- Open Access
Promoter methylation of IGFBP-3 and p53 expression in ovarian endometrioid carcinoma
© Torng et al; licensee BioMed Central Ltd. 2009
Received: 12 August 2009
Accepted: 11 December 2009
Published: 11 December 2009
Insulin-like growth factor binding protein (IGFBP-3) is an antiproliferative, pro-apoptotic and invasion suppressor protein which is transcriptionally regulated by p53. Promoter methylation has been linked to gene silencing and cancer progression. We studied the correlation between IGFBP-3 and p53 expression as well as IGFBP-3 promoter methylation in ovarian endometrioid carcinoma (OEC) by immunohistochemical staining and quantitative methylation-specific PCR (qMSP). Additionally, we assessed the molecular regulatory mechanism of wild type (wt) p53 on IGFBP-3 expression using two subclones of OEC, the OVTW59-P0 (low invasive) and P4 (high invasive) sublines.
In 60 cases of OEC, 40.0% showed lower IGFBP-3 expression which was significantly correlated with higher IGFBP-3 promoter methylation. p53 overexpression was detected in 35.0% of OEC and was unrelated to clinical outcomes and IGFBP-3. By Kaplan-Meier analysis, patients with lower IGFBP-3, higher IGFBP-3 promoter methylation, and normal p53 were associated most significantly with lower survival rates. In OEC cell line, IGFBP-3 expression was correlated with IGFBP-3 promoter methylation. IGFBP-3 expression was restored after treatment with a DNA methy-transferase inhibitors (5-aza-deoxycytidine) and suppressed by a p53 inhibitor (pifithrin-α). The putative p53 regulatory sites on the promoter of IGFBP-3 were identified at -210, -206, -183 and -179 bases upstream of the transcription start site. Directed mutagenesis at these sites quantitatively reduced the transcription activity of IGFBP-3.
Our data suggests that IGFBP-3 silencing through IGFBP-3 promoter methylation in the absence of p53 overexpression is associated with cancer progression. These results support a potential role of IGFBP-3 methylation in the carcinogenesis of OEC.
Epithelial ovarian cancer is the most lethal gynecological cancer among women worldwide. Metastasis occurs in most patients before diagnosis, and the majority of patients have high subsequent recurrent rate after completing surgery and chemotherapy . Survival is at about 3 years for patients in advanced stages . The poor prognosis of ovarian cancer is due to the difficulty in early diagnosis and the detrimental processes of invasion and metastasis. A better understanding of the molecular mechanisms of cancer development and progression will help to improve the diagnosis and treatment of the disease.
Insulin-like growth factor binding proteins (IGFBPs) are circulating transport proteins for IGF, with IGFBP-3 being the predominant IGFBP in circulation . IGFBP-3 can regulate cell growth and death, either dependent or independent of its interaction with IGF [3, 4]. Recently, we have further identified IGFBP-3 as an invasion suppressor gene using an established ovarian cancer cell line OVTW59-P0 and its sublines P1 to P4, which were obtained from a gradual invasion model with sequential increase in invasiveness . Many tumor suppressor genes involved in cancer formation and progression are frequently epigenetically silenced through aberrant hypermethylation at their promoter regions . In ovarian cancer, genes with promoter hypermethylation are frequently found to be related to cancer progression [7, 8]. Aberrant promoter hypermethylation of IGFBP-3 and gene silencing are observed in many cancers, such as lung, hepatocellular, gastric, colorectal, breast, and ovarian cancers [9–13]. However, the association of IGFBP-3 promoter hypermethylation with poor clinical outcome was identified only at early stages in lung and ovarian cancers [9, 10].
p53 is a known transcription factor for IGFBP-3 expression . Induction of IGFBP-3 by p53 has been shown to cause cell apoptosis in an IGF-independent manner . Eleven p53 binding sites have been identified within IGFBP-3 gene based on the homology to the p53 binding consensus sequence, and confirmation by electrophoretic mobility shift analyses . Promoter hypermethylation at these p53 binding sites caused gene silencing and resistance to p53 . Therefore, it has been suggested that p53 could mediate cross-talk to the IGF axis through IGFBP-3 regulation .
