Non-human mammals containing at least two human leukocyte antigen (HLA) class I alleles in their genomes, methods for producing such mammals, and uses thereof - Patents.com

Abstract

The present invention relates to a non-human mammal comprising in its genome at least two human leukocyte antigen (HLA) class I alleles, said at least two human HLA alleles being functionally expressed to express corresponding MHC I polypeptides on the surface of cells of said mammal to present MHC antigens that provide an antigen-specific CD8+ T cell response by the non-human mammal, and optionally said at least two human HLA class I alleles being functionally expressed to express corresponding MHC I polypeptides on the surface of cells of said mammal to present MHC antigens that provide an antigen-specific CD8+ T cell response by the non-human mammal, (a) at least one human HLA-A allele, and / or (b) at least one human HLA-B allele, and / or (c) at least one human HLA-C allele, and / or (d) (a) and (b), and / or (e)(a) and (c), and / or (f)(b) and (c), and / or (g) A non-human mammal comprising (a), (b), and (c).

Description

[Technical field]

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS This application references European Patent Application No. 22152603.1, filed January 21, 2022, the contents of which are incorporated herein by reference in their entirety for all purposes.

[0002] Sequence Listing This application contains a sequence listing in computer readable format, which is incorporated herein by reference.

[0003] Technical Field The present invention relates to a non-human mammal comprising at least two human leukocyte antigen (HLA) class I alleles in its genome, which can be used as a new tool to isolate human T cell receptors that are selected against specific human HLA class I haplotypes for adoptive T cell therapy of cancer. The present invention further relates to a nucleic acid construct comprising a nucleic acid encoding at least two human HLA class I alleles, an expression vector comprising such a nucleic acid construct, and a respective host cell comprising such an expression vector. The present invention further relates to a method for modifying endogenous HLA alleles of a non-human mammal, and a method for producing a non-human mammal oocyte having a modified target sequence in its genome. The present invention also relates to a method for producing a non-human mammal having a modified target sequence in its genome. The present invention also relates to a method for producing one or more T cell receptors capable of binding to an antigen of interest, and a method for identifying an epitope capable of eliciting an immune response. [Background technology]

[0004] For effective adoptive T cell therapy (ATT), the appropriate selection of tumor-specific antigens (TSAs) is crucial. To completely eliminate tumors without recurrence as much as possible, recurrent somatic driver mutations provide ideal TSAs for ATT. In silico algorithms can predict epitopes as TSAs for ATT, but the probability is low. Thus, the accuracy of prediction algorithms in describing processed neoepitopes still needs to be validated.

[0005] Simultaneous validation of the predicted proteasomal processing of TSA and identification of neoantigen-specific T cell receptors (TCRs) is possible in transgenic mice. One such model, developed by Li et al. (Nature Medicine, Vol. 16, No. 9 (2010), pages 1029-1034), is the ABabDII mouse, which has a complete human TCR locus, knockout of the mouse TCR locus and MHC class I locus, and is transgenic with the human HLA class I allele HLA-A*02:01 (also referred to herein as HLA-A2). Although it is possible to isolate human TCRs from ABabDII mice, this model is a limited source of obtaining only HLA-A2-restricted TCRs, thereby restricting epitope identification to HLA-A2 antigens. In addition, Moore et al., "Humanization of T cell-mediated immunity in mice", Sci.Immunol.6, eabj4026 (2021), report a transgenic mouse called "VelociT" mouse that has been modified with one human MHC-I (HLA-A2) and one MHC-II (HLA-DR2) allele while excluding the respective mouse MHC. Summary of the Invention [Problem to be solved by the invention]

[0006] Thus, there is a need to expand the TCR repertoire for the isolation of HLA-restricted TCRs. It is therefore an object of the present invention to provide tools and methodologies that allow for the expansion of the TCR repertoire. [Means for solving the problem]

[0007] This object is achieved, inter alia, by a transgenic non-human mammal, a nucleic acid construct, a composition, a method and a use having the features of the respective independent claims.

[0008] In a first aspect, the present invention provides a non-human mammal comprising in its genome at least two human leukocyte antigen (HLA) class I alleles (e.g. at least three HLA class I alleles, at least four HLA class I alleles, at least five HLA class I alleles, or more), wherein said at least two human HLA alleles are functionally expressed to express corresponding MHC I polypeptides on the surface of cells of said mammal to present MHC antigens by the non-human mammal that provide an antigen-specific CD8+ T cell response. In some embodiments, the at least two human HLA class I alleles include (a) at least one human HLA-A allele, and / or (b) at least one human HLA-B allele, and / or (c) at least one human HLA-C allele, and / or (d)(a) and (b), and / or (e)(a) and (c), and / or (f)(b) and (c), and / or (g)(a), (b), and (c). In embodiments, the non-human mammal comprises at least three human leukocyte antigen (HLA) class I alleles in its genome, and said at least three human HLA class I alleles include at least one human HLA-A allele, at least one human HLA-B allele, and at least one human HLA-C allele.

[0009] In a second aspect, the present invention provides a nucleic acid construct comprising a nucleic acid encoding at least two human HLA class I alleles, wherein said nucleic acid construct is capable of functionally expressing said at least two human HLA alleles such that said corresponding MHC I polypeptides are expressed on the surface of a cell of said mammal to present MHC antigens providing an antigen-specific CD8+ T cell response by the non-human mammal, and optionally said at least two human HLA class I alleles comprise (a) at least one human HLA-A allele, and / or (b) at least one human HLA-B allele, and / or (c) at least one human HLA-C allele, and / or (d) (a) and (b), and / or (e) (a) and (c), and / or (f) (b) and (c), and / or (g) (a), (b), and (c). In an embodiment, the nucleic acid construct comprises a nucleic acid encoding at least three HLA class I alleles, wherein the at least three human HLA class I alleles comprise at least one human HLA-A allele, at least one human HLA-B allele, and at least one human HLA-C allele.

[0010] In a third aspect, the present invention provides an expression vector comprising a nucleic acid construct as described herein.

[0011] In a fourth aspect, the present invention provides a host cell (e.g., an isolated and / or recombinant host cell) comprising a nucleic acid construct as described herein and / or an expression vector as described herein.

[0012] In a fifth aspect, the present invention provides a method for modifying endogenous HLA alleles of a non-human mammal, the method comprising transducing and / or transplanting a nucleic acid construct and / or expression vector as described herein into a non-human mammal as described herein.

[0013] In a sixth aspect, the present invention provides a method for producing a non-human mammalian oocyte having a modified target sequence in its genome, the method comprising the step of introducing into the non-human mammalian oocyte a nucleic acid construct and / or an expression vector as described herein.

[0014] In a seventh aspect, the present invention provides a method for producing a non-human mammal having a modified target sequence in its genome, comprising the steps of: (a) producing an oocyte as described herein; (b) analyzing offspring delivered by the non-human female host into which the oocyte obtained in (a) was transferred for the presence of the modification.

[0015] In an eighth aspect, the present invention provides a non-human mammal produced and / or modified by a method of producing a non-human mammal having a modified target sequence in its genome as described herein.

[0016] In a ninth aspect, the present invention provides a method of producing one or more T cell receptors capable of binding to an antigen of interest, the method comprising administering to a non-human animal as described herein the antigen of interest and / or a nucleic acid encoding the antigen of interest.

[0017] In a tenth aspect, the present invention provides a T cell receptor obtained or obtainable by each of the methods as described herein.

[0018] In an eleventh aspect, the present invention provides a method for identifying an epitope capable of eliciting an immune response, the method comprising administering to a non-human animal as described herein an antigen of interest suspected of containing an epitope capable of eliciting an immune response.

[0019] In a twelfth aspect, the present invention provides epitopes that can induce immune responses and are identified by the respective methods as described herein.Epitopes can be natural or non-natural.Non-natural application of epitopes can be, for example, ex vivo mixtures that include (i) peptides having the amino acid sequence of epitopes identified by the methods described herein, and (ii) one or more MHC proteins (e.g., one or more MHC class I or class II proteins).

[0020] In a thirteenth aspect, the present invention provides the use of a non-human animal as described herein for generating one or more T cell receptors capable of binding an antigen of interest.

[0021] In a fourteenth aspect, the present invention provides the use of a non-human animal as described herein for identifying epitopes capable of eliciting an immune response.

[0022] In a fifteenth aspect, the present invention provides the use of a non-human animal as described herein to identify one or more T cell receptors capable of binding an antigen of interest.

[0023] The invention will be better understood with reference to the detailed description considered in conjunction with the non-limiting examples and drawings. [Brief description of the drawings]

[0024] [Figure 1]Figure 1 shows a schematic of the pMHC-I / HLA-I antigen processing pathway (this simplified version does not represent all the codomains and adaptor proteins involved as chaperones in the antigen processing pathway): 1. Antigen uptake by the proteasome in the cytoplasm. 2. Protein processing by protease activity inside the proteasome core. 3. TAP protein complex facilitates transport of peptides to the ER. PLC, Erp57, and CRT proteins direct peptide loading onto MHC / HLA. 4. Transport of peptides via the secretory pathway to the cell surface for presentation to CTLs. [Diagram 2] FIG. 2 shows an illustration of tumor-specific neo-antigens that bind specific human HLA class I haplotypes. Step A is a schematic showing recurrent neo-antigens derived from cancer driver point mutations. Step B shows wild-type and mutant constructs screened against 18 class I alleles using affinity prediction software NetMHC Server-DTU, and step C shows that the HLA binding strength of 266 mutants revealed non-HLA-A2 binders. Step D shows the in silico screen that provided the basis for the generation of the ABab.I mouse described herein as an example of a non-human transgenic mammal as described herein with six new HLA class I human alleles. [Diagram 3]Figures 3A-C show the HLA-A2 restricted TRRAP-S722F neoantigen as a target for ATT. Figure 3A shows the TRRAP gene. The asterisk indicates the location of the recurrent mutation in the proline-rich N-terminal domain. Figure 3B shows the nonamer mutant epitope-S722F (with the "aa" changes highlighted in red (F)) and wild-type epitope-S722 (in green (S)) with predicted IC50 (for HLA-A2) of 29 nM and 55 nM, respectively. Targeting the TRRAP-S722F mutation for ATT in melanoma could benefit approximately 911 patients / year. Figure 3C shows the presence of the human TRRAP orthologue in mice. The location of the mutation is underlined. The nonamer epitopes are highlighted in red (black) boxes: TRAP nonamer wild type (SEQ ID NO: 27), TRAP nonamer mutant (SEQ ID NO: 30), TRRAP_human (SEQ ID NO: 28), TRRAP_mouse (SEQ ID NO: 29). [Figure 4]Figure 4A-4F show the generation and characterization of TRRAP S722F TCR in ABabDII mice. Figure 4A shows the TCR detection pipeline with the immunization schedule used for the illustrative example. Figure 4B shows that CD8+ T cells from ABabDII showed S722F peptide reactivity by IFN-γ production (gated on CD3+ cells, n=3). Controls were with wild-type S722 peptide or without peptide (w / o). A representative flow cytometry plot of one of three responder mice is shown. Figure 4C shows that splenic CD8+ T cells (CD4 depleted) from responder mice cultured in vitro produced IFN-γ to 10-9M peptide. Figure 4D shows enrichment of S722F-specific T cells (CD8+IFN-γhigh) by IFN-γ capture assay. Figure 4E shows amplification of dominant and subdominant TCR variable chains in cloning by 5'-RACE RT-PCR. Figure 4F shows, on the left, that dominant Vαβ chains transduced into human T cells were co-cultured with TAP- / - T2 cells (1×105) loaded with titrated peptide nonamers (concentrations: 10-6M to 10-12M) and IFN-γ release was measured by ELISA. Figure 4F shows, on the right, an endogenous processing assay using a cell line naturally expressing mutant TRRAP. IFN-γ levels were measured after overnight co-culture of PBMCs (1×105) with A375 cells (mutated S722F+, NY-ESO+) or SK-Mel-37 (wild-type S722+, NY-ESO+). TCR-ESO recognizes the NY-ESO157-165 epitope. Bar graphs represent mean values ​​± SD (n=2). One representative of three co-culture experiments is shown. [Diagram 5]Figures 5A-5D show the design and construction of the polycistronic ABab.I transgene. Figure 5A shows an overview of the total number of mutational epitopes predicted in silico across 18 HLA alleles. The top 6 HLA alleles (shaded in blue (grey)) reflect the natural human HLA haplotypes (genes present in the ABab.I transgene). Figure 5B shows the three data sets were multiplied to estimate the number of individuals / years with mutations. Figure 5C shows the total number of individuals / years with mutations (likelihood of ATT treatment) was calculated for 175 mutation combinations for 18 alleles using the data in A and B. Red (black) asterisks highlight the top high ranking HLA alleles. Figure 5D shows the polycistronic ABab.I transgene designed to encode 6 HLA alleles separated via 5 different "self-cleaving" viral 2A peptide linkers, driven by a single H-2 Db promoter and polyadenylation signal at the top. Figure 5D shows at the bottom the single single chain construct designs. Each HLA is a chimeric human-mouse fusion single chain containing the α1 and α2 heavy chain cDNA regions from each human HLA allele and the α3, transmembrane, and cytoplasmic domains from the mouse H-2 Db gene fused to human β2m by a glycine-serine polylinker. [Figure 6]Figure 6A-6B show HLA protein expression analysis of polycistronic ABab.I transgenes. Figure 6A shows ABab.I transgenes that are primary constructs (I) and modified versions (II, III) from the initial design. I. Primary transgene with a viral peptide 2A linker separating HLA. II. Modified transgenes that are H-2 Db promoter-driven eGFP gene and CMV promoter-driven HLA. III. Modified transgenes with different polyadenylation signal sequences. Dotted boxes represent modifications to the initial primary construct. Figure 6B shows surface staining of MCA205 cells transiently transfected with the constructs in A. I and II are shown (III not shown). Cells were stained using a pan-HLA-ABC antibody (clone: ​​W6 / 32) 48 hours after transfection. Controls were pMP71-eGFP (GFP vector, measured in the same channel) and LCL-BM14 (lymphoblastoid cells) positive for HLA ABC. One representative of multiple flow cytometry experiments is shown (n>5). [Figure 7] Figure 7A-7B show the mRNA expression profile of polycistronic ABab.I transgenes. Figure 7A shows the ABab.I primary transgene at the top of I. At the bottom of I are polyadenylation (pA) signal modified constructs driven by the CMV viral promoter in H-2 Db or II. Figure 7B shows quantitative RT-PCR analysis of mRNA transcripts from the constructs shown in I and II of Figure 7A. The cDNA region of the ABab.I transgene was amplified from MCA205 cell lysates 48 hours after transfection using primers flanking 2A peptide linkers (P2A, F2A, and C2A). The percentage change of the homeostatic H-2 Db relative to the mouse MHC class I gene (red (black) dotted line) is compared. Bar graphs represent the mean ± SD (n=3). P values ​​are ****, P ≤ 0.0001; ***, P ≤ 0.001; ns, not significant (unpaired two-tailed t test). [Figure 8]Figures 8A-C show HLA expression analysis of a single monocistrone from the ABab.I transgene. Figure 8A shows that in I, a single single chain with or without β2m was cloned from the ABab.I primary transgene. All HLA alleles as monocistrons are driven by their own H-2Db promoter and pA signal. The dotted box represents further modifications from the primary construct. II. The ABab.I transgene expresses six monocistrons (shown in I of Figure 8A) as a single haplotype (5' to 3') by its own regulators. Figures 8B and 8C show MCA205 cells expressing the six single chains, surface stained with or without IFN-γ pretreatment. HLA-A2 protein was stained for reference. Cells were stained 48 hours after gene transfer using pan-HLA-ABC antibody (clone: ​​W6 / 32) in Figure 8B and human-β2 microglobulin antibody (clone: ​​TUE99) in Figure 8C. Representative flow cytometry plots of one of three experiments are shown. [Figure 9]Figures 9A-E show the HLA haplotype expression profile of the ABab.I transgene in the MCA205 cell line. Figure 9A shows MCA205 cells stained using a pan-HLA-ABC antibody. At the top, the six alleles expressing HLA class I haplotypes (shown in II in Figure 8A) were stained 48 hours after transfection. At the bottom, transfected MCA205 cells were enriched for HLA+ cell populations and stained for HLA alleles with or without IFN-γ pretreatment (48 hours before staining). Figures 9B, 9C, and 9D show that after enrichment, HLA-specific antibodies were used to stain MCA205 cells that were or were not pretreated with IFN-γ. Figure 9B shows the HLA-A3 allele stained with HLA-A*03:01 specific antibody (clone: ​​GAP.A3). Figure 9C shows HLA-B7 alleles stained with HLA-B*07:02 specific antibody (clone: ​​BB7.1). Figure 9D shows HLA-C alleles stained with pan-HLA-C antibody (clone: ​​H-5). Figure 9E shows MCA205 / ABab.I stable cell line stained with pan-HLA-ABC antibody after multiple rounds of cell sorting by flow cytometry. One representative example of multiple flow cytometry staining experiments is shown (n>5). [Figure 10] Figure 10 shows that the piggyback transposon targets the ABab.I transgene to oocytes. A schematic diagram of the piggyback (PB) transposon strategy for generating ABab.I mice using pronuclear microinjection technology is shown. The PB-targeted ABab.I-targeted transgene (flanked by ITRs for PB transposase catalysis) has repeats of six chimeric HLA genes, each with its own promoter and 3'-UTR components, constructed sequentially (5' to 3') as shown in Figure 7A II. [Figure 11]Figure 11A-B show the genotyping profiles and breeding schemes of the ABab.I transgenic founders. Figure 11A shows, on the left, the HLA-PCR reactions, which are the PCR genotyping data of the two founder mice (F0), Q4115 and Q4118. HLA-A*03, A*11, B*07, B*15, C*04, and C*07 PCR shows the genomic (g)DNA of the MCA205 / ABab.I stable cell line (lane 1), the ABab.I transgene as plasmid DNA (lane 2), control HO (lane 3), gDNA of a C57BL6 / N mouse (lane 4), the first founder Q4115 with positive ABab.I (lane 5), two negative mice (lanes 6 and 7), and the second founder Q4118 with positive ABab.I (lane 8). Figure 11A shows, on the right, the internal control PCRs, which are 18s rRNA and human TCRαβ PCR, genomic (g) DNA of C57BL6 / N mice (lane a), gDNA of MCA205 / ABab.I stable cell line (lane b), ABab.I transgene as plasmid DNA (lane c), control HO (lane d), gDNA of ABabDII mice (lane e), first founder Q4115 (lane f), two negative mice (lanes g and h), and second founder Q4118 of positive ABab.I (lane i). Figure 11B shows a diagram of the breeding scheme of F0-F4 homozygous mice. F0 was backcrossed with ABabDII for germline transmission. NMRI backcrosses were used for homozygosity testing. [Figure 12]Figures 12A-D show human HLA class I haplotype expression in ABab.I mice. Figure 12A shows HLA staining of peripheral blood cells from C57BL6 / N, ABab.I, and ABabDII mice using β2 microglobulin (β2m) antibody. Two representative plots (2 mice / strain) out of multiple experiments are shown (n>10). Figure 12B shows HLA ABC positive LCL-BM14 (lymphoblastoid cells) stained with β2m antibody. Figure 12C is a histogram showing the mean fluorescence intensity (MFI) of β2m staining in ABabDII and ABab.I mouse strains, as well as the LCL-BM14 cell line. One representative staining out of multiple mice is shown (n>20). Figure 12D shows the MFI of β2m compared between ABabDII and ABab.I mice. Bar graphs represent the mean ± SD (n=5). P values ​​are ***, P≦0.001 (unpaired two-tailed t test). [Figure 13]Figures 13A-D show phenotypic characterization of peripheral blood T cells from ABab.I mice. Figure 13A shows CD8+CD4+ profiling (in CD3+ cells) of peripheral blood cells from young (8-12 weeks) mice by flow cytometry, where the indicated mouse strains were stained with antibodies specific for CD3, CD8, and CD4 chains and analyzed by flow cytometry (gated on CD3+ lymphocytes). Two representative plots (2 mice / strain) out of multiple experiments are shown (n>30). Figures 13B and 13C show absolute T cell counts / μl blood, where Figure 13B shows CD3+CD8+ and Figure 13C shows CD3+CD4+ quantified by flow cytometry using CountBright® beads. Each data point represents one young mouse of the indicated strain. Horizontal intervals in the scatter plots represent the mean ± SD. Summary data for C57BL6 / N (n=9), HHD (n=9), ABabDII (n=32), and ABab.I (n=60) mice. Figure 13D shows the CD8 / CD4 ratio based on absolute counts in peripheral blood in different mouse strains. Bar graphs represent mean ± SD (n=7). P values ​​in Figures 13B, 13C, and 13D indicate ****, P≦0.0001, ns, not significant (unpaired two-tailed t-test). [Figure 14]Figures 14A-D show phenotypic characterization of T cells from lymphoid organs of ABab.I mice. Figure 14A shows spleens and lymph nodes with whole blood from young (8-12 weeks) mice of the indicated mouse strains stained with antibodies specific for CD3, CD8, and CD4 chains and analyzed by flow cytometry (gating on CD3+ lymphocytes). Two representative plots (2 mice / strain) out of multiple experiments are shown (n>10). Figures 14B and 14C show absolute T cell counts / mouse, with Figure 14B showing CD3+CD8+ and Figure 14C showing CD3+CD4+ quantified by flow cytometry using CountBright® beads. Each data point represents one young mouse of the indicated strain. Horizontal intervals in the scatter plots represent the mean ± SD. Summary data for ABabDII (n=7) and ABab.I (n=5) mice. Figure 14D shows the CD8 / CD4 ratio based on absolute numbers in pooled organs in different mouse strains. Bar graphs represent mean ± SD (n = 5). P values ​​in Figure 14B, Figure 14C, and Figure 14D indicate ****, P < 0.0001, ***, P < 0.001, ns, not significant (unpaired two-tailed t-test). [Figure 15]Figures 15A-15C show the comparison of the TCRβ repertoires in ABabDII, ABab.I, and human donors. Figure 15A shows the absolute numbers of unique TCRβ amino acid (aa) clone types in 3×105 (average of mouse strains) or 1.8×105 human CD8+ T cells. Figure 15B shows the clone distributions of TCRβ aa clone types of different sizes: rare, 0 < x ≤ 0.001%, small, 0.001% < x ≤ 0.01%, medium, 0.01% < x ≤ 0.1%, large, 0.1% < x < 1%, highly expanded, 1% < x < 10%. Human donors were excluded from this analysis. Figure 15C shows that TCRβ diversity was calculated using the iChao1 estimator (lower abundance limit of the total number of unique templates within an individual repertoire). The horizontal bars in the scatter plot represent the mean ± SD. The data are from ABabDII (n = 5), ABab.I (n = 5) mice, and humans (n = 2 for Figure 15B, n = 3 for Figures 15A and 15C). The P values in Figures 15A and 15C indicate **, P ≤ 0.01, ns, not significant (unpaired two-sided t-test). [Figure 16] Figures 16A-16B show the usage frequencies of Vβ and Jβ genes in in-frame and out-of-frame TCRβ clone types. Figures 16A and 16B show the usage frequencies of Vβ and Jβ genes of unique TCRβ clone types in ABabDII and ABab.I mice, and human CD8+ T cells. Figures 16A and 16B show the out-of-frame frequencies at the top and the in-frame frequencies at the bottom. The arrangement of Vβ gene (A) or Jβ (B) segments is depicted on the x-axis according to their positions on human chromosomes from 5' to 3'. The dotted line in Figure 16A represents the usage frequency of random Vβ genes (TRBV, 2.1%). The bar graphs represent the mean ± SD. The data are from ABabDII (n = 5), ABab.I (n = 5) mice, and humans (n = 3). *Vβ gene expression was not observed in both ABabDII and ABab.I mice. [Figure 17]Figures 17A-B show CDR3β region analysis in ABabDII, ABab.I, and human donors. Figure 17A shows the distribution of CDR3 lengths of TCRβ clonotypes. Bar graphs show the in-frame frequency of different CDR3β lengths in ABabDII and ABab.I mice, and humans. Figure 17B shows the average CDR3β length compared between ABabDII and ABab.I mice, and humans. Bar graphs represent mean ± SD. Data are from ABabDII (n=5), ABab.I (n=5) mice, and humans (n=3). P values ​​in Figures 17A and 17B indicate ****, P≦0.0001 (unpaired two-tailed t-test). [Figure 18] Figures 18A-B show shared repertoire analysis in ABabDII, ABab.I, and human donors. Figure 18A shows the Jaccard index, which represents a score based on the number of TCRβ clones shared within and between repertoire groups of ABabDII and ABab.I mice, and humans. Bar graphs represent mean ± SD. Data are from ABabDII (n=5), ABab.I (n=5) mice, and humans (n=3). P values ​​indicate ****, P≦0.0001, ***, P≦0.001, ns, not significant (unpaired two-tailed t-test). Figure 18B shows the total number of unique TCRβ clones in ABabDII and ABab.I from pooled data. Five mice per lineage were pooled for analysis. [Figure 19]Figures 19A-C show functional characterization of HLA alleles present in ABab.I mice. Figure 19A shows a list of model TCRs restricted to HLA alleles of ABab.I with known peptide binders. In addition to this list, the HLA-A2-restricted T1367 MAGE-A1 TCR was used as an internal reference in all presentation assays. Figures 19B and 19C show the functional activity of HLA alleles of ABab.I through surface expression and model peptide presentation, where in Figure 19B, the MCA205 cell line stably expresses six HLA alleles, except for HLA-A2, similar to ABab.I mice, and in Figure 19C, ex vivo isolated peripheral blood or lymphoid organ cells were co-cultured with model TCR-transduced or untransduced T cells. After overnight co-culture, cytokine levels were measured by ELISA assay for mouse IFN-γ. pHLA-TCR pairs are listed in Figure 19A. Bar graphs represent the mean ± SD of within-assay duplicates, one representative of multiple co-culture experiments is shown (n>5). [Figure 20]Figure 20 shows CD8+ T cell responses to a panel of human antigens in ABab.I mice. Specific CD8+ T cell responses in ABab.I mice were observed after immunization with a panel of model antigens (listed along the x-axis). Young ABab.I mice were immunized (minimum 2 times) with at least 4 weeks between peptide injections. Seven days after injection, peripheral blood was stimulated in vitro with the same specific peptide used for immunization (right), or with a non-specific peptide (left), or with CD3 / CD28 beads (middle, positive control). Cells were analyzed for CD3, CD8, and intracellular IFN-γ expression by flow cytometry. Dot plots show CD8+IFN-γ+ T cells (gated on CD3+ lymphocytes). The non-specific NY-ESO-APRGPHGGAASGL peptide (SEQ ID NO: 172) was used only for in vitro stimulation. Representative flow cytometry plots of one of the multiple responders are shown (out of 6-7 mice / antigen, 6 mice responded to the KRAS peptide, 3 mice responded to the MAGE-A1 and HPV peptides, and 1 mouse responded to the mCALR, CMV, and MAGE-A12 peptides). [Figure 21]Figure 21A shows the immune response to HLA-A*02-restricted MAGE-A1 in ABab.I mice after vaccination. ABab.I mice were immunized with full-length MAGE-A1 mutated at position 279 (V→D) to disrupt the dominant HLA-A*02-restricted epitope KVLEYVIKV (278-286). On day 7 after the third immunization, peripheral blood was stimulated with MAGE-A1-expressing cells and intracellular staining for IFNg was performed. Figure 21B shows that TCR1 isolated from ABab.I mice recognizes the new HLA-A*02-restricted MAGE-A1 epitope. TCR1 was isolated from ABab.I mice immunized with full-length MAGE-A1 mutated at position 279 (V→D) to disrupt the dominant HLA-A*02-restricted epitope KVL278-286. TCR1-transduced human T cells were co-cultured with K562-A2 cells expressing MAGE-A1, fragments of MAGE-A1 (F1-F3), or loaded with KVLEYVIKV (KVL). F1-F3 span the following fragments of MAGE-A1: F1 - 97, F2 - 98-222, and F3 - 223-309. Figure 21C shows that MAGE-A1-reactive TCR1 recognizes the nonamer GTLEEVPTA. TCR1-transduced human T cells were co-cultured with K562-A2 cells loaded with the predicted nonamer conjugates (Table 18), or loaded with KVLEYVIKV (KVL). [Figure 22] Figure 22 shows flow cytometry analysis. CD8+ T cells were measured in vitro for CD3, CD8, and intracellular IFN-γ expression by flow cytometry. The dot plots in the figure show CD8+IFN-γ+ T cells (gated on CD3+ lymphocytes). As a non-specific peptide, NY-ESO peptide (APRGPHGGAASGL) was used only for in vitro stimulation (as a negative control). Flow cytometry plots of one of the multiple responders are shown (out of 6-7 mice / epitopes, 6 mice responded to KRAS peptide, 3 mice responded to MAGE-A1 and HPV peptides, and 1 mouse responded to mCALR, CMV, and MAGE-A12 peptides). [Figure 23] Figure 23 shows the extent of humanization in the HuTCR mice of the invention. The chimeric HLA in the HuTCR mice promotes the CD8 / CD4 coreceptor, which has been extensively validated to function in human T cells. [Figure 24] Figure 24 shows that the number of CD8 T cells is higher in the HuTCR mice of the present invention compared to single HLA class I mice. The higher number of CD8 T cells in the HuTCR mice of the present invention suggests that the probability of isolating a TCR against any known tumor target is increased and that the TCR-HLA interaction is more optimal compared to that of single HLA class I mice. [Diagram 25] FIG. 25 shows active T cell responses in the blood of multi-HLA HuTCR mice of the present invention immunized multiple times with DNA encoding full-length MAGE-A1 mutated to delete epitopes 278-286. In all 11 mice used in this experiment, T cell responses were examined separately for each of HLA-A*02:01, HLA-A*03:01, and HLA-A*11:01, the HLA combinations HLA-B*07:02 and HLA-B*15:01, and the HLA combinations HLA-C*04:01 and HLA-C*07:02. Mice with positive T cell responses are indicated by circles located above the dashed background line, and mice with negative T cell responses are indicated by circles located below the dashed background line. For every HLA allele or combination of HLA alleles tested, mice were found to have active T cell responses against MAGE-A1 derived epitopes in the blood. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0025] The present invention relates to a new non-human mammal, which comprises at least two or three human leukocyte antigen (HLA) class I alleles in its genome. Optionally, the at least two or at least three human HLA class I alleles comprise at least one human HLA-A allele, at least one human HLA-B allele, and at least one human HLA-C allele. Such transgenic non-humans have been found to provide several advantages. First, it has been found herein that by incorporating such multiple human HLA class I alleles into the genome of non-human mammals, TCRs can be isolated from a wide range of TCR repertoires, in particular TCRs that are non-HLA-A2 restricted. For example, a transgenic mouse carrying two human HLA-A alleles, two HLA-B alleles, and two human HLA-C alleles (HLA-A*03:01, HLA-A*11:01, HLA-B*07:02, HLA-B*15:01, HLA-C*04:01, and HLA-C*07:02, see Experimental Section) represents approximately 80% of the American or European population with respect to its HLA class I gene profile. In another example, a transgenic mouse with the following two human HLA-A alleles, two HLA-B alleles, and two human HLA-C alleles (HLA-A*24:02, HLA-A*33:03, HLA-B*40:01, HLA-B*46:01, HLA-C*01:02, and HLA-C*03:04) represents more than 80% of the ethnic Chinese population in terms of HLA class I gene profile. It is noted herein that such a mouse closely resembles the class I haplotype as in humans, since every (human) individual has six HLA class I alleles. Thus, as a second advantage of providing a non-human mammal with multiple human HLA class I alleles in its genome, the present invention allows the TCR repertoire to be specifically adapted to any given human (sub)population for which a TCR needs to be provided.As also shown by the further examples given herein for sets of HLA class I alleles corresponding to ethnic Japanese or Indian populations, the class I alleles to be integrated into the genome of the transgenic animal can be selected according to their prevalence in the respective human population. Thus, the present invention provides for the first time a "universal toolbox" to mimic any human population of interest in terms of TCR repertoire. In this context, it is also noted that: when multiple such non-human mammals (e.g., 5, 10, or even up to 20 transgenic mice) with the same set of, for example, 6 human HLA class I alleles are used for immunization, the present invention allows, as a third advantage, to more closely represent / mimic the selected human (sub)population, thereby further increasing the chance of obtaining a suitable TCR against a given antigen. Moreover, it is also found herein, as a fourth advantage, that transgenic animals with multiple human HLA class I alleles allow a direct comparison of the strength of the T cell response associated with each of the human alleles against a given antigen (see Example 4), thereby also identifying possible dominant T cell responses. Thus, the transgenic animals of the present invention simultaneously evaluate the T cell responses associated with multiple human HLA alleles, allowing the strength of the T cell responses to be directly compared. This "multiplexity" is a further advantage of the present invention. Finally, as a further advantage, non-human transgenic animals as described herein can be used to detect new epitopes (also HLA-A2 restricted epitopes) that have not yet been identified or that have been theoretically predicted but for which TCRs have not yet been experimentally generated (see Example 2 of the present application).

