WO2017203264A1

WO2017203264A1 - Hla targeting

Info

Publication number: WO2017203264A1
Application number: PCT/GB2017/051483
Authority: WO
Inventors: Philip Michael Savage
Original assignee: Philip Michael Savage
Priority date: 2016-05-27
Filing date: 2017-05-25
Publication date: 2017-11-30
Also published as: GB201609380D0

Abstract

The invention relates to a complex comprising a) a polypeptide comprising an effector domain, wherein said effector domain comprises a HLA class I polypeptide, or a fragment comprising a T-cell binding portion thereof; and an attachment means for selectively attaching the polypeptide to a target cell; and b) an antigenic peptide bound or attached to said HLA class I polypeptide or fragment, characterised in that said HLA class I polypeptide or fragment thereof comprises a mutation or mutations conferring an enhanced binding affinity of said HLA class I polypeptide or fragment thereof for a CD8 molecule. The invention also relates to uses, compositions and medical uses.

Description

HLA TARGETING

BACKGROUND TO THE INVENTION The invention relates to materials and methods useful for inducing immune responses against target cells. In particular, the invention is concerned with recruiting existing cytotoxic T-cells to direct an immune response against a target cell. The invention uses modified HLA molecules as part of a larger complex to achieve these beneficial effects. There is a large diversity of HLA types. Even common HLA types such as HLA-A2 are only carried by approximately 25% of patients. This is low, and is therefore

problematic when trying to induce a response restricted to even common HLA types such as HLA-A2. Only 1 - 2% of T-cells typically react with a presented peptide. This is a low proportion and is therefore problematic.

HLA molecules comprise three domains - ai, a2 and 03. The ai and 02 domains are involved in the interaction with the T-cell receptor. The 03 domain is involved in the interaction with the CD8 molecule on the T-cell surface. Prior art approaches have been focussed on the 01/02 part of the HLA molecule, together with the peptide presented by the 01/02 part of the HLA molecule. This is entirely logical since the current understanding is that T-cells are targeted via their TCR (T-cell Receptor), which recognises the presented peptide in the context of the ai/ 02 domains of the HLA molecule presenting that peptide.

Wooldridge et al. 2010 (Journal of Immunology, Volume 184, Pages 3357 to 3366) disclose that MHC Class 1 molecules with superenhanced CD8 binding properties bypass the requirement for cognate TCR recognition and nonspecifically activate CTLs. This document is focussed purely on stimulation of T-cells. There is no disclosure of killing of any cancer cells. There is no teaching of any therapeutic molecule. There is no teaching of any targeting of the engineered HLAs to a target cell. There is no delivery system to take the engineered HLA molecules to a target cell. Thus, there is no disclosure of the multipart molecules and complexes which are the subject of the invention. Wooldridge et al. 2007 (European Journal of Immunology Volume 37, Pages 1323 to 1333) discloses enhanced immunogenicity of CTL antigens through mutation of the CD8 binding MHC Class 1 invariant region. In particular, a Q115E substitution in the o2 domain of HLA-A*020i is described. This Q115E substitution is disclosed to enhance CD8 binding by approximately 50% without altering TCR/PMHCi

interactions. The authors disclose that the enhanced CD8 binding of the mutated HLA- A2 molecule results in improved antigen recognition at the cell surface. It is further disclosed that the incremental increase in HLA-A2/CD8 affinity enhances CTL priming, and induces better zeta chain phosphorylation. It is disclosed that the enhanced binding does not affect T-cells' specificity. This document is focussed on antigen mediated interactions. There is no disclosure of killing of cancer cells. There is no teaching of any therapeutic molecule. There is no teaching of any targeting of the engineered HLA molecules to a target cell. Thus, there is no disclosure of the multipart molecules and complexes which are the subject of the invention.

The present invention seeks to overcome problem(s) associated with the prior art. SUMMARY OF THE INVENTION In stark contrast to the prior art, the present invention focusses instead on the 03 domain of the HLA molecule. This 03 domain can interact directly with CD8 on the T- cell. The inventors have realised that by engineering the HLA molecule, in particular engineering the 03 domain of that HLA molecule, a high affinity interaction between the 03 domain of HLA and CD8 of the T-cell can be induced. This interaction can be made so strong/ of such high affinity so that almost all T-cells can be potentially recruited via the CD8-HLA 03 domain interaction. This breaks the prior art idea of recruitment via the presented peptide in the context of the ai/ 02 domains of HLA. In effect, the invention "overrides" the conventional recruitment via presentation of peptides and instead demonstrates direct recruitment by direct interaction of the 03 domain of HLA with the CD8 on the T-cell surface.

The invention is based upon these surprising findings. In more detail, the invention presents molecules and complexes of molecules featuring engineered 03 HLA sequences so as to achieve the direct recruitment of CD8⁺ T-cells by interaction of the CD8 molecule with the 03 domain of the HLA molecule. The engineered HLA molecule is targeted to the target cell such as a cancer cell. In this way, CD8⁺ T-cells can be directly recruited to the killing of the target cell such as a cancer cell via the engineered 03 HLA domain.

Thus, in one aspect the invention provides a complex comprising

a) a polypeptide comprising

an effector domain, wherein said effector domain comprises a HLA class I polypeptide, or a fragment comprising a T-cell binding portion thereof; and

an attachment means for selectively attaching the polypeptide to a target cell; and b) an antigenic peptide bound or attached to said HLA class I polypeptide or fragment,

characterised in that said HLA class I polypeptide or fragment thereof comprises a mutation or mutations conferring an enhanced binding affinity of said HLA class I polypeptide or fragment thereof for a CD8 molecule. Suitably said antigenic peptide is arranged in the peptide binding groove of said HLA class I polypeptide or fragment. Suitably said antigenic peptide is arranged to be presented by said HLA class I molecule or fragment thereof. Suitably said antigenic peptide is arranged to be presented for T cell recognition by said HLA class I molecule or fragment thereof.

Suitably the fragment of the HLA class I polypeptide comprises at least the T-cell binding portion of the HLA class I polypeptide. Suitably the fragment of the HLA class I polypeptide comprises at least the alpha 2 and alpha 3 domains of the HLA class I polypeptide, suitably at least the alpha 3 domain of the HLA class I polypeptide.

As used herein, the term "enhanced", such as in the context of "enhanced binding affinity of said HLA class I polypeptide or fragment thereof for a CD8 molecule", means enhanced or increased with reference to the wild-type HLA class I polypeptide or fragment thereof. Suitably the wild-type HLA class I polypeptide used for comparison is the same type (allele) as the HLA class I polypeptide, or a fragment comprising a T-cell binding portion thereof. In other words, enhanced means when comparing the equivalent HLA polypeptide, or equivalent fragment thereof, having the wild type sequence, the mutated sequence possesses a greater or improved affinity for CD8. Suitably the CD8 is cognate for the HLA polypeptide or fragment thereof, for example suitably the CD8 and HLA are from the same species.

