TGF [beta] conversion receptor, nucleic acid encoding same, cell and pharmaceutical composition comprising same

Abstract

The present invention relates to the field of adoptive immune cell therapy, in particular adoptive T cell therapy. The present invention provides a conversion receptor which is capable of effectively converting an immunosuppressive signal of TGF [beta], which is normally present in the hostile tumor environment of a solid tumor, into a co-stimulatory signal, thereby improving T cell activation, and which can be safely used in therapy. Also provided are nucleic acids encoding the transforming receptors, cells expressing the receptors and pharmaceutical compositions comprising the cells, in particular for the treatment of cancer or infectious diseases, such as adoptive T cell therapy by solid tumors.

Description

[0001] This invention relates to the field of adoptive immunotherapy, particularly adoptive T-cell therapy. The invention provides a conversion receptor that effectively converts the immunosuppressive signal of TGFβ, normally present in the harsh tumor environment of solid tumors, into a co-stimulatory signal, thereby improving T-cell activation, and is safe for therapeutic use. Also provided are nucleic acids encoding the conversion receptor, cells expressing said receptor, and pharmaceutical compositions comprising said cells, particularly for treating cancer or infectious diseases, such as adoptive T-cell therapy of solid tumors.

[0002] The tumor microenvironment is a harsh environment in which T cells may dysfunction due to various metabolic and immunosuppressive mechanisms. 1、2 Transforming growth factor β (TGFβ / TGFb) is a pleiotropic cytokine commonly found in the tumor microenvironment of solid tumors, where it acts as a potent immunosuppressant. 3 It is produced by cancer-associated fibroblasts and tumor cells themselves. TGFβ has multiple effects on T cells, leading to decreased function, such as inhibiting proliferation, exhaustion, and inhibiting cell lysis. It also affects transcription factors (IFNγ, granzyme B, perforin, Tbet, and Eomesodermin). 3 These effects collectively lead to T cell dysfunction and may result in a poor anti-tumor response. In fact, the presence of TGFβ is associated with disease progression ( 4-6 ) and failure of immune checkpoint inhibitor therapy ( 7、8 This is related to the fact that next-generation cell immunotherapies for solid tumors need to take this adverse factor into account to ensure the function of T-cell products after infusion.

[0003] Different approaches have been adopted in the existing technology to solve this problem. For example, Bendle GM et al. ( 9 The study utilized a dominant-negative TGFβ receptor construct based on the TGFβRII, which lacks the intracellular signaling domain of TGFβ signaling, thereby blocking its inhibitory signaling. In a clinical trial, four out of eight patients demonstrated an objective clinical response to EBV-specific adoptive T-cell therapy expressing the construct. 10 ).

[0004] In contrast to this blockade, other groups used signal transduction receptors or conversion receptors to convert immunosuppressive TGFβ signals into cellular immunostimulatory signals. For this purpose, the extracellular domain of TGFβRII fused with the intracellular signaling domain of CD137 (4-1BB). This improved the production of effector cytokines in T cells and led to increased tumor eradication rates in in vivo models of solid tumors. 11-13 Similarly, WO2021244486 A1 ( 14This study taught the development of a fusion protein that comprises an extracellular portion of an immunosuppressive cytokine (such as TGFβ) receptor and an intracellular region of one of a variety of co-stimulatory molecules.

[0005] In addition, a TGFβ-responsive peptide containing an scFv-binding domain derived from the PET1074B9 antibody and intracellular domains of CD28 and CD3ζ was also used. 15、16 Since signaling mediated solely by this receptor is suitable for activating T cells, it can be considered a chimeric antigen receptor (CAR). Therefore, the safety of using T cells expressing this switching receptor for adoptive T cell transfer is questionable, as TGFβ can also be found in healthy tissues outside the tumor environment.

[0006] Based on existing technology, the inventors have solved the problem of providing a favorable conversion receptor that can effectively convert the immunosuppressive signal of TGFβ into a co-stimulatory signal that improves T cell activation through its TCR and can be safely used for human treatment.

[0007] The present invention, and particularly the claimed subject matter, solves this problem. The present invention provides a nucleic acid encoding a conversion receptor comprising the following domains. a) TGFβ binding domain, i.e., scFv, b) Hinge domain, c) Transmembrane domains, d) Intracellular costimulatory signaling domain, wherein, after the switching receptor binds to TGFβ, the intracellular costimulatory signaling domain cannot activate T cells without further stimulation.

[0008] The conversion receptor encoded by nucleic acid is a continuous single-chain polypeptide. As is standard in the art, the domains are arranged in order from the N-terminus to the C-terminus of the polypeptide. Upon expression in T cells expressing the T cell receptor (TCR) and typical CD28, the conversion receptor of the present invention, in the presence of TCR activation and TGFβ, is able to increase the secretion of effector cytokines, such as IFNγ, from said T cells compared to TCR activation in the absence of TGFβ.

[0009] The inventors developed a novel conversion receptor (also called a conversion receptor) of this invention and compared it with three prior art receptors that bind TGFβ, providing positive signaling to T cells or not providing signaling at all. These receptors all have structures and domains different from the conversion receptor. The conversion receptor was benchmarked against prior art receptors BM (benchmark receptor) 1, BM2, and BM3. BM1 is a chimeric antigen receptor (CAR) against soluble TGFβ, possessing an scFv-binding domain derived from the PET1074B9 antibody and intracellular domains of CD28 and CD3ζ.16 BM2 is composed of the extracellular domain of TGFβ receptor II (TGFβRII) and the transmembrane and intracellular domains of CD137. 12 BM3 is a truncated version of TGFβRII that binds to soluble TGFβ but does not transmit any signal. 17 ).

[0010] The preferred structure of the conversion receptor of the present invention is shown in Table 1.

[0011] Table 1: 15 conversion receptor (SR) constructs. The individual components of each receptor are shown in the corresponding column.

[0012] The inventors discovered that all conversion receptors have better co-stimulatory effects than BM and can be safely used for human T-cell therapy, and that cells transduced with the conversion receptors of this invention have higher IFNγ secretion in the presence of TGFβ.

[0013] As explained in more detail below, BM1 and BM2 receptors protect T cells from the negative effects of TGFβ, with fold changes in IFNγ secretion of 3.4 and 1.2, respectively, in the presence of TGFβ. BM3 cells do not completely avoid TGFβ, with a fold change of 0.87. The switching receptors of this invention all show fold changes greater than 1, meaning they convert negative signals to positive signals and lead to increased IFNγ secretion. All switching receptors contain a TGFβ-binding domain, namely scFv.

[0014] Among the target prior art receptors, the optimal immunostimulatory effect of TGFβ is found in BM1, the only prior art receptor that also contains a TGFβ binding domain (i.e., scFv). However, it differs from the conversion receptor of the present invention, specifically in that it contains intracellular domains of CD28 and CD3ζ, while the receptor of the present invention contains a co-stimulatory domain of CD137. They do not contain intracellular signaling domains of CD3ζ and / or CD28.

[0015] Except for BM1, all tested switching receptors of the present invention outperformed other prior art benchmark receptors. It is not surprising that BM1 provides a strong signal, as it contains an intracellular signaling domain that is even independent of TCR stimulation, compared to the receptor of the present invention. However, the switching receptor of the present invention further preferably contains different scFvs capable of binding TGFβ. SR01-04 outperformed other constructs, including the benchmark receptor, showing a fold change in IFNγ secretion of 3.5 to 5 in the presence of TGFβ upon TCR stimulation.

[0016] WO2006086469A2 compared the functional affinity of different anti-TGFβ-scFvs. Interestingly, PET1074B9 showed the highest functional affinity, while PET1073G12 and PET1287A10 performed poorly. However, this experiment surprisingly shows that the aforementioned characteristics do not predict the most favorable scFv for converting the receptor in this invention. The inventors can demonstrate that, contrary to expectations, optimal results are obtained using a PET1073G12-based scFv in the context of the constructs of this invention.

[0017] The conversion receptor of the present invention preferably has a functional affinity for TGFβ of up to 50 ng / mL, for example in the ranges of 0.1-50 ng / mL, 0.4-10 ng / mL, and 0.5-5 ng / mL, preferably 0.7-2 ng / mL. In the context of the present invention, TGFβ can be TGFβ1, TGFβ2, or TGFβ3, preferably TGFβ1. Typically, the human system has the highest relevance, and TGFβ is human.

[0018] When T cells expressing the switching receptor are activated by anti-CD3 / anti-CD28 beads in the presence of TGFβ, the functional affinity of the switching receptor itself can be measured by an increase in IFNγ secretion, for example, as shown in Materials and Methods. Figure 7 As shown in Figure A.

[0019] In a preferred embodiment of the invention, scFv comprises variable reconnection (VH) and variable light chain (VL) structural domains, wherein

[0020] a) The VH domain contains CDR1 shown in SEQ ID NO:1, CDR2 shown in SEQ ID NO:2, and CDR3 shown in SEQ ID NO:3, and the VL domain contains CDR1 shown in SEQ ID NO:4, CDR2 shown in SEQ ID NO:5, and CDR3 shown in SEQ ID NO:6. Based on this, the scFv containing these CDRs is named PET1073G12, and the present invention's conversion receptor having this scFv has proven particularly advantageous. Suitable frame regions are disclosed, for example, in WO2006086469A2. They can be human or mouse frame regions, preferably human. PET1073G12 preferably contains the variable heavy chain (VH) domain shown in SEQ ID NO:7 and / or the variable light chain (VL) domain shown in SEQ ID NO:8, preferably both. The VH and VL domains can be in any order, but the VH domain is preferably N-terminal, as this leads to better expression. The scFv can, for example, have the sequence shown in SEQ ID NO:9, as in the conversion receptors in Table 1. The scFv is optionally encoded by SEQ ID NO:10.