[0026] It is noted in this context that the illustrative experimental example of such a transgenic animal of the invention, a mouse called "ABab.I mouse" (also referred to herein as "multi-HLA mouse" or "HuTCR mouse"), containing six (additional) human HLA alleles as a single haplotype, was developed starting from the background of an ABabDII mouse with a single human HLA-A2 allele (Li et al., 2010, Nature Medicine, Vol. 16, No. 9 (2010), pages 1029-1034). This "multi-HLA mouse" thus diversifies the set of HLA genes, allowing epitope detection for non-HLA-A2 restricted TCR isolation and a broader TCR repertoire. For this mouse, the six alleles in the ABab.I mouse were selected by in silico screening for putative neoepitopes predicted to bind high-ranking HLA alleles. In the course of the present invention, 23 and 152 high affinity TSAs (IC50<50nM) were predicted by in silico screening from recurrent point mutations (n=266) that bind HLA-A2 and other high frequency class I alleles (n=18), respectively. In particular, epitope immunogenicity was tested in 4 of the 23 HLA-A2 binders in ABabDII mice. Only one of the four immune epitopes elicited a CD8+ T cell (CTL) response. However, this was confirmed to be not endogenously processed when tested with TCRs generated from ABabDII mice. However, this form of paradoxical immunology is very challenging. Previous work in identifying epitopes from vaccinia virus by Assarsson et al. in 2007 is consistent with the predicted data of in silico screening generated and examined in the course of the present invention. Thus, the prediction algorithm has a low probability of selecting immunogenic epitopes.Therefore, in the course of the present invention, an ABab.I mouse model with a set of new chimeric class I fusion alleles, namely HLA-A*03:01, A*11:01, B*07:02, B*15:01, C*04:01, and C*07:02, was developed using a piggyback transposon strategy. By introducing these six HLA alleles as a single genotype resembling the natural state in humans, we expanded the peripheral pool with five times more unique and rare V(D)J-TCRβ clonotypes than ABabDII mice, broadening the TCR repertoire four-fold. In other words, the ABab.I mouse of the present invention has a high affinity for HLA-derived tumor-specific antigens (ICs) derived from recurrent somatic point mutations. 50 In addition to being a useful tool for identifying new TCRs against T cells (<50 nM), it can also be used as a versatile in vivo model system for epitope discovery.

[0027] definition As used herein, the term "mammals" may refer to a group of vertebrate animals characterized by the presence of mammary glands in the females that produce milk for feeding (nursing) their young.

[0028] As used herein, the term "non-human mammal" may refer to any mammal that is not a human.Non-limiting examples of non-human mammals of the present invention include, but are not limited to, rodents (e.g., mice or rats), dogs, felines, primates, rabbits, pigs, and ruminants (e.g., rodents, dogs, felines, primates, rabbits, pigs, and ruminants).Non-limiting examples of "rodents" are mice, rats, squirrels, chipmunks, gophers, porcupines, beavers, hamsters, gerbils, guinea pigs, degus, chinchillas, prairie dogs, and groundhogs.Non-limiting examples of "dogs" include members of the canis lupus familiaris subspecies, as well as wolves, foxes, jackals, and coyotes. Non-limiting examples of "felidae" include members of two subfamilies: Pantherinae, which includes lions, tigers, jaguars, and leopards, and Felinae, which includes cougars, cheetahs, servals, lynxes, caracals, ocelots, and domestic cats. The term "primates" as used herein refers to all monkeys, including, for example, Cercopithecus cercopithecus (Old World monkeys) or Platyrrhine simians (New World monkeys), as well as lemurs, tarsiers, apes, and marmosets (common marmosets). A further example of a non-human mammal of the present invention is the transgenic ABab.I mouse produced in the course of the present invention, which may also be interchangeably referred to as "multi-HLA mouse" or "HuTCR mouse" elsewhere in this specification.

[0029] As used herein, the term "genome" can refer to the genetic material of an organism (e.g., a non-human mammalian, e.g., mouse genome). For example, in this context, the term "genome" can refer to a set of haploid chromosomes and the genes they contain.

[0030] As used herein, the term "human leukocyte antigen" or "HLA" can refer to any of a variety of (polymorphic) proteins encoded by genes of the human major histocompatibility complex and found on the surface of many cell types, such as white blood cells. HLA class I alleles include HLA-A, HLA-B, and HLA-C alleles (http: / / hla.alleles.org / genes / index.html). The human major histocompatibility complex (MHC) is divided into three regions on chromosome 6p21.3: class II (centromere), class III, and class I (telomere), flanked by class I and class II extension regions.

[0031] As used herein, the term "major histocompatibility complex" or "MHC" may refer to a group of mammalian genes that code for cell surface polymorphic glycoprotein molecules that present antigenic peptide fragments for T cell recognition and aid the immune system's ability to distinguish self from non-self. The MHC codes for highly polymorphic proteins, many of which are associated with the immune system. The products of the classical polymorphic class I genes human leukocyte antigen A (HLA-A), HLA-B, and HLA-C interact with T cell receptor (TCR) molecules and killer immunoglobulin-like receptors (KIRs) expressed on natural killer cells and some T cells.

[0032] As used herein, the term "allele" may refer to any of the alternative forms of a gene (e.g., a human HLA gene) that may occur at a given locus. Each HLA allele name has a unique number that corresponds to a set of up to four numbers separated by colons. The length of the allele name is determined by the sequence of the allele and its closest relatives. Typically, an HLA allele name includes the following elements: an "HLA" prefix, a hyphen ("-") used to separate the gene name from the HLA prefix, a gene name (e.g., "A", "B", "C"), a delimiter "*", a field 1 indicating the allele group (e.g., 03, 11, 01, 26, 24, 32, 02, 07, 15, 58, 40, 35, 08, 04, 07, 16, 03, 07, or 06, etc.), a field delimiter ":", and a field 2 indicating a particular HLA protein (e.g., 01 or 02, etc.). Exemplary human class I HLA genes of the present invention include HLA-A (class I alpha chain), HLA-B (class I alpha chain), and HLA-C (class I alpha chain). Additional human class I HLA alleles of the present invention include HLA-E, HLA-F, HLA-G alleles, as well as pseudogenes HLA-H, HLA-J, HLA-K, HLA-L, HLA-N, HLA-P, HLA-S, HLA-T, HLA-U, HLA-V, HLA-W, and HLA-Y (http: / / hla.alleles.org / alleles / class1.html). Exemplary alleles of the present invention include any human class I and class II HLA alleles, e.g., selected from the group consisting of class I HLA-A, HLA-B, and HLA-C alleles. Such alleles may be selected from the group of exemplary human HLA alleles which are: HLA-A*01, HLA-A*02, HLA-A*03, HLA-A*11, HLA-A*24, HLA-A*26, and HLA-A*32; HLA-B*07, HLA-B*08, HLA-B*15, HLA-B*27, HLA-B*35, HLA-B*40, and HLA-B*58; HLA-C*03, HLA-C*04, HLA-C*06, HLA-C*07, and HLA-C*16 alleles.Exemplary human HLA alleles in the present invention include, but are not limited to, the HLA-A*01:03, HLA-A*03:01, HLA-A*24:02, HLA-A*11:01, HLA-B*07:02, HLA-B*08:01, HLA-B*15:01, HLA-B*27:05, HLA-B*35:01, HLA-B*44:02, HLA-C*04:01, HLA-C*07:01, HLA-C*07:02, and HLA-A*02:01 alleles. Currently, there are 32330 HLA and related alleles described by the HLA nomenclature and included in the IPD-IMGT / HLA database release 3.46 (2021-10) build 2d19adf (https: / / www.ebi.ac.uk / ipd / imgt / hla / ) according to http: / / hla.alleles.org / alleles / index.html. Any human class I HLA allele (e.g., HLA-A, HLA-B, or HLA-C allele) is within the contemplation of the invention. Any human class II HLA allele (e.g., any allele selected from the HLA-DRB1*13 group of alleles, e.g., HLA-DRB1*13:01) is within the contemplation of the invention. Exemplary class II HLA alleles in the present invention further include, but are not limited to, alleles of HLA-DRA, HLA-DRB1, HLA-DRB2-9, HLA-DQA1, HLA-DQA2, HLA-DQB1, HLA-DPA1, HLA-DPB1, HLA-DMA, HLA-DMB, HLA-DOA, HLA-DOB proteins (e.g., http: / / hla.alleles.org / alleles / class2.html). Additionally, human class I and class II HLA alleles can also be searched using the "Allele Query Tool" in the IPD-IMGT / HLA database as described above.

[0033] In this context, it is again noted that one advantage of the present invention is that it allows the mimicking of any given human (sub)population by incorporating into the genome of a non-human animal any desired set of class I HLA alleles, e.g. class I HLA alleles that have a high prevalence in a given population or region.

[0034] In one such illustrative example, the transgenic animals of the invention contain integrated into their genome the following HLA class I alleles: HLA-A*03:01, HLA-A*11:01, HLA-B*07:02, HLA-B*15:01, HLA-C*04:01, HLA-C*07:02, and HLA-A*02:01. As noted above, such transgenic animals represent approximately 80% of the American or European population with respect to their HLA class I gene profile.

[0035] In another illustrative example, the transgenic animal of the invention may comprise, integrated into its genome, a set of HLA class I alleles corresponding to the (ethnic) Chinese population. Such a transgenic animal may comprise two or more of the following sets of human HLA class I alleles: HLA-A*11:01, which has a prevalence (i.e., % of individuals carrying that allele) of 47.5% in the ethnic Chinese population, and / or HLA-A*24:02, which has a prevalence of 28.1% in the ethnic Chinese population, and / or HLA-A*33:03, which has a prevalence of 19.2% in the ethnic Chinese population, and / or HLA-B*40:01, which has a prevalence of 28.4% in the ethnic Chinese population, and / or HLA-B*50:01, which has a prevalence of 28.5% in the ethnic Chinese population, and / or HLA-B*60:01, which has a prevalence of 28.5% in the ethnic Chinese population. In one embodiment, HLA-B*46:01, which has a frequency of 25.1% in the ethnic Chinese population, and / or HLA-B*58:01, which has a frequency of 16.7% in the ethnic Chinese population, and / or HLA-B*15:02, which has a frequency of 12.5% ​​in the ethnic Chinese population, and / or HLA-C*07:02, which has a frequency of 35.1% in the ethnic Chinese population, and / or HLA-C*01:02, which has a frequency of 34.6% in the ethnic Chinese population, and / or HLA-C*03:04, which has a frequency of 21.9% in the ethnic Chinese population, may be selected. In this illustrative example, the transgenic animal may include two of the human HLA-A alleles described above, two of the human HLA-B alleles described above, and two of these human HLA-C alleles, e.g., HLA-A*11:01, HLA-A*24:02, HLA-B*40:01, HLA-B*46:01, HLA-C*01:02, and HLA-C*03:04.

[0036] In another illustrative example, a transgenic animal of the invention may contain, integrated into its genome, a set of HLA class I alleles that represent the (ethnic) Japanese population. Such transgenic animals may contain two or more of the specific HLA class I alleles: HLA-A*24:02 (frequency of 58.1% in the Japanese population), and / or HLA-A*02:01 (27.4%), and / or HLA-B*52:01 (18.8%), and / or HLA-B*51:01 (17.0%), and / or HLA-B*35:01 (16.6%), and / or HLA-B*15:01 (14.8%), and / or HLA-B*40:02 (14.7%), and / or HLA-C*01:02 (31.6%), and / or HLA-C*03:03 (27.5%), and / or HLA-C*03:04 (23.8%). Also, in this illustrative example, the transgenic animal can include two of the human HLA-A alleles described above, two of the human HLA-B alleles described above, and two of these human HLA-C alleles, e.g., HLA-A*24:02, HLA-A*02:01, HLA-B*52:01, HLA-B*51:01, HLA-C*01:02, and HLA-C*03:03.

[0037] In yet a further illustrative example, a transgenic animal of the invention may comprise, integrated into its genome, a set of HLA class I alleles representing the (ethnic) Indian population. Such transgenic animals may comprise two or more of the HLA class I alleles: HLA-A*01:01 (28.5%), and / or HLA-A*11:01 (26.0%), and / or HLA-A*24:02 (25.4%), and / or HLA-B*40:06 (17.4%), and / or HLA-B*51:01 (14.4%), and / or HLA-B*52:01 (14.3%), and / or HLA-B*44:03 (14.3%), and / or HLA-C*06:02 (25.9%), and / or HLA-C*04:01 (25.4%), and / or HLA-C*07:02 (20.4%). Also, in this illustrative example, the transgenic animal can include two of the human HLA-A alleles described above, two of the human HLA-B alleles described above, and two of these human HLA-C alleles, e.g., HLA-A*01:01, HLA-A*24:02, HLA-B*40:06, HLA-B*51:01, HLA-C*06:02, and HLA-C*04:01.

[0038] As used herein, the term "functionally expressed" may refer to an HLA allele that is expressed (e.g., at the cell surface) and functions as an HLA protein, e.g., presents a peptide derived from an endogenous protein (e.g., from inside the cell). The term "functionally expressed" may particularly refer to an HLA allele that is expressed (e.g., at the cell surface of a mammal of the present invention) and functions as an HLA protein, e.g., presents a peptide derived from an endogenous protein (e.g., from inside the cell of a mammal of the present invention), since the corresponding MHC I polypeptide is expressed, e.g., at the cell surface of a mammal of the present invention, and the non-human mammal presents MHC antigens that provide an antigen-specific CD8+ T cell response, and optionally provides an antigen-specific CD4+ T cell response.

[0039] As used herein, the term "monocistronic" can refer to a nucleic acid (eg, fully processed, eg, RNA) that encodes a single protein (eg, an HLA allele).

[0040] As used herein, the term "promoter" can refer to a regulatory region of DNA (e.g., located upstream of a gene) that provides a control point for regulated gene transcription. An example of a promoter of the present invention is the mammalian H-2 Db promoter. Another example of a promoter of the present invention is the viral CMV promoter. Other suitable promoters that drive MHC I transgene expression and can be used herein include, but are not limited to, the H-2Kb promoter (see Kimura et al., "Detailed analysis of the mouse H-2Kb promoter: enhancer-like sequences and their role in the regulation of class I gene expression," Cell 1986 Jan 31;44(2):261-72), the pROSA-26 promoter (see Zambrowicz et al., "Disruption of Overlapping Transcripts in the Rosa Beta Geo 26 Gene Trap Strain Leads to Widespread Expression of Beta-Galactosidase in Mouse Embryos and Hematopoietic Cells," Proc Natl Acad Sci USA, 94(1997), 3789-94), the viral SV40 promoter, or the mouse β2m promoter.

[0041] As used herein, the term "2A peptide linker" may refer to a class of 18-22 amino acid long peptides that can induce ribosomal skipping during translation of a protein in a cell. A non-limiting example 2A peptide linker in the present invention is the E2A peptide linker from Equine rhinitis A virus (e.g., SEQ ID NO: 229).

[0042] As used herein, the term "epitope" may refer to a portion of an antigen that is recognized by the immune system, particularly an antibody, a B cell, or a T cell. An epitope is the specific portion of an antigen that binds to an antibody.

[0043] In the present context, the term "antigen" may refer to any substance that can be specifically bound by a component of the immune system. Only antigens that are capable of eliciting (or eliciting or inducing) an immune response are considered immunogenic and are called "immunogens."

[0044] The term "polypeptide" may be used herein in the same manner as the term "protein". Proteins (usually fragments thereof, preferably biologically active fragments, and peptides having less than 30 amino acids, e.g., up to 10 or more amino acids, e.g., 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 25, or 30 amino acids) comprise one or more amino acids bound to each other via covalent peptide bonds (resulting in an amino acid chain). The term "polypeptide" as used herein refers to a group of molecules, e.g., more than 30 amino acids. Polypeptides can further form multimers, such as dimers, trimers, and higher oligomers, i.e., multimers consisting of two or more polypeptide molecules. The polypeptide molecules forming such dimers, trimers, etc. may be identical or non-identical. The corresponding higher order structures of such multimers are consequently referred to as homo- or heterodimers, homo- or heterotrimers, etc. An example of a heteromultimer is an antibody molecule, which in its natural form consists of two identical light polypeptide chains and two identical heavy polypeptide chains. The terms "polypeptide" and "protein" also refer to naturally modified polypeptides / proteins, where the modification is effected by post-translational modifications such as, for example, glycosylation, acetylation, and phosphorylation. Such modifications are well known in the art.

[0045] The term "amino acid" or "amino acid residue" typically refers to an amino acid having its art-recognized definition, such as an amino acid selected from the group consisting of alanine (Ala or A), arginine (Arg or R), asparagine (Asn or N), aspartic acid (Asp or D), cysteine ​​(Cys or C), glutamine (Gln or Q), glutamic acid (Glu or E), glycine (Gly or G), histidine (His or H), isoleucine (Ile or I), leucine (Leu or L), lysine (Lys or K), methionine (Met or M), phenylalanine (Phe or F), proline (Pro or P), serine (Ser or S), threonine (Thr or T), tryptophan (Trp or W), tyrosine (Tyr or Y), and valine (Val or V), although modified, synthetic, or rare amino acids can be used as desired. In general, amino acids can be grouped as having nonpolar side chains (e.g., Ala, Cys, He, Leu, Met, Phe, Pro, Val), negatively charged side chains (e.g., Asp, Glu), positively charged side chains (e.g., Arg, His, Lys), or polar uncharged side chains (e.g., Asn, Cys, Gln, Gly, His, Met, Phe, Ser, Thr, Trp, and Tyr).

[0046] For ease of reference, variants (or mutants) of the polypeptides of the invention may be referred to by using the nomenclature original amino acid:position:substituted amino acid, e.g., a substitution of S (serine) with F (phenylalanine) at position 722 may be designated as "S722F."

[0047] Generally, as used herein, the terms "polynucleotide" and "nucleic acid" or "nucleic acid molecule" are taken to be synonymous. Generally, nucleic acid molecules can include DNA molecules, RNA molecules, oligonucleotide thiophosphates, substituted ribooligonucleotides, or PNA molecules, among others. Furthermore, the term "nucleic acid molecule" can refer to DNA or RNA, or their hybrids, or any modifications thereof known in the art (for examples of modifications, see, for example, US Pat. No. 5,525,711, US Pat. No. 4,711,955, US Pat. No. 5,792,608, or EP Pat. No. 302,175). Polynucleotide sequences can be single-stranded or double-stranded, linear or circular, natural or synthetic, without any size restrictions. For example, the polynucleotide sequence may be genomic DNA, cDNA, mRNA, antisense RNA, ribozyme RNA, or DNA encoding such RNA, or a chimeroplast (Gamper et al., Nucleic Acids Res (2000), 28(21):4332-4339). The polynucleotide sequence may be in the form of a vector, a plasmid, or viral DNA or RNA.

