WO2009051555A2

WO2009051555A2 - Modified mhc class i binding peptides

Info

Publication number: WO2009051555A2
Application number: PCT/SE2008/051164
Authority: WO
Inventors: Daniel Badia-Martinez; Adnane Achchour; Thorbald Van Hall; Marianne Van Stipdonk; Rienk Offringa
Original assignee: Akademisch Ziekenhuis Leiden
Priority date: 2007-10-15
Filing date: 2008-10-15
Publication date: 2009-04-23
Also published as: WO2009051555A3; GB0720118D0

Abstract

The present invention relates to a method of producing a polypeptide or peptide, or nucleic acid encoding the polypeptide or peptide, wherein the polypeptide or peptide comprises a modified peptide sequence that has increased binding affinity for an MHC Class I molecule and/or generates a higher CD8+ T cell immune response, the method comprising: identifying a peptide sequence ('parent peptide') of a CD8+ T cell epitope of an antigen, wherein the peptide sequence is capable of being displayed on an MHC Class I molecule; and producing a polypeptide or peptide comprising a modified peptide sequence of the CD8+ T cell epitope or a nucleic acid molecule encoding said polypeptide or said peptide, wherein the modified peptide sequence has a mutation compared with the parent peptide, the mutation providing a cyclic or pseudo-cyclic amino acid residue at position 3 of the modified peptide sequence, defined such that the residues are sequentially numbered with position 1 being the first residue displayed within the peptide- binding groove of the MHC Class I molecule, and wherein the modified peptide has greater binding affinity for the MHC Class I molecule and/or generates a higher CD8+ T cell immune response, compared with the parent peptide.

Description

Modified MHC Class I Binding Peptides

The present invention relates to peptides modified to enhance their immunogenicity, particularly binding affinity to Class I major histocompatibility complex (MHC-I) , and methods for modifying peptides to enhance their immunogenicity, particularly MHC-I binding affinity. Such modified peptides may enhance CD8⁺ T lymphocyte response useful in the treatment of various pathologies including infections and cancer.

MHC-I molecules are plasma membrane proteins expressed by virtually all mammalian cells. They bind peptides derived by intracellular processing of viral, bacterial or endogenous proteins. Their main function is to transport and present these peptides to CD8+ T lymphocytes. The interaction of peptide-MHC protein complexes on the surface of professional antigen presenting cells (APC) with antigen receptors on the surface of T-cells causes T-cell activation and stimulation of an immune response.

MHC-I molecules play a crucial role in immune surveillance by selectively binding to intracellular peptide antigens and presenting them at the cell surface to CD8⁺ T lymphocytes, including cytotoxic T lymphocytes (CTLs) , through their T cell receptor (TCR) . To prevent autoimmunity, high-affinity T cells specific for MHC molecules presenting 'self peptides are normally eliminated in the thymus

(Central tolerance) . Therefore, in healthy individuals the majority of the T cell population is thought to be composed of those T cells recognizing high-affinity 'non-self peptides. Tight and long- lasting binding of T cells to cells presenting MHC molecules with non-self peptides elicits a cytotoxic response that results in the killing of the infected cells and prevents further spread of the disease (Kourilsky and Fazilleau, Int. Rev. Immunol., Oct; 20 (5) : 575- 91, 2001; Yewdell and Bennink, Curr. Opin. Immunol., Feb; 13 (1) : 13-8 , 2001; Davis et al, Nature Immunology, Mar; 4 (3) : 217-24, 2003; Pardoll, Ann. Rev. Immunol., Vol. 21: 807-839, 2003; Van der Burg et al, J. Immunology, vol.156, 33083, 1996) . The body is left with T cells capable of binding to self-epitopes at low affinity, and these are down-regulated by peripheral tolerance mechanisms including ^ignorance' in order to prevent autoimmunity (Sprent and Kishimoto, Immunol. Rev., JuI; 185 : 126-35, 2002) . Vaccination in the presence of strong adjuvants may help to overcome tolerance by stimulating Λself -specific T cells.

The identification of tumour-associated antigens (TAAs) recognised by CTLs at the molecular level opens new possibilities for the design of well-defined therapeutic cancer vaccines. Eradication of tumours is often associated with a robust CTL response to TAAs. But since many TAAs are self-proteins or closely related to self-proteins, they tend to be poorly immunogenic. Moreover, many TAA-derived peptides are not strong binders to MHC-I molecules making them poor inducers of CD8 T-cell immunity.

The first two domains (αl and α2 ) of the MHC Class I proteins form an extremely polymorphic peptide-binding groove with closed ends, approximately 30 A long and 12 A wide in the middle. Each α-helical domain contributes half of the eight-stranded β-sheet floor of the groove and an α-helix wall. The most variable residues point into this groove and up from the tops of both helices, conferring unique peptide- and TCR-binding properties to each MHC molecule. The majority of these variable residues are located in the central portion of the cleft, whereas clusters of highly conserved residues that hold on to the peptide termini occur at both ends of the groove. Some of these conserved, primarily aromatic, residues block the ends of the binding groove, while others create a conserved network of hydrogen-bonding ligands and water molecules whereby the termini of the peptides are rigidly fixed.

The relative arrangement of the polymorphic residues that line the peptide-binding cleft creates pockets that may accommodate predominant amino acid side-chains of the peptide, thereby anchoring the peptide on to the Class I molecule. In the analysis of the structure of the first determined MHC Class I structure, HLA-A2, six major pockets were identified and labelled A through F (Bjorkman et al, Oct 8-14;329 (6139) :506-12, Nature, 1987 and Bjorkman et al, Nature, Oct 8-14 ; 329 (6139) : 512-8 , 1987) . Residues that comprise each region are approximately the same in all the defined MHC-I structures, so that each section is located at the same place in the binding cleft in any MHC Class I molecule. All the structures have two distinct and fairly conserved pockets A and F at each end of the peptide-binding groove. Pockets B through E have distinct sizes and character in different allelic variants of MHC Class I molecules, thereby imposing different sequence constraints and requirements on the peptide that is bound. A consequence of this is that MHC-I binding peptides contain allele-specific sequence motifs, defined by the position and the identity of at least a couple of anchoring residues, one of which is the C-terminus . In addition, there are secondary pockets that can enhance and further tune the affinity of a particular peptide. Owing to the large variation of binding constraints of MHC-I molecules, each individual presents an unique pool of peptides to T-lymphocytes . Therefore, improving binding affinity of peptides through replacement of such anchor residues is a complicated and tailor-based approach.

gplOO is an enzyme involved in pigment synthesis that is expressed by the majority of malignant melanoma cells, as well as by normal melanocytes. gplOO is a member of a family of "self" (i.e., unmutated) , melanoma/melanocyte differentiation antigens that are widely expressed by melanoma cells. Overwijk et al, (J. Exp. Med. JuI 20;188 (2) :277-86, 1998) demonstrated that recombinant vaccinia virus (rW) encoding the mouse homologue of gplOO (mgplOO) was non- immunogenic in mice, whereas mice vaccinated with rVV encoding human gplOO (hgplOO) developed a specific CD8+ T cell immune response, which was found to be cross-reactive with mgplOO. The basis for the greater immunogenicity of hgplOO was attributed to differences within the MHC-I T cell epitope. Overwijk et al (J. Exp. Med. JuI 20; 188 (2) : 277-86, 1998 and J. Exp. Med. Aug 18; 198 (4) : 569-80, 2003) describe sequence differences in the three N-terminal amino acids between peptide fragments of mouse gpl00₂5-33 (mgpl00₂5-33) and human gpl00₂5-33 (hgpl00₂5-33) resulting in a 2- log increase in stabilisation of the mouse MHC-I molecules H-2D^b on

TAP-deficient RMA/S cells and a 3-log increase in IFNγ release by recognising T-cells. The two peptides are identical in 6 positions of 9, but are different in positions 1, 2 and 3. (Conventionally the residues of an MHC-displayed peptide are numbered sequentially with position 1 being the first residue localised within the MHC-I binding groove.) Peptides that bind to H-2D^b commonly use positions 5 and 9 as main anchor positions (with most often an asparagine at position 5 and a hydrophobic residue at position 9) .

It has been reported that peptides binding to the MHC Class I molecule H-2D^d exploit a four residue peptide binding motif (G2, P3, R/K/H5 and I/L/F9 or 10) (Corr et al . , J. Exp. Med., Dec 1;178 (6) :1877-92, 1993) . The crystal structure of H-2D^d MHC Class I (Achour et al . , Immunity, Aug; 9 (2) : 199-208 , 1998) confirmed that the H-2D^d peptide binding motif comprised these four anchor residues.

Despite extensive investigation into TAAs and therapeutic vaccines for cancer and infectious diseases, the mechanism (s) by which heteroclitic peptides elicit enhanced immunity are unclear. The dogma has been that in order to increase the binding affinity of peptides to MHCI, one has to increase the structural complementarities between the peptides and their cognate MHC-I molecules. Thus the most common procedure for increasing the complementarities between a specific peptide and a specific MHC-I molecule has been to substitute key residues that act as anchor positions with residues that display more appropriate properties. Examples of such successful modifications have been published using both structural and functional approaches (e.g. Chen et al, J. Exp. Med., Apr 18 ; 201 (8 ) : 1243-552005; Webb et al, J. Biol. Chem., May; 279: 23438-23446, 2004; Borbulevych et al, J. Immunology, Apr 15;174 (8) :4812-20, 2005) . The present inventors have determined modifications to peptides that increase their affinity for binding MHC-I molecules and/or that generate an enhanced CD8+ CTL response.

The present inventors have found that one or more modifications may be introduced in a peptide sequence of a CD8+ T cell epitope of an antigen, wherein the modifications enhance binding affinity of the peptide for MHC-I, and which may moreover generate an enhanced CD8+ T cell immune response. Thus, the modifications may be such that the conformation of the modified peptide, when presented, preserves T cell recognition.

Administration of a polypeptide or peptide comprising the modified peptide sequence to an individual may induce a stronger response from CD8⁺ cytotoxic T lymphocytes (CTLs) towards the polypeptide or peptide comprising the modified peptide sequence as compared with the parent peptide. This is of particular advantage where the parent peptide might otherwise not provoke sufficient response from CTLs and/or might otherwise not bind with enough affinity to the cognate MHC Class I molecule. Further, by inducing a stronger response from CTLs, the modified peptide sequence may also indirectly generate a stronger response from other immune cells. Such a response may be directed towards both the modified and the original peptides. Thus, the modified peptide sequence may generate an enhanced immune response in an individual, which may be of therapeutic advantage for the treatment of pathologies such as cancer or infection, including use in vaccines.

Increased binding affinity of a peptide for MHC-I is additionally advantageous for in vitro applications which exploit or require stability of the peptide-MHC-I interaction, such as the use of MHC tetramers e.g. in fluorescence activated cell sorting (FACS) . In a first aspect, the present invention provides a method of producing a polypeptide or peptide, or nucleic acid encoding the polypeptide or peptide, wherein the polypeptide or peptide comprises a modified peptide sequence that has increased binding affinity for an MHC-I molecule and/or generates a higher CD8+ T cell immune response, the method comprising: identifying a peptide sequence ("parent peptide") of a CD8 T cell epitope of an antigen, wherein the peptide sequence is capable of being displayed on an MHC Class I molecule; and producing a polypeptide or peptide comprising a modified peptide sequence of the CD8⁺ T cell epitope or a nucleic acid molecule encoding said polypeptide or said peptide, wherein the modified peptide sequence has a mutation compared with the parent peptide, the mutation providing a cyclic or pseudo-cyclic amino acid residue at position 3 of the modified peptide sequence, defined such that the residues are sequentially numbered with position 1 being the first residue displayed within the peptide- binding groove of the MHC-I, and wherein the modified peptide has greater binding affinity for the MHC-I molecule and/or generates a higher CD8+ T cell immune response, compared with the parent peptide.

The cyclic amino acid residue may, for example, be a proline or an aromatic residue, such as tyrosine.

Most of the peptides that bind to human MHC-I make use of two main anchor residues, localized at peptide position 2 and at the C- terminal section of the peptide that is exposed within the peptide binding cleft. Similarly most of the peptides that bind to mouse MHC-I make use of one anchor position at peptide residue 5 and an additional anchor position at their C-termini. Thus, most of the current and previous studies aiming at increasing the affinity of peptides to their cognate MHC-I molecules have been based on the introduction of modifications that would increase the sterical fit of the peptide in the binding cleft of that precise MHC class I molecule. The finding that modification of position 3 to provide, for example, a proline, an aromatic (such as tyrosine) or a cyclic residue could increase binding affinity for the modified peptide to an MHC-I and enhance the CD8+ T cell immune response generated by the peptide was completely unexpected in view of the dogma described above that in order to increase the binding affinity of a peptide to an MHC-I, one has to increase the structural complementarities between the peptide and its cognate MHC-I. Thus, previous studies have focused on substitutions at main anchor positions, such as peptide residues 2 and 9 (see Chen et al, J. Exp. Med., Apr 18;201(8) , 1243-1255, 2005; and Webb et al, J. Biol. Chem. , May; 279: 23438 - 23446, 2004) . The findings underlying the present invention are, therefore, surprising.

Without wishing to be bound by theoretical considerations, work of the present inventors suggests that the introduced cyclic residue, such as a proline residue, interacts through so-called CH-π interactions (Bhattacharyya and Chakrabarti, J. MoI. Biol., 331, 925- 940, 2003) with the side chain of the MHC-I tyrosine residue Y159, conserved among most known MHC-I molecules, as well as CH-O interactions with the side chain of the MHC-I glutamate residue E9 (Bhattacharyya and Chakrabarti, J. MoI. Biol., 331, 925-940, 2003) . Introduced aromatic cyclic side chains as well as pseudo-cyclic residues at position 3 of the peptide may further stabilize the peptide's interaction with the peptide-binding cleft of the MHC molecule through so-called ring-ring stacking interactions with the side chain of the MHC class I tyrosine residue Y159 (Burley and Petsko, Science, 5 July, pp.23-29, 1985) . This approach of binding improvement should be applicable to all MHC-I independent of specific binding pockets, since the Y159 is conserved among all human, and even mammalian, MHC-I alleles.

In a further aspect the invention provides a method of producing a polypeptide or peptide, or nucleic acid molecule encoding the polypeptide or peptide, wherein the polypeptide or peptide comprises a modified peptide sequence that has increased binding affinity for an MHC Class I molecule and/or generates a higher CD8+ T cell immune response, the method comprising: identifying a peptide sequence ("parent peptide") of a CD8 T cell epitope of an antigen, wherein the peptide sequence is capable of being displayed on an MHC Class I molecule and wherein the peptide sequence has a proline, an aromatic (such as tyrosine) , a cyclic or pseudo-cyclic amino acid at position 3, defined such that the residues are sequentially numbered with position 1 being the first residue displayed within the peptide-binding groove of the MHC Class I molecule; and producing a polypeptide or peptide comprising a modified peptide sequence of the CD8⁺ T cell epitope or an nucleic acid molecule encoding said polypeptide or encoding said peptide, wherein the modified peptide sequence has a mutation compared with the parent peptide, the mutation providing a glycine or a hydrophobic amino acid residue at position 2 or any appropriate modification at position 2 according to the requirements for the specific MHC class I allele (see for example Table 3) of the modified peptide sequence and wherein the modified peptide has greater binding affinity for the MHC Class I molecule and/or generates a higher CD8+ T cell immune response, compared with the parent peptide.

Thus, in certain cases the mutation may comprise providing a proline at position 3 of the peptide, providing a combination of a glycine at position 2 and a proline at position 3 of the peptide, or (for example when the MHC-I molecule comprises HLA-B53) providing a proline at position 2 and a proline at position 3. Similarly, for example when the MHC-I molecule comprises HLA-B27, the mutation may comprise providing arginine or a lysine at position 2 and a proline at position 3.

