealed the following annotation for the integration site in CHO-K1: NC_048596:34375010, located in the intergenic region. This integration site, designated S4F6, represents the insertion site of a foreign gene into an endogenous gene.

[0291] The sequence fragment of NC_048596 after integration of target gene 6 is shown as SEQ ID No.14.

[0292] The GBB-01-H8 cell genome was amplified using the above-mentioned breakpoint upstream primer F (SEQ ID No. 15): tgcagccacactaagatgg and the target gene 6 downstream primer R (SEQ ID No. 16): ccagagacaggctcaaggac on the exogenous target gene 6 expression fragment to obtain a fragment that met the theoretical size (2876 bp), as shown in Figure 28 (corresponding band in the GBB-01-H8 lane).

[0293] The amplified bands were recovered by gel excision and purified, and multiple pairs of primers were designed for DNA sequencing. The sequence after splicing the sequencing results was shown in SEQ ID No. 14, which was consistent with the theoretical sequence, thus confirming that the target gene 6 was correctly integrated into the corresponding position of the marker gene expression fragment.

[0294] Comparative Example 1. Performance comparison of the exogenous target gene expression vector microparticles prepared by the present method and the exogenous target gene expression vector microparticles constructed by CRISPR / Cas9

[0295] (1) Construction of Cas9 editing plasmid:

[0296] An sgRNA (SEQ ID No. 17: gatagtcacccacaccagaggttttagagctagaaatagcaagttaaaataaggctagtccgttatcaacttgaaaaagtggcaccgagtcggtgc; GC content = 55%, no secondary) was designed to target the 1678-1744 base interval of the nucleic acid fragment represented by SEQ ID No. 12. Double-stranded fragments were then merged and annealed. The reaction system consisted of 5 μL each of forward and reverse oligos, 5 μL of 10× NEB buffer 2, and water to 50 μL. The system was incubated at 95°C (measured with a thermometer) for 5 minutes, then cooled to 25°C. The resulting plasmid was ligated with the pDonor CRISPR / Cas9 vector digested with PvuI. The ligation product was transformed into a DH5α competent culture medium and plated on ampicillin-resistant LB plates. Single colonies were picked the next day, and the plasmid was isolated. Finally, the correct Cas9 encoding plasmid was obtained by first-generation sequencing verification.

[0297] (2) Construction of donor plasmid:

[0298] The donor plasmid consists of a 5' homology arm, an anti-PD1 monoclonal antibody encoding gene, a hygromycin encoding gene and a 3' homology arm connected in series on a pUC19 plasmid.

[0299] (3) Electroporation transfection of Cas9 encoding plasmid and anti-PD1 donor plasmid:

[0300] Suspension Chinese hamster ovary cancer cells, namely CHO-K1 cell line, were used. The cell density was adjusted to 5×105 cells / mL the day before transfection. On the second day, transfection was performed using electroporation with the conditions of voltage 1575V, pulse duration 10ms, and number of pulses 3. A 6-well plate for cell culture was used for transfection, 3×106 cells, 2mL of culture medium, and 15μg of the required plasmid. After transfection, the cells were cultured in an incubator at 37°C and 5% CO2. Among them, adding Cas9 encoding plasmid and anti-PD1 donor plasmid for co-transfection can obtain a pool of site-specific integrated cells, and adding only anti-PD1 donor plasmid for transfection can obtain a pool of randomly integrated cells expressing anti-PD1.

[0301] (4) Pressure screening and enrichment of the cell pool with site-specific integration.

[0302] (5) 5' / 3' junction PCR was used to preliminarily identify the site-specific integrated cell pool, and the amplified target fragment was sequenced. After confirming that the site-specific integrated cell pool was obtained, the supernatant was collected.

[0303] (6) Western blot was used to detect the expression of anti-PD1 antibodies in the site-specific integrated cell pool, confirming that the site-specific integrated cell pool was obtained. After 3 weeks of drug addition screening, the random integrated cell pool showed almost no expression of anti-PD1 antibodies by Western blot, while the site-specific integrated cell pool had a higher expression of anti-PD1 antibodies.