[0021] In an alternative embodiment of the invention, scFv comprises variable heavy chain (VH) and variable light chain (VL) structural domains, wherein

[0022] b) The VH domain comprises CDR1 shown in SEQ ID NO:11, CDR2 shown in SEQ ID NO:12, and CDR3 shown in SEQ ID NO:13, and the VL domain comprises CDR1 shown in SEQ ID NO:14, CDR2 shown in SEQ ID NO:15, and CDR3 shown in SEQ ID NO:16. Based on this, the scFv is named PET1074B9. Suitable frame regions are disclosed, for example, in WO2006086469A2. They can be human or mouse frame regions, preferably human. PET1074B9 preferably comprises the VH domain of SEQ ID NO:17 and / or the VL domain of SEQ ID NO:18, preferably both. The VH and VL domains can be in any order, but the VH domain is preferably N-terminal, as this leads to better expression. The scFv can, for example, have the sequence of SEQ ID NO:19, as shown in the conversion receptors in Table 1. The scFv is optionally encoded by SEQ ID NO:20.

[0023] In an alternative embodiment of the invention, scFv comprises variable heavy chain (VH) and variable light chain (VL) structural domains, wherein

[0024] c) The VH domain comprises CDR1 shown in SEQ ID NO:21, CDR2 shown in SEQ ID NO:22, and CDR3 shown in SEQ ID NO:23, and the VL domain comprises CDR1 shown in SEQ ID NO:24, CDR2 shown in SEQ ID NO:25, and CDR3 shown in SEQ ID NO:26. Based on this, the scFv is named PET1287A10. Suitable frame regions are disclosed, for example, in WO2006086469A2. They can be human or mouse frame regions, preferably human. PET1287A10 preferably comprises the VH domain of SEQ ID NO:27 and / or the VL domain of SEQ ID NO:28, preferably both. The VH and VL domains can be in any order, but the VH domain is preferably N-terminal because this leads to better expression. The scFv can, for example, have the sequence of SEQ ID NO:29, as shown in the conversion receptors in Table 1. The scFv is optionally encoded by SEQ ID NO:30.

[0025] The scFv contained in the conversion receptor of the present invention may, for example, include PET1073G12 of SEQ ID NO:9, PET1074B9 of SEQ ID NO:19, and PET1287A10 of SEQ ID NO:29, or variants of any of the scFv having at least 90% sequence identity with any of SEQ ID NO:9, 19, or 29. Of course, the variants include the CDR region described herein and specifically bind to TGFβ.

[0026] Typically, the different functional portions or domains of the transconverting receptor of the present invention can be derived, for example, from humans, mice, rats, dogs, donkeys, goats, or rabbits. Preferably, and particularly, if intended for use in treating human subjects, at least the hinge, transmembrane, and intracellular domains are human; optionally, all domains are human. It is also preferred that the frame region of the scFv is human. Using human sequences is advantageous for minimizing human immunogenicity. Similarly, in treating mutine / mouse subjects or in mouse cells or mouse experiments, optionally, these domains are preferably mouse-derived.

[0027] In the conversion receptor of the present invention, the scFv has a hinge domain at its C-terminus. This hinge domain can also be considered a spacer region, allowing the scFv to better access TGFβ. Its length can be 3-100 amino acids, 4-50 amino acids, 5-30 amino acids, 6-25 amino acids, 7-20 amino acids, 8-15 amino acids, 9-14 amino acids, 10-13 amino acids, or 11-12 amino acids. The hinge domain can be any suitable hinge region, for example, as known in the art in the context of CAR. However, since CARs often bind to cell surface antigens, a “long” hinge, such as one with 30 or more amino acids, is typically chosen. Such a hinge can be used, for example, as in EP3769816A1 (… 18 The extracellular portion of CD8α taught by [the protein ID: AAB04637.1]. ​​However, in the present case, a shorter hinge is sufficient because the ligand is a soluble cytokine. Therefore, the hinge can preferably have about 7-15 amino acids, for example 10-13 amino acids, or for example 12 amino acids.

[0028] In some embodiments, the hinge is the hinge region of the IgG molecule. For example, it may include the CH2CH3 region of the IgG molecule. In some embodiments, the spacer region includes one or more of the hinge region, CH1, CH2, and CH3 regions. In some embodiments, the spacer region is derived from the hinge, CH1, CH2, and / or CH3 regions or other regions (preferably hinge regions) of IgG1, IgG2, IgG3, IgG4, IgA1, IgA2, IgD, IgE, or IgM from human, mouse, rat, dog, donkey, goat, or rabbit (preferably human). In some embodiments, the spacer region includes the hinge region of the IgG molecule. If it further includes the CH2CH3 region of the IgG molecule, it may have additional L235E / N297Q or L235D / N297Q mutations to prevent Fc receptor binding. In some embodiments, the hinge is composed of the hinge region of the IgG molecule. In a preferred embodiment, the hinge region may include or be composed of the human IgG4 hinge domain, for example (SEQ ID NO:31). For example, this is the case in the preferred switching receptors SR02 and SR04 of the present invention. SR02 differs from SR03 only in the hinge region and has a stronger, reproducible co-stimulatory effect. Therefore, the hinge domain is preferably the IgG4 hinge domain.

[0029] Other conversion receptors of the present invention have a CD8 hinge domain, such as CD8α or CD8β, preferably human CD8α. It may contain or consist of the sequence of SEQ ID NO:32. Since adjacent transmembrane domains can also be derived from CD8, using this combination, such as SR03, simplifies cloning and reduces the risk of forming potential immunostimulatory epitopes.

[0030] At the C-terminus of the hinge domain, the transmembrane domain anchors the transconjugation receptor to the cell membrane. It allows signal transduction from the extracellular portion of the molecule to the intracellular portion. It is located between the hinge domain and the intracellular domain and may be directly adjacent to them. Other linkers may be used, but are not required. Preferably, the transmembrane domain is derived from a co-stimulatory molecule. For example, the transmembrane domain may be selected from transmembrane domains of CD8 (CD8α or CD8β, preferably CD8α) (e.g., SEQ ID NO:33), CD28 (e.g., SEQ ID NO:34), or CD137 (e.g., SEQ ID:35). It may also be a transmembrane domain as described in WO2014172584A1, such as a transmembrane domain of CD134 or CD7. A transmembrane domain that allows TGFβ signal transduction from the extracellular portion of a molecule having a specific sequence variant described herein to the intracellular portion may be used, for example, having at least 90% sequence identity with any of SEQ ID NO:33-34. The CD137 transmembrane domain appears to negatively impact expression and is therefore generally less desirable. Switching receptors with CD8 or CD28 transmembrane domains exhibit superior expression.

[0031] Interestingly, the inventors can demonstrate that the use of the transmembrane domain of CD8, particularly CD8α, is most advantageous in the conversion receptor of the present invention. For example, SR02 still outperforms SR1-2 and SR04, which have transmembrane domains derived from CD137 and CD28, respectively.

[0032] In the intracellular portion, all conversion receptors of the present invention include an intracellular signaling domain that is a co-stimulatory signaling domain. Therefore, the intracellular signaling domain is capable of providing a co-stimulatory T cell signal, i.e., this signal can activate T cells together with the activated TCR signal. However, it cannot activate T cells without further stimulation, especially without stimulation via TCR or CAR. Therefore, it does not include an intracellular domain of the CD3 chain (particularly CD3ζ).

[0033] Intracellular signal transduction domains are preferably intracellular domains of immunoglobulin superfamily or TNF receptor superfamily proteins, for example ( 12The intracellular domain may be, for example, an intracellular signaling domain of CD137 (4-1BB), CD134 (OX40), CD27, TNFRSF18 (GITR or AITR), CD270 (HVEM or TNFRSF14), CD154 (CD40L), CD28, CD278 (ICOS), CD226 (DNAM-1), CD4, or CD8. Intracellular domains of T cell-stimulating cytokine receptors, such as IL-2, IL-7, IL-12, or IL-15, may also be used. It may also be an intracellular domain of TIM (T cell immunoglobulin mucin domain, such as TIM-1), CD2 / SLAM, or a lactoprotein, such as BTN or BTN-like. It may also be a combination of these signaling domains. In a preferred embodiment, the intracellular signaling domain includes at least one of CD137, CD28, CD4, and / or CD8, for example, one or two intracellular signaling domains. It may optionally contain at least one immune receptor tyrosine-based activation motif (ITAM) and / or be capable of binding TRAF (TNF receptor-associated factor). Preferably, the switching receptor of the present invention contains an intracellular signaling domain, optionally combined with a spacer region (e.g., a portion of the intracellular domain of CD4 or CD8 (preferably CD8α)).

[0034] Preferably, the intracellular signal transduction domain is the intracellular signal transduction domain of CD137, or a combination of the intracellular portion of human CD8 and the intracellular signal transduction domain of human CD137. It can also be the intracellular signal transduction domain of CD28, or a combination of the intracellular portion of human CD8 and the intracellular signal transduction domain of human CD28.

