Targeted Ligand-Payload-Based Drug Delivery for Cell Therapy

Abstract

This disclosure relates to ligand - payload based drug delivery for cell therapy, and a drug delivery platform is disclosed herein that provides for flexible fine - tuning of cell therapy. In particular, an engineered fusion protein is conjugated to a high - affinity ligand carrying at least one drug payload that is internalized by the cells to be transplanted, to observe or modulate the effects of the transplanted cell therapy.

Description

[0001] This patent application is a divisional application of the patent application with application number 2018800243086, application date February 17, 2018, and invention title "Targeted Ligand-Payload Based Drug Delivery for Cell Therapy".

[0002] Cross References

[0003] This application claims the benefit of U.S. Provisional Application 62 / 460,118, filed February 17, 2017, under 35 U.S.C. § 119(e). The content of this application is hereby incorporated by reference in its entirety. Technical Field

[0004] The present disclosure provides a drug delivery platform for cell therapy. In particular, an engineered protein is conjugated with a high-affinity ligand carrying at least one drug payload to be internalized by cells transplanted with the engineered protein, thereby modulating the effect of transplanted cell therapy. Background Art

[0005] In the past few decades, great progress has been made in the field in terms of cell types, delivery methods, and suitable disease models. In terms of cell types, current cell therapies can be broadly classified into chimeric antigen receptor (CAR), cells for tumor models, and stem cell-based regenerative medicine.

[0006] CAR T, also known as chimeric T cell receptor, chimeric immunoreceptor, or artificial T cell receptor, enables immune effector cells (usually T cells or NK cells) to recognize target cells with corresponding antigens and exercise their cytotoxic activity. The emergence and development of CAR-T technology have provided hope for certain types of cancer, making CAR-T a superstar in both the fields of biomedical research and clinical research.

[0007] Regenerative medicine is a game-changing medical field with the potential to completely cure damaged tissues and organs, providing solutions and hope for those suffering from currently irreparable conditions. Advances in developmental and cell biology, immunology, and other fields have provided new opportunities to improve existing regenerative therapies and develop new ones.

[0008] Stem cells have the ability to develop into many different types of cells through a process called differentiation, such as skin cells, brain cells, lung cells, etc. Stem cells are a key component of regenerative medicine because they open the door to new clinical applications.

[0009] Multiple types of stem cells, including adult and embryonic stem cells, can be used in regenerative medicine. In addition, various types of progenitor cells are used in regenerative medicine, such as those found in umbilical cord blood and bioengineered cells called induced pluripotent stem cells. Each type has unique properties, and some are more versatile than others.

[0010] Many of the regenerative therapies being developed start with the cells of a specific patient. For example, a patient's own skin cells can be collected, reprogrammed in the laboratory to give them certain characteristics, and delivered back to the patient to treat his or her disease.

[0011] Although anti-CD19 CAR T has achieved great success in the clinical application of leukemia treatment, lethal side effects such as cytokine storms resulting from the rapid lysis of tumor cells, and the killing of normal CD19+ B cells by rapidly proliferating anti-CD19 CAR T cells require more refined control of CAR T cells. In stem cell-based regenerative therapies, efforts have been made to better understand the differentiation process and trophic effects of transplanted cells in target tissues. At the same time, these processes can be altered by some small molecule drugs specifically delivered to stem cells, thereby further promoting the regeneration of target tissues.

[0012] Another long-term concern for CAR T cells and other stem cell-based regenerative therapies is the tumorigenic potential of these transplanted cells. In summary, having a private gateway to control the activity of transplanted cells (CAR T cells or stem cells) after they are transplanted would be ideal. Summary of the Invention

[0013] The present disclosure provides a drug delivery platform for fine-tuning cell therapy. The drug delivery system comprises:

[0014] a. An engineered protein on a target cell for transplantation, wherein the fusion protein comprises a first component and a second component, the first component and the second component are linked by a peptide linker, the first component is a non-membrane protein, and the second component is a membrane-anchored peptide or protein;

[0015] b. At least one small ligand conjugated to the linker, wherein the at least one small ligand has an inherent high affinity for at least one component of the engineered protein; and

[0016] c. At least one drug payload conjugated to the linker, wherein when the small ligand binds to at least one component of the engineered protein, the drug payload associates with the target cell.

[0017] In some embodiments, the foregoing drug delivery platform has a drug payload of an imaging agent. Such imaging agents can be selected from the group consisting of the fluorescent dyes rhodamine, fluorescein, and S0456. Alternatively, such imaging agents are selected from the group consisting of radioisotope chelated imaging moieties, EC 20 chelating heads, NOTA, and DOTA.

[0018] In some embodiments, the foregoing drug delivery platform has a drug payload of a cytotoxic drug. Such cytotoxic drugs can be selected from the group consisting of tubulysin, DM1, DM4, and auristatin.

[0019] In some embodiments, the foregoing drug delivery platform has a drug payload of a gene expression modulator.

[0020] In some embodiments, the foregoing drug delivery platform has a drug payload of a cell activity modulator.

[0021] In some embodiments, the foregoing modulators can be selected from the group consisting of dasatinib, MEK1 / 2 inhibitors, and PI3K inhibitors; HDAC inhibitors, kinase inhibitors, and metabolic inhibitors; GSK3β inhibitors, MAO-B inhibitors, and Cdk5 inhibitors.

[0022] In some embodiments, the foregoing modulator is a phosphatase inhibitor, RORγt agonist, or siRNAmi181a1.

[0023] In some embodiments, the foregoing drug payload is a phosphatase inhibitor, including but not limited to inhibitors against SHP1 / 2 and TC-PTP.

[0024] In some embodiments, when a small ligand binds to at least one component of the engineered protein, the foregoing drug payload in the drug delivery platform is further internalized by the target cell.

[0025] In some embodiments, the foregoing drug delivery platform has a cleavable linker to link the small ligand and the payload drug. The linker can be selected from the group consisting of:

[0026]

[0027] In some embodiments, the foregoing engineered protein components are selected from the group consisting of folate receptor alpha (FRa), folate receptor beta (FRb), urokinase receptor (uPAR), FK506 binding protein (FKBP), dihydrofolate reductase (DHFR), single-chain variable fragment against fluorescein isothiocyanate (scFv against FITC), and single-chain variable fragment against dinitrophenol (scFv against DNP).

[0028] In some embodiments, the foregoing small ligands are selected from the group consisting of:

[0029]

[0030] In some embodiments, the foregoing drug delivery platform has a first component that is an FKBP, a second component that is a peptide that confers a glycosylphosphatidylinositol (GPI) anchor to the first component, and the small ligand is FK506 or a derivative thereof. In some embodiments, the FK506 derivative eliminates the calcineurin binding site.

[0031] In some embodiments, the foregoing second component is a full-length or truncated folate receptor (FR).

[0032] In some embodiments, the foregoing drug delivery platform has at least one segment of the flexible peptide linker SGGGS to link the first and second components of the engineered protein.

[0033] In some embodiments, the foregoing drug delivery platform comprises an engineered protein selected from SEQ ID NO: 1-2 (the amino acid sequences of murine FKBP-FRα and human FKBP-FRα, respectively).

