BTN2A1-binding peptide

Abstract

The present invention relates to a butyrophilin subfamily 2 member A1 (BTN2A1)-binding peptide having an amino acid sequence comprising an amino acid other than glutamic acid, such as tryptophan, at the position corresponding to position 22 of SEQ ID NO: 1, an amino acid other than arginine, such as tyrosine, at the position corresponding to position 73 of SEQ ID NO: 1, and / or an amino acid other than threonine, such as histidine or tryptophan, at the position corresponding to position 81 of SEQ ID NO: 1. The BTN2A1-binding peptide may be contained in an (exogenous) immune receptor or its extracellular domain, preferably a γδ T cell receptor or its extracellular domain, more preferably a γ9δ2 T cell receptor or its extracellular domain. The present invention particularly relates to use in therapy, preferably in the treatment of cancer and / or infection, or in diagnostic methods.

Description

[Technical Field]

[0001] The present disclosure relates to BTN2A1-binding peptides, particularly V gamma 9 TCR domain, and more particularly Vy9V52 TCR-based therapies, that can be used in the treatment of cancer, autoimmune disease, or in the treatment of infection. [Background technology]

[0002] Targeting tumors with immune cells has accelerated over the past few decades, and many different immune cell subsets have been investigated for immunotherapy. One of the immune subsets with the most successful clinical translation has been T cells. Many different strategies have been developed to harness the power of T cells, including immune checkpoint blockade, bispecific T cell engagers, and genetically engineered T cells, or CAR T cells [1, 2]. Many of these strategies have primarily considered the major T cell lineage, αβ T cells.

[0003] However, a smaller T cell lineage, γδ T cells, is currently attracting much attention due to its unique characteristics. They are at the interface of the innate and adaptive immune systems, combining both features. They possess innate-like responses but are also capable of forming immunological memory. The γδ T cell lineage can be divided into two subsets: Vγ9Vδ2 T cells and Vδ2-negative γδ T cells. The latter subset is associated with more adaptive immune features [3]. Consequently, T cells belonging to the Vδ2-negative T cell subset have been shown to recognize multiple antigens in a clonal manner [4]. Meanwhile, Vγ9Vδ2 T cells are classified as non-variant T cells; all use the same TCR V gene and recognize the same antigen complex [5], resembling iNKT cells that recognize CD1d [6] or MAIT cells that recognize MR1 [7]. A commonality between both γδ T cells is their ability to recognize and kill a wide range of tumors, which has led to many different strategies for using these γδ T cells and their γδ T cell receptors (TCRs) as immunotherapies [8].

[0004] T cells bearing the Vγ9Vδ2 TCR can recognize intracellular phosphoantigens arising from microbial pathogens or, in the case of stressed or malignant cells, from a dysregulated mevalonate pathway [9]. These phosphoantigens bind to the intracellular domain B30.2 of BTN3A1 [10, 11], resulting in the formation of a complex between the intracellular domains of BTN3A1 and BTN2A1

[12] . Recent studies have reported that the Vγ9 domain of the Vγ9Vδ2 TCR interacts with BTN2A1 using germline-encoded residues, independent of the CDR3γ residues, and that this interaction is essential for Vγ9Vδ2 T cell activation [13, 14]. Both studies determined the affinity of this interaction to be 40–50 μM, a typical TCR-ligand affinity

[15] , but interaction with BTN2A1 alone may not be sufficient for T cell activation. The previously identified phosphoantigen-sensing molecule BTN3A1 remains important for Vγ9Vδ2 T cell activation, second only to BTN2A1 [13, 14, 16]. Although an interaction between the extracellular domain of BTN3A1 and the extracellular domain of the Vγ9Vδ2 TCR has not been formally demonstrated, it was recently reported that a fusion protein composed of the extracellular domains of BTN2A1 and BTN3A1 can activate Vγ9Vδ2 T cells in the presence of costimulatory signals

[17] .

[0005] Vγ9Vδ2 TCRs have varying potencies when used as Vγ9Vδ2 TCR-based therapeutic or diagnostic agents [16, 18-20]. This potency is related to the binding strength (affinity) of the Vγ9Vδ2 TCR to tumor cells

[16] , with "strongly" binding Vγ9Vδ2 TCRs resulting in better and higher functional avidity, as demonstrated by increased tumor control in vitro and in xenograft mouse models [19, 20]. Vγ9Vδ2 TCRs are often not sufficiently potent to induce (and sustain) tumor control in most preclinical in vivo models.

[0006] Several therapeutic strategies utilizing the tumor recognition capabilities of the Vγ9Vδ2 TCR have been developed. One of these strategies, the TEG concept [19, 21, 22], is currently being tested in phase I clinical trials, while γδTCR-antibody fusion proteins

[23] and γδTCR-anti-CD3 bispecific molecules

[20] are in preclinical trials. In many TCR-based therapeutic strategies, the receptor-ligand interaction is typically enhanced to submicromolar affinities to enhance tumor clearance in vivo [24-26]. Because tumor reactivity is directly related to the binding strength of the Vγ9Vδ2 TCR to its ligand complex, affinity maturation of the Vγ9Vδ2 TCR has been considered to increase the efficacy of the above-mentioned Vγ9Vδ2 TCR-based therapies. However, approaches to affinity enhancement of Vγ9Vδ2 TCRs, for example using phage display, have proven unpromising due to, for example, the poor expression of these TCRs or TCR variable fragments in E. coli

[27] . [Prior art documents] [Patent documents]

[0007] [Patent Document 1] WO99 / 42077 [Patent Document 2] W006 / 040153 [Patent Document 3] WO06 / 122825 [Patent Document 4] WO94 / 04678 [Patent Document 5] U.S. Patent No. 4,554,101 [Patent Document 6] WO2023102615A1 [Non-patent literature]

[0008] [Non-Patent Document 1] https: / / www.ebi.ac.uk / pdbe / entry / pdb / 1hxm Summary of the Invention [Problem to be solved by the invention]

[0009] Thus, there remains a need to develop novel approaches for affinity enhancement of Vy9V52 and / or to identify Vy9V52 TCR variants with increased ligand-complex affinity, in particular Vy9 domain variants with increased binding affinity to BTN2A1. It is an object of the present disclosure to solve one or more of the above or other problems. [Means for solving the problem]

[0010] The inventors reasoned that an alternative strategy for affinity maturation of the Vy9 domain for binding to BTN2A1 is necessary to obtain more potent Vy9V52 TCR-based therapeutics. Previous attempts to affinity mature Vy9V52 TCRs, for example using phage display, have been unsuccessful due to, for example, poor expression in E. coli.

[0011] The present inventors have developed an efficient screening method based on the recently developed gamma delta TCR anti-CD3 bispecific molecule (GAB) to identify more potent Vy9V52 TCR variants for immunotherapy and to characterize Vy9V52 TCR variants that mediate improved functional avidity and increased stability

[20] . This disclosure can be used to obtain more potent diagnostic and therapeutic agents, if based on the Vy9 chain, particularly its BTN2A1-binding peptide.

[0012] In particular, mutant BTN2A1-binding amino acid sequences (i.e., peptides) within the Vγ9 domain have been identified that can increase affinity for BTN2A1 by approximately 10-fold, e.g., via GAB or TEG (TEG = T cells engineered to express a defined GdTCR), and enable more potent T cell activation and tumor cell killing. E22W , Vγ9 T81H , and / or Vγ9 T81Whave found that this enhances the affinity of the peptide for BTN2A1 and can be used, for example, to enhance the activity of GAB or other g9d2 TCR-based immunotherapies, or when the g9d2 TCR is used as a diagnostic tool.

[0013] Thus, the present disclosure relates to mutations (combinations of) in the BTN2A1-binding peptide within the V gamma 9 chain of the gamma-delta T cell receptor (gdTCR) that enhance the efficacy of gdTCR-based therapeutics, such as alpha-beta T cells (TEGs) engineered to express a defined gdTCR and gdTCR anti-CD3 bispecific molecules (GABs), and / or, in the case of GABs, increase the expression levels of gdTCRs bearing recombinant V gamma 9.

[0014] While the V gamma 9 chain can pair with any V delta (V delta 1, ... V delta 8) chain to form a gdTCR, most commonly, the V gamma 9 chain pairs with the V delta 2 chain to generate the Vy9V52 TCR. This combination of gamma and delta TCR chains senses intracellular phosphoantigen levels in the cell, which can increase when the cell's metabolism is perturbed, such as in the case of infection, or when the cell undergoes malignant transformation. Increased intracellular phosphoantigen levels result in the translocation and complexation of the butyrophilin molecules BTN2A1 and BTN3A1-BTN3A2 / A3 heterodimer, which triggers Vy9V52 TCR-mediated T cell activation. While potent anticancer agents are anticipated, the present disclosure can also be used as a therapeutic agent to eliminate infected cells or as a diagnostic tool, i.e., to determine BTN2A1 expression by cells.

[0015] Furthermore, mutant BTN2A1-binding peptides were identified that increase the stability of, for example, GAB. R73YWe found that this enhances the stability of the BTN2A1-binding peptide within the Vγ9 chain, thereby enhancing the stability of GAB. This modification can increase cell expression and therefore production yield per culture volume, which is currently a major obstacle to GAB production, making it relatively expensive. The mutation increases production yield by at least two-fold and is also useful for stabilizing the β-chain.

[0016] Additionally, the present inventors have identified mutant BTN3A1 / A2 / A3-binding peptides that confer enhanced binding affinity to BTN3A1 / A2 / A3, e.g., via GAB or TEG, and / or enable more potent T cell activation and tumor cell killing. For example, a BTN3A1 / A2 / A3-binding peptide according to the present disclosure may be combined with a BTN2A1-binding peptide according to the present disclosure, e.g., a γδ T cell receptor or its extracellular domain, more preferably a γ9δ2 T cell receptor or its extracellular domain, wherein preferably the BTN2A1-binding peptide is a T cell receptor γ chain domain, preferably a T cell receptor γ9 chain domain, and / or the BTN3A1 / A2 / A3-binding peptide is a T cell receptor δ chain domain, preferably a T cell receptor δ2 chain domain. DETAILED DESCRIPTION OF THE INVENTION

[0017] In a first aspect, the present disclosure relates to butyrophilin subfamily 2 member A1 (BTN2A1) binding peptides.

[0018] BTN2A1 is a member of the butyrophilin and butyrophilin family of transmembrane (TM) proteins, eight of which (BTN1A1, BTN2A1 / 2A2, BTN3A1 / 3A2 / 3A3, MOG, and BTNL2) are located in the major histocompatibility complex (MHC) class I region of human chromosome 6.

[0019] BTN2A1 binding is important for the recognition of intracellular phosphoantigens arising from microbial pathogens or, in the case of stressed or malignant cells, from a dysregulated mevalonate pathway, such as occurs in cancer or infectious diseases. [9] These phosphoantigens bind to the intracellular domain B30.2 of BTN3A1. [10, 11] This results in the formation of a complex between the intracellular domains of BTN3A1 and BTN2A1.

[12] This leads to the formation of BTN2A1 homodimers and BTN3A1-BTN3A2 / A3 heterodimers.

[77] Recent studies have reported that BTN2A1-binding peptides from the Vγ9 domain of the Vγ9Vδ2 TCR interact with BTN2A1 using germline-encoded residues, independent of the CDR3γ residues, and that this interaction is important for Vγ9Vδ2 T cell activation. [13, 14] A BTN2A1-binding peptide according to the present disclosure can be combined with a BTN3A1-binding moiety, such as the delta(2) chain, for example, in a γδ TCR or its extracellular domain (e.g., a Vγ9Vδ2 TCR or its extracellular domain).

[0020] The BTN2A1-binding peptides of the present disclosure comprise an amino acid sequence that binds to BTN2A1, wherein the amino acid sequence preferably has at least 50, 60, 70, 80, 90, 91, 92, 93, 94, 95, 96, 97, 98, 99, or 100% sequence identity with SEQ ID NO:1, and the amino acid sequence is characterized by including an amino acid other than glutamic acid at a position corresponding to position 22 of SEQ ID NO:1.

[0021] Additionally or alternatively, a BTN2A1-binding peptide can include an amino acid other than arginine at a position corresponding to position 73 shown in SEQ ID NO:1.

[0022] Additionally or alternatively, a BTN2A1-binding peptide may contain an amino acid other than threonine at a position corresponding to position 81 set forth in SEQ ID NO:1.

[0023] Additionally or alternatively, the BTN2A1-binding peptides of the present disclosure comprise an amino acid sequence that binds to BTN2A1, wherein the amino acid sequence preferably has at least 50, 60, 70, 80, 90, 91, 92, 93, 94, 95, 96, 97, 98, 99, or 100% sequence identity with SEQ ID NO:43, and wherein the amino acid sequence contains an amino acid other than glutamic acid at a position corresponding to position 22 of SEQ ID NO:43.

[0024] Additionally or alternatively, a BTN2A1-binding peptide can include an amino acid other than arginine at a position corresponding to position 73 set forth in SEQ ID NO:43.

[0025] Additionally or alternatively, a BTN2A1-binding peptide can include an amino acid other than threonine at a position corresponding to position 81 of SEQ ID NO:43.

[0026] Mutation of the naturally occurring amino acids at positions 22 and / or 81 in a BTN2A1-binding peptide (SEQ ID NO: 1 or 43) in the variable domain of the human TCR γ9 chain surprisingly increased the affinity of the peptide for BTN2A1, with the mutation at position 22 resulting in a greater than 10-fold increase in affinity for BTN2A1. The term affinity refers to the strength of the binding reaction between a binding domain and an epitope, e.g., between a peptide and BTN2A1. It is the sum of the attractive and repulsive forces acting between a binding domain and an epitope. The term affinity, as used herein, refers to the apparent binding affinity, determined as the equilibrium dissociation constant (Kd). High affinity binding domains preferably have a binding affinity of 10 or more. -8 Less than M, preferably 10 -9 A low affinity binding domain preferably has a binding affinity Kd for its specific epitope of less than 10 -8 More than M, preferably 10 -7 It has a binding affinity, Kd, ​​for its specific epitope that exceeds M.

[0027] Additionally, it has been surprisingly found that mutating the amino acid at position 73 naturally occurring in the BTN2A1-binding peptide (SEQ ID NO: 1 or 43) within the variable domain of the human TCR γ9 chain can increase the stability of the BTN2A1-binding peptide and increase production yield (e.g., grams of BTN2A1-binding peptide or constructs containing same) per culture volume by more than two-fold.

[0028] The present disclosure particularly relates to the BTN2A1-binding peptides described herein, wherein the amino acid other than glutamic acid at the position corresponding to position 22 is tryptophan (or a conservative substitution thereof), e.g., the amino acid other than glutamic acid at the position corresponding to position 22 is tryptophan or phenylalanine, and most preferably, the amino acid other than glutamic acid at the position corresponding to position 22 is tryptophan.

[0029] Additionally or alternatively, the amino acid other than arginine at the position corresponding to position 73 is tyrosine (or a conservative substitution thereof), e.g., the amino acid other than arginine at the position corresponding to position 73 is tyrosine, tryptophan, or phenylalanine.

[0030] Additionally or alternatively, the amino acid other than arginine at the position corresponding to position 81 is histidine (or a conservative substitution thereof) or tryptophan (or a conservative substitution thereof), for example, the amino acid other than arginine at the position corresponding to position 81 is histidine, arginine, lysine, tryptophan, phenylalanine, or tyrosine.

