WO2005077980A2

WO2005077980A2 - Soluble tcr-like molecules and their uses

Info

Publication number: WO2005077980A2
Application number: PCT/IB2005/001297
Authority: WO
Inventors: Ruprecht Jules Joost Van Neerven
Original assignee: Bioceros Bv
Priority date: 2004-02-13
Filing date: 2005-02-11
Publication date: 2005-08-25
Also published as: WO2005077980A3

Abstract

The invention provides single domain proteins that are capable of specifically binding to an MHC-peptide complex. Pharmaceutical compositions comprising the single domain proteins according to the invention can be used to treat infectious diseases, cancer, autoimmune diseases and transplant rejection.

Description

SOLUBLE TCR-LIKE MOLECULES AND THEIR USES RELATED APPLICATIONS This application claims the benefit of U.S. Patent Application Serial No.

60/544,878 filed February 13, 2004, the contents of which are hereby incorporated by reference.

FIELD OF THE INVENTION

The invention relates to the field of immunology and medicine. More specifically, the invention relates to molecules capable of specifically interacting with a peptide/MHC molecule complex. A molecule according to the invention can be used to detect intracellular protein antigens on the cell surface and/or may be used to deliver a moiety to a cell, resulting in cell death, decreased cell proliferation or inhibition of protein synthesis in said cell. A molecule of the invention can therefore be used as a tool in a diagnostic procedure, or as a therapeutic agent. Furthermore the invention relates to the preparation of a pharmaceutical composition comprising molecules capable of specifically interacting with a peptide/MHC molecule complex, for the therapeutic treatment of an individual with a tumour and/or a virus infection and/or an autoimmune disease.

BACKGROUND OF THE INVENTION

The immune system is capable of responding to agents (e.g. infectious agents, proteins, (tumour) cells etc.) that pose a threat to an organism. The specificity of this immune response is dependent on the antigen-specific recognition of (protein) antigens. B lymphocytes and T lymphocytes have antigen-specific receptors, and are thus capable of specifically reacting to antigens. Whereas B lymphocytes do so by producing soluble antibodies that specifically bind to proteins, T lymphocytes have a membrane bound T cell receptor (TCR) that can recognize antigen-derived peptides in the context of a major histocompatibility complex (MHC) protein. Antibodies are highly specific soluble proteins that have successfully been developed to treat a wide range of human diseases (Maini R.N., and Taylor PC. 2000 Annu Rev Med. 51:207-29, Chang T.W. 2000 Nat. Biotechnol. 18:157-62. von Mehren M. et al. 2003 Annu.Rev. Med. 2003 54:343-369). Therapeutic uses of antibodies are based on the fact that antibodies can readily neutralize soluble proteins (e.g. enzymes, cytokines and growth factors), and can bind to cell-surface expressed proteins, resulting in modulation or blocking of the function of the cell surface expressed protein. In addition, antibodies can be used to specifically target compounds to cells. However, therapeutic targeting and visualization of cancer cells, virus-infected cells, and/or cells presenting autoantigens is hampered because many of the specific antigens present inside of these cells are not available on their cell membrane, and as a result can not be targeted by conventional antibodies. As mentioned above, T lymphocytes recognize antigen-derived peptides in the context of an MHC molecule. The genomic organisation and the generation of diversity of TCR encoding genes is very similar to that of antibodies, yet TCRs do not recognize intact proteins, but rather recognize an antigen-derived peptide that is embedded in the groove of an MHC molecule (Fremont D. H. et al. 1992 Science 257, 919-927; Madden, D. R et al. 1993, Cell 75, 693-708; Stern, L.J. et al. 1994, Nature 368, 215-221; Young, A. et al. 1994, Cell 16, 39-50). They do not recognize MHC molecules that have bound an irrelevant peptide or empty MHC molecules. This process is termed MHC-restriction (Zinkeπiagel, R.M. and Doherty, P.C. 1974, Nature 248, 701-702). The peptides bound by MHC proteins are derived from either extracellular proteins that are taken up by antigen presenting cells, or from proteins that are synthesized inside the cell (Townsend, A.R.M. et al. 1985, Cell 42, 457-467; Falk, K. et al. 1991, Nature 351, 290-296; Rudensky A. et al., 1992 Nature 359, 429-31.). Thus, non-excreted, non-cell surface proteins can be recognized by TCRs, but not by antibodies. TCRs recognize MHC molecules and peptide simultaneously. Structural studies have shown that TCRs contain quite rigid complementarity determining regions (CDR) that are recognize the MHC molecules. In contrast, the CDR regions that interact with the peptide bound in the MHC molecule have the structural flexibility that is needed to recognize the peptide (Garcia K.C. et al., 1998 Science 20;279:1166-72; Garboczi, D. N. and W. E. Biddison 1999 Immunity 10: 1-7.)

The major histocompatibility complex consists of a range of MHC class I and MHC class II molecules, all grouped together on the same chromosome. The human MHC, also called the Human Leucocyte Antigens (HLA) is located on the short arm of chromosome 6. The diversity of MHC alleles expressed in the general population is very large, and can be explained by the notion that diversity of MHC molecule expression ensures that different individuals have a different ability to present potential pathogens to their immune system. This diversity provides an important evolutionary advantage because a portion of the population will have an MHC haplotype that enables them to fight an infectious agent better than other individuals, thus preventing eradication of the whole population. However, this also means that MHC haplotype (i.e., the combination of all MHC class I and MHC class II molecules expressed in the body) can be linked to disease. The most striking example is HLA-B29, a human MHC class I antigen, that is strongly linked to birdshot chorioretinopathy (Priem H.A. et al.,1988 Am J Ophthalmol. 105:182- 185). More than 95% of all birdshot chorioretinopathy patients are HLA-B29-positive. Likewise, correlations between autoimmune diseases and MHC class II antigens have also been described (i.e., HLA-DR4 with rheumatoid arthritis: Khan M.A. et al., 1988 Tissue Antigens 31:254-258).

