EP1121435A1 - Camel, an alternative translation product of the tumour antigen lage-1 - Google Patents

Abstract

The tumor-associated antigen CAMEL and DNA encoding it. The tumor-associated antigen is encoded by an open reading frame of the LAGE-1 gene. The tumor-associated antigen, immunogenic (poly)peptides derived therefrom and DNAs encoding them, are useful for cancer immunotherapy.

Description

CAMEL, AN ALTERNATIVE TRNSLATION PRODUCT OF THE TUMOUR ANTIGEN LAGE-1

The present invention relates to the field of cancer therapy, more specifically to tumor-associated antigens.

Cytotoxic T lymphocytes (CTL_S) play an important role in the defense against melanoma. Melanoma-specific CTL clones have been obtained from either tumor infiltrating lymphocytes (TIL) in vitro stimulated with cytokines, or from peripheral blood mononuclear cells (PBMC) cultured with (autologous) tumor cells. T cell responses against tumor cells are enhanced by cytokine transfection of the tumor cells, both in animal models and in in vitro human culture systems, (van Elsas et al., 1997; Gansbacher et al., 1990; Tepper et al., 1989; Fearon et al., 1990; Dranoff et al., 1993)

The antigens recognized by the tumor-specific T cells become better defined by the development of molecular cloning techniques. These T cell targets can be divided in three groups: 1) tumor-specific antigens, not expressed in healthy tissues, except testis and placenta (e.g., MAGE, BAGE, GAGE, NY-ESO-1 , LAGE-1); 2) antigens that are lineage-specific and expressed in both melanoma and melanocytes (e.g., MART-1/ Melan- A, gp100, tyrosinase) and 3) unique, mutated antigens (e.g., β-catenin, CDK4, MUM-1) (reviewed by Van den Eynde and Brichard, 1995).

By means of Representational Difference Analysis (RDA), a PCR-based method that has been used to identify genes with tissue-specific or tumor- specific expression, the LAGE-1 and NY-ES01 genes were identified as being tumor specific by screening cDNA libraries from melanoma cell lines with a primer from a cDNA clone enriched in melanoma-specific sequences (Lethe et al., 1998).

NY-ESO-1 is a gene originally identified by SEREX technology (Chen et al., 1997). It was shown to have two different reading frames (DNA sequences and derived protein sequences given in SEQ ID NO: 7 - 10), translation products of which were shown to contain epitopes of tumor specific T-cells (Jager et al., 1998; Wang et al., 1998).

It was an object of the present invention to provide a novel tumor- associated antigen.

To solve the problem underlying the invention, melanoma cell line 518A2 and its IL-2- or GM-CSF-transfectants were compared for their CTL stimulating capacity in vitro. Stimulation of autologous PBMC with the IL-2 producing melanoma cells resulted in a melanoma-specific CTL response (van Elsas et al., 1997). CTL clones derived from this culture recognized, besides autologous melanoma cell lines, also a panel of HLA-A^*0201 positive melanoma cell lines, but were not reactive with normal melanocytes. Although 518A2 was shown to express a number of antigens previously identified to be recognized by anti-melanoma CTL (van Elsas et al., 1996), the CTL clones available recognize a new melanoma-specific antigen that is immunodominant in 518A2.

In the experiments of the present invention, the target structure that was recognized by one of these CTL clones was fully characterized and named CAMEL (CTL-recognized Antigen on Melanoma). These sequences are described in the attached sequence listing as SEQ ID NO: 1 and SEQ ID NO: 2.

It was surprisingly found that CAMEL is encoded by a reading frame of the LAGE-1 ^s-cDNA (SEQ ID NO:3) that is distinct from that encoding the putative LAGE-1 protein (SEQ ID NO: 4). (This reading frame is designated ORF-1.)

In the present invention, a cDNA clone was identified that lacks the first 84 bp of the LAGE-1³ sequence (SEQ ID NO: 3) which means that it is devoid of the initiation codon at position 54 of that sequence (Fig. 2a). The first possible translation initiation site in this clone (4H8) is the ATG at position 94 of LAGE-1 ^s (SEQ ID NO: 3), which is however, not in frame with the first ATG at position 54. Therefore, the CAMEL protein (SEQ ID NO: 2) translated from the 4H8 cDNA clone is different from the putative LAGE-1 ^s protein (SEQ ID NO: 4).

In a first aspect, the present invention is directed to the tumor-associated antigen CAMEL (SEQ ID NO: 2) which is encoded by an isolated DNA molecule with the sequence as defined in SEQ ID NO: 1.

The coding sequence of CAMEL corresponds to the ORF-1 of LAGE-1 cDNA (Lethe et al., 1997; WO 98/32855).

In the present invention „ORF-1" is defined as the open reading frame starting with ATG at position 94 of SEQ ID NO:3 (LAGE-1 ^s), which corresponds to position 10 in SEQ ID NO: 1 (CAMEL), to position 96 in SEQ ID NO: 5 (LAGE-1 ^L)

In a further aspect, the present invention relates to immunogenic (poly)peptides derived from CAMEL. A first group of peptides is selected from peptides inducing a humoral immune response (induction of antibodies). Such peptides are selected by randomly choosing continuous stretches of amino acids (at least 12-15), applying them to an individual and confirming the generation of antibodies by standard immunological assays, e.g. ELISA. This group of immunogenic (poly)peptides also encompasses the entire CAMEL antigen or larger fragments thereof.

The second group of peptides, which is preferred, can be presented by MHC molecules (in humans: HLA molecules), they have the potential to induce an immune response, in particular by eliciting a CTL response.

In a preferred embodiment, immunogenic peptides which have the ability to elicit a CTL response, are selected from peptides with the sequence set forth in SEQ ID NO: 11 , 12, 24, 25 and 26. To obtain peptides that have the ability to elicit a cellular immune response, the selection of peptide sequences from a given antigen is, in the first place, based on the requirement for such peptide to bind to an MHC molecule present in the repertoire of the patient to be treated. Two classes of MHC molecules are distinguished, class I and class II. Class I molecules consist of a membrane-inserted heavy chain and a non-covalently attached light chain. In their structure, MHC class I molecules resemble a moose's head, the antlers forming a groove which is recognized by the peptide. In humans, HLA-A, B and C are the "classical" MHC class I molecules.

Additional immunogenic peptides may be identified by methods known in the art which rely on the correlation between MHC-binding and CTL induction, e.g. those used by Stauss et al., 1992, who identified candidate T-cell epitopes in human papilloma virus.

Since immunogenic peptides can be predicted based on their "peptide binding motif" synthetic peptides which represent CTL epitopes may be designed and synthesized. Several methods, which are useful in the present invention for designing peptides, have been used to identify CTL epitopes from known protein antigens.