Wild type HLA-A2 and CD8 have a Kd of ΐ3θμΜ. The invention uses HLA class I polypeptide(s) having Kd for CD8 lower than wild-type, suitably lower than ΐ3θμΜ, suitably lower than ιοομΜ, suitably about 85μΜ, suitably 85μΜ, suitably even lower than 85 μΜ.

Suitably said HLA class I polypeptide or fragment has a Kd for CD8 of less than ΐ3θμΜ. Suitably said HLA class I polypeptide or fragment has a Kd for CD8 of about 85μΜ.

Suitably said HLA class I polypeptide or fragment comprises amino acid sequence corresponding to at least the T-cell binding portion of SEQ ID NO:i. Most suitably said HLA class I polypeptide or fragment comprises the full length amino acid sequence of SEQ ID NO:i. Most suitably said HLA class I polypeptide does not comprise a fragment, but rather comprises the full length amino acid sequence of SEQ ID NO:i. Suitably said HLA class I polypeptide or fragment comprises amino acid sequence corresponding to SEQ ID NO:i except for the alpha 3 domain.

Suitably said HLA class I polypeptide or fragment comprises amino acid sequence of the alpha 1 and alpha 2 domains corresponding to those of SEQ ID NO:i.

Suitably said HLA class I polypeptide or fragment comprises amino acid sequence having at least 80% sequence identity to the corresponding residues of SEQ ID NO:i (i.e. suitably the residues other than the alpha 3 domain); more suitably at least 90% sequence identity, more suitably at least 95% sequence identity, more suitably at least 98% sequence identity, most suitably 100% sequence identity.

The differences between the alpha domain of human A2 and murine KB are shown in alignment below. The differences are all of substitution.

Suitably said HLA class I polypeptide or fragment comprises an alpha 3 domain having the amino acid sequence of SEQ ID NO:3.

Suitably said HLA class I polypeptide or fragment comprises the amino acid sequence of SEQ ID NO: 2. Suitably said HLA class I polypeptide or fragment further comprises a substitution relative to the sequence of SEQ ID NO: 1 in the alpha 2 domain of said HLA polypeptide or fragment, wherein said substitution comprises Q115E.

Suitably said HLA class I polypeptide or fragment and said antigenic peptide are present as a single polypeptide. Suitably said HLA class I polypeptide or fragment and said antigenic peptide are present as a single chain trimer. Suitably said single chain trimer has the amino acid sequence of SEQ ID NO: 5. Suitably said HLA class I polypeptide or fragment comprises a HLA-A2 polypeptide.

Suitably said HLA-A2 polypeptide or fragment comprises an HLA-A*20i polypeptide.

Suitably the attachment means comprises an antibody or antigen binding fragment thereof or an aptamer.

In one embodiment suitably the attachment means comprises:

(a) a linking polypeptide with specific affinity for a molecule on the surface of the target cell

(b) a coupling system for coupling the linking polypeptide to the HLA class I molecule or fragment thereof,

wherein the coupling system comprises:

a first small molecule joined to the linking polypeptide; and

a second small molecule joined to the HLA class I molecule

wherein interaction of the small molecules forms a stable bridge between the linking polypeptide and the HLA class I molecule.

Suitably said target molecule comprises a cell surface antigen present on the surface of said target cell.

Suitably said target molecule comprises a tumour associated antigen (TAA).

Suitably said CD8 comprises human CD8. In one aspect, the invention relates to a complex as described above, further comprising a CD8 molecule bound thereto.

Suitably said CD8 molecule is attached to the surface of a CD8+ T cell. In one aspect, the invention relates to a cell comprising a complex as described above bound to a target molecule on said cell. In one aspect, the invention relates to a method of targeting of a cytotoxic T cell to a target cell, said method comprising contacting said target cell with a complex as described above in the presence of said cytotoxic T cell, said cytotoxic T cell comprising CD8 at the cell surface.

In one aspect, the invention relates to a method of treating a subject comprising administering to said subject a complex as described above in the presence of cytotoxic T cell(s) of said subject, said cytotoxic T cell(s) comprising CD8 at the cell surface. Suitably said subject has a tumour and said target molecule comprises a tumour associated antigen cognate to said tumour.

In one aspect, the invention relates to use of a HLA class I polypeptide, or a fragment comprising a T-cell binding portion thereof, said HLA class I polypeptide or fragment comprising a mutation conferring an enhanced binding affinity of said HLA class I polypeptide or fragment for a CD8 molecule, in targeting of a cytotoxic T cell to a target cell.

In one aspect, the invention relates to a method for producing or enhancing an immunological response against a target cell, comprising the step of attaching a complex as described above to the target cell.

In one aspect, the invention relates to use of a complex as described above in the preparation of a medicament for induction of a cytotoxic T cell response against a cell recognised by the attachment means.

In one aspect, the invention relates to a pharmaceutical composition comprising

(i) a complex as described above, and

(ii) a pharmaceutically acceptable excipient or carrier.

The pharmaceutical composition may be administered using any suitable means known in the art, such as injection.

In one aspect, the invention relates to a kit comprising one or more pharmaceutical compositions as described above and written instructions for the administration of said composition(s) to a subject. In one aspect, the invention relates to a complex as described above, or a polypeptide as defined above, for treatment of cancer.

Suitably the HLA molecule of the invention is delivered to the target cell via a targeting moiety or attachment means. The targeting moiety or attachment means may be an antibody, or an antigen binding fragment thereof such as a Fab, scFv, or other such molecule which recognises an antigen borne by the target cell such as a cancer cell or tumour cell. Alternatively, the HLA molecule of the invention may be delivered via gene transfer or gene therapy type approach in which a nucleic acid encoding the HLA molecule of the invention is transfected, transduced or otherwise delivered to the target cell.

COMBINATION THERAPIES

The invention may usefully be combined with immuno-stimulatory antibodies such as ipilimumab, PDi inhibitor, or any other known agent capable of uprating or enhancing the immune response. The targeting moiety of the molecule of the invention may comprise high affinity T-cell receptor, or may comprise one half of a receptor-ligand pair (i.e. a receptor or a ligand) or may comprise an antibody or fragment thereof directed at a tumour antigen.

HLA MOLECULES

A key advance made by the invention is that by changing the HLA structure, the peptide presentation by the HLA molecule may be rendered no longer important, or of reduced importance, to recruitment of CD8+ T-cells. Recruitment of CD8⁺T-cells via the non-peptide presenting part of the HLA molecule is itself new and key to this advance.

