Novel T-Cell Receptor and Ligand

Abstract

The present disclosure relates to a new T-cell receptor (TCR), in particular at least one complementarity-determining region (CDR) thereof; a T-cell expressing said TCR; a clone expressing said TCR; a vector encoding said TCR; a soluble version of said TCR; a pharmaceutical composition or bispecific comprising said TCR, said cell, said clone or said vector; use of said TCR or said cell or said clone or said vector or said pharmaceutical composition or bispecific to treat cancer; and a method of treating cancer using said TCR, said cell, said clone, said vector, said pharmaceutical composition or bispecific comprising said TCR.

Claims

1. A T cell receptor (TCR) or a binding fragment of a TCR which binds a tumour antigen comprising: a) a γ chain that comprises a CDR3 (CDR3γ) comprising an amino acid sequence that is at least 88% identical to the amino acid sequence of SEQ ID NO: 1; and/or b) a δ chain that comprises a CDR3 (CDR3δ) comprising an amino acid sequence that is at least 88% identical to the amino acid sequence of CALGVLPTVTGGGLIF (SEQ ID NO: 2).

2-5. (canceled)

6. The TCR or binding fragment of claim 1, wherein: a) the γ chain comprises CDR1γ, CDR2γ, and CDR3γ amino acid sequences that are at least 88% identical to the amino acid sequences of SEQ ID NOs: 3, 4, and 1, respectively; and/or b) the δ chain comprises CDR1δ, CDR2δ, and CDR3δ amino acid sequences that are at least 88% identical to the amino acid sequences of SEQ ID NOs: 5, 6, and 2, respectively.

7. (canceled)

8. (canceled)

9. The TCR or binding fragment of claim 1 wherein: a) the γ chain comprises an extracellular region comprising an amino acid sequence that is at least 88% identical to the amino acid sequence of SEQ ID NO: 7 or 17; and/or b) the δ chain comprises an extracellular region comprising an amino acid sequence that is at least 88% identical to the amino acid sequence of SEQ ID NO: 8, 14, or 18.

10-12. (canceled)

13. The TCR or binding fragment of claim 1, wherein: a) the amino acid sequence of the TCR or binding fragment is artificial; b) at least one amino acid is substituted, added or deleted relative to the wildtype sequence; c) at least one amino acid is substituted, added or deleted in a framework region, a CDR or a constant region relative to the wildtype sequence; d) at least one amino acid is substituted, added or deleted relative to the wildtype sequence, wherein the at least one amino acid is not located in any CDR; e) the TCR is of soluble form; f) the TCR or tumour-specific binding fragment is specific for at least one SCNNA1 gene product isoform; g) the TCR or tumour-specific binding fragment of a TCR binds to at least one of the SCNNA1 gene product isoforms encoded by the amino acid sequence of SEQ ID NOs: 29 to 34; and/or h) the TCR or tumour-specific binding fragment of a TCR binds to the extracellular domain of at least one SCNNA1 gene product isoform.

14-20. (canceled)

21. The TCR or binding fragment of claim 1, wherein: a) the γ chain comprises an amino acid sequence that is at least 88% identical to amino acids 19-307 of SEQ ID NO: 15; b) the δ chain comprises an amino acid sequence that is at least 88% identical to amino acids 21-290 of SEQ ID NO: 16.

22. The TCR or binding fragment of claim 1, wherein the γ chain and the δ chain each comprise a constant region in which the constant region is replaced by the corresponding sequence of the constant region of a murine TCR or a variant thereof.

23. A polynucleotide encoding the TCR or binding fragment of claim 1, optionally wherein the polynucleotide is comprised within a vector, optionally a viral vector.

24-27. (canceled)

28. A cell comprising the polynucleotide of claim 23, optionally wherein the cell is a T cell.

29-31. (canceled)

32. An ex vivo process comprising (i) obtaining T-cells from a patient, (ii) optionally expanding the T-cells (iii) transforming the T-cells with the vector of claim 25; and (iv) reintroducing said transduced T-cells into the patient.

33. A method of treating cancer in a subject in need thereof comprising administering to the subject the cell of claim 28.

34. A pharmaceutical composition comprising the cell of claim 28 and a pharmaceutically acceptable carrier.

35. (canceled)

36. A bispecific construct comprising the TCR or binding fragment of claim 1 and an immune cell activating component or ligand that binds to and activates an immune cell, optionally wherein the immune cell activating component or ligand that binds to and activates an immune cell binds to CD3.

37. (canceled)

38. A fusion protein comprising the TCR or binding fragment of claim 1 and a heterologous protein.

39-49. (canceled)

50. A polynucleotide encoding the bispecific construct of claim 36, optionally wherein the polynucleotide is comprised within a vector, optionally wherein the vector is a viral vector.

51. (canceled)

52. (canceled)

53. A cell comprising the polynucleotide of claim 50, optionally wherein the cell is a T cell.

54. (canceled)

55. (canceled)

56. A method of treating cancer in a subject in need thereof, comprising administering to the subject a therapeutically effective amount of the cell of claim 53.

57. (canceled)

58. A polynucleotide encoding the fusion protein of claim 38, optionally wherein the polynucleotide is comprised within a vector, optionally wherein the vector is a viral vector.

59. A cell comprising the polynucleotide of claim 58, optionally wherein the cell is a T cell.

60. A method of treating cancer in a subject in need thereof comprising administering to the subject a therapeutically effective amount of the cell of claim 59.

61. A method of producing a TCR, the method comprising: a) introducing into a host cell the polynucleotide of claim 23; and b) culturing the host cell so that the polynucleotide is expressed and the TCR is produced.

62. A method of producing a bispecific construct, the method comprising: a) introducing into a host cell the polynucleotide of claim 50; and b) culturing the host cell so that the polynucleotide is expressed and the bispecific construct is produced.