[0035] In a preferred embodiment, the conversion receptor of the present invention comprises an intracellular signal transduction domain of CD137, preferably human CD137. More preferably, the conversion receptor of the present invention comprises the intracellular signal transduction domain of SEQ ID NO:36. This domain may also be a variant having at least 90% sequence identity, capable of initiating CD137-mediated signal transduction. CD137 is a member of the TNF receptor (TNFR) superfamily, and its intracellular domain lacks known intrinsic enzymatic activity. It relies on adaptor proteins of the TNFR-associated factor (TRAF) family to construct the CD137 signaling body, transducing the signal into the cell. Therefore, upon ligand binding, in this example, upon binding of TGFβ to the scFv portion of the conversion receptor, TRAF1, TRAF2, and TRAF3 are recruited to the intracellular signal transduction domain of CD137, possibly homo and / or heterotrimers with different conformations, thereby initiating the construction of the CD137 signaling body. Thus, when TGFβ binds to the scFv portion of the conversion receptor, the signaling domain can recruit TRAF1, TRAF2, and TRAF3.

[0036] Including a spacer region between the transmembrane domain and the intracellular signaling domain (e.g., the CD137 intracellular signaling domain) may be advantageous. Specifically, the switching receptor may comprise a combination of the intracellular portion of CD8 (preferably CD8α) and the intracellular signaling domain (e.g., human CD137). Combinations of the intracellular portions of CD8 and CD137 disclosed in EP3769816A1 can be used in different contexts. The intracellular portion of CD8α may be human and may have, for example, SEQ ID NO:37. The intracellular portion of CD8 may also be a variant of SEQ ID NO:37 having at least 90% sequence identity, wherein the combination of the said intracellular portion of CD8 and the intracellular signaling domain of human CD137 is capable of initiating CD137-mediated signal transduction. The intracellular portion of CD8β may also be used.

[0037] The following data suggest that the intracellular portion containing CD8 is advantageous in SR02 (compared to SR01) and other scFv-based switching receptors. This is particularly preferred if the transmembrane domain is also derived from CD8. However, interestingly, the co-stimulatory effect does not appear to be better if the hinge domain is also derived from CD8.

[0038] Therefore, the preferred conversion receptor of the present invention comprises a hinge domain derived from IgG4, a transmembrane domain derived from CD8, and a combination of an intracellular portion of CD8 and an intracellular signaling domain of human CD137. This appears to be particularly advantageous in the case of a conversion receptor comprising PET1073G12. Interestingly, for constructs comprising other tested scFvs, SR08 and SR12 have been shown to convert TGFβ signaling into co-stimulatory signals more efficiently than SR06 and SR10, respectively. Notably, SR04 also exhibits excellent function. Therefore, another preferred conversion receptor comprises a hinge derived from IgG4, a transmembrane domain derived from CD28, and an intracellular signaling domain of human CD137. While the intracellular portion may include a combination with the intracellular portion of CD8, as disclosed herein, preferably, the construct does not include binding to the intracellular portion of CD8.

[0039] Furthermore, the inventors have demonstrated that, compared to other conversion receptors of the present invention, the use of the CD137 transmembrane domain and the intracellular signaling domain of CD137 (without the CD8 portion in the middle) results in reduced expression of the construct.

[0040] To simplify the detection of receptor expression of the present invention on cells, the transconverting receptor may also include an extracellular tag, optionally a FLAG tag (SEQ ID NO:38), a c-myc tag, a CD34 tag, a His tag, or an HA tag, preferably a FLAG tag. The tag may be located at the N-terminus of the TGFβ-binding domain. Optionally, a linker, such as a glycine-serine linker, or a GGS, GSS, or GSG linker, may be attached between the tag and the TGFβ-binding domain. In principle, each tag may use a ligand or preferably an antibody. Preferably, the tag is not highly immunogenic or not immunogenic in human subjects, which is most important in the therapeutic setting of human subjects. However, often no pre-existing antibody is available, and a single dose may be sufficient. However, in another embodiment, the receptor of the present invention does not contain an additional extracellular tag (or detectable label). This is not necessary because the TGFβ-binding domain and / or hinge domain can be used to detect expression. For example, the TGFβ-binding domain can be detected with labeled TGFβ. Antibodies against the receptor, such as antibodies against the frame or hinge region of scFv, may also be used. Therefore, the risk of immune responses against metastatic T cells carrying the conversion receptor of this invention can be minimized.

[0041] To enable transmembrane expression of the conversion receptor of the present invention, it typically comprises an N-terminal signal peptide, also referred to herein as a leader or leader sequence. This sequence can be, for example, a signal peptide for expressing a transmembrane protein, such as an antibody chain, such as mouse κ light chain, mouse λ light chain, mouse heavy chain, human κ light chain, human λ light chain, human heavy chain, human HLA, human TCRα, human TCRβ, human TCRγ, or human TCRδ. Interestingly, expression is improved using a mouse κ light chain leader sequence compared to using a human κ light chain leader sequence or a human heavy chain leader sequence. Therefore, preferably, the conversion receptor further comprises an N-terminal mouse κ light chain leader sequence (e.g., SEQ ID NO:40). Optionally, a tag and scFv sequence follow the C-terminus of the leader sequence. A spacer region, such as a glycine-serine linker, may also be present between the leader and tag and / or tag and scFv and / or leader and scFv domains.

[0042] The conversion receptor of the present invention can be any one of SR01 to SR12 or SR1-2, SR2-2 or SR3-2. Therefore, it can have a sequence of any one of SEQ ID NO:41-55, or a sequence having at least 90%, for example, at least 95% or at least 99% sequence identity with any one of SEQ ID NO:41-55. Preferably, the receptor has a sequence of SEQ ID NO:42 (SR02) or a sequence having at least 90%, for example, at least 95% or at least 99% sequence identity with it. Most preferably, the sequence identity is 100%.

[0043] In the nucleic acid of the present invention encoding the conversion receptor, the sequence encoding the conversion receptor is typically functionally linked to a promoter region, particularly a promoter that allows the receptor to be expressed in T cells, preferably human T cells. The promoter can be constitutive or inducible. For example, the promoter can be human EF1α, PGK, β-actin, IF4A1, GAPDH, GRP78, HSP70, β-kinin, ubiquitin B, CD45, MP71, or CMV. Preferably, the promoter is EF1α. Thus, the nucleic acid can be considered an expression construct of the conversion receptor. The sequence encoding the conversion receptor of the present invention can be directly linked to the promoter. Alternatively, functional linking to the promoter can occur in the context of a polycistronic operon, where different coding polypeptides are encoded under the control of a single promoter and separated by, for example, an internal ribosomal entry side (IRES) or a 2A element. Such other polypeptides can, for example, encode the α and β chains of the TCR. Optionally, a CD8 co-receptor can also be encoded.

[0044] The 2A element can be, for example, P2A, E2A, F2A, or T2A. Different 2A peptides have different self-cleavage efficiencies, with T2A and P2A being the most efficient and F2A the least efficient. In the context of this invention, the P2A element is preferred, for example, in combination with a glycine-serine linker (e.g., a GSG linker) located directly at the N-terminus of the P2A element. Adding the optional "GSG" N-terminus linker to the 2A peptide helps improve efficiency and is therefore the preferred choice for all P2A elements. Linkers, such as the C-terminus of a GSG linker with a cleavage site, may also be present.

[0045] The nucleic acid of this invention can be a vector, such as an expression vector. It is typically DNA. The nucleic acid can be, for example, microcircular DNA, plasmid, nanoparticle, doggybone DNA, or RNA or modified RNA.

[0046] Recently, DNA constructs using doggybone DNA (dbDNA) have been developed; dbDNA is a synthetic, closed linear DNA. Cell-free processes rely on the use of amplification, such as with Phi29 DNA polymerase, where the template is amplified and then incubated with the proteolytic enzyme TelN to complete a single closed linear DNA. The resulting dbDNA may contain the desired sequence, promoter, and polyA tail, but lacks bacterial sequences, such as antibiotic resistance genes.

[0047] Microcircular DNA can be derived from plasmids in which bacterial sequences (such as origins of replication) have been excised. Microcircular DNA is a small (typically less than 5 kb, preferably about 4 kb) circular plasmid derivative that has been isolated from all prokaryotic vector fractions. Microcircular DNA has been used as a transgenic vector for gene editing in mammalian cells, with the advantage that, because it does not contain bacterial DNA sequences, it is less likely to be considered foreign and destroyed. The smaller size of microcircular DNA also expands its cloning capabilities and facilitates its delivery into cells, for example, using transposases. Suitable nucleic acids and protocols are disclosed, for example, in WO 2017 / 158019 A1 (…). 19 )middle.

[0048] Microcircular DNAs lack origins of replication, so they typically do not replicate within target cells, and the encoded genes are lost with cell division. A new addition to this field is nonviral, self-replicating microcircular DNA, thanks to the presence of S / MAR elements. Nanoparticles are small circular DNA constructs without antibiotic resistance genes, with a backbone of less than 500 bp.

[0049] Plasmids (containing circular DNA at the origin of replication) can also be used in this invention, but it is preferred that less DNA needs to be transferred. Therefore, the nucleic acid is preferably doggybone DNA, nanoplasmids or microcircular DNA, with microcircular DNA being the most preferred.

[0050] In one embodiment, the DNA is a transposon or contains transposons. Therefore, the nucleic acid encoding the desired sequence may be flanked by inverted repeats that can be mobilized by a transposase, preferably a Sleeping Beauty transposase such as SB100X. Suitable transposases are disclosed, for example, in WO 2009 / 003671 A2 (… 20 Suitable transposons are disclosed, for example, in WO 2017 / 158029 A1 (…). 21 For example, the carrier can be a p2 carrier, or preferably a p4 or p5 carrier.

[0051] The vector can also be a viral vector, such as a retrovirus, adenovirus, AAV, or lentivirus vector.

[0052] The present invention also provides a conversion receptor encoded by the nucleic acid of the present invention. The conversion receptor may be an isolated protein or expressed as a cell surface molecule.