It is a benefit of the invention that it can be used to increase the number of T-cells hitting the target. By HLA we mean human leukocyte antigen, (i.e. human major histocompatibility complex (MHC)) molecules. There are three classes - class I, class II and class III; the invention is concerned with HLA class I molecules. There are numerous loci/alleles - the invention may be applied to alternate HLA class I molecules provided they have a domain corresponding to the alpha 3 domain which is taught as being important herein.

Suitably the invention is applied to HLA- A.

More suitably the invention is appled to HLA-A2.

Most suitably the invention is applied to HLA-A2 based on or derived from SEQ ID NO: 1.

Suitably the HLA comprises a mutation enhancing the interaction with CD8; suitably the interaction with CD8 is enhanced by approximately 35-65%, suitably 50%.

Suitably the HLA has a Kd for the interaction with CD8 of approximately 85μΜ (for comparison, the Kd of interaction of a normal HLA-A2 molecule with CDS is approximately 130UM.)

Suitably the molecule of the invention is not a naturally occurring human HLA molecule.

Suitably the HLA comprises a human 03 domain having high affinity binding to CD8. Suitably the 03 HLA domain of a human HLA molecule is swapped for, or mutated to, the sequence of a mouse 03 domain.

Suitably the HLA comprises a mouse 03 domain. This has the advantage of having high affinity binding to human CD8. With reference to the sequence of the wild type human HLA-A2 molecule of SEQ ID NO: 1, the alpha 1, alpha 2 and alpha 3 domains are defined as in UniProt accession number P01892 (human HLA class I histocompatibility antigen, A-2 alpha chain). In more detail, the domains are as follows (numbering includes the signal sequence):

Note the numbering used in some prior art such as Wooldridge et al refers to the amino acid number according to the final protein starting at 1 (after the signal sequence is removed). To convert a Wooldridge amino acid sequence number to an absolute number based on the Uniprot sequence it is necessary to add 24.

It should be noted that the amino acid positions referred to are for the final protein.

The invention is focussed on an engineered HLA molecule, in particular with an engineered 03 domain. Suitably the HLA molecule used in the invention comprises human amino acid sequence, together with murine amino acid sequence of the 03 domain.

It will be noted that the human and murine 03 domains are very similar. Construction of a HLA molecule according to the invention may be done by recombinantly replacing the human 03 domain with a murine 03 domain. Alternatively, point mutations may be made to a human 03 domain in order to adjust its sequence to conform to the murine 03 domain sequence. Whichever route is used, suitably the HLA molecule of the invention comprises 03 domain sequence which conforms to the mouse 03 domain sequence. An exemplary mouse 03 domain sequence is provided as SEQ ID NO: 3.

It should be noted that the 03 domain of HLA is a conserved sequence. In other words, the 03 domain is shared across different HLA types. Therefore, the invention relates to HLA molecules having murine 03 domains, wherein the type or allele of the HLA is determined by operator choice and wherein the 03 domain is conformed to the murine 03 domain sequence. In one embodiment, the HLA molecule of the invention comprises human HLA-A2 sequence together with the murine 03 domain sequence. An exemplary sequence (complete A2/KB hybrid sequence - (i.e. human HLA-A*20i with particular mutation(s) in the alpha 3 domain) is given below (The leader sequence is shown underlined (only 21 amino acids in this version)):

Most suitably the sequence of the HLA molecule useful in the invention corresponds to the HLA molecule disclosed in Wooldridge et al. 2010.

It should be noted that the designation "HLA-A2/Kb" refers to the mouse equivalent of human HLA-A2. Human HLA-A2 and the mouse version of this (mouse H2-Kb, usually referred to as "A2/Kb") are exemplary molecules which have been used to illustrate the invention. The invention is not intended to be limited only to these HLA types/all eles. In particular, the invention is focussed on engineered 03 domains, especially 03 domains caused to conform to the murine 03 domain amino acid sequence. Since the 03 domain is conserved across different HLA molecules, so the invention can be applied across different HLA molecules. In order to do this, the skilled worker simply "transplants" or otherwise engineers the sequence of the HLA molecule of interest to conform to the murine sequence of the 03 domain.

'Transferring' or 'transplanting' mutations onto an alternate HLA molecule of interest can be accomplished by site directed mutagenesis of a nucleotide sequence encoding the HLA 'backbone' or parent sequence. This technique is well known in the art.

Essentially the backbone HLA sequence is selected and aligned with the exemplary

HLA sequence(s) herein, and the selected mutations are transferred to (i.e. made in) the corresponding/homologous positions.

This may be done by cloning/ligation/PCR based recombinant nucleic acid techniques, or may be done by de novo synthesis and translation of a nucleic acid encoding the

HLA molecule of interest, or synthesis of the polypeptide itself, or any other technique known in the art. Suitably the HLA is produced by recombinant means such as by translation of a nucleic acid encoding said HLA using well known commercially available expression systems such as bacterial or other expression host systems e.g. the pET plasmid + expression system which is widely available e.g. from Agilent Technologies 5301 Stevens Creek Blvd Santa Clara, CA 95051 United States. Suitably the expression system is a eukaryotic system such as a rabbit reticulocyte translation system, or expression in eukaryotic cells such as Sf9 insect cells, or using a Cos cell expression system or other similar technique.

If necessary, the polypeptide/complex of the invention may be purified, for example by including a 6His tag and purifying by Ni-NTA chromatography or similar.

Suitably the complex of the invention maybe assembled in vitro or in vivo.

Below we present an alignment of the human and mouse (HLA-A2 (SEQ ID NO: 6) and A2/Kb (SEQ ID NO: 7)) sequences in which the differences between the human and mouse sequences, including between the human 03 domain and the mouse 03 domain, can be seen

As can be seen, this alignment highlights the amino acid positions which differ between mouse and human HLA 03 domain sequences.

Most suitably the complex/polypeptide of the invention will need the intact (but engineered) full size HLA class I molecule. By 'intact' is meant full-length. The alpha 3 domain is the area for engineering i.e. changes from the wild- type human sequence.

In more detail, we present an alignment between the amino acid sequence of the standard wild type alpha 3 domain (dark shaded) and the preferred mouse alpha 3 domain.

The amino acid numbering is based on the full size mature protein without the leader sequence. This is the same approach used in the art e.g. Wooldridge et al (see above). The Alpha 3 domain starts at amino acid number 183 (upper / boxed) is A2 - SEQ ID NO: 8 ; lower (unboxed) is A2/KB - SEQ ID NO: 9):

In principle, the invention requires a HLA molecule having a 03 domain with enhanced affinity for CD8 compared to the wild-type human HLA molecule. This may be conveniently achieved in one embodiment by swapping the human 03 domain for the murine 03 domain, which replacement results in an enhanced affinity of the HLA molecule for CD8 when compared to human wild-type HLA molecule affinity for CD8. However, it may be that the skilled operator wishes to use a different mutation or set of mutations in order to achieve this property. By reference to the above alignment showing differences between the human and murine 03 domain sequences, the skilled worker may select different mutations or combinations of mutations and test them for their effect on affinity of the resulting HLA molecule for CD8. Measuring the affinity of two molecules, such as the affinity of HLA for CD8, is a straightforward matter which is well known in the art.