63. A method of producing a fusion protein, the method comprising: a) introducing into a host cell the polynucleotide of claim 58; and b) culturing the host cell so that the polynucleotide is expressed and the bispecific construct is produced.

Description

[0316] FIGS. 1A-B show how a γδ T-cell line reactive to autologous and non-autologous lymphoblastoid cell lines (LCLs) was clonotyped and found to express a TCR comprised of the TRGV3 and TRDV1 genes with the CDR3s CATWDRRDYKKLF (SEQ ID NO: 1) and CALGVLPTVTGGGLIF (SEQ ID NO: 2), respectively. A clone was grown by limited dilution that expressed this TCR and named SW.3G1. (FIG. 1A) Purified γδ T-cells from a healthy donor, 9909, were primed (day 0) and re-stimulated (day 14) with a pool of LCLs from three donors (0439, pt146 and Hom-2). On day 28 the T-cell line was incubated with the LCLs used for priming and also autologous LCL-9909 for 4 h with activation assessed by inclusion of TAPI-0, anti-CD107a and anti-TNFα antibodies. The activated cells were sorted by flow cytometry and the T-cell receptors (TCRs) analysed by next generation sequencing. Pie charts depict the proportion of the displayed TCR chains and CDR3s (complementarity determining regions) that were present in the sorted cells. The percentage of activated cells for the flow cytometry plots is shown above each gate. T-cell clone SW.3G1 obtained from the lines expresses the highlighted TCR chains, with the TCR extracellular domain less the connecting peptide sequences shown in FIG. 2 (SEQ ID NOs: 7 and 8; δ chain mutant version). (FIG. 1B) Clone SW.3G1 was phenotyped with the antibodies and confirmed to express a TCRδ1 chain. SW.3G1 did not express an αβTCR or CD8 or CD4 glycoproteins associated with recognition of conventional peptide-HLA antigens.

[0317] FIG. 2 shows the T-cell receptor sequence of the γ and δ TCR chains of clone SW.3G1 (extracellular regions less the connecting peptide only, mutant version of δ chain, Example 2a, SEQ ID NOs: 7 and 8). The mRNA structures (top) show that for each chain CDR1 and CDR2 are encoded in the germline. CDR3 is the product of junctional diversity at V-J joins of T cell receptor (TCR)-γ chain and V-D-J joins in TCR-δ chain. CDR3 is consequently hypervariable. The order adopted for the CDR loops is maintained throughout the figure. The panel on the right shows the expected protein fold. TCRs adopt similar tertiary structures that position the complementarity-determining regions (CDR) loops at the membrane distal end of the molecules. Together the six CDR loops form the antigen binding site.

[0318] FIGS. 3A-D show SW.3G1 can recognize and kill autologous and non-autologous LCLs, but not healthy cells of various tissue origins. (FIG. 3A) Co-incubation of SW.3G1 with LCLs for 4 h, with activation assessed by inclusion of TAPI-0, anti-CD107a and anti-TNFα antibodies. (FIG. 3B) 6.5 h chromium release cytotoxicity assay using the same LCLs as in (FIG. 3A). (FIG. 3C) Autologous healthy B-cells magnetically purified directly ex-vivo from donor 9909 were used in activations assays as in FIG. 3A, with LCL-9909 used as a positive control for activation. (FIG. 3D) Activation assays as in FIG. 3A, using LCL-9909 and the healthy cell lines, CIL-1 (non-pigmented ciliary epithelium) and Hep2 (hepatocyte). Percentage of gated cells is shown.

[0319] FIGS. 4A-B show SW.3G1 mediated lysis of LCLs from multiple donors that share no common HLA and an array of cancer cell lines from different tissues. SW.3G1 was used in 6.5 h chromium release cytotoxicity assays. (FIG. 4A) A panel of lymphoblastic cell lines (LCLs) from 24 donors (named on the x-axis). The first three bars (narrow hatch) depict LCLs that were used to generate the T-cell lines from donor 9909 from which SW.3G1 was cloned. T-cell to LCL ratio of 1:1. (FIG. 4B) SW.3G1-mediated killing of panel of cancer cell lines (named on the x-axis) of different tissue origin (key) at a T-cell to cancer cell ratio of 10:1. Note that the key for (FIG. 4B) does not correspond to (FIG. 4A).

[0320] FIGS. 5A-B show that SW.3G1 does not recognize target cells by known mechanisms. (FIG. 5A) γδ SW.3G1 clone was co-incubated for 4 h with HMB-PP, the lymphoblastic cell line (LCL)-9909 and phytohaemagluttinin (PHA). T-cells were also incubated alone. T-cell activation was assessed by inclusion of TAPI-0, anti-CD107a and anti-TNFα antibodies, with the percentage of activated cells shown above the gated cells. (FIG. 5B) Using the same activation assay as in (FIG. 5A), SW.3G1 was incubated with LCLs that had been pre-labelled with antibodies (Abs) that bind the proteins named on the x-axis. SW.3G1 was also incubated with the LCLs without Ab (no Ab control). The percentage of reactivity is shown graphically (y-axis). MICA/B (Major Histocompatibility Complex (MHC) Class-I related chain A/B) and EPCR (Endothelial protein C receptor). Anti-MHC class I and II Abs were also included.

[0321] FIG. 6 shows the whole genome CRISPR/Cas9 approach used to identify candidate genes/proteins involved in target cell recognition by SW.3G1.

