WO2004012817A2 - Use of genes identified to be involved in tumor development for the development of anti-cancer drugs - Google Patents

Abstract

The present invention relates to the use of inhibitors directed against the expressed proteins as well as the murine genes and/or their human orthologues listed in Table 1 for the preparation of a therapeutic composition for the treatment of cancer, in particular for the treatment of solid tumors of lung, colon, breast, prostate, ovarian, and pancreas as well as leukemia and lymphoma. The invention also relates to the therapeutic compositions comprising the inhibitors and to methods for development of the inhibitor compounds. Such therapeutic compositions can consist of small molecule inhibitors, antibodies, antisense molecules, RNA interference (RNAi) molecules and gene therapies against these genes and/or their expression products.

Description

USE OF GENES IDENTIFIED TO BE INVOLVED IN TUMOR DEVELOPMENT FOR THE DEVELOPMENT OF ANTI-CANCER DRUGS

The present invention relates to the use of the murine genes identified by retroviral insertional tagging as well as their human orthologues for the identification and development of anti-cancer drugs, like small molecule inhibitors, antibodies, antisense molecules, RNA interference (RNAi) molecules and gene therapies against these genes and/or their expression products, and especially anti-cancer drugs effective against solid tumors of e.g. lung, colon, breast, prostate, ovarian, and pancreas as well as leukemia and lymphoma. The invention further relates to pharmaceutical preparations comprising one or more of said inhibitors and methods for the treatment of cancer using said pharmaceutical preparations .

After a quarter century of rapid advances, cancer research has generated a rich and complex body of knowledge revealing cancer to be a disease involving dynamic changes in the genome. Cancer is thought to result from at least six essential alterations in cell physiology that collectively dictate malignant growth: self-sufficiency in growth signals, insensitivity to growth-inhibitory (anti-growth) signals, evasion of programmed cell death (apoptosis) , limitless replicative potential, sustained angiogenesis, and tissue invasion and metastasis.

In general, these essential alterations are the result of mutations in genes involved in controlling these cellular processes. These mutations include deletions, point mutations, inversions, amplifications, and translocations . The mutations result in either an aberrant level, timing, and/or location of expression of the encoded protein or a change in function of the encoded protein. These alterations can affect cell physiology either directly, or indirectly, for example via signaling cascades.

Identifying genes which promote the transition from a normal cell into a malignant cell provides a powerful tool for the development of novel therapies for the treatment of cancer.

One of the most common therapies for the treatment of cancer is chemotherapy. The patient is treated with one or more drugs which function as inhibitors of cellular growth and which is thus intrinsically toxic. Since cancer cells are among the fastest growing cells in the body, these cells are severely affected by the drugs used. However, also normal cells are affected resulting in, besides toxicity, very severe side-effects like loss of fertility. ^• Another commonly used therapy to treat cancer is radiation therapy. Radiotherapy uses high energy rays to damage cancer cells and this damage subsequently induces cell cycle arrest. Cell cycle arrest will ultimately result in programmed cell death (apoptosis) . However, also normal cells are irradiated and damaged. In addition, it is difficult to completely obliterate, using this therapy, all tumor cells. Importantly, very small tumors and developing metastases cannot be treated using this therapy. Moreover, irradiation can cause mutations in the cells surrounding the tumor which increases the risk of developing new tumors. Combinations of both therapies are frequently used and a subsequent accumulation of side-effects is observed.

The major disadvantage of both therapies is that they do not discriminate between normal and tumor cells. Furthermore, tumor cells have the tendency to become resistant to these therapies, especially to chemotherapy. Therapies directed at tumor specific targets would increase the efficiency of the therapy due to i) a decrease in the chance of developing drug resistance, ii) the drugs used in these tumor specific therapies are used at much lower concentrations and are thus less toxic and iii) because only tumor cells are affected, the observed side-effects are reduced.

The use of tumor specific therapies is limited by the number of targets known. Since tumors mostly arise from different changes in the genome, their genotypes are variable although they may be classified as the same disease type. This is one of the main reasons why not a single therapy exists that is effective in all patients with a certain type of cancer. Diagnosis of the affected genes in a certain tumor type allows for the design of therapies comprising the use of specific anti-cancer drugs directed against the proteins encoded by these genes .

Presently, only a limited number of genes involved in tumor development are known and there is a clear need for the identification of novel genes involved in tumor development to be used to design tumor specific therapies and to define the genotype of a certain tumor.

In the research that led to the present invention, a number of genes were identified by proviral tagging to be involved in tumor development. Proviral tagging is a method that uses a retrovirus . to infect normal vertebrate cells. After infection, the virus integrates into the genome thereby disrupting the local organization of the genome. This integration is random and, depending on the integration site, can affect the expression or function of nearby genes. If a gene involved in tumor development is affected, the cell has a selective advantage to develop into a tumor as compared to the cells in which no genes involved in tumor development are affected. As a result, all cells within the tumor originating from this single cell will carry the same proviral integration. Through analysis of the region nearby the retroviral integration site, the affected gene can be identified. Due to the size of the genome and the total number of integration sites investigated in the present invention, a gene that is affected in two or more independent tumors must provide a selective advantage and therefore contribute to tumor development. Such sites of integration are designated as common insertion sites (CIS). Importantly, several genes identified using this approach are well-known cancer genes (e.g. Myc, Myb, Tpl2, Ccndl , and Ccnd2) with demonstrated functional relevance for human cancer. This finding validates the approach indicating that novel genes that are involved in tumor development can be identified using this strategy.

The genes claimed in the present invention are all identified by common insertion sites and are so far not reported to be involved in tumor development. The novel cancer genes that were identified in this manner are the following: Cclδ , Cd83 Lγl 08_f Sdc4 Selpl , and Sema 4b, encoding cell-surface proteins; Ggtal _r Pla2g7 _f and Rabggtb, encoding enzymes; Ak4 _r Camk2d, Camkk2 , Dgke, Mknk2 , mouse orthologue of PSK Nori2 , Ntkl _f Pim3 , and Ptkθl , encoding kinases; DUSP8, mouse orthologue of DUSP5 , and Ptpnl , encoding phosphatases; Wisp2 and Wnt5b, encoding secreted factors; Cabp2 Calm2, Coxol c, Fbxw4 , Fkbpl O _f Gnbl , Hbsll , Kifl3a _r mouse orthologue of PRAX-1 , Swap^"/ '0 , and Tia 2, encoding signaling proteins; Elf 4 , Gfilb, Hivepl , Klf3, Maz , Mef2d_r Supt4h _r Zfhxlb, and Znfnl a3, encoding proteins involved in transcriptional regulation; Cil -pending, Tsga2, genes with the following Celera identification codes mCG10088, mCG14584, mCG15383, mCG16286, mCG16752, mCG16756, mCG18310, mCG19525, mCG21612, mCG2332, mCG49753, mCG50456, mCG5049, mCG55300, mCG55520, mCG55784, mCG57225, mCG57228, mCG57816, mCG59306, mCG59312, mCG60024, mCG60113, mCG60609, mCG61858, mCG62190, mCG62286, mCG62490, mCG63480, mCG65233, mCG7764, and mCG9361. These genes are also listed in Table 1 as well as their human orthologues, if known. The term "human orthologue" as used herein should be interpreted as a human gene that is homologous to the gene identified in mouse as a result of divergence from a common ancestor. In Table 1, the genes are identified by their official murine and human gene symbol and name (unless otherwise stated) , their gene identification code as referred to in the Celera Discovery System Database (www. celeradiscoverysystem. com) , and their chromosomal localization. The genes are classified into broad functional groups according to their proposed function.

The cDNA and protein sequences of these genes, as well as their known splice variants, are given in the figures (see Figures 5, 9, 12, and 13) . The sequences were obtained from the following databases: Celera Discovery System (www. celeradiscoverysystem. com) , Ensembl (www. ensembl . org) , and National Center for Biotechnology Information (NCBI; www.ncbi .nlm.nih. gov) . Primarily, Celera sequences are given except if additional variants (e.g. due to alternative splicing) are available at the other databases. The genes claimed in the present invention comprise the cDNA and protein sequences of the genes listed in Table 1 (also given in Figures 5, 9, 12, and 13) as well as any other sequences that have diverged from a common ancestor, either naturally or as a result of utagenesis, and as a consequence share at least 90% identity to the given cDNA sequence determined using standard software packages such as BLAST software (e.g. using standard settings) from NCBI (www.ncbi .nlm.nih. gov/BLAST) .