It is well established that every MHC class I allelic product has allele-specif ic requirements for the peptide ligand that binds to its groove and that it ultimately presents. These requirements were summarized as a motif by Falk et al., 1991. A large number of MHC peptide motifs and MHC ligands have become known to date. A method to search a known protein sequence for epitopes fitting to a given class I molecule, which is based on this knowledge and which may be used in the present invention, was reviewed by Rammensee et al., 1995. It comprises the following steps: first, the protein sequence is screened for stretches fitting to the basic anchor motif (two anchors in most cases), whereby allowance should be made for some variation in peptide lengths as well as in anchor occupancy. If a motif, for example calls for 9mers with lie or Leu at the end, 10mers with a fitting C-terminus should be considered as well, and other aliphatic residues at the C-terminus, like Val or Met, should also be considered. In this way, a list of peptide candidates is obtained. These are inspected for having as many non-anchor residues as possible in common with ligand already known, or with the residues listed among the "preferred residues" or "others" on top of each motif (Table, given by Rammensee et al., 1995), for various HLA molecules. Binding assays can be performed at this stage to exclude weak binders which occur frequently among peptides conforming to a basic motif. If a detailed study on peptide binding requirements is available, the candidates can also be screened for non-anchor residues detrimental or optimal for binding (Ruppert et al., 1993). One should keep in mind, however, that pure peptide binding motifs can be misleading in the search for natural ligands, since other constraints, such as enzyme specificity during antigen processing and specificity of transporters or chaperones, are likely to contribute to ligand identity in addition to the MHC binding specificity.

This approach was successfully applied by, inter alia, Kawakami et al., 1995, to identify gp100 epitopes based on known HLA-A2.1 motifs. The validity of the method was confirmed by identifying, in parallel, the epitope regions by using COS cells transfected with cDNA fragments generated by sequential deletion and testing for T-cell reactivity, as described above.

Recently, data bases and prediction algorithms have become available that enable to predict, with great reliability, peptide epitopes that bind to HLA molecules of interest.

Examples for peptide candidates with potential immunogenicity that can be derived from the tumor-associated antigen of the present invention, are the CAMEL-derived peptides with the sequence HLSPDQGRF and LMAQEALAF for HLA-A3 or RMAVPLLRβ for HLA-A3101. Similarly, other peptides for these or for other alleles can be determined by the method mentioned above. The peptide binding can be tested in peptide binding assays. In order to determine the immunogenicity of the selected peptide or peptide equivalent, as defined below, which is the crucial parameter for peptide-based vaccine development and which in most cases strongly correlates with the stability of the peptide-MHC interaction (van der Burg et al., 1996), the methods described by Sette et al., 1994, in combination with quantitative HLA-binding assays, may be used. Alternatively, immunogenicity of the selected peptide may be checked by performing in vitro CTL induction by known methods e.g. as described below for ex vivo CTL induction.

Alternatively to peptides derived from the naturally expressed tumor antigens, functional equivalents thereof, i.e. peptides with partially altered sequences or substances mimicking peptides, e.g. "peptidomimetics" or retro-inverso peptides, may be obtained by the following methods:

To enhance the immunogenicity of the peptides, amino acid substitutions may be introduced at anchor positions to increase peptide MHC class I- binding affinity. The modified peptides are subsequently evaluated for enhanced binding and immunogenicity by screening for recognition by TIL (tumor-infiltrating lymphocytes) and CTL induction as described by Parkhurst at al, 1996, and Bakker et al., 1997.

Another method useful in the present invention to find more immunogenic peptides by screening peptide libraries with a known CTL was described by Blake et al. 1996; it suggests the use of combinatorial peptide libraries for constructing functional mimics of tumor epitopes recognized by MHC class l-restricted CTLs.

In principle, the selection of peptides capable of eliciting a cellular immune response is carried out in several steps, as described in WO 97/30721 , which disclosure is incorporated herein by reference. In short, the candidates are first tested for their binding ability to an MHC molecule; subsequently good binders are tested for immunogenicity. A general strategy for obtaining efficient immunogenic peptides has been described by Schweighoffer, 1997.

The polypeptide of the present invention or immunogenic peptides derived from its sequence, respectively, can be produced recombinantly or by peptide synthesis, as described in WO 96/10413, the disclosure of which is incorporated herein by reference. For recombinant production, a DNA molecule encoding the antigen or the CTL peptide of interest is inserted into an expression vector, transformed or transfected into an appropriate host cell, cultivated under conditions suitable for expression, recovered and purified. For chemical synthesis, various conventional techniques may be used, e.g. commercially available automatic synthesizers.

The tumor antigen of the present invention and the immunogenic peptides derived therefrom or the respective peptide equivalents are useful in cancer therapy, e.g. to induce, in the context of the appropriate MHC presenting molecule, an immunological response to tumors which express the corresponding antigen determinants. The induction of CTLs can be accomplished in vivo or ex vivo.

For in vivo induction of CTLs, a pharmaceutical composition comprising the peptide/antigen is administered to an individual suffering from a tumor associated with the respective tumor antigen in an amount sufficient to elicit an effective CTL response to the antigen-bearing tumor. Thus, the present invention provides pharmaceutical compositions for therapeutic treatment which are intended for parenteral, topical, oral or local administration. Preferably, the compositions are for parenteral administration, e.g. for intravenous, subcutaneous, intradermal or intramuscular application. The peptides/antigens are dissolved or suspended in a pharmaceutically acceptable carrier, preferably an aqueous carrier. The composition may contain additional auxiliary substances, e.g. buffering agents, etc. The peptides may be used alone or in combination with adjuvants, e.g. saponins, alumn, or, in a particularly preferred embodiment, polycations, like polyarginine or polylysine. The peptides may also be linked to components assisting CTL priming, e.g. T helper peptides, lipids or liposomes or coadministered with such components or with immunostimulating substances, e.g. cytokines (IL-2, IFN-γ). Methods and compositions for preparing and administering pharmaceutical compositions for therapeutic treatments are described in WO 95/04542 and WO 97/30721 the disclosures of which are herein incorporated by reference.

The immunogenic peptides may also be used to elicit a CTL response ex vivo. An ex vivo CTL response to a tumor expressing the antigen is induced by incubating a patient's CTL precursor cells together with antigen presenting cells and the immunogenic peptide. The thus activated CTLs are allowed to mature and expand to effector CTLs which are then readministered to the patient. Alternatively, the tumor antigen may be pulsed onto APCs which present MHC class ll-reactive peptides

(Mayordomo et al., 1995; Zitvogel et al., 1996). A suitable method for loading peptides onto cells, e.g. dendritic cells, is disclosed in WO 97/19169.

The peptides of the invention are preferably applied as a combination of peptides, e.g. different CAMEL-peptides. In an even more preferred embodiment, the peptides of the invention are combined with peptides derived from other tumor antigens, e.g. LAGE-1 and ESO-NY-1. The selection of the peptides is optimized towards covering multiple HLA types in order to be useful for a broad population of patients and/or towards a broad variety of malignancies, which is taken into account by combining peptides from a large variety of tumor antigens. The number of peptides suitable to be combined to yield an efficient therapy may vary within a broad range, e.g. from about 2 to approximately 100.

In a further aspect, the present invention is directed to an isolated DNA molecule with the sequence set forth in SEQ ID NO: 1 encoding CAMEL. This DNA molecule, which is designated "CAMEL-DNA", contains the ORF-1 of LAGE-1 cDNA which is defined by nucleotides 54 - 336 of the sequence set forth in SEQ ID NO: 3.