[0322] FIGS. 7A-C show the results of the whole genome CRISPR/Cas9 approach that identified multiple candidate genes for target cell recognition by SW.3G1. (FIG. 7A) Autologous LCL-9909 and cancer cell line KBM7s were transduced with a whole genome CRISPR/Cas9 library. The libraries were put though several selections using the SW.3G1 T-cell clone to generate a target cell line that was resistant to lysis. The surviving target cells (post-selection) were tested alongside the pre-selected cell lines in activation assays with SW.3G1. Activation assessed by inclusion of TAPI-0, anti-CD107a and anti-TNFα antibodies. (FIG. 7B) Sequencing of the post-selection libraries revealed enriched guide RNAs corresponding to the key shown in (FIG. 7C), identifying genes of interest. (FIG. 7C) Candidate genes seen in both the LCL-9909 and KBM7 libraries, or seen only for LCL-9909 or KBM7s. Gene and (protein) names are shown with website links for further information. The hatched key refers to the hatching of the pie charts in (FIG. 7B).

[0323] FIG. 8 shows information about the candidate gene/protein SCNN1A, which was identified by the whole genome CRISPR/cas9 library approach.

[0324] FIG. 9 shows the canonical protein sequence of SCNN1A which aligned with five expressed splice variants. Isoform 1 is the canonical sequence (Isoform 1 UniProt P37088-1). Boxed amino acids in BLACK in solid lines: region used to generate the polyclonal Ab used in this study. Boxed amino acids in dashed line: Sites of protein variants due to different amino acid residues to the ones shown. The amino acid residues of the protein variants and are not displayed here, but can be found at http://www.uniprot.org/uniprot/P37088. Boxed amino acids in alternating dash and dot lines: amino acid differences between splice isoforms.

[0325] FIGS. 10A-D show the results of experiments to validate the role of SCNN1A in target cell recognition by SW.3G1. (FIG. 10A) Schematic of the SCNN1A gene and protein, with guide RNA (gRNA) sites from the whole genome GeCKO library (gRNA-1, SEQ ID NO: 12) and a different validation gRNA SCNN1A sequence we designed (gRNA-2, SEQ ID NO: 13). Figure adapted from Chen 2014. (FIG. 10B) Long term killing assay using SW.3G1 with LCL 0.174 wild-type, gRNA-1 (GeCKO gRNA) and gRNA-2 (our gRNA) knock-out LCL.174 cells. (FIG. 10C) Western blot analysis of the breast cancer cell line MDA-MB-231 that had received the gRNA-2 (our gRNA) for SCNN1A. Wild-type cells used for comparison, with red arrow indicating the 76 kDa SCNN1A protein. (FIG. 10D) MDA-MB-231 cells from (FIG. 10C) and melanoma MM909.24, wild-type and SCNN1A knock-out (KO) cell lines used in long term killing assays with SW.3G1. Cancer cells were used as lines and selected by puromycin treatment for those expressing the gRNAs, with no subsequent cell cloning.

[0326] FIGS. 11A-B show that transfer of the TCR from SW.3G1 confers target cell reactivity to αβ T-cells from three healthy donors. (FIG. 11A) Purified CD8+ T-cells from three donors were co-transduced with the SW.3G1 T-cell receptor chains (marker and purification via rat(r)CD2) and a gRNA to render the recipient T-cells TCRβ chain negative (selected by puromyicn treatment) (Legut et al 2017). Purity of the cells was checked with rCD2 antibody (Ab) and anti-γδ TCR Ab. (FIG. 11B) The cells from one donor were tested in long-term killing assays (lower graph). LCL 0.174 cell lines were used: wildtype, SCNN1A knockout (using gRNA-1 and -2) and SCNN1A knock-in (KO cells that had received a codon optimized SCNN1A transgene) cells.

EXAMPLES

Example 1

Methods and Materials

(a) T-Cell Line Generation and Clonotyping

[0327] Peripheral blood mononuclear cells (PBMCs) were purified from the blood of a healthy donor (code 9909) by standard density gradient separation. The dominant population of γδ T-cells in peripheral blood express a Vγ9Vδ2 TCR and typically respond to antigens derived from bacteria. In order to enrich γδTCR+/δ2− T-cells thereby increasing the likelihood of finding cancer reactive T-cells, we modified a magnetic based purification protocol. The first adaptation was to stain the PBMCs with a PE conjugated anti-Vδ2 antibody (Ab) (clone B6, BioLegend, San Diego, Calif.). Next, γδ TCR+ T-cells were negatively enriched by positively removing γδ TCR− cells according the manufacturer's instructions (Miltheyi Biotec, Bergish Gladbach, Germany). The second adaptation involved adding anti-PE microbeads (Miltneyi Biotec) to the beads of the γδ TCR purification kit, thereby removing δ2+ cells at the same time as the γδ TCR− cells. The purified cells were co-incubated with irradiated (3000-3100 rad) LCLs from three donors that had been generated from PBMC by immortalizing B-cells with Epstein-Barr Virus (EBV). All LCLs were grown in R10 media (RPMI-1640, 10% heat-inactivated fetal calf serum, 2 mM L-glutamine, 100 U/mL Penicillin and 100 μg/mL Streptomycin, all Life Technologies, Carlsbad, Calif.) as suspension cells. After 14 days the T-cells were restimulated with irradiated LCLs from the same donors. On day 28 the T-cells were harvested and used in activation assays to assess reactivity towards LCLs. T-cells (30,000) were incubated for 4 h in 96 U well plates with an equivalent number of LCLs. 30 mM of the TNFα Processing Inhibitor-0 (TAPI-0 from Sigma Aldrich) (Haney et al., 2011), anti-CD107a Ab (H4A2, Becton Dickinson (BD), Franklin Lakes, N.J.) and anti-TNFα Ab (cA2, Miltenyi Biotec) were added to the assay media at the start of the assay, with the cells subsequently stained with the cell viability dye, Vivid (Life Technologies, 1:40 dilution in PBS then 2 μL per stain in 50 μL) and anti-CD3 antibody (Ab) (BW264/56, Miltenyi Biotec). Activated cells were sorted on a BD FACS Aria in to RLT Plus buffer (supplemented with 40 mM DTT) (Qiagen) ready for sequencing of the TCR chains. RNA was extracted using the RNEasy Micro kit (Qiagen, Hilden, Germany). cDNA was synthesized using the 5′/3′ SMARTer kit (Clontech, Paris, France) according to the manufacturer's instructions. The SMARTer approach used a Murine Moloney Leukaemia Virus (MMLV) reverse transcriptase, a 3′ oligo-dT primer and a 5′ oligonucleotide to generate cDNA templates, which were flanked by a known, universal anchor sequence. PCRs were performed using anchor-specific forward primers and reverse primers of the constant regions of the γ or δ TCR chains. The final PCR products were gel purified and prepared for next generation sequencing (Donia et al., 2017).