Table 1.

The first object of the present invention to provide novel genes involved in tumor development for use in the design of tumor specific therapies is thus achieved by using the human orthologues of the murine genes of Table 1 to develop inhibitors directed against these genes and/or their expression products and to use these inhibitors for the preparation of pharmaceutical compositions for the treatment of cancer. Preferably, such therapeutic compounds and therapies consist of small molecule inhibitors, antibodies, antisense molecules, RNA interference (RNAi) molecules and gene therapies against these genes and/or their expression products .

A method for the development of therapeutic compounds according to the present invention comprises the steps of: a) identification of genes involved in cancer, in particular by using retroviral insertional tagging, optionally in a specific genetic background; b) validation of one or more of the identified genes as relevant target genes for therapeutic compounds by one or more of the following methods: determination of the expression profile of the identified genes in human tumors and normal tissues; - determination of the functional importance of the identified genes for cancer; c) development of therapeutic compounds by one or more of the following methods: production of expression products of the validated genes and use of these products for production and/or design of therapeutic compounds ; use of the gene sequence to design therapeutic compounds . In one embodiment of the present invention, the inhibitors are antibodies and/or antibody derivatives directed against the expression products of the genes listed in Table 1. Such antibodies and/or antibody derivatives such as scFv, Fab, chimeric, bifunctional and other antibody- derived molecules can be obtained using standard techniques generally known to the person skilled in the art. Therapeutic antibodies are particularly useful against gene expression products located on the cellular membrane. Antibodies may influence the function of their target proteins by for example steric hindrance or blocking at least one of the functional domains of those proteins. In addition, antibodies may be used for deliverance of at least one linked toxic compound to a tumor cell .

In a second embodiment of the present invention, the inhibitor is a small molecule capable of interfering with the function of the proteins encoded by the genes listed in Table 1. In addition, small molecules can be used for deliverance of at least one linked toxic compound to a tumor cell. Small molecule inhibitors are usually chemical entities that can be obtained by high-throughput screening of already existing libraries of compounds and/or by designing compounds based on the structure of the protein encoded by a gene involved in tumor development. Briefly, the structure of at least a fragment of the protein is determined by either Nuclear Magnetic Resonance or X-ray crystallography. Based on this structure, virtual screening of compounds can be performed using specific software packages. The selected compounds are synthesized using medicinal and/or combinatorial chemistry and thereafter analyzed for their inhibitory effect on the protein in vitro and in vivo . This step can be repeated until a compound is selected with the desired inhibitory effect. After optimization of the compound identified either by high- throughput or virtual screening, its toxicity profile and efficacy as cancer therapeutic is tested in vivo using appropriate animal model systems .

On a different level of inhibition nucleic acids can be used to block the production of proteins by destroying the mRNA transcribed from the gene. This can be achieved by antisense drugs or by RNA interference (RNAi) . By acting at this early stage in the disease process, these drugs prevent the production of a disease-causing protein. The present invention relates to antisense drugs, such as antisense RNA and antisense oligodeoxynucleotides, directed against the genes listed in Table 1. Each antisense drug binds to a specific sequence of nucleotides in its mRNA target to inhibit production of the protein encoded by the target mRNA. The invention furthermore relates to RNAi molecules. RNAi refers to the introduction of homologous double stranded RNA to specifically target the transcription product of a gene, resulting in a null or hypomorphic phenotype. RNA interference requires an initiation step and an effector step. In the first step, input double-stranded (ds) RNA is processed into 21-23-nucleotide 'guide sequences". These may be single- or double-stranded. The guide RNAs are incorporated into a nuclease complex, called the RNA-induced silencing complex (RISC) , which acts in the second effector step to destroy mRNAs that are recognized by the guide RNAs through base-pairing interactions. RNAi molecules are thus double stranded RNAs (dsRNAs) that are very potent in silencing the expression of the target gene. The present invention relates to dsRNAs complementary to the genes listed in Table 1.

The invention relates further to gene therapy, in which the genes listed in Table 1 are used for the design of dominant-negative forms of these genes which inhibit the function of their wild-type counterparts following their directed expression in a cancer cell. Alternatively, RNAi approaches can be used for gene therapy, e.g. by introducing a dsRNA-producing sequence into a cancer cell. Another object of the present invention is to provide a pharmaceutical composition comprising the inhibitors according to the present invention as active ingredient for the treatment of cancer. The composition can further comprise at least one pharmaceutical acceptable additive like for example a carrier, an emulsifier, or a conservative. In addition, it is the object of the present invention to provide a method for treatment of cancer patients which method comprises the administration of the pharmaceutical composition according to the invention to cancer patients.

The invention will be further illustrated in the examples that follow and which are not given to limit the invention. In the examples, reference is made to the following figures: Figure 1. RT-PCR analysis of SELPLG expression in a panel of 20 different human tumor cell lines demonstrating SELPLG expression in DU145, SW480, MDA-MB-435s, and MDA-MB- 231, as well as in the controls, i.e. eosinophils and leukocytes . Figure 2. Q-PCR analysis of SELPLG expression in a panel of different human tumor cell lines derived from colon (1, LS174T; 2, HCT116; 3, HT-29; 4, DLD-1; 5, S 480) , lung (6, A549; 7, EKNX; 8, HOP-62) , breast (9, MCF7; 10, T-47D; 11, MDA-MB-231; 12, MDA-MB-361; 13, MDA-MB-435s; 14, MDA-MB- 468), prostate (15, DU145; 16, PC-3) , ovarian (17, OVCAR-4; 18, IGROV-1), and melanoma (19, Ml4; 20, SK-MEL-5) . All expression levels, depicted in arbitrary units (A.U.), are related to the average expression detected in HCT116 cells (CT = 37.3). Duplicate measurements are shown. Note: both values measured for SELPLG expression in S 480 are not shown to scale; these values are 284 and 252, respectively. Figure 3. FACS analysis of SELPLG expression on the cell surface of two selected human tumor cell lines, i.e. MCF7 and MDA-MB-231. Cells were stained with anti-SELPLG antibodies (bold line) or with isotype control antibodies (thin line) . A shift of the SELPLG peak to the right relative to the isotype control peak indicates expression of SELPLG protein.

Figure 4. Perfusion assay to determine the rolling capacity of selected human tumor cell lines. Images depict computerized analyses of rolling MDA-MB-231 and MDA-MB-468 cells, respectively, on endothelium. Lines indicate the rolling tracks of the cells in one field during 1 second. Figure 5. Sequences of murine Selpl and its human orthologue SELPLG according to the Celera Discovery System database, i.e. cDNA sequence of murine Selpl (mCT2027) , protein sequence of murine Selpl (mCP14202) , cDNA sequence of human SELPLG (hCT1953870) , and protein sequence of human SELPLG (hCP1765130) .

Figure 6. RT-PCR analysis of SEMA4B expression in a panel of 20 different human tumor cell lines. Figure 7. Q-PCR analysis of SEMA4B expression in a panel of different human tumor cell lines derived from colon (1, LS174T; 2, HCT116; 3, HT-29; 4, DLD-1; 5, SW480) , lung (6, A549; 7, EKVX; 8, HOP-62), breast (9, MCF7 ; 1.0, T-47D; 11, MDA-MB-231; 12, MDA-MB-361; 13, MDA-MB-435s; 14, MDA-MB- 468), prostate (15, DTJ145; 16, PC-3), ovarian (17, OVCAR-4; 18, IGROV-1), and melanoma (19, Ml4; 20, SK-MEL-5) . All expression levels, depicted in arbitrary units (A.U.), are related to the average expression detected in HCT116 cells (CT = 27.4). Duplicate measurements are shown.