[0034] In some embodiments, the foregoing drug delivery platform comprises an engineered protein selected from SEQ ID NO: 12-15.

[0035] In some embodiments, the foregoing target cells for transplantation are immune cells. For example, the immune cells can be NK cells or chimeric antigen receptor T (CAR T) cells. Such CAR T cells can express an amino acid sequence selected from SEQ ID NO: 3-4.

[0036] In some embodiments, the foregoing drug delivery platform has a small ligand conjugate of Formula I.

[0037]

[0038] In some embodiments, the foregoing drug delivery platform has target cells for transplantation that are CAR T cells expressing SEQ ID NO: 3 (the amino acid sequence of a murine anti-CD19 CAR T construct) or SEQ ID NO: 4 (the amino acid sequence of a human anti-CD19 CAR T construct).

[0039] In some embodiments, the foregoing small ligand is further conjugated to a fluorescent dye or a radioactive probe for tracking drug internalization.

[0040] In some embodiments, the foregoing drug delivery platform comprises a small ligand that is further conjugated to a modulator of endogenous gene expression or a modulator of transduced transgene expression.

[0041] In some embodiments, for the foregoing drug delivery platform, the target cells for transplantation are stem cells, progenitor cells, or transplanted cells designed to synthesize biochemical substances lacking in the patient.

[0042] The present disclosure also provides CAR T cells comprising a construct expressing an amino acid sequence selected from SEQ ID NOs: 12-15.

[0043] The present invention also provides a DNA construct encoding an amino acid sequence selected from SEQ ID NOs: 12-15.

[0044] The present disclosure also provides a DNA construct encoding an FKBP-FRa fusion receptor, which comprises any one of SEQ ID NOs: 1-2 operably linked to an EF1a promoter in an expression vector. In some embodiments, such an expression vector is pWPI having SEQ ID NO: 5.

[0045] The present invention also provides a DNA construct comprising any one of SEQ ID NOs: 6-8.

[0046] The present disclosure also provides a transplanted cell comprising the inserted genes hFKBP-FR (SEQ ID NO: 7) and human anti-CD19 CAR (SEQ ID NO: 9).

[0047] The present disclosure also provides a transplanted cell comprising the inserted genes mFKBP-FR (SEQ ID NO: 8) and mouse anti-CD19 CAR (SEQ ID NO: 10).

[0048] The present disclosure also provides a method for modulating the effect of cell therapy. The method includes:

[0049] a. Identifying target cells for transplantation, wherein the transplanted target cells have cell therapy functions;

[0050] b. Providing an engineered fusion protein on the surface of the target cells for transplantation, the fusion protein comprising a first component and a second component, the first component and the second component being linked by a flexible peptide linker, the first component being a non-membrane protein, and the second component being a glycosylphosphatidylinositol (GPI)-anchored peptide or protein;

[0051] c. Providing an effective payload of a drug conjugate to the target cells, wherein the drug payload is conjugated to a small ligand via a linker and optionally conjugated to a fluorescent dye, wherein the small ligand binds to at least one component of the engineered fusion protein with high affinity and is internalized by the target cells together with the drug payload;

[0052] d. Releasing the drug within the target cell to modulate the therapeutic function of the target cell.

[0053] In some embodiments, the aforementioned cellular therapy function is to provide optical guidance surgery to a subject.

[0054] In some embodiments, the aforementioned cellular therapy function is to control target cell proliferation.

[0055] In some embodiments, the aforementioned cellular therapy function is to perform cytotoxicity on cancer cells conjugated to the target cell.

[0056] In some embodiments, the aforementioned transplanted target cell is an immune cell. For example, the target cell is a CAR T cell.

[0057] In some embodiments, the aforementioned transplanted target cell is a stem cell, progenitor cell, or transplanted cell designed to synthesize a biochemical substance that the patient lacks.

[0058] In some embodiments, the aforementioned drug payload is an imaging agent that is a fluorescent dye selected from rhodamine and FITC, or a radioisotope imaging agent selected from EC20 chelating head, NOTA, and DOTA.

[0059] In some embodiments, the aforementioned drug payload is a cytotoxic drug selected from the group consisting of tubulysin, DM1, DM4, and auristatin.

[0060] In some embodiments, the aforementioned drug payload is a gene expression regulator selected from kinase inhibitors consisting of dasatinib, MEK1 / 2 inhibitor, and PI3 kinase inhibitor, or siRNA of mi181a1.

[0061] In some embodiments, the aforementioned transplanted target cell comprises a fusion protein selected from SEQ ID NO: 12 - 15.

[0062] In some embodiments, the aforementioned engineered protein component is selected from FRa, FRb, uPAR, FKBP, DHFR, scFv against FITC, and scFv against DNP.

[0063] In some embodiments, the aforementioned small ligand is selected from the group consisting of:

[0064]

[0065] In some embodiments, the aforementioned linker connecting the small ligand and the payload drug is selected from the group consisting of:

[0066]

[0067] In some embodiments, the transplanted target cells comprise an engineered FKBP-linker-FRa fusion protein selected from the group consisting of SEQ ID NO:1 and SEQ ID NO:2

[0068] In some embodiments, the transplanted target cells are CAR T cells that comprise an engineered anti-CD19 CAR T construct selected from SEQ ID NO:3 and SEQ ID NO:4.

[0069] In some embodiments, the drug conjugate is an FK506 releasable linker of Formula I, wherein the binding domain of FK506 has an affinity for FKBP of from about 4 pM to about 100 pM.

[0070]

[0071] In some embodiments, the transplanted target cells are CAR T cells, and the drug conjugate is selected from the group consisting of GSK3b inhibitors, MAPK inhibitors to control the over cytokine storm of the transplanted CAR T cells.

[0072] In some embodiments, the transplanted target cells are CAR T cells, and the drug conjugate is a regulator designed to control unwanted T cell proliferation.

[0073] In some embodiments, the transplanted target cells are stem cells or progenitor cells, and the drug conjugate is a GSK3b inhibitor to enhance fracture repair.

[0074] In some embodiments, the transplanted target cells are stem cells, progenitor cells or transplanted cells designed to synthesize biochemicals that the patient lacks; and the drug conjugate is selected from the group consisting of MAO-B inhibitors and cdk5 inhibitors to treat Parkinson's disease or other neurodegenerative diseases.

[0075] In some embodiments, the transplanted target cells are NK cells, and the drug conjugate is a RORγt agonist to control Th17 cell-mediated immune responses.

[0076] These and other features, aspects, and advantages of the present invention will be better understood with reference to the following drawings, related description, and claims. BRIEF DESCRIPTION OF THE DRAWINGS

[0077] Figure 1: A. Overview of an FKBP-FRa and FK506-payload based drug delivery platform for cell therapy; B. Illustration of a stealth pathway platform for CAR T cell payload delivery.

[0078] Figure 2A. Left: Chemical structure of FK506, with the FKBP binding site (yellow) and derivatization site (red) highlighted. Right: Crystal structure of the ternary complex of calmodulin A fragment (green), calmodulin B (cyan), FKBP12 (purple

[0079] ), and FK506 (yellow), PDB: 1TCO

[0080] Figure 2B . Various combinatorial selections of the two modules in the engineered fusion protein and their respective ligand selections, highlighting potential derivatization sites.

