WO2003097843A1 - Tumor suppressor locus in prostate cancer - Google Patents

Abstract

The long arm of chromosome 6 is frequently deleted in diverse human neoplasm. Studies showed a minimum deletion region between markers D6S1056 and D6S300 on chromosome 6q in primary CaP (prostate cancer). Significant association of LOH (loss of heterozygosity) of 6q16 with aging CaP patients and patients with seminal vesicle invasion was shown using LCM (laser capture microdissection)-derived DNA samples from 48 CaP patients. A novel contig designated as 6qTSG1 was generated within a minimal 200 kb region centered around D6S1013. The 6qTSG1 exhibited complex multiple splicing variants with low or absent expression in CaP cells. However, none of the transcripts identified contained a significant open reading frame. Expression of 6qTSG1 was induced in LNCaP cells that were cultured in charcoal-stripped serum medium suggesting an up-regulation of 6qTSG1 by androgen ablation. Induction of 6qTSGl expression in response to androgen ablation was abrogated in androgen independent derivatives of LNCaP cells. The 6qTSG1 inhibited tumor cell growth after transfection into PC3 cells. A candidate CaP suppressor locus was identified on chromosome 6q16.1, of which alteration linked to CaP progression.

Description

TUMOR SUPPRESSOR LOCUS IN PROSTATE CANCER Rights in the Invention

This invention was made, in part, with support from a grant by the Center for Prostate Disease Research, a program of the Henry M. Jackson Foundation for the Advancement of Military Medicine, funded by the United States Army, Medical Research and Material Command, and the United States Government may have certain rights in this application- Reference to Related Applications

This application claims priority to U.S. Provisional application number 60/379,758, entitled "Characterization of Prostate Cancer Associated Tumor Suppressor Gene Locus," filed May 14, 2002, which is completely and entirely incorporated herein by reference. Background

1. Field of the Invention

This invention relates to tools and methods for the identification of a tumor suppressor locus and, in particular, to a tumor suppressor locus of human chromosome 6. The invention further relates to the tumor suppressor locus of human chromosome 6 and to product expressed from such loci, and to methods for the treatment and prevention of cancer that relate to or are derived from the tumor suppressor locus.

2. Description of the Background

Prostate cancer is the most common cancer and the second leading cause of cancer- related deaths in American men. IN 2002, the American Cancer Society has estimated 189,000 new CaP cases and 30,200 deaths due to CaP ( Jemal et al., 2002). However, the understanding of the molecular genetics that cause this malignancy lags behind other major types of human cancers because of the complexity of genetic alterations and highly heterogeneous nature of tumor. Chromosomal hot spots associated with sporadic as well as inherited prostate cancers are being intensely analyzed for putative TSGs (tumor suppressor genes). A number of chromosomal regions have been identified for their frequent deletions in CaP, including 6ql6-q22, 7q31, 8p21-p22, lOp, 10q23-q24, 12pl2-13, 13ql4-q21, 16q22-24, and 18q21-q24 (Srikantan et al, 1999; Hyytinen et al, 2002; Zenklusen et al, 1994; Bova et al, 1993; Cabeza-Arvelaiz et al, 2001a; Narla et al, 2001 ; Li et al, 1997; Cabeza-Arvelaiz et al, 2001b; Kibel et al, 1998; Hyytinen et al, 1999; Suzuki et al, 1996; Latil et al, 1994). These chromosome regions are suspected of harboring critical TSGs for prostate cancer. There has been extensive work from several laboratories on the characterization of chromosome 8p21-22 locus but the identity of the TSG remains elusive (Sato et al., 1999; Dahiya et al, 1998; Noeller et al, 1997; Cabeza-Arvelaiz et al, 2001a) . NKX3.1 on the 8p21 locus is a promising candidate TSG based on studies of frequent 8p21 LOH, ΝKX3.1 knockout mice, loss of expression in advanced CaP and its tumor cell growth inhibitory functions. However, no tumor-associated mutations have been reported (Noeller et al, 1997), which suggests that it is not a classical TSG. PTEΝ/MMACI on chromosome 10q25 has been discovered as a TSG frequently altered in advanced CaP (Li et al., 1997). PTEΝ is by far the most frequently mutated gene in advanced CaP (Gray et al, 1998). Recently described KLF6 (Kruppel-like factor 6) gene on chromosome 1 Op 15 is frequently mutated in sporadic CaP (Νarla G. et al, 2001). With KLF6 mutation in over 50% of prostate tumors, the original report has placed KFL6 at the top of the list of genetic changes in CaP. However, further study is needed to confirm these initial exiting findings. In familial CaP, chromosome lq is identified as the primary suspected locus (Issacs, W. et al., 2001). Recent reports have described germline mutations of ELAC2 on 17pl 1 and RNASELl on lq25 in a small subset of CaP prone families (Xu et al, 2001; Carpten et al, 2002; Tavtigian et al, 2001). Overall, each of the TSG alterations is associated with only a subset of the CaP suggesting for low penetrance alleles. Identification of specific genes on frequently altered chromosome loci in CaP remains a major challenge and represents one of the most important areas in developing a mutation profile of prostate cancer. Summary of the Invention

The present invention overcomes the problems and disadvantages associated with current strategies and designs, and provides new tools and methods for identifying and applying tumor suppressor loci in the treatment of cancer.