Figure 8. Western blot analysis of SEMA4B protein expression in selected human tumor cell lines using polyclonal antibodies generated against peptide 2

(DPAFNASAYIPESLGSLC) derived from human SEMA4B. Similar results were obtained using antibodies generated against peptide 1 ( SVPVI ISTSRNSAPC) also derived from human SEMA4B (data not shown) . Figure 9. Sequences of murine Sema4b and its human orthologue SEMA4B according to the Celera Discovery System database, i.e. cDΝA sequence of murine Sema4b (mCT20086), protein sequence of murine Sema4b (mCP7317), cDΝA sequences of 3 different splice variants of human SEMA4B (hCT2269854, hCT2269851, and hCT18586) , and corresponding protein sequences of these splice variants of human SEMA4B (hCP1851384, hCP1851383, and hCP43026, respectively) .

Figure 10. RT-PCR analysis of PSK expression in a panel of 20 different human tumor cell lines. Figure 11. Q-PCR analysis of PSK expression in a panel of different human tumor cell lines derived from colon (1, LS174T; 2, HCT116; 3, HT-29; 4, DLD-1; 5, S 480), lung (6, A549; 7, EKVX; 8, HOP-62), breast (9, MCF7; 10, T-47D; 11, MDA-MB-231; 12, MDA-MB-361; 13, MDA-MB-435s; 14, MDA-MB- 468), prostate (15, DU145; 16, PC-3), ovarian (17, OVCAR-4; 18, IGROV-1), and melanoma (19, M14; 20, SK-MEL-5). All expression levels, depicted in arbitrary units (A.U.), are related to the average expression detected in HCT116 cells (CT = 27.0) . Duplicate measurements are shown.

Figure 12. Sequences of the murine orthologue of PSK and of human PSK according to the Celera Discovery System database, i.e. cDNA sequence of the murine orthologue of PSK (mCT22502) , protein sequence of the murine orthologue of PSK (mCP10412), cDNA sequences of 3 different splice variants of human PSK (hCT2260670, hCT1956039, and hCT9408), and corresponding protein sequences of these splice variants of human PSK (hCP1900108, hCP1763475, and hCP36195, respectively) .

Figure 13. cDNA and protein sequences of all murine genes of the present invention listed in Table 1, as well as their known human orthologues, and the known splice variants. The sequences were obtained from the following databases: Celera Discovery System (www. celeradiscoverysystem. com) , Ensembl (www . ensembl . org) , and National Center for Biotechnology Information (NCBI; www.ncbi.nlm.nih.gov) . Primarily, Celera sequences are given except if additional variants (e.g. due to alternative splicing) are available at the other databases .

EXAMPLES

EXAMPLE 1

Identification of genes involved in tumor development using Eμ yc and Eμ yc Pim ^/~ Pim2^~/~ mice

INTRODUCTION Retroviral insertions in the genome can transform host cells by activation of proto-oncogenes or inactivation of tumor suppressor genes. Retroviral insertions near such genes are instrumental in the clonal outgrowth of the incipient tumor cell. A full-blown tumor then results from multiple rounds of retroviral insertional mutagenesis in which proviral insertions mark genes collaborating in stepwise tumor progression. To modify the sensitivity of the retroviral screen, different genetic backgrounds can be used to identify oncogenes hardly or not found in retroviral screens using "wild type" background. On the basis of strong cooperation between c-Myc and Pim in tumor development, genes acting downstream of, or parallel to Pim are likely to be selected for in tumors originating from mice deficient for Pim but expressing high levels of Myc .

MATERIALS AND METHODS Mice and M-MuLN infection

The EμMyc mice were bred with Pi l deficient Piml neo59 mice and Pim2 deficient Pim2 K180 mice to generate EμMyc Pimr^f~ , EμMyc Pim2^'/~ and EμMyc Piml^''^' Pim2^~/~ mice. Νeonates were infected with 1.10⁵ infectious units of M-MuLV. Moribund mice were sacrificed and lymphomas were isolated.

Southern blot analysis of common insertion sites (CIS)

Tumor DΝA was isolated. Genomic tumor DΝA (10 μg) was restricted with the appropriate enzyme, separated on a 0.7 % agarose gel and subsequently transferred to Hybond-Ν membranes (Amersham) . The number of proviral insertions and the insertions into the known CIS Piml , Pim2, Bmil and Gfil were analyzed. Genomic fragments, free of repetitive sequences, flanking the proviruses and hybridizing to a CIS were used as probes to analyze the frequency at which these loci were inserted by a provirus . o

Isolation of the proviral insertion sites

1. Ligation

Tumor DNA (3 μg) was restricted with BstYI (New England Biolabs) after which the enzyme was inactivated. The splinkerette adaptor was generated by annealing the splinkerette oligos, HMSpAA: 5' cgaagagtaaccgttgctaggagagaccgtggtgaatgagactggtgtcgacactagtgg 3' (SEQ ID NO. 1) and HMSpBB: 5' gatccactagtgtgacacagtctctaatttttttttttaaaaaaa 3' (SEQ ID NO. 2). Both oligos contain modifications of a splinkerette. The oligos (150 pmol each) were denatured at 95 °C for 3' and subsequently cooled to room temperature at a rate of 1^"C per 15" using a thermocycler (PTC100, Perkin Elmer) . 600 ng of genomic tumor DNA restricted with BstYI was ligated to the splinkerette oligo (molar ratio 1:10) with 4 U T4 DNA ligase (Roche Diagnostics) in a final volume of 40 μl . To avoid amplification of the internal 3' M-MuLV fragment, the ligated fragments were restricted with 10 U of EcoRV in a total volume of 100 μl . Ligation mixtures were desalted in a Microcon YM-30 (Amicon BioSeparations) according to the manufacturer .

2. PCR Amplification

M-MuLV flanking sequences were amplified with a radioactive LTR-specific primer, AB949 (5' gctagcttgccaaactcaggtgg 3' (SEQ ID NO. 3)), and a splinkerette-primer, HMSpl (5' cgaagagtaacgttgctaggagagacc 3' (SEQ ID NO. 4)). Primer AB949 (10 pmol) was radioactively labeled with .γ-³²P ATP (3 μCu) using T4 PNK (0.2U) (Roche Diagnostics) . The 50 μl PCR mixture contained 150 ng ligated tumor DNA, 10 pmol primer (each) , 300 nmol dNTPs, 1 U PfuITurbo™ and IX PfuITurbo™ buffer (Stratagene) . The hot start PCR conditions were 3' 94 °C, 2 cycles 15" 94 °C, 30" 68°C, 3' 30" 72°C, 27 cycles 15" 94°C, 30" 66°C, 3' 30" 72°C, and 5' 72 °C. Radioactive PCR fragments were concentrated using a microcon-30 (Amicon BioSeparations) and subsequently separated on a 3.5% denaturing polyacrylamide gel. The gels were dried onto 3MM Wattman paper and exposed O/N to X-Omat AR films (Kodak) . Amplified fragments were excised from the gel and boiled for 30' in 100 μl TE . 1 μl of the DNA solution was used for a nested amplification with a ³²P labeled virus specific primer HMOOl (5' gccaaacctacaggtggggtcttt 3' (SEQ ID NO. 5) ) and a non-radioactive splinkerette-specific primer HMSp2 (5' gtggctgaatgagactggtgtcgac 3' (SEQ ID NO. 6)). The nested PCR was performed with 5 pmol of primers (each) , 200 nM dNTPs (each), 1.75 mM Mg, 1 U Taq polymerase (Gibco BRL), IX PCR buffer (Gibco BRL) in a final volume of 20 μl . The PCR conditions were 15" 94°C, 30" 60°C, 3' 72°C for 25 cycles (fragments < 400 bp) or for 28 cycles (fragments > 400 bp) . The re-amplified fragments were separated on a 3.5% denaturing polyacrylamide gel and isolated as described above. 1 μl of the amplified fragments were again re- amplified in a non-radioactive PCR of 25 cycles under the conditions as described for the radioactive nested PCR.

3. Sequence analysis The nested PCR mixture was treated with 0.5 U exonuclease and 0.5 U shrimp alkaline phosphatase according to the manufacturer (Amersham) . About 25 ng of the PCR product was used in the sequence reaction containing BigDye terminator mix (Perkin Elmer) and primer HMOOl. In addition, HMSp2 was used as primer for sequencing of the amplified fragments larger than 500 bp. Automated sequencing was performed on an ABI 377 (Perkin Elmer) . The sequences were processed with Sequencher 3.1.1™ and blasted against the annotated mouse genomic database at Celera using the Celera Discovery System™.