[0081] Figure 3 : Left: Negative and positive regulation of CAR T cell activity, adapted from The quest for spatio - temporal control of CAR T cells, Sun J. etc. 2015. Right: Mechanism of FKBP - caspase 9 - mediated apoptosis induced by AP1903 (FK506 dimer) and structure of AP1903, adapted from Inducible Apoptosis as a Safety Switch for Adoptive Cell Therapy, Malcolm K.B. etc. 2011

[0082] Figure 4: a. Map of the pWPI expression vector with FKBP - FRa insert (hFRa 1 - 24: red, FKBP: yellow, hFRa 25 - 258: red) b. FKBP - FRa - transduced K562 cells showed a higher band of approximately 50 kDa (37 kDa of FRa plus 12 kDa of FKBP) compared to FRa - positive KB cells and untransduced K562 cells. c. Payload vector construct and CAR T construct design. d. Construct design of FKBPFR3GS (recorded as FF3). From the N - terminus to the C - terminus, it has the 1 - 24 aa of human FRa as a signal peptide, human FKBP protein, a linker of three Gly - Ser, and then the 25 - 258 aa of human FRa. In FKBPFR1GS (recorded as FF1), the linker of three Gly - Ser in FF3 is replaced with a linker of one Gly - Ser, and the other parts remain unchanged. e. Construct design of 4m5.3FR. From the N - terminus to the C - terminus, it has the hCD8 signal peptide, scFv of the 4M5.3 antibody against FITC, GS linker, and the 25 - 258 aa of human FRa.

[0083] Figure 5.Interference between FR and FKBP in the FKBPFR1GS fusion receptor. Binding of folic acid in the FKBPFR1GS fusion protein blocks the binding of FK506-rhodamine at as low as 0.01 nM and completely eliminates FK506-rhodamine binding at 50 nM.

[0084] Figure 6 .FKBPFR1GS jurkat cells show reduced FK506-rhodamine intensity after binding OTL38. FRET from FK506-rhodamine (donor) to OTL38 (FA-S0456, acceptor, ex / em: 774 / 794 nm) within the fusion receptor indicates the interaction between FR and FKBP.

[0085] Figure 7 .Increasing the linker length between FKBP and FR significantly reduces the interference between the two moieties. Compared with FF1 (1GS between FKBP and FR), FF3 (3GS between FKBP and FR) retains the binding of FK506-rhodamine in the presence of 10 nM FA, which is comparable to the physiological concentration of FA in the human body.

[0086] Figure 8 .PI-PLC treatment releases the GPI-anchored fusion receptor FF3. Jurkat T cells with the FF3 fusion receptor show saturated binding with 20 nM FA-FITC (EC17), while after treatment with 5 mU PI-PLC or 50 mU PI-PLC, FA-FITC loses its binding to the cells, indicating the release of the GPI-anchored FF3 fusion receptor.

[0087] Figure 9 .FA-rhodamine binding curve in the FKBPFR3GS fusion receptor. The FKBPFR3GS fusion receptor stably expressed on human T cells can bind the folic acid derivative (FA-rhodamine) with high affinity (Kd = 0.95 nM), and this affinity is comparable to the FA-rhodamine affinity (Kd ~ 1 nM) in FR+KB cells. Therefore, the binding characteristics of FR in the fusion receptor are retained.

[0088] Figure 10 .FK506-rhodamine binding curve in the FKBPFR3GS fusion receptor. The FKBPFR3GS fusion receptor stably expressed on human T cells is able to bind the FK506 derivative (FK506-rhodamine) with high affinity (Kd = 3.93 nM), which means the binding characteristics of FKBP in the fusion receptor are retained.

[0089] Figure 11. SLF-FITC binds to the FKBPFR3GS fusion receptor with a relatively high binding affinity (Kd = 62 nM), and this binding is competitively blocked by free SLF (100x, pre-incubated). Compared with the parental ligand FK506, SLF (a mimic of FK506) exhibits a 10-fold lower binding affinity for the FKBPFR fusion receptor, which is consistent with previous reports.

[0090] Figure 12 . The FA-rhodamine binding curve in the 4M5.3FR fusion receptor. FA-rhodamine can bind to the 4M5.3FR fusion receptor stably expressed on human T cells with a high affinity (Kd = 2.25 nM), and this affinity is comparable to the FA-rhodamine affinity (Kd ~1 nM) in FR+KB cells. Therefore, the FR binding characteristics are retained in the 4M5.3FR fusion receptor.

[0091] Figure 13 . The FITC-AF647 binding curve in the 4M5.3FR fusion receptor. FITC-AF647 can bind to the 4M5.3FR fusion receptor stably expressed on human T cells with a high affinity (Kd = 8.03 nM). 100x comp represents free sodium fluorescein. The binding characteristics of scFv 4M5.3 to FITC are retained in the 4M5.3FR fusion receptor.

[0092] Figure 14 . FA-tubulin can mediate receptor-specific killing effects against FF3+ human T cells. This effect is blocked by FA compensation (100x pre-incubation with FA). This means the successful internalization and release of free drug tubulin through the FF3 fusion receptor system.

[0093] Figure 15 . FA-tubulin specifically kills the hFF3+ population in mixed human T cell cultures. As the FA-Tub concentration increases, the absolute number of hFF3+ cells decreases, and at high concentrations, hFF3- cells are also killed by the released drug and the bystander effect.

[0094] Figure 16 . SLF-Tub specifically kills hFF3+ Jurkat cells with an IC 50 = 138 nM. This indicates the successful internalization of SLF-Tub through the FKBPFR3GS fusion receptor and the release of tubulin intracellularly.

[0095] Figure 17 . Both FITC-DM4 and FITC-Tub can specifically kill 4M5.3FR+ human T cells, and FITC-Tub has a higher IC 50. Compensation blocking of receptor-mediated killing effect by free FITC sodium (pre-warmed 100x). This means the successful internalization and release of FITC-cytotoxic drug into T cells via the 4M5.3FR fusion receptor.

[0096] Figure 18 FITC-Tubulin specifically kills the 4M5.3FR+ population in mixed human T cell cultures. As the concentration of FITC-Tub increases, the absolute number of 4M5.3FR+ cells decreases, and at high concentrations, 4M5.3FR- cells are also killed by the released drug and the bystander effect.

[0097] Figure 19 . FITC-DM4 specifically kills the 4M5.3FR+ population in mixed human T cell cultures. As the concentration of FITC-DM4 increases, the absolute number of 4M5.3FR+ cells decreases, and at high concentrations, 4M5.3FR- cells are also killed by the released drug and the bystander effect.

[0098] Figure 20 . Dasatinib (Lck inhibitor) and ibrutinib (ITK inhibitor) at a concentration of 10 nM reduce the lysis effect of anti-CD19 CART cells (FMC63 CAR T, effector) on CD19+ Raji tumor cells (target). Two effector:target ratios (E:T) were tested. Normal T cells and anti-CD19 CAR T with CD19-K562 cells were used as negative controls.