[0304] (7) Obtaining monoclonal cell lines by limiting dilution method: The previously site-directed integration cell pool was used to obtain monoclonal cell lines by limiting dilution method. The cell pool was plated on a 96-well plate with a cell density of 0.8 cells / well. The plate was checked on the same day to confirm the monoclonality. When the cells were grown for about 5 weeks, 1×106 monoclonal cells were aspirated and genomic DNA was extracted. The 5' / 3' junction PCR test was performed again using the same method as step (5). The positive cell lines obtained by junction PCR test were further subjected to Western blot test to determine whether the anti-PD1 antibody was correctly expressed.

[0305] (8) Evaluation of the expression of anti-PD1 antibodies in site-specific integrated cell lines: The expression of anti-PD1 antibodies in site-specific integrated cell lines is approximately 2.3 times that of random integrated cell lines. Site-specific integrated cell lines can also be understood as targeted integrated cell lines, where the target site is a fixed site.

[0306] Table 2

[0307] According to the steps of Example 3, exogenous target gene expression vector microparticles for GBB-01 cells with the anti-PD1 antibody editing gene as the target gene and the S4F6 site (NC_048596:34375010) as the fixed integration site were constructed. The cell line with a single copy of the anti-PD1 antibody editing gene randomly integrated was used as a control. The expression levels of the anti-PD1 antibody in the 6-well plate batch culture in the basal medium are shown in Table 2. The single copy random integration cell line expressed 285 mg / L of anti-PD1 antibody in the 6-well plate batch culture, while the expression level of the exogenous target gene expression vector microparticles for GBB-01 cells with the anti-PD1 antibody editing gene as the target gene and the S4F6 site as the fixed integration site was 705 mg / L, which was 2.47 times the expression level of the single copy random integration cell line, which was superior to the exogenous target gene expression vector microparticles constructed by CRISPR / Cas9.

[0308] Example 4: Screening of the S2C4 site and preparation of site-specific integration vector particles and exogenous target gene expression vector particles

[0309] This example screens the S2C4 site and prepares site-specific integration vector particles and exogenous target gene expression vector particles according to the protocol of Example 2. The difference from Example 2 is that:

[0310] The specific steps of differentiation are:

[0311] (1) The consumables in the specific embodiment include: Neon Resuspension Buffer R (ThermoFisher), electroporation solution E (ThermoFisher), recovery medium EX-CELL Advanced CHO Fed-batch Medium (Sigma-Aldrich), expansion medium EX-CELL Advanced CHO Fed-batch Medium plus 200 μg / ml hygromycin (ThermoFisher) and 1% GlutaMAX (ThermoFisher), shake flask medium EX-CELL Advanced CHO Fed-batch Medium plus 1% GlutaMAX, basal medium in the fed-batch medium uses 100% EX-CELL Advanced CHO Fed-batch Medium plus 1% GlutaMAX, and feed medium is 4% Cell Boost 7a / 7b (HyClone).

[0312] (2) In step 1, the cells in the stable cell pool are screened for high fluorescence cells, and the 1% cells with the highest fluorescence are obtained.

[0313] (3) In step 2, the growth of the enriched cells was observed at any time. When the coverage rate reached more than 50%, the cells were expanded in the manner of 96-well plate to 6-well plate.

[0314] (4) In step 6, the monoclonal fluorescent site-specific integration vector particles inserted into the endogenous gene are amplified and cultured for stable passage. Passaging is performed every 3 or 4 days, with a cell density of 3 x 105 / ml at 4 days and 5 x 105 / ml at 3 days. The fluorescence value of the cell line is monitored during passage, and the total passage is cultured for 90 days. The highly fluorescent cell line is retained as the preferred site-specific integration vector particle, designated GBBc001 cells, and used as the mother cell for subsequent site-specific integration of the exogenous target gene.

[0315] (5) In step 7, the target gene 3 (fusion protein) or target gene 1 (monoclonal antibody) fragment is cloned into the plasmid to construct a target gene 3 recombinant plasmid containing RMCE (as shown in FIG7 ) and a target gene 1 recombinant plasmid containing RMCE (as shown in FIG6 ). The target gene 3 recombinant plasmid contains the exogenous target gene 3 expression fragment with target gene 3 between attB. The target gene 1 recombinant plasmid contains the exogenous target gene 1 expression fragment with target gene 1 between attB.

[0316] (6) In step 8, the proportion of non-fluorescent cells in the cell pool is counted, and after the cell viability recovers to above 90%, the cells are enriched and cultured. After the viability recovers to 90%, the cells are plated using the limiting dilution method to monoclonize the cell pool.