[0031] Additionally or alternatively, a BTN2A1-binding peptide according to the present disclosure may comprise: - a threonine (or a conservative substitution thereof), such as threonine, alanine, serine, or glycine, at the position corresponding to position 18 of SEQ ID NO: 1 (or 43); - arginine (or a conservative substitution thereof), such as arginine, histidine, or lysine, at the position corresponding to position 20 of SEQ ID NO: 1 (or 43), - glutamic acid (or a conservative substitution thereof) or leucine (or a conservative substitution thereof), such as glutamic acid, aspartic acid, asparagine, and glutamine, leucine, methionine, isoleucine, valine, or cysteine, at a position corresponding to position 70 of SEQ ID NO: 1 (or 43); - aspartic acid (or a conservative substitution thereof), such as aspartic acid, glutamic acid, asparagine, or glutamine, at a position corresponding to position 72 of SEQ ID NO: 1 (or 43), - arginine (or a conservative substitution thereof), histidine (or a conservative substitution thereof), or tryptophan (or a conservative substitution thereof), such as arginine, histidine, lysine, histidine, arginine, lysine, tryptophan, phenylalanine, or tyrosine, at a position corresponding to position 81 of SEQ ID NO: 1 (or 43); - threonine (or a conservative substitution thereof) or isoleucine (or a conservative substitution thereof), such as threonine, alanine, serine, glycine, isoleucine, methionine, leucine, valine, or cysteine, at a position corresponding to position 83 of SEQ ID NO: 1 (or 43); and / or - a histidine (or a conservative substitution thereof), e.g., histidine, arginine, or lysine, at a position corresponding to position 85 of SEQ ID NO: 1 (or 43); include.

[0032] Preferably, a BTN2A1-binding peptide according to the present disclosure comprises: - threonine or leucine at the position corresponding to position 18 of SEQ ID NO: 1, - arginine or histidine at the position corresponding to position 20 of SEQ ID NO: 1, - glutamic acid or a conservative substitution thereof selected from aspartic acid or leucine at the position corresponding to position 70 of SEQ ID NO: 1, - aspartic acid or asparagine at the position corresponding to position 72 of SEQ ID NO: 1, - threonine or a conservative substitution thereof selected from alanine, serine and glycine, or histidine, or tryptophan or phenylalanine at the position corresponding to position 81 of SEQ ID NO: 1; - threonine or a conservative substitution thereof selected from alanine, serine and glycine, or isoleucine, or a conservative substitution thereof selected from methionine, leucine, valine, and cysteine ​​at the position corresponding to position 83 of SEQ ID NO: 1; and / or - histidine at the position corresponding to position 85 of SEQ ID NO: 1 include.

[0033] The (human) BTN2A1-binding peptide according to the present disclosure is preferably the (variable) domain of the (human) T cell receptor γ chain, preferably the (variable) domain of the (human) T cell receptor γ9 chain.

[0034] While most natural T cell receptors (TCRs) contain complete alpha (α) and beta (β) chains, the majority of natural immune cells express alternative receptors formed by complete gamma (γ) and delta (δ) chains. TCR chains typically consist of two extracellular domains, the variable (V) domain and the constant (C) domain, which together form an antiparallel β-sheet from the immunoglobulin superfamily (IgSF). The constant domain lies near the cell membrane and is followed by a transmembrane domain and a short cytoplasmic tail, while the variable domain is capable of binding to antigens or target moieties. The variable domains of both the TCR α and β chains, or both the TCR γ and δ chains, may each have three hypervariable or complementarity-determining regions (CDRs).

[0035] The BTN2A1-binding peptides of the present disclosure may be contained in an (exogenous) immunoreceptor or its extracellular domain, such as a human (exogenous) immunoreceptor or its extracellular domain. The immunoreceptor may be a T cell receptor or a chimeric antigen receptor, preferably a γδ T cell receptor or its extracellular domain. The γ(9) chain (variable) domain comprising the BTN2A1-binding peptide of the present invention may pair with any δ1-8 chain.

[0036] The present disclosure also provides a butyrophilin subfamily 3 member A1 (BTN3A1) and / or BTN3A2 and / or BTN3A3 binding peptide comprising an amino acid sequence that binds to BTN3A1 and / or BTN3A2 and / or BTN3A3, wherein the amino acid sequence has at least 70, 75, 80, 85, 90, 95, 99, 100% sequence identity to SEQ ID NO: 52 (and / or SEQ ID NOs: 53, 54, 55, 56), and the amino acid sequence is - at the position corresponding to position 31 of SEQ ID NO: 52 (and / or SEQ ID NOs: 53, 54, 55, 56), valine (or a conservative substitution thereof), methionine (or a conservative substitution thereof), alanine (or a conservative substitution thereof), isoleucine (or a conservative substitution thereof), leucine (or a conservative substitution thereof), serine (or a conservative substitution thereof), or threonine (or a conservative substitution thereof), preferably valine (or a conservative substitution thereof), methionine (or a conservative substitution thereof), alanine (or a conservative substitution thereof), and / or - serine (or a conservative substitution thereof) or alanine (or a conservative substitution thereof) at the position corresponding to position 53 of SEQ ID NO: 52 (and / or SEQ ID NOs: 53, 54, 55, 56); The invention is characterized in that it comprises Preferably, the BTN3A1 / A2 / A3 binding peptide is a T cell receptor δ chain domain, preferably a T cell receptor δ2 chain domain.

[0037] Additionally or alternatively, the modification at position 53 of SEQ ID NO: 52 (and / or SEQ ID NOs: 53, 54, 55, 56) is preferably selected from a substitution of lysine (K) with alanine (A), a substitution of lysine (K) with cysteine ​​(C), a substitution of lysine (K) with methionine (M), a substitution of lysine (K) with serine (S), a substitution of lysine (K) with tryptophan (W), a substitution of lysine (K) with valine (V), or a substitution of lysine (K) with proline (P).

[0038] BTN3A1 / A2 / A3-binding peptides according to the present disclosure may provide enhanced binding affinity to BTN3A1 / A2 / A3, for example, via GAB or TEG, and / or may enable more potent T cell activation and tumor cell killing.

[0039] A BTN3A1 / A2 / A3-binding peptide according to the present disclosure and a BTN2A1-binding peptide according to the present disclosure can be combined in an (exogenous) immune receptor, T cell receptor, or chimeric antigen receptor, γδ T cell receptor, or an extracellular domain thereof, more preferably a γ9δ2 T cell receptor or an extracellular domain thereof, wherein preferably the BTN2A1-binding peptide is a T cell receptor γ chain domain, preferably a T cell receptor γ9 chain domain, and / or the BTN3A1 / A2 / A3-binding peptide is a T cell receptor δ chain domain, preferably a T cell receptor δ2 chain domain.

[0040] Preferably, the (exogenous) immune receptor, T cell receptor or chimeric antigen receptor disclosed herein, preferably a γδ T cell receptor, or an extracellular domain thereof, is capable of binding to or binds to a tumor cell, e.g. an antigen present on the surface of a tumor cell.

[0041] The exogenous immune receptor according to the present disclosure is preferably not an endogenous T cell receptor. For example, the exogenous immune receptor can be a specific selected γδ T cell receptor useful in cancer treatment. The sequence may be similar to the endogenous γδ T cell receptor. The difference is that the exogenous immune receptor is preselected against a specific target, e.g., an antigen present on the surface of a tumor cell. The exogenous immune receptor is expressed, for example, from a transgene construct and not from an endogenous locus. The exogenous immune receptor according to the present disclosure can be of a different origin, i.e., derived from another species, compared to the origin of the T cells engineered to obtain the engineered T cells bearing the exogenous immune receptor. The exogenous immune receptor can also be of the same origin, i.e., derived from the same species, compared to the origin of the T cells engineered to obtain the engineered T cells bearing the exogenous immune receptor. The exogenous immune receptor can also be an engineered γδ T cell receptor or an engineered αβ T cell receptor.

[0042] Any of the immune receptors according to the present disclosure may be chimeric antigen receptors (CARs), which are recombinant receptors that combine the specificity of an antigen-specific antibody with the activation function of T cells.

[42] CARs may be fusion molecules between an antibody and a transmembrane domain, allowing expression of the antibody on the cell surface of immune cells and signal transduction to the cell.

[0043] In one aspect of the present disclosure, any of the immune receptors according to the present disclosure may be selected from the group consisting of an (engineered) γδ T cell receptor, an (engineered) αβ T cell receptor, or a chimeric antigen receptor (CAR).

[0044] In particular, the BTN2A1-binding peptide according to the present disclosure may be a (human) γ9δ2 T cell receptor or its extracellular domain.

[0045] The present disclosure also provides i) a γδ T cell receptor or an extracellular domain thereof as disclosed herein (including a BTN2A1-binding peptide according to the present disclosure and / or a BTN3A1 / A2 / A3-binding peptide according to the present disclosure), and ii) Toxins and / or labels The present invention provides a combination or construct comprising:

[0046] The toxin is preferably any compound or combination of compounds effective for killing cancer cells or infected cells. The toxin may be, for example, diphtheria toxin, pseudomonas toxin, and / or saporin. The toxin and / or label may be fused to the γδ T cell receptor or its extracellular domain, for example, via a linker.

[0047] The label can be any label useful in diagnostic situations. The label can allow visualization of the binding of the BTN2A1-binding peptide to BTN2A1 and / or the BTN3A1 / A2 / A3-binding peptide to BTN3A1 / A2 / A3. The label can be, for example, a fluorophore. Fluorophores are highly sensitive and generally do not affect the properties of the target molecule. The process can involve binding of the fluorophore to BTN2A1 / BTN3A1 / A2 / A3 proteins expressed by, for example, cancer cells or infected cells, via the BTN2A1 / BTN3A1 / A2 / A3-binding peptides of the present invention. Once binding is complete, fluorescence can be visualized by excitation, for example, via a fluorescent microscope. Fluorescent labels can be used in assays such as ELISA, FISH, and fluorescent microscopy. The label can be linked to the BTN2A1 / BTN3A1 / A2 / A3-binding peptides of the present invention by any linker.

[0048] The present disclosure further provides: i) a γδ T cell receptor or an extracellular domain thereof as disclosed herein (including a BTN2A1-binding peptide according to the present disclosure and / or a BTN3A1 / A2 / A3-binding peptide according to the present disclosure), and ii) an (effector) cell binding domain, preferably an immune cell binding domain, preferably a B cell binding domain, a macrophage binding domain, or a fibroblast binding domain, more preferably a T cell binding domain and / or a natural killer (NK) cell binding domain; A construct comprising:

[0049] The γδ T cell receptor or its extracellular domain is preferably a γ9δ2 T cell receptor or its extracellular domain. As used herein, the term "extracellular domain" of a gamma or delta TCR chain includes the extracellular portions of the V gamma and C gamma domains, or the extracellular portions of the V delta and C delta domains.

[0050] This construct preferably combines a (low) affinity TCR interaction with its ligand on tumor cells, by binding to CD3 on T lymphocytes and / or CD16 on NK cells, with a (high) affinity interaction with, for example, T lymphocytes and / or NK cells. This allows the bispecific construct to first bind to T cells and / or NK cells, which are then subsequently recruited to the tumor. This concept can be further elaborated by creating a trispecific construct, to which tumor binding depends both on the γδ TCR and a second molecule, such as a checkpoint ligand.

[0051] As demonstrated herein, constructs according to the present disclosure can attach to infected or cancer cells and bind to immune cells, such as T cells and / or natural killer (NK) cells, to mount an immune response against the infected or cancer cells, thereby reducing or even eliminating the cells. The small size of constructs according to the present disclosure, combined with their acellular nature, makes them ideal treatment tools for infectious diseases and cancer. These constructs hold great promise for the treatment of cancer and infectious diseases.

[0052] Fusion of the extracellular domain of the γδ TCR as a tumor-binding domain to, for example, an anti-CD3 scFv can effectively target T cells to tumor cells without the need for TEG engineering. The γδ TCR anti-CD3 bispecific molecule (abbreviated as GAB) can redirect CD3+ effector cells to several tumor cell lines, both hematopoietic and solid, while retaining the tumor recognition mode of action described for certain γ9δ2 TCRs [43, 44, 45], thereby opening up new antigenic domains for the bispecific format.

[0053] Therefore, in the above construct, the T cell binding domain is capable of binding to Cluster of Differentiation 3 (CD3), CD4, CD8, CD16, CD56, CD103, CD134, CD154, and / or CD314, and / or is a single chain Fv anti-CD3, CD4, CD8, CD16, CD56, CD103, CD134, CD154, and / or the natural killer (NK) cell binding domain may bind to CD16, NKG2D, NKp30, NKp44, NKp46, and / or DNAM, and / or is a single chain Fv anti-CD16, NKG2D, NKp30, NKp44, NKp46, and / or DNAM.

[0054] Additionally or alternatively, in the constructs, the binding domain can be modified to inhibit T cell activation. For example, the construct can bind to a T cell inhibitory domain. Specifically, in the above constructs, the T cell binding domain can bind to PD1 (expressed on the surface of T cells), e.g., to reduce T cell activation. This embodiment can be useful, for example, in the prevention or treatment of autoimmune diseases, where it is desirable to block, but not enhance, the activity of T cells (e.g., gdT cells or abT cells).

[0055] Therefore, in the above construct, - the T cell binding domain may bind to PD1, LAG3, CTLA4, TIGIT, CD96, BTLA, VISTA, TIM3, LAIR1, (inhibitory) KIR, CD160, and / or an immunoreceptor binding domain having an intracellular ITIM or ITSM motif, and / or may be a single chain Fv anti-PD1, LAG3, CTLA4, TIGIT, CD96, BTLA, VISTA, TIM3, LAIR1, (inhibitory) KIR, CD160, and / or an immunoreceptor binding domain having an intracellular ITIM or ITSM motif, and / or - the natural killer (NK) cell binding domain may bind to NKG2A, CD96, TIGIT, (inhibitory) KIR, PD1, TIM3, LAG3, CD112R, CD160, LAIR1, and / or an immunoreceptor having an intracellular ITIM or ITSM motif, and / or may be a single chain Fv anti-NKG2A, CD96, TIGIT, (inhibitory) KIR, PD1, TIM3, LAG3, CD112R, CD160, LAIR1, and / or an immunoreceptor having an intracellular ITIM or ITSM motif binding domain.

[0056] The gamma delta T cell receptor (or its extracellular domain) and the immune cell binding domain, T cell binding domain, and / or natural killer (NK) cell binding domain are preferably fused via a linker or linking group that preferably provides conformational flexibility so that the extracellular domain of the gamma delta TCR can interact with its epitope, while the T cell binding domain and / or the NK cell binding domain can interact with their cognate epitope. Preferred linker groups are linker polypeptides comprising 1 to 60 amino acid residues, preferably 5 to 40 amino acid residues, and most preferably about 15 amino acid residues, e.g., 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, or 20 amino acid residues. Some preferred examples of such amino acid sequences include Gly-Ser linkers of the type (GlyxSery)z, such as (Gly4Ser)3, (Gly4Ser)7, or (Gly3Ser2)3, as described, for example, in WO99 / 42077, and the GS30, GS15, GS9, and GS7 linkers, as described, for example, in WO06 / 040153 and WO06 / 122825, and hinge-like regions, such as the hinge region of a native heavy chain antibody or similar sequences (as described in WO94 / 04678). The most preferred linker is the (Gly4Ser)3 linker.

[0057] The immune cell binding domain, T cell binding domain, and / or natural killer (NK) cell binding domain are preferably antibodies, preferably single-chain heavy chain variable domain antibodies, such as camelid VHHs, variable neoantigen receptors derived from shark immunoglobulins, scFvs, tandem scFvs, scFabs, improved scFabs

[58] , or antibody mimetics, such as engineered ankyrin repeat proteins, binding proteins based on the Z domain of protein A, binding proteins based on fibronectin type III domains, engineered lipocalins, and binding proteins based on the human Fyn SH3 domain

[59] ,

[60] . For example, Liao et al., 2000

[61] , describe single-chain antibodies (scFvs) against CD3 expressed on the plasma membrane of tumor cells. Furthermore, single-chain antibodies against CD3 are commercially available, for example from Creative Biolabs.