Pathologically, epithelial ovarian cancer is classified into four major histological subtypes: serous, mucinous, endometrioid and clear cell carcinoma. Each subtype is associated with distinct molecular alterations . In our previous study, we found 51.4% of the ovarian endometrioid carcinoma (OEC) subtype showing lower IGFBP-3 expression, which is associated significantly with poor patient outcome . From literature review, p53 overexpression is more frequently reported in the serous subtype of ovarian cancer [18–20]. The association of altered p53 expression in tumor tissue with patient survival in ovarian cancer is still under debate [19–21], but a significant correlation is often reported in the serous subtype [18, 19]. Despite the close molecular regulation between p53 and IGFBP-3 being known for more than a decade , little information concerning p53 regulation of IGFBP-3 or the effect of epigenetic inactivation of IGFBP-3 exist with defined clinical endpoints. In the current study, we analyzed the association between p53 and IGFBP-3 expression in ovarian cancer progression. We explored the clinical significance of aberrant promoter hypermethylation at the p53 binding sites of IGFBP-3 in OEC. Functional analysis of p53 regulation on IGFBP-3 expression was further assessed using the OVTW59-P0 and P4 cell lines. Our study indicates that methylation at the p53 transcription factor binding sites in the IGFBP-3 promoter can silence IGFBP-3 expression and hence lead to cancer progression in OEC.
IGFBP-3 expression was correlated with IGFBP-3 promoter methylation, but not p53 expression
Tumor expressions of IGFBP-3, p53 and IGFBP-3 promoter methylation and the clinicopathological features in patients with OEC.
N = 36
N = 24
N = 39
N = 21
N = 20
N = 20
Age (mean ± SD)
51.7 ± 11.9
52.2 ± 12.7
49.3 ± 12.3
55.7 ± 11.7
51.8 ± 12.9
51.5 ± 13.3
The amount of methylation at IGFBP-3 promoter in these patient samples was measured by quantitative real time methylation specific PCR (qMSP). Twenty samples were unable to be bisulfite converted probably due to failed in DNA extraction from very old paraffin blocks. In total, qMSP analysis was performed on 40 patient samples. There were no differences in IGFBP-3 and p53 expressions or in clinicopathological features between the samples which were unable to be analyzed by qMSP and those which were (Additional file 1, Table S1). Among these 40 samples, the calculated promoter methylation of IGFBP-3 ranged from 0.01 to 59.8%. By using receiver operating characteristic (ROC) curve and area under the curve (AUC) analysis methods, the optimal cutoff value was set at a threshold of > 3% as hypermethylation. At this threshold, the sensitivity of IGFBP-3 methylation with low IGFBP-3 expression was 0.86 and the specificity was 0.73 (Additional file 1, Figure S1).
Table 1 shows the correlation among IGFBP-3 and p53 expression and qMSP results with the clinical parameters. High IGFBP-3 expression was found in tumors of lower grades and lower stages, and p53 tended to be overexpressed in tumors with higher grades. No correlation was found between p53 overexpression and IGFBP-3 expression or IGFBP-3 promoter methylation. Cases that showed higher IGFBP-3 methylation (at threshold > 3%) were significantly correlated with lower IGFBP-3 expression and higher histological grades.
Relation among the four quartiles of IGFBP-3 promoter methylation by qMSP analysis with IGFBP-3 expression and clinicopathological features in OEC patients.
Number of patients (%)
0.01 -- 1.07
1.12 -- 2.83
3.10 -- 6.36
6.88 -- 59.79
Survival Analysis for patients with OEC.