RESULTS

471 provirus flanks were sequenced from 38 EμMyc Piml^''^' Pim2^~/' and 18 control EμMyc lymphomas . This number corresponds to approximately 60% of all retroviral insertions in these tumors. 47 loci showed proviral integrations in more than one tumor and were therefore designated as common insertion sites (CIS) . Based on sequence comparisons with the Celera annotated mouse genomic database, the gene located nearby the CIS was identified. Of these genes, the ones that were so far not described to be involved in tumor development are listed in Table 1 combined with the novel cancer genes from Example 2.

DISCUSSION

Although proviral integrations occur randomly, they may affect the expression or function of nearby genes. If a gene is affected in two or more independent tumors, this indicates that these integrations provide a selective advantage and therefore contribute to tumor development. Multiple of these common insertion sites were identified of which a large number are demonstrated for the first time to play a role in cancer. Importantly, several of the other genes identified are well-known cancer genes (e.g. Myj, B il , Tpl2, and Ccnd2) validating the approach. This example shows that the pursued strategy can be successfully used to identify novel genes that are involved in tumor development. EXAMPLE 2

Identification of genes involved in tumor development using Cdkn2a-deficient mice

INTRODUCTION

To identify genes that cooperate with the combined loss of pl6^{Ink a} and pl9^Arf, encoded by the Cdkn2a locus, in tumorigenesis, neonatal mice deficient for the second exon of Cdkn2a were infected with Moloney murine leukemia virus (MoMLV) . This retrovirus induces tumors by insertional activation of proto-oncogenes or, though inherently more rarely, through inactivation of tumor suppressor genes.

MATERIALS AND METHODS Mo-MLN tumorigenesis

Cdkn2a -/- mice as well as +/- and +/+ littermates were infected intraperitoneally with 10⁵ infectious units of Mo-MLV within 72 hours of birth. Diseased mice were euthanized, necropsied, and then CdJcn2a-genotyped post-mortem by PCR. Most of each tumor was frozen for genetic studies. Fragments of tumors were fixed in either 10% formalin or Bouin' s fixative, paraffin embedded, sectioned, and stained with hematoxylin and eosin. Lymphomas were analyzed by Southern blotting to assess T-cell receptor and immunoglobulin gene rearrangements. Cell surface markers for hematopoietic lineages were assessed by immunohistochemistry . Histiocytic sarcoma transplantation into immunodeficient mice was accomplished by subcutaneous injection near the scapulae of BALB/c SCID mice of approximately 4.5 X 10⁵ disaggregated cells from macroscopically visible histiocytic liver nodules in virally infected Cdkn2a -/- mice. Tumors appearing in the livers of recipient mice histologically resembled those in the donors. They were confirmed to be of donor origin by PCR genotyping for the exon 2-deleted Cdkn2a allele.

Retrieval and analysis of Mo-MLV integration site sequences 463 viral insertion sites were amplified using splinkerette-aided amplification procedures. 284 insertion site sequences were retrieved using inverse PCR on SacII- digested tumor DNA. The MoMLV-specific primers used for the nested inverse PCR were 5' atcggacagacacagataagtt 3' (SEQ ID NO. 7), 5' gccaaacctacaggtggggtcttt 3' (SEQ ID NO. 8), 5' aaacctgtgatgcctgaccagt 3' (SEQ ID NO. 9), 5' gcaactgagacctgcaaagcttgt 3' (SEQ ID NO. 10) . Amplification products were purified from agarose gels and subjected to direct automated sequencing. For candidate gene identification, insertion site sequences were filtered for the presence of repetitive DNA and homology searches using BLASTn were performed in GenBank databases as well as in the mouse genome sequence (Celera Genomics) . An unambiguous match was defined as having 1 homology region only and a BLASTn probability value of 10^"25 or less. While CISs traditionally have been demonstrated using Southern blotting, the presence of the complete mouse genome allows for in silico CIS determination. Following the statistical analysis retroviral common insertion sites were defined as 2 or more integrations within 26 kb or 3 insertions within 300 kb. For a set of 500 insertions, these windows give a tolerable statistically calculated background of ~2.5 CISs occurring at random. When flanking a gene, the accepted distance between insertions was set to 100 kb. While the functional distance between viral insertions and candidate oncogenes is known to differ between loci, the statistical threshold set here is in accordance with viral integration patterns surrounding previously characterized common insertion sites.

RESULTS

From a panel of 104 lymphoid (55%) and myeloid (45%) tumors, a total of 747 unique MoMLV integration sites were isolated and directly sequenced using a combination of inverse PCR and splinkerette-aided insertion site amplification. Homology searches in GenBank and in the mouse genome database (Celera Genomics, Rockville, MD) unambiguously mapped 565 viral insertions. 172 of the sequences were clustered in 46 common insertion sites (CISs) . Using the Celera annotated mouse genomic database, the genes located nearby the CISs were identified. Of these genes, the ones that were so far not described to be involved in tumor development are listed in Table 1 combined with the novel cancer genes from Example 1.

DISCUSSION

Although proviral integrations occur randomly, they may affect the expression or function of nearby genes. If a gene is affected in two or more independent tumors, this indicates that these integrations provide a selective advantage and therefore contribute to tumor development. Multiple of these common insertion sites were identified of which a large number are demonstrated for the first time to play a role in cancer. Importantly, several of the other genes identified are well-known cancer genes (e.g. Myc, Myb, Tpl2, and Ccndl ) validating the approach. This example shows that the pursued strategy can be successfully used to identify novel genes that are involved in tumor development. EXAMPLE 3

Development of anti-cancer drugs on the basis of the genes involved in tumor development from Examples 1 and 2

VALIDATION OF THE IDENTIFIED GENES AS RELEVANT TARGETS

Expression profile of the identified genes in human tumors

Expression of the described genes, that were originally identified by genome-wide functional screens involving retroviral insertional tagging in mouse models (see Examples 1 and 2) , is determined in a panel of different human tumor cell lines by semi-quantitative RT-PCR and/or Quantitative PCR (Q-PCR) . Genes that are overexpressed in human tumor cell lines as determined by Q-PCR can be further examined by other techniques such as Northern blot analysis and, if gene-specific antibodies are available, Western blot and/or FACS analysis. Moreover, expression of these genes can be investigated in primary human tumors, e.g. using microarray analysis. From a set of primary human tumors of various organs, RNA is prepared using standard laboratory techniques to investigate the expression of the described genes in these samples relative to their expression in normal, unaffected tissues from the same origin by using microarrays on which these genes, or parts thereof, are spotted. Microarray analysis allows rapid screening of a large set of genes in a single experiment (DeRisi et al . Use of a cDNA microarray to analyze gene expression patterns in human cancer. Nat Genet 14:457-60, 1996; Lockhart et al . Expression monitoring by hybridization to high-density oligonucleotide arrays. Nat Biotechnol 14:1649, 1996). Confirmation of differential expression of these genes further suggests their involvement in human tumors . Functional importance of the identified genes for human cancer

Subsequently, the functional importance of the identified genes for human tumor cells is determined by either interfering with the expression of the genes or by affecting the function of the expressed proteins encoded by the genes. For example, selected human tumor cell lines can be transfected with plasmids encoding cDNA of the genes or with plasmids encoding RNA interference probes for the genes. RNA interference is a recently developed technique that involves introduction of double-stranded oligonucleotides designed to block expression of a specific gene (see Elbashir et al . Duplexes of 21-nucleotide RNAs mediate RNA interference in cultured mammalian cells. Nature 411:494-8, 2001; Brummelkamp et al . A system for stable expression of short interfering RNAs in mammalian cells. Science 296:550-3, 2002). If available, function-blocking antibodies against cell-surface expressed proteins encoded by the identified genes can be used to interfere with the protein's function. Using standard laboratory techniques and assays, the cell lines are extensively checked for effects of transfection and/or function-blocking antibody treatment on relevant phenotypes, e.g. cell cycle status, proliferation, adhesion, apoptosis, invasive abilities, etc.