[0058] A preferred single-chain antibody against CD3 present in a (bispecific) construct according to the present disclosure comprises a single-chain Fv anti-CD3 binding domain, preferably derived from the chimeric mouse-human OKT3 antibody, as described, for example, in

[62] . Preferred scFvs derived from the OKT3 antibody have been described

[63] ,

[64] .

[0059] Constructs according to the present disclosure are preferably bispecific fusion proteins, such as constructs comprising i) a γδ T cell receptor or its extracellular domain as disclosed herein (including a BTN2A1-binding peptide according to the present disclosure and / or a BTN3A1 / A2 / A3-binding peptide according to the present disclosure), and ii) a toxin and / or a label, or more preferably an (effector) cell-binding domain, preferably an immune cell-binding domain, more preferably a T cell-binding domain and / or a natural killer (NK) cell-binding domain. Additionally or alternatively, the γδ T cell receptor or its extracellular domain is fused to the T cell-binding domain and / or the natural killer (NK) cell-binding domain.

[0060] The disclosed constructs can combine tumor targeting with immune cell recruitment, thus avoiding the major drawback of engineered immune cells such as CAR-T and TEG, which are associated with the challenging logistics of such advanced therapeutic medicines (ATMPs): production of ATMPs is a personalized, cumbersome, and expensive process that takes weeks in most cases and can be associated with production failures

[65] .

[0061] A preferred recombinant bispecific protein according to the present disclosure comprises the extracellular domain of a gamma delta TCR, preferably a gamma 9 delta 2 TCR. A preferred recombinant bispecific protein according to the present disclosure comprises the extracellular domain of a TCR gamma chain, preferably gamma 9 (comprising the BTN2A1 peptide disclosed herein and / or the BTN3A1 / A2 / A3 binding peptide disclosed herein), linked at its C-terminus to a CD3 binding domain, preferably an scFv derived from the OKT3 antibody, e.g., as described in

[64] , and preferably the extracellular domain of a delta TCR, preferably the extracellular domain of a delta 2 TCR. Additionally or alternatively, the extracellular domain of the gamma TCR and / or delta TCR, preferably the extracellular domain of the delta TCR, preferably the extracellular domain of a delta 2 TCR, may be fused at its N- or C-terminus to the extracellular domain of a checkpoint-associated molecule, e.g., the extracellular domain of the PD-1 receptor.

[0062] The present disclosure also provides a nucleic acid or combination of nucleic acids encoding a BTN2A1-binding peptide according to the present disclosure and / or a BTN3A1 / A2 / A3-binding peptide according to the present disclosure, or (either) a construct according to the present disclosure. The nucleic acid may be contained in a vector and / or may be contained in a cell, which may or may not be an immune cell. Preferably, the nucleic acid is expressed in the cell.

[0063] An immunoreceptor (or its extracellular domain) according to the present disclosure can be, for example, a gamma delta T-cell receptor (or its extracellular domain), which comprises a first chain that is a gamma chain and a second chain that is a delta chain. The chains can be present on one nucleic acid or on two separate nucleic acids. A first nucleic acid encodes the first chain and a second nucleic acid encodes the second chain, or one nucleic acid encodes both the first and second chains. The one or more nucleic acids can be DNA or RNA. The exogenous immunoreceptor amino acid sequence that it encodes is introduced into a cell and expressed therein such that it is expressed on the surface of the cell.

[0064] Preferably, in one embodiment, the nucleic acid encoding an immunoreceptor (or its extracellular domain) encodes an immunoreceptor (or its extracellular domain) in which different chains, e.g., gamma and delta chains, are expressed as a single translated protein product that includes an F2A or T2A peptide linker sequence between the coding sequences for both chains, such that the translated protein undergoes self-cleavage, resulting in the formation of separate chains.

[0065] The one or more nucleic acids encoding an immunoreceptor (or its extracellular domain) according to the present disclosure may be mRNA, which can be directly translated into an immunoreceptor (or its extracellular domain) when introduced into the cytoplasm of a T cell, e.g., via transfection. Preferably, the nucleic acid(s), e.g., encoding a T cell receptor chain, is / are included within a genetic construct. The genetic construct(s) can express mRNA encoding the immunoreceptor (or its extracellular domain), which is then expressed on the surface of the engineered T cell. The genetic construct can be included within a DNA or viral vector. Introduction of the one or more nucleic acids can be via transfection or transduction methods, depending on the type of nucleic acid(s) used. It is understood that the genetic construct can be composed of DNA or RNA, depending on the type of genetic construct(s) used. For example, when the genetic construct is incorporated into a retroviral or lentiviral vector, the genetic construct is included within the RNA vector genome (i.e., the sequence encoding the genetic construct). Retroviral and lentiviral vectors are well known in the art to have an RNA genome, which is reverse transcribed into DNA upon entry into the cell, and the DNA is then integrated into the host genome. Reverse transcription thus converts the genetic information, i.e., the gene construct, from RNA to double-stranded DNA, thereby enabling expression from the DNA. Integration is advantageous because it allows the proliferation of transduced cells while maintaining the viral vector genome containing the gene construct. The gene construct may also be contained within a DNA vector, such as a plasmid DNA. A suitable DNA vector may be a transposon. Suitable transposon systems (e.g., class I or class II-based) are well known in the art. As mentioned above, if an immune receptor contains two chains, such as a gamma T-cell receptor chain and a delta T-cell receptor chain, two separate gene constructs may be generated, for example, on one or two separate retroviral or DNA vectors. Alternatively, one gene construct may express one mRNA encoding the two chains.Such an mRNA may encode the two chains separately, for example, via an IRES or using a self-cleaving peptide sequence as described herein.

[0066] The nucleic acid(s) used can result in expression of the encoded immunoreceptor (or its extracellular domain), for example, through high level expression of the immunoreceptor (or its extracellular domain), for example, by using a strong promoter.

[0067] Thus, the present disclosure also provides cells that express a BTN2A1-binding peptide according to the present disclosure and / or a BTN3A1 / A2 / A3-binding peptide according to the present disclosure, a construct according to the present disclosure, and / or a nucleic acid or combination of nucleic acids according to the present disclosure, as described above.

[0068] The cell can be a bacterial cell, such as an Escherichia coli cell, or a eukaryotic cell, such as a fungal cell, including a yeast cell, such as Saccharomyces cerevisiae or a methylotrophic yeast, such as Pichia pastoris, or a mammalian cell. The eukaryotic cell is preferably a cell that is easily infected and / or transfected using standard methods known to those skilled in the art, such as yeast cells and chicken fibroblast cells. The eukaryotic cell is preferably an insect cell or a mammalian cell. Suitable insect cells include, for example, ovarian Spodoptera frugiperda cells, such as Sf9 and Sf21, Drosophila Schneider 2 cells, and Aedes albopictus C6 / 36 cells. Suitable mammalian cells include, for example, baby hamster kidney cells, human embryonic kidney cells such as HEK293 and freestyle HEK293F™ cells (ThermoFisher Scientific), VERO cells, MDCK cells, CHO cells, HeLa, and PER.C6 cells.

[66] Preferred cells are human embryonic kidney cells such as HEK293 and freestyle HEK293F™ cells.

[0069] Cells expressing a BTN2A1-binding peptide and / or a BTN3A1 / A2 / A3-binding peptide according to the present disclosure can be immune cells, more preferably human T cells or human NK cells, more preferably αβ T cells or γδ T cells. Immune cells according to the present disclosure can be immune cells engineered to contain, and preferably express, an exogenous immune receptor. Immune cells according to the present disclosure can be human immune cells, preferably human T cells or human NK cells. The exogenous immune receptor can have the same function as the corresponding endogenous T cell receptor with respect to antigen recognition and T cell action. Unengineered immune cells are cells that express an endogenous immune receptor, i.e., a T cell receptor.

[0070] Cells from a subject, such as immune cells, e.g., T cells or NK cells, can be isolated, or an established immune cell line can be used. The subject can be afflicted with cancer (patient) or a healthy subject. These immune cells can be genetically modified in vitro to express the immunoreceptors (or their extracellular domains) disclosed herein. These engineered cells can be activated in vitro and expanded to therapeutically effective expressing cell numbers. In cell therapy, these engineered cells can be infused as a pharmaceutical composition into a recipient in need thereof. The infused cells can kill (or at least stop the proliferation of) cancerous cells in the recipient that express the antigen recognized by the immunoreceptor disclosed herein. The recipient can be the same subject from which the cells were obtained (autologous cell therapy) or another subject of the same species (allogeneic cell therapy).

[0071] Also envisioned are αβ T cells (TEGs) bearing γδ TCRs according to the present disclosure, which combine the robust proliferative capacity of αβ T cells (which is active even in late-stage cancer patients, see

[46] ) with the broad tumor reactivity of γδ TCRs.

[0072] The present disclosure further relates to a method of producing a BTN2A1-binding peptide according to the present disclosure and / or a BTN3A1 / A2 / A3-binding peptide according to the present disclosure, or a construct according to the present disclosure, comprising expressing a BTN2A1-binding peptide according to the present disclosure and / or a BTN3A1 / A2 / A3-binding peptide according to the present disclosure, or a construct according to the present disclosure in a host cell to produce a BTN2A1-binding peptide according to the present disclosure and / or a BTN3A1 / A2 / A3-binding peptide according to the present disclosure, or a construct according to the present disclosure. The BTN2A1-binding peptide according to the present disclosure and / or a BTN3A1 / A2 / A3-binding peptide according to the present disclosure, or a construct according to the present disclosure, may be secreted into the growth medium of the host cell.

[0073] Also provided are pharmaceutical compositions comprising a BTN2A1-binding peptide and / or a BTN3A1 / A2 / A3-binding peptide according to the present disclosure (or a γδ TCR or extracellular domain thereof comprising the same), a construct according to the present disclosure, or a cell according to the present disclosure. The pharmaceutical composition preferably comprises a pharmaceutically acceptable carrier. As used herein, a carrier refers to a non-toxic material that does not interfere with the effectiveness of the biological activity of the active ingredient. The term "physiologically acceptable" refers to a non-toxic material that is compatible with a biological system such as a cell, cell culture, tissue, or organism. The characteristics of the carrier depend on the route of administration.

[0074] A formulation containing a therapeutically effective number of cells or constructs according to the present disclosure may contain a pharmaceutically acceptable excipient (carrier or diluent). The excipients included in the formulation have different purposes depending, for example, on the nature of the construct, the (portion) number of immune cells used, and the mode of administration. Examples of commonly used excipients include, but are not limited to, saline, buffered saline, dextrose, water for injection, glycerol, ethanol, and combinations thereof, stabilizers, solubilizers, and surfactants, buffers and preservatives, isotonicity agents, bulking agents, and lubricants.

[0075] Formulations containing a therapeutically effective number of cells or constructs according to the present disclosure can be administered to a subject using modes and techniques known to those skilled in the art. Typical modes include, but are not limited to, intravenous injection. Other modes include, but are not limited to, intratumoral, intradermal, subcutaneous (sc, sq, sub-Q, hypo), intramuscular (im), intraperitoneal (ip), intraarterial, intramedullary, intracardiac, intraarticular / joint, intrasynovial (synovial fluid area), intracranial, intraspinal, and intrathecal (spinal fluid) administration.

[0076] Usually about 1×10 4 From about 1 x 10 10 A formulation containing between 1×10 immune cells, or 0.1-10, or 1-100, or 10-1000 mg of construct may be administered. In most cases, the formulation will contain about 1×10 5 to approximately 1 × 10 9immune cells between approximately 5 x 10 5 to about 5 × 10 8 of immune cells, or approximately 1 x 10 6 to approximately 1 × 10 7 The physician can ultimately determine the appropriate dosage to be used.

[0077] A BTN2A1-binding peptide and / or a BTN3A1 / A2 / A3-binding peptide according to the present disclosure (or a γδ TCR or its extracellular domain comprising said peptide), a construct according to the present disclosure, or a cell according to the present disclosure can be administered by injection or by gradual infusion over time. Administration of the construct is preferably parenteral, e.g., intravenous, intraperitoneal, intranasal, or intramuscular. Preparations for parenteral administration include sterile aqueous or non-aqueous solutions, suspensions, and emulsions. Examples of non-aqueous solvents are propylene glycol, polyethylene glycol, vegetable oils such as olive oil, and injectable organic esters such as ethyl oleate. Aqueous carriers include water, alcoholic / aqueous solutions, emulsions, or suspensions, including saline and buffered media.

[0078] Parenteral vehicles include sodium chloride solution, Ringer's dextrose, dextrose and sodium chloride, lactated Ringer's, or fixed oils. Intravenous vehicles include fluid and nutrient replenishers, electrolyte replenishers (such as those based on Ringer's dextrose), and the like. Preservatives and other additives may also be present, such as antimicrobials, antioxidants, chelating agents, and inert gases and the like.

[0079] The present disclosure particularly provides a BTN2A1-binding peptide according to the present disclosure and / or a BTN3A1 / A2 / A3-binding peptide according to the present disclosure (or a (γδ) TCR or extracellular domain thereof comprising said peptide), any one of the constructs according to the present disclosure, or a cell according to the present disclosure for use in therapy, preferably for use in the treatment of cancer and / or infection (such as an infectious disease).

[0080] Accordingly, there is provided a method for treating cancer and / or infection (such as an infectious disease), comprising administering an effective amount of a BTN2A1-binding peptide according to the present disclosure and / or a BTN3A1 / A2 / A3-binding peptide according to the present disclosure (or a (γδ) TCR or extracellular domain thereof comprising said peptide), any one of the constructs according to the present disclosure, or a cell according to the present disclosure to, for example, a subject in need thereof. Preferably, the subject is human.

[0081] As will be apparent, a γδ TCR or its extracellular domain comprising a BTN2A1-binding peptide according to the present disclosure and / or a BTN3A1 / A2 / A3-binding peptide according to the present disclosure can preferably bind to or bind to a tumor cell, e.g., an antigen present on the surface of a tumor cell.

[0082] Occasional off-target toxicity (due to tight binding of the BTN2A1 / BTN3A1 / A2 / A3-binding peptide or the Vγ9 / Vδ2 chain containing it to BTN2A1 / BTN3A1 / A2 / A3 in healthy tissues) can be overcome by reducing the dose of the BTN2A1 / BTN3A1 / A2 / A3-binding peptide (or the γδ TCR or its extracellular domain containing it).

[0083] The uses or methods according to the present disclosure do not require the co-administration of an aminobisphosphonate, such as pamidronate or zoledronate, to increase intracellular levels of phosphoantigens, for example in tumor cells or infected cells, or at least little aminobisphosphonate, such as pamidronate or zoledronate, is required for such co-administration, for example up to 10, 9, 8, 7, 6, 5, 4, 3, 2, 1 mg, as applied in the present methods or uses. One of the main obstacles to current Vy9V52 TCR-based therapies is the low levels of phosphoantigens in tumor cells, which leads to suboptimal activation of Vy9V52 T cells or Vy9V52-TCR-based therapies. The present disclosure may overcome this challenge by enhancing binding to BTN2A1 / BTN3A1 / A2 / A3.

[0084] Treatment in the present disclosure can involve prophylactic or therapeutic administration in humans suffering from, for example, cancer or an infectious disease. Thus, a BTN2A1-binding peptide according to the present disclosure and / or a BTN3A1 / A2 / A3-binding peptide according to the present disclosure (or a (γδ) TCR or extracellular domain thereof comprising the same), any one of the constructs according to the present disclosure, or a cell according to the present disclosure can be administered to an individual suspected of having cancer or an infection, or can be administered to an individual already exhibiting active infection or cancer to reduce the signs and symptoms of said cancer or infection.