[0048] The term "nucleic acid construct" can refer to a single- or double-stranded nucleic acid molecule, which is isolated from a gene of natural origin or which has been modified to contain a segment of nucleic acid in a manner that would not occur in nature, or which has been synthetically produced, and which includes one or more regulatory sequences.

[0049] The term "control sequences" may refer to nucleic acid sequences necessary for expression of a polynucleotide encoding an HLA allele of the present invention. Such control sequences may include, but are not limited to, a leader, a polyadenylation signal, a promoter, and a transcription terminator.

[0050] The term "expression" may refer to any step involved in making the HLA protein of the present invention, for example by a non-human mammal of the present invention, including but not limited to transcription, post-transcriptional modification, translation, and post-translational modification.

[0051] The term "polyadenylation signal" (or "pA") can refer to a polyA tail that is added to a nucleic acid (e.g., RNA) at the end of transcription. Exemplary pAs in the present invention are bGH pA (bovine growth hormone polyadenylation signal) and full-length H-2 Db pA.

[0052] The term "expression vector" can refer to a linear or circular nucleic acid molecule (e.g., DNA) that contains a polynucleotide encoding an HLA allele of the present invention and is operably linked to a control sequence that effects its expression.

[0053] The term "host cell" can refer to any cell type (e.g., a bacterial cell) that is susceptible to transformation, transduction, or transduction, etc., with a nucleic acid construct or expression vector comprising an HLA allele of the present invention. The term "host cell" encompasses any progeny of a parent cell that is not identical to the parent cell due to mutations that occur during replication.

[0054] The term "oocyte", as used herein, may refer to a female germ cell involved in reproduction, i.e., an egg or ovum. In the present invention, the term "oocyte" includes both pre-fertilized oocytes and fertilized oocytes, also called zygotes. Thus, pre-fertilized oocytes contain only maternal chromosomes, whereas fertilized oocytes contain both maternal and paternal chromosomes. After fertilization, oocytes remain in a doubled haploid state for several hours, for example, up to 18 hours after fertilization in mice. For example, the oocyte of the present invention may be a fertilized oocyte.

[0055] The term "fertilized oocyte" as used herein may refer to an oocyte after fusion with a fertilizing sperm. For many hours after fertilization (e.g., up to 18 hours in mice), the oocyte is in a state of double haploidy, containing one maternal haploid pronucleus and one paternal haploid pronucleus. After the two pronuclei migrate together, their membranes break and the two genomes condense into chromosomes, thereby reconstituting a diploid organism. This fertilized oocyte, also called single-cell zygote, two-cell zygote and four-cell zygote, is encompassed by the term "fertilized oocyte" as used herein.

[0056] The term "fragment" can refer to a polypeptide having one or more (eg, several, eg, up to 10-20) amino acids deleted from the amino and / or carboxyl terminus of a mature polypeptide.

[0057] As used herein, the term "TCR" may refer to a T cell receptor.

[0058] As used herein, the term "TCR" may refer to a T cell receptor.

[0059] As used herein, the term "pMHC" may refer to peptide-binding major histocompatibility complex.

[0060] As used herein, the term "pHLA" may refer to peptide-linked human leukocyte antigen.

[0061] As used herein, the term "CDR" may refer to complementarity determining region.

[0062] As used herein, the term "ATT" may refer to adoptive transfer therapy.

[0063] As used herein, the term "TSA" may refer to a tumor-specific antigen.

[0064] As used herein, the term "TAA" may refer to a tumor-associated antigen.

[0065] As used herein, the term "BCR" may refer to the B cell receptor.

[0066] As used herein, the term "PD-1" may refer to programmed cell death protein 1.

[0067] As used herein, the term "MAGE-1" may refer to melanoma-associated antigen-1.

[0068] As used herein, the term "NY-ESO-1" may refer to New York esophageal squamous cell carcinoma 1.

[0069] As used herein, the term "APC" may refer to an antigen-presenting cell.

[0070] As used herein, the term "mIL-2 / 15" may refer to murine interleukin 2 / 15.

[0071] The adaptive arm of the immune system The immune system provides a protective barrier to the body by fighting against foreign invaders such as bacteria, viruses, and parasites using innate and adaptive defense mechanisms. Innate immunity is the body's nonspecific primary defense in detecting and destroying pathogens. Toll-like receptors (TLRs) sense pathogens and recruit immune cells to the site for elimination by creating an inflammatory environment along with macrophage-induced phagocytosis. Dendritic cells (DCs) are professional antigen-presenting cells (APCs) that present endogenous and exogenous antigens to major histocompatibility molecules (MHC) I and II to act as a bridge in the recruitment of the adaptive immune system. Cells of the adaptive immune system are recruited by several proinflammatory cues through the activation of multiple pathways such as nuclear factor-κB (NF-κB), interferon (IFN) regulatory factors (IRFs), and activator protein-1 (AP-1). Although innate immune mechanisms act quickly by releasing proinflammatory cytokines such as IFN-α, -β (type I), and -γ (type II), cell-mediated adaptive immunity is the most effective arm of the immune system in eliminating pathogens. The adaptive arm of the immune system consists of two cell-mediated responses: T-cell and B-cell mediated immunity. T lymphocytes and B lymphocytes are precursors of antibody-secreting cells that differ in their specificity and mechanism of action. B cells recognize pathogenic antigens via the B-cell receptor (BCR) and helper CD4 T cells via MHC class II proteins. + Helps recruit T cells to the site of infection. Naive CD8 + T cell activation occurs via the classical MHC class I presentation pathway by antigen-presenting DCs. + T cells promote cytotoxic CD8 T cells by providing a strong signal to antigen-presenting DCs via CD40L-CD40 interactions. + In addition, CD4 + T cells express cytotoxic CD8 by direct interaction with the CD40 receptor. + Supports T cell memory formation.

[0072] Conventional naive CD8 + T cells and CD4+ Both T cells interact with cognate antigens presented on polymorphic MHC class I and II proteins via the αβ T cell receptor (TCR). The naive TCR repertoire formed in the thymus largely determines the precursor frequency and distribution of specific T cell clones. Quantifying the TCR repertoire can provide insight into the diversity of T cell-mediated responses an organism can mount.

[0073] T cell receptors (TCRs) and their diverse repertoires shape T cell-mediated immunity The αβTCR is generated by random and abundant variable (V), diversity (D), and joining (J) domain recombination events in the thymus. 20 Theoretically, combinations up to TCRαβ can occur. + T cells undergo positive and negative selection, with thymocytes undergoing positive selection when their self-MHC interactions are adequate, and negative selection when their affinity for any self-peptide MHC (pMHC) is too high. Approximately 7.5% of thymocytes produce thymic signaling, of which approximately 3-5% survive selection and become naive T cells in the periphery. These T cells, which have diverse αβTCRs, interact intrinsically with pMHC (by costimulation via CD3-γ, -δ, -ε, -ζ, CD4, and CD8 chains) and become naive CD8 + / CD4 + TCR V forms a repertoire of T cells that encounter foreign cognate antigens for T cell-mediated responses. MHC reactivity for approximately 5–20% of the preselection pool and approximately 10% of the peripheral T cell pool describes the intrinsic affinity for TCR V. α and V β Additional factors such as complementarity determining region (CDR) 1 and CDR2 encoded by (germline encoded) further direct the germline bias of the TCR towards the MHC.

[0074] There are some exceptions where CDR1 and CDR2 are not the main contact domains with the MHC. α and V βCrystallographic structures have shown a special diagonal arrangement in the MHC where CDR1 and CDR2 of V bind to the α2 and α1 helices of MHC class I or the β and α helices of MHC II. α and V β The hypervariable CDR3 loops of TCR are the primary binding points for peptide-binding MHC (pMHC). Existing evidence indicates that TCR and MHC genes have essentially co-evolved to interact with each other, whereas specific antigenic peptides processed and presented via the class I / II antigen presentation pathway precisely define potential T cell effector functions through the interaction of CDR3α and CDR3β.

[0075] Thus, the abundant combination of CDR3-V α and -V β Clonotypes generate diverse TCR repertoires that enable unique pMHC interactions that promote T cell-mediated immunity.

[0076] pMHC-I / HLA-I restricted antigen presentation pathway MHC glycoproteins, known in humans as human leukocyte antigens (HLA), process antigens and present them on their surface as 8-15 amino acids (aa) for class I peptides or 12-25 aa (8-10 aa core motif) for class II peptides. Two antigen presentation pathways, peptide-loaded HLA-I (pHLA-I) and pHLA-II, support the surface presentation of processed epitopes. HLA class I presents endogenously processed epitopes, such as somatic point mutants, virus-derived and aberrant frameshift proteins, in its α2 and α1 domains to the cytotoxic CD8 + HLA class II alleles present exogenously phagocytosed and processed epitopes, such as extracellular matrix proteins from the stroma and ruptured viral proteins, in their β1 and α1 helices to helper CD4 T cells (T HProfessional APCs can incorporate exogenous antigens into the endogenous HLA I pathway by cross-presentation. Coreceptor interactions between CD8 and the α3 domain of HLA I, and between CD4 and the β2 helix of HLA II, are important for T cell function.

[0077] Cellular, viral, aberrantly mutated, and misfolded post-translational proteins in the cytoplasm undergo proteasomal degradation. After proteasomal processing, fragmented peptides enter the endoplasmic reticulum (ER) via peptide loading complex (PLC) proteins such as transporter-associated protein (TAP) and the chaperone tapasin. Antigen processing-associated ER aminopeptidase (ERAAP) proteins trim the amino termini of ER-resident peptides before loading onto MHC. Calreticulin (CRT), Erp57, and PLC load the processed peptides onto MHC / HLA. The pMHC-I / HLA-I complex is released into early endosomes and then transported to CD8 + It translocates to the cell surface via late endosomal compartments for T cell recognition by T lymphocytes (CTLs).

[0078] Tumor Immunology Immune System and Cancer Cancer is widely recognized as an inherited disease, since oncogenic mutations, chromosomal rearrangements, and abnormalities in gene expression are the most common causes detected in approximately one million cases. Total sequencing studies of primary lung tumor samples identified more than 50,000 point mutations (with adjacent non-tumor tissues as controls), including, for example, Gly1 in the RAS gene. 12The tumor microenvironment, which includes fibroblasts, inflammatory cells, vascular endothelial cells, and extracellular matrix, contains several mutational hotspots, such as the endothelial cell-specific markers. Emerging evidence has demonstrated that all cancers evolve by accumulating a small subset of driver mutations that confer adaptive benefits and a large number of passenger mutations that do not confer growth advantages. For decades, research to understand the characteristics of cancer has focused primarily on tumor cells alone. Oncogenic driver mutations allow cancers to grow, but they do not grow independently without additional factors that support their growth. Studies have confirmed that the heterogeneous properties of tumor tissues lead to dynamic interactions with the surrounding microenvironment, often modulating cancer progression. The tumor microenvironment, which includes fibroblasts, inflammatory cells, vascular endothelial cells, and extracellular matrix, comprises the so-called “stroma”. It has been described in the literature that a chronic inflammatory state in the stroma, which accumulates with inflammatory cells, promotes cancer development. One example is stromal fibroblasts, which suppress transforming growth factor-β (TGF-β) signaling to induce gastric and prostate tumors. In the ongoing inflammatory state of tumors, similar to wound healing, fibroblasts, macrophages, and endothelial cells (as part of the stroma) actively participate in supporting the formation of new blood vessels to nourish the tumor, a process known as "angiogenesis."

[0079] CD8 + and CD4 + CD8 T cells require tumor-restricted peptide-loaded HLA-I (pHLA-I) and pHLA-II recognition, respectively, for tumor rejection. + Although T cells (CTLs) can elicit cytotoxic activity alone in most cases, in some models, CD4 +T cells reject tumors in the absence of CTLs. CTLs release cytokines such as interferon-γ (IFN-γ) and tumor necrosis factor-α (TNF-α) to reject tumors by attacking the tumor stroma, and in some cases, release cytotoxic molecules such as perforin to eliminate tumors, especially in virus-transformed tumors. From the study of many analytical models, T cells use IFN-γ as an important effector molecule for tumor rejection. In 2017, Kammertoens et al. inferred the mechanism of IFN-γ in tumor stroma, especially in endothelial cells. T cells, through IFN-γ induction, cause tumor ischemia, which indicates a physiological state of vascular regression, thereby inhibiting angiogenesis. Another topic linking the immune system to cancer is the concept of "immunosurveillance", which is still a controversial topic, requiring further evidence. In 2002, Dunn et al.'s study based on a methylcholanthrene (MCA)-induced spontaneous tumor model claimed to have demonstrated cancer immunosurveillance of tumor rejection and growth control by T cells. However, subsequent studies have disproved the concept of T cell-mediated surveillance of tumor rejection. In the initial report, several other reasons for the change in tumor development were proposed, including tissue repair at the MCA site, differences in steady-state IFN-γ concentrations between control and experimental groups, and biased IFN-γ protection due to differences in the rearing conditions of the animal facility. Furthermore, tumor development in such models was questioned, since either chemical-induced chronic inflammation or opportunistic infections could be the reason for spontaneous growth, which does not resemble tumors that grow naturally in humans. In 2005, Willimsky et al. developed a sporadic tumor model induced by viral transgenic dormant oncogenes (reflecting physiological tumors) to study the influence of T cells recognizing tumor cells in the context of immune surveillance. This study demonstrated that immunogenic sporadic tumors avoid T cell destruction by inducing tolerance rather than escaping T cell recognition, and suggested a possible tumor immunosurveillance effect against virus-associated cancers. Besides the SV40 model, EBV- and HPV-associated tumors can be monitored by T cells, but more solid experimental evidence is needed for the immunosurveillance of non-viral tumors.Furthermore, reports demonstrating the concept of immune surveillance in non-virus-related cancers require careful interpretation of the data.

[0080] In summary, the role of T cells in attacking tumor cells is clear. Independent of T cell immune surveillance and its control over tumor growth, TCR-modified CD8 + and CD4 + T cell-based treatment options may offer the best chance of tumor rejection.

[0081] Adoptive T cell therapy (ATT) for cancer To date, most immunologists have focused on targeting non-tumor-specific self-antigens using therapeutic vaccination, with notable exceptions. Therapeutic vaccines against tumor-associated common antigens by active and passive cancer immunotherapy have not met the expected success due to well-established tolerance mechanisms, suppression, and tumor microenvironment factors. In contrast to cancer vaccination, adoptive T cell therapy (ATT) has the potential to reject large established tumors, in particular by targeting TCR-modified CTLs (FasL) that release effector molecules such as IFN-γ but not perforin. + ) could destroy the surrounding stroma with a complete eradication effect. Besides TCR-modified CTLs, chimeric antigen receptor (CAR)-linked T cells (CAR T cells) have shown success in the use in non-solid tumor clinical studies. Early phase clinical studies with CAR T cells targeted the CD19 surface antigen in hematological malignancies. Although early results targeting surface antigens in liquid tumors have been noteworthy, there are still challenges to achieve similar success by targeting solid tumors.

[0082] CAR T cells have several obstacles to overcome to demonstrate efficacy in solid tumors without causing toxicity. These include single-chain variable fragments (scFv) in CAR T cells targeting unique tumor-associated antigens (TAA) on the tumor cell surface without deleting the antigen. Secondly, CAR-bound T cells need to migrate to distant tumor sites for efficient homing. Besides this, evasion of the immunosuppressive tumor microenvironment by T cells poses further challenges. In clinical studies targeting folate receptor-α in ovarian cancer, limited T cell homing to solid tumor sites was reported, which could be enhanced by co-expressing chemokine receptor 2 (CXCR2) in CAR T cells against the Gro-α chemokine. Blocking TGF-β signaling in CAR T cells targeting prostate-specific membrane antigen by co-expressing a dominant-negative RII receptor enhanced their ability to invade tumor stroma. In other studies of clinically relevant murine pleural mesothelioma, CD28 CAR T cells restored effector function by cotransducing a dominant-negative programmed death-1 (PD-1) receptor or by knocking down PD-1 expression by more than 60% with short hairpin RNA (shRNA).

[0083] Lessons learned from clinical studies on antigen selection, T cell migration, homing, and efficacy improvement may be applicable to both CAR-based and TCR-modified adoptive T cell therapy, but the most important concern is always "safety". TCR-modified T cells stand out in terms of safety because they can accurately recognize endogenously processed antigens on tumor cells or stroma via HLA-I or HLA-II restricted methods. In ATT, the use of TCR-modified T cells offers the distinct advantage of targeting tumor-specific antigens to avoid harmful autoimmune effects by reducing the chance of cross-reactivity against unintended self-targets. A unique class of such truly tumor-specific antigens may arise from nonsynonymous somatic point mutations.

[0084] Somatic point mutations can result in tumor-specific neoantigens Cancer development is an evolutionary, multi-step process that occurs through a series of genomic instability events. Genomic instability can lead to uncontrolled cell proliferation, invasion, and metastasis. Such genomic alterations include gene amplifications, chromosomal translocations, deletions, and collectively any somatic mutations.

[0085] Most tumor cells produce and process antigenic peptides, which are expressed on the cell surface or released into the extracellular matrix, the stroma. Studies have reported that a wide variety of peptide antigens are associated or shared among several cancer types, including colon and prostate cancer, malignant melanoma, kidney cancer, breast cancer, and lung cancer. In all cancer entities, tumor antigens largely fall into two categories: tumor-specific antigens (TSAs) and tumor-associated antigens (TAAs). TSAs are a class of antigens exclusively expressed by tumor cells, including all abnormal protein structures derived from somatic mutations that may be the primary cause of malignant transformation. On the other hand, TAAs are antigens derived from overexpressed abnormal proteins that are expressed not only by tumor cells but also by a subpopulation of normal cells.

[0086] Recent advances in next-generation sequencing of whole genomes and exomes (all protein-coding regions) have made the identification and screening of mutations in TSAs easy and of great importance. Such TSAs are considered to be ideal and truly tumor-specific (called neo-antigens) targets in adoptive T-cell therapy (ATT). The properties of TSAs derived from mutations vary from cancer to cancer and from individual to individual due to inter- and intra-tumor heterogeneity, respectively. Adult solid tumor elements have about 50 mutations at the time of primary diagnosis, while childhood cancers and leukemias have acquired fewer mutations (10 on average). UV-associated melanomas of the skin have 100-200 mutations. However, genomically unstable colorectal cancers can contain more than 1000 mutations. Notably, the majority of mutations that frequently alter the encoded proteins are nonsynonymous somatic point mutations caused by single amino acid substitutions, such as R175H and R248Q in the TP53 gene, and G12V and G13D in the RAS gene.

[0087] Regardless of the total number of somatic mutations in a tumor, or the recurrent occurrence of certain mutations (hot spots), the abnormal proteins produced from driver mutations may or may not form epitopes and induce immunogenicity. Many factors contribute to the formation of epitopes, such as antigen expression level, peptide processing by the proteasome, and trafficking (e.g., Figure 1), and binding affinity to MHC / HLA determines whether a mutant peptide can become a suitable epitope. The probability that a nonsynonymous somatic mutation will go through these steps to produce a new tumor-specific antigen is low. The probability prediction proved true in a study of vaccinia virus that reported that only one of 14 predicted epitopes was actually immunogenic. In detail, HLA-A*02:01 (HLA-A2) and IC 50 Of the approximately 100 vaccinia-encoded epitopes predicted to bind at <100 nM, only 50% of the epitopes bind to CD8 +induced a T cell response, from which only 15% were demonstrated for endogenous processing. Of the 15% processed epitopes, 11% induced a CTL response, which explains that 1 of all 14 predicted epitopes forms an immunogenic epitope. Comparing vaccinia virus with physiological cancers, which have a similar probability of somatic mutation, it can be estimated that there are about 4-5 mutated epitopes per cancer with 50 mutations. Since every human individual has 6 HLA alleles, 20-25 epitopes per individual can be expected.

[0088] Immunodominance also plays a role in dominant vs. recessive mutated epitopes. Recessive epitopes can only be targeted if dominant epitopes are lost. One factor that recessive epitopes may not show immunodominance may be affinity for HLA. Dominant epitopes may show strong affinity for HLA alleles. On the other hand, some HLA alleles show predictions that bind more epitopes than others, making the estimation difficult. In addition to point mutations, frameshift mutations that result in new reading frames may increase the probability of possible epitopes being generated per individual.

[0089] As a boost, in cancer patients, various forms of TSAs inhibit CD8 + They have been identified as targets of T cells and have also been explored in preclinical trials. There is ample evidence reporting that cytotoxic T lymphocytes can effectively recognize tumor antigens resulting from mutations in various forms of cancer, such as colorectal cancer, head and neck cancer (CASP-8), melanoma (β-catenin), lung squamous cell carcinoma (NFYC), and leukemia. Taking all the above understanding together, somatic mutations, if proven to be naturally processed and presented, may provide potential neoepitopes for ATT of cancer.

[0090] Recurrent somatic driver mutations as targets of ATT in cancer Adoptive transfer of tumor antigen-specific T cells is one of the most promising advanced pharmaceutical preparations (ATMPs) in the field of cancer immunotherapy. However, the appropriate selection of antigens for potent ATT remains problematic.

[0091] During the past two decades, several tumor-specific preclinical studies have led to the development of a variety of biological agents that block or inhibit various molecular targets derived from the genetic abnormalities of cancer. Targeted cancer therapies, such as monoclonal antibodies (mAbs), small molecule drugs (SMDs), and cancer vaccines, have used tumor-associated and self-antigens as targets.

[0092] Because they target and attack cells expressing self-antigens, these therapies exhibited high toxicity, frequent drug resistance, and relapse, thereby weakening the overall therapeutic benefit. Therefore, careful selection of appropriate tumor antigens may still be a key challenge for effective ATT. Therefore, compared with SMD and mAb, using TCR-binding T cells to target recurrent somatic driver mutations (TSAs) may be a more precise strategy.

[0093] Anders et al. have reported that monospecific CD8 + T cells were detected in large, established tumors (≥500 mm 3 We have shown that complete rejection of such tumors can be achieved by elimination of escape mutants. Targeting driver mutations (evolutionary ancestors of cancer cells) that produce homogeneously expressed epitopes can also avoid the resulting antigen-deleting mutants.

[0094] Not all mutations are driver mutations and can form epitopes. Driver mutations confer a growth advantage to tumors. With respect to tumor heterogeneity, driver mutations can occur recurrently across tumors (intertumor) and among tumors (intratumor). Recurrent neoepitopes are often derivatives of driver mutations (ancestral mutations) shared by all cancer cells and are ideal target antigens that cannot be lost. However, the selection and validation of the most promising neoepitope candidates for ATT is a daunting task.

[0095] Recently, in silico-based identification of neoantigens from tumor whole genome or exome sequencing datasets is believed to aid in neoepitope selection. In 2019, this bioinformatics study published a predictive screen using TCGA datasets to identify and rank the binding affinity of putative neoepitopes to HLA complexes. A machine learning-based affinity prediction algorithm was used as an opportunity to devise off-the-shelf T cell therapies for subgroups of cancer patients sharing recurrent neoepitopes. Immunogenicity validation for efficient process and presentation is inevitable for all predicted recurrent mutated epitopes. For this reason, humanized TCR-HLA mouse models may be utilized as a useful tool to validate the immunogenicity of predicted epitopes as well as to detect new epitopes by full-length protein-coding gene immunization.

[0096] Humanized Mammalian Models The fact that repetitive neoepitopes are ideal targets with homogeneous expression in all cancer cells has contributed to the emerging methods for isolating TCRs for cancer ATT. Identification of neoantigen-specific TCRs is possible from humans or from transgenic mammals such as transgenic mice, rats, or pigs, to name a few.

[0097] In humans, the α- and β-chains of TCRs can be isolated by in vitro priming with mutant peptides from HLA-matched healthy individuals or from immunized candidates. In either case, in the former case, the precursor TCR repertoire against the neoepitope of the healthy donor is not affected by the central tolerance mechanism of deletion of T cells that bind with high affinity only to the wild-type self-counterpart. Thus, although the isolation of TCRs is in principle possible by autologous priming, the percentage of neoantigen-specific T cells is low, with only about 1.2% of mutations being naturally recognized, as shown in patients with melanoma, gastrointestinal cancer, lung cancer, and ovarian cancer. Moreover, the level of expression of neoantigens in the periphery is a determining factor for whether T cells are deleted or anergic.

[0098] In transgenic animals, as exemplified experimentally by transgenic mice herein, isolation of TCRs from the naive repertoire can be achieved after multiple immunizations to generate memory responses in vivo, and the TCRs are derived from tumor-free hosts, potentially selecting high-affinity T cells from the naive precursor pool that may be deleted by chronic antigen stimulation in tumor-bearing human individuals.

[0099] Transgenic HLA-I mice presenting human HLA antigens An example of a tumor-free human HLA host is the HLA-A2 / D host, which has served as a model for monitoring human immune responses for TCR isolation and epitope identification. b Transgenic mouse models using diverse chimeric HLA-D antigens to understand inter-virus heterologous immunity in comparison to humans. b / K bModels were used. Chimeric MHC / HLA transgenic models have existed for some time, but questions remain regarding the education of mouse TCRs against chimeric single human HLA alleles, the overall TCR repertoire, and the specific Vas specificities that are gained or lost during thymic selection. Furthermore, severe toxicity on target combined with lethality was reported in cancer patients treated with affinity matured (in the CDR3α region) anti-MAGE-A3 TCRs from HLA-A2 transgenic mice that cross-reacted with other shared epitopes of MAGE-A9 and -A12 antigens. As mouse TCRs are not negatively selected in the thymus against the patient's HLA alleles and the human proteome, precise on- and off-toxicity assessment is required for such mouse-derived TCRs that may cross-react with normal human proteins. Also, partially humanized mouse TCRs are indeed immunogenic when transferred to humans.