DEVELOPMENT OF THERAPEUTIC COMPOUNDS Development of antibody-based inhibitors

Validated cancer genes that encode membrane-bound proteins are selected as targets for conventional antibody- based therapies. Antibodies are generated against functionally relevant domains of the proteins and subsequently screened for their ability to interfere with the target's function using standard techniques and assays (Schwartzberg. Clinical experience with edrecolomoab : a monoclonal antibody therapy for colorectal carcinoma. Crit . Rev Oncol He atol 40:17-24, 2001; Herbst et al . Monoclonal antibodies to target epidermal growth factor receptor- positive tumors: a new paradigm for cancer therapy. Cancer 94:1593-611, 2002) .

Development of small molecule inhibitors Validated cancer genes that do not encode membrane- bound proteins are selected as targets for the development of small molecule inhibitors. To identify putative binding sites or pockets for small molecules on the surface of the target proteins, the three-dimensional structure of those targets are determined by standard Nuclear Magnetic Resonance or crystallization techniques (de Vos et al . Three-dimensional structure of an oncogene protein: catalytic domain of human c-H-ras p21. Science 239:888-93, 1988; Williams et al . Crystal structure of the BRCT repeat region from the breast cancer-associated protein BRCA1. Nat Struct Biol 8:838-42, 2001) . Mutational analysis is performed to investigate the functional importance of the identified binding sites for the protein's function. Subsequently, computer programs such as Cerius2 (Molecular Simulations Inc., San Diego, CA, USA) and Ludi/ACD (Accelrys Inc., San Diego, CA, USA) are used for virtual screening of small molecule libraries (Boh . The computer program Ludi: A new method for the de novo design of enzyme inhibitors. J Comp Aided Molec Design 6:61-78, 1992) . The compounds identified as potential binders by these programs are synthesized by combinatorial and/or medicinal chemistry and screened for binding affinity to the targets as well as for their inhibitory capacities of the target protein's function by standard in vitro and in vivo assays. In addition to the rational development of small molecule inhibitors, existing small molecule compound libraries are tested using high-throughput screening assays to generate lead compounds. Lead compounds identified are subsequently co-crystallized with the target to obtain information on how the binding of the small molecule can be improved (Zeslawska et al . Crystals of the urokinase type plasminogen activator variant beta (c) -uPAin complex with small molecule inhibitors open the way towards structure-based drug design. J Mol Biol 301:465-75, 2000) . Based on these findings, novel compounds are designed, synthesized, tested, and co-crystallized. This optimization process is repeated for several rounds leading to the development of a high-affinity compound of the invention that successfully inhibits the function of its target protein.

Development of antisense molecule inhibitors

These inhibitors are either antisense RNA or antisense oligodeoxynucleotides (antisense ODNs) and are prepared synthetically or by means of recombinant DNA techniques. Both methods are well within the reach of the person skilled in the art. ODNs are smaller than complete antisense RNAs and have therefore the advantage that they can more easily enter the target cell. In order to avoid their digestion by DNAse, ODNs but also antisense RNAs are chemically modified. For targeting to the desired target cells, the molecules are linked to ligands of receptors found on the target cells or to antibodies directed against molecules on the surface of the target cells. Development of RNAi molecule inhibitors

Double-stranded RNA corresponding to a particular gene is a powerful suppressant of that gene. The ability of dsRNA to suppress the expression of a gene corresponding to its own sequence is also called post-transcriptional gene silencing or PTGS . The only RNA molecules normally found in the cytoplasm of a cell are molecules of single-stranded mRNA. If the cell finds molecules of double-stranded RNA, dsRNA, it uses an enzyme to cut them into fragments containing 21-25 base pairs (about 2 turns of a double helix) . The two strands of each fragment then separate enough to expose the antisense strand so that it can bind to the complementary sense sequence on a molecule of mRNA. This triggers cutting the mRNA in that region thus destroying its ability to be translated into a polypeptide. Introducing dsRNA corresponding to a particular gene will knock out the cell's endogenous expression of that gene. This can be done in particular tissues at a chosen time. A possible disadvantage of simply introducing dsRNA fragments into a cell is that gene expression is only temporarily reduced.

However, a more permanent solution is provided by introducing into the cells a DNA vector that can continuously synthesize a dsRNA corresponding to the gene to be suppressed. RNAi molecules are prepared by methods well known to the person skilled in the art.

Evaluation of candidate therapeutic compounds

The toxicity of the candidate therapeutic compounds is tested using standard assays (e.g. via commercially available services such as those provided by MDS Pharma Services, Montreal, Quebec, Canada) . The efficacy of the compounds is tested in an appropriate animal model system before entry into clinical development (e.g. Brekken et al . Selective inhibition of vascular endothelial growth factor (VEGF) receptor 2 (KDR/Flk-1) activity by a monoclonal anti-VEGF antibody blocks tumor growth in mice. Cancer Res 60:5117-24, 2000; Wilkinson et al . Antibody targeting studies in a transgenic murine model of spontaneous colorectal tumors . Proc Natl Acad Sci USA 98:10256-60, 2001; Laird et al . SU6668 inhibits Flk- 1/KDR and PDGFRbeta in vivo, resulting in rapid apoptosis of tumor vasculature and tumor regression in mice. FASEB J 16:681-90, 2002) .

EXAMPLE 4 Validation of identified genes in human tumor cell lines

EXPRESSION OF IDENTIFIED GENES IN HUMAN TUMOR CELL LINES RT-PCR

A panel of 20 different human cancer cell lines derived from colon (LS174T, HCT116, HT-29, DLD-1, SW480), lung (A549, EKVX, HOP-62), breast (MCF7, T-47D, MDA-MB-231, MDA-MB-361, MDA-MB-435s, MDA-MB-468), prostate (DU145, PC-3), ovarian (OVCAR-4, IGROV-1), and melanoma (Ml4, SK-MEL-5) were lysed and total RNA was extracted using Trizol^Tm reagent according to manufacturer's instructions. First strand cDNA was obtained by reverse transcriptase (RT) reactions with oligo dT primers using 1 μg of RNA. Subsequently, PCRs were performed on the RT reactions to amplify fragments corresponding to specific regions of the target gene using primers spanning exon-intron boundaries. GAPDH cDNA fragments were amplified from each cDNA sample as an internal control for cDNA quality. The following oligonucleotides were used (FW = forward; RV = reverse) :

SELPLG: FW 5' cagctgtcccatgctctg 3' (SEQ ID NO. 11)

RV 5' gcctcagaagtccgtcactc 3' (SEQ ID NO. 12)

SEMA4B: FW 5' agaatctgctcctggacacc 3' (SEQ ID NO. 13)

RV 5' accacctctgggcagtagc 3' (SEQ ID NO. 14)

PSK: FW 5' atgacccctaccagccaga 3' (SEQ ID NO. 15)

RV 5' gaggctcttgggctgctg 3' (SEQ ID NO. 16)

Quantitative PCR (Q-PCR)

Total RNA was prepared from the 20 human tumor cell lines described above using Trizol^Tm reagent according to manufacturer's instructions. One microgram of RNA was reverse transcribed to generate the corresponding cDNA, which was used as a template for Q-PCR. The reverse transcription step was performed in 96-well plates using the TaqMan reverse transcription kit (Applied Biosystems) according to the manufacturer's recommendations. The cDNA was quantified by the SyBR green method using a SyBR Green PCR master Mix kit (Applied Biosystems) according to the manufacturer's recommendations. For each reaction, 8 ng of cDNA was used as a template and 300 nM of specific forward and reverse oligonucleotides added. Duplicate experiments were carried out using an Applied Biosytems 7000 SDS. Values were normalized according to expression of the β-gl ucoronidase { GUS) gene, which was measured as internal control. Oligonucleotides were designed using the Primer Express software (Applied Biosystems) . These oligonucleotides were validated by Q-PCR experiments to obtain a quantitative measurement (quantification of serially diluted cDNA and determination of PCR efficiency) . The following oligonucleotides were used (FW = forward; RV = reverse) :

GUS : FW 5' cccgcggtcgtgatgt 3' (SEQ ID NO. 17) RV 5' tgagcgatcaccatcttcaagt 3' (SEQ ID NO. 18)

SELPLG: FW 5' ggaccttgtcactaaagcagagaag 3' (SEQ ID NO. 19)

RV 5' tgtcccacagctgcaagct 3' (SEQ ID NO. 20)

SEMA4B: FW 5' gtgcagcttcaagggcaag 3' (SEQ ID NO. 21)

RV 5' ctgagcggcaggaggatct 3' (SEQ ID NO. 22) PSK: FW 5' acagtgggaagcagtccaatga 3' (SEQ ID NO. 23)

RV 5' aggcaatactccattaccagcca 3' (SEQ ID NO. 24)

Western blotting

Lysates from selected human tumor cell lines were prepared and subjected to Western blot analysis. Two independent polyclonal antibodies generated against peptide 1 (SVPVIISTSRVSAPC; SEQ ID NO. 25) and peptide 2 (DPAFVASAYIPESLGSLC; SEQ ID NO. 26), respectively, both derived from human SEMA4B, were used to determine expression levels of SEMA4B in these cell lines.

