Antigen binding polypeptides binding to a HLA-e:peptide complex

Abstract

The invention relates to antibodies which bind to a HLA-E:peptide complex, wherein the peptide is a WT1 derived peptide. The invention also relates to uses of such antibodies in the treatment or prevention of infection or cancer.

Description

[0001] ANTIGEN BINDING POLYPEPTIDES

[0002] FIELD OF INVENTION

[0003] The present invention relates to antibodies and CAR-T cells, and their use in treating infection and cancer.

[0004] BACKGROUND HLA-E is a non-polymorphic HLA class I molecule. There are two major alleles in the population differing only in one amino acid at position 107 which is outside the peptide binding groove (Strong et al., Correlating differential expression, peptide affinities, crystal structures, and thermal stabilities. J Biol Chem. 2003;278(7):5082-90). The primary function of HLA-E is to bind a peptide usually termed ‘VL9’ which is derived from the signal peptide of classical HLA class I A, B, C molecules and HLA-G, but not HLA-E.

[0005] The HLA-VL9 complex in turn binds to the NKG2A-CD94 inhibitory or NKG2C-CD94 -activating receptors on natural killer cells and a subset of T cells (Braud et al., HLA-E binds to natural killer cell receptors CD94 / NKG2A, B and C. Nature.

[0006] 1998;391 (6669): 795 -9).

[0007] Other peptides derived from ‘self’ proteins, which may be abnormally expressed or mutated in cancer cells, or peptides derived from viruses or bacteria can also bind to HLA-E, but the large majority do so with lower binding affinity so cannot compete effectively with the VL9 peptide (Walters et al., Detailed and atypical HLA-E peptide binding motifs revealed by a novel peptide exchange binding assay. Eur J Immunol.

[0008] 2020). However, in certain circumstances, presentation of the VL9 peptide is disturbed, for instance in cytomegalovirus (CMV) infection, in mycobacterial infection or in cancer cells, leading to HLA-E bound to other peptides being presented on the cell surface. These cells can then be recognised by CD8+ T cells, and a response initiated through their classical Major Histocompatibility complex class I (MHC-I)-restricted T cell receptor (TCR). As HLA-E is non-polymorphic and HLA-E restricted responses to pathogens have thus far been poorly characterised, further identification of immune responses to peptide antigen bound to HLA-E may be useful for universal immunotherapies in the population due to the lack of HLA-E genetic polymorphism. Therefore, the generation of antibodies or T cells which recognise HLA-E bound to peptide antigens derived from a cancer or pathogen, or even autoantigens has considerable therapeutic potential.

[0009] SUMMARY OF INVENTION

[0010] In an aspect, there is provided an antigen binding polypeptide which is capable of binding to a HLA-E:peptide complex. The antigen binding polypeptide may specifically bind to a HLA-E:peptide complex. The antigen binding polypeptide may preferentially bind to a HLA-E:peptide complex. The antigen binding polypeptide may induce cell death of cells displaying a HLA-E:peptide complex.

[0011] In another aspect, there is provided an antigen binding polypeptide which competes for binding to a HLA-E:peptide complex with an antigen binding molecule of the invention.

[0012] In another aspect, there is provided an antigen binding polypeptide which binds to a HLA-E:peptide complex, in which the peptide has an arginine at position one (when described N-terminal to C-terminal).

[0013] In another aspect, there is provided an antigen binding polypeptide which binds to a HLA-E:peptide complex, wherein the antigen binding polypeptide interacts with one or more of, such as all of, the following residues in HLA-E: Asp69, Gln72, Ile73, Arg75, Val76, Thr80, Tyr84, Serl47, Glul52. The antigen binding polypeptide may also interact with one or more of, such as all of the following residues in HLA-E: Glul9, Asn38, Asp39, Ala40, Arg68, Gln72 and Arg75.

[0014] The antigen binding polypeptide may be an antibody or antigen binding fragment thereof. The antigen binding polypeptide may be a chimeric antigen receptor (CAR) or antigen binding fragment thereof. The antigen binding polypeptide may be a synthetic T cell receptor and antigen receptor (STAR). The antigen binding polypeptide may be a T cell receptor fusion construct (TRuC). The antigen binding polypeptide may comprise:

[0015] a) a heavy chain variable region comprising a complementarity determining region CDR3 of SEQ ID NO: 3 or a sequence having at least 80%, 90%, 95%, 98%, 99% or 100% identity thereto; and / or

[0016] a light chain variable region comprising a CDR3 of SEQ ID NO: 6 or a sequence having at least 80%, 90%, 95%, 98%, 99% or 100% identity thereto; or

[0017] b) a heavy chain variable region comprising a complementarity determining region CDR3 of SEQ ID NO: 29 or a sequence having at least 80%, 90%, 95%, 98%, 99% or 100% identity thereto; and / or

[0018] a light chain variable region comprising a CDR3 of SEQ ID NO: 32 or a sequence having at least 80%, 90%, 95%, 98%, 99% or 100% identity thereto; or

[0019] c) a heavy chain variable region comprising a complementarity determining region CDR3 of SEQ ID NO: 37 or a sequence having at least 80%, 90%, 95%, 98%, 99% or 100% identity thereto; and / or

[0020] a light chain variable region comprising a CDR3 of SEQ ID NO: 40 or a sequence having at least 80%, 90%, 95%, 98%, 99% or 100% identity thereto.

[0021] The antigen binding polypeptide may comprise or consist of:

[0022] a) a heavy chain variable region comprising:

[0023] a CDR1 of SEQ ID NO: 1,

[0024] a CDR2 of SEQ ID NO: 2, and

[0025] a CDR3 of SEQ ID NO: 3,

[0026] or sequences having at least 80%, 90%, 95%, 98%, 99% or 100% identity thereto; and / or

[0027] a light chain variable region comprising:

[0028] a CDR1 of SEQ ID NO: 4,

[0029] a CDR2 of SEQ ID NO: 5, and

[0030] a CDR3 of SEQ ID NO: 6,

[0031] or sequences having at least 80%, 90%, 95%, 98%, 99% or 100% identity thereto; or b) a heavy chain variable region comprising:

[0032]

[0033] or sequences having at least 80%, 90%, 95%, 98%, 99% or 100% identity thereto; and / or

[0034] a light chain variable region comprising:

[0035] a CDR1 of SEQ ID NO: 30,

[0036] a CDR2 of SEQ ID NO: 31, and

[0037] a CDR3 of SEQ ID NO: 32,

[0038] or sequences having at least 80%, 90%, 95%, 98%, 99% or 100% identity thereto; or c) a heavy chain variable region comprising:

[0039] a CDR1 of SEQ ID NO: 35,

[0040] a CDR2 of SEQ ID NO: 36, and

[0041] a CDR3 of SEQ ID NO: 37,

[0042] or sequences having at least 80%, 90%, 95%, 98%, 99% or 100% identity thereto; and / or

[0043] a light chain variable region comprising:

[0044] a CDR1 of SEQ ID NO: 38,

[0045] a CDR2 of SEQ ID NO: 39, and

[0046] a CDR3 of SEQ ID NO: 40,

[0047] or sequences having at least 80%, 90%, 95%, 98%, 99% or 100% identity thereto.

[0048] The CDRs may be associated with any framework region. Preferably, the framework region is of human origin.

[0049] The antigen binding polypeptide may comprise or consist of:

[0050] a) a heavy chain variable region comprising or consisting of SEQ ID NO: 7 or a sequence having at least 80%, 90%, 95%, 98%, 99% or 100% identity thereto; and / or

[0051] a light chain variable region comprising or consisting of SEQ ID NO: 8 or a sequence having at least 80%, 90%, 95%, 98%, 99% or 100% identity thereto; or

[0052] b) a heavy chain variable region comprising or consisting of SEQ ID NO: 33 or a sequence having at least 80%, 90%, 95%, 98%, 99% or 100% identity thereto; and / or a light chain variable region comprising or consisting of SEQ ID NO: 34 or a sequence having at least 80%, 90%, 95%, 98%, 99% or 100% identity thereto; or

[0053] c) a heavy chain variable region comprising or consisting of SEQ ID NO: 41 or a sequence having at least 80%, 90%, 95%, 98%, 99% or 100% identity thereto; and / or

[0054] a light chain variable region comprising or consisting of SEQ ID NO: 42 or a sequence having at least 80%, 90%, 95%, 98%, 99% or 100% identity thereto.

[0055] The antigen binding polypeptide may consist of:

[0056] a) a heavy chain variable region comprising or consisting of SEQ ID NO: 7; and a light chain variable region comprising or consisting of SEQ ID NO: 8; or b) a heavy chain variable region comprising or consisting of SEQ ID NO: 33; and a light chain variable region comprising or consisting of SEQ ID NO: 34; or c) a heavy chain variable region comprising or consisting of SEQ ID NO: 41; and a light chain variable region comprising or consisting of SEQ ID NO: 42.

[0057] The antigen binding polypeptide may consist of:

[0058] a) a heavy chain comprising or consisting of SEQ ID NO: 43; and

[0059] a light chain comprising or consisting of SEQ ID NO: 44; or

[0060] b) a heavy chain comprising or consisting of SEQ ID NO: 45; and

[0061] a light chain comprising or consisting of SEQ ID NO: 46; or

[0062] c) a heavy chain comprising or consisting of SEQ ID NO: 47; and

[0063] a light chain comprising or consisting of SEQ ID NO: 48.

[0064] The peptide of the HLA-E:peptide complex may be, or be derived from, any of SEQ ID NOs: 13-16, or a sequence with at least 80%, 90%, 95%, 98%, or 99% identity thereto. The peptide of the HLA-E:peptide complex may be between seven and 11 amino acids in length. Preferably the peptide is 9 amino acids in length. Most preferably, the peptide comprises an arginine at position one (when described N-terminal to C-terminal).

[0065] The HLA-E may be HLA-E*0101 or HLA-E*0103, or both.

[0066] An antigen binding polypeptide of the invention may be isolated. The invention is based upon previous findings that HLA class I molecules are down-regulated or absent in many tumour types, whereas the non-classical HLA-E is significantly up-regulated and can be used as a biomarker of tumour invasiveness. HLA-E overexpression has been shown to negatively interfere with innate immune surveillance, thereby protecting tumour cells from NK and CD8 T cell- mediated cytotoxicity. The inventors developed a universal antibody from a phage antibody library that targets the HLA-E*0101 and the very closely related E*0103 molecule bound to peptides derived from both pathogens or tumour. This antibody can surprisingly recognise tumour or pathogen specific peptides bound to HLA-E and therefore has high therapeutic potential for cancer and other infectious disease treatment, for example as a bi-specific antibody or engineered as a chimeric antigen receptor (CAR) which can be expressed on T cells, or other immuno-responsive cells such as NK cells.

[0067] In another aspect, there is provided a nucleic acid encoding an antigen binding polypeptide of the invention. Such nucleic acids may be provided by any of SEQ ID NOs: 10-12. The skilled person will understand that due to codon redundancy, a number of DNA sequences may be used to encode an antigen binding polypeptide of the invention. Alternatively, codon optimization of the nucleotide sequence can be used to improve the efficiency of translation in expression systems for the production of an antigen binding polypeptide of the invention.