In general, MHC class II molecules are heterodimers of a transmembrane MHC class II alpha and a transmembrane MHC class II beta chain that bind peptides derived from extracellular antigens that are taken up and processed by antigen presenting cells such as monocytes, macrophages and/or dendritic cells (Neefjes J.J., and F. Momburg 1993 Curr Opin Immunol. 1993 5:27-34; Braciale TJ., and N . Braciale 1991 Immunol Today 12:124-129). The expression of MHC class II proteins is limited to antigen presenting cells and activated cells of the immune system. Occasionally MHC class II can be expressed on other cell types under non-physiological conditions (e.g. in autoimmunity and chronic inflammation). Antigen presenting cells can take up antigens via pinocytosis or via receptor mediated endocytosis. This is followed by degradation of the proteins into peptides that can bind to MHC class II proteins in intracellular vesicles. Finally, the loaded MHC class II molecules are transported to the cell membrane. The peptides that bind to MHC class II proteins are rather large, and can range from 12 to approximately 25 amino acids in length. This can be explained by the three dimensional structure of the MHC class II heterodimer (Brown J.H. et al. 1993 Nature 364:33-39). The dimer forms a peptide-binding groove that is open on both ends, thus allowing a peptide overhang on both ends. As a result, a peptide that comprises a 12 amino acid stretch that has the correct MHC-binding motif (i.e. has the correct three anchor residues necessary to bind to a given MHC class II protein) can bind to the MHC class II molecule; excess amino acids will stick out of the binding groove. These MHC class II- peptide complexes are recognized by CD4+ T helper cells.

For MHC class I proteins the situation is quite different. MHC class I proteins are also heterodimers, but consist of a transmembrane alpha chain that binds beta-2 microglobulin to form a heterodimer. MHC class I proteins are expressed on virtually all cells of the body, with very few exceptions (as in the eye, and on sperm cells). The MHC class I proteins bind relatively short (8-10 amino acids) peptides that are derived from the cytoplasm. These peptides bind in the peptide binding groove of an MHC class I molecule, that is large enough to accommodate a peptide of 8-10 amino acids as is clear from the molecular structure of MHC class I molecules (Fremont, D. H. et al,

1992 Science 257, 919-927; Madden, D. R. et al., 1993 Cell 75, 693-708; Young, A. C. M. et al, 1994 Cell 16, 39-50). These peptides are generated as a result of the proteolytic degradation of cytoplasmic proteins by the proteasome (Neefjes J.J., and F. Momburg

1993 Curr Opin Immunol. 1993 5:27-34; Braciale T.J., and N.L. Braciale 1991 Immunol Today 12:124-129). After degradation, peptides are transported via the TAP transporter into the endoplasmic reticulum, where the peptide binds to the MHC class I protein, after which it is transported to the cell membrane. This means that many of the peptides bound to MHC class I molecules are derived from self proteins, but in the case of cancers or infections that MHC class I molecules are able to present peptides of cancer- specific or viral proteins to the immune system. Such MHC class I-peptide complexes are recognized by CD8+ cytotoxic T cells. Studies have shown that one way to target these MHC-peptide complexes (to recognize intracellular proteins on the cell surface) is by using soluble TCR molecules (Garcia K.C. et al., 1996 Science 274:209-219; Gregoire C. et al., 1991 Proc Natl Acad Sci U S A. 88:8077-8081; US patent 6,080,840; and US patent publication 20020142389; US patent publication 20020058253). The use of soluble T-cell receptors is, however, limited due to their low affinity to MHC-peptide complexes, their low stability, and because it is very difficult to produce high levels of sTCR molecules. This is in contrast to antibodies that are very specific, have high affinity, are stable and production methods are available.

Even though the structure and genomic organisation of human antibodies resemble T cell receptors, antibodies specifically recognizing MHC-peptide complexes have only been reported in a very limited number of studies after conventional immunization (Wylie D. E. et al., 1982, J Exp. Med. 155, 403-414; van Leeuwen A. et al. 1979 J. Exp. Med. 150, 1075-1083; Froscher B.G and N.R. Klinman 1986, J Exp. Med. 164, 196-210; Murphy D.B. et al., 1989 Nature 338, 765-768; Due H.T. et al, 1993 Int. Immun. 5, 427-431; Aharoni, R.et al., 1991, Nature 351, 147-150). This is most likely because the antibodies have to recognize the complex of MHC molecule + peptide. As the peptide is embedded in the binding groove of an MHC molecule, it is not highly accessible to conventional antibodies.

Another approach to making antibodies that recognize MHC-peptide complexes employs human monomeric Fab-fragments or scFv antibody fragemnts from phage display libraries (Reviewed in Cohen CJ. et al., 2003 J Mol. Recogn. 16:324-332 and disclosed in US patent publication 20020150914, the contents of which are incorporated herein by reference). However, in general these monomeric antibody fragments have significantly lower affinities than conventional antibodies, as a result of which their applicability to be used as targeting molecules in vivo is limited. Moreover, conventional antibodies are relatively large proteins that consist of two heavy chains and two light chains, linked to each other by disulphide links. Fragments of antibodies such as F(ab)2 fragments, F(ab), or single chain Fv antibodies (ScFv) are smaller versions of antibodies that lack the constant regions of heavy and light chains, yet they are still dependent on the correct folding of the heavy and light chain-derived variable regions. As a result production levels of these antibodies are very low, and the resulting proteins are still quite large (appr. 260 amino acids for ScFv). In addition, F(ab) fragments and ScFv molecules generally have a lower affinity than the bivalent parent antibody. In addition, antibodies and antibody fragments may be simply too large to penetrate deeply into biological tissues. This is even more of a problem, when the antibody is coupled to a moiety to kill or inhibit the growth of tumour cells.

Accordingly, there exists a need in the art for improved therapeutically useful antigen recognition molecules that are capable of binding with high affinity to MHC- peptide complexes, that are easily produced in commercially useful quantities, and that are small enough to penetrate deeply into biological tissues, such as for example solid tumors.

