Nucleic acid molecule encoding car and vector containing same, car, immune cell containing car and pharmaceutical composition containing said cell, and method for producing immune cell containing car

Abstract

The present invention relates to a nucleic acid molecule having a nucleotide sequence that encodes a chimeric antigen receptor (CAR). The present invention relates to a vector containing said nucleic acid molecule. The present invention relates to a CAR. The present invention relates to an immune cell containing a CAR. The present invention relates to a pharmaceutical composition containing an immune cell containing a CAR. The present invention relates to a method that produces an immune cell containing a CAR.

Description

Nucleic acid molecule encoding CAR and vector containing same, CAR, immune cell containing CAR and pharmaceutical composition containing the cell, and method for producing immune cell containing CAR

[0001] The present invention relates to a nucleic acid molecule having a nucleotide sequence encoding a chimeric antigen receptor (hereinafter also referred to as "CAR"). The present invention relates to a vector comprising the nucleic acid molecule. The present invention relates to a CAR. The present invention relates to an immune cell comprising a CAR. The present invention relates to a pharmaceutical composition for treating malignant tumors comprising an immune cell comprising a CAR. The present invention relates to a method for producing an immune cell comprising a CAR.

[0002] CARs are receptor proteins engineered by genetically fusing an extracellular domain containing an antigen-binding region, a transmembrane domain, and an intracellular domain that transmits immune cell activation signals. For example, as described in Patent Document 1, the antigen-binding region of CARs uses a single-chain antibody that recognizes antigens expressed on tumor cells. In recent years, cancer immunotherapy has attracted attention. This involves introducing a gene encoding a CAR into immune cells, expressing the CAR on their surface, and then transplanting the immune cells into patients to treat cancer. Upon recognizing antigens on tumor cells in vivo, CAR-expressing immune cells become activated and express cytotoxic molecules such as Fas ligand, perforin, and granzymes, as well as cytokines, thereby exerting antitumor effects. However, cytokine production by CAR-expressing immune cells can sometimes cause cytokine release syndrome (CRS).

[0003] U.S. Patent No. 7,741,465

[0004] An object of the present invention is to provide a CAR that, when expressed in immune cells, enables the suppression of cytokine production by the immune cells. Another object of the present invention is to provide a nucleic acid molecule encoding such a CAR and a vector comprising the nucleic acid molecule. A further object of the present invention is to provide immune cells comprising such a CAR, a pharmaceutical composition comprising the immune cells, and a method for producing the immune cells.

[0005] The present inventors discovered that cytokine production by immune cells containing CAR can be suppressed by substituting basic amino acid residues for specific amino acid residues in framework region 3 (FR3) of the light chain variable region of the antigen-binding domain of CAR, and thus completed the present invention. Accordingly, the following inventions [1] to

[21] are provided.

[0006] [1] A nucleic acid molecule having a nucleotide sequence encoding a CAR, wherein the nucleic acid molecule comprises a segment encoding an extracellular domain, a segment encoding a transmembrane domain, and a segment encoding an intracellular domain, wherein the segment encoding the extracellular domain comprises a nucleotide sequence encoding an antigen-binding region comprising a light chain variable region and a heavy chain variable region, and wherein at least three codons in the nucleotide sequence encoding FR3 of the light chain variable region as defined by the Kabat method are codons encoding basic amino acid residues.

[0007] [2] The nucleic acid molecule according to [1] above, wherein the at least three codons in the light chain variable region include at least three selected from the group consisting of a codon encoding the 60th amino acid residue, a codon encoding the 63rd amino acid residue, a codon encoding the 65th amino acid residue, a codon encoding the 67th amino acid residue, a codon encoding the 70th amino acid residue, a codon encoding the 72nd amino acid residue, a codon encoding the 74th amino acid residue, a codon encoding the 76th amino acid residue, a codon encoding the 77th amino acid residue, a codon encoding the 79th amino acid residue, and a codon encoding the 81st amino acid residue.

[0008] [3] The nucleic acid molecule according to [1] or [2] above, wherein three to five codons selected from the group consisting of a codon encoding the 60th amino acid residue, a codon encoding the 63rd amino acid residue, a codon encoding the 65th amino acid residue, a codon encoding the 67th amino acid residue, a codon encoding the 70th amino acid residue, a codon encoding the 72nd amino acid residue, a codon encoding the 74th amino acid residue, a codon encoding the 76th amino acid residue, a codon encoding the 77th amino acid residue, a codon encoding the 79th amino acid residue, and a codon encoding the 81st amino acid residue, as defined by the Kabat method, are codons encoding basic amino acid residues.

[0009] [4] The nucleic acid molecule according to any one of [1] to [3] above, wherein the antigen-binding region comprises a single-chain antibody, and the single-chain antibody is a single-chain antibody that binds to a complex of a MAGE-A4-derived peptide and HLA-A2, a complex of a PRAME-derived peptide and HLA-A24, CD19, BCMA, or CEA.

[0010] [5] The nucleic acid molecule according to any one of [1] to [4] above, wherein the transmembrane domain comprises a transmembrane region of any one protein selected from the group consisting of the T cell receptor α chain, the T cell receptor β chain, CD3ε, CD3ζ, CD4, CD5, CD8, CD9, CD16, CD22, CD28, CD33, CD37, CD45, CD64, CD80, CD86, CD134, CD154, 4-1BB, ICOS, and GITR.

[0011] [6] The nucleic acid molecule according to any one of [1] to [5] above, wherein the intracellular domain comprises a signaling domain of at least one protein selected from the group consisting of CD3ζ, CD3γ, CD3δ, CD3ε, CD5, CD22, CD66d, CD79a, CD79b, FcRγ, and FcRβ.

[0012] [7] The nucleic acid molecule according to [6] above, wherein the segment encoding the intracellular domain further comprises a nucleotide sequence encoding a costimulatory domain, and the costimulatory domain is a costimulatory domain of at least one protein selected from the group consisting of 4-1BB, CD28, GITR, CD2, CD5, CD8, CD9, CD16, CD22, CD33, CD37, CD64, CD80, CD86, CD134, CD154, and ICOS.

[0013] [8] The nucleic acid molecule according to any one of [1] to [7] above, further comprising a segment encoding a hinge domain between the nucleotide sequence encoding the antigen-binding region and the segment encoding the transmembrane domain.

[0014] [9] The nucleic acid molecule according to any one of [1] to [8] above, wherein the nucleic acid molecule is DNA or RNA.

[0015]

[10] A vector comprising the nucleic acid molecule according to any one of [1] to [9] above.

[0016]

[11] A CAR comprising an extracellular domain, a transmembrane domain, and an intracellular domain, wherein the extracellular domain comprises an antigen-binding region comprising a light chain variable region and a heavy chain variable region, and wherein at least three amino acid residues in FR3 of the light chain variable region as defined by the Kabat method are basic amino acid residues.

[0017]

[12] The CAR described in

[11] above, wherein the at least three amino acid residues include at least three selected from the group consisting of the 60th amino acid residue, the 63rd amino acid residue, the 65th amino acid residue, the 67th amino acid residue, the 70th amino acid residue, the 72nd amino acid residue, the 74th amino acid residue, the 76th amino acid residue, the 77th amino acid residue, the 79th amino acid residue, and the 81st amino acid residue in the light chain variable region.

[0018]

[13] A CAR according to

[11] or

[12] above, in which 3 to 5 amino acid residues selected from the group consisting of the 60th, 63rd, 65th, 67th, 70th, 72nd, 74th, 76th, 77th, 79th, and 81st amino acid residues in the light chain variable region as defined by the Kabat method are basic amino acid residues.

[0019]

[14] A CAR according to any one of

[11] to

[13] above, wherein the antigen-binding region comprises a single-chain antibody, and the single-chain antibody is a single-chain antibody that binds to a complex of a MAGE-A4-derived peptide and HLA-A2, a complex of a PRAME-derived peptide and HLA-A24, CD19, BCMA, or CEA.

[0020]

[15] The CAR according to any one of

[11] to

[14] above, wherein the transmembrane domain comprises a transmembrane region of any one protein selected from the group consisting of the T cell receptor α chain, the T cell receptor β chain, CD3ε, CD3ζ, CD4, CD5, CD8, CD9, CD16, CD22, CD28, CD33, CD37, CD45, CD64, CD80, CD86, CD134, CD154, 4-1BB, ICOS, and GITR.

[0021]

[16] The CAR according to any one of

[11] to

[15] above, wherein the intracellular domain comprises a signaling domain of at least one protein selected from the group consisting of CD3ζ, CD3γ, CD3δ, CD3ε, CD5, CD22, CD66d, CD79a, CD79b, FcRγ, and FcRβ.

[0022]

[17] The CAR described in

[16] above, wherein the intracellular domain further comprises a costimulatory domain, and the costimulatory domain is a costimulatory domain of at least one protein selected from the group consisting of 4-1BB, CD28, GITR, CD2, CD5, CD8, CD9, CD16, CD22, CD33, CD37, CD64, CD80, CD86, CD134, CD154, and ICOS.

[0023]

[18] The CAR according to any one of

[11] to

[17] above, further comprising a hinge domain between the antigen-binding region and the transmembrane domain.

[0024]

[19] An immune cell comprising the chimeric antigen receptor according to any one of

[11] to

[18] above.

[0025]

[20] A pharmaceutical composition for treating malignant tumors, comprising the immune cells described in

[19] above.

[0026]

[21] A method for producing immune cells containing a chimeric antigen receptor, comprising introducing the nucleic acid molecule according to any one of [1] to [9] above or the vector according to

[10] above into immune cells and causing the immune cells to express the chimeric antigen receptor.

[0027] According to the present invention, immune cells containing a CAR in which cytokine production is suppressed can be obtained.

[0028]

[0033] Figure 1A is a schematic diagram of an example of a nucleic acid molecule of this embodiment. A magnified view of the segment (VL) encoding the light chain variable region in the nucleic acid molecule is also shown. In the figure, A indicates the segment encoding the extracellular domain, B indicates the segment encoding the transmembrane domain, and C indicates the segment encoding the intracellular domain. D indicates the segment encoding the antigen-binding region contained in the extracellular domain. Among these segments, VH indicates the segment encoding the heavy chain variable region, L indicates the segment encoding the linker, HD indicates the segment encoding the hinge domain, TMD indicates the segment encoding the transmembrane domain, co-STD indicates the segment encoding the costimulatory domain, and SD indicates the segment encoding the signaling domain. In the magnified view of VL, dashed lines marked with * indicate codons encoding basic amino acid residues. In Figure 1A, the segment encoding FR3 has three codons encoding basic amino acid residues, but the present invention is not limited to this. Figure 1B is a schematic diagram of an example of a CAR of this embodiment. In the figure, VH indicates a heavy chain variable region, VL indicates a light chain variable region, HD indicates a hinge domain, TMD indicates a transmembrane domain, co-STD indicates a costimulatory domain, and SD indicates a signal transduction domain. The curve connecting VH and VL indicates a linker. In VL, the dashed line marked with an * indicates a basic amino acid residue. In Figure 1B, there are three basic amino acid residues in FR3, but the present invention is not limited to this. Figure 1B is a schematic representation of nucleic acid molecules encoding CARs in Examples 1 to 3. Figure 1C is a schematic representation of a nucleic acid molecule encoding a CAR in Example 4. Figure 1D is a graph showing the tumor area (average value) of each mouse after administration of PBS or CAR-T cells to mice transplanted with tumor cells.

[0029] 1. Nucleic Acid Molecule The nucleic acid molecule of this embodiment has a nucleotide sequence encoding a CAR. As used herein, "nucleotide sequence" is synonymous with "base sequence" or "nucleic acid sequence." A nucleotide sequence refers to the linear arrangement (order) of nucleotides within a nucleic acid molecule. Thus, a nucleic acid molecule encoding a polypeptide has a nucleotide sequence that encodes the polypeptide. As used herein, the expression "having a nucleotide sequence" means both consisting of the nucleotide sequence and including the nucleotide sequence. As used herein, the term "polypeptide" encompasses protein molecules, partial regions within protein molecules, and fragments of protein molecules. As used herein, a partial region within a nucleic acid molecule may be referred to as a "segment," and a partial region within a protein molecule may be referred to as a "domain."

[0030] As illustrated in FIG. 1A , the nucleic acid molecule of this embodiment comprises, from the 5' end, a segment encoding an extracellular domain, a segment encoding a transmembrane domain, and a segment encoding an intracellular domain. The segment encoding the extracellular domain comprises a nucleotide sequence encoding an antigen-binding region comprising a light chain variable region and a heavy chain variable region. As used herein, an "antigen-binding region comprising a light chain variable region and a heavy chain variable region" refers to a domain comprising at least one light chain variable region and at least one heavy chain variable region, and capable of binding to a predetermined antigen via these. Examples of antigen-binding regions include single-chain antibodies. Single-chain antibodies, also known as scFvs, comprise a light chain variable region and a heavy chain variable region linked via a peptide linker and are a part of the CAR construct of this embodiment. In the nucleic acid molecule of FIG. 1A , the segment encoding the antigen-binding region comprises a nucleotide sequence encoding a single-chain antibody consisting of a VH, L, and VL. Details of each segment of the nucleic acid molecule of this embodiment and the CAR encoded by this nucleic acid molecule are described below.

[0031] As exemplified in FIG. 1A, the nucleic acid molecule of this embodiment is characterized in that at least three codons in the nucleotide sequence encoding the FR3 of the light chain variable region (hereinafter also referred to as "light chain FR3") are codons that encode basic amino acid residues. That is, the CAR encoded by the nucleic acid molecule of this embodiment has an antigen-binding region in which at least three amino acid residues in the light chain FR3 are basic amino acid residues. This antigen-binding region is also referred to as a "modified antigen-binding region" hereinafter. A CAR having a modified antigen-binding region is also referred to as a "modified CAR" hereinafter. The nucleic acid molecule of this embodiment can be said to be a nucleic acid molecule encoding a modified CAR. As used herein, a "codon" refers to three consecutive nucleotides in DNA or RNA.

[0032] A nucleic acid molecule encoding an altered antigen-binding region can be obtained by modifying the codons described below in a nucleic acid molecule encoding the original antigen-binding region. As used herein, the term "original antigen-binding region" refers to an antigen-binding region before modification in which the number of basic amino acid residues in the light chain FR3 is two or less. That is, among the nucleotide sequences encoding the original antigen-binding region, the number of codons encoding basic amino acid residues in the nucleotide sequence encoding the light chain FR3 is two or less. Hereinafter, a CAR having the original antigen-binding region is also referred to as an "unmodified CAR."

[0033] Framework regions (FRs) are regions present in the light chain variable region and heavy chain variable region of an antibody, other than the complementarity-determining regions (CDRs). FRs act as scaffolds connecting the three CDRs and contribute to the structural stability of the CDRs. Therefore, the amino acid sequences of FRs are highly conserved among antibodies of the same species. Each heavy chain and light chain variable region contains three CDRs, CDR1, CDR2, and CDR3, and four FRs, FR1, FR2, FR3, and FR4. These are arranged in the following order from the N-terminus of the variable region: FR1, CDR1, FR2, CDR2, FR3, CDR3, and FR4. Hereinafter, heavy chain CDRs will be referred to as "HCDRs" and light chain CDRs as "LCDRs."

[0034] In the art, methods for numbering amino acid residues in CDRs (hereinafter also referred to as "numbering methods") are known to define the boundaries and lengths of CDRs. When amino acid residues in CDRs are numbered according to the numbering method, the amino acid residues in FRs are also numbered. The numbers assigned to amino acid residues according to the numbering method indicate the position of the amino acid residue in the light chain or heavy chain amino acid sequence. Examples of numbering methods include the Kabat method (Kabat E A. et al., Sequences of Proteins of Immunological Interest., NIH publication No. 91-3242), the Chothia method (Chothia C. and Lesk A M., Canonical Structures for the Hypervariable Regions of Immunoglobulins., J Mol Biol., vol. 196, pp. 901-917, 1987), the IMGT method (Lefranc M P. et al., Developmental and Comparative Immunology 29 (2005) 185-203), the Honergger method (Honegger A. et al., Yet Another Numbering Scheme for Immunoglobulin Variable Domains: An Automatic Modeling and Analysis Tool., J Mol Biol., vol. 309, pp. 657-670, 2001), the ABM method, and the Contact method.

[0035] With respect to the antigen-binding region, the boundaries and lengths of the CDRs and FRs in the light and heavy chain variable regions may be defined using any numbering system. Herein, the boundaries and lengths of the CDRs and FRs are defined according to the Kabat system. For example, when the antigen-binding region comprises or consists of a single-chain antibody, according to the Kabat system, light chain FR1 of the single-chain antibody is defined as the region consisting of amino acid residues 1 to 23 of the light chain variable region. Light chain FR2 of the single-chain antibody is defined as the region consisting of amino acid residues 35 to 49 of the light chain variable region. Light chain FR3 of the single-chain antibody is defined as the region consisting of amino acid residues 57 to 88 of the light chain variable region. Light chain FR4 of the single-chain antibody is defined as the region consisting of amino acid residues 98 to 109 of the light chain variable region. Herein, when the positions of amino acid residues in the light chain variable region of the antigen-binding region are described, the positions of those amino acid residues refer to the positions defined according to the Kabat system, unless otherwise specified.

