A high-throughput screening method for bidirectional recognition of TCR and pMHC based on cell munching

Abstract

This invention discloses a high-throughput screening method based on cytokinesis for bidirectional recognition of TCR and pMHC. The method involves displaying both pMHC and TCR on the cell membrane, with the TCR simultaneously fused with the MS2 phage capsid protein for capturing barcode and MS2 RNA. The two cell types are co-cultured, and cytokinesis-mediated membrane transfer occurs between cells exhibiting bidirectional TCR and pMHC recognition. Cells displaying pMHC are isolated and collected for single-cell transcriptome sequencing analysis. The sequencing results are grouped according to different pMHCs, and the corresponding TCR is determined using the barcode sequence, thus identifying the TCR that specifically recognizes pMHC bidirectionally. This high-throughput screening method based on cytokinesis for bidirectional specific recognition of TCR and pMHC overcomes, to some extent, the limitations of existing technologies in terms of screening throughput, physiological relevance, and bidirectional matching ability.

Description

Technical Field

[0001] This invention belongs to the field of biotechnology, specifically relating to a high-throughput screening method based on bidirectional recognition of TCR and pMHC using cytokinesis. Background Technology

[0002] T cells are the core effector cells of the adaptive immune system. They specifically recognize antigenic peptides (pMHC) presented by the major histocompatibility complex (MHC) molecule through their surface-expressed T-cell receptors (TCRs). This molecular recognition process directly determines the specificity of the T-cell immune response and its functional fate. However, at the individual level, the diversity of TCR clones is extremely rich; the potential number of TCR clones in the human peripheral T cell repertoire can be as high as 10^60^76. 7 -10 8 Meanwhile, immunogenic pMHCs constitute a large and unresolved molecular library. Therefore, constructing a high-throughput mapping of the correspondence between TCRs and pMHCs is of significant scientific and practical value for revealing the T cell immune recognition mechanism, advancing TCR-T cell therapy and vaccine development, and optimizing TCR-specific computational prediction models.

[0003] However, existing large-scale screening technologies for TCR-pMHC interactions still have many shortcomings. Current mature technologies typically screen for corresponding target pMHCs based on known functionally active TCRs, or screen for TCRs capable of recognizing the antigen based on known highly immunogenic pMHCs. For example, T-scan technology, pMHC tetramer staining technology, and in vitro display systems such as phage or yeast display all rely on the aforementioned one-way screening strategies.

[0004] The above-mentioned technologies have the following limitations: on the one hand, the screening process is highly dependent on prior information and is difficult to use for the systematic discovery of unknown TCRs or unknown antigens; on the other hand, existing screening systems are mostly limited to "one-to-many" or "many-to-one" detection modes and lack the ability to simultaneously achieve "many-to-many" bidirectional parallel interactive screening between large-scale pMHC libraries and large-scale TCR libraries in the same experimental system.

[0005] In addition, recent years have also reported on the use of viral vectors to display pMHCs and combined with single-cell sequencing technology to achieve "many-to-many" identification of TCR-pMHCs. However, the efficiency of viral entry into target cells in this type of technology essentially depends on the binding affinity between the receptor and ligand, and the screened TCRs may not possess optimal biological functional characteristics. Furthermore, because the viral surface lacks interactions with other co-stimulatory or co-receptor-ligand interactions in the natural antigen presentation environment, the screening results may still deviate from the actual in vivo immune recognition process. Moreover, to ensure that only one specific pMHC molecule is displayed on a single viral vector, different pMHCs usually need to be constructed and individually packaged into corresponding viral vectors, which objectively increases the operational complexity and limits the overall throughput of the screening system.

[0006] In summary, there is an urgent need in this field to develop a new high-throughput technology system that can more realistically simulate the immune recognition process under physiological conditions and support efficient bidirectional interactive screening between large-scale pMHC libraries and TCR libraries.

[0007] Cell phagocytosis is the process by which a cell selectively "bites off" a small piece of its own cell membrane from another cell through direct physical contact, during which functional molecules are exchanged. This process plays a crucial role in immune responses, development, and diseases, particularly cancer. Summary of the Invention

[0008] To address the aforementioned shortcomings in the prior art, this invention provides a high-throughput screening method based on bidirectional TCR and pMHC recognition using cytokinesis, achieving screening based on bidirectional specific recognition of TCR and pMHC.

[0009] The cytokinesis utilized in this application refers to the biological phenomenon in which T cells and antigen-presenting cells undergo specific interaction, and through the molecular recognition of TCR and pMHC and the synergistic effects of multiple co-stimulatory or co-receptor-ligand interactions, the two types of cells come into full contact and form a stable immune synapse, thereby triggering the specific directional transfer of TCR, pMHC and other related membrane proteins between cells.