(b) Clone SW.3G1 Procurement and Phenotyping

[0328] T-cells were cloned directly from the T-cell line by limiting dilution (Theaker et al., 2016). After 4 weeks of culture, 50% of each clone by culture volume was harvested and used for the activation assays with LCLs as above. Prior to performing activation assays, T-cell clones were washed and incubated for 24 h in reduced serum media. Clones that exhibited reactivity towards the LCLs were grown to sufficient numbers for TCR sequencing (below). Clone SW.3G1 was stained with Abs for surface expression of CD3 (Miltenyi Biotec), CD8 (BW135/80, Miltenyi Biotec), CD4 (M-T466, Miltenyi Biotec), αβ TCR (BW242/412, Miltenyi Biotec) and TCR Vδ1 chain (REA173, Miltenyi Biotec).

(c) Sequencing of the SW.3G1 TCR

[0329] As above for sequencing the T-cell lines with the purified PCR products after the final PCR being cloned into Zero-Blunt TOPO and transformed into One Shot Chemically Competent E. coli cells for standard sequencing (both from Life Technologies). The sequences of the γ and δ chains of the natural SW.3G1 clone are SEQ ID NOs: 21 and 22, respectively.

(d) SW.3G1 Recognized LCLs but not Healthy Cells

[0330] To confirm SW.3G1 reactivity towards LCLs, activation assays as above, and chromium release cytotoxicity assays were performed. Healthy B-cells were purified from donor 9909 using a PE conjugated anti-CD19 Ab (HIB19, Miltenyi Biotec) and positive capture with anti-PE microbeads (Miltenyi Biotec) and used immediately in assays. Other healthy cell lines and their proprietary culture media were obtained from Sciencell (Carlsbad, Calif.): CIL-1 (human non-pigmented ciliary epithelium) and Hep2 (human hepatocyte) were used in activation as above.

(e) SW.3G1 Killed all Immortalized and Cancer Cell Lines Tested

[0331] LCLs and tumour cells were labelled with chromium 51 for cytotoxicity assays (Ekeruche-Makinde et al., 2012), with T-cell to target cell ratios of 1:1 (LCLs) or 10:1 (cancer cells). LCLs were maintained as above. Cancer cells lines (ATCC® reference for background and culture information)/tissue of origin: SiHa (HTB-35) and MS751 (HTB-34)/cervical; MCF7 (HTB-22), MDA-MB-231 (CRM-HTB-26) and SKBR3 (HTB-30)/breast; TK143 (CRL-8303) and U20S (HTB-96)/bone; HCT-116 (CCL-247) and Colo205 (CCL-222)/colon; Jurkat (TIB-152), K562 (CCL-243), THP-1 (TIB-202), U266 (TIB-196) and Molt-3 (CRL-1552)/blood; Caki-1 (HTB-46)/kidney; A549 (CCL-185) and H69 (HTB-119)/lung. MM909.11, MM909.12, MM909.15, MM909.46 and MM909.24 are skin melanomas obtained from cancer patients treated at the Center for Cancer Immune Therapy (CCIT, Herlev Hospital, Copenhagen, Denmark). The ‘MM’ cell lines and melanomas Mel 526 and Mel 624 were maintained as adherent cells in R10, passaged once weekly or when required, aiming for 20-80% confluence. Cells were detached from tissue culture flasks by rinsing with D-PBS followed by incubation with D-PBS and 2 mM EDTA at 37° C. until detached.

(f) SW.3G1 Did not Recognize Target Cells by Known Mechanisms

[0332] The Vγ9Vδ2 T-cell activator (E)-4-Hydroxy-3-methyl-but-2-enyl pyrophosphate (HMB-PP) (Sigma Aldrich) was reconstituted in DMSO and added directly to assay wells. The following monoclonal Abs were used for blocking assays: anti-HLA, -B, -C(clone W6/32, Biolegend), anti-HLA-DR, -DP, -DQ (clone Tu39, Bioloegend), anti-EPCR (polyclonal, R&D systems), anti-MICA/MICB (clone 6D4, BioLegend) and anti-CD1d (clone 51.1, Miltenyi Biotech) were used at a final concentration of 10 μg/mL.

(g) Gene Trapping by Whole Genome CRISPR

[0333] A whole genome CRISPR/Cas9 library approach was used (FIG. 5 for an overview). Whole genome targeted LCL-9909 and KBM7s using the GeCKO v2 sub-libraries A and B (Adgene plasmid, #1000000048, kindly provided by Dr. Feng Zhang (Patel et al., 2017)) were used for selection by SW.3G1. Briefly, successfully transduced target cells selected with puromycin were co-incubated with SW.3G1 at a predefined ratio for 2-3 weeks in 96 U well plates. Activation assays (as above) were performed with pre- and post-selected target cells to confirm loss of SW.3G1 activity towards the selected cells. Genomic DNA from the target cells that had survived two rounds of selection with SW.3G1 was used for next generation sequencing to reveal inserted guide RNAs and candidate genes.