One embodiment of the invention is directed to isolated tumor suppressor loci derived from human chromosome 6 and, preferably, sequence derived from the region identified as 6qTSGl . Loci may be purified or recombinant, and may comprise one or more sequences selected from SEQ ID ΝOS 57-64. Promoters that are selectively active in cancerous cells may be isolated and coupled to therapeutically useful antineoplastic genes for expression in cancerous cells. Alternatively, sequence selectively expressed in cancerous cell can be isolated and functionally coupled to recombinant promoters, which may be inducible. Another embodiment of the invention is directed to methods for the treatment and/or prevention of cancer by stimulation of 6qTSGl expression. Agents that can be used to stimulate expression include, but are not limited to, the presence or absence of steroids, such as testosterone and progesterone.

Another embodiment of the invention is directed to methods for the treatment and/or prevention of cancer by directly, administering sequence of 6qTSGl into cells. Introduction may be as recombinant vectors that express 6qTSGl, or parts of 6qTSGl.

Another embodiment of the invention is directed to diagnostic kits and methods for the detection of cancer in tissues suspected of containing cancer. Kits may contain sequences that will hybridize to or antibodies to will bind to expression products from 6qTSGl . Such methods can be used to detect the presence or absence of cancer and also may be used to determine the severity of disease in patients.

Other embodiments and advantages of the invention are set forth in part in the description, which follows, and in part, may be obvious from this description, or may be learned from the practice of the invention. Description of the Figures

Figure 1. Association of LOH at chromosome 6ql6.1 with patient age. (A) The frequency of LOH at 6ql6.1 significantly increased in patients with older age. (B) LOH on 6ql6.1 is significantly increased in the patients at or over 60 years old compared to the patients less than 60 years old (pθ.0001).

Figure 2. Physical map and transcript map of the LOH candidate region on chromosome 6ql6.1. The physical map shows a 700 kb segment of human chromosome 6ql6.1 between markers D6S1056 and D6S300. The 200 kb minimal LOH region between markers p991 and D6S1003 is shown as a hatched box that is covered by three overlapping contigs: AL591519, AL590305 and AL356094. The transcript map shows the gene and ESTs mapped in this region. Genes are depicted by arrows to represent transcriptional orientation. The exon-intron structure oiόqTSGl was deduced by comparison of cDNA and genomic sequences. The trapped exons are shown by arrowheads. Figure 3. Schematic representation of the splicing variants of the 6qTSGl gene. Alignment of expressed 6qTSGl cDNAs isolated in cDNA library screening, RACE- and RT-PCR experiments revealed a total of eight splicing variants ranging in size from 2.9 kb to 4.8 kb represented a contig of at least 15 exons spanning 80 kbp of genomic DNA. The sites of polyadenylation at the 3 '-terminal exons are denoted with A's. The exon-intron boundaries were determined by aligning the cDNA sequences against the published genomic sequence. Notably, none of the transcripts contains a long ORF, indicating that 6qTSGl may function as a non-protein coding RNA molecule. The asterisks show those that were chosen for functional assay.

Figure 4. Northern blot analysis of the 6qTSGl gene. The 435 bp core sequence of 6qTSGl cDNA was used to probe multiple human tissue blots including prostate tissues (Clontech). Two main transcripts, 4.4kb and 3.4kb, were found ubiquitously expressed in almost all tissues tested.

Figure 5. RT-PCR analysis of the candidate transcripts present in minimal region of 6ql6.1 in prostate cancer cell lines. Total RNA was isolated from tumor cells cultured in regular fetal bovine serum (FBS) growth media. Total RNA was treated with DNase I followed by reverse transcription and subjection to PCR analysis. A housekeeping gene, GAPDH, was used as a loading control.

Figure 6. Induction of expression of 6qTSGl and EST (AI814228) and 6qTSGl by androgen ablation. A. Up-regulation of 6qTSGl and EST (AI814228) expression by androgen ablation. LNCaP cells were grown in medium supplemented with regular FBS, and then switched to the medium containing charcoal-stripped FBS. Cells were collected for total RNA isolation as indicated by time. The expression of EST (AI814228) and 6qTSGl was analyzed by RT-PCR. A housekeeping gene, GAPDH, was used as a loading control. B. The negative response of the expression of 6qTSGl to androgen. LNCaP cells were cultured in charcoal-stripped FBS medium for five days, and indicated concentration of androgen analog R1881 was added for 48 hours. The expression of 6qTSGl was analyzed by RT-PCR. The DD3 gene was used as a control that showed increased expression when LNCaP cells were cultured in androgen-containing medium.

Figure 7. Exogenous expression of 6qTSGl C form, but not H form, inhibits colony formation of PC3 cells. PC3 cells (2x 10⁵ cells) were plated in a 60 mm culture dish and incubated at 37°C one day before transfection. For transfection, 5 μg of each plasmid DNA: pcDNA3.1(+), pcDNA-6qTSG-C (sense) or pcDNA-6qTSG-C (antisense), pcDNA-6qTSG- H (sense) or pcDNA-6qTSG-H (antisense) was transfected using TransFast reagent (Promega) according to the protocols provided. After 24 hours, cells were cultured in medium containing 400-500 μg ml of geneticin (G-418) for 3-4 weeks. Cell colonies were stained with crystal violet (lmg/ml) and counted.

Figure 8. Sequence of Isoform A (contains exon 12; 86556-86645)(4798 bp).

Figure 9. Sequence of Isoform B (4708 bp).

Figure 10. Sequence of Isoform C from όqTSlGl (contains primers 1109 and 1026 without pl l l l) (2840bp).

Figure 11. Sequence of Isoform D (4607 bp).

Figure 12. Sequence of Isoform E from όqTSlGl (contains primers 1044 and 1170 with pi 042; 2949bp) (4347bp).

Figure 13. Sequence of Isoform G from όqTSlGl (contains primers 1044 and 1170 with pi 042 (4822bp).

Figure 14. Sequence of Isoform H from όqTSlGl (contains primers 1044 and 1170 without pl042 (2949bp).