[0085] For example, in the treatment of leukemia, patients undergoing allogeneic stem cell transplantation may also benefit from the infusion of a BTN2A1-binding peptide according to the present disclosure and / or a BTN3A1 / A2 / A3-binding peptide according to the present disclosure (or a (γδ) TCR or extracellular domain thereof comprising said peptide), any one of the constructs according to the present disclosure, or cells according to the present disclosure. In this way, elimination of the leukemia may be facilitated.

[0086] The present disclosure also provides a BTN2A1-binding peptide according to the present disclosure and / or a BTN3A1 / A2 / A3-binding peptide according to the present disclosure for use in a diagnostic method, preferably for determining BTN2A1 / BTN3A1 / A2 / A3 expression in cells in a sample.

[0087] In particular, a method or use for determining BTN2A1 / BTN3A1 / A2 / A3 expression in cells in a sample, comprising: a) providing a sample comprising cells obtained from a subject (e.g., a tumor sample); b) combining the cells with a BTN2A1-binding peptide in accordance with the present disclosure and / or a BTN3A1 / A2 / A3-binding peptide in accordance with the present disclosure (under conditions suitable for binding); and c) determining binding of the BTN2A1-binding peptide and / or BTN3A1 / A2 / A3-binding peptide to the cells, thereby determining BTN2A1 / BTN3A1 / A2 / A3 expression in the cells in the sample. The present invention discloses a method or use comprising:

[0088] The method or use may be useful, for example, in determining susceptibility to a BTN2A1-binding peptide and / or BTN3A1 / A2 / A3-binding peptide therapeutic strategy according to the present disclosure in a patient suffering from, for example, cancer and / or infection (infectious disease).

[0089] In particular, the present disclosure provides an in vitro method for determining cellular BTN2A1 / BTN3A1 / A2 / A3 expression in a sample, the method comprising: a) providing a sample (e.g., a tumor sample) comprising cells obtained from a subject, preferably a cancer or infectious disease patient; b) combining the cells with a BTN2A1-binding peptide in accordance with the present disclosure and / or a BTN3A1 / A2 / A3-binding peptide in accordance with the present disclosure (under conditions suitable for binding); and c) determining binding of the BTN2A1-binding peptide and / or BTN3A1 / A2 / A3-binding peptide to the cells, thereby determining BTN2A1 / BTN3A1 / A2 / A3 expression in the cells in the sample. wherein preferably said binding is indicative of cellular BTN2A1 / BTN3A1 / A2 / A3 expression and absence of said binding is indicative of absence of cellular BTN2A1 / BTN3A1 / A2 / A3 expression.

[0090] Step c) may involve determining the degree of binding of the BTN2A1-binding peptide and / or BTN3A1 / A2 / A3-binding peptide to the cells, thereby determining the degree of cellular BTN2A1 / BTN3A1 / A2 / A3 expression in the sample; preferably, the degree of binding is indicative of the degree of cellular BTN2A1 / BTN3A1 / A2 / A3 expression.

[0091] The present disclosure also provides a β-chain peptide having at least 30, 40, 50, 60, 70, 80, 90, 95, 99, 100% sequence identity to any one of SEQ ID NOs: 12-42, e.g., SEQ ID NOs: 13, 14, 15, 16, 17, 18, 19, and / or 20, wherein the amino acid sequence is: an amino acid other than arginine at a position corresponding to position 69 in SEQ ID NO: 12 (or an amino acid other than arginine at a position corresponding to position 69 in any one of SEQ ID NOs: 13 to 42, e.g., SEQ ID NOs: 13, 14, 15, 16, 17, 18, 19, and / or 20); The present invention provides a β-chain peptide comprising:

[0092] The position corresponding to position 69 of SEQ ID NO: 12 is - a position corresponding to position 69 of SEQ ID NO: 12 - a position corresponding to position 68 of SEQ ID NO: 13 - a position corresponding to position 68 of SEQ ID NO: 14 - a position corresponding to position 68 of SEQ ID NO: 15 - a position corresponding to position 68 of SEQ ID NO: 16 - a position corresponding to position 68 of SEQ ID NO: 17 - a position corresponding to position 68 of SEQ ID NO: 18 - a position corresponding to position 67 of SEQ ID NO: 19 - a position corresponding to position 68 of SEQ ID NO: 20 - a position corresponding to position 69 of SEQ ID NO: 21 - a position corresponding to position 69 of SEQ ID NO: 22 - a position corresponding to position 69 of SEQ ID NO: 23 - a position corresponding to position 69 of SEQ ID NO: 24 - a position corresponding to position 69 of SEQ ID NO: 25 - a position corresponding to position 69 of SEQ ID NO: 26 - a position corresponding to position 69 of SEQ ID NO: 27 - a position corresponding to position 68 of SEQ ID NO: 28 - a position corresponding to position 68 of SEQ ID NO: 29 - a position corresponding to position 68 of SEQ ID NO: 30 - a position corresponding to position 69 of SEQ ID NO: 31 - a position corresponding to position 69 of SEQ ID NO: 32 - a position corresponding to position 69 of SEQ ID NO: 33 - a position corresponding to position 69 of SEQ ID NO: 34 - a position corresponding to position 68 of SEQ ID NO: 35 - a position corresponding to position 68 of SEQ ID NO: 36 - a position corresponding to position 68 of SEQ ID NO: 37 - a position corresponding to position 68 of SEQ ID NO: 38 - a position corresponding to position 68 of SEQ ID NO: 39 - a position corresponding to position 68 of SEQ ID NO: 40 - a position corresponding to position 70 of SEQ ID NO: 41 - a position corresponding to position 68 of SEQ ID NO: 42 Related to.

[0093] The beta chain peptides of the present disclosure can have antigen-binding functionality, for example, tumor antigen-binding functionality.

[0094] It has been surprisingly found that mutating the amino acid at this position that naturally occurs in the beta chain peptide increases the stability of the peptide and can increase the production yield (e.g., grams of peptide or construct containing it) per culture volume by more than two-fold.

[0095] In the β-strand peptides of the present disclosure, the amino acid other than arginine at the position corresponding to position 69 of SEQ ID NO: 12 (or the amino acid other than arginine at the position corresponding to position 69 of any one of SEQ ID NOs: 13 to 42) can be tyrosine or a conservative substitution thereof. Alternatively, it can be phenylalanine or a conservative substitution thereof, tyrosine or a conservative substitution thereof, or tryptophan or a conservative substitution thereof.

[0096] The β chain peptide of the present disclosure may be a β chain (variable) domain of a T cell receptor, preferably a β chain (variable) domain of a T cell receptor. Additionally or alternatively, the β chain peptide of the present disclosure may be comprised in an (exogenous) immune receptor or its extracellular domain, preferably an αβ T cell receptor or its extracellular domain.

[0097] The present disclosure also provides a nucleic acid or a combination of nucleic acids encoding the β chain peptides of the present disclosure. Additionally, a cell expressing the β chain peptide of the present disclosure is provided, wherein the cell is preferably an immune cell, more preferably a human T cell or a human NK cell, more preferably an αβ T cell or a γδ T cell.

[0098] The present disclosure further contemplates a method of producing a beta chain peptide of the present disclosure, comprising expressing a beta chain peptide of the present disclosure, thereby producing a beta chain peptide of the present disclosure, which may be secreted into the growth medium of the host cell.

[0099] Also contemplated are pharmaceutical compositions comprising the beta chain peptides of the present disclosure and / or the beta chain peptides of the present disclosure for use in therapy, preferably for use in treating cancer and / or infection. Such treatment can involve, for example, administering an effective amount of the beta chain peptides of the present disclosure to a subject in need thereof. Preferably, the subject is human.

[0100] In the context of the present disclosure, the term "(poly)peptide" is equivalent to the term "protein" and / or a (poly)peptide can be a part of a protein, i.e., comprised in a protein or protein domain. A (poly)peptide has a specific amino acid sequence. A "variant" of a polypeptide of the present disclosure preferably has an amino acid sequence with at least 25% sequence identity to the reference polypeptide. A (poly)peptide of the present disclosure is isolated if it is no longer present in its natural environment. Peptides according to the present disclosure can have a length between 1-500 bp, 10-500 bp, 50-250 bp, or at least 5, 10, 20, 30, 40, 50 bp, and / or at most 100, 150, 200, 300, 400, 500, 1000 bp.

[0101] T cells or T lymphocytes belong to a group of white blood cells called lymphocytes that play a role in cell-mediated immunity. T cells originate from hematopoietic stem cells in the bone marrow, mature in the thymus (where T cells originate), and acquire their full function in peripheral lymphoid tissues. During T cell differentiation, CD4 - CD8 - T cells (negative for both CD4 and CD8 co-receptors) are committed to either αβ (alpha beta) or γδ (gamma delta) fate as a result of initial β or γ TCR gene rearrangement. Cells that undergo early β chain rearrangement express a pre-TCR structure on the cell surface, consisting of an intact β chain and a pre-TCR α chain. Such cells express CD4 + CD8 +CD4 TCRs successfully complete γ gene rearrangement before β gene rearrangement, resulting in the rearrangement of the TCR α chain locus and the expression of the αβ TCR on the surface. - CD8 - T cells express γδTCR and CD4 - CD8 - The T cell receptor binds to the CD3 protein to form the T cell receptor complex. T cells, i.e., those expressing either the αβTCR or the γδTCR, express the T cell receptor complex on their cell surface. γδT cells account for approximately 1–5% of the total T cell population. The extracellular region of the T cell receptor chain contains the variable region. The variable regions of the three complementarity-determining regions (CDR1, CDR2, and CDR3) of the T cell receptor chain are located. These regions are typically the most variable and contribute to the diversity among TCRs. The CDR regions are constructed during T cell differentiation, during which the so-called variable (V), diverse (D), and joining (J) gene segments randomly combine to generate diverse TCRs. The constant regions of the T cell receptor chain, i.e., alpha, beta, gamma, or delta chains, do not change significantly. Likewise, the framework regions of the T cell receptor chain, whether it be an alpha, beta, gamma, or delta chain, are largely unchanged.

[0102] "γδ T cells" or "gamma delta T cells" represent a small subset of T cells, but the antigenic molecules that activate them remain largely unknown. Gamma delta T cells can be considered a component of adaptive immunity, in that they rearrange TCR genes to confer binding diversity and exhibit a memory phenotype. However, various subsets can also be considered part of innate immunity, in which restricted TCRs are used as pattern recognition receptors. For example, Vγ9 / Vδ2 T cells are specifically and rapidly activated by a set of nonpeptidic phosphorylated isoprenoid precursors, collectively referred to as phosphoantigens. γδ T cells can be identified using antibodies specific for the γδ T cell receptor. Antibodies suitable for FACS are widely available. Conditions that allow for the selection of negative and / or positive cells are selected and are provided, for example, by the antibody manufacturer. Examples of antibodies that may be suitable are γδTCR-APC (clone B1, #555718) available from BD Pharmingen (BD, 1 Becton Drive, Franklin Lakes, NJ, USA) or pan-γδTCR-PE (clone IMMU510, #IM1418U) available from Beckman Coulter. Furthermore, nucleic acid sequences (or amino acid sequences) corresponding to the γT cell receptor chain and / or the δT cell receptor chain can be determined from such selected cells. Thus, γδ T cells may also be defined as cells containing nucleic acid (or amino acid) sequences corresponding to the γT cell receptor chain and / or the δ2T cell receptor chain.

[0103] Natural killer cells (NK cells), defined as large granular lymphocytes (LGLs), constitute the third cell type differentiated from a common lymphoid precursor cell that gives rise to B and T lymphocytes. NK cells are known to differentiate and mature in the bone marrow, lymph nodes, spleen, tonsils, and thymus, where they then enter the circulation. NK cells do not express the T cell antigen receptor (TCR), the Pan T marker CD3, or the surface immunoglobulin (Ig) B cell receptor; however, they typically express the surface markers CD16 (FcγRIII) and CD56 in humans and NK1.1 or NK1.2 in C57BL / 6 mice. Up to 80% of human NK cells also express CD8.

[0104] The term "conservative substitution" as used herein may refer to the replacement of one or more amino acids in a polypeptide without substantial loss of functionality. It is common general knowledge that certain amino acids can be replaced with other amino acids without loss of activity of the polypeptide. For example, the following amino acids can typically be exchanged with each other: Ala, Ser, Thr, Gly (small aliphatic, non-polar or slightly polar residues) Asp, Asn, Glu, Gln (polar negatively charged residues and their amides) His, Arg, Lys (polar positively charged residues) Met, Leu, Ile, Val (Cys) (large aliphatic non-polar residues) Phe, Tyr, Trp (large aromatic residues) (See, for example,

[67] ).

[0105] Preferred "substitutions" are conservative substitutions, i.e., substitutions in which a residue is replaced by another residue of the same general type. In making these changes, the hydropathic index of an amino acid may be considered (see, e.g.,

[68] ). It is known in the art that certain amino acids can be substituted with other amino acids having a similar hydropathic index or score and still result in a polypeptide with similar biological activity. In making such changes, substitutions of amino acids within a hydropathic index of ±2 are preferred, those within ±1 are more preferred, and those within ±0.5 are even more preferred. Similarly, selected amino acids can be substituted with other amino acids having similar hydrophilicity, as set forth in U.S. Pat. No. 4,554,101. In making such changes, as with the hydropathic index, substitutions of amino acids within a hydrophilic index of ±2 are preferred, those within ±1 are more preferred, and those within ±0.5 are even more preferred.

[0106] As used herein, the term "sequence identity" refers to a measure of the identity of nucleotide or amino acid sequences. Typically, sequences are aligned to obtain the highest degree of match. "Identity" itself has an art-recognized meaning and can be calculated using published techniques. See, for example,

[48] ,

[49] ,

[50] ,

[51] , and

[52] . While there are many methods for measuring identity between two polynucleotide or polypeptide sequences, the term "identity" is well known to those skilled in the art.

[53] Commonly used methods for determining identity or similarity between two sequences include, but are not limited to, those disclosed in

[54] and

[55] . Methods for determining identity and similarity are codified in computer programs, such as NCBI Nucleotide Blast (blastn, https: / / blast.ncbi.nlm.nih.gov / ) with standard settings. Preferred computer program methods for determining identity and similarity between two sequences include, but are not limited to, the GCS program package.

[56] ,

[57]

[0107] By way of illustration, a reference to a nucleotide sequence or amino acid sequence having at least, for example, 95% "identity" to a reference sequence means that the nucleotide sequence or amino acid sequence is identical to the reference sequence, except that there may be up to five point mutations per 100 nucleotides or amino acids in the reference sequence. In other words, to obtain a nucleotide sequence or amino acid sequence that is at least 95% identical to the reference sequence, up to 5% of the nucleotides or amino acids in the reference sequence may be deleted and / or substituted with other nucleotides or amino acids, and / or up to 5% of the total nucleotides or amino acids in the reference sequence may be inserted into the reference sequence. Preferably, sequence identity refers to sequence identity over the entire length of the sequence. It is further understood that when a "sequence" is referred to herein, it generally refers to an actual physical molecule having a particular subunit (e.g., amino acid or nucleotide) sequence.