IGFBP-3 expression and IGFBP-3 methylation
Methylation ≤ 3%
Methylation > 3%
IGFBP-3 expression and IGFBP-3 methylation with normal p53
Methylation ≤ 3%
Methylation > 3%
Clinical correlation of IGFBP-3 expression, p53 overexpression and IGFBP-3 promoter methylation with survival rates
Cox-regression analysis revealed low IGFBP-3 expression to be associated with lower progression-free survival (PFS) and lower overall survival (OS) rates. p53 overexpression was not associated with patient survivals while high IGFBP-3 promoter methylation was marginally associated with lower OS. When combining these variables for survival analysis, low IGFBP-3 expression, high IGFBP-3 promoter methylation and normal p53 expression were correlated most significantly with lower PFS and OS. However, since IGFBP-3 expression and IGFBP-3 promoter methylation were correlated with disease stages and tumor grades (Table 1), the associations of these variables with survival rates was not independent of age, stage and grade (Table 3).
Correlation of IGFBP-3 promoter methylation and IGFBP-3 expression in ovarian cancer cell lines
Transcriptional regulation of p53 on IGFBP-3 promoter
Both P0 and P4 expressed similar levels of wild type (wt) p53 by direct sequencing after amplifying exons 5, 6, 7, and 8 of p53 gene, and Western blot analysis (data not shown). To test the transcriptional regulation of IGFBP-3 by p53, the -253 ~+61 portion of the promoter and first exon of IGFBP-3 were transfected into P0 and P4. After treatment with a known p53 inhibitor, pifithrin-α, the luciferase activity representing IGFBP-3 expression was completely abolished in these two sublines (Figure 3c). This result indicates that promoter sites between -253 ~+61 are essential for transcriptional regulation by p53 on IGFBP-3 expression
Identification of methylation hotspots at the p53 consensusbinding sequence in IGFBP-3 promoter
Since p53 binding at these sequences has been reported to be diminished by methylation , we established a series of site-directed mutation constructs to test the importance of these four loci on the transcriptional regulation by p53 of IGFBP-3 (Figure 4c). In addition to using P0 and P4, a human embryonic 293T kidney cell line with documented wt p53 was also included in analysis. The transcription activity of IGFBP-3 was significantly reduced after mutagenesis at these loci and the amount of luciferase activity corresponding to IGFBP-3 expression decreased as the number of mutated loci increased (Figure 4d). These findings indicate that these four loci are critical spots for transcriptional regulation by p53 of IGFBP-3. In addition, methylation at these sites could quantitatively silence the process of IGFBP-3 induction by p53.
We demonstrated previously that IGFBP-3 is an invasion suppressor in OEC . In this study, we evaluated the molecular mechanism of IGFBP-3 regulation in cancer cells and found that methylation at the CpG sites of the IGFBP-3 promoter was significantly correlated with IGFBP-3 silencing in OEC tumors. Clinically, patients with lower IGFBP-3 expression and higher IGFBP-3 promoter methylation have significantly lower PFS and OS. This association was strongest in cases with normal p53 expression. In an in vitro study using OEC cell line OVTW59, we identified a quantitative correlation between IGFBP-3 suppression and methylation in the first 200 base pairs surrounding the transcription start site of IGFBP-3. Our data suggest that IGFBP-3 silencing through aberrant hypermethylation at its promoter is a major regulatory pathway for OEC tumorigenesis.
Methylation specific PCR (MSP) has been widely used to study promoter methylation in clinical specimens. By MSP, aberrant methylation at the IGFBP-3 promoter was found in 44% of epithelial ovarian cancer and in 28.6% (12/42) of OEC subtype [7, 12]. However, IGFBP-3 silencing has been reported unrelated to IGFBP-3 promoter methylation [7, 12]. A significant association between IGFBP-3 promoter methylation and disease progression has been reported, but only in early-stage ovarian cancer . It was concluded that methylation at IGFBP-3 promoter is a common tumorigenesis process in the early steps of ovarian cancer progression . By setting the calculated percentage of IGFBP-3 methylation at 3% as a cutoff point for qMSP analysis, we found methylation at the IGFBP-3 promoter significantly correlated with IGFBP-3 silencing in OEC. Furthermore, we found patients with lower IGFBP-3 expression and higher IGFBP-3 promoter methylation to be significantly associated with poor outcomes. A possible reason is that we have used a more sensitive method, qMSP, to study these epigenetic events. In addition, we have focused our study on the OEC subtype of ovarian cancer. It has been observed that promoter methylation of specific genes in cancer occur frequently in a tumor-type and cell-type specific manner . Our results demonstrate that the clinical significance of aberrant IGFBP-3 promoter methylation is more commonly observed in the OEC subtype of ovarian cancer.