Figure 15. Sequence of Isoform F from όqTSGl (contains primers 1044 and 1042. The

3 'end is expanded by exonlO without splice) (3578bp).

Description of the Invention

As embodied and broadly described herein, the present invention is directed to tools and methods for the identification and use of tumor suppressor loci and products derived therefrom in the diagnosis, treatment and prevention of prostate cancer.

Cancer, in one sense, can be considered an alteration in the expression of a number and variety of different genes leading to uncontrolled and undesired tissue growth. Types of genes that are typically altered include dominant-acting oncogenes, tumor suppressor genes ("TSG"), genes affecting cell cycle and genes affecting genomic stability, to name a few.

Tumor suppressor genes often function as negative regulators of cell growth. As the term implies, expression of a TSG in a target cell typically evokes a loss of cell proliferative ability. TSG are often difficult to identify and isolate. Most methods of identification require extensive mapping of known genetic sites containing known tumor suppressor function.

It has been surprisingly discovered that a TSG exists on human chromosome 6 and that suppressor function was effective in reducing the proliferation of cancer cells. This was in part surprising because the locus, referred to as 6qTSGl (about 80 kbp), does not contain a major opening reading frame, but instead, at least fifteen small open reading frames (see Figure 3), that can be expressed as at least eight different transcripts (at least isoforms A - H; see Figures 8-15), based on splice variations. Experiments have been performed analyzing each of the isoforms to identify open reading frames, expression products (RNA and protein), and regions expressed in cancerous, pre-cancerous and normal cells, and to correlate those results with the ability to reduce tumor cell proliferation.

One embodiment of the invention is directed to a tumor suppressor loci isolated from human chromosome 6q. TSG function can be localized to positions 6ql6.1 to 6q25, which is capable of expressing at least eight different isoforms. This sequence has been cloned and introduced into recombinant vectors for analysis and experimentation. A further embodiment of the invention is directed to the isoforms expressed from 6qTSGl and, preferably, the isoform sequences of SEQ ID NOS 57-64 including but not limited to, functionally active, but otherwise neutral nucleotide substitutions thereof. A further embodiment of the invention is directed to the peptide sequences expressed from 6qTSFl including, but not limited to, neutral amino acid substitutions thereof. Preferably the peptide and nucleotide sequences, which may comprise DNA, RNA, or PNA, function as tumor suppressors that, when present at an effective level, reduces cell proliferation in cell cultures or in vivo.

Another embodiment of the invention is directed to methods for the treatment and/or prevention of cancer by stimulating expression of the 6q TSG locus. Expression of this locus has been shown to correlate with reduced cell proliferation and, thus, tumor formation. Agents whose presence or absence may stimulate expression include, but is not limited to, commercially available transcription inducers, pharmaceuticals used for chemotherapy of cancer patients, androgens, steroids such as testosterone or progesterone (or available derivatives thereof), or antagonists or combination of such agents.

Alternatively, 6qTSGl expression can be increased directly by genetic therapy. Vectors (e.g. prokaryotic, eukaryotic, shuttle) containing the 6qTSGl sequence, or parts thereof, can be introduced into cells suspected of being cancerous or pre-cancerous, and induced to express the TSG function. That function may be carried upon the entire region such as, for example, on an RNA product, or on significant portions such as translation products expressed from one or more exons.

Another embodiment of the invention is directed to promoters isolated from 6qTSGl (as can be identified from known promoter sequences by those of ordinary skill in the art), which can be functionally coupled to recombinant genes. Thus, the recombinant genes would be preferentially expressed in neoplastic cells. Examples of recombinant genes which would be useful to express in cancerous cells include, but are not limited to, genes that encode toxic products, anticancer pharmaceuticals, immune system modulators (e.g. an interleukin, a growth factor, a proliferation suppressor, an interferon, etc.), and combinations thereof. Sequences within 6qTSGl that become expressed may also be functionally coupled to a recombinant promoter (e.g. inducible promoters) and introduced to cells to increase the presence of that expression product. All of these combinations may be useful for the treatment and/or prevention of cancer.

Another embodiment of the invention is directed to a diagnostic for detecting the presence or absence of one or more of these isoforms in tissues suspected of being or becoming cancerous. Typical tissues include, but are not limited to, breast, prostate, hard and soft tissue tumors, bone, lymph, skin, and organ tissue. Diagnostic kits may contain DNA, RNA or PNA sequences designed to hybridize with RNA products of 6qTSGl, antibodies that specifically bind to protein or peptides expressed from RNA isoforms, or both. Antibodies may be derived from human, murine, or any mammalian source, or may be humanized to minimize undesirable immunological affects. Antibodies may be polyclonal or monoclonal (IgG, IgM, IgD, IgA, IgE), or portions of antibodies (e.g. Fv portions), which demonstrate specificity to translation products of 6qTSGl that correlate with the presence of disease. Diagnosis involves obtaining a biological sample (e.g. tumor tissue, biopsy sample, blood, plasma, cells) from a patient suspected of having a cancer or pre-cancerous condition and contacting that sample with a diagnostic kit of the invention. Results from multiple testing and comparisons with normal levels in non-diseased patients, can be used to create a database of expression levels for 6qTSGl expression (either nucleic acid or peptide), which can be used as a standard from which to determine the presence or absence of disease. Results of individual tissue testing may be used to detect disease or simply determine the severity of disease. Identification of 6qTSGl

The existence of this TSG was supported directly by microcell-mediated chromosomal transfer experiments in numerous types of tumor cells. Introduction of the entire normal chromosome 6 into a melanoma cell suppresses the tumorigenesis (Trent et al, 1990). In breast tumors, transfer of chromosome 6 into MDA-MB-231 cells altered cell growth in vitro and inhibits tumor formation in vivo. In addition, the tumor cell that retained chromosome 6q underwent senescence. The region has been mapped to 6q21-q23 (Negrini et al, 1994). In another study, the chromosome segment which conferred loss of tumorigenicity and reversion of other neoplastic properties, has been defined on a small donor chromosome fragment on 6q23.3-q25 flanked by D6S292 and D6S311 (Theile et al, 1996). In an SV40-immortalized fibroblast cell line, introduction of chromosome 6 caused cell senescence suggesting that chromosome 6 also contains senescence-related genes (Sandhu et al, 1994). Further study in ovarian tumor cells led to the localization of a senescence gene on chromosomal region 6ql4- 21 (Sandhu et al, 1996).