[0108] In this specification and the claims, the verb "comprise" and its conjugations are used in their open-ended sense, meaning that items following the word are included, but not that items not specifically mentioned are excluded. In addition, the reference to an element by the indefinite article "a" or "an" does not exclude the possibility that more than one of the element is present, unless the context clearly requires that only one of the element is present. The indefinite article "a" or "an" therefore normally means "at least one." [Brief explanation of the drawings]

[0109] [Figure 1] Figure 1A shows an analysis of the Vγ9-BTN2A1 interface to predict affinity-enhancing mutations. The Vγ9-BTN2A1 interface is as predicted by Haddock. BTN2A1 is shown in cartoon form, and the Vγ9 TCR is shown in surface form. Vγ9 residues at the binding interface with BTN2A1 are shown in dark gray. Panel A was generated using PyMol. Figure 1B shows an analysis of the Vγ9-BTN2A1 interface to predict affinity-enhancing mutations. This is a cartoon representation of the Vγ9Vδ2 TCR (pdb 1hxm), with residues predicted to be involved in binding to BTN2A1 shown in dark gray and their side chains represented as stick figures. Panel B was generated using PyMol. Figure 1C shows an analysis of the Vγ9-BTN2A1 interface to predict affinity-enhancing mutations. The amino acid sequences of the CDR residues of the Vγ9 domain are shown in italic font, and the interface residues are shown in black. (D) Analysis of the Vγ9-BTN2A1 interface to predict affinity-enhancing mutations. Limited overlap of the top 25 predicted mutations from four different software packages: EvoEF, Elaspic2 (ddG score and EL2 score), and mCSM-PPI2. Venn diagrams were generated using R studio. [Figure 2]This figure shows that the Vγ9R73Y mutation increases GAB production without affecting activity. (A) GAB expression levels in HEK293F culture medium were determined by Western blot 6 days after transfection. The C-terminal poly-His tag of GAB was used for detection with a 6xHis antibody. Western blots of two separate transfections are shown. (B) GAB yield after purification of GAB without and with the Vγ9R73Y mutation. Significance was calculated using Welch's t-test in GraphPad Prism (v8.3.0). (C) IFNγ release by T cells after overnight coculture with RPMI8226 cells in the presence of GAB and 30 μM pamidronate. IFNγ concentrations in the coculture medium were determined by ELISA and plotted as mean + SD (n = 2). (D) GAB expression levels in HEK293F culture medium were determined by Western blot 6 days after transfection for additional mutations at the Vγ9R73′ position as in A. (E) Relative quantification of GAB expression levels compared to GAB-AJ8-Vγ9R73Y (n=2). [Figure 3] This figure shows that the buried arginine in Vγ9 is also present in half of the Vβ genes. (A) Plot of the sequence probability of all functional Vβ genes consisting of IMGT residues 75-84. The probability plot was generated using WebLogo 3.7.4. (B) Buried arginine "R73" in TRGV9 (pdb 1hxm) and (C) the equivalent arginine in TRBV6 (pdb 2bnu). The figure was generated using PyMol. [Figure 4]Figure 1 shows the effect of a single Vγ9 mutation on GAB expression levels and GAB activity. (A) GAB expression levels in HEK293F medium were analyzed by dot blot using an anti-6xHis antibody. Results are shown as the relative intensity of the dot blot signal compared to that of GAB-AJ8s. Each dot represents the GAB expression level of a HEK293F transfectant compared to GAB-AJ8s on the same dot blot. (B) Relative IFNγ secretion induced by the GAB-AJ8s-Vγ9 mutant compared to GAB-AJ8s. IFNγ was measured after coculture of T cells with either MZ1851RC or SCC9 tumor cells in the presence of GAB expression medium and 30 μM pamidronate. Dots represent individual measurements normalized to the corresponding GAB-AJ8s condition in the same assay. [Figure 5] Figure 1 shows that the Vy9E22W mutation substantially enhances the activity of purified GAB. IFNγ release was measured using ELISA after 24 hours of co-culture of T cells with tumor cells. T cells were co-cultured with three different tumor cell lines at a 1:1 E:T ratio in the presence of different concentrations of GAB and 30 μM pamidronate. [Figure 6] This figure shows that the Vγ9E22W mutation in GAB-AJ8s induced greater tumor cell killing compared to GAB-AJ8s. (A-C) T cells and the indicated luciferase-transduced target cells were cocultured for 20 hours at an effector-target ratio of 10:1 in the presence of GAB and 30 μM pamidronate. The percentage of surviving cells was quantified by measuring luciferase activity, and the signal was normalized to that of T cells with target cells without GAB. Means (+SD) are plotted. n=3. (D) For each GAB titration, EC50 values ​​were calculated using GraphPad Prism 9.3.0. Three EC50 values ​​for GAB-AJ8s and GAB-AJ8sγE22W were plotted, and significance was calculated using a proportion t-test in GraphPad Prism 9.3.0. [Figure 7]This figure shows that the Vγ9E22W mutation substantially enhances the activity of GAB with different CDR3δ sequences. IFNγ release was measured using ELISA after 24 hours of coculture of T cells with tumor cells. T cells were cocultured with two different tumor cell lines at a 1:1 E:T ratio in the presence of different concentrations of GAB and in the absence or presence of 10 or 30 μM pamidronate. The GAB dilution series was as follows: (A) GAB-AJ8s with CDR3δ, which resulted in intermediate activity; (B) GAB-A3s with CDR3δ, which resulted in strong activity; and (C) GAB-LM1s with CDR3δ, which resulted in very low activity. Dots represent the mean of two biological replicates, and error bars represent standard deviation. [Figure 8] Figure 1 shows that the Vγ9E22W mutation significantly enhances the activity of αβ T cells transduced with a functional γ9δ2 TCR. Killing of RMPI8226-lucGFP cells by αβ T cells transduced with four different γ9δ2 TCRs after 20 hours of co-culture. Dots represent the mean of two replicates at different E:T ratios (10:1, 3:1, and 1:1) and in the presence or absence of 30 μM pamidronate. Significance was calculated using one-way ANOVA in GraphPad Prism (v8.3.0). *p<0.05, **p<0.01, ***p<0.005. [Figure 9-1] Figure 9A shows the effect of single Vγ9 mutations at critical residues on GAB activity. Residues were divided into different classes based on the reactivity of the first set of Vγ9 mutations. Critical residues are clustered together on the Vγ9 domain. Figure 9B shows the effect of single Vγ9 mutations at critical residues on GAB activity. The amino acid sequences of the CDR residues of the Vγ9 domain are shown in italic font, Vγ9R73 is underlined, Vγ9E22 is shown in dark gray bold, critical residues are shown in dark gray, and allowed / untested residues are shown in medium gray. [Figure 9-2]Figure 9C shows the effect of a single Vγ9 mutation at a critical residue on GAB activity. Relative IFNγ secretion induced by the GAB-AJ8s-Vγ9 mutant compared to GAB-AJ8s. IFNγ was measured after co-culture of T cells with tumor cells MZ1851RC in the presence of GAB expression medium and 30 μM pamidronate. Dots represent individual measurements normalized to the corresponding GAB-AJ8s condition in the same assay. [Figure 9-3] Figure 9D shows the effect of single Vγ9 mutations at key residues on GAB activity. Top 5 Vγ9 mutations that induced the highest relative IFNγ release. [Figure 10-1] Figure 10A shows that the Vγ9T81H and Vγ9T81W mutations in GAB-AJ8s induced better T cell responses than GAB-AJ8s. IFNγ release was measured using ELISA after 24 hours of coculture of T cells with tumor cells. T cells were cocultured with three different tumor cell lines at an effector-target ratio of 1:1 in the presence of different concentrations of GAB and 30 μM pamidronate. Figure 10B shows that the Vγ9T81H and Vγ9T81W mutations in GAB-AJ8s induced better T cell responses than GAB-AJ8s. T cells were cocultured with luciferase-transduced target cells at an effector-target ratio of 5:1 in the presence of GAB and 30 μM pamidronate for 20 hours. The percentage of viable cells was quantified by measuring luciferase activity, and the signal was normalized to T cells with target cells without GAB. Means (+SD) are plotted. n=2. [Figure 10-2]Figure 10C shows that the Vγ9T81H and Vγ9T81W mutations in GAB-AJ8s induced better T cell responses compared to GAB-AJ8s. For each GAB titration, EC50 values ​​were calculated using GraphPad Prism 9.3.0. The EC50 values ​​of the GAB-AJ8sγT8IH, GAB-AJ8sγT81W, and GAB-AJ8sγT83I mutations were compared to the EC50 value of GAB-AJ8s for significance using a ratio t-test in GraphPad Prism 9.3.0. [Figure 11] Figure 1 shows the effect of a single Vγ9-E22 mutation on GAB activity. (A) IFNγ was measured after coculture of T cells with tumor cells MZ1851RC in the presence of GAB expression medium of the indicated GAB-AJ8s mutants and 30 μM pamidronate. Dots represent measurements of unique cocultures. (B) IFNγ release was measured using ELISA after 24 hours of coculture of T cells with tumor cells. T cells were cocultured with either RPMI8226 or SCC9 at a 1:1 E:T ratio in the presence of different concentrations of the indicated purified GAB and 30 μM pamidronate. [Figure 12] Figure 1 shows that the Vδ2 mutations K53S and K53S enhance the activity of GAB-AJ8s. (A) IFNγ release was measured using ELISA after 24 hours of coculture of T cells with tumor cells. T cells were cocultured with RPMI8226-lucGFP at a 1:1 effector-to-target ratio in the presence of different concentrations of the indicated purified GAB and 30 μM pamidronate. (B) T cells were cocultured with luciferase-transduced RPMI8226 tumor cells at a 5:1 effector-to-target ratio in the presence of GAB and 10 μM pamidronate for 20 hours. The percentage of viable cells was quantified by measuring luciferase activity, and the signal was normalized to that of T cells with target cells without GAB. [Figure 13]Figure 1 shows that the hydrophobic Vδ2-G31 mutation enhances GAB activity. (A) IFNγ was measured after co-culture of T cells with tumor cells MZ1851RC in the presence of GAB expression medium of the indicated GAB-AJ8sδG31 mutants and 30 μM pamidronate. Dots represent measurements of unique co-cultures. (B) IFNγ release was measured using ELISA after 24 h of co-culture of T cells with tumor cells. T cells were co-cultured with either RPMI8226 or SCC9 at a 1:1 E:T ratio in the presence of different concentrations of the indicated purified GAB and 30 μM pamidronate. [Figure 14] Figure 14A shows that combining Vy9E22W with V52G31V or V52K53S in GAB-AJ8s results in superior GAB activity. FNγ release was measured using ELISA after 24 h of coculture of T cells with tumor cells. T cells were cocultured with either RPMI8226-lucGFP or SCC9-lucGFP at a 1:1 E:T ratio in the presence of different concentrations of the indicated purified GAB and 30 μM pamidronate. Figure 14B shows that combining Vy9E22W with V52G31V or V52K53S in GAB-AJ8s results in superior GAB activity. T cells were cocultured with luciferase-transduced RPMI8226 or SCC9 tumor cells at a 5:1 effector-target ratio in the presence of GAB and 10 μM pamidronate for 20 h. The percentage of viable cells was quantified by measuring luciferase activity, and the signal was normalized to T cells with target cells without GAB.

[0110] array

[0111] [Table 1]

[0112] Overview of the V gamma 9 sequence. The CDR regions are shown in italic font, with key residues in bold. The position of the stabilizing residue is underlined in bold (γR73Y mutation).

[0113] SEQ ID NO:1:

[0114] [ka]

[0115] (TRGV9*01 germline used in GAB-AJ8, TEG-MNP2, and TEG-LM1)

[0116] Exemplary sequences with specific CDR regions are shown below, in which position 73, which affects yield, is underlined. Position E22, which affects potency, is shown in bold in dark grey, key residue positions T18, R20, E70, D72, T81, T83, H85 are shown in dark grey, and allowed residues / untested positions S31, G56, V58, S66, 174, K60, E76, T79 are shown in medium grey.

[0117] SEQ ID NO:43: >TRGV9*01

[0118] [ka]

[0119] SEQ ID NO: 2 (GAB-AJ8s = GAB-AJ8 γR73Y Used in Vγ9-R73Y)

[0120] [ka]

[0121] SEQ ID NO: 3 (GAB-AJ8s γE22W , GAB-A3s γE22W , GAB-LM1s γE22W Used in Vγ9-E22W-R73Y)

[0122] [ka]

[0123] SEQ ID NO: 4 (GAB-AJ8s γT81H Used in Vγ9-R73Y-T81H)

[0124] [ka]

[0125] SEQ ID NO: 5 (GAB-AJ8s γT81W Used in Vγ9-R73Y-T81W)

[0126] [ka]

[0127] SEQ ID NO: 6 (γδTCR MNP2 γE22W or TEG-LM1 γE22W used in transduced T cells (Vγ9-E22W)

[0128] [ka]

[0129] Sequence of the Vγ9 acceptor cassette Nucleotide sequence of the Vγ9 acceptor cassettes. These cassettes were subcloned into the GAB expression vector using the BglII and EcoNI restriction sites. The introduced BsmBI acceptor site is highlighted in bold.

[0130] Sequence number 44 (>Vg9_18_22_acc)

[0131] [ka]

[0132] Sequence number 7 (>Vg9_29_33_acc)

[0133] [ka]

[0134] Sequence number 8 (>Vg9_56_60_acc)

[0135] [ka]

[0136] Sequence number 9 (>Vg9_66_70_acc)

[0137] [ka]

[0138] Sequence number 10 (>Vg9_72_76_acc)

[0139] [ka]

[0140] Sequence number 11 (>Vg9_81_85_acc)

[0141] [ka]

[0142] List of human TCR V genes containing arginine corresponding to VγR73 (IMGT numbering V Gamma 9-R80)

[0143] SEQ ID NO: 12 (TRBV2)

[0144] [ka]

[0145] SEQ ID NO: 13 (TRBV6-1)

[0146] [ka]

[0147] SEQ ID NO: 14 (TRBV6-2)

[0148] [ka]

[0149] SEQ ID NO: 15 (TRBV6-3)

[0150] [ka]

[0151] SEQ ID NO: 16 (TRBV6-4)

[0152] [ka]

[0153] SEQ ID NO: 17 (TRBV6-5)

[0154] [ka]

[0155] SEQ ID NO: 18 (TRBV6-6)

[0156] [ka]

[0157] SEQ ID NO: 19 (TRBV6-8)

[0158] [ka]

[0159] SEQ ID NO: 20 (TRBV6-9)

[0160] [ka]

[0161] SEQ ID NO: 21 (TRBV7-2)

[0162] [ka]

[0163] SEQ ID NO: 22 (TRBV7-3)

[0164] [ka]

[0165] SEQ ID NO: 23 (TRBV7-4)

[0166] [ka]

[0167] SEQ ID NO: 24 (TRBV7-6)

[0168] [ka]

[0169] SEQ ID NO: 25 (TRBV7-7)

[0170] [ka]

[0171] SEQ ID NO: 26 (TRBV7-8)

[0172] [ka]

[0173] SEQ ID NO: 27 (TRBV7-9)

[0174] [ka]

[0175] SEQ ID NO: 28 (TRBV10-1)

[0176] [ka]

[0177] SEQ ID NO: 29 (TRBV10-2)

[0178] [ka]

[0179] SEQ ID NO: 30 (TRBV10-3)

[0180] [ka]

[0181] SEQ ID NO: 31 (TRBV11-1)

[0182] [ka]

[0183] SEQ ID NO: 32 (TRBV11-2)

[0184] [ka]

[0185] SEQ ID NO: 33 (TRBV11-3)

[0186] [ka]

[0187] SEQ ID NO: 34 (TRBV14)

[0188] [ka]

[0189] SEQ ID NO: 35 (TRBV15)

[0190] [ka]

[0191] SEQ ID NO: 36 (TRBV19)

[0192] [ka]

[0193] SEQ ID NO: 37 (TRBV24-1)

[0194] [ka]

[0195] SEQ ID NO: 38 (TRBV25-1)

[0196] [ka]

[0197] SEQ ID NO: 39 (TRBV27)

[0198] [ka]

[0199] SEQ ID NO: 40 (TRBV28)

[0200] [ka]

[0201] SEQ ID NO: 41 (TRBV29-1)

[0202] [ka]

[0203] SEQ ID NO: 42 (TRBV30)

[0204] [ka]

[0205] Overview of V-Delta 2 Array The CDR regions are shown in italic font. Important residues are shown in bold.