In some cases the binding motifs of peptides that bind to some specific MHC-I alleles make use of anchor residues other than at position 2 of the peptide. In such cases, the peptide may be adapted to the specific allele requirements through the introduction of a proline, an aromatic (such as tyrosine), a cyclic or a pseudo-cyclic residue at position 3 of the presented peptide in accordance with the present invention. Furthermore, the peptide may be modified at other positions so as to provide optimal complementarities between the peptide and the MHC-I allele. Modifications and combinations of modifications may be selected with reference to Table 3.

The finding that modification of position 2 of the peptide sequence to provide a hydrophobic residue, particularly a glycine residue, when a proline residue, an aromatic (such a tyrosine) or a cyclic residue is present, or has been introduced, at position 3 can increase binding affinity for an MHC-I and may generate a higher CD8+ T cell immune response also runs against the prevailing approach of previous attempts to increase peptide immunogenicity of MHC Class I- binding peptides. As explained above, previous studies have focused on substitutions at the main anchor positions, primarily at peptide residues 2, 5 and 9 in mouse and human, respectively.

Methods of the invention may further comprise determining binding affinity of the modified peptide for an MHC-I and comparing the binding affinity with the binding affinity of the parent peptide for the MHC-I molecule. The method may comprise determining that the modified peptide has a higher binding affinity than the parent peptide for the MHC-I molecule.

Alternatively, or additionally, methods of the invention may further comprise determining ability of the modified peptide to generate a CD8+ T cell immune response, and comparing the level of CD8+ T cell immune response generated by the modified peptide with the level of CD8+ T cell immune response generated by the parent peptide. The method may comprise determining that the modified peptide generates a higher CD8+ T cell immune response than the parent peptide. Examples of assays for determining binding affinity and for determining CD8+ T cell immune responses are described elsewhere herein .

Methods of the invention may be used to determine effects of certain mutations in peptides, in terms of affinity for binding MHC-I and/or for generating CD8+ T cell immune responses.

Thus, methods of the invention may be used to produce a polypeptide or peptide, or nucleic acid encoding the polypeptide or peptide, wherein the modified polypeptide or peptide comprises a modified peptide sequence, the method comprising identifying a parent peptide and producing a polypeptide or peptide comprising a modified peptide sequence, or an isolated nucleic acid encoding said polypeptide or peptide, wherein the modified peptide sequence has a mutation compared with the parent peptide as discussed elsewhere herein. For example mutation may provide a proline residue, an aromatic residue (such as tyrosine) , a cyclic or pseudo-cyclic residue at position 3 and/or may provide glycine, or a hydrophobic residue, or any required residue in order to complement the B-pocket, at position 2. The method may comprise determining whether the modified peptide has greater binding affinity for MHC-I and/or generates a higher CD8+ T cell immune response, as compared with the parent peptide.

The effect of mutation on binding affinity and/or immune response may vary according to the nature of the mutation, the identity of the MHC-I, and the combination of these two variables. Thus, methods of the invention may be used to determine which of the mutations described herein provide the greatest increase in binding affinity and/or immune response, in relation to one or more MHC-I.

In a further aspect the invention provides modified polypeptides and/or peptides, and/or nucleic acid molecules encoding modified polypeptides and/or peptides, obtained by methods of the invention. In a further aspect the invention provides D/L polypeptides and/or peptides obtained by methods of the invention.

In a further aspect the invention provides polypeptides and/or peptides with reduced peptide-bonds obtained by methods of the invention .

In a further aspect the invention provides poly-N-acylated polypeptides and/or peptides obtained by methods of the invention.

In a further aspect the invention provides beta amino acid- substituted polypeptides and/or peptides obtained by methods of the invention .

In a further aspect the invention provides partially modified retro- inverso pseudo-polypeptides and/or pseudo-peptides obtained by methods of the invention.

In a further aspect the invention provides an isolated polypeptide or peptide comprising a mutant peptide sequence which comprises a CD8⁺ T cell epitope capable of being displayed on an MHC-I, wherein said mutant peptide sequence has the sequence of a naturally occurring CD8⁺ T cell epitope of an antigen, except that said mutant peptide sequence has a proline, an aromatic residue (such as tyrosine) , cyclic or pseudo-cyclic amino acid residue at position 3 which is not present at position 3 of the sequence of said naturally occurring CD8⁺ T cell epitope, and optionally has a hydrophobic amino acid residue at specific anchor positions such as position 2 or 9 which is not present at position 2 or 9 of the sequence of said naturally occurring CD8+ T cell epitope, and wherein the position numbering is defined such that the residues are sequentially numbered with position 1 being the first residue displayed within the peptide- binding groove of the MHC Class I molecule. In a further aspect the invention provides an isolated polypeptide or peptide comprising a mutant peptide sequence which comprises a CD8⁺ T cell epitope capable of being displayed on an MHC-I, wherein said mutant peptide sequence has the sequence of a naturally occurring CD8⁺ T cell epitope of an antigen, except that said mutant peptide sequence has a proline, an aromatic residue (such as tyrosine) , cyclic or pseudo-cyclic amino acid residue at position 3 which is not present at position 3 of the sequence of said naturally occurring CD8⁺ T cell epitope, and optionally has a appropriate amino acid residue at specific anchor positions such as an arginine at position 2 of a peptide that binds to HLA-B27 (Lopez de Castro J. A., Eur. J. Immunology 35, 336-340 (2005), which is not present at position 2 of the sequence of said naturally occurring CD8+ T cell epitope, or a lysine at position 9 of the peptide that binds to HLA-AIl (Li et al, J. Immunology 172, 6157-6184 (2004)) , and wherein the position numbering is defined such that the residues are sequentially numbered with position 1 being the first residue displayed within the peptide- binding groove of the MHC Class I molecule.

In further aspects the invention provides: an isolated nucleic acid encoding a polypeptide or peptide according to the invention; a vector comprising a nucleic acid according the invention; a host cell comprising a vector according to the invention; and a pharmaceutical composition comprising a polypeptide or peptide of the invention or a nucleic acid of the invention or a vector of the invention and a pharmaceutically acceptable excipient.

In further aspects the invention provides a polypeptide or peptide according to the invention or an isolated nucleic acid encoding a polypeptide or peptide according to the invention or a vector comprising a nucleic acid according the invention or a pharmaceutical composition according to the invention for use in a method of therapy of the human or animal body, particularly for use in generating an immune response in a subject or preventing an infection or a cancer, such as melanoma, in a subject. In further aspects the invention provides use of a polypeptide or peptide according to the invention or an isolated nucleic acid according to the invention or a vector comprising a nucleic acid according the invention or a pharmaceutical composition according to the invention in the manufacture of a medicament for generating an immune response in a subject or preventing an infection or a cancer, such as melanoma, in a subject.

In further aspects the invention provides methods of treating an individual comprising administering a polypeptide or peptide according to the invention or an isolated nucleic acid according to the invention or a vector comprising a nucleic acid according the invention or a pharmaceutical composition according to the invention in order to generate an immune response in the subject or to prevent or treat an infection or a cancer, such as melanoma. The individual may be a human or a non-human animal, e.g. mammal or bird.

In a further aspect the invention provides use of a polypeptide or peptide according to the invention or an isolated nucleic acid according to the invention or a vector comprising a nucleic acid according the invention or a pharmaceutical composition according to the invention in the manufacture of a medicament for generating immunological tolerance in a subject. The subject may have a neurological disease, such as Parkinson's disease or Alzheimer's disease .

In a further aspect the invention provides use of a polypeptide or peptide according to the invention or an isolated nucleic acid according to the invention or a vector comprising a nucleic acid according the invention or a pharmaceutical composition according to the invention in the manufacture of a medicament for preventing or reducing an autoimmune reaction in a subject, wherein the subject has a disease having an autoimmune component. The disease having an autoimmune component may be a disease selected from multiple sclerosis, rheumatoid arthritis, systemic lupus erythematosus, type I diabetes, Guillain-barre syndrome and Crohn's disease.

In a further aspect the invention provides a complex comprising an MHC-I and a peptide according to the invention. The invention may comprise one or more dimerical, tetramerical or oligomerical complexes comprising an MHC-I molecule and a modified peptide according to the invention.

These and further aspects of the invention are described in further detail below and with reference to the accompanying examples.

Description of the Figures

Figure 1 shows the electron density maps of the three peptides, the mouse homologue non-immunogenic mgpl00₂₅-₃₃ (EGS) (A), the intermediate immunogenic human peptide hgplOU₂₅-₃₃ (KVP) (B), and the highly immunogenic heteroclitic peptide modified-gplOO (EGP) (C), all in complex with H-2D^b. The annealed omit 2F_O-F_C electron density map of EGS, KVP and EGP when bound to H-2D^b (upper (A) , middle (B) and lower (C) panels, respectively), contoured at 1.0 σ. The final models of

MHC-peptide complexes are displayed for comparison. The peptides are depicted with their N termini to the left and their C termini to the right. The heavy chain of H-2D^b is omitted for simplicity.

Figure 2A shows superimposed side views of the peptides mgpl00₂5-33

(EGS) (in white) , hgpl00₂₅-₃₃ (KVP) (in dark grey) and the heteroclitic peptide modified-gplOO (EGP) (in grey) , in the crystal structures of their complexes with H-2D^b. The peptides are depicted with their N termini to the left and their C termini to the right. This figure suggests that the conformation of the three peptides is overall similar. A square indicates the region that is enlarged and displayed in Figure 2B. Figure 2B: The side chain of the proline residue at position 3 of the EGP peptide (in grey) has shifted by about 0.8A towards the C-terminal of the peptide binding cleft, allowing for a better interaction with heavy chain residues Y159 and E9.

Figure 3A shows the interaction of the low-affinity peptide mgpl00₂₅-₃₃ (EGS) in complex with H-2D^b, depicting the interaction between residues 1-5 of the peptide (plE-p5N, indicated in bold in the figure) and residues Yl₁ E9, K66, Q70, Q97, S99 and Y159 from the heavy chain of H-2D^b. Distances (in A) between the side chain of the serine residue at position 3 of the peptide (p3S) and surrounding MHC-I residues such as Y159, E9 and S99 are indicated beside the dashed lines. Residues Y45, K66 and Q70, the side chains of which form parts of the binding B-pocket are depicted.

Figure 3B shows the interaction of the intermediate-binding affinity peptide hgplOU₂₅-₃₃ (KVP) with H-2D^b, identifying potential binding interactions with residues Yl₁ Y22, Y45, K66, Q70 and Y159. Distances (in A) between the side chain of the proline residue at position 3 of the peptide (p3S) and surrounding MHC-I residues such as Y159, E9 and S99 are indicated besides the dashed lines. Note that the side chain of the proline residue p3P in the peptide forms CH-π and CH-O interactions with the side chain of the heavy chain residues Y159, conserved amongst most of the known MHC Class I molecules .

Figure 3C shows the interaction of the very high-binding affinity heteroclitic peptide gplOO (EGP) with H-2D^b. The position of the proline residue has shifted by about 0.8A, allowing for a better CH-π interaction with the side chain of residue Y159. Furthermore, the side chain also interacts through CH-O interactions with the side chain of the MHC Class I glutamate residue E9. Hydrogen bond interactions are depicted as dashed lines. Finally, it should be noted that the side chain of proline residue at position 3 of the EGP peptide binds deeper in the cleft of H-2D^b, allowing for more extensive van der Waals interactions with MHC-I residues.

Figure 4 shows RMA-S binding curves for mgpl00₂₅-₃₃ (EGS) and hgpl00₂₅-₃₃ (KVP), as well as peptides modified at position 3, with H-2D^b. The modification of peptide residue p3 from a serine in EGS to a proline in EGP results in a strongly enhanced binding affinity (as indicated by the larger arrow) .

Figure 5 shows A) A comparison of the binding affinity of the EGS, EGP and KVP to two other peptides (ElA and MuLV) , known from previous studies to bind very well (ElA) and not at all (MuLV) , respectively, to H-2D^b. The substitution of p3S to p3P in EGP results in the formation of a peptide that binds much better than ElA. B) A comparative analysis of the impact of different mutations on the capacity of hgplOO peptide and modified variants to bind to H-2D^b. The progressive shortening of the side chain of the residue at position 2 of the peptide (in combination with the introduction of a proline at position 3) increases the binding affinity of the modified peptides to H-2D^b (compare KVP, KAP and KGP, where KVP is an intermediary binder to H-2D^b while KGP shows better binding) . C) Similarly to Figure 8B, the introduction of a glycine at position 2 of the peptide increases the binding affinity of the modified peptide to H-2D^b. The binding affinity increases from a relatively poor binder (EGS) to a very high affinity-binder (EGP) .

Figure 6 shows a comprehensive functional analysis of the impact of the different peptide variants KVA, KVP, KAP and KGP on a) the proliferation, b) Interferon-gamma production and c) expression of CD62L. Each column is summarised for the draining lymph nodes, the non draining lymph nodes and the spleen. Introduction of a proline at position 3 of the peptide, conjugated with the introduction of a glycine at position 2, results in an enhanced proliferation, an enhanced production of interferon gamma and an enhanced expression of the maturity marker CD62L (compare KVA and KGP) .

Figure 7 shows a comprehensive functional analysis of the impact of the different peptide variants EGS, EVS, EVP and EGP on a) the proliferation, b) Interferon-gamma production and c) expression of CD62L. Each column is summarised for the draining lymph nodes, the non draining lymph nodes and the spleen. Introduction of a proline at position 3 of the peptide, conjugated with the introduction of a glycine at position 2, results in an enhanced proliferation, an enhanced production of interferon gamma and an enhanced expression of the maturity marker CD62L (compare EGS and EGP) .

Figure 8 shows a comparative analysis of the capacity of CD8 T cells to bind to tetramers of H-2D^b in complex with the non-immunogenic low binding affinity peptide EGS. Upon immunization with the EGP peptide, the CD8 T cells bind to tetramers of H-2D^b/EGS and produce large amounts of interferon gamma. Conversely, immunization with EGS does not result in any binding to tetramers of H-2D^b/EGS and/or expression of interferon gamma (when compared to control PBS) .

Figure 9 shows a comparison of the binding affinity of the Epstein- Barr Virus-associated peptide BMLF-I (259-267, with sequence GLCTLVAML) and modified variants to HLA-A2 on the surface of T2-A2 cells. GLC is the unmodified BMLF-I peptide. P3 is the BMLF-I peptide modified at position 3 to a proline. Y3 is the BMLF-I peptide modified at position 3 to a tyrosine. G2P3 is the BMLF-I peptide modified at positions 2 and 3 to a glycine and a proline, respectively. Both proline and tyrosine modifications at position p3 of the peptide analogues resulted in a better binding when compared to the wild-type peptide. The combined p2Gp3P modification does not bind at all.

Figure 10 shows a comparison of the binding affinity of the Epstein Barr virus-associated peptide EBNA-4 (416-424, with sequence IVTDFSVIK) and modified variants to HLA-AIl on the surface of T2-A11 cells, using an MHC class I-specific antibody. IVT is the unmodified EBNA-4 peptide. P3 is the EBNA-4 peptide modified at position 3 to a proline. Y3 is the EBNA-4 peptide modified at position 3 to a tyrosine. G2P3 is the EBNA-4 peptide modified at positions 2 and 3 to a glycine and a proline, respectively. The p3P modification results in a better binding affinity to HLA-AIl when compared to the wild-type peptide.

Figure 11a shows the amino acid sequence of human gplOO (SEQ ID NO: 1) . gi I 639590 | gb | AAC60634.1 | gplOO [Homo sapiens]

Figure lib shows the amino acid sequence of mouse gplOO (SEQ ID NO: 2) . gi 131981217 I ref |NP 068682.2| silver [Mus musculus]

Figure 12. The combined p2Gp3P substitution results in higher binding affinity and improved stabilization of MHC class I expression

A) The human orthologue gplOO-derived peptide KVP (KVPRNQWDL) is more immunogenic than the natural mouse epitope EGS (EGSRNQWDL) . Groups of C57BL/6 mice (n=4) were injected with saline, longer versions of the EGS or KVP peptides (AVGALEGSRNQDWLGVPRQL and AVGALKVPRNQDWLGVPRQL, respectively) and frequencies of EGS-directed, IFN-γ-producing CD8+ T cells were measured by flow cytometry. Means and standard deviation of the groups are plotted for one out of three comparable experiments. Statistical analyses were performed using Graph Pad software .