[0099] Figure 21 . FITC-dasatinib reduces the lysis effect of FMC63+4M5.3FR+ hT cells on Raji cells. This means the successful internalization and release of FITC-dasatinib into T cells via the 4M5.3FR fusion receptor and the release of dasatinib into T cells.

[0100] Figure 22 . A TC-PTP inhibitor at a concentration of 100 nM reduces the co-inhibitor molecule population in exhausted anti-CD19 CAR T cells (generated by stimulating with CD19+ Raji cells 7 times, see detailed procedure below). PD-1, LAG3, and double-positive populations are all reduced after treatment. This means that phosphatase inhibitors, such as TC-PTP inhibitors, can be used as payloads for a secret gateway platform to rejuvenate exhausted CAR T cells.

[0101] Table 1. Potential applications of the FKBP-FRa cell therapy platform and corresponding payloads.

[0102] Sequence Listing

[0103] Amino acid sequence of mouse FKBP-FRa, SEQ ID NO:1

[0104] Amino acid sequence of human FKBP-FRa, SEQ ID NO:2

[0105] Amino acid sequence of mouse anti-CD19 CAR T construct, SEQ ID NO:3

[0106] Amino acid sequence of human anti-CD19 CAR T construct, SEQ ID NO:4

[0107] Vector pWPI for human T cell transduction, SEQ ID NO:5

[0108] pMP71 gb NotIEcoRI mouse anti-CD19 for mouse T cell transduction, SEQ ID NO:6

[0109] pWPI-FRa 1-24FKBP FRa, SEQ ID NO:7

[0110] pWPI mFKBP-mFRa SGGGS, SEQ ID NO:8

[0111] pHR EcorI hAnti cd19 1D3 myc hinge cd28 cd3zeta, SEQ ID NO:9

[0112] pWPI pmei mAnti cd19 1D3 myc hinge cd28 cd3zeta, SEQ ID NO:10

[0113] FKBP-1SG-FR with GPI anchor amino acid sequence, SEQ ID NO:11

[0114] FKBP-3SG-FR with GPI anchor amino acid sequence, SEQ ID NO:12

[0115] 4M5.3-FR with GPI anchor amino acid sequence, SEQ ID NO:13

[0116] FMC63-T2A-FKBP3SGFR, SEQ ID NO:14

[0117] FMC63-T2A-4M5.3SGFR, SEQ ID NO:15

[0118] FRb with signal peptide, SEQ ID NO:16

[0119] uPAR with signal peptide, SEQ ID NO:17

[0120] SEQ ID NO:18 DHFR

[0121] SEQ ID NO:19 scFv against FITC:4M5.3 (Kd = 200 pM)

[0122] SEQ ID NO:20 scFv against FITC 4D5Flu (Kd = 10 nM)

[0123] SEQ ID NO:21 scFv against DNP SPE7 Detailed implementation manners

[0124] Although the concepts of the present disclosure are described in detail in the drawings and the specification herein, the results and their descriptions in the drawings should be considered exemplary rather than restrictive; it should be understood that only illustrative embodiments are shown and described, and all changes and modifications falling within the spirit of the present disclosure are desired to be protected.

[0125] Unless otherwise defined, scientific and technical nomenclature has the same meaning as commonly understood by those of ordinary skill in the art to which the present disclosure pertains.

[0126] The present disclosure provides a new platform for controlling the activity of transplanted cells by genetically incorporating a fusion receptor on the surface of the transplanted cells. Then, the high intrinsic affinity between a small molecule ligand and the fusion receptor on the surface of the transplanted cells is used to specifically target these transplanted cells with a drug payload conjugated to the small molecule ligand. A portion of the fusion receptor is responsible for internalizing the conjugate, and once the payload is inside the transplanted cell, it will be released through a cleavable linker. Depending on the type of transplanted cell and the desired regulation to be imposed on the transplanted cell, the drug payload can be of various functions. By varying the payload in the conjugate, such as being a cytotoxic drug or a kinase inhibitor, the drug payload can be used to control multiple aspects of the transplanted cells, such as proliferation, differentiation, or cytokine release profile.

[0127] The peptidyl-prolyl isomerase (PPIase) family consists of FK506-binding proteins (FKBPs), cyclophilins, and parvulins. There are 18 FKBPs, 24 cyclophilins, and 3 parvulins in humans. Among these, compared with FKBP12 (KD FK506 ≈0.2 nM), FKBP51 and FKBP52 share high to moderate FK506-binding affinities, with KD FK506 ≈104 nM and KD FK506 ≈23 nM, respectively. Additionally, neither of these two FKBPs is expressed on the cell membrane, resulting in little cross-binding activity in our system. Efforts have also been made to design for FKBP12 F36Vand FKBP51 F67V with a synthetic ligand that has a higher affinity than FKB PWT while retaining the overall structure of the wild-type protein. All homologs and mutant proteins, as well as their ligands as described above, can be applicable to the present disclosure.

[0128] In particular, in one embodiment, a pair of exemplary small molecule ligands and a fusion receptor are selected as FK506-FKBP. The overall process can be summarized in Figure 1A wherein the FK-506-payload first binds to the FKBP-FRa engineered cells; after this binding, the transmembrane fusion protein internalizes the payload-linker-FK506. Next, the internalized FK506-payload is cleaved at the linker and the payload is released in the cell. Depending on the cell type and payload type, the released payload drug can exert its desired function.

[0129] In Figure 1B a specific chimeric antigen receptor T cell-mediated cell therapy is shown. In this model, CAR T cells expressing a fusion protein are presented to cancer cells, the fusion protein having a structure that includes a suitable signal peptide from the N-terminus to a C-terminus, a protein module 2 linked to a protein module 1 of a GPI anchor. In some embodiments, the cancer cells have a CD19 surface protein that will be recognized by the CAR T cells and conjugated with a payload associated with the CAR T cells when at least one module of the targeting ligand binds to the fusion receptor. Generally, by virtue of the high affinity of the targeting ligand for any one of these modules, the payload associated with the ligand can be internalized into the target cells through the chimeric antigen receptor and released to conjugate with the cancer cells.

[0130] There can be many different combinations of GPI-anchored proteins, designated as module 1 and its ligand, and module 2 target cell surface protein and its respective ligand, presented in Figure 2B wherein either the module 1 protein or the module 2 protein or both can conjugate a high-affinity targeting ligand to facilitate payload delivery of the ligand conjugation. For example, a fusion protein having a structure that includes FRa-linker-FKBP is feasible, wherein FRa is conjugated with an FA derivative while FKBP is conjugated with an FK506 derivative, and either the FA derivative or the FK 506 derivative or both can be linked to a payload, such as a cytotoxic drug, an imaging agent, or a modulator. By virtue of this flexibility of carrying the same or different payloads, some unexpected synergistic or regulatory effects of different or the same payloads can be achieved, or the target cells can be observed if the payload is an imaging agent. The advantages of the flexibility and diversity of the payload delivery of this system will be understood through more examples.