[0100] Therefore, it seems inevitable to comprehensively humanize mouse models expressing diverse human HLA with human TCR loci in order to carefully predict and transplant mouse-derived therapeutics into the human system. On the other hand, transgenic HLA-A2 / K mice with a complete human TCR locus are being developed to efficiently isolate TCRs with optimal affinity. b Mice have been generated and established (Li, L.-P. et al. Transgenic mice with a diverse human T cell antigen receptor repertoire. Nat. Med. 16, 1029 (2010)).

[0101] Human TCR transgenic mice as an ideal tool to isolate TCRs for cancer ATT The need for humanization led to the development of a new humanized transgenic mouse named ABabDII. ABabDII mice have a complete human TCR locus, knockout mouse TCR locus and MHC I locus, and are transgenic with human class I allele of HLA-A2 (Li, L.-P. et al. Transgenic mice with a diverse human T cell antigen receptor repertoire. Nat. Med. 16, 1029 (2010)). Thus, this model has become a useful tool for TCR isolation against TSAs, including TAAs and viral epitopes. Its high degree of humanization allows the identification of functional CD8 for optimal affinity TCR isolation. + Even if it provides a T cell population, the ABabDII model is a limited source of only HLA-A2 restricted TCR. Therefore, this is actually a bottleneck that ABabDII mice do not provide other HLA alleles. Therefore, this requires humanized TCR mice expressing other class I alleles for epitope detection, which benefits TCR isolation for many non-HLA-A2 individuals.

[0102] Aims and Objectives One of the main goals was to expand the allelic diversity to study human TCR repertoire expansion against full human HLA haplotypes in mice. By a strategy of piggyback transposon (or PB, a mobile genetic element that efficiently transposes between vectors and chromosomes via a "cut and paste" mechanism), six HLA alleles were introduced into the transcriptionally active site as a single haplotype, as in humans (every individual has six HLA class I alleles). Another aim was to ensure the surface expression of the newly introduced HLA alleles and analyze their presentation capacity. Finally, our new mouse model could be an in vivo epitope detection system to identify various tumor antigens restricted to high frequency (allele-wise) HLA genes and thereby isolate T cell receptors for the ATT of cancer.

[0103] The ABab.I transgenic mouse model of the present invention is a new tool for isolating broadly HLA-restricted TCRs.

[0104] To date, the ABabDII transgenic mouse model described by Li et al. has been a useful system for isolating human TCRs, as it contains the entire human T cell receptor locus. However, this model presents the limitation of isolating only HLA-A2-restricted TCRs. In order to target diverse non-HLA-A2-binding antigens and broaden the HLA repertoire of the existing system for parallel epitope detection, it is necessary to include multiple human ABabDII mice. In this regard, one of the aims of the present invention was to generate "ABab.I transgenic mice" by introducing a set of new chimeric class I fusion alleles (HLA-A*03:01, A*11:01, B*07:02, B*15:01, C*04:01, C*07:02) as a single genotype into existing ABabDII mice (HLA-A2). Furthermore, the use of ABab.I mice was to detect previously undescribed immunogenic epitopes that were selected to be presented on any of the six HLA alleles after processing.

[0105] In the process of the present invention, the goal is to develop tumor-specific antigens with high HLA affinity (IC) derived from recurrent somatic point mutations. 50 Our objective was to develop a new transgenic ABab.I mouse line that, besides being a useful tool for identifying new TCRs against IgG antibodies (<50 nM), also serves as an in vivo model system for epitope detection.

[0106] In one embodiment, the invention provides a non-human mammal (e.g., a transgenic ABab.I mouse of the invention) comprising in its genome at least two or at least three human leukocyte antigen (HLA) class I alleles, wherein said at least two or at least three human HLA alleles are functionally expressed to express corresponding MHC I polypeptides on the surface of cells of said mammal to present MHC antigens that provide an antigen-specific CD8+ T cell response by the non-human mammal, wherein said at least two or at least three human HLA class I alleles comprise: (a) at least one human HLA-A allele, and / or (b) at least one human HLA-B allele, and / or (c) at least one human HLA-C allele, and / or (d)(a) and (b), and / or (e)(a) and (c), and / or (f)(b) and (c), and / or (g)(a), (b), and (c). For clarity, possibility (g) means that the transgenic mammal comprises in its genome at least one human HLA-A allele, at least one human HLA-B allele, and at least one human HLA-C allele.

[0107] In a further embodiment, the present invention provides a nucleic acid construct comprising a nucleic acid encoding at least two or at least three human HLA class I alleles, wherein said nucleic acid construct is capable of functionally expressing said at least two or at least three human HLA alleles such that corresponding MHC I polypeptides are expressed on the surface of cells of said mammal to present MHC antigens providing an antigen-specific CD8+ T cell response by the non-human mammal, and optionally said at least two or at least three human HLA class I alleles comprise (a) at least one human HLA-A allele, and / or (b) at least one human HLA-B allele, and / or (c) at least one human HLA-C allele, and / or (d)(a) and (b), and / or (e)(a) and (c), and / or (f)(b) and (c), and / or (g)(a), (b), and (c). For clarity, possibility (g) means that the nucleic acid construct comprises a nucleic acid encoding a transgenic mammal whose genome comprises at least one human HLA-A allele, at least one human HLA-B allele, and at least one human HLA-C allele.

[0108] In a further embodiment, the present invention provides an expression vector comprising a nucleic acid construct as described herein.

[0109] In further embodiments, the present invention provides host cells (e.g., isolated and / or recombinant host cells) comprising a nucleic acid construct as described herein and / or an expression vector as described herein.

[0110] In a further embodiment, the present invention provides a method of modifying endogenous HLA alleles of a non-human mammal, comprising transducing and / or transplanting a nucleic acid construct and / or expression vector as described herein into a non-human mammal as described herein.

[0111] In a further embodiment, the present invention provides a method of producing a non-human mammalian oocyte having a modified target sequence in its genome, the method comprising the step of introducing into the non-human mammalian oocyte a nucleic acid construct and / or an expression vector as described herein.

[0112] In a further embodiment, the invention provides a method of producing a non-human mammal having a modified target sequence in its genome, comprising: (a) producing an oocyte as described herein; and (b) analyzing offspring delivered by a non-human female host into which the oocyte obtained in (a) was transferred for the presence of the modification.

[0113] In a further embodiment, the invention provides a non-human mammal generated and / or modified by a method of generating a non-human mammal having a modified target sequence in its genome as described herein. In a further embodiment, the invention provides a method of generating one or more T cell receptors capable of binding an antigen of interest, the method comprising administering the antigen of interest to a non-human animal as described herein.

[0114] In a further embodiment, the present invention provides a T cell receptor obtained or obtainable by each of the methods as described herein.

[0115] In a further embodiment, the invention provides a method for identifying an epitope capable of eliciting an immune response, comprising administering to a non-human animal as described herein an antigen of interest suspected of containing an epitope capable of eliciting an immune response.

[0116] In further embodiments, the present invention provides epitopes capable of eliciting an immune response and identified by the respective methods as described herein.

[0117] In a further embodiment, the invention provides for the use of a non-human animal as described herein to generate one or more T cell receptors capable of binding an antigen of interest.

[0118] In a further embodiment, the invention provides for the use of a non-human animal as described herein to identify epitopes capable of eliciting an immune response.

[0119] In a further embodiment, the invention provides for the use of a non-human animal as described herein to identify one or more T cell receptors capable of binding an antigen of interest.

[0120] The invention as described herein, in particular the non-human mammal of the invention (i.e. the transgenic ABab.I mouse of the invention, which may also be referred to interchangeably herein as a "multi-HLA mouse" or "HuTCR mouse"), and the TCR isolation method based thereon, provide for TCR isolation: - greater TCR diversity and therefore a higher probability of success in isolating a TCR against any known tumor target; - a wider HLA range, therefore the non-human mammal of the invention (which may also be referred to as a multi-HLA mouse) corresponds to approximately 80% of the European / American population, - improved and / or more extensive methods for producing TCRs based thereon; - Proof of concept for generating high affinity TCRs.

[0121] Furthermore, in some aspects / embodiments of the invention, the non-human mammal of the invention (i.e., the transgenic ABab.I mouse of the invention, which may also be referred to interchangeably herein as a "multi-HLA mouse" or a "HuTCR mouse") and / or methods based thereon, comprises: - the extent of humanization (e.g. in mice) includes HLA and TCR; - the interaction of CD8 and CD4 promoted by the chimeric HLA as described herein, for example the chimeric HLA in the non-human mammal of the invention promotes the interaction of the CD8 / CD4 coreceptor. The non-human mammal of the invention is extensively validated for function in human T cells, - random integration-based gene editing (e.g., via HLA allele integration as described herein); - the non-human mammal of the invention and the method based thereon may be collectively referred to as the "platform" of the invention, which has already been validated for TCR isolation for tumor targeting in the course of the invention, - broad HLA coverage, e.g., seven class I alleles (corresponding to approximately 80% coverage in Europe / USA) and one class II allele as described herein; - For example, higher numbers of T cells (e.g., CD8) compared to single HLA class I allele mice (approximately 40%), - broader and / or higher TCR diversity (e.g. very broad TCR repertoire), - the possibility of isolating high affinity TCRs, which has for example been demonstrated against multiple tumor targets in the course of the invention; - comprising / providing one or more of the following features / advantages: fully human TCR genes.

[0122] The present invention is also characterized by the following items (eg, embodiments). 1. A non-human mammal comprising at least two (e.g., at least three, at least four, at least five, at least six, or at least seven, etc.) human leukocyte antigen (HLA) class I alleles in its genome, wherein said at least two human HLA alleles are associated with corresponding MHC and wherein the HLA class I polypeptide is functionally expressed on the surface of a cell of said mammal to present MHC antigens that provide an antigen-specific CD8+ T cell response by the non-human mammal, and optionally wherein said at least two human HLA class I alleles include (a) at least one (e.g., at least two, or at least three) human HLA-A allele, and / or (b) at least one (e.g., at least two) human HLA-B allele, and / or (c) at least one (e.g., at least two) human HLA-C allele, (d) (a) and (b), and / or (e) (a) and (c), and / or (f) (b) and (c), and / or (g) (a), (b), and (c). 2. The non-human mammal of any one of the preceding clauses, wherein the mammal further comprises at least one human HLA class II allele in its genome. 3. A non-human mammal according to any one of the preceding clauses, wherein the mammal comprises at least four, five, six, or seven human HLA class I alleles in its genome. 4. Mammals are (a) at least two different human HLA-A alleles, preferably at least three different human HLA-A alleles, and / or (b) at least two different human HLA-B alleles, and / or (c) A non-human mammal according to any one of the preceding paragraphs, comprising in its genome at least two different human HLA-C alleles. 5. At least two human HLA class I alleles are selected from the group consisting of HLA-A*03 (e.g., HLA-A*03:01), HLA-A*11 (e.g., HLA-A*11:01), HLA-A*01 (e.g., HLA-A*01:01), HLA-A*26 (e.g., HLA-A*26:01), HLA-A*24 (e.g., HLA-A*24:02), HLA-A*32 (e.g., HLA-A*32:01), HLA-A*32 (e.g., HLA-A*32 ... 10. The non-human mammal according to any one of the preceding clauses, comprising one or more HLA-A class I alleles selected from the group (Europe / USA) consisting of HLA-A*03, HLA-A*11, and HLA-A*02 (e.g., HLA-A*02:01), ... preferably the three HLA-A class I alleles are selected from the group of human HLA alleles consisting of HLA-A*03, HLA-A*11, and HLA-A*02. 6. The non-human mammal according to any one of the preceding clauses, wherein the at least two human HLA class I alleles comprise one or more HLA-A class I alleles selected from the group consisting of HLA-A*11 (e.g., HLA-A*11:01), HLA-A*24 (e.g., HLA-A*24:02), and HLA-A*33 (e.g., HLA-A*33:03), preferably HLA-A*24 (e.g., HLA-A*24:02) and HLA-A*33 (e.g., HLA-A*33:03) (China). 7. The non-human mammal according to any one of the preceding paragraphs, wherein the at least two human HLA class I alleles comprise one or more HLA-A class I alleles selected from the group consisting of HLA-A*24 (e.g., HLA-A*24:02), HLA-A*02 (e.g., HLA-A*02:01), HLA-A*11 (e.g., HLA-A*11:01), and HLA-A*31 (e.g., HLA-A*31:01), preferably HLA-A*24 (e.g., HLA-A*24:02) and HLA-A*31 (e.g., HLA-A*31:01) (Japan). 8. The non-human mammal according to any one of the preceding clauses, wherein the at least two human HLA class I alleles comprise one or more HLA-A class I alleles selected from the group consisting of HLA-A*01 (e.g., HLA-A*01:01), HLA-A*11 (e.g., HLA-A*11:01), and HLA-A*24 (e.g., HLA-A*24:02), preferably HLA-A*01 (e.g., HLA-A*01:01) and HLA-A*24 (e.g., HLA-A*24:02) (India). 9. The non-human mammal according to any one of the preceding clauses, wherein the at least three HLA-A class I alleles are selected from the group of human HLA alleles consisting of HLA-A*03 (e.g., HLA-A*03:01), HLA-A*11 (e.g., HLA-A*11:01), HLA-A*01 (e.g., HLA-A*01:01), HLA-A*26 (e.g., HLA-A*26:01), HLA-A*24 (e.g., HLA-A*24:02), HLA-A*32 (e.g., HLA-A*32:01), and HLA-A*02 (e.g., HLA-A*02:01), preferably the three HLA-A class I alleles are selected from the group of human HLA alleles consisting of HLA-A*03, HLA-A*11, and HLA-A*02. 10. The non-human mammal according to any one of the preceding clauses, wherein the at least two HLA class I alleles comprise one or more HLA-B class I alleles selected from the group (Europe / USA) consisting of HLA-B*07 (e.g., HLA-B*07:02), HLA-B*15 (e.g., HLA-B*15:01), HLA-B*58 (e.g., HLA-B*58:01), HLA-B*40 (e.g., HLA-B*40:01), HLA-B*35 (e.g., HLA-B*35:01), and HLA-B*08 (e.g., HLA-B*08:01), preferably the at least two HLA-B class I alleles are selected from the group of human HLA alleles consisting of HLA-B*07 and HLA-B*15. 11. The non-human mammal according to any one of the preceding clauses, wherein the at least two HLA class I alleles comprise one or more HLA-B class I alleles selected from the group consisting of HLA-B*40 (e.g., HLA-B*40:01), HLA-B*46 (e.g., HLA-B*46:01), HLA-B*58 (e.g., HLA-B*58:01), and HLA-B*15 (e.g., HLA-B*15:02), preferably HLA-B*40 (e.g., HLA-B*40:01) and HLA-B*46 (e.g., HLA-B*46:01) (China). 12. The non-human mammal according to any one of the preceding clauses, wherein the at least two HLA class I alleles comprise one or more HLA-B class I alleles selected from the group consisting of HLA-B*52 (e.g., HLA-B*52:01), HLA-B*51 (e.g., HLA-B*51:01), HLA-B*35 (e.g., HLA-B*35:01), HLA-B*15 (e.g., HLA-B*15:01), and HLA-B*40 (e.g., HLA-B*40:02), preferably HLA-B*52 (e.g., HLA-B*52:01) and HLA-B*51 (e.g., HLA-B*51:01) (Japan). 13. The non-human mammal according to any one of the preceding clauses, wherein the at least two HLA class I alleles comprise one or more HLA-B class I alleles selected from the group consisting of HLA-B*40 (e.g., HLA-B*40:06), HLA-B*51 (e.g., HLA-B*51:01), HLA-B*52 (e.g., HLA-B*52:01), and HLA-B*44 (e.g., HLA-B*44:03), preferably HLA-B*40 (e.g., HLA-B*40:06) and HLA-B*51 (e.g., HLA-B*51:01) (India). 14. The non-human mammal according to any one of the preceding clauses, wherein the at least two HLA-B class I alleles are selected from the group of human HLA alleles consisting of HLA-B*07 (e.g., HLA-B*07:02), HLA-B*15 (e.g., HLA-B*15:01), HLA-B*58 (e.g., HLA-B*58:01), HLA-B*40 (e.g., HLA-B*40:01), HLA-B*35 (e.g., HLA-B*35:01), and HLA-B*08 (e.g., HLA-B*08:01), preferably the at least two HLA-B class I alleles are selected from the group of human HLA alleles consisting of HLA-B*07 and HLA-B*15. 15. The non-human mammal according to any one of the preceding clauses, wherein the at least two HLA class I alleles comprise one or more HLA-C class I alleles selected from the group (Europe / USA) consisting of HLA-C*04 (e.g., HLA-C*04:01), HLA-C*07 (e.g., HLA-C*07:02), HLA-C*16 (e.g., HLA-C*16:01), HLA-C*03 (e.g., HLA-C*03:04), HLA-C*07 (e.g., HLA-C*07:01), and HLA-C*06 (e.g., HLA-C*06:02), preferably the two HLA-C class I alleles are selected from the group of human HLA alleles consisting of HLA-C*04 and HLA-C*07. 16. The non-human mammal according to any one of the preceding clauses, wherein the at least two HLA class I alleles comprise one or more HLA-C class I alleles selected from the group consisting of HLA-C*07 (e.g., HLA-C*07:02), HLA-C*01 (e.g., HLA-C*01:02), and HLA-C*03 (e.g., HLA-C*03:04), preferably HLA-C*01 (e.g., HLA-C*01:02) and HLA-C*03 (e.g., HLA-C*03:04) (China). 17. The non-human mammal according to any one of the preceding clauses, wherein the at least two HLA class I alleles comprise one or more HLA-C class I alleles selected from the group consisting of HLA-C*01 (e.g., HLA-C*01:02), HLA-C*03 (e.g., HLA-C*03:03), and HLA-C*03 (e.g., HLA-C*03:04), preferably HLA-C*01 (e.g., HLA-C*01:02) and HLA-C*03 (e.g., HLA-C*03:03) (Japan). 18. The non-human mammal according to any one of the preceding clauses, wherein the at least two HLA class I alleles comprise one or more HLA-C class I alleles selected from the group consisting of HLA-C*06 (e.g., HLA-C*06:02), HLA-C*04 (e.g., HLA-C*04:01), HLA-C*07 (e.g., HLA-C*07:02), and HLA-C*15 (e.g., HLA-C*15:02), preferably HLA-C*06 (e.g., HLA-C*06:02) and HLA-C*15 (e.g., HLA-C*15:02) (India). 19. The non-human mammal according to any one of the preceding clauses, wherein the at least two HLA-C class I alleles are selected from the group of human HLA alleles consisting of HLA-C*04 (e.g., HLA-C*04:01), HLA-C*07 (e.g., HLA-C*07:02), HLA-C*16 (e.g., HLA-C*16:01), HLA-C*03 (e.g., HLA-C*03:04), HLA-C*07 (e.g., HLA-C*07:01), and HLA-C*06 (e.g., HLA-C*06:02), preferably the two HLA-C class I alleles are selected from the group of human HLA alleles consisting of HLA-C*04 and HLA-C*07. 20. Mammals are: (a) HLA-A*03, (b) HLA-A*11, (c) HLA-B*07, (d) HLA-B*15, (e) HLA-C*04, (f) HLA-C*07, and (g) A non-human mammal according to any one of the preceding paragraphs, comprising in its genome six or seven human leukocyte antigen (HLA) class I alleles selected from the group of human HLA alleles comprising or consisting of HLA-A*02. 21. The seven human HLA alleles are: (a) HLA-A*03:01, (b) HLA-A*11:01, (c) HLA-B*07:02, (d) HLA-B*15:01, (e) HLA-C*04:01, (f) HLA-C*07:02, and (g) A non-human mammal according to any one of the preceding clauses, comprising or consisting of HLA-A*02:01. 22. A non-human mammal according to any one of the preceding clauses, wherein at least two human HLA alleles are functionally expressed such that corresponding MHC I polypeptides are expressed on the surface of cells of the mammal. 23. The non-human mammal according to any one of the preceding clauses, wherein the corresponding MHC I polypeptide is expressed on the surface of peripheral blood cells of the mammal. 24. The non-human mammal according to any one of the preceding clauses, wherein the corresponding MHC I polypeptide expressed on the surface of a cell of the mammal presents an MHC-presented antigen that provides an antigen-specific CD8+ T cell response by the non-human mammal (e.g., said antigen-specific CD8+ T cell response is measured by any suitable means, e.g., FACS). 25. The non-human mammal according to any one of the preceding clauses, wherein at least two human HLA alleles encode a chimeric human / non-human mammal MHC molecule. 26. The non-human mammal according to any one of the preceding clauses, wherein the at least two HLA alleles encode a chimeric human / non-human mammal MHC I polypeptide, wherein the human portion of the chimeric polypeptide comprises the α1 domain and the α2 domain of a human MHC I polypeptide, and wherein the non-human mammal portion of the chimeric polypeptide comprises the α3 domain, the transmembrane domain, and the cytoplasmic domain of an endogenous non-human mammal MHC I polypeptide, and wherein the non-human mammal expresses the chimeric human / non-human mammal MHC I polypeptide. 27. The non-human mammal according to any one of the preceding clauses, wherein the at least two HLA alleles encode a chimeric human / non-human MHC I polypeptide, wherein the human portion of the chimeric polypeptide comprises the α1, α2, and α3 domains of a human MHC I polypeptide, and wherein the non-human mammal portion of the chimeric polypeptide comprises the transmembrane and cytoplasmic domains of an endogenous non-human mammal MHC I polypeptide, and wherein the non-human mammal expresses the chimeric human / non-human mammal MHC I polypeptide. 28. Each of the at least two HLA alleles is monocistronic, preferably each of the at least two HLA alleles monocistronic (a) HLA-A*03, (b) HLA-A*11, (c) HLA-B*07, (d) HLA-B*15, (e) HLA-C*04, and (f) A non-human mammal according to any one of the preceding clauses, which is selected from the group of human HLA alleles comprising or consisting of HLA-C*07. 29. The non-human mammal of any one of the preceding clauses, wherein each HLA allele comprises its own endogenous mouse H-2Db promoter (e.g., having or comprising SEQ ID NO: 227) and polyadenylation signal (e.g., having or comprising SEQ ID NO: 228). 30. A non-human mammal according to any one of the preceding clauses, wherein the mammal comprises a complete human T cell receptor (TCR) locus in its genome. 31. The non-human mammal of any one of the preceding clauses, wherein the mammal has a defect in its endogenous T cell receptor (TCR) locus. 32. The non-human mammal according to any one of the preceding clauses, wherein the non-human mammal is selected from the group consisting of rodents, dogs, felines, primates, rabbits, pigs, and ruminants. 33. The non-human mammal according to any one of the preceding clauses, wherein the rodent is a mouse or a rat. 34. A nucleic acid construct comprising a nucleic acid encoding at least two (e.g., at least three, at least four, at least five, at least six, or at least seven, etc.) human HLA class I alleles, wherein said nucleic acid construct is capable of functionally expressing said at least two human HLA alleles (e.g., in a non-human mammal, e.g., as described herein) such that corresponding MHC I polypeptides are expressed on the surface of a cell of the non-human mammal comprising the nucleic acid construct to present MHC antigens that provide an antigen-specific CD8+ T cell response by the non-human mammal, and optionally wherein said at least two human HLA class I alleles are (a) at least one (e.g., at least two or at least three) human HLA-A alleles, and / or (b) at least one (e.g., at least two) human HLA-B alleles, and / or (c) at least one (e.g., at least two) human HLA-C alleles, and / or (d) (a) and (b), and / or (e)(a) and (c), and / or (f)(b) and (c), and / or (g) A nucleic acid construct comprising (a), (b), and (c). 35. The nucleic acid construct of any one of the preceding clauses, wherein the nucleic acid further encodes at least one human HLA class II allele. 36. The nucleic acid construct of any one of the preceding clauses, wherein the nucleic acid construct comprises four, five, six, or seven human HLA class I alleles. 37. The nucleic acid construct comprises: (a) at least two different human HLA-A alleles, preferably at least three different human HLA-A alleles, and / or (b) at least two different human HLA-B alleles, and / or (c) The nucleic acid construct of any one of the preceding clauses, comprising at least two different human HLA-C alleles. 38. The nucleic acid construct according to any one of the preceding clauses, wherein the at least two human HLA class I alleles comprise one or more HLA-A class I alleles selected from the group consisting of HLA-A*03 (e.g., HLA-A*03:01), HLA-A*11 (e.g., HLA-A*11:01), HLA-A*01 (e.g., HLA-A*01:01), HLA-A*26 (e.g., HLA-A*26:01), HLA-A*24 (e.g., HLA-A*24:02), HLA-A*32 (e.g., HLA-A*32:01), and HLA-A*02 (e.g., HLA-A*02:01), and preferably the three HLA-A class I alleles are selected from the group of human HLA alleles consisting of HLA-A*03, HLA-A*11, and HLA-A*02. 39. The nucleic acid construct according to any one of the preceding clauses, wherein the at least two human HLA class I alleles comprise one or more HLA-A class I alleles selected from the group consisting of HLA-A*11 (e.g., HLA-A*11:01), HLA-A*24 (e.g., HLA-A*24:02), and HLA-A*33 (e.g., HLA-A*33:03), preferably HLA-A*24 (e.g., HLA-A*24:02) and HLA-A*33 (e.g., HLA-A*33:03). 40. The nucleic acid construct according to any one of the preceding clauses, wherein the at least two human HLA class I alleles comprise one or more HLA-A class I alleles selected from the group consisting of HLA-A*24 (e.g., HLA-A*24:02), HLA-A*02 (e.g., HLA-A*02:01), HLA-A*11 (e.g., HLA-A*11:01), and HLA-A*31 (e.g., HLA-A*31:01), preferably HLA-A*24 (e.g., HLA-A*24:02) and HLA-A*31 (e.g., HLA-A*31:01). 41. The nucleic acid construct according to any one of the preceding clauses, wherein the at least two human HLA class I alleles comprise one or more HLA-A class I alleles selected from the group consisting of HLA-A*01 (e.g., HLA-A*01:01), HLA-A*11 (e.g., HLA-A*11:01), and HLA-A*24 (e.g., HLA-A*24:02), preferably HLA-A*01 (e.g., HLA-A*01:01) and HLA-A*24 (e.g., HLA-A*24:02). 42. The nucleic acid construct according to any one of the preceding clauses, wherein said at least two HLA-A class I alleles are selected from the group of human HLA alleles consisting of HLA-A*03 (e.g., HLA-A*03:01), HLA-A*11 (e.g., HLA-A*11:01), HLA-A*01 (e.g., HLA-A*01:01), HLA-A*26 (e.g., HLA-A*26:01), HLA-A*24 (e.g., HLA-A*24:02), HLA-A*32 (e.g., HLA-A*32:01), and HLA-A*02 (e.g., HLA-A*02:01), and preferably said at least three HLA-A class I alleles are selected from the group of human HLA alleles consisting of HLA-A*03, HLA-A*11, and HLA-A*02. 43. The nucleic acid construct according to any one of the preceding clauses, wherein the at least two HLA class I alleles comprise one or more HLA-B class I alleles selected from the group consisting of HLA-B*07 (e.g., HLA-B*07:02), HLA-B*15 (e.g., HLA-B*15:01), HLA-B*58 (e.g., HLA-B*58:01), HLA-B*40 (e.g., HLA-B*40:01), HLA-B*35 (e.g., HLA-B*35:01), and HLA-B*08 (e.g., HLA-B*08:01), preferably the at least two HLA-B class I alleles are selected from the group of human HLA alleles consisting of HLA-B*07 and HLA-B*15. 44. The nucleic acid construct according to any one of the preceding clauses, wherein the at least two HLA class I alleles comprise one or more HLA-B class I alleles selected from the group consisting of HLA-B*40 (e.g., HLA-B*40:01), HLA-B*46 (e.g., HLA-B*46:01), HLA-B*58 (e.g., HLA-B*58:01), and HLA-B*15 (e.g., HLA-B*15:02), preferably HLA-B*40 (e.g., HLA-B*40:01) and HLA-B*46 (e.g., HLA-B*46:01) (China). 45. The nucleic acid construct according to any one of the preceding clauses, wherein the at least two HLA class I alleles comprise one or more HLA-B class I alleles selected from the group consisting of HLA-B*52 (e.g., HLA-B*52:01), HLA-B*51 (e.g., HLA-B*51:01), HLA-B*35 (e.g., HLA-B*35:01), HLA-B*15 (e.g., HLA-B*15:01), and HLA-B*40 (e.g., HLA-B*40:02), preferably HLA-B*52 (e.g., HLA-B*52:01) and HLA-B*51 (e.g., HLA-B*51:01) (Japan). 46. ​​The nucleic acid construct according to any one of the preceding clauses, wherein the at least two HLA class I alleles comprise one or more HLA-B class I alleles selected from the group consisting of HLA-B*40 (e.g., HLA-B*40:06), HLA-B*51 (e.g., HLA-B*51:01), HLA-B*52 (e.g., HLA-B*52:01), and HLA-B*44 (e.g., HLA-B*44:03), preferably HLA-B*40 (e.g., HLA-B*40:06) and HLA-B*51 (e.g., HLA-B*51:01) (India). 47. The nucleic acid construct according to any one of the preceding clauses, wherein said at least two HLA-B class I alleles are selected from the group of human HLA alleles consisting of HLA-B*07 (e.g., HLA-B*07:02), HLA-B*15 (e.g., HLA-B*15:01), HLA-B*58 (e.g., HLA-B*58:01), HLA-B*40 (e.g., HLA-B*40:01), HLA-B*35 (e.g., HLA-B*35:01), and HLA-B*08 (e.g., HLA-B*08:01), preferably, the two HLA-B class I alleles are selected from HLA-B*07 and HLA-B*15. 48. The nucleic acid construct according to any one of the preceding clauses, wherein the at least two HLA class I alleles comprise one or more HLA-C class I alleles selected from the group consisting of HLA-C*04 (e.g., HLA-C*04:01), HLA-C*07 (e.g., HLA-C*07:02), HLA-C*16 (e.g., HLA-C*16:01), HLA-C*03 (e.g., HLA-C*03:04), HLA-C*07 (e.g., HLA-C*07:01), and HLA-C*06 (e.g., HLA-C*06:02), and preferably the two HLA-C class I alleles are selected from the group of human HLA alleles consisting of HLA-C*04 and HLA-C*07. 49. The nucleic acid construct according to any one of the preceding clauses, wherein the at least two HLA class I alleles comprise one or more HLA-C class I alleles selected from the group consisting of HLA-C*07 (e.g., HLA-C*07:02), HLA-C*01 (e.g., HLA-C*01:02), and HLA-C*03 (e.g., HLA-C*03:04), preferably HLA-C*01 (e.g., HLA-C*01:02) and HLA-C*03 (e.g., HLA-C*03:04) (China). 50. The nucleic acid construct according to any one of the preceding clauses, wherein the at least two HLA class I alleles comprise one or more HLA-C class I alleles selected from the group consisting of HLA-C*01 (e.g., HLA-C*01:02), HLA-C*03 (e.g., HLA-C*03:03), and HLA-C*03 (e.g., HLA-C*03:04), preferably HLA-C*01 (e.g., HLA-C*01:02) and HLA-C*03 (e.g., HLA-C*03:03) (Japan). 51. The nucleic acid construct according to any one of the preceding clauses, wherein the at least two HLA class I alleles comprise one or more HLA-C class I alleles selected from the group consisting of HLA-C*06 (e.g., HLA-C*06:02), HLA-C*04 (e.g., HLA-C*04:01), HLA-C*07 (e.g., HLA-C*07:02), and HLA-C*15 (e.g., HLA-C*15:02), preferably HLA-C*06 (e.g., HLA-C*06:02) and HLA-C*15 (e.g., HLA-C*15:02) (India). 52. The nucleic acid construct according to any one of the preceding clauses, wherein said at least two HLA-C class I alleles are selected from the group of human HLA alleles consisting of HLA-C*04 (e.g., HLA-C*04:01), HLA-C*07 (e.g., HLA-C*07:02), HLA-C*16 (e.g., HLA-C*16:01), HLA-C*03 (e.g., HLA-C*03:04), HLA-C*07 (e.g., HLA-C*07:01), and HLA-C*06 (e.g., HLA-C*06:02), preferably said at least two HLA-C class I alleles are selected from the group of human HLA alleles consisting of HLA-C*04 and HLA-C*07. 53. Mammals contain seven human leukocyte antigen (HLA) class I alleles in their genome. The seven human HLAs are: (a) HLA-A*03, (b) HLA-A*11, (c) HLA-B*07, (d) HLA-B*15, (e) HLA-C*04, (f) HLA-C*07, and (g) The nucleic acid construct according to any one of the preceding clauses, wherein the nucleic acid construct is selected from the group of human HLA alleles comprising or consisting of HLA-A*02. 54. The seven human HLA alleles are (a) HLA-A*03:01, (b) HLA-A*11:01, (c) HLA-B*07:02, (d) HLA-B*15:01, (e) HLA-C*04:01, (f) HLA-C*07:02, and (g) A nucleic acid construct according to any one of the preceding clauses, comprising or consisting of HLA-A*02:01. 55. The nucleic acid construct according to any one of the preceding clauses, wherein the HLA allele encodes a chimeric human / non-human mammalian MHC molecule. 56. The nucleic acid construct of any one of the preceding clauses, wherein the at least two HLA alleles encode a chimeric human / non-human mammalian MHC I polypeptide, wherein the human portion of the chimeric polypeptide comprises the α1 domain and the α2 domain of a human MHC I polypeptide, and the non-human mammalian portion of the chimeric polypeptide comprises the α3 domain, the transmembrane domain, and the cytoplasmic domain of an endogenous non-human mammalian MHC I polypeptide. 57. The nucleic acid construct of any one of the preceding clauses, wherein the at least two HLA alleles encode a chimeric human / non-human MHC I polypeptide, wherein the human portion of the chimeric polypeptide comprises the α1, α2, and α3 domains of a human MHC I polypeptide, and the non-human mammalian portion of the chimeric polypeptide comprises the transmembrane and cytoplasmic domains of an endogenous non-human mammalian MHC I polypeptide. 58. Each of the at least two HLA alleles is monocistronic, preferably each of the at least three HLA alleles monocistronic: (a) HLA-A*03, (b) HLA-A*11, (c) HLA-B*07, (d) HLA-B*15, (e) HLA-C*04, and (f) The nucleic acid construct according to any one of the preceding clauses, which is selected from the group of human HLA alleles comprising or consisting of HLA-C*07. 59. Each HLA allele is provided with its own endogenous H-2Db promoter and polyadenylation signal, and preferably, at least two of the above-mentioned HLA alleles are (a) HLA-A*03, (b) HLA-A*11, (c) HLA-B*07, (d) HLA-B*15, (e) HLA-C*04, and (f) The nucleic acid construct according to any one of the preceding clauses, which is selected from the group of human HLA alleles comprising or consisting of HLA-C*07. 60. The nucleic acid construct of any one of the preceding clauses, wherein the endogenous and own H-2Db promoter is an MHC H-2 class I promoter (e.g., having or comprising SEQ ID NO: 227). 61. The polyadenylation signal (pA) is a bovine growth hormone (bGH) polyadenylation signal (e.g., having or comprising SEQ ID NO: 228), and preferably, at least two of the above-mentioned HLA alleles are (a) HLA-A*03, (b) HLA-A*11, (c) HLA-B*07, (d) HLA-B*15, (e) HLA-C*04, and (f) The nucleic acid construct according to any one of the preceding clauses, which is selected from the group of human HLA alleles comprising or consisting of HLA-C*07. 62. The nucleic acid construct encodes a 2A peptide linker (e.g., having or including SEQ ID NO: 229) such that the encoded at least two human MHC class I molecules are separated by a 2A peptide linker, and preferably the at least two HLA alleles described above are (a) HLA-A*03, (b) HLA-A*11, (c) HLA-B*07, (d) HLA-B*15, (e) HLA-C*04, and (f) HLA-C*07 and / or HLA-C*08. 63. The nucleic acid construct of any one of the preceding clauses, further encoding a leader sequence (e.g., having or comprising SEQ ID NO: 230). 64. The nucleic acid construct according to any one of the above clauses, wherein the nucleic acid construct is according to any one of Figures 1 to 22 disclosed herein (e.g., the nucleic acid construct is flanked by an ITR, e.g., a 5'-piggyback-ITR, e.g., having or comprising SEQ ID NO: 231, and a 3'-piggyback-ITR, e.g., having or comprising SEQ ID NO: 232). 65. The nucleic acid construct of any one of the preceding clauses, wherein the leader sequence is a human β2-microglobulin (β2m) leader sequence (e.g., having or comprising SEQ ID NO: 233). 66. The nucleic acid construct according to any one of the preceding clauses, wherein the nucleic acid construct is codon-optimized for expression in said non-human mammal. 67. An expression vector comprising the nucleic acid construct of any one of the preceding clauses. 68. A host cell (e.g. a recombinant and / or isolated and / or non-human host cell) comprising a nucleic acid construct according to any one of the preceding clauses, and / or an expression vector according to any one of the preceding clauses. 69. A method for modifying endogenous HLA alleles of a non-human mammal, comprising transducing and / or transplanting into said non-human mammal a nucleic acid construct according to any one of the above clauses and / or an expression vector according to any one of the above clauses. 70. The method according to any one of the preceding clauses, wherein transduction is carried out by pronuclear microinjection. 71. A method for generating a non-human mammalian oocyte having a modified target sequence in its genome (e.g., by means of transposon-mediated targeting, e.g., as described in the experimental section herein), comprising the step of introducing into the non-human mammalian oocyte a nucleic acid construct (e.g., an ITR-flanked targeting vector (e.g., an HLA-bearing cassette)) described in any one of the preceding paragraphs, and / or an expression vector described in any one of the preceding paragraphs. 72. The method according to any one of the preceding clauses, wherein the nucleic acid construct according to any one of the preceding clauses and / or the expression vector according to any one of the preceding clauses is injected into the nucleus / pronucleus of the oocyte or introduced into the oocyte by electroporation. 73. A method for producing a non-human mammal having a modified target sequence in its genome, comprising: (a) producing an oocyte as described in any one of the preceding paragraphs; and (b) analyzing offspring delivered by a non-human female host into which the oocyte obtained in (a) was transferred for the presence of the modification. 74. The method according to any one of the preceding clauses, wherein the method comprises a transposon-mediated targeting step. 75. The method according to any one of the preceding clauses, wherein the non-human mammal is selected from the group consisting of rodents, dogs, felines, primates, rabbits, pigs, and ruminants. 76. The method according to any one of the preceding clauses, wherein the rodent is a mouse or a rat. 77. A non-human mammal produced and / or modified by the method according to any one of the preceding clauses. 78. A method for producing one or more T cell receptors capable of binding to an antigen of interest, the method comprising administering the antigen of interest to a non-human animal as described in any one of the preceding paragraphs. 79. The method of any one of the preceding clauses, wherein the antigen of interest is a full-length polypeptide or a fragment of a full-length polypeptide. 80. The method according to any one of the preceding clauses, wherein the antigen is a tumor-specific antigen (TSA) or a tumor-associated antigen. 81. The method of any one of the preceding clauses, further comprising isolating T cells that bind the antigen of interest from cells of the non-human animal. 82. The method of any one of the preceding clauses, further comprising isolating nucleic acid encoding a T cell receptor that binds the antigen of interest from a cell of the non-human animal. 83. The method according to any one of the preceding clauses, wherein the T cell receptor is a human T cell receptor. 84. A T cell receptor obtained or obtainable by a method according to any one of the preceding clauses. 85. A method for identifying an epitope capable of eliciting an immune response, comprising administering to a non-human animal as described in any one of the preceding paragraphs an antigen of interest suspected of containing an epitope capable of eliciting an immune response. 86. The method according to any one of the preceding clauses, wherein the antigen of interest is a full-length polypeptide or a fragment of a full-length polypeptide. 87. The method according to any one of the preceding clauses, wherein the epitope is selected from the group consisting of a tumor-specific epitope, a viral epitope, a neoepitope, and a human self-epitope. 88. The method of any one of the preceding clauses, further comprising isolating T cells that bind the antigen of interest from cells of the non-human animal. 89. The method of any one of the preceding clauses, further comprising isolating nucleic acid encoding a T cell receptor that binds the antigen of interest from a cell of the non-human animal. 90. The method of any one of the preceding clauses, further comprising recombinantly producing the T cell receptor and determining the epitope to which the T cell receptor binds. 91. The method according to any one of the preceding clauses, wherein the method is a method of unbiased epitope detection. 92. The method of any one of the preceding clauses, wherein the epitope is an MHC I-restricted epitope. 93. An epitope capable of eliciting an immune response and identified by a method according to any one of the preceding clauses, preferably the epitope GTLEEVPTA of MAGE-A1 (SEQ ID NO: 200). 94. Use of a non-human animal according to any one of the preceding clauses for generating one or more T cell receptors capable of binding to an antigen of interest. 95. Use of a non-human animal according to any one of the preceding clauses to identify one or more T cell receptors capable of binding to an antigen of interest. 96. The use according to any one of the preceding clauses, wherein the antigen is selected from the group consisting of a tumor-specific antigen, a viral antigen, a neoepitope, and a human self-antigen. 97. Use of a non-human animal according to any one of the preceding clauses for identifying epitopes capable of eliciting an immune response. EXAMPLES