[0010] A high-throughput screening method based on bidirectional recognition of TCR and pMHC using cytokinesis includes the following steps: (1) pMHC was displayed on the cell membrane of a first mammalian cell, and cells displaying different pMHC were mixed to obtain pMHC mixed cells; (2) The TCR and MS2 phage capsid protein were fused and expressed and displayed on the cell membrane of a second mammalian cell. At the same time, barcode and MS2 coding sequences were introduced into the second mammalian cell. The barcode RNA sequences introduced into cells with different TCRs were different. The stem-loop structure RNA of the MS2 domain could be captured by the MS2 phage capsid protein. Cells displaying different TCRs were mixed to obtain TCR mixed cells. (3) The pMHC mixed cells in step (1) and the TCR mixed cells in step (2) are mixed and cultured together. Membrane transfer mediated by cytokinesis occurs between cells that have bidirectional recognition between TCR and pMHC. (4) Cells displaying pMHC were isolated and collected, and single-cell transcriptome sequencing analysis was performed. The sequencing results were grouped according to different pMHCs, and the corresponding TCRs were determined by barcode sequences, thereby identifying the TCRs that specifically recognize pMHCs.

[0011] Preferably, in step (1), when pMHC is displayed on the cell membrane of a first mammalian cell, the introduced nucleotide sequence is in the form of signal peptide-antigen peptide-flexible linker peptide-β2M-MHC-I heavy chain, wherein the signal peptide is used to guide the directional transport of the expressed product to the cell membrane, the antigen peptide is the specific peptide epitope to be presented, β2M is β-2-microglobulin, and the MHC-I heavy chain is a class I major histocompatibility complex heavy chain.

[0012] The membrane display construction strategy described above is not limited to specific species, but can be extended to pMHC-I molecular expression systems from other species, such as humans, mice, non-human primates, and dogs. In specific implementation, the antigenic peptide, β2M, and MHC-I heavy chain sequences can be designed and replaced according to the MHC-I allele characteristics of the target species.

[0013] Preferably, in step (1), the first mammalian cell is selected from the K562 cell line or other cell lines that do not endogenously present pMHC; In step (2), the second mammalian cell is selected from T cells with endogenous TCR knockout or T cell-derived cell lines, such as the Jurkat cell line.

[0014] Preferably, in step (1), the number of different pMHC types in the pMHC mixed cells is ≤100; in step (2), the number of different TCR types in the TCR mixed cells is ≤50.

[0015] Preferably, in step (1), the proportion of each type of cell containing different pMHC in the pMHC mixed cells is ≥1% of the total cells; in step (2), the proportion of each type of cell containing different TCR in the TCR mixed cells is ≥2% of the total cells.

[0016] In other words, in pMHC mixed cells and TCR mixed cells, only a small proportion of cells containing each specific sequence are needed for successful recognition. After mixing multiple different sequences, pMHC mixed cells containing different pMHC sequences are constructed to form a pMHC library, and TCR mixed cells containing different TCR sequences are constructed to form a TCR library. High-throughput bidirectional recognition and screening can be achieved between the two libraries.

[0017] Preferably, in step (1), the coding sequence of the first fluorescent protein is also introduced into the first mammalian cell.

[0018] Preferably, in step (2), when the TCR is fused with the MS2 phage coat protein and displayed on the cell membrane of a second mammalian cell, the introduced nucleotide sequence is in the form of TCR α chain - co-expression mediator sequence - TCR β chain - flexible linker peptide - MCP, wherein the co-expression mediator sequence is used to mediate the co-expression of the TCR α chain and the TCR β chain, the MCP is the MS2 phage coat protein, and the number of MCP domains is one, two or more.

[0019] Preferably, in step (2), when the barcode and MS2 coding sequence is introduced into the second mammalian cell, the introduced nucleotide sequence is in the form of barcode-MS2 or MS2-barcode, the number of MS2 binding sites is one, two or more, and the barcode is a random nucleotide sequence including multiple nucleotide sequences.

[0020] More preferably, in step (2), when the barcode and MS2 coding sequence are introduced into the second mammalian cell, the introduced nucleotide sequence also includes the coding sequence of the second fluorescent protein.

[0021] In step (2), the TCR is fused with the MS2 phage capsid protein and expressed, then displayed on the cell membrane of a second mammalian cell. Simultaneously, barcodes and MS2 coding sequences are introduced into the second mammalian cell. When introducing sequences into the second mammalian cell in these two steps, if there are only a few TCR types (e.g., just a few), the barcodes of known sequences can be introduced one-to-one with the TCR sequences, thus eliminating the need to confirm the one-to-one correspondence between TCR sequences and barcode sequences. However, if there are many TCR types, TCR sequences can be introduced in a mixed manner, and barcode sequences can be randomly introduced. In this case, it is necessary to confirm the one-to-one correspondence between TCR sequences and barcode sequences. During sequencing, in addition to performing single-cell transcriptome sequencing analysis on the isolated and collected cells displaying pMHC, single-cell transcriptome sequencing analysis is also performed on the cells displaying TCRs to confirm the one-to-one correspondence between TCR sequences and barcode sequences, and finally, the corresponding sequence relationship between pMHC and TCR is obtained.