SUMMARY OF THE INVENTION

The present invention teaches novel classes of single chain proteins that resemble soluble TCR-like molecules and their applications. In one aspect, the invention provides single domain proteins that are capable of specifically binding to an MHC-peptide complex. These proteins are preferably small, i.e., consisting of less than 240 amino acids, preferably consisting of 60 to 200 amino acids, more preferably consisting of 80 to 180 amino acids, even more preferably consisting of 100 to 140 amino acids, and most preferably consisting of 110 to 135 amino acids. The single domain protein of the invention can be an antibody, a single domain antibody, a camelid antibody, a camelid NHH antibody fragment or any functional fragment thereof.

In certain embodiments, the protein can recognize an MHC-peptide complex, in which the MHC molecule can be either an MHC class I molecule complexed with a pathogen-derived peptide (e.g., a virus-derived peptide), a tumor derived peptide, or an alloantigen-derived peptide. In other embodiments the single-domain protein of the invention can recognize an MHC-peptide complex in which the MHC molecule is a Class II molecule complex with an autoantigen-derived peptide or a tumor-derived peptide.

In other embodiments the single domain proteins of the invention that are capable of specifically binding to an MHC-peptide complex can be coupled to a moiety. A moiety can confer additional properties to a molecule of the invention. For example, the moiety can be used to inhibit cell growth and/or viability. Such moieties include, but are not limited to toxic molecules, cytostatic drugs, cytotoxic drugs, a radiolabel, antiviral drugs and/or immunomodulatory drugs. In certain embodiments, single domain proteins linked to such a moiety are capable of specifically binding to an MHC-peptide complex and can be used to kill and/or inhibit proliferation of and/or deliver a moiety to a human pathogen-infected cell or a human tumour cell. Alternatively, the moiety can be used to detect a molecule of the invention. In these embodiments, the moiety includes, but is not limited to, fluorescent molecules, chemoluminescent molecules, radionucleotides, enzymes and/or colorimetric molecules. In further embodiments, the single domain protein of the invention, when linked to a detectable moiety, can be used to detect the presence of an MHC-peptide to detect a pathogen-infected human cell a human tumour cell in vitro or in vivo. A molecule according to the invention can also be used to prevent the activation of an autoantigen-specific T cell and/or an alloantigen-specific T cell.

Also described are nucleic acids, functional part, functional derivative and/or analogue thereof encoding a single domain protein of the invention that is capable of specifically binding to an MHC-peptide complex according, and a method for the production of such a molecule by inserting into the genome of said organism one or more copies of a nucleic acid.

In yet another embodiment, a pharmaceutical composition comprising a single domain protein according to the invention can be used to treat an infectious disease (eg a virus infection), cancer, an autoimmune disease or allotransplant rejection, either as a single-agent therapy or in combination with a further treatment for such diseases. DETAILED DESCRIPTION OF THE INVENTION The present invention provides small monovalent single chain proteinaceous molecules capable of specifically interacting with an MHC-peptide complex present on the surface of a cell, and uses thereof.

I. Definitions In order that the present invention may be more readily understood, certain terms are first defined below, and additional definitions are set forth throughout the Detailed Description.

As used herein, the term "antibody" refers to polyclonal antibodies, monoclonal antibodies, humanized antibodies, deimmunized antibodies, single-chain antibodies, and fragments thereof such as F.ab, F.(ab')2, Fv, single chain Fv (scFv) and other fragments which retain the antigen binding function of the parent antibody.

As used herein, the term "monoclonal antibody" refers to an antibody composition having a homogeneous antibody population. The term is not limited regarding the species or source of the antibody, nor is it intended to be limited by the manner in which it is made. The term encompasses whole immunoglobulins as well as bivalent fragments as F.(ab')2, or monovalent fragments such as Fab, Fv fragments,

NHH fragments or scFv fragments and others which retain the antigen binding function of the antibody. By definition monovalent antibodies or antibody fragments contain a single antigen-binding unit. Monoclonal? antibodies of any eukaryotic species can be used in this invention. It is however preferred that the antibodies are of camel, dromedary, llama, or of shark origin. Alternatively, the parent antibody can be of murine, rat or human origin and is camelized to obtain characteristics specific for camelid antibodies.

"Camelid antibodies" refer to antibodies derived from a camel, dromedary, llama, or a shark, and/or a camelized antibody. Camelid antibodies can be any antibody derived from the abovementioned species, and can be an antibody dimer, a heavy chain monomer, or a NHH antibody fragment or a fraction thereof. Antigen-specifc camelid antibodies can be induced through vaccination, or alternatively, through screening naϊve antibody libraries. Monoclonal camelid antibodies or fragments thereof can be generated by using hybridoma technology, or can be generated using phage display techniques as described in detail in, for example, WO 97/49805, the contents of which are incorporated herein by reference.

"Anticalins" are single domain antigen recognition molecules that are derived from natural lipocalins or related proteins. When a lipocalin (described in detail in WO99/16873) or a related protein (for example, as described in WO03/029471) is engineered to specifically bind to a protein antigen it is defined here as an anticalin.

An "MHC-peptide complex" as used herein is defined as a complex that comprises an MHC molecule and a peptide. An MHC molecule can be either an MHC class I molecule or an MHC class II molecule. A peptide can be a small peptide that ranges from 7-11 amino acids (binding to MHC class I) or a larger peptide of 10-30 amino acids (that can bind to MHC class II molecules). Preferably the MHC molecule is an HLA molecule (i.e., a human MHC molecule). As termed herein, "single-domain protein capable of specifically binding to an

MHC-peptide complex" refers to any antibody or anticalin or a functional fragment thereof that is capable of specifically binding to a complex of an antigen-derived peptide bound in the groove of a major histocompatibility (MHC) protein. The single domain protein does not bind to the MHC molecule in the absence of the antigen-derived peptide or in the presence of an irrelevant peptide.

"Pathogen-derived peptides" and "Tumour-derived peptides" refer to peptides derived from pathogens or tumours, respectively. Such peptides can range from 8 to 30 amino acids and can be isolated from a pathogen source, generated via recombinant DΝA technology or synthesized synthetically. For example a pathogen-derived peptide can be a peptide derived from a viral protein, a peptide derived from a protein expressed by an intracellular microorganism, or a peptide derived from an allergen; a tumour- derived peptide can be a peptide derived from a tumour-specific protein (i.e., a protein that is not or at very low levels expressed on normal cells, but is expressed on a tumour cell).

