Cloning and Characterization of UROC28, a Novel Gene Overexpressed in Prostate, Breast, and Bladder Cancers¹

Prostate cancer is the most common malignancy among men in the United States, affecting over 179,300 men and resulting in about 37,000 deaths in 1999 (1). A small percentage (about 10–15%) of newly diagnosed cancers are actually metastatic at the time of diagnosis (2, 3). Approximately 30% of men who are treated for localized disease will recur and a subset of these men will progress to androgen-independent metastatic prostate cancer(1, 2, 3). The mechanism(s) for disease progression and development of the androgen-independent state remains poorly understood. It is not clear why some patients with prostate cancer progress so quickly and others do not. It is possible that multiple genetic and/or epigenetic factors contribute to the biological heterogeneity of prostate cancer and the variability in the rate of progression and disease-specific mortality (4, 5). Identification of genetic and epigenetic factors that may play important roles in prostate cancer progression and metastasis is of great significance to prostate cancer management.

Like many other cancers, the development of prostate cancer is a multistage process involving initiation, progression, invasion, and metastasis (6, 7). Studies have demonstrated that transformation of a normal cell to a fully malignant cell requires a series of genetic changes including mutations of DNA and changes of gene expression at the RNA and protein levels (8, 9). Recently, several laboratories including ours have been actively involved in identifying genes associated with prostate cancer progression and metastasis. These efforts have resulted in the discovery of several genes involved in different biochemical pathways related to the pathogenesis of prostate cancer. Examples of genes identified include but are not limited to HER2/neu(10), prostate-specific transglutaminase(pTGase; Refs. 11, 12), PSMA(13), caveolin (14), PTEN (15), PSCA (16), POV1 (17), NKX3.1 (18),and ETS-2 (19).

In search of potential new gene markers for prostate cancer, we have applied a modified, agarose gel-based differential display method(20, 21) to isolate genes differentially expressed among normal prostate, prostate cancer, and metastatic prostate cancer tissues. We report here the cloning and characterization of a novel gene, UROC28, that is overexpressed in prostate, breast, and bladder cancer. The full cDNA sequence, chromosomal localization of the gene, the development of a specific polyclonal antibody, and detection of the UROC28 protein in serum are also described. The data indicate a correlation between overexpression and the pathogenic determinants of prostate cancer, which may support its eventual application in diagnosis and treatment of prostate cancer.

Frozen tissues used in the experiments were obtained from the CHTN³ (Birmingham,AL), the Virginia Prostate Center Tissue Bank at Eastern Virginia Medical School, (Norfolk, VA), and the Department of Urology,University of Washington, (Seattle, WA). Pathological reports were provided by the organizations for tissue specimens. The specimens were quick frozen in liquid nitrogen immediately after surgery and stored at −70°C until processed for RNA isolation. Total RNAs were isolated from the specimens as described previously (22). Total RNAs (10 μg) from each tissue were treated with 5 units of RNase-free DNase I (Life Technologies) in the presence of 20 mm Tris-HCl (pH 8.4), 50 mm KCl, 2 mm MgCl₂ and 20 units of RNase inhibitor (Boehringer Mannheim). After extraction with phenol/chloroform and ethanol precipitation, the RNA was redissolved in diethylpyrocarbonate-treated H₂O.