B) KVP binds with higher affinity to H-2Db than EGS as measured by MHC surface stabilization on RMA-S cells. The ElA and MuLV peptides were used as positive and negative controls, respectively. Binding affinity to H-2Db is expressed as an index of mean fluorescence in presence and absence of peptide.

C-E) The peptides EGP, KGP and KGS bind with higher affinity to H-2Db when compared to all the other peptide variants. The comparative data is presented according to the mutated peptide residues; 1, 2 and 3 in C, D and E, respectively. Similar results were obtained in four separate experiments.

F-H) The super-peptides EGP and KGP display a superior capacity at stabilizing H-2Db expression levels on the surface of cells when compared to all other peptide variants. The comparative data is presented according to the mutated peptide residues; 1, 2 and 3 in F, G and H, respectively. Similar results were obtained in four separate experiments .

Figure 13. Immunization with the super-peptide gplOO epitope EGP efficiently elicited endogenous anti-tumor CD8+ T cell responses

A) C57BL/6 mice were immunized with KVP, KGP, EGS or EGP peptides. After 10 days spleens were harvested, stimulated once with the respective peptides in vitro and tested against the immunizing peptide (vaccine peptide) as well as titrated concentrations of the original peptide EGS. IFN-γ production was measured intracellularly and quantified by flow cytometry. Three mice were included in each group. One representative experiment out of three is displayed.

B) Frequencies of EGS-specific CD8+ T-cells were directly monitored from the blood of immunized C57BL/6 mice. Immunizations were performed with 20-mer long peptides comprising KVP, EGS or EGP sequences and topical imiquimod as adjuvants. A total number of 38 mice per group were tested.

* means p<0.05, ** means p<0.001 and student T-test were performed using GraphPad software.

C) Spleens and lymph nodes from peptide-immunized mice from (B) were harvested, stimulated once in vitro with the same peptide and killing capacity was tested in a DNA fragmentation assay against B16 melanoma cells. All statistical analyses were performed using Graph Pad software.

D) In vivo killing assay in which immune mice were injected with two differentially CFSE-labeled target cells in equal numbers. CFSElow- and CFSEhigh-labeled targets were exogenously pulsed with EGS and a control peptide, respectively. Percentage of killing was calculated by the ratio of the two targets in spleens of mice. Groups of four mice were analyzed. Representative mean results and SEM of three mice per group are presented. * means p <0.05, ** means p<0.001 and student T-test were performed using GraphPad software.

E) B16 tumor cells were injected subcutaneously into naive C57BL/6 mice. After eight days, mice were treated with one peptide vaccination and non-stimulated pmel cells. Mice were sacrificed when tumor sizes exceeded above 1 cm3. Kaplan-Meier survival curves of EGP-treated group (n=9) and controls were statistically significant different (log-rank test in GraphPad software) p = 0.0007. The arrow indicates the start of treatment (day 9) .

Figure 14. TCR Vb usage of H-2D^b/EGS-reactive T-cell pool is diverse and comparable for peptide variants

C57BL/6 mice were immunized with peptides KVP, KGP or EGP in combination with the adjuvant imiquimod. After 10 days spleens were harvested, stimulated once with respective nanomeric peptides in vitro and then activated with lμg/ml EGS peptide overnight. Cells were stained for CD8, IFN-γ and with the indicated TCR Vb-specific antibody. IFN-γ-producing CD8⁺ T-cells were gated and expression of TCR Vb segments was determined. The percentages of Vb-positive cells are depicted as the proportion out of the total pool of reactive cells in each mouse, which was set to 100%. Mean and standard error of the mean (six mice per group) are plotted. The EGS-immunized animals yielded too few reactive cells in order to analyse the Vb usage (less then 5%) .

Detailed description of the invention Numbering of peptide residues

Conventionally, residues in amino acid sequences are sequentially numbered from the N to the C terminus. For MHC-I-binding peptides, position 1 in the peptide is defined as the first residue localised within the peptide-binding groove of the MHC Class I molecule. This may be the N-terminal residue of the peptide, with the N-terminus forming hydrogen bond interactions with the hydroxyl groups of heavy chain residues tyrosine 7 and tyrosine 171. In some cases, such as in the case of the peptide KAVYNFATM (SEQ ID NO: 20) , the first residue (lysine residue K) extends out of the peptide-binding cleft of the MHC molecule H-2K^b (Achour et al, Immunity, 17, 757-758, 2002; Velloso et al., J. Immunology, 172, 5504-5511, 2004) . In this case, the second residue, alanine, acts as binding residue number 1, the valine residue acts as residue number 2 and the tyrosine residue acts as peptide residue number 3. Thus residue 1 will always be the first residue displayed within the peptide-binding cleft. If one or more N-terminal residues of the displayed peptide sit outside the cleft, the residue number 1 will be the first position of the peptide that occupies the A-pocket within the cleft.

Numbering of heavy chain residues

Numbering of MHC-I heavy chains follows the commonly used definition that states that the αl-domain is usually initiated by a motif of three residues ^ΛGSH' (Glycine-Serine-Histidine) and that the glycine residue is considered as the amino acid residue number 1 in the heavy chain.

Mutations

In certain aspects the present invention provides a mutation of the peptide sequence of an MHC Class I epitope. The present inventors have established the importance of the interaction between the amino acid residue at position 3 of the peptide sequence and the tyrosine residue 159 on the heavy chain, conserved among most known classical and non-classical MHC Class I molecules. Although performance of aspects of the invention does not require an understanding of the molecular or thermodynamic reasons for the advantages conferred by mutations disclosed herein, and therefore without wishing to be bound by theory, it is noted that a mutation to provide a proline, an aromatic (such as tyrosine), a cyclic amino acid or a synthetic cyclic or pseudo-cyclic amino acid at position 3 of the peptide sequence is believed to cause stronger binding to the MHC Class I Molecule heavy chain through combined CH-π and CH-O interactions with surrounding residues such as Y159 and E9 (Bhattacharyya and Chakrabarti, J. MoI. Biol., 331, 925-940, 2003) or through ring-ring stacking interaction between the aromatic, the cyclic or the pseudo- cyclic natural or synthetic amino acid at position 3 of the peptide sequence and the side chain of the tyrosine residue 159 on the heavy chain (Burley and Petsko, Science, 5 July, pp. 23-29, 1985), conserved among most known classical and non-classical MHC Class I molecules.

Typically the mutation will provide a cyclic residue at position 3, such as proline, tyrosine, phenylalanine, histidine or tryptophan, or a synthetic cyclic amino acid such as a hydroxylated or a methylated histidine (Anderton S., Current Opinion in Immunology, 16, 753-758 (2004) ) . Preferably, the modification will provide a proline or tyrosine at position 3.

Mutation to provide tyrosine at position 3 is preferred in the case of peptides that are restricted to the MHC Class I molecule HLA-A2. Mutation to provide tyrosine at position 3 of such HLA-A2-restricted peptides may increase the binding affinity of the modified peptide for HLA-A2 , when compared to the mutation to provide proline at position 3 of the same peptide. However, it should be noted that mutation to a proline still results in an enhancement of binding affinity when compared to wild-type peptide binding affinity.

Mutation to provide proline at position 3 is preferred in the case of peptides that are restricted to the MHC Class I molecule HLA-AIl. Mutation to provide proline at position 3 may increase the binding affinity of the modified peptide for HLA-AIl, when compared to the mutation to provide tyrosine at position 3 of the same peptide (Figures IOA and lOB) . The modification may provide a "pseudo-cyclic" amino acid (also known as a cyclic-like amino acid) . As used herein "pseudo-cyclic amino acid" means an amino acid which has an unclosed ring structure. Examples of pseudo-cyclic amino acids are modified amino acid residues, such as N-alkylated amino acid residues (e.g. N-propyl alanine, N-ethyl glycine, and derivatives) . The unclosed ring structure may be able to interact with the conserved heavy chain residue Y159 of an MHC Class I molecule in order to increase binding affinity of the MHC Class I binding peptide. For example, the pseudo-cyclic residue may have a hydrophobic interaction with the ring of the side chain of the heavy chain residue Y159, increasing the binding affinity of the modified peptide to the MHC-I molecule.

The cyclic amino acid may be aromatic. Examples of aromatic amino acid residues include tyrosine, tryptophan and phenylalanine.

The mutation to provide a cyclic or pseudo-cyclic amino acid at position 3, as defined further herein, may be an amino acid modification, addition, deletion or substitution, and specifically includes the possibility of an amino acid substitution of one cyclic or pseudo-cyclic residue for a different cyclic or pseudo-cyclic residue. Thus, the parent peptide may comprise a cyclic or pseudo- cyclic residue at position 3, and the modified peptide sequence may comprise a different cyclic or pseudo-cyclic residue. Thus, mutation of a tryptophan residue W to a proline residue P or a tyrosine residue Y at position 3 is an example of a mutation providing a cyclic amino acid residue at position 3.

Further mutations

Optionally, the specified mutations may be the only mutations relative to the parent peptide. In such cases the rest of the peptide sequence is unchanged relative to parent. In other cases, further, unspecified mutations may also be present in the modified peptide as compared with the parent peptide. An example of such additional modifications is the introduction of a glycine residue at position 2 of the peptide. Glycine at position 2 may be of particular advantage in the case of peptides that are restricted to the MHC molecule H-2D^b. The coupling of a glycine at position 2 with a cyclic residue, such as a proline or a tyrosine, at position 3 of the peptide will further increase the binding affinity of the peptide to H-2D^b.

Another example of such modifications is the introduction of a valine, leucine, isoleucine, phenylalanine or any other natural or unnatural hydrophobic residue at position 2 and/or position 9 of the peptide. "Hydrophobic residue" as used herein means an amino acid (naturally occurring or synthetic) that has a non-polar side chain. Examples include: glycine, alanine, valine, leucine, isoleucine, proline, phenylalanine, methionine, tryptophan and cysteine. Preferred hydrophobic residues are: glycine, valine, leucine, isoleucine and phenylalanine. A hydrophobic residue at position 2 and/or 9 may be particularly advantageous in the case of peptides that are restricted to a human MHC Class I molecules such as HLA-A2. A combined hydrophobic residue at position 2 and a lysine residue at position 9 may be particularly advantageous in the case of peptides that are restricted to a human MHC Class I molecules such as HLA-AIl. The introduction of a mutation may be at position 2 and/or position 9 of a peptide that is restricted to HLA-A2 or HLA-AIl, i.e. at both positions 2 and 9, or only at position 2 or only at position 9 of the modified peptide. The provision of a mutation at position 2 of the modified peptide (that interacts with the so-called B-pocket) may provide improved complementarity between the modified peptide and the peptide-binding groove within the so-called B-pocket of the MHC Class I molecule. Similarly, the provision of a mutation at position 9 (that interacts with the so-called F-pocket) , or the provision of a mutation at each of the positions 2 and 9 of the modified peptide may provide improved complementarity between the modified peptide and the peptide-binding groove within the so-called B-pocket of the MHC Class I molecule, and the so-called F-pocket, respectively. Further modifications at other positions that increase complementarity between the residue in question and the residues that form any of the pockets with which it interacts will be apparent to the skilled person .

A number of peptide sequences of CD8⁺ T cell epitopes of antigens have a cyclic residue at position 3 naturally. For example, the human gpl00₂₅-₃₃ epitope sequence KVPRNQDWL (SEQ ID NO: 7) has proline at position 3. In the case of such peptide sequences it may not be necessary to modify the amino acid at position 3 in order to increase binding affinity for an MHC Class I molecule. When a peptide sequence of a CD8⁺ T cell epitope is identified as having a cyclic residue at position 3 the peptide sequence may be modified at position 2 and/or 9, or in any position of the modified peptide that would result in an increased complementarity between the residue in question and the residues that form any of the pockets with which it interacts. Accordingly, in certain embodiments of the invention

(generally when the parent peptide already has a cyclic amino acid at position 3) the modified peptide sequence may comprise a mutation to provide a glycine residue or a natural or unnatural hydrophobic amino acid residue at position 2 of the modified peptide, or any other fitting residue. Preferably, the modification is to provide glycine at position 2 in the case of peptides that are restricted to the MHC molecule H-2D^b. Preferably, the modification is to provide a natural or unnatural hydrophobic amino acid residue (such as valine, leucine, isoleucine or phenylalanine) at position 2 and/or position 9 in the case of peptides that are restricted to HLA-A2. Preferably, the modification is to provide a natural or unnatural hydrophobic amino acid residue (such as valine, leucine, isoleucine or phenylalanine) at position 2 and a lysine residue at position 9 in the case of peptides that are restricted to HLA-AIl.

Also contemplated in the case of a parent peptide having a cyclic or pseudo-cyclic amino acid residue at position 3 is to modify the amino acid at position 3 of the parent peptide to provide a different cyclic or pseudo-cyclic amino acid residue at position 3. For example, where the parent peptide has a proline at position 3, the mutation may be to provide a tyrosine at position 3. It has been found that mutation to provide glycine at position 2 of the modified peptide when a proline, a cyclic or cyclic-like residue is present at or is provided at position 3 dramatically increases the binding affinity of the modified peptide for the MHC Class I molecule H-2D^b. However, the preferred mutation at position 2, or any main or secondary anchor position may differ depending on the intended MHC Class I molecule to be bound by the modified peptide. For example, mutation to provide leucine, isoleucine, phenylalanine or valine at position 2 when a cyclic or cyclic-like residue is present at position 3 may increase the binding affinity of the modified peptide for human MHC Class I molecules such as HLA-A2 or HLA-AIl.

It has been found that mutation of position 2 of the modified peptide to provide glycine, when a proline, a cyclic or pseudo-cyclic residue is present at position 3 leads to enhanced T cell responses as described in further detail below.

Except where the context clearly dictates otherwise, any mutation described herein may be combined with any other mutation described herein .

Antigen

The parent peptide sequence may be a CD8+ T cell epitope from any antigen of any species that is capable of being displayed, even with an extremely low binding affinity, on an MHC Class I molecule which may be from any of a broad array of species. The antigen may be associated with infection or with a type of cancer or other disease, or associated with an allergic response. In all cases, the antigen has an epitope that can bind, even with an extremely low binding affinity, to an MHC Class I molecule. There may, of course, be more than one such epitope present in the antigen. The peptide sequence of any such epitope or epitopes may be modified in accordance with the above methods in order to enhance immunogenicity . Furthermore, the antigen may be a synthetic peptide or polypeptide having an epitope that can bind to an MHC Class I molecule.

Protein databases of examples of antigens having epitopes that are able to bind an MHC Class I molecule are described in Novellino L, Castelli C, Parmiani G. ^ΛA listing of human tumour antigens recognised by T cells: March 2004 update' published in Cancer Immunol Immunother. 2005 Mar; 54 (3) : 187-207. See also, ^ΛMHC ligands and peptide motifs' by H. Rammensee, J. Bachmann and S. Stevanovic; ISBN 3540631259 9783540631255 1570594600 9781570594601 0412133318

9780412133312. Further lists can be found at websites of various institutes or research facilities. At the time of filing at least the following provided such further lists:

European Bioinformatics Institute: http : //www. ebi . ac . uk/ Patabases /protein .html

Immune Epitope Database IEDB, sponsored by NIH, NIAID and HHS: http : / /www . immuneepitope . org/home . do

Table 1 provides a non-exhaustive list of tumour-associated antigens which may comprise parent peptide sequences for modification in accordance with the present invention. Polypeptides or peptides comprising modified peptide sequences of these antigens are useful for treatment of the cancers with which the antigens are associated.