II Molecules recognizing MHC-peptide complexes. A protein capable of specifically interacting with an MHC-peptide complex as described in this invention can be any antigen recognition protein, preferably a monovalent (i.e. has a single antigen-recognition site) single domain protein. The protein is preferably small, i.e., consisting of less than 240 amino acids, preferably consisting of 60 to 200 amino acids, more preferably consisting of 80 to 180 amino acids, more preferably consisting of 100 to 140 amino acids, and most preferably consisting of 110 to 135 amino acids.

A protein capable of specifically interacting with an MHCrpeptide complex can be an antibody. Preferably the antibody is selected from the group of a camelid heavy chain monomer, camelid NHH antibody fragment, murine or human single domain antibody fragments affybody, camelized ScFv, or any functional fragment thereof.

Also disclosed are chimeric, humanized and/or deimmunized versions of the abovementioned monovalent single domain proteins molecule capable of specifically interacting with an MHC-peptide complex. Chimeric antibodies are produced by recombinant processes well known in the art, and have an animal variable region and a human constant region. Humanized antibodies correspond more closely to the sequence of human antibodies than do chimeric antibodies. In a humanized antibody, only the complementarity determining regions (CDRs), which are responsible for antigen binding and specificity, are non-human derived and have an amino acid sequence corresponding to the non-human antibody, and substantially all of the remaining portions of the molecule (except, in some cases, small portions of the framework regions within the variable region) are human derived and have an amino acid sequence corresponding to a human antibody. See L. Riechmann et al, Nature; 332: 323-327 1988; U.S. Pat. No.

5,225,539 (Medical Research Council); U.S. Pat. Nos. 5,585,089; 5,693,761; 5,693,762

(Protein Design Labs, Inc.). Deimmunized antibodies are antibodies in which the antibody sequence is screened for potential MHC-binding and/or T-cell epitope encoding amino acid sequences, followed by the introduction of amino acid substitutions to minimize the number of such potential MHC-binding and/or T-cell epitope encoding amino acid sequences. This method is described in detail in WO9852976 (Biovation Ltd).

In a preferred embodiment of the invention, the single-domain protein capable of specifically binding to an MHC-peptide complex is a camelid NHH antibody fragment. Camelidae (camels, dromedaries and llamas) as well as some sharks have unusual antibodies without light chains, termed heavy chain antibodies (Hamers-Casterman C. et al., 1993 Nature 363:446-468). These antibodies are highly stable antibodies that exist as a dimer of two heavy chains that lack the CHI domain. However, they can also exist as single chain antibodies or as Nπ antibody fragments (NHH). Importantly, the CDR3 regions of the antibodies are much longer than the CDR3 regions of conventional antibodies, resulting in large protruding loops (Desmyter A. et al., 1996 Nat. Struct. Biol. 3:803-811). Camelid NHH antibody fragments are the smallest fragment of naturally occurring single-domain antibodies that have evolved to be fully functional in the absence of a light chain. Because these molecules have evolved in nature to be fully functional having a high affinity they can be discovered relatively easy. In addition, being a single-domain protein they have a unique loop structure by which they can bind into enzyme active sites and receptor clefts that make them extremely suited for application in inhibition of the active site of enzymes (Lauwereys M. et al., 1998 EMBO J. 17:3512-3520; WO 97/49805).

Using conventional immunization techniques it has been proven extremely difficult to generate antibodies that specifically bind MHC-peptide complexes. This may be explained by the fact that peptides are deeply buried inside the cleft of an MHC molecule. As a result, a protein recognizing an MHC-peptide complex will recognize a large part of a self MHC molecule, and only a small part of said protein will recognize the peptide buried in the cleft of the MHC molecule. Conventional antibodies have CDR regions that are not flexible enough to mimic T cell receptors that interact with a peptide using a flexible CDR region, and interact with an MHC molecule using relatively rigid CDR regions (Garcia K.C. et al., 1998 Science 20;279:1166-72; Garboczi, D. N. and W. E. Biddison 1999 Immunity 10: 1-7.). As an integral part of the invention, it is disclosed here that because of the long flexible protruding CDR3 region of camelid antibodies as well as the more rigid CDR1 and CDR2 regions it is now possible to easily generate camelid NHH antibody fragments that resemble T cell receptors and that are capable of binding with high affinity to MHC-peptide complexes, especially because the CDR3 region can easily reach the peptides that are embedded in the cleft of MHC molecules.

Camelid antibodies have a high degree of format flexibility and can be easily be linked to other moieties using recombinant methods. The camelid NHH antibody fragments have a high natural similarity with human antibodies and can be humanized if needed and no immunogenicity has been observed in relevant animal studies. Finally, as camelid NHH antibody fragments are small proteins, it is relatively easy to produce large amounts of these proteins. The method to generate and produce camelid NHH antibody fragments are described, for example, in WO97/49805 and in Ghahroudi M.A et al., 1997 FEBS Lett .414:521 -526., the contents of which are incorporated herein by reference. hi another preferred embodiment of the invention, the single domain protein capable of specifically interacting with an MHC-peptide complex is an anticalin. Anticalins are single domain antigen recognition molecules that are derived from natural lipocalins (Beste G. et al., 1999 Proc Natl Acad Sci USA 96:1898-1903; Korndorfer IP. et al., 2003 Proteins 53:121-129). The method to generate and use anticalins is described in detail in WO99/16873 and WO03/029471 (Pieris Proteolab AG the contents of which are incorporated herein by reference).

It should be noted that a single-domain protein capable of specifically binding to an MHC-peptide complex can recognize either an MHC-peptide complex in which the MHC molecule is an MHC-class I molecule or an MHC-peptide complex in which the MHC molecule is an MHC-class II molecule. Traditionally such peptides have been identified by studying T-cell responses to synthetic peptides. Relevant peptides can also be identified or obtained in a variety of methods known in the art. For example, but not as a way of limitation, it is possible to isolate MHC molecules from pathogen-infected cells an/or tumour cells, elute the peptides bound to the MHC molecule isolated, followed by identification of the eluted peptides (Falk K. et al., 1991 Nature 351:290- 296; Rotzchke O. and Falk K. 1991 Immunol Today 12:447-455; Rudenski AN. et al., 1991 Nαtwre 353:622-627; Chicz R.M. et al., 1992 Nature 358:764-768).