A modified, agarose gel-based differential display (20, 21) was used to identify genes differentially expressed in prostate cancer. RNA (10 μg) from each tissue was treated with RNase-free DNase I as described above. Five μg from each of the RNA samples was reverse transcribed into cDNA using random hexamers and Moloney-murine leukemia virus (M-MLV) reverse transcriptase(Life Technologies) following manufacturer’s instructions. The reaction mixture contained 50 mm Tris-HCl (pH 8.3), 75 mm KCl, 3 mm MgCl₂, 10 mm DTT, 500 μm dNTP, 2 μmrandom hexamers and 400 units of M-MLV reverse transcriptase. PCR was performed with one arbitrary 10mer. The primer used for identifying UROC28 was 5′-TGGAGGTTGT-3′. PCR conditions were as follows: 1× PCR buffer [50 μm dNTPs, 0.2 μm arbitrary primer(s), 1/20 volume (1 μl) of the cDNA, 1 unit of Taq DNA polymerase (Life Technologies)] in a final 20-μl mixture. The amplification parameters included 40 cycles of reaction with 30 s denaturing at 94°C, 1 min 30 s annealing at 38°C, and 1 min extension at 72°C. A final extension at 72°C was performed for 15 min. The PCR products were then separated on a 2% agarose gel with 0.5μg/ml ethidium bromide, and positive bands were identified, excised,purified by Qiaex resin (Qiagen), and cloned into plasmid by TA cloning (Promega). The differential expression of positive bands was confirmed by relative quantitative RT-PCR (10, 12). A gene, designated UROC28, was found up-regulated in prostate cancer.

A human prostate cDNA library constructed in λgt10 vector was purchased from Clontech and used for full-length cDNA cloning of UROC28. The method of Benton and Davis (23) was followed for cDNA library screening. A 0.6-Kb UROC28 cDNA fragment was labeled with ³²P using High Prime system(Boeringer Mannheim). Two to three rounds of rescreening were carried out to obtain a pure positive clone. Both strands of the isolated cDNA clones were sequenced by the dideoxynucleotide-chain termination method(24) using a primer walking strategy.

The procedure for FISH chromosomal localization of UROC28 was performed according to Heng et al. (25). Briefly,lymphocytes were cultured in a MEM supplemented with 10% FCS and phytohemagglutinin at 37°C for 68–72 h. The lymphocyte cultures were then treated with bromodeoxyuridine (0.18 mg/ml; Sigma) to synchronize the cell population. The cells were then washed with serum-free medium to release the block and recultured at 37°C for 6 h in MEM with thymidine (2.5 μg/ml; Sigma). Cells were harvested and slides were made by using standard procedures. Slides were baked at 55°C for 1 h. After RNase treatment, the slides were denatured in 70% formamide in 2x SSC for 2 min at 70°C followed by ethanol dehydration. The 0.6-Kb UROC28 cDNA probe was biotinylated with dATP for 1 h at 15°C using BioNick labeling kit (Life Technologies). The labeled probe was denatured at 75°C for 5 min in a hybridization solution containing 50% formamide, 10% dextran sulfate,and human cot I DNA. The denatured probe was loaded onto the slides and subjected to overnight hybridization. Slides were then washed,detected, and amplified. FISH signals and DAPI banding pattern were recorded separately by taking photographs, and the assignment of the FISH mapping data with chromosomal bands was achieved by superimposing FISH signals with DAPI-banded chromosomes.

Five μg of the DNA-free total RNA was reverse transcribed into cDNA using random hexamers and M-MLV reverse transcriptase (Life Technologies) following manufacturer’s instructions. The reaction mixture contained 50 mm Tris-HCl (pH 8.3), 75 mm KCl, 3 mm MgCl₂, 10 mm DTT, 500 μm dNTP, 2 μmrandom hexamers, and 400 units M-MLV reverse transcriptase. The reaction was incubated at 22°C for 10 min, then at 37°C for 50 min. The synthesized cDNA was used for PCR. The primers used for PCR and their sequences are as follows: UROC28-P1, 5′-GCT TCA GGG TGG TCC AAT TAG AGT T-3′; and UROC28-P2: 5′ TCC AAC AAC GAC ACA TTC AGG AGT T 3′.

The primers amplified a 446-bp PCR product from the human UROC28 cDNA. The relative abundance of UROC28 in the tissues was studied by a relative quantitative RT-PCR using β-actin as a control. The PCR mixture contained 2 μl of cDNA, 10 mm Tris-HCl (pH 9.3),50 mm KCl, 3 mm MgCl₂,200 μm dNTP, 1.25 units of Taq DNA polymerase (Life Technologies), and 200 nm of sense and antisense primers in a total of 50-μl reaction. The amplification was performed in a thermal cycler (MJ Research), which included 1-min denaturing at 94°C, 1-min annealing at 56°C, and 1-min extension at 72°C. The PCR was run for 35 cycles for UROC28 and 22 cycles for β-actin. The PCR products were run on a 1.2% agarose gel with ethidium bromide. The UROC28 bands were quantitated by an IS-1000 image analyses system(Alpha Innotech) and normalized with that of β-actin control. All of the normalized values are presented as the mean ± SD.