[0068] In another aspect, the invention provides a vector comprising a nucleic acid of the invention. Suitable vectors of the invention can be chosen or constructed, containing appropriate regulatory sequences, including promoter sequences, terminator sequences, polyadenylation sequences, enhancer sequences, marker genes and other sequences as appropriate. Vectors may be for example plasmids or viral. For further details see, for example, (Sambrook, J., E. F. Fritsch, and T. Maniatis. (1989), Molecular cloning: a laboratory manual, 2nd ed. Cold Spring Harbor Laboratory, Cold Spring Harbor, New York). Many known techniques and protocols for manipulation of nucleic acid, for example in preparation of nucleic acid constructs, mutagenesis, sequencing, introduction of DNA into cells and gene expression, and analysis of proteins, are described in detail in (Ausubel et al., Current protocols in molecular biology. New York: Greene Publishing Association; Wiley-Interscience, 1992). The vector may be an expression vector. The vector or expression vector may be a plasmid. A nucleic acid molecule or vector of the invention may be expressed using any suitable expression system, for example in a suitable host cell or in a cell-free system.

[0069] In another aspect, the invention provides a host cell comprising an antigen binding polypeptide of the invention, nucleic acid, and / or vector of the invention. The host cell may be selected from bacterial host cells (prokaryotic systems) such as E. Coli, or eukaryotic cells such as those of yeasts, fungi, insect cells or mammalian cells. Preferably a host cell of the invention is capable of producing the antigen binding polypeptide of the invention. The produced antigen binding polypeptide may be enriched by means of selection and / or isolation. In the case of a CAR, the host cell may be an autologous T-cell or NK cell derived from a subject to be treated.

[0070] Thus, in another aspect, the invention provides a CAR-T cell expressing an antigen binding polypeptide of the invention, or comprising a nucleic acid and / or vector of the invention. The CAR-T cell may be an autologous T-cell derived from a subject to be treated. The T-cell may be a CD8+ T-cell. The CAR-T cell may be a CD4+ T cell. The invention may provide a population of CAR-T cells expressing an antigen binding polypeptide of the invention, or comprising a nucleic acid and / or vector of the invention. The population may comprise or consist of CD4+ and / or CD8+ T-cells expressing an antigen binding polypeptide of the invention, or comprising a nucleic acid and / or vector of the invention. Upon binding to the HLA-E:peptide complex, the CAR-T cell can exhibit effector functions and / or cytolytic effects towards cells bearing the HLA-E:peptide complex and / or undergo proliferation and / or cell division. Activated CAR-T cells can secrete anti-tumor or pathogen cytokines which can include, but are not limited to, TNFalpha, IFNy and IL2.

[0071] In any aspect, the CAR-T or CAR-NK cell may be-dual targeting. For example, the CAR-T or CAR-NK cell may in addition to targeting a HLA-E:peptide complex referred to herein, also target another distinct protein or peptide, such as a tumour antigen.

[0072] In another aspect, the invention provides a CAR-NK cell expressing an antigen binding polypeptide of the invention, or comprising a nucleic acid and / or vector of the invention. The CAR-NK cell may be an autologous NK-cell derived from a subject to be treated. A CAR-T cell of the invention may further comprise one or more exogenous or recombinant co -stimulatory ligand, such as one, two, three or four co -stimulatory ligands. The CAR-T cells may co-express the CAR and the one or more exogenous or recombinant co -stimulatory ligand. The interaction between CAR and one or more exogenous or recombinant co-stimulatory ligand may provide a non-antigen-specific signal and activation of the cell. Co-stimulatory ligands include, but are not limited to, members of the tumour necrosis factor (TNF) superfamily, and immunoglobulin (Ig) superfamily ligands.

[0073] A CAR-T cell of the invention may further express or present one or more heterologous co-receptor. The one or more heterologous co-receptor may comprise or consist of a CD8 co-receptor. The CD8 co-receptor may comprise a dimer or pair of CD8 chains which comprises a CD8-a and CD8-P chain or a CD8-a and CD8-a chain. Preferably, the CD8 co-receptor is a CD8aa co-receptor comprising a CD8-a and CD8-a chain. The CD8 co-receptor binds to class 1 MHCs and potentiates TCR signalling. The one or more co-receptor may comprise or consist of a CD2 co-receptor.

[0074] An antigen binding polypeptide of the invention may also be produced by chemical synthesis. The obtained antigen binding polypeptide may be enriched by means of selection and / or isolation.

[0075] According to a further aspect, the invention provides a pharmaceutical composition comprising one or more antigen binding polypeptide, nucleic acid, vector, host cell and / or CAR-T cell, or CAR-NK cell of the invention, optionally together with one or more pharmaceutically acceptable excipients or diluents. For example, the composition may comprise a CD4+ CAR T-cell and a CD8+ CAR-T cell of the invention.

[0076] In another aspect, an antigen binding polypeptide, nucleic acid, vector, host cell, CAR-T cell, CAR-NK cell or pharmaceutical composition of the invention may be for use in the treatment or prevention of one or more disease or disorder in a subject.

[0077] In another aspect, there is provided a method of treating or preventing one or more disease or disorder in a subject, comprising administering to the subject an effective amount of an antigen binding polypeptide, nucleic acid, vector, host cell, CAR-T cell, CAR-NK cell or composition of the invention. The subject may or may not have received a previous treatment for the disease or disorder. If the subject has received previous treatment for the disease or disorder, the subject may not have responded, or responded poorly to the previous treatment. Response to the previous treatment may be calculated in a number of ways known to the skilled person, such as but not limited to tumour size, or viral or bacterial load.

[0078] In an aspect, there is provided the use of an antigen binding polypeptide, nucleic acid, vector, host cell or pharmaceutical composition of the invention in the manufacture of a medicament for the treatment or prevention of one or more diseases or disorders in a subject.

[0079] Antigen binding polypeptides, nucleic acids, vectors, host cells = CAR-T cells, or CAR-NK cells of the invention can be formulated into pharmaceutical compositions using established methods of preparation (Gennaro, A. L. and Gennaro, A. R. (2000) Remington: The Science and Practice of Pharmacy, 20th Ed., Lippincott Williams & Wilkins, Philadelphia, PA). To prepare the pharmaceutical compositions, pharmaceutically inert inorganic or organic excipients can be used. To prepare for example pills, powders, gelatin capsules or suppositories, lactose, talc, stearic acid and its salts, fats, waxes, solid or liquid polyols, natural and hardened oils are examples of pharmaceutically acceptable excipients which can be used. Suitable excipients for the production of solutions, suspensions, emulsions, aerosol mixtures or powders for reconstitution into solutions or aerosol mixtures prior to use include water, alcohols, glycerol, polyols, and suitable mixtures thereof as well as vegetable oils.

[0080] A pharmaceutical composition of the invention may be administered via any parenteral or non-parenteral (enteral) route that is therapeutically effective. Parenteral application methods include, for example, intracutaneous, subcutaneous, intramuscular, intratracheal, intranasal, intravitreal or intravenous injection and infusion techniques, e.g. in the form of injection solutions, infusion solutions or mixtures, as well as aerosol installation and inhalation, e.g. in the form of aerosol mixtures, sprays or powders. A pharmaceutical composition of the invention can be administered systemically or topically in formulations containing conventional non-toxic pharmaceutically acceptable excipients or carriers, additives and vehicles as desired. A combination of intravenous and subcutaneous infusion and / or injection might be most convenient in case of compounds with a relatively short or long serum half-life or needing rapid onset of action. Preferably, the pharmaceutical composition is administered subcutaneously or intravenously. The pharmaceutical composition may be an aqueous solution, an oilin water emulsion or a water-in-oil emulsion.

[0081] For intravenous injection, or injection at the site of affliction, or other site of administration, the active ingredient will be in the form of a parenterally acceptable aqueous solution which is pyrogen-free and has suitable pH, isotonicity and stability. Those of relevant skill in the art are well able to prepare suitable solutions using for example, isotonic vehicles such as Sodium Chloride Injection, Ringer’s Injection, Lactated Ringer’s Injection. Preservatives, stabilisers, buffers, antioxidants and / or other additives may be included, as required.

[0082] The compositions of the invention are preferably administered to an individual in a “therapeutically effective amount”, this being sufficient to show benefit to the individual. The optimal dosage will depend on the biodistribution of the antigen binding polypeptide, the mode of administration, the severity of the disease / disorder being treated as well as the medical condition of the patient. If desired, the antigen binding polypeptide may be given in a sustained release formulation, for example liposomal dispersions or hydrogel-based polymer microspheres, like PolyActiveTM or OctoDEXTM (cf. Bos et al., Business Briefing: Pharmatech 2003: 1-6). Other sustained release formulations available are for example PLGA based polymers (PR pharmaceuticals), PLA-PEG based hydrogels (Medincell) and PEA based polymers (Medivas). Prescription of treatment, e.g., decisions on dosage etc, is within the responsibility of a medical practitioner, and typically takes account of the disorder to be treated, the condition of the individual patient, the site of delivery, the method of administration and other factors known to practitioners.

[0083] The pharmaceutical composition may also contain additives, such as, for example, fillers, binders, wetting agents, glidants, stabilizers, preservatives, emulsifiers, and furthermore solvents or solubilizers or agents for achieving a depot effect. The latter is that fusion proteins may be incorporated into slow or sustained release or targeted delivery systems, such as liposomes and microcapsules.

[0084] In any aspect, the subject may be a mammal. Preferably, the subject is a human. The one or more disease or disorder to be treated may be cancer or an infection. The cancer or infection may be associated with increased WT1 expression. Many solid tumours are associated with increased WT1 expression (Xaio-wei Qi et al., 2015, Sci. reports, 5:8924). The cancer may be acute myeloid leukaemia, breast cancer, endometrial carcinoma, ovarian cancer, hepatocellular cancer, Non-small cell lung cancer, colorectal cancer, colon cancer soft tissue sarcoma, glioblastoma, astrocytoma, melanoma, or mesothelioma. The infection may be from HIV1 or HIV2. The infection may be from Mycobacterium tuberculosis (Mtb). The infection may be from Epstein Barr virus (EBV). The infection may be from Human Papillomavirus (HPV). The infection may be from SARS-CoV-2.

[0085] The infection may be caused my any microbe which has one or more peptide which is bound by HLA-E. Preferably the one or more peptide will contain an arginine at position one (when described N-terminal to C-terminal).

[0086] A cancer or infection said to be associated with increased WT1 expression may refer to a cancer or infection which induces the expression and / or presentation of a WT1 peptide in the subject to be treated. The WT1 peptide may be that of SEQ ID NO: 13.

[0087] An antigen binding polypeptide, nucleic acid, vector, host cell, CAR-T cell, CAR-NK cell or pharmaceutical composition of the invention may be administered alone or in combination with one or more other therapeutic agent, either simultaneously, sequentially or separately, dependent upon the condition to be treated. The one or more other therapeutic agent may be selected from the group comprising cytotoxic agents, immune activation agents such as checkpoint inhibitors or TLR agonists, antiinflammatory agents such as steroids, CAR-T cells such as regulatory or cytolytic CAR-T cells. The therapeutic may be conjugated to an antigen binding polypeptide as described herein.

[0088] In another aspect, there is provided a method of monitoring treatment efficacy or disease status in a subject diagnosed with cancer or an infection, comprising:

[0089] i. providing a biological sample obtained from the subject;

[0090] ii. determining the tumour volume, viral or bacterial load, or level of binding of one or more antigen binding polypeptides of the invention to HLA-E:peptide complex bearing cells in the sample obtained from the subject before treatment, or at intervals between treatments, or at time intervals in the absence of treatment;

[0091] iii. determining that the treatment is effective, or that the disease status is improving, if the tumour volume, viral or bacterial load, or level of binding of one or more antigen binding polypeptides of the invention to HLA-E:peptide bearing cells, is reduced after treatment or between treatment intervals or at time intervals in the absence of treatment.

[0092] Preferably, in cancer, the one or more cancer cell expresses WT1 and / or a peptide which has arginine at the N-terminus, binds HLA-E and cross reacts with an antigen binding polypeptide of the invention. Preferably, in an infection, the pathogen may encode a peptide that has arginine at position one (when described N-terminal to C-terminal), binds HLA-E and cross reacts with an antigen binding polypeptide of the invention.