[0131] Similarly, another embodiment of a ligand paired with a GPI-anchored protein can be

[0132] paired with uPAR

[0133] Also covered is that FITC or its derivative can bind to the single-chain variable fragment of an antibody against FITC

[0134]

[0135] Some advantages of choosing FK506-FKBP as an exemplary ligand-protein pair in this delivery system are worth mentioning. 1. FKBP is not a membrane protein naturally present on mammalian cell membranes, so the FK506-payload conjugate will specifically bind to target cells; 2. The FKBP protein is a relatively small protein with a molecular weight of 12 kDa, which makes it easier to fuse with other receptors with minimal perturbation to the receptor structure and internalization properties. 3. FK506 does not naturally occur in the human body, so the fusion receptor will not be blocked; 4. The binding affinity between FK506 and FKBP is about 4 pM, so the payload drug can be delivered with high affinity; 5. The co-crystal structure of FK506-FKBP is available, and the well-established derivatization sites of FK506 retain the FK506-FKBP binding while eliminating the unwanted binding between FK506 and calcineurin (see Figure 2A ).

[0136] Sequences that can modify FKBP are covered, and the corresponding FK506 ligand can be modified accordingly such that the modified form of FKBP and the modified form of FK506 still have the desired affinity as exemplified herein or better than the present disclosure.

[0137] For the other part of the fusion protein, the folate receptor (FR) is chosen as a monomer due to its well-understood internalization process. Previous studies have shown that "magic carbonate"-linked folate conjugates can be internalized by FR and cleaved by the reducing environment inside the cell. Using this mechanism, the FK506-FKBP-conjugated drug payload is internalized by FR and then released into the cytoplasm.

[0138] Due to the great potential and severe side effects of CAR T therapy, several controlled CAR T cell designs have been reported. Most of them focus on on / off switching by incorporating Boolean gates or cascade pathways for T cell activation (see Figure 3, the left figure, which depicts the negative and positive regulation of CAR T cell activity, is adapted from The quest for spatio - temporal control of CAR T cells, Sun J. etc. (2015). The Malcolm K.B. group designed an FKBP - caspase 9 fusion protein and used an FK506 dimer to induce apoptosis in target cells (see Figure 3 , the right figure, which depicts the mechanism of AP1903 (FK506 dimer) - induced FKBP - caspase 9 - mediated apoptosis and the structure of AP1903, is adapted from Inducible Apoptosis as a Safety Switch for Adoptive Cell Therapy, Malcolm K.B. etc. (2011).

[0139] The present disclosure has several advantages compared with these reported methods: 1. Replacing the binary on / off switch, our platform can deliver multiple types of regulatory payloads and modify many aspects of target cells, so it has great flexibility compared with the binary on / off switch; 2. The control moiety FK506 - payload is a small molecule, which allows for linear control and dose optimization compared with pre - engineered cells. 3. The platform can be used not only for CAR T cells but also for many other stem - cell - based regenerative therapies.

[0140] The greatest novelty and the most important part of this platform is the multifunctional payload, which can be selected to address potential side effects or improve the efficiency of cell therapy. For example, cytotoxic drugs can be delivered to transplanted cells if the following situations occur: 1. Cells proliferate excessively and affect normal organs or systems, such as anti - CD19 CAR T cells. 2. Cells become tumorigenic, which lies in the lentivirus - based gene modification and the intrinsic characteristics of stem cells.

[0141] On the other hand, due to the inhibitory microenvironment of the target tissue, some cell therapies are less successful. For example, in CAR T cell therapy against solid tumors, in addition to the low penetration rate, the proliferation and activation of CAR T are highly inhibited by MDSC and tumor cells. This can be potentially alleviated by T cell activation induced by RORrt agonists or MAP kinase inhibitors and the expression of granzyme B in CAR T cells induced by TLR8 agonists. Although intracellular targets (such as RORrt) may be more suitable for our payload, due to the proximity on the cell membrane, membrane receptors such as TLR8 can also be accessible.

[0142] In stem cell regenerative therapies, payloads are more diverse according to the disease model. Instead of delivering pre-fixed genes developed by using stem cells as a gene delivery platform, the present disclosure provides fine-tuning for the transplanted cells and their microenvironment and obtains the desired phenotype through a variety of small molecule payloads. Since the small molecules are conjugated with FK506 that specifically targets FKBP-FRa overexpressed in transplanted cells, non-specific targeting of normal tissues by the small molecules is also avoided.

[0143] One of many examples is the induction of BMP2 overexpression in mesenchymal stem cells (MSCs) for fracture repair in bone regeneration therapy. At the same time, by introducing GSK3β inhibitors into transplanted cells through such a drug payload delivery system, the expression levels of BMP2 and / or VEGF can be increased. GSK3β inhibitors are drugs desired for fracture repair. Therefore, GSK3β inhibitors are ideal as potential payloads for further regulating the function of transplanted MSCs and the microenvironment within the fracture site.

[0144] Another example of this drug payload delivery system involves neurodegenerative diseases, including Alzheimer's disease, Parkinson's disease, etc., where MSC-based therapies have a promising future. MSCs have been modified to overexpress GDNF, VEGF, and many other cytokines to promote neuron regeneration. At the same time, it has been confirmed that small molecules such as MAO-B inhibitors increase the expression of GDNF, NGF, and BDNF in astrocytes. Several kinase inhibitors have also been proposed for the treatment of Alzheimer's disease, such as PI3K inhibitor (BEZ235), Cdk5 inhibitor (roscovitine), and GSK3b inhibitor (NP-12). The targeted and specific delivery of these small molecules to MSCs using the FK506-FKBP pair in neurodegenerative models will improve the regenerative efficacy while avoiding the side effects of these effective inhibitors and agonists in other non-target tissues.

[0145] Materials and Methods:

[0146] Compounds and synthetic procedures:

[0147] The targeting ligand is connected to the payload through a linker. Linker optimization options are listed as follows. The payloads are characterized into 3 categories: I imaging, II cytotoxic drugs, III regulatory small molecule drugs.

[0148] Linker optimization:

[0149]

[0150] Compound classification:

[0151]

[0152] Detailed compound structure and synthesis route

[0153] FK506 - Rhodamine:

[0154]

[0155] Procedure: React rhodamine - NHS ester (1.0 equivalent) in dimethylformamide with Boc - NH - PEG3 - NH2 (1.2 equivalents) and diisopropylethylamine (3.0 equivalents) at room temperature for 2 hours. Purify the product by preparative reverse - phase HPLC with a UV detector. Deprotect the Boc group of the purified rhodamine - PEG3 - NH - Boc conjugate (1.0 equivalent) by stirring in a 1:10 TFA - dichloromethane system for 2 hours. Then dissolve the crude free - amine product in dimethylformamide and activate it with EDC (2.0 equivalents) and HOBT (2.0 equivalents) in the presence of diisopropylethylamine (3.0 equivalents). After 15 minutes, add FK506 - CO2H (1.2 equivalents, synthesized using the procedure in the following reference: Bioorg. Med. Chem., 17(2009)5763 - 5768) and stir the reaction mixture overnight. After purification on preparative reverse - phase HPLC with a UV detector (monitored at 280 nm wavelength), isolate the final FK506 - rhodamine conjugate. Load the crude product onto an Xterra RP18 preparative HPLC column (Waters) and elute with gradient conditions starting from 95% 5 mM sodium phosphate (mobile phase A, pH 7.4) and 5% acetonitrile (mobile phase B), and reach 0% A and 50% B at a flow rate of 12 mL / min within 35 minutes. The retention time of the product peak during the gradient (0 - 50% B) in a 7 - minute analytical HPLC - MS analysis = 2.5 minutes. ESI m / z = 1539.6. Abbreviations: PEG = polyethylene glycol; EDC = 1 - ethyl - 3 - (3 - dimethylaminopropyl)carbodiimide; HOBT = hydroxybenzotriazole; HPLC = high - performance liquid chromatography.