The filter for Northern hybridization was purchased from ClonTech and was hybridized to the 446-bp [³²P]-labeled UROC28 cDNA probe. The ClonTech Multiple Tissue Northern blots contained 2 μg of oligo(dT)-purified mRNA from different specific normal human tissues. Hybridization, washing, and X-ray film exposure were performed as described previously (26). After stripping, the same filter was hybridized to the β-actin probe.

LnCaP cell line was obtained from American Type Culture Collection(Rockville, MD). The cells were cultured in RPMI 1640 with 10%charcoal-stripped serum for 48 h first, then incubated with RPMI 1640 with 10% charcoal-stripped serum in the presence of 0, 0.1, 1,10, or 100 nm DHT for 24 h, respectively. RNA was isolated and subjected to RT-PCR analyses as described above.

UROC28-specific antisense nucleotide probe (5′-TCT TAA CTC GGG GCA TTT GGT CTT C-3′) and the corresponding sense probe were synthesized and labeled with biotin at the 3′ ends. Sense probe was used as the negative control. Hybridization was performed on formalin-fixed paraffin sections using a MicroProbe System (Fisher Scientific). Paraffin-embedded and formalin-fixed tissues and their corresponding pathological diagnoses were obtained from the CHTN and Department of Pathology, Johns Hopkins Medical Institutions (Baltimore, MD). All of the reagents and diluents used were obtained from Research Genetics(Huntsville, AL). Briefly, paraffin sections in 5-μm thickness were deparaffinized using Auto Dewax solution and dehydrated with alcohol. Sections were then treated with Auto Blocker to block any endogenous peroxidase activity. A pretreatment with pepsin solution for 3 min at 105°C was followed by probe (100-ng/ml) incubation. Hybridization was carried out at 105°C for 5 min and then at 45°C for one h. Sections were then washed with PostHyb Wash solution for 5 min at 45°C before incubation with strepavidin-HRP for 5 min at 50°C and followed by two changes of DAB substrate incubation for 5 min (50°C) each. Sections were counterstained with hematoxylin for 1 min before dehydration and mounting. Polydeoxythymidylic acid hybridization control was performed in parallel to confer the general mRNA integrity in the paraffin sections.

Anti-UROC28 rabbit polyclonal antibody was produced using a synthetic peptide as the immunogen, which corresponds to the predicted amino acid 54 to 74 of UROC28 (Research Genetics, Huntsville, AL). The antibody was peptide-affinity purified, and immunohistochemistry was performed on formalin-fixed paraffin sections using a MicroProbe System (Fisher Scientific). After dewaxing and dehydration, sections were microwaved with citrate buffer (pH 6.0) for two time for 5 min each. The sections were washed with deionized water and PBS (pH 7.4), then incubated with 0.5% Triton X-100 and 0.5% milk in PBS for 5 min at room temperature. The sections were blocked with 5% milk in PBS containing 0.1% Triton X-100 for 20 min, then incubated with the rabbit polyclonal antibody diluted 1:1000 with PBS containing 0.5%milk and 0.1% Triton X-100 at 4°C overnight in a humidified chamber. After washing three times with PBS/0.1% Triton X-100, the sections were sequentially incubated for 20 min each with SuperSensitive biotinylated MultiLink secondary antibody, streptavidin-alkaline phosphatase (Biogenex), and freshly prepared Vector Red chromogen substrate (Vector Laboratories). The sections were counterstained with hematoxylin for one min followed by dehydration and mounting. Rabbit preimmune serum was used as the negative control. Sequential tissue sections used for UROC28 in situhybridization were used for immunohistochemistry.