[0093] A biological sample may be a blood or serum sample, tissue biopsy, cerebrospinal fluid, saliva, or urine sample. Preferably, the biological sample may be a blood or serum sample.

[0094] The level of binding of one or more antibodies or antigen binding fragments of the invention to HLA-E-WTl-expressing cells in the sample may be determined using any method known to the skilled person. One such method is for example using flow cytometry or any other technique utilising a detectable label, to be able to determine the number of HLA-E-WT1 expressing cells in the sample.

[0095] Tumour volume may be determined by any suitable technique known to the skilled person. Viral or bacterial load may be determined by any suitable technique known to the skilled person.

[0096] The reduction in tumour volume, viral or bacterial load, or level of binding of one or more antibodies or antigen binding fragments of the invention HLA-E-WTl-expressing cells may be by 10% or more, such as 25% or more, 50% or more, 75% or more, or 90% or more. The treatment intervals or time intervals in the absence of treatment may be two weeks or more, such as four weeks or more, 8 weeks or more, 12 weeks or more, six months or more, or 12 months or more.

[0097] In another aspect, there is provided a use of a polypeptide of SEQ ID NO: 13 in a method of identifying an antigen binding polypeptide which specifically binds to an HLA-E:peptide complex as described herein.

[0098] The method may include a screening step to identify one or more antigen binding polypeptides which are thought to specifically bind to a HLA-E:peptide complex as described herein. The screening step may include a step to remove non-specific antigen binding polypeptides.

[0099] The method may include a validation step in which the antigen binding polypeptide is demonstrated to bind to a HLA-E:peptide complex as described herein.

[0100] The term “antibody” as referred to herein refers to a glycoprotein comprising at least two heavy (H) chains and two light (L) chains inter-connected by disulfide bonds. Each heavy chain is comprised of a heavy chain variable region (VH) and a heavy chain constant region. Each light chain is comprised of a light chain variable region (VL) and a light chain constant region. The variable regions of the heavy and light chains contain a binding domain that interacts with an antigen. The VH and VL regions can be further subdivided into regions of hypervariability, termed complementarity determining regions (CDR), interspersed with regions that are more conserved, termed framework regions (FR). The constant regions of the antibodies may mediate the binding of the immunoglobulin to host tissues or factors, including various cells of the immune system (e.g effector cells) and the first component (Clq) of the classical complement system.

[0101] The term "antigen-binding fragment thereof refers to one or more fragments of an antibody or CAR that retain the ability to selectively bind to an antigen.

[0102] An antibody or antigen binding fragment thereof may be a monoclonal antibody, bispecific antibody, multi-specific antibody, scFv or other single chain or modified format, Fab, (Fab’)2, Fv, dAb, Fd, nanobody, camelid antibody, diabody, domain antibodies or bispecific antibodies, and epitope-binding fragments of any of the above. Techniques for the production of antibodies and antigen binding fragments thereof are well known in the art.

[0103] A bispecific antibody or antigen binding fragment thereof of the invention, may in addition to binding a HLA-E:peptide complex as described herein, also bind CD3 and / or CD28, thereby recruiting and activating bystander T-cells to exert their effector functions on cells with an HLA-E:peptide complex as described herein on their surface, such as cancer cells.

[0104] A trispecific antibody or antigen binding fragment thereof of the invention, may in addition to binding a EILA-E: peptide complex as described herein, also bind HLA-E not in complex with a peptide as described herein, and also CD3 and / or CD28, thereby recruiting and activating bystander T-cells to exert their effector functions on cells with an HLA-E:peptide complex as described herein on their surface, such as cancer cells.

[0105] The term "antibody" also includes immunoglobulins (Ig's) of different classes (i.e. IgA, IgG, IgM, IgD and IgE) and subclasses (such as IgGl, lgG2 etc.).

[0106] The antigen binding polypeptide may be modified to change in vivo stability and / or half-life. The modification for example may be PEGylation. The antigen binding polypeptide may also be modified to enhance or reduce its binding to other receptors such as Fc receptors.

[0107] The antibody or antigen binding fragment thereof may be an antibody-like molecule which includes the use of CDRs separately or in combination in synthetic molecules such as SMIPs and small antibody mimetics.

[0108] The percent identity of two amino acid sequences or of two nucleic acid sequences is generally determined by aligning the sequences for optimal comparison purposes (e.g., gaps can be introduced in the first sequence for best alignment with the second sequence) and comparing the amino acid residues or nucleotides at corresponding positions. The "best alignment" is an alignment of two sequences that results in the highest percent identity. The percent identity is determined by comparing the number of identical amino acid residues or nucleotides within the sequences (i.e., % identity = number of identical positions / total number of positions x 100). The determination of percent identity between two sequences can be accomplished using a mathematical algorithm known to those of skill in the art. An example of a mathematical algorithm for comparing two sequences is the algorithm of Karlin and Altschul, 1990, PNAS, 87(6):2264-8, modified as in Karlin and Altschul, 1993, PNAS, 90( 12): 5873 -5877 The NBLAST and XBLAST programs of Altschul et al., 1990, J. Mol. Biol., 215:403-10 have incorporated such an algorithm. BLAST nucleotide searches can be performed with the NBLAST program, score = 100, word length = 12 to obtain nucleotide sequences homologous to a nucleic acid molecules of the invention. BLAST protein searches can be performed with the XBLAST program, score = 50, word length = 3 to obtain amino acid sequences homologous to a protein molecules of the invention. To obtain gapped alignments for comparison purposes, Gapped BLAST can be utilized as described in Altschul et al. (1997). Alternatively, PSI-Blast can be used to perform an iterated search that detects distant relationships between molecules (Id.). When utilizing BLAST, GappedBLAST, and PSI-Blast programs, the default parameters of the respective programs (e.g., XBLAST and NBLAST) can be used. See http: / / www.ncbi.nlm.nih.gov. Another example of a mathematical algorithm utilized for the comparison of sequences is the algorithm of Myers and Miller. The ALIGN program (version 2.0) which is part of the GCG sequence alignment software package has incorporated such an algorithm. Other algorithms for sequence analysis known in the art include ADVANCE and ADAM as described in Torellis and Robotti (1994); and FASTA described in Pearson and Lipman (1988). Within FASTA, ktup is a control option that sets the sensitivity and speed of the search.

[0109] An antigen binding polypeptide of the invention may comprise one or more mutated amino acid residues. The terms "mutated", "mutant" and "mutation" in reference to a nucleic acid or an antigen binding polypeptide of the invention refers to the substitution, deletion, or insertion of one or more nucleotides or amino acids, respectively, compared to the "naturally" occurring nucleic acid or polypeptide, i.e. to a reference sequence that can be taken to define the wild-type.

[0110] The amino acid variations in the CDR sequences may be conservative amino acid substitutions.

[0111] A mutation may be a substitution wherein the substitution is a conservative substitution. Conservative substitutions are generally the following substitutions, listed according to the amino acid to be mutated, each followed by one or more replacement(s) that can be taken to be conservative: Ala — > Gly, Ser, Vai; Arg — > Lys; Asn — > Gin, His; Asp Glu; Cys —> Ser; Gin —> Asn; Glu —> Asp; Gly —> Ala; His —> Arg, Asn, Gin; lie —> Leu, Vai; Leu —> lie, Vai; Lys —> Arg, Gin, Glu; Met —> Leu, Tyr, He; Phe —> Met, Leu, Tyr; Ser —> Thr; Thr —> Ser; Trp —> Tyr; Tyr —> Trp, Phe; Vai —> He, Leu. Other substitutions are also permissible and can be determined empirically or in accord with other known conservative or non-conservative substitutions.

[0112] 1, 2 or 3 conservative substitutions may be made in the CDRs of the antigen binding polypeptide of the invention.

[0113] Methods of making an antigen binding polypeptide are well known in the art. The skilled person may use hybridoma technology for example, or may use recombinant DNA technology to clone the respective antibody sequence into a vector, such as an expression vector. Methods of making a bispecific antibody molecule are known in the art, e.g. recombinant DNA technology, chemical conjugation of two different monoclonal antibodies or for example, also chemical conjugation of two antibody fragments, for example, of two Fab fragments. Alternatively, bispecific antibody molecules are made by quadroma technology, which is by fusion of the hybridomas producing the parental antibodies. Because of the random assortment of H and L chains, a potential mixture of ten different antibody structures are produced of which only one has the desired binding specificity. A bispecific antibody molecule of the invention can act as a monoclonal antibody (mAb) with respect to each target. The antibody or antigen binding fragment thereof may be chimeric, humanized or fully human. The antibody or antigen binding fragment thereof may be a human IgGl isotype or other natural or modified isotype. A bispecific antibody molecule or multi-specific antibody may for example be a bispecific tandem single chain Fv, a bispecific Fab2, or a bispecific diabody.

[0114] An scFv may be joined to human heavy and light chain constant regions to give IgG antibodies. An antibody of the invention may also be joined to TCR hinge, transmembrane and signalling domains, such as 4-1BB, zeta and CD28 signalling domains.

[0115] All of the features disclosed in this specification may be combined in any combination, including with any aspect or any embodiment. Table 1 - Sequences

[0116] SEQ ID Description Sequence

[0117] 1 5B11CDR- GETESSYA

[0118] H1

[0119] 2 5B11CDR- ISGSGGST

[0120] H2

[0121] 3 5B11 CDR- AKDWGLWFGELPRQGYFDY

[0122] H3

[0123] 4 5B11CDR- NIGSKS

[0124] L1

[0125] 5 5B11 CDR- YDF

[0126] L2

[0127] 6 5B11 CDR- QVWDSSSDHVI

[0128] L3

[0129] 7 5B11 VH EVQLVESGGGLVQPGGSLRLSCAASGFTFSSYAMSWVR QAPGKGLEWVSAISGSGGSTYYADSVKGRFTISRDNSK NTLYLQMNSLRAEDTAVYYCAKDWGLWFGELPRQGYF DYWGQGTLVTVSS

[0130] 8 5B11 VL TSYVLTQPPSVSVAPGKTARIPCGGNNIGSKSVHWYQQ KPGQAPVLVIYYDFDRPSGIPERFSGSNSGNTATLTISRV EAGDEADYYCQVWDSSSDHVIFGGGTKLTVL

[0131] 9 Linker GGGGSGGGGSGGGGS

[0132] 10 5B11 VH GAGGTGCAGCTGGTGGAGTCTGGGGGAGGCTTGGTA DNA CAGCCTGGGGGGTCCCTGAGACTCTCCTGTGCAGCCT CTGGATTCACCTTTAGTAGCTATGCCATGAGCTGGGT CCGCCAGGCTCCAGGGAAGGGGCTGGAGTGGGTCTC AGCTATTAGTGGTAGTGGTGGTAGCACATACTACGCA GACTCCGTGAAGGGCCGGTTCACCATCTCCAGAGACA ATTCCAAGAACACGCTGTATCTGCAAATGAACAGCCT GAGAGCCGAGGACACGGCCGTATATTACTGTGCGAA AGATTGGGGACTATGGTTCGGGGAGTTACCACGACAA GGATACTTTGACTACTGGGGCCAGGGAACCCTGGTCA CCGTCTCGAGC

[0133] 11 5B11 VL ACGTCCTATGTGCTGACTCAGCCACCCTCAGTGTCAG DNA TGGCCCCAGGAAAGACGGCCAGGATTCCCTGTGGGG

[0134]