Further, it is also possible to use T-cell epitope and/or MHC-binding motif prediction programs to predict which relevant peptides of a given pathogen-derived or tumour-derived will be presented to the immune system by MHC molecules (Singh- Jasuja H. et al., 2004 Cancer Immunol Immunother. Jan 31 [Epub ahead of print]). Finally, in vitro peptide binding studies have identified relevant peptides by studying the binding of pathogen-derived peptides to isolated MHC molecules (Chen B.P., and P. Parham 1989 Nature 337:743-745; Dornmair K. et al, 1991 Proc Natl Acad Sci USA 88:1335-1338; Stevens J. et al., 1998 J Biol Chem. 273:2874-2884; Sospedra M. et al 2003, Methods 29:236-247; Koelle DM 2003 Methods 29:213-226). Thus, peptides can be generated synthetically or recombinantly, be isolated from degraded proteins, or be isolated from peptide libraries.

In certain embodiments of the invention, an MHC-peptide complex comprises a pathogen derived peptide bound to an MHC class I protein. The pathogen-derived peptide can be a virus-derived peptide or a tumour-derived peptide. Pathogen-derived peptide can be derived from a virus HTLN-1, HIN-1, HIN-II, Hepatitis-A, -B, and -C virus, CMN, EBN, influenza virus, HPN. A range of other human pathogenic viruses are listed on http://www.ncbi.nlm.nih.gov/ICTNdb/. Many virus-derived peptides that can bind to MHC molecules have been identified and are known to those skilled in the art.

In other embodiments of the invention, the MHC-peptide complex comprises a tumour-derived peptide. The tumour-derived peptide can be derived from a tumour antigen that is widely expressed in many tumours, including but not limited to, MAGE- 1, MAGE-3, H-Y antigen, ΝY-ES0, telomerase, and HER-2/neu, (Andersen M.H. et al., 2001 Cancer Res. 61:5964-5968; Pedersen L.O. et al., 1995 Current Opinion in Immunology 7, 85-96; Rammensee H.G. et al., 1995 Immunogenetics 41, 178-228; US patent publication 20030181370; Jager E. et al., 1998 J Exp Med. 187:265-270; Nonderheide R.H. et al., 1999 Immunity 10:673-679; Disis MX. et al., 2002 J Clin Oncol. 20:2624-2632; Nan den Eynde B.J. et al, 1997 Curr Opin Immunol. 9:684-693). The tumour-derived peptide can also be derived from a protein that is specifically expressed in a certain tumour, such as for example, gplOO, Melan-A, MART-1, and tyrosinase (Rammensee, H. G., 1995 Cur Op. Immunol. 7, 85-96; Rammensee H.G. et al., 1995 Immunogenetics 41,178-228; Kawakami Y. et al., 1995 J. Immunol. 154,3961- 3968), MUCl (breast cancer, Apostolopoulos N. et al., 1999 Curr Opin Mol Ther. 1:98- 103), PSA (prostate cancer, Correale P. et al, 1997 J Natl Cancer Inst.;89:293-300), CEA (Schlom j. et al., 1996 Breast Cancer Res Treat. 38:27-39) or P53 (colon cancer Houbiers J.G et al., 1993 Eur J Immunol 23, 2072-2077) and a range of other tumour- specific proteins known in the art.

In further embodiments, a pathogen-derived peptide can be derived from an alloantigen, such as those expressed by allotransplants. An alloantigen is a minor- or a major-histocompatibility antigen that is recognized by the immune system of the recipient of the transplant, resulting in transplant rejection.

In still further embodiments, MHC-peptide complexes can also comprise an autoantigen- or a tumour- derived peptide complexed to an MHC class II molecule.

In further embodiments, the invention provides a single domain protein capable of specifically binding to an MHC-peptide complex coupled to a moiety. For example, a NHH camelid antibody or functional fragment thereof, can be linked to a second or subsequent moiety to form a fusion molecule. Preferably, the moiety can be linked to a single domain protein capable of specifically binding to an MHC-peptide complex using recombinant DΝA technology, or can be linked via a chemical process. In terms of the invention, moiety can serve to confer additional properties to the protein (e.g., toxicity, inhibition of proliferation, or inhibition of protein synthesis) or to improve its biological activity (i.e., pharmacokinetic properties). Suitable moieties include a molecule with toxic activity (e.g. a ribosome inactivating protein, a radiolabel, and/or a cytostatic drug), an immunomodulatory drug, a protection molecule like polyethylene glycol (PEG), an antiviral drug, or a moiety that can be used to detect the molecule of the invention (e.g. a fluorescent label, a chemoluminescence label, a radiolabel, an enzyme or colorimetric molecule).

Ill Uses of the Single Domain Proteins of the Invention In yet another aspect, the invention provides a method to determine the binding activity of a single domain protein to an MHC-peptide complex in a biological sample. The method includes the steps of detecting the presence of a molecule using a method according the invention, and determining the levels (i.e., amount) of binding molecule- molecule (i.e., complex) in the sample. Using the methods of the invention, it is possible to monitor the efficacy of anti-viral or cancer therapies by detecting the presence of virus-infected human cells or tumour cells at different time points during treatment. Furthermore, especially in the case of leukemias, the method can be used to detect minimal residual disease. Screening technologies are known in the art for example proteomic technologies. For example see Srinivas P.R. et al., 2001 Clin Chem. 47:1901- 1911.

The invention further provides a method for the production of a diagnostic kit containing reagents useful in the methods of the invention. For example, the kit can contain reagents (e.g., single domain protein of the invention) for use in a method to detect the presence of virus-infected human cells or tumour cells in a sample, and instructions for use. Suitable basis for a diagnostic kit are known in the art.

The invention further provides a method of administering a molecule of the invention coupled to a moiety, such as a radiolabel intravenously, to detect the presence of a pathogen-infected human cell or a tumour cell in vivo.