[0156] FK506 - NIR dye:

[0157]

[0158] Synthesis procedure:

[0159]

[0160] SLF - FITC:

[0161]

[0162]

[0163]

[0164]

[0165]

[0166]

[0167]

[0168]

[0169] Procedure: React DM4 (1.0 equivalent) in dimethyl sulfoxide with 2-(pyridin-2-yldithio)ethan-1-amine (1.0 equivalent) and diisopropylethylamine (3.0 equivalents) at room temperature for 1 hour. Then react the resulting crude product with FITC (1.0 equivalent), and stir the reaction mixture for 1 hour. The final FITC-DM4 conjugate is isolated after purification on a preparative reverse-phase HPLC equipped with a UV detector (monitored at a wavelength of 280 nm). The crude product is loaded onto an Xterra RP18 preparative HPLC column (Waters) and eluted with gradient conditions starting with 95% 5 mM sodium phosphate (mobile phase A, pH 7.4) and 5% acetonitrile (mobile phase B), and reaching 0% A and 100% B at a flow rate of 12 mL / min within 10 minutes. In a 7-minute analytical HPLC-MS analysis, the retention time of the product peak during the gradient (0 - 100% B) = 4.23 minutes. ESI m / z = 1244.8. Abbreviations: FITC = fluorescein isothiocyanate; HPLC = high performance liquid chromatography.

[0170]

[0171]

[0172]

[0173]

[0174]

[0175] Experimental procedure:

[0176] Cell culture

[0177] 293TN cells were cultured in DMEM containing 10% FBS and no antibiotics for lentivirus packaging. Raji and Jurkat cells were cultured in RPMI-1640 containing 10% FBS and 10% penicillin / streptomycin. Primary human T cells were isolated from hPBMC using ficoll and enriched by negative selection using the EasySep TM Human T Cell Enrichment Kit (19051, Stemcell Tech), activated for 1 day with Dynabeads CD3 / CD28 (11161D, Thermo Fisher), and cultured in TexMACS medium supplemented with 30 IU hIL2 (130-097-745, Miltenyi Biotec Inc.). T cells were cryopreserved in RPMI-1640 containing 20% human AB serum (HP1022, Valley Biomedical) and 10% DMSO.

[0178] Lentivirus packaging

[0179] Pantropic VSV-G pseudotyped lentiviruses were produced by transfecting 293TN cells with a mixture of transgenic expression vectors and packaging plasmids (CPCP-K2A, Cellecta) using Lipofectamine 2000. At 24 hours, the virus supernatant was harvested, concentrated, and then added to certain cell lines or primary T cells thawed on the same day. For T cell transduction, after adding the virus supernatant and 8 μg / ml polybrene, the cells were centrifuged at 2500 rpm for 90 minutes at 37°C.

[0180] Binding assay

[0181] For the binding assay, cells were incubated at 4°C for 30 minutes with ligand-dye alone or with ligand-dye and free ligand (100x, pre-incubated for 30 minutes). After incubation, the cells were washed 3 times and resuspended in 2% FBS PBS, and 7-AAD was added to remove dead cells. FRET imaging of FK506-rhodamine and FA-S0456: To understand the occupancy of the fusion receptor, FKBP-FRα+ jurkat cells were incubated with FA / FA-S0456, FK506 / FK506-rhodamine in the indicated order and concentrations. FRET was manifested by the loss of intensity of FA-rhodamine (due to its energy transfer to FA-OTL38 on the same or nearby receptors) and detected by a BD Fortessa flow cytometer. The results were analyzed using FlowJo software. PI-PLC treatment to release GPI-anchored proteins

[0182] 1x10 5Cells were incubated with 5 mU or 50 mU PI-PLC (P5542-5UN, Sigma) in digestion buffer (2% BSA) at 37 °C for 30 minutes; after incubation, the cells were washed three times with PBS and then incubated with ligand-dye on ice for 30 minutes.

[0183] Cell viability assay

[0184] Cells were seeded into 96-well plates and incubated with different concentrations of certain ligand-cytotoxic drugs for 2 hours, with or without 100x pre-incubated free ligand competition. After 2 hours of incubation, the cells were washed 3 times with warm medium and fresh medium was added. After 72 hours, the cell number was tested by assay (G7570, Promega) or the ligand-dye staining of receptor-positive cells was quantified.

[0185] CAR T cell lysis effect

[0186] According to the E:T ratio, 1x10 5 CAR T cells and a certain number of Raji cells were co-incubated in 96-well plates with or without treatment drugs. After 24 hours, 100 ul of the supernatant was taken for LDH assay. The lysis percentage was calculated as (treatment group - CAR T only) / (maximum lysis - CAR T only) % Depletion of CAR T cells

[0187] 1x10 6 Raji cells were repeatedly added to 1x10 6 CAR T cells in 24-well plates every 12 hours without changing the medium. The depleted state was characterized by a lower lysis effect and higher expression of co-inhibitory molecules: PD-1, LAG-3 and Tim-3.

[0188] In vivo ablation of fusion receptor-positive CAR T by ligand-cytotoxic drugs.

[0189] 4x10 5 luc+Raji cells were injected intravenously into NSG mice. Six days later, 1x10 7 fusion receptor-positive CAR T cells were injected intravenously. On day 8, the ligand-cytotoxic drug (0.5 umole / kg, 1 umole / kg) was injected intravenously once. IL2 and INFr were measured by ELISA using serum samples taken every 3 days after CAR T injection. CAR T cells in peripheral blood were counted by flow cytometry.

[0190] In vivo regulation of fusion receptor-positive CAR T by ligand-drug

[0191] 4x105 luc+Raji cells were intravenously injected into NSG mice. Six days later, 1x10 7 fusion receptor-positive CAR T cells were intravenously injected. On day 7, the ligand-drug (0.5 umole / kg, 1 umole / kg) was intravenously injected once. In the case of the CAR T depletion model, 1x10 5 CAR T was intravenously injected on day 6. Once the CAR T population was shown in the peripheral blood and the tumor burden was unstable and continued to increase, the ligand-drug (0.5 umole / kg, 1 umole / kg) was intravenously injected.

[0192] In vivo imaging and tracking of fusion receptor-positive CAR T.

[0193] 2x10 6 Raji was subcutaneously injected into the right flank of NSG mice. Fourteen days later, 1x10 7 fusion receptor-positive CAR T was intravenously injected. On day 16, the animals were administered a 99mTc-conjugated conjugate (10 nmol, 150 μCi) by intravenous injection and imaged by a SPECT imager.

[0194] Example

[0195] 1. Design and expression of the fusion protein

[0196] 1.1 Design of the FKBP-FR fusion receptor (SEQ ID NO:2)

[0197] Synthesis of FK506 derivatives: FK506-rhodamine and FK506-microtubule lysin. The synthesis is described in the Materials and Methods section.