[0135] GAAACAACATCGGAAGTAAGAGTGTGCACTGGTACC AGCAGAAGCCAGGCCAGGCCCCTGTGTTGGTCATCTA TTATGATTTCGACCGGCCCTCAGGGATCCCTGAGCGA TTCTCTGGCTCCAACTCTGGGAACACGGCCACCCTGA CCATCAGCAGGGTCGAGGCCGGGGATGAGGCCGACT ATTACTGTCAGGTGTGGGATAGTAGTAGTGATCATGT CATATTCGGCGGAGGGACCAAGCTCACCGTCCTA

[0136] Linker GGTGGAGGCGGTTCAGGCGGAGGTGGCAGCGGCGGT DNA GGCGGGT WT1 RMFPNAPYL

[0137] peptide

[0138] Mtb 44 RLPAKAPLL

[0139] peptide

[0140] Mtb 14 RMAATAQVL,

[0141] peptide

[0142] HIV-1 Gag RMYSPTSIL

[0143] peptide

[0144] VL9 peptide VMAPRTLVL

[0145] VL9 peptide VMAPRTLLL

[0146] HCMV NLVPMVATV

[0147] pp65

[0148] peptide

[0149] Phagemid CAG GAA ACA GCT ATG AC

[0150] forward

[0151] primer

[0152] VH forward GCA ACC GGT GTA CAT TCT GAG GTG CAG CTG GTG primer GAG

[0153] VH reverse TGG GCC CTT GGT CGA CGC GCT CGA GAC GGT GAC primer CAG

[0154] VL forward GCA ACC GGT GTA CAT TCT ACG TCC TAT GTG CTG primer ACT 3

[0155] VL reverse TGG TGC AGC CAC CGT ACG TAG GAC GGT GAG CTT primer GGT

[0156]

[0157] CAR GGT TCC ACA GGT GAC GAG GTG CAG CTG GTG forward

[0158] primer

[0159] CAR CAC GAA TGC GGC CGC TAG GAC GGT

[0160] reverse GAG CTT

[0161] primer

[0162] 2E9 CDR- GDSVSSNSAA

[0163] H1

[0164] 2E9 CDR- TYYRSKWYN

[0165] H2

[0166] 2E9 CDR- ALQRGWYPNYFDS

[0167] H3

[0168] 2E9 CDR- GGSIAKNY

[0169] L1

[0170] 2E9 CDR- EDN

[0171] L2

[0172] 2E9 CDR- QSYDSSSHGV

[0173] L3

[0174] 2E9 VH QVQLQQSGPGLVKPSQTLSLTCAISGDSVSSNSAAWNW IRQSPSRGLEWLGRTYYRSKWYNDYAESVKSRITINPDT SKNQFSLQLSSVTAADTAVYYCALQRGWYPNYFDSWG QGTLVTVSS

[0175] 2E9 VL NFMLTQPRSVSASPGKTVTISCTGSGGSIAKNYVQWYQ QRPGSAPTTIIYEDNQRPSGVPDRFSGSIDSSSNSASLTIS GLRTEDEADYYCQSYDSSSHGVFGGGTKLTVL

[0176] 4E5CDR- GDSVSSNSAA

[0177] H1

[0178] 4E5CDR- TYYRSKWYN

[0179] H2

[0180] 4E5 CDR- ARSGYYGSGGMDV

[0181] H3

[0182] 4E5 CDR- NIGSK

[0183] L1

[0184]

[0185] 4E5 CDR- YDS

[0186] L2

[0187] 4E5 CDR- QVWDSSSDHVV

[0188] L3

[0189] 4E5 VH QVQLQQSGPGLVKPSQTLSLTCAISGDSVSSNSAAWNW IRHSPSRGLEWLGRTYYRSKWYNDYAVSVKSRITINPDT SKNQFSLQLNSVTPEDTAVYYCARSGYYGSGGMDVWG QGTTVTVSS

[0190] 4E5 VL TSYVLTQPPSVSVAPGKTARITCGGNNIGSKSVHWYQQ KPGQAPVLVIYYDSDRPSGIPERFSGSNSGNTATLTISRV EAGDEADYYCQVWDSSSDHVVFGGGTKVTVL

[0191] 2E9 Heavy QVQLQQSGPGLVKPSQTLSLTCAISGDSVSSNSAAWNW chain IRQSPSRGLEWLGRTYYRSKWYNDYAESVKSRITINPDT SKNQFSLQLSSVTAADTAVYYCALQRGWYPNYFDSWG QGTLVTVSSASTKGPSVFPLAPSSKSTSGGTAALGCLVK DYFPEPVTVSWNSGALTSGVHTFPAVLQSSGLYSLSSVV TVPSSSLGTQTYICNVNHKPSNTKVDKRVEPKSCDKTHT CPPCPAPELLGGPSVFLFPPKPKDTLMISRTPEVTCVVVD VSHEDPEVKFNWYVDGVEVHNAKTKPREEQYNSTYRV VSVLTVLHQDWLNGKEYKCKVSNKALPAPIEKTISKAK GQPREPQVYTLPPSREEMTKNQVSLTCLVKGFYPSDIAV EWESNGQPENNYKTTPPVLDSDGSFFLYSKLTVDKSRW QQGNVFSCSVMHEALHNHYTQKSLSLSPGK

[0192] 2E9 Light NFMLTQPRSVSASPGKTVTISCTGSGGSIAKNYVQWYQ chain QRPGSAPTTIIYEDNQRPSGVPDRFSGSIDSSSNSASLTIS GLRTEDEADYYCQSYDSSSHGVFGGGTKLTVLRTVAAP SVFIFPPSDEQLKSGTASVVCLLNNFYPREAKVQWKVD NALQSGNSQESVTEQDSKDSTYSLSSTLTLSKADYEKH KVYACEVTHQGLSSPVTKSFNRGEC

[0193] 4E5 Heavy QVQLQQSGPGLVKPSQTLSLTCAISGDSVSSNSAAWNW chain IRHSPSRGLEWLGRTYYRSKWYNDYAVSVKSRITINPDT SKNQFSLQLNSVTPEDTAVYYCARSGYYGSGGMDVWG QGTTVTVSSASTKGPSVFPLAPSSKSTSGGTAALGCLVK DYFPEPVTVSWNSGALTSGVHTFPAVLQSSGLYSLSSVV TVPSSSLGTQTYICNVNHKPSNTKVDKRVEPKSCDKTHT

[0194]

[0195] CPPCPAPELLGGPSVFLFPPKPKDTLMISRTPEVTCVVVD VSHEDPEVKFNWYVDGVEVHNAKTKPREEQYNSTYRV VSVLTVLHQDWLNGKEYKCKVSNKALPAPIEKTISKAK GQPREPQVYTLPPSREEMTKNQVSLTCLVKGFYPSDIAV EWESNGQPENNYKTTPPVLDSDGSFFLYSKLTVDKSRW QQGNVFSCSVMHEALHNHYTQKSLSLSPGK

[0196] 46 4E5 Light TSYVLTQPPSVSVAPGKTARITCGGNNIGSKSVHWYQQ chain KPGQAPVLVIYYDSDRPSGIPERFSGSNSGNTATLTISRV EAGDEADYYCQVWDSSSDHVVFGGGTKVTVLRTVAAP SVFIFPPSDEQLKSGTASVVCLLNNFYPREAKVQWKVD NALQSGNSQESVTEQDSKDSTYSLSSTLTLSKADYEKH KVYACEVTHQGLSSPVTKSFNRGEC

[0197] 47 5B11 Heavy EVQLVESGGGLVQPGGSLRLSCAASGFTFSSYAMSWVR chain QAPGKGLEWVSAISGSGGSTYYADSVKGRFTISRDNSK NTLYLQMNSLRAEDTAVYYCAKDWGLWFGELPRQGYF DYWGQGTLVTVSSASTKGPSVFPLAPSSKSTSGGTAAL GCLVKDYFPEPVTVSWNSGALTSGVHTFPAVLQSSGLY SLSSVVTVPSSSLGTQTYICNVNHKPSNTKVDKRVEPKS CDKTHTCPPCPAPELLGGPSVFLFPPKPKDTLMISRTPEV TCVVVDVSHEDPEVKFNWYVDGVEVHNAKTKPREEQY NSTYRVVSVLTVLHQDWLNGKEYKCKVSNKALPAPIEK TISKAKGQPREPQVYTLPPSREEMTKNQVSLTCLVKGFY PSDIAVEWESNGQPENNYKTTPPVLDSDGSFFLYSKLTV DKSRWQQGNVFSCSVMHEALHNHYTQKSLSLSPGK

[0198] 48 5B11 Light TSYVLTQPPSVSVAPGKTARIPCGGNNIGSKSVHWYQQ chain KPGQAPVLVIYYDFDRPSGIPERFSGSNSGNTATLTISRV EAGDEADYYCQ VWD S S SDHVIFGGGTKLTVLRTVAAP S VFIFPPSDEQLKSGTASVVCLLNNFYPREAKVQWKVDN ALQSGNSQESVTEQDSKDSTYSLSSTLTLSKADYEKHK VYACEVTHQGLSSPVTKSFNRGEC

[0199]

[0200] All primers are referred to in the 5’ to 3’ orientation.

[0201] Reference binding to a HLA-E:peptide complex herein may refer to binding to a peptide which is in complex with HLA-E. BRIEF DESCRIPTION OF THE DRAWINGS

[0202] Figure 1 - shows a DNA sequence encoding a single chain trimer of WT1 peptide-linkerl32microglobulin-linker HLA-E heavy chain. The upper diagram shows the overall design of the single chain construct and the lower part indicates the amino acid sequence (single code). SP: signal peptide; WT1: WT1 peptide RMFPNAPYL; EGFP: enhanced green fluorescent protein

[0203] Figure 2 - is a schematic representation of the scFv-CAR constructs. A. The diagram represents the functional elements involved in the CAR construct. B. The map of MSCV-derived p5Bl l retroviral vector. The scFv may be either the 5B11 scFv, the 2E9 ScFv or the 4E5 scFv.

[0204] Figure 3 - shows the results of a monoclonal phage ELISA for selected phage clones against HLA-E / WT1 complex and controls. A. A representative of 38 phage clones were analysed by ELISA for their ability to bind HLA-E / WT1 with a low background against control HLA-E / VL9 monomer. B. The two clones p5Bl 1 and p4B6 were further confirmed with a panel of monomers including HLA-E / WT1, HLA / E-VL9, HLA-E / SARS-CoV2, HLA-A2402 / WT1 and HLA-E / Mtb44. The nucleotide sequences of the p5Bl l and p4B6 were identical.

[0205] Figure 4 - demonstrates binding specificity of the generated IgGl 5B11 mAb against a panel of monomers validated via ELISA. Both anti-HLA-E / WTl wildtype and LALA version (mutated to abrogate Fc receptor binding) antibodies showed high specificity for HLA-E / WT1 and HLA-E / Mtb44. The SDS-PAGE gel images reveal the integrity and purity of the produced IgGl antibodies.

[0206] Figure 5 - shows the cell surface staining using 5Bll-scFv. Top- Staining with a control antibody (2M2, BioLegend) that recognises p2-microglobulin of P2-microglobulin deficient 293T cells to confirm surface expression of selected HLA-E*01:03 SCT (presenting the canonical VL9 [VMAPRTLLL] peptide, the WT1 peptide [RMFPNAPYL], two M. tuberculosis peptides [mtb44, RLPAKAPLL; mtbl4, RMAATAQVL], or a HIV-1 Gag peptide [RL9HIV, RMYSPTSIL]), or an HLA-A*02:01 SCT presenting the WT1 peptide (far right). Bottom panel represents cells stained with 5Bl l-scFv. The 5B11 binds to HLA-E containing peptides that share arginine (R) at the N terminus, position 1.