Microcell-mediated chromosomal transfer experiments also provided evidence that chromosome 6 was implicated in the regulation of tumor metastasis. A potential tumor metastasis suppressor locus was functionally linked to 6ql6.3-q23 (Miele et al, 2000). KiSSl was identified by differential gene expression in chromosome 6 introduced-melanoma cells. Expression of KiSSl inhibits metastasis of human melanoma and breast tumor cells (Lee et al, 1997). It has been demonstrated that a gene localized on 6ql6.3-q23 possibly regulated expression of KiSSl (Miele et al, 2000). Shirashki et al. showed that loss of KiSSl expression was associated with LOH of 6ql6.3-q23 in melanoma progression (Shirasaki et al., 2001). A highly metastatic melanoma cell line C8161 lost its metastatic ability, but retained its tumorigenetic ability after it received chromosome 6. The chromosome 6- transferred melanoma cell line expressed higher levels of Manganese Superoxide Dismutase (MnSoD) and sensitized the cells to UN-induced expression of the p53 gene (Alvarez et al, 1998). Both MnSoD and p53 positively regulate the expression of MASPIN, which suppressed tumor cell invasion in vitro and tumor cell metastasis in vivo (Zou et al, 1994). Taken together, these studies may indicate that chromosome 6 harbors a gene or genes, which suppresses tumorigenesis and metastasis directly or indirectly through its target.

Further delineation of the tumor suppressor locus on chromosome 6q has been mainly achieved using LOH assays and comparative genomic hybridization. There has been substantial clinical and experimental evidence that consistently supports the involvement of genes located within human chromosome 6ql6-21 that control tumorigenesis. Allelic deletions in this region were detected by loss of heterozygosity (LOH) analysis in many neoplasms, including carcinomas of the prostate (Srikantan et al, 1999; Cooney et al, 1996), breast (Orphanos et al, 1995), ovarian (Wan et al, 1999), salivary gland (Queimado et al., 1998), and melanoma (Millikin et al, 1991). In general, the deletion on chromosome 6q involves a wide chromosome segment. Studies in different types of tumors yielded different deletion patterns. Prostate tumors showed LOH clustering around 6ql6-q21 with a frequency of 30% to 50% from three independent studies (Srikantan et al, 1999; Cooney et al, 1996).

The precise map of the LOH region was discovered on chromosome 6ql6.1 region. The identification of putative candidate tumor suppressor was discovered to be contig(s) within the minimal 200 kb deleted region on chromosome 6ql6.1.

At least eight isoforms of 6qTSgl have been identified and sequences. The exact open reading frames for these isoforms can be easily translated by those of ordinary skill in the art, and are schematically represented in Figure 3. Exon positions in these isoforms as well as splice donor and acceptor sites are shown in Table 4.

The following examples illustrate embodiments of the invention, but should not be viewed as limiting the scope of the invention. Examples

Patient specimens and DNA preparation

LCM-derived Matched normal prostate and tumor prostate epithelial cells were obtained from frozen section of radical prostatectomies performed at Walter Reed Army Medical Center (WRAMC) (under an IRB-approved protocol). Clinical information was retrieved from the CPDR Tri-service Multicenter Prostate Cancer Patient Database. Both normal and tumor cells of each patient were taken from the same slides by Laser Capture Microdisection (LCM) of frozen tissue sections.

DNA was prepared by incubating the LCM-derived cells in buffer (10mm Tris, 1 mM EDTA, 1% Tween 20) at 37°C for overnight, and then heating at 98°C for 10 minutes. The crude DNA preparation was used directly for polymerase chain reaction. Cell Cultures

Prostate tumor cell lines: DU145, LNCaP, PC3 and 293 kidney embryonic cells were obtained from American Type Culture Collection (ATCC, Rockville, MD), and were maintained in growth medium recommended by the supplier. Immortalized human primary prostate epithelial cells, RC58T/h and 957E/h, were provided by Dr. John Rhim (CPDR, USUHS). All cell lines were grown at 37°C in a humidified incubator with 5% CO₂. Loss of Heterozygosity analysis of 6ql6.1 locus

Polymorphic microsatellite between chromosome 6ql6.3-21 markers were selected from public Database as defined previously (Srikantan et al, 1999). Eight additional microsatellite markers were selected from centromere to telomere, based on the information from the Whitehead/MIT Database (http://www.genome.wi.mit.edu/): D6S275-D6S1274-D6S1056- D6S361-D6S1013-D6S300-D6S1054-D6S424. The physical distance between markers D6S275 and D6S424 was found to be 2.5 Mb based on the sequence map. Primers for new polymorphic markers p991, p978 and pi 003 were designed based on the complete genomic sequence and analyzed for polymorphism.