Serum specimens from 18 normal individuals, 15 biopsy-confirmed patients with NEM, and 14 CaP patients with the clinical cancer stage ranging from T_1a to T₄ and Gleason scores ranging from 3 to 7 (average,6), were studied. The normal sera were residual sera obtained from healthy male blood donors from the Oklahoma Blood Institute (OBI) of Oklahoma City, OK, and donors’ confidentiality is strictly held under guidelines at OBI. The patients’ sera (NEM and PCa) were residual samples from Institutional Review Board-approved cancer biomarker studies previously conducted at UroCor with collaborators from Johns Hopkins Medical Institutions and University of Michigan Cancer Center. The clinical diagnoses of these patients were provided by our urologist collaborators. All of the patients’ personal identifications were kept confidential and remain unknown to UroCor staff. Each serum specimen was assayed in duplicate, and the CV was recorded. Twenty μl of serum from each normal or patient test sample was diluted to 100μl with Tris-buffered solution (TBS) and blotted in duplicate onto nitrocellulose filter using a slot blot apparatus from Bio-Rad Laboratories (Hercules, CA). After blocking with 5% nonfat milk, the filter was incubated with the polyclonal UROC28 antibody described in“Immunohistochemistry” in this section in 1:500 dilution overnight at 4°C. The filter was then washed and incubated with 1:1000 alkaline phosphate-conjugated goat antirabbit immunoglobulins (DAKO,Carpinteria, CA) for 1 h., washed, and then incubated with 5-bromo-chloro-3-indolylphosphate petoluidine salt/nitroblue tetrazolium chloride chromogen/substrate (Kirkegaard & Perry Laboratories, Gaithersburg, MD). The alkaline phosphate signal intensity of the UROC28 protein-antibody immune complexes bands were quantitated by an IS-1000 image analyses system (Alpha Innotech). The positive signal intensity was quantitated using relative absorbance units based on optical density. The average of duplicate test samples was used for analysis. The serially diluted synthetic UROC28 peptide was used as the positive control standard for each slot blot immunoassay. Any patient’s sera with an absorbance value outside the linear range of detection from the peptide standard curve were diluted appropriately and reassayed when necessary. Statistical analyses were performed using the Stata v5.0 statistical software program (STATA Corp., College Station, TX).

A modified agarose gel-based mRNA differential display method(21) was used to identify genes differentially expressed in prostate cancer tissue. UROC28 was identified as one of several genes overexpressed in prostate cancer by comparing display band patterns between normal prostate and prostate cancer. The original identified UROC28 cDNA fragment was determined to be ∼0.6 kb. After cloning into pGEM-T plasmid vector, the fragment was then fully sequenced. A GenBank search indicated that the UROC28 fragment did not match any known genes in the database. The human tissue specificity and the mRNA transcript size of UROC28 were evaluated using Northern blot analysis. Northern hybridization of the UROC28 fragment to mRNAs from eight different organs showed a major 2-kb band in colon, prostate,small intestine, testes, and spleen, the expression was minimal in thymus, ovary, and peripheral blood leukocytes. An additional band at 2.4 kb was seen in prostate, and also in spleen but with less intensity, which indicated the possibility of two alternative splicing variants for this gene (Fig. 1).

The full-length cDNAs of the two alternative splicing mRNA variants were cloned by cDNA library screening and sequencing. As shown in Fig. 2, the two mRNA variants are identical for 1.96 kb of their 5′ sequences, including the 5′-untranslated region, the complete open reading frame, and part of the 3′ untranslated region. They differ only at the end of the 3′untranslated region starting at 1960 bp. Both mRNAs have polyadenylic acid tails and the predicted polyadenylation sites. The two mRNAs have the same open reading frame, encoding a protein of 135 amino acids (Fig. 2). A strong Kozak consensus sequence is found preceding the ATG initiation codon. Bioinformatics analyses indicate that there is a possible transmembrane domain from aa34 to aa50, three PKC phosphorylation sites at aa62 (SQK), aa89 (TMK), and aa94 (SMK), and one myristylation site at aa118(GLECCL). In vitrotranslation experiments using rabbit reticulocyte lysate with both cDNA variants generated a single M_r17,000 protein product, which is the predicted size from the open reading frame (data not shown).