[0207] Figure 6 - confirms specificity of 5B11 antibody for arginine at position 1 of the peptide. Top: Relative staining with the HLA-E / WT1 antibody of p2-microglobulin-deficient 293T cells expressing HLAE*01:03 SCT encoding the canonical VMAPRTLLL peptide with position 1 changed to every other amino acid. Bottom:

[0208] Staining of p2-microglobulin-deficient 293T cells expressing HLA-E*01:03 SCT encoding the VMAPRTLLL, RMAPRTLLL and KMAPRTLLL peptides.

[0209] Figure 7 - demonstrates the binding of HLA-E-WT1 tetramer to 5B11 CAR on CD8+ T cells. Left panel show staining with the HLA-E-WT1 tetramer and HLA-E VL9 (negative control) tetramer on CAR-anti-CD19 transduced CD 8+ T cells and CAR-anti HLA-E-WT1 transduced T-cells. The plot on the right shows the HLA-E WT1 tetramer and anti-EGFR (CAR marker) staining of the CAR transduced T cells.

[0210] Figure 8 - shows Anti-HLA-E-WTl CAR-T killing of an AML cell line P31FUJ.

[0211] Upper panels: CAR-T killing ofP31FUJ AML cell line cells cocultured with 5B11 CAR T at E: T of 1: 1 for 6h and 24h. 5B11 CAR T cells were cocultured with WT1 peptide-pulsed P31FUJ. At 6 hours, killing of AML cells was greater when WT1 peptide added. HLA-E-WT1 antibody 5B11 partially blocked killing of AML by CAR T cells. Reduction of AML cells was greater when the CAR transduced-T cells were a mixture of both CD8 and CD4 T cells.

[0212] Figure 9 - shows the structure of HLA-E:peptide complex bound by antibody 5B11 of the invention bound to HLA-E*0103 -WT1 peptide at 3.2A. A: shows the orientation of the Fab (pink and yellow) over the N-terminus of the peptide (red) in the binding groove of HLA-E (green). B: shows the hydrogen bonds formed between the Fab light chain and the indicated side chains of the HLA-E alpha helix C: shows the single H-bond between the arginine at position 1 of the peptide and the carbonyl group of glycine 29 in the Fab Heavy chain CDR1. D shows the CDR1, CDR2 and CDR3 contacts between the Fab heavy chain and the alpha -2 helix of HLA-E. The interactions are tabulated in 9E. Figure 10A - demonstrates different AML cell line killing by a CAR-T cell comprising the 5B11 scFv of the invention in the presence and absence of added WT110. B shows the CAR-T cell mediated % reduction of the AML cells relative to the irrelevant T cell, at effector:target cell ratios of 1:2, 1:1 and 2:1 in the absence of added WTlpeptide.

[0213] Figure 11- demonstrates Lysis of Ovarian Cancer Cells and 2E9 CAR-T cells lyse WT1+ / HLA-E+ ovarian cancer cell lines. The ovarian cancer cell lines were seeded at low density onto the xCELLigence (Agilent) wells and then cultured at 37°C in the absence of added WT1 peptide. As the cells grew they adhered to the surface of the well which altered impedance which was measured. After 24 hours CAR-T cells, or identically cultured control CD8 T cells, were added, at the indicated effectortarget ratios, and impedance was measured for up to 100 hours. The impedance signal reached a plateau at 40-60 hours. The loss of adherent cells indicated CAR-T cell mediated lysis. The figure shows that both ovarian cancer line cells were killed by the CAR-T cells though Kuramochi cells were more sensitive. These cells expressed higher levels ofWTl and HLA-E.

[0214] Figure 12 - demonstrates lysis of Acute Myeloid Leukaemia (AML) cells. 2E9 CAR-T cells lyse WT1+ / HLA-E+ AML cell lines. The AML cells P31FUJ and M0L13 were seeded onto xCELLigence wells that had been coated with anti-CD71 (anti-Transferrin receptor monoclonal antibody) to enable them to stick to the surface. After 4-6 hours the WT1 CAR-T 2E9 cells were added at different effector Target ratios as indicated and impedance measured for up to 80 hours of co-culture. Lysis of target cells was indicated by loss of impedance. The figure shows lysis of both AML cell lines, in the absence of added WT1 peptide though to different degrees.

[0215] Figure 13 - shows that lysis of AML cells was HLA-E dependent. The HLA-E gene was targeted for CRISPR-Cas9 knock-down in the AML cell lines P31FUKI and M0LM13. The AML cells were then cultured in the xCELLigence assay with WT1 CAR-T 2E9 cells at a ratio of 5:1. Lysis of both P31FUJ and M0LM13 was completely inhibited by HLA-E knock-down, showing HLA-E specificity.

[0216] Figure 14 - shows the specificity of 2E9 antibody binding in an ELISA to HLA-E*01:01-WTl peptide and to HLA-E*01:03-WTl peptide, but not to HLA-E*01:03 bound to VL9 VMAPRTLLL, an HPV16 El peptide, Mtb44 peptide, a SARS-CoV2 peptide, all of which bind well to HLA-E, nor to HLA-A* 0:201 bound to the WT1 peptide.

[0217] Figure 15 - shows the crystal structure at 2.2A of the 2E9 Fab fragment (pink heavy chain and white light chain) bound to the WT1 peptide (red) in the groove of HLA-E*01:03. B2microglobulin is shown in blue and the stabilizing camelid nanobody specific for B2m in yellow, at 2.2A.

[0218] Figure 16 - A shows the interactions between the 2E9 Fab heavy chain CDR3 (pink) and the tyrosine at position 8 and the C-terminal carbonyl group of the WT1 peptide, which also forms an H-bond with the lysine in the CDR3. Hydrogen bonds are formed between the carbonyl group of the asparagine at position 5 of the peptide and the tryptophan at residue 105 in the heavy chain CDR3. The Pi stacking of the aromatic ring of tyrosine at position 8 of the peptide and the tryptophan at position 105 of the heavy chain CDR3 are also shown. B shows that in the interactions between 2E9 Fab Heavy Chain and HLA-E Heavy Chain, CDR2 and CDR3 that equally contribute to binding.

[0219] Figure 17 - shows that in the interactions between 2E9 Fab Light Chain and HLA-E Heavy Chain, CDR1 seems to dominate.

[0220] MATERIALS AND METHODS

[0221] Primary cells, cell lines and antibodies

[0222] Human peripheral blood mononucleated cells (PBMCs) were purified by FicoII-Hypaquedensity centrifugation (Sigma) from cord blood samples provided by the NHS Blood and Transplant. Platinum-GP (Plat-GP) retroviral packaging cell line was purchased from Cell Biolabs, PG-13 stable retroviral packing cell line and HEK-293F were obtained by ATCC, various Esptein-Barr virus (EBV) infected human B cells were generated from SARS-CoV-1 or HIV patients in the inventors’ lab, and the acute myeloid leukaemia cell line P31FUJ

[0223]

[0224] (httgs7 / dej^^ Monoclonal antibodies against human HLA-E (clone 3D 12) conjugated to APC and anti-human EGFR / APC were purchased from Biolegend, and the isotype control mouse antibody IgGlK / APC was obtained from Invitrogen. Other anti-human CD3 / APC-H7, CD19 / BV510, CD4 / PerCP-Cy5.5, CD8 / BV421, and Streptavidin / PE were purchased from BD Biosciences.

[0225] Generation of HLA-E / WT1 monomer and tetramer

[0226] The WT1 peptide, residuesl26-135, RMFPNAPYL was predicted to bind to HLA-E using an algorithm described in Walters et al (Walters et al., 2020) It is also one of the possible binding peptides predicted by NET-MHC. A DNA oligonucleotide encoding this sequence was incorporated into a single chain trimer to encode: WT1 peptide-linker-p2microglobulin-linker HLA-E heavy chain (Figure 1). This was transfected into 293T cells and surface expression of HLA-E was detected with the antibody 3D12 which is specific for correctly folded HLA-E (Biolegend). The peptide was then synthesised and shown to bind to HLA-E in an ELISA assay, as described by Walters et al (Walters et al., 2020). Binding was similar to the level of binding of the canonical HLA class la signal peptide VMAPRTLVL which is the natural ligand for HLA-E in most cells (Braud et al., 1997). Thermal stability of HLA-E refolded with the WT1 peptide was determined by differential scanning fluorimetry giving a thermal melt value of 50oC, which is comparable to HLA-E bound to VL9 (Walters et al., 2022). HLA-E-WT1 tetramers were made using previously described methods (Walters et al., 2018; Walters et al., 2020; Walters et al., 2022). HLA-E protein with a C terminal BirA tag was expressed in inclusion bodies in E.coli, then solubilised in urea. Then it was refolded with beta-2-microglobulin (P2m) and biotinylated using the Bir A enzyme (Altman et al., 1996) (Braud et al., 1998).

[0227] Phage display screening for HLA-E / WT1 complex

[0228] For the biopanning of biotinylated HLA-E / WT1 monomer, a single-chain Fv library for M13 phage display (pre-immune library constructed from pooled donor lymphocyte RNA). Bacteriophages were first preincubated in 2% skimmed milk in PBS (MPBS) for Ih at room temperature (RT). Streptavidin Dynabeads were washed three time with PBS- 0.1% Tween 20 (PBST) and then incubated in 2% PBSM for Ih at RT. The library was first panned against the M-280 Streptavidin Dynabeads to remove non-specific binders. The unbound phages were then incubated with HLA-E biotinylated monomer bound to the canonical HLA class I Signal peptide VMAPRTLVL (HLA-E / VL9) (Braud et al., 1997) and captured by M-280 Streptavidin Dynabeads to dispose of binders against HLA-E / VL9 complex. The remaining phages were incubated with biotinylated HLA-E / WT1 monomer and pulled down by Streptavidin Dynabeads. After 15 washes with 0.1% PBST and further 5 washes with PBS, bound phages were eluted by 1 mg / mL trypsin-PBS. The eluted phages were diluted in PBS and used to infect Escherichia coli TGI cells (Agilent Technologies) and grown on TYE plates containing 100 pg / mL ampicillin and 1% glucose overnight at 37 °C. The phage infected TGI lawn was collected for phage amplification with helper phage M13K07 (New England BioLabs). After overnight growth, the phages in supernatant were precipitated using 20% PEG / NaCl (Polyethylene glycol 8000, 2.5 M NaCl) before subjecting to further selection rounds. Five rounds of biopanning were performed with a consistent amount of 30 pg / mL HLA-E / VL9 and decreasing amounts of biotinylated HLA-E / WT1 monomer (30, 10, 5, 1 and 0.5 pg / mL for rounds 1-5, respectively).

[0229] Characterization of the full-length hlgGl for HLA-E / WT1 complex Monoclonal phage ELISA - Individual phage clones from fourth and fifth bio-panning rounds on HLA-E / WT1 complex were tested for binding to biotinylated HLA-E / WT1 monomer by ELISA. Selected clones were grown in 96-well deep well plates in 2TY medium containing 100 pg / mL ampicillin and 1% glucose followed by phage production using M13K07 helper phage (New England BioLabs) rescue. Phage were produced overnight in 2TY-AmpKan, then the culture supernatant used for ELISA. Pierce Streptavidin coated High-Capacity Plates (ThermoFisher Scientific) were coated with 2 pg / mL either HLA-E / WT1 -biotin or HLA-E / VL9-biotin monomers in PBS overnight at 4°C. The plates were washed three times with 200pl 0.1 % PBST. Phage supernatant from the same well was then transferred to both HLA-E / WT1 and HLA-E / VL9- coated plates and incubated for 1 h at room temperature. The wells were washed three times with 0.1% PBST and then 100 pL secondary antibody added (1 / 5000 dilution of HRP / Anti-M13 Monoclonal Antibody Conjugated- SinoBiological) and incubated for 1 h at room temperature. The plates were washed four times with 0.1% PBST, followed by addition of 100 pL of Tetramethylbenzidine (TMB) Substrate Solution (Invitrogen). The reaction was stopped after 5-10 min using 50 pL Stop Solution (Abeam). Absorbance was measured at 450nm using GloMax Microplate Reader (Promega).