PCR were performed with genomic DNA from each paired tumor and normal sample using fluorescence labeled primers. The PCR product was analyzed on a Perkin Elmer ABI PRISM 310 Genetic Analyzer (Applied Biosystems, CA). LOH was analyzed using GeneScan and Genotype software. όqTSGl Transcript identification

Transcripts were identified by database searches utilizing human genome resources on chromosome 6 (http://www.ncbi.nlm.nih.gov), exon-trapping of BAG (bacterial artificial chromosome) DNA clones within the minimum deletion region of interest, computational prediction of exons from genomic DNA sequences using GeneScan (http://genes.mit.edu/GENSCAN.html). Exon trapping

BAC DNA was purchased from Research Genetics (Huntsville, AL). Restriction fragments from BAC DNA (accession number: AL356094) were shotgun subcloned into the EcoRI site of exon-trapping vector pSPL3 and transfected into COS-1 cells following the procedure provided by the manufacturer (Invitrogen). Spliced products obtained by RT-PCR were cloned into pAMPIO using the UDG cloning kit provided in the exon-trapping kit (Invitrogen) and sequenced. cDNA library screening

Normal prostate cDNA libraries were screened by Edge BioSystems (Gaithersburg, MD) using a 435bp core sequence of όqTSGl as the probe. DNA sequencing was performed using the AmpliCycle sequencing kit from Applied Biosystems (Foster City, CA) on the 3100 DNA Genetic Analyzer. Complete cDNA sequence was aligned with the published genomic sequence.

Rapid amplification of cDNA ends (RACE)

5' or 3' RACE (rapid amplification of cDNA ends) was performed on mRNA obtained from LNCaP cells according to the manufacturer's instructions (Invitrogene). Marathon ready cDNA libraries from human prostate and cDNAs derived from human testis, prostate, and skeletal muscle (Ambion) were also used as templates for RACE reactions. Typically, two antisense or sense gene-specific primers were designed from the partial cDNA sequences. Adaptor primer was used in the primary RACE reaction with the 3' or 5' outer primer. Primary PCR product was 1 :100 diluted and used as a template for secondary RACE reactions using the nested adaptor primer and nested gene-specific primers. Products were analyzed by gel electrophoresis and subcloned into the TA Cloning vector (Invitrogen). The clones that contained the insert were sequenced using the AmpliCycle sequencing kit from Applied Biosystems (Foster City, CA) on the 3100 DNA Genetic Analyzer. Northern analysis

Northern blots containing multiple human tissue mRNA including prostate were purchased from Clontech. To determine the transcript size(s) and tissue distribution pattern of potential transcripts, blots were hybridized witli either gel-purified inserts from available IMAGE cDNA clones or PCR products generated from marathon-ready cDNA libraries (Clontech) or cDNAs derived from LNCaP cells. The latter was used if no IMAGE clone was available or if available clones contained a repetitive sequence. The probes were labeled with ³²P-dCTP by random priming (Stratagene) following the manufacturer's instructions. Hybridization was performed at 68°C for 12-14 hours in ExpressHyb Solution (Clontech) containing a lxl0⁶/ml ³²P-labled probe followed by stringent washing in 0.1%SDS; 2XSSC at room temperature, and 0.1%SDS; 0.1XSSC at 68°C. The blots were exposed to a Kodak XR film for autoradiography. Colony formation assay

The cDNA of two transcript variants of the putative gene, όqTSGl was obtained by RT-PCR from prostate mRNA (Clontech) using forward primer pi 109 (5' TCCAA GATGA GGTAG TCTCA GATG 3'; SEQ ID NO 1) or pl044 (5' GGCAT CATTG CAGTG ATGCT TTGGG 3'; SEQ ID NO 2) and reverse primer pl026 (5' AAGAT CTGCA TCGCT ATCTC TATGG 3'; SEQ ID NO 3) or pi 170 (5' TTTCA CAGGG TGATT GTGGG AATAC A 3'; SEQ ID NO 4), respectively, followed by cloning into PCR-blunt II TOPO vector (Invitrogen). An insert was excised from PCR-blunt II by EcoRI digestion and subcloned into mammalian expression vector pcDNA3.1(+) (Invitrogen) to generate pcDNA-1109+1026 or pcDNA- 1044+1170 sense and antisense constructs.

Cells (2x10⁵) were seeded in a 60 mm dish one day before transfection. The expression plasmid DNA (5 μg) was transfected into tumor cells using TransFast reagent (Promega) according to the protocols provided by the manufacturer. The cells were then cultured in medium containing 400-500 μg/ml of geneticin (G-418) for 2-4 weeks. Drug resistant colonies were stained with crystal violet and the counted. Frequent LOH on 6ql6.1 in LCM-derived Prostate Tumor Samples LOH on chromosome 6ql6-21 between markers D6S1056 and D6S300 is associated with CaP (Srikantan et al, 1999). The region runs about 700 kb covered by 7 overlapping BAC DNA clones with complete genomic sequence and is located on 6q 16.1. To further verify the association of the LOH with CaP in this region, eight microsatellite markers were analyzed on LCM-derived DNAs obtained from matched normal and tumor prostate epithelial cells of 48 CaP patients. Frequency of LOH was 54% by accumulative loss at one or more microsatellite markers analyzed. The frequent LOH was found between markers D6S1056 and D6S300. The frequency of LOH was 40%, 47%, and 40% for the markers D6S1056, D6S1013, and D6S300, respectively. The highest rate of LOH was at marker D6S1013 (47%) (Table 1). The high frequency of LOH on 6q 16.1 strongly indicated that 6ql6.1 is one of the chromosomal hotspots involved in prostate tumorigenesis.