[0053] This invention also discloses a method for preparing cells containing the nucleic acids of this invention. A vector, such as purified microcircular DNA, can be transferred to recipient cells via lipid transfection, jet injection, extrusion, nanoneedles, or other transfection methods, or preferably electroporation. Infection with a viral vector is another preferred method for preparing the cells of this invention. For example, RNA can be readily transferred via electroporation.

[0054] The present invention also provides a cell comprising the nucleic acid of the present invention, wherein the cell preferably also expresses the conversion receptor of the present invention. The cell may be a bacterial cell, such as *Escherichia coli*, but is preferably a eukaryotic cell, such as a mammalian cell. Typically, for research purposes, the cell is a mouse or human cell. The therapeutic cell is typically a human cell to which a human subject will receive treatment. The cell is preferably a T cell, NK cell, or NKT cell, with T cells being the most preferred. The T cells may be αβ T cells, and in particular they may comprise or be composed of CD8αβ T cells. More preferably, the T cells are a mixture of CD4 and CD8 T cells. The T cells may also be γδ T cells.

[0055] The cells of the present invention preferably further express a class I restricted TCR construct. The TCR may target tumor antigens, such as tumor antigens expressed by solid tumors, like MAGE antigens (e.g., MAGEA1, MAGEA2, MAGEA3, or MAGEA4). The TCR may also be directed to, for example, EBV antigens or another tumor antigen.

[0056] For treatment, T cells, such as αβ or γδ T cells engineered with the nucleic acids of this invention, can be allogeneic to allow the use of off-the-shelf solutions. In particular, allogeneic adoptive T cell therapy is advantageous for the γδ T cells of this invention. Typically, T cells are syngeneic, such as autologous. This avoids immune rejection or the need for immunosuppression to prevent graft-versus-host disease (GvHD).

[0057] The present invention also provides a pharmaceutical composition comprising the nucleic acid of the present invention, the conversion receptor of the present invention, or the cell of the present invention. In particular, the pharmaceutical composition may comprise the cell of the present invention, which is a T cell, NK cell, or NKT cell. Optionally, the pharmaceutical composition further comprises a suitable buffer and / or excipient.

[0058] The pharmaceutical compositions of the present invention are typically intended for intravenous administration. They may include pharmaceutically acceptable carriers, such as buffers, for example, physiological saline or PBS. They may include excipients, such as stabilizers, such as SPGA, carbohydrates (such as sorbitol, mannitol, starch, sucrose, glucose, dextran), proteins, such as albumin or casein, or protein-containing reagents, such as bovine serum or skim milk. If the composition does not contain cells but nucleic acids, it may be a dried composition, such as a lyophilized composition. Such compositions typically contain fillers, such as non-reducing sugars like trehalose. They may also contain buffers.

[0059] T cells typically occur at a rate of 1 × 10⁻⁶ per kilogram of body weight. 5 -5×10 9 Administered to a concentration of approximately 1 × 10⁻⁶ cells to the subject or patient. 6 –5×10 11 Cells, for example, 1×10 8 –2×10 10 Individual cells can be administered to human subjects as a single intravenous dose. These parameters can be adjusted by the physician based on the patient's age, sex, weight, and medical condition.

[0060] In one embodiment, the pharmaceutical composition of the present invention comprises autologous T cells of a subject, which are engineered in vitro to express the nucleic acids of the present invention, such as autologous CD4 and / or CD8 T cells.

[0061] Alternatively, the nucleic acids of the present invention, particularly the expression vectors, can be administered to subjects for in vivo engineering, i.e., in vivo transduction of T cells. The expression vectors can target T cells, for example, by recognizing surface molecules expressed only by the cells to be engineered, such as CD3, CD4, CD8, TCRα, TCRβ, TCRγ, and / or TCRδ, via specific antibodies.

[0062] The present invention also provides pharmaceutical compositions for treating subjects in need, particularly those suffering from cancer (including cancer caused by infectious agents) or infectious diseases, preferably cancers such as solid tumors. Subjects suffering from diseases such as cancer are also designated herein as patients. Preferably, the pharmaceutical composition is used for adoptive T-cell therapy, such as adoptive T-cell therapy for cancer.

[0063] It can be adoptive T-cell therapy or T-cell receptor (TCR) gene therapy, for example, targeting HLA-A. Epitopes of tumor testis antigen MAGEA4 presented on 01, for example, TCR CTC127 (e.g., WO 2022 / 2243514 A1 ( 22 The variants thereof disclosed in ( ) have at least 90% sequence identity in the CDR1, CDR2 and CDR3 regions, and optionally have at least 98% or at least 99% sequence identity with CTC127 on the complete sequence of the variable region.

[0064] In adoptive T-cell therapy, in addition to TCR transgenic T cells, T cells expressing CAR (chimeric antigen receptor) can also be used.

[0065] A method for treating subjects in need, such as those with cancer, preferably solid cancer, is also disclosed. Treatment typically involves intravenous injection of the pharmaceutical composition of the invention, preferably infusion. Typically, patients undergo pretreatment prior to T-cell administration, for example, on days 5, 4, and 3 before T-cell administration, by administration of fludarabine and cyclophosphamide (e.g., infusion of 30 mg / m²). 2 Fludarabine and 500 mg / m 2 Cyclophosphamide).

[0066] Subjects are typically mammalian subjects, such as mice or humans. Human patients are preferred.

[0067] Solid tumors can include, for example, melanoma, gastric cancer, head and neck cancer, lung cancer, breast cancer, ovarian cancer, cervical cancer, bladder cancer, synovial sarcoma, liposarcoma, and esophageal cancer. TGFβ is highly expressed in all of these solid tumors. Cancers can also be caused by CMV, Kaposi's sarcoma virus, Helicobacter pylori, Merkel cell polyomavirus, or human papillomavirus (HPV). The infectious diseases can be caused by intracellular protozoa, such as Trypanosoma, Leishmania, Toxoplasma, or Mycobacterium, HCV, HIV, and parainfluenza virus. TGFβ expression plays a role in downregulating the immune response in related infectious diseases.

[0068] The present invention is further described in the following embodiments with reference to the accompanying drawings and sequences, but is not limited thereto. For the purposes of this invention, all references cited herein are incorporated herein by reference in their entirety. Attached Figure Description

[0069] Figure 1 IFNγ release from T cells in the presence of TGFβ. T cells from two donors were expanded for 9 days using a laboratory-scale protocol and then restimulated for 24 hours with αCD3 / αCD28 magnetic beads (1:100 dilution) in or without 10 ng / ml human recombinant TGFβ. IFNγ in the supernatant was measured by ELISA. The figures show the mean ± SEM concentration of IFNγ from replicate experiments from the two donors.

[0070] Figure 2 Modular building blocks for the switching receptor. Leader: Mouse κ light chain. FLAG: Tag for specific detection and / or selection of the switching receptor. TGFβ binding domain: VH and VL chains of one of three human anti-TGFβ antibodies. Hinge: Human IgG4 or CD8. Transmembrane (TM) domain: Human CD137, CD8, or CD28. Signal transduction domain: Human CD137 or truncated human CD8 with CD137.

[0071] Figure 3 : Cell surface expression of the switching receptor and BM receptor. Human CD8 + T cells were retrovirally transduced using 15 candidate conversion receptors or 3 BM receptors. Transduction efficiency and surface expression levels (BM3 only) were assessed by flow cytometry staining with FLAG tags or TGFβRII on day 5 post-transduction. The dot plot shows live T cells from one of the two donors. Numbers indicate the frequency of cells expressing the engineered receptor.

[0072] Figure 4 Functional assessment of candidate switching receptors and baseline receptors. A total of 10 ng / ml of recombinant human TGFβ was evaluated in the presence or absence of TGFβ. 4 T cells were stimulated with αCD3 / αCD28 magnetic beads (1:100 dilution) for 24 hours. IFNγ in the supernatant (1:5 dilution) was detected by ELISA. A) IFNγ concentration under two experimental conditions (with and without TGFβ). B) fold change in IFNγ secretion in the presence of TGFβ. Dashed lines indicate a fold change of 1, i.e., no difference between the two conditions. Data are the mean ± SEM from duplicate experiments with two T cell donors.

[0073] Figure 5 Functional assessment of candidate switching receptors and baseline receptors. A total of 10 ng / ml of recombinant human TGFβ was evaluated in the presence or absence of TGFβ.4 T cells were stimulated with αCD3 / αCD28 magnetic beads (1:300 dilution) for 24 hours. IFNγ in the supernatant (1:5 dilution) was detected by ELISA. A) IFNγ concentration under two experimental conditions (with and without TGFβ). B) fold change in IFNγ secretion in the presence of TGFβ. Dashed lines indicate a fold change of 1, i.e., no difference between the two conditions. Data are the mean ± SEM from duplicate experiments with two T cell donors.

[0074] Figure 6 T cell activation was assessed by receptor switching in the presence of TGFβ but without TCR (CD3 / CD28) stimulation. (Total 10) 4 T cells were cultured for 24 hours with or without 10 ng / ml recombinant human TGFβ. IFNγ in the supernatant (1:5 dilution) was detected by ELISA. Data are presented as mean ± SEM from duplicate experiments from two T cell donors.

[0075] Figure 7 In-depth functional evaluation of the preferred candidate switching receptor. A total of 10 [preferred] TGFβs were evaluated in the presence or absence of human TGFβ at a range of concentrations (0, 0.4, 0.8, 1.6, 3.1, 6.2, 12.5, 25, and 50 ng / mL). 4 T cells were stimulated for 24 hours with suboptimal concentrations of αCD3 / αCD28 magnetic beads (1:300 dilution). IFNγ in the supernatant (1:5 dilution) was detected by ELISA. A) IFNγ concentration (values ​​at TGFβ 0 are not displayed on a logarithmic scale). B) fold change in IFNγ secretion in the presence of TGFβ compared to secretion without TGFβ. Data are the mean ± SEM from duplicate experiments with two T cell donors.