Infectious Disease Infectious disease and cancer are significant health problems throughout the world. Although advances have been made in detection and therapy of these diseases, no vaccine or other universally successful method for prevention or treatment is currently available, underlining the need for novel forms of therapy to manage these diseases. The present invention describes the use of single domain proteins that are capable of specifically binding to an MHC-peptide complex to eliminate tumour cells and/or cells infected with intracellular pathogens.

As described herein, infection is defined by the invasion and multiplication of foreign microorganisms such as viruses, intracellular pathogens, bacteria, fungi including yeast, and parasites, in the body. Infections are generally harmful to the host, resulting in local cellular injury. A local infection may persist and spread by extension to become an acute, subacute or chronic clinical infection or disease state. Many of the frequently occurring infectious diseases of today are caused by viruses.

Only a small number of antiviral drugs are currently available for treatment of virus infections. A complication to the development of such drugs is that mutant strains of virus which are resistant to currently available antiviral drugs develop readily. In addition, for a number of viral diseases vaccines are available that are aimed at the induction of neutralizing, virus-specific antibodies. These antibodies are highly specific for the virus, and through binding to the virus prevent the virus from binding to and entering cells. Once the cell has become infected, these anti-virus antibodies cannot bind to the virus anymore.

Another clear advantage of single domain proteins of the invention which are capable of specifically binding to an MHC-peptide complex is that they are a class of proteins that can be used to eliminate cells infected with intracellular pathogens. The single domain proteins that are capable of specifically binding to an MHC-peptide complex can therefore bind to virus-infected cells and, either through ADCC or through other immunological mechanisms such as phagocytosis, kill and remove infected cells. It is also possible to couple the single domain protein recognizing a MHC-peptide complex to a moiety to directly kill the infected cell or to prevent its function (i.e., protein synthesis) or proliferation. As a result the viral load in an infected individual may decrease significantly, leading to a much quicker recovery. Cancer As with the antiviral agents described above, the development of anticancer agents for treating cancer effectively has also been problematic. Cancer therapy currently relies on a combination of early diagnosis and aggressive treatment, which may include radiotherapy, chemotherapy or hormone therapy. However, the toxicity of such treatments limits the use of presently available anticancer agents. The high mortality rate for many cancers indicates that improvements are needed in cancer prevention and treatment. In recent years, major advances in tumour immunology have led to an increased understanding of the immune responses against tumours. It is now well established that human tumour cells express antigens that are recognized by cytotoxic T lymphocytes (CTLs) (Coulie P.G. et al., 2003 Curr Opin Immunol. 15:131-137). The molecular recognition events associated with these tumour-associated immune responses involves the expression of specific peptides in complex with MHC class I molecules on the cancer cells. These tumour-specific MHC-peptide complexes present on the surface of tumour cells may also offer a unique and specific target for an antibody-based therapeutic approach. Another approach for the treatment of cancer is the use of anti-tumour antibodies

(von Mehren M. et al., 2003 Annu Rev Med. 54:343-369). However it has been shown in a number of clinical trials that the application of these antibodies is limited because of poor penetration of solid tumours (Jain R.K., and L.T. Baxter 1988 Cancer Res. 1988 48:7022-7032). This is even more of a problem, when the antibody is coupled to a moiety to kill or inhibit the growth of tumour cells. Therefore, a very small antigen- recognition molecule is highly desirable for the treatment of solid tumours.

However, conventional antibodies are relatively large proteins that consist of two heavy chains and two light chains, linked to each other by disulphide links. Fragments of antibodies such as F(ab)2 fragments, F(ab), or single chain Fv antibodies (ScFv) are smaller versions of antibodies that lack the constant regions of heavy and light chains, yet they are still dependent on the correct folding of the heavy and light chain-derived variable regions. As a result production levels of these antibodies are very low, and the resulting proteins are still quite large (appr. 260 amino acids for ScFv). In addition, F(ab) fragments and ScFv molecules generally have a lower affinity than the bivalent parent antibody. Small single domain antigen recognition molecules consisting of a single antigen-recognition chain (and not as a recombinant molecule comprised of a heavy and a light chain) have also been studied. (See Cortez-Retamozo N. et al., IntJ.Cancer 2002, 98:456-462; Skerra A. J.Mol.Recogn. 2000;13:167-187).

The single domain antigen recognition molecules of the invention, which recognizes an MHC-peptide complex, avoid the low affinity and production problems associated with Fab fragments and scFv molecules. This is because the single chain antigen recognition molecules of the invention, (e.g., a camelid antibody or an anticalin) are selected for affinity as single chain molecules and do not need to be recombinantly modified.

Accordingly, to overcome these problems in cancer therapy with antibodies, the present invention provides use of small (i.e. consisting of less than 240 amino acids) monovalent single domain proteins capable of specifically interacting with an MHC- peptide complex.to kill or inhibit the proliferation of a tumour cell. The single domain protein can have this activity alone, or may be coupled to a moiety. Suitable moieties include, but are not limited to, a molecule with toxic activity, a radiolabel, or a cytostatic drug.

Autoimmunity In autoimmune diseases, and transplant rejection the activation of the immune system is either unwanted or excessive. These diseases are highly specific for the antigens that are recognized by the immune system such as auto-antigens (autoimmune diseases), allo-antigens (transplant rejection). The current invention provides use of a single domain protein capable of specifically binding to an MHC-peptide complex to prevent the interaction of an auto- or alloantigen-specific T lymphocyte with a cell expressing an auto- or alloantigen, thus preventing the said T cells. IV. Additional Composition and Methods of the Invention In another aspect, the invention provides a nucleic acid functional part, functional derivative and/or analogue thereof encoding a monovalent single domain protein molecule capable of specifically interacting with an MHC-peptide complex. The invention also provides a vector comprising a nucleic acid according to the invention. Suitable vectors are known to one of skill in the art, for example plasmid vectors, viral vectors etc. In another aspect the invention provides a cell comprising a vector according to the invention. The invention also provides a method for the production of a molecule according to the invention, comprising inserting into the genome of said organism one or more copies of a nucleic acid according to the invention. An organism in the context of the present invention can be for example a micro-organism (e.g. Archaea, Bacteria, Cyanobacteria, Microalgae, Fungi, Yeast, Viruses, Protozoa, Rotifers, Nematodes, Micro-Crustaceans, Micro-Molluscs, Micro-Shellfish, Micro-insects etc.), a plant, a non-human animal and a plant or animal cell (e.g. artificial cell, cell culture or protoplast etc.).