Therefore, a subset of the above mutations may be used in the HLA molecule of the invention. Moreover, a superset and/or alternative mutations maybe used in order to achieve the enhanced affinity of the HLA for CD8.

Most suitably, conservative substitutions are chosen relative to the mouse sequence. In other words, if the change from the human sequence to the mouse sequence is considered, then suitably an alternative amino acid which might be substituted is one which is conservative to the amino acid found in the mouse sequence. For example, if the mouse sequence has a positively charged amino acid then suitably a different positively charged amino acid is used at that position.

Conservative substitutions may be made for example according to the table below: Amino acids in the same block in the second column and preferably in the same line in the third column may be substituted for each other:

Suitably amino acids which are common to the human and mouse 03 domains (i.e. conserved between the human and mouse 03 domains) are not substituted. However, if it is desired to substitute amino acids at those positions, then suitably conservative substitutions are made.

In all cases, it is a straightforward matter for the skilled operator to test the resulting mutated HLA molecule for the property of enhanced interaction with CD8 (suitably human CD8) to decide whether or not that mutated HLA molecule is useful in the invention. Such tests are well known in the art, such as measuring the Kd of the mutated HLA for human CD8 and comparing the value to the Kd of the wild type HLA for human CD8. Lower Kd value means enhanced interaction.

It is an advantage of the invention that every T cell in the patient which recognises HLA (irrespective of the allele) will be targeted by the chimeric molecule/ complex of the invention. In this way, approximately all T cells in the subject may be targeted.

It is an advantage that the invention provides a genuine "one for all" reagent.

The invention has the advantage of providing a universal reagent which can be used in subjects regardless of their HLA type.

By CD8 we mean human CD8 (cluster of differentiation 8). An exemplary sequence of the CD8 alpha chain is UniProt accession number P01732. UniProt (Universal Protein Resource) is a comprehensive catalogue of information on proteins ('UniProt: a hub for protein information' Nucleic Acids Res. 43: D204-D212 (2015).). Suitably the UniProt release at the date of filing is relied upon. For the avoidance of doubt, UniProt Release 20i5_n is relied upon.

In more detail, the UniProt consortium European Bioinformatics Institute (EBI), SIB Swiss Institute of Bioinformatics and Protein Information Resource (PIR)'s UniProt Knowledgebase (UniProtKB) Release 20i5_i2, (9-Dec-20i5) is relied upon.

REFERENCE SEQUENCES

The polypeptide components of the complexes of the invention are based on HLA polypeptide sequences such as the exemplary HLA-A2 sequence. In particular, suitably amino acid addresses given in the application correspond to the numbering of the wild type human HLA-A2 of SEQ ID NO:i. In some contexts, the amino acid numbering may be based on the full size mature protein without the leader sequence - this is the same approach as used in some of the prior art e.g. Wooldridge et al (ibid.). Where truncated or extended forms of HLA are used as polypeptides in complexes of the invention (e.g. where a 6his tag or antigenic peptide is added or where a section of the polypeptide is deleted) then the amino acid numbering should be treated as

corresponding to the equivalent section of the full length HLA reference sequence and not as an 'absolute' or rigidly inflexible numeric address. By way of explanation, if the description mentions a substitution of Q115, this means amino acid 115 of the HLA reference sequence of SEQ ID NO: 1. If the polypeptide used is truncated by deletion of the first 10 amino acids, the address given will still be Q115 (rather than e.g. Q105) - this will be easily understood by the skilled reader to refer to the amino acid of the corresponding context with reference to the full length HLA sequence of SEQ ID NO:i, as is conventional in the art.

Clearly there are elements of the HLA wild type sequence which we teach are important to mutate, such as by substitution, to achieve certain advantages. However, there are also numerous residues which may or may not be mutated depending on operator choice. Typically it would be expected that if the skilled operator had a concern whether or not a particular residue could be mutated or not, they could make the mutation and then test the resulting polypeptide to ensure that the desired property, such as enhanced affinity for CD8, was retained in the mutated version. It will be understood that the invention may equally make use of fragment(s) such as epitopes or segments, variant(s) or mutant(s) of the target molecule bound by the attachment means. Suitably any such fragment(s), variant(s) or mutant(s) have at least 80% sequence identity to the reference sequences along the whole length of said fragment(s), variant(s) or mutant(s), suitably 90%, suitably 95%, suitably 98% sequence identity along the whole length of said fragment(s), variant(s) or mutant(s).

Sequence identity may be calculated using any suitable technique known in the art. For example, a suitable computer program for carrying out such an alignment is the GCG Wisconsin Bestfit package (University of Wisconsin, U.S.A.; Devereux et al., 1984, Nucleic Acids Research 12:387). Examples of other software than can perform sequence comparisons include, but are not limited to, the BLAST package (see Ausubel et al., 1999 ibid - Chapter 18), FASTA (Atschul et al., 1990, J. Mol. Biol., 403-410) and the GENEWORKS suite of comparison tools. Suitably default settings are used for any gap penalties or other such values required for the calculation.

ANTIGENIC PEPTIDE

HLA class I molecules are cell surface molecules which possess a peptide binding groove

exposed on the external surface of the cell, which groove is arranged under normal circumstances to bind a peptide derived from the interior of the cell. When a recognition receptor on a cytotoxic T cell binds to an HLA class I molecule on the surface of a scanned cell, the recognition receptor is enabled to contact the peptide binding groove of the HLA class I molecule and interact with any peptide contained therein. If this peptide matches the specificity of the recognition receptor, the T cell is said to recognise the scanned cell, and may consequently trigger an immunological response against said scanned cell. It is an advantage of the invention that this conventional 'scanning-recognition' approach is circumvented as T-cells are recruited regardless of their antigenic specificity by virtue of the interaction between the engineered HLA part of the complex described and the CD8 on the T cell being recruited. However, it is important to note that HLA molecules described herein still require a peptide in their binding groove for structural and/or functional stability of the molecule. A HLA molecule having an unoccupied peptide binding groove can be unstable or can "fall apart". Therefore, suitably the complex of the invention comprises antigenic peptide arranged in the peptide binding groove of the HLA molecule. However, it is important to note that the identity of this peptide does not compromise the universal application of the invention. From an understanding of classical immunology, it would be expected that the antigenic peptide presented in the peptide binding groove would determine the specificity of the T cell which is then recruited. This is how ordinary immune interactions operate. However, the invention advantageously drives recruitment of T cells regardless of the antigenic specificity of those T cells, this advantageous effect being achieved by the engineering of the 03 domain of the HLA molecule according to the invention. Thus, the choice of the antigenic peptide situated in the peptide binding groove of the HLA molecule is matter for operator choice. The only requirement of the invention is that a peptide is located in this groove in order to provide a stable and functional HLA molecule. The precise identity of that peptide is not a factor important to the invention, because the technical benefits which the invention delivers address the problem of antigenic specificity and render it unimportant. Thus, the importance of having the peptide binding groove occupied is confined to the structural/functional stability of the HLA molecule and is only minimally, or not at all, related to the specificity of the recruited T cells.