[0076] Figure 8 In the absence of TCR stimulation, in-depth functional evaluation of the preferred candidate switching receptor and benchmark receptor 1 (BM1) was conducted. A total of 10 [units / mL] of human TGFβ were evaluated in the presence of a range of concentrations (0.4, 0.8, 1.6, 3.1, 6.2, 12.5, 25, and 50 ng / mL). 4 T cells were cultured for 24 hours. IFNγ in the supernatant was detected by ELISA. Data showed mean ± SEM values ​​from duplicate experiments with two T cell donors.

[0077] Figure 9 : Modular structure of the main candidate SR02.

[0078] Figure 10 Multifunctionality assessment. Total 10 4 SR02 +T cells were stimulated (A) or unstimulated (B) with αCD3 / αCD28 magnetic beads (1:300 dilution) and cultured for 24 hours in the presence or absence of 10 ng / ml recombinant human TGFβ. Cytokines in the supernatant were detected using a BioLegend cytokine bead array (CBA) according to its instructions.

[0079] Figure 11 Cell proliferation and survival were assessed under repeated TGFβ stimulation. Unengineered T cells and SR02+ T cells were repeatedly stimulated with αCD3 / αCD28 magnetic beads (1:300 dilution) in the presence or absence of TGFβ. A total of 7 stimulations were performed over 17 days (days 0, 3, 5, 7, 10, 12, and 14). Cell counts and viability were assessed between each round of stimulation, and cells were adjusted to 8 x 10⁸ cells / cells. 5 Cells / mL for the next round of stimulation. A) Cell count in the absence of TGFβ. B) Cell viability in the absence of TGFβ. C) Cell count in the presence of TGFβ. D) Cell viability in the presence of TGFβ.

[0080] Figure 12 Functional assessment of cancer-specific TCR-T cells expressing the switching receptor in the presence of TGFβ. T cells were transduced with cancer-specific TCR (CTC127); cancer-specific TCR and SRO2 switching receptor (CTC127 + SWITCH); or untransduced (unengineered). 4x10 3 One transduced T cell (or an equivalent number of unengineered T cells) and 1x10 4 Target cancer cells expressing the CTC127-recognized homologous antigen and HLA allele were co-cultured. In A, the target cancer cells were A375; in B, the target cancer cells were H1703. Co-culture was performed overnight for 18 hours with or without exogenous TGFβ (10 ng / ml). After co-culture, IFNγ in the supernatant was measured by ELISA at a 1:10 dilution (to avoid supersaturation). Data are presented as mean ± SEM from duplicate experiments from two T cell donors.

[0081] Figure 13 In vivo efficacy of therapeutic T cells expressing the switching receptor and CTC127 cancer-specific TCR. (5x10) 6 A375 cancer cells were subcutaneously implanted into immunodeficient NSG mice. Eight days later, when the average volume of the solid tumor reached 100 mm... 3At that time, therapeutic human T cells were administered intravenously. These T cells were transduced with CTC127x and co-receptor (CTC127x + CoR); transduced with CTC127x, co-receptor, and SRO2 switching receptor (CTC127x + CoR + SWITCH); or untransduced (unengineered). Tumor growth was monitored with calipers every 2-3 days. If the tumor volume reached >1500 mm, treatment was initiated. 3 Mice were then euthanized. The numbers in parentheses represent the number of mice with a complete response within the group divided by the total number of mice. A complete response was defined as a mouse completely rejecting the tumor and remaining tumor-free throughout the observation period. In A, 1x10⁻⁶ mice were injected. 7 A dose of transduced T cells. In B, 0.2 x 10⁻⁶ cells were injected. 7 A dose of transduced T cells.

[0082] Figure 14 In vivo efficacy of therapeutic T cells expressing the switching receptor and CTC45 cancer-specific TCR. (5x10) 6 A375 cancer cells engineered to express the CTC45 homologous antigen were subcutaneously implanted into immunodeficient NSG mice. Nine days later, when the average solid tumor volume reached 50 mm... 3 At that time, therapeutic human T cells were administered intravenously. These T cells were transduced with CTC45; transduced with CTC45 and SR02 switching receptors; or untransduced (unengineered). Tumor growth was monitored every 2-3 days using calipers. If the tumor volume reached >1500 mm, the tumor was considered closed. 3 Mice were then euthanized. The numbers in parentheses represent the number of mice with a complete response within the group divided by the total number of mice. A complete response was defined as a mouse completely rejecting the tumor and remaining tumor-free throughout the observation period. In A, 1x10⁻⁶ mice were injected. 7 A dose of transduced T cells. In B, 0.2 x 10⁻⁶ cells were injected. 7 A dose of transduced T cells. sequence SEQ ID NO: 1 HCDR1 / PET1073G12 SEQ ID NO: 2 HCDR2 / PET1073G12 SEQ ID NO: 3 HCDR3 / PET1073G12 SEQ ID NO: 4 VCDR1 / PET1073G12 SEQ ID NO: 5 VCDR2 / PET1073G12 SEQ ID NO: 6 VCDR3 / PET1073G12 SEQ ID NO: 7 Heavy chain variable domain / PET1073G12 SEQ ID NO: 8 Light chain variable domain / PET1073G12 SEQ ID NO: 9 scFv / PET1073G12 containing (G4S)3 connectors but excluding hinge structural domain SEQ ID NO: 10 Preferred nt sequence / PET1073G12 SEQ ID NO: 11 HCDR1 / PET1074B9 SEQ ID NO: 12 HCDR2 / PET1074B9 SEQ ID NO: 13 HCDR3 / PET1074B9 SEQ ID NO: 14 VCDR1 / PET1074B9 SEQ ID NO: 15 VCDR2 / PET1074B9 SEQ ID NO: 16 VCDR3 / PET1074B9 SEQ ID NO: 17 Heavy chain variable domain / PET1074B9 SEQ ID NO: 18 Light chain variable domain / PET1074B9 SEQ ID NO: 19 scFv / PET1074B9 containing (G4S)3 connector but excluding hinge structure domain SEQ ID NO: 20 Preferred nt sequence / PET1074B9 SEQ ID NO: 21 HCDR1 / PET1287A10 SEQ ID NO: 22 HCDR2 / PET1287A10 SEQ ID NO: 23 HCDR3 / PET1287A10 SEQ ID NO: 24 VCDR1 / PET1287A10 SEQ ID NO: 25 VCDR2 / PET1287A10 SEQ ID NO: 26 VCDR3 / PET1287A10 SEQ ID NO: 27 Heavy chain variable domain / PET1287A10 SEQ ID NO: 28 Light chain variable domain / PET1287A10 SEQ ID NO: 29 scFv / PET1287A10 containing (G4S)3 connector but excluding hinge structure domain SEQ ID NO: 30 Preferred nt sequence / PET1287A10 SEQ ID NO: 31 Human IgG4 hinge domain SEQ ID NO: 32 PCD8 hinge structural domain. SEQ ID NO: 33 human CD8 transmembrane domain SEQ ID NO: 34 human CD28 transmembrane domain SEQ ID NO: 35 human CD137 transmembrane domain SEQ ID NO: 36 Intracellular signal transduction domain of human CD137 SEQ ID NO: 37 Intracellular fraction of human CD8 SEQ ID NO: 38 FLAG tag SEQ ID NO: 39 -- SEQ ID NO: 40 Mouse κ light chain leader sequence SEQ ID NO: 41 Conversion receptor (SR01) SEQ ID NO: 42 Preferred switching receptor (SR02) SEQ ID NO: 43 Switching receptor (SR03) SEQ ID NO: 44 Switching receptor (SR04) SEQ ID NO: 45 Switching receptor (SR05) SEQ ID NO: 46 Switching receptor (SR06) SEQ ID NO: 47 Switching receptor (SR07) SEQ ID NO: 48 Switching receptor (SR08) SEQ ID NO: 49 Switching receptor (SR09) SEQ ID NO: 50 Switching receptor (SR10) SEQ ID NO: 51 Switching receptor (SR11) SEQ ID NO: 52 Switching receptor (SR12) SEQ ID NO: 53 Switching receptor (SR1-2) SEQ ID NO: 54 Switching receptor (SR2-2) SEQ ID NO: 55 Switching receptor (SR3-2) Nucleic acid encoding SR02 (SEQ ID NO: 56) Example Materials and methods

[0083] Production of transduced T cells

[0084] To generate transduced T cells, PBMCs were isolated from fresh blood by gradient centrifugation and the T cells contained therein were stimulated in a medium supplemented with T cell-stimulating cytokines with bead-fixed anti-CD3 and anti-CD28 antibodies (TransAct, Miltenyi GmbH). For example, IL-2, IL-7, and / or IL-15 could be used. On days 2 and 3 post-stimulation, the cells were transduced with a transgenic retrovirus (e.g., SEQ ID NO 41-55). GALV cells were lipid-transfected with transgenic plasmid DNA using a Lipofectamine transfection kit (Invitrogen) according to the manufacturer's instructions to generate retroviruses. The transduced T cells were further cultured for 9 to 12 days and used for assays. The expression of the conversion receptor was assessed by flow cytometry using an antibody targeting the FLAG tag.