[0036] The nucleic acid molecule of this embodiment can be obtained by linking, from the 5' end, a nucleic acid molecule encoding the extracellular domain, a nucleic acid molecule encoding the transmembrane domain, and a nucleic acid molecule encoding the intracellular domain. Linking of these nucleic acid molecules can be carried out using known gene recombination techniques and other molecular biology techniques. The nucleic acid molecule encoding the extracellular domain may consist of a nucleic acid molecule encoding a modified antigen-binding region. Preferably, the nucleic acid molecule encoding the extracellular domain is obtained by linking a nucleic acid molecule encoding the modified antigen-binding region with a nucleic acid molecule encoding the hinge domain using the above-mentioned techniques.

[0037] In the nucleotide sequence encoding the light chain FR3 of the original antigen-binding region, changing three or more codons encoding non-basic amino acid residues to codons encoding basic amino acid residues is hereinafter also referred to as "codon modification" or "codon modification." Nucleic acid molecules encoding the modified antigen-binding region can be obtained by modifying the codons in the nucleic acid molecule encoding the original antigen-binding region. Non-basic amino acid residues are neutral amino acid residues and / or acidic amino acid residues, preferably neutral amino acid residues. The nucleotide sequence encoding the modified antigen-binding region is preferably the same as the nucleotide sequence encoding the original antigen-binding region except for the codon-modified sites.

[0038] By such codon modification, at least three codons in the nucleotide sequence encoding the light chain FR3 of the original antigen-binding region are changed to codons encoding basic amino acid residues, thereby making it possible to obtain a nucleic acid molecule encoding a modified antigen-binding region. Codon modification can be performed by substituting or inserting codons in the nucleic acid molecule encoding the original antigen-binding region.

[0039] The basic amino acid residues are lysine, arginine, and histidine residues. Among them, arginine and lysine residues are preferred. The neutral amino acid residues are alanine, asparagine, cysteine, glycine, glutamine, isoleucine, leucine, methionine, phenylalanine, proline, serine, threonine, tryptophan, tyrosine, and valine residues. The acidic amino acid residues are aspartic acid and glutamic acid residues.

[0040] In the nucleotide sequence encoding the light chain FR3, the at least three codons introduced by codon modification may all be codons encoding arginine residues, or all may be codons encoding lysine residues. Alternatively, in the nucleotide sequence encoding the light chain FR3, some of the at least three codons introduced by codon modification may be codons encoding arginine residues, and the remaining codons may be codons encoding lysine residues.

[0041] The nucleic acid molecule of this embodiment may be DNA or RNA. In the nucleotide sequence encoding the light chain FR3, the type of codon introduced by codon modification is not particularly limited as long as it encodes a basic amino acid residue. When the nucleic acid molecule of this embodiment is DNA, examples of codons encoding basic amino acid residues include AGA, AGG, CGA, CGC, CGG, and CGT, which encode arginine residues; AAA and AAG, which encode lysine residues; and CAC and CAT, which encode histidine residues. When the nucleic acid molecule of this embodiment is RNA, examples of codons encoding basic amino acid residues include AGA, AGG, CGA, CGC, CGG, and CGU, which encode arginine residues; AAA and AAG, which encode lysine residues; and CAC and CAU, which encode histidine residues.

[0042] In the nucleic acid molecule of this embodiment, the number of codons encoding basic amino acid residues in the nucleotide sequence encoding the light chain FR3 is, for example, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, or 14. The number of codons encoding basic amino acid residues in the nucleotide sequence encoding the light chain FR3 is preferably 3 to 6, more preferably 3 to 5.

[0043] In the modified light chain FR3 of the antigen-binding region, at least three basic amino acid residues resulting from the codon modification are preferably located at amino acid residue positions excluding Vernier zone residues and non-exposed residues from the amino acid sequence of the light chain FR3. "Vernier zone residues" are amino acid residues in the FR amino acid sequence that contribute to the structural stability of the CDR. "Non-exposed residues" are amino acid residues that are folded internally and not exposed on the surface. Modification of non-exposed residues is expected to have little or no effect. For example, the amino acid residues excluding Vernier zone residues and non-exposed residues from the amino acid sequence of the light chain FR3 are amino acid residues 57, 58, 59, 60, 61, 62, 63, 65, 67, 70, 72, 74, 76, 77, 79, 80, and 81 in the light chain variable region.

[0044] In the nucleotide sequence encoding the altered antigen-binding region, it is preferred that at least three codons selected from the group consisting of the codon encoding the 60th amino acid residue, the codon encoding the 63rd amino acid residue, the codon encoding the 65th amino acid residue, the codon encoding the 67th amino acid residue, the codon encoding the 70th amino acid residue, the codon encoding the 72nd amino acid residue, the codon encoding the 74th amino acid residue, the codon encoding the 76th amino acid residue, the codon encoding the 77th amino acid residue, the codon encoding the 79th amino acid residue, and the codon encoding the 81st amino acid residue in the light chain variable region encode basic amino acid residues. More preferably, three to six codons selected from the above group encode basic amino acid residues. Even more preferably, three to five codons selected from the above group encode basic amino acid residues. For example, each of the codons described in any one of 1) to 21) below encodes a basic amino acid residue.

[0045] 1) a codon encoding the 60th amino acid residue, a codon encoding the 63rd amino acid residue, and a codon encoding the 65th amino acid residue in the light chain variable region; 2) a codon encoding the 60th amino acid residue, a codon encoding the 63rd amino acid residue, and a codon encoding the 76th amino acid residue in the light chain variable region; 3) a codon encoding the 60th amino acid residue, a codon encoding the 74th amino acid residue, and a codon encoding the 76th amino acid residue in the light chain variable region; 4) a codon encoding the 63rd amino acid residue, a codon encoding the 65th amino acid residue, and a codon encoding the 67th amino acid residue in the light chain variable region; 5) a codon encoding the 63rd amino acid residue, a codon encoding the 65th amino acid residue, and a codon encoding the 70th amino acid residue in the light chain variable region; 6) a codon encoding the 63rd amino acid residue, a codon encoding the 65th amino acid residue, and a codon encoding the 72nd amino acid residue in the light chain variable region; 7) a codon encoding the 63rd amino acid residue, a codon encoding the 67th amino acid residue, and a codon encoding the 70th amino acid residue in the light chain variable region; 8) a codon encoding the 63rd amino acid residue, a codon encoding the 67th amino acid residue, and a codon encoding the 72nd amino acid residue in the light chain variable region; 9) a codon encoding the 63rd amino acid residue, a codon encoding the 70th amino acid residue, and a codon encoding the 72nd amino acid residue in the light chain variable region; 10) a codon encoding the 65th amino acid residue, a codon encoding the 67th amino acid residue, and a codon encoding the 70th amino acid residue in the light chain variable region; 11) a codon encoding the 65th amino acid residue, a codon encoding the 67th amino acid residue, and a codon encoding the 72nd amino acid residue in the light chain variable region; 12) a codon encoding the 65th amino acid residue, a codon encoding the 70th amino acid residue, and a codon encoding the 72nd amino acid residue in the light chain variable region; 13) a codon encoding the 67th amino acid residue, a codon encoding the 70th amino acid residue, and a codon encoding the 72nd amino acid residue in the light chain variable region;14) a codon encoding the 70th amino acid residue, a codon encoding the 72nd amino acid residue, and a codon encoding the 74th amino acid residue in the light chain variable region; 15) a codon encoding the 77th amino acid residue, a codon encoding the 79th amino acid residue, and a codon encoding the 81st amino acid residue in the light chain variable region; 16) a codon encoding the 63rd amino acid residue, a codon encoding the 65th amino acid residue, a codon encoding the 67th amino acid residue, and a codon encoding the 70th amino acid residue in the light chain variable region; 17) a codon encoding the 63rd amino acid residue, a codon encoding the 65th amino acid residue, a codon encoding the 67th amino acid residue, and a codon encoding the 72nd amino acid residue in the light chain variable region; 18) a codon encoding the 63rd amino acid residue, a codon encoding the 65th amino acid residue, a codon encoding the 70th amino acid residue, and a codon encoding the 72nd amino acid residue in the light chain variable region; 19) 20) a codon encoding the 65th amino acid residue, a codon encoding the 67th amino acid residue, a codon encoding the 70th amino acid residue, and a codon encoding the 72nd amino acid residue in the light chain variable region; and 21) a codon encoding the 63rd amino acid residue, a codon encoding the 65th amino acid residue, a codon encoding the 67th amino acid residue, a codon encoding the 70th amino acid residue, and a codon encoding the 72nd amino acid residue in the light chain variable region;

[0046] In the nucleotide sequence encoding the modified antigen-binding region, it is preferred that the codons encoding at least three amino acid residues selected from the group consisting of positions 63, 65, 67, 70, and 72 in the light chain variable region and the codon encoding at least one amino acid residue selected from the group consisting of positions 60, 74, 76, 77, 79, and 81 in the light chain variable region are codons encoding basic amino acid residues.

[0047] When three to six codons in the nucleotide sequence encoding the light chain FR3 of the modified antigen-binding region are codons encoding basic amino acid residues, the three to six codons preferably include codons encoding three, four, or five amino acid residues selected from the group consisting of positions 63, 65, 67, 70, and 72 in the light chain variable region, and codons encoding one, two, or three amino acid residues selected from the group consisting of positions 60, 74, 76, 77, 79, and 81 in the light chain variable region.

[0048] When three to five codons in the nucleotide sequence encoding the light chain FR3 of the modified antigen-binding region are codons encoding basic amino acid residues, the three to five codons preferably include codons encoding three or four amino acid residues selected from the group consisting of positions 63, 65, 67, 70, and 72 in the light chain variable region, and codons encoding one or two amino acid residues selected from the group consisting of positions 60, 74, 76, 77, 79, and 81 in the light chain variable region.

[0049] In the nucleotide sequence encoding the modified antigen-binding region, it is preferred that the codons encoding at least three amino acid residues selected from the group consisting of positions 60, 74, 76, 77, 79, and 81 in the light chain variable region and the codon encoding at least one amino acid residue selected from the group consisting of positions 63, 65, 67, 70, and 72 in the light chain variable region are codons encoding basic amino acid residues.

[0050] When three to six codons in the nucleotide sequence encoding the light chain FR3 of the modified antigen-binding region are codons encoding basic amino acid residues, the three to six codons preferably include codons encoding three, four, or five amino acid residues selected from the group consisting of positions 60, 74, 76, 77, 79, and 81 in the light chain variable region, and codons encoding one, two, or three amino acid residues selected from the group consisting of positions 63, 65, 67, 70, and 72 in the light chain variable region.

[0051] When three to five codons in the nucleotide sequence encoding the light chain FR3 of the modified antigen-binding region are codons encoding basic amino acid residues, the three to five codons preferably include codons encoding three or four amino acid residues selected from the group consisting of positions 60, 74, 76, 77, 79, and 81 in the light chain variable region, and codons encoding one or two amino acid residues selected from the group consisting of positions 63, 65, 67, 70, and 72 in the light chain variable region.

[0052] In the nucleotide sequence encoding the modified antigen-binding region, when a codon other than the codons encoding amino acid residues 60, 63, 65, 67, 70, 72, 74, 76, 77, 79, and 81 in the light chain variable region is modified, the modified codon is preferably a codon encoding an amino acid residue at a site selected from the group consisting of 57, 58, 59, 61, and 62.

[0053] Because CDRs are involved in the affinity of the antigen-binding region for the antigen, it is preferable that the nucleotide sequence encoding the CDR of the antigen-binding region is not altered in the nucleic acid molecule of this embodiment. That is, the amino acid sequence of the CDR of the modified antigen-binding region and the nucleotide sequence encoding it are preferably the same as the amino acid sequence of the CDR of the original antigen-binding region and the nucleotide sequence encoding it.

[0054] Hereinafter, immune cells containing a CAR before modification will also be referred to as "immune cells before modification," and immune cells containing a CAR after modification will also be referred to as "immune cells after modification." When the amino acid sequence of a CAR after modification is the same as that of the CAR before modification, except that at least three amino acid residues in the light chain FR3 are basic amino acid residues, the immune cells after modification can be compared with the immune cells before modification in terms of cytokine production and cytotoxicity. Cytokine production in the immune cells after modification is suppressed compared to the immune cells before modification. The type of cytokine is not particularly limited, but examples include IFNγ, tumor necrosis factor (TNF)-α, and interleukin (IL)-6. As used herein, "cytokine production is suppressed" means that, after a predetermined period of time has elapsed since mixing tumor cells and immune cells at a predetermined ratio, the amount of cytokines released by the immune cells after modification is less than the amount released by the immune cells before modification. For example, the amount of cytokines released by the modified immune cells is 93% or less, 92% or less, 91% or less, 90% or less, 80% or less, 70% or less, 60% or less, 50% or less, 40% or less, 30% or less, or 20% or less of the amount released by the unmodified immune cells. The amount of cytokines released by the CAR-containing immune cells can be determined, for example, by co-culturing the immune cells with tumor cells and then measuring the cytokines in the culture supernatant using a known method such as enzyme-linked immunosorbent assay (ELISA). Specifically, the method described in Example 1 below is exemplified.

[0055] Preferably, the modified immune cells maintain their cytocidal effect against tumor cells compared to the unmodified immune cells, and more preferably, the cytocidal effect is improved. As used herein, "the cytocidal effect is maintained" means that, after a predetermined period of time has elapsed since mixing tumor cells and immune cells at a predetermined ratio, the cytocidal effects of the unmodified and modified immune cells are substantially equivalent. In a preferred embodiment, "the cytocidal effect is maintained" means that the cytocidal effect of the modified immune cells is 95% to 105% of that of the unmodified immune cells. Even if the cytocidal effect of the modified immune cells is reduced compared to that of the unmodified immune cells, a significantly reduced amount of cytokines may result in a better therapeutic effect. When such immune cells are administered to a patient, increasing the dose of the immune cells can improve the cytocidal effect against tumor cells while maintaining low levels of cytokines. In other words, the dose of the immune cells can be appropriately adjusted to maintain or improve the cytocidal effect, as long as cytokine levels are maintained at a low level. The "cytocidal effect" can be evaluated in vitro. For example, the cytocidal effect can be evaluated by indicators such as the number and survival rate of tumor cells when CAR-containing immune cells are co-cultured with tumor cells. Specifically, the method described in Example 1 below is exemplified. Furthermore, when targeting solid cancers, evaluation can also be performed by transplanting tumor cells into a non-human animal such as a mouse, administering CAR-containing immune cells, and measuring the size of the tumor. Specifically, the method described in Example 5 below is exemplified. When targeting blood cancers, evaluation can also be performed by transplanting tumor cells into a non-human animal such as a mouse, administering CAR-containing immune cells, and measuring the number of tumor cells in the blood using a microscope, flow cytometer, or the like.

[0056] As described above, nucleic acid molecules encoding modified antigen-binding regions can be obtained from nucleic acid molecules encoding the original antigen-binding region using known gene recombination techniques and other molecular biology techniques. First, a primer set for codon modification is prepared based on the nucleotide sequence of the original nucleic acid molecule encoding the antigen-binding region. For example, if the codon modification is a codon substitution, a primer set designed to replace at least three codons in the nucleotide sequence encoding the light chain FR3 with codons encoding basic amino acid residues is prepared. This primer set is then used to amplify the original nucleic acid molecule encoding the antigen-binding region by PCR, thereby obtaining a nucleic acid molecule encoding an antigen-binding region in which at least three amino acid residues in the light chain FR3 have been replaced with basic amino acid residues. Alternatively, if the codon modification is a codon insertion, a primer set designed to insert codons encoding basic amino acid residues into at least three positions in the nucleotide sequence encoding the light chain FR3 is prepared. Then, by performing PCR using this primer set, the nucleic acid molecule encoding the original antigen-binding region is amplified as a template to obtain a nucleic acid molecule encoding an antigen-binding region in which basic amino acid residues have been inserted into at least three positions in the light chain FR3.

[0057] As described above, since the nucleic acid molecule encoding the modified antigen-binding region is obtained based on the nucleic acid molecule encoding the original antigen-binding region, it is preferable that a nucleic acid molecule containing a nucleotide sequence encoding the original antigen-binding region is available. For example, if an E. coli clone containing plasmid DNA encoding the original antigen-binding region is available, a nucleic acid molecule containing a nucleotide sequence encoding the original antigen-binding region can be obtained by extracting the plasmid DNA from the E. coli clone. Furthermore, for preparing the above-mentioned primer set, it is preferable that the nucleotide sequence encoding the original antigen-binding region is publicly known or can be confirmed. If the antigen-binding region comprises or consists of a single-chain antibody, the nucleotide sequence encoding the single-chain antibody can be found in publicly known databases such as PDB, GeneBank, abYsis, and IMGT. If a nucleic acid molecule containing a nucleotide sequence encoding the original antigen-binding region is available, the nucleotide sequence encoding the original antigen-binding region can be determined by sequencing the nucleic acid molecule.