Ten cases in our study showed unparallel results between IGFBP-3 expression and IGFBP-3 promoter methylation, i.e. either high IGFBP-3 and IGFBP-3 promoter methylation > 3% or low IGFBP-3 and IGFBP-3 promoter methylation ≤ 3%. Though threshold of > 3% as hypermethylation was selected based on ROC curve analysis that showed 0.86 sensitivity and 0.73 specificity of low IGFBP-3 expression, we found five (50%) of these unparallel cases were at the third quartile of methylation, i.e. high IGFBP-3 expression and promoter methylation levels between 3% and 6.36% (Table 2). This could be due to inadequate sensitivity and specificity of qMSP assay. Alternatively, the unparallel result could be explained by multiple mechanisms of IGFBP-3 regulation. IGFBP-3 could also interact with several other growth-inhibitory agents to mediate wide varieties of growth suppression signal in the absence of IGF [24, 25]. In addition, other CpG methylation sites in the IGFBP-3 promoter that were not studied might also contribute to IGFBP-3 silencing
We further explored the association between p53 and IGFBP-3 promoter methylation and the mechanism of p53 regulation on IGFBP-3. Thirty-five percent of these cases showed p53 overexpression that was correlated with higher tumor grade at marginal significance. However, overexpression of p53 was not correlated with IGFBP-3 expression or IGFBP-3 promoter methylation and it did not correlate with OEC patient survival. This suggested that p53 is not important in the progression of OEC, and that IGFBP-3 silencing through promoter methylation is a mechanism different from p53 overexpression. Since OEC is similar in histological pattern to the endometrioid (EC) subtype of endometrial cancer, we reviewed the molecular characteristics of different subtypes of endometrial cancer. In endometrial cancer, p53 alteration was specifically reported to be present in the non-endometrioid (non-EC) serous subtype, which has been classified as a type 2 endometrial cancer. The EC subtype of endometrial cancer is classified as type 1 endometrial cancer . Accordingly, Kurman et al. proposed an epithelial ovarian cancer model composed of type 1 and type 2 tumors based on: the histological patterns, molecular features, and the process of tumor progression. Type 1 tumors are slow growing, genetically stable, and are characterized by mutations in KRAS, BRAF, PTEN, and beta-catenin. Type 2 tumors are highly aggressive, rapidly growing, genetically instable, and are characterized by mutation of TP53 . Based on this model, the OEC subtype is classified as a type 1 tumor and serous subtype, with characteristic TP53 mutations, is classified as a type 2 tumor . In our study, we confirmed that p53 alteration is not a characteristic of OEC. Instead, low IGFBP-3 and high IGFBP-3 promoter methylation are major prognostic factors for OEC. By survival analysis, patient survival is lowest in cases with low IGFBP-3, high IGFBP-3 promoter methylation, and normal p53. These suggest OEC is a distinct subtype of ovarian cancer that IGFBP-3 silencing through IGFBP-3 promoter methylation could play an important role in cancer development and progression. We found p53 alteration is not important for the tumorigenesis of OEC, but IGFBP-3 silencing through IGFBP-3 promoter methylation might subsequently interrupt the communication network between p53 and the IGF axis, and hence lead to cancer progression of OEC.