Table 1

Examples of MHC-I-restricted human cancer antigens recognised by CD8⁺ lymphocytes :

A. Melanoma-melanocyte differentiation antigens MART-I (Melan-A) NP_005502 Kawakami, PNAS, 1994 gplOO (pmel-17) AAC60634 Kawakami, PNAS, 1994

Tyrosinase AAD13984, AAK00805, AAG38762, AAB60319,

AAA61244, AAA61242 Britchard, J. Exp. Med., 1993 Tyrosinase related protein-1 AAC154 I Wang, J. Exp. Med., 1995

Tyrosinase related protein-2 CAB93531, CAB93530 Wang,

J. Exp. Med., 1996

Melanocyte-stimulating hormone receptor CAA46588 Salazar-Onfray,

Cancer Res.1997

B. Cancer-testes antigens

MAGE-I AAL69948 van der Bruggen, Science, 1991

MAGE-2 AAA17729, P43356 Visseren, Int. J. Cancer, 1997

MAGE-3 AAA17446 Gaugler, J. Exp. Med., 1994

MAGE-12 AAA19023 Panelli, J. Immunol. 2000

BAGE AAI13811, AAI07040, NP_001178 Boel, J. Immunity, 1995 GAGE NP_001459 van den Eynde, J. Exp. Med., 1995

NY-ESO-I CAA05908, NP_001318 Jager, J. Exp. Med., 1998; Wang, J. Immunol. 1998

C. Non-mutated shared antigens overexpressed on cancers α-Fetoprotein CAA79592 Butterfield, Cancer Res. 1999

Telomerase catalytic protein Q99973 Vonderheide, Immunity,

1999

G-250 P41870 Vissers, Cancer Res. 1999 MUC-I P15941 Jerome, Cancer Res. 1991

Carcinoembryonic antigen AAB93379 Tsang, J. Natl. Cancer Inst

1995 p53 BAC16799, AAK76359, AAK76358, BAD11806, AAC12971, CAA25652, AAB20140, AAZ05887 Theobald, PNAS, 1995

Her-2/neu AAA58637 Ioannides, Cell Immunol.1993 PRAME CAG30435 Kessler, J. Exp. Med., 2001

D. Mutated antigens β-catenin CAA79497, CAA61107, AAL89457, AAD32267 Robbins, J. Exp. Med., 1996

MUM-I NP_116242, AAI30444, AAH19585, AAH08098, AAH82987 Chiari, Cancer Res. 1999

CDK-4 NP_000066 Wolfel, Science, 1995 Caspase-8 AAD24962, BAB32555 Mandruzzato, J. Exp. Med., 1997

KIA 0205 Gueguen, J. Immunol. 1998 HLA-A2-R1701 Brandle, J. Exp. Med., 1996

E. Virus-encoded antigens E6 and E7 from HPV van Driel, Eur. J. Cancer 1999 and Morishima, Int. J. Cane. 2007.

EBNA-3, -4, -6 and BMLF-I from EBV, Hill, J. Exp. Med. 1995.

Core- and surface-antigen and polymerase from HBV, Yang, Vaccine,

2007.

The above cancer-associated antigens are believed to be associated with the following cancers:

MAGE, MAGE family testis and tumor-specific protein, AAK00357; associated with breast, lung, bladder, oesophagus, prostate, colorectal, head and neck cancers.

ESO-I Cancer/testis antigen IB (L antigen family member 2) (LAGE-2 protein) , (Autoimmunogenic cancer/testis antigen NY-ESO-I) , P78358; associated to Breast, lung, bladder, oesophagus, prostate head and neck cancers. EBNA-3, -4 and -6; EBNA-3, Epstein-Barr nuclear antigen 3 (EBV nuclear antigen 3) (EBNA-3) (EBNA- 3A), P12977; EBNA-4, Epstein-Barr nuclear antigen 4 (CAA373981) , EBNA-6, Epstein-Barr nuclear antigen 6 (P03204), BMLF-I, (ABU55428); associated to nasopharyngeal carcinoma, Burkitt ' s and Hodgkin's lymphoma.

HBV surface antigen, HBV surface proteins [Hepatitis B virus], CAA48355, HBV core antigen [Hepatitis B virus], CAA59535 and polymerase HBV polymerase [Hepatitis B virus], CAA48354; associated to Hepatoma.

HPV E6, E7 [Human papillomavirus type 16], AAD33252 and [Human papillomavirus type 16], AAD33253; associated to anogenital cancers and head-and neck cancers.

HER-2/neu, Receptor tyrosine-protein kinase erbB-2 precursor (pl85erbB2) (C-erbB-2) , (NEU proto-oncogene) (Tyrosine kinase-type cell surface receptor HER2), (MLN 19) (CD340 antigen), P04626; associated to breast, ovary, pancreas cancers.

ras and p53; Ras, muscle RAS oncogene homolog [Homo sapiens], NP 036351; p53, tumor protein p53 [Homo sapiens], NP 000537; associated to many kinds of cancers .

MUC-I [Homo sapiens], CAA56734; associated to breast, pancreas, colon, ovary, lung cancers.

SART-I, squamous cell carcinoma antigen recognized by T cells 1 [Homo sapiens], NP 005137; associated to squamous head, neck, lung cancers.

PRAME, preferentially expressed antigen in melanoma [Homo sapiens] NP 996839 and RAGE-I, preferentially expressed antigen in melanoma [Homo sapiens], Q9UQ07; associated with kidney cancer. Telomerase, Telomerase protein component 1 (Telomerase-associated protein 1) , (Telomerase protein 1) (p240) (p80 telomerase homolog) , Q99973; associated with many kinds of cancer

In addition, a large number of pathogen-related antigens having MHC-I epitopes are known. Such pathogen-related antigens may comprise parent peptide sequences for modification in accordance with the present invention. Examples of such pathogen-related antigen MHC-I epitopes are shown in Table 9 of patent application GB0720118.9.

Particular examples of antigens include the human gplOO antigen having the amino acid sequence of SEQ ID NO: 1 (Figure lla) and the mouse gplOO antigen having the amino acid sequence of SEQ ID NO: 2 (Figure lib) . Sequences for hgplOO and mgplOO are also available under database Accession numbers gi | 639590 | gb | AAC60634.1 | gplOO [Homo sapiens] and gi | 31981217 | ref | NP_068682.2 | silver [Mus musculus], respectively.

Iden ti fying a pepti de sequence of a CD8+ T cel l epi t ope of an anti gen

A parent peptide sequence of a CD8+ T cell epitope of an antigen may be identified by any suitable method. For example, electronic sequence information may be searched for known antigen sequences, published sequences may be evaluated to identify candidate MHC Class I epitopes, e.g. using an algorithm for the prediction of an MHC Class I epitope. Such algorithms are described further below.

A number of epitopes that are capable of binding an MHC Class I molecule are known to be present in human and mouse gplOO. For example, human gpl00₂5-3₃: KVPRNQDWL (SEQ ID NO: 7) and gplθθ₂o₉-2i7 : ITDQVPFSV (SEQ ID NO: 23) .

Many different predictive programs are available for the identification of potential MHC Class I epitopes in any given polypeptide. Any suitable method may be used to identify within a given polypeptide or peptide sequence an epitope that is capable of being displayed on an MHC Class I molecule. Examples of epitope- identification programs can be found at the websites of the following organizations and facilities:

SYFPEITHI: http : //www . syfpeithi . de/

Immune Epitope Database: http://www.immuneepitope.org/home.do Bioinformatics and Molecular Analysis Section, NIH: http : //www-bimas . cit . nih . gov/molbio/hla bind/ Max-Planck-Institute for Infection Biology: www.mpiib-berlin .mpg . de/MAPPP

Jenner Institute, Bioinformatics group: http : //www . j enner . ac . uk/MHCPred/

Technical University of Denmark, NetChop Server www. cbs . dtu . dk/ services /NetChop

University of Tubingen, Prediction Algorithm for Proteasomal Cleavages: www.paproc . de

The Hebrew University of Jerusalem, computational molecular biology http : //margalit .huji.ac.il/ Institute of Microbial Technology (India), ProPred-I tool www. imtech . res . in/raghava/propredl/ index . html Stockholm Bioinformatics Center, SVMHC tool: www. sbc . su . se/svmhc/new. cgi

The algorithms used within these prediction web sites can be found at the same sites. The following is a non-exhaustive list of references describing suitable prediction algorithms:

BIMAS: Parker, K. C, Bednarek,M. A. and Coligan,J.E. (1994) ^ΛScheme for ranking potential HLA-A2 binding peptides based on independent binding of individual peptide side-chains' . J. Immunol., 152, 163- 175. SYFPEITHI: Hans-Georg Rammensee, Jutta Bachmann, Niels Nikolaus Emmerich, Oskar Alexander Bachor, Stefan Stevanovic (1999) ΛSYFPEITHI: database for MHC ligands and peptide motifs' . Immunogenetics 50: 213-219

RANKPEP: Reche PA, Glutting JP, Reinherz EL. (2002) 'Prediction of MHC Class I binding peptides using profile motifs' . Hum Immunol. 63 (9) :701-9.

MHC Class I-restricted peptides are described at the website http://www.immuneepitope.org/home.do. See also Greenbaum et al . Journal of Molecular Recognition 2007 Mar-Apr; 20 (2) : 75-82 , which describes algorithms for epitope identification.

Identifying a peptide sequence of a CD8+ T cell epitope may comprise selecting the peptide sequence for modification, and need not comprise predicting a peptide sequence of an epitope, since many peptide sequences of CD8+ T cell epitopes have already been identified, and many examples can be found in the published literature .

Table 2 provides a non-exhaustive list of MHC-I epitopes from tumour- associated antigens (TAAs) which may be employed as a parent peptide in accordance with the present invention. Polypeptides or peptides comprising modified sequences of these parent peptides are useful for treatment of the cancers with which the TAAs are associated.

MHC-I epitopes from various pathogen sources are known in the art and a non-exhaustive list is compiled i.a. in Table 9 of patent application GB0720118.9, filed on 15 October 2007. Polypeptides or peptides comprising modified sequences of the parent peptides of Table 9 of GB0720118.9 may be useful for treatment of pathogen- related disease, especially disease associated with the pathogen from which the parent peptide is associated.

Table 2

Gene HLA Allele Peptide epitope

BAGE HLA-CwI 6 AARAVFLAL (SEQ ID NO: 25)

CAMEL HLA-A2 MLMAQEALAFL(SEQ ID NO: 26)

DAM- 6, -10 (MAGE-Bl, -B2) HLA-A2 FLWGPRAYA (SEQ ID NO: 27)

GAGE-I, -2, -! HLA-Cw6 YRPRPRRY (SEQ ID NO: 28) GAGE-3,-4,-5,-6,-7B HLA-A29 YYWPRPRRY (SEQ ID NO: 29)

IL-13Ra2 HLA-A* 0201 WLPFGFILI (SEQ ID NO: 30) MAGE-Al HLA-Al EADPTGHSY (SEQ ID NO: 31)

HLA-A3 SLFRAVITK (SEQ ID NO: 32)

HLA-A24 NYKHCFPEI (SEQ ID NO: 33)

HLA-A28 EVYDGREHSA (SEQ ID NO: 34)

HLA-B37 REPVTKAEML (SEQ ID NO: 35)

HLA-B53 DPARYEFLW (SEQ ID NO: 36)

HLA-Cw2 SAFPTTINF (SEQ ID NO: 37)

HLA-Cw3 SAYGEPRKL (SEQ ID NO: 38)

HLA-CwI 6 SAYGEPRKL (SEQ ID NO: 39)

MAGE-A2 HLA-A2 KMVELVHFL (SEQ ID NO: 40)

HLA-A2 YLQLVFGIEV (SEQ ID NO: 41)

HLA-A24 EYLQLVFGI (SEQ ID NO: 42)

HLA-B37 REPVTKAEML (SEQ ID NO: 43) MAGE-A3 HLA-Al EADPIGHLY (SEQ ID NO: 44)

HLA-A2 FLWGPRALV (SEQ ID NO: 45)

HLA-A24 TFPDLESEF (SEQ ID NO: 46)

HLA-A24 IMPKAGLLI (SEQ ID NO: 47) HLA-B44 MEVDPIGHLY (SEQ ID NO: 48)

HLA-B52 WQYFFPVIF (SEQ ID NO: 49)

HLA-B37 REPVTKAEML (SEQ ID NO: 50)

HLA-B*3501 EVDPIGHLY (SEQ ID NO: 51) MAGE-A4 HLA-A2 GVYDGREHTV (SEQ ID NO: 52) MAGE-A6 HLA-A34 MVKISGGPR (SEQ ID NO: 53)

HLA-B37 REPVTKAEML (SEQ ID NO: 54)

HLA-B*3501 EVDPIGHVY (SEQ ID NO: 55)

MAGE-AlO HLA-A2 GLYDGMEHL (SEQ ID NO: 56) MAGE-A12 HLA-Cw7 VRIGHLYIL (SEQ ID NO: 57) NA88-A HLA-B13 MTQGQHFLQKV (SEQ ID NO: 58) NY-ESO-I HLA-A2 SLLMWITQCFL (SEQ ID NO: 59)

HLA-A2 SLLMWITQC (SEQ ID NO: 60)

HLA-A2 QLSLLMWIT (SEQ ID NO: 61)

HLA-B*3501 MPFATPMEA (SEQ ID NO: 62)

NY-ESO-Ia (CAG-3; HLA-A31 ASGPGGGAPR (SEQ ID NO: 63)

SSX-2 HLA-A2 KASEKIFYV (SEQ ID NO: 64)

TRAG-3 HLA-A* 0201 ILLRDAGLV (SEQ ID NO: 65)

CEA (CAP-I) HLA-A2 YLSGANLNL (SEQ ID NO: 66)

HLA-A3 HLFGYSWYK (SEQ ID NO: 67)

Ep-CAM HLA-A2 GLKAGVIAV (SEQ ID NO: 68) GpIOO HLA-A2 KTWGQYWQV (SEQ ID NO: 69)

HLA-A2 AMLGTHTMEV (SEQ ID NO: 70)

HLA-A2 MLGTHTMEV (SEQ ID NO: 71)

HLA-A2 SLADTNSLAV (SEQ ID NO: 72)

HLA-A2 ITDQVPFSV (SEQ ID NO: 73)

HLA-A2 LLDGTATLRL (SEQ ID NO: 74)

HLA-A2 YLEPGPVTA (SEQ ID NO: 75) HLA-A2 VLYRYGSFSV (SEQ ID NO: 76)

HLA-A2 RLMKQDFSV (SEQ ID NO: 77)

HLA-A2 RLPRIFCSC (SEQ ID NO: 78)

HLA-A3 LIYRRRLMK (SEQ ID NO: 79)

HLA-A3 ALNFPGSQK (SEQ ID NO: 80)

HLA-A3 SLIYRRRLMK (SEQ ID NO: 81)

HLA-A3 ALLAVGATK (SEQ ID NO: 82)

HLA-A24 VYFFLPDHL (SEQ ID NO: 83)

HLA-A* 6801 HTMEVTVYHR (SEQ ID NO: 84)

HLA-B*3501 VPLDCVLYRY (SEQ ID NO: 85)

HLA-Cw8 SNDGPTLI (SEQ ID NO: 86)

Mammaglobin-A HLA-A3 PLLENVISK (SEQ ID NO: 87)

KLLMVLMLA (SEQ ID NO: 88)

TTNAIDELK (SEQ ID NO: 89)

AIDELKECF (SEQ ID NO: 90)

Melan-A/MART-lb HLA-A2 AAGIGILTV (SEQ ID NO: 91)

HLA-A2 EAAGIGILTV (SEQ ID NO: 92)

HLA-A2 AND B*350 ILTVILGVL (SEQ ID NO: 93)

HLA-B45 AEEAAGIGIL (SEQ ID NO: 94)

HLA-B45 AEEAAGIGILT (SEQ ID NO: 95)

MClR HLA-A2 TILLGIFFL (SEQ ID NO: 96)

HLA-A2 FLALIICNA (SEQ ID NO: 97)

OAl HLA-A*2402 LYSACFWWL (SEQ ID NO: 98)