[0230] Cloning and Human IgGl Expression

[0231] Phagemid DNA was extracted from individual phage clones that have showed specificity to HLA-E / WT1 complex. Clones were grown overnight in 2TY-AmpGlu media, then phagemids isolated using Purelink Hipure Plasmid Mini Prep kit (Invitrogen). Selected phagemids were sequenced by Sanger sequencing at Genewiz (Genomics from Azenta Life Sciences, UK), using phagemid-derived forward primer (5’ CAG GAA ACA GCT ATG AC 3’).

[0232] Reactions were performed (PCR) using primers (Forward- 5’ GCA ACC GGT GTA CAT TCT GAG GTG CAG CTG GTG GAG 3’; Reverse- 5’ TGG GCC CTT GGT CGA CGC GCT CGA GAC GGT GAC CAG 3’) for VH and (Forward- 5’ GCA ACC GGT GTA CAT TCT ACG TCC TAT GTG CTG ACT 3’; Reverse- 5’ TGG TGC AGC CAC CGT ACG TAG GAC GGT GAG CTT GGT 3’) for VL. The PCR products were separately cloned into respective IgGl heavy chain and IgGl light chain expression vector using In-Fusion HD Cloning kit (TaKaRa). Plasmid DNA for transfection was prepared using PureLink HiPure Plasmid Maxiprep kit (Invitrogen) and expressed in HEK-293F cells. Suspension 293F cells were cultured in Erlenmeyer Flask (Corning) with FreeStyle 293 expression media (ThermoFisher Scientific) at 37 °C, 8% CO2 with shaking at 125 rpm. Transfection was performed using 293fectin Transfection reagent (ThermoFisher Scientific) following manufacturer’s protocol. The culture was incubated at 37 °C, 8% CO2 with shaking at 125 rpm for up to 5 days. The supernatants were purified by Protein A affinity chromatography. Bound IgGl antibodies were eluted with 0.2 M Glycine, pH 2.5 (ThermoFisher Scientific) and neutralized with IM Tris-HCl, pH 8.0 (Invitrogen). The mixture was buffer-exchanged and concentrated using Ultra Centrifugal filter-3 OK (Amicon, Millipore, Sigma). To assess the purity, 2 pg of obtained mAbs were loaded onto Sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) in the presence and absence of lx reducing agent (ThermoFisher Scientific). The gel was stained with InstantBlue Coomassie protein stain (Abeam) and analysed using Gel Doc XR+ with Image Lab software (Bio-Rad).

[0233] IgGl ELISA

[0234] Pierce Streptavidin coated High-Capacity Plates (ThermoFisher Scientific) were washed three times with 0.1% PBST before coating with 2 pg / mL HLA-E / VL9, HLA-E / WT1, HLA-E / SARS-CoV-2, HLA-E / Mtb44 and HLA-A2402 / WT1 monomers overnight at 4°C. Following three washes of 0.1% PBST, the p5B 11, p5B 11-LALA and p4B6 mAbs were added from a start concentration of 10 pg / mL with a serial dilution (1 in 5) to 0.00064 pg / mL. The plates were then incubated for 1 h at room temperature and washed three times with 0.1% PBST. HRP-conjugated goat antihuman

[0235] IgG secondary antibody (Abeam) was diluted at 1 / 5000 in PBS and 100 pL was added per well. After Ih incubation and three times 0.1% PBST washes, 100 pL of Tetramethylbenzidine (TMB) Substrate Solution (Invitrogen) was added and incubated 10 min in dark before adding 50 pL Stop Solution (Abeam). Absorbance was read at 450 nm using GloMax Microplate Reader (Promega).

[0236] Single Chain Trimers with the WT1 peptide

[0237] A single chain trimer (SCT) or HLA-E*01:03 was constructed that encoded the HLA-E signal sequence, the WT1 peptide, a flexible glycine / serine linker, the mature coding sequence of P2microglobulin, a second glycine / serine flexible linker, the HLA-E*01:03 heavy chain with a mutation (position 84 changed from tyrosine to alanine to open the end of the peptide binding groove to better accommodate the linker that follows the peptide), a short linker, and Enhanced Green Fluorescent protein (EGFP) (Figure 1). This SCT was expressed in 293T cells, and cells were stained with the HLA-E-specific antibody 3D 12 (Biolegend). The WT1 SCT expressed as well as the HLA-E SCT encoding the canonical high-affinity VMAPRTLLL peptide, suggesting that the WT1 peptide is also bound by HLA-E with high affinity (Figure 5). In contrast, a HLA-*02:01 SCT with the WT1 peptide did not express as well as an HLA-A*02:01 SCT with the HCMV pp65 peptide NLVPMVATV (Figure 3), suggesting the WT1 peptide is not an optimal peptide for HLA-A*02:01.

[0238] Generation of construct of WT1 Chimeric Antigen Receptor (CAR) and retrovirus production

[0239] The MSCV CAR expression retroviral vector was modified from MSCV-IRES-GFP vector (Addgene) by replacing IRES-GFP with human CD8 transmembrane domain and third generation CAR intracellular signalling domain (co-stimulatory domains of CD28 and 4-1BB, CD3^ cytoplasmic signalling domain) with an addition of a truncated epithelial growth factor receptor EGFR as a surface co-expression marker protein (Figure 2). The identified p5Bl l scFv was cloned into the CAR vector using In-Fusion HD Cloning kit (TaKaRa) (Forward-5’ GGT TCC ACA GGT GAC GAG GTG CAG CTG GTG 3’; Reverse-5’ CAC GAA TGC GGC CGC TAG GAC GGT GAG CTT 3’).

[0240] The Plat-GP retroviral packaging cell line was first used by co-transfecting with 0.5 pg pCMVVSV- G envelope vector and 2 pg MSCV-p5Bl l CAR vector using 7 pL X-tremeGENE HP Transfection Reagent (Roche) in a six-well tissue culture plate. The supernatant containing retroviral particles was collected after 48h and 72h posttransfection and used to infect PG-13 for generating a stable virus-producing cell line. PG-13 cells were infected using Plat-GP derived viral particles together with 8 pg / mL polybrene (Sigma) and centrifuged at 1000 x g at 32°C for 2 h. After 24 h incubation, the transduction efficiency was assessed by flow cytometry analysis of EGFR expression level. The PG13 cells showed high efficiency (>98%) of EGFR expression, was maintained to constantly produce retroviruses. The supernatant containing CARV retrovirus particles was harvested every 24-48 h and concentrated for T cell transduction.

[0241] Transduction of primary T cells and CAR-T expansion

[0242] Human PBMCs were prepared as described previously. Human T-Activator CD3 / CD28 (ThermoFisher Scientific) Dynabeads were mixed with PBMCs in RPMI media (THermoFisher Sceintific) with 10% fetal calf serum (RIO) supplemented with 50 lU / mL IL-2 (Miltenyi) at a density of 3 x 106cells / mL in a bead-to-cell ratio of 1: 1 for 48-72 h. Recombinant Human Fibronectin Fragment (RetroNectin) from TaKaRa was coated at 20 pg / mL on non-treated tissue culture six-well plate to bind viruses with centrifugation at 1000 x g for 2 h at 32 °C. The plate was rinsed once with PBS before adding pre-activated PBMCs at a density of 1 x 106cells / mL. The mixture of RetroNectin-bound virus and PBMCs was then centrifuged at 500 x g for 5 min followed by incubating at 37°C. Transduced cells were harvested 3 days post-transduction for sorting using anti-CD3 / APC-H7 and anti-EGFR / APC antibodies. Sorted CAR-T cells were expanded in a six-well G-Rex culture plate (Wilson Wolf) with a minimum amount of 1 x 106cells per well in 30 mL of R10 supplied with 10 ng / mL human IL-7 and 14 ng / mL IL- 15 (Miltenyi). The media was replenished every two to three days. CAR-T cells and autologous non-transduced T cells were harvested on Day 12 and frozen at -80°C for 48h followed by storing in liquid nitrogen.

[0243] Antibody and Tetramer Staining

[0244] Soluble HLA-E / WT1 and HLA-E / VL9 monomers were produced using methods previously described (Braud et al., 1998; Walters et al., 2020; Walters et al., 2022). 20 pg biotinylated monomers were thawed and prepared for trimerization with Streptavidin / PE at a molar ratio of 4: 1 at room temperature, 1 / 10 of the Streptavidin / PE solution was added to the monomer solution every 10 min. Once assembled, the generated WT1 CAR-T cells, CD 19 CAR-T cells and autologous non-transduced T cells were stained with 0.5 pg / mL PE-conjugated tetramers for Ih at 4°C, followed by antibodies: anti-CD3 / APC-H7 and anti-EGFR / APC and then acquired using Flow Cytometry and analysed with FlowJo software. Cytotoxicity assays for WT1 CAR-T function

[0245] CAR-T Killing of Acute Myeloid Leukaemia (AML ) cell lines

[0246] WT1 CAR-T cells were sorted for EGFR+ / CD8+ double positive or EGFR+ / CD4+ double positive cells to enrich CD8+ CAR-T or CD4+ CAR-T. CD8+ T cells or CD4+ T cells were also sorted from the untransduced PBMC from the same donor which have been cultured under the same condition as CAR-T to be used as negative controls. 1 x 105AML P31FUJ cells were cocultured with WT1 CD8+ CAR T or non-transduced CD8+ T cells at E: T ratio of 1:1 in 96-well tissue culture plate without or with WT1 peptide at final concentration of 50pM. After 6hrs or 24hrs coculture, 6 pL CountBright beads (Invitrogen) were added to each well and mixed with cells. Cells and CountBright beads were then harvested and washed with PBS followed by staining with viability LIVE / DEAD Fixable Aqua / CD3 / CD8 antibodies. Cells were acquired on a LSRFortessa (BD Biosciences) flow cytometer and analyzed using FlowJo v10.4 (Tree Star). P31FUJ cells were gated on live / CD3- / CD8- cells. The number of P31FUJ cells were normalised by the number CountBright beads acquired from the same well. % Reduction of P31FUJ cells was calculated to data obtained with non-transduced CD8+ T cells using the following formula:

[0247]

[0248] wmber &>fF31FUJ ceffs co-cutored WTI -CAR-T ce / fe _ X 100 r ofced nu feer of PJIFiJJ cefe co-cuit ns w?t tw? transduced CDS+ Tce&

[0249] CAR-T Killing of B Lymphoblastoid Cell Line (BLCL) cells.

[0250] CAR-T cells were thawed and recovered in RIO for at least 2 h at 37 °C. The Epstein Barr Virus (EBV)-transformed B lymphoblastoid cell line (BLCL) cells were stained with 1 pM CFSE (ThermoFisher Scientific) for 15 min at 37°C and followed by incubating with RIO for 5 min at room temperature. The stained cells were then rinsed with RIO media for three times. 1 x 105BLCL were co-cultured with CAR-T cells, or nontransduced T control at ratio 1:1 in 200 pL media at 37°C. WT1 peptide was pulsed at a concentration of 50 pM and kept in the media during the assay. The co-cultured samples were analysed by Flow Cytometry (LSRII, BD Biosciences) at 4h and 8h timepoint. Anti-CD3 / APC-H7 was used to stain CAR-T cells for 20 min at room temperature. IpL of 4’,6-diamidino-2-phenylindole (DAPI, Miltenyi) and 6 pL CountBright beads (Invitrogen) were added to each sample and mixed properly. A fixed number of beads (5000) was acquired for flow cytometry, the absolute number of cells was extrapolated as follows: cells in the tube= (number of collected cells / number of collected beads) x total beads added to the tube. The percentage of survival was calculated as:

[0251] % survival = absolute number of viable target celts after culture with CAR-T. X 100 absolute number of viable target cells before adding CAR I

[0252] The BLCL cells were gated on live / CD3- / CFSE+.