Thus, as can be clearly appreciated, it is an enormous advantage delivered by the invention that the targeting system overcomes specificity of both the antigenic peptide and/or the HLA allele present in the recipient.

In contrast, prior art approaches have to be carefully tailored to each individual patient. For example, using the wrong HLA type in a patient might result in no T cells being targeted. For example, use of an antigenic peptide to which the patient had no preexisting response in the art might have led to no T cells being recruited. Prior art techniques depend on T cells capable of recognising common antigenic peptides which most patients have previously been exposed to and therefore most patients have previously acquired an immune response against. In contrast, the present invention circumvents both these requirements by directly recruiting T cells on the basis of interaction between the engineered 03 domain of HLA according to the invention and their cognate CD8 molecules present on the T cells being recruited.

Suitably the antigenic peptide ('recognition peptide') may comprise, or may consist of, a tumour specific peptide, a viral peptide, a bacterial peptide, or a parasitic peptide. More suitably the antigenic peptide comprises a peptide selected from the group consisting of an influenza virus peptide, a measles virus peptide, an Epstein-Barr virus (EBV) peptide, a peptide comprising the RAKFFQLL (SEQ ID NO: 10) epitope of the lytic protein BZLFi, a Cytomegalovirus (CMV) peptide and a tetanus toxoid peptide.

Suitably the antigenic peptide ('recognition peptide') has good stability.

Most suitably the antigenic peptide ('recognition peptide') comprises or consists of a viral peptide, most suitably an influenza peptide or a CMV peptide.

When necessary, the antigenic peptide ('recognition peptide') comprises the

appropriate HLA-matched peptide.

Commonly used HLA-A2 antigenic peptides include:

CMV-PP65 (aa495-503): NLVPMVATV (SEQ ID NO: 11)

MART-i:26-35(27L) analog: ELAGIGILTV (SEQ ID NO: 12)

Influenza (Lys-Gly-Ileu-Leu-Gly-Phe-Val-Phe-Thr-Leu-Thr-Val) (SEQ ID NO: 13) The HLA and antigenic peptide are arranged such that the antigenic peptide is located in the antigen presenting groove of the HLA molecule, as is normal and well

understood. The association may be by ordinary binding (e.g. involving hydrogen bonding, van-der-waals forces etc) or may be by covalent attachment. Suitably the recognition peptide may be attached to the HLA class I molecule or fragment thereof in accordance with the method described in Garboczi (PNAS 89, 1992, ΡΡ3429-3433)·

Suitably the recognition peptide may be attached to the HLA class I molecule by encoding them both in a single open reading frame and thereby fusing the two polypeptides (thereby attaching the recognition peptide to the HLA class I molecule).

Suitably the antigenic peptide (recognition peptide) may be attached to the HLA class I molecule by producing them as single chain trimers. These molecules have (e.g.) the antigenic peptide-Beta2M-HLA molecule all in one long polypeptide and are much more stable (Hansen TH, Connolly JM, Gould KG, Fremont DHet al., 2010, Basic and translational applications of engineered MHC class I proteins, TRENDS IN

IMMUNOLOGY, Vol: 31, Pages: 363-369, ISSN: 1471-4906). This publication is specifically incorporated herein by reference for the teachings of attachment/single chain trimer production. The publication exemplifies wild type HLA molecules, but the process is the same for the engineered HLA molecules of the invention. Suitably the HLA class I polypeptide and the antigenic peptide are provided as a single chain trimer, or as part of a single chain trimer.

An example of a single chain trimer which has the HLA/Kb components is below. The A2/Kb single chain trimer sequence is shown below (This links the antigenic peptide, B2M and HLA alpha 1, alpha 2 and Kb together in one long polypeptide. Kb is the relevant mouse alpha 3.):

ATTACHMENT MEANS The invention provides a route to target pathogenic cells such as a tumour cell or any diseased or foreign cell the presence of which is undesired in a patient, such as a cancer cell, leukaemia cell, a cell infected with the HIV virus or with any other pathogen or parasite such as a microbe (e.g. bacterium or fungus) or virus, or a cell responsible for detrimental activity in auto-immune disease, tumour cells or cancer cells. Thus suitably the invention also relates to the treatment of such diseases including bacterial or viral or fungal infection or auto-immune disease or cancer or leukaemia or similar diseases as outlined. Most suitably the invention is applied to tumour cells or cancer cells. Most suitably the invention relates to treatment of tumours or cancers. This targeting is achieved through the attachment means. The invention is not concerned with targeting antigen presenting cells such as cells comprising a T cell receptor. Suitably the attachment means has affinity for a target molecule other than the T cell receptor. Suitably the attachment means does not bind to the T cell receptor (TCR). Suitably the target molecule is not the T cell receptor (TCR). HLA molecules can interact with the TCR. The attachment means suitably comprises a polypeptide with high specific affinity for a target cell specific molecule on the surface of the target cell. By "target cell specific molecule" herein is meant any molecule that is characteristically expressed or over- expressed on the surface of the target cell. By way of example, in cancer cells said "target cell specific molecule" could include a tumour associated antigen (see below). Suitably, the linking polypeptide will comprise an antibody, suitably a monoclonal antibody, or antigen binding fragment or derivative thereof, raised against and/ or capable or specifically binding said target cell specific molecule. Of course antibodies may be raised against peptides or epitope strings or using other well known strategies such as library selection or anti-idiotype approaches; equally other non-antibody specific binders may be used as the attachment means such as for example aptamers - in all cases the most important property for the attachment means is specific binding to the target molecule.

