ssDNA-Binding Protein 2 Is Frequently Hypermethylated and Suppresses Cell Growth in Human Prostate Cancer

In 2007, it is estimated that more than 27,000 deaths will be due to prostate cancer and more than 200,000 new cases will be diagnosed. Prostate cancer is the most common malignant neoplasm and the second cause of cancer death among men in the United States (1). The current screening method for prostate cancer relies predominantly on the serum level of the prostate-specific antigen. However, its poor specificity limits this approach; about one third of patients with an elevated prostate-specific antigen undergo further unnecessary medical procedures as they do not have a malignant form of the disease (2). Therefore, it is urgent to discover novel molecular markers for prostate cancer to augment current detection approaches.

Unlike genetic alterations, gene promoter methylation, one type of epigenetic change, is commonly found in prostate cancer (3). More than 30 genes, including GSTP1, APC, RASSF1, and RARB, have been reported to be hypermethylated in prostate cancer (3–7). Promoter methylation contributes to tumorigenesis through silencing of tumor suppressor genes by down-regulating gene expression (8). However, many of the specific molecular events involved in prostate cancer progression still remain elusive.

Sequence-specific ssDNA-binding protein, the chicken orthologue of human SSBP, shares 97% identity with both mouse and human expressed sequence tags (9). Human ssDNA-binding protein 2 (SSBP2) was identified as a target of unbalanced translocations between 5q and 17p in acute myeloid leukemia (10, 11). A recent study on SSBP2 has shown that it is a novel regulator of hematopoietic growth and differentiation (12). However, the role of SSBP2 in human solid tumors remains unexplored. In this article, we report that SSBP2 is hypermethylated in human prostate cancer and that this epigenetic alteration is closely related to SSBP2 protein expression. Forced SSBP2 expression was able to inhibit cell growth through cell cycle arrest in prostate cancer cells.

Cell lines and primary tissues. Four prostate cancer cell lines, DU145, 22Rv1, LNCaP, and PC3, were included in the study. All cell lines were obtained from the American Type Culture Collection. Cells were maintained in RPMI 1640 supplemented with 10% fetal bovine serum.

Eighty-eight primary prostate adenocarcinomas with no metastatic lesions detected were obtained from patients who were consecutively diagnosed and treated primarily with radical prostatectomy at the Portuguese Oncology Institute-Porto, Porto, Portugal. Thirty paired high-grade prostatic intraepithelial neoplasia (PIN) lesions identified in 30 of these specimens were also collected. Twenty-three randomly selected patients with benign prostatic hyperplasia (BPH) who underwent transurethral resection of the prostate were used as controls. The patient characteristics, such as Gleason score and grade, are shown in Table 1. All tissue specimens were obtained immediately after surgical resection and stored at −80°C for further analysis. Then, 5-μm-thick sections were cut from frozen tissue fragments in cryostat to identify areas of adenocarcinoma and high-grade PIN in the radical prostatectomy specimens, and BPH in the control tissue samples was obtained from transurethral resections of the prostate after H&E staining. The areas of interest were identified and the frozen tissue block was trimmed to allow the specific collection of the selected areas [containing at least 70% neoplastic cells (either PIN or adenocarcinoma)]. An average of fifty 12-μm-thick sections was then cut in the cryostat and processed for DNA extraction (4).

5-Aza-2′-deoxycytidine treatment of prostate cancer cells. Cells were seeded in T-75 flasks at a density of 1 × 10⁶ per flask at 37°C with 5% CO₂. The cells were treated with 5 μmol/L 5-aza-2′-deoxycytidine (5-Aza-dC) every 24 h for 4 d. The cells were harvested for DNA and RNA preparation.

RNA extraction and reverse transcription-PCR. RNA extraction and reverse transcription-PCR were done as described previously (13). Four micrograms of total RNA were reverse transcribed with SuperScript II reverse transcriptase (Invitrogen) and amplified by PCR. The primer sequences of SSBP2 were 5′-GCCTGAGCGACAAAAAGTTC-3′ (forward) and 5′-CATCTCCCCATCTCCAAAGA-3′ (reverse). The PCR conditions were 1 min at 95°C, 1 min at 58°C, and 1 min at 72°C for 30 cycles. β-Actin was used as a control, and its primer and amplification were described previously (13).