[0022] Preferably, in step (3), the cell ratio of TCR mixed cells to pMHC mixed cells is 1:4~5.

[0023] Beneficial effects of this invention: This application presents a high-throughput screening method for bidirectional recognition of TCR and pMHC based on cytokinesis. This method constructs a high-throughput screening technology based on cytokinesis to achieve bidirectional specific recognition between TCR and pMHC, which to some extent overcomes the limitations of existing technologies in terms of screening throughput, physiological relevance, and bidirectional matching ability. Attached Figure Description

[0024] Figure 1 Construction of the K562-pMHC cell line.

[0025] Figure 2 Construction of the Jurkat cell line with endogenous TCR knockout. Figure 2 In the figure, A represents the successful TCR knockout verified by flow cytometry. Figure 2 B in the figure represents the successful TCR knockout verified by gene sequencing.

[0026] Figure 3 This study aims to detect the specificity of TCR and pMHC intercellular transfer mediated by cytokinesis.

[0027] Figure 4 To detect the specific cytokinesis effect when 2% of the TCR cell bank contained paired TCRs and 1% of the pMHC cell bank contained paired pMHCs, 0% of the paired TCRs were used as the control cell bank.

[0028] Figure 5 To detect the expression level of TCR in the cell membrane of Jurkat-1G4α95LY and Jurkat-1G4α95LY-MCP cells.

[0029] Figure 6 After co-culturing K562-NY-ESO-1 with Jurkat-1G4α95LY and Jurkat-1G4α95LY-MCP, the TCR membrane transfer level on K562-NY-ESO-1 was detected.

[0030] Figure 7 This is an example of an RNA barcode structure with an MS2 domain.

[0031] Figure 8 To achieve bidirectional specific recognition and screening of TCR-pMHC based on cytokinesis and single-cell sequencing technology. Figure 8 In the figure, A represents the proportion of cells in the three groups of K562-NY-ESO-1, K562-p53R175H, and K562-MAGEA10 corresponding to the EGFP detection frequency intervals at different sequencing depths. Figure 8 B in the table represents the EGFP detection frequency of three cell groups: K562-NY-ESO-1, K562-p53R175H, and K562-MAGEA10. Detailed Implementation

[0032] Example 1: Display of pMHC mammalian cell membrane pMHC is displayed on mammalian cell membranes in the form of signal peptide-polypeptide-(G4S)3-β2M-HLA, which is inserted into the pSIN vector. The signal peptide sequence is shown in SEQ ID No. 1; (G4S)3 is a flexible linker peptide with the sequence GGGGSGGGGSGGGGS; and the β2M sequence is shown in SEQ ID No. 2, representing β-2-microglobulin, UniProt ID: P61769.

[0033] Constructing Expressions (p53) R175H / HLA-A0201-pMHC neoantigen plasmid (peptide sequence as shown in SEQ ID No. 3, HLA-A0201 sequence as shown in SEQ ID No. 4, p53 R175H The HLA-A0201-pMHC sequence is shown in SEQ ID No. 5, and the NY-ESO-1 sequence is shown in SEQ ID No. 5. 157-165 / HLA-A0201-pMHC plasmid (peptide sequence as shown in SEQ ID No. 6, NY-ESO-1) 157-165 The HLA-A0201-pMHC sequence is shown in SEQ ID No. 7, and the MAGE-A10 sequence is shown in SEQ ID No. 7. 254-262 / HLA-A0201-pMHC plasmid (peptide sequence as shown in SEQ ID No. 8, MAGE-A10) 254-262 The / HLA-A0201-pMHC sequence is shown in SEQ ID No. 9.

[0034] The pSIN empty vector plasmid was linearized by double enzyme digestion, confirmed by electrophoresis, and then recovered by gel extraction. The linearized vector and insert fragment were ligated using a homologous recombination ligation system, followed by transformation into Trans5α. Single clones were picked, amplified, and sequenced to confirm correct sequence and reading frame. Subsequently, pSIN-pMHC plasmids were extracted. The pSIN-pMHC plasmids were packaged with lentiviruses. Using HEK293T cells as packaging cells, when cell confluence was approximately 80%, the transfer vector and packaging / enveloping plasmids were co-transfected with transfection reagents. 72 h post-transfection, the supernatant was collected to obtain viral fluid. The obtained viral fluid was stably used to infect K562 cell lines (purchased from the Cell Bank of the Chinese Academy of Sciences, catalog number: THu191), and polybrene was added for transduction at MOI=10. To verify the expression of HLA-A02 molecules on the cell membrane, staining was performed using a commercially available HLA-A02 flow cytometry antibody (clone BB7.2, purchased from BioLegend, catalog number: 343307), and the expression was detected by flow cytometry. All transduced cells effectively expressed HLA-A02. Further sorting of HLA-A02-positive cell populations yielded the K562-pMHC cell line stably expressing the corresponding pMHC complex. Flow cytometry results are shown below. Figure 1 As shown.