The CAMEL-DNA of the present invention, or the corresponding RNA, may be used, as an alternative to the use of the protein or the peptide, for cancer immunotherapy. Alternatively to using the natural sequence or fragments thereof, engineered derivatives may be utilized. These include sequences modified to encode (poly)peptides with improved immunogenicity, e.g. taking into account the modifications described above for the peptides. Another form of modification is the assembly of multiple sequences encoding immunologically relevant peptides in a so-called string-of-beads fashion, as described by Toes et al., 1997. The sequences may also be modified by adding auxiliary coding elements, e.g. targeting functions that ensure more efficient delivery and processing of the immunogen (e.g. Wu et al., 1995).

The nucleic acid molecules may be delivered either directly or as part of a recombinant virus or bacterium. In principle, any method that is known for gene therapy may be applied for nucleic acid-based cancer immunotherapy, both in vivo and ex vivo.

Examples for in vivo delivery are direct injection (injection of "naked" DNA) either intramuscularly or by "gene gun", which has been shown to result in the generation of CTLs against tumor antigens. Examples for recombinant organisms are vaccinia virus, fowlpox virus and adenovirus or Listeria monocytogenes (see Coulie, 1997 for a comprehensive review). Furthermore, synthetic nucleic acid carriers like cationic lipids, microspheres, microbeads, liposomes may be useful for in vivo delivery of the sequence encoding respective antigen/peptide. Similarly as for peptides, various auxiliary agents that enhance the immune response may be co-applied, e.g. cytokines, either as proteins or as plasmids encoding these. Examples for ex vivo delivery are transfection of dendritic cells (Tuting, T., 1997) or other antigen presenting cells which are applied as a cellular cancer vaccine.

The present invention is also directed to the use of cells that express the tumor-associated antigen of the invention, either naturally or upon transfection with the respective coding sequence, for the preparation of a tumor vaccine.

In the present invention, it has been shown that CTL clones raised against IL-2 producing melanoma cell line 518/IL-2.14 are reactive against two alternatively spliced variants of LAGE-1 , LAGE-1 ^s (SEQ ID NO: 3) and LAGE-1 ^L (SEQ ID NO: 5) and NY-ESO-1 (SEQ ID NO: 9). NY-ESO-1 is a recently described tumor antigen, identified by screening a cDNA library of an esophagus carcinoma with autologous patient serum (SEREX-method (Chen et al., 1997)). NY-ESO-1 is expressed in different tumor types but not in healthy tissues except the testis.

In the present invention, the epitope of specific CTL 1/29 was determined by cDNA expression cloning and a truncated LAGE-1 cDNA clone was found. This truncation led to the identification of the peptide epitope in an alternative reading frame, since the "normal" translation initiation site of LAGE-1 was absent. However, COS/HLA-A*0201 cells transfected with full length LAGE-1 or NY-ESO-1 cDNA clones could stimulate the CTL clone to TNF- production as well. This probably means that two different proteins can be translated from one single mRNA.

NY-ESO-1 also has been described as the target of melanoma-specific HLA-A*0201 restricted CTL clones, which recognize an epitope translated in ORF3, located between aa 155 and 167 (Jager et al., 1998). Therefore, it is very likely that also LAGE-1 ^s will be recognized by these clones, but not LAGE-1 ^L, since the protein sequence is different at this part of the molecule. The CAMEL-specific CTL clones recognize a peptide in an alternative reading frame, which is encoded in both LAGE-1 and NY-ESO- 1. This means that tumor cells expressing either LAGE-1 or NY-ESO-1 can be recognized by MLMAQEALAFL-specific CTL, which might enlarge the number of tumors that can be treated with immunotherapy based on this peptide.

Brief description of the Figures:

Figure 1 : COS-7 transfection experiments with cDNA clone CAMEL and deletion constructs

a) COS-7 cells were transfected with cDNAs as indicated and tested with CTL 1/29 in a TNF-α release assay.

b) Deletion constructs of CAMEL cDNA were cotransfected with HLA- A^*0201 cDNA in COS-7, followed by a TNF-α release assay with CTL 1/29. The PCR clones contain the numbers of nucleotides of the CAMEL cDNA as indicated.

Figure 2:

a) Nucleotide alignment of cDNA clones CAMEL, LAGE-1 ^s, LAGE-1 ^Land NY-ESO-1.

b) Protein translations of the cDNA clones LAGE-1 ^s, LAGE-1 ^L and NY- ESO-1. The translation of CAMEL is identical to the translation of

LAGE-1 ^S/L in ORF1. Although ORF3 seems the most putative one, the CTL epitope is encoded in ORF1 (underlined).

Figure 3: Characterisation of peptides recognized by CTL clone 1/29

a) TNF-α release assay with predicted HLA-A^*0201 binding CAMEL peptides. Peptides as indicated were loaded on BLM, an HLA-A*0201⁺ melanoma cell line, at a concentration of 10 μg/ml and tested with CTL 1/29 in a TNF-α release assay.

b) The effects of increasing concentrations of peptides, derived from the major target epitope MLMAQEALAFL on recognition by CTL 1/29. Various concentrations of peptides as indicated were loaded on

HLA-A*0201⁺ cells and tested in a TNF-α release assay with CTL 1/29.

Figure 4: LAGE-1 ^{S L} and NY-ESO-1 encode the CTL epitope

COS/HLA-A*0201 cells were transfected with these cDNA clones and reactivity with CTL 1/29 was measured in a TNF-α release assay.

Figure 5: His-tagged CAMEL protein, synthesized in E.coli

Figure 6: Expression of LAGE-1 ^S/L and NY-ESO-1 in healthy human tissues and melanoma cell lines

a) Northern Blot analysis of the expression of LAGE-1/NY-ESO-1 in a panel of healthy human tissues as indicated. The Blot was hybridised with ³²P-dCTP-labeled LAGE-1 ^s cDNA.

b) RT-PCR for LAGE-1 and NY-ESO-1. To discriminate between LAGE-1 and NY-ESO-1 mRNA, the same panel of melanoma cell lines was analysed by RT-PCR with gene-specific primers. Melanoma cell lines as indicated were used as targets in a TNF-α release assay with CTL 1/29.

Figure 7: Immunohistochemical analysis of CAMEL expression in human tumors

Figure 8: Stabilization of HLA-A2 surface expression by synthetic peptides on T2-cells Brief description of the sequences:

SEQ ID NO: 1 : CAMEL (4H8) cDNA sequence and translation

SEQ ID NO: 2: CAMEL protein sequence SEQ ID NO: 3: LAGE-1 ^s cDNA sequence and translation

SEQ ID NO: 4: LAGE-1 ^s protein sequence

SEQ ID NO: 5: LAGE-1 ^L cDNA sequence and translation

SEQ ID NO: 6: LAGE-1 ^L protein sequence

SEQ ID NO: 7: NY-ESO-1 cDNA sequence and translation SEQ ID NO: 8: NY-ESO-1 protein sequence

SEQ ID NO: 9: NY-ESO-1 cDNA and alternative translation

SEQ ID NO: 10: protein sequence of alternatively translated NY-ESO-1

SEQ ID NO: 11 : peptide sequence of the CAMEL CTL epitope (11 -mer)

SEQ ID NO: 12: peptide sequence of the CAMEL CTL epitope (10-mer) SEQ ID NO: 13: oligonucleotide SP6F-pSV