FACS analysis

Single-cell suspensions of selected human tumor cell lines were analyzed by flow cytometry using a FACScan. The cells were labeled with anti-SELPLG antibodies (PL-1; Im unotech, Marseille, France) Immnudetection was performed using donkey-anti-mouse antibodies coupled to fluorescein isothiocyanate (FITC) .

FUNCTIONAL IMPORTANCE OF IDENTIFIED GENES

Perfusion assay to determine rolling capacity To investigate whether SELPLG proteins expressed on the surface of human tumor cell lines are capable of mediating rolling on endothelial cells, perfusion assays under steady flow were performed in a modified form of a transparent parallel plate perfusion chamber as previously described (Sakariassen et al . A perfusion chamber developed to investigate platelet interaction in flowing blood with human vessel wall cells, their extracellular matrix, and purified components. J Lab Clin Med 102:522-35, 1983) The micro-chamber has a slit height of 0.2 mm and a width of 2 mm and contains a circular plug on which a coverslip with confluent HUNEC or L cells expressing selectin proteins can be mounted. Perfusions using selected human tumor cell lines and neutrophils as controls were performed as previously described (Ulfman et al . Characterization of eosinophil adhesion to TΝF-alpha-activated endothelium under flow conditions: alpha 4 integrins mediate initial attachment, and E-selectin mediates rolling. J Immunol 163:343-50, 1999). Video images of the perfusions were recorded and evaluated using image-analysis software Optimas 6.1 (Media Cybernetics Systems, Silver Springs, MD, USA) . The cut-off value to distinguish between rolling and static adherent cells was set at 1 μm/s. With this method, static adherent, rolling, and freely flowing cells (which were not in focus) could be clearly distinguished. To investigate the involvement of P- selectin and SELPLG in rolling of the selected human tumor cell lines on endothelial cells, function-blocking monoclonal antibodies were used in the perfusion assays directed against P-selectin (WASP12.2; Endogen, Boston, MA, USA) and SELPLG (PL-1; Immunotech, Marseille, France) .

EXAMPLE 5 SELPLG

INTRODUCTION

Selectins function in the initial step of recruitment of leukocytes, and primarily neufrophils, to the site of an inflammatory reaction. To enter the site of inflammation, leukocytes have to leave the bloodstream and pass the endothelium. A widely accepted paradigm for leukocyte extravasation is referred to as the multi-step model. Firstly, specific ligands expressed on the surface of leukocytes interact with selectins that are expressed on endothelium activated by inflammatory agents. This allows the leukocytes to slow down from the circulation and roll on the endothelium. Subsequently, other proteins such as integrins are involved in firm adhesion and transendothelial migration of the leukocytes.

P-selectin, a cell-surface glycoprotein expressed on activated endothelium, is critical for mediating leukocyte rolling in response to inflammatory signals by interacting with its ligand expressed on the surface of leukocytes called P-selectin ligand (official gene name: SELPLG) , demonstrated e.g. using P-selectin knockout mice (Broide et al . Inhibition of eosinophil rolling and recruitment in P-selectin- and intracellular adhesion molecule-deficient mice. Blood 91:2847-56, 1998). Thus, SELPLG is essential for leukocyte rolling which is one of the earliest steps of an acute inflammatory response and, as such, contributes to many inflammatory diseases.

Specific in vitro binding of P-selectin to several types of cancer cells, such as melanoma, neuroblastoma, colon, lung, and breast tumor cells, has been demonstrated as well suggesting involvement of P-selectin in metastasis. Pronounced inhibition of metastasis in P-selectin-deficient mice compared with wild-type controls in a colon carcinoma cell model provides compelling evidence for the role of P- selectin in the metastatic process (Borsig et al . Heparin and cancer revisited: mechanistic connections involving platelets, P-selectin, carcinoma ucins, and tumor metatasis. Proc Natl Acad Sci USA 98:3352-7, 2001; Borsig et al . Synergistic effects of L- and P-selectin in facilitating tumor metastasis can involve non-mucin ligands and implicate leukocytes as enhancers of metastasis. Proc Natl Acad Sci USA 99:2193-8, 2002). Characterization of the proteins on the surface of tumor cells that interact with P-selectin might provide a novel strategy to develop anti-metastatic therapeutics .

RESULTS AND DISCUSSION

To determine whether SELPLG is expressed in human tumors, RT-PCR analysis was performed on a panel of 20 different human tumor cell lines of various origin. SELPLG expression was observed in the prostate cancer cell line DU145, the colon carcinoma cell line SW480, and the breast tumor cell lines MDA-MB-435s and MDA-MB-231 (Figure 1) . Subsequently, Q-PCR analysis was performed to quantify the relative expression levels in the different cell lines. As expected, based on the restricted expression of SELPLG, most cell lines do not or modestly express SELPLG (e.g. HCT116: CT = 37.3; T-47D: CT > 40). However, overexpression of SELPLG was clearly identified in 7 of the 20 tumor cell lines investigated. In addition to the already mentioned cell lines demonstrated to express SELPLG by RT-PCR (i.e. DU145, SW480, MDA-MB-435s, and MDA-MB-231) , Q-PCR also revealed SELPLG overexpression in the prostate tumor cell line PC-3 and in both ovarian carcinoma cell lines OVCAR-4 and IGROV-1 (Figure 2) . The highest expression was detected in the colon carcinoma cell line SW480. Due to usage of different primers in RT-PCR versus Q-PCR analysis, alternative splicing or other organ and/or tumor specific modifications might have affected amplification of products in both assays possibly explaining the observed differences between the RT-PCR and Q- PCR results. Using a monoclonal antibody directed against SELPLG protein, FACS analysis was performed on the breast tumor cell lines MCF7 and MDA-MB-231. SELPLG protein could not be detected on MCF7 cells in agreement with the absence of detectable SELPLG transcripts in both RT-PCR and Q-PCR analyses. In contrast, MDA-MB-231 clearly showed expression of SELPLG protein on the cell surface (Figure 3) confirming the observed SELPLG mRNA expression as obtained by RT-PCR and Q-PCR analyses.

To investigate whether the expressed SELPLG proteins on the surface of human tumor cell lines are also functionally active, perfusion assays with these cells were performed on endothelium. No rolling was observed of the breast tumor cell lines MCF7 and T-47D (data not shown) in agreement with the absence of detectable SELPLG transcripts in RT-PCR (Figure 1) and Q-PCR (Figure 2) analyses as well as of SELPLG protein in FACS analysis (Figure 3) in these two cell lines. Importantly, both MDA-MB-231 and MDA-MB-468 cells showed significant rolling on endothelium (Figure 4) . This could be completely blocked by the addition of function- blocking monoclonal antibodies directed against P-selectin demonstrating that P-selectin interactions are crucial for rolling of these cells. However, addition of function- blocking monoclonal antibodies directed against SELPLG only abrogated rolling of MDA-MB-231 cells demonstrating that SELPLG is essential for rolling of these cells, whereas rolling of MDA-MB-468 cells was not affected (data not shown) . These observations are again completely in agreement with expression data showing SELPLG expression in MDA-MB-231 cells by both RT-PCR and Q-PCR analyses, but not in MDA-MB- 468 cells (Figures 1 and 2) . In conclusion, these results demonstrate that SELPLG, most likely via interactions with P- selectin, is crucial for rolling of at least the human breast tumor cells MDA-MB-231 potentially regulating its metastatic capacity. Apparently, rolling of MDA-MB-468 cells occurs independently of SELPLG, although P-selectin interactions are essential. Other cell-surface proteins on these cells might be involved in interacting with P-selectin as has been shown for CD24 (Aigner et a-Z . CD24, a mucin-type glycoprotein, is a ligand for P-selectin on human tumor cells. Blood 89:3385-95, 1997) . cDNA and protein sequences of murine Selpl as well as of its human orthologue SELPLG are given in Figure 5 (SEQ ID NO. 27-30) .