P polypeptide HLA-A2 IMLCLIAAV (SEQ ID NO: 99) PSA HLA-Al VSHSFPHPLY (SEQ ID NO: 100)

HLA-A2 FLTPKKLQCV (SEQ ID NO: 101)

HLA-A2 VISNDVCAQV (SEQ ID NO: 102)

TRP-I (or gp751 HLA-A31 MSLQRQFLR (SEQ ID NO: 103) TRP-2 HLA-A2 SVYDFFVWL (SEQ ID NO: 104)

HLA-A2 TLDSQVMSL (SEQ ID NO: 105)

HLA-A31 LLGPGRPYR (SEQ ID NO: 106)

HLA-A33 LLGPGRPYR (SEQ ID NO: 107)

HLA-Cw8 ANDPIFWL (SEQ ID NO: 108)

Tyrosinase HLA-Al KCDICTDEY (SEQ ID NO: 109)

HLA-Al SSDYVIPIGTY (SEQ ID NO: 110)

HLA-A2 YMDGTMSQV (SEQ ID NO: 111)

HLA-A2 MLLAVLYCL (SEQ ID NO: 112)

HLA-A24 AFLPWHRLF (SEQ ID NO: 113)

HLA-B44 SEIWRDIDF (SEQ ID NO: 114)

HLA-B*3501 TPRLPSSADVEF (SEQ ID NO: 115) α-Actinin-4 HLA-A2 FIASNGVKLV (SEQ ID NO: 116) β-Catenin HLA-A24 SYLDSGIHF (SEQ ID NO: 117)

Caspase-8 HLA-B35 FPSDSWCYF (SEQ ID NO: 118)

CDK-4 HLA-A2 ACDPHSGHFV (SEQ ID NO: 119)

ELF2 HLA-A68 ETVSEQSNV (SEQ ID NO: 120)

HSP70-2 M HLA-A2 SLFEGIDIY (SEQ ID NO: 121)

KIAA0205 HLA-B44*03 AEPINIQTV (SEQ ID NO: 122)

Malic enzyme HLA-A2 FLDEFMEGV (SEQ ID NO: 123)

MART-2 HLA-Al FLEGNEVGKTY (SEQ ID NO: 124)

MUM-I HLA-B44 EEKLIVVLF (SEQ ID NO: 125)

MUM- 2 HLA-B44 SELFRSGLDY (SEQ ID NO: 126)

HLA-Cw6 FRSGLDSYV (SEQ ID NO: 127)

MUM- 3 HLA-A28 EAFIQPITR (SEQ ID NO: 128) Myosin HLA-A3 KINKNPKYK (SEQ ID NO: 129) OS-9 HLA-B44 KELEGILLL (SEQ ID NO: 130) BING-4 HLA-A2 MCQWGRLWQL (SEQ ID NO: 131) K-RAS HLA-B35 WVGAVGVG (SEQ ID NO: 132)

N-RAS HLA-Al ILDTAGREEY (SEQ ID NO: 133)

OGT HLA-A2 SLYKFSPFPL (SEQ ID NO: 134)

TGFaRII HLA-A2 RLSSCVPVA (SEQ ID NO: 135)

TRP-2/INT2 HLA-A68 EVISCKLIKR (SEQ ID NO: 136)

TRP-2-6b HLA-A2 ATTNILEHY (SEQ ID NO: 137)

Adipophilin HLA-A2 SVASTITGV (SEQ ID NO: 138)

AIM- 2a HLA-Al RSDSGQQARY (SEQ ID NO: 139)

AFP HLA-A2 GVALQTMKQ (SEQ ID NO: 140) ART- 4 HLA-A24 AFLRHAAL (SEQ ID NO: 141)

HLA-A24 DYPSLSATDI (SEQ ID NO: 142)

CLCA2 HLA-A2 LLGNCLPTV (SEQ ID NO: 143) Cyp-B HLA-A24 KFHRVIKDF (SEQ ID NO: 144)

HLA-A24 DFMIQGGDF (SEQ ID NO: 145) EphA2 HLA-A* 0201 IMNDMPIYM (SEQ ID NO: 146)

HLA-A* 0201 VLAGVGFFI (SEQ ID NO: 147)

FGF-5 HLA-A3 NTYASPRFKb (SEQ ID NO: 148) G250 HLA-A2 HLSTAFARV (SEQ ID NO: 149) GnT-V HLA-A2 VLPDVFIRC (SEQ ID NO: 150) HER-2/neu HLA-A2 KIFGSLAFL (SEQ ID NO: 151)

HLA-A2 11 SAWGI L (SEQ ID NO: 152)

HLA-A2 RLLQETELV (SEQ ID NO: 153)

HLA-A2 WLGWFGI (SEQ ID NO: 154)

HLA-A2 ILHNGAYSL (SEQ ID NO: 155)

HLA-A2 YMIMVKCWMI (SEQ ID NO: 156)

HLA-A24 TYLPTNASL (SEQ ID NO: 157)

HLA-A3 VLRENTSPK (SEQ ID NO: 158)

HST-2 (FGF-6) HLA-A31 YSWMDISCWI (SEQ ID NO: 159) hTERT HLA-A2 ILAKFLHWL (SEQ ID NO: 160)

HLA-A2 ILAKFLHWL (SEQ ID NO: 161)

HLA-A2 RLVDDFLLV (SEQ ID NO: 162)

HLA-A3 KLFGVLRLK (SEQ ID NO: 163) iCE HLA-B7 SPRWWPTCL (SEQ ID NO: 164)

Livin (ML-IAP) HLA-A2 SLGSPVLGL (SEQ ID NO: 165)

HLA-A2 RLASFYDWPL (SEQ ID NO: 166)

M-CSF HLA-B*3501 LPAVVGLSPGEQEY (SEQ ID NO: 167; MUCl HLA-AIl STAPPAHGV (SEQ ID NO: 168)

HLA-A2 STAPPVHNV (SEQ ID NO: 169)

MUC2 HLA-A2 LLNQLQVNL (SEQ ID NO: 170)

HLA-A2 MLWGWREHV (SEQ ID NO: 171)

PRAME HLA-A24 LYVDSLFFL (SEQ ID NO: 172)

HLA-A2 VLDGLDVLL (SEQ ID NO: 173)

HLA-A2 SLYSFPEPEA (SEQ ID NO: 174)

HLA-A2 ALYVDSLFFL (SEQ ID NO: 175)

HLA-A2 SLLQHLIGL (SEQ ID NO: 176)

PSMA HLA-Al HSTNGVTRIY (SEQ ID NO: 177)

HLA-A24 LYSDPADYF (SEQ ID NO: 178)

HLA-A24 NYARTEDFF (SEQ ID NO: 179)

P15 HLA-A24 AYGLDFYIL (SEQ ID NO: 180) P53 HLA-A24 AIYKQSQHM (SEQ ID NO: 181)

HLA-B46 SQKTYQGSY (SEQ ID NO: 182)

RAGE HLA-B7 SPSSNRIRNT (SEQ ID NO: 183) RUl HLA-B51 VPYGSFKHV (SEQ ID NO: 184) RU2 HLA-B7 LPRWPPPQL (SEQ ID NO: 185) SART-I HLA-A24 EYRGFTQDF (SEQ ID NO: 186)

HLA-A* 2601 KGSGKMKTE (SEQ ID NO: 187) SART -2 HLA-A24 DYSARWNEI (SEQ ID NO: 188) HLA-A24 AYDFLYNYL (SEQ ID NO: 189) HLA-A24 SYTRLFLIL (SEQ ID NO: 190)

SART -3 HLA-A24 VYDYNCHVDL (SEQ ID NO: 191) HLA-A24 AYIDFEMKI (SEQ ID NO: 192) HLA-A2 LLQAEAPRL (SEQ ID NO: 193) HLA-A2 RLAEYQAYI (SEQ ID NO: 194)

SOXlO HLA-A2 SAWISKPPGV (SEQ ID NO: 195) Survivin HLA-A2 ELTLGEFLKL (SEQ ID NO: 196) HLA-A2 TLPPAWQPFL (SEQ ID NO: 197)

Survivin-2Bg HLA-A24 AYACNTSTL (SEQ ID NO: 198)

TRG HLA-B52 YQLCLTNIF (SEQ ID NO: 199)

WTl HLA-A2 RMFPNAPYL (SEQ ID NO: 200) HLA-A24 CMTWNQMNL (SEQ ID NO: 201) HLA-A24 RWPSCQKKF (SEQ ID NO: 202)

707-AP HLA-A2 RVAALARDA (SEQ ID NO: 203) abl-bcr alb-b3(b2a2) HLA-A*0201 FVEHDDESPGL (SEQ ID NO: 204)

abl- bcr alb-b4(b3a2) HLA-A*0201 FVEHDLYCTL (SEQ ID NO: 205) bcr-abla HLA-A2 FMVELVEGA (SEQ ID NO: 206)

HLA-A2 KLSEQESLL (SEQ ID NO: 207)

HLA-A2 MLTNSCVKL (SEQ ID NO: 208) bcr-abl p210 (b3a2) HLA-A2 SSKALQRPV (SEQ ID NO: 209)

HLA-A3 ATGFKQSSK (SEQ ID NO: 210)

HLA-A3 KQSSKALQR (SEQ ID NO: 211)

HLA-A3, HLA-AIl HSATGFKQSSK (SEQ ID NO: 212;

HLA-A3 KQSSKALQR (SEQ ID NO: 213)

HLA-B8 GFKQSSKAL (SEQ ID NO: 214)

ETV6/AML HLA-A2 RIAECILGM (SEQ ID NO: 215) NPM/ALK HLA-A2*0201 SLAMLDLLHV (SEQ ID NO: 216)

HLA-A2*0201 GVLLWEIFSL (SEQ ID NO: 217)

SYT/SSX HLA-B7, HLA-B42 QRPYGYDQIM (SEQ ID NO: 218) HPVl 6 E6 HLA-A2 TIHDIILECV (SEQ ID NO: 219)

HLA-A24 VYDFAFRDL (SEQ ID NO: 220)

HLA-A24 QYNKPLCDLL (SEQ ID NO: 221)

HLA-A24 EYRHYCYSL (SEQ ID NO: 222)

HLA-B7 QERPRKLPQL (SEQ ID NO: 223)

HLA-Bl 4 SSRTRRETQL (SEQ ID NO: 224)

HLA-B27 GRWTGRCMSC (SEQ ID NO: 225)

HLA-B57 FAFRDLCIVY (SEQ ID NO: 226)

HPVl 6 E7 HLA-A2*0201 YMLDLQPET (SEQ ID NO: 227)

HLA-B8 TLHEYMLDL (SEQ ID NO: 228)

Core antigen HBV HLA-A2 FLPSDFFPSV (SEQ ID NO: 229)

HLA-AIl YVNVNMGLK (SEQ ID NO: 230)

HLA-A24 EYLVSFGVW (SEQ ID NO: 231)

HLA-A24 KYTSFPWLL (SEQ ID NO: 232)

HLA-A31 STLPETTVVRR (SEQ ID NO: 233)

Surface antigen HBV HLA-A2 GLSPTVWLSV (SEQ ID NO: 234)

HLA-A2 WLSLLVPFV (SEQ ID NO: 235)

HLA-A2 FLLTRILTI (SEQ ID NO: 236)

Polymerase HBV HLA-A2 GLSRYVARL (SEQ ID NO: 237) EBNA-3 EBV HLA-A2 SVRDRLARL (SEQ ID NO: 238)

HLA-A3 RLRAEAQVK (SEQ ID NO: 239)

HLA-A24 RYSIFFDY (SEQ ID NO: 240)

HLA-A29 VFSDGRVAC (SEQ ID NO: 241)

HLA-A30 AYSSWMYSY (SEQ ID NO: 242)

HLA-B7 VPAPAGPIV (SEQ ID NO: 243) HLA-B7 RPPIFIRRL (SEQ ID NO: 244)

HLA-B8 FLRGRAYGL (SEQ ID NO: 245)

HLA-B8 QAKWRLQTL (SEQ ID NO: 246)

HLA-B8 AYPLHEQHG (SEQ ID NO: 247)

HLA-B8 YIKSFVSDA (SEQ ID NO: 248)

EBNA- 4 EBV HLA-A2, A68 AVFDRKSDAK (SEQ ID NO: 2491

HLA-A3, All AVFDRKSDAK (SEQ ID NO: 250 )

HLA-AIl IVTDFSVIK (SEQ ID NO: 251)

EBNA- 6 EBV HLA-A2 LLDFVRFMGV (SEQ ID NO: 252;

HLA-A2 SLREWLLRI (SEQ ID NO: 253)

HLA-B7 QPRAPIRPI (SEQ ID NO: 254)

BMLF-I EBV HLA-A2 GLCTLVAML (SEQ ID NO: 255)

HLA-Bl 8 DEVEFLGHY (SEQ ID NO: 256)

Si ze of MHC Cl ass I-binding epi t opes

In general, MHC-I-binding epitopes have a peptide sequence of 8 to 10 amino acids, with epitopes of 9 amino acids in length being common. However, it has been found that sequence lengths outside this range can also bind to an MHC Class I molecule. For example, Glithero et al., J. Biol. Chem., 2005, Vol. 281, No. 18, pp. 12699-12704 describes binding between a pentapeptide NYPAL and H-2D^b. Peptides longer than 10 amino acids in length have also been reported to bind to MHC Class I molecules (Tynan et al, Nature Immunology, 2007, Mar; 8 (3) :268-76; Burrows et al, Trends in Immunology, 2006 Jan; 27 (1) : 11-6; Tynan et al, Nature Immunology, 2005, Nov; 6 (11 ) : 1114- 22, Miles et al, J. Immunology, 2005, Sep 15; 175 (6) : 3826-34; Tynan et al, J. Biol. Chem., 2005, 280 (25) : 23900-9) . Therefore, in certain embodiments the parent peptide sequence is from 3 to 25 amino acids in length, such as 5 to 15 amino acids in length. Preferably, the parent peptide is from 8 to 10 amino acids in length. MHC Class I Molecule

Methods described herein may be used to improve binding affinity to any MHC Class I molecule (MHC-I) . The MHC-I may be human. Preferably, the MHC-I is one having the conserved tyrosine 159 residue. Preferred MHC-I include human MHC-I such as HLA-A2 and HLA- AIl as well as mouse MHC-I such as H-2D^b.

MHC molecules in different species are listed for example in the following internet sites mentioned above: http : //www. syfpeithi . de/ http : //www. immuneepitope . org/home . do

The MHC Class I molecule may be other than H-2D^d. For example, the MHC-I may be a classical MHC-I such as H-2D^b, HLA-A2 and HLA-AIl. Alternatively, the MHC Class I molecule may be a non-classical MHC-I such as HLA-G, HLA-F, HLA-E, or any novel MHC Class I-like molecule discovered in humans, as well as all MHC Class I-like molecules in vertebrates and/or those produced by viruses, bacteria and any other pathogens .

The polypeptide or peptide of the invention may bind to an MHC Class I molecule other than H-2D^d. The polypeptide or peptide of the invention may have greater binding affinity for an MHC Class I molecule other than H-2D^d than the binding affinity it has for H-2D^d. Preferably, the polypeptide or peptide of the invention has at least 2-fold, such as 5-fold or 10-fold, greater binding affinity for an MHC-I other than H-2D^d than the binding affinity it has for H-2D^d.

Binding affinity

Binding affinity of a peptide for an MHC-I can be defined using a competition-based cellular binding assay (van der Burg et al, Human Immunology, 1995; Kessler et al, J. Exp. Med. 2001) . High affinity binding peptides were identified as those with an IC₅₀ <6 μM, peptides that bound with intermediate affinity corresponded to 6 μM < IC₅₀ < 15 μM, peptides that displayed a low binding affinity corresponded to 15 μM < IC₅₀ < 100 μM) , and finally peptides with undetectable binding capacity corresponded to an IC₅₀ > 100 μM. To define more precisely binding characteristics, peptide-MHC stability can be assessed by measuring the dissociation rate of high affinity binding peptides complexed with the MHC Class I molecule at 37°C (van der Burg et al, J. Immunology, 1996) . A modified peptide may be considered as having increased binding affinity for an MHC-I as compared with the parent peptide if the modified peptide exhibits a statistically significant increase in binding affinity as measured in a binding assay as described herein. Preferably, a modified peptide will have binding affinity for an MHC-I which is at least 10% greater, more preferably at least 50% greater, more preferably at least 100% greater, more preferably at least 5-fold greater, most preferably at least 10-fold greater than the binding affinity of the parent peptide for the same MHC-I.