LOH on 6ql6.1 Is Associated with Increased Age of CaP patients and aggressive CaP The association of LOH on 6ql6.1 with various clinico-pathologic features of CaP patients was statistically analyzed. LOH at one or more 6ql6.1 markers was taken as positive. The incidence of LOH on 6ql6.1 increased significantly from 16.7% for patients under 60 years old to 68.2% for patients at or over 60 years old (pO.OOOl, Figure la). The frequency of LOH increased with age from 7.1% for patients younger than 54 years old to 87.5% for patients older than 65 years old (Figure lb). Five out of seven CaP patients (71%) with seminal vesicle invasion showed LOH on 6ql6.1 while only 14 out of 39 CaP patients (36%) without seminal vesicle invasion exhibited LOH (Table 2). This result indicated that 6ql6.1 deletion potentially plays a role in the development of sporadic prostate tumors and links to development of more aggressive cancer.

Determination of Minimum Deletion Region on 6ql6.1 in CaP To further narrow down the candidate region, three additional new microsatellite markers were designed and two known markers identified, which were not used in previous studies, between markers of D6S1056 and D6S300 using the publicly available genomic sequence information. Eight tumor samples that retained allelic balance at either D6S300 or D6S1056 were analyzed for LOH (Table 3). All eight tumor samples showed LOH at D6S1013, but retained heterozygosity at either D6S1003 or p300. One tumor retained heterozygosity at marker p991 and two tumors retained heterozygosity at D6S361 (Table 3). These data indicate that the minimal deletion region was defined between markers of pi 003 and p991, which span 200 kb (Figure 2). Further, the parental LNCaP cells were compared with their androgen-independent sublines. The parental LNCaP showed heterozygosity at markers pi 003, pi 013, and p991, however, its derivatives exhibited a LOH pattern at marker D6S1013 and retained heterozygosity at flanking markers (Table 3). The first generation of LNCaP derivatives, LNCaP-C4, showed LOH at pi 013 and retained heterozygosity at pi 003 and p991, respectively; whereas, the second generation LNCaP-C4-2 and the bone metastatic line LNCaP-C4-2B contained wider deletions beyond pi 003. These results implicated the potential involvement of 6q 16.1 in prostate tumor progression. Transcript Mapping of the Minimum Deletion Region on 6ql6.1 A physical map of the minimum deletion region was constructed according to the public database and the Ensembl Human ContigView (http://www.ensembl.org/Homo_sapiens) (Figure 2). This region was covered by 3 overlapping BACs (AL591519, AL590305, and AL356094). Potential transcripts in the 200kb region covered by two BAC clones were identified by various strategies such as DNA sequencing and DNA sequence database searching, computational prediction, exon trapping, cDNA selection and RNA RACE. No known gene from the genome sequence data was present in this 200kb region. Examination of the public database showed 5 unique ESTs within the 200kbp minimum deletion region (Figure 2). RT-PCR analysis demonstrated that three ESTs expressed in normal prostate and LNCaP cells (accession numbers AI814228, aw014383, and aa910261) and the other two EST did not express in normal prostate tissue (aw827194, bg924424). Complete analysis of the EST sequences indicated that all three that expressed EST were intronless and contained no significant open reading frames. Repeated attempts to extend the EST sequences failed. Identification of a Candidate Tumor Suppressor Gene in the Minimal Deletion Region on Chromosome 6ql6.1

Exon trapping using the BAC clone AL356094 yielded two unique exons (exon 5 and 10), which led to the identification of a 435 bp nucleotide by RT-PCR. The full-length transcripts of the gene were obtained by 5' and 3' RACE-PCR and cDNA library screening. The gene was designated as όqTSGl. Detailed cDNA analysis revealed a total of eight different mRNA variants ranging from 4.8 kb to 2.9 kb composed of 15 exons (Figure 3 and Table 4), which share the same 435 bp core sequence. Each form of transcript was confirmed by sequencing the RT-PCR products amplified from 5' to 3' ends. DNA sequence analysis of all the όqTSGl variants did not yield any significant open reading frame except for a 113 amino acids ORF present in the repetitive sequence (exon 1). Large amounts of repetitive sequence are present in all the transcript variants (Table 4). The result indicated that όqTSGl may either represent a novel non-coding RNA or a partial sequence. To evaluate the expression profile of όqTSGl in human tissue, northern blot analysis of όqTSGl in human tissues was performed. The 435 bp core sequence (exons 5 -10 except for exon 7) of όqTSGl cDNA was used to probe human multiple tissue blots including prostate tissues (Clontech). Two main transcripts which were 4.4 kb and 3.4 kb in size were found ubiquitously expressed in almost all tissues (Figure 4). However, multiple bands are detected on northern blot. The όqTSGl expression was relatively abundant in tissues from muscle, testis, brain and adrenal gland. Prostate tissue expressed low levels of όqTSGl on northern blot. όqTSGl is Expressed in Normal Prostate Tissue and CaP Cell Line loss or Decreased Expression of όqTSGl

RT-PCR assays were performed to assess the expression of the two EST representing genes, AA910261, Ai 814228, as well as the όqTSGl within the minimal candidate region in prostate tumor cells. Expression of AA910261, Ai 814228, and the core sequence of όqTSGl were found in normal prostate tissue, however, there was no expression or reduced expression in prostate tumor cell lines (Figure 5). όqTSGl Expression Is Induced in LNCaP Cells by Androgen Deprivation Expression of Ai 814228 and όqTSGl was increased when LNCaP cells were cultured in charcoal-stripped fetal bovine serum (FBS) medium (Figure 6A).