(h) Confirming SCNN1A Role in Target Cell Recognition

[0334] Lentiviral particles were generated by calcium chloride transfection of HEK 293T cells and concentrated by ultracentrifugation prior to transduction of target cells using 8 μg/mL of polybrene and spinfection. gRNAs were cloned into the pLentiCRISPR v2 plasmid (kindly provided by Dr. Feng Zhang, Addgene plasmid 52961), which encodes the SpCas9 protein and a puromycin resistance marker gene (pac, puromycin N-acetyltransferase), and co-transfected with packaging and envelope plasm ids pMD2.G and psPAX2 (all from Addgene). Full-length codon optimized SCNN1A transgene (Isoform 1, UniProt P37088-1) was cloned in to a 3.sup.rd generation lentiviral transfer vector pELNS (kindly provided by Dr. James Riley, University of Pennsylvania, PA). The pELNS vector contains rat CD2 (rCD2) gene for selection of cells using an anti-rCD2 PE Ab (OX-34, BioLegend). SCNN1A expression in target cells was assessed using the rabbit anti-SCNN1A polyclonal antibody (PA1-902A, ThermoFisher Scientific) for flow cytometry (data not sown) and western blot analysis according to the manufacturer's instructions.

(i) Transduction of Polyclonal T-Cells with the SW.3G1 TCR Confers Target Cell Recognition

[0335] Codon optimized, full length TCR chains, separated by a self-cleaving 2A sequence, were synthesized (Genewiz) (SEQ ID NO: 23; corresponding protein sequence SEQ ID NO: 24) and cloned into the 3.sup.rd generation lentiviral transfer vector pELNS (kindly provided by Dr. James Riley, University of Pennsylvania, PA). The pELNS vector contains a rat CD2 (rCD2) marker gene separated from the TCR by another self-cleaving 2A sequence. Additionally, cells were co-transduced with a gRNA to ablate TCRβ chain expression in recipient cells by targeting both TCR-β constant domains (manuscript currently at Blood for publication). Lentiviral particles were generated by calcium chloride transfection of HEK293T cells. TCR transfer vectors were co-transfected with packaging and envelope plasmids pMD2.G, pRSV-Rev and pMDLg/pRRE. Lentiviral particles were concentrated by ultracentrifugation prior to transduction of CD8+ T-cells using 5 μg/ml of polybrene, with the CD8+ T-cells purified by magnetic separation (Miltenyi Biotec) from three healthy donors 24 h in advance and activated overnight with CD3/CD28 beads (Dynabeads, Life Technologies) at 3:1 bead:T-cell ratio. T-cells that had taken up the virus were selected by incubation with 2 μg/ml puromycin (TCRβ chain knock-out) and enriched with anti-rCD2 PE Ab (OX-34, BioLegend) followed by anti-PE magnetic beads (Miltenyi Biotec). 14 d post transduction T-cells were expanded with allogeneic feeders and PHA. TCR transduced cells were used in longterm killing assays whereby LCL.174 targets were plated in duplicate at the density of 50,000 cells/well in 96 U well plates. SW.3G1 was added to the target and incubated for 7 days. Target cells were also plated without T-cells, to serve as a 100% survival control. Cells were harvested, washed with PBS, and stained with Vivid and anti-CD3 antibody (to exclude T-cells). As an internal control, CountBright™ Absolute Counting Beads (Life Technologies) were added to each well prior to harvesting/washing (approximately 10,000 beads/well). The samples were the acquired on FACS Canto II, and at least 1,000 bead events were acquired per sample. The survival of target cells was calculated according to the following formula:

[00001]$\%\mathrm{survival}=\frac{\begin{array}{c}\mathrm{number}\mathrm{of}\mathrm{experimental}\mathrm{cell}\mathrm{events}/\\ \mathrm{number}\mathrm{of}\mathrm{experimental}\mathrm{bead}\mathrm{events}\end{array}}{\mathrm{number}\mathrm{of}\mathrm{control}\mathrm{cell}\mathrm{events}/\mathrm{number}\mathrm{of}\mathrm{control}\mathrm{bead}\mathrm{events}}\times 100\%$

(j) Results

Clone Characterisation

[0336] 1. Purified γδ T-cells from a healthy donor (9909) primed and re-stimulated with a pool of three non-autologous lymphoblastoid cell lines (-0439, -pt146 and -HOM-2). Reactivity towards each of the cells lines was tested at day 28 (FIG. 1). The T-cell line also recognized autologous LCL-9909 (FIG. 1).

[0337] 2. T-cells from the aforementioned line were flow cytometry sorted based on reactivity to each of the LCLs and their TCRs analysed by next generation sequencing (FIG. 1). For the γ-chain sequencing, two unique CDR3s were present with variable chains TRGV9 and TRGV3. For the δ-chains, three CDR3s were present with variable chains TRDV1 and TRDV2.

[0338] 3. T-cells clones procured from the donor 9909 T-cell line expressed a γ3δ1 TCR and CDR3s CATWDRRDYKKLF and CALGVLPTVTGGGLIF for each respective chain (FIG. 1 and FIG. 2). All the clones that grew expressed the same TCR. This clone was named SW.3G1.

[0339] Ab staining of SW.3G1 confirmed expression of the Vδ1 chain, and αβ TCR−/CD8 low/CD4− (FIG. 1B).