MHC Class I peptide-binding groove

The A-, B-, C-, D-, E- and F-pockets are corresponding to the initial description in the pioneering publication of Prof. Pamela Bjorkman (Bjorkman et al, Nature, 1987 (a) ; Bjorkman et al, Nature, 1987 (b) and Saper et al J. MoI. Biol., Bjorkman et al, J. Immunology, 2005) . However, structural knowledge acquired since 1987 demonstrated that in some cases, the pockets and in particular pocket B is formed by other residues on the heavy chain of the MHC Class I molecules.

Thus, although the words 'B-pocket' are used herein, this does not mean that such a denomination is restricted to the residues initially described in the above publications.

Each MHC allele has a distinct peptide specificity that is a reflection of the chemical composition of the residues that comprise the antigen-binding site. This peptide specificity may be summarized by a peptide motif that demonstrates the restrictions and preferences of the individual pockets that comprise the antigen-binding site. Residues that protrude into specificity pockets are termed anchors, and these anchors are characterized by residues or residue classes that are conserved at equivalent positions of the peptide. The term anchor originated from the strong belief that these positions conferred significant binding energy. These anchors may be dominant, whereby an inherent inflexibility in the type of residue chelated suggests highly specific and efficient binding, or secondary or promiscuous whereby a large number of residue types may be accommodated, but some preferences are evident. Each allelic form of a class I molecule preferentially binds peptides that conform to a particular binding sequence motif, often defined by the second (P2) and the last residues (Pω) , usually P8, P9 or PlO in the human and by the fifth (P5) and the last one (Pω) in the mouse.

Table 3

Non-exhaustive list of peptide motifs for common MHC class I alleles

Allele Peptide Motif

HLA-A0201- (nanomeric peptides) AILTVM at P2 and AILTVM at P9 HLA-A0201- (decameric peptides) AILTVM at P2 and AILTVM at PlO HLA-AIlOl (nanomeric peptides) KR at P9 HLA-A6801 (nanomeric peptides) AILTVM at P2 and RK at P9 HLA-A6801 (decameric peptides) AILTVM at P2 and RK at PlO HLA-B27 (nanomeric peptides) RK at P2 and LFYRKHMI at P9 HLA-B35 (octameric peptides) P at P2 and YFMLI at P8 HLA-B35 (nanomeric peptides) P at P2 and YFMLI at P9 HLA-B53 (nanomeric peptides) P at P2 H-2D^b (nanomeric peptides) N at P5 and AILTVM at P9 H-2D^d (decameric peptides) G at P2, P at P3, RK at P5 and AILTVMFYW at

P9 H-2K^b (octameric peptides) FY at P5 and AILTVM at P8 H-2K^b (nanomeric peptides) FY at P5 and AILTVM at P9 H-2Ld (nanomeric peptides) PS at P2 and AILTVMFYW at P9

Footnote : A: alanine, F: phenylalanine, H. histidine, I: isoleucine, K: lysine, L: leucine, M: methionine, N: asparagine, P: proline, R: arginine, S: serine, T: threonine, V: valine, W. tryptophane, Y: tyrosine.

Producing a polypepti de or pepti de compri sing a modi fi ed pepti de sequence

Any suitable method of producing a polypeptide or peptide comprising the modified peptide sequence or an isolated nucleic acid molecule encoding the polypeptide or peptide may be used in the present invention. The polypeptide or peptide or isolated nucleic acid molecule may be synthesised de novo using knowledge of the sequence of the parent peptide or may be synthesised by mutating the parent peptide sequence e.g. using site-directed mutagenesis. Therefore, a polypeptide or peptide comprising modified peptide sequence may be provided directly with the mutation (s) or indirectly by first producing the parent peptide sequence and mutating it.

In order to provide peptides it may be convenient to synthesise the entire peptide comprising the modified peptide sequence using standard Fmoc chemistry. Another convenient way of providing a longer peptide or a polypeptide is to express nucleic acid encoding it, by use of the nucleic acid in an expression system. Systems for cloning and expression of a polypeptide in a variety of different host cells are well known. Suitable host cells include bacteria, eukaryotic cells such as mammalian and yeast, and baculovirus systems. Suitable vectors can be chosen or constructed, containing appropriate regulatory sequences, including promoter sequences, terminator fragments, polyadenylation sequences, enhancer sequences, marker genes and other sequences as appropriate. Vectors may be plasmids, viral e.g. 'phage, or phagemid, as appropriate. For further details see, for example, Molecular Cloning: a Laboratory Manual: 3rd edition, Sambrook and Russell, 2001, Cold Spring Harbor Laboratory Press. 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 Current Protocols in Molecular Biology, Ausubel et al . eds., John Wiley & Sons, 1992. Nucleic acid encoding a polypeptide or peptide comprising the modified peptide sequence may be produced synthetically making use of the known sequence of the parent peptide. Alternatively, the nucleic acid encoding the parent peptide may be cloned and subjected to site-directed mutagenesis so as to provide an altered nucleic acid sequence encoding the modified peptide sequence.

In certain embodiments the polypeptide provided comprises the full length sequence of the antigen with the modified peptide. For example, where the antigen is full length mouse gplOO of SEQ ID NO: 2, the method may include providing a full length mouse gplOO polypeptide sequence wherein the serine at position 3 of the gpl00₂₅-₃₃ epitope (position 27 of the mouse gplOO sequence) is mutated to proline. In this example the polypeptide provided by the method may be designated mgplOO S27P.

In other embodiments the method may provide a fragment of the full length sequence of the antigen wherein the fragment comprises the modified peptide sequence. The fragment may be at least 100 amino acids in length. The modified peptide sequence may be at any suitable position within the polypeptide sequence. For example, the modified peptide sequence may be N-terminal or C-terminal or may lie at any position therebetween. Preferably, the position of the modified peptide sequence will be chosen such that, upon antigen processing of the polypeptide, the modified peptide sequence is displayed on an MHC Class I molecule.

In further embodiments the method may provide a peptide of up to 25 amino acids in length wherein the peptide comprises the modified peptide sequence. The method may provide a peptide having one or more amino acids N-terminal and/or C-terminal of the modified peptide sequence. In other embodiments the method may provide a peptide consisting of the modified peptide sequence. As used herein a polypeptide or peptide comprising a modified peptide sequence is said to "correspond to" the full length sequence of the antigen or a fragment of the antigen even though the polypeptide or peptide will not have the exact same sequence as the antigen or a fragment of the antigen; the polypeptide or peptide will have at least one modification in accordance with the invention. Preferably, the only difference between the sequence of the polypeptide or peptide comprising the modified peptide sequence and the antigen or a fragment of the antigen is one or more modifications in accordance with the invention.

In certain embodiments the polypeptide or peptide may comprise a plurality of MHC Class I epitopes. The plurality may include multiple instances of the same modified epitope and/or a number of differing epitopes. In certain embodiments the polypeptide may comprise an "epitope string" comprising multiple repeats of the modified peptide sequence.

In still further embodiments the method may provide an isolated nucleic acid encoding the polypeptide (s) or peptide (s) comprising the modified peptide sequence as described herein. Such an isolated nucleic acid may be in the form of a vector in order to produce the polypeptide or peptide comprising the modified peptide sequence in a host cell and/or in order to deliver the polypeptide (s) or peptide (s) to an organism, including a human. Therefore, the isolated nucleic acid provided by the method may be useful for delivering the polypeptide (s) or peptide (s) comprising the modified peptide sequence to a patient. Typically a vector will comprise a nucleic acid encoding the polypeptide (s) or peptide (s) comprising the modified peptide sequence as described herein operably linked to one or more regulatory sequences.

T cell responses

The polypeptide or peptide comprising the modified peptide sequence may elicit an increased response from CD8⁺ CTLs as compared with the parent peptide. The increased CTL response may include increased CTL proliferation and/or IFNγ production and/or expression of maturity markers such as CD62L. Such CTL responses may be assessed using assays described in further detail below. It has been found that mutation to provide a cyclic or pseudo-cyclic amino acid at position 3 of the modified peptide increases binding affinity of the modified peptide for an MHC Class I molecule while preserving T cell recognition. Therefore, the conformation of the modified peptide when presented on an MHC Class I molecule may be essentially unchanged as compared with the conformation of the parent peptide when presented on the MHC Class I molecule, such that T cell recognition is preserved. The polypeptide or peptide comprising the modified peptide sequence may be capable of acting as a T helper peptide-epitope .

CTL Proliferation

Proliferation will lead to the expansion of stimulated CD8+ T cells. The proliferation of T lymphocytes, like that of other cells, may be measured in vitro by determining the amount of 3H-labelled thymidine incorporated into the DNA of cultured cells. Thymidine incorporation provides a quantitative measure of the rate of DNA synthesis, which is usually directly proportional to the rate of cell division.

IFN-gamma producti on

The production levels of interferon gamma reflect the levels of activation of CD8+ T cells.

Expression of maturity markers

CD8+ T cell maturity may be assessed by measuring the expression of maturity markers such as CD62L. Thus, an increased CD8+ T cell response may be detected by determining the level of (i) IFN-γ (ii) 3H-labelled thymidine incorporated into DNA and/or (iii) expression of maturity markers, in cultured CD8+ T cells contacted with an antigen presenting cell displaying the modified peptide as compared with cultured CD8+ T cells contacted with an antigen presenting cell displaying the parent peptide, and detecting a higher level of (i) , (ii) and/or (iii) for T cells stimulated with the modified peptide as compared with T cells stimulated with the parent peptide.

Suitable methods for measuring CTL proliferation, IFN-gamma production and CD62L expression are described in detail in Example 4 below. A modified peptide may be considered as causing an enhanced CTL response as compared with the parent peptide if the modified peptide exhibits a statistically significant increase in CTL proliferation, IFN-gamma production and/or expression of maturity markers such as CD62L, as measured using assay methods described herein. Preferably, a modified peptide will induce a CTL response which is at least 10% greater, more preferably at least 50% greater, yet more preferably at least 100% greater, yet more preferably at least 5-fold greater, most preferably at least 10-fold greater than the corresponding CTL response of the parent peptide.

Therapeutic methods

In a still further aspect the present invention provides a method of treating a subject which comprises administering an effective amount of a polypeptide or peptide according to the invention or an isolated nucleic acid according to the invention or a vector comprising a nucleic acid according the invention or a pharmaceutical composition according to the invention. Conditions which may be treated in accordance with the invention include tumours (preferably, cancerous tumours such as melanoma) and infections and conditions alleviated by generation of an immune response. In the case of treatment of a tumour, the parent peptide sequence may be a CD8+ T cell epitope from a tumour-associated antigen which is found on the tumour to be treated. For example, in the case of treatment of melanoma the parent peptide sequence may be from gplOO.

In a still further aspect the present invention provides a method of treating a subject which comprises administering an effective amount of a polypeptide or peptide according to the invention or an isolated nucleic acid according to the invention or a vector comprising a nucleic acid according the invention or a pharmaceutical composition according to the invention. Conditions which may be treated in accordance with the invention include induction of tolerance, in order to stop, slow down and/or impair the progression of autoimmune reactions. In the case of treatment of such a disease, the parent peptide sequence may be a CD8+ T cell epitope from a self, a virus, a bacteria or any other pathogen-associated antigen which is correlated to the treated disease. For example, in the case of treatment of multiple sclerosis, the parent peptide sequence may be from e.g. myelin basic protein (MBP) , proteolipid protein (PLP) and myelin oligodendrocyte glycoprotein (MOG) .

Therapeutic agents will be formulated appropriately for their desired route of administration. The agent or pharmaceutical composition comprising the agent may be administered to a subject by any convenient route of administration, whether systemically/ peripherally or topically (i.e., at the site of desired action) . For example, for parenteral administration, e.g. injection, either subcutaneously, intramuscularly or intravenously, injectables can be prepared in conventional forms, either as liquid solutions or suspensions, solid forms suitable for solution or suspension in liquid prior to injection, or as emulsions. Suitable excipients are, for example, water, saline, dextrose, glycerol, ethanol or the like. In addition, if desired, the pharmaceutical compositions to be administered may also contain minor amounts of non-toxic auxiliary substances such as wetting or emulsifying agents, pH buffering agents and the like, such as for example, sodium acetate, sorbitan monolaurate, triethanolamine oleate, triethanolamine sodium acetate, etc . Adjuvant s

In certain embodiments a polypeptide or peptide of the invention or an isolated nucleic acid of the invention is provided in the form of a pharmaceutical composition comprising an adjuvant. For example, the composition may comprise the polypeptide (s) or peptide (s) comprising the modified peptide (s) sequence (s) and CpG oligodeoxynucleotide as described in Example 3. Non-exclusive examples of adjuvants as well as listings of adjuvants that may be suitable for use with the peptides of the invention can be found the following references: Cox and Coulter, ^ΛAdjuvants - a classification and review of their modes of action', Vaccine, 1997; Singh and O'Hagan, 'Advances in vaccine adjuvants', Nature Biotechnology, 1999.

The following non-exhaustive list of potential adjuvants are useful in accordance with the present invention:

Alum; aluminium-based mineral salts aluminium hydroxide, aluminium phosphate, calcium phosphate.

Immunostimulatory adjuvants Cytokines (e.g. IL-2, IL- 12 and GM-CSF), Saponins (e.g. QS21), CpG oligos, Lipopolysaccharide (LPS) , Monophosphoryl Lipid A (MPL) , Polyphosphazenes .

Lipid particles

Emulsions (e.g. Freund's, SAF, MF59) , Liposomes, Virosomes, Iscoms Cochleates.

Nano and Microparticulate adjuvants

PLG polylactide co-glycolide microparticles, Poloxamer particles, Virus-like particles.

Mucosal adjuvants Heat-labile enterotoxin (LT) , Cholera toxin (CT) , Mutant toxins (e.g. LTK63 and LTR72) , microparticles, polymerised liposomes.

Non-particulate adjuvants Muramyl dipeptide (MDP) and derivatives. Other adjuvants

Calcium salts, Proteosomes, Virosomes, Stearyl tyrosine γ-inulin, Algammulin, Non-ionic block copolymers, e.g. polyoxypropylene/polyoxyethylene .

Complexes

A soluble, multimeric complex of MHC Class I molecules and peptides according to the invention, particularly a tetramer, may be used in identification and characterisation of antigen-specific T cells. The tetramers may bind peptide-specific cytotoxic T cells in vitro. The use of tetramer assays to assess immunisation responses is described in further detail in Example 5 below.

Experimental

Example 1 - Crystal structures of MHC-I molecule in complex with gplOO peptides

The present inventors initiated a structural study involving the determination of the crystal structures of an MHC-I molecule from the mouse, H-2D^b, in complex with two 9-mer variants of a peptide derived from the melanoma-associated protein gplOO. The reasons underlying the differences in binding affinity and immunogenicity of mouse gpl00₂₅-₃₃ (mgplOO corresponding to the sequence EGSRNQDWL (SEQ ID NO: 3) and human gpl00₂₅-₃₃ (hgplOO corresponding to the sequence KVPRNQDWL, SEQ ID NO: 7) were studied. Both main anchor positions at positions 5 (asparagine) and 9 (leucine) are conserved between the two peptides.

Complexes of H-2D^b with mgplOO and H-2D^b with hgplOO were produced, refolded and isolated. Crystallisation conditions were determined for both complexes. Thereafter, complexes of H-2D^b with modified analogue gplOO (EGP) were also produced, refolded and isolated. Crystallisation conditions were also determined for the latter complex.