[0123] In order that the present invention may be readily understood and put into practice, certain aspects of the invention are described by the following non-limiting examples.

[0124] Example 1 material and method material In silico software algorithms Table 1 lists web-based resources used for T cell epitope prediction screening. [Table 1]

[0125] Databases and Datasets Table 2 lists the obtained data sets for HLA frequency, ranking, and cancer incidence. [Table 2]

[0126] Cell media, flasks, tubes, and plates Table 3 shows a list of cell culture media. * Standard media preparation: DMEM / RPMI + 10% (i.e., 50 ml per 500 ml) heat-inactivated sterile-filtered fetal bovine serum + 100 IU / ml penicillin and 100 μg / ml streptomycin. ♯ Mouse T cell medium (mTCM): * Standard RPMI medium + 1 mM sodium pyruvate + 0.1 mM non-essential amino acid solution + 2 mM HEPES buffer + 50 μM β-mercaptoethanol. [Table 3]

[0127] Table 4 shows the list of consumables. [Table 4]

[0128] Chemical Reagents Table 5 shows a list of chemical reagents. [Table 5] JPEG2025502421000006.jpg152149

[0129] buffer solution Table 6 shows a list of buffers. [Table 6]

[0130] cell line Table 7 shows a list of cell lines. [Table 7]

[0131] Cytokines Table 8 shows a list of cytokines. [Table 8]

[0132] Antibodies and primers Table 9 shows a list of antibodies. [Table 9]

[0133] Table 10 shows a list of primers. [Table 10] JPEG2025502421000012.jpg224149JPEG2025502421000013.jpg179149

[0134] mouse Table 11 shows a list of transgenic and wild type mice with relevant TCR loci and HLA genotypes. [Table 11]

[0135] research equipment Table 12 shows the list of equipment. [Table 12] JPEG2025502421000016.jpg147149

[0136] software Table 13 shows a list of commercially available software. [Table 13]

[0137] method Design and construction of chimeric HLA class I alleles in silico ABab.I polycistronic transgenes encoding six HLA alleles separated by 2A peptide linkers were designed in silico and synthesized by GeneArt (Thermo Fisher Scientific). The HLA alleles of the transgenes are chimeric human-mouse fusion proteins. The α1 and α2 domains are human cDNAs from the respective HLA alleles, and the α3 to cytoplasmic domains are mouse cDNAs from the H-2 Db gene. Human β2m is linked to the N-terminus by a 15 amino acid glycine-serine peptide linker. Modified HLA constructs with H-2 Db or CMV promoters, 3'-bGH polyadenylation signal mutants, and HLA single chains with or without β2m were custom cloned in-house. Phusion® DNA polymerase combined the HLA amplicons by splicing by overhang extension PCR (SOE). The six HLA alleles were sequentially cloned as a single haplotype using unique restriction enzyme sites into pre-integrated multiple cloning sites. A piggyback ITR-flanked ABab.I transgene was generated using 5'-AscI and 3'-PacI sites. The long HLA plasmid containing the repeat sequences was propagated by Stbl3 E. coli (Thermo Fisher Scientific, Rockford, USA).

[0138] In vitro characterization of HLA-ABab.I transgenes by flow cytometric analysis Surface HLA class I expression was analyzed by flow cytometry 48 hours after all transient transfections. In transfection experiments, MCA205 cells were treated or not with recombinant IFN-γ (100 ng / ml). The primary polycistronic ABab.I transgene and all modified HLA constructs were transfected with 10 μl of Lipofectamine® 2000. In all cases, cells were stained with fluorochrome-conjugated antibodies against human pan-HLA-ABC and β2-microglobulin 2 days after transfection. MCA205 / ABab.I stable cell lines were transfected with AscI-PacI-digested linearized DNA constructs into MCA205 cells, followed by HLA expression through flow cytometry-based cell sorting using pan-HLA-ABC antibodies. + It was generated by multiple enrichment of the cell population.

[0139] Quantification of transgene mRNA using quantitative RT-PCR analysis H-2D b MCA205 cells were transiently transfected with the polyadenylation signal modified construct driven by the CMV promoter or CMV promoter as described above. After 2 days, total RNA in cell lysates was reverse transcribed using anchored oligo(dT) and random hexamers. cDNA amplification was quantified using SYBR® green dye with HLA-specific primers in quantitative RT-PCR (qRT-PCR) reactions. As controls, 18s rRNA (housekeeping) and homeostatic H-2 D b The reaction was analyzed. From the Ct values, the difference in expression ΔCt = Ct(HLA / H-2 D b The percentage of expression change was calculated using the 18s rRNA-Ct(18s rRNA) ... b The measurements were made by normalizing to (set at 100%).

[0140] Construction of genome targeting vectors Six chimeric HLA alleles, HLA-A*03:01, A*11:01, B*07:02, B*15:01, C*04:01, and C*07:02, were cloned into the minimal mammalian expression vector pMA plasmid as natural HLA haplotypes similar to humans. For mouse genome targeting, the 5'-AscI and 3'-PacI sites were sequentially digested and blunted to insert the highly active PiggyBac restriction inverted terminal repeat (ITR) to generate the oocyte targeting vector.

[0141] Generation of ABab.I transgenic mice using pronuclear injection technique In vitro transcribed (ivt) piggyback (PB) transposase mRNA targeted the ITR-flanked ABab.I transgene cassette into recipient oocytes. 10-12 week old ABabDII male mice served as mating partners. To collect zygotes, 6-10 week old ABabDII female mice were superovulated before mating. NMRI mice served as pseudopregnant foster mothers. Donor HLA plasmids (ABab.I transgene) were co-prepared with PB transposase (ivt mRNA) in a 2:1 molar ratio. DNA mixed with ivt mRNA samples was pronuclear microinjected into fertilized ABabDII oocytes as described above. All mouse experiments were performed in accordance with standard guidelines approved by the Berlin State Agency for Occupational, Health and Technical Safety (Landesamt fur Arbeitsschutz, Gesundheitsschutz und technische Sicherheit).

[0142] Genotyping of HLA I haplotypes in ABab.I Tg mice Genomic DNA was extracted from ear biopsies of transgenic mice at 95°C in the presence of 0.05M NaOH. Fragments were amplified using conventional PCR with Taq DNA polymerase with HLA-specific primers for all six chimeric fusion alleles. 18s rRNA and human TCRαβ served as control reactions. PCR amplicons were run on 1.5% gels at 100V for 45 min using an agarose gel electrophoresis system. The PCR primers were as follows: [Table 14] JPEG2025502421000019.jpg10158

[0143] Estimation of absolute CD3+CD8+ / CD4+ counts in blood and lymphoid organs Blood preparation: Fc receptor III / II was blocked by anti-CD16 / 32 antibody in 50 μl blood per mouse (each strain). Red blood cells were lysed by red blood cell lysis buffer (ACK buffer) before staining and flow cytometry. Lymphoid organ preparation: Spleens and lymph nodes (axillary, brachial, mediastinal, inguinal, and mesenteric) from 8-12 week old mice were crushed and filtered through a 70 μm cell strainer. ACK-lysed spleens and lymph nodes were passed through a 40 μm strainer and then pooled together. Blood and lymphoid organ staining for T cell counting: Samples were stained using fluorescent dye-conjugated CD3, CD8, and CD4 antibodies. CountBright® counting beads were used in a volume of 11 μl (11,880 beads) per sample for T cell quantification by flow cytometry.

[0144]

number

[0145] Enrichment of T cell populations (CD3+CD8+) using non-binding FACS sorting Lymphocytes and whole blood from young ABabDII or ABab.I mice (8–12 weeks) were prepared using the protocol described above (see 2.2.7 "Preparation of blood and lymphoid organs") and pooled. CD3 + Cells were enriched from the pooled cell suspension by column-free bead-based magnetic separation using the EasySep® Mouse T Cell Nonbinding Isolation Kit (Stemcell Technologies GmbH, Cologne, Germany). Briefly, cell suspensions (2 × 10 8 Fc receptors in the 100% IgG4-dependent IgG4 antibodies (IgG4-dependent IgG4 antibodies) were blocked by incubation with 50 μl / ml of rat serum. Pooled cells from spleen, lymph nodes, and blood were cultured to identify non-T cells (CD3 - Streptavidin-coated magnetic spheres (75 μl / ml) were used to stain these CD3 - The cells were captured on an EasySep® magnet while CD3 + The cells passed through a magnetic field and were collected in the flow-through solution. + Cell populations were surface stained with fluorescent antibodies specific for CD3, CD8, and CD4 markers. CD3+CD8+ T cells (>=97% pure) were enriched by flow cytometry-based sorting for organic DNA (genomic) extraction using the phenol-chloroform method.

[0146] Deep sequencing of the TCRβ repertoire in ABab.I mice using ImmunoSEQ® technology Deep sequencing of the TCRβ immune repertoire was performed at Adaptive Biotechnologies (Seattle, USA). The ImmunoSEQ® platform accurately quantifies all VJ gene combinations in a cell population using a multiplex PCR-based assay that is bias-corrected in the first step to minimize amplification bias. The assay also provides quantitative abundance data with a sensitivity of 1 in 200,000 T cells. 3×10 5From the mice, 2–4 μg of genomic DNA / sample (average for each mouse) and approximately 1.8 × 10 5 Human CD8 + 1.5 μg of genomic DNA / sample obtained from T cells were deep sequenced. Three healthy human donors (ages: 30, 48, and 60 years) donated blood with informed consent. Blood collection and processing were performed in accordance with the human experimentation guidelines under license EA4 / 046 / 10 (Ethics Committee). Computational analysis was performed on the sequencing data by Adaptive Biotechnologies. The ImmunoSEQ® Analyzer Portal was accessed. TCR sequences were named according to the Immunogenetics (IMGT) nomenclature. The R program was used to identify CD4 + Statistical analysis was performed as described for T cells.