[0317] (7) In step 9, 14 days later, the confirmed non-fluorescent and normally growing monoclonal exogenous target gene expression vector particles (i.e., monoclonal cells) were expanded to a 24-well plate for further amplification (Advanced+1% Glutamax) and then expanded to the shake flask stage culture (Advanced+1% Glutamax). About 1 month later, the cell line can enter the feeding stage and be cultured by fed-batch culture (Advanced+1% Glutamax for the first three days, 3% 7a and 0.3% 7b (cytiva) on the 3rd and 5th days, 5% 7a and 0.5% 7b (cytiva) on the 7th, 9th, 11th, and 13th days). iva), sugar was supplemented based on cell sugar consumption on culture days 3, 5, 7, 9, 11, and 13, and expression levels were assessed. The results of a comparison of expression levels of exogenous target gene 3 expression vector microparticles with a single copy of target gene 3 site-specifically integrated and randomly integrated cell lines cultured under the same conditions are shown in FIG29 (in this figure, GBBc001 represents exogenous target gene 3 expression vector microparticles). The results of a comparison of expression levels of exogenous target gene 1 expression vector microparticles with a single copy of target gene 1 site-specifically integrated and randomly integrated cell lines cultured under the same conditions are shown in FIG29 (in this figure, GBBc001 represents exogenous target gene 1 expression vector microparticles).

[0318] (13) In step 10, in order to investigate the stability of the cell product clone corresponding to the exogenous target gene 3 expression vector microparticles, the product clone was continuously passaged for about 90 days for stability assessment. During the continuous passage process, the exogenous target gene 3 expression vector microparticles of the 1st, 13th and 26th generations were batch cultured for 7 days (Advanced+1% GlutaMAX), and sugar was supplemented on the 3rd, 5th and 7th days of culture according to the sugar consumption of the cells. As shown in Figure 31, the results showed that the exogenous target gene 3 expression vector microparticles showed passage stability and high expression level of target gene 3. At this time, the exogenous target gene 3 expression vector microparticles were called GBBc001-14 cells. The exogenous target gene 1 expression vector microparticles were called GBBc001-3 cells.

[0319] Next-generation whole-genome sequencing of GBBc001, the mother cell line corresponding to GBBc001-14 cells, which highly express both the fluorescent gene EGFP and target gene 3, revealed the following annotation for the integration site on CHO-K1: NC_048595:28250698, located in intron 3of11 of Mtf1. The insertion site of the exogenous gene within the endogenous gene in the mother cell GBBc001 is NC_048595:28250698, designated as the S2C4 site.

[0320] The sequence fragment after integration of target gene 3 is shown as SEQ ID No.19.

[0321] The GBBc001-14 cell genome was amplified using the breakpoint upstream primer F1 (SEQ ID No. 20: gtctgggcatggtggttgca, 131 bp upstream of the breakpoint in the genome) and the target gene 3 downstream primer R1 (SEQ ID No. 21: gaggacactccacgcaaca, approximately 600 bp downstream of the translation start site of the target gene 3 within the exogenous target gene expression fragment. A fragment that met the theoretical size (2113 bp) was obtained (131 bp in the genome, 696 bp in the original vector integrated into the genome, and 1286 bp in the product sequence after site replacement), as shown in Figure 32 (corresponding bands in lane GBBc001-14).

[0322] The amplified bands were recovered by gel excision and purified, and multiple pairs of primers were designed for DNA sequencing. The sequence after splicing was shown as SEQ ID No. 19, which was consistent with the theoretical sequence, thus confirming that the target gene 3 was correctly integrated into the corresponding position of the marker gene expression fragment.

[0323] Similarly, the sequence fragment of the cell line GBBc001-3, which highly expresses both the fluorescent gene EGFP and the target gene 1, after the target gene 1 is integrated is shown in SEQ ID No. 22.

[0324] The GBBc001-3 cell genome was amplified using the breakpoint upstream primer F3 (SEQ ID No. 20): gtctgggcatggtggttgca and the target gene 1 downstream primer R3 (SEQ ID No. 23): actcgcctctattgaagct on the exogenous target gene 1 expression fragment to obtain a fragment that met the theoretical size (3198 bp), as shown in Figure 33 (corresponding band in lane GBBc001-3).