Preparation and crystallisation of the H-2D^b/peptide MHC complexes Both the H-2D^b heavy chain and the murine β₂m were expressed in Escherichia coli. The H-2D^b/KVP, H-2D^b/EGP and H-2D^b/EGS MHC complexes were refolded by dilution. All complexes were isolated following concentration using size-exclusion chromatography. The isolated refolded MHC complexes were thereafter concentrated to 6 mg/ml. Crystals were obtained in hanging drops by vapour diffusion. Crystal screens (Hampton Research, Laguna Niguel, CA, USA) were used to establish initial crystallisation conditions that were then refined in a finer grid. The best crystals for all H-2D^b/peptide complexes were obtained in 1.8 M ammonium sulphate, 0.1 M Tris Cl pH 9.0, at 4°C. Typically, 4 μl of a 6 mg/ml protein solution in 20 mM Tris Cl pH 7.5 were mixed at a 4:2 ratio with the crystallisation reservoir solution. The drops were allowed to equilibrate at 4°C.

Data collection and processing Data collection was performed under cryogenic conditions (T=IOOK) , at beam line 1711 in MaxLab (Lund, Sweden) and beam lines ID14-EH1 and ID29 (ESRF, Grenoble, France) (lambda=0.934A) using an ADSC Q210 CCD detector. X-ray data for the H-2D^b/KVP, the H-2D^b/EGP and the H- 2D^b/EGS complexes were collected to 1.9, 2.3 and 2.6A resolution, respectively. Crystals were preliminarily soaked in a cryoprotectant solution containing 25% glycerol before data collection. The diffraction data were processed using the XDS program package. Crystals for the H-2D^b/KVP, H-2D^b/EGP and H-2D^b/EGS complexes belong to space group P2i.

Structure determination and refinement

The space group and unit cell parameters were determined using the program XDS, and scaling and reduction of the data were performed using XSCALE. The structure of H-2D^b/KVP, H-2D^b/EGP and H-2D^b/EGS were solved by molecular replacement using the program PHASER. The atomic coordinates of the previously solved crystal structure of H-2D^b in complex with the LCMV peptide gp33 (pdb code 1N5A) were used as the search model. Four solutions were unambiguously identified. Further crystallographic refinement was conducted using the program REFMAC5 from the CCP4 package. Five percent of the reflections were set aside for validation and monitoring the refinement by R_free • The structures were first subjected to rigid body refinement. Heavy weighted (Wa=300 kcal/mol A²) , restrained non-crystallographic symmetry (NCS) was used throughout the first stage of refinement.

Later in the refinement, the weight was significantly decreased for regions that did not obey to NCS. After each round of refinement, errors in the model were identified by careful examination of the (2F_O-F_C) and (F_o-F_c) maps with the program COOT. Isotropic, individual B-factor refinement was used throughout the procedures. Refinement cycles of TLS parameters were also performed using REFMAC5. The stereochemistry of the model was analyzed using PROCHECK. Solvent- accessible areas were analyzed within the CNS program package. The final models consist of 276 residues for the heavy chain, 99 residues for the β₂m subunit and 9 residues for the peptides. A total of 270 water molecules were added to well-defined peaks found between 2 and 4 A from 0 or N atoms in the protein, using the program ARP in REFMAC. Structural comparisons were carried out with Swiss PDB- Viewer using default parameters.

Structural models of the peptides hgplOO (KVP) , mgplOO (EGS) , and modified gplOO (EGP) , respectively, in complex with H-2D^b are shown in Figures 1 to 3.

Upon scrutinizing and comparing the three structures, the inventors concluded that a number of explanations were possible for the observed higher affinity binding of hgplOO when compared to mgplOO. Among those considered further were: a) The presence of a valine at position 2 (p2V) in hgplOO (instead of a glycine in mgplOO) could help to fill the so-called pocket B. This hypothesis corresponded well to the dogma stating that in order to increase the binding affinity of peptides to a specific MHC Class I molecule, one has to improve the fit of such peptides through, surface complementarity, charge-charge interactions, or both.

b) The presence of a proline residue at position 3 (p3P) could stabilise the binding of the peptide through ring-ring stacking interactions with the side chain of tyrosine 159 (Y159) in the heavy chain of H-2D^b. In order to explore this hypothesis, sequence comparison was performed. The comparison of known sequences of MHC Class I molecules in the human, the mouse and other mammalians revealed that residue Y159 is conserved.

c) The combination of a larger hydrophobic residue at position 2 that would better complement the peptide/MHC heavy chain interactions, and of a proline (or any other aromatic residue, such as a phenylalanine or a tyrosine, or any other cyclic natural or synthetic compound) at position 3 of the peptide might also be responsible for the improved binding of the gplOO with position 2 hydrophobic and/or position 3 aromatic .

4) The combination of a glycine residue at position 2 and of a proline (or any other aromatic residue, such as a phenylalanine or a tyrosine, or any other cyclic natural or synthetic compound) at position 3 of the peptide might also be responsible for the improved binding of the gplOO peptide with position 2 with a high level of conformational freedom and position 3 aromatic.

Example 2 - Synthesis of MHC-I binding peptides

In order to test the effects of modification in positions 1, 2 and 3 of gpl00₂₅-₃₃ peptides on MHC-I binding affinity and on the induction of specific cytotoxic T lymphocyte responses, peptides were produced using standard methods.

The following peptides, shown in Table 4 below, were produced for testing in the context of H-2D^b:

Table 4

Name Sequence

EGS EGSRNQDWL (SEQ ID NO: 3)

EVS EVSRNQDWL (SEQ ID NO: 4)

EVP EVPRNQDWL (SEQ ID NO: 5)

EGP EGPRNQDWL (SEQ ID NO: 6)

KVP KVPRNQDWL (SEQ ID NO: 7)

KVA KVARNQDWL (SEQ ID NO: 8)

AVP AVPRNQDWL (SEQ ID NO: 9)

KAP KAPRNQDWL (SEQ ID NO: 10

KGP KGPRNQDWL (SEQ ID NO: 11

Peptides and CpG

The EGS, EVS, EVP, EGP, KVP, KVA, AVP, KAP and KGP peptides were synthesised in the Leiden University Medical Center. Variants of gpl00₂₅-₃₃ (plp2p3RNQDWL) are designated by the amino acid sequence of position 1, 2 and 3 of the peptide. CpG oligodeoxynucleotides 1826 TCCATGACGTTCCTGACGTT (SEQ ID NO: 24) were synthesised in the Leiden Institute of Chemistry. The following peptides (shown in Table 5 below) were used for testing in the context of HLA-A2 and HLA-AIl and were purchased from the Genscript, USA. GLCTLVAML (SEQ ID NO: 12) is the Epstein-Barr Virus- associated peptide BMLF-I (259-267) . IVTDFSVIK (SEQ ID NO: 16) is the Epstein Barr virus-associated peptide EBNA-4 (416-424) .

Table 5

Name Sequence

For HLA-A2 :

GLC GLCTLVAML (SEQ ID NO: 12

GLY GLYTLVAML (SEQ ID NO: 13

GLP GLPTLVAML (SEQ ID NO: 14

GGP GGPTLVAML (SEQ ID NO: 15

For HLA-AIl:

IVT IVTDFSVIK (SEQ ID NO: 16

IVY IVYDFSVIK (SEQ ID NO: 17

IVP IVPDFSVIK (SEQ ID NO: 18

IGP IGPDFSVIK (SEQ ID NO: 19

Exampl e 3 - Binding affini ty of pepti des t o MHC-I

The binding affinity of various peptides for MHC-I molecule H-2D -Φ , HLA-A2 and HLA-AIl was tested.

H-2D^b binding assay

Binding of peptides to H-2D^b was determined using a RMA-S binding assay. Briefly, RMA-S cells were cultured for 2 days at 26°C to achieve expression of ^Λempty' Class I molecules on the cell surface, Serial dilutions of peptides were added to duplicate cultures of cells to allow for stabilization of MHC molecules. Peptides corresponding to Adenovirus 5 ElA •_:234-243 D^b binding epitope SGPSNTPPEI (SEQ ID NO: 21) and MuLV envi₈₉-i₉₆ K^b binding epitope SSWDFITV (SEQ ID NO: 22) were used as positive and negative controls, respectively. After 4 h incubation at 37°C, cells were stained with antibodies (BD biosciences) specific for H-2D^b (28.14.8S) followed by allophycocyanin-labeled goat-anti-mouse Ig. Fluorescence was detected using a FACS Calibur cytometer (BD biosciences) . The H-2D^b binding index is expressed as (mean fluorescence with peptide - mean fluorescence without peptide) /mean fluorescence without peptide.

The results are shown in Table 6 below:

Table 6

Name Sequence Binding affinity

mgpl00₂5-33 peptides EGS EGSRNQDWL

EVS EVSRNQDWL + ( + )

EVP EVPRNQDWL +++

EGP EGPRNQDWL

hgpl00₂5-33 peptides

KVP KVPRNQDWL +++

KVA KVARNQDWL + ( + )

AVP AVPRNQDWL +++

KAP KAPRNQDWL ++++

KGP KGPRNQDWL ++++

These results indicate that the introduction of a proline at position 3 of the peptide increased in a significant manner the binding affinity of the modified peptides. It should be noted that the modified peptides were recognised by the T-cell receptors on the surface of the tested T lymphocytes.

The results also indicate that the filling of the pocket B by a larger residue at position 2 such as an alanine or valine only slightly increased the binding affinity of the modified peptides. It was found that the presence of a glycine at position 2 increased binding affinity of the peptides for MHC-I .

Significantly, the results indicate that the combined introduction of a proline at position 3 of the peptide and the use of a glycine at position 2 increased binding affinity of the modified peptides to H-2D^b very significantly (by an amplitude not previously observed) . Furthermore, it was found that the increase in binding affinity for MHC-I was achieved without altering the recognition of the peptides by T-cells (as demonstrated by the comparative analysis of the crystal structures of H-2D^b/EGS, H-2D^b/KVP and H-2D^b/EGP) .

The enhanced binding affinity of modified peptides having proline at position 3 or a combination of glycine at position 2 and proline at position 3 is shown with reference to the binding-concentration curves of Figures 4-5 and 12.

HLA-A2 and HLA-AIl binding affinity assays

Peptide binding to HLA-A2 and HLA-AIl molecules was measured by using the T2-A2 cell line (Nijman et al . , 1993 and Oh et al . , 2004), and the similar cell line T2-A11. T2 cells (3^χlO⁵/well) were incubated overnight in 96-well plates with culture medium (a 1:1 mixture of RPMI 1640-Eagle-Hank's amino acid (EHAA) media containing 2.5% FCS, 100 U/ml penicillin, and 100 μg/ml streptomycin) with 10 μg/ml human β2-microglobulin (Sigma-Aldrich, St. Louis, MO) and peptides. Cells were washed twice with cold PBS containing 2% FCS and then incubated for 30 min at 4 ⁰C with anti-HLA-A2.1 BB7.2 mAb (1/100 dilution of hybridoma supernatant) or the anti-HLA-All antibody 4i90. After washing, cells were stained with 5 μg/ml FITC-labelled goat anti- mouse Ig (BD PharMingen, San Diego, CA) and expression levels of HLA- A2 or HLA-AIl were measured by flow cytometry (FACSCaliber; BD Biosciences, Mountain View, CA) . HLA-A2 and HLA-AIl expression were quantified as fluorescence index (FI) according to the formula: FI= [ (mean fluorescence with peptide-mean fluorescence without peptide) /mean fluorescence without peptide] . Background fluorescence without BB7.2 or 4i90 as subtracted for each individual value. To compare the different peptides, FI₅₀, the peptide concentration, μM, that increases HLA-A2 or HLA-AIl expression by 50% over no peptide control background, was calculated from the titration curve for each peptide .

The results are shown in Table 7 below:

Table 7

Name Sequence Binding affinity

HLA-A2 :

GLC GLCTLVAML ++(+)

GLY GLYTLVAML +++++

GLP GLPTLVAML ++++ GGP GGPTLVAML HLA-AI l :

IVT IVTDFSVIK -(+)

IGP IGPDFSVIK +++ IVY IVYDFSVIK ++++

IVP IVPDFSVIK +++++(+)

The results indicate, inter alia, that introduction of a cyclic residue at position 3 (Proline or Tyrosine) increased binding affinity of the modified peptides to HLA-A2 and HLA-AIl compared with the wild-type sequences.

Exampl e 4 - Induction of T-cell response by MHC-I binding peptides

The ability of various peptides to induce a T-cell response was tested by measuring: the proliferation rate of T-cells, the expression of the maturity marker CD62L and the expression of IFNγ. The T-cell response assays employed pmel cells specific for H- 2D^b/mgpl00 and H-2D^b/hgpl00. Importantly, the modified peptides enhanced proliferation of T cells in the draining lymph nodes as well as the non-draining lymph nodes and the spleen (Figures 9 and 10) .

Mice

Male C57BL/6 mice were obtained from Iffa Credo (France) . gpl00₂₅-₃₃/D^b specific TCR transgenic pmel mice were a kind gift of Dr. N. P. Restifo (Bethesda) and were bred to express the congenic marker

Thyl .1. All mice were housed in the animal facility of the Leiden University Medical Center under specific pathogen free conditions and used between 6 and 12 weeks of age. Experiments were performed in accordance with national legislation and institutional guidelines. In vivo experiments

Lymphocytes from spleen and lymph nodes of naive pmel mice (with T cells specific for H-2D^b/EGS that crossreact to H-2D^b/KVP) were isolated and enriched for T lymphocytes by passing the cell suspension over a nylon wool column. Pmel cells were labelled with 5 μM CFSE (Molecular Probes, Leiden) and 3 x 10^δ CD8⁺ T lymphocytes were adoptively transferred by injection into the tail vein. One day later the mice were immunised by subcutaneous injection of a mixture of 50 μg peptide and 25 μg CpG in PBS. After 4 days, spleen and draining lymph nodes were removed and single cell suspensions were prepared for analysis. Lymphocytes were incubated with fluorochrome- labelled antibodies (BD biosciences) specific for CD8 (53-6.7) , Thyl.l (HIS51) , and CD62L (MEL-14) . A part of the lymphocytes was incubated for 3 h in the presence of 1 μg/ml EGS and GolgiPlug™ and thereafter stained for intracellular IFN-γ (XMGl.2) in combination with CD8 and Thyl.l according to the manufacturer's protocols (BD biosciences) . Cells were analyzed using a FACS Calibur™ flow cytometer and Cell Quest Pro software (BD biosciences) . Plots of CFSE, IFN-γ and CD62L are gated on CD8⁺, Thyl.l÷ live lymphocytes.

The ability of the tested peptides to stimulate T-cell proliferation and IFNγ expression is shown in Table 8 below:

Table 8 Name Sequence T-cell proliferation IFNγ expression mgpl00₂5-33 peptides

EGS EGSRNQDWL

EVS EVSRNQDWL EVP EVPRNQDWL

EGP EGPRNQDWL

hgpl00₂5-33 peptides

KVP KVPRNQDWL ++ +

KVA KVARNQDWL

AVP AVPRNQDWL KAP KAPRNQDWL

KGP KGPRNQDWL

The results show that the T-cell responses correspond very well to the enhanced binding affinity of the modified peptides described above in Example 3. The introduction of proline at position 3 increased the T-cell response induced by the modified peptides. The combined introduction of proline at position 3 and glycine at position 2 further enhanced T-cell responses.

Example 5 - Immunization assays

Immunization assays with mgpl00₂₅-₃₃ peptide EGS (SEQ ID NO: 3) or modified mpgl00₂₅-₃₃ peptide EGP (SEQ ID NO: 6) were carried out using an optimised vaccination protocol. The vaccination protocol involved priming the endogenous gplOO-specific T-cell repertoire. Directly ex vivo no CTL response could be detected, but following in vitro re- stimulation with peptide-loaded dendritic cells it was possible to select murine gplOO specific responses using tetramerical constructs and checking for IFNy expression in the EGP-immunised mice but not in the EGS-immunised group. These results show that the EGP peptide, with a glycine at position 2 and a proline at position 3, induces a stronger immune response than the unmodified EGS peptide having serine at position 3. This also demonstrates that the activated T cells cross-react and recognise the non-immunogenic peptide EGS in complex with H-2D^b.