[0147] Retroviral transduction of mouse splenocytes Ecotropic retrovirus (infects mouse / rat cells only) packaging cell line Platinum-E 188 (Plat-E) was transfected with pMP71 retroviral plasmid (3 μg / construct) encoding T cell receptor genes (construct: TCRβ-mCβ-P2A-TCRα-mCα) using lipofectamine at 80% cell culture density. 3 ml of viral supernatant was collected twice at different time points using a 0.45 μm filter membrane to infect mouse ABabDII spleen cells. Spleen and lymph nodes were prepared and pooled using the protocol described above (see 2.2.7 Preparation of lymphoid organs). Spleen cells were preactivated with anti-CD3 (1 μg / ml) and anti-CD28 (0.1 μg / ml) antibodies in the presence of 40 IU / ml recombinant mouse IL-2 (cell concentration was 2 × 10 6Transduction was performed twice, 48 and 72 hours after gene transfer. In the first transduction, the supernatant containing the viral particles was first centrifuged (3000g, 90 min at 4°C) on a retronectin-coated non-tissue culture 24-well plate. 2 × 10 cells were cultured in mTCM medium supplemented with 8 μg / ml protamine sulfate (ps), 10 μl CD3 / CD28 Dynabeads®, and 40 IU / ml IL-2. 6 of preactivated splenocytes were added onto the virus capture plate and centrifuged at 800g for 30 min at 32°C. For the second transduction, 1 ml of the supernatant was removed from the top layer of the 24-well plate and replaced with a fresh virus batch (8 μg / ml ps, 40 IU / ml IL-2) taken at 72 h and centrifuged at 800g for 90 min at 32°C. After the second transduction, 2 × 10 transduced splenocytes were added to the virus capture plate and centrifuged at 800g for 90 min at 32°C. 6 The cell concentration was adjusted to 50 cells / ml, and 50 ng / ml mouse IL-15 was added.

[0148] Functional characterization of HLA alleles expressed in ABab.I Tg mice in vitro IFN-γ production was monitored in the supernatants of the coculture assays by enzyme-linked immunosorbent assay (ELISA). 4 5 x 10 TCR-transduced effector T cells and 4 Overnight (16 h) co-culture was performed with target cells (MCA205 / ABab.I tumor cells or peripheral blood cells / lymphoid organs of ABab.I mice) in a volume of 1000 μg / mL. Target cells were pulsed with peptides (1 μM / peptide) or phorbol myristate acetate (PMA) and ionomycin (1 μM each) as positive controls.

[0149] Characterization of HLA-ABab.I transgenes in vivo by peptide immunization 200 μl of injection suspension containing 100 μg of short peptide, 100 μl of incomplete Freund's adjuvant, and 50 μg of CpG oligonucleotide in PBS was injected subcutaneously into both lateral sides of the hind paws. Young ABabDII or ABab.I mice (6–7 mice / group) aged 10–20 weeks were immunized with a 4-week interval between prime and boost. Similar peptide mixtures were prepared and injected for further boosting regimes of all peptides. ABabDII was used as a standard control. 1 and ABab.I mice were injected with 100 μg of MAGE-A1 in the same injection mixture. 278 Mice were immunized with KVLEYVIKV (SEQ ID NO: 26) peptide. Seven to ten days after each boost injection, peripheral blood cells from immunized mice were ACK lysed and cultured with 1 μM of each specific or non-specific peptide for 5 to 6 hours in the presence of protein transport inhibitors. After cell fixation, cells were stained intracellularly for IFN-γ and surface stained for CD3, CD8 markers and analyzed by flow cytometry.

[0150] statistical analysis GraphPad Prism software was used to generate standard curves, plot graphs, and calculate statistical significance levels (P values) using means and standard deviations. FlowJo was used to analyze the flow cytometry raw data (.fcs files) including statistics of mean fluorescence intensity values.

[0151] result Cancer mutagenesis screening yields potential neoantigens for adoptive T cell therapy Recurrent hotspot mutations that may give rise to putative tumor-specific antigens (TSAs) were screened in silico using an open-source server (Figure 2, Table 1, and Table 2). Using a literature survey, 266 "nonsynonymous somatic point mutations" previously described in 40 different genes were selected as TSAs (Figure 2A). The selected point mutations (occurring in ≥2 patients) were primarily detected by whole-exome sequencing analysis of 21 different human cancer elements. The mutation frequencies ranged from 0.1% to 39%. The NetMHC program predicted the affinity of the mutant and wild-type epitopes for HLA alleles under the applied selection conditions (Figure 2B). Of these 266 mutants, 175 epitopes were found to have strong binding affinity (IC) in silico to at least one of the 18 prevalent HLA class I alleles. 50 :<50 nM) (Figure 2C). The HLA-A2-expressing ABabDII mouse model provided a platform to analyze the immunogenicity of 23 HLA-A2-restricted binders (Figure 2D, left). From 152 non-HLA-A2 binders, 68 putative epitopes bound the human-like novel class I haplotypes "HLA-A*0301-A*11:01, B*07:02-B*15:01, C*04:01-C*07:02". The restrictive state of these six HLA alleles for multiple (>5) mutant epitopes provided the rationale for generating ABab.I mice (Figure 2D, right).

[0152] Human T cell receptors (TCRs) isolated from ABabDII and ABab.I mice can be applied in the clinic for adoptive T cell therapy (ATT) of cancer against endogenously processed mutant antigens.

[0153] The ABabDII mouse model induces HLA-A2-restricted neoantigen-specific T cell responses HLA-A2-restricted neoantigen-specific CD8 +To address T cell responses, we selected a frequently occurring mutation, S722F, in the N-terminus of the transformation / transcription domain-associated protein (TRRAP) gene. A serine to phenylalanine mutation at position 722 had an IC50 of 55 nM in a predictive screen. 50 We generated HLA-A2 epitopes bearing the TRRAP protein (Figure 3A and Figure 3B). The TRRAP protein is a transcriptional cofactor involved in cell transformation via activation of MYC. 4% of melanoma cases, totaling 86,000 onsets / year, express the S722F driver mutation. Considering that this 4% with S722F express HLA-A2 at an allele frequency of 26.5% in Europe, we estimated that approximately 911 individuals per year could have ATT (theoretical) after stratification from the previous treatment group (Figure 3B). Similarly, we calculated global estimates of individuals / year carrying the TRRAP-S722F mutation using data sets from individual continents, such as the United States, South America, and Asia (data not shown). Mice carry a TRRAP orthologue to ensure thymic deletion of wild-type epitope-restricted T cells (Figure 3C).

[0154] To detect TCR candidates, young ABabDII mice (n=7, 12-16 weeks) were immunized with the mutant 9-mer peptide (Figure 3B, KLVFGSV(F)L, SEQ ID NO: 30) at 4-week intervals. + T cell responses were analyzed in blood one week after each boost using intracellular staining, followed by culture of T cells and selection of peptide-specific T cells in an IFN-γ capture assay (Figure 4A). One week after the fourth injection (third boost) regimen, approximately 0.8% of T cells responded in blood (Figure 4B), which corresponds to the peptide (10 -9 After culturing T cells for 10 days under low concentrations of IL-2 (20 IU / ml) and 50 IU / ml, the T cells expanded to approximately 17% in the spleen (Figure 4C). After T cell culture, 12,000 IFN-γ specific for S722F were expressed. + CD8 + Cells were cultured at 10 -6M peptide concentration after 2 h of stimulation (Figure 4D). No wild-type specific T cell responses were observed after in vitro restimulation (Figures 4B, 4C, and 4D). 5'-RACE PCR yielded one dominant TCRαβ pair (Figure 4E). 5'-RACE amplification work used primers annealing to the constant TCRαβ and anchor 5'-adapter regions as previously described.

[0155] In the figure (Figure 4E), the following SEQ ID NOs correspond to the listed CDR3 regions: [Table 15]

[0156] To ensure TCR interaction and affinity for S722F, transduced T cells were titrated with peptide titration T2 (TAP - / - ) cells. α 17-V s 27 pairs produced IFN-γ without recognizing the wild-type epitope, resulting in a 10 -11 The epitope S722F was recognized with sufficient sensitivity up to the M peptide concentration (Figure 4F, left). + Human cell lines were co-cultured with transduced T cells, which did not recognize A375 or SK-Mel-37 cells, which are positive for mutant and wild-type TRRAP, respectively. 157-165 is expressed in melanoma cells A375 and SK-Mel-37. TCR-ESO (control TCR) inhibits NY-ESO by producing IFN-γ. 157-165 epitope was recognized (Fig. 4F, right side).

[0157] Therefore, the neoantigen S722F derived from the TRRAP mutation 715-723 It has been demonstrated that the nonamer is not endogenously processed and presented to HLA-A2.

[0158] ABab.I transgenes designed to reflect natural human HLA haplotypes Twenty-three mutant epitopes from the TRRAP, C-KIT, XPO1, FOXA1, RAC1, RAC2, RHOT1, TP53, MAP2K1, EGFR, TRAF7, SMO, SF3B1, FBXW7, SPOP, EZH2, FLT3, NFE2L2, TSHR, PTPN11, and NOTCH1 genes are predicted to bind HLA-A2 (Figure 2C, pie chart). Mutant peptides from the first four genes were analyzed for immunogenicity as well as TRRAP-S772F (Figure 4A). ABabDII mice increased CD8 expression after immunization with epitopes from C-KIT (K642E), XPO1 (E571K), and FOXA1 (D226N). + (Data not shown) ABabDII mice provide the opportunity to isolate TCRs restrictive to only these 23 HLA-A2 binders.

[0159] 8 / 152 non-HLA-A2 binders (Figure 2C, pie chart) are predicted to bind at least one allele of the novel HLA haplotypes A*0301-A*11:01, B*07:02-B*15:01, and C*04:01-C*07:02, which are similar to the human HLA status (all individuals have 6 genes). To understand the biology of the entire human haplotype in mice and to target the other 68 high affinity mutants (Table 16), we designed the ABab.I transgene based on the results of the in silico screen to reflect the human HLA haplotype (Figure 5).

[0160] Table 16 shows in silico predicted mutant epitopes that bind the ABab.I HLA haplotype. [Table 16] JPEG2025502421000023.jpg228161JPEG2025502421000024.jpg228161JPEG2025502421000025.jpg228161 JPEG2025502421000026.jpg225161JPEG2025502421000027.jpg232161JPEG2025502421000028.jpg184161

[0161] Prediction programs suggested that HLA-A*03:01 binds a total of 8 mutant epitopes, and A*11:01 binds up to 21 epitopes, B*07:02 binds 6 epitopes, B*15:01 binds 12 epitopes, C*04:01 binds 13 epitopes, and C*07:02 showed restriction to 8 epitopes in silico (Figure 5A and Table 16).

[0162] The total number of individuals / year with the mutation based on the three factors predicts the global ATT treatment probability (Figure 5B). + We estimated that approximately 70,389 individuals / year carry a mutation that creates the targeted epitope. + , about 1,381 B*07:02 + , about 2,651 people B*15:01 + , about 21,532 C*04:01 + , and approximately 10,451 C*07:02 + were theoretically calculated as individuals likely to carry the mutation (Figure 5C). In total, approximately 117,624 individuals / year worldwide were estimated to carry these 68 recurrent point mutations in cancer, predicted epitopes as listed (Table 16) that bind to at least one allele of the new class I haplotype. Similar to the human HLA architecture, the six HLA alleles of high occurrence in the ABab.I transgene design were encoded as polycistronic expression cassettes (Figure 5D).

[0163] Polycistronic HLA transgenes are transcribed to produce mRNA but are not translated to express protein The polycistronic primary ABab.I and all modified constructs (Figure 6A) did not express any HLA alleles after staining with pan-HLA ABC antibody (Figure 6B). HLA expression was measured 48 hours after transient gene transfer. b Promoted ABab.I showed no surface HLA expression in MCA205 cells. 98% of LCL-BM14 cells showed endogenous HLA expression, with 39% of cells expressing GFP (Figure 6B,I). LCL-BM14 are EBV-transformed lymphoblastoid B cells.

[0164] To access promoter activity, six HLA were transfected with the enhanced GFP (eGFP) gene or H-2 D b The H-2 D promoter was switched to that of the CMV promoter (Fig. 6A, II). b In contrast to 53% of cells expressing GFP induced by H-2 D, 53% of cells expressed GFP from the control pMP71-eGFP retroviral plasmid (Figure 6B, II). b Promoter switching to CMV did not restore expression of the six HLAs (Fig. 6B, II).

[0165] Next, 500bp H-2 D b The polyadenylation (pA) sequence was amplified by the 1 kb full-length (FL)H-2 D b or replaced with a 350 bp synthetic bovine growth hormone (bGH) pA and analyzed for defects in transcription termination. b Alternatively, the CMV promoter was driven by pA modified constructs, which were analyzed for surface expression using flow cytometry staining (FIG. 6A, III). None of the pA modified constructs showed significant HLA surface expression after transient gene transfer (data not shown).

[0166] H-2 D with all individual pA modifications (Figure 7A) bThe H-2D promoter and CMV promoter-driven primary ABab.I constructs produced mRNA in MCA205 cells (Figure 7B). HLA mRNA was detected using cDNA amplification with specific primers for the three viral peptide linkers P2A (front region), F2A (middle region), and C2A (terminal region). b The promoter-driven bGH pA construct suppressed mouse MHC expression (endogenous D in MCA205 cells) after transient transfection. b H-2 D resulted in up to 7% HLA expression compared to H-2 D b The promoter-driven FL pA and primary ABab.I constructs yielded 6% and 2% HLA expression, respectively (Fig. 7B, I). The CMV-driven bGH pA construct did not express endogenous D b A significant relative change of 193% in mRNA production versus expression was measured. FL pA and the primary ABab.I construct of CMV significantly suppressed the basal H-2 D b showed relative changes of 138% and 75%, respectively (Fig. 7B, II).

[0167] HLA alleles are stably expressed when prompted by their own 5' and 3' regulatory elements H-2 D in a polycistronic configuration with bGH pA b The six promoter driven HLA alleles made HLA mRNA but did not produce protein (FIGS. 6 and 7).

[0168] Six HLA alleles were cloned as monocistronic with or without β2 microglobulin (β2m) (Figure 8A, I), their expression was tested, and then they were arranged together in a single class I haplotype with the respective regulators (Figure 8A, II). The single chains without β2m (linked) were exogenously co-transfected with β2m plasmid. The respective leader sequences translocate HLA in the single chains without β2m linkage, for example, in the HLA-A*03:01 construct, the A*03 leader peptide was cloned in place of the β2m leader sequence.

[0169] MCA205 cells were transfected with or without β2m-linked single chains (Fig. 8A, I) in the presence or absence of IFN-γ. HLA single chains transfected without β2m (co-transfected with β2m) did not show any significant difference in surface expression of HLA alleles (data not shown). And all six monocistronic (as β2m-linked single chains) were stained using pan-HLA ABC and anti-human β2m antibodies and measured by flow cytometry in transiently transfected MCA205 cells (Fig. 8B).

[0170] Pan-HLA ABC antibodies stained transiently transfected MCA205 cells and showed that increases in HLA expression ranged from 9- to 23-fold in single chains of A*03 (5.4%-52%), A*11 (6%-59%), B*07 (5.6%-57%), B*15 (5.6%-56%), C*04 (2.3%-52%), and C*07 (2.6%-48%) with 100 ng / ml IFN-γ treatment for 48 h at the time of transfection (Figure 8B). Similarly, anti-human β2m stained transiently transfected MCA205 cells and showed increases in HLA expression ranging from 1.5- to 8-fold in single chains of A*03 (8.6%-15%), A*11 (9.3%-22%), B*07 (5.4%-20%), B*15 (3.9%-18%), C*04 (1.6%-12%), and C*07 (3.7%-14.6%) treated with 100 ng / ml IFN-γ for 48 h at the time of transfection (Figure 8C).

[0171] The ABab.I transgenes encoding the six chimeric monocistrons were transfected into the MCA205 cell line and subjected to HLA expression analysis. The HLA haplotype of ABab.I encodes six consecutive alleles in a head-to-tail configuration. All six genes contain their own promoter and polyadenylation signal (Figure 8A, II).

[0172] Two days after gene transfer, 16% of the cells were positive for HLA alleles (Figure 9A, top).+ The population was enriched. After 2 days of sorting and in vitro culture, cells were pretreated with or without IFN-γ and stained using a pan-HLA ABC antibody. While 41% of cells stained for HLA without IFN-γ, 77% of MCA205 cells stained for HLA alleles after 48 h in culture and enrichment in the presence of IFN-γ (Figure 9A, bottom). At the same time, HLA-specific antibodies were used to stain the sorted cells (HLA + ) were stained for HLA expression. The increase in HLA expression ranged from 2- to 4-fold after staining with A*03-specific (19%-73%), B*07-specific (17%-69%), and pan-HLA-C (32%-69%) antibodies after 48 h of 100 ng / ml IFN-γ treatment (Figure 9B, Figure 9C, and Figure 9D). AscI-PacI digested linearized ABab.I DNA was used for MCA205 transfection to generate the stable cell line MCA205 / ABab.I. HLA was identified by flow cytometry sorting at different time points. + After multiple rounds of cell enrichment, 92% of the cells expressed the ABab.I transgene (Figure 9E).

[0173] ABab.I founder mice carrying six HLA genes as a single haplotype were generated by piggyBac transposon-mediated targeting. The ABab.I HLA cassette was shown to be expressed transiently (FIGS. 9A and 9B) and stably (FIG. 8C) on the surface of mouse MCA205 cells in vitro.

[0174] In vivo integration of HLA haplotypes was achieved using the hyperactive PiggyBac (PB) transposon system. The PB transposase functions on an ITR-flanked targeting vector (a cassette carrying HLA) by a cut-paste mechanism. The enzyme was co-transfected as mRNA and excised (cut) the 5' and 3' ITR regions, releasing the insert (HLA) which was then integrated (pasted) into the transcriptionally active sites of the mouse genome (Figure 10).

[0175] After successful pronuclear injection, allele-specific PCR confirmed the presence of the six HLA alleles in two founder animals (Figure 11A). Genotyping PCR was used to confirm genomic integration of the alleles in F0_Q4115 and F0_Q4118 by ear biopsy DNA (Figure 11A, left). As controls, the 18s rRNA housekeeping gene and the human TCRαβ gene were examined (Figure 11A, right). Three additional founders, F0_Q6844, F0_Q8552, and F0_Q8556, were positively genotyped (data not shown). A stable homozygous ABab.I mouse line from the Q6844 line was established following standard breeding patterns (Figure 11B).

[0176] Anti-human β2m antibodies were used to stain and quantitate HLA expression in peripheral blood cells from C57BL6 / N, ABabDII, and ABab.I mice, and in LCL-BM14 cells (99% β2m + ) served as controls (Figures 12A and 12B). 3% and 7% of lymphocytes in ABabDII mice and 35% and 40% of lymphocytes in ABab.I mice were stained for HLA expression with β2m antibody (Figure 12A). Two ABab.I mice of founder line Q6844 are shown here. Other founders tested for HLA expression include Q4115, Q4118, Q8552, and Q8556 (data not shown). Mouse-to-mouse variation was analyzed by staining multiple mice per line. Mean fluorescence intensity (MFI) comparison of β2m revealed 4-fold more HLA expression in ABab.I compared to ABabDII. A 120-fold higher MFI was observed in human HLA (non-chimeric)-expressing LCL-BM14 cells (Figures 12C and 12D). One possible explanation is that the folding properties of human HLA proteins are different from those of the chimeric fusion molecules, which may hinder the accessibility of the epitope to the β2m antibody, but this requires further analysis.

[0177] ABab.I mice have higher CD8+ T cell counts compared to single HLA-bearing ABabDII mice Phenotypic characterization of T cells showed increased thymic output / peripheral survival in ABab.I mice (founder F0_Q6844 line) compared to ABabDII and HHD mice (Figure 13). HHD mice have a murine TCR repertoire selected against human HLA-A2. CD8 expression of peripheral blood cells of the four mouse lineages by flow cytometry was significantly increased in ABab.I mice (founder F0_Q6844 line) compared to ABabDII and HHD mice (Figure 13). + CD4 + Profiling (CD3 + In ABabDII, 13% and 14% of CD8 + T cells, about 6% in both HHD mice, 36% and 45% in ABab.I, and 36% and 41% in C57BL6 / N mice (Figure 13A). Two mice / strains are shown here. CD8 in ABab.I mice + The percentage of T cells was similar to that in C57BL6 / N mice. On average, CD3 + T cells were 90% CD4 + ABabDII mice had 64%, ABab.I mice had 45%, and C57BL6 / N mice had 55% (Figure 13A). A significant number of ABab.I mice had 4-fold more CD8 + The number of T cells was accumulated (Figure 13B). + ABab.I mice with high T cell counts were crossed in F2 to generate stable mouse lines (Figure 11B). + No significant differences in T cell numbers were observed (Fig. 13C). The CD8 / CD4 ratio was high in ABab.I mice, similar to that in C57BL6 / N mice, and threefold higher than that in ABabDII mice (Fig. 13D).

[0178] CD8 in pooled lymphoid organs + and CD4 + T cell population (CD3 + The comparison of CD8+ cells showed that ABabDII inhibited 14% and 12% of CD8+ cells. + T cells, 62% and 67% CD4 +T cells, and 47% and 44% of CD8 + T cells, 39% CD4 + On average, ABab.I mice had 3.5-fold more CD8 T cells in major secondary lymphoid organs (Figure 14A). + The absolute number of T cells was estimated (Figure 14B), and CD4 + There was no significant difference in absolute T cell numbers (Fig. 14C). The CD8 / CD4 ratio was higher in ABab.I mice, as in peripheral blood, and 2.5-fold higher than that in ABabDII mice (Fig. 14D).

[0179] Broader CD8+ T cell repertoire in ABab.I mice compared with ABabDII mice Deep sequencing of the TCRβ repertoire revealed that CD8 + In T cells, 11.2 ± 2.7 × 10 4 and 4.8 ± 0.80 × 10 4 The unique in-frame amino acid (aa) clonotypes of ABab.I were quantified (Figure 15A). From three age-unmatched humans, a 30-year-old donor had 11.4 ± 4.1 × 10 compared to ABab.I mice. 4 The total number of unique aa clonotypes in human donors was highly variable. Given the large difference in age (30, 50, and 65 years), human data were excluded from the analysis of clone size. ABab.I mice had 8.5% rare clones, whereas ABabDII mice had 5.7% rare clones (Fig. 15B). However, ABabDII mice had 3-fold more small clones (4.1%) than ABab.I mice with 1.4% small clones (Fig. 15B). Entire repertoire sequencing was not possible using ImmunoSEQ® deep sequencing. Thus, mouse and human CD8 +An estimate of the total TCR repertoire in T cells was determined using computational statistics. The iChao1 estimator calculated the total number of clonotypes (observed and undetected) per mouse using information on clones that occurred only once or twice. iChao1 uses the lower bound of rare occurrence to estimate the true species richness. The first human donor, a 30-year-old individual, had a maximum of 2 × 10 6 They have the most diverse repertoire of clonal types, with an average of 1 × 10 6 (n=3). Following humans, ABab.I mice had a diverse repertoire of 0.66×10 6 (n=5). ABabDII had 1.8×10 5 The least common clonotype was V (Figure 15C). β (Figure 16A) and J β (FIG. 16B) Gene usage was analyzed. The latter represents the preselection pool, which is an unbiased measure of the diversity of all possible V(D)J recombination events that could have occurred before positive selection. All detected V(D)J genes were 100% expressed, except for TRBV5-1 and TRBV6-1, whose absence or lack of expression was previously reported in ABabDII mice and, therefore, in ABab.I mice. β The majority of the segments were found to be rearranged (Figure 16A). ABabDII and ABab.I mice were not able to express any favorable V in the preselection pool. β Similarly, in humans, the use of out-of-frame V β Gene usage patterns were comparable without any selective preference.

[0180] In both the pre- and post-selection pools, some V β Genes were over- or under-represented in mice. β The genes include TRBV20, TRBV21, TRBV23, TRBV27, and TRBV28. βThe genes include TRBV6-2 / 6-3, TRBV6-4, TRBV7-9, TRBV9-1, and TRBV10-1, most of which are closer to the 5' end (Figure 16A, top). Single nucleotide polymorphisms may be the reason for this difference in appearance. Compared to humans, in ABabDII and ABab.I mice, two Vs, such as TRBV7-3 and TRBV12-3 / 12-4, are more abundant than in ABabDII. β Preferential overrepresentation of genes was observed (Fig. 16A, top).

[0181] In-frame V s Some genes in the 5' region, such as TRBV2-1 and TRBV4-1, and TRBV7-9, are highly selected by positive selection, while V such as TRBV7-3, TRBV20 / 21 / 23 / 24, and TRBV25 are highly selected by positive selection. s These represent post-selection pools of similar frequency, except that the genes were not selected in the thymus (Fig. 16A, bottom). s 4-1, and V in ABabDII mice. s 12-3 / 12-4, one V in each system s Only genes appear to be predominantly represented in the post-selection pool of either mouse (FIG. 16A, bottom).

[0182] Also, J β Gene usage was also nonrandom and similar between transgenic mice and humans (Figure 16B), although some J genes, TRBJ1-2, TRBJ2-1, and TRBJ2-3, were not randomly used. β The segment was more frequently used in mice than in humans (Fig. 16B, top). In particular, the J β 2-07 was more highly represented in mice than in humans, with approximately 30% total usage in the post-selection pool (FIG. 16B, bottom).

[0183] Longer CDR3-containing T cell receptors are enriched in ABab.I mice The V(D)J recombination events of the TCR β chain generate CDR3 diversity, which, together with the VJ events of the TCR α chain, is responsible for antigen recognition. In ABab.I mice, significantly longer CDR3 β regions were enriched than in ABabDII mice.