Bisulfite treatment of DNA. Genomic DNA was extracted using the phenol chloroform method, and bisulfite modification of genomic DNA was done as described (13). Briefly, 2 μg of genomic DNA were used for chemical treatment. Then, the sample was purified using the Wizard DNA Clean Up System (Promega) followed by incubation with sodium hydroxide and ammonium acetate separately. The sample was precipitated in 100% ethanol. The treated DNA was resuspended in 100 μL LoTE [10 μmol/L Tris-HCl (pH 8), 2.5 μmol/L EDTA (pH 8)] and stored at −80°C for further analysis.

PCR amplification of bisulfite-treated DNA and sequencing analysis. Bisulfite-modified genomic DNA was amplified using primer sets that were designed to recognize DNA alterations after bisulfite treatment. The primer sequences of SSBP2 for bisulfite-treated DNA were 5′-TGGGTGGATTAGGAGGAATG-(649-668)-3′ and 5′-TGTGAAGGGCTGACTCACTG-(785-804)-3′. A touchdown PCR program was used to amplify the promoter region of SSBP2. All PCR products were gel extracted (Qiagen) and sequenced using the ABI BigDye cycle sequencing kit (Applied Biosystems).

Real-time quantitative methylation-specific PCR. Bisulfite-modified genomic DNA was amplified using Taqman methylation-specific PCR (MSP) primer sets that were designed based on the sequencing results of the SSBP2 promoter in tumor and BPH. The Taqman MSP primer sequences of SSBP2 were 5′-ATTTTTGCGGTCGTAGCGGT-(694-714)-3′ (forward) and 5′-TTCTACGACAAATCTAACGAA-(763-784)-3′ (reverse). The fluorescent probe was designed to hybridize to a region of SSBP2 promoter that was determined previously to be methylated in prostate cancer cell lines by sequencing. The probe sequence of SSBP2 is 6FAM-5′-ATATCCAAAACGCCGCGAAACTCC-3′-TAMRA. The annealing temperature is 60°C. β-Actin was used as a control (13). For all MSP reactions, 3 μL of bisulfite-treated DNA were added to a final volume of 20 μL. All reactions were done in duplicate. Serial dilutions of in vitro methylated human leukocyte genomic DNA were used to construct a calibration curve. The methylation ratio was defined as the Taqman value derived from SSBP2 promoter amplification divided by Taqman value from β-actin amplification and multiplied by 100 (Taqman methylation value: TaqMeth V).

Construction of recombinant mammalian expression vector of SSBP2. A human cDNA clone of SSBP2 was purchased from OriGene. The full length of SSBP2 was amplified using primer sets that were designed to cover the whole open reading frame. The primers used were 5′-TCGGCACAAGCATGTACGGCAA-3′ (forward) and 5′-TGGATCACACGCTCATTGTCA-3′ (reverse). The PCR amplifications were done as follows: 1 min at 95°C, 1 min at 54°C, and 1 min at 72°C for 40 cycles followed by a 5-min denaturation at 95°C. A 7-min elongation step at 72°C completed the PCR amplification. The PCR products were gel extracted and cloned into a PCR2.1-TOPO TA cloning vector (Invitrogen) through TA ligation. Then, the fragment of SSBP2 was subcloned into a pcDNA3.1 vector in the sense direction using restriction enzymes HindIII and XholI whose sites flanked the fragment. The sequence and direction of the insert was confirmed during each step by ABI BigDye cycle sequencing.

Transfection and colony formation assay. Three prostate cancer cell lines (DU145, 22Rv1, and PC3) were transfected with SSBP2 expression plasmids using Fugene6 reagent following the manufacturer's protocol (Roche). After 24 h, the medium was changed to complete RPMI 1640 (Invitrogen). Forty-eight hours after transfection, cells were selected using G418 (1.2 mg/mL). Colonies were photographed 2 wk after selection.

A stable cell line of PC3 overexpressing SSBP2 was generated and maintained in complete RPMI 1640 containing G418 (1.2 mg/mL). The overexpression of SSBP2 was then confirmed by Western blot analysis.

Cell cycle analysis. A stable cell line overexpressing SSBP2 was maintained in complete medium containing 1.2 mg/mL G418. For flow cytometry analysis, after the cells were synchronized for 48 h in medium containing 1% serum, the cells were then cultured in complete medium. After 24 h, the cells were harvested and washed in PBS. Then, they were fixed in 70% ethanol overnight at 4°C. Subsequently, the cells were stained with propidium iodide (20 mg/mL) solution for 30 min at room temperature. DNA content was measured by flow cytometry (BD FACSCalibur), and data were analyzed using CellQuest Pro Software (BD Technology).