Of the eleven p53 binding sites in IGFBP-3 gene that has been identified , Hanafusa et al. observed that at least four sites between -210 to -150 are essential for p53 induced expression of IGFBP-3 in human hepatocarcinoma cell line HepG2. They also found that hypermethylation of these sequences could selectively suppress p53 induced IGFBP-3 expression . Using our OEC cell line, we identified four sites in the IGFBP-3 promoter at the -210, -206, -183, and -179 loci, as methylation hot spots. Functional analyses have been performed to study the influence of methylation at these sites on the p53 regulation of IGFBP-3. Hanafusa et al. showed diminished p53 binding and IGFBP-3 repression associated with hypermethylation of these sequences by luciferase assay after co-transfection of p53 and mutant IGFBP-3 promoter constructs, and by electrophoresis mobility shift assay . In our study, we used pifithrin-α to inhibit p53 and found a subsequent decrease in IGFBP-3 expression in OVTW59-P4 and 293T cell lines, both of which contain wt p53. These functional analyses support the observation that p53 can up-regulate IGFBP-3 expression , and methylation at these sites suppress the p53 activation of IGFBP-3.
Furthermore, our results show a quantitative association between methylation at the IGFBP-3 promoter and IGFBP-3 expression. The amount of methylation at these hot spots were positively related to IGFBP-3 suppressions in the P0 and P4 sublines. Using site-directed mutagenesis, we demonstrated a linear correlation between IGFBP-3 suppression and the amount of methylation at these loci. Our in vitro data support our clinical qMSP finding that IGFBP-3 expression was correlated to IGFBP-3 promoter methylation in OEC tumors.
Our observation that normal p53 is a required factor along with lower IGFBP-3 and higher IGFBP-3 promoter methylation for a significant survival result also suggests that wt p53 is an important molecular characteristic of OEC subtype. Recently, Kawasaki et al. reported an inverse association between IGFBP-3 promoter methylation and microsatellite instability in patients with methylator phenotype colorectal cancers, particularly under the condition of wt p53 . We postulate that the biological environment of IGFBP-3 promoter methylation is strictly regulated, such that wt p53 and un-methylated IGFBP-3 promoter region are necessary factors to maintain a homeostatic condition.
Progression of tumorigenesis involves multiple steps of genetic alteration. From OEC clinical specimens and cell lines, our data indicates that aberrant methylation at p53 consensus sequence binding region of IGFBP-3 promoter could contribute to low IGFBP-3 expression and subsequently to a poor patient survival outcome. Normal p53 plays an important role in this regulation. In conclusion, our studies show clinical evidence on methylation-dependent epigenetic silencing of IGFBP-3 expression regulated by p53. This regulatory pathway represents an important mechanism for loss of IGFBP-3 expression during OEC tumorigenesis and/or progression to metastasis.
General laboratory reagents were purchased from Sigma-Aldrich (St. Louis, MO, USA) and cell culture reagents from Gibco/BRL (Grand Island, NY, USA), unless otherwise specified. 5-aza-dC was obtained from Sigma-Aldrich and pifithrin-α from Calbiochem (Gibbstown, NJ, USA).
Tumor specimen and cell lines
Sixty cases of OEC specimens were collected from the archive of Department of Pathology, National Taiwan University (NTU) Hospital from May, 1994 to September 2006. These patients received surgeries and chemotherapies according to our previous report . Informed consents were obtained for each patient before receiving operations. OEC cell lines OVTW59-P0 and P4 were established in our laboratory previously . Cell lines A549, H1299 and 293T were obtained from Taiwan National Health Research Institute cell bank. All cell lines were maintained in DMEM solution with 5% FCS.
Staining was performed in 4 μm paraffin sections as described previously . In brief, sections were dewaxed, rehydrated, heated by microwave and then blocked with 1% H2O2 and normal horse serum. Antibodies against IGFBP-3 (Sigma-Aldrich) and p53 (clone DO-7; DAKO, Glostrup, Denmark), as the primary antibodies, were incubated, followed by biotinylated secondary antibodies, avidin-biotin-peroxidase complex and peroxidase substrate were then added for microscopic observation. Interpretations of immunostaining were performed independently by two authors (PLT and CTL). Immunostaining of IGFBP-3 was defined as low or high when less or more, respectively, than 25% of tumor cells exhibited strong staining of cytoplasm. Immunostaining of p53 was defined as normal or overexpressed when less or more, respectively, than 25% of tumor cells exhibited strong nuclear staining.