[0325] The amplified bands were recovered by gel excision and purified, and multiple pairs of primers were designed for DNA sequencing. The sequence after splicing the sequencing results was shown in SEQ ID No. 22, which was consistent with the theoretical sequence, thereby confirming that the target gene 1 was correctly integrated into the corresponding position of the marker gene expression fragment.

[0326] Example 5: Targeted integration site screening system

[0327] As shown in FIG34 , the fixed integration site screening system is an electronic device and / or a non-transitory computer-readable storage medium;

[0328] The electronic device includes a display, a microscope, a robotic arm, a stage, a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, the steps of the method for screening site-specific integration sites as described in Examples 1 to 4 are implemented.

[0329] The non-transitory computer-readable storage medium stores a computer program thereon, and when the computer program is executed by the processor, the steps of the method for screening site-specific integration sites as described in Examples 1 to 4 are implemented.

[0330] Example 6: Site-directed integration system of exogenous target genes

[0331] As shown in FIG33 , the exogenous target gene site-specific integration system is an electronic device and / or a non-transitory computer-readable storage medium;

[0332] The electronic device includes a display, a microscope, a robotic arm, a stage, a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, the steps of the method for site-directed integration of an exogenous target gene as described in Examples 1 to 4 are implemented.

[0333] The non-transitory computer-readable storage medium stores a computer program thereon, and when the computer program is executed by the processor, the steps of the method for site-directed integration of the exogenous target gene as described in Examples 1 to 4 are implemented.

[0334] It will be easily understood by those skilled in the art that the above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A method for screening site-specific integration sites, characterized in that: The following steps are involved: (1) randomly integrating an exogenous gene containing two recognition sequences and a marker gene between the two recognition sequences into an endogenous gene of a vector particle to obtain a vector particle carrying the marker gene; (2) amplifying and culturing the vector particles carrying the marker gene obtained in step (1) to form a vector particle pool to be screened; (3) performing monoclonal screening on the vector particle pool to be screened obtained in step (2), and retaining monoclonal vector particles whose expression levels of the corresponding marker genes and / or exogenous target genes are higher than a preset threshold value according to the expression levels of the marker genes and / or exogenous target genes during the monoclonal screening, to obtain monoclonal vector particles with high protein expression; (4) Subculturing the high protein expression monoclonal vector particles obtained in step (3), determining the subculture stability of the monoclonal vector particles based on the marker gene expression levels of the vector particles over multiple generations, obtaining high expression monoclonal vector particles with subculture stability, and using the insertion position of the exogenous gene on the endogenous gene as the fixed integration site.

2. The method for screening site-specific integration sites according to claim 1, characterized in that: The length of the marker gene is comparable to that of the exogenous target gene to be integrated; The marker gene has a plurality of optional lengths, preferably 0.75k, 1k, 3k, and 15k; It is preferred to use marker genes of 3k or less and predict the expression levels of exogenous target genes of other lengths through intelligent learning algorithms.

3. The method for screening site-specific integration sites according to claim 1, characterized in that: The marker gene is selected from one or more of a neomycin resistance gene, a thymidine kinase gene, a hygromycin phosphotransferase gene, a dihydrofolate reductase gene, a thymidine kinase gene, a glutamine synthetase gene, an asparagine synthetase gene, a tryptophan synthetase gene, a histidinol dehydrogenase gene, an aminoglycoside phosphotransferase gene, a tryptophan synthetase gene and a fluorescent protein gene; and / or, The carrier particles are used to express one or more of amino acids, polypeptides, proteins, nucleotides, nucleic acids, secretions, and are selected from cells, bacteria, and viruses, and are commonly carrier cells; the carrier cells are eukaryotic cells; optionally, the eukaryotic cells are mammalian cells; optionally, the mammalian cells include Chinese hamster ovary CHO cells and human embryonic kidney HEK293 cells; and / or, The recognition sequence is independently selected from one or more of the following sequences: LoxP sequence, LoxPL3 sequence, LoxP 2L sequence, LoxFas sequence, Lox511 sequence, Lox2272 sequence, Lox2372 sequence, Lox5171 sequence, Loxm2 sequence, Lox71 sequence, Lox66 sequence, FRT sequence, Bxb1 attP sequence, Bxb1 attB sequence, attP sequence, attB sequence, attP-GA sequence and attB-GA sequence.