To further demonstrate the negative response of the expression of Ai 814228 and όqTSGl to androgen, the androgen analog R1881 was added into medium after LNCaP cells were previously cultured in charcoal-stripped FBS medium for five days. The expression of Ai 814228 and όqTSGl decreased whereas the expression of DD3 increased (Figure 6B). The similar expression pattern of these two genes indicated that this EST represents a transcript that might be one of the different splicing variants, however, the possibility remained which warrants further investigate. Expression of όqTSGl Inhibits Prostate Cancer Cells Growth The tumor suppressor function of όqTSGl was assessed in prostate tumor cells steadily transfected with constructs expressing όqTSGl. The 6qTSGl variant forms C and H was selected for the initial study. The pcDNA-6qTSG-C or H sense, pcDNA-6qTSG-C or H antisense constructs and pcDNA3.1(+) carrying the neo-resistant gene was transfected into PC3 cells and then maintained the cells in medium containing G418 for 3-4 weeks. The drug resistant colonies were counted by staining with crystal violet. Results from three independent experiments revealed that the colony numbers in cells transfected with pcDNA- 6qTSG-C sense reduced by about 40-50% relative to its antisense and the vector only controls (Figure 7). However, όqTSGl variant H form did not show inhibition of colony formation under the same experimental condition (Figure 7).

Allelic losses of chromosomal 6q have been implicated in diverse types of cancers including breast, ovarian, melanoma, renal and prostate cancers (Theile et al., 1996; Hankins et al, 1996; Millikin et al, 1991; Morita et al, 1991, Srikantan et al 1999). However, most of the studies have stopped short of precisely locating the candidate tumor suppressor genes. A positional cloning approach was taken to systemically map the candidate gene locus and cloned a putative tumor suppressor gene όqTSGl from a long suspected chromosomal hot spot 6ql6! for TSGs. Using LOH analysis in LCM-derived prostate tumor DNA samples, a possible location of a tumor suppressor locus was substantially narrowed down to a 200kbp minimum deletion region between markers D6S1003 and D6S991 on 6ql6.1. A significant association was found between LOH of 6q 16.1 and the increasing age of CaP patients. LOH on this locus was also correlated with high risk of seminal vesicle invasion. In addition, 6ql6.1 deletion was linked to conversion from androgen-dependence to androgen- independence in a LNCaP prostate cancer cell model. The candidate tumor suppressor gene, όqTSGl, was identified within the 200 kb region. The preliminary data showed that όqTSGl inhibited prostate tumor cell growth. Expression of όqTSGl was induced with androgen ablation in LNCaP. These observations led to the hypothesis that the όqTSGl defect on chromosome 6ql6.1 may be involved in the progression of prostate cancer.

A fine map of the minimum deletion region on 6q 16.1 was determined using eight microsatellite markers on 48 pairs of DNA samples derived from laser capture microdissected (LCM) cells obtained from matched tumor and normal specimens of prostate cancer patients. The frequency of LOH at one or more microsatellite markers between markers D6S1056 and D6S300 was 54% which is higher than the data we obtained previously (29%). This is probably because the tumor samples obtained by LCM were purer than the manually microdissected ones. Furthermore, five out of seven patients (71%) with seminal vesicle invasion showed the LOH on 6q 16.1 while only 14 out of 39 (36%) CaP patients without seminal vesicle invasion exhibited LOH. Although it is not statistically significant (p=0.07), there is a trend that LOH on 6ql6.1 was associated with more aggressive tumors. This observation also strengthens findings that the risk for 6q LOH to non-organ-confined disease was 5-fold higher than for organ-confined disease (Srikantan et al, 1999). Recent studies showed a similar high rate of LOH on 6ql4 to 6q22 in primary prostate tumors and xenografts (Hyytinen et al, 2002; Verhagen et al, 2002). The defined minimum deletion regions were in close vicinity (within 10 million base pair sequence) to the minimum deletion region identified in this study. Hyytinen et al also found a more frequent LOH in higher stage tumors, but no correlation of LOH with other clinical pathological parameters of prostate cancer patients. In an experimental LNCaP tumor progression model developed for human prostate cancer, deletion of 6q24-qter was associated with androgen independence and tumorigenicity (Hyytinen et al, 1997; Nupponen et al, 1998). LOH analysis of these cell lines revealed a minimum deletion region that coincided with the primary tumor samples. Although the genetic changes are often complicated in cell cultures, LOH analysis on LNCaP tumor progression models provided further evidence that 6ql6! alterations were associated with prostate tumor progression as well as the conversion from androgen-dependence to androgen-independence. With about 50% deletion of 6ql6 -21 in prostate tumors, these studies, in combination, demonstrated the potential involvement of 6ql6 in prostate tumorigenesis. Also 6q deletion may offer a selective growth advantage for tumor cell progression. Furthermore, the strikingly higher frequency of LOH on 6ql6.1 in elderly patients emphasizes the importance of 6ql6 in sporadic tumor development, given the natural history that prostate cancer is more insidious at onset and takes more than a decade to become clinically significant. Accumulated oxidative stress during aging contributes to the genomic instability

Aided by the physical mapping and sequence data generated by the Human Genome Project, όqTSGl on 6ql6.1 was identified as a candidate for a tumor suppressor gene of prostate cancer. Expression analysis by reverse transcription uncovered a total of eight splicing variants. This variability involves selection, number and combination of exons. All the recognition motifs of exon-intron boundary sequences correspond to the human splice site consensus (Zhang, 1998). However, no ORF of significant length could be observed in όqTSGl transcripts and since all transcripts contain internal exons without a contiguous ORF, this indicated that όqTSGl is likely transcribed into a non-coding RNA. Thus, όqTSGl may represent a novel non-coding RNA although the possibility exists that όqTSGl encodes a short peptide.