[0184] Considering the shorter CDR3 lengths, in ABabDII 5.5% of clonotypes expressed TCRs with CDR3 sequences of 33 bp, 11.5% of 36 bp, and 21% of 39 bp. In ABab.I, the lengths corresponded to CDR3 sequences of 4.9% 33 bp, 10% 36 bp, and 19% 39 bp. In ABab.I mice, a higher percentage of clones expressed longer CDR3 sequences than in ABabDII mice: 23% vs. 21% at 48 bp, 10% vs. 8.3% at 51 bp, and 3.7% vs. 3.4% at 54 bp (Figure 17A). However, in humans, the longest CDR3β sequences are produced compared to mice (Figure 17A). On average, ABab.I mice had TCRs with a CDR3 length of 42±0.1 bp, compared with a CDR3 length of 41.5±0.06 bp in ABabDII mice, but less than humans, which have an average CDR3 of 43.5±0.18 bp (Figure 17B).

[0185] Natural human HLA haplotypes in ABab.I mice educate diverse unique clones Shared clones were compared between mouse strains and humans using Jaccard similarity index estimation. The Jaccard index (J) evaluates the immune repertoire shared between samples. The J index ranges from 0 to 1, and the score is calculated based on the formula of the total number of shared clones between two samples divided by the total number of unique clones, i.e., J(A,B)=A∩B / A∪B. We detected higher shared clones within strains (ABab.I and ABab.I, or ABabDII and ABabDII) than between strains (ABab.I and ABabDII, and vice versa). ABab.I mice shared about 6.4% of clones with each other, with a Jaccard index of 0.0636 ± 0.003, and ABabDII mice shared about 5% of clones within their group, with a Jaccard index of 0.0498 ± 0.004 (Figure 18A). Humans shared more clones with ABab.I mice (Jaccard index: 0.006±0.001) than with ABabDII mice (Jaccard index: 0.005±0.001) or each other, but this difference was not statistically significant (FIG. 18A).

[0186] Both ABabDII and ABab.I generated more shared TCRβ clones with each other than with humans, likely reflecting the similarity of the human TCR locus and genetic background. Notably, ABab.I mice shared fewer clones with ABabDII (Jaccard index: 0.04 ± 0.003) than within the group (0.0636 ± 0.003) (Figure 18A).

[0187] ABab.I mice generated more unique clones compared to ABabDII mice. The total number of shared and unique clones was determined by pooled analysis from five mice per line. ABabDII and ABab.I mice generated fewer than 100,000 clones between each other (0.8 × 10 5 ) were shared between the ABab.I mice and the ABabDII mice. 5 generated nearly four times more unique and rare clones than TCRβ clones (ABab.I: 7.5 × 10 5 ) (Figure 18B).

[0188] HLA alleles in ABab.I mice efficiently present epitopes and induce T cell responses ex vivo To determine epitope presentation by HLA alleles in newly generated ABab.I mice, peripheral blood or lymphoid organ cells isolated from mice ex vivo were co-cultured with TCR-transduced T cells recognizing model epitopes. Previously described TCRs against a panel of model antigens obtained from literature sources (A3-, A11-, B7-, B15-, and C7-restricted) or an in-house generated TCR (C4-restricted) were used in the co-culture assays (Figure 19A). The following SEQ ID NOs apply to Figure 19A:

[0189] [Table 17]

[0190] Co-culture supernatants were analyzed for mouse IFN-γ to evaluate effector T cells that recognize HLA alleles bound to each model epitope. As a positive control, MCA205 / ABab.I cells, which express six HLA alleles similar to ABab.I mice, were used as target cells to present the model epitopes to the transduced T cells. After 16 h of co-culture, IFN-γ was measured in the supernatants to confirm functional activity. All five model TCRs recognized their respective peptide-HLA combinations (pHLA-TCR) and produced IFN-γ ranging from 9437 ± 863 pg / ml to 11,877 ± 299 pg / ml (Figure 19B).

[0191] To measure pHLA-TCR interactions in ex vivo ABab.I mice, peripheral blood and lymphoid organ cells were used as target cells. After recognizing blood pHLA, all TCR-transduced T cells produced IFN-γ, with the lowest release being 4168 ± 308 pg / ml and the highest release being 7873 ± 212 pg / ml (Figure 19C), proving functional activity. Similarly, pHLA from secondary lymphoid organs (spleen, lymph nodes as target cells) was recognized, and all model TCR-transduced T cells released IFN-γ in the range of 5540 ± 216 pg / ml to 8054 ± 270 pg / ml (Figure 19C).

[0192] PMA / ionomycin activates protein kinase C and increases calcium (Ca2+) levels in transduced T cells to maximize IFN-γ release. + ) channel. In contrast, no IFN-γ release was measured in non-transduced T cells (only small background) (Figures 19A and 19B). T1367-transduced T cells, which are HLA-A2 (present in ABab.I mice) restricted TCR, were used in co-culture as an internal reference control in all presentation assays. A model TCR against the CMV pp65 epitope was isolated in-house from immunized HuTCR mice expressing HLA-C*04:01 as a single chain (see Discussion section 4.5). All model epitopes used in the co-culture assays (Figure 19A) were carried forward to in vivo immunization of ABab.I mice and were shown to be upregulated by IFN-γ release in CD8 T cells using intracellular staining. + T cell responses were analyzed.

[0193] ABab.I mice mount immune responses to peptide antigens in vivo. Antigen-specific CD8 expression was observed in ABab.I mice against the model antigens described (Figure 19A) using peptide immunization. + T cell responses were analyzed (Figure 20). All six model antigens were predicted in silico to be strong binders to their respective HLA alleles. Approximately 1.8% of CD8 +It was measured that T cells produced IFN-γ against the mutant calreticulin (mCALR) protein-derived epitope, KMRMRRMRR (SEQ ID NO: 173). Approximately 6.9% of CD8 + T cells released IFN-γ to the mutant KRAS protein-derived G12V epitope VVGAVGVGK (SEQ ID NO: 174). Approximately 1.4% released IFN-γ to the B*07:02-binding CMV epitope TPRVTGGGAM (SEQ ID NO: 175), approximately 0.2% to the B*15:01-binding HPV epitope SAFRCFIVY (SEQ ID NO: 176), approximately 0.3% to the C*04:01-binding CMV epitope QYDPVAALF (SEQ ID NO: 177), and approximately 0.23% to the MAGE-A12 epitope VRIGHLYIL (SEQ ID NO: 178) (Figure 20). The HLA-A2-bound MAGE-A1 epitope KVLEYVIKV (SEQ ID NO: 179) was released by approximately 1% of CD8 + T cell responses were elicited. The non-specific peptide used in the in vitro restimulation was the NY-ESO epitope APRGPHGGAASGL (SEQ ID NO: 180). No T cell responses to the non-specific peptide were observed in any case, demonstrating a specific response to the immunizing peptide. CD8 + T cell responses ranged from 1.7% to 14% (Figure 20). Thus, in the ABab.I mouse model, CD8 T cells were expressed against the group of epitopes described. + T cell responses demonstrate the functionality of human HLA haplotypes in vivo.

[0194] Consider In the course of this invention, an "ABab.I transgenic mouse model" was generated using a piggyback transposon strategy with six HLA class I alleles in a "humanized TCR" mouse background. Human HLA haplotypes in mice were confirmed by genotyping the six HLAs using allele-specific PCR. Phenotypes were confirmed by staining with all available pan-HLA and HLA-specific antibodies using flow cytometry analysis. CD8 + and CD4 + We provided evidence supporting increased thymic output in ABab.I mice by profiling the T cell receptors of ABab.I mice. We used ImmunoSEQ® deep sequencing to characterize the immune repertoire and observed that ABab.I mice have a four-fold broader T cell receptor repertoire than ABabDII mice. We characterized the antigen-presenting efficacy of all six HLA alleles present in ABab.I mice by expressing effector CD8 T cells ex vivo and in vivo. + This was demonstrated using an IFN-γ release assay by T cells.

[0195] The need for computational neoantigen prediction and epitope immunogenicity validation - Mutation neoepitope S722F is not endogenously processed Of the 266 mutations, 175 class I mutant epitopes were predicted to bind 18 highly ranked (with respect to population) HLA alleles. All 266 mutations were somatic point mutations that occurred recurrently more than twice in different cancer entities. We focused only on recurrent driver mutations, as they are present in all cancer cells throughout the disease and provide truly tumor-specific antigens. Early evidence suggests that chemotherapy with agents that inactivate cancer driver oncogenes and tumor-specific CD8 +We compared ATT treatment and demonstrated that eradication of escape mutants along with destruction of tumor stroma resulted in tumor elimination without recurrence. Based on this, we hypothesized that the production of TCRs against recurrent neo-antigens derived from driver mutations may enable complete eradication of tumors without antigen loss and recurrence.

[0196] In silico screening, epitopes are identified by IC 50 pMHC affinity was recorded as IC 50 between <1 nM and >20,000 nM. Engels et al. investigated pMHC affinity along with treatment outcomes in tumor rejection versus recurrence. This study was carried out using IC 50 Strong pMHC affinity of <10 nM led to tumor rejection in all cases, and IC 50 reported that relapse occurred when the concentration was >100 nM. Two reviews subsequently reported that CD8 + High affinity for ATT treatment (IC 50 Based on these studies, this predictive screen used ABabDII and ABab.I (produced as part of the present invention) to generate TCRs against these repetitive antigens, highlighting the importance of selecting class I epitopes with a repetitive antigen concentration of 100 nM. 50 We selected mutated neoepitopes with a molecular weight of less than 50 nM. Not all predicted neoantigens yield immunogenic epitopes. Epitope immunogenicity depends on the natural endogenous processing for pMHC presentation and the interaction with CD8 +This is combined with the ability to induce T cell responses. Epitopes defined in silico should be evaluated with caution as they may result in false positives. In this study, we predicted that the TRRAP-S722F driver mutation contains a 9-mer epitope that is a putative neo-antigen. The S722F mutant epitope induced an immune response in ABabDII mice. The TCR isolated from the mice was specific for S722F and recognized small amounts of exogenously loaded peptide. However, it did not react to melanoma cell lines that naturally express the mutant gene, confirming that the 9-mer epitope is not endogenously processed and presented. This paradoxical immunology approach demonstrated that the S722F predicted epitope is not endogenously processed. In addition to the TRRAP mutant (S722F), the genetically predicted neoepitopes C-KIT (K642E), XPO1 (E571K), and FOXA1 (D226N) upregulated CD8 expression in ABabDII mice. + did not induce a T cell response (proprietary data not shown and not shown).

[0197] Several other findings confirmed the notion that paradoxical immunology (predicting epitopes first and investigating endogenous processing last) is not the best way to select target antigens for adoptive T cell therapy. In epitope immunogenicity studies of vaccinia virus, about 100 epitopes were found to be IC10-specific for HLA-A2. 50 They showed good predicted binding affinities of <100 nM. Only 50% of them expressed CD8 + induced T cell responses. 15% of the response-eliciting peptides were further processed by the proteasome. Of these, 11% were expressed in spontaneous CD8 T cells upon vaccinia virus infection. + They elicited responses, demonstrating that only 1 / 14 of the predicted subset generated immunogenic epitopes. 126 In other melanoma RNA vaccination trials, CD4 + CD8 compared with T cell responses +T cell responses were relatively low, which may be due to more promiscuous class II peptide presentation by the MHC. Furthermore, in another long peptide melanoma vaccination trial, only 16% of predicted neoepitopes were identified by CD8 + It induced a T cell response.

[0198] Software programs do not adequately predict endogenous epitope modifications in antigen presentation pathways using their algorithms. These changes within the proteasomal, post-proteasomal, and peptide transport compartments are important factors for determining epitopes as immunogenic and targetable. In one study, we demonstrated epitope destruction by the proteasome of a proposed epitope. In another example, it was reported that epitopes trimmed by ERAP were more prone to escape than non-trimmed epitopes. In a third study, human individuals showed CD8 expression against a CDK4 mutant antigen. + Although it demonstrated a T cell response, this was later validated in experimental cancer models as a relatively poor target for T cell therapy.

[0199] Thus, artificial network algorithms cannot accurately predict proteasomal processing, peptide transport, N-terminal trimming by ERAP enzymes, or splicing mechanisms to define human neoantigens for ATT in cancer.

[0200] Polycistronic transgene design requires consideration of mRNA stability and protein misfolding-Lessons learned The polycistronic primary ABab.I transgene separated by viral 2A linker sequences did not show expression of surface proteins from any of the six HLA alleles. Thus, the three polyadenylation signals (short D b , total length D b , and bGH pA) were used to bWe engineered the primary constructs driven by the promoter. At the same time, the promoter sequence in these three constructs was replaced with the viral CMV promoter. All six of these constructs were found to be H-2 D by qRT-PCR. b Although we quantified significant amounts of HLA mRNA in the primary constructs, more than 30-fold higher in the CMV primary constructs, we did not detect cell surface protein expression by flow cytometry. This untranslatable state of HLA mRNA, affecting cell surface presentation, could be caused by multiple reasons. Such considerations could be mRNA instability of long HLA transcripts, inefficient excision of the viral 2A linker, and unfavorable post-translational modifications causing misfolding of the nascent polypeptide chain.

[0201] We considered that mRNA stability is influenced by a single common 3'-UTR region that regulates all six HLA alleles. Furthermore, recent evidence confirmed that the native 2A linker functions inefficiently by not self-cleaving to release the protein. Of note, the primary transgene contained the native 2A peptide linker. Sequence divergence versus homology studies reported that transient misfolding occurs in proteins derived from polycistronic mRNAs but is abolished when sequence identity is low. HLA class I chimeric chains maintain 90-98% homology. In line with this, a recent review postulated that there is an open question as to how nascent polypeptides from polycistrons are folded without forming aggregates for cellular clearance.

[0202] Based on these factors, it was decided to clone and express all six HLA alleles as monocistronic driven by their own 5'- and 3' regulatory elements (DNA sequence of ABab.I transgene construct in founder mice - SEQ ID NO: 208).

[0203] Protein characterization of chimeric HLA in ABab.I mice and the requirement for single-chain and specific antibodies - Human lymphoblastoid cell lines express high levels of HLA proteins Single HLA alleles were cloned from the primary polycistronic transgene and recloned as monocistronic for successful surface expression. To confirm cell surface expression of HLA, we always stained with pan antibodies or antibodies against the human β2m chain present in all chimeric alleles. With the exception of HLA-A*03:01 and HLA-B*07:02, there are no specific antibodies present against unique HLA alleles in ABab.I mice that can be used for flow cytometric analysis, making validation of HLA proteins difficult.

[0204] Notably, both pan-HLA and β2m antibodies stained different HLA chains with different intensities, and in particular β2m antibodies stained HLA genes of lymphoblastoid cell lines (LCLs) with a higher fluorescence intensity profile (MFI) than the chimeric alleles of ABab.I mice. Notably, the HLA genes of LCL lines are fully human, where pHLA stabilization occurs by antigens derived from the human proteome. The HLA alleles of ABab.I mice are chimeric, where pHLA stabilization occurs by antigens derived from the mouse proteosome. This difference in expression patterns raised the question of whether it was the importance of the peptide to HLA stability that allowed β2m antibodies to better stain fully human HLA and increase the shift in mean fluorescence intensity.

[0205] Interestingly, studies using HLA antibodies from parous women confirmed that peptides influence HLA reactivity by antibodies. Based on this, we speculate that the more stable pHLA is at the surface, the better it can be detected by antibody staining. Thus, we could correlate the difference in alloantibody staining with the difference in β2m staining between LCL and ABab.I mice. Another possible explanation could be that the chimeric HLA of ABab.I mice is folded differently in an unnatural way, which prevents epitope accessibility by β2m antibodies, which requires further analysis.

[0206] Another point of note is that expression of fully human HLA in transgenic mice without β2m linked as a fusion protein may increase surface stability and therefore improve detection. However, the maintenance of the interaction between mouse CD8α and the α3 domain of the chimeric mouse-human HLA heavy chain is not consistent with the detection of CD8α. + It is important for co-activation of T cells.

[0207] Generation of ABab.I transgenic mice using pronuclear injection technique Two ABab.I transgene targeting vectors were generated for injection into oocytes of ABabDII donors: the first vector was flanked by Rosa26 guide RNA sequences for CRISPR-Cas9 recognition, and the second vector was flanked at both the 5' and 3' ends by terminal inverted repeat sequences for hyperactive PiggyBac recognition.

[0208] Exploratory attempts to use CRISPR-Cas9 technology by homology-independent targeted integration (HITI) were unsuccessful after several attempts. Injection of naked DNA transgenes without a vector backbone to randomly integrate was similarly unsuccessful. We considered that using homology arms flanking the transgene cassette for efficient integration into the Rosa26 locus in HITI using CRISPR-Cas9 could be a better strategy. Thus far, the HITI method has only shown high integration efficiency in postmitotic neurons, not in oocytes, so successful attempts in mouse embryos will require careful optimization. DNA strand breaks in the mouse genome during the pronuclear microinjection procedure are the primary cause of random transgene integration in a head-to-tail fashion. During chromosome end joining, the breakpoint ends take up the available DNA transgene as a donor template and ligate the junctions. It is speculated that the large size (~17 kb) of the ABab.I transgene with homologous sequences could be the reason for the failure of integration of the naked transgene as a concatemer. Using a piggyback transposon strategy, one copy per locus could be targeted.

[0209] Therefore, at the same time, we utilized the hyperactive PiggyBac transposase to catalyze the ABab.I transgene cassette with ITRs for integration into the mouse genome. Five founders, Q4115, Q4118, Q6844, Q8552, and Q8556, were successfully generated. The first founder, Q4115, did not transmit its genomic HLA alleles in the germline to the F1 generation of nearly 250 genotyped mice. It is believed that PiggyBac catalyzed the transgene to integrate into the genome after multiple cell divisions of the embryo or to integrate at a later stage, which resulted in mosaic founder mice that could not transmit the germline to the F1 generation as per Mendelian genetics. Nevertheless, we were able to achieve germline transmission in all other founders, Q4118, Q6844, Q8552, and Q8556.

[0210] CD8 in blood and secondary lymphoid organs + and CD4 + The phenotypes of the different founder mice were investigated based on the absolute number of T cells (data for the other founder mice except for the Q6844 line are not shown). The Q6844 line had the highest and most consistent CD8 T cell counts from F1 to F4 generations. + These results provide indirect evidence for HLA expression in the thymus and possibly in the periphery. High levels of HLA expression are required for positive selection in the thymus and for T cell maintenance in the periphery. Thus, CD8 + We hypothesized that the high number of T cells (Q6844 lineage) reflects efficient thymic positive selection and increased diversity of the TCR repertoire. Furthermore, the impact on T cell homeostatic proliferation is another interesting question to address. Therefore, it is important to consider CD44 staining as an activation marker to understand the thymic output and peripheral homeostasis based on the number of precursor T cells entering the circulation.

[0211] However, using next-generation ImmunoSEQ repertoire sequencing, we found that CD8 + We were able to deep sequence the TCRβ immune repertoire of T cells.

[0212] Next-generation deep sequencing of ABab.I and ABabDII mice-CD8 + TCRβ immunosequencing CD8 in ABab.I and ABabDII mice + After observing the difference in absolute numbers of T cells, CD8 + To understand whether the increased T cell numbers led to a more extensive TCR repertoire, we performed deep sequencing of the TCRβ repertoire in these lineages. If so, we hypothesized that CD8β expression in ABab.I mice compared with ABabDII mice was significantly greater in ABab.I mice than in ABabDII mice. +The increase in T cell numbers could be excluded as a result of homeostatic proliferation. Of note, ABabDII mice have a single HLA-A2, whereas ABab.I mice additionally have a fully human HLA haplotype with six alleles. Thus, we investigated whether the introduction of multiple class I alleles into humanized TCR mice would result in a broader repertoire, and how broader it would be compared to ABabDII mice.

[0213] Using iChao1 estimation, we demonstrated that ABab.I mice have a diverse repertoire that is, on average, four times more extensive than ABabDII mice. We deep sequenced approximately two-fold more in-frame amino acid clonotypes in ABab.I vs. ABabDII mice (11.2 ± 2.7 × 10 4 vs. 4.8±0.80×10 4 ), the iChao1 estimator theoretically extrapolated the repertoire to be four-fold more extensive and back-calculated to the full mouse repertoire by including clones that occurred only once or twice (lower limit events). By theoretical extrapolation, ABab.I mice use diverse V and J genes, up to 2.57 × 10 5 clonotype (approximately 4-fold less), compared with ABabDII mice that generated only 1 × 10 6We showed that ABab.I mice generate clonotypes that are closer to the original V(D)J TCRβ combinations. The additional diversity in ABab.I mice reflects a higher percentage of rare and unique clones, and that thymic selection has indeed selected for novel V(D)J TCRβ combinations against multiple HLA alleles. Thus, expression of six HLA alleles in ABab.I mice selects for a highly diverse human TCR repertoire, likely due to enhanced positive selection and overall thymic output that is reduced in ABabDII mice. The out-of-frame TCR repertoire represents the pre-selection pool before positive selection and provides an accurate estimate of how frequently individual V and J segments are rearranged in ABabDII and ABab.I mice. As expected, the pre-selection (out-of-frame) TCR repertoires were roughly identical between ABabDII and ABab.I mice, since the human TCR loci and recombinases were the same. The (in-frame) repertoire after positive selection reflects, on the one hand, the frequency with which individual V and J segments are rearranged and, on the other hand, the frequency with which some V β We found that gene segments, such as TRBV4-1, are over-represented in ABab.I mice but under-represented in ABabDII mice. Similarly, TRBV12-3 / 12-4 are over-represented in ABabDII mice but not in ABab.I mice.

[0214] We found a clear relationship between multiple HLA expression and their impact on repertoire diversity. On the TCR side, some V β It is speculated that genes may have a higher inherent affinity for some HLA alleles than for others. The key requirement for positive selection is that the CDR1 and CDR2 regions interact with the selected MHC class I molecule. β The genes differ in the CDR1 and CDR2 regions, and MHC class I genes are highly polymorphic. βIt can be hypothesized that the intrinsic affinity for gene segments is higher for some MHC I alleles and lower for others, and that the MHC and TCR loci have co-evolved to ensure more efficient positive selection. This hypothesis was supported by Chen et al., who showed that human CD4 in mice selected against a single mouse or a single human MHC class II molecule. + They sequenced the TCR repertoire and found that the species-compatible human MHC class II was mediated by the species-incompatible mouse MHC molecule IA. b Compared with + We found that TCR repertoires were selective for TCR expression, which was explained by the high inherent affinity of human TCRs for human MHC class II, on average.

[0215] Similarly, some V β The gene segment has a higher specific affinity for HLA-A2, whereas other V β It may be hypothesized that the gene segment has a lower intrinsic affinity for HLA-A2, but as a consequence a higher intrinsic affinity for any of the other six MHC class I alleles in ABab.I mice. This may mean that certain TCRs are less likely to be selected by HLA-A2, but are more efficiently selected by other MHC I alleles in ABab.I mice.

[0216] Chen et al. also observed that TCRs selected against mouse MHC II have, on average, shorter CDR3 sequences compared to human TCRs selected against human MHC II molecules. β The different specific affinities of gene segments are modulated by CDR3 length, i.e., certain Vs with relatively low specific affinity for MHC βThe finding is that gene segments are more likely to be selected if they have shorter CDR3s. TCR repertoires selected against multiple class I alleles contain, on average, slightly larger CDR3 regions compared to TCR repertoires selected against only a single MHC class I allele, in this case HLA-A2. β The segment has increased specific affinity for any of the six new MHC class I alleles in ABab.I mice, allowing the selection of TCRs with slightly larger CDR3 regions. This may explain the diversity of ABab.I mice compared with ABabDII mice, but also result in more specific TCRs. This hypothesis is based on studies in mice with a defect in the expression of terminal deoxynucleotidyl transferase (TdT), resulting in TCRs with shorter CDR3s. These TCRs tended to be more promiscuous and cross-reactive. Despite the relative diversity of the TCR repertoire in ABabDII, it is speculated that the TCR repertoire in ABabDII mice is suboptimal due to the relatively short CDR3 region. 21 On the other hand, thymocytes of ABab.I mice can choose one of six additional class I alleles to be selected. It is therefore hypothesized that different MHC alleles in the thymus compete with each other to select a given T cell clone. If this is correct, a T cell clone that cannot be selected by HLA-A2 could be selected by any of the other six MHC class I alleles with their increased specific affinity, and could also accommodate the longer CD3 by not interfering with positive selection.

[0217] However, it should be noted that the hypothesis that T cells from ABab.I mice find HLA with optimal unique affinity at the expense of reduced selection against HLA-A2 remains speculative at present: it is difficult to prove this hypothesis, since it is not known whether different MHC class I alleles can select for a similarly diverse TCR repertoire.

[0218] Therefore, we are in the process of generating and characterizing mice that carry a single HLA allele from those contained in the ABab.I mice. By sequencing the TCR repertoire of mice expressing a single HLA allele in the same background, we can identify specific V β It can be examined whether segments have different specific affinities for different HLA alleles. 21、248 For this reason, single ABab-A*03, ABab-A*11, ABab-B*07, ABab-B*15, ABab-C*04, and ABab-C*07 founder lines were generated and require characterization prior to deep sequencing (proprietary data not shown). β It can prove or disprove whether gene segments are preferentially selected or preferentially ignored by individual MHC I alleles.

[0219] However, mice generally select TCRs with shorter CD3 than humans. This may be due to higher TdT and exonuclease activity or longer time window of expression in humans compared to mice. The TdT-exonuclease machinery with recombination activating genes (RAGs) has evolved to amplify (make more edits) the addition or deletion of nucleotides. This increased activity creates a greater number of unique TCRs with longer CDR3s that affect the entire repertoire.