Microarray analysis. Ten micrograms of total RNA from separate stable PC3 cell line overexpressing SSBP2 or pcDNA3.1 were used for microarray analysis on the U133 plus gene chip (Affymetrix). The cDNA preparation, RNA labeling, hybridization, and data analysis were done per the Affymetrix protocol. Only genes with a 1.5-fold increase or decrease in expression were considered for further analysis. The genes associated with cell cycle pathways and apoptosis, such as p15, p16, p27, and Fas, were presented.

Western blotting. For Western blotting, whole-cell lysates were prepared in radioimmunoprecipitation assay buffer containing a protease inhibitor cocktail (Sigma). Protein concentrations were measured by MicroBCA kit (Pierce). For each sample, 30 μg of whole-cell lysate were loaded onto a 10% precast SDS-polyacrylamide gel (Bio-Rad). After gel electrophoresis and membrane transfer, the membrane was probed with rabbit anti-SSBP2 polyclonal antibody (Geneway), mouse anti-c-Myc monoclonal antibody (BD PharMingen), and anti-actin monoclonal antibody. Secondary antibodies used were goat anti-rabbit or anti-mouse IgG conjugated to horseradish peroxidase. Protein bands were displayed by enhanced chemiluminescence.

Tissue microarrays and immunohistochemical analysis. Prostate adenocarcinoma tissue microarrays were purchased from U.S. Biomax, Inc. The arrays had 40 prostate mucinous adenocarcinomas (pT1-pT4) along with 8 nonmalignant normals and tumor adjacent normals. Immunohistochemical analysis was done as described previously (13). Briefly, the tissue arrays were deparaffinized, rehydrated, and microwaved. Then, the tissue array was incubated with rabbit anti-SSBP2 polyclonal antibody (1:100; Geneway) for 16 h at 4°C followed by biotinylated secondary antibody and peroxidase-labeled streptavidin complex detection (Dako). The slides were then analyzed by microscope and quantified based on the intensity of the staining with Mayer's hematoxylin. Parallel sections with nonimmune rabbit IgG of the same isotype were examined as negative control. The slides were examined under microscope and the pictures were taken by Nikon ACT-1 software.

Statistical analysis. Statistical analyses were done using Statistical Analysis System software. The significance level used P ≤ 0.05.

SSBP2 is identified as a candidate methylation gene in prostate cancer. Our laboratory has previously described a whole-genome approach to identify methylated genes using probabilistic search algorithms in combination with an established pharmacologic unmasking strategy (14). We found that the SSBP2 promoter was methylated in DU145, 22Rv1, and PC3 cell lines but not in LNCaP by direct sequencing of bisulfite-treated DNA (Fig. 1A). Therefore, DU145, 22Rv1, and PC3 were used in subsequent experiments. To examine whether the methylation of SSBP2 is tumor specific, we tested tissues from five primary human prostate cancers and six BPH. SSBP2 promoter methylation was observed in four of five (80%) primary prostate cancers compared with zero of six (0%) BPH (Fig. 1A). The representative sequencing data of bisulfite-treated DNA from primary prostate cancer and BPH were shown in Fig. 1B. The cytosines of the SSBP2 promoter in DNA from BPH were converted to thymines after bisulfite treatment, whereas the cytosines in DNA from prostate cancer were protected due to methylation (Fig. 1B).

SSBP2 is hypermethylated in prostate cancer but not in BPH. To determine the quantity of SSBP2 methylation, we designed MSP primers and a Taqman probe conjugated with specific fluorescence based on the sequencing data of bisulfite-treated DNA (Materials and Methods). The primers were located in a region 694 to 784 bp upstream of the transcription start site in the SSBP2 promoter (Fig. 2A). Both the primers and probe sequences detected CpG methylated sites in cancer but not in BPH.

To confirm that SSBP2 methylation is actually cancer cell specific, we did quantitative MSP in 88 samples from prostate cancer lesions with tumor grades from pT2a to pT4 as well as 23 benign prostatic lesions from BPH patients that were used as normal control. We also analyzed SSBP2 promoter methylation levels in 30 paired PIN samples that were microdissected from the same set of 88 prostate cancer patients. The Taqman methylation values in primary prostate cancer were significantly higher than in BPH and PIN (Fig. 2B). The frequency of hypermethylation in primary prostate cancer was 61.4% (54 of 88 cases; cutoff value of 120) compared with 0% (0 of 23 cases) in BPH. PIN showed an intermediate hypermethylation frequency (30%, 9 of 30 cases), consistent with its premalignant characteristics. The average methylation level of the SSBP2 promoter was much higher in prostate cancer than in BPH or PIN (Fig. 2C). The above data suggest that hypermethylation of SSBP2 is cancer cell specific in prostate cancer.