[0206] SEQ ID NO: 52>TRDV2*03 germline (used as delta chain in GABs AJ8 and A3)

[0207] [ka]

[0208] SEQ ID NO: 53>Vδ2-G31V (used as the delta chain in GAB-AJ8s-δG31V)

[0209] [ka]

[0210] SEQ ID NO: 54>Vδ2-G31M (used as the delta chain in GAB-AJ8s-δG31M)

[0211] [ka]

[0212] SEQ ID NO: 55>Vδ2-K53S (used as the delta chain in GAB-AJ8s-δK53S and GAB-A3s-δK53S)

[0213] [ka]

[0214] SEQ ID NO: 56>Vδ2-K53A (used as delta chain in GAB-AJ8s-δK53A)

[0215] [ka]

[0216] SEQ ID NO: 69>GAB-AJ8s-Vδ2-CDR1_2acc (KpnI+BamHI, double BsmBI stuffer for subcloning)

[0217] [ka]

[0218] Alignment TRDV2

[0219] [ka]

[0220] Vd2 CLY31 Preferred substitutions: VAL, NET, ALA (hydrophobic); alternative substitutions (slightly better compared to wt): ILE, LEU, SER, and THR Vd2 LYS53 Preferred substitutions: CER, ALA

[0221] In cases where there is no match between the above sequence and the corresponding sequence in the sequence listing, the above sequence may be used. Alternatively, the sequence in the sequence listing may be used.

[0222] The following examples describe different embodiments of the present disclosure. Unless otherwise stated, all recombinant DNA techniques are performed according to standard protocols, e.g., as described in

[69] ,

[70] ,

[71] .

[0223] Experimental Section Example 1

[0224] Materials and Methods Molecular modeling and affinity prediction Predictions of Vγ9-BTN2A1 interactions were performed using the docking program HADDOCK [28, 29], using an online server (version 2.2). The input model for the Vγ9 domain was derived from the pdb file 1hxm, B chain residues 1–230 (see, for example, https: / / www.ebi.ac.uk / pdbe / entry / pdb / 1hxm). For BTN2A1, a homology model was created using the structure prediction program phyre2

[30] based on residues 29–248 of the extracellular domain of the uniprot entry Q7KYR7 and used as input for HADDOCK. For each input model, a list of active residues was prepared based on the mutagenesis dates described in the literature [13, 14]. For Vγ9, the list included residues 20, 22, 60, 70, 76, and 85, based on the mutations R20A, E22A, K60A, E70A, E76A, and H85A reported by Rigau et al.

[13] . For BTN2A1, the selected residues were 37, 41, 96, 98, and 107 in the homology model, corresponding to the previously reported mutations R65A, K79A, R124A, Y126A, and E135A

[14] . The largest cluster (cluster 1) had a HADDOCK score of -103.0 + / - 6.5 and contained 69 models; cluster 1_1 was selected for further analysis. The Vγ9-BTN2A1 complex models were examined and analyzed using Open-Source Pymol (Schrodinger, LLC). A list of Vγ9 residues that could potentially bind to BTN2A1 was created based on the following criteria: 1. Vγ9 residues with a distance from the Vγ9-Cα atom to any atom of BTN2A1 <9.0 Å, and 2. Vγ9 residues with surface-exposed side chains. This list was used as input for several programs that predict changes in binding affinity between wild-type and mutant proteins. All prediction programs were run using online web servers with default settings. The programs used were EvoEF [31, 32], ELASPIC2 [33, 34], and mCSM-PPI2

[35] .

[0225] Introduction of selected Vγ9 mutants into GAB expression plasmids To introduce mutations into the Vγ9 domain of GAB, a previously described GAB expression vector

[20] was modified to facilitate rapid introduction of mutations using Golden Gate cloning. For this purpose, the Vγ9 domain in the GAB expression vector was replaced with a so-called "Vγ9 acceptor cassette." A total of six plasmids were constructed, each containing a specific length of DNA within the Vγ9 domain that can be digested with the BsmBI restriction enzyme. The nucleotide sequence of the BsmBI adapter is (A / T)GAGACGTTTCCTGATGGACTGACGTCTCA (SEQ ID NOs: 45 and 46). The sequences of the six different Vγ9 acceptor cassettes are listed in Appendix 1.

[0226] For selected mutations, corresponding acceptor DNA oligo pairs were designed to reintroduce the Vγ9 residue containing a single point mutation. DNA oligos were ordered from Integrated DNA Technologies Europe. Both the sense and antisense oligos had a 5' extension consisting of four nucleotides, compatible with the overhang remaining after BsmBI digestion. After annealing the oligo pairs, these double-stranded DNA fragments with compatible overhangs were ligated to the BsmBI-digested Vγ9 acceptor vector using T4 ligase. After a short incubation at room temperature (5–15 min), the ligation reaction mixture was used to transform E. coli TOP10 cells using a standard heat shock protocol. Transformed bacteria were plated on ampicillin-containing LB agar plates for selection. Single colonies were used to inoculate liquid LB cultures, and plasmid preparations were performed using the Macherey-Nagel NucleoSpin Plasmid EasyPure kit. The purified plasmid was sent for Sanger sequencing to confirm successful introduction of the point mutations.

[0227] Expression of GAB in 293F cells 293F cells were cultured in Gibco Freestyle Expression medium and transfected with DNA:polyethyleneimine (PEI) complexes to express GAB. 25 kDa linear PEI (Polysciences, Germany) was added to the DNA in Freestyle medium at a 3:1 ratio. 1.25 μg of DNA and 3.75 μg of PEI per million cells were premixed in 1 / 10 or 1 / 30 of the culture volume, vortexed briefly, and incubated at room temperature for 10–15 minutes. The DNA:PEI mixture was added dropwise to the culture at a cell density of 10^6 cells / ml. The culture was incubated on a shaking platform at 37°C and 8% CO2 for 6 days. On day 6, the expression medium was collected after centrifugation at 3000 × g for 10 minutes.

[0228] Western blot analysis 293F expression medium was mixed with 4x Laemmli non-reducing sample buffer (Biorad) and incubated at 95°C for 5 minutes. Samples were loaded onto a 4-20% Mini-Protean TGX gel (Biorad). After electrophoresis, the gel was placed on a nitrocellulose membrane (Transblot Turbo, Bioorad), and proteins were transferred to the membrane. The membrane was incubated with a 5% milk powder solution in PBS for 45 minutes at room temperature. The membrane was then incubated with 5 ml of PBS-Tween containing 1 μl of anti-6xHis (BD Pharmingen) and 0.75 μl of goat anti-mouse HRP (Southern Biotech) for 30 minutes at room temperature. The blot was washed extensively with PBS-T, incubated with ECL reagent (Cytiva Amersham) for 30 seconds, and imaged using a ChemiDoc MP Imaging System (Biorad).

[0229] Purification of GAB GAB was purified from the expression medium using a two-step purification protocol previously described

[20] . Briefly, the expression medium was concentrated and diafiltered using a Vivaflow 200 cassette (10 kDa MWCO, Sartorius, Germany), then loaded onto a 1 ml HisTrap column (Cytivia), and His-tagged GAB was eluted using a linear imidazole gradient. The GAB-containing fractions were buffer-exchanged into a low ionic strength buffer, loaded onto a 1 ml HiTrap Q column (Cytivia), and eluted using a linear NaCl gradient. The GAB-containing fractions were concentrated and stored at -80°C in 20 μg aliquots until further use. All buffers were pH 8.0.

[0230] Cells and cell lines Buffy coats obtained from Sanquin Blood Bank (Amsterdam, The Netherlands) were used to isolate PBMCs using Ficoll-Paque Plus (Cytivia) containing 2.5% heat-inactivated pooled human serum (Sanquin Blood Bank, The Netherlands), 1% penicillin / streptomycin (Invitrogen), and 1.7 × 10 3 IU / ml of rhlL-7 and 1.5 × 10 2 T cells were expanded from PBMCs using CD3 / CD28 Dynabeads (Thermo Fisher Scientific) in RPMI 1640 medium (Gibco) containing 1.1U / ml rhlL-15 (both Milteny Biotec, Germany). 48–72 h before T cell assays, the medium was replaced with cytokine-free medium.

[0231] The adherent cell lines SCC9(-lucGFP) and MZ1851RC(-lucGFP) were cultured in DMEM medium (Gibco) supplemented with 10% FCS (Bodinco, The Netherlands) and penicillin / streptomycin (Invitrogen), while RMPI8226(-lucGFP) was cultured in RPMI 1640 medium supplemented with 10% FCS and 1% penicillin / streptomycin.

[0232] TEGs were generated as previously described

[19] . Briefly, T cells were transduced with retroviral particles containing DNA encoding the γTCR chain, δTCR chain, and antibiotic resistance genes. After antibiotic selection, TEGs were expanded using a rapid expansion protocol and then subjected to functional testing.

[0233] IFNγ release assay using GAB expression medium 293F expression medium containing GAB (mutants) was serially diluted (3-fold dilution series) in complete RPMI1640 medium and 50 μl was transferred to a 96-well U-bottom plate.

[0234] The remaining T cells were resuspended in complete RPMI 1640 at a concentration of 10^6 cells / ml, and 50 μl was added to the GAB cells in a 96-well U-bottom plate. Tumor cell lines were resuspended in complete RPMI 1640 medium supplemented with 120 μM pamidronate at a concentration of 10^6 cells / ml for SCC9 or 5 x 10^5 cells / ml for MZ1851RC, and 50 μl of the tumor cell line suspension was added to wells containing GAB and T cells. Cocultures were incubated at 37°C and 5% CO2 for 20 hours, cells were pelleted using centrifugation, and IFN-γ concentrations were determined using a Human IFN gamma Uncoated ELISA Kit (Invitrogen). Each plate contained dilutions of the GAB-AJ8 mutant expression medium as well as GAB-AJ8s and GAB-LM1 expression medium as controls. The IFNγ concentration of each GAB-AJ8 mutant was normalized to the corresponding dilution of GAB-AJ8s.

[0235] IFNγ release assay using purified GAB 5x10^4 effector cells and 5x10^4 target cells or 2.5x10^4 MZ1851RC cells were co-incubated in the presence or absence of GAB (at different concentrations as indicated) at 37°C and 5% CO2 for 20 hours. To increase intracellular phosphoantigen levels, pamidronate was added to the co-cultures at the concentrations indicated in the figure legends. Supernatants were collected after 20 hours, and IFN-γ levels were determined using an IFN-gamma human uncoated ELISA kit (Invitrogen).

[0236] Luciferase-based killing assay 10^4 target cells stably expressing luciferase were incubated with T cells at a target cell / effector cell ratio of 1:5 in the absence or presence of pamidronate (Calbiochem, USA) at different GAB concentrations (as indicated). Alternatively, 10^4 RPMI8226-lucGFP cells were incubated with TEG at different target cell / effector cell ratios in the absence or presence of pamidronate. After 20 hours, firefly luciferin (Promega, USA) was added to the wells (125 μg / ml), and bioluminescence was measured using a SoftMax Pro plate reader. Signals in treated wells were normalized to the signal measured with target and T cells alone, which was assumed to represent 100% viable cells.

[0237] result Prediction of Vγ9 mutations that increase affinity for BTN2A1 The first step in our approach to improving the affinity of the Vγ9 domain for BTN2A1 was to generate a molecular interaction model of the Vγ9 domain-BTN2A1 complex. To this end, we used the affinity-affecting mutations reported by Rigau et al. and Karunakaran et al. [13, 14] as input for the docking program HADDOCK [28, 29]. As expected, all input residues were located at the predicted interface of the complex, and the overall interdomain orientation was similar to the model presented by Karunakaran et al.

[14] . We identified 16 residues within the Vγ9 domain that met two criteria for further investigation: they were in close proximity to BTN2A1, with a distance of ≤9.0 Å from Vγ9-Cα to BTN2A1 and possessed surface-exposed side chains (Figure 1A). Mutating and testing all 16 residues to other amino acids using experimental methods would be a daunting task, resulting in 304 possibilities. Therefore, we investigated whether computational affinity prediction methods could reduce the number of mutations tested. Three different affinity prediction methods available online, EvoEF, ELASPIC2, ​​and mCSM-PPI2, were used to predict the effects of single mutations at any of the 16 residues. For ELASPIC2, ​​we used both ddG and EL2 scores in our further analysis. We found that predictions of the most favorable mutations differed substantially between affinity prediction methods. To limit the analysis to the most favorable mutations, we used the top 25 of each prediction output for further comparison. Within this mutation set, 15 of the 16 interface residues were present, and only T79 was not present in any of the top 25 with any mutations, likely due to its large distance of 9.0 Å to BTN2A1. Only 13 mutations were shared among the top 25 of at least two predicted outputs, and three of these mutations were included in the top 25 of three predicted outputs ( Fig. 1B ), leaving 84 unique mutations for further testing.The limited overlap in predicted affinity-enhancing mutations is due, in part, to the fact that the input models were not experimentally determined and therefore have a relatively high degree of uncertainty. The quality of the input models is crucial for the performance of affinity prediction methods.

[36] A second reason for the differences in the top 25 variants among different prediction methods is the differences in the algorithms underlying the prediction methods.

[0238] Mutant Vγ9 R73Y increases GAB production yield by 2-fold While the primary goal was to increase the potency of the Vy9 domain in Vy9V52 TCR-based therapeutics, because low protein yields can be a limiting step for translating soluble therapeutic formats like GAB into the clinic, we were also interested in mutations that would increase the production yield of soluble Vy9V52 TCR. While examining the modeled Vy9-BTN2A1 interface for Vy9 residues involved in the interaction with BTN2A1, we found that the arginine Vy9, whose charged side chain is inserted into the hydrophobic core of the Vy9 domain, R73 We found that such buried charged residues can be present in proteins, but in many cases they form salt bridges with oppositely charged residues or have important functions in protein function

[37] . In general, buried ionizable amino acids are tolerated, but they tend to reduce protein stability. We therefore substituted arginine with bulky tyrosine and observed that expression increased in the initial study, as analyzed by Western blot (Figure 2A). Because the change in Western blot intensity of GAB may not translate to an increase in protein yield after purification, we investigated the effect of tyrosine ... R73Y We performed parallel expression and purification analyses of GAB with and without the mutation, which significantly increased the yield of GAB after purification by two-fold (Figure 2B).

[0239] Vγ9 R73Y The mutation is located within / close to the BTN2A1 binding site, thus Vγ9R73Y Mutations can affect the activity of GAB. Since increasing GAB yield is only useful if activity is not lost, we used an IFNγ release assay targeting the RPMI8226 cell line to determine the activity of GAB-AJ8. wt and GAB-AJ8 γR73Y We compared the T cell activation abilities of GAB-AJ8 and GAB-AJ8. wt and GAB-AJ8 γR73Y No difference in activation was observed between Vγ9 and Vγ9. R73Y This demonstrates that mutations can be used without negatively impacting GAB efficacy (Figure 2C).

[0240] Another hydrophobic residue at position Vγ9-R73 is Vγ9 R73Y To test whether Vγ9-R73 could induce the increased expression seen in GAB-AJ8, we substituted Vγ9-R73 with the hydrophobic residues alanine, valine, isoleucine, leucine, histidine, phenylalanine, tryptophan, and methionine in GAB-AJ8 and assessed their expression levels in small-scale HEK293F cultures using Western blot analysis (Figure 2D). Although expression levels were reasonably good, none of the other substitutions in Vγ9-R73 could induce the increased expression seen in GAB-AJ8. R73Y did not result in higher expression levels compared with (Fig. 2E).