[0058] The modified and original antigen-binding regions preferably bind to at least an antigen expressed in tumor cells. The antigen may be an antigen also expressed in normal cells, or an antigen specifically expressed in tumor cells. The antigen includes not only full-length antigens, but also fragments of the antigen and complexes of the antigen fragments with MHC (major histocompatibility complex) proteins. Examples of antigen fragments include portions of the antigen presented by antigen-presenting cells or tumor cells themselves, and synthetic peptides consisting of a portion of the amino acid sequence of the antigen. MHC proteins are called HLA (human leukocyte antigens) in humans, and include HLA-A, HLA-B, HLA-C, HLA-DR, HLA-DQ, and HLA-DP. For example, when the antigen is a protein expressed in tumor cells, the modified and original antigen-binding regions can bind to complexes of the antigen fragments with MHC proteins.

[0059] The antigens recognized by the antigen-binding region are preferably antigens present on the surface of tumor cells and antigens present inside tumor cells. Examples of such antigens include CD19, CD20, CD30, CD44, CD133, MAGE-A1, MAGE-A2, MAGE-A3, MAGE-A4, PRAME, NY-ESO-1, SSX2, GP100, MART-1, HER2, carcinoembryonic antigen (CEA), MUC-1, CA125, glypican 3 (GPC3), B-cell maturation antigen (BCMA), prostate-specific membrane antigen (PSMA), and ganglioside GM2. The antigen recognized by the antigen-binding region may also be a complex of an antigen fragment and an MHC protein. This complex is formed when the antigen is fragmented within tumor cells and the fragment binds to the MHC of the tumor cells. The formed complex is then presented on the surface of the tumor cells. Examples of such complexes include a complex between a MAGE-A4-derived peptide and HLA-A2 (hereinafter also referred to as a "MAGE-A4 / HLA-A2 complex") and a complex between a PRAME-derived peptide and HLA-A24 (hereinafter also referred to as a "PRAME / HLA-A24 complex"). Examples of HLA-A2 that form complexes with MAGE-A4-derived peptides include HLA-A*02:01, HLA-A*02:02, HLA-A*02:03, HLA-A*02:05, HLA-A*02:06, HLA-A*02:07, and HLA-A*02:11. Of these, HLA-A*02:01 is preferred. Examples of HLA-A24 that form complexes with PRAME-derived peptides include HLA-A*24:02 and HLA-A*24:03. Of these, HLA-A*24:02 is preferred. MAGE-A4-derived peptides and PRAME-derived peptides may form complexes with HLAs different from those described above depending on the sequence of the fragmented peptides. MAGE-A4-derived peptides and PRAME-derived peptides are oligopeptides consisting of a portion (e.g., 8 to 20 amino acids) of the amino acid sequence of MAGE-A4 and PRAME, respectively. An example of a MAGE-A4-derived peptide is an oligopeptide consisting of the amino acid sequence GVYDGREHTV (SEQ ID NO: 1).An example of a PRAME-derived peptide is an oligopeptide consisting of the amino acid sequence LYVDSLFFL (SEQ ID NO: 2).

[0060] The nucleic acid molecule encoding the original antigen-binding region itself can be obtained using known genetic recombination techniques and other molecular biology techniques. When the antigen-binding region comprises or consists of a single-chain antibody, nucleic acid molecules encoding single-chain antibodies that bind to tumor cell antigens can be isolated, for example, by phage display using an antibody phage library. Alternatively, hybridomas that produce antibodies that bind to tumor cell antigens can be generated, and then nucleic acid molecules encoding single-chain antibodies can be generated by reverse transcription and PCR using RNA extracted from the hybridomas. Hybridomas can be generated by known methods, such as the method described in Kohler G. and Milstein C., Nature, vol. 256, pp. 495-497, 1975.

[0061] In the segment encoding the extracellular domain (hereinafter also referred to as the "extracellular segment"), the nucleotide sequence encoding the antigen-binding region comprises a nucleotide sequence encoding a light chain variable region and a nucleotide sequence encoding a heavy chain variable region. The nucleotide sequence encoding the antigen-binding region may also comprise a nucleotide sequence encoding part or all of the constant region. The constant region may be either a heavy chain or light chain constant region, but is preferably a light chain constant region. In the nucleotide sequence encoding the antigen-binding region, the order of the nucleotide sequence encoding the light chain variable region and the nucleotide sequence encoding the heavy chain variable region is not particularly limited. For example, the nucleotide sequence encoding the antigen-binding region may comprise, from the 5' end, a nucleotide sequence encoding a light chain variable region and a nucleotide sequence encoding a heavy chain variable region. Alternatively, the nucleotide sequence encoding the antigen-binding region may comprise, from the 5' end, a nucleotide sequence encoding a heavy chain variable region and a nucleotide sequence encoding a light chain variable region. When a nucleotide sequence encoding part or all of the light chain constant region is included, the nucleotide sequence preferably follows the nucleotide sequence encoding the light chain variable region. When a nucleotide sequence encoding part or all of a heavy chain constant region is included, the nucleotide sequence preferably follows the nucleotide sequence encoding the heavy chain variable region.

[0062] The segment encoding the antigen-binding region comprises or consists of a nucleotide sequence encoding a single-chain antibody that binds to any one of the above-mentioned tumor cell antigens. Preferably, the segment encoding the antigen-binding region comprises or consists of a nucleotide sequence encoding a single-chain antibody that binds to a complex of a MAGE-A4-derived peptide and HLA-A2, a complex of a PRAME-derived peptide and HLA-A24, CD19, or CEA.

[0063] Preferably, the nucleotide sequence encoding the single-chain antibody that binds to a complex of a MAGE-A4-derived peptide and HLA-A2 comprises the nucleotide sequence set forth in SEQ ID NOs: 153 to 158. SEQ ID NO: 153 represents the nucleotide sequence encoding HCDR1, SEQ ID NO: 154 represents the nucleotide sequence encoding HCDR2, SEQ ID NO: 155 represents the nucleotide sequence encoding HCDR3, SEQ ID NO: 156 represents the nucleotide sequence encoding LCDR1, SEQ ID NO: 157 represents the nucleotide sequence encoding LCDR2, and SEQ ID NO: 158 represents the nucleotide sequence encoding LCDR3. A specific example of a nucleotide sequence encoding a single-chain antibody comprising SEQ ID NOs: 153 to 158 is set forth in SEQ ID NO: 159. Nucleotide sequences encoding single-chain antibodies based on SEQ ID NO: 159, with modified FR3, are set forth in, for example, SEQ ID NOs: 160 to 169.

[0064] Preferably, the nucleotide sequence encoding the single-chain antibody that binds to a complex of a PRAME-derived peptide and HLA-A24 comprises the nucleotide sequence set forth in SEQ ID NOs: 170 to 175. SEQ ID NO: 170 is the nucleotide sequence encoding HCDR1, SEQ ID NO: 171 is the nucleotide sequence encoding HCDR2, SEQ ID NO: 172 is the nucleotide sequence encoding HCDR3, SEQ ID NO: 173 is the nucleotide sequence encoding LCDR1, SEQ ID NO: 174 is the nucleotide sequence encoding LCDR2, and SEQ ID NO: 175 is the nucleotide sequence encoding LCDR3. A specific example of a nucleotide sequence encoding a single-chain antibody comprising SEQ ID NOs: 170 to 175 is set forth in SEQ ID NO: 176. Nucleotide sequences encoding single-chain antibodies based on SEQ ID NO: 176, with modified FR3, are set forth in, for example, SEQ ID NOs: 177 to 179.

[0065] Preferably, the nucleotide sequence encoding the single-chain antibody that binds to CD19 comprises the nucleotide sequence set forth in SEQ ID NOs: 180 to 185. SEQ ID NO: 180 is the nucleotide sequence encoding HCDR1, SEQ ID NO: 181 is the nucleotide sequence encoding HCDR2, SEQ ID NO: 182 is the nucleotide sequence encoding HCDR3, SEQ ID NO: 183 is the nucleotide sequence encoding LCDR1, SEQ ID NO: 184 is the nucleotide sequence encoding LCDR2, and SEQ ID NO: 185 is the nucleotide sequence encoding LCDR3. A specific example of a nucleotide sequence encoding a single-chain antibody comprising SEQ ID NOs: 180 to 185 is set forth in SEQ ID NO: 186. Nucleotide sequences encoding single-chain antibodies based on SEQ ID NO: 186, with modified FR3, are set forth in, for example, SEQ ID NOs: 187 to 205.

[0066] The nucleotide sequence encoding a single-chain antibody preferably contains a nucleotide sequence encoding a peptide linker between the nucleotide sequence encoding the light chain variable region and the nucleotide sequence encoding the heavy chain variable region. In a single-chain antibody, the peptide linker is the moiety that connects the light chain variable region and the heavy chain variable region. The amino acid sequence of the peptide linker is not particularly limited, but a sequence of 15 to 20 amino acids in length containing repeats of an amino acid sequence consisting of glycine and serine residues (e.g., GGGGS: SEQ ID NO: 206) is generally used. For example, the nucleotide sequence encoding a single-chain antibody may contain, from the 5' end, a nucleotide sequence encoding a light chain variable region, a nucleotide sequence encoding a peptide linker, and a nucleotide sequence encoding a heavy chain variable region. Alternatively, the nucleotide sequence encoding a single-chain antibody may contain, from the 5' end, a nucleotide sequence encoding a heavy chain variable region, a nucleotide sequence encoding a peptide linker, and a nucleotide sequence encoding a light chain variable region.

[0067] In a CAR, the transmembrane domain is a site for anchoring the CAR to the cell membrane of an immune cell. The transmembrane domain of a CAR can be derived from a transmembrane protein. For example, in the nucleic acid molecule of this embodiment, a segment having a nucleotide sequence encoding a transmembrane domain (hereinafter also referred to as a "transmembrane segment") may comprise a nucleotide sequence encoding the entire transmembrane protein. Alternatively, the transmembrane segment may comprise a nucleotide sequence encoding a portion of the transmembrane protein, as long as the function as a transmembrane domain is maintained. The portion of the transmembrane protein preferably comprises all or a portion of the transmembrane region of the protein. Examples of transmembrane proteins include, but are not limited to, the T cell receptor α chain, the T cell receptor β chain, CD3ε, CD3ζ, CD4, CD5, CD8α, CD9, CD16, CD22, CD28, CD33, CD37, CD45, CD64, CD80, CD86, CD134, CD154, 4-1BB (also known as CD137), ICOS (Inducible T-cell co-stimulator), and GITR (Glucocorticoid-induced TNF receptor). Preferably, the transmembrane segment comprises a nucleotide sequence encoding the transmembrane region of any one protein selected from this group of transmembrane proteins. Among these, a nucleotide sequence encoding the transmembrane region of CD8α or CD28 is particularly preferred. A specific example of a nucleotide sequence encoding the transmembrane region of CD28 is set forth in SEQ ID NO: 207.

[0068] The nucleic acid molecule of this embodiment preferably contains a segment encoding a hinge domain (hereinafter also referred to as a "hinge segment") between the nucleotide sequence encoding the antigen-binding region and the transmembrane segment. In a CAR, the hinge domain can provide the extracellular domain with the length necessary for the antigen-binding region to access the antigen and flexibility to avoid steric hindrance. The hinge domain is also called a spacer region. The hinge domain of a CAR can be derived from, for example, a transmembrane protein or IgG. Specifically, in the nucleic acid molecule of this embodiment, the hinge segment may comprise a nucleotide sequence encoding the entire extracellular region of a transmembrane protein or the entire constant region of IgG. Alternatively, the hinge segment may comprise a nucleotide sequence encoding a portion of the extracellular region of a transmembrane protein or a portion of the constant region of IgG, as long as its function as a hinge domain is maintained. Hereinafter, the portion of the extracellular region of a membrane protein that can be used as a hinge domain will also be referred to as a "hinge region." Preferably, the hinge segment comprises a nucleotide sequence encoding any one region selected from the group consisting of an IgG light chain constant region, a CD8α hinge region, and a CD28 hinge region. The IgG is preferably IgG4. A specific example of a nucleotide sequence encoding an IgG light chain constant region is shown in SEQ ID NO: 208. A specific example of a nucleotide sequence encoding a CD28 hinge region is shown in SEQ ID NO: 209.

[0069] In a CAR, the intracellular domain comprises a signaling domain. The signaling domain of a CAR is a site for inducing signal transduction that activates immune cells expressing the CAR when the antigen-binding region binds to an antigen. In the nucleic acid molecule of this embodiment, the segment encoding the intracellular domain (hereinafter also referred to as the "intracellular segment") may comprise a nucleotide sequence encoding the signaling domain. The signaling domain of a CAR may be derived from a membrane protein having an intracellular region, such as a cell membrane receptor or a transmembrane protein. The intracellular region of the membrane protein may contain a signaling domain. That is, the intracellular domain of a CAR may comprise the signaling domain of a membrane protein. Thus, the intracellular segment may comprise a nucleotide sequence encoding the entire membrane protein. Alternatively, the intracellular segment may comprise a nucleotide sequence encoding a portion of the membrane protein, as long as the function as the signaling domain is maintained. Preferably, the portion of the membrane protein comprises all or part of the signaling domain of the protein. Examples of membrane proteins include, but are not limited to, CD3ζ, CD3γ, CD3δ, CD3ε, CD5, CD22, CD66d, CD79a, CD79b, FcRγ, and FcRβ. Preferably, the intracellular segment comprises a nucleotide sequence encoding the signaling domain of at least one protein selected from this group of membrane proteins. Among them, CD3ζ is preferred. A specific example of a nucleotide sequence encoding the signaling domain of CD3ζ is shown in SEQ ID NO: 210.

[0070] The intracellular segment preferably has a nucleotide sequence encoding a costimulatory domain in addition to the nucleotide sequence encoding the signaling domain. It is known that the transmission of a signal from the costimulatory domain to immune cells (particularly T cells) together with the signal from the signaling domain improves the proliferation ability, cytotoxic activity, survival rate, etc. of the immune cells. The costimulatory domain of a CAR can be derived from a membrane protein having an intracellular region, such as a cell membrane receptor or a transmembrane protein. The intracellular region of the membrane protein may contain a costimulatory domain. That is, the intracellular domain of a CAR may contain a costimulatory domain of a membrane protein. Thus, the intracellular segment may contain a nucleotide sequence encoding the entire membrane protein. Alternatively, the intracellular segment may contain a nucleotide sequence encoding a portion of the membrane protein, as long as the function as a costimulatory domain is maintained. The portion of the membrane protein preferably contains all or part of the costimulatory domain of the protein. Examples of membrane proteins include, but are not limited to, 4-1BB (also known as CD137), CD28, GITR, CD2, CD5, CD8, CD9, CD16, CD22, CD33, CD37, CD64, CD80, CD86, CD134, CD154, and ICOS. Preferably, the intracellular segment comprises a nucleotide sequence encoding the costimulatory domain of at least one protein selected from these membrane proteins. Among these, at least one selected from the group consisting of 4-1BB, CD28, and GITR is preferred. A specific example of a nucleotide sequence encoding the costimulatory domain of CD28 is set forth in SEQ ID NO: 211. A specific example of a nucleotide sequence encoding the costimulatory domain of GITR is set forth in SEQ ID NO: 212.

[0071] In the intracellular domain segment, the order of the nucleotide sequence encoding the signaling domain and the nucleotide sequence encoding the costimulatory domain is not particularly limited. For example, the nucleotide sequence encoding the intracellular domain may include, from the 5' end, a nucleotide sequence encoding the signaling domain and a nucleotide sequence encoding the costimulatory domain. Alternatively, the nucleotide sequence encoding the intracellular domain may include, from the 5' end, a nucleotide sequence encoding the costimulatory domain and a nucleotide sequence encoding the signaling domain.

[0072] The nucleic acid molecule of this embodiment may contain various nucleotide sequences in addition to the nucleotide sequence encoding the CAR, as necessary. Examples of such nucleotide sequences include a leader sequence, a restriction enzyme recognition sequence, a nucleotide sequence encoding a peptide tag, a stop codon, etc. The peptide tag can be appropriately selected from known peptide tags, such as a histidine tag, a glutathione-S-transferase (GST) tag, or a FLAG (registered trademark) tag.

[0073] A further embodiment of the present invention relates to a vector comprising the nucleic acid molecule described in 1 above. Specifically, the vector of this embodiment may be in a form in which the nucleic acid molecule of this embodiment is incorporated into a known vector. The type of vector is not particularly limited, and examples thereof include a plasmid vector and a viral vector. The vector may be linear or circular. The type of plasmid vector is not particularly limited, and examples thereof include an expression vector, a vector for producing a viral vector, a transposon vector, and a cloning vector. An expression vector is a vector that enables the expression of a protein encoded by the nucleotide sequence of a nucleic acid molecule incorporated into the vector in an appropriate host cell such as a mammalian cell, an insect cell, a yeast, or an Escherichia coli. A transposon vector is a vector that, when introduced into an appropriate host together with an expression vector incorporating a gene encoding a transposase, enables the nucleic acid molecule incorporated into the transposon vector to be integrated into the genome of the host cell.