[0198] hFRα is a GPI-anchored membrane protein that has 24 amino acids as a signal peptide at the N-terminus. To maintain the membrane presentation and internalization properties, we chose to use the full-length FRα and incorporated the hFKBP sequence and a flexible peptide linker (SGGGS) between T24 and R25 of hFRa (Figure 4a). The flexible linker was chosen to be resistant to common enzymatic digestion in the human body. Then the entire sequence was inserted into the pWPI lentiviral expression vector, where EF1a was used as the desired promoter for protein expression in transduced T cells.

[0199] 1.2 Expression of the FKBP-FR fusion receptor in transduced cells

[0200] The expression of FKBP-FRα was confirmed by Western blotting in lentivirus-transduced K562. Compared with non-transduced cells, the lysates of transduced K562 cells showed a specific band of approximately 50 kDa for the anti-hFRα antibody (Figure 4b).

[0201] 1.3 Construction of FKBPFR3GS (recorded as FF3) fusion protein

[0202] See Figure 4d (SEQ ID NO:12). From the N-terminus to the C-terminus, it has 1-24aa of human FRα as the signal peptide, human FKBP protein, a linker of three Gly-Ser, and then 2-258aa of human FRα. In the construct design of FKBPFR1GS (labeled as FF1), the linker of three Gly-Ser in FF3 was replaced with a linker of one Gly-Ser, and the other parts remained unchanged. As will be shown in the binding assay, increasing the linker length reduces the interference between the two components in the fusion protein.

[0203] 1.4 Construction of FITC-svFv-FR fusion protein with GS linker.

[0204] See Figure 4e. The construct is also called 4M5.3FR (SEQ ID NO:13). From the N-terminus to the C-terminus, it has hCD8 signal peptide, svFv of 4M5.3 (against FITC), GS linker, and 25-258aa of human FRα. 1.5. In vivo non-invasive tracking of FKBP-FRa / FKBPtFRa positive cells by FK506-99mTc PET imaging

[0205] For the CAR T cell model, 1x10 6 KB cells were subcutaneously implanted into NSG (Jackson laboratory). When the tumor reached 100mm 3 in size, 15 million anti-FITC CAR+FKBP-FRa+ or anti-FITC CAR+FKBP-FRa- human T cells were intravenously injected into the mice. FITC-FA was injected at the designated days to induce the proliferation of CAR T. The mice were imaged every two days after CAR T implantation by the following procedure. On the imaging day, according to previous reports, FK506-EC20 was formulated with 99mTc. 200 μCi of 99mTc in 100 μl solution was intravenously injected into each mouse, and whole-body images were taken by MiLab PET / CT, focusing on the tumor region, spleen, and lymph nodes. 3D images were reconstructed by PMOD software. After the last imaging at about day 10 after CAR T implantation, the mice were euthanized, and the 99mTc distribution in each organ was counted by a gamma counter.

[0206] For the hematopoietic stem cell transplantation model, humanized NSG mice were generated as previously reported. Ten million CD34+FKBP-FRa+ hHSCs were intravenously injected into humanized NSG mice. Four months later, whole-body images were taken using FK506-99mTc as described above, focusing on the bone marrow and spine.

[0207] 2. The FKBP-FRa fusion receptor specifically binds and internalizes FK506-payload

[0208] 2.1 The FKBP-FRa fusion receptor specifically binds FK506-rhodamine

[0209] For the binding assay, cells were incubated at 4 degrees for 30 minutes with ligand-dye alone or with ligand-dye and free ligand (100x, pre-incubated for 30 minutes). After incubation, the cells were washed 3 times and resuspended in 2% FBS PBS, and 7-AAD was added to remove dead cells. A BD Fortessa flow cytometer was used. The results were analyzed using FlowJo software.

[0210] 2.2. Binding of the FKBP-FRa fusion receptor to FK506-rhodamine (FRET imaging of FK506-rhodamine and FA-S0456)

[0211] To understand the occupancy of the fusion receptor, FKBP-FRa+ jurkat cells were incubated with FA / FA-SO456, FK506 / FK506-rhodamine in the indicated order and concentrations. FRET was manifested by the loss of intensity of FA-rhodamine (due to its energy transfer to FA-OTL38 on the same or nearby receptors), and was detected by a BD Fortessa flow cytometer.

[0212] Figure 5 It was shown that the binding of folic acid in the FKBPFR1GS fusion protein blocked the binding of FK506-rhodamine at as low as 0.01 nM and completely eliminated FK506 rhodamine binding at 50 nM.

[0213] Figure 6 It was shown that FKBPFR1GS jurkat cells showed decreased FK506-rhodamine intensity after binding OTL38 (folate receptor-targeted dye). FRET from FK506-rhodamine (donor) to OTL38 (FA-S0456, receptor, ex / em: 774 / 794 nm) indicated the interaction between FR and FKBP within the fusion receptor.

[0214] Figure 7It is shown that increasing the linker length between FKBP and FR significantly reduces the interference between the two components of the fusion protein. Compared with FF1 (1GS between FKBP and FR), FF3 (3GS between FKBP and FR) retains the binding of FK506-rhodamine in the presence of 10 nM FA, which is comparable to the physiological concentration of FA in the human body.

[0215] 2.3. Release of GPI-anchored FF3 fusion receptor

[0216] Treatment of T cells with FF3 fusion protein using PI-PLC results in the release of the GPI-anchored receptor FF3. Jurkat T cells with the FF3 fusion receptor show saturable binding to 20 nM FA-FITC (EC17), while after treatment with 5 mU PI-PLC or 50 mU PI-PLC, FA-FITC loses its binding to the cells, indicating the release of the GPI-anchored FF3 fusion receptor. See Figure 8 。

[0217] 2.4 The fusion protein FKBPFR3GS in human T cells retains FR binding properties

[0218] The FA-rhodamine binding curve in the FKBPFR3GS fusion receptor is shown. The FKBPFR3GS fusion receptor stably expressed on human T cells can bind the folic acid derivative (FA-rhodamine) with high affinity (Kd = 0.95 nM), and this affinity is comparable to the FA-rhodamine affinity (Kd ~ 1 nM) in FR+KB cells. Therefore, the binding properties of FR in the fusion receptor are retained. See Figure 9 。

[0219] 2.5 The fusion protein FKBPFR3GS in human T cells retains FKBP binding properties

[0220] The FK506-rhodamine binding curve in the FKBPFR3GS fusion receptor. The FKBPFR3GS fusion receptor stably expressed on human T cells is able to bind the FK506 derivative (FK506-rhodamine) with high affinity (Kd = 3.93 nM), which means that the binding properties of FKBP are retained in the fusion receptor. See Figure 10 。

[0221] 2.6 SLF-FITC binds to the FKBPFR3GS fusion receptor with a relatively high binding affinity (Kd = 62 nM)

[0222] Competition blockade of SLF-FITC binding by free SLF (100, pre-incubated). Compared with the parental ligand FK506, SLF (a mimic of FK506) exhibits 10-fold lower binding affinity for the FKBPFR fusion receptor. SeeFigure 11 and compare with Figure 10 .