[0241] Although R73 in Vγ9 does not occur at the equivalent position in other human Vγ genes, we analyzed human Vβ genes deposited at IMGT

[40] and found that an arginine is present at this position in more than half of the functional Vβ genes (Table 1, Figure 3A). Comparison of the structures between the TCR-bearing TRGV9 and TRBV6 showed that the arginine is buried in the hydrophobic core of the variable domain in both the γ and β chains (Figures 3B and 3C). Similarly, mutating this arginine to tyrosine in the Vβ gene would most likely result in a more stable β chain.

[0242] [Table 2]

[0243] A limited number of predicted Vγ9 mutations enhance T cell activation relative to Vγ9 To assess the impact of predicted mutations on the activity of the Vγ9Vδ2 TCR, we utilized the recently developed GAB

[20] . All 84 unique mutations from the prediction output were assembled into the Vγ9 chain of TCR AJ8 using the Golden Gate assembly strategy

[41] . R73Y This was introduced as an additional mutation following the Vγ9 mutation, which enhances expression. R73Y GABs with the α-Gamma variant are indicated by adding an "s" after the TCR name, such as GAB-AJ8s. After sequence verification, the plasmids were used to transiently transfect HEK293F cells for GAB production. Five days after transfection, expression media was collected and the expression levels of GAB were determined using dot blot analysis. Although all mutants were expressed to levels comparable to wild-type GAB, the dot blot signals varied between replicates, and further analysis of expression levels was not considered appropriate (Figure 4A).

[0244] We tested the T cell activation ability of GAB mutants in cocultures of T cells with tumor cells MZ1851RC or SCC9 using expression medium containing GAB. 30 μM pamidronate was included in the cocultures to boost intracellular phosphoantigen levels in tumor cells. IFNγ release was used as a marker of T cell activation and was measured using an IFNγ ELISA. Following the AJ8 mutant and AJ8s GAB

[20] , we also included a less activating GAB based on the Vγ9Vδ2 TCR LM1 as a negative control

[19] . To enable comparison of results between different assays, IFNγ release was normalized to the GAB-AJ8s condition on the same assay plate. Nearly half of the mutants, similar to GAB-LM1s, did not or poorly induce IFNγ release by T cells. As expected, except for the mutation E70L, which had activity comparable to that of GAB-AJ8s, mutations introduced into Vγ9 residues identified by Rigau et al. as essential for BTN2A1 binding, such as R20, E70, and H85

[13] , also abolished T cell activation in this assay. Beyond these previously reported residues, we found additional Vγ9 residues in our screens where mutagenesis strongly reduced T cell activation potential with most or all of the substitutions tested. These newly identified Vγ9 residues important for interaction with BTN2A1 were, in numerical order, T18, E22, D72, T81, and T83. For other Vγ9 residues tested in our screen, such as S31, S66, and I74, none of the mutations introduced reduced the ability of GAB to activate T cells, making it unlikely that these residues are essential for interaction with BTN2A1 (Figure 4B).

[0245] Of the 84 mutants we tested in our GAB screen, only a few induced more T cell activation than GAB-AJ8s. One of these mutants, Vγ9 E22Wshowed the most potent induction of T cell activation in the IFNγ release assay, with a mean fold increase of 5.6. Some other mutations, such as G56N, V58F, S66K, S66N, E70L, and I74Q, induced slightly higher T cell activation, with increases between 1.2 and 2.3-fold, with greater variability between assays (Figure 4B).

[0246] Mutant Vγ9 E22W increases the efficacy of GAB by 10-fold In our screening method, we used expression media containing GAB to identify potential activity-enhancing mutations. One limitation of this screening method is that it could not correct for differences in GAB concentration; therefore, the observed differences in cytokine secretion may simply reflect differences in protein yield. For a more reliable comparison between GAB-AJ8s and the Vy9 mutants, we therefore expressed and purified selected Vy9 mutants, focusing on mutants that showed enhanced activity during screening: E22W, G56N, V58F, S66K, E70L, and I74Q. We used this purified GAB set in a T cell activation assay to titrate the GAB mutants, allowing for a more thorough evaluation of potency differences.

[0247] First, we evaluated IFNγ release from T cells co-cultured with tumor cell lines and GAB in the presence of 30 μM pamidronate. As seen in the initial screening, GAB-AJ8 γE22W induced higher levels of IFNγ compared with GAB-AJ8s and was also able to induce T cell activation at lower GAB concentrations in all three cell lines tested. The other GAB-Vγ9 mutants had activity comparable to that of GAB-AJ8s (Fig. 5).

[0248] Next to IFNγ release, we were also interested in examining the effect of Vγ9 mutants on tumor cytotoxicity induced by T cells. Luciferase-expressing tumor cell lines were co-cultured with T cells, GAB, and 30 μM pamidronate, and after 20 hours, luciferase activity was used to determine the percentage of surviving cells as a readout. Again, GAB-AJ8 γE22W The Vγ9 mutation was superior, increasing activity by more than 10-fold in all three tumor cell lines (Figures 6A-6C). The other Vγ9 mutations did not consistently enhance potency across different tumor cell lines (Figures 6A-6C), but this was more evident when calculating EC50 values. γE22W Analysis of the difference in EC50 values ​​between GAB-AJ8 γE22W showed a significant increase in efficacy at 100 mg / kg / day, with a mean difference of 15.5-fold (95% CI, 5.8-41.9) (Figure 6D).

[0249] Mutant Vγ9 E22W reduces dependence on pamidronate for tumor cell recognition One of the major obstacles to Vγ9Vδ2 TCR-based therapy is the low level of phosphoantigens in tumor cells, which leads to suboptimal activation of Vγ9Vδ2 T cells or Vγ9Vδ2-TCR-based therapy. Many current strategies require the co-administration of aminobisphosphonates, such as pamidronate or zoledronate, to increase intracellular levels of phosphoantigens [8]. In a previous report, we observed differences in the pamidronate dependence of GAB between tumor cell lines: SCC9 required high concentrations of pamidronate to be targeted by GAB, while MZ1851RC was already recognized without the addition of pamidronate

[20] . Here, we investigated the effect of GAB-AJ8s on T cell activation by assessing T cell activation in the absence and presence of 10 and 30 μM pamidronate. γE22WWe tested the effect of pamidronate on the activation ability of GAB-AJ8s. Consistent with previous findings, GAB-AJ8s failed to induce any T cell activation when combined with the highly pamidronate-dependent cell line SCC9. However, all three pamidronate conditions inhibited GAB-AJ8s activation. γE22W GAB-AJ8s could induce T cell activation, and there was no significant difference in IFNγ release between the pamidronate conditions (Figure 7A). In contrast to previous results, using MZ1851RC as a target cell line, we were unable to confirm that any T cell activation was induced by GAB-AJ8s without the addition of pamidronate, whereas at 30 μM pamidronate, GAB-AJ8s induced T cell activation. Similar to what was seen with SCC9, GAB-AJ8s γE22W Using MZ1851RC, we also induced T cell activation in all three pamidronate conditions. With this target cell line, we observed a significant effect of pamidronate concentration, with IFNγ increasing more than two-fold at the highest GAB concentration (15 μg / ml) between the no pamidronate and 30 μM pamidronate conditions (Figure 7A).

[0250] Previously, we have shown that CDR3δ can have a major impact on the activation capacity of GAB. GAB-AJ8s has intermediate potency, whereas two other reported Vγ9Vδ2 TCRs used in the GAB format, GAB-CI5 and GAB-A3, showed high potency

[20] . E22W To test whether the mutation also enhanced the potency of GAB-A3, we performed GAB-A3s γE22W We generated a mAb containing GAB-AJ8 and compared its efficacy with GAB-A3s in a T cell activation assay. Here, we again used SCC9 and MZ1851RC as target cell lines and incubated them with T cells, three concentrations of pamidronate, and a dilution series of GAB. Similar to the results seen with GAB-AJ8, GAB-A3s did not induce any T cell activation when combined with SCC9, whereas GAB-A3s γYE22W However, GAB-AJ8s induced T cell activation (Fig. 7B). YE22Wand GAB-A3s γE22W No clear difference in activation ability was observed between GAB-A3s and MZ1851RC (Figures 7A and 7B). When MZ1851RC was used as a target cell line, GAB-A3s induced T cell activation in the presence of 10 and 30 μM pamidronate, but GAB-A3s γYE22W induced higher levels of IFNγ release at each pamidronate concentration (Fig. 7B).

[0251] GAB-AJ8s γE22W and GAB-A3s γE22W The limited difference in T cell activation capacity between GAB-LM1s and GAB-LM1s raises the question of whether the CDR3δ sequence is important at all. To investigate this, we used the less functional Vγ9Vδ2 TCR LM1

[19] and compared GAB-LM1s and GAB-LM1s using the tumor cell lines SCC9 and MZ1851RC as described above. γE22W We compared the T cell activation abilities of GAB-LM1s and GAB-LM1s. Consistent with previous observations, GAB-LM1s did not induce IFNγ release by T cells under either condition (Figure 7C)

[20] . 9E22W Introduction of the mutation slightly increased INFγ release by T cells in both tumor cell lines SCC9 and MZ1851RC, regardless of pamidronate concentration (Fig. 7C). However, GAB-LM1s γE22W and the more powerful GAB-AJ8s γE22W and GAB-A3s γE22W There is a clear difference in T cell activation capacity between Vγ9 and GAB, which suggests that the efficacy of GAB is E22W This confirms that after the introduction of the mutations, the dependence on the composition of CDR3δ was still evident, although not as clear as in GAB, which is composed of the wild-type Vγ9 domain.

[0252] Mutant Vγ9 E22W increases the potency of Vγ9Vδ2 TCR-transduced αβ T cells Vγ9 E22WTo investigate whether the mutation also increases the potency of Vγ9Vδ2 TCR-transduced αβ T cells [19, 21], we used retroviral transduction to investigate whether the Vγ9Vδ2 TCR-transduced αβ T cells also increase the potency of Vγ9Vδ2 TCR-transduced αβ T cells [19, 21]. E22W We generated αβ T cells expressing Vγ9Vδ2 TCRs with or without mutations. Four different Vγ9Vδ2 TCRs were transduced: the low-activating Vγ9Vδ2 TCR LM1 (wt and Vγ9 E22W ) and functional Vγ9Vδ2 TCR MNP2 (wt and Vγ9 E22W These transduced T cells were assessed for their killing capacity using luciferase-transduced RPMI8226 cells. As expected, Vγ9Vδ2 TCR LM1wt did not induce any killing, whereas functional Vγ9Vδ2 TCR MNP2wt killed approximately 60% of RPMI8226 cells. Similar to GAB-induced activation, Vγ9Vδ2 TCR LM1wt YE22W showed little activity, whereas Vγ9Vδ2 TCR MNP2 γE22W significantly increased the killing capacity of the transduced T cells, killing more than 90% of RPMI8226 cells (Fig. 8).

[0253] Screening for additional enhancing mutations at key residues Screening based on affinity prediction yielded only one Vγ9 mutant with significantly higher activation capacity using purified GAB. However, the results of screening this Vγ9 mutant panel provided further insight into the binding interface (Figure 4B). Several Vγ9 residues could be mutated to various different amino acids without altering GAB efficacy. These residues, S31, G56, V58, S66, and I74, are called permissive residues and are located at the edge of the BTN2A1 binding interface on the Vγ9 domain (Figure 9A). Another group of residues, called critical residues, T18, R20, E70, D72, T81, T83, and H85, are not amenable to extensive amino acid substitutions and are therefore strongly involved in the interaction with BTN2A1. These residues cluster together on the Vγ9 domain (Figure 9A) and include residues R20, E70, and H85, previously identified as important for interaction with BTN2A1

[13] .

[0254] To investigate whether additional potency-enhancing mutations could be identified at these critical residues, we generated a new Vγ9 mutant panel in GAB-AJ8s encompassing the remaining amino acid substitutions not tested in the initial screen. Again, many of the introduced mutations resulted in activation of the less functional Vγ9Vδ2 TCR LM1s similar to GAB (Figure 9C). However, two mutations, Vγ9 mutation T81H (average 4.2-fold difference) and T83I (average 8.8-fold difference), had significantly higher IFNγ production compared to GAB-AJ8s, while several other mutations, Vγ9 mutations T81I, T81W, and T83L, induced a 2-fold increase in IFNγ production (Figure 9D).

[0255] Mutant Vγ9 T81H and Vγ9 T81W increases the efficacy of GAB by two-fold The five selected GAB-AJ8s-Vγ9 mutants, T81H, T81I, T81W, T83I, and T83L, were expressed in HEK293F cells and purified from the expression medium using a two-step purification strategy. The potency of the GAB-Vγ9 mutants was evaluated in cocultures of T cells and tumor cell lines in the presence of 30 μM pamidronate. γT81H and GAB-AJ8s γT81W showed a slight increase in potency compared to GAB-AJ8s when assessed for IFNγ secretion by T cells (Figure 10A). Three other GAB-Vγ9 mutants, Vγ9T81I, T83I, and T83L, were equally or less potent than GAB-AJ8s (Figure 10A).

[0256] When T cell-mediated tumor cell killing was assessed, the same trend was observed (Figure 10B). To formally assess the difference in efficacy, GAB-AJ8s, GAB-AJ8s, and GAB-AJ8s were tested in each individual experiment. γT81H , GAB-AJ8s γT81W , and GAB-AJ8sγ T83I The EC50 value of GAB-AJ8s was determined by paired analysis. γT81H and GAB-AJ8s γT81W demonstrated a significant increase in potency compared to GAB-AJ8s, with EC50 values ​​differing by an average of 2.4-fold and 2.8-fold, respectively, whereas GAB-AJ8s γT83I This latter observation is consistent with the fact that screening with crude GAB expression media did not reveal any difference in efficacy between GAB-AJ8s and GAB-AJ8s (Fig. 10C). γT83I This was unexpected because induced the highest fold difference in relative IFNγ secretion. As previously described, differences in expression of GAB aggregates in the crude expression medium can result in false-positive signals.

[0257] In summary, the inventors have surprisingly found that mutant Vγ9 R73YWe characterized that this nucleotide sequence can stabilize protein expression, e.g., enhance GAB yield and expression, and is most likely to stabilize TCRs in the TEG format. We found that arginine is present at this position in more than half of functional Vβ genes, suggesting that this strategy could also be used to stabilize Vβ genes for diagnostic and therapeutic purposes. We confirmed that mutations introduced into Vγ9 residues identified as essential for BTN2A1 binding, such as R20, E70, and H85

[13] , also induce low T cell activation, but interestingly, Vγ9-E70 can be replaced with leucine without any loss of GAB potency. We applied these findings to the GAB-AJ8s γE22W The present inventors have also expanded on this by showing that GAB-AJ8s has 10-fold increased activity in both IFNγ release and tumor killing compared to GAB-AJ8s. γE22W We showed that can induce T cell activation in the absence of pamidronate, and that this activation is further enhanced by increasing pamidronate concentrations.

[0258] Furthermore, Vγ9 E22W Introducing mutations into the more potent GAB-A3s similarly increased the activity of this GAB. αβ T cells transduced with a bispecific GAB format followed by a Vγ9Vδ2 TCR exhibited similar behavior, i.e., Vγ9 combined with a functional Vγ9Vδ2 TCR. E22W The mutation increased the killing capacity of these transduced αβ T cells. E22W The mutation has the potential to enhance Vy9V52 TCR-based immunotherapy.

[0259] Finally, we identified two additional mutations that increased the potency of GAB-AJ8 by 2-fold: Vγ9 T81H and Vγ9 T81W was identified.