[0220] ABab.I mice generated 4 times more unique clones than ABabDII mice and therefore also shared more TCRβ clonotypes with the repertoire of human donors (approximately 1000 compared to approximately 500). The number of clones shared between ABab.I mice is surprisingly high, since they were not HLA-matched. The high clonal overlap is likely explained by nonrandom V and J usage. In humans, the theoretical number of TCRβ sequences is 5 × 10 11If VJ rearrangements were random, one would expect fewer than five TCRβ clones to be shared between any two individuals; however, in practice, approximately 10,000 shared clones have been detected, exceeding the theoretical 5 × 10 11 This is consistent with the detection of only 0.1% of TCRα / TCRβ genes. The data in ABab.I mice support the data in humans and suggest that the diversity of TCRα / TCRβ combinations is greater than previously predicted. We also deep sequenced the TCRα repertoires of these two mouse groups. Preliminary data analysis showed that the TCRα repertoire was also more diverse in ABab.I mice compared to ABabDII mice (own data not shown and not published). Further analysis is still required to clarify the combinatorial diversity between these mice.

[0221] Taken together, by expanding the humanization of human TCR loci mouse models, a more human-like and more diverse repertoire may be obtained.

[0222] Functional characterization of ABab.I mice ex vivo and in vivo We showed that ex vivo APCs from ABab.I mice could efficiently present model epitopes in all six HLA alleles and induce effector T cells to produce IFN-γ. In the absence of HLA-specific antibodies, we devised this in vitro co-culture assay to estimate both HLA expression and its functionality by activating model epitope-restricted TCR-modified T cells. APCs from blood, spleen, and lymph node cells were used as effector targets to recapitulate the physiological circulating immune system for immunization.

[0223] The purpose of generating ABab.I mice was to isolate T cell receptors for treating cancer patients that target tumor-restricted antigens. Herein, we demonstrate ex vivo as well as in vivo by immunizing ABab.I mice with a wide range of model antigens, each of which is known to bind to one of the six HLA alleles present in the mice. These epitopes were selected to correspond to various antigen categories, such as frameshift-derived mutant (mCALR for A*03), point mutant (KRAS-G12V for A*11), viral (CMV-pp65 for B*07 and C*04, HPV-E5 for B*15), and tumor-associated (MAGE-A12 for HLA-C*07:02) epitopes.

[0224] Although most epitopes selected for the group were demonstrated to be endogenously processed, mice did not express the predicted IC 50 Regardless of the nM value of mAb, mutant KRAS and viral epitopes responded more strongly than other epitopes. Although there is no concern for processed epitopes, for epitopes that have not been demonstrated to be processed, the ABab.I mouse model can be leveraged as a platform to discover new epitopes by immunization with DNA or adenovirus (encoding the antigen) expressing the complete protein sequence.

[0225] In conclusion, the ABab.I mouse model expresses six HLA alleles similar to the natural human situation and may represent a unique new tool for isolating novel T cell receptors for clinical use in ATT of cancer.

[0226] Understanding the biology behind fully human HLA haplotypes in mouse models The ABab.I model, with its diverse TCRβ repertoire, has proven to be a useful tool for isolating unique T cell receptors for therapy. However, understanding the repertoire of the TCRα locus may determine the overall combinatorial diversity of TCR combinations that ABab.I mice can generate. This would allow the TCR diversity of ABab.I mice to be closer to the theoretically possible TCR diversity combinations (approximately 10 15 It is also possible to study how similar the genotypes are to the genotypes (clonal types).

[0227] V β The different intrinsic affinity of ABab.I mice for some human HLA alleles but not others may be a possible next step to understand the skewness in positive selection in the thymus. Thus, by comparing the deep sequencing data of ABab.I mice with single ABab animals in humanized TCR backgrounds, ABab-A*03 mice, ABab-A*11 mice, ABab-B*07 mice, ABab-B*15 mice, ABab-C*04 mice, and ABab-C*07 mice, we can clearly understand whether there is a bias in TCR affinity for HLA and whether different HLA alleles can select similar or dissimilar TCR repertoires.

[0228] Knocking out HLA-A2 should be the next event in single ABab and ABab.I mice to generate a pure line lacking the A2 gene, an ideal representation of the normal HLA class I locus in humans.

[0229] Epitope detection of non-HLA-A*02:01 restricted antigens using ABab.I mice Paradoxical immunology based on epitope prediction is a challenge. Thus, ABab.I mice provide an opportunity to identify immunogenic epitopes and simultaneously isolate T cell receptors. Direct immunology by immunizing ABab.I mice with DNA, adenovirus, or mRNA encoding complete proteins for epitope detection could result in TCRs restricted to such newly detected epitopes for ATT.

[0230] ABab.I mice as a new tool to isolate human TCRs for ATT in cancer In predictive screening, we noted the number of individuals per year carrying the mutation. In principle, immunization of ABab.I mice with such mutant epitopes would result in a range of TCRs from high to optimal affinity that could be advanced to clinical applications, but information on endogenous affinity is needed.

[0231] Thus, the ABab.I mouse is not only suitable for understanding the human repertoire formed in the natural class I HLA environment, but also represents a new model for isolating T cell receptors against a wide variety of antigens for adoptive T cell therapy.

[0232] Example 2 Unbiased detection of the novel MAGE-A1 epitope GTLEEVPTA (SEQ ID NO: 200) In this example, new MAGE-A1 epitopes were screened in ABab.I mice.

[0233] First, the immune response to HLA-A*02-restricted MAGE-A1 in ABab.I mice was studied after vaccination (Figure 21A). ABab.I mice were immunized with full-length MAGE-A1 mutated at position 279 (V→D) to disrupt the dominant HLA-A*02-restricted epitope KVLEYVIKV(278-286) (SEQ ID NO: 199). On day 7 after the third immunization, peripheral blood was stimulated with MAGE-A1 expressing cells and intracellular staining for IFNg was performed (Figure 21A).

[0234] We then localized the epitope on MAGE-A1 bound by one of the TCRs, designated TCR1, isolated from each ABab.I mouse. For this purpose (location of the epitope), human T cells transduced with TCR1 were co-cultured with K562-A2 cells expressing MAGE-A1, fragments of MAGE-A1 (F1-F3), or loaded with KVLEYVIKV (KVL). F1-F3 span the following fragments of MAGE-A1: F1 - 97, F2 - 98-222, and F3 - 223-309. We found that TCR1 recognizes an HLA-A*02-restricted MAGE-A1 epitope located in the N-terminal fragment formed by amino acid residues 1-97 of MAGE-A1 (Figure 21B).

[0235] Finally, to determine the exact epitope, TCR1-transduced human T cells were co-cultured with K562-A2 cells loaded with the predicted nonamer conjugates (Table 18) or KVLEYVIKV (KVL). MAGE-A1-reactive TCR1 was found to recognize the nonamer GTLEEVPTA (SEQ ID NO: 200) (Figure 21C). It is noted here that although the nonamer GTLEEVPTA was predicted as a MAGE-A1 epitope in US2003 / 0148973, US2003 / 0148973 was unable to generate HLA-A2-restricted TCRs against this epitope, and only TCRs against the four epitopes shown in Table IV of US2003 / 0148973 could be generated. Notably, in US Patent Application No. 2003 / 0148973, genetically modified mice expressing human HLA-A2 MHC molecules were immunized with mature MAGE-A1 protein, meaning that essentially the same technique of immunization with MAGE-A1 used herein is also used in US Patent Application No. 2003 / 0148973. This means that the non-human mammal of the invention, such as the mouse of the invention, will generate CD8 T cells against MHC I-presented peptides / antigens for which it was unable to generate TCRs. +This clearly demonstrates that it is possible to generate TCRs.

[0236] Table 18 shows the predicted nonamer binders. [Table 18]

[0237] In summary, a new MAGE-A1 epitope has been identified in an unbiased manner, which is distinct from the known dominant HLA-A*02 restricted epitope KVL278-286. This further demonstrates that ABab.I mice allow for the identification of new epitopes and shows that the ABab.I mice of the present invention can generate TCRs against epitopes not generated by other means (e.g., US Patent Application Publication No. 2003 / 0148973).

[0238] Example 3 Immunization experiments (in vivo immune response analysis) CD8 in vivo + To analyze killer T cell responses, multi-HLA mice were immunized with the model epitopes described as peptides (as shown in the table, nonamers and decamers). All six model epitopes were predicted in silico (using software) to be strong binders to their respective HLA alleles.

[0239] Table 19 shows model peptides presented by HLA alleles present in multi-HLA ABab.I mice and used for immunization. [Table 19]

[0240] In the procedure, young mice were immunized with peptides (minimum 2 times) with at least 4 weeks between injections. Seven days after the booster injection, PBMCs from multi-HLA mice were stimulated in vitro with either the same peptide used for immunization (right), or a non-specific peptide (left), or CD3 / CD28 beads (center, positive control). Intracellular flow cytometry staining revealed that CD8 + The amount of IFN-γ cytokine released by T cells was measured. CD3, CD8, and intracellular IFN-γ expression were analyzed by flow cytometry. + T cells were measured in vitro. The dot plots in the figure show CD8 + IFN-γ + Indicates T cells (CD3 + Gating on lymphocytes). As a non-specific peptide, the NY-ESO peptide (APRGPHGGAASGL) was used only for in vitro stimulation (as a negative control).

[0241] Flow cytometry plots of one of the multiple responders are shown. Of 6-7 mice / epitopes, 6 mice responded to KRAS (HLA-A*11:01 restricted) peptide, 3 mice responded to MAGE-A1 (HLA-A*02:01 restricted, as described in Example 2 above) and HPV (HLA-B*07:02 and / or HLA-B*15:01 restricted) peptides, and 1 mouse responded to mCALR (HLA-A*03:01 restricted), CMV (HLA-C*04:01 restricted), and MAGE-A12 (HLA-C*07:02 restricted) peptides. Immune responses are analyzed in the blood (PBMCs) of immunized mice by flow cytometry. To this end, intracellular IFN-γ cytokine proteins are stained using antibodies along with CD3 and CD8 antibodies. In vitro restimulation of PBMCs in blood (7 days after booster injection) with each peptide used in the initial immunization demonstrated responses to the specific peptides but not to the non-specific peptides. Beads were used as a positive control to demonstrate T cell competence in the blood of multi-HLA mice.

[0242] Example 4 Analysis of T cell responses of HuTCR mice to individual HLA class I alleles In this example, T cell responses in ABab.I mice to individual HLA class I alleles integrated into their genome were analyzed.

[0243] For this purpose, 11 ABab.I mice were immunized multiple times with DNA encoding full-length MAGE-A1 mutated at position 279 (V→D) to disrupt the dominant HLA-A*02-restricted epitope KVLEYVIKV(278-286) (SEQ ID NO: 199). In this regard, see Example 2. Blood T cells from the mice were then co-cultured with cell lines expressing different single or dual HLA and MAGE-A1. CD8 + Interferon-γ-secreting cells (IFNg +) were measured by intracellular staining. By doing so, the T cell responses of HLA-A*02:01, HLA-A*03:01, HLA-A*11:01 were analyzed individually, whereas the T cell responses of the alleles HLA-B*07:02 and HLA-B*15:01 and the alleles HLA-C*04:01 and HLA-C*07:02 were analyzed in combination.

[0244] The results of this experiment are shown in Figure 25. Mice with positive T cell responses are indicated by circles located above the background dashed line, and mice with negative T cell responses are indicated by circles located below the background dashed line. As can be seen from Figure 25, ABab.I mice were found to have active T cell responses against MAGE-A1 derived epitopes in the blood for all HLA class I alleles tested individually (HLA-A*02:01, HLA-A*03:01, HLA-A*11:01) and also for the HLA class I alleles tested jointly (HLA-B*07:02 and HLA-B*15:01, and HLA-C*04:01 and HLA-C*07:02, respectively).

[0245] Thus, and in this experiment, the ABab.I mice of the present invention are CD8+ to all human HLA class I alleles that they carry. + It has been confirmed that the antibody can induce a T cell response. In this context, it is also noted that the amount of activated T cells in the blood is usually less than the amount of activated T cells found in the spleen of mice. Thus, based on the results of blood T cell counts, the amount of activated CD8 + It may be hypothesized that the number of T cells, and therefore the active T cell response, is higher.

[0246] Furthermore, in the results shown in Figure 25, transgenic animals carrying multiple human HLA class I alleles allow direct comparison of the strength of T cell responses associated with each of the human alleles. +The T cell responses were assessed by quantity (7 of 11 immunized mice showed a positive T cell response) and biological activity (IFNg of T cells). + The transgenic animals of the invention thus simultaneously assess T cell responses associated with multiple human HLA alleles, allowing for a direct comparison of the strength of T cell responses. This "multiplexity" is a further advantage of the invention.

[0247] Those skilled in the art will readily appreciate that the present invention is well adapted to carry out the objects and obtain the ends and advantages set forth above, as well as the essentials thereof. Moreover, it will be readily apparent to those skilled in the art that various substitutions and modifications can be made to the invention disclosed herein without departing from the scope and spirit of the invention. The compositions, methods, procedures, procedures, molecules, and specific compounds described herein represent certain embodiments and are exemplary and are not intended to limit the scope of the invention. Modifications herein and other uses that fall within the spirit of the invention as defined by the claims will occur to those skilled in the art. The listing or description of a document previously listed in this specification should not necessarily be taken as an admission that the document is part of the state of the art or is common general knowledge.

[0248] The invention described herein for illustrative purposes can be suitably practiced without any element or limitation not specifically disclosed herein. Thus, for example, terms such as "include", "includes", "comprise", and the like are to be interpreted broadly and without limitation. Furthermore, the terms and expressions used herein are used as descriptive terms and not as limiting terms, and there is no intention in the use of such terms and expressions to exclude any equivalents of the features shown and described, or portions thereof, but it is recognized that various modifications are possible within the scope of the invention described. Thus, although the present invention has been specifically disclosed by example embodiments and optional features, it should be understood that modifications and variations of the invention embodied herein may be adopted by those skilled in the art, and such modifications and variations are considered to be within the scope of the present invention.

[0249] The present invention is described broadly and generically herein. Also, each of the narrower species and subgroups falling within the generic disclosure are part of the present invention. This includes the generic description of the invention with a predicate or negative limitation excluding any limitation from the genus, regardless of whether the implementation is specifically described herein.

[0250] Other embodiments are within the scope of the following claims. Furthermore, those of skill in the art will recognize that when features or aspects of the invention are described in terms of a Markush group, the invention is also thereby described in terms of any individual member or subgroup member of the Markush group.

Claims

1. A non-human mammal comprising in its genome at least two (e.g., at least three, at least four, at least five, at least six, or at least seven) human leukocyte antigen (HLA) class I alleles, wherein the at least two human HLA alleles are functionally expressed to express corresponding MHC I polypeptides on the surface of cells of the mammal to present MHC antigens by the non-human mammal that provide an antigen-specific CD8+ T cell response, wherein the at least two human HLA class I alleles are (a) at least one human HLA-A allele, and (b) at least one human HLA-B allele, and / or (c) at least one human HLA-C allele, and / or (d) (a) and (b), and / or (e) (a) and (c), and / or (f) (b) and (c), and / or (g) A non-human mammal comprising (a), (b), and (c).

2. The non-human mammal of claim 1 , wherein the mammal further comprises at least one human HLA class II allele in its genome.

3. The non-human mammal of claim 1 , wherein the mammal comprises at least four, five, six, or seven human HLA class I alleles in its genome.

4. The mammal is (a) at least two different human HLA-A alleles, preferably at least three different human HLA-A alleles, and / or (b) at least two different human HLA-B alleles, and / or (c) the non-human mammal of claim 1, which comprises at least two different human HLA-C alleles in its genome;

5. The mammal is (a) HLA-A*03, (b) HLA-A*11, (c) HLA-B*07, (d) HLA-B*15, (e) HLA-C*04, (f) HLA-C*07, and (g) The non-human mammal according to claim 1, comprising in its genome six or seven human leukocyte antigen (HLA) class I alleles selected from a group of human HLA alleles comprising or consisting of HLA-A*02.

6. The seven human HLA alleles are: (a) HLA-A*03:01, (b) HLA-A*11:01, (c) HLA-B*07:02, (d) HLA-B*15:01, (e) HLA-C*04:01, (f) HLA-C*07:02, and (g) The non-human mammal of claim 1, comprising or consisting of HLA-A*02:

01.

7. The non-human mammal of claim 1, wherein the at least two HLA alleles encode a chimeric human / non-human mammal MHC I polypeptide, wherein the human portion of the chimeric polypeptide comprises the α1 domain and the α2 domain of the human MHC I polypeptide, and the non-human mammal portion of the chimeric polypeptide comprises the α3 domain, the transmembrane domain, and the cytoplasmic domain of an endogenous non-human mammal MHC I polypeptide, and the non-human mammal expresses the chimeric human / non-human mammal MHC I polypeptide.

8. 2. The non-human mammal of claim 1, wherein the mammal comprises a fully human T cell receptor (TCR) locus in its genome, and optionally, the mammal is deficient in its endogenous T cell receptor (TCR) locus.

9. 2. The non-human mammal of claim 1, wherein the non-human mammal is selected from the group consisting of rodents, dogs, felines, primates, rabbits, pigs, and ruminants, and optionally, the rodent is a mouse or a rat.

10. 1. A nucleic acid construct comprising nucleic acids encoding at least two (e.g., at least three, at least four, at least five, at least six, or at least seven, etc.) human HLA class I alleles, wherein the nucleic acid construct is capable of functionally expressing the at least two human HLA alleles (e.g., in a non-human mammal, e.g., as described herein) such that corresponding MHC I polypeptides are expressed on the surface of cells of the non-human mammal comprising the nucleic acid construct to present MHC antigens that provide an antigen-specific CD8+ T cell response by the non-human mammal, wherein the at least two human HLA class I alleles are (a) at least one (e.g., at least two or at least three) human HLA-A alleles, or (b) at least one (e.g., at least two) human HLA-B alleles, and / or (c) at least one (e.g., at least two) human HLA-C alleles, and / or (d) (a) and (b), and / or (e) (a) and (c), and / or (f) (b) and (c), and / or (g) A nucleic acid construct comprising (a), (b), and (c).

11. The nucleic acid construct of claim 10 , wherein the nucleic acid further encodes at least one human HLA class II allele.

12. An expression vector comprising a nucleic acid construct comprising nucleic acids encoding at least two (e.g., at least three, at least four, at least five, at least six, or at least seven, etc.) human HLA class I alleles, The nucleic acid construct is capable of functionally expressing the at least two human HLA alleles (e.g., in a non-human mammal, e.g., as described herein) such that corresponding MHC I polypeptides are expressed on the surface of cells of the non-human mammal comprising the nucleic acid construct to present MHC antigens that provide an antigen-specific CD8+ T cell response by the non-human mammal, and the at least two human HLA class I alleles are (a) at least one (e.g., at least two or at least three) human HLA-A alleles, or (b) at least one (e.g., at least two) human HLA-B alleles, and / or (c) at least one (e.g., at least two) human HLA-C alleles, and / or (d) (a) and (b), and / or (e) (a) and (c), and / or (f) (b) and (c), and / or (g) An expression vector comprising (a), (b), and (c).

13. A host cell (e.g., a recombinant and / or isolated and / or non-human host cell) comprising a nucleic acid construct and / or an expression vector, The nucleic acid construct comprises nucleic acids encoding at least two (e.g., at least three, at least four, at least five, at least six, or at least seven, etc.) human HLA class I alleles, and is capable of functionally expressing the at least two human HLA alleles (e.g., in a non-human mammal, e.g., as described herein) such that corresponding MHC I polypeptides are expressed on the surface of cells of the non-human mammal comprising the nucleic acid construct to present MHC antigens that provide an antigen-specific CD8+ T cell response by the non-human mammal, and the at least two human HLA class I alleles are (a) at least one (e.g., at least two or at least three) human HLA-A alleles, or (b) at least one (e.g., at least two) human HLA-B alleles, and / or (c) at least one (e.g., at least two) human HLA-C alleles, and / or (d) (a) and (b), and / or (e) (a) and (c), and / or (f) (b) and (c), and / or (g) comprising (a), (b), and (c); The expression vector comprises the nucleic acid construct in a host cell (e.g., a recombinant and / or isolated and / or non-human host cell).

14. 1. A method for modifying endogenous HLA alleles in a non-human mammal, comprising transducing and / or transplanting a nucleic acid construct and / or an expression vector into said non-human mammal; The nucleic acid construct comprises nucleic acids encoding at least two (e.g., at least three, at least four, at least five, at least six, or at least seven, etc.) human HLA class I alleles, and is capable of functionally expressing the at least two human HLA alleles (e.g., in a non-human mammal, e.g., as described herein) such that corresponding MHC I polypeptides are expressed on the surface of cells of the non-human mammal comprising the nucleic acid construct to present MHC antigens that provide an antigen-specific CD8+ T cell response by the non-human mammal, and the at least two human HLA class I alleles are (a) at least one (e.g., at least two or at least three) human HLA-A alleles, or (b) at least one (e.g., at least two) human HLA-B alleles, and / or (c) at least one (e.g., at least two) human HLA-C alleles, and / or (d) (a) and (b), and / or (e) (a) and (c), and / or (f) (b) and (c), and / or (g) comprising (a), (b), and (c); The method, wherein the expression vector comprises the nucleic acid construct.

15. 1. A method of producing a non-human mammalian oocyte having a modified target sequence in its genome (e.g., by means of transposon-mediated targeting, e.g., as described in the Experimental Section herein), comprising the steps of introducing into the non-human mammalian oocyte a nucleic acid construct (e.g., an ITR-flanked targeting vector (e.g., an HLA-bearing cassette)) and / or an expression vector; The nucleic acid construct comprises nucleic acids encoding at least two (e.g., at least three, at least four, at least five, at least six, or at least seven, etc.) human HLA class I alleles, and is capable of functionally expressing the at least two human HLA alleles (e.g., in a non-human mammal, e.g., as described herein) such that corresponding MHC I polypeptides are expressed on the surface of cells of the non-human mammal comprising the nucleic acid construct to present MHC antigens that provide an antigen-specific CD8+ T cell response by the non-human mammal, and the at least two human HLA class I alleles are (a) at least one (e.g., at least two or at least three) human HLA-A alleles, or (b) at least one (e.g., at least two) human HLA-B alleles, and / or (c) at least one (e.g., at least two) human HLA-C alleles, and / or (d) (a) and (b), and / or (e) (a) and (c), and / or (f) (b) and (c), and / or (g) comprising (a), (b), and (c); The method, wherein the expression vector comprises the nucleic acid construct.

16. 1. A method for producing a non-human mammal having a modified target sequence in its genome, comprising: (a) producing an oocyte; and (b) analyzing offspring delivered by a non-human female host into which the oocyte obtained in (a) has been transferred for the presence of said modification; producing said oocyte by a method for producing a non-human mammalian oocyte having a modified target sequence in its genome (e.g., by means of transposon-mediated targeting, e.g., as described in the Experimental Section herein), the method comprising the step of introducing into the non-human mammalian oocyte a nucleic acid construct (e.g., an ITR-flanked targeting vector (e.g., an HLA-bearing cassette)) and / or an expression vector; The nucleic acid construct comprises nucleic acids encoding at least two (e.g., at least three, at least four, at least five, at least six, or at least seven, etc.) human HLA class I alleles, and is capable of functionally expressing the at least two human HLA alleles (e.g., in a non-human mammal, e.g., as described herein) such that corresponding MHC I polypeptides are expressed on the surface of cells of the non-human mammal comprising the nucleic acid construct to present MHC antigens that provide an antigen-specific CD8+ T cell response by the non-human mammal, and the at least two human HLA class I alleles are (a') at least one (e.g., at least two or at least three) human HLA-A alleles, or (b') at least one (e.g., at least two) human HLA-B alleles, and / or (c') at least one (e.g., at least two) human HLA-C alleles, and / or (d') (a') and (b'), and / or (e'), (a') and (c'), and / or (f'), (b') and (c'), and / or (g') comprising (a'), (b'), and (c'); The method, wherein the expression vector comprises the nucleic acid construct.

17. A non-human mammal produced and / or modified by a method for producing a non-human mammal oocyte having a modified target sequence in its genome, comprising: producing said oocyte by a method for producing a non-human mammalian oocyte having a modified target sequence in its genome (e.g., by means of transposon-mediated targeting, e.g., as described in the Experimental Section herein), the method comprising the step of introducing into the non-human mammalian oocyte a nucleic acid construct (e.g., an ITR-flanked targeting vector (e.g., an HLA-bearing cassette)) and / or an expression vector; The nucleic acid construct comprises nucleic acids encoding at least two (e.g., at least three, at least four, at least five, at least six, or at least seven, etc.) human HLA class I alleles, and is capable of functionally expressing the at least two human HLA alleles (e.g., in a non-human mammal, e.g., as described herein) such that corresponding MHC I polypeptides are expressed on the surface of cells of the non-human mammal comprising the nucleic acid construct to present MHC antigens that provide an antigen-specific CD8+ T cell response by the non-human mammal, and the at least two human HLA class I alleles are (a') at least one (e.g., at least two or at least three) human HLA-A alleles, or (b') at least one (e.g., at least two) human HLA-B alleles, and / or (c') at least one (e.g., at least two) human HLA-C alleles, and / or (d') (a') and (b'), and / or (e'), (a') and (c'), and / or (f'), (b') and (c'), and / or (g') comprising (a'), (b'), and (c'); The expression vector comprises the nucleic acid construct.

18. A method for producing one or more T cell receptors capable of binding to an antigen of interest, comprising administering the antigen of interest to a non-human animal described in claim 1.

19. A T cell receptor obtained or obtainable by a method of producing one or more T cell receptors capable of binding to an antigen of interest, comprising: The method comprises administering an antigen of interest to the non-human animal of claim 1.

20. A method for identifying an epitope capable of inducing an immune response, comprising administering to a non-human animal described in claim 1 an antigen of interest suspected of containing an epitope capable of inducing an immune response.

Images