[0035] Example 2: Construction of Jurkat cell line with endogenous TCR knockout CRISPR sgRNAs for TCR knockout (KO) Jurkat cells were designed using the website http: / / crispor.tefor.net / (TRAC-sgRNA and TRBC-sgRNA sequences are shown in SEQ ID No. 10 and SEQ ID No. 11). Complementary oligonucleotides were annealed to form double-stranded sgRNA inserts. Subsequently, the pSpCas9(BB)-2A-GFP (PX458) plasmid (purchased from Miaoling Plasmid Platform, catalog number: P0107) was linearized by double restriction endonuclease digestion, and the linearized vector gel recovery product was obtained. The annealed sgRNA inserts were then ligated to the linearized PX458 vector using a homologous recombination ligation system. The ligation product was transformed into Trans5α competent cells, ultimately obtaining PX458-TRAC and PX458-TRBC recombinant plasmids. The PX458-TRAC plasmid was transfected into Jurkat cells (purchased from the Cell Bank of the Chinese Academy of Sciences, catalog number: SCSP-513) using electroporation (electroporation kit purchased from Lonza, Cell Line Nucleofector® Kit V, catalog number: VCA-1003). Jurkat cells expressing green fluorescent protein were then isolated by flow cytometry and cultured in 96-well plates. After identifying CD3-negative monoclonal antibodies by flow cytometry (CD3 flow cytometry antibody purchased from BioLegend, cloneOKT3, catalog number: 317317), the PX458-TRBC plasmid was transfected again by electroporation to achieve TCRβ chain knockout. TCRβ knockout monoclonal antibodies were screened by reintroducing the α chain of TCR-1G4 and detecting CD3 expression by flow cytometry (TCR-1G4-α chain sequence shown in SEQ ID No. 12). Finally, gene sequencing confirmed the expected editing of the TRAC and TRBC gene regions; the results are shown in [link to results]. Figure 2 This confirmed the successful construction of Jurkat cells with double knockout of TCRα and TCRβ chains, and named them Jurkat-TCR-KO.

[0036] Example 3: Cytokinesis-mediated TCR and pMHC-specific intercellular transfer In the construction of the TCR plasmid, to improve expression efficiency, the EF-1α promoter in the pSIN lentiviral vector was replaced with the MND promoter (the MND promoter sequence is shown in SEQ ID No. 13). Simultaneously, the constant regions of the human TCR α and β chains were replaced with the corresponding mouse TCR constant regions (the mouse TCR α chain constant region sequence is shown in SEQ ID No. 14, and the mouse TCR β chain constant region sequence is shown in SEQ ID No. 15). The core structure of this plasmid is: MND promoter - human TCR α variable region - mouse TCR α constant region - P2A - human TCR β variable region - mouse TCR β constant region. Lentiviral particles carrying the target TCR were prepared using a second-generation lentiviral packaging system. After transfection, the viral supernatant was collected, centrifuged to remove cell debris, and filtered or concentrated as needed. Subsequently, the viral supernatant was used to transduce Jurkat-TCR-KO cells; these cells lacked endogenous TRAC / TRBC, which reduced the mismatch between exogenous and endogenous TCRs and improved the stable expression of exogenous TCRs. Flow cytometry was used to detect TCR / CD3 surface expression to obtain Jurkat-TCR-KO cells that stably expressed the target TCR.

[0037] Stable expression of NY-ESO-1 targeting in the Jurkat-TCR-KO cell line 157-165 The 1G4α95LY TCR of / HLA-A0201-pMHC (TCR sequence shown in SEQ ID No. 16) was used to obtain the Jurkat-1G4α95LY cell line expressing this TCR by flow cytometry sorting of CD3-positive cells. To verify the specificity of the cytokinesis response, KRAS-targeting cells were stably expressed in the Jurkat-TCR-KO cell line. G12D The 4095A TCR of / HLA-C0802-pMHC was extracted and CD3-positive cells were sorted (the 4095A TCR sequence is shown in SEQ ID No. 17) to obtain the control cell line Jurkat-4095A expressing this TCR.