[0260] conclusion We identified four mutations within the Vγ9 domain that favor GAB. R73Y increases GAB production yield by 2-fold, presumably by stabilizing the Vγ9 domain. This mutation is most likely also useful for engineering approaches using β-strands. E22W increases the efficacy of GAB by 10-fold and enhances other Vγ9Vδ2-TCR-based therapeutics such as TEG. T81H and Vγ9 T81W increases the potency of GAB by a factor of 2. Thus, the relationship between the substitution of specific residues and the effect of said substitution on the BTN2A1 binding ability or stability of peptides according to the present disclosure is shown in Tables 2 and 3.

[0261] [Table 3]

[0262] [Table 4] Example 2

[0263] Substitution of Vγ9-E22 with phenylalanine increases the potency of GAB-AJ8s Above, we have described Vγ9 E22W We identified position Vγ9 as one of the mutations that increased the potency of GAB and TEG, however, the number of substitutions tested at this position was limited. Using the Golden Gate cloning strategy described above, we identified position Vγ9 as one of the mutations that increased the potency of GAB and TEG. E22 This number was increased by introducing eight additional substitutions in the GAB-AJ8s variants. These GAB-AJ8s variants were expressed in HEK293F cells, and the expression medium was used in a T cell activation assay. After 20 hours of co-culture of T cells, MZ1851RC tumor cells, 30 μM pamidronate, and GAB expression medium, IFNγ levels in the co-culture medium were determined using ELISA. These eight GAB-AJ8s-Vγ9 variants were E22Among the mutants, one in particular, GAB-AJ8s-Vγ9 E22F However, GAB-AJ8s-Vγ9 outperformed GAB-AJ8s (Figure 11A). Several other hydrophobic substituents, such as alanine, leucine, isoleucine, methionine, and tyrosine, resulted in similar or slightly reduced activity compared to GAB-AJ8s, whereas histidine at position 22 completely abolished activity (Figure 11A). However, the use of unpurified GAB expression medium yielded slightly skewed results because no correction was made for differences in expression levels or aggregation. Therefore, GAB-AJ8s-Vγ9 E22F The purified GAB-AJ8s and GAB-AJ8s were expressed and purified using an IFNγ release assay using RPMI8226 and SCC9 tumor targets. γE22W Substitution of glutamic acid at position 22 with phenylalanine resulted in more potent GAB-AJ8s, but GAB-AJ8s γE22W The performance was not as good as that of the previous method (Figure 11B).

[0264] Certain Vδ2 variants enhance the potency of GAB-AJ8s Recently, affinity-enhancing mutations at residue K53 of Vδ2-CDR2 have been described [72 and patent document WO2023102615A1]. To determine whether mutations at this position also enhance the potency of GAB-AJ8s, the mutations K53S and K53A were introduced into the AJ8 δ chain and GAB-AJ8s was potently inhibited. δK53S and GAB-AJ8s δK53A was expressed in HEK293F and purified as described above. Purified GAB was used in activity assays using RPMI8226 as the tumor target. Consistent with the reported increased affinity of these Vδ2-K53 mutants, GAB-AJ8s δK53S and GAB-AJ8s δK53A Both V52-K53 mutants were more potent than GAB-AJ8s in inducing T cell activation as measured by IFNγ release (Fig. 12A) and in T cell-mediated tumor cell killing (Fig. 12B). No substantial differences in potency were observed between the two V52-K53 mutants in these assays.

[0265] The influence of residues within V52-CDR1 has also been previously investigated by limited mutagenesis or by combining human and primate Vγ9 and Vδ2 chains [73, 74]. V52-CDR1 is highly conserved among primates, with the exception of V52 residue 31, which is glycine in humans (Homo sapiens), chimpanzees (Pan troglodytes), and some gorillas (Gorilla gorilla), but serine or threonine in other primate species (see "Alignment TRDV2" disclosed herein). To better understand which amino acids at V52 residue 31 could generate functional GAB-AJ8s, we introduced mutations at V52 residue 31 into the GAB expression vector harboring GAB-AJ8s-V52-CDR1_2_acc using Golden Gate cloning as described above for the Vγ9 mutant. After sequence verification, the expression plasmid was used to transfect HEK293F cells for GAB production. After 5 days of expression, supernatants were collected and used in co-culture assays with donor-derived T cells and the MZ1851RC cancer cell line to assess dose-dependent T cell activation by GAB. GAB-AJ8s, a small polar amino acid found in the primate Vδ2 chain, is a nucleotide sequence encoding GAB-AJ8s. δG31S and GAB-AJ8s δG31TSubstitution of Vδ2-G31 resulted in similar activity compared to GAB-AJ8s, whereas using bulkier polar amino acids such as asparagine or glutamine, or charged amino acids such as aspartic acid, reduced activity (Figure 13A). Substitution of Vδ2-G31 with any hydrophobic amino acid (alanine, valine, leucine, isoleucine, or methionine) resulted in similar or increased potency of GAB-AJ8s variants compared to GAB-AJ8s, indicating that hydrophobic amino acids are preferred at this position (Figure 13A). To confirm that this increase in potency was due to the introduced mutations and not differential expression, small or large hydrophobic amino acids, valine or methionine, respectively, were used as a substitute for glycine. These GAB-AJ8s variants, GAB-AJ8s δG31V and GAB-AJ8s δG31M were expressed, purified, and used in T cell activation assays together with GAB-AJ8s to evaluate their efficacy under controlled conditions using RPMI8226 and SCC9 tumor targets. δG31V and GAB-AJ8s δG31M Both were more potent than GAB-AJ8s, as indicated by increased IFNγ release by T cells at low GAB concentrations after co-culture with tumor cells (Figure 13B). In conclusion, mutating V52-G31 to a hydrophobic amino acid will yield increased potency for GAB, or any other V52-based therapeutic.

[0266] Vγ9 E22W Combination of GAB with a potency-enhancing Vδ2 mutant yields GAB with superior efficacy As shown above (Fig. 7), Vγ9 E22W not only increases the potency of GAB-AJ8s but also that of GAB-A3s, due to the inherently more potent TCR A3 [75, 76]. However, TCR-A3 is composed of germline Vδ2-CDR1 and Vδ2-CDR2 sequences, and therefore affinity-enhancing mutations in these regions are not readily apparent in Vγ9. E22WThe question remains: will combining GABs with Vγ9 and Vδ2 produce a GAB with even greater potency? To investigate this, GABs with affinity-enhancing mutations in both the Vγ9 and Vδ2 chains were expressed, purified, and tested for their ability to activate T cells when co-cultured with tumor cells, along with single mutations. The GAB panel tested included GAB-AJ8s, GAB-AJ8s, and GAB-AJ8s. γE22W , AJ8s δG31V , GAB-AJ8s γE22WδG31V , GAB-AJ8s δK53S , and GAB-AJ8sγ E22WδK53S First, IFNγ levels after 20 hours of co-culture were measured using ELISA. All single mutant GAB-AJ8s γE22W , AJ8s δG31V , and GAB-AJ8s δK53S were equally potent and, as established above, were an order of magnitude more potent than GAB-AJ8s (Figure 14A). γE22WδG31V and GAB-AJ8s γE22WδK53S The combined Vγ9 and V62 mutations in GAB-AJ8s resulted in a GAB variant that was 10-fold more potent than the single mutants. These differences were observed in both hematological and solid tumor cell lines, RPMI8226 and SCC9, respectively. Finally, T cell-induced cytotoxicity was determined using luciferase-expressing tumor cells. Consistent with the results obtained using the IFNγ release assay, GAB-AJ8s γE22WδG31V and GAB-AJ8s γE22WδK53S was again over 10-fold more potent than the single Vγ9 or V62 mutants and over 100-fold more potent than GAB-AJ8s in T cell-mediated killing of RPMI8226 and SCC9 tumor cells (Figure 14B).

[0267] conclusion Affinity maturation has been used for decades to improve the efficacy and potency of therapeutic proteins such as antibodies and αβ TCRs. Here, the affinity of the Vγ9 chain for BTN2A1 and the Vδ2 chain for BTN3A was enhanced by mutagenesis. Combining the most potent Vγ9 chain mutant, E22W, with the Vδ2 chain mutants G31V or K53S in a GAB format generated a GAB with unprecedented potency. Potentially, triple or quadruple mutants combining the Vγ9 mutations E22W and / or T81H / W with the Vδ2 mutations at positions G31 and K53 described herein will generate even more potent GABs or TEGs.

[0268] (References) TIFF2026501215000057.tif224150TIFF2026501215000058.tif224149TIFF2026501215000059.tif187150

Claims

1. 1. A butyrophilin subfamily 2 member A1 (BTN2A1)-binding peptide comprising an amino acid sequence that binds to BTN2A1, wherein the amino acid sequence has at least 70% sequence identity with SEQ ID NO: 1, and the amino acid sequence is: - tryptophan or phenylalanine at the position corresponding to position 22 in SEQ ID NO: 1, - a tyrosine at the position corresponding to position 73 as set forth in SEQ ID NO: 1, or a conservative substitution thereof selected from tryptophan and phenylalanine, and / or - histidine, or a conservative substitution thereof selected from arginine and lysine, or a conservative substitution thereof selected from tryptophan, or phenylalanine and tyrosine, at the position corresponding to position 81 of SEQ ID NO: 1; A butyrophilin subfamily 2 member A1 (BTN2A1) binding peptide comprising:

2. - threonine at the position corresponding to position 18 as set forth in SEQ ID NO: 1, or a conservative substitution thereof selected from alanine, serine, glycine, and leucine; - arginine at the position corresponding to position 20 as set forth in SEQ ID NO: 1, or a conservative substitution thereof selected from histidine and lysine; - glutamic acid or a conservative substitution thereof selected from aspartic acid, asparagine, and glutamine, or leucine or a conservative substitution thereof selected from methionine, isoleucine, valine, and cysteine, at the position corresponding to position 70 of SEQ ID NO: 1; - aspartic acid or asparagine at the position corresponding to position 72 in SEQ ID NO: 1, - threonine at the position corresponding to position 81 of SEQ ID NO: 1, or a conservative substitution thereof selected from alanine, serine, and glycine, or a conservative substitution thereof selected from histidine, arginine, and lysine, or a conservative substitution thereof selected from tryptophan, phenylalanine, and tyrosine; - threonine at the position corresponding to position 83 of SEQ ID NO: 1, or a conservative substitution thereof selected from alanine, serine, and glycine, or isoleucine, or a conservative substitution thereof selected from methionine, leucine, valine, and cysteine, and / or - histidine at the position corresponding to position 85 of SEQ ID NO: 1 2. The BTN2A1-binding peptide of claim 1, comprising:

3. 3. The BTN2A1-binding peptide of claim 1 or 2, wherein the amino acid sequence has at least 80%, 90%, or 95% sequence identity with SEQ ID NO:

1.

4. 4. The BTN2A1-binding peptide of claim 1, which is a T cell receptor gamma chain domain, preferably a T cell receptor gamma 9 chain domain.

5. The BTN2A1-binding peptide is combined with a BTN3A1 and / or BTN3A2 and / or BTN3A3-binding peptide comprising an amino acid sequence that binds to butyrophilin subfamily 3 member A1 (BTN3A1) and / or BTN3A2 and / or BTN3A3, the amino acid sequence having at least 70% sequence identity to SEQ ID NO: 52, the amino acid sequence being: - valine, methionine, alanine, isoleucine, or leucine at the position corresponding to position 31 of SEQ ID NO: 52, and / or - serine or alanine at the position corresponding to position 53 of SEQ ID NO: 52 The invention is characterized in that it comprises 5. A BTN2A1-binding peptide according to any one of claims 1 to 4, wherein the BTN3A1 and / or BTN3A2 and / or BTN3A3-binding peptide is a δ chain domain of a T cell receptor, preferably a δ2 chain domain of a T cell receptor.

6. A BTN2A1-binding peptide described in any one of claims 1 to 5, which is contained in an (exogenous) immune receptor or its extracellular domain, preferably a γδ T cell receptor or its extracellular domain, more preferably a γ9δ2 T cell receptor or its extracellular domain.

7. i) the γδ T cell receptor or its extracellular domain according to claim 6, and ii) Toxins and / or labels A construct containing

8. i) the γδ T cell receptor or its extracellular domain according to claim 6, and ii) an effector cell-binding domain, preferably a T cell-binding domain and / or a natural killer (NK) cell-binding domain; A construct containing

9. - the T cell binding domain binds to CD3, CD4, CD8, CD16, CD56, CD103, CD134, CD154, and / or CD314 and / or is a single chain Fv anti-CD3, CD4, CD8, CD16, CD56, CD103, CD134, CD154, and / or CD314 binding domain; and / or - the natural killer (NK) cell binding domain binds to CD16, NKG2D, NKp30, NKp44, NKp46, and / or DNAM and / or is a single chain Fv anti-CD16, NKG2D, NKp30, NKp44, NKp46, and / or DNAM binding domain; The construct of claim 8.

10. - the T cell binding domain binds to PD1, LAG3, CTUX4, TIGIT, CD96, BTLA, VISTA, TIM3, LAIR1, (inhibitory) KIR, CD160, and / or an immunoreceptor having an intracellular ITIM or ITSM motif, and / or is a single chain Fv anti-PD1, LAG3, CTLA4, TIGIT, CD96, BTLA, VISTA, TIM3, LAIR1, (inhibitory) KIR, CD160, and / or an immunoreceptor having an intracellular ITIM or ITSM motif binding domain, and / or - the natural killer (NK) cell binding domain binds to NKG2A, CD96, TIGIT, (inhibitory) KIR, PD1, TIM3, LAG3, CD112R, CD160, LAIR1, and / or an immunoreceptor having an intracellular ITIM or ITSM motif, and / or is a single chain Fv anti-NKG2A, CD96, TIGIT, (inhibitory) KIR, PD1, TIM3, LAG3, CD112R, CD160, LAIR1, and / or an immunoreceptor having an intracellular ITIM or ITSM motif binding domain; The construct of claim 8.

11. 11. The construct of any one of claims 7 to 10, wherein the construct is a fusion protein, preferably a bispecific fusion protein, and / or wherein the γδ T cell receptor or its extracellular domain is fused to a T cell binding domain and / or a natural killer (NK) cell binding domain.

12. 12. A nucleic acid or combination of nucleic acids encoding a BTN2A1-binding peptide according to any one of claims 1 to 6, or a construct according to any one of claims 7 to 11.

13. A cell expressing a BTN2A1-binding peptide according to any one of claims 1 to 6, a construct according to any one of claims 7 to 11, or a nucleic acid or combination of nucleic acids according to claim 12.

14. A cell expressing a BTN2A1-binding peptide according to any one of claims 1 to 6, which is an immune cell, more preferably a human T cell or a human NK cell, more preferably an αβ T cell or a γδ T cell.

15. 12. A method for producing a BTN2A1-binding peptide of any one of claims 1 to 6 or a construct of any one of claims 7 to 11, comprising expressing the BTN2A1-binding peptide of any one of claims 1 to 6 or the construct of any one of claims 7 to 11 in a host cell to produce the BTN2A1-binding peptide of any one of claims 1 to 6 or the construct of any one of claims 7 to 11.

16. A pharmaceutical composition comprising a BTN2A1-binding peptide described in any one of claims 1 to 6, a construct described in any one of claims 7 to 11, or a cell described in claim 13 or 14.

17. A BTN2A1-binding peptide according to any one of claims 1 to 6, a construct according to any one of claims 7 to 11, or a cell according to claim 13 or 14, for use in therapy, preferably for use in the treatment of cancer, autoimmune disease, and / or infection.

18. The BTN2A1-binding peptide is for use in a diagnostic method, preferably for determining BTN2A1 expression in cells in a sample, the method comprising: a) providing a sample comprising cells obtained from a subject; b) combining the cells with a BTN2A1-binding peptide of any one of claims 1 to 6; and c) determining binding of the BTN2A1-binding peptide to the cells, thereby determining cellular BTN2A1 expression in the sample.

7. The BTN2A1-binding peptide of any one of claims 1 to 6, comprising:

Images