[0223] Binding curve of 2.7 FA - rhodamine in the 4M5.3FR fusion receptor.

[0224] FA - rhodamine can bind to the stably expressed 4M5.3FR fusion receptor on human T cells with high affinity (Kd = 2.25 nM), and this affinity is comparable to that of FA - rhodamine in FR+KB cells (Kd ~ 1 nM). Thus, the FR - binding property is retained in the 4M5.3FR fusion receptor. See Figure 12 .

[0225] Binding curve of 2.8 FITC - AF647 in the 4M5.3FR fusion receptor.

[0226] FITC - AF647 can bind to the stably expressed 4M5.3FR fusion receptor on human T cells with high affinity (Kd = 8.03 nM). 100x comp represents free sodium fluorescein. The binding property of scFv 4M5.3 to FITC is retained in the 4M5.3FR fusion receptor. See Figure 13 .

[0227] 3.1. Killing effect of FA - tubulin on FF3+ human T cells.

[0228] FA - tubulin can mediate receptor - specific killing effect on FF3+ human T cells. This effect is blocked by FA compensation (pre - incubation with 100x of FA). This means the successful internalization and release of free drug tubulin through the FF3 fusion receptor system. See Figure 14 .

[0229] 3.2 Killing effect of FA - tubulin on the hFF3+ population in mixed human T cell cultures.

[0230] FA - tubulin specifically kills the hFF3+ population in mixed human T cell cultures. As the concentration of FA - Tub increases, the percentage of hFF3+ cells decreases. See Figure 15 .

[0231] 3.3 SLF - Tub specifically kills hFF3+ Jurkat cells with an IC 50 = 138 nM.

[0232] Incubating SLF - Tub with hFF3 Jurkat cells for 2 hours can kill receptor - positive cells. This indicates that the FKBPFR3GS fusion receptor successfully internalizes SLF - Tub and releases tubulin intracellularly. See Figure 16 .

[0233] 3.4 Killing effects of FITC-DM4 and FITC-Tub on 4M5.3FR+ human T cells.

[0234] Both FITC-DM4 and FITC-Tub can specifically kill 4M5.3FR+ human T cells, while FITC-Tub has a higher IC 50 . Compensation blocking of receptor-mediated killing effect by free FITC sodium (100x pre-warmed). This means the successful internalization and release of FITC-cytotoxic drugs into T cells through the 4M5.3FR fusion receptor. See Figure 17 .

[0235] 3.5 FITC-Tubulin specifically kills the 4M5.3FR+ population in mixed human T cell cultures.

[0236] As the concentration of FITC-Tub increases, the absolute number of 4M5.3FR+ cells decreases, and at high concentrations, 4M5.3FR- cells are also killed by the released drugs and bystander effects. See Figure 18 .

[0237] 3.6 FITC-DM4 specifically kills the 4M5.3FR+ population in mixed human T cell cultures.

[0238] As the concentration of FITC-DM4 increases, the absolute number of 4M5.3FR+ cells decreases, and at high concentrations, 4M5.3FR- cells are also killed by the released drugs and bystander effects. See Figure 19 .

[0239] 3.7 Regulatory effects of kinase inhibitors on anti-CD19 CAR T cells against CD19+ Raji

[0240] Dasatinib (Lck inhibitor) and ibrutinib (ITK inhibitor) at a concentration of 10 nM reduce the lysis effect of anti-CD19 CAR T cells (FMC63 CAR T, effector) on CD19+ Raji tumor cells (target). Two effector:target ratios (E:T) were tested. Normal T cells and anti-CD19 CAR T cells with CD19-K562 cells were used as negative controls. See Figure 20 .

[0241] 3.8 FITC-Dasatinib can reduce the lysis effect of FMC63+4MFR+ hT cells on Raji cells.

[0242] This means the successful internalization and release of FITC-dasatinib into T cells through the 4M5.3FR fusion receptor and the release of dasatinib into T cells. See Figure 21 .

[0243] The TC-PTP inhibitor at a concentration of 100 nM reduces the co-inhibitor molecule population in exhausted anti-CD19 CAR T cells

[0244] Exhausted anti-CD19 CAR T cells were generated by stimulating with CD19+ Raji cells 7 times (see the detailed procedure in the Materials and Methods section). The PD-1 positive, LAG3 positive, and double positive populations were reduced after treatment. See Figure 22 。

[0245] 4. Other FK506-payloads to control the activity of cell therapies

[0246] The technical advantageous feature of this drug payload delivery system is its versatility. The potential payloads and the corresponding effects are listed below (Table 1). Small molecule payloads were selected based on the following parameters: 1. The functional assays for free drug in vitro and in vivo have been confirmed by published literature or work in our laboratory. 2. The chemical structure of the drug has a free amine that is relatively more accessible for derivatization. 3. Any of the following is preferred: a drug approved by the FDA; commercially available at a reasonable price. The FK506-payloads were first tested for in vitro experiments, and T cell activation and stem cell cytokine release were monitored by multiplex immunoassays. For in vivo disease models, we have well-established the CAR T therapy and the fracture mouse model in our laboratory, and several potential collaborators have established the neurodegenerative mouse model.

[0247]

Claims

1. A drug delivery platform for cell therapy, comprising: a. An engineered protein on a target cell for transplantation, wherein the engineered protein consists of a first component and a second component, the first component and the second component are linked by a peptide linker, the first component is a non-membrane protein, and the second component is a membrane-anchored peptide or protein; b. A small ligand conjugated to the linker, wherein the small ligand has an intrinsic binding affinity for at least one component of the engineered protein; and c. A drug payload conjugated to the linker, wherein when the small ligand binds to at least one component of the engineered protein, the drug payload associates with the target cell; wherein the first component is the FK506-binding protein FKBP, wherein the second component is a peptide that confers a glycosylphosphatidylinositol GPI anchor on the first component and consists of folate receptor alpha FRa or folate receptor beta FRb, wherein the small ligand is FK506, an FK506 derivative, SLF as a synthetic ligand for FKBP, an SLF derivative, FA folic acid or an FA derivative: wherein the peptide linker is at least one segment of SGGGS, wherein the linker is one selected from alkyl, polyethylene glycol PEG, polyproline, oligo-(4-piperidinecarboxylic acid), oligopiperidine, peptide, glycopeptide, and combinations thereof, and wherein the drug payload is selected from rhodamine, fluorescein isothiocyanate, S0456, EC20 chelating head, NOTA 1,4,7-triazacyclononane-1,4,7-triacetic acid, DOTA 1,4,7,10-tetraazacyclododecane-1,4,7,10-tetraacetic acid, tubulysin, DM1, DM4, auristatin, dasatinib, MEK1 / 2 inhibitor, PI3K inhibitor, HDAC inhibitor, kinase inhibitor, metabolic inhibitor, GSK3β inhibitor, MAO-B inhibitor, Cdk5 inhibitor, RORγt agonist, SHP1 / 2 inhibitor, TC-PTP inhibitor, and siRNA mi181a1.

2. The drug delivery platform according to claim 1, wherein the target cell for transplantation is an immune cell.

3. The drug delivery platform according to claim 2, wherein the immune cell is a chimeric antigen receptor CAR T cell.

Images