SSBP2 hypermethylation is associated with advanced prostate cancer. Based on our quantification data, of the 88 cases of prostate cancer, there were 54 cases with a higher SSBP2 methylation level and 34 cases with a lower SSBP2 methylation level than the cutoff value of 120. Statistical analysis showed that SSBP2 hypermethylation status was closely related to prostate tumor staging (P = 0.016; Table 1). All patients with tumor staging higher than pT3b (100%, eight of eight) were detected as hypermethylated (Table 1). No significance was found between SSBP2 methylation and other clinicopathologic factors, such as age, serum prostate-specific antigen level, or Gleason score.

SSBP2 expression is down-regulated in primary prostate cancer cells and can be reactivated by 5-Aza-dC. To examine the expression pattern of SSBP2 in primary prostate tissues, we did immunohistochemistry on a prostate adenocarcinoma tissue array with an anti-SSBP2 polyclonal antibody. The immunohistochemical results of primary prostate tissue showed that SSBP2 was predominantly expressed in normal prostate epithelial cells (Fig. 3A). All eight normal prostate epithelium examined showed intense SSBP2 expression, whereas SSBP2 was expressed at much lower levels in epithelial cells from prostate cancer. Of 40 prostate cancer samples, only 5 (12.5%) showed the same expression level as normal prostate, whereas the remaining 35 (87.5%) showed absent or lower expression (Table 2). Interestingly, all six tumors examined with staging pT3 and pT4 showed absent or decreased expression compared with normal prostate (Table 2). This expression pattern was consistent with hypermethylation in advanced prostate cancers. The reverse transcription-PCR data also showed reduced mRNA levels of SSBP2 in prostate cancer cell lines (Fig. 3B). We treated the three SSBP2 methylated prostate cancer cell lines (DU145, 22Rv1, and PC3) with the demethylating agent 5-Aza-dC and found that SSBP2 could be robustly reactivated (Fig. 3B). The above results suggest that SSBP2 is down-regulated through promoter hypermethylation.

SSBP2 inhibits prostate cancer cell growth through suppression of cell cycle. To examine whether SSBP2 has the ability to suppress cell proliferation in prostate cancer cell lines, we did the colony formation assay separately in all three prostate cancer cell lines (DU145, 22Rv1, and PC3) in which endogenous SSBP2 was methylated and its expression was silenced. The results showed that the colony formation ability was significantly suppressed by ectopic expression of SSBP2 in all of the above prostate cancer cell lines (Fig. 4A). We then established a stable cell line overexpressing SSBP2 in PC3 cells. We did cell cycle analysis in PC3 cells stably expressing SSBP2 and found that the cells with overexpressing SSBP2 were arrested in the G₁ phase. The percentage of cells in S and G₂-M phase was decreased from 18.69% to 11.55% and 23.31% to 11.62% separately by SSBP2 (Fig. 4B). Overexpression of SSBP2 in stable PC3 cell lines was confirmed by Western blotting (Fig. 4C).

To explore the molecular events involved in cell cycle regulation by SSBP2, we did microarray analysis with total RNA from SSBP2-transfected PC3 stable cell line using PC3 cells transfected with pcDNA3.1 alone as a control. From the microarray data, we found that genes that inhibit cell cycle progression (such as p15, p16, p27, GDF15, and CCNG2) were up-regulated at the transcriptional level in SSBP2-overexpressed PC3 cells. On the other hand, genes that promote cell cycle (such as CDC25A, CDC20, MCM4, and MCM5) were down-regulated (Table 3). Previous work in leukemia cells suggested decreased c-Myc protein as a target of cell cycle arrest induced by SSBP2 (12). We also found that the c-Myc protein level was reduced in overexpressed PC3 cells by Western blot (Fig. 4C), whereas microarray data showed minimal changes at the mRNA level. The above data suggest that SSBP2 acts as a tumor suppressor gene in prostate cancer cells and that one mechanism of suppression is due to its inhibition of normal cell cycle progression.