The differential expression of UROC28 gene in prostate cancer was first confirmed using primers unique to each variant in a relative quantitative RT-PCR experiment. RT-PCR was performed on five to six frozen tissues each from normal prostate, BPH, and prostate cancer at different Gleason scores as well as from metastatic prostate cancer specimens. As shown in Fig. 3, the expression of both UROC28 variants is low in all of the normal and BPH tissues and is up-regulated in all of the prostate cancer and metastatic cancer tissues tested in 35 PCR cycles.

The differential expression of both UROC28 transcript variants in prostate cancer was further investigated in an expanded panel of frozen tissues. RNA was isolated from 8 normal prostate tissues, 20 BPH tissues, 28 prostate cancer tissues with different Gleason scores, and 7 metastatic prostate cancer tissues. RT-PCR was performed using primers unique to both variants, and the expression levels were normalized with that of β-actin. As shown in Fig. 4, both variants were similarly up-regulated in prostate cancer and metastatic prostate cancer tissues. A low level of expression for both variants was observed in normal tissues and NMPH tissues. Both UROC28 transcript variants were up-regulated 3- to 4-fold in prostate cancer with varying Gleason scores, and 4- to 6-fold in metastatic prostate cancer tissues.

In situ hybridization with biotin-labeled UROC28-specific oligonucleotide probe with streptavidin-HRP and DAB substrate demonstrated that UROC28 mRNA was preferentially expressed in prostate and breast glandular or ductal epithelial cells (Fig. 5, E–H). The reporter HRP enzyme reacted with DAB and yielded a brown precipitate reflecting the level of hybridized signals. Paraffin sections from prostate cancer with different Gleason scores ranging from 6 to 9 were analyzed for the cellular localization and relative expression level of UROC28 mRNA. Elevated, yet heterogeneous, UROC28 mRNA hybridization signal was observed in prostate cancer glandular epithelium. Fig. 5,Fshowed the elevated UROC28 mRNA signal in Gleason score 8 prostate cancer glandular epithelia as compared with the adjacent or distant prostate acini without evidence of malignancy (Fig. 5,E). UROC28 mRNA was expressed primarily in the basal cells of the benign prostate acini (Fig. 5,E). Polydeoxythymidylic acid hybridization was performed on these prostate sections as a positive control for the integrity of the mRNA in the tissue sections. In situ hybridization analysis also showed the up-regulation of UROC28 mRNA in intraductal breast cancer cells (Fig. 5,H) as compared with breast tissue without pathological evidence of malignancy(Fig. 5 G). These observations are in concordance with the PCR results described below.

Rabbit polyclonal anti-UC 28 antibody immunostaining also demonstrated the expression of UROC28 protein in prostate glandular epithelial cells. This polyclonal antibody was raised against the putative cytoplasmic region of the UROC28 protein. Positive immunostaining is indicated (Fig. 5) by the red precipitate from alkaline phosphatase reporter enzyme and the substrate. Serial tissue sections used in the in situ localization of UROC28 mRNA were analyzed for the expression of UROC28 protein. Similar to the in situ hybridization findings, increased level of UROC28 protein was observed in prostate cancer glandular epithelial cells(Fig. 5,B) as compared with the prostate glandular epithelium without pathological evidence of malignancy (Fig. 5,A). Elevated UROC28 protein was also noted in ductal epithelial cells in breast cancer sections (Fig. 5,D) as compared with the breast counterpart without pathological evidence of malignancy (Fig. 5,C). UROC28 protein was localized primarily in the cytoplasm of prostate and breast cancer glandular epithelial cells (Fig. 5,A–D). However, distinct nuclear localization of UROC28 protein was also noted in prostate cancer glandular epithelia (Fig. 5 B).