SEQ ID NO: 14: oligonucleotide R1

SEQ ID NO: 15: oligonucleotide R2

SEQ ID NO: 16: oligonucleotide T7R-pSV

SEQ ID NO: 17: oligonucleotide F3 SEQ ID NO: 18: oligonucleotide ESO-1 B

SEQ ID NO: 19: oligonucleotide ESO-1A

SEQ ID NO: 20: oligonucleotide 4H8-A

SEQ ID NO: 21 : oligonucleotide 4H8-C

SEQ ID NO: 22 oligonucleotide CAMEL-XHO SEQ ID NO: 23 oligonucleotide CAMEL-KPN SEQ ID NO: 24 peptide CAMEL10 SEQ ID NO: 25 peptide CAMEL16 SEQ ID NO: 26 peptide CAMEL17 SEQ ID NO: 27 tyrosinase peptide SEQ ID NO: 28 MAGE-3 peptide

In the Examples, if not stated otherwise, the following materials and methods were used

a) Cell cultures

Melanoma cell lines and COS-7 cells were maintained in DMEM containing 4.5 mM glucose supplemented with 8% FCS, 2 mM L-glutamine, 100 μg/ml of each penicillin and streptomycin. Melanoma cell line 518A2 was established from the dissected metastasis of a male patient in 1985, as described before (Versteeg et al., 1988). An IL-2 producing variant, 518/IL- 2.14, was obtained by transfection of 518A2 with the IL-2 cDNA (Osanto et al., 1993). Other melanoma cells that were used as targets in TNF-α release assay are FM3.29, FM6, FM28.4 and FM55P (gifts from J. Zeuthen, Denmark), MM127, MM415, MM485 (gifts from N. Hayward, Australia), SK- MEL-23, SK-MEL-29 (obtained from T. Wolfel, Mainz), MM0221 , Mi3046/2, NA8, BLM (obtained from M. Visseren, Leiden). EBV-transformed B-LCL and the TNF-α-sensitive WEHI-164 clone 13 (a gift from Dr. P. Coulie, Brussels) were cultured in RPMI-1640, supplemented with L-glutamine and antibiotics as above, and 10% FCS.

With the IL-2 producing cell line 518/IL-2.14 and autologous peripheral blood mononuclear cells (PBMC) a CTL induction was performed, resulting in melanoma-specific HLA-A^*0201 restricted CTL clones (van Elsas et al., 1997). The identification of the epitope of one of these clones, CTL 1/29, is reported here. b) cDNA expression cloning

A cDNA library of 518/IL-2.14 was constructed in the expression vector pSVsportl (GIBCO, BRL) using the Superscript Plasmid System (GIBCO, BRL). As to that purpose, poly-A⁺ mRNA was isolated using the Fast-Track system (Invitrogen), followed by reverse-transcription with an oligo-dT/Notl primer. Sail adapters were ligated to ds-cDNA and after Notl digestion and size fractionation, cDNA fragments were cloned into the pSVsportl vector digested with Sail and Notl. After electroporation into ElectroMAX-DH10B (GIBCO, BRL) (following the manufacturers instructions) and selection for ampicilin resistance, 50-100 colonies were pooled for mini DNA isolation (QIAprep 8 plasmid kit, Qiagen). The in this way obtained cDNA pools were transfected in duplicate into COS-7 cells, together with the restriction element HLA-A^*0201 (pBJ1.neo/HLA-A^*0201 , (Lin et al., 1990)), using the DEAE-dextran method. Briefly, COS-7 cells were seeded in 96-wells flatbottom plates at 1.5x10⁴ cells per well in 100 μl DMEM, 8% FCS. After 2 hours, medium was replaced by 30 μl transfection mixture, containing 100 ng cDNA pool, 100 ng HLA-A^*0201 cDNA, 400 ng/ml DEAE-dextran and 100 μM chloroquine in serum free DMEM. Cells were incubated for 4 hours at 37°C and shocked for 2 minutes by the addition of 50 μl 10% DMSO in PBS. The shock medium was replaced by 200 μl DMEM, 8% FCS, and 48 hours later the cells were used as target cells for CTL in a TNF-α release assay.

c) Deletion constructs

Deletion constructs of cDNA clone 4H8 were obtained by PCR. PCR products were cloned in vector pCR3.uni (TA cloning system, Invitrogen). The constructs pCR-246 and pCR-464 were made with the vector-based forward primer, SP6F-pSV (SEQ ID NO: 13) and the reverse primers in cDNA 4H8, R1 (SEQ ID NO: 14) and R2 (SEQ ID NO: 15) respectively. As a control the complete 679 bp cDNA was cloned by PCR with two primers on the pSVsport vector, SP6F-pSV (SEQ ID NO: 13) and T7R-pSV (SEQ ID NO: 16), resulting in pCR-679.

d) TNF-α release assay

CTL reactivity against tumor target cells, transfected COS-7 or peptide loaded cells was measured in a TNF-α release assay. Target cells were seeded in duplicate or triplicate at 1.5-2x10⁴ cells per well in a 96-wells flat bottom plate and 1500-2000 CTL were added to each well, in a total volume of 100 μl / well (IMDM, supplemented with antibiotics and 5% FCS). After 24 hours of co-culturing of effector and target cells, 50 μl out of each well was added to a fresh 96-wells flatbottom plate, containing 50 μl

(4.5x10⁴) TNF-α-sensitive WEHI-164 cells per well in IMDM, supplemented with antibiotics, 5% FCS, 2 μg/ml Actinomycin D and 40 mM LiCI. A viability staining was performed 24 hours later by the addition of 50 μl of 3-(4,5dimethylthiazol-2-yl)-2,5 diphenyltetrazolium bromide (MTT) solution (2.5 mg/ml in PBS). After incubation for 2-4 hours at 37°C the OD5₅₀-₆5₀ was measured. TNF-α release in pg/ml was calculated from a standard with known TNF-α concentrations.

e) Northern Blot analysis

To determine expression in healthy tissues a Multiple Tissue Northern Blot was obtained commercially (Clontech). As a probe, LAGE-1 cDNA was used, labeled with γ-³²P-dCTP by use of the Mega-Prime Labeling kit (Amersham).

f) RT-PCR

cDNA synthesis was performed using oligo-dT and M-MLV reverse transcriptase (Promega). Primers used for LAGE-1 specific PCR were the F3 (SEQ ID NO: 17) and ESO-1 B primer (SEQ ID NO: 18). ESO-1 B was also used as a reverse primer in the NY-ESO-1 -specific PCR, while ESO-1A (SEQ ID NO: 19) was the forward primer in this reaction (Chen et al., 1997). Reactions were performed in a Biometra-Uno or -Trio programmed as follows: 5 minutes 95°C, 30 cycles of 1 min. 95°C, 1 min. 58°C, 1 min. 72°C, followed by 10 minutes 72°C.

g) Expression of CAMEL in E. Coli

A fragment containing the coding sequence of CAMEL was made by PCR with the following primers:

4H8-A: GAAGAACATATGCTGATGGCCCAGGAGGC (SEQ ID NO: 20) 4H8-C: TTAAAGATCTCAGAACCGCCCCTGGTCG (SEQ ID NO: 21)

This fragment was digested with Ndel and Bglll and cloned in the Ndel and BamHI sites of vector pET19b (Novagen, Madison, Wl). This vector contains a 6xHis-tag coding sequence, allowing detection of the His-tagged protein with an anti-His antibody. The pET19b-CAMEL construct was transformed into BL21 (DE3)pLysS E. coli bacteria (Novagen, Madison, Wl). After culturing the bacteria at 30°C until an OD₆oo =0.5 , IPTG (1 mM) was added to induce overexpression of the His-tagged CAMEL protein.