Suitably the target cell does not comprise T cell receptor(s). Suitably the target cell is not an antigen presenting cell. Suitably the target cell is not a T cell. Suitably the target cell is not a CD8+ T cell. In one embodiment the attachment means may specifically bind a target molecule on the target cell which target molecule may have been previously introduced. For example the target cell might be contacted with a recognition reagent; said recognition reagent might comprise a monoclonal antibody or aptamer capable of specifically binding a molecule on said target cell, said monoclonal antibody or aptamer being linked to a coupling system. Said linking polypeptide may comprise an antibody raised against a target cell specific molecule and a coupling system for coupling said antibody to said HLA class I molecule or fragment thereof. The coupling system may comprise a two- or three-step chain of well-characterised paired small molecules, joined to the antibody and the HLA class I molecule so as to form a stable bridge between the two. Examples of paired small molecules which might be used in this connection include (but are not limited to) biotin and avidin/streptavidin (Moro, 1997 Cancer Res, 57, 1922-1928; Altman et al, Science 274, 1996, 94-96), and calmodulin and calmodulin binding peptides (Neri, 1996, J. Invest. Dermatol. 107, 164-170). Alternatively, said linking polypeptide may comprise an antibody raised against a target cell specific molecule, which antibody is adapted to be attached directly to said HLA class I molecule or fragment thereof. Alternatively, the connection may be made using the DOCK-AND-LOCK® (or DNL®) method, which can be used for making a considerable number of bioactive molecules of increasing complexity. DNL® utilizes the natural interaction between two proteins, cyclic AMP-dependent protein kinase, or PKA, and A-kinase anchoring proteins, or AKAPs. The region that is involved in such interaction for PKA is called the

dimerization and docking domain, or DDD, which always appears in pairs. Its binding partner in AKAPs is the anchoring domain, or AD. When mixed together, DDD and AD will bind with each other spontaneously to form a binary complex, a process termed docking. Once "docked," certain amino acid residues incorporated into DDD and AD will react with each other to "lock" them into a stably-tethered structure. Since DDD always appears in pairs, any component that is linked to DDD will have two copies present in the final products. The outcome of DNL® is the exclusive generation of a stable complex, in a quantitative manner that retains the full biological activities of its individual components. DNL is available from Immunomedics, Inc., 300 The American Road, Morris Plains, NJ 07950, USA. Rossi EA, Goldenberg DM, Chang CH. Complex and defined biostructures with the dock-and-lock method. Trends Pharmacol Sci. 2012 Sep;33(9):474-8i.

As applied in the invention, the dock and lock technology allows two protein components to be joined securely via a non-covalent bond using a non-immunogenic protein interaction. This has the advantage that an attachment means such as an antibody derived molecule can be joined simply in vitro or in vivo to the engineered HLA class I without the need to make a fusion protein or use a potentially

immunogenic coupling system (such as biotin and streptavidin). In more detail, we refer to Rossi et al 2014 (mAbs volume 6 pages 381-391 Ά new class of bispecific antibodies to redirect T cells for cancer immunotherapy'). This document is hereby incorporated herein by reference, specifically for the teachings regarding the 'dock and lock' coupling; in particular in Figure 1 of Rossi et al the HLA molecule of the invention takes the place of the scFv of Rossi et al. The DDD2 of Rossi et al joins the two antibody portions and the single AD2 peptide attaches the monomelic HLA. Thus this system may be easily applied to the molecules and complexes of the invention as described.

Suitably the attachment means and the effector domain are heterologous. In other words, suitably the attachment means and the effector domain are not part of the same naturally occurring molecule. Suitably the polypeptide of the invention is a non- naturally occurring molecule. Suitably the effector domain and the attachment means are fused to form the polypeptide of the invention.

In one embodiment, when the HLA polypeptide is, or is based on, human HLA polypeptide sequence, suitably the attachment means comprises non-human sequence such as mouse monoclonal antibody sequence, or artificial sequence such as an aptamer sequence, or may comprise a humanised antibody sequence (or fragment of any thereof). Typically humanised antibody sequences (or fragments thereof) comprise human framework or constant regions together with heterologous CDRs. Suitably the CDRs may be from a non-human organism such as a mouse. Thus, it will be clear to the skilled reader that the invention does not embrace naturally occurring polypeptides such as wild-type polypeptides occurring in nature. Thus it should be clear that the invention comprises subject matter eligible for patent protection. When the invention is applied to the treatment of cancer, suitably the cancer cells comprise said target molecule. Cancer is a broad term referring to a range of neoplastic diseases which can arise from different tissues and therefore can comprise different cell types. When operating the invention the skilled worker will pay attention to the particular cancer or tumour type being treated. The skilled person will select an attachment means for the polypeptide of the invention which recognises a target molecule on the surface of the cancer cells being treated. This selection process is well within the ability of the person skilled in the art since cancer antigens/tumour antigens and their associations with particular cancer/tumour types are widely published and well known in the art. For example, we refer to Buonaguro et al. 2011 (Clinical and Vaccine Immunology, Volume 18, pages 23 to 24) which highlights a range of tumour associated antigens - the table below is taken from Buonaguro at al. 2011:

Other examples of antigens useful to target (i.e. using an attachment means which specifically binds to such antigen(s)) in the invention include carcinoembryoic antigen, placental alkaline phosphatase, polymorphic epithelial mucin, human chorionic gonadotrophin, CD20, prostate specific antigen, ca-125, HMW-MAA and others.

Other examples of tumour antigens useful in the invention include: heteroclinic melanoma epitope (Melan-A) (ELAGIGILTV) (SEQ ID NO: 12); SP14 (EBV denved EBNA-i epitope) (HPVGEADYFE) (SEQ ID NO: 14).

Other examples of antigens useful in the invention include: CD20 , CD19, CD45, CD52, PDL-i, EGFR, CDna.

In case any further guidance is needed, we refer to the examples section below. Suitably attachment means (e.g. targeting systems) may include one part of a ligand/receptor pair (e.g. to target a particular receptor, the polypeptide of the invention may be coupled to (such as fused with) a ligand for that receptor), ImmTACS (Immune mobilising monoclonal TCRs Against Cancer - available from Immunocore Limited, 101 Park Drive, Milton Park, Abingdon, Oxon, OX14 4RY, UK; e.g. the polypeptide of the invention may be coupled to, or may comprise, an ImmTAC as the attachment means) as well as antibody based attachment means.

Suitably the complex or polypeptide of the invention may be delivered by direct injection, or maybe produced in vivo in the subject e.g. by gene delivery of a nucleic acid sequence encoding the polypeptide of interest. Gene delivery systems for mammalian use are well known, such as using retroviral vectors to deliver the coding sequence which is then expressed in situ in the subject. Suitable vectors such as retroviral vectors are well known in the art, and may be obtained for example from Addgene (LGC Standards, Queens Road, Teddington, Middlesex, TW11 oLY, UK).

COMBINATION MUTATIONS

Suitably the HLA further comprises an 02 domain having high affinity binding to CD8. Suitably the HLA further comprises a human a2 domain having high affinity binding to CD8.