4. The method for screening site-specific integration sites according to claim 1, characterized in that: Step (3) uses image processing technology to screen monoclonal vector particles, and specific schemes include but are not limited to deep learning target detection model YOLO series and SSD series algorithm models.

5. The method for screening site-specific integration sites according to claim 4, characterized in that: Step (3) predicts the expression level of the marker gene and / or the exogenous target gene by an intelligent algorithm based on the cell image; specifically comprises the following steps: (3-1) imaging the vector particle pool to be screened obtained in step (2) to obtain an image of the vector particle to be screened; preferably, high-throughput scanning imaging is used, and when the marker gene is a fluorescent protein, high-throughput fluorescent scanning imaging is preferably used; (3-2) performing image processing on the carrier particle images to be screened obtained in step (3-1) to obtain images of each carrier particle, and using an intelligent algorithm to predict the expression level of the marker gene and / or the exogenous target gene of the carrier particle according to the image features; the image features preferably used include: cell morphological features and / or optical features, including but not limited to roundness, size, and grayscale of specific cell sites; Preferably, the expression levels of the marker gene and / or the exogenous target gene of the carrier microparticle predicted by the multiple intelligent algorithms are evaluated by using the evaluation standard normalized cumulative loss gain of the ranking algorithm to evaluate the ranking accuracy of the protein expression levels predicted by the multiple intelligent algorithms, and according to the rule that the larger the standard normalized cumulative loss gain, the better the ranking accuracy of the protein expression level, the intelligent algorithm with the best ranking accuracy of the protein expression level prediction is selected; For a specific intelligent algorithm for predicting protein expression, the standard normalized loss cumulative gain calculation formula is as follows: Wherein IDCG is the DCG value of the real ranking of the protein expression of the sample monoclonal vector particles, i represents the predicted ranking number of a monoclonal vector particle in the sample monoclonal vector particles according to the predicted protein expression of the intelligent algorithm, and rel(i) represents the real protein expression of the i-th monoclonal vector particle; k is the sample size of sample monoclonal vector particles; When the predicted protein is a fluorescent protein, rel(i) indicates the fluorescence value; when the predicted protein is a non-fluorescent protein, rel(i) indicates the value of the protein expression amount tested by Western blot, Elisa, mass spectrometry, electrophoresis, etc. (3-3) The carrier particles whose expression levels of the marker gene and / or the exogenous target gene exceed a preset threshold are acquired according to image positioning and moved to a preset position in the culture medium for retention.

6. The method for screening site-specific integration sites according to claim 1, characterized in that: Step (3) uses semi-solid culture medium to screen monoclonal vector particles.

7. The method for screening site-specific integration sites according to claim 1, characterized in that: Step (4) The passage stability of the monoclonal vector particles is determined by an intelligent algorithm; the specific steps are as follows: (4-1) obtaining the marker gene expression levels of the monoclonal vector particles of multiple generations; (4-2) using an intelligent algorithm to evaluate the differences in marker gene expression levels of multiple generations of monoclonal vector particles according to the marker gene expression levels of multiple generations of monoclonal vector particles obtained in step (4-1); (4-3) judging whether the monoclonal vector microparticles have generation stability according to the difference in marker gene expression levels of the multiple generations of monoclonal vector microparticles obtained in step (4-2) according to the principle that the smaller the difference in expression level, the greater the probability of having generation stability.

8. The site-specific integration vector particles obtained by screening the site-specific integration site screening method according to any one of claims 1 to 7, characterized in that: An exogenous gene is inserted into the endogenous gene of the vector particle; the exogenous gene includes two first recognition sequences and a marker gene expression segment for expressing a marker protein and arranged between the two first recognition sequences; The expression level of the marker gene of the site-specific integration vector particle is higher than the preset threshold value; The site-specific integration vector particles have stability in passage.

9. A fixed integration site screening system, characterized in that: The system is an electronic device and / or a non-transitory computer-readable storage medium; The electronic device comprises a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the steps of the method for screening site-specific integration sites according to any one of claims 1 to 7 are implemented; The non-transitory computer-readable storage medium stores a computer program thereon, and when the computer program is executed by a processor, the steps of the method for screening site-specific integration sites according to any one of claims 1 to 7 are implemented.