Samples were taken at Oh and 4h after IPTG induction and lysates of these samples were tested on a Western Blot with the Penta-His Antibody (Qiagen) according to the Western and Colony Blot protocol of the supplier. The His-tagged protein was visualized using the SuperSignal Substrate system for Western blotting (Pierce, Rockford, US).

h) Preparation of anti-CAMEL antisera

Antibodies against the CAMEL protein were raised by immunizing two rabbits with three synthetic peptides derived from hydrophobic regions of this molecule :

F4: (K)GAMLAAQERRVPRAAEV(K) (pos. 15-31 of SEQ ID NO: 2) A5: (K)GQQGPRGREEAPRGVRM(K) (pos. 36 -52 of SEQ ID NO: 2) B5: (K)KRRMEGAPAGPGGRTAA(K) (pos. 58 -73 of SEQ ID NO: 2) (The lysine residues at both termini enable the peptides to be linked to KLH.)

For rabbit no. 1 , 1 mg of each peptide was linked chemically to 2.5 mg of the carrier molecule KLH (Keyhole Limpet Hemocyanin) and after dialyzing, 0.8 mg of this protein in CFA (Complete Freund's Adjuvants) was weekly injected subcutaneously. Another rabbit (no. 2) was injected six times with the three peptides, not linked to KLH, following the scheme of 300 μg s.c. in CFA, 300 μg s.c. in IFA, 4x boost of 150 μg. The reactivity and specificity of the antisera were confirmed in ELISA and Western blot experiments. After four immunizations, antisera of both rabbits were reactive with the recombinant CAMEL protein synthesized in E.coli, but differed in their precise epitope: rabbit no. 1 produced antibodies against the CAMEL-B5 peptide, whereas the serum of rabbit no. 2 reacted with peptide F4. The antisera will further be referred to as "antiserum B5" and "antiserum F4".

j) Preparation of CAMEL-EGFP fusion proteins

The CAMEL coding sequence was fused to the Aequorea victoria -derived Green Fuorescent Protein (GFP). The CAMEL cDNA molecule was cloned into the pEGFP-N1 vector (Clontech), which contains a cDNA encoding the Enhanced, red shifted variant of GFP. In order to clone the cDNA molecule in frame with the EGFP cDNA and unidirectional, two primers were designed. The forward primer designated CAMEL-XHO (TTACJCGAGATGCTGATGGCCCAGG; SEQ ID NO: 22 ) covers the initiation codon ATG and contains an Xho1 site and the reverse primer CAMEL-KPN (AAGGJACCTTGAACCGCCCCTGGTCG; SEQ ID NO: 23) contains a mutation of the stop-codon and a Kpn1 site. The vector carrying the fusion construct was transfected into COS cells by calcium phosphate precipitation, protein lysates of the cells were used for Western blotting using CAMEL antisera against the CAMEL peptides B5 and F4, and anti- EGFP antibodies to detect the fusion protein according to standard protocols. Example 1

cDNA clone 4H8 (CAMEL) encodes the target for melanoma-specific CTL1/29

The antigenic epitope of melanoma-specific CTL 1/29 was identified by the expression of cDNA library 518/IL2.14 and the restriction element HLA- A*0201 in COS-7 cells, followed by CTL screening in a TNF-α release assay. A positive pool of cDNAs was subcloned and clone 4H8, called CAMEL (SEQ ID NO: 1), was found to stimulate TNF-α release by the CTL to a similar extent as the original 518/IL2.14 cell line (Fig. 1). COS-7 cells or COS-7 cells transfected with HLA-A^*0201 or the 4H8 cDNA only were not recognized. The isolated 4H8 cDNA clone has a 679 bp insert, which shows strong homology with NY-ESO-1 (SEQ ID NO: 7), a tumor antigen originally identified as a target for humoral immune responses by serum screening methods (SEREX) (Chen et al., 1997). Colony hybridization of the cDNA library, using clone 4H8 as a probe resulted in the detection of 2 types of full length clones which were called LAGE-1 ^s (SEQ ID NO: 3) and LAGE-1 ^L (SEQ ID NO: 5) (Fig. 2a). LAGE-1 ^L contains a 229 bp insertion at position 457, which has the consensus sequences for an intron, starting with a 5' GT and ending 3' AG. This indicates alternative splicing of LAGE-1 mRNA. However, cDNA clone 4H8 lacks the first 84 bp of the LAGE-1 cDNA sequence.

Example 2

The peptide epitope of CTL 1/29 is coded in an alternative reading frame of LAGE-1 or NY-ESO-1

To identify which peptide was recognized by CTL 1/29, deletion constructs of cDNA 4H8 were transfected in HLA-A*0201⁺ COS-7 cells and tested in a TNF-α release assay. CTL reactivity was measured with all constructs (Fig. 1 b), indicating that the epitope was coded within the first 330 bp of clone 4H8. An HLA-A^*0201 binding motif search was performed on the predicted protein sequence of that region (Drijfhout et al., 1995; D'Amaro et al., 1995), presuming that the ATG at position 10 in 4H8 functions as the translation initiation site. Predicted strong binding peptides at regions 1-11 , 2-11 , 1 -9, 10-18, 11-19, 16-25, 17-25, 49-57, 55-63 and 70-78 of the CAMEL protein sequence (Fig. 2b) were added to HLA-A*0201⁺ BLM melanoma cells, and tested for CTL reactivity in a TNF-α release assay (Fig. 3b).

At a peptide concentration of 10 μg/ml only the N-terminal 11- and 10-mer peptides (M)LMAQEALAFL (SEQ ID NO: 11 and NO: 12) induced preponderant recognition by CTL 1/29 (Fig. 3a), indicating that the epitope recognized by the CTL is located in the first 11 amino acids of the CAMEL- encoded protein. Closer inspection of peptides derived of this N-terminal 11-mer in a peptide concentration dependent TNF-α release assay (Fig. 3b) revealed that the methionine at position 1 as well as the leucine at position 11 are of crucial importance for reconstituting CTL reactivity. Deletion of either of these amino acids leads to an at least 5 decades higher peptide concentration required for comparable TNF-α release. The only other peptide showing weak activity is the 3-11 MAQEALAFL. In contrast, the MLMAQEALA has no activity at all (Fig. 3b), suggesting that the C-terminal amino acids FL do significantly contribute to the recognition by the CTL.