Tumor parts from 10 μm paraffin sections were dissected under a stereomicroscope for DNA isolation. An estimation of greater than 80% purity of tumor cells was obtained in each case. Genomic DNA was extracted using QlAamp DNA kit (Qiagene, Valencia, CA, USA) and subjected to EpiTect Bisulfite kit (Qiagene) according to the manufacture's protocol. In brief, 2 μg of genomic DNA was subjected to a denature/incubation repeat cycle in sodium bisulfite solution.
Quantitative real time methylation specific PCR (qMSP)
Primer sequences for IGFBP-3 analysis.
Annealing °C (Cycle)
Quantitative real time methylation-specific PCR
F: 5'- AGGTGATTCGGGTTTCGGGC -3'
R: 5'- GACCCGAACGCGCCG -3'
R: 5'-CGAC CCGAACGCGCCGACC-3'
Wild type IGFBP3
Mutant (M12) IGFBP3
Mutant (M34) IGFBP3
Methylation-specific PCR (MSP)
Primers for methylated and unmethylaled region of IGFBP-3 promoter, as previously reported , are listed in Table 3. The schematic diagram showing the locations of primers used are shown in Figure 1a. Twenty μL of reaction volume containing one-twentieth of the modified DNA, 4 deoxynucleoside triphosphates, PCR primers (-MSP) and HotStar TaqDNA MasterMix (Qiagen, Valencia, CA) were set for first PCR. The PCR product was then used as DNA template for a second methylated (-M) or unmethylated (-U) IGFBP-3 PCR reaction. Results were resolved by electrophoresis using 2% agarose gels. Cell line A549 treated with Sss I was used as a positive methylated control.
Bisulfite-PCR and sequencing (BSP)
The -372 ~+140 bp from the first exon of IGFBP-3 promoter region was amplified from one-twentieth of the modified DNA by PCR using HotStarTaq DNA Master Mix Kit (Qiagene, Valencia, CA). The primer sequences and PCR conditions are shown in Table 4. The pGEM-T Easy Vector (Promega Corp., Madison, WI, USA) was used for TA cloning and high-efficiency competent cells DH5α (Yeastern, Taipei, Taiwan) was used for transformation. Colonies were selected and rechecked by electrophoresis. The recovered plasmids were sequenced at the Sequencing Core Facility of the NTU College of Medicine. Cytosines in CpG dinucleotides that remained unconverted after bisulfite treatment were considered as methylated.
Twenty-fold concentrated conditioned media, collected from cells cultured in FCS-free DMEM media for 24 hours, were subjected to 8% SDS-PAGE (Millipore, Bedford, MA, USA) and then transferred to membranes (Schleicher & Schuell, Germany) using MilliBlot-SDS semi-dry electroblotting system (Millipore). After blocking, the membrane were probed with antibodies against IGFBP-3 (Research Reagents, Texas, USA). Reactions were amplified by biotinylated second antibody, followed by streptavidin-HRP. Results were exposed using ECL system.
cDNA preparation and quantitative real time reverse-transcriptase PCR (QRT-PCR)
Total RNA was extracted using TRIzol reagent (Invitrogen, Carlsbad, CA, USA) and cDNA was prepared using SuperScript III First-strand Synthesis of the Oligo (dT) primer system (Invitrogen). The cDNA samples were subjected to QRT-PCR as described previously  using 7700 Sequence Detector (Applied Biosystems, Foster City, CA, USA) and SYBR green Master Mix Kit (Alppied Biosystems). Primer sequences for IGFBP-3 were forward: 5' TGTGGCCATGACTGAGGAAA, reverse: 5' TGCCAGACCTTCTTGGGTTT; and GAPDH were forward: 5' TGGTATCGTGGAAGGACTCA, reverse: 5' AGTGGGTGTCGCTGTTGAAG. Each target gene was normalized to GAPDH mRNA expression for comparison.