EXAMPLE 6 SEMA4B

INTRODUCTION Invasive growth is a complex program in which cell proliferation combines with cell-cell dissociation and movement, matrix degradation and survival. It occurs under physiological conditions (such as organ development and regeneration, axon guidance and wound healing) as well as in carcinoma progression, in which it is essential for tumor invasion and metastasis. Scatter factors (e.g. hepatocyte growth factor, HGF) and their receptors (e.g. tyrosine kinase receptors MET and RON) are the main regulators of normal and neoplastic invasive growth. However, two other families of proteins that are structurally related to MET, i.e. ^λsemaphorins' and ^xplexins', are involved in the control of invasive growth as well (Trusolino et al . Scatter factor and semaphorin receptors: cell signaling for invasive growth. Nature Rev Cancer 2:289-300, 2002). Semaphorins are a large family of secreted and membrane-bound proteins . Two distinct types of semaphorin receptors are known, i.e. unassociated plexins, that can bind to membrane-bound semaphorins, and plexins associated with neuropilins, that can bind to secreted semaphorins .

MET activity is deregulated in many human cancers due to mutations, gene amplification, protein overexpression, or production of HGF-dependent autocrine loops. MET's growth- promoting activity causes cellular transformation, whereas its ability to enhance motility and survival accounts for invasion and metastasis. Although semaphorins and plexins might also be involved in cancer, their role in neoplastic growth is not well established yet. Preliminary studies have demonstrated that SEMA3C is overexpressed in metastatic lung adenocarcinomas and in recurrent squamous cell carcinomas resistant to radiation and cytostatic drugs (Yamada et aJ . Identification of semaphorin E as a non-MDR drug resistance gene of human cancer. Proc Natl Acad Sci USA 94:14713-8,

1997; Martin-Satue et al . Identification of semaphorin E gene expression in metastatic human lung adenocarcinoma cells by mRNA differential display, J Surg Oncol 72:18-23, 1999). Another secreted semaphorin, SEMA3E, is overexpressed in metastatic cell lines in comparison with the non-metastatic parental population (Christensen et al . Transcription of a novel mouse semaphorin gene, M-semaH, correlates with the metastatic ability of mouse tumor cell lines. Cancer Res 58:1238-44, 1998) .

RESULTS AND DISCUSSION To determine whether SEMA4B is expressed in human tumors, RT-PCR analysis was performed on a panel of 20 different human tumor cell lines of various origin. Significant expression of SEMA4B was detected 11 of the 20 tumor cell lines investigated, i.e. in 4 of 5 colon carcinoma cell lines, 2 of 3 lung, 3 of 6 breast, 1 of 2 ovarian, and 1 of 2 prostate tumor cell lines (Figure 6) . Subsequently, Q- PCR analysis was performed to quantify the relative expression levels in the different cell lines (Figure 7) . SEMA4B appeared to be relatively abundantly expressed (mean CT = 27.7). In general, the relative expression levels measured by Q-PCR analysis correlated with the RT-PCR results. A clear exception is the high expression of SEMA4B detected in the prostate cancer cell line DU145 by Q-PCR that revealed no detectable expression by RT-PCR analysis. Due to usage of different primers in RT-PCR versus Q-PCR analysis, alternative splicing or other organ and/or tumor specific modifications might have affected amplification of products in both assays possibly explaining the observed differences between the RT-PCR and Q-PCR results. Two independent SEMA4B-specific antibodies were generated against peptide 1 (SVPNIISTSRVSAPC) and peptide 2 (DPAFVASAYIPESLGSLC) , respectively, both derived from human SEMA4B (see also Example 8) . Different human tumor cell lines derived from breast (MCF7, T-47D, MDA-MB-231, MDA-MB-435s, MDA-MB-468) and colon (SW480) were selected based on differential expression of SEMA4B as determined by RT-PCR analysis (Figure 6) . Interestingly, RT-PCR data from these cell lines correlated with the Western blot results obtained with both antisera (Figure 8) further confirming expression of SEMA4B in different human tumor cell lines and suggesting relevance of SEMA4B in human cancer. cDNA and protein sequences of murine Sema4B as well as of its human orthologue SEMA4B are given in Figure 9 (SEQ ID NO. 31-38) .

EXAMPLE 7 PSK

INTRODUCTION

Mammalian STE20/mitogen-activated protein kinase kinase kinase kinase (MAP4K) family consists of approximately 30 serine/threonine kinases related in their catalytic domains. By analogy with the prototype STE20 kinase in Saccharomyces cerevisiae, mammalian MAP4K kinases are likely to regulate changes in transcription, cytoskeletal organization, and cell cycle progression in response to extracellular signals.

Recently, it was demonstrated that the STE20/MAP4K family member HGK (hepatocyte progenitor kinase-like/germinal center kinase-like kinase) is overexpressed in multiple human tumor cells and is actively involved in cell transformation and invasion as well as in reducing the adhesive properties of tissue culture cells (Wright et al . The STE20 kinase HGK is broadly expressed in human tumor cells and can modulate cellular transformation, invasion, and adhesion. Mol Cell Biol 23:2068-82, 2003). Another STE20/MAP4K family member was isolated from primary prostate carcinomas and therefore named PSK, for prostate-derived STE20-like kinase (Moore et al . PSK, a novel STE20-like kinase derived from prostatic carcinoma that activates the c-Jun N-terminal kinase mitogen- activated protein kinase pathway and regulates actin cytoskeletal organization. J Biol Chem 275:4311-22, 2000).

RESULTS AND DISCUSSION

To determine whether PSK is expressed in human tumors, RT-PCR analysis was performed on a panel of 20 different human tumor cell lines of various origin. Expression of PSK was detected in most of the tumor cell lines investigated (Figure 10) . Subsequently, Q-PCR analysis was performed to quantify the relative expression levels in the different cell lines (Figure 11) . PSK appeared to be relatively abundantly expressed (mean CT = 27.6) . Highest expression levels were detected in the breast tumor cell lines MCF7 and MDA-MB-435s. In general, the relative expression levels measured by Q-PCR analysis did not show a clear correlation with the RT-PCR results. Due to usage of different primers in RT-PCR versus Q-PCR analysis, alternative splicing or other organ and/or tumor specific modifications might have affected amplification of products in both assays possibly explaining the observed differences between the RT-PCR and Q-PCR results. cDNA and protein sequences of the murine orthologue of PSK as well as of human PSK are given in Figure 12 (SEQ ID NO. 39-46) .

EXAMPLE 8

Development of antibodies directed against SEMA4B

PRODUCTION OF SEMA4B PEPTIDES

SEMA4B encoding a membrane-bound protein was selected from the identified genes as a target for antibody development. Gene-specific peptides of SEMA4B predicted to be immunogenic and adopt a conformation similar to that of the corresponding region of the native protein were synthesized for immunization purposes. Software programs such as Antigen Prediction' within the EMBOSS package of the UK HGMP Resource Center website were employed.

The following peptides were synthesized and conjugated to BSA (bovine serum albumin) :

SEMA4B: peptide 1 SVPVIISTSRVSAPC (SEQ ID NO. 25) peptide 2 DPAFVASAYIPESLGSLC (SEQ ID NO. 26)

PRODUCTION OF SEMA4B ANTIBODIES

Polyclonal antibodies directed against SEMA4B were generated by immunization of rabbits with gene-specific, BSA- conjugated peptides described above using standard protocols. The presence of antibodies directed against the target antigens was confirmed by screening sera from immunized rabbits against selected human tumor cell lines expressing SEMA4B using Western blot analysis.