[0074] The type of viral vector is not particularly limited, and examples thereof include retroviral vectors, lentiviral vectors, adenoviral vectors, adeno-associated virus (AAV) vectors, vaccinia virus vectors, Epstein-Barr virus (EBV) vectors, etc. The viral vector is preferably replication-deficient so that the virus does not self-replicate in infected cells.

[0075] The vector may contain appropriate control sequences as necessary, such as a promoter sequence, an operator sequence, an enhancer sequence, a nucleotide sequence encoding a drug resistance marker, a multicloning site, and the like.

[0076] As described above, the nucleic acid molecule of this embodiment may be either DNA or RNA. DNA is a more stable substance than RNA, and various DNA vectors are commercially available. Therefore, the nucleic acid molecule of this embodiment, which is DNA, is advantageous in that it is easy to store and handle. It is known that RNA encoding a protein can express the protein without being affected by the transcriptional regulatory process when introduced into a cell. Therefore, the nucleic acid molecule of this embodiment, which is RNA, is advantageous in that it enables rapid CAR expression in immune cells.

[0077] 2. CAR Another embodiment of the present invention relates to a CAR. The CAR of this embodiment is a protein molecule encoded by the nucleotide sequence of the nucleic acid molecule described in 1 above. The CAR of this embodiment is a protein produced by genetic engineering techniques and includes, in order from the N-terminus, an extracellular domain, a transmembrane domain, and an intracellular domain. The extracellular domain includes an antigen-binding region comprising a light chain variable region and a heavy chain variable region. The CAR of this embodiment is characterized in that at least three amino acid residues in the light chain FR3 of the antigen-binding region are basic amino acid residues. That is, the CAR of this embodiment includes an antigen-binding region in which at least three amino acid residues in the light chain FR3 are basic amino acid residues. This antigen-binding region is the same as the modified antigen-binding region described above. Hereinafter, the antigen-binding region of the CAR of this embodiment will also be referred to as the "modified antigen-binding region." The CAR of this embodiment can also be considered a variant of a CAR having an original antigen-binding region. The CAR of this embodiment is preferably expressed in immune cells and anchored to the cell membrane of the immune cells. Immune cells expressing the CAR exhibit suppressed cytokine production compared to immune cells expressing a CAR containing the original antigen-binding region, and preferably maintain, or more preferably improve, the cytocidal effect of the immune cells against tumor cells.

[0078] In the CAR of this embodiment, at least three basic amino acid residues in the light chain FR3 are derived from the above-mentioned codon modification. In the light chain FR3 of the modified antigen-binding region, the at least three basic amino acid residues derived from the codon modification may all be arginine residues or all lysine residues. Alternatively, some of the at least three basic amino acid residues derived from the codon modification may be arginine residues, with the remainder being lysine residues.

[0079] The at least three amino acid residues in the light chain FR3 before being changed to basic amino acid residues are neutral or acidic amino acid residues, preferably neutral amino acid residues. That is, the at least three basic amino acid residues in the light chain FR3 of the modified antigen-binding region are changed from at least three residues selected from neutral and / or acidic amino acid residues in the light chain FR3 of the original antigen-binding region.

[0080] In the light chain FR3 of the modified antigen-binding region, the number of at least three basic amino acid residues resulting from codon modification is, for example, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, or 14. In the light chain FR3 of the modified antigen-binding region, the number of at least three basic amino acid residues resulting from codon modification is preferably 3 to 6, more preferably 3 to 5. As described above, the at least three basic amino acid residues in the light chain FR3 of the modified antigen-binding region are preferably located at positions selected from amino acid residues 57, 58, 59, 60, 61, 62, 63, 65, 67, 70, 72, 74, 76, 77, 79, 80, and 81 in the light chain variable region.

[0081] In the modified antigen-binding region, it is preferred that at least three amino acid residues selected from the group consisting of amino acid residues at positions 60, 63, 65, 67, 70, 72, 74, 76, 77, 79, and 81 in the light chain variable region are basic amino acid residues derived from codon modification. More preferably, three to six amino acid residues selected from the above group are basic amino acid residues. Even more preferably, three to five amino acid residues selected from the above group are basic amino acid residues. For example, the basic amino acid residues derived from codon modification are any of the amino acid residues listed in 1) to 21) below.

[0082] 1) amino acid residues 60, 63, and 65 of the light chain variable region; 2) amino acid residues 60, 63, and 76 of the light chain variable region; 3) amino acid residues 60, 74, and 76 of the light chain variable region; 4) amino acid residues 63, 65, and 67 of the light chain variable region; 5) amino acid residues 63, 65, and 70 of the light chain variable region; 6) amino acid residues 63, 65, and 72 of the light chain variable region; 7) amino acid residues 63, 67, and 70 of the light chain variable region; 8) amino acid residues 63, 67, and 72 of the light chain variable region; 9) amino acid residues 63, 70, and 72 of the light chain variable region; 10) amino acid residues 65, 67, and 70 of the light chain variable region; 11) amino acid residues 65, 67, and 72 of the light chain variable region; 12) amino acid residues 65, 70, and 72 of the light chain variable region. 13) amino acid residues 67, 70, and 72 in the light chain variable region; 14) amino acid residues 70, 72, and 74 in the light chain variable region; 15) amino acid residues 77, 79, and 81 in the light chain variable region; 16) amino acid residues 63, 65, 67, and 70 in the light chain variable region; 17) amino acid residues 63, 65, 67, and 72 in the light chain variable region; 18) amino acid residues 63, 65, 70, and 72 in the light chain variable region; 19) amino acid residues 63, 67, 70, and 72 in the light chain variable region; 20) amino acid residues 65, 67, 70, and 72 in the light chain variable region; and 21) amino acid residues 63, 65, 67, 70, and 72 in the light chain variable region.

[0083] In the altered antigen-binding region, it is preferred that at least three amino acid residues selected from the group consisting of positions 63, 65, 67, 70, and 72 in the light chain variable region and at least one amino acid residue selected from the group consisting of positions 60, 74, 76, 77, 79, and 81 in the light chain variable region are basic amino acid residues.

[0084] When three to six amino acid residues in the light chain FR3 of the altered antigen-binding region are basic amino acid residues, the three to six amino acid residues preferably include three, four, or five amino acid residues selected from the group consisting of amino acid residues at positions 63, 65, 67, 70, and 72 in the light chain variable region, and one, two, or three amino acid residues selected from the group consisting of amino acid residues at positions 60, 74, 76, 77, 79, and 81 in the light chain variable region.

[0085] When three to five amino acid residues in the light chain FR3 of the altered antigen-binding region are basic amino acid residues, the three to five amino acid residues preferably include three or four amino acid residues selected from the group consisting of amino acid residues at positions 63, 65, 67, 70, and 72 in the light chain variable region, and one or two amino acid residues selected from the group consisting of amino acid residues at positions 60, 74, 76, 77, 79, and 81 in the light chain variable region.

[0086] In the altered antigen-binding region, it is preferred that at least three amino acid residues selected from the group consisting of positions 60, 74, 76, 77, 79, and 81 in the light chain variable region and at least one amino acid residue selected from the group consisting of positions 63, 65, 67, 70, and 72 in the light chain variable region are basic amino acid residues.

[0087] When three to six amino acid residues in the light chain FR3 of the altered antigen-binding region are basic amino acid residues, the three to six amino acid residues preferably include three, four, or five amino acid residues selected from the group consisting of amino acid residues at positions 60, 74, 76, 77, 79, and 81 in the light chain variable region, and one, two, or three amino acid residues selected from the group consisting of amino acid residues at positions 63, 65, 67, 70, and 72 in the light chain variable region.

[0088] When three to five amino acid residues in the light chain FR3 of the altered antigen-binding region are basic amino acid residues, the three to five amino acid residues preferably include three or four amino acid residues selected from the group consisting of amino acid residues at positions 60, 74, 76, 77, 79, and 81 in the light chain variable region, and one or two amino acid residues selected from the group consisting of amino acid residues at positions 63, 65, 67, 70, and 72 in the light chain variable region.

[0089] In the modified antigen-binding region, when an amino acid residue other than amino acid residues 60, 63, 65, 67, 70, 72, 74, 76, 77, 79, and 81 in the light chain variable region is modified, the modified amino acid residue is preferably an amino acid residue at a site selected from the group consisting of amino acid residues 57, 58, 59, 61, and 62.

[0090] In the CAR of this embodiment, the extracellular domain comprises an antigen-binding region comprising a light chain variable region and a heavy chain variable region. Examples of antigen-binding regions include those of single-chain antibodies. The antigen-binding region may comprise a part or all of the constant region of an antibody. The constant region may be either a heavy chain or a light chain constant region, but is preferably a light chain constant region. The order of the light chain variable region and the heavy chain variable region in the antigen-binding region is not particularly limited. For example, the antigen-binding region may comprise, in order from the N-terminus, a light chain variable region and a heavy chain variable region. Alternatively, the antigen-binding region may comprise, in order from the N-terminus, a heavy chain variable region and a light chain variable region. In the antigen-binding region, it is preferable that a part or all of the light chain constant region is included after the light chain variable region. In the antigen-binding region, it is preferable that a part or all of the heavy chain constant region is included after the heavy chain variable region.

[0091] In the CAR of this embodiment, the antigen-binding region may comprise or consist of a single-chain antibody that binds to any one of the above-mentioned antigens present on the surface of tumor cells and antigens present inside tumor cells. Preferably, the antigen-binding region comprises or consists of a single-chain antibody that binds to a complex of a MAGE-A4-derived peptide and HLA-A2, a complex of a PRAME-derived peptide and HLA-A24, CD19, or CEA.

[0092] Preferably, the amino acid sequence of the single-chain antibody that binds to a complex of a MAGE-A4-derived peptide and HLA-A2 comprises the amino acid sequence set forth in SEQ ID NOs: 213 to 218. SEQ ID NO: 213 is the amino acid sequence of HCDR1, SEQ ID NO: 214 is the amino acid sequence of HCDR2, SEQ ID NO: 215 is the amino acid sequence of HCDR3, SEQ ID NO: 216 is the amino acid sequence of LCDR1, SEQ ID NO: 217 is the amino acid sequence of LCDR2, and SEQ ID NO: 218 is the amino acid sequence of LCDR3. A specific example of the amino acid sequence of a single-chain antibody comprising SEQ ID NOs: 213 to 218 is set forth in SEQ ID NO: 219. The amino acid sequences of single-chain antibodies based on SEQ ID NO: 219 but with modified FR3 are set forth in, for example, SEQ ID NOs: 220 to 229.

[0093] Preferably, the amino acid sequence of the single-chain antibody that binds to a complex of a PRAME-derived peptide and HLA-A24 comprises the amino acid sequence set forth in SEQ ID NOs: 230 to 235. SEQ ID NO: 230 is the amino acid sequence of HCDR1, SEQ ID NO: 231 is the amino acid sequence of HCDR2, SEQ ID NO: 232 is the amino acid sequence of HCDR3, SEQ ID NO: 233 is the amino acid sequence of LCDR1, SEQ ID NO: 234 is the amino acid sequence of LCDR2, and SEQ ID NO: 235 is the amino acid sequence of LCDR3. A specific example of the amino acid sequence of a single-chain antibody comprising SEQ ID NOs: 230 to 235 is shown in SEQ ID NO: 236. The amino acid sequences of single-chain antibodies based on SEQ ID NO: 236 but with modified FR3 are shown, for example, in SEQ ID NOs: 237 to 239.

[0094] Preferably, the amino acid sequence of the single-chain antibody that binds to CD19 comprises the amino acid sequence set forth in SEQ ID NOs: 240 to 245. SEQ ID NO: 240 is the amino acid sequence of HCDR1, SEQ ID NO: 241 is the amino acid sequence of HCDR2, SEQ ID NO: 242 is the amino acid sequence of HCDR3, SEQ ID NO: 243 is the amino acid sequence of LCDR1, SEQ ID NO: 244 is the amino acid sequence of LCDR2, and SEQ ID NO: 245 is the amino acid sequence of LCDR3. A specific example of the amino acid sequence of the single-chain antibody comprising SEQ ID NOs: 240 to 245 is set forth in SEQ ID NO: 246. The amino acid sequences of single-chain antibodies based on SEQ ID NO: 246 but with modified FR3 are set forth in, for example, SEQ ID NOs: 247 to 265.

[0095] Preferably, the single-chain antibody comprises a peptide linker between the light chain variable region and the heavy chain variable region. The peptide linker is as described above. For example, the single-chain antibody may comprise, from the N-terminus, a light chain variable region, a peptide linker, and a heavy chain variable region. Alternatively, the single-chain antibody may comprise, from the N-terminus, a heavy chain variable region, a peptide linker, and a light chain variable region.

[0096] In the CAR of this embodiment, the transmembrane domain may comprise the entire length of a transmembrane protein. Alternatively, the transmembrane domain may comprise a portion of the transmembrane protein, as long as its function is maintained. Preferably, the portion of the transmembrane protein comprises all or part of the transmembrane region of the protein. Examples of transmembrane proteins that can be used for the transmembrane domain include the proteins exemplified in 1. above. Preferably, the transmembrane domain is the transmembrane region of any one protein selected from the group consisting of the T cell receptor α chain, the T cell receptor β chain, CD3ε, CD3ζ, CD4, CD5, CD8α, CD9, CD16, CD22, CD28, CD33, CD37, CD45, CD64, CD80, CD86, CD134, CD154, 4-1BB (also referred to as CD137), ICOS, and GITR. Among these, the transmembrane region of CD8α or CD28 is preferred. A specific example of the amino acid sequence of the transmembrane region of CD28 is shown in SEQ ID NO: 266.

[0097] The CAR of this embodiment preferably includes a hinge domain between the antigen-binding region and the transmembrane domain. The hinge domain may include the entire extracellular region of a transmembrane protein or the entire IgG constant region. Alternatively, the hinge domain may include a portion of the extracellular region of a transmembrane protein or a portion of the IgG constant region, as long as its function is maintained. Proteins that can be used for the hinge domain include the proteins exemplified in 1. above. Preferably, the hinge domain is any one region selected from the group consisting of the IgG light chain constant region, the CD8α hinge region, and the CD28 hinge region. The IgG is preferably IgG4. A specific example of the amino acid sequence of the IgG light chain constant region is shown in SEQ ID NO: 267. A specific example of the amino acid sequence of the CD28 hinge region is shown in SEQ ID NO: 268.

[0098] In the CAR of this embodiment, the intracellular domain may comprise the entire length of a membrane protein having an intracellular region. Alternatively, the intracellular domain may comprise a portion of the above membrane protein, as long as its function as a signaling domain is maintained. The signaling domain may be present in the intracellular region of the membrane protein. The portion of the membrane protein having an intracellular region preferably comprises all or part of the signaling domain of the protein. Examples of membrane proteins that can be used for the intracellular domain include the proteins exemplified in 1. above. Preferably, the intracellular domain comprises the signaling domain of at least one protein selected from the group consisting of CD3ζ, CD3γ, CD3δ, CD3ε, CD5, CD22, CD66d, CD79a, CD79b, FcRγ, and FcRβ. Among these, the signaling domain of CD3ζ is preferred. A specific example of the amino acid sequence of the signaling domain of CD3ζ is shown in SEQ ID NO: 269.

[0099] In the CAR of this embodiment, the intracellular domain preferably has a costimulatory domain in addition to the signaling domain. The intracellular domain may include the entire length of a membrane protein having an intracellular region. Alternatively, the intracellular domain may include a portion of the membrane protein, as long as the function as a costimulatory domain is maintained. The costimulatory domain may be present in the intracellular region of each of the above proteins. The portion of the membrane protein having an intracellular region preferably includes all or a portion of the costimulatory domain of the protein. Examples of membrane proteins that can be used as costimulatory domains include the proteins exemplified in 1. above. Preferably, the costimulatory domain includes the costimulatory domain of at least one protein selected from the group consisting of 4-1BB (also known as CD137), CD28, GITR, CD2, CD5, CD8, CD9, CD16, CD22, CD33, CD37, CD64, CD80, CD86, CD134, CD154, and ICOS. Among these, the costimulatory domains of 4-1BB, CD28, and GITR are preferred. A specific example of the amino acid sequence of the costimulatory domain of CD28 is shown in SEQ ID NO: 270. A specific example of the amino acid sequence of the costimulatory domain of GITR is shown in SEQ ID NO: 271.

[0100] In the intracellular domain, the order of the signaling domain and the costimulatory domain is not particularly limited. For example, the intracellular domain may include, in order from the N-terminus, the signaling domain and the costimulatory domain. Alternatively, the intracellular domain may include, in order from the N-terminus, the costimulatory domain and the signaling domain.

[0101] The CAR of this embodiment may contain additional oligopeptides or polypeptides as needed. Examples of such oligopeptides and polypeptides include signal peptides and peptide tags. The peptide tag can be appropriately selected from known peptide tags such as a histidine tag, a GST tag, and a FLAG (registered trademark) tag.