[0085] Determination of functional affinity

[0086] To determine functional affinity, a total of 10 individuals were stimulated with aCD3 / aCD28 beads diluted 1:300 (TransAct, Miltenyi GmbH) in the presence of different concentrations of recombinant human TGFβ1 (Peprotech). 4 T cells with a conversion receptor transduction efficiency >70% were selected after 24 hours. The assay was performed in 96-well plates with a total volume of 200 µl of serum-free medium. IFNγ was detected in the supernatant (1:5 dilution) by ELISA (BDBiosciences). IFNγ values ​​were then normalized to % of the maximum IFNγ and plotted on a logarithmic concentration scale. Nonlinear curve fitting was performed to determine the EC50 of the conversion receptor. 50 This means that IFNγ is released at 50% of the maximum TGFβ concentration.

[0087] Cancer cell co-culture

[0088] To determine the role of TGFβ in TCR-engineered T cells recognizing target cancer cells and the ability of the switching receptor (SR02) to overcome it, T cells transduced with CTC127 (such as the cancer-specific TCR disclosed in WO / 2022 / 243514 A1) with or without the switching receptor were co-cultured with cancer cell lines for 18 hours, and TCR-mediated activation was measured as IFNγ released into the supernatant. 4 x 10 T cells were cultured in the presence or absence of 10 ng / ml human recombinant TGFβ1 (Peprotech). 3 Transduced T cells (or an equivalent number of unengineered T cells) with 1x10 HLA alleles expressing the homologous antigen and CTC127 recognition (A375 or H1703) 4 Target cancer cells were co-cultured. The assay was performed in 96-well plates with a total volume of 200 µl of serum-free medium. IFNγ was detected in the supernatant (1:10 dilution) by ELISA (BD Biosciences) after 18 hours.

[0089] Cancer rejection reaction in the body

[0090] For an in vivo cancer rejection model, immunodeficient NSG mice (NOD.Cg-Prkdc) <scid>Il2rg <tm1wji> / SzJ) using 5x10 6 A375 cancer cells were subcutaneously implanted. After 8 to 9 days, when the solid tumor reached a volume of 50-100 mm... 3 Administer therapeutic human T cells intravenously. Monitor tumor growth with calipers every 2-3 days. If the tumor volume >1500 mm... 3 If so, the mice are euthanized humanely. result

[0091] The immunosuppressive effect of TGFβ on T cells

[0092] To demonstrate the inhibitory effect of TGFβ on T cell function, the inventors designed an experiment in which T cells were polyclonally restimulated via their TCRs using magnetic beads (TransAct, Miltenyi GmbH) coated with anti-CD3 and anti-CD28 antibodies, with or without TGFβ. Twenty-four hours later, the cytotoxic mediator IFNγ, a substitute indicator of T cell function, was detected in the supernatant. Prior to stimulation, the T cells were expanded for nine days using a laboratory-scale manufacturing protocol; however, at this point, they were not engineered with tumor-specific TCRs but carried their native TCRs.

[0093] In this experiment ( Figure 1 The study observed an immunosuppressive effect of TGFβ on T cells, as activated T cells in the presence of TGFβ secreted less IFNγ than T cells not exposed to TGFβ (a 40% reduction). Therefore, this experimental setup was well-suited to assess the ability of candidate switching receptors to reverse the inhibitory effects of TGFβ.

[0094] Transforming receptor construct design

[0095] The inventors' goal was to design a favorable transmembrane switching receptor that binds to extracellular TGFβ and transduces the signal to T cells, thereby reversing the inhibitory effect of TGFβ. As an important safety feature, the co-stimulatory signal alone is insufficient to activate T cells in the absence of TCR signaling. Therefore, T cells carrying the switching receptor remain quiescent in the presence of TGFβ before encountering target tumor cells. For this purpose, the inventors selected a fusion of the co-stimulatory CD137 intracellular domain or a portion of the CD8 intracellular domain with the CD137 intracellular domain.

[0096] The conversion receptor was designed as a modular construct containing multiple domains. Figure 2 ), and synthesized 15 candidate constructs with different domains (Table 1). The TGFβ binding domain is a single-chain variable fragment (scFv) constructed based on three anti-TGFβ antibodies (PET1073G12 (scFv sequence SEQ ID NO: 9), PET1074B9 (scFv sequence SEQ ID NO: 19), and PET1287A10 (scFv sequence SEQ ID NO: 29)). These antibodies were developed by Genzyme in patent (WO2006086469A2 ( 23 As described in )).

[0097] Other released builds

[0098] A literature search identified three other engineered receptors that bind to TGFβ, providing positive or no signal to T cells. These receptors all possess structures and domains different from the candidate conversion receptor of this invention. These receptors were used to benchmark the conversion receptor of this invention. BM (benchmark receptor) 1 is a chimeric antigen receptor (CAR) targeting soluble TGFβ, possessing an scFv-binding domain derived from the PET1074B9 antibody and intracellular domains of CD28 and CD3ζ. 16 BM2 is composed of the extracellular domain of TGFβ receptor II (TGFβRII) and the transmembrane and intracellular domains of CD137. 12 BM3 is a truncated version of TGFβRII that binds to soluble TGFβ but does not send any signal. 17 ).

[0099] Candidate switching receptor selection

[0100] To assess expression levels and function, the candidate conversion receptors of this invention and three benchmark receptors were cloned into a retroviral vector backbone, and activated human CD8+ T cells were retrovirally transduced. The expression levels of the conversion receptors were assessed by flow cytometry staining for FLAG tags (or TGFβRII of BM3) on the cell surface. Except for SR1-2, SR2-2-, and SR3-2-, most candidate conversion receptors exhibited excellent transduction rates and surface expression levels (…). Figure 3 In the baseline receptor, BM1 and BM3 showed high expression, while BM2 showed moderate expression.

[0101] To evaluate the efficacy of the conversion receptor in reversing TGFβ inhibition, the aforementioned polyclonal activation assay was performed using αCD3 / αCD28 magnetic beads in the presence and absence of TGFβ. TGFβ reduced IFNγ secretion from unengineered T cells; T cells expressing the conversion receptor were not only immune to this inhibitory effect but also secreted more IFNγ upon activation in the presence of αCD3 / αCD28 and TGFβ. Figure 4 A). To normalize the data and compare constructs, we calculated the fold change in IFNγ secretion for each construct under αCD3 / αCD28 + TGFβ versus αCD3 / αCD28 activation alone ( Figure 4 Figure 1 T cell release of IFNγ in the presence of TGFβ. T cells from two donors were expanded for 9 days using a laboratory-scale protocol, and then restimulated for 24 hours with αCD3 / αCD28 beads (1:100 dilution) in or without 10 ng / ml human recombinant TGFβ. IFNγ concentration in the supernatant was measured by ELISA. The figure shows the mean ± SEM concentration of IFNγ from duplicate samples from the two donors.

[0102] Figure 2 Modular building blocks for the switching receptor. Leader: Mouse κ light chain. FLAG: Tag for specific detection and / or selection of the switching receptor. TGFβ binding domain: VH and VL chains of one of three human anti-TGFβ antibodies. Hinge: Human IgG4 or CD8. Transmembrane (TM) domain: Human CD137, CD8, or CD28. Signal transduction domain: Human CD137 or truncated human CD8 with CD137.

[0103] Figure 3 Cell surface expression of the conversion receptor and BM receptor. Human CD8+ T cells were retrovirally transduced with 15 candidate conversion receptors or 3 BM receptors. On day 5 post-transduction, transduction efficiency and surface expression levels were assessed by flow cytometry staining for FLAG tags or TGFβRII (BM3 only). Dot plots show live T cells from one of two donors. Numbers indicate the frequency of cells expressing engineered receptors.

[0104] Figure 4 B) TGFβ resulted in a 40% reduction in IFNγ secretion in unengineered cells (fold change 0.6). BM1 and BM2 receptors protected T cells from this negative effect, resulting in fold changes of 3.4 and 1.2, respectively. BM3 cells did not completely avoid TGFβ, with a fold change of 0.87. Our switching receptors all showed fold changes greater than 1, meaning they converted negative signals to positive signals and led to increased IFNγ secretion. SR01-04 outperformed other constructs, including the baseline receptor, with fold changes between 3.5 and 5.

[0105] To further explore the differences between candidate constructs and select a leader, experiments were repeated using 3-fold lower concentrations of αCD3 / αCD28 magnetic beads, designed to provide suboptimal TCR stimulation and thus highlight the contribution of switching receptor signaling to T cell activation. In this setting, compared to stimulation with the optimal concentration of αCD3 / αCD28 magnetic beads (… Figure 4 Compared to A), unengineered T cells had a 40% reduction in absolute IFNγ secretion. Figure 5 A), and the inhibitory effect of TGFβ was stronger (fold change of 0.55). BM1 and BM2 receptors were able to protect T cells from this negative effect, with fold changes of 3.4 and 1.9, respectively. BM3 cells failed to completely avoid TGFβ, with a fold change of 0.7. All novel conversion receptors protected T cells from TGFβ, but the differences between the best-performing constructs (SR01-SR04) became more pronounced in this setting. SR02 and SR04 both showed the highest fold changes, approximately 6-fold ( Figure 5 B). Importantly, these constructs outperform all three benchmark receptors.

[0106] To assess the safety of the conversion receptor construct, IFNγ release in the absence of αCD3 / αCD28 magnetic beads was also measured to evaluate whether the conversion receptor could independently stimulate T cells upon exposure to TGFβ, an undesirable characteristic. Conversion receptor-engineered T cells circulating in specific tissues in vivo were not activated upon encountering TGFβ. The conversion receptor should only synergize with TCRs to specifically enhance the antitumor effects of T cells.