Example 3

Comparison of CAMEL, LAGE-1 ^{S L}, NY-ESO-1

As already mentioned, cDNA clone 4H8 lacks the first 84 bp of the LAGE-1 ^s sequence, which means that it is devoid of the initiation codon at position 54 of that sequence (Fig. 2a). The first possible translation initiation site in 4H8 corresponds with the ATG at position 94 of LAGE-1 ^s, which is however, not in frame with the first ATG at position 54. Therefore, the protein translated from the 4H8 cDNA clone is different from the putative LAGE-1 protein, since translation takes place in another reading frame (Fig. 2a and b). 4H8 encodes a protein of 109 amino acids (SEQ ID NO: 2) with a predicted molecular weight of 11.7 kD. The LAGE-1 ^s protein translated from the first ATG will be a 180 aa protein of 18.2 kD (SEQ ID NO: 4), while the unspliced variant, LAGE-1 ^L, encodes a 210 aa protein of 21.1 kD (SEQ ID NO: 6). NY-ESO-1 protein (SEQ ID NO: 8) is probably of the same size as LAGE-1 ^s, but differs at 26 amino acids. However, if translation of LAGE- 1^S/L starts at the second ATG, proteins will be translated in another reading frame and are in that case identical to the protein translated from cDNA 4H8. Alternative translation of NY-ESO-1 (SEQ ID NO: 9 and NO: 10) results in a shorter variant of this protein (58 amino acids), because of an earlier stop codon (Fig. 2b), which differs from the CAMEL protein sequence only in its last 5 amino acids (Fig. 2b).

It was examined whether cells transfected with the complete LAGE-1 (or NY-ESO-1) cDNA clones are able to stimulate CTL 1/29. Remarkably,

COS/HLA-A^*0201 cells transfected with LAGE-1 ^s, the alternatively spliced LAGE-1 ^L (as well as with the NY-ESO-1) cDNA are able to stimulate CTL 1/29 (Fig. 4). This indicates that, at least in COS-7 cells, protein translation also starts from the second start codon at nucleotide 94 in LAGE-1 ^s, notwithstanding the presence of the first ATG at position 54. Also in this case, this results in the "alternative reading frame" peptide, MLMAQEALAFL, recognized by CTL 1/29.

Example 4

Expression of CAMEL in E. Coli

To investigate whether CAMEL is indeed translated from the ORF-1 of the CAMEL (4H8) cDNA, the CAMEL cDNA (SEQ ID No: 1) was cloned in a bacterial expression vector (pET19b) (Studier et al., 1990). This vector contains a 6xHis-tag coding sequence, allowing detection of the His-tagged protein with an anti-His antibody. The pET19b-CAMEL construct was transformed into E.coli and the bacteria were treated with IPTG to induce expression of the His-tagged CAMEL protein. Extracts were analyzed by Western blotting using the Penta-His antibody. Western blotting of a lysate shows a 15.5 kD protein, only slightly higher than the expected 14.5 kD of the His-tagged CAMEL protein after staining with a anti-His antibody (Hg. 5).

The CAMEL cDNA (SEQ ID No: 1) was cloned in pET19b and expressed in E.Coli. Lanes 1 and 2 represent the samples taken at Oh, lanes 3 and 4 at 4h after induction with IPTG. Because CAMEL might be an unstable protein, induction of protein expression was performed in the absence

(lanes 1 and 3) or presence (lanes 2 and 4) of PMSF (a protease inhibitor). At the left the positions of the molecular weight marker proteins are indicated.

Example 5

Expression of LAGE-1 and NY-ESO-1 in healthy human tissues and melanoma cell lines

Hybridisation of Multiple Tissue Northern blots containing RNA of healthy human tissues with the LAGE-1 ^s cDNA showed high expression in testis and placenta and low, (but clear) expression in heart, skeletal muscle and pancreas (Fig. 6a). The positive signals exist of two bands, probably reflecting LAGE-1 ^s/NY-ESO-1 (750 bp) and LAGE-1 ^L (1000 bp).

Several melanoma cell lines were tested for expression of LAGE-1 and NY-ESO-1 by(Northern Blot analysis and) RT-PCR (Fig. 6b). Because of the strong homology between both genes, it is not possible to discriminate between LAGE-1 and NY-ESO-1 on Northern Blot. Therefore RT-PCR was performed with specific primers. In most cell lines a correlated expression of LAGE-1 and NY-ESO-1 was found; only cell line FM3.29 had expression of LAGE-1 , but was negative for NY-ESO-1. Other cell lines expressed either both or none of the two genes (Fig. 6b). There was a good correlation between the level of expression and the recognition by CTL 1/29 (Fig. 6b).

Example 6

Determination of LAGE1/CAMEL expression of in human tumors by RT- PCR

In order to evaluate the percentage of LAGE1/CAMEL positive human tumors, individual tumor tissues from breast or lung cancer patients were subjected to RT-PCR, as described in the Method section f), using the LAG E1 -specific primers F3 (SEQ ID NO: 17) and ESO-1 B (SEQ ID NO: 18); reactions were performed in a Perkin Elmer 9600 thermocycler with 35 (instead of 30) cycles.

As shown in Table 1 , approximately 50% (3/6) of the tested breast cancer specimens and 80% (10/12) of the lung cancer specimens were shown to be positive for LAGE1 /CAMEL mRNA .

Table 1 :

Example 7

Immunohistochemical analysis of CAMEL expression in human tumors

As described in Examples 5 and 6, LAGE-1 mRNA was detected in a panel of tumor cell lines and tumor tissues and in a restricted number of healthy tissues by means of RT-PCR and Northern blot experiments. However, it remained to be determined whether in cells expressing the LAGE-1 mRNA also the alternatively translated CAMEL is produced.

Frozen sections from a panel of different human tumors were analyzed by immunohistochemistry using a CAMEL-specific rabbit antiserum B5 which was affinity-purified against the B5 peptide. (For the preparation of the antisera, see h) in the Methods section. The B5 antiserum was used because B5 is CAMEL-specific, while the F4 antiserum may also recognize an epitope present on a protein expressed from the ORF-1 of NY-ESO-1). Specificity of the purified serum was demonstrated by peptide ELISA and Western blotting against COS cells transfected with a CAMEL-GFP fusion protein. Immunohistochemistry was performed using a 3-step avidin-biotin- peroxidase staining procedure. The results are summarized in Table 2, examples are shown in Figure 7.

Table 2:

AC: adenocarcinoma

SCC: squamous cell carcinoma

A total of 48 different specimen was investigated. In the majority of cases (38/48) CAMEL expression was detected. About half of the positive cases showed expression of CAMEL in 40% or more of the tumor cells, in some of the cases close to 100% of the tumor cells showed CAMEL-specific staining (an example is shown Figure 7, Colon AC). In the majority of tumor specimens expression was heterogeneous ranging from less than 10% of positive tumor cells to more than 70% of positive tumor cells (Table 2; examples are shown in Figure 7, arrows indicate positive tumor cell staining).

Example 8

Identification of HLA-A2 binding peptides within the CAMEL ORF

In order to identify further HLA-A2 epitopes besides the CTL-epitope (M)LMAQEALAFL (SEQ ID NO:11 and 12), CAMEL (SEQ ID NO:2) was examined according to the algorithms and motifs published by Parker et al.; 1994, and Rammensee et al., 1995. The result of this analysis indicated that several further peptides within the CAMEL protein have the potential to bind to HLA-A2, and three of these candidates -CAMEL10: FLMAQGAML (SEQ ID NO: 24), CAMEL16: AMLAAQERRV (SEQ ID NO: 25) and CAMEL17: MLAAQERRV (SEQ ID NO:26) - were synthesized.