Here, we show that SSBP2 may be a potential DNA marker in prostate cancer. We confirmed SSBP2 promoter methylation in 88 primary prostate cancers by quantitative MSP analysis. Of 88 cases, 54 (61.4%) were hypermethylated in the SSBP2 promoter region, whereas it was not found to be hypermethylated in BPH (0 of 23), confounding condition that can mimic prostate cancer (Fig. 2B). Additionally, we found that the methylation levels of PIN were between those of tumor and BPH, consistent with its pathologic characteristics as a preinvasive form of prostate cancer. This finding is different from other reported methylation genes, such as GSTP1, APC, RASSF1, and MGMT, which were found to have similar methylation levels in PIN and prostate cancer (4, 7). Interestingly, statistical analysis revealed that SSBP2 hypermethylation was correlated with tumor stage in prostate cancer. Our quantitative and clinical data suggest that SSBP2 promoter methylation is a specific and frequent event in prostate cancer.

SSBP2 was found to be hypermethylated in three prostate cancer cell lines (DU145, 22Rv1, and PC3) and reactivated after treatment with the demethylating agent 5-Aza-dC, indicating that SSBP2 was physiologically inactivated through epigenetic silencing in prostate cancer cell lines. We further examined SSBP2 expression pattern in an independent set of primary human prostate tissue from cancer patients and nonneoplastic tissues using immunohistochemistry. Our results showed that SSBP2 was significantly down-regulated in most of the primary prostate cancers (35 of 40) compared with normal prostate tissues. This down-regulation was again more obvious in cancers with higher tumor stages.

SSBP2 belongs to a gene family that shares high identity in its open reading frames (9). SSBP2 localizes to chromosome 5q13.1, which was found to be translocated and deleted in myelodysplasia and acute myelogenous leukemia (10). A recent study showed that SSBP2 might be a candidate tumor suppressor gene in hematopoietic malignancies. It is a regulator of cell growth and differentiation in myeloid leukemia cells (12). Our functional analysis showed that SSBP2 contained the ability to suppress cell proliferation in prostate cancer cells, showing its tumor-suppressive activity in solid tumor as well as in hematopoietic malignancy.

Further experiments revealed that SSBP2 induced cell cycle arrest in PC3 cells. The cell cycle arrest induced by SSBP2 seemed to be through inhibition of progression from the G₁ to S and G₂-M phase in prostate cancer cells. In contrast, overexpression of SSBP2 in leukemia cells (U937) induced a predominantly S-phase change. The difference between cell cycle arrest effects by SSBP2 on prostate and hematopoietic cells indicates that the effects of SSBP2 might be cell type dependent. Our microarray data revealed that several genes involved in cell cycle progression were dysregulated, confirming the cell cycle arrest caused by SSBP2 overexpression in PC3 cells. We also observed changes in expression of genes involved in apoptosis, Wnt signaling, and cell adhesion. Generally, genes associated with proapoptosis and cell adhesion inhibition were up-regulated in SSBP2-transfected cells, whereas genes related to antiapoptosis, cell adhesion promotion, and Wnt signaling were down-regulated (Table 3). The reduction of c-Myc protein level in SSBP2-overexpressed PC3 cells as well as in hematopoietic cells indicates that SSBP2 may share a common molecular target in both cell types. The cross-talk between SSBP2 and the above gene products needs to be further explored.

We thus report that the SSBP2 promoter is hypermethylated in primary human prostate cancers compared with nonneoplastic prostate lesions (BPH). To our knowledge, this is the first report that SSBP2 expression is down-regulated in primary prostate cancer and cancer cell lines and is tightly correlated with its promoter hypermethylation. Moreover, we found that SSBP2 plays a role in tumor suppression through inhibition of cell cycle progression, revealing clues to its role in tumorigenesis. Our findings identify SSBP2 as a novel biomarker in prostate cancer and further support its role through inactivation in human prostate tumorigenesis. SSBP2 hypermethylation and its expression need to be tested in a larger numbers of patients of prostate cancer to see if it predicts poor outcome in a future study.

Under a licensing agreement between OncoMethylome Sciences, S.A. and the Johns Hopkins University, Dr. Sidransky is entitled to a share of royalty received by the University on sales of products described in this article. Dr. Sidransky owns OncoMethylome Sciences, S.A. stock, which is subject to certain restrictions under University policy. Dr. Sidransky is a paid consultant to OncoMethylome Sciences, S.A. and is a paid member of the company's Scientific Advisory Board. The term of this arrangement is being managed by the Johns Hopkins University in accordance with its conflict of interest policies.