10. A method for site-specific integration of an exogenous target gene, characterized in that: The following steps are involved: Obtain the site-specific integration vector particle as described in claim 8, and use the first recognition sequence in the exogenous gene in the site-specific integration vector particle to perform homologous recombination, recombining the exogenous target gene expression fragment into the gene of the site-specific integration vector particle while removing the marker gene expression fragment.

11. The method for site-specific integration of an exogenous target gene according to claim 10, characterized in that: The exogenous target gene expression fragment is arranged in a recombinant vector, and the exogenous target gene is site-specifically integrated into the site of the site-specific integration vector particle through the recombinant vector; the recombinant vector is a lentiviral vector, an adenoviral vector, an adeno-associated viral vector, a herpes virus vector, a poxvirus vector, a baculovirus vector, a papillomavirus vector, a papillomatosis virus vector, an integrating phage vector, a non-viral vector, a transposon, a transposase, an integrase substrate or a plasmid.

12. The method for site-specific integration of an exogenous target gene according to claim 10, characterized in that: The length of the marker gene expression segment of the site-directed integration vector particle is comparable to that of the exogenous target gene expression segment.

13. The method for site-specific integration of an exogenous target gene according to claim 10, characterized in that: The method of performing homologous recombination using the first recognition sequence of the site-specific integration fragment of the site-specific integration vector particle is specifically as follows: Constructing an exogenous target gene recombinant fragment containing an exogenous target gene; wherein the exogenous gene recombinant fragment contains two second recognition sequences and an exogenous target gene expression fragment; the exogenous target gene expression fragment is between the two second recognition sequences; the exogenous target gene expression fragment includes a promoter for regulating the expression of the exogenous target gene and an exogenous target gene sequence; The exogenous gene is integrated into the site of the site-directed integration vector particle, and the two first recognition sequences on both sides of the marker gene are respectively changed into a third recognition sequence and a fourth recognition sequence; the third recognition sequence is formed by the reaction of the first recognition sequence and the second recognition sequence, and the fourth recognition sequence is formed by the reaction of the first recognition sequence and the second recognition sequence, and the third recognition sequence is different from the fourth recognition sequence.

14. The method for site-specific integration of an exogenous target gene according to claim 13, characterized in that: The third recognition sequence and the fourth recognition sequence will not become the second recognition sequence or the first recognition sequence in the same reaction system.

15. The method for site-specific integration of an exogenous target gene according to claim 13, characterized in that: The third recognition sequence and the fourth recognition sequence are generated by a recombinase, and the recombinase is one of Bxb1 integrase, ΦC31 integrase, Cre recombinase or FLP recombinase; and / or The second recognition sequence is independently selected from one or more of the following sequences: LoxP sequence, LoxPL3 sequence, LoxP 2L sequence, LoxFas sequence, Lox511 sequence, Lox2272 sequence, Lox2372 sequence, Lox5171 sequence, Loxm2 sequence, Lox71 sequence, Lox66 sequence, FRT sequence, Bxb1 attP sequence, Bxb1 attB sequence, attP sequence, attB sequence, attP-GA sequence and attB-GA sequence; and / or, The third recognition sequence and the recognition sequence are independently selected from one or more of the following sequences: LoxP sequence, LoxPL3 sequence, LoxP 2L sequence, LoxFas sequence, Lox511 sequence, Lox2272 sequence, Lox2372 sequence, Lox5171 sequence, Loxm2 sequence, Lox71 sequence, Lox66 sequence, FRT sequence, attL sequence and attR sequence.

16. The method for site-specific integration of an exogenous target gene according to claim 10, characterized in that: The following steps are involved: S1. Obtain a plurality of site-directed integration vector particles as described in claim 8, and use the first recognition sequence in the exogenous gene in the site-directed integration vector particle to perform homologous recombination, so as to recombinantly transfer the exogenous target gene expression fragment to the gene of the site-directed integration vector particle and remove the marker gene expression fragment, thereby obtaining a plurality of candidate exogenous target gene expression vector particles; S2, amplifying and subculturing the multiple candidate exogenous target gene expression vectors obtained in step S1; S21, retaining the monoclonal exogenous target gene expression vector particles whose expression amount of the corresponding exogenous target gene is higher than a preset threshold value according to the expression amount of the exogenous target gene; S22 determines the generation stability of the exogenous target gene expression vector microparticles according to the exogenous target gene expression amount of the multiple generations of monoclonal exogenous target gene expression vector microparticles; The exogenous target gene expression vector particles with high protein expression and stable passage are obtained.