For priming of endogenous T cells, naive C57BL/6 mice were shaved on the flank and injected subcutaneously on day 0 and day 7 with PBS alone or 50 μg EGS or 50 μg EGP peptide. Immediately following injection, Aldara creme containing 5% imiquimod (3M Health Care Ltd., Leicestershire, UK) was applied to the skin at the injection site. On day 10, splenocytes were harvested and restimulated in vitro with LPS-matured, irradiated dendritic cells (Dl) loaded with 0.5 μg/ml EGS. After 6 days, viable cells were isolated by ficoll density gradient and directly stained with CD8 and EGS/D^b tetramer (a kind gift of Dr. T Schumacher, Amsterdam) or stained for CD8 and intracellular IFN-γ after 3 h of incubation with EGS peptide in the presence of Golgi blockade. Plots are gated on live, CD8⁺ lymphocytes (Figure 11) .

Vaccination with super-peptide EGP induces robust CTL responses from endogenous T-cell repertoire

The obvious disadvantage of a TCR transgenic system is that the monoclonal CTL population triggered by peptide analogs does not reflect the polyclonal character of endogenous responses. In addition, peptide analogs may give rise to CTL responses with different fine-specificities that do not necessarily cross-react with the original peptide target. Endogenous CTL responses against EGS, KVP and the super-peptides EGP and KGP were analyzed in order to address this issue. Groups of C57BL/6 mice were vaccinated with each peptide in combination with the TLR7 ligand imiquimod as an adjuvant (Rechtsteiner et al . , Journal of Immunology, 174, 2476-2480, 2005) . The reactivity of the vaccine-induced CD8⁺ T-cells isolated from spleens and lymph nodes was analyzed in vitro (Figure 13A) .

Vaccination and in vitro stimulation with EGS did not yield any detectable peptide-specific T-cells while e.g. EGP-induced T-cell responses displayed the most favorable characteristics: they were present at high frequencies and cross-reacted to even very low concentrations of the natural target EGS.

Vaccination with the super-peptide EGP induces melanoma-reactive CTL

Peptide vaccination for treatment of cancer has so far not met clinical success, although vaccine-induced CTL responses can be detected in some patients (Rosenberg et al . , Nature Medicine, 10, 909-915, 2004) . Peptide vaccination can be strongly improved through the use of longer peptides that comprise minimal CTL epitopes in combination with adequate adjuvants that activate the innate immune system (Bijker et al . , Journal of Immunology, 179, 5033-5040, 2007) . The immunotherapeutic potential of the super-peptide EGP was assessed against the aggressive B16 melanoma that expresses the gplOO antigen EGS. Ex vivo analyses of C57BL/6 mice immunized with EGS, KVP or EGP combined with topical application of imiquimod (obtained as Aldara™, 5% imiquimod; 3M Health Care Ltd.) resulted in very high frequencies of EGS-specific CD8⁺ T-cells in the blood of EGP-injected animals, with numbers up to 50% of the total CD8⁺ cell population (Figure 13B) . The potency of this peptide clearly overruled that of the other groups, in which only low frequencies against the natural mouse peptide EGS were detected. Furthermore, analysis with tetramerical constructs of H-2D^b in complex with wild-type EGS corroborated these results (Figure 8) . Importantly, the CD8⁺ T cells raised by EGP efficiently killed B16 melanoma cells in vitro (Figure 13C) as well as peptide-loaded targets in vivo (Figure 13D) . In all of these experiments, the human orthologue KVP performed better than the mouse EGS, but was far less efficient when compared to the analog super- peptide EGP. In all animals immunized with EGP, the induced CTL population displayed a relatively broad repertoire of TCR variable regions that was comparable to that of the other peptide variants (Figure 14), indicating that super-peptide agonists are different from bacterial superantigens that activate single TCR Vβ families (Sundberg et al . , Current Opinion in Immunology, 14, 36-44, 2002) . Finally, we tested the immunotherapeutic potential of the improved gplOO super-peptide EGP against established B16 mouse melanomas. C57BL/6 mice were inoculated with a tumorigenic dose of melanoma and treated from day 9, when sub-clinical tumors were established.

Therapeutic vaccination with the peptide vaccine led to significant numbers of CTLs in blood and within tumor beds, resulting in a delay of tumor outgrowth. Addition of naive pmel cells at day 8 in the treatment scheme was sufficient to eradicate established melanomas in half of the EGP-treated group (Figure 13E) . Interestingly, no overt sign of autoimmune pathology or ocular autoimmunity was observed in the surviving mice during the time frame of the experiments (four months) . Nonetheless, all surviving mice turned white at the site of vaccination and whiskers after approximately two months, so far to a less pronounced extent than previously described in this model

(Overwijk et al . , Journal of Experimental Medicine, 198, 569-580, 2003) .

In conclusion, the introduction of a proline residue at position 3 of an otherwise very weak tumor CTL epitope (EGS) transformed it into a super-peptide (EGP) that can be used as a powerful vaccine capable of inducing effective anti-tumor CTL responses.

Claims

1. A method of producing a polypeptide or peptide, or nucleic acid encoding the polypeptide or peptide, wherein the polypeptide or peptide comprises a modified peptide sequence that has increased binding affinity for an MHC Class I molecule and/or generates a higher CD8+ T cell immune response, the method comprising: identifying a peptide sequence ("parent peptide") of a CD8⁺ T cell epitope of an antigen, wherein the peptide sequence is capable of being displayed on an MHC Class I molecule; and producing a polypeptide or peptide comprising a modified peptide sequence of the CD8 T cell epitope or a nucleic acid molecule encoding said polypeptide or said peptide, wherein the modified peptide sequence has a mutation compared with the parent peptide, the mutation providing a cyclic or pseudo-cyclic amino acid residue at position 3 of the modified peptide sequence, defined such that the residues are sequentially numbered with position 1 being the first residue displayed within the peptide- binding groove of the MHC Class I molecule, and wherein the modified peptide has greater binding affinity for the MHC Class I molecule and/or generates a higher CD8+ T cell immune response, compared with the parent peptide.

2. A method according to claim 1, wherein the mutation provides a cyclic amino acid residue at position 3 of the modified peptide sequence .

3. A method according to claim 2, wherein the cyclic amino acid residue is proline.

4. A method according to claim 2, wherein the cyclic amino acid residue is a natural or unnatural amino acid residue having an aromatic group.

5. A method according to claim 2, wherein the cyclic amino acid residue is tyrosine, phenylalanine, histidine or tryptophan.

6. A method according to claim 1, wherein the mutation provides a pseudo-cyclic amino acid residue at position 3 of the modified peptide sequence.

7. A method according to claim 6, wherein the pseudo cyclic amino acid is N-propyl alanine.

8. A method according to any of the preceding claims, wherein the modified peptide has a hydrophobic residue at position 2.

9. A method according to any of the preceding claims, wherein the modified peptide has a glycine at position 2.

10. A method according to any one of claims 1 to 7, wherein the modified peptide has a residue at position 2, 5 and/or 9 selected from an MHC-I-binding motif shown in Table 3.

11. A method of producing a polypeptide or peptide, or nucleic acid molecule encoding the polypeptide or peptide, wherein the polypeptide or peptide comprises a modified peptide sequence that has increased binding affinity for an MHC Class I molecule and/or generates a higher CD8+ T cell immune response, the method comprising: identifying a peptide sequence ("parent peptide") of a CD8⁺ T cell epitope of an antigen, wherein the peptide sequence is capable of being displayed on an MHC Class I molecule and wherein the peptide sequence has a cyclic or pseudo-cyclic amino acid at position 3, defined such that the residues are sequentially numbered with position 1 being the first residue displayed within the peptide- binding groove of the MHC Class I molecule; and producing a polypeptide or peptide comprising a modified peptide sequence of the CD8⁺ T cell epitope or an nucleic acid molecule encoding said polypeptide or encoding said peptide, wherein the modified peptide sequence has a mutation compared with the parent peptide, the mutation providing a hydrophobic amino acid residue at position 2 of the modified peptide sequence and wherein the modified peptide has greater binding affinity for the MHC Class I molecule and/or generates a higher CD8+ T cell immune response, compared with the parent peptide.

12. A method according to claim 8 or claim 11, wherein said hydrophobic residue at position 2 is selected from glycine, valine and leucine.

13. A method according to any preceding claim wherein the antigen is a tumour associated antigen or an antigen from a pathogen.

14. A method according to claim 13, wherein the antigen is selected from the antigens listed in Table 1.

15. A method according to claim 14, wherein the antigen is human gplOO, mouse gplOO, Epstein-Barr Virus-associated peptide BMLF-I or NY-ESO-I .

16. A method according to any of the preceding claims, wherein the parent peptide sequence is 8 to 10 amino acids in length.

17. A method according to claim 13, wherein the parent peptide comprises a peptide epitope having a peptide sequence shown in Table 2.

18. A method according to claim 13 , wherein the parent peptide comprises a peptide epitope having a peptide sequence selected from SEQ ID NO: 3, SEQ ID NO: 7, SEQ ID NO: 12 and SEQ ID NO: 16.

19. A method according to any of the preceding claims, comprising producing a polypeptide comprising the modified peptide sequence, or producing nucleic acid encoding said polypeptide, wherein the polypeptide corresponds to the full length sequence of the antigen.

20. A method according to any of claims 1 to 18, comprising producing a polypeptide comprising the modified peptide sequence, or producing nucleic acid encoding said polypeptide, wherein the polypeptide corresponds to a fragment of at least 100 amino acid residues of the antigen.

21 A method according to any of claims 1 to 18 comprising producing a peptide comprising the modified peptide, or producing nucleic acid encoding said peptide, wherein the peptide comprising the modified peptide is up to 25 amino acids in length.

22 A method according to any of claim 21 wherein the peptide comprising the modified peptide is capable of acting as a T helper peptide-epitope .

23 A method according to any of claims 1 to 18, comprising producing a peptide consisting of the modified peptide sequence, or producing nucleic acid encoding a peptide consisting of the modified peptide sequence.

24 A method according to any one of claims 1 to 23 comprising producing an nucleic acid molecule encoding the polypeptide or peptide comprising the modified peptide sequence.

25 A method according to claim 24 wherein the nucleic acid molecule is a vector.

26 A method according to any preceding claim, wherein the MHC Class I molecule is H-2D^b, HLA-A2 or HLA-AIl.

27 A method of producing a composition comprising a polypeptide or peptide, or an isolated nucleic acid encoding the polypeptide or peptide, wherein the polypeptide or peptide comprises a modified peptide sequence that has increased binding affinity for an MHC Class I molecule and/or that generates a higher CD8+ T cell immune response, the method comprising producing a polypeptide, peptide or nucleic acid according to the method of any of claims 1 to 26 and formulating the polypeptide, peptide or nucleic acid into a composition comprising at least one additional component.

28 A method according to claim 27 wherein the composition comprises a pharmaceutically acceptable excipient .

29 A polypeptide, peptide or nucleic acid produced by a method according to any of claims 1 to 26

30.A composition produced by a method according to claim 27or claim 28

31 An isolated polypeptide or peptide comprising a mutant peptide sequence which comprises a CD8⁺ T cell epitope capable of being displayed on an MHC Class I molecule, wherein said mutant peptide sequence has the sequence of a naturally occurring CD8⁺ T cell epitope of an antigen, except that said mutant peptide sequence has a cyclic or pseudo-cyclic amino acid residue at position 3 which is not present at position 3 of the sequence of said naturally occurring CD8⁺ T cell epitope, and wherein the position numbering is defined such that the residues are sequentially numbered with position 1 being the first residue displayed within the peptide-binding groove of the MHC Class I molecule.

32 An isolated polypeptide or peptide comprising a mutant peptide sequence which comprises a CD8⁺ T cell epitope capable of being displayed on an MHC Class I molecule, wherein said mutant peptide sequence has the sequence of a naturally occurring CD8⁺ T cell epitope of an antigen, except that said mutant peptide sequence has a cyclic or pseudo-cyclic amino acid residue at position 3 which is not present at position 3 of the sequence of said naturally occurring CD8⁺ T cell epitope, and has a hydrophobic amino acid residue at position 2 which is not present at position 2 of the sequence of said naturally occurring CD8+ T cell epitope, and wherein the position numbering is defined such that such that the residues are sequentially numbered with position 1 being the first residue displayed within the peptide-binding groove of the MHC Class I molecule .

33 A polypeptide or peptide according to claim 3lor 32 wherein said naturally occurring CD8⁺ T cell epitope comprises an epitope having a sequence shown in Table 2.

34 A polypeptide or peptide according to claim 33 wherein said naturally occurring CD8⁺ T cell epitope comprises an epitope having a sequence selected from SEQ ID NO: 3, SEQ ID NO: 7, SEQ ID NO: 12 and SEQ ID NO: 16.

35 An isolated polypeptide or peptide comprising a CDS' T cell epitope capable of being displayed on an MHC Class I molecule, wherein the epitope comprises a peptide sequence selected from: SEQ ID NO: 11, SEQ ID NO: 10, SEQ ID NO: 9, SEQ ID NO: 6, SEQ ID NO: 5, SEQ ID NO: 18, SEQ ID NO: 13, SEQ ID NO: 14, SEQ ID NO: 17 and SEQ ID NO: 19.

36. A polypeptide according to claim 29 or any one of claims 31 to 35 having a plurality of MHC Class I epitopes.

37. An isolated nucleic acid molecule encoding a polypeptide or peptide according to any of claims 31 to 36.

38. A vector comprising a nucleic acid according to claim 37.

39. A host cell comprising a vector according to claim 38.

40. A pharmaceutical composition comprising: a polypeptide, peptide or nucleic acid according to claim 29, or a polypeptide or peptide according to any one of claims 31 to 36, or a nucleic acid molecule according to claim 37, or a vector according to claim 38; and a pharmaceutically acceptable excipient.

41. A polypeptide, peptide or nucleic acid according to claim 29, or a polypeptide or peptide according to any one of claims 31 to 36, or a nucleic acid molecule according to claim 37, or a vector according to claim 38, for use in a method of treatment of the human or animal body.

42. A polypeptide, peptide or nucleic acid according to claim 29, or a polypeptide or peptide according to any one of claims 31 to 36, or a nucleic acid molecule according to claim 37, or a vector according to claim 38, for use in generating an immune response against the antigen in a subject.

43. A polypeptide, peptide or nucleic acid according to claim 29, or a polypeptide or peptide according to any one of claims 31 to 36, or a nucleic acid molecule according to claim 37, or a vector according to claim 38 , for use in a method of treating or preventing an infection or a cancer in a subject.

44. A polypeptide, peptide or nucleic acid according to claim 29, or a polypeptide or peptide according to any one of claims 31 to 36, or a nucleic acid molecule according to claim 37, or a vector according to claim 38, for use in a method of treating or preventing melanoma.

45. Use of a polypeptide, peptide or nucleic acid according to claim 29, or a polypeptide or peptide according to any one of claims 31 to 36, or a nucleic acid molecule according to claim 37, or a vector according to claim 38, for the manufacture of a medicament for generating an immune response in a subject.

46. Use of a polypeptide, peptide or nucleic acid according to claim 29, or a polypeptide or peptide according to any one of claims 31 to 36, or a nucleic acid molecule according to claim 37, or a vector according to claim 39, for the manufacture of a medicament for treating or preventing an infection or a cancer in a subject.

47. Use according to claim 46, wherein the medicament is for treating or preventing melanoma.

48. An isolated complex comprising an MHC Class I molecule and a peptide according to claim 25 or any one of claims 31 to 35.

49. A complex according to claim 48, wherein the complex comprises a dimer, tetramer or an oligomer of MHC Class I molecules.