Evidence supports that non-coding RNA plays an important role in the regulation of genome imprinting, oxidative stress response, gene expression, cell growth and differentiation (Eddy, 1999; Askew et al., 1999; Kelley et al., 2000; Takeda et al., 1998). Overexpression of non-coding RNA such as DD3 and PCGEM1 genes have been linked to prostate tumors. A novel non-coding RNA of a prostate-specific gene (PCGEM1) showed overexpression in about half of CaP (Srikantan et al, 2000). Recent study further shows biologic function associated with PCGEM1 as exogenous overexpression of the PCGEM1 gene in LNCaP cells or in NIH3T3 cells which resulted in increased cell growth in culture. Clinical utilization of DD3 (de Kok et al, 2002) or PCGEM1 as biomarkers for CaP is promising. On the other hand, The HI 9, codes for a tumor suppressing non-coding mRNAs (Askew et al, 1999). It has been demonstrated that HI 9 has tumor-suppressing potential (Hao et al, 1993). When human HI 9 was introduced into embryonal tumor lines, it caused growth inhibition, morphological changes and abrogation of clonogenicity in soft agar as well as tumorigenicity in nude mice (Hao et al., 1993). Disruption of normal imprinting of the H19 gene may contribute to tumorigenesis (Hibi et al, 1996; Kondo et al, 1995). όqTSGl expression in the PC3 prostate cancer cell lines caused a significant decrease in the number of colonies in the colony-forming assay, indicating that expression of όqTSGl inhibited cell proliferation and/or cell survival function(s). Expressional analysis in prostate cancer cell lines detected an absence expression of όqTSGl in PC3, DU145, and the LNCaP sublines. The expression of όqTSGl was induced in LNCaP when cultured in charcoal- stripped FBS medium. Expression of όqTSGl was not induced in PC3, DU145 and the androgen-independent LNCaP derivatives. όqTSGl is identified herein as a novel transcription unit from a chromosomal hotspot locus, 6ql6.1, associated with CaP development and progression. Results show a striking association of LOH on 6ql6.1 with the aging CaP patients and advanced stage of disease (T3c), as well as an increased risk for tumor metastasis. Androgen-ablation induced expression of όqTSGl in LNCaP cells underscores the potential role of όqTSGl in prostate cell growth and apoptosis.

Other embodiments and uses of the invention will be apparent to those skilled in the art from consideration of the specification and practice of the invention disclosed herein. All references cited herein, including all publications, U.S. and foreign patents and patent applications including the provisional priority document, are specifically and entirely incorporated herein by reference. It is intended that the specification and examples be considered exemplary only with the true scope and spirit of the invention indicated by the following claims.

Claims

Claims

1. An isolated tumor suppressor genetic locus derived from human chromosome 6.

2. The tumor suppressor locus of claim 1 , wherein the isolated locus is a recombinant gene.

3. The tumor suppressor locus of claim 1 , wherein the isolated locus is purified from eukaryotic cells.

4. The tumor suppressor locus of any of claims 1-3, which is localized to about 200 Kbp of human chromosome 6ql6.1.

5. The tumor suppressor locus of any of claims 1-4, which is localized between positions 6ql6.3 and 6q23 of human chromosome 6.

6. The tumor suppressor locus of any of claims 1-5, which comprises a sequence selected from the group consisting of SEQ ID NOS 57-64.

7. The tumor suppressor locus of any of claims 1 -5, which comprises the sequence of SEQ ID NO 57.

8. The tumor suppressor locus of any of claims 1 -5, which comprises the sequence of SEQ ID NO 58.

9. The tumor suppressor locus of any of claims 1-5, which comprises the sequence of SEQ ID NO 59.

10. The tumor suppressor locus of any of claims 1-5, which comprises the sequence of SEQ ID NO 60.

11. The tumor suppressor locus of any of claims 1-5, which comprises the sequence of SEQ ID NO 61.

12. The tumor suppressor locus of any of claims 1-5, which comprises the sequence of SEQ ID NO 62.

13. The tumor suppressor locus of any of claim 1 - 5 , which comprises the sequence of SEQ ID NO 63.

14. The tumor suppressor locus of any of claims 1-5, which comprises the sequence of SEQ ID NO 64.

15. The tumor suppressor locus of any of claims 1-14, which is functionally linked to a recombinant promoter.

16. The tumor suppressor locus of any of claims 1-14, which is functionally linked to a recombinant gene. 17 The tumor suppressor locus of claim 16, wherein the recombinant gene is a toxin. 18 A recombinant cell containing the tumor suppressor locus of any of claims 1-17. 19 The recombinant cell of claim 18, which is a human cell. 20 An expression product of the tumor suppressor locus of any of claims 1-17. 21 The expression product of claim 20 which comprises one or more peptides. 22 The expression product of claim 20 which comprises one or more RNAs. 23, An antibody that specifically binds to the expression product of any of claims 20-22. 24 An antibody that specifically binds to the tumor expression locus of any of claim 1- 17 25 A kit for the detection or determination of disease in a tissue sample obtained from a patient, said kit comprising a sequence that hybridizes to a sequence expressed from the tumor suppressor locus of any of claims 1-14.

26. A kit for the detection or determination of disease in a tissue sample obtained from a patient, said kit comprising an antibody that binds to a sequence expressed from the tumor suppressor locus of any of claims 1-14.

27. A method for treating a cancer comprising administering an agent to increase expression of the tumor suppressor locus of any of claims 1-17.

28. The method of claim 27, wherein the cancer is breast cancer.

29. The method of claim 27, wherein the cancer is prostate cancer.

30. The method of claim 27, wherein the cancer is ovarian cancer.

31. The method of any of claims 27-30, wherein the agent is an androgen, a steroid, or an antibody.

32. The method of any of claims 27-30, wherein the agent is estrogen, progesterone, testosterone, or a derivative thereof.

33. The method of any of claims 27-30, wherein the agent is an androgen antagonist.