Suitably the HLA comprises a mutation such as a substitution which narrows the distance between the Q115 of the HLA molecule and the R4 of the CD8 molecule. The distance between these two components is typically approximately 3.18A when both molecules are wild-type; suitably the HLA useful in the invention comprises a mutation such as a substitution leading to a reduction in this distance so that the distance is less than 3.18A, suitably approximately 2.5A, suitably about 2.56A. Suitably the HLA comprises a substitution at the position corresponding to Q115; suitably the position corresponding to Q.1.15 is E (Q1.15E). Suitably the substitution is as disclosed in Wooldridge et al. 2007 (European Journal of Immunology Volume 37, Pages 1323 to 1333)·

FORMULATION AND ADMINISTRATION

Suitably the pharmaceutical composition comprises an effective amount of the complex and/ or polypeptide of the invention.

The administration of said pharmaceutical composition may be by way of oral, sublingual, transdermal or parenteral administration. Said effective amount of the pharmaceutical composition will depend on factors such as the nature and severity of the disorder being treated and on the weight, age and condition of the patient. Dosage is suitably determined by the physician.

For oral or parenteral administration, suitably the pharmaceutical composition is administered in the form of a unit-dose composition, such as a unit dose oral or parenteral composition.

Such compositions are prepared by admixture and are suitably adapted for oral or parenteral administration, and as such maybe in the form of tablets, capsules, oral preparations, powders, granules, lozenges, reconstitutable powders, injectable and liquid infusible solutions or suspensions or suppositories.

Tablets and capsules for oral administration are usually presented in a unit dose, and contain conventional excipients such as binding agents, fillers, diluents, tabletting agents, lubricants, disintegrants, colourants, flavourings, and wetting agents. The tablets may be coated according to well known methods in the art.

Suitable fillers for use include cellulose, mannitol, lactose and other similar agents.

Suitable disintegrants include starch, polyvinylpyrrolidone and starch derivatives such as sodium starch glycolate. Suitable lubricants include, for example, magnesium stearate. Suitable pharmaceutically acceptable wetting agents include sodium lauryl sulphate.

These solid oral compositions may be prepared by conventional methods of blending, filling or tabletting. Repeated blending operations may be used to distribute the active agent throughout those compositions employing large quantities of fillers. Such operations are, of course, conventional in the art.

Oral liquid preparations may be in the form of, for example, aqueous or oily

suspensions, solutions, emulsions, syrups, or elixirs, or may be presented as a dry product for reconstitution with water or other suitable vehicle before use. Such liquid preparations may contain conventional additives such as suspending agents, for example sorbitol, syrup, methyl cellulose, gelatin, hydroxyethylcellulose,

carboxymethyl cellulose, aluminium stearate gel or hydrogenated edible fats, emulsifying agents, for example lecithin, sorbitan monooleate, or acacia; non-aqueous vehicles (which may include edible oils), for example, almond oil, fractionated coconut oil, oily esters such as esters of glycerine, propylene glycol, or ethyl alcohol;

preservatives, for example methyl or propyl p-hydroxybenzoate or sorbic acid, and if desired conventional flavouring or colouring agents.

Oral formulations also include conventional sustained release formulations, such as tablets or granules having an enteric coating. For parenteral administration, fluid unit dose forms may be prepared comprising a sterile vehicle. The components of the composition, depending on the vehicle and the concentration, can be either suspended or dissolved. Parenteral solutions are normally prepared by dissolving the components of the composition in a vehicle and filter sterilising before filling into a suitable vial or ampoule and sealing. Advantageously, adjuvants such as a local anaesthetic, preservatives and buffering agents are also dissolved in the vehicle. To enhance the stability, the composition may be frozen after filling into the vial and the water removed under vacuum. Parenteral suspensions are prepared in substantially the same manner except that the compound may be suspended in the vehicle instead of being dissolved and sterilised by exposure to ethylene oxide before suspending in the sterile vehicle. Advantageously, a surfactant or wetting agent may be included in the composition to facilitate uniform distribution of the complex of the invention.

The compositions will suitably be accompanied by written or printed directions for use in the treatment concerned.

Most suitably the complex of the invention is formulated for injection and is administered by direct injection e.g. to the site of the cancer or tumour being targeted, or intravenously (i.v.).

FURTHER APPLICATIONS

The invention is useful to greatly widen the pool of T cells that can see the HLA molecule by making the engineered changes taught herein, which has the benefit of making it a much better way of getting T cells to kill tumour cells in vivo.

Further particular and preferred aspects are set out in the accompanying independent and dependent claims. Features of the dependent claims may be combined with features of the independent claims as appropriate, and in combinations other than those explicitly set out in the claims.

Where an apparatus feature is described as being operable to provide a function, it will be appreciated that this includes an apparatus feature which provides that function or which is adapted or configured to provide that function. BRIEF DESCRIPTION OF THE DRAWINGS

Embodiments of the present invention will now be described further, with reference to the accompanying drawings, in which:

Figure l shows assessment of CD8 and HLA-A2 allele expression by sorted CD8⁺ T-cell populations used in the flow cytometric killing assays. CD8⁺ T-cells were sorted from PBMCs of HLA-A2⁺ and HLA-A2^" healthy donors by magnetic bead separation, before being subjected to a round of in vitro expansion.

Figure 2 shows representative FATAL cytotoxicity assay flow cytometry histograms showing the lysis of the HLA-A2Kb⁺ CiR target cells (eF670⁺ peak) relative to an internal HLA-A2wt⁺ CiR control population (eF670^~ peak), following a 12 hour incubation period. Specific lysis of HLA-A2Kb-expressing targets by HLA-A2⁺ and HLA-A2- CD8⁺ T-cells (middle panels) and HLA-A2 and non-HLA-A2 restricted CD8⁺ T-cell clones MelCs and SB14 (bottom panels), respectively are shown at the optimal E:T of 5:1. Specific lysis was assessed against a reference assay containing no CD8⁺ T- cell effectors (top panel).

Figure 3 shows graphical representation of the specific lysis of HLA-A2Kb-expressing CiR targets in the presence of either sorted CD8⁺ T-cells (A) or CD8⁺ T-cell clones MelC5 and SB14 (B ), at a full range of E:Ts (0.5:1 to 5:1). Specific lysis was assessed against a reference assay containing no CD8⁺ T-cells. Error bars indicate SD from the mean of 2 replicates.

EXAMPLES We demonstrate killing of cancer cells bearing (transfected) A2/Kb by both HLA-A2 and HLA-non-A2 specific T cells.

A2Kb vs. A2wt killing:

Reagents:

1) CD8⁺ T-cells, obtained through microbead magnetic sorting of HLA-A2⁺ and HLA- A2- healthy donor PBMCs.