[0038] To detect cytokinesis-mediated TCR and pMHC-specific membrane translocation, Jurkat-1G4α95LY or Jurkat-4095A cells were compared with cells stably expressing NY-ESO-1. 157-165K562 target cells (K562-NY-ESO-1) containing / HLA-A0201-pMHC were mixed at a 1:5 ratio and co-cultured in 24-well plates at 37°C and 5% CO2 for 45 min. To facilitate flow cytometry differentiation between the two cell types, mtagBFP (blue fluorescent protein, sequence shown in SEQ ID No. 18) was stably expressed in K562-NY-ESO-1 cells, and K562 cells were identified via the Pacific Blue / BFP channel. After co-culture, cells were washed with pre-chilled PBS / 2 mM EDTA and then stained for BFP detection. + CD3 levels and BFP levels on the surface of K562 cells - HLA-AO2 levels on the surface of Jurkat cells. CD3 signaling acquired by K562 cells reflects the membrane translocation of TCR / CD3 from Jurkat cells to target cells, while HLA-AO2 signaling acquired by Jurkat cells reflects the membrane translocation of pMHC from K562 cells to T cells. Results showed that significant bidirectional membrane translocation was only detected under co-culture conditions of Jurkat-1G4α95LY and K562-NY-ESO-1, while NY-ESO-1 was not recognized. 157-165 No significant membrane transfer was observed in the control group Jurkat-4095A of / HLA-A0201-pMHC, indicating that this cytokinesis-mediated membrane transfer is highly dependent on the specific recognition of TCR and pMHC.

[0039] Example 4: Based on cytokinesis, library-to-library screening of at least 50 TCR libraries and 100 pMHC libraries can be achieved in a single reaction. To verify the feasibility of screening for interacting TCRs and pMHCs based on cytokinesis in a single reaction, we transduced two different pMHCs in K562 cells using lentiviruses and constructed a pMHC cell bank with two stable cell populations, including 1% of cells expressing NY-ESO-1. 157-165 K562 cells expressing the HLA-A0201-pMHC complex (mtagBFP positive) and K562 cells expressing HLA-A0201 (mtagBFP negative) at 99% cellular percentage. Consistent with the above procedure, the TCR cell bank consisted of Jurkat-1G4α95LY cells expressing positive TCR at 2% cellular percentage and Jurkat-4095-10 cells expressing negative control TCR at 98% cellular percentage (4095-10 TCR sequence is shown in SEQ ID No. 19); at the same time, a control TCR cell bank consisted of Jurkat-1G4α95LY cells at 0% cellular percentage and Jurkat-4095-10 cells at 100% cellular percentage.

[0040] The pMHC and TCR cell libraries were co-cultured at a 1:5 ratio in 24-well plates at 37°C and 30 rpm for 120 minutes on a horizontal shaker. The CD3 level on the surface of K562 cells was then detected by flow cytometry. Results are as follows: Figure 4 As shown, a small subset of mtagBFP-positive K562 cells exhibited mild cytokinesis (Q2 gate). All CD3-positive K562 cell banks (Q2 gate + Q3 gate) were sorted by flow cytometry, and after expanded culture, the proportion of mtagBFP-positive cells was re-examined. The results showed that NY-ESO-1 expression was present. 157-165 The K562 content of the / HLA-A0201-pMHC complex increased from 1% in the original library to 8.98%, while no enrichment was observed in the control group with 0% positive TCRs. This result indicates that a detectable specific cytokinesis reaction can occur between the TCR library (2% of paired TCRs) and the pMHC library (1% of paired pMHCs), suggesting that library-to-library screening of at least 50 TCR libraries and 100 pMHC libraries can be achieved through a single reaction.

[0041] Example 5: Demonstration of the TCR-MCP fusion protein in mammalian cell membranes The MS2 bacteriophage coat protein (MCP) can specifically recognize and efficiently bind to the stem-loop structure (MS2 stem-loop) in MS2 bacteriophage RNA. By fusing the MCP sequence with the intracellular region of the TCR, high-affinity capture of RNA carrying the corresponding MS2 stem-loop structure can be achieved (MCP sequence and MS2 stem-loop structure sequence are shown in SEQ ID No. 20 and SEQ ID No. 21, respectively).