The invention further provides a method of treatment of cancer, infectious diseases, autoimmune disease or allotransplant rejection by administering a single domain protein capable of specifically binding to an MHC-peptide complex according to the invention.

Molecules according to the invention can be formulated, separately or together, with a variety of pharmaceutically acceptable carriers prior to administration. As used herein, "pharmaceutically acceptable carriers" include any and all solvents, dispersion media, coatings, antibacterial and antifungal agents, isotonic and absorption delaying agents, and the like that are physiologically compatible. Preferably, the carrier is suitable for intravenous, intramuscular, subcutaneous, parenteral, spinal or epidermal administration {e.g., by injection or infusion).

Suitable basis for pharmaceutical compositions are known in the art. Pharmaceutically acceptable carriers are well known in the art and include, but are not limited to, saline, buffered saline, dextrose, water, glycerol, sterile isotonic aqueous buffer, and combinations thereof. One example of such an acceptable carrier is a physiologically balanced culture medium containing one or more stabilizing agents such as stabilized, hydrolyzed proteins, lactose, etc. The carrier is preferably sterile. In yet another aspect the invention provides administering a pharmaceutical composition according to the invention with a carrier to a suitable recipient to treat cancer, infectious diseases, autoimmune disease or allotransplant rejection. Depending on the use a therapeutic effective amount of a second substance (i.e. anti-viral drug, various cancer treatments, or vaccines comprising tumour antigens and/or antigens derived from infectious agents) can be added to the pharmaceutical composition.

EXAMPLES This invention is further illustrated by the following examples, which should not be construed as limiting. The contents of all references, patents and published patent applications cited throughout this application are hereby incorporated in entirety by reference.

Example 1 Generation of a camelid NHH antibody fragment that specifically binds to HLA- A2 complexed with CMN peptide ρp65 ΝLNPMNAT.

Llama's are immunized with 50 μg of a monomer of HLA-A2 loaded with CMN peptide pp65 ΝLNPMNAT, and are subsequently boosted on five occasions with the same amount of material. Antibody responses are checked in the serum of the immunized animals on ELISA. 10⁷ peripheral blood cells are harvested from the immunized animals with a positive antibody titer against the peptide-HLA-A2 complex, and NHH phage display libraries are constructed as described in Ghahroudi M.A et al., 1997 EERS Lett.414:521-526. The phage display libraries are panned for several rounds on immobilized peptide-HLA-A2 complex. Specific NHH clones are further selected based on their ability to bind peptide-HLA-A2 on intact cells, on their affinity, and on their inability to bind to HLA-A2 that is complexed with an irrelevant peptide. Example 2 Generation of a camelid NHH antibody fragment that specifically binds to HLA- A2 complexed with H-Y peptide FIDSYICON. Llama's are immunized with 50 μg of a monomer of HLA-A2 loaded with H-Y peptide FIDSYICQN, and are subsequently boosted on five occasions with the same amount of material. Antibody responses are checked in the serum of the immunized animals on ELISA. 10⁷ peripheral blood cells are harvested from the immunized animals with a positive antibody titer against the peptide-HLA-A2 complex, and NHH phage display libraries are constructed as described in Ghahroudi M.A et al., 1997 EERS Lett.414:521 -526. The phage display libraries are panned for several rounds on immobilized peptide-HLA-A2 complex. Specific NHH clones are further selected based on their ability to bind peptide-HLA-A2 on intact cells, on their affinity, and on their inaboility to bind to HLA-A2 that is complexed with an irrelevant peptide..

Example 3 Generation of a camelid NHH antibody fragment that specifically binds to HLA- A2 in complex with MART-1 peptide AAGIGILTN. Llama's are immunized with 50 μg of a monomer of HLA-A2 loaded with MART-1 peptide AAGIGILTN, and are subsequently boosted on five occasions with the same amount of material. Antibody responses will be checked in the serum of the immunized animals on ELISA. 10⁷ peripheral blood cells are harvested from the immunized animals with a positive antibody titer against the peptide-HLA-A2 complex, and NHH phage display libraries are constructed as described in Ghahroudi M.A et al., 1997 EERSJett.414:521-526. The phage display libraries are panned for several rounds on immobilized peptide-HLA-A2 complex. Specific NHH clones are further selected based on their ability to bind peptide-HLA-A2 on intact cells, on their affinity, and on their inaboility to bind to HLA-A2 that is complexed with an irrelevant peptide.

Example 4 Generation of an anticalin that specifically binds to HLA-A2 complexed with

CMN peptide pp65 ΝLNPMNAT. According to the methods described in Beste G. et al, 1999 Proc Natl Acad Sci USA 96:1898-1903, a random phage display library encoding bilin-binding protein (BBP) are selected for binding specifically to an immobilized monomer of HLA-A2 loaded with CMN peptide pp65 ΝLNPMNAT. The libraries are panned for several rounds on immobilized peρtide-HLA-A2 complex. Specific anticalin clones are further selected based on their ability to bind peptide-HLA-A2 on intact cells, on their affinity, and on their inability to bind to HLA-A2 that is complexed with an irrelevant peptide.

Example 5 Generation of an anticalin that specifically binds to HLA-A2 complexed with H- Y peptide FIDSYICON. According to the methods described in Beste G. et al., 1999 Proc Natl Acad Sci

USA 96:1898-1903, a random phage display library encoding bilin-binding protein (BBP) are selected for binding specifically to an immobilized monomer of HLA-A2 loaded with CMN peptide pp65 ΝLNPMNAT. The libraries are panned for several rounds on immobilized peptide-HLA-A2 complex. Specific anticalin clones are further selected based on their ability to bind peptide-HLA-A2 on intact cells, on their affinity, and on their inability to bind to HLA-A2 that is complexed with an irrelevant peptide.