Construction of wild and mutant expression vectors
The promoter region of IGFBP-3 (from -253 ~+61) and the site-directed mutagenesis constructs at 4 different mutant sites (C to A at -210, -206, -183 and -179) were generated by PCR amplification with primers and PCR conditions as shown in Table 4. Wild type pGL3-IGFBP-3 and the mutant constructs were generated by subcloning of the PCR fragments to pGL3-Basic plasmid (Promega Corp.). The mutant constructs were named as Mut A (mutation at -210 and -206), Mut B (mutation at -183 and -179) and Mut A+B (mutant at all of the four sites), respectively, as shown in Figure 3c. All constructs were confirmed by DNA sequencing.
Transient transfection and luciferase assays
Cells were transiently transfected using Arrest-In transfection reagents (Open Biosystems, Huntsville, USA). Cells at a density of 1 × 105/well were seeded in 6-well plate. Transfection was performed using 2 μg of plasmid DNA in serum free medium for 6 hours. Cells were then incubated in serum containing medium for another 48 hours. After washing twice with PBS, cells were collected for luciferase assays (Promega Corp.) using a dual luciferase reporter assay system according to the manufacture's instructions. The relative luciferase activity was normalized against renilla activity by co-transfection with 1 μg of pRL-TK (Promega Corp.).
The cutoff value of IGFBP-3 methylation as a threshold of hypermethylation was identified using ROC curve and AUC. The expressions of IGFBP-3 and p53, and IGFBP-3 methylation status were compared among patients with different clinical and pathologic features using t test and χ2 test. Associations among IGFBP-3 expression, p53 overexpression and IGFBP-3 promoter methylation with progression-free survival (PFS) and overall survival (OS) rates were assessed by Cox-regression analysis. Kaplan-Meier analysis with logrank test was used to estimate survival probabilities and to compare survival distributions categorized by IGFBP-3 expression, IGFBP-3 promoter methylation and p53 overexpression. Statistical analysis was carried out using Statistical Analysis System (SAS) version 8.0 (SAS Institute, Cary, NC). Probability values less than 0.05 were regarded as significant.
We thank Dr. Chi-Ling Chen from the Graduate Institute of Clinical Medicine for her assistance in statistical analysis and Mr. Jian-Liang Chou from the National Chung Cheng University for his technical assistance. We also thank Mr. Daniel Deatherage from the Ohio State University for his kindly help in English editing.
Grant support: National Science Council (NSC95-2314-B-002-290, 96-2314-B-002-097) and National Health Research Institute (NHRI-EX96-9416B1).
- Ohwada M, Suzuki M, Suzuki T, Hiratsuka M, Kawai T, Saito K, Sato I: Problems with peritoneal cytology in second-look laparotomy performed in patients with epithelial ovarian carcinoma. Cancer. 2001, 93: 376-380. 10.1002/cncr.10137View ArticlePubMedGoogle Scholar
- McGuire WP, Hoskins WJ, Brady MF, Kucera PR, Partridge EE, Look KY, Clarke-Pearson DL, Davidson M: Cyclophosphamide and cisplatin versus paclitaxel and cisplatin: a phase III randomized trial in patients with suboptimal stage III/IV ovarian cancer (from the Gynecologic Oncology Group). Semin Onco. 1996, 23: 40-47.Google Scholar
- Firth SM, Baxter RC: Cellular actions of the insulin-like growth factor binding proteins. Endocr Rev. 2002, 23: 824-854. 10.1210/er.2001-0033View ArticlePubMedGoogle Scholar