[0042] The constructed TCR-MCP expression plasmid is driven by the MND promoter, and its core expression framework is: MND promoter-TCRα chain-P2A-TCRβ chain-(G4S)3-MCP-MCP-HA tag. The TCR α chain and TCR β chain are co-expressed via the P2A sequence. The C-terminus of the TCR β chain is further tandemly linked with a flexible linker peptide (G4S)3, two MCP domains, and an HA tag to construct the 1G4α95LY TCR-MCP fusion expression vector. The sequence of the 1G4α95LY TCR-MCP fusion protein is shown in SEQ ID No. 22. The expression plasmid was packaged using a lentiviral system to obtain lentiviral particles carrying the target fusion gene, which were then used to transduce Jurkat-TCR-KO cells. After transduction, cells were cultured and their surface CD3 expression levels were detected by flow cytometry. Cells that successfully reconstructed the TCR / CD3 complex were sorted and enriched based on the CD3-positive signal, resulting in a Jurkat cell line stably expressing the 1G4α95LY TCR-MCP fusion protein, named Jurkat-1G4α95LY-MCP. Further, Jurkat-1G4α95LY cells stably expressing the unfused MCP 1G4α95LY TCR were used as a control, and the TCR expression levels on the cell membrane surface of the two groups were compared by flow cytometry. Results are as follows: Figure 5 As shown, a clear TCR expression signal was detected on the surface of Jurkat-1G4α95LY-MCP cells, indicating that the 1G4α95LY TCR-MCP fusion protein can be effectively transported and displayed on the cell membrane surface. At the same time, compared with the control cells without MCP fusion, the TCR expression level on the membrane surface was only slightly decreased, indicating that MCP fusion has little effect on the efficiency of TCR membrane loading.

[0043] Example 6: Cell membrane transfer of TCR-MCP fusion protein mediated by cytokinesis Jurkat cells stably expressing the 1G4α95LY TCR-MCP fusion protein (Jurkat-1G4α95LY-MCP) and Jurkat cells stably expressing the non-fused MCP 1G4α95LY TCR (Jurkat-1G4α95LY) were used as the experimental and control groups, respectively, and were compared with cells stably loaded with NY-ESO-1 157-165K562 target cells (K562-NY-ESO-1) containing the HLA-A0201-pMHC complex were co-cultured at a 1:5 ratio for 45 min. To facilitate accurate differentiation between effector and target cells in subsequent flow cytometry analysis, mtagBFP fluorescent protein was pre-stably expressed in K562-NY-ESO-1 cells, enabling specific recognition in the Pacific Blue channel. After co-culture, cells were collected and stained with surface antibodies. Gated analysis was performed using forward scatter, side scatter, and BFP signals. The intensity of CD3 signal on the surface of BFP-positive K562 cells was detected to assess the level of TCR / CD3-related membrane component transfer from Jurkat cells to K562 cells mediated by cytokinesis. Results are as follows: Figure 6 As shown, compared with the control group Jurkat-1G4α95LY cells, Jurkat-1G4α95LY-MCP cells expressing the 1G4α95LY TCR-MCP fusion protein were still able to specifically recognize NY-ESO-1. 157-165 The MCP fusion mediated significant cytokinesis-related membrane translocation, indicating that the MCP fusion did not impair the TCR's ability to recognize the target pMHC. At the same time, its cytokinesis efficiency was only slightly lower than that of the control TCR without MCP fusion, indicating that the MCP fusion had little impact on the TCR-mediated intermembranous translocation function.

[0044] Example 7: Construction of RNA barcodes with MS2 domains To enable the capture and intercellular transfer of characteristic information by the TCR-MCP fusion protein, this invention constructs an RNA barcode containing an MS2 domain. The design logic of this barcode system is based on the following technical principles: 1. Minimizing the mRNA sequence by reducing the length of non-core functional regions to significantly reduce its molecular weight while maintaining barcode integrity, thereby effectively reducing the impact of steric hindrance on intracellular macromolecular interactions. 2. Selecting the EF-1α promoter with high transcriptional activity to drive target gene expression, ensuring that the barcode molecule maintains a high steady-state expression level in the host cell. 3. Integrating the fluorescent protein coding sequence into the expression framework, utilizing its luminescent properties as a reporter gene, facilitating the screening and identification of successfully transduced positive cell populations by flow cytometry. 4. To avoid adverse effects of the barcode structure on the spatial conformation and fluorescence function of the fluorescent protein, a flexible linker peptide, preferably a (4S)3 flexible linker peptide composed of glycine and serine, is placed between the barcode and the fluorescent protein, thereby ensuring the normal functioning of the fluorescent protein. 5. By introducing multiple tandemly repeated MS2 domains into a single RNA barcode sequence, the binding site density of MCP is increased, thereby significantly enhancing the binding efficiency and detection signal-to-noise ratio between the two.