2) CD8⁺ T-cell clones: MelC5, an HLA-A*20i-restricted clone specific for the heteroclinitc melanoma epitope, Melan-A (ELAGIGILTV) (SEQ ID NO: 12). SB14, an HLA-B*35o8-restricted clone specific for the EBV-derived EBNA-i epitope

(HPVGEADYFE) (SEQ ID NO: 14). Both CD8⁺ T-cell clones and bulk-sorted CD8⁺ T-cells were expanded in the same culture conditions: i.e. with mixed irradiated allogeneic PBMCs sourced from three unrelated donors, in RPMI-1640 media (Life Technologies, Carlsbad, CA, USA) supplemented with antibiotics (Penicillin, 100 U/ml; and Streptomycin, 100 μg/ml), L- glutamine (2 mM), human AB serum (10%; all from Sigma Aldrich St. Louis, MO, USA), interleukin-2 (IL-2, 200 IU/ml, the pharmacy), IL-15 (25 ng/ml, Peprotech, Rocky Hill, NJ, USA), and phytohemagglutinin (PHA, ^g/ml, Fisher Scientific, Loughborough, UK).

Fluorom etric assessm ent of T-lymphocyte antigen-specific lysis (FATAL) cytotoxicity assay:

In order to measure specific lysis by the sorted CD8⁺ T-cells and clones, target HLA- A2Kb⁺ CiR cells were labelled with 5μΜ eF670 dye to allow for flow cytometric separation from the eF670^" HLA-A2wt control cells, which were left unstained. 3x1ο⁴ eF670⁺ HLA-A2kb⁺ CiRs and an equal number of eF670^~ HLA-A2wt⁺ CiRs cells were subsequently combined with specific CD8⁺ T-cells at the following effector to target (E:T) ratios: 0:1, 0.5:1, 1:1, 2:1, 3:1, 4:1 and 5:1. Assays containing no CD8⁺ effector T- cells (E:T of 0:1) were used as a baseline to measure killing. Following a 12 hour incubation period (37°C, 0.5% C0₂) period, the cells were harvested, stained with anti- CD8-BV785 (BD) and the viability dye Aqua (Life Technologies), prior to acquisition on a Fortessa flow cytometer (BD Biosciences). The flow cytometric assessment of lytic activity was based on the quantification of eF670-labelled HLA-A2kb⁺ target cell elimination relative to the internal eF670^" HLA-A2wt⁺ control cells (see Figure 1).

This primarily shows that the differences in killing by the sorted CD8s were not related to differences in the purity of these populations.

Figure 2 shows flow cytometry results demonstrating lysis of target cells relative to a control.

Figure 3 shows graphical representation of specific lysis of cells.

Panel A shows results for pools of T cells. The shaded bars include specificity for HLA- A2. The open bars show killing only via the engineered alpha 3 domain of the HLA according to the invention.

Panel B shows results for T cell clones. The shaded bars relate to A2-T cells; the open bars relate to non-A2 T cells. Although illustrative embodiments of the invention have been disclosed in detail herein, with reference to the accompanying drawings, it is understood that the invention is not limited to those precise embodiments and that various changes and modifications can be effected therein by one skilled in the art without departing from the scope of the invention as defined by the appended claims and their equivalents.

Claims

1. A complex comprising

a) a polypeptide comprising

characterised in that said HLA class I polypeptide or fragment thereof comprises a mutation or mutations conferring an enhanced binding affinity of said HLA class I polypeptide or fragment thereof for a CD8 molecule.

2. A complex according to claim 1 wherein said HLA class I polypeptide or fragment has a Kd for CD8 of less than 130μΜ.

3. A complex according to claim 2 wherein said HLA class I polypeptide or fragment has a Kd for CD8 of about 85μΜ. 4. A complex according to any of claims 1 to 3 wherein said HLA class I polypeptide or fragment comprises an alpha 3 domain having the amino acid sequence of SEQ ID NO:3.

5. A complex according to any of claims 1 to 4 wherein said HLA class I polypeptide or fragment comprises the amino acid sequence of SEQ ID NO:2.

6. A complex according to any preceding claim wherein said HLA class I polypeptide or fragment further comprises a substitution relative to the sequence of SEQ ID NO: 1 in the alpha 2 domain of said HLA polypeptide or fragment, wherein said substitution comprises Q115E.

7. A complex according to any preceding claim wherein said HLA class I polypeptide or fragment and said antigenic peptide are present as a single polypeptide. 8. A complex according to claim 7 wherein said HLA class I polypeptide or fragment and said antigenic peptide are present as a single chain trimer.

9. A complex according to claim 8 wherein single chain trimer has the amino acid sequence of SEQ ID NO: 5.

10. A complex according to any preceding claim wherein said HLA class I polypeptide or fragment comprises a HLA-A2 polypeptide.

11. A complex according to any preceding claim wherein the attachment means comprises an antibody or antigen binding fragment thereof or an aptamer. 12. A complex according to any preceding claim wherein the attachment means comprises:

wherein the coupling system comprises:

a first small molecule joined to the linking polypeptide; and

a second small molecule joined to the HLA class I molecule

13. A complex according to any preceding claim wherein said target molecule comprises a cell surface antigen present on the surface of said target cell. 14. A complex according to any preceding claim wherein said target molecule comprises a tumour associated antigen (TAA).

15. A complex according to any of claims 1 to 14 wherein said CD8 comprises human CD8.

16. A complex according to any of claims 1 to 14, further comprising a CD8 molecule bound thereto.

17. A complex according to claim 16 wherein said CD8 molecule is attached to the surface of a CD 8+ T cell.

18. A cell comprising a complex according to any of claims 1 to 15 bound to a target molecule on said cell.

19. A method of targeting of a cytotoxic T cell to a target cell, said method comprising contacting said target cell with a complex according to any of claims 1 to 15 in the presence of said cytotoxic T cell, said cytotoxic T cell comprising CD8 at the cell surface.

20. A method of treating a subject comprising administering to said subject a complex according to any of claims 1 to 15 in the presence of cytotoxic T cell(s) of said subject, said cytotoxic T cell(s) comprising CD8 at the cell surface.

21. A method according to claim 19 wherein said subject has a tumour and wherein said target molecule comprises a tumour associated antigen cognate to said tumour.

22. Use of a HLA class I polypeptide, or a fragment comprising a T-cell binding portion thereof, said HLA class I polypeptide or fragment comprising a mutation conferring an enhanced binding affinity of said HLA class I polypeptide or fragment for a CD8 molecule, in targeting of a cytotoxic T cell to a target cell.

23. A method for producing or enhancing an immunological response against a target cell, comprising the step of attaching a complex according to any of claims 1 to 15 to the target cell. 24. Use of a complex according to any of claims 1 to 15 in the preparation of a medicament for induction of a cytotoxic T cell response against a cell recognised by the attachment means.

A pharmaceutical composition comprising

a complex according to any of claims 1 to 15, and

a pharmaceutically acceptable excipient or carrier.

26. A kit comprising one or more pharmaceutical compositions according to claim 25 and written instructions for the administration of said composition(s) to a subject.

27. A complex according to any of claims 1 to 15, or a polypeptide as defined in any of claims 1 to 15, for treatment of cancer.