[0340] 4. Activation assays using TNFα and CD107a as the readouts confirmed SW.3G1 reactivity towards autologous LCL-9909 and non-autologous LCL-0439 (FIG. 3A). Donors 9909 and 0439 are completely HLA mismatched for both MHC class I and class II alleles, therefore SW.3G1 is recognizing target cells in an HLA-independent manner. SW.3G1 was also able to lyse LCL-9909 and -0439 and is therefore cytotoxic (FIG. 3B). The recognition of the LCLs was dependent on the immortalization process when EBV infects a B-cell, as autologous healthy B-cells purified directly ex-vivo from 9909 did not act as targets for SW.3G1 (FIG. 3C). Similarly, the healthy cells CIL-1 (epithelial cell) and Hep2 (hepatocyte) did not elicit SW.3G1 activation (FIG. 3D).

[0341] 5. SW.3G1 was able to lyse LCLs from all 24 donors tested (FIG. 3B for the autologous LCL and FIG. 4A for 23 non-autologous LCLs) providing further confirmation that SW.3G1 is acting in a HLA independent manner. Furthermore, LCL 0.174 (FIG. 4A, 4th bar from the left) only expresses one copy of chromosome 6, the human chromosome that carries the MHC locus. The chromosome 6 in cell LCL 0.174 contains a large deletion and does not carry genes for MHC class II and many components involved in MHC class I antigen processing.

[0342] 6. SW.3G1 killed 23 cancer cell lines that originate from 8 different tissues: skin/melanoma, kidney, colon, breast, blood/leukemia, lung, cervix and bone. (FIG. 4B).

[0343] 7. SW.3G1 did not respond to the known γδ T-cell antigen, HMB-PP (FIG. 5A), which leads to recognition of pathogen infected cells by the Vγ9Vδ2 TCR T-cell subset in a similar manner to recognition of the self pyrophosphate. This pathway requires target cells to express Butyrophillin 3A1. SW.3G1 reactivity towards LCL-9909 (autologous) -0439 and -pt146 was not hindered by inclusion of blocking antibodies that bind known γδ T-cell ligands: Major Histocompatibility Complex (MHC) Class-I related chain A and B (MICA/MICB), EPCR Endothelial Protein C Receptor (EPCR) and CD1d (FIG. 5B). MHC class I and class II also failed to block SW.3G1 activation. Although not an extensive exclusion process, these data suggested that SW.3G1 might recognize an unknown γδ TCR ligand at the surface of cancer cells. Therefore, a whole genome CRISPR/Cas9 library approach was adopted to find candidate genes/proteins involved in SW.3G1 recognition of target cells (FIG. 6).

[0344] 8. Whole genome CRISPR/Cas9 libraries were used to create gene knockouts in autologous LCL-9909 and the haploid myeloid leukaemia cell line KBM7. Both libraries were co-incubated with SW.3G1 for successive rounds of selection to enrich for target cells containing gRNAs that allowed escape from SW.3G1-mediated lysis (FIG. 7A). SW.3G1 reactivity dropped from 59% for pre-selected LCL-9909 to 12% post-selection. For KMB7s reactivity went from 12% to 4.2%. The post-selected LCL-9909s and KBM7s were used for next generation sequencing to identify gRNAs that had been enriched. Key genes were identified with 4 of the total 7 genes shared between the LCL-9909 and KBM7 libraries (FIGS. 7B and 7C). Guides specific for the gene SCNN1A (also used here to describe the encoded protein), which encodes for the protein Sodium Channel Epithelial 1 Alpha Subunit, were highly enriched present in both libraries (FIGS. 7B&C). SCNN1A gene and protein aliases are shown in FIG. 8. The protein is cell surface expressed and therefore a good candidate for further exploration. SCNN1A has 6 splice variant isoforms and various naturally occurring mutations (FIGS. 8 and 9).

[0345] 9. LCL.174 transduced with SCNN1A gRNA from the whole genome library (GeCKO, gRNA-1, SEQ ID NO: 12) or a different guide designed in-house (gRNA-2, SEQ ID NO: 13) (FIG. 10A) were no longer targets of SW.3G1, thereby confirming SCNN1A's role in target cell recognition (FIG. 10B), with lysis falling to below 5% for the knockout cell lines compared to 100% killing for the wildtype cells. SCNN1A gene knockout lines were created in two cancer cells, which either partially or completely escaped lysis by SW.3G1 (FIG. 10C). It is noteworthy that the SCNN1A knockout cells created throughout this study were used as lines, and not cloned before performing assays. This may account for the residual reactivity seen for some of the ‘knockout’ cell lines as a minority proportion of the cells within a knockout line probably still express SCNN1A, due to escape from puromycin selection and/or unsuccessful ablation of the SCNN1A gene. Western blot analysis of the SCNN1A knockout MDA-MB-231 cells used for SW.3G1 activation assays revealed a substantial reduction of the SCNN1A protein in the knockout cell line compared to the wildtype cells (FIG. 10C) confirming the gene knockout. The Ab used can recognise all SCNN1A isoforms (FIG. 9). We also noted that the SCNN1A knockout cells became less viable with extensive culture (3+ weeks) and in some cases cell division halted completely. This observation was unique to the SCNN1A gRNA as the same cell lines transduced with gRNAs for many other genes did not exhibit the same change in cell growth and vitality. This result suggested that the SCNN1A gene is essential for long-term growth of cells in culture.

[0346] 10. Transfer of the SW.3G1 TCR in to polyclonal CD8+ T-cells from three healthy donors conferred reactivity to target cell LCL-pt146 (FIG. 11A). TCR transduced cells exhibited the same functional profile to SCNN1A knockout cells as described above for the SW.3G1 clone (FIG. 11B). To compliment the SCNN1A knockout data, and to further confirm the role of SCNN1A in target cell recognition, we transduced the knockout cells with the SCNN1A gene. The introduction of a native SCNN1A gene to knockout cells expressing the SCNN1A gRNAs would lead to gene ablation of the transgene. Therefore, a codon-optimized gene was introduced, different to the DNA sequence of the native gene (Isoform 1 UniProt P37088-1, FIG. 9), but expressing the same protein. Killing of the gRNA-1 or gRNA-2 transduced cells was ablated but could be restored by expressing the SCNN1A gene in the knockout cells (FIG. 11B).