[0045] The RNA barcode expression vector constructed in this invention can be divided into two configurations: barcodes located at the 5′ end and 3′ end of the transcript. Taking the 5′ end barcode configuration as an example, its expression framework is as follows: Figure 7 As shown, the barcode is (N8)-(G4S)3-EGFP-(MS2)6, where barcode (N8) is an 8 nt random nucleotide barcode sequence, (G4S)3 is a flexible linker peptide, and the EGFP sequence is shown in SEQ ID No. 23. Six MS2 binding sites are tandemly connected at the end for subsequent specific recognition with the MCP system (the barcode (N8)-(G4S)3-EGFP-(MS2)6 plasmid DNA sequence is shown in SEQ ID No. 24). Correspondingly, the expression frame with the 3′ barcode configuration is EGFP-(MS2)6-Barcode (N8), which places the random barcode sequence at the end of the transcript, thereby constructing two RNA barcode modules with different spatial layouts (the EGFP-(MS2)6-Barcode (N8) plasmid DNA sequence is shown in SEQ ID No. 25). Based on the above design, oligonucleotide fragments containing random barcode (N8) regions were synthesized and inserted into a pre-constructed vector backbone using molecular cloning methods to form plasmid libraries with diverse barcode sequences. The resulting barcode plasmid libraries were then packaged with lentiviruses to prepare viral libraries for infecting Jurkat cell lines. The multiple infection index (MOI) was controlled at 0.3 during infection to ensure that each cell integrates only a single barcode vector, thus establishing a one-to-one correspondence between the barcode and the cell. Cells were cultured post-infection, and positive cells were detected and enriched based on the EGFP reporter signal in the vector. Ultimately, a stable Jurkat chassis cell line expressing RNA barcodes was obtained, providing a foundation for subsequent RNA barcode-based cell tracking, cluster identification, and membrane transfer detection.

[0046] Example 8: Bidirectional specific recognition and screening of TCR-pMHC based on cytokinesis and single-cell sequencing technology. To demonstrate that RNA barcodes with the MS2 domain can facilitate intercellular transfer based on cytokinesis and to identify single-cell-to-single-cell interactions using single-cell transcriptome sequencing, this invention stably expressed Barcode (N8)-(G4S)3-EGFP-(MS2)6 in Jurkat-1G4α95LY-MCP cells and sorted EGFP-positive cell lines (Barcode DNA sequence: ATCGATCG). These cell lines were then co-cultured with a K562 cell bank stably expressing mtagBFP fluorescent protein at a 1:4 ratio for 45 minutes. The K562 cell bank was loaded with NY-ESO-1...157-165 K562 cells containing HLA-A0201-pMHC (i.e., K562-NY-ESO-1) and loaded with MAGE-A10 254-262 K562 cells containing HLA-A0201-pMHC (i.e., K562-MAGEA10), loaded with p53 R175H K562 cells of HLA-A0201-pMHC (i.e., K562-p53R175H) were mixed in a 1:1:1 ratio.

[0047] K562 cells positive for mtagBFP fluorescent protein were sorted by flow cytometry. After collecting the positive cell population, single-cell 5′ transcriptome sequencing was performed using the 10x Genomics platform. Library construction was performed using the Chromium GEM-XSingle Cell 5′ Reagent Kits v3, following the manufacturer's instructions. To significantly improve the sequencing depth of EGFP-(MS2)6 specific transcripts, this invention constructed a targeted EGFP enrichment library based on nested PCR technology. Specifically, using a single-cell cDNA library as a template, the first round of amplification was performed using primer pair F1 / R1, followed by fragment purification and recovery via agarose gel electrophoresis. Subsequently, using this recovered product as a template, a second round of specific amplification and gel purification was performed using primer pair F1 / R2 (F1 is complementary to the Read 1T sequence, and R1 and R2 are complementary to the EGFP sequence. The F1 sequence is shown in SEQ ID No. 26, the R1 sequence in SEQ ID No. 27, and the R2 sequence in SEQ ID No. 28). The resulting amplified products were subjected to high-throughput deep sequencing to accurately resolve and quantify the intercellular transfer behavior of EGFP-(MS2)6 mRNA in the K562 cell line.

[0048] Data analysis of single-cell transcriptome sequencing results was performed to remove cells expressing the CD3D or CD3G genes, thereby minimizing interference from Jurkat cell contamination. NY-ESO-1 cells were retained. 157-165 / HLA-A0201-pMHC, p53 R175H / HLA-A0201-pMHC、MAGE-A10 254-262Cells expressing only one of the three genes / HLA-A0201-pMHC were identified and divided into three groups based on expression: K562-NY-ESO-1, K562-p53R175H, and K562-MAGEA10. Statistical analysis was performed on the single-cell EGFP detection frequency in these three cell populations. The counting principle was defined as follows: In deep sequencing data, when a single sequence simultaneously contains an EGFP gene-specific fragment and a TSO (Template Switch Oligo) sequence, the single cell corresponding to the 10× Barcode carried by that sequence is considered to have detected an EGFP positive event (the EGFP gene-specific fragment sequence is shown in SEQ ID No. 29, and the TSO sequence is shown in SEQ ID No. 30). The analysis results are as follows: Figure 8 As shown, the data exhibits a certain amount of non-specific background signal in the low-frequency range, manifested as a low frequency of non-specific EGFP detection. However, with the increase in EGFP detection frequency (i.e., increased signal enrichment), the proportion of the K562-NYESO1 cell population significantly increases and becomes dominant, while the signal in the K562-p53R175H and K562-MAGEA10 groups almost disappears in the corresponding high-frequency range.