These synthetic peptides were evaluated for their ability to increase surface HLA-A2 expression on the transport defective cell Nne174CEM.T2 (Nijman et al., 1993). Briefly, 5x10⁵ cells/0.2 ml/well were seeded in 96-well V-bottom plates and incubated for 16 hours with increasing ammounts (0-320 μg/ml) of peptide at 37°C in a humidified atmosphere. HLA-A2 surface expression was measured by FACS analysis (Becton Dickinson) using purified BB7.2 as primary antibody and a goat-anti-mouse IgG RPE conjugate(DAKO) as detection antibody.

As positive controls the known HLA-A2 restricted CTL-epitopes from CAMEL (MLMAQEALAFL, SEQ ID NO:11) or tyrosinase (Wδlfel et al., 1994; YMNGTMSQV, SEQ ID NO:27) were applied. Negative controls included an HLA-A1 binding (and therefore irrelevant) peptide from MAGE-3 (Gaugler et al. , 1994; EVDPIGHLY, SEQ ID NO:28) or no peptide at all.

The results from these experiments suggest that the nonamer CAMEL10 binds with similar affinity to HLA-A2 as compared to the positive controls used in the assay. The two other peptides (CAMEL16 and CAMEL17) showed only low affinity in this assay. Therefore in particular CAMEL10 represents a potential new HLA-A2 restricted CTL-epitope derived from CAMEL protein (FIG. 8).

The testing of the immunogenicity of CAMEL10 and if it represents a naturally processed and presented ligand can be done as described in WO 97/30721 and Schweighoffer, 1997. Bibliography

Bakker, A.B.H., van der Burg, S.H., Huijbens, R.J.F., Drijfhout, J.-W., Melief, J.M., Adema, G.J., and Figdor, C.G. (1997). Analogues of CTL epitopes with improved MHC Class-I Binding capacity elicit anti-melanome CTL recognizing the wild-type epitope. Int. J. Cancer 70, 302-309.

Blake, J., Johnston, JN., Hellstrόm, K.E., Marquardt, H., and, Chen, L. (1996). Use of Combinatorial Peptide Libraries to Construct

FunctionalMimics of Tumor Epitopes Recognized by MHC Class I- Restricted Cytolytic T Lymphocytes. J. Exp. Med. 184, 121-130.

Chen, Y.T., Scanlan, M.J., Sahin, U., Tureci, O., Gure, A.O., Tsang, S., Williamson, B., Stockert, E., Pfreundschuh, M., and Old, L.J. (1997). A testicular antigen aberrantly expressed in human cancers detected by autologous antibody screening. Proc. Νatl. Acad. Sci. U. S. A. 94, 1914-1918.

Coulie P.G. (1997) Human tumour antigens recognized by T cells: new perspectives for anti cancer vaccines. Mol. Med. Today 3, 261-268

D'Amaro, J., Houbiers, J.G., Drijfhout, J.W., Brandt, R.M., Schipper, R., Bavinck, J.Ν., Melief, C.J., and Kast, W.M. (1995). A computer program for predicting possible cytotoxic T lymphocyte epitopes based on HLA class I peptide-binding motifs. Hum. Immunol. 43, 13-18.

Dranoff, G., Jaffee, E., Lazenby, A., Golumbek, P., Levitsky, H., Brose, K., Jackson, V., Hamada, H., Pardoll, D., and Mulligan, R. (1993). Vaccination with irradiated tumor cells engineered to secrete murine granulocyte-macrophage colony-stimulating factor stimulates potent, specific, and long-lasting anti-tumor immunity. Proc. Natl. Acad. Sci. USA 90, 3539-3543.

Drijfhout, J.W., Brandt, R.M., D'Amaro, J., Kast, W.M., and Melief, C.J. (1995). Detailed motifs for peptide binding to HLA-A*0201 derived from large random sets of peptides using a cellular binding assay. Hum. Immunol. 43, 1-12.

Falk, K., Rόtzschke, O., Stevanovi^'c, S., Jung, G., and Rammensee, H-G (1991). Allele-specific motifs revealed by sequencing of self-peptides eluted from MHC molecules. Nature 351, 290-296.

Fearon, E.R., Pardoll, D.M., Itaya, T., Golumbek, P., Levitsky, H.I., Simons, J.W., Karasuyama, H., Vogelstein, B., and Frost, P. (1990). lnterleukine-2 production by tumor cells bypasses T helper function in the generation of an antitumor response. Cell 60, 397-403.

Gansbacher, B., Zier, K., Daniels, B., Cronin, K., Bannerji, R., and Gilboa, E. (1990). Interleukin 2 gene transfer into tumor cells abrogates tumorigenicity and induces protective immunity. J. Exp. Med. 172, 1217-1224.

Gaugler B, van den Eynde B, van der Bruggen P, Romero P, Gaforio JJ, De Plaen E, Lethe B, Brasseur F, Boon T (1994) J. Exp. Med. 179:921-30

Jager, E., Chen, Y.T., Drijfhout, J.W., Karbach, J., Ringhoffer, M., Jager, D., Arand, M., Wada, H., Noguchi, Y., Stockert, E., Old, L.J., and Knuth, A. (1998). Simultaneous humoral and cellular immune response against cancer-testis antigen NY-ESO-1 : definition of human histocompatibility leukocyte antigen (HLA)-A2-binding peptide epitopes. J. Exp. Med. 187, 265-270. Kawakami, Y., Eliyahu, S., Jennings, C, Sakaguchi, K., King, X., Southwood, S., Robbins, P.F., Sette, A., Appella, E., and Rosenberg, S.A. (1995). Recognition of multiple epitopes in the human melanoma antigen gp100 by tumor-infiltrating T lymphocytes associated with in vivo tumor regression. J. Immunol. 154, 3961-3968.

Lethe, B., Lucas, S., Michaux, L., Desmet, C, Godelaine, D., Serrano, A., Deplaen, E. and Boon, T. (1998). LAGE-1 , a new gene with tumor specificity. Int. J. Cancer 76 (6):903-908.

Lin, A.Y., Devaux, B., Green, A., Sagerstrom, C, Elliott, J.F., and Davis, M.M. (1990). Expression of T Cell Antigen Receptor Heterodimers in a Lipid-Linked Form. Science 249, 677-679.

Mayordomo, J.I., Zorina, T., Storkus, W.J., Zitvogel, L., Celluzzi, C, Falo, L.D., Melief, C.J., lldstad, ST., Kast, W.M., DeLeo, A.B., and Lotze, M.T. (1995). Bone marrow-derived dendritic cells pulsed with synthetic tumour peptides elicit protective and therapeutic antitumour immunity. Nature Medicine 1, 1297-1302.

Nijman HW, Houbiers JG, Vierboom MP, van der Burg SH, Drijfhout JW, D^'Amaro J, Kenemans P, Melief CJ, Kast WM (1993) Eur. J. Immunol. 6: 1215-9

Osanto, S., Brouwenstyn, N., Vaessen, N., Figdor, C.G., Melief, C.J., and Schrier, P.I. (1993). Immunization with interleukin-2 transfected melanoma cells. A phase l-ll study in patients with metastatic melanoma. Hum. Gene Ther. 4, 323-330.