(k) Conclusion

[0347] The SW.3G1 TCR enables T-cells to recognise a wide range of tumours. Recognition occurs via the SCNN1A gene product. SW.3G1 T-cell clone recognises a cancer-cell specific SCNN1A ligand in the absence of MHC restriction.

[0348] This invention centres around the TCR identified in T-cell clone SW.3G1. This TCR recognises a wide range of cancer cells through the expression of SCNN1A. This TCR does not recognise non-tumour cells. CRISPR/Cas9 knockout of SCNN1A from tumour lines or antibody blocking confirmed there TCR requires the SCNN1A gene product for recognition of tumour cells. The SW.3G1 TCR can be used in a variety of different cancer immunotherapy strategies. The broad tumour recognition and human leukocyte antigen (HLA)-independence of recognition unlocks exciting possibilities for pan-cancer, pan-population immunotherapies using this TCR.

Example 2

[0349] Design of a mutant TCR

Example 2a

[0350] A mutant version of the SW.3G1 TCR was designed and prepared in which a R to N mutation was introduced into the constant region within the extracellular region of the δ chain. This mutation was made in order to improve stability of the domain. See SEQ ID NOs: 7 and 8.

Example 3

Preparation of an Artificial T-Cell Expressing the TCR of the Invention

[0351] Expression of the T cell receptor of the invention (“SW.3G1 T cell receptor”) in patients' T cells is a potential therapeutic approach for cancer immunotherapy. DNA encoding the γ and δ chains of the T cell receptor, separated with a cleavable read-through linker, such as those of the 2A family can be expressed in a lentiviral vector (see Example 1(i)) This vector when transduced into patients' CD8 T cells will drive expression of the SW.3G1 T cell receptor γ and δ chains at a 1:1 stoichiometry and thus allow a mature functional SW.3G1 TCR to be expressed on the cell surface.

[0352] To produce CD8 T cells expressing a SW.3G1 recombinant receptor the following procedures are followed. Firstly, the DNA sequences of the TCR (such as the mutant of Example 2) are synthesised from assembled oligonucleotides and cloned into a lentiviral vector. Features of the lentiviral system to be utilised include a lentiviral vector incorporating self-inactivating long terminal repeats (LTR). The viral vector encodes the entire coding sequence, including the signal sequences, of the TCR γ and δ chains separated by a read-through self-cleaving linker of the 2A family. The transcription of the mRNA encoding the TCR chains is driven by a constitutive promoter such as the cytomegalovirus (CMV) promoter or elongation factor 1a (EF1a) promoter.

[0353] The SW.3G1 encoding lentiviral plasmid vector is packaged into a virus using three additional packaging plasmids. The packaging plasmids encode 1) gag-pol 2) rev and 3) the envelope protein of vesicular stomatitis virus (VSV-G). Separation of the virus genome into 4 plasmids increases the safety of the virus and minimises the chances of recombination events leading to replication competent lentivirus (RCL) formation. Using cationic lipid transfection reagents, the vector and three packaging plasmids are simultaneously transfected into the packaging cell line HEK293T, in which replication-incompetent, SW.3G1 TCR-encoding lentivirus is assembled and secreted into the packaging cell culture supernatant. HEK293T cells may be grown attached to plastic or adapted for growth in suspension. Virus particles are concentrated from the cell supernatant by high speed centrifugation, and in some cases may be filtered through 0.45 μm pore filters to remove cellular debris and cryopreserved prior to transduction of CD8 T cells.

[0354] To purify patient peripheral blood CD8 T cells, apheresis is first performed on patients to isolate large numbers of peripheral blood mononuclear cells (PMBC) typically in the range of 10-20 billion mononuclear cells per patient. CD8 T cells can isolated from these PBMCs using magnetic bead isolation. Microscopic paramagnetic beads coated with antibodies against CD8 are incubated with the PBMC, followed by separation on columns within a strong magnetic field. CD8-negative cells are eluted from the column by washing with serum albumin-containing phosphate buffered saline (SA/PBS). Following washing, the CD8 bead-labelled cells are released by removal from the magnet and washing with SA/PBS. The CD8 T cells are then enumerated and incubated in cell culture medium in advance of activation and transduction with lentivirus encoding the TCR of the invention.

[0355] Transduction of CD8 T cells with TCR-encoding lentiviral particles is achieved by incubating CD8 T cells in cell culture media into which lentiviral particles are added. CD8 cells may be in the resting state or pre-activated with antibodies against CD3 and CD28 with addition of cytokines including interleukin-2 and interleukin-7. Following 24-48 hours after transduction the cells are expanded for infusion back into patients. Expansion is performed with antibodies against CD3 and CD28, either in soluble or bead-bound form with addition of interleukin-2 for 14 days. In addition to interleukin-2, other cytokines such as interleukin-7 may also be added to the cultures. Expansion may also be performed using autologous feeder PBMCs which are irradiated with 30 Gray gamma irradiation to inhibit proliferation plus anti-CD3 antibody plus interleukin-2 at 3000-6000 IU/ml for 14 days.

Example 4

Applications of a Soluble Form of the TCR of the Invention

[0356] Solubilisation of TCRs opens up the ability to utilise TCRs in a number of applications, to include but not limited to, precise affinity measurements of the TCR for its ligand, crystallography-based structure solving, multimerization, and the generation of clinical therapies utilising the TCR as a targeting mechanism.