[0049] These results fully demonstrate that EGFP-(MS2)6 mRNA can achieve highly targeted intercellular transfer through cytokinesis. By combining enrichment frequency analysis with deep sequencing, non-specific background interference can be effectively eliminated, significantly improving the ability to identify specifically interacting cells, thus achieving high-precision screening based on TCR-pMHC bidirectional specific recognition.

Claims

1. A high-throughput screening method based on bidirectional recognition of TCR and pMHC using cytokinesis, characterized in that, Includes the following steps: (1) pMHC was displayed on the cell membrane of a first mammalian cell, and cells displaying different pMHC were mixed to obtain pMHC mixed cells; (2) The TCR and MS2 phage capsid protein were fused and expressed and displayed on the cell membrane of a second mammalian cell. At the same time, barcode and MS2 coding sequences were introduced into the second mammalian cell. The barcode RNA sequences introduced into cells with different TCRs were different. The stem-loop structure RNA of the MS2 domain could be captured by the MS2 phage capsid protein. Cells displaying different TCRs were mixed to obtain TCR mixed cells. (3) The pMHC mixed cells in step (1) and the TCR mixed cells in step (2) are mixed and cultured together. Membrane transfer mediated by cytokinesis occurs between cells that have bidirectional recognition between TCR and pMHC. (4) Cells displaying pMHC were isolated and collected, and single-cell transcriptome sequencing analysis was performed. The sequencing results were grouped according to different pMHCs, and the corresponding TCRs were determined by barcode sequences, thereby identifying the TCRs that specifically recognize pMHCs.

2. The high-throughput screening method for bidirectional recognition of TCR and pMHC based on cytokinesis according to claim 1, characterized in that, In step (1), when pMHC is displayed on the cell membrane of the first mammalian cell, the introduced nucleotide sequence is signal peptide-antigen peptide-flexible linker peptide-β2M-MHC-I heavy chain. The signal peptide is used to guide the directional transport of the expressed product to the cell membrane, the antigen peptide is the specific peptide epitope to be presented, β2M is β-2-microglobulin, and the MHC-I heavy chain is the class I major histocompatibility complex heavy chain.

3. The high-throughput screening method for bidirectional recognition of TCR and pMHC based on cytokinesis according to claim 1, characterized in that, In step (1), the first mammalian cell is selected from the K562 cell line or other cell lines that do not endogenously present pMHC; In step (2), the second mammalian cell is selected from T cells with endogenous TCR knockout or cell lines derived from T cells.

4. The high-throughput screening method for bidirectional recognition of TCR and pMHC based on cytokinesis according to claim 1, characterized in that, In step (1), the number of different pMHC types in the pMHC mixed cells is ≤100; In step (2), the number of different TCRs in the TCR mixed cells is ≤50.

5. The high-throughput screening method for bidirectional recognition of TCR and pMHC based on cytokinesis according to claim 1, characterized in that, In step (1), the proportion of cells containing each different pMHC in the pMHC mixture is ≥1% of the total cells; In step (2), the proportion of each type of TCR containing different TCRs in the TCR mixed cells is ≥2% of the total cells.

6. The high-throughput screening method for bidirectional recognition of TCR and pMHC based on cytokinesis according to claim 1, characterized in that, In step (1), the coding sequence of the first fluorescent protein is also introduced into the first mammalian cell.

7. The high-throughput screening method for bidirectional recognition of TCR and pMHC based on cytokinesis according to claim 1, characterized in that, In step (2), when TCR is fused with MS2 phage capsid protein and expressed and displayed on the cell membrane of a second mammalian cell, the introduced nucleotide sequence is TCR α chain - co-expression mediator sequence - TCR β chain - flexible linker peptide - MCP, wherein the co-expression mediator sequence is used to mediate the co-expression of TCR α chain and TCR β chain, MCP is MS2 phage capsid protein, and the number of MCP domains is one, two or more.

8. The high-throughput screening method for bidirectional recognition of TCR and pMHC based on cytokinesis according to claim 1, characterized in that, In step (2), when the barcode and MS2 coding sequence is introduced into the second mammalian cell, the introduced nucleotide sequence is in the form of barcode-MS2 or MS2-barcode, the number of MS2 binding sites is one, two or more, and the barcode is a random nucleotide sequence including multiple nucleotide sequences.

9. The high-throughput screening method for bidirectional recognition of TCR and pMHC based on cytokinesis according to claim 8, characterized in that, In step (2), when the barcode and MS2 coding sequence are introduced into the second mammalian cell, the introduced nucleotide sequence also includes the coding sequence of the second fluorescent protein.

10. The high-throughput screening method for bidirectional recognition of TCR and pMHC based on cytokinesis according to claim 1, characterized in that, In step (3), the ratio of TCR mixed cells to pMHC mixed cells is 1:4~5.

Images