[0102] The CAR of this embodiment can be produced using a protein expression system with the nucleic acid molecule of this embodiment. The protein expression system may be an expression system using host cells or a cell-free protein synthesis system. When produced using an expression system using host cells, for example, the nucleic acid molecule of this embodiment incorporated into an expression vector suitable for the host cell to be used can be introduced into the host cell to express the CAR of this embodiment. Examples of cell-free protein synthesis systems include wheat germ-derived synthesis systems, Escherichia coli-derived synthesis systems, and reconstituted cell-free protein synthesis systems. When CAR is produced in host cells, the host cells can be lysed in a solution containing an appropriate solubilizing agent to release the CAR into the solution. In a cell-free protein synthesis system, the synthesized CAR is contained in the reaction solution. The released CAR can be recovered by known methods such as column chromatography. For example, when the produced CAR has a histidine tag or a GST tag as a peptide tag, it can be recovered by affinity chromatography using a carrier containing Ni-NTA (nitrilotriacetic acid chelated with nickel ions) or glutathione. If necessary, the recovered CAR may be purified by known methods such as gel filtration and dialysis.

[0103] 3. Immune Cells Comprising a CAR A further embodiment of the present invention relates to immune cells comprising a CAR. The CAR-comprising immune cells of this embodiment are immune cells that express the CAR described in 2. above by introducing the nucleic acid molecule described in 1. above into the immune cells. Details of the CAR and nucleic acid molecule are as described above. Examples of immune cells before introduction of the nucleic acid molecule described in 1. above include immune cells collected from mammals, including humans, immune cells prepared from stem cells, and immune cells established as cell lines. Specific examples include immune cells obtained by leukapheresis and immune cells isolated from blood.

[0104] Examples of immune cells separated or isolated from biological samples include peripheral blood mononuclear cells (PBMCs), T cells, and natural killer (NK) cells. Among these, T cells and NK cells are preferred. T cells include CD8-positive T cells, CD4-positive T cells, cytotoxic T cells, helper T cells, regulatory T cells, and tumor-infiltrating lymphocytes. Preferred T cells are CD8-positive T cells and cytotoxic T cells. Preferred immune cell cell lines are lymphocyte-derived cell lines, such as Jurkat cells, MOLT-4 cells, and U-937 cells. Hereinafter, CAR-expressing T cells may be referred to as "CAR-T cells."

[0105] The method for introducing the nucleic acid molecule described in 1. above into immune cells is not particularly limited and can be appropriately selected from known gene transfer methods. Examples of gene transfer methods include lipofection, electroporation, calcium phosphate, cationic polymer-mediated gene transfer, viral vector-mediated gene transfer, and transposon-mediated gene transfer. For lipofection and cationic polymer-mediated gene transfer, commercially available transfection reagents such as FuGENE (registered trademark) and JetPEI (registered trademark) may be used.

[0106] Immune cells containing the CAR of this embodiment can be cultured in the same manner as immune cells before the introduction of the nucleic acid molecule of this embodiment. The medium, serum, additives, etc. can be appropriately determined depending on the type of immune cells used. Examples of media include MEM, DMEM, and RPMI-1640. When the immune cells are lymphocytes, commercially available lymphocyte media such as GT-T502, GT-T503, and GT-T551 (Takara Bio Inc.) may be used. Examples of serum include fetal bovine serum (FBS) and human AB serum. Examples of additives include L-glutamine, insulin, and IL-2. Immune cell culture conditions include, for example, 37°C in a 5% CO2 atmosphere.

[0107] In immune cells containing the CAR of this embodiment, the CAR is expressed as a transmembrane protein. When the extracellular domain of the CAR binds to an antigen on a tumor cell, the immune cells are activated by a signal from the intracellular domain of the CAR. The activated immune cells containing the CAR of this embodiment exert a cytocidal effect by releasing cytotoxic proteins (e.g., perforin, granzymes, etc.) and antitumor cytokines (e.g., tumor necrosis factor (TNF)-α, lymphokines, etc.), and expressing cell surface molecules that induce cell death, such as Fas ligand. Preferably, immune cells containing the CAR of this embodiment maintain their cytocidal effect, and more preferably, have an improved cytocidal effect, compared to immune cells containing a CAR with the original antigen-binding domain. The cytocidal effect of immune cells containing a CAR can be examined by known methods, such as a cytotoxicity assay. Specifically, the method described in Example 3 below is exemplified. Alternatively, as described in Example 4 below, the cytocidal effect can be evaluated by administering immune cells containing a CAR to a non-human animal transplanted with tumor cells and measuring the tumor size. On the other hand, immune cells containing the CAR of this embodiment release reduced amounts of cytokines (e.g., IFNγ, granulocyte-macrophage colony-stimulating factor (GM-CSF), etc.) that cause CRS, compared to immune cells containing a CAR with the original antigen-binding region. The amount of cytokines released by immune cells containing a CAR can be determined by known methods such as enzyme-linked immunosorbent assay (ELISA). Specifically, the method described in Example 5 below is exemplified.

[0108] 4. Method for Producing Immune Cells Containing CAR A further embodiment of the present invention relates to a method for producing immune cells containing CAR (hereinafter also referred to as the "production method of this embodiment"). In the production method of this embodiment, the nucleic acid molecule described in 1. above is introduced into immune cells, and the immune cells are caused to express the CAR described in 2. above, thereby producing immune cells containing the CAR described in 3. above. Details of the CAR and nucleic acid molecule are as described above. Details of the immune cells before introduction of the nucleic acid molecule described in 1. above are as described above. Preferred are PBMCs, T cells, and NK cells. When immune cells containing CAR obtained by the production method of this embodiment are transplanted into a living organism, it is preferable to use immune cells separated or isolated from a biological sample collected from the living organism itself or another living organism of the same species as the living organism.

[0109] The method for introducing the nucleic acid molecule described in 1. above into immune cells is as described above. After introducing the nucleic acid molecule into immune cells, the immune cells are preferably cultured for a predetermined period of time. The method for culturing immune cells is as described above. The predetermined period may be at least the period until CAR is expressed in the immune cells. The period until CAR expression is determined depending on the method for introducing the nucleic acid molecule, but may be, for example, 3 hours to 72 hours, preferably 6 hours to 48 hours. When the method for introducing the nucleic acid molecule is a method that enables stable expression of CAR in immune cells, the predetermined period may be a period sufficient for the proliferation of immune cells expressing CAR. Such a period is not particularly limited, but may be, for example, 48 hours to 20 days, preferably 72 hours to 14 days.

[0110] By introducing the nucleic acid molecule described in 1. above into immune cells, the CAR described in 2. above is expressed in the immune cells. If necessary, the expression of CAR in immune cells may be confirmed. The expression of CAR in immune cells may be confirmed by known protein detection methods, such as ELISA, flow cytometry, immunoprecipitation, polyacrylamide gel electrophoresis, and Western blotting.

[0111] 5. Pharmaceutical Composition A further embodiment of the present invention relates to a pharmaceutical composition for treating malignant tumors, comprising immune cells containing CAR (hereinafter also referred to as the "pharmaceutical composition of this embodiment"). The pharmaceutical composition of this embodiment contains the immune cells containing CAR described in 3. above as an active ingredient. Details of the CAR and the immune cells containing the same are as described above. The pharmaceutical composition of this embodiment may further contain a pharmaceutically acceptable additive. Examples of such additives include an aqueous medium for stably preserving immune cells, D-glucose, dextran, serum albumin, dimethyl sulfoxide (DMSO), etc. Examples of the aqueous medium include physiological saline, phosphate-buffered saline (PBS), and complex electrolyte solutions. The pharmaceutical composition of this embodiment is preferably administered parenterally to patients. That is, the pharmaceutical composition is preferably in a form suitable for parenteral administration, such as an injection or infusion.

[0112] The malignant tumors targeted for treatment by the pharmaceutical composition of this embodiment are tumors containing tumor cells bearing an antigen recognized by CAR. The malignant tumors may be blood cancers or solid cancers. Examples of malignant tumors include acute leukemia (acute myeloid leukemia, B-cell acute lymphoblastic leukemia, acute monocytic leukemia, acute erythroleukemia, acute megakaryoblastic leukemia, etc.), chronic leukemia (chronic myeloid leukemia, chronic lymphocytic leukemia, chronic monocytic leukemia, etc.), lymphomas (diffuse large B-cell lymphoma, follicular lymphoma, mantle cell lymphoma, Burkitt lymphoma, Hodgkin lymphoma, lymphoplasmacytic lymphoma, marginal zone lymphoma, etc.), multiple myeloma, melanoma, breast cancer, prostate cancer, bladder cancer, uterine cancer, liver cancer, kidney cancer, pancreatic cancer, spleen cancer, lung cancer, stomach cancer, esophageal cancer, ovarian cancer, osteosarcoma, and neuroblastoma.

[0113] 6. Treatment Method A further embodiment of the present invention relates to a method for treating malignant tumors using immune cells containing the CAR described in 3 above (hereinafter also referred to as the "treatment method of this embodiment"). The method comprises administering the above-described pharmaceutical composition to a patient with malignant tumor. The malignant tumors to be treated are as described above.

[0114] In the administration step, the immune cells are preferably administered to the patient parenterally. Parenteral administration includes, for example, intravenous, intraarterial, intramuscular, intraperitoneal, and subcutaneous administration. Among these, intravenous administration is preferred.

[0115] The administration of the pharmaceutical composition of this embodiment can be appropriately determined depending on the type of cancer, the patient's condition, age, body weight, etc. The dose is, for example, 1 x 10 immune cells containing CAR per administration for an adult weighing 50 kg or more. 4 cells or more 1×10 10 cells or less, preferably 1 x 10 5 cells or more 1×10 9 cells or less, preferably 1 x 10 6 cells or more 5×10 8 The number of administrations is 1 to 4 times per day, or less. The number of administrations may be one or more. After administration, immune cells containing CARs engraft and proliferate in the patient's body, so the number of administrations is usually one. However, if it is determined that the engraftment and proliferation of immune cells is insufficient, multiple administrations may be performed. The administration interval can be, for example, one to four times per day, weekly, every 10 to 30 days, monthly, every 3 to 6 months, or once a year.

[0116] The pharmaceutical composition of this embodiment can be used in combination with other anticancer drugs. The administration of the other anticancer drug and the administration of the pharmaceutical composition described in 5. above may be on the same day or on different days. Examples of other anticancer drugs include, but are not limited to, alkylating agents such as cyclophosphamide, metabolic antagonists such as pentostatin, molecularly targeted drugs such as rituximab, kinase inhibitors such as imatinib, proteasome inhibitors such as bortezomib, calcineurin inhibitors such as cyclosporine, anticancer antibiotics such as idarubicin, plant alkaloids such as irinotecan, platinum preparations such as cisplatin, hormone therapy drugs such as tamoxifen, and immunosuppressants such as nivolumab and pembrolizumab.

[0117] Prior to administration of the pharmaceutical composition of this embodiment, the patient may be treated with lymphocyte-depleting chemotherapy to reduce the number of white blood cells in the patient, using, for example, fludarabine, cyclophosphamide, or bendamustine.

[0118] 7. Method for Reducing Cytokine Production in Immune Cells A further embodiment of the present invention relates to a method for reducing cytokine production in immune cells containing a CAR (hereinafter also referred to as the "reducing method of this embodiment"). The reducing method of this embodiment comprises the steps of preparing the nucleic acid molecule described in 1. above and introducing the nucleic acid molecule described in 1. above into immune cells to allow the immune cells to express the CAR described in 2. above. The obtained immune cells containing a CAR express a CAR containing a modified antigen-binding region. Immune cells expressing the CAR exhibit reduced cytokine production compared to immune cells expressing a CAR containing the original antigen-binding region. Meanwhile, the cytocidal effect of immune cells expressing a CAR containing the modified antigen-binding region is maintained or improved compared to immune cells expressing a CAR containing the original antigen-binding region. The reducing method of this embodiment can contribute to further reducing cytokine-induced side effects in therapeutic methods using immune cells expressing CARs. Details of the nucleic acid molecule and CAR are as described above. The method for introducing a nucleic acid molecule into immune cells and the method for measuring cytokines produced by immune cells containing a CAR are as described above.

[0119] The present invention will be described in more detail below with reference to examples, but the present invention is not limited to these examples.

[0120] Example 1: Generation of CAR-T cells that bind to the MAGE-A4 / HLA-A2 complex and confirmation of their effects (1)

[0121] (1) Obtaining a nucleic acid molecule encoding a template CAR. Similar to the example in US2020 / 0276237 (which is incorporated herein by reference), two plasmid DNAs for viral vector construction containing a gene encoding a CAR with a single-chain antibody that binds to the MAGE-A4 / HLA-A2 complex were obtained. The complex recognized by the single-chain antibody was a complex of MAGE-A4 and HLA-A*02:01. Hereinafter, the CAR encoded by the gene in one plasmid DNA will be referred to as "MAGE-A4-zG," and the CAR encoded by the gene in the other plasmid DNA will be referred to as "MAGE-A4-zG-s1." The amino acid sequence of the MAGE-A4-derived peptide in the above MAGE-A4 / HLA-A2 complex was GVYDGREHTV (SEQ ID NO: 1). Each of the above plasmid DNAs was used as a template to generate nucleic acid molecules encoding MAGE-A4-zG and MAGE-A4-zG-s1 mutants by PCR.

[0122] Referring to Figure 2A, the gene encoding MAGE-A4-zG contained, from the 5' end, a leader sequence (Leader), a nucleotide sequence encoding a heavy chain variable region (VH), a nucleotide sequence encoding a linker (L), a nucleotide sequence encoding a light chain variable region (VL), a nucleotide sequence encoding a light chain constant region (CL), a nucleotide sequence encoding the CD28 transmembrane domain (CD28™), a nucleotide sequence encoding CD3ζ, and a nucleotide sequence encoding the GITR intracellular domain (GITRICD), linked in this order. VH, L, VL, and CL constituted segments having nucleotide sequences encoding the extracellular domain of CAR. VH, L, and VL constituted segments having nucleotide sequences encoding a single-chain antibody that specifically binds to the MAGE-A4 / HLA-A2 complex. CD3ζ and GITRICD constituted segments having nucleotide sequences encoding the intracellular domain of CAR. CL was a segment having a nucleotide sequence encoding the hinge domain of CAR.

[0123] The gene encoding MAGE-A4-zG-s1 had the same structure as the gene encoding MAGE-A4-zG, except that an extra nucleotide sequence was added to the 3' end of the gene encoding MAGE-A4-zG (3' side of GITRICD). Searches of publicly known databases confirmed that this extra nucleotide sequence did not match any nucleotide sequence encoding any protein. Furthermore, computer analysis of the amino acid sequence corresponding to this extra nucleotide sequence predicted that it would form a linear polypeptide without any secondary structure.

[0124] MAGE-A4-zG and MAGE-A4-zG-s1 were CARs containing a single-chain antibody consisting of VH, VL and a linker connecting them, and CL as extracellular domains, CD28TM as transmembrane domains, and CD3ζ and GITRICD as intracellular domains.

[0125] (2) Obtaining nucleic acid molecules encoding MAGE-A4-zG and MAGE-A4-zG-s1 mutants [Reagents] QIAprep Spin Miniprep Kit (QIAGEN), PrimeSTAR® Max Premix (Takara Bio Inc.), Ligation high ver.2 (Toyobo Co., Ltd.), T4 Polynucleotide Kinase (Toyobo Co., Ltd.), Dpn I (Toyobo Co., Ltd.), Competent high DH5α (Toyobo Co., Ltd.)

[0126] (2.1) Primer design and PCR A primer set for obtaining a polynucleotide encoding a CAR comprising a single-chain antibody in which the amino acid residue shown in a) or b) below in MAGE-A4-zG has been substituted with an arginine residue was designed based on the nucleotide sequence of SEQ ID NO: 3. Furthermore, a primer set for obtaining a polynucleotide encoding a CAR comprising a single-chain antibody in which the amino acid residue shown in a) below in MAGE-A4-zG-s1 has been substituted with an arginine residue was designed based on the nucleotide sequence of SEQ ID NO: 9.

[0127] a) amino acid residues 63, 65, 67, and 70 of the VL as defined by Kabat; and b) amino acid residues 63, 65, 67, and 72 of the VL as defined by Kabat.

[0128] The above-mentioned plasmid DNAs containing the genes encoding MAGE-A4-zG and MAGE-A4-zG-s1 were used as template DNA. A PCR reaction solution containing the template DNA, a primer set (SEQ ID NOs: 77 and 78), and PrimeSTAR® Max Premix was prepared, and PCR reactions were performed. This resulted in PCR products with restriction enzyme recognition sites (hereinafter referred to as restriction sites) added to the 5' and 3' ends. DpnI was added to the resulting PCR product to fragment the template plasmid DNA. Using the DpnI-treated PCR product as a template, a PCR reaction solution containing a primer set for mutant generation and PrimeSTAR® Max Premix was prepared, and PCR reactions were performed. The resulting PCR product was mixed with restriction enzyme-treated plasmid DNA for viral vector production (empty vector), Ligation High ver. 2, and T4 polynucleotide kinase, and the mixture was incubated at 16°C for 1 hour to perform a ligation reaction.