Parker KC, Bednarek MA, Coligan JE (1994), J. Immunol. 152:163 Parkhurst, M.R., Salgaller, M.L, Southwood, S., Robbins, P.F., Sette, A., Rosenberg, S.A., and Kawakami, Y. (1996). Improved induction of melanoma-reactive CTL with peptides from the melanoma antigen gp100 modified at HLA-A^*0201 -binding residues. J. Immunol. 157, 2539-2548.

Rammensee, H.G. and Friede, T. (1995). MHC ligands and peptide motifs: first listing. Immunogenetics 41, 179-228.

Ruppert, J., Sidney, J., Celis, E., Kubo, R.T., Grey, H.M., and Sette, A. (1993). Prominent role of secondary anchor residues in peptide binding to HLA-A2.1 molecules. Cell 74, 929-937.

Sambrook, J., Fritsch, E.F., and Maniatis, T. (1989). Molecular Cloning, a Laboratory Manual (New York: Cold Spring Harbor Laboratory Press).

Schweighoffer, T. (1997) Molecular cancer vaccines: tumor therapy using antigen-specific immunizations. Path. One. Res. 3, 164-176

Sette, A., Vitiello, A., Reherman, B., Fowler, P., Nayersina, R., Kast, W.M., Melief, C.J.M., Oseroff, C, Yuan, L., Ruppert, J., Sidney, J., del Guercio, M.-F., Southwood, S., Kubo, R.T., Chesmut, R.W., Grey, H.M., and Chisari, F.V. (1994). The relationship between class I binding affinity and immunogenicity of potentioal cytotoxic T Cell epitopes. J. Immunol. 753, 5586-5592.

Stauss, H.J., Davies, H., Sadovnikova, E., Chain, B., Horowitz., and Sinclair, C. (1992). Induction of cytotoxic T lymphocytes with peptides in vitro: Identification of candidate T-cell epitopes in human papilloma virus. Proc. Natl. Acad. Sci. USA. 89, 7871 -7875. Studier, F.W., Rosenberg, A.H., Dunn, J.J. and Dubendorf, J.W. (1990). Use of T7 RNA polymerase to direct expression of cloned genes. Methods in Enzymology 185, 60-89

Tepper, R.I., Pattengale, P.K., and Leder, P. (1989). Murine interleukine-4 displays potent anti-tumour activity in vivo. Cell 57, 503-512.

Toes, R.E., Hoeben, R.C., Van der Voort, E., Ressing, M.E., Van-der-Eb, A.J. Melief, C.J.M., and Offringa, R. 1997 Protective anti-tumor immunity induced by vaccination with recombinant adenoviruses encoding multiple tumor-associated cytotoxic T lymphocyte epitopes in a string-of-beads fashion. Proc. Natl. Acad. Sci. U.S.A. 94 (26): 14660-14665

Tuting, T., DeLeo, A.B., Lotze, M.T., and Storkus, W.J. Genetically modified bone marrow-derived dendritic cells expressing tumor-associated viral or "self" antigens induce antitumor immunity in vivo. Eur. J. Immunol. 27, 2702-2707, 1997.

van der Burg, S.H., et al., (1996), Immunogenicity of peptides bound to MHC class I molecules depends on the MHC-peptide complex stability. J. Immunol. 756, 3308-3314

Van den Eynde, B. and Brichard, V.G. (1995). New tumor antigens recognized by T cells. Current Opinion in Immunology 7, 674-681.

van Elsas, A., van der Minne, C.E., Borghi, M., van der Spek, C.W., Braakman, E., Osanto, S., and Schrier, P.I. (1996). CTL Recognition of an IL-2 Producing Melanoma Vaccine. In: Immunology of human melanoma. Tumor-host interaction and immunotherapy, edited by M. Maio, Amsterdam: IOS, 1996, p. 165-173. van Elsas, A., Aamoudse, C, van der Minne, C.E., van der Spek, C.W., Brouwenstijn, N., Osanto, S., and Schrier, P.I. (1997). Transfection of IL-2 augments CTL response to human melanoma cells in vitro: immunological characterization of a melanoma vaccine. Journal of Immunotherapy 20, 343-353.

Versteeg, R., Noordermeer, I.A., Krϋse-Wolters, K.M., Ruiter, D.J., and Schrier, P.I. (1988). c-Myc downregulates class I HLA expression in human melanomas. EMBO J. 7, 1023-1029.

Wang, R.F., Johnston, S.L., Zeng,G., Topalian,S.L., Schwartzentruber.D.J., and Rosenberg,S.A. (1998). A breast and melanoma-shared tumor antigen: T cell responses to antigenic peptides translated from different open reading frames. J. Immunol. 161, 3596-3606.

Wόlfel T, van Pel A, Brichard V, Schneider J, Seliger B, Meyer zum Bϋschenfelde KH, Boon T (1994) Eur. J. Immunol. 24:759-64

Wu, T.C., Guamieri, F.G., Staveley-O'Carroll, K.F., Viscidi, R.P., Levitsky, H.I., Hedrick, L, Cho, K.R., August, J.T., and Pardoll, D.M. (1995).

Engineering an intracellular pathway for major histocompatibility complex class II presentation of antigens. Proc. Natl. Acad. Sci. U.S.A. 02(25): 11671 -11675.

Zitvogel, L., Mayordomo, J.I., Tjandrawan, T., DeLeo, A.B., Clarke, M.R., Lotze, M.T., and Storkus, W.J. (1996). Therapy of murine tumors with tumor peptide-pulsed dendritic cells: dependence on T cells, B7 costimulation, and T helper cell 1 -associated cytokines. J. Exp. Med. 183, 87-97.

Claims

Claims

1. Tumor-associated antigen CAMEL which is encoded by the ORF-1 of LAGE-1 cDNA and has the amino acid sequence set forth in SEQ ID NO: 2.

2. The tumor-associated antigen of claim 1 for use in cancer therapy.

3. Immunogenic (poly)peptide derived from the tumor-associated antigen of claim 1.

4. An immunogenic peptide of claim 3, characterized in that it has the amino acid sequence Met Leu Met Ala Gin Glu Ala Leu Ala Phe Leu

(SEQ ID NO: 11).

5. An immunogenic peptide of claim 3, characterized in that it has the amino acid sequence Leu Met Ala Gin Glu Ala Leu Ala Phe Leu (SEQ ID NO: 12).

6. An immunogenic peptide of claim 3, characterized in that it has the amino acid sequence Phe Leu Met Ala Gin Gly Ala Met Leu (SEQ ID NO: 24).

7. An immunogenic peptide of claim 3, characterized in that it has the amino acid sequence Ala Met Leu Ala Ala Gin Glu Arg Arg Val (SEQ ID NO: 25).

8. An immunogenic peptide of claim 3, characterized in that it has the amino acid sequence Met Leu Ala Ala Gin Glu Arg Arg Val

(SEQ ID NO: 26).

9. An immunogenic peptide of any one of claims 3 to 8 for use in cancer immunotherapy.

10. Pharmaceutical composition containing the tumor-associated antigen CAMEL of claim 1.

11. Pharmaceutical composition containing an immunogenic peptide of any one of claims 3 to 8.

12. Isolated DNA molecule comprising the sequence set forth in SEQ ID NO: 1.

13. Recombinant DNA molecule comprising the DNA molecule of claim 12.

14. A DNA molecule of claim 12 or 13 for use in cancer immunotherapy.