WESTERN BLOT ANALYSIS USING SEMA4B ANTIBODIES

Different human tumor cell lines derived from breast (MCF7, T-47D, MDA-MB-231, MDA-MB-435s, MDA-MB-468) and colon (SW480) were selected based on differential expression of SEMA4B as determined by RT-PCR analysis (Figure 6) . Protein lysates from these cell lines were prepared and subjected to Western blot analysis using two independent polyclonal antibodies generated against peptide 1 (SVPVIISTSRVSAPC) and peptide 2 (DPAFVASAYIPESLGSLC), respectively, both derived from human SEMA4B. Interestingly, RT-PCR data from these cell lines correlated with the Western blot results obtained with both antisera (Figure 8) indicating that the described peptides can be successfully used for the development of antibodies directed against SEMA4B .

Claims

1. Use of inhibitors of the expressed proteins of the murine genes and/or their human orthologues listed in Table 1 for the preparation of a therapeutic composition for the treatment of cancer, in particular for the treatment of solid tumors of lung, colon, breast, prostate, ovarian, and pancreas as well as leukemia and lymphoma.

2. Use as claimed in claim 1, wherein the inhibitors are antibodies or derivatives thereof directed against the expression products of the genes that are expressed on the cell membrane.

3. Use as claimed in claim 1 or 2, wherein the antibodies or derivatives thereof are directed against one of the following peptides: SVPVIISTSRVSAPC (SEQ ID NO. 25) or DPAFVASAYIPESLGSLC (SEQ ID NO. 26) .

4. Use as claimed in claim 1, 2 or 3, wherein the derivatives are selected from the group consisting of scFv fragments, Fab fragments, chimeric antibodies, bifunctional antibodies, and other antibody-derived molecules.

5. Use as claimed in claim 1, wherein the inhibitors are small molecules interfering with the biological activity of the protein expressed by the gene.

6. Use of inhibitors of the mRNA transcripts of the genes listed in Table 1 for the preparation of a therapeutic composition for the treatment of cancer.

7. Use as claimed in claim 6, wherein the inhibitors are antisense molecules, in particular antisense RNA or antisense oligodeoxynucleotides .

8. Use as claimed in claim 6, wherein the inhibitors are double stranded RNA molecules for RNA interference .

9. Use as claimed in claim 1 and claim 6, wherein the treatment comprises gene therapy.

10. Use as claimed in any one of the claims 1-9, wherein the therapeutic composition is for treatment of inflammatory diseases.

11. Use as claimed in any one of the claims 1-10, wherein the gene is SELPLG, comprising a cDNA sequence which shares at least 90% identity to the cDNA sequences shown in Figure 5 (SEQ ID NO. 27 and 29) .

12. Use as claimed in any one of the claims 1-10, wherein the gene is SEMA4B, comprising a cDNA sequence which shares at least 90% identity to the cDNA sequences shown in Figure 9 (SEQ ID NO. 31, 33, 35, and 37) .

13. Use as claimed in any one of the claims 1-10, wherein the gene is PSK, comprising a cDNA sequence which shares at least 90% identity to the cDNA sequences shown in Figure 12 (SEQ ID NO. 39, 41, 43, and 45) .

14. Use as claimed in any one of the claims 1-10, wherein the expressed protein is SELPLG, comprising a sequence which shares at least 90% identity to the protein sequences shown in Figure 5 (SEQ ID NO. 28 and 30) .

15. Use as claimed in any one of the claims 1-10, wherein the expressed protein is SEMA4B, comprising a sequence which shares at least 90% identity to the protein sequences shown in Figure 9 (SEQ ID NO. 32, 34, 36, and 38) .

16. Use as claimed in any one of the claims 1-10, wherein the expressed protein is PSK, comprising a sequence which shares at least 90% identity to the protein sequences shown in Figure 12 (SEQ ID NO. 40, 42, 44, and 46) .

17. Inhibitor compound directed against the expressed proteins of a murine gene and/or its human orthologue listed in Table 1 for use in the treatment of cancer.

18. Inhibitor compound as claimed in claim 17, which is an antibody or derivatives thereof directed against the expression products of a gene that is expressed on the cell membrane.

19. Inhibitor compound as claimed in claim 18, wherein the antibodies or derivatives thereof are directed against one of the following peptides: SVPVIISTSRVSAPC (SEQ ID NO. 25) or DPAFVASAYIPESLGSLC (SEQ ID NO. 26) .

20. Inhibitor compound as claimed in claim 17, 18 or 19, wherein the derivative is selected from the group consisting of scFv fragments, Fab fragments, chimeric antibodies, bifunctional antibodies, or other antibody- derived molecules.

21. Inhibitor compound as claimed in claim 17, which is a small molecule interfering with the biological activity of the protein expressed by the gene.

22. Inhibitor compound directed against the transcription product (mRNA) of a murine gene and/or its human orthologue listed in Table 1 for use in the treatment of cancer.

23. Inhibitor compound as claimed in claim 22, which is an antisense molecule, in particular an antisense RNA or an antisense oligodeoxynucleotide .

24. Inhibitor compound as claimed in claim 23, which is a double stranded RNA molecule for RNA interference.

25. Inhibitor compound as claimed in any one of the claims 17-24 directed against the SELPLG gene, comprising a cDNA sequence which shares at least 90% identity to the cDNA sequences shown in Figure 5 (SEQ ID NO. 27 and 29) .

26. Inhibitor compound as claimed in any one of the claims 17-24 directed against the SEMA4B gene, comprising a cDNA sequence which shares at least 90% identity to the cDNA sequences shown in Figure 9 (SEQ ID NO. 31, 33, 35, and 37) .

27. Inhibitor compound as claimed in any one of the claims 17-24 directed against the PSK gene, comprising a cDNA sequence which shares at least 90% identity to the cDNA sequences shown in Figure 12 (SEQ ID NO. 39, 41, 43, and 45) .

28. Inhibitor compound as claimed in any one of the claims 17-24 directed against the expressed protein SELPLG, comprising a sequence which shares at least 90% identity to the protein sequences shown in Figure 5 (SEQ ID NO. 28 and 30) .

29. Inhibitor compound as claimed in any one of the claims 17-24 directed against the expressed protein SEMA4B, comprising a sequence which shares at least 90% identity to the protein sequences shown in Figure 9 (SEQ ID NO. 32, 34, 36, and 38) .

30. Inhibitor compound as claimed in any one of the claims 17-24 directed against the expressed protein PSK, comprising a sequence which shares at least 90% identity to the protein sequences shown in Figure 12 (SEQ ID NO. 40, 42, 44, and 46) .

31. Therapeutic composition for the treatment of cancers in which one or more of the murine genes and/or their human orthologues listed in Table 1 are involved, comprising a suitable excipient, carrier or diluent and one or more inhibitor compounds as claimed in claims 17-30.

32. Compositions as claimed in claim 31, wherein the cancer is a solid tumor of e.g. lung, colon, breast, prostate, ovarian, and pancreas.

33. Compositions as claimed in claim 31, wherein the cancer is leukemia or lymphoma.

34. Therapeutic composition for the treatment of inflammatory diseases in which one or more of the murine genes and/or their human orthologues listed in Table 1 are involved, comprising a suitable excipient, carrier or diluent and one or more inhibitor compounds as claimed in claims 17- 30.

35. Method for the development of therapeutic inhibitor compounds as claimed in claims 17-30, which method comprises the steps: a) identification of genes involved in cancer, in particular by using retroviral insertional tagging, optionally in a specific genetic background; b) validation of one or more of the identified genes as relevant target genes for therapeutic compounds by one or more of the following methods: determination of the expression profile of the identified genes in human tumors and normal tissues;

- determination of the functional importance of the identified genes for cancer; c) development of therapeutic compounds by one or more of the following methods: - production of expression products of the validated genes and use of these products for production and/or design of therapeutic compounds; use of the gene sequence to design therapeutic compounds.

36. Method as claimed in claim 35, wherein the gene identified in step a) is selected from the murine genes and/or their human orthologues listed in Table 1.