[0129] Hereinafter, CARs having a single-chain antibody in which the amino acid residues shown in a) and b) above in MAGE-A4-zG have been replaced with arginine residues will be referred to as "MAGE-A4-zG-m1" and "MAGE-A4-zG-m2," respectively. Hereinafter, a CAR containing a single-chain antibody in which the amino acid residue shown in a) above in MAGE-A4-zG-s1 has been replaced with arginine residues will be referred to as "MAGE-A4-zG-m1-s1." Table 1 shows the correspondence between each mutant and the sequence number of the primer set used to create it. In the table, "F" indicates the sequence number of the nucleotide sequence of the forward primer, and "R" indicates the sequence number of the nucleotide sequence of the reverse primer. The same explanations for the tables also apply to the tables in the following examples.

[0130]

[0131] (2.2) Transformation, Plasmid DNA Extraction, and Sequencing Using the ligation solution and DH5α, E. coli transformants were obtained by the heat shock method. Single colonies on agar medium were cultured in ampicillin-containing LB liquid medium. Plasmid DNA was extracted from the resulting E. coli using a QIAprep Spin Miniprep Kit. Each of the resulting plasmid DNAs was sequenced. Sequencing confirmed that nucleic acid molecules encoding the MAGE-A4-zG and MAGE-A4-zG-s1 mutants had been obtained.

[0132] (3) Generation of T cells expressing MAGE-A4-zG, MAGE-A4-zG-s1, and their mutants. (3.1) Preparation of human lymphocytes. PBMCs were isolated from blood donated by healthy donors using Ficoll-Paque™ PLUS (GE Healthcare). The obtained PBMCs were used as human lymphocytes. The collection and analysis of human peripheral blood and other samples used in this study were conducted in accordance with the Declaration of Helsinki. All studies were conducted according to protocols approved by the Mie University School of Medicine Research Ethics Committee and with written consent from the subjects. Collected samples were encrypted to prevent personal identification and stored in an anti-theft refrigerator and liquid nitrogen tank. The subjects' personal information was anonymized, and strict precautions were taken to prevent the disclosure of personal privacy and genetic analysis results.

[0133] (3.2) Separation and Culture of PBMCs Anti-CD3 antibodies OKT-3 (eBioscience) and RetroNectin® (Takara Bio Inc.) were added to ACD-A solution (Terumo Corporation) to final concentrations of 5 μg / mL and 25 μg / mL, respectively. 400 μL of the resulting solution was placed in each well of a 12-well plate (Nunc) and left to stand overnight at 4°C. The solution in each well was removed, and the plate was washed three times with PBS to prepare a PBMC culture plate. AB serum (300 μL, Veritas) was added to lymphocyte medium GT-T503 (50 mL, Takara Bio Inc.), and human IL-2 (NIPRO) was further added to a final concentration of 300 IU / mL to prepare a PBMC culture medium. The PBMCs obtained in (3.1) above were added to the prepared medium at a concentration of 2.5 × 10 5 ~3.0×10 5 The cells were suspended at a concentration of 1000 cells / mL. 2 mL of the cell suspension was placed in each well of a PBMC culture plate and cultured at 37°C in a CO2 incubator for 3 days. On the third day, 1 mL of medium was replaced and the cells were cultured for an additional day.

[0134] (3.3) Retrovirus-mediated introduction of CAR genes into immune cells (i) Production of retroviruses using packaging cells Plat-A Plat-A Plat-A packaging cells were transfected with the plasmid DNA for producing each viral vector containing the genes encoding MAGE-A4-zG, MAGE-A4-zG-s1, and their mutants obtained in (2) above using FuGENE (registered trademark) (Promega) and cultured for two days. The culture supernatant was then collected to obtain a solution containing the viral vector (hereinafter also referred to as "viral solution").

[0135] (ii) Preparation of retrovirus dilution solution and washing solution Albuminar (registered trademark) 25% (CSL Behring) was added to a mixture of PBS (40 mL) and ACD-A solution (2.4 mL) to a final concentration of 2.5% to prepare a retrovirus dilution solution. Albuminar (registered trademark) 25% was added to PBS (39 mL) to a final concentration of 1.5% to prepare a retrovirus washing solution.

[0136] (iii) Preparation of retrovirus infection plate RetroNectin® (Takara Bio Inc.) was diluted with retrovirus diluent to a final concentration of 20 μg / mL. 250 μL of the RetroNectin® diluent was added to each well of a 24-well plate and allowed to stand overnight at 4°C. The solution in each well was removed, and the plate was washed twice with retrovirus washing solution. 1 mL of the virus solution prepared in (3.1) above was added to each well, and the plate was centrifuged at 2000 × g for 2 hours at 32°C to coat the inside of each well with retrovirus. The virus solution in each well was removed, and the plate was washed twice with retrovirus washing solution to prepare a retrovirus infection plate.

[0137] (iv) Generation of CAR-T cells by retroviral infection. PBMCs cultured in (3.2) above were collected and 1.3 × 10 5 The cells were suspended in PBMC culture medium at a concentration of 1.5 mL / mL. 1.5 mL of the cell suspension was placed in each well of a retroviral infection plate and centrifuged at 1000 × g for 10 minutes at 32°C. The plate was then placed in a CO2 incubator and cultured at 37°C. This yielded T cells (CAR-T cells) expressing various CARs, including MAGE-A4-zG, MAGE-A4-zG-s1, and their mutants. CAR-T cells were used for various assays 12 days after PBMC isolation.

[0138] (4) Confirmation of CAR-T cell cytotoxicity and cytokine secretion levels (4.1) Cytotoxicity assay The various CAR-T cells prepared in (3) above were used as effector cells, and their cytotoxicity was measured using the N-SPC® Non-RI Cytotoxicity Assay Kit (Techno Suzuta Co., Ltd.). Human melanoma cell lines SK-MEL-37 and NW-MEL-38 were used as target cells. These target cells were A2-positive, MAGE-A4-positive tumor cells.

[0139] The measurement principle of the above assay kit is as follows: When the BM-HT Reagent included in the assay kit is added to target cells, the BM-HT Reagent is hydrolyzed by intracellular esterases to produce HT chelate within the target cells. When target cells are killed by effector cells, the HT chelate leaks into the culture supernatant. When the Eu Solution included in the assay kit is added to the culture supernatant containing the HT chelate, an Eu / HT complex is formed. When this complex is excited with laser light, time-resolved fluorescence is generated. Because the leakage of HT chelate depends on the cytotoxic activity of effector cells, the cell-killing effect can be quantitatively measured by measuring time-resolved fluorescence.

[0140] 1 x 10 target cells 4 The cells were suspended in medium at a concentration of 1000 cells / mL, and BM-HT Reagent was added. Effector cells (CAR-T cells) and target cells were mixed at a cell number ratio of 3:1 or 1:1 and placed in each well of a 96-well plate for 2 hours at 37°C. For comparison, two control wells containing target cells without effector cells were prepared and cultured in the same manner. At 1.5 hours after the start of culture, the detergent provided with the kit was added to one of the control wells, and the fluorescence detected in this control well was taken as the maximum fluorescence. The other control well was left untreated, and the fluorescence detected in this control well was taken as the minimum fluorescence. Eu Solution (120 μL) was added to the culture supernatant (12 μL) of the co-cultured cells, and the cells were allowed to stand at room temperature for 15 minutes, after which time-resolved fluorescence was measured.

[0141] (4.2) Measurement of IFNγ (i) Sample Preparation CAR-T cells containing MAGE-A4-zG-s1 and CAR-T cells containing MAGE-A4-zG-m1-s1 prepared in (3) above were used as effector cells. For comparison, the control cells prepared in (3) above were also used. SK-MEL-37 and NW-MEL-38, A2-positive MAGE-A4-positive tumor cells, were used as target cells. The effector and target cells were each 1 × 10 5 The cells were suspended in medium at a concentration of 1000 cells / mL, placed in each well of a 96-well plate, and co-cultured for 12 hours at 37° C. The culture supernatant was collected and used as a sample.

[0142] (ii) Reagent Preparation: IFNγ in each sample was measured using Invitrogen® Human IFN gamma Uncoated ELISA with Plates (Thermo Fisher Scientific). The 10x Coating Buffer was diluted 10-fold with purified water to prepare a coating buffer. A capture antibody (anti-human IFNγ antibody) (48 μL) was added to the coating buffer (12 mL) to prepare a capture antibody dilution. 100 μL of this capture antibody dilution was added to each well of a 96-well flat-bottom Costar® 9018 plate (Corning). The plate was left overnight at 4°C. The solution in the wells was removed, and the wells were washed five times with 0.05% PBS-T (PBS, 0.05% Tween™-20). The 5x Assay Diluent was diluted 5-fold with purified water to prepare the Assay Diluent. 200 μL of Assay Diluent was added to each well and blocked for 1 hour at room temperature. The Assay Diluent was removed from the wells, and the wells were washed five times with 0.05% PBS-T to obtain plates coated with capture antibodies. Recombinant human IFN-γ was dissolved in Assay Diluent to a final concentration of 1000 pg / mL. This solution was diluted two-fold in seven steps to prepare standards. A detection antibody (biotin-labeled anti-human IFN-γ antibody) (48 μL) was added to 12 mL of coating buffer to prepare a diluted solution of the detection antibody. An enzyme (streptavidin-HRP) (48 μL) was added to 12 mL of coating buffer to prepare a diluted solution of the enzyme.

[0143] (iii) IFNγ Measurement. Samples and standards were added to plate wells and incubated for 2 hours at room temperature. The wells were washed five times with 0.05% PBS-T. 100 μL of the detection antibody dilution was added to each well. The plate was incubated at room temperature for 1 hour. The solution in the wells was removed, and the wells were washed five times with 0.05% PBS-T. 100 μL of the enzyme dilution was added to each well. The plate was incubated at room temperature for 30 minutes. The solution in the wells was removed, and the wells were washed seven times with 0.05% PBS-T. 100 μL of TMB substrate solution was added to each well. The plate was incubated in the dark at room temperature for 15 minutes. 50 μL of 0.18 M H2SO4 was added to each well to stop the reaction, and the absorbance at 450 nm was immediately measured using a Model 680 microplate reader (Bio-Rad). A standard curve was created from the standard measurement results. The absorbance of each sample was applied to the standard curve to obtain the IFNγ concentration in each sample. The obtained values ​​were used as the IFNγ secretion levels of each effector cell and control cell.

[0144] (2) Results Table 2 shows the results of cytotoxic activity and IFNγ production when SK-MEL-37 was used as the target cell and T cells containing MAGE-A4-zG or its mutants were used as the effector cells. Table 3 shows the results of cytotoxic activity and IFNγ production when SK-MEL-37 was used as the target cell and T cells containing MAGE-A4-zG-s1 or its mutants were used as the effector cells. In the table, "CAR type" refers to the code assigned to each CAR by the inventors and includes information about the CAR mutation. "WT" indicates an unmodified CAR (hereinafter, WT CAR is also referred to as "wild type"). Codes containing "R" in the "CAR type" column indicate CARs in which the amino acid residue shown in the "mutation site" in the table has been replaced with an arginine residue. In the "SEQ ID NO:" column, "NA" indicates the sequence number of the nucleotide sequence encoding the CAR, and "AA" indicates the sequence number of the amino acid sequence of the CAR. The numerical values ​​for "cytotoxic activity" are shown as a ratio when the proportion of target cells killed by CAR-T cells whose CAR type is WT is set to 1.00. "3:1" and "1:1" indicate the ratio of the number of effector cells to the number of target cells (hereinafter, this ratio is also referred to as the "cell mixing ratio"). The numerical values ​​for "IFNγ production" are shown as a ratio when the amount of IFNγ produced by CAR-T cells whose CAR type is WT is set to 1.00. The explanations for the tables also apply to the tables in the following Examples.

[0145] As shown in Table 2, the cytotoxic activity of each mutant was as follows, assuming that the cytotoxic activity of T cells containing MAGE-A4-zG was 1.00. The cytotoxic activity of T cells containing MAGE-A4-zG-m1 was 1.59 at a cell mixing ratio of 3:1 and 1.14 at a cell mixing ratio of 1:1. The cytotoxic activity of T cells containing MAGE-A4-zG-m2 was 1.41 at a cell mixing ratio of 3:1 and 1.60 at a cell mixing ratio of 1:1. The cytotoxic activity of T cells containing MAGE-A4-zG-m1 or MAGE-A4-zG-m2 was higher than that of T cells containing MAGE-A4-zG. On the other hand, the IFNγ production of each mutant, assuming that the IFNγ production of T cells containing MAGE-A4-zG was 1.00, was as follows: The IFNγ production of T cells containing MAGE-A4-zG-m1 was 0.67, and that of T cells containing MAGE-A4-zG-m2 was 0.47. The IFNγ production of T cells containing MAGE-A4-zG-m1 or MAGE-A4-zG-m2 was lower than that of T cells containing MAGE-A4-zG.

[0146] As shown in Table 3, the cytotoxic activity of T cells containing MAGE-A4-zG-s1 was 2.34 at a cell mixing ratio of 3:1 and 2.25 at a cell mixing ratio of 1:1, when the cytotoxic activity of T cells containing MAGE-A4-zG-s1 was set to 1.00. On the other hand, the IFNγ production of T cells containing MAGE-A4-zG-m1-s1 was 0.66, when the IFNγ production of T cells containing MAGE-A4-zG-s1 was set to 1.00. The cytotoxic activity of T cells containing MAGE-A4-zG-m1-s1 was higher than that of T cells containing MAGE-A4-zG-s1, but the IFNγ production was lower than that of T cells containing MAGE-A4-zG-s1.

[0147]

[0148]

[0149] Table 4 shows the cytotoxic activity and IFNγ production when NW-MEL-38 cells were used as target cells and T cells containing MAGE-A4-zG-s1 or MAGE-A4-zG-m1-s1 were used as effector cells. As shown in Table 4, when the cytotoxic activity of T cells containing MAGE-A4-zG-s1 was set to 1.00, the cytotoxic activity of T cells containing MAGE-A4-zG-m1-s1 was 1.35 at a cell mixing ratio of 3:1 and 1.65 at a cell mixing ratio of 1:1. The cytotoxic activity of T cells containing MAGE-A4-zG-m1-s1 was higher than that of T cells containing MAGE-A4-zG-s1. On the other hand, when the IFNγ production of T cells containing MAGE-A4-zG-s1 was set to 1.00, the IFNγ production of T cells containing MAGE-A4-zG-m1-s1 was 0.40. The T cells containing MAGE-A4-zG-m1-s1 were lower than those containing MAGE-A4-zG-s1.

[0150]

[0151] The immune cells in this example were suggested to be CAR-T cells with suppressed cytokine production. Furthermore, the CAR-T cells had improved cytotoxicity.

[0152] Example 2: Generation of CAR-T cells that bind to the MAGE-A4 / HLA-A2 complex and confirmation of their effects (2)

[0153] (1) Obtaining a nucleic acid molecule encoding a mutant of MAGE-A4-zG Using the nucleic acid molecule encoding MAGE-A4-zG prepared in Example 1 as template DNA, a nucleic acid molecule encoding a mutant different from that of Example 1 was prepared by PCR. Specifically, the procedure was as follows.

[0154] (1.1) Primer design, PCR, and ligation reaction A primer set for obtaining a polynucleotide encoding a CAR comprising a single-chain antibody in which the amino acid residues in MAGE-A4-zG shown in a), c), e), f), or h) below have been substituted with arginine residues was designed based on the nucleotide sequence of SEQ ID NO: 3. Furthermore, a primer set for obtaining a polynucleotide encoding a CAR comprising a single-chain antibody in MAGE-A4-zG in which the amino acid residues in b), d), or g) below have been substituted with lysine residues was designed based on the nucleotide sequence of SEQ ID NO: 3. PCR and ligation reactions were performed using these primer sets and template DNA in the same manner as in Example 1.

[0155] a) amino acid residues 63, 65, and 72 of the VL as defined by Kabat; b) amino acid residues 63, 67, and 70 of the VL as defined by Kabat; c) amino acid residues 63, 70, and 72 of the VL as defined by Kabat; d) amino acid residues 65, 67, and 70 of the VL as defined by Kabat; e) amino acid residues 65, 70, and 72 of the VL as defined by Kabat; f) amino acid residues 67, 70, and 72 of the VL as defined by Kabat; g) amino acid residues 63, 65, 67, 70, and 72 of the VL as defined by Kabat; and h) amino acid residues 60, 74, and 76 of the VL as defined by Kabat.

[0156] (1.2) Transformation, extraction of plasmid DNA, and sequencing As in Example 1, E. coli transformants were obtained using the solution after the ligation reaction and DH5α. Then, as in Example 1, plasmid DNA was extracted from E. coli cultured in liquid medium and sequenced. The sequencing results confirmed that a nucleic acid molecule encoding a MAGE-A4-zG mutant had been obtained.