(a) Generation of Soluble, Monomeric SW.3G1 TCR for Further Application

[0357] Expression of a soluble SW.3G1 TCR (“SW.3G1 sTCR”), or a soluble mutant form as described in Example 2, could be used to characterise binding kinetics to its cognate SCNNA1 gene ligand. Briefly, SW.3G1 sTCR can be tested by surface plasmon resonance (SPR), biolayer interferometry (BLI), or similar, to determine biophysical kinetics of the interaction between the soluble TCR and a derivative of its cognate SCNNA1 gene ligand. For example, in the case of SPR, the its cognate SCNNA1 gene ligand is immobilised on specialised biosensor chips and concentrated SW.3G1 sTCR is passed over the chip and the interaction measured by SPR. From this, the affinity and binding half-life can be determined by calculation of the on- (k.sub.a or k.sub.on) and off-rates (k.sub.d or k.sub.off).

(b) Generation of Multimeric SW.3G1 sTCRs for Further Application

[0358] The DNA sequence for SW.3G1 sTCR can also be modified for example to contain an acceptor tag peptide for biotin at the C-terminus of one of the chains. Following expression of such a construct, this acceptor tag can specifically be labelled with biotin via an enzymatic reaction using a biotin ligase such as BirA. Such a method will create a uniform biotinylation of the SW.3G1 sTCR which will allow it to be multimerised via binding to streptavidin, for generation of tetramers, pentamers or dextramers. These multimers can be further modified via the addition of a molecule such as phycoerythrin (PE), a fluorophore that allows detection of the multimer via fluorescence. The SW.3G1 sTCR multimeric complex could then be used to identify SCNNA1 gene ligand expressing cells via incubation of the molecule with target cells, and detection and/or sorting via methods such as fluorescent activated cell sorting (FACS).

[0359] (c) Generation of modified SW.3G1 sTCRs for potential therapeutic application

[0360] The SW.3G1 sTCR binding to its cognate ligand in vivo is unlikely to confer any therapeutic benefit. In order to generate a therapeutic agent with clinical effect, further modifications of the sTCR are required. The SW.3G1 TCR recognises its target on a number of different cancer cells and cancer cell-derived cell lines. Therefore, to generate a reagent with therapeutic utility in the oncology setting, a SW.3G1 sTCR would need to have a function introduced that would induce cell death of the target cell, either directly or indirectly.

[0361] To modify SW.3G1 sTCR for this purpose, an effector function could be introduced, such as an anti-CD3 agonistic antibody, to engage and redirect T cell activity. An anti-CD3 antibody optionally in scFv format could be added to the N or C terminus of either the γ or δ chain of the SW.3G1 sTCR, via a short, non-immunogenic linker sequence (such as GGGGS, SEQ ID NO: 25). Such a fusion protein would have bispecific properties.

REFERENCES

[0362] Chen, J., T R Kleyman and S. Sheng. 2014 Deletion of α-subunit exon 11 of the epithelial Na+ channel reveals a regulatory module. Am J Physiol. Renal Physiol 306, F561-7. [0363] Donia, M., Kjeldsen, J. W., Andersen, R., Westergaard, M. C. W., Bianchi, V., Legut, M., Attaf, M., Szomolay, B., Ott, S., Dolton, G., Lyngaa, R., Hadrup, S. R., Sewell, A. K., Svane, I. M., 2017. PD-1 polyfunctional T cells dominate the periphery after tumor-infiltrating lymphocyte therapy for cancer. Clin. Cancer Res. clincanres.1692.2016. [0364] Ekeruche-Makinde, J., Clement, M., Cole, D. K., Edwards, E. S. J., Ladell, K., Miles, J. J., Matthews, K. K., Fuller, A., Lloyd, K. A., Madura, F., Dolton, G. M., Pentier, J., Lissina, A., Gostick, E., Baxter, T. K., Baker, B. M., Rizkallah, P. J., Price, D. A., Wooldridge, L., Sewell, A. K., 2012. T-cell receptor-optimized peptide skewing of the T-cell repertoire can enhance antigen targeting. J. Biol. Chem. 287, 37269-81. [0365] Haney, D., Quigley, M. F., Asher, T. E., Ambrozak, D. R., Gostick, E., Price, D. A., Douek, D. C., Betts, M. R., 2011. Isolation of viable antigen-specific CD8+ T cells based on membrane-bound tumor necrosis factor (TNF)-alpha expression. J. Immunol. Methods 369, 33-41. [0366] Legut, M. G Dolton, A. A. Mian, O. Ottmann and A. K. Sewell. 2017 CRISPR-mediated TCR replacement generates superior anticancer transgenic T-cells. Blood [Epub ahead of print] doi: https://doi.org/10.1182/blood-2017-05-787598 [0367] Patel, S. J., Sanjana, N. E., Kishton, R. J., Eidizadeh, A., Vodnala, S. K., Cam, M., Gartner, J. J., Jia, L., Steinberg, S. M., Yamamoto, T. N., Merchant, A. S., Mehta, G. U., Chichura, A., Shalem, O., Tran, E., Eil, R., Sukumar, M., Guijarro, E. P., Day, C.-P., Robbins, P., Feldman, S., Merlino, G., Zhang, F., Restifo, N. P., 2017. Identification of essential genes for cancer immunotherapy. Nature 548, 537-542. [0368] Theaker, S. M., Rius, C., Greenshields-Watson, A., Lloyd, A., Trimby, A., Fuller, A., Miles, J. J., Cole, D. K., Peakman, M., Sewell, A. K., Dolton, G., 2016. T-cell libraries allow simple parallel generation of multiple peptide-specific human T-cell clones. J. Immunol. Methods 430, 43-50.