[0253] EXAMPLES

[0254] Example 1 - Phage screening of scFvs specific for HLA-E: WT1 complex

[0255] The inventors initially developed a monoclonal antibody against HLA-E bound to a peptide in the Wilms Tumour-1 protein (WT1). WT1 is a transcription factor that is present at low levels or absent in most normal tissue but is over expressed in several cancers, including acute myeloid leukaemia (AML), as well as in many infections, therefore making it a good target for immunotherapy.

[0256] The biotinylated HLA-E / WT1 monomers were used as the bait in five rounds of biopanning. Non-specific and HLA-E / VL9 phage binders were captured and removed by M-280 Streptavidin Dynabeads from each round. In total, 480 clones were randomly picked and screened by ELISA. The majority of the selected clones were cross reactive betwe en both monomers, of which two clones (p5Bll and p4B6) showed distinct specificity to HLA-E / WT1 complex with a low binding ability against HLA-E / VL9. The two phage clones were then isolated and tested with a panel of monomers including HLA-E / SARS-CoV2, HLA-A2402 / WT1 and HLA-E / Mtb44. The results showed that both p5Bll and p4B6 were also reactive to HLAE / Mtb44 complex while remained a clear background to HLA-E / VL9, HLA-E / SARS-CoV2 and

[0257] HLA-A2402 / WT1. The individual phage genomic DNA was extracted and analysed, the sequencing results showed the two phage clones have identical sequence (figure 3).

[0258] Example 2 - Specificity of the IgGl mAb 5B11

[0259] Although the two phage clones showed the same sequence, they were cloned into a human immunoglobulin G1 (IgGl) vector individually to validate the binding specificity as an IgGl monoclonal antibody (mAb). The mAbs were produced by HEK293F cells and purified from the cell culture supernatants via protein A affinity chromatography. The purity and integrity of p5B11 and p4B6 mAbs were analysed on SDS-PAGE. Under reducing conditions, two bands were observed which corresponding to the IgGl heavy and light chains. Overall, the IgGl mAbs were successfully produced. The binding specificity of the generated mAbs was assessed by ELISA analysis. A serial dilution (10, 2, 0.4, 0.08, 0.016, 0.0032, 0.00064 pg / mL) of both p5Bl 1 and p4B6 mAbs was performed against a panel of monomers including HLAE / WT1, HLA-E / VL9, HLA-E / SARS-CoV2, HLA-A2402 / WT1 and HLA-E / Mtb44. Consistent with previous phage ELISA results, both p5B11 and p4B6 mAbs showed specific binding to HLAE / WT1 and HLA-E / Mtb44 but not to the other complexes. Additionally, a double mutated (Leu234Ala and Leu234Ala) IgGl mAb was generated and purified using the same methods that revealed the matching binding pattern. Collectively, these data (figure 4) confirmed the specificity of the anti-HLA-E / WT1 mAbs (p5B11 and p4B6) against HLA-E / WT1 and HLA-E / Mtb44 complexes. Further analysis (Figs 5 and 6) showed that the antibody had specificity for HLA-E bound to peptides with arginine at their amino terminus. This specificity was confirmed by the crystal structures of figure 9.

[0260] Example 3 - Anti-HLA-E / WT1 CAR-T binds to HLA-E / WT1 tetramer

[0261] The p5Bl 1 scFv was cloned into a previously engineered third generation-CAR vector. Following retroviral transduction, T cells that co-expressed EGFR proteins were sorted for further expansion in G-rex plate. After 12 days, cells were harvested and the WT1-CAR expression was assessed by flow cytometric analysis at day 7 and day 12 utilizing assembled HLA-E / WT1 tetramer and anti-EGFR antibody. The CAR expression was 58.8% on the representative figure 7 right panel. Moreover, results demonstrated the generated HLA-E / WT1 CAR-T cells show high specificity to HLA-E / WT1 tetramer but not binding to the control HLA-E / VL9 tetramer. The CAR-anti-CD19 cells reacted with neither complex.

[0262] Example 4 - 5B11 CAR-T cells kill AML cells

[0263] The anti-cancer cytotoxic capacity of 5B11 WT1 specific CAR-T was evaluated by killing of AML cell line P31FUJ in vitro. P31FUJ cells were cocultured with enriched CD8+ CAR-T at E: T ratio of 1: 1 for 6 hours or 24 hours either with WT1 peptide at 50pM in culture or without. The anti-AML activity was quantified by proportional reduction of P31FUJ cells when they were incubated with CD8+ CAR-T in comparison to untransduced CD8+ T cells. CD8+ CAR-T reduced P31FUJ cells by 47.9% after 6hr coculture in the presence of added WT1 peptide. The killing of P31FUJ by CD8+ CAR-T increased to 72.9% after 24hr coculture.

[0264] The killing of P31FUJ cells by CD8+ CAR-T was blocked by 50% by the addition of anti-HLA-E WT1 antibody 5B11 at lOOpM in culture, confirming WT1 specific killing by CAR-T. CD4+ CART enhanced CD8+ CAR-T killing by 26.7% when CD8+ CAR T, CD4+ CAR-T and P31FUJ were cocultured at 1:1:1 ratio compared to no CD4+ CAR-T.

[0265] These experiments indicate (figure 8) that the transferred 5B11 CAR transduced into allogeneic CD8 T cells confers specificity for the WT1 peptide presented by HLA-E and that these T cells can kill AML cells. In these short term (4-6 hour) assays added peptide was needed but in longer assays lysis of AML cells in the absence of added peptide was seen.

[0266] Example 5- Breadth of specificity of 5B11 CAR T cells

[0267] The inventors also assessed killing activities of WT1 CAR-T with stimulation of canonical HLA-E binding signal peptide VMAPRTLVL (VL9), HLA-E binding HIV peptide RMYSPTSIL (GAGRL9) and Mycobacterium tuberculosis peptide RLPAKAPLL (Mtb44), with arginine at position 1 for the latter two peptides. The killing of P31FUJ by CD8+ WT1 CAR-T were observed for all 4 peptides tested after 24hr coculture, with WT1 peptide at highest killing of 52.6%, VL9 peptide at lowest killing of 32%, whilst GAG RL9 and Mtb44 showed intermediate killing of 37.6% and 41.5% respectively. The data suggested that the arginine at position 1 was crucial for the function specificity of WT1 CAR-T (Figure 3).

[0268] Firstly, the 5B11 CARs expression was validated by flow cytometric analysis of which the expression level is above 50%. p5B11 CAR-T and CD19 CAR-T cells were cocultured with BLCL at E: T ratio of 1:1 in the presence or absence of WT1 peptide added to the media at 50 pM. CD 19 CAR-T cells specifically and significantly lysed the BLCL with 47% lysis at 4h and 72% at 8h. p5B11 CAR-T cells exhibited specific killing in BLCL with pulsed WT1 peptide (28% reduction at 4h and 50% at 8h), and a reduced killing rate (21% reduction at 4h and 37% at 8h) was also observed of BLCL without pulsed peptide which probably due to native WT1 is presented on HLA-E of the EBV-infected BCL. Non-transduced T cells showed non-specific killing activity at a limited level (Figure 9 bottom). Overall, the data demonstrated that the p5B11 CAR-T cells potently and specifically lysed WT1 peptide-pulsed target cells.

[0269] Example 7- structure determination of binding of an antigen binding polypeptide 5B11 of the invention.

[0270] The inventors sought to understand the spatial binding characteristics of the antigen binding polypeptide of the invention, so X-ray crystallography studies were undertaken on a HLA-E: WT1 peptide complex bound to a Fab of the invention. Figure 9A shows the orientation of the Fab light chain (pink) and heavy chain (yellow) with HLA-E (green, bottom) and the peptide (red, running N-C from right to left in the left part of the figure). The Fab sits over the N-terminus of the peptide.

[0271] Figure 9B shows the Fab heavy chain (yellow, top) interactions with HLA-E. These involve the CDR2 (amino acids 57, 59) and CDR3 region (amino acids 103, 108 and 109) of the heavy chain. Figure 10C shows the contacts between the light chain CDR1 amino acid G29 and the arginine at position one of the peptide. This is the only peptide antibody direct contact, which is consistent with its specificity for peptides with arginine at position one. Figure 9D shows contacts between amino acids of the light chain CDR2 (51 and 52) plus N66 with the HLA-E alpha 1 helix.

[0272] Figure 9E shows the light chain contacts with the HLA-E alpha-2 helix, involving the light chain CDR2 residues 50, 53, CDR1 residues 303, 32 and the CDR3 amino acid 93. These contacts with the alpha helices of HLA-E give the Fab specificity for HLA-E.

[0273] Figure 9E shows the hydrogen bonds between the HLA-E:peptide complex and the Fab, and the distances of those contacts. Only single peptide contact is seen with the R at the N terminus.

[0274] Example 8 - killing of AML cells using a CAR-T cell comprising the antigen binding polypeptide 5B11 of the invention.

[0275] The inventors sought to demonstrate the utility of the antigen binding polypeptide of the invention in tumour cell killing, in the context of CAR-T cells, using the AML cell line P31FUJ in the xCELLigence assay (Figure 11). In the upper panels WT1 peptide was added to the cells in the assay. In the lower panels no peptide was added showing that at the 2: 1 CAR-T cell to target cell ration there was recognition of the AML cells in the absence of added peptide. Figure 12 summarises results with other AML cell lines. In both Figures 11 and 12, AML cell killing is demonstrated with a CAR-T utilising an antigen binding polypeptide of the invention. Figures 12-14 also demonstrate this with a CAR comprising clone 2E9.

[0276] Example 9 - Specificity of antibody 2E9 isolated from the phage scFv library. A second antibody was isolated from the same phage display library using HLA-E*01:01 bound to the WT1 peptide for positive selection and HLA-E*01:01 bound to the VL9 peptide for negative selection. The antibody showed specificity for just the WT1 peptide bound to either HLA-E*01:01 or HLA-E*01:03 in the initial screening against virus and mycobacterial derived peptides (Figure 14).

[0277] Example 10 - CAR-T made with 2E9 using the construct described in Figure 2 lysed ovarian cancer cell lines that express HLA-E and WT1 and AML cell lines expressing HLA-E and WT1 (Figs 11, 12). Knock-down of HLA-E using CRISPR-Cas9 reduced susceptibility to of the cell lines to CAR-T killing (Fig 13). Lysis was seen in the absence of added peptide for both the ovarian cancer cells and AML cells.

[0278] Example 10 - Structure Determination of the Fab fragment of antibody 2E9 of the invention bound to HLA-E*01:03 bound to the WT1 peptide. The inventors sought to understand the spatial binding characteristics of the antigen binding polypeptide 2E9 of the invention, so X-ray crystallography studies were undertaken on a HLA-E: WT1 peptide complex bound to a Fab of the invention. Figure 15 shows the orientation of the Fab light chain (pink) and heavy chain (yellow) with HLA-E (green, bottom) and the peptide (red, running N-C from right to left in the left part of the figure). The Fab sits over the C-terminal half of the peptide with a particular focus on the tyrosine at position 9 of the peptide. Figures 16 and 17 show the amino acids in the Fab that contact amino acids of HLA-E and the peptide, identifying critical interactions that could be mutated to alter peptide specificity or binding affinity.