A Western slot blot protocol was used to investigate whether UROC28 protein is present in sera of normal and prostate cancer individuals. As shown in Fig. 6, the mean serum UROC28 protein level in individuals with prostate cancer is significantly higher than both normal and NEM individuals at 95%confidence interval (P < 0.001). Student’s t test analysis demonstrated that the mean serum UROC28 protein level between BPH and prostate cancer was significantly different (P = 0.0003). Also, the mean URCO28 levels between normal versus NEM and normal versus prostate cancer were both significantly different with a P < 0.0001. These Western slot blot assays performed reproducibly demonstrating an interassay CV of 11% and an intraassay CV of 8%.

To investigate whether UROC28 gene is also differentially expressed in cancer tissues other than prostate, RNAs from six each of both frozen normal and cancer tissues of breast,colon, lung, and bladder origins were obtained from the CHTN, and RNA was prepared as described above. Relative quantitative RT-PCR was performed to evaluate expression of the smaller UROC28 transcript,which was shown to be expressed in several tissues by Northern hybridization (Fig. 1). As shown in Fig. 7, the expression of UROC28gene was up-regulated more than 4-fold in breast cancer, and 2.5-fold in bladder cancer, when compared with corresponding normal tissues. However, the gene did not show significant differential expression in lung and colon cancers.

Because UROC28 is up-regulated in both prostate and breast cancers, and hormones similarly regulate both prostate and mammary glands, we tested whether the expression of this gene is regulated by androgen. LnCaP cells were initially culture in RPMI 1640 supplied with 10%charcoal-stripped serum for 48 h; then different amounts of DHT(0, 0.1, 1, 10, and 100 nm) were added to the medium and incubated for 24 h. RNAs were then isolated and subjected to RT-PCR analyses. As shown in Fig. 8,expression of UROC28 mRNA is stimulated about 2- to 4- fold by DHT, and the stimulation is DHT dose dependent.

FISH chromosomal mapping was performed to determine the chromosomal localization of UROC28 gene. As shown in Fig. 9, based on the match of DAPI banding and UROC28 FISH hybridization signal, UROC28 was mapped to the long arm of chromosome 6. The detailed position was further determined based on the summary from 10 photos (Fig. 9,C), which mapped the gene to chromosome 6, region q23-q24. Both FISH mapping (Fig. 9) and human genomic DNA Southern hybridization (data not shown) indicated that a single copy gene encodes UROC28.

We report the cloning of a novel gene that is overexpressed in prostate cancer and two other cancers (breast and bladder). The gene transcribes two mRNA variants that share most of their sequence including the whole open reading frame. They differ only at the end of the 3′ untranslated region. Whereas the small mRNA is expressed in most of the tissue types tested, the larger mRNA variant is confined mainly to prostate, with some expression in spleen (Fig. 1). In prostate, the expression level of the smaller transcript is about twice that of the larger transcript (Fig. 1). The fact that expression of the larger mRNA transcript is much higher in prostate when compared with other tissues implies a possible delicate control mechanism of the gene in this tissue. Because both transcripts are similarly up-regulated in prostate cancer (Fig. 4), they may be equally important in prostate carcinogenesis. The exact roles of the different 3′ untranslated regions of the two transcripts remains to be determined.

The single copy of UROC28 gene is mapped to chromosome 6q23–24 region. Genes in this chromosomal region that are associated with prostate cancer have been reported previously(27, 28, 29). Hyytinen et al. (27)showed that loss of 6q24-qter was associated with androgen independence and tumorigenicity. Srikantan et al.found that loss of 6q23–24 might be associated with some prostate cancers (28). Cooney et al. indicated that the proximal 6q deletions are related to prostate cancer progression (29). Furthermore, the 6q23–24region has also been implicated in other cancers. For example,amplification of c-myb in 6q24 was shown to correlate with pancreatic tumor progression (30); loss of heterozygosity in this region was associated with progression of breast and cervical cancers (31, 32, 33). Using RT-PCR, in situ hybridization, and immunohistochemistry, we found that the expression of both UROC28 mRNA and protein is also overexpressed in breast and bladder cancers. The fact that UROC28 is overexpressed in multiple cancer types and that the gene is localized to chromosome 6q23–24 implies that the gene may represent another oncogene candidate in this large (megabases) chromosomal region. However, loss of heterozygosity studies are often used to locate tumor suppressor genes; thus, finding UROC28, an oncogene candidate, in a region of chromosomal loss is somewhat surprising.