[0107] Most conversion receptors did not activate T cells; only SR1-2, SR01, SR04, and SR12 showed a low level of IFNγ secretion in the presence of TGFβ alone. Figure 6 It is worth noting that BM1 (a TGFβ-CAR containing two intracellular signaling domains, CD28 and CD3ζ) activates T cells in the presence of TGFβ alone. Therefore, the BM1 construct is not suitable as a conversion receptor in tumor-specific TCRs or CAR-T cells, because T cells carrying BM1 may be activated in tissue sites outside the tumor, which may lead to toxicity.

[0108] To select a leader, two preferred candidates, SR02 and SR04, were evaluated more thoroughly. For this purpose, their function upon activation (via αCD3 / αCD28 magnetic beads) in the presence of a range of TGFβ concentrations (0.4–50 ng / mL) was assessed. Total IFNγ secreted after overnight activation was measured, and the fold change relative to the absence of TGFβ activation was calculated. Across a wide range of TGFβ concentrations (0.4–12.5 ng / mL), SR02+ T cells secreted more IFNγ than SR04+ T cells. Figure 7 A). At lower TGFβ concentrations (0.4 to 6.2 ng / mL, a range that may be close to physiological TGFβ concentrations), the fold change in SRO2 was also greater. Figure 7 B). On the other hand, SR04+ T cells showed a tendency to release higher levels of IFNγ under non-physiological high TGFβ concentrations.

[0109] To further evaluate the safety of the switching receptor and whether SR02 or SR04 induced any background activation of T cells, IFNγ secretion at a range of TGFβ concentrations was measured in the absence of TCR (CD3 / CD28) stimulation. BM1 was also included in this experiment because it can activate T cells in the absence of TCR signaling. Figure 6 T cells expressing SR02 and SR04 did not secrete significant levels of IFNγ in response to any different concentrations of TGFβ. Figure 8 In contrast, BM1 exhibits significant TGFβ dose-dependent IFNγ secretion. TGFβ may be present in healthy tissues; therefore, activation of engineered cells solely by TGFβ may lead to undesirable toxicity. The conversion receptor of this invention does not activate T cells without accompanying TCR stimulation, thus offering a more favorable safety profile than the classic BM1 designed using a classic CAR.

[0110] Based on the experiments conducted, SR02 was selected as the preferred lead candidate. SR02 exhibited excellent surface expression and effectively rescued T cells from the harmful effects of TGFβ, while no background activation of T cells was detected at physiological concentrations of TGFβ. Notably, the scFv of SR02 differs from that used by BM1. The modular composition of SR02 is as follows... Figure 9 As shown.

[0111] Functional analysis of the switching receptor (SR02) – multifunctionality

[0112] In previous experiments, IFNγ was measured as a surrogate indicator of T cell activation, and it was observed that cells expressing SRO2 resisted the inhibitory effects of TGFβ, and even secreted higher levels of IFNγ upon activation in the presence of TGFβ. Next, these findings were extended to other cytokines important for the anti-tumor function of T cells. For this purpose, a multi-cytokine bead array (CBA) covering 13 cytokines and cytotoxic mediators related to CD8+ T cell function was used. The CBA (Biolegend's LEGENDplex) was used to measure the levels of IFNγ from... Figure 5 (αCD3 / αCD28 magnetic beads diluted 1:300) and Figure 6 (Without TCR stimulation) SR02 supernatant of the culture shown. Besides IFNγ, T cells expressing SR02, when activated with αCD3 / αCD28 in the presence of TGFβ, also secrete higher levels of IL-2 and TNFα. Figure 10 A). IL-2 is a potent stimulant for T cell function and survival, while TNFα is a pro-inflammatory cytokine with intrinsic tumor-killing activity. This suggests that the protective effect of SR02 against TGFβ inhibition is not limited to a single cytokine but broadly enhances cytotoxic T cell function. Interestingly, in this 24-hour experiment, the secretion of granzyme A and granzyme B (GrzA and GrzB, respectively) was unaffected by TGFβ. This is likely because these enzymes are already present in the granules and are secreted immediately upon activation, while IFNγ, IL-2, and TNFα require prior expression and translation. Other cytokines measured by CBA were detected only at low concentrations (IL-4, IL-10, IL-6, IL-17A, sFas, sFasL, perforin, and Gnly), thus the effect of SR02 could not be measured. T cells expressing SR02 did not exhibit autonomous cytokine secretion beyond basal levels when exposed to TGFβ without TCR stimulation. Figure 10 B).

[0113] Functional analysis of T cells expressing switching receptors – repetitive stimulation

[0114] Besides its short-term effects on cytokine secretion, TGFβ can reduce long-term T cell proliferation and survival. To address this issue, a repeated stimulation assay was used to test whether SR02 could protect T cells from this harmful effect. This assay simulated the conditions of persistent antigen and TGFβ exposure in the tumor microenvironment. Unengineered and SR02-expressing T cells were stimulated every 2–3 days with or without 10 ng / ml TGFβ using αCD3 / αCD28 magnetic beads, and cell number and viability were monitored between each stimulation (stimulation was performed on days 0, 3, 5, 7, 10, 12, and 14). In the absence of TGFβ, the number of unengineered and SR02+ T cells increased after 6 stimulations (d14) and then began to decline. Figure 11 A). The viability of both cell populations remained stable over four stimulations (d10), then slowly declined until the end of the observation period (d17). Figure 11 B). In the presence of TGFβ, unengineered T cells proliferate only after two stimulations (d5), then their number and viability rapidly decline. By day 10 (after 4 stimulations), only a very small number of viable T cells remain (approximately 1.2% viability), making further stimulation impossible. Figure 11 C&D). In contrast, SR02+ T cells expanded after four stimulations in the presence of TGFβ, after which their numbers slowly declined. Their viability began to decline slowly after the third stimulation (day 7), reaching approximately 8% at the end of the experiment. Figure 11 C&D).

[0115] This study demonstrated that SR02+ T cells significantly outperformed unengineered T cells under harsh conditions of repeated stimulation and TGFβ exposure, mirroring the in vivo situation within the tumor microenvironment. By day 10, the number of SR02+ T cells was 8.7 times higher than the initial cell number, while unengineered cells had shrunk to 0.08 times the initial cell number, a difference of approximately 100-fold. This significant improvement in T cell durability provided by SR02 can significantly improve clinical outcomes.

[0116] Regarding safety, the trial also showed that SR02+ T cells do not proliferate indefinitely when exposed to TGFβ, and are therefore unlikely to develop into autonomously dividing T-cell leukemia and / or lymphoma.

[0117] Functional analysis of therapeutic T cells expressing cancer-specific TCRs and switching receptors

[0118] To evaluate the immunosuppressive effect of TGFβ on TCR-engineered T cells and the ability of the switching receptor to overcome it, we generated T cells expressing CTC127 (a TCR targeting a cancer-specific antigen) alone or in combination with the switching receptor. We co-cultured the cells with cancer cell lines that naturally expressed the CTC127 homolog and HLA (A375 and H1703). To mimic the solid tumor microenvironment, we added 10 ng / mL of recombinant TGFβ to the cells at the start of co-culture. After 18 hours, we measured IFNγ in the supernatant as a correlation with T cell activation in target cancer cells. CTC127-T cells recognized both cell lines in the absence of TGFβ, but the recognition strength decreased in the presence of TGFβ. Figure 12 A&B). TGβ leads to a 30% to 20% reduction in IFNγ secretion. Conversely, CTC127-T cells, which also express the switching receptor, are insensitive to inhibition of exogenous TGFβ. Figure 12 A&B cells, even in the absence of exogenous TGFβ, secreted more IFNγ than CTC127 cells. Figure 12 (A / B). In the absence of exogenous TGFβ, the increased IFNγ secretion of CTC127-converting receptor T cells compared to CTC127T cells can be explained by the cancer cells' own endogenous production of TGFβ, which is known, at least for the A375 cell line. These data suggest that the conversion receptor effectively supports target recognition by TCR-engineered T cells in the presence of TGFβ in vitro.

[0119] Next, we tested the effect of the receptor conversion in a therapeutic in vivo model. We subcutaneously injected A375 cancer cells into NSG mice and waited 8 days for the tumors to establish to an average size of 100 mm. 3 Mice were then intravenously injected with engineered or unengineered T cells. These T cells were transduced with CTC127x and a co-receptor (CTC127x + CoR, as disclosed in European patent application 23191753.5); or with CTC127x, a co-receptor, and an SRO2 switching receptor (CTC127x + CoR + SWITCH); or untransduced (unengineered). We then tracked tumor growth in mice for a total of 35 days. We tested two therapeutic T cell doses: 1x102 7 And a dose 5 times lower, 0.2 x 10⁻⁶. 7 The aim was to create harsher conditions for T cells and to fully analyze the functional differences between T cells expressing the switching receptor (SR02) and those not expressing it. Receive 1x10 7 Mice with engineered T cells rapidly rejected tumors and remained tumor-free for the remainder of the experiment. Figure 13 A). At this dose, we did not see a difference between the CTC127x + CoR and CTC127x + CoR + SWITCH groups; both resulted in complete remission with similar kinetics. We observed a difference at the lower 0.2x10⁻¹⁰ dose. 7 Differences in efficacy were observed at different cell doses. CTC127x-T cells expressing the co-receptor controlled growth but failed to completely reject the tumor; 3 out of 4 mice survived, with one showing complete remission. Figure 13 B). In contrast, T cells expressing the switching receptor completely rejected the tumor in all mice, resulting in 100% complete remission (B). Figure 13 B).