Example 6 Generation of an anticalin that specifically binds to HLA-A2 complexed with MART-1 peptide AAGIGILTN. According to the methods described in Beste G. et al., 1999 Proc Natl Acad Sci USA 96:1898-1903, a random phage display library encoding bilin-binding protein (BBP) are selected for binding specifically to an immobilized monomer of HLA-A2 loaded with CMN peptide pp65 ΝLNPMNAT. The libraries are panned for several rounds on immobilized peptide-HLA-A2 complex. Specific anticalin clones are further selected based on their ability to bind peptide-HLA-A2 on intact cells, on their affinity, and on their inability to bind to HLA-A2 that is complexed with an irrelevant peptide.

Claims

CLAIMSWhat is claimed is:

1. A single domain protein that is capable of specifically binding to an MHC- peptide complex.

2. A single domain protein according to claim 1 that consists of less than 240 amino acids.

3. A single domain protein according to claims 1 that consists of about 60 to 200 amino acids.

4. A single domain protein according to claim 1 that consists of about 80 to 180 amino acids.

5. A single domain protein according to claim 1 that consists of about 100 to 140 amino acids.

6. A single domain protein according to claim 1 that consists of about 110 to 135 amino acids.

7. A single domain protein according to any of the preceding claims that is an antibody.

8. A single domain protein according to claim 7 that is a single domain antibody.

9. A single domain protein according to claims 8 that is a camelid antibody.

10. A single domain protein according to claims 9 that is a camelid NHH antibody fragment.

11. A single domain protein according to claims 1 that is an anticalin.

12. A single domain protein according to any of the preceding claims wherein said MHC-peptide complex comprises an MHC-class I molecule.

13. A single domain protein according to claim 12, wherein the MHC-peptide complex comprises a pathogen-derived peptide.

14. A single domain protein according to claims 12, wherein the MHC-peptide complex comprises a virus derived peptide.

15. A single domain protein according to claim 12, wherein the MHC-peptide complex comprises a tumor derived peptide.

16. A single domain protein according to claim 12, wherein the MHC-peptide complex comprises an alloantigen-derived peptide.

17. A single domain protein according any of claims 1-11, wherein said MHC- peptide complex comprises an MHC-class II molecule.

18. A single domain protein according to claims 17, wherein the MHC-peptide complex comprises an autoantigen-derived peptide.

19. A single domain protein according to claim 17, wherein the MHC-peptide complex comprises a tumor-derived peptide.

20. A single domain protein according to any of the preceding claims that is coupled to a moiety.

21. A single domain protein according to claim 20, wherein said moiety is selected from the group of toxic single domain proteins, cytostatic drugs, cytotoxic drugs, a radiolabel, antiviral drugs and immunomodulatory drugs

22. A single domain protein according to claim 20, wherein said moiety is a detectable molecule selected from the group of fluorescent single domain proteins, chemoluminescent single domain proteins, radionucleotides, enzymes and colorimetric single domain proteins.

23. A method to detect the presence of an MHC-peptide complex comprising, contacting the sample with a single domain protein according to any of claims

1-22; and detecting said complex in the sample to detect a pathogen-infected human cell.

24. A method to detect the presence of an MHC-peptide complex comprising, contacting the sample with a single domain protein according to any of claims 1-

22; and detecting said complex in the sample to detect a human tumour cell.

25. A diagnostic kit to detect the presence of a pathogen-infected cell or a tumour cell comprising a single domain protein according to any of claims 1-22.

26. Use of a single domain protein according to any one of claims 1-22 to detect virus-infected cells or tumor cells using a method according to claims 23-24 for a diagnostic procedure in vivo.

27. Use of a single domain protein according to any one of claims 1-14 and 20-22 to kill and/or inhibit proliferation of and/or deliver a moiety to a human pathogen-infected cell.

28. Use of a single domain protein according to any one of claims 1-12, 15, and 19- 22 to kill and/or inhibit proliferation of and/or deliver a moiety to a human tumour cell.

29. Use of a single domain protein according to any one of claims 1-12 and 16-22 to prevent the activation of an autoantigen-specific T cell and/or an alloantigen-specific T cell.

30. A nucleic acid or a functional part, functional derivative and/or analogue thereof encoding the single domain protein according to any one of claims 1-22.

31. A method for the production of a single domain protein according to any one of claims 1-22 in an organism, comprising inserting into the genome of said organism one or more copies of a nucleic acid according to claim 30.

32. Use of a single domain protein according to claims 1-14 and 20-22, to treat an infectious disease.

33. Use according to claim 30, wherein said infectious disease comprises a virus infection.

34. Use of a single domain protein according to claims 1-12, 15, 19 and 20-22 to treat cancer.

35. Use of a single domain protein according to claims 1-12, 16, 18 and 20-22 to treat an autoimmune disease.

36. Use of a single domain protein according to claims 1-12, 16, 18 and 20-22, to prevent and/or treat the rejection of an allotransplant.

37. A pharmaceutical composition comprising a single domain protein according to anyone of claims 1-22.

38. Use of a pharmaceutical composition according to claim 37 for treating an infectious disease.

39. Use of a pharmaceutical composition according to claim 37 for treating cancer

40. Use of a pharmaceutical composition according to claim 37 for treating an autoimmune disease.

41. Use of a pharmaceutical composition according to claim 37 to prevent and/or treat the rejection of an allotransplant.

42. A method of treatment of a disease comprising administrating a therapeutically effective amount of the pharmaceutical composition according to claim 37 to a suitable recipient.

43. A method of treatment according to claim 42, wherein the disease is selected from the group consisting of an infectious disease, cancer, autoimmune disease and transplant rejection.

44. A method of treatment according to claim 42, further comprising providing a further treatment for said disease.

45. A method of treatment according to claim 44, wherein further treatment comprises a cancer treatment.

46. A method of treatment according to claim 44 wherein further treatment comprises a treatment for an infectious disease.

47. A method of treatment according to claim 44 wherein further treatment comprises a treatment for a virus infection.