[0157] Hereinafter, CARs having single-chain antibodies in which the amino acid residues in MAGE-A4-zG shown in a), c), e), f), or h) above have been replaced with arginine residues will be referred to as "MAGE-A4-zG-m5," "MAGE-A4-zG-m7," "MAGE-A4-zG-m9," "MAGE-A4-zG-m10," and "MAGE-A4-zG-m12," respectively. Furthermore, CARs having single-chain antibodies in which the amino acid residues in b), d), or g) above have been replaced with lysine residues will be referred to as "MAGE-A4-zG-Km6," "MAGE-A4-zG-Km8," and "MAGE-A4-zG-Km11," respectively. Tables 5 and 6 show the correspondence between each mutant and the sequence number of the primer set used to create it. In Table 6, "1st" indicates the primer set for substituting the 74th and 76th amino acid residues of VL, and "2nd" indicates the primer set for substituting the 60th amino acid residue of VL.

[0158]

[0159]

[0160] (2) Preparation of CAR-T cells expressing MAGE-A4-zG and its mutants As in Example 1, genes encoding MAGE-A4-zG and its mutants were introduced into PBMCs isolated from the blood of healthy donors and cultured using retroviruses to obtain CAR-T cells.

[0161] (3) Confirmation of CAR-T Cell Cytotoxicity and Cytokine Secretion Amounts The cytotoxic activity and IFNγ production of each CAR-T cell were measured in the same manner as in Example 1. SK-MEL-37 cells were used as target cells. Measurements of MAGE-A4-zG-m5, MAGE-A4-zG-Km6, MAGE-A4-zG-m7, MAGE-A4-zG-Km8, MAGE-A4-zG-m9, and MAGE-A4-zG-m10 were performed on separate days from measurements of MAGE-A4-zG-Km11 ​​and MAGE-A4-zG-m12. The results are shown in Tables 7 and 8. In the "CAR type" column in the table, a symbol containing "K" indicates a CAR in which the amino acid residue indicated in the "mutation site" column in the table was replaced with a lysine residue. This also applies to the tables in the following examples.

[0162]

[0163]

[0164] As shown in Table 7, when the cytotoxic activity of T cells containing MAGE-A4-zG was set to 1.00, the cytotoxic activity of each mutant was as follows. The cytotoxic activity of T cells containing MAGE-A4-zG-m5 was 1.62 at a cell mixing ratio of 3:1 and 1.05 at a cell mixing ratio of 1:1. The cytotoxic activity of T cells containing MAGE-A4-zG-Km6 was 1.52 at a cell mixing ratio of 3:1 and 1.61 at a cell mixing ratio of 1:1. The cytotoxic activity of T cells containing MAGE-A4-zG-m7 was 1.77 at a cell mixing ratio of 3:1 and 0.99 at a cell mixing ratio of 1:1. The cytotoxic activity of T cells containing MAGE-A4-zG-Km8 was 1.47 at a cell mixing ratio of 3:1 and 1.36 at a cell mixing ratio of 1:1. The cytotoxic activity of T cells containing MAGE-A4-zG-m9 was 1.61 at a cell mixing ratio of 3:1 and 1.23 at a cell mixing ratio of 1:1. The cytotoxic activity of T cells containing MAGE-A4-zG-m10 was 1.77 at a cell mixing ratio of 3:1 and 0.99 at a cell mixing ratio of 1:1. The IFNγ production of T cells containing MAGE-A4-zG was 1.00, and the IFNγ production of each mutant was as follows: T cells containing MAGE-A4-zG-m5, 0.43; T cells containing MAGE-A4-zG-Km6, 0.93; T cells containing MAGE-A4-zG-m7, 0.22; T cells containing MAGE-A4-zG-Km8, 0.91; T cells containing MAGE-A4-zG-m9, 0.29; and T cells containing MAGE-A4-zG-m10, 0.32.

[0165] As shown in Table 8, the cytotoxic activity of each mutant was as follows, assuming that the cytotoxic activity of T cells containing MAGE-A4-zG was 1.00. The cytotoxic activity of T cells containing MAGE-A4-zG-Km11 ​​was 2.33 at a cell mixing ratio of 3:1 and 1.68 at a cell mixing ratio of 1:1. The cytotoxic activity of T cells containing MAGE-A4-zG-m12 was 1.23 at a cell mixing ratio of 3:1 and 0.90 at a cell mixing ratio of 1:1. On the other hand, the IFNγ production of each mutant was as follows, assuming that the IFNγ production of T cells containing MAGE-A4-zG was 1.00. The IFNγ production of T cells containing MAGE-A4-zG-Km11 ​​was 0.89, and the IFNγ production of T cells containing MAGE-A4-zG-m12 was 0.40.

[0166] The immune cells in this example were suggested to be CAR-T cells with suppressed cytokine production. Furthermore, many clones had improved cytocidal effects. MAGE-A4-zG-m12 showed a slightly reduced cytocidal effect compared to the wild-type, but significantly suppressed cytokine production. It was suggested that increasing the dose of this clone could improve or maintain the same cytocidal effect as the wild-type, while suppressing cytokine production more than the wild-type.

[0167] Example 3: Generation of CAR-T cells that bind to the PRAME / HLA-A24 complex and confirmation of their effects

[0168] (1) Obtaining a nucleic acid molecule encoding a template CAR Plasmid DNA for constructing a viral vector containing a gene encoding a CAR with a single-chain antibody that binds to the PRAME / HLA-A24 complex was obtained in the same manner as in Example 1. The complex recognized by the single-chain antibody was a complex of PRAME and HLA-A*24:02. Hereinafter, the CAR encoded by the gene in the plasmid DNA will be referred to as "PRAME-zG." The amino acid sequence of the PRAME-derived peptide in the PRAME / HLA-A24 complex was LYVDSLFFL (SEQ ID NO: 2). The above plasmid DNA was used as a template to prepare a nucleic acid molecule encoding a PRAME-zG mutant by PCR.

[0169] The structure of the gene encoding PRAME-zG itself was the same as that of the gene encoding MAGE-A4-zG in Example 1. That is, the gene encoding PRAME-zG contained, from the 5' end, a leader sequence, a nucleotide sequence encoding VH, a nucleotide sequence encoding a linker, a nucleotide sequence encoding VL, a nucleotide sequence encoding CL, a nucleotide sequence encoding CD28TM, a nucleotide sequence encoding CD3ζ, and a nucleotide sequence encoding GITRICD, linked in this order (see Figure 2A). PRAME-zG was a CAR containing a single-chain antibody consisting of VH, VL, and a linker connecting them, and CL as extracellular domains, CD28TM as a transmembrane domain, and CD3ζ and GITRICD as intracellular domains.

[0170] (2) Obtaining a nucleic acid molecule encoding a PRAME-zG mutant The above-mentioned plasmid DNA was used as a template. A PCR reaction solution containing the template DNA, a primer set (SEQ ID NOs: 101 and 102), and PrimeSTAR® Max Premix was prepared, and a PCR reaction was performed. This resulted in a PCR product with restriction sites added to the 5' and 3' ends. DpnI was added to the resulting PCR product to fragment the template plasmid DNA. Using the DpnI-treated PCR product as a template, a nucleic acid molecule encoding a PRAME-zG mutant was prepared by PCR. Specifically, the procedure was as follows.

[0171] (2.1) Primer design, PCR, and ligation reaction A primer set for obtaining a polynucleotide encoding a CAR comprising a single-chain antibody in which the amino acid residue shown in a) or b) below in PRAME-zG has been substituted with an arginine residue was designed based on the nucleotide sequence of SEQ ID NO: 29. Furthermore, a primer set for obtaining a polynucleotide encoding a CAR comprising a single-chain antibody in which the amino acid residue shown in c) below in PRAME-zG has been substituted with a lysine residue was designed based on the nucleotide sequence of SEQ ID NO: 29. PCR and ligation reactions were performed in the same manner as in Example 1 using these primer sets for producing mutants and the above-mentioned DpnI-treated PCR product.

[0172] a) amino acid residues 63, 65, and 72 of the VL as defined by Kabat; b) amino acid residues 63, 70, and 72 of the VL as defined by Kabat; and c) amino acid residues 65, 70, and 72 of the VL as defined by Kabat.

[0173] (2.2) Transformation, extraction of plasmid DNA, and sequencing As in Example 1, an E. coli transformant was obtained using the solution after the ligation reaction and DH5α. Then, as in Example 1, plasmid DNA was extracted from the E. coli cultured in liquid medium and sequenced. The sequencing results confirmed that a nucleic acid molecule encoding a PRAME-zG mutant had been obtained.

[0174] Hereinafter, CARs having a single-chain antibody in which the amino acid residues shown in a) or b) above in PRAME-zG have been replaced with arginine residues will be referred to as "PRAME-zG-m1" and "PRAME-zG-m2," respectively. Furthermore, a CAR having a single-chain antibody in which the amino acid residue shown in c) above has been replaced with a lysine residue will be referred to as "PRAME-zG-Km3." Table 9 shows the correspondence between each mutant and the sequence number of the primer set used to create it.

[0175]

[0176] (3) Preparation of CAR-T cells expressing PRAME-zG and its mutants As in Example 1, genes encoding PRAME-zG and its mutants were introduced using a retrovirus into PBMCs isolated from the blood of a healthy donor and cultured to obtain each type of CAR-T cell.

[0177] (4) Confirmation of CAR-T cell cytotoxicity and cytokine secretion levels The cytotoxic activity and IFNγ production of each CAR-T cell were measured in the same manner as in Example 1. The human melanoma cell line SK-MEL-124 was used as the target cell. SK-MEL-124 was a PRAME-positive tumor cell. Measurements of PRAME-zG-m1 and PRAME-zG-m2 were performed on different days from the measurement of PRAME-zG-Km3. The results are shown in Tables 10 and 11.

[0178]

[0179]

[0180] As shown in Table 10, the cytotoxic activity of each mutant was as follows, assuming that the cytotoxic activity of T cells containing PRAME-zG was 1.00. The cytotoxic activity of T cells containing PRAME-zG-m1 was 2.96 at a cell mixing ratio of 3:1 and 2.37 at a cell mixing ratio of 1:1. The cytotoxic activity of T cells containing PRAME-zG-m2 was 1.84 at a cell mixing ratio of 3:1 and 1.54 at a cell mixing ratio of 1:1. On the other hand, the IFNγ production level of each mutant, assuming that the IFNγ production level of T cells containing PRAME-zG was 1.00, was as follows: The IFNγ production level of T cells containing PRAME-zG-m1 was 0.49, and the IFNγ production level of T cells containing PRAME-zG-m2 was 0.56.

[0181] As shown in Table 11, when the cytotoxic activity of T cells containing PRAME-zG was defined as 1.00, the cytotoxic activity of T cells containing PRAME-zG-Km3 was 1.16 when the cell mixing ratio was 3:1 and 1.71 when the cell mixing ratio was 1:1. On the other hand, when the IFNγ production level of T cells containing PRAME-zG was defined as 1.00, the IFNγ production level of T cells containing PRAME-zG-Km3 was 0.87.

[0182] The immune cells in this example were suggested to be CAR-T cells with suppressed cytokine production. Furthermore, the CAR-T cells had improved cytotoxicity.

[0183] Example 4: Generation of CAR-T cells that bind to CD19 and confirmation of their effects

[0184] (1) Obtaining a nucleic acid molecule encoding a template CAR A polynucleotide (SEQ ID NO: 186) encoding a single-chain antibody that binds to CD19 (hereinafter also referred to as "anti-CD19-WT") was obtained by gene synthesis. In addition, polynucleotides (SEQ ID NOs: 187, 188, and 190) encoding single-chain antibodies in which any of the amino acid residues in anti-CD19-WT shown in a) to c) below were substituted with arginine residues were synthesized as variants of anti-CD19-WT. Hereinafter, the single-chain antibodies in which the amino acid residues shown in a) to c) below have been modified will be referred to as "anti-CD19-m1," "anti-CD19-m2," and "anti-CD19-m4," respectively.

[0185] a) amino acid residues 63, 65, 67, and 70 of the VL defined by Kabat; b) amino acid residues 63, 65, 67, and 72 of the VL defined by Kabat; and c) amino acid residues 65, 67, 70, and 72 of the VL defined by Kabat.

[0186] (2) Obtaining Nucleic Acid Molecules Encoding CD19-28z Mutants (2.1) Preparation of the First and Second Inserts Using the polynucleotides encoding anti-CD19-WT, anti-CD19-m1, anti-CD19-m2, and anti-CD19-m4 as templates, a PCR reaction mixture containing a primer set (SEQ ID NOs: 109 and 110) and PrimeSTAR® Max Premix was prepared, and PCR was performed. This resulted in the first inserts encoding anti-CD19-WT, anti-CD19-m1, anti-CD19-m2, and anti-CD19-m4. A polynucleotide (SEQ ID NO: 111) encoding the hinge domain, transmembrane domain, and intracellular domain of CD28 was obtained by gene synthesis. Using this polynucleotide as a template, a PCR reaction mixture containing a primer set (SEQ ID NOs: 112 and 113) and PrimeSTAR® Max Premix was prepared, and PCR was performed. This resulted in a second insert encoding a region consisting of the hinge domain, transmembrane domain, and intracellular domain of CD28.

[0187] (2.2) Preparation of Linearized Vector and Fusion with Insert: Using a viral vector preparation plasmid DNA (SEQ ID NO: 114) containing a polynucleotide encoding the transmembrane and intracellular domains of CD28 as a template, a PCR reaction solution containing a primer set (SEQ ID NOs: 115 and 116) and PrimeSTAR® Max Premix was prepared, and PCR was performed. DpnI was added to the resulting PCR product to fragment the template plasmid DNA. This resulted in a linearized vector. This linearized vector was fused with the first and second inserts described above using In-Fusion® HD Enzyme Premix (Takara Bio Inc.). This resulted in plasmid DNA containing genes encoding CARs containing anti-CD19-WT, anti-CD19-m1, anti-CD19-m2, and anti-CD19-m4, respectively. Hereinafter, CARs carrying the single-chain antibodies anti-CD19-WT, anti-CD19-m1, anti-CD19-m2, and anti-CD19-m4 will be referred to as "CD19-28z," "CD19-28z-m1," "CD19-28z-m2," and "CD19-28z-m4," respectively.

[0188] As an example, the structure of the gene encoding CD19-28z is described below. Referring to Figure 2B, the gene encoding CD19-28z contains, from the 5' end, a leader sequence, a nucleotide sequence encoding a VL, a nucleotide sequence encoding a linker, a nucleotide sequence encoding a VH, a nucleotide sequence encoding a CD28 hinge domain (CD28 hinge), a nucleotide sequence encoding a CD28TM, a nucleotide sequence encoding a CD28 intracellular domain (CD28ICD), and a nucleotide sequence encoding CD3ζ, all linked in this order. CD19-28z was a CAR comprising a single-chain antibody consisting of a VH, a VL, and a linker connecting them as the extracellular domain, and the CD28 hinge domain, CD28TM as the transmembrane domain, and CD28ICD and CD3ζ as the intracellular domain.

[0189] (2.3) Primer design, PCR, and ligation reaction A primer set for obtaining a polynucleotide encoding a CAR comprising a single-chain antibody in which any of the amino acid residues shown in d), e), g) to k), and m) to q) below in CD19-28z has been substituted with an arginine residue was designed based on the nucleotide sequence of SEQ ID NO: 37. Furthermore, a primer set for obtaining a polynucleotide encoding a CAR comprising a single-chain antibody in which the amino acid residue shown in d), f), l), or m) below in CD19-28z has been substituted with a lysine residue was designed based on the nucleotide sequence of SEQ ID NO: 37. PCR and ligation reactions were performed using these primer sets and template DNA in the same manner as in Example 1.

[0190] d) amino acid residues 63, 65, 70 and 72 of the VL as defined by Kabat; e) amino acid residues 60, 63 and 65 of the VL as defined by Kabat; f) amino acid residues 63, 65 and 70 of the VL as defined by Kabat; g) amino acid residues 63, 65 and 72 of the VL as defined by Kabat; h) amino acid residues 63, 67 and 70 of the VL as defined by Kabat; i) amino acid residues 63, 70 and 72 of the VL as defined by Kabat; j) amino acid residues 65, 67 and 70 of the VL as defined by Kabat; k) amino acid residues 65, 67 and 72 of the VL as defined by Kabat; l) amino acid residues 65, 67, 70 and 72 of the VL as defined by Kabat; m) amino acid residues 63, 65, 67, 70 and 72 of the VL as defined by Kabat; n) amino acid residues 70, 72 and 74 of the VL as defined by Kabat. o) amino acid residues 60, 63, and 76 of the VL as defined by Kabat; p) amino acid residues 60, 74, and 76 of the VL as defined by Kabat; and q) amino acid residues 77, 79, and 81 of the VL as defined by Kabat.

[0191] (2.2) Transformation, Extraction of Plasmid DNA, and Sequencing As in Example 1, E. coli transformants were obtained using the solution after the ligation reaction and DH5α. Then, as in Example 1, plasmid DNA was extracted from E. coli cultured in liquid medium and sequenced. As a result of sequencing, it was confirmed that a nucleic acid molecule encoding a CD19-28z mutant had been obtained.