[0120] These results were confirmed using different cancer-specific TCRs:CTC45. In this model, A375 cancer cells were engineered to express the CTC45 homolog. At high dose levels (1x10⁻¹), 7 , Figure 14 At dose A), we did not see a difference between the CTC45 and CTC45+ receptor switching groups; both resulted in complete remission with similar kinetics. At low dose levels (0.2 x 10⁻⁶), 7 , Figure 14 In B), CTC45+ receptor conversion resulted in 100% complete remission, while CTC45 alone resulted in 75% complete remission (3 out of 4 mice). The increased remission response in the CTC45 group in this model can be explained by the high levels of antigen expressed by engineered A375 cancer cells.

[0121] These data demonstrate the advantages of combating TGFβ in the solid tumor microenvironment. TCR-engineered T cells expressing the switching receptor (at a 5-fold lower dose) showed improved function and cleared tumors, while TCR-engineered T cells without the switching receptor were ultimately suppressed by TGFβ and failed to effectively control tumor growth.

[0122] in conclusion

[0123] For patients receiving receptor-engineered immune cell therapy, clinical benefit may depend on the ability of immune cells to overcome TGFβ inhibition in the solid tumor tumor mesenchyme (TME). With this in mind, the inventors developed 15 candidate switching receptors to convert negative signals from TGFγ to positive signals. They tested each candidate switching receptor, along with three benchmark receptors, in an experiment designed to evaluate the constructs' ability to rescue T cell function from TGFγ inhibition. Furthermore, they tested whether candidate SRs induced T cell activation when exposed to TGFγ in the absence of TCR stimulation. SR01–SR04 significantly outperformed all state-of-the-art benchmark receptors and the constructs tested elsewhere.

[0124] Based on its expression level and ability to effectively convert negative TGFβ signals into stimulatory signals in the absence of background activation, SR02 was selected as the lead conversion receptor construct. Further characterization of SR02+ T cells showed that they maintained their function in the presence of TGFβ and secreted high levels of cytokines (IL-2, IFNγ, and TNFα) only when their TCRs were stimulated. SR02+ T cells were better suited to proliferate and survive under harsh conditions of repeated stimulation in the presence of TGFβ than conventional T cells. The improved function and increased persistence of SR02+ T cells together constitute a decisive feature that may help improve the treatment efficacy and prognosis of patients with solid tumors.

[0125] The conversion receptor of the present invention, in particular SRO2, in combination with tumor-specific TCRs, represents a next-generation approach with great potential for the effective treatment of solid tumors with highly inhibitory tumor microenvironments. References 1. F. Cavallo, C. De Giovanni, P. Nanni, G. Forni, P.-L. Lollini, Cancer Immunol Immunother .60, 319–26 (2011). 2. D. Hanahan, R. a Weinberg, Cell .144, 646–74 (2011). 3. DVF Tauriello, E. Sancho, E. Batlle, Nat Rev Cancer .22, 25–44 (2022). 4. SF Shariat et al. , Clinical Cancer Research .10, 1992–1999 (2004). 5. H. Tsushima et al. , Gastroenterology .110, 375–382 (1996). 6. S. Desruisseau et al. , Br J Cancer .94, 239–246 (2006). 7. S. Mariathasan et al. , Nature .554, 544–548 (2018). 8. C. J. Martin et al. , Sci Transl Med .12, 1–16 (2020). 9. G. M. Bendle, C. Linnemann, L. Bies, J.-Y. Song, T. N. M.Schumacher, J Immunol .191, 3232–9 (2013). 10. C. M. Bollard et al. , Journal of Clinical Oncology .36, 1128–1139(2018). 11. S. Sukumaran et al. , Cancer Discov .8, 972–987 (2018). 12. T. L. Roth et al. , Cell .181, 728-744.e21 (2020). 13. WO2014172584A1. 14. WO2021244486 A1. 15. WO2017075433 A1. 16. Z. L. Chang et al. , Nat Chem Biol .14, 317–324 (2018). 17. J. D. Silk et al. , The Journal of Immunology .208, 169–180 (2022). 18. EP3769816A1. 19. WO 2017 / 158019 A1. 20. WO 2009 / 003671 A2. 21. WO 2017 / 158029 A1. 22. WO 2022 / 243514 A1. 23. WO2006086469A2. 24. European patent application 23 191 753.5. < / scid>

Claims

1. A nucleic acid encoding a switching receptor comprising a) a TGFb binding domain, namely a scFv, b) a hinge domain, c) a transmembrane domain, d) an intracellular costimulatory signaling domain, wherein, Upon binding of the switching receptor to TGFp, the intracellular costimulatory signaling domain is not able to activate the T cell in the absence of further stimulation. wherein, optionally, the intracellular signaling domain comprises the intracellular signaling domain of CD137, CD28, CD4 and / or CD8.

2. The nucleic acid according to claim 1, wherein the intracellular signaling domain is the intracellular signaling domain of CD137 or a combination of the intracellular part of human CD8 and the intracellular signaling domain of human CD137.

3. The nucleic acid according to any one of the preceding claims, wherein the scFv comprises a variable heavy chain (VH) and a variable light chain (VL) domain, wherein a) the VH domain comprises CDR1 as set forth in SEQ ID NO: 1, CDR2 as set forth in SEQ ID NO: 2 and CDR3 as set forth in SEQ ID NO: 3 and the VL domain comprises CDR1 as set forth in SEQ ID NO: 4, CDR2 as set forth in SEQ ID NO: 5 and CDR3 as set forth in SEQ ID NO: 6 b) the VH domain comprises CDR1 as set forth in SEQ ID NO: 11, CDR2 as set forth in SEQ ID NO: 12 and CDR3 as set forth in SEQ ID NO: 13 and the VL domain comprises CDR1 as set forth in SEQ ID NO: 14, CDR2 as set forth in SEQ ID NO: 15 and CDR3 as set forth in SEQ ID NO: 16 c) the VH domain comprises CDR1 as set forth in SEQ ID NO: 21, CDR2 as set forth in SEQ ID NO: 22 and CDR3 as set forth in SEQ ID NO: 23 and the VL domain comprises CDR1 as set forth in SEQ ID NO: 24, CDR2 as set forth in SEQ ID NO: 25 and CDR3 as set forth in SEQ ID NO: 26, preferably a).

4. The nucleic acid according to claim 3, wherein the scFv comprises a) a variable heavy chain (VH) domain as set forth in SEQ ID NO: 7 and a variable light chain (VL) domain as set forth in SEQ ID NO: 8, wherein the scFv optionally has SEQ ID NO: 9, or b) a variable heavy chain (VH) domain as set forth in SEQ ID NO: 17 and a variable light chain (VL) domain as set forth in SEQ ID NO: 18, wherein the scFv optionally has SEQ ID NO: 19, or c) a variable heavy chain (VH) domain as set forth in SEQ ID NO: 27 and a variable light chain (VL) domain as set forth in SEQ ID NO: 28, preferably a).

5. The nucleic acid according to claim 4, wherein the scFv is PET1073G12 as set forth in SEQ ID NO: 9 or a variant thereof having at least 90% sequence identity to SEQ ID NO: 9, preferably PET1073G12 as set forth in SEQ ID NO:

9.

6. The nucleic acid according to any one of the preceding claims, wherein the hinge domain is an IgG hinge domain, such as an IgG4 hinge domain, a CD28 hinge domain or a CD8 hinge domain, preferably an IgG4 hinge domain.

7. The nucleic acid according to any one of the preceding claims, wherein the transmembrane domain is selected from the transmembrane domain of CD8, CD28 or CD137, preferably the transmembrane domain of CD8.

8. The nucleic acid according to any one of the preceding claims, wherein the switching receptor comprises the intracellular signaling domain of CD137.

9. The nucleic acid according to any one of the preceding claims, wherein the switching receptor comprises a combination of the intracellular part of human CD8 and the intracellular signaling domain of human CD137.

10. The nucleic acid according to any one of the preceding claims, wherein the switching receptor further comprises an extracellular tag, optionally a FLAG tag N-terminal of a TGFp binding domain.

11. The nucleic acid according to any one of the preceding claims, wherein the switching receptor further comprises an N-terminal leader sequence, optionally a mouse kappa light chain leader sequence.

12. The nucleic acid according to any one of the preceding claims, wherein the switching receptor has the sequence of any one of SEQ ID NOs: 41-55, or at least 90% sequence identity to any one of SEQ ID NOs: 41-55, preferably the sequence of SEQ ID NO:

42.

13. A switching receptor encoded by the nucleic acid of any one of claims 1-12.

14. A cell comprising the nucleic acid of any one of claims 1-12, wherein the cell is optionally a human T cell, NK cell or NKT cell, such as a T cell.

15. The cell according to claim 14, wherein the cell expresses a switching receptor, preferably the switching receptor has at least 90%, optionally 100%, sequence identity to SEQ ID NO:

42.

16. The cell according to claim 15, wherein the cell further expresses a transgenic T cell receptor.

17. The cell according to claim 15, wherein the cell further expresses a chimeric antigen receptor.

18. The cell according to any one of claims 14-17, which is a gd T cell.

19. The cell according to any one of claims 14-17, which is an aP T cell.

20. A pharmaceutical composition comprising the nucleic acid of any one of claims 1-12, the switching receptor of claim 13 and / or the cell of any one of claims 14-19, which cell is a human T cell, NK cell or NKT cell, wherein the pharmaceutical composition optionally further comprises a pharmaceutically acceptable buffer or excipient.

21. The pharmaceutical composition according to claim 20, for use in the treatment of cancer and infectious diseases, optionally for adoptive T cell therapy.

22. The pharmaceutical composition according to any one of claims 20-21, for use in allogeneic adoptive T cell therapy, wherein preferably the T cells are gd T cells.

23. The pharmaceutical composition according to any one of claims 20-21, for use in autologous adoptive T cell therapy, wherein preferably the T cells are ab T cells.

Images