[0279] Conclusion

[0280] The inventors have developed monoclonal antibodies that are specific for the WT1 peptide residues 126-135, RMFPNAPYL peptide bound to HLA-E. One of the antibodies 5B11 also binds other peptides that have arginine at position 1 in their sequence, bound to HLA-E, another antibody 2E9 has specificity for the penultimate tyrosine of the peptide. Crucially, neither antibodys binds to the dominant self peptide VMAPRTLVL (VL9) bound to HLA-E,. The inventors developed chimeric antigen receptors based on these two antibodies and show that rhey bind and kills cells expressing HLA-E and the WT1 peptide, including AML cells in the context of CAR-T cells. The antibody therefore has potential as a therapeutic or prophylactic reagent specific for pathogen or pathology associated peptides bound to the non-polymorphic HLA class I molecule HLA-E. These peptides include the above WT1 peptide and previously described peptides RL9, derived from HIV-1 gag and Two peptides Mtbl4 and Mtb44 derived from Mycobacterium tuberculosis, as well as others with an are bound by HLA-E and which have an arginine at position one.

[0281] References

[0282] Altman, J. D., P. A. Moss, P. J. Goulder, D. H. Barouch, M. G. McHeyzer-Williams, J. I. Bell, A. J. McMichael, and M. M. Davis. 1996. Phenotypic analysis of antigen-specific T lymphocytes. Science 274:94-96.

[0283] Braud, V., E. Y. Jones, and A. McMichael. 1997. The human major histocompatibility complex class lb molecule HLA-E binds signal sequence-derived peptides with primary anchor residues at positions 2 and 9. Eur J Immunol 27: 1164-1169.

[0284] Braud, V. M., D. S. Allan, C. A. O'Callaghan, K. Soderstrom, A. D'Andrea, G. S. Ogg, S. Lazetic, N. T. Young, J. I. Bell, J. H. Phillips, L. L. Lanier, and A. J. McMichael. 1998. HLA-E binds to natural killer cell receptors CD94 / NKG2A, B and C. Nature 391:795-799.

[0285] Walters, L. C., K. Harlos, S. Brackenridge, D. Rozbesky, J. R. Barrett, V. Jain, T. S. Walter, C. A.

[0286] O'Callaghan, P. Borrow, M. Toebes, S. G. Hansen, J. B. Sacha, S. Abdulhaqq, J. M. Greene, K. Früh, E. Marshall, L. J. Picker, E. Y. Jones, A. J. McMichael, and G. M. Gillespie. 2018. Pathogen-derived HLA-E bound epitopes reveal broad primary anchor pocket tolerability and conformationally malleable peptide binding. Nat Commun 9:3137.

[0287] Walters, L. C., A. J. McMichael, and G. M. Gillespie. 2020. Detailed and atypical HLA-E peptide binding motifs revealed by a novel peptide exchange binding assay. Eur J Immunol Walters, L. C., D. Rozbesky, K. Harlos, M. Quastel, H. Sun, S. Springer, R. P. Rambo, F. Mohammed, E. Y. Jones, A. J. McMichael, and G. M. Gillespie. 2022. Primary and secondary functions of HLA-E are determined by stability and conformation of the peptide-bound complexes. Cell Rep 39:110959.

Claims

CLAIMSCLAIMS1. An antigen binding polypeptide which is capable of specifically binding to a HLA-E:peptide complex.

2. An antigen binding polypeptide which competes for binding to a HLA-E:peptide complex with an antigen binding molecule of claim 1.

3. The antigen binding polypeptide of claim 1 or claim 2, wherein the peptide of the HLA-E:peptide complex has an arginine at position 1, and / or a tyrosine at position 8.

4. The antigen binding polypeptide of claim 1 or claim 2, wherein the peptide of the HLA-E:peptide complex is, or is derived from SEQ ID NO: 13.

5. The antigen binding polypeptide of any of claims 1-4, wherein the antigen binding polypeptide interacts with one or more of, such as all of, the following residues in HLA-E: Asp69, Gln72, Ile73, Arg75, Val76, Thr80, Tyr84, Serl47, Glu52; and / or wherein the antigen binding polypeptide interacts with one or more of, such as all of the following residues in HLA-E: Glul9, Asn38, Asp39, Ala40, Arg68, Gln72 and Arg75.

6. The antigen binding polypeptide of any of claims 1-5, wherein the antigen binding polypeptide is an antibody or antigen binding fragment thereof, or a Chimeric antigen receptor (CAR).

7. The antigen binding polypeptide of any of claims 1-6, wherein the antigen binding polypeptide comprises:a) a heavy chain variable region comprising a complementarity determining region CDR3 of SEQ ID NO: 29 or a sequence having at least 80%, 90%, 95%, 98%, 99% or 100% identity thereto; and / ora light chain variable region comprising a CDR3 of SEQ ID NO: 32 or a sequence having at least 80%, 90%, 95%, 98%, 99% or 100% identity thereto; orb) a heavy chain variable region comprising a complementarity determining region CDR3 of SEQ ID NO: 3 or a sequence having at least 80%, 90%, 95%, 98%, 99% or 100% identity thereto; and / ora light chain variable region comprising a CDR3 of SEQ ID NO: 6 or a sequence having at least 80%, 90%, 95%, 98%, 99% or 100% identity thereto; orc) a heavy chain variable region comprising a complementarity determining region CDR3 of SEQ ID NO: 37 or a sequence having at least 80%, 90%, 95%, 98%, 99% or 100% identity thereto; and / ora light chain variable region comprising a CDR3 of SEQ ID NO: 40 or a sequence having at least 80%, 90%, 95%, 98%, 99% or 100% identity thereto.

8. The antigen binding polypeptide of any of claims 1-7, wherein the antigen binding polypeptide comprises:a) a heavy chain variable region comprising: a CDR1 of SEQ ID NO: 27, a CDR2 of SEQ ID NO: 28, and a CDR3 of SEQ ID NO: 29; and / ora light chain variable region comprising: a CDR1 of SEQ ID NO: 30, a CDR2 of SEQ ID NO: 31, and a CDR3 of SEQ ID NO: 32; orb) a heavy chain variable region comprising: a CDR1 of SEQ ID NO: 1, a CDR2 of SEQ ID NO: 2, and a CDR3 of SEQ ID NO: 3; and / ora light chain variable region comprising: a CDR1 of SEQ ID NO: 4, a CDR2 of SEQ ID NO: 5, and a CDR3 of SEQ ID NO: 6; orc) a heavy chain variable region comprising: a CDR1 of SEQ ID NO: 35, a CDR2 of SEQ ID NO: 36, and a CDR3 of SEQ ID NO: 37; and / ora light chain variable region comprising: a CDR1 of SEQ ID NO: 38, a CDR2 of SEQ ID NO: 39, and a CDR3 of SEQ ID NO: 40,or sequences having at least 80% identity thereto.

9. The antigen binding polypeptide of any of claims 1-8, wherein the antigen binding polypeptide comprises:a) a heavy chain variable region comprising or consisting of SEQ ID NO: 33; and / or a light chain variable region comprising or consisting of SEQ ID NO: 34; orb) a heavy chain variable region comprising or consisting of SEQ ID NO: 7; and / or a light chain variable region comprising or consisting of SEQ ID NO: 8; orc) a heavy chain variable region comprising or consisting of SEQ ID NO: 41; and / or a light chain variable region comprising or consisting of SEQ ID NO: 42,or sequences having at least 80% identity thereto.

10. The antigen binding polypeptide of any of claims 1-9, wherein the antigen binding polypeptide comprises:a) a heavy chain comprising or consisting of SEQ ID NO: 43; anda light chain comprising or consisting of SEQ ID NO: 44; orb) a heavy chain comprising or consisting of SEQ ID NO: 45; anda light chain comprising or consisting of SEQ ID NO: 46; orc) a heavy chain comprising or consisting of SEQ ID NO: 47; anda light chain comprising or consisting of SEQ ID NO: 48.

11. A nucleic acid encoding the antigen binding polypeptide of any of claims 1-1012. A vector comprising the nucleic acid of claim 11.

13. A host cell comprising an antigen binding polypeptide of any of claims 1-10, the nucleic acid of claim 11, and / or the vector of claim 12.

14. A CAR-T cell or CAR-NK cell expressing an antigen binding polypeptide of any of claims 1-10, or comprising a nucleic acid and / or vector of claim 11 or 12.

15. A pharmaceutical composition comprising one or more antigen binding polypeptide of any of claims 1-10, nucleic acid of claim 11, vector of claim 12, host cell of claim 13 and / or CAR-T cell or CAR-NK cell of claim 14.

16. The antigen binding polypeptide of any of claims 1-10, nucleic acid of claim 11, vector of claim 12, host cell of claim 13, CAR-T cell or CAR-NK cell of claim 14 or composition of claim 15, for use in the treatment or prevention of one or more disease or disorder in a subject.

17. A method of treating or preventing one or more disease or disorder in a subject, comprising administering to the subject an effective amount of the antigen binding polypeptide of any of claims 1-10, nucleic acid of claim 11, vector of claim 12, host cell of claim 13, CAR-T cell or CAR-NK cell of claim 14, or composition of claim 15.

18. Use of the antigen binding polypeptide of any of claims 1-10, nucleic acid of claim 11, vector of claim 12, host cell of claim 13, CAR-T cell or CAR-NK cell of claim 14, or composition of claim 15, in the manufacture of a medicament for the treatment or prevention of one or more diseases or disorders in a subject.

19. The antigen binding polypeptide, nucleic acid, vector, host cell, CAR-T cell or CAR-NK cell, or composition for use according to claim 16, or the method of claim 17, or the use of claim 18, wherein the disease or disorder is an infection or cancer.

20. The antigen binding polypeptide, nucleic acid, vector, host cell, CAR-T cell or CAR-NK cell for use according to claim 19, the method of claim 19, or the use of claim 19, wherein the cancer is acute myeloid leukaemia, breast cancer, endometrial carcinoma, ovarian cancer, hepatocellular cancer, Non-small cell lung cancer, colorectal cancer, colon cancer soft tissue sarcoma, glioblastoma, astrocytoma, melanoma, or mesothelioma; orwherein the infection is caused by HIV1, HIV2, Mycobacterium tuberculosis (Mtb), Epstein Barr virus (EBV), Human Papillomavirus (HPV), or SARS-CoV-2.

21. The antigen binding polypeptide, nucleic acid, vector, host cell, CAR-T cell or CAR-NK cell for use according to claim 19, the method of claim 19, or the use of claim 19, wherein the cancer is characterised by increased WT1 expression, optionally wherein the cancer is characterised by expression and / or presentation of SEQ ID NO: 13.

22. A method of monitoring treatment efficacy or disease status in a subject diagnosed with cancer or an infection, comprising:i. providing a biological sample obtained from the subject;ii. determining the tumour volume, viral or bacterial load, or level of binding of one or more antigen binding polypeptides of the invention to HLA-E:peptide complex bearing cells in the sample obtained from the subject before treatment, or at intervals between treatments, or at time intervals in the absence of treatment;iii. determining that the treatment is effective, or that the disease status is improving, if the tumour volume, viral or bacterial load, or level of binding of one or more antigen binding polypeptides of the invention to HLA-E:peptide bearing cells, is reduced after treatment or between treatment intervals or at time intervals in the absence of treatment.

23. The antigen binding polypeptide of any of claims 1-10, wherein the antigen binding polypeptide is a bispecific or trispecific antibody, wherein in addition to binding a HLA-E:peptide complex, also binds CD3 and / or CD28.

Images