[0192] Hereinafter, CARs having a single-chain antibody in which any of the amino acid residues shown in d), e), g) to k), and m) to q) above in CD19-28z have been replaced with arginine residues will be referred to as "CD19-28z-m3," "CD19-28z-m5," "CD19-28z-m7," "CD19-28z-m8," "CD19-28z-m10," "CD19-28z-m11," "CD19-28z-m12," "CD19-28z-m16," "CD19-28z-m17," "CD19-28z-m18," "CD19-28z-m19," and "CD19-28z-m20," respectively. Furthermore, CARs containing single-chain antibodies in which the amino acid residues shown in d), f), l), or m) above have been replaced with lysine residues are referred to as "CD19-28z-Km13," "CD19-28z-Km6," "CD19-28z-Km14," and "CD19-28z-Km15," respectively. Tables 12 and 13 show the correspondence between each mutant and the sequence numbers of the primer sets used to create it. For CD19-28z-m18 in Table 13, "first round" refers to the primer set used to replace the 60th and 63rd amino acid residues of the VL, and "second round" refers to the primer set used to replace the 76th amino acid residue of the VL. For CD19-28z-m19, "first round" refers to the primer set used to replace the 74th and 76th amino acid residues of the VL, and "second round" refers to the primer set used to replace the 60th amino acid residue of the VL.

[0193]

[0194]

[0195] (3) Generation of CAR-T cells expressing CD19-28z and its variants As in Example 1, genes encoding CD19-28z and its variants were introduced using retroviruses into PBMCs isolated from the blood of healthy donors and cultured to obtain CAR-T cells.

[0196] (4) Confirmation of CAR-T cell cytotoxicity and cytokine secretion levels The cytotoxic activity and IFNγ production of each CAR-T cell were measured in the same manner as in Example 1. In some experiments, the ratio of effector cells to target cells was 2:1 (see Table 16). The target cells used were the human precursor B cell leukemia cell line Nalm-6. Nalm-6 were CD19-positive tumor cells. The measurement results are shown in Tables 14 to 17. The results in each table were based on measurements performed on different days.

[0197]

[0198]

[0199]

[0200]

[0201] As shown in Table 14, the cytotoxic activity of each mutant was as follows, assuming that the cytotoxic activity of T cells containing CD19-28z was 1.00. The cytotoxic activity of T cells containing CD19-28z-m1 was 1.03 at a cell mixing ratio of 3:1 and 1.30 at a cell mixing ratio of 1:1. The cytotoxic activity of T cells containing CD19-28z-m2 was 0.90 at a cell mixing ratio of 3:1 and 1.19 at a cell mixing ratio of 1:1. The cytotoxic activity of T cells containing CD19-28z-m3 was 0.91 at a cell mixing ratio of 3:1 and 1.26 at a cell mixing ratio of 1:1. On the other hand, the IFNγ production of each mutant, assuming that the IFNγ production of T cells containing CD19-28z was 1.00, was as follows. The IFNγ production of T cells containing CD19-28z-m1 was 0.25, that of T cells containing CD19-28z-m2 was 0.24, and that of T cells containing CD19-28z-m3 was 0.26.

[0202] As shown in Table 15, the cytotoxic activity of each mutant was as follows, assuming that the cytotoxic activity of T cells containing CD19-28z was 1.00. The cytotoxic activity of T cells containing CD19-28z-m1 was 1.00 at a cell mixing ratio of 3:1 and 1.56 at a cell mixing ratio of 1:1. The cytotoxic activity of T cells containing CD19-28z-m3 was 1.09 at a cell mixing ratio of 3:1 and 1.37 at a cell mixing ratio of 1:1. The cytotoxic activity of T cells containing CD19-28z-m4 was 1.17 at a cell mixing ratio of 3:1 and 1.49 at a cell mixing ratio of 1:1. On the other hand, the IFNγ production of each mutant, assuming that the IFNγ production of T cells containing CD19-28z was 1.00, was as follows: The IFNγ production of T cells containing CD19-28z-m1 was 0.31, that of T cells containing CD19-28z-m3 was 0.27, and that of T cells containing CD19-28z-m4 was 0.29.

[0203] As shown in Table 16, when the cytotoxic activity of T cells containing CD19-28z was set to 1.00, the cytotoxic activity of each mutant was as follows. The cytotoxic activity of T cells containing CD19-28z-m5 was 1.09 at a cell mixing ratio of 2:1 and 0.95 at a cell mixing ratio of 1:1. The cytotoxic activity of T cells containing CD19-28z-Km6 was 1.50 at a cell mixing ratio of 2:1 and 1.13 at a cell mixing ratio of 1:1. The cytotoxic activity of T cells containing CD19-28z-m7 was 1.58 at a cell mixing ratio of 2:1 and 1.11 at a cell mixing ratio of 1:1. The cytotoxic activity of T cells containing CD19-28z-m8 was 1.61 at a cell mixing ratio of 2:1 and 1.19 at a cell mixing ratio of 1:1. The cytotoxic activity of T cells containing CD19-28z-m10 was 1.59 at a cell mixing ratio of 2:1 and 0.98 at a cell mixing ratio of 1:1. The cytotoxic activity of T cells containing CD19-28z-m11 was 1.40 at a cell mixing ratio of 2:1 and 1.06 at a cell mixing ratio of 1:1. The cytotoxic activity of T cells containing CD19-28z-m12 was 1.47 at a cell mixing ratio of 2:1 and 1.39 at a cell mixing ratio of 1:1. The cytotoxic activity of T cells containing CD19-28z-Km13 was 1.68 at a cell mixing ratio of 2:1 and 1.37 at a cell mixing ratio of 1:1. The cytotoxic activity of T cells containing CD19-28z-Km14 was 1.72 at a cell mixing ratio of 2:1 and 1.34 at a cell mixing ratio of 1:1. The cytotoxic activity of T cells containing CD19-28z-Km15 was 1.29 at a cell mixing ratio of 2:1 and 1.23 at a cell mixing ratio of 1:1. The cytotoxic activity of T cells containing CD19-28z-m16 was 1.93 at a cell mixing ratio of 2:1 and 1.32 at a cell mixing ratio of 1:1. On the other hand, when the IFNγ production of T cells containing CD19-28z was set at 1.00, the IFNγ production of each mutant was as follows:The IFNγ production of T cells containing CD19-28z-m5 was 0.34, the IFNγ production of T cells containing CD19-28z-m6 was 0.20, the IFNγ production of T cells containing CD19-28z-m7 was 0.13, the IFNγ production of T cells containing CD19-28z-m8 was 0.08, the IFNγ production of T cells containing CD19-28z-m10 was 0.21, the IFNγ production of T cells containing CD19-28z-m11 was 0.17, the IFNγ production of T cells containing CD19-28z-m12 was 0.16, the IFNγ production of T cells containing CD19-28z-Km13 was 0.13, the IFNγ production of T cells containing CD19-28z-Km14 was 0.13. The IFNγ production level of T cells containing CD19-28z-Km15 was 0.14, and the IFNγ production level of T cells containing CD19-28z-m16 was 0.15.

[0204] As shown in Table 17, the cytotoxic activity of each mutant was determined as follows, assuming that the cytotoxic activity of CD19-28z-containing T cells was 1.00: the cytotoxic activity of CD19-28z-m17-containing T cells was 1.43 at a cell mixing ratio of 3:1 and 2.43 at a cell mixing ratio of 1:1; the cytotoxic activity of CD19-28z-m18-containing T cells was 1.27 at a cell mixing ratio of 3:1 and 1.42 at a cell mixing ratio of 1:1; the cytotoxic activity of CD19-28z-m19-containing T cells was 1.38 at a cell mixing ratio of 3:1 and 1.76 at a cell mixing ratio of 1:1; and the cytotoxic activity of CD19-28z-m20-containing T cells was 1.57 at a cell mixing ratio of 3:1 and 1.90 at a cell mixing ratio of 1:1. On the other hand, when the IFNγ production of T cells containing CD19-28z was set at 1.00, the IFNγ production of each mutant was as follows: IFNγ production of T cells containing CD19-28z-m17 was 0.34, IFNγ production of T cells containing CD19-28z-m18 was 0.28, IFNγ production of T cells containing CD19-28z-m19 was 0.68, and IFNγ production of T cells containing CD19-28z-m20 was 0.39.

[0205] As shown in Examples 1 to 4, the IFNγ production levels of the immune cells of the present invention were 0.08 to 0.93, with the wild-type being set at 1.00, and all were lower than those of the wild-type. The cytotoxic activity of the CAR-T cells was 0.90 to 2.96, with the wild-type being set at 1.00, and the cytotoxic activity was maintained or improved in many clones compared to the wild-type.

[0206] Example 5: In vivo cancer treatment model experiment using CAR-T cell administration (1) Materials and methods (1.1) Experimental animals NOG mice (NOD / Shi-scid, IL-2RγKO Jic) were purchased from CLEA Japan, Inc. Female mice aged 7 to 8 weeks were used in the experiment.

[0207] (1.2) Transplantation of human tumors and administration of CAR-T cells NW-MEL-38 were used as tumor cells. CAR-T cells containing MAGE-A4-zG-s1 and CAR-T cells containing MAGE-A4-zG-m1-s1 prepared in Example 1 were used as effector cells. NW-MEL-38 cells were implanted at 5 × 10 in the lateral side of the body of 12 NOG mice. 6 Four days after tumor cell implantation, PBS, T cells containing MAGE-A4-zG-s1, or CAR-T cells containing MAGE-A4-zG-m1-s1 were subcutaneously injected at 5 × 10 cells / mouse. 6 The tumors were infused via the tail vein at a dose of 1000 cells / mouse. Each group received PBS, CAR-T cells containing MAGE-A4-zG-s1, and CAR-T cells containing MAGE-A4-zG-m1-s1, with n=4. Tumor diameters were measured every 2 or 3 days.

[0208] (2) Results Figure 3 shows the mean tumor area for each group. As can be seen from Figure 3, tumor size was reduced in the groups administered with CAR-T cells containing MAGE-A4-zG-s1 and those containing MAGE-A4-zG-m1-s1 compared with the group administered with PBS. Furthermore, tumor size was reduced in the group administered with CAR-T cells containing MAGE-A4-zG-m1-s1 compared with the group administered with CAR-T cells containing MAGE-A4-zG-s1. These results suggest that the antitumor activity of immune cells containing CARs can be improved by substituting basic amino acid residues for at least three amino acid residues in the light chain FR3 of the single-chain antibody contained in the CAR.

Claims

1. An immune cell comprising a chimeric antigen receptor comprising an extracellular domain, a transmembrane domain, and an intracellular domain, the extracellular domain comprises an antigen-binding region comprising a light chain variable region and a heavy chain variable region; at least three amino acid residues in framework region 3 of the light chain variable region as defined by the Kabat method are basic amino acid residues; the at least three amino acid residues include at least three selected from the group consisting of the 60th, 63rd, 65th, 67th, 70th, 72nd, 74th, 76th, 77th, 79th, and 81st amino acid residues of the light chain variable region; The immune cell has suppressed cytokine production compared to an immune cell comprising a chimeric antigen receptor having an antigen-binding region in which the at least three amino acid residues are neutral amino acid residues or acidic amino acid residues. An immune cell containing a chimeric antigen receptor.

2. The immune cell according to claim 1, wherein 3 to 5 amino acid residues selected from the group consisting of the 60th, 63rd, 65th, 67th, 70th, 72nd, 74th, 76th, 77th, 79th, and 81st amino acid residues in the light chain variable region as defined by the Kabat method are basic amino acid residues.

3. The immune cell according to claim 1, wherein the antigen-binding region comprises a single-chain antibody, and the single-chain antibody is a single-chain antibody that binds to a complex of a MAGE-A4-derived peptide and HLA-A2, a complex of a PRAME-derived peptide and HLA-A24, CD19, BCMA, or CEA.

4. The immune cell according to claim 1, wherein the transmembrane domain comprises a transmembrane region of any one protein selected from the group consisting of a T cell receptor α chain, a T cell receptor β chain, CD3ε, CD3ζ, CD4, CD5, CD8, CD9, CD16, CD22, CD28, CD33, CD37, CD45, CD64, CD80, CD86, CD134, CD154, 4-1BB, ICOS, and GITR.

5. The immune cell of claim 1, wherein the intracellular domain comprises a signal transduction domain of at least one protein selected from the group consisting of CD3ζ, CD3γ, CD3δ, CD3ε, CD5, CD22, CD66d, CD79a, CD79b, FcRγ, and FcRβ.

6. the intracellular domain further comprises a costimulatory domain; The immune cell according to claim 5, wherein the costimulatory domain is a costimulatory domain of at least one protein selected from the group consisting of 4-1BB, CD28, GITR, CD2, CD5, CD8, CD9, CD16, CD22, CD33, CD37, CD64, CD80, CD86, CD134, CD154, and ICOS.

7. The immune cell of claim 1 , further comprising a hinge domain between the antigen-binding region and the transmembrane domain.

8. A pharmaceutical composition comprising the immune cell according to any one of claims 1 to 7.

9. The pharmaceutical composition according to claim 8, which is for treating a malignant tumor.

10. A method for producing immune cells comprising a chimeric antigen receptor, comprising: the production method comprises introducing a nucleic acid molecule having a nucleotide sequence encoding a chimeric antigen receptor into an immune cell and allowing the immune cell to express the chimeric antigen receptor; the nucleic acid molecule comprises a segment encoding an extracellular domain, a segment encoding a transmembrane domain, and a segment encoding an intracellular domain; the segment encoding the extracellular domain comprises a nucleotide sequence encoding an antigen-binding region comprising a light chain variable region and a heavy chain variable region; at least three codons in the nucleotide sequence encoding framework region 3 of the light chain variable region defined by the Kabat method are codons encoding basic amino acid residues; the at least three codons include at least three selected from the group consisting of a codon encoding the 60th amino acid residue, a codon encoding the 63rd amino acid residue, a codon encoding the 65th amino acid residue, a codon encoding the 67th amino acid residue, a codon encoding the 70th amino acid residue, a codon encoding the 72nd amino acid residue, a codon encoding the 74th amino acid residue, a codon encoding the 76th amino acid residue, a codon encoding the 77th amino acid residue, a codon encoding the 79th amino acid residue, and a codon encoding the 81st amino acid residue of the light chain variable region; The immune cell has suppressed cytokine production compared to an immune cell comprising a chimeric antigen receptor having an antigen-binding region in which the amino acid residues encoded by the at least three codons are neutral amino acid residues or acidic amino acid residues. A method for producing immune cells containing a chimeric antigen receptor.

11. The method according to claim 10, wherein in the nucleic acid molecule, 3 to 5 codons selected from the group consisting of a codon encoding the 60th amino acid residue, a codon encoding the 63rd amino acid residue, a codon encoding the 65th amino acid residue, a codon encoding the 67th amino acid residue, a codon encoding the 70th amino acid residue, a codon encoding the 72nd amino acid residue, a codon encoding the 74th amino acid residue, a codon encoding the 76th amino acid residue, a codon encoding the 77th amino acid residue, a codon encoding the 79th amino acid residue, and a codon encoding the 81st amino acid residue in the light chain variable region defined by the Kabat method are codons encoding basic amino acid residues.

12. the nucleic acid molecule, wherein the nucleotide sequence encoding the antigen-binding region comprises a nucleotide sequence encoding a single chain antibody; The method according to claim 10, wherein the single chain antibody is a single chain antibody that binds to a complex of a MAGE-A4-derived peptide and HLA-A2, a complex of a PRAME-derived peptide and HLA-A24, CD19, BCMA, or CEA.

13. The method according to claim 10, wherein in the nucleic acid molecule, the segment encoding the transmembrane domain comprises a nucleotide sequence encoding a transmembrane region of any one protein selected from the group consisting of a T cell receptor α chain, a T cell receptor β chain, CD3ε, CD3ζ, CD4, CD5, CD8, CD9, CD16, CD22, CD28, CD33, CD37, CD45, CD64, CD80, CD86, CD134, CD154, 4-1BB, ICOS, and GITR.

14. The method of claim 10, wherein in the nucleic acid molecule, the segment encoding the intracellular domain comprises a nucleotide sequence encoding a signal transduction domain of at least one protein selected from the group consisting of CD3ζ, CD3γ, CD3δ, CD3ε, CD5, CD22, CD66d, CD79a, CD79b, FcRγ and FcRβ.

15. the nucleic acid molecule, wherein the segment encoding the intracellular domain further comprises a nucleotide sequence encoding a costimulatory domain; The method of claim 14, wherein the costimulatory domain is a costimulatory domain of at least one protein selected from the group consisting of 4-1BB, CD28, GITR, CD2, CD5, CD8, CD9, CD16, CD22, CD33, CD37, CD64, CD80, CD86, CD134, CD154, and ICOS.

16. The method of claim 10, wherein the nucleic acid molecule further comprises a segment encoding a hinge domain between the nucleotide sequence encoding the antigen-binding region and the segment encoding the transmembrane domain.

17. The method according to claim 10, wherein the nucleic acid molecule is DNA or RNA.

18. A method for producing immune cells containing a chimeric antigen receptor, comprising introducing a vector containing the nucleic acid molecule according to any one of claims 10 to 17 into immune cells and causing the immune cells to express the chimeric antigen receptor.