The similar up-regulation of the gene in both prostate and breast cancers deserves special attention, because of some common characteristics of breast and prostate cancers. It is well known that the growth and proliferation of both breast and prostate cancer cells are modulated by androgens via common androgen receptor(AR; Refs. 34, 35, 36). Recently, several genes have been reported to be regulated similarly in these two cancers,including AR (34), BRCA1(37), E-cadherin (38), PSA (39), FGF-1 (40), EGFR (41, 42), HER2/neu (10, 41, 42), and Kai 1 (43, 44). Furthermore, expression of some of these genes is regulated by the same mechanism, such as hypermethylation (38), in both prostate and breast cancers. The up-regulation of UROC28mRNA and protein levels in both prostate and breast cancers and the stimulation of the gene by DHT further support the notion that similar pathways may be involved in modulating the growth and progression of these two cancers.

The correct diagnosis and prognosis of prostate cancer is critical in definitive and curative management of this disease. It is agreed that patients diagnosed early with organ-confined tumors are curable ∼90–95% of the time with radical prostatectomy(45) or about 85–95% with radiation therapy(46). Current clinical diagnostic dilemmas created for prostate cancer detection surround the changing natural history of the disease produced by PSA screening (47, 48). There is a significant amount (∼60–70%) of clinical stage T_1c disease (PSA >2.5 ng/ml and nonpalpable disease) presenting at diagnosis that has variable pathology present in the prostate organ (47, 48). The latter provides a new diagnostic and prognostic pretreatment challenge at the time of diagnosis in terms of providing a more precise determination of the extent of the patient’s disease(45). The present widely used PSA assay cannot reliably distinguish between prostate cancer and BPH, nor predict which prostate cancer will progress rapidly. We have shown in this communication that the serum UROC28 protein level is significantly different between normal and BPH, and between BPH and prostate cancer individuals. These preliminary results suggest that UROC28 may provide an alternative serum marker either alone or in combination with other markers such as PSA for a more accurate diagnosis of prostate cancer. The wide range of UROC28 protein detected in the NEM sera might be attributable to the possibility that some of these NEM cases might have contained occult cancer, because an average of 25% of prostate cancer may be missed at the first biopsy (49). More studies are under way to further explore the potential clinical utility of UROC28 as a new serum marker for prostate cancer.

In conclusion, we have demonstrated that the expression of UROC28 gene is significantly up-regulated in primary as well as metastatic prostate cancer tissues, with higher expression of the gene observed in cancer tissues of higher Gleason scores and metastatic tissues. We also demonstrated that UROC28 protein could be detected in serum, and a higher serum UROC28 protein level was detected in prostate cancer individuals compared with normal individuals. Results from in situ hybridization and antibody immunostaining confirm that the gene is up-regulated at the mRNA and protein levels in the glandular epithelial cells of prostate cancer. Basal cells of the prostate acini have been referred to as progenitor cells for prostate glandular tissues. The loss of the basal cell layer and the overexpression of UROC28 mRNA in cancer glandular epithelium may imply a regulatory role for UROC28 and basal cells in prostate carcinogenesis. The observation of nuclear localization of UROC28 protein in prostate cancer glandular epithelia may imply unique tissue-specific regulatory mechanism of UROC28. Our findings support the possibility that UROC28 gene may play a role in prostate cancer progression, and that the increased expression of UROC28 mRNA and protein may serve as potential new markers for better management of prostate cancer.

We would like to thank Sheryl Christofferson and Lei Gong for providing technical support.