17. The method for site-specific integration of an exogenous target gene according to claim 16, wherein: Step S21 predicts the expression level of the exogenous target gene through an intelligent learning algorithm; The expression level of the exogenous target gene is retained and predicted by an intelligent algorithm based on the cell image; specifically comprising the following steps: S211, imaging the monoclonal exogenous target gene expression vector microparticles obtained in step S21 to obtain an image of a candidate monoclonal exogenous target gene expression vector microparticle; preferably, high-throughput scanning imaging is used; S212, predicting the expression level of the exogenous target gene of the monocloned exogenous target gene expression vector microparticles using an intelligent algorithm according to the image features; preferably, the image features used include: cell morphological features and / or optical features, including but not limited to roundness, size, and grayscale of specific cell sites; Preferably, the expression levels of the exogenous target gene of the monoclonal exogenous target gene expression vector particles predicted by multiple intelligent algorithms are evaluated by using the evaluation standard normalized cumulative loss gain of the ranking algorithm to evaluate the ranking accuracy of the protein expression levels predicted by the multiple intelligent algorithms, and according to the rule that the larger the standard normalized cumulative loss gain, the better the ranking accuracy of the protein expression level, the intelligent algorithm with the best ranking accuracy of the protein expression level is selected for the prediction; For a specific intelligent algorithm for predicting protein expression, the standard normalized loss cumulative gain calculation formula is as follows: Wherein IDCG is the DCG value of the real ranking of the protein expression of the sample monoclonal vector particles, i represents the predicted ranking number of a monoclonal vector particle in the sample monoclonal vector particles according to the predicted protein expression of the intelligent algorithm, rel(i) represents the real protein expression of the i-th monoclonal vector particle; k is the sample size of the sample monoclonal vector particles; When the predicted protein is a fluorescent protein, rel(i) indicates the fluorescence value; when the predicted protein is a non-fluorescent protein, rel(i) indicates the value of the protein expression amount tested by Western blot, Elisa, mass spectrometry, electrophoresis, etc.

18. The method for site-specific integration of an exogenous target gene according to claim 16, wherein: In step S22, the passage stability of the monoclonal exogenous target gene expression vector microparticles is determined by using an intelligent algorithm; the specific steps are as follows: S221, obtaining the expression levels of the exogenous target gene of multiple generations of the monoclonal exogenous target gene expression vector particles; S222, using an intelligent algorithm to evaluate the difference in the expression amount of the exogenous target gene of the multiple generations of monoclonal exogenous target gene expression vector particles according to the expression amount of the exogenous target gene of the multiple generations of monoclonal exogenous target gene expression vector particles obtained in step S221; S223, judging whether the monoclonal exogenous target gene expression vector microparticles have generation stability according to the difference in the expression amount of the exogenous target gene of the multiple generations of monoclonal exogenous target gene expression vector microparticles obtained in step S222, according to the principle that the smaller the difference in the expression amount, the greater the probability of having generation stability.

19. The exogenous target gene expression vector microparticle obtained by screening the method for site-directed integration of the exogenous target gene according to any one of claims 10 to 18, characterized in that: An exogenous target gene is inserted into the endogenous gene of the vector particle; the exogenous gene includes a third recognition sequence, a fourth recognition sequence, and an exogenous target gene expression segment arranged between the third recognition sequence and the fourth recognition sequence for expressing a target protein; The target gene expression level of the exogenous target gene expression vector microparticle is higher than the preset threshold value; The exogenous target gene expression vector microparticles have passage stability.

20. A site-specific integration system for an exogenous target gene, characterized in that: The system is an electronic device and / or a non-transitory computer-readable storage medium; The electronic device comprises a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the steps of the method for site-directed integration of an exogenous target gene according to any one of claims 10 to 18 are implemented; The non-transitory computer-readable storage medium stores a computer program thereon, and when the computer program is executed by a processor, the steps of the method for site-specific integration of an exogenous target gene as claimed in any one of claims 10 to 18 are implemented.

Images