- Bhattacharyya N, Pechhold K, Shahjee H, Zappala G, Elbi C, Raaka B, Wiench M, Hong J, Rechler MM: Nonsecreted insulin-like growth factor binding protein-3 (IGFBP-3) can induce apoptosis in human prostate cancer cells by IGF-independent mechanisms without being concentrated in the nucleus. J Biol Chem. 2006, 281: 24588-24601. 10.1074/jbc.M509463200View ArticlePubMedGoogle Scholar
- Torng PL, Lee YC, Huang CY, Ye JH, Lin YS, Chu YW, Huang SC, Cohen P, Wu CW, Lin CT: Insulin-like growth factor binding protein-3 (IGFBP-3) acts as an invasion-metastasis suppressor in ovarian endometrioid carcinoma. Oncogene. 2008, 27: 2137-2147. 10.1038/sj.onc.1210864View ArticlePubMedGoogle Scholar
- Mulero-Navarro S, Esteller M: Epigenetic biomarkers for human cancer: The time is now. Crit Rev Oncol Hematol. 2008, 68: 1-11. 10.1016/j.critrevonc.2008.03.001View ArticlePubMedGoogle Scholar
- Wiley A, Katsaros D, Chen H, Rigault de la Longrais IA, Beeghly A, Puopolo M, Singal R, Zhang Y, Amoako A, Zelterman D, Yu H: Aberrant promoter methylation of multiple genes in malignant ovarian tumors and in ovarian tumors with low malignant potential. Cancer. 2006, 107: 299-308. 10.1002/cncr.21992View ArticlePubMedGoogle Scholar
- Watts GS, Futscher BW, Holtan N, Degeest K, Domann FE, Rose SL: DNA methylation changes in ovarian cancer are cumulative with disease progression and identify tumor stage. BMC Med Genomics. 2008, 1: 47- 10.1186/1755-8794-1-47PubMed CentralView ArticlePubMedGoogle Scholar
- Buckbinder L, Talbott R, Velasco-Miguel S, Takenaka I, Faha B, Seizinger BR: Induction of the growth inhibitor IGF-binding protein 3 by p53. Nature. 1995, 377: 646-649. 10.1038/377646a0View ArticlePubMedGoogle Scholar
- Chang YS, Wang L, Liu D, Mao L, Hong WK, Khuri FR, Lee HY: Correlation between insulin-like growth factor-binding protein-3 promoter methylation and prognosis of patients with stage I non-small cell lung cancer. Clin Cancer Res. 2002, 8: 3669-375.PubMedGoogle Scholar
- Hanafusa T, Yumoto Y, Nouso K, Nakatsukasa H, Onishi T, Fujikawa T, Taniyama M, Nakamura S, Uemura M, Takuma Y, Yumoto E, Higashi T, Tsuji T: Reduced expression of insulin-like growth factor binding protein-3 and its promoter hypermethylation in human hepatocellular carcinoma. Cancer Lett. 2002, 176: 149-158. 10.1016/S0304-3835(01)00736-4View ArticlePubMedGoogle Scholar
- Wiley A, Katsaros D, Fracchioli S, Yu H: Methylation of the insulin-like growth factor binding protein-3 gene and prognosis of epithelial ovarian cancer. Int J Gynecol Cancer. 2006, 16: 210-218. 10.1111/j.1525-1438.2006.00299.xView ArticlePubMedGoogle Scholar
- Tomii K, Tsukuda K, Toyooka S, Dote H, Hanafusa T, Asano H, Naitou M, Doihara H, Kisimoto T, Katayama H, Pass HI, Date H, Shimizu N: Aberrant promoter methylation of insulin-like growth factor binding protein-3 gene in human cancers. Int J Cancer. 2007, 120: 566-573. 10.1002/ijc.22341View ArticlePubMedGoogle Scholar
- Grimberg A, Coleman CM, Burns TF, Himelstein BP, Koch CJ, Cohen P, El-Deiry WS: P53-dependent and p53-independent induction of insulin-like growth factor binding protein-3 by deoxyribonucleic acid damage and hypoxia. J Clin Endocrinol Metab. 2005, 90: 3568-3574. 10.1210/jc.2004-1213PubMed CentralView ArticlePubMedGoogle Scholar
- El-Deiry WS, Kern SE, Pietenpol JA, Kinzler KW, Vogelstein B: Definition of a consensus binding site for p53. Nat Genet. 1992, 1: 45-49. 10.1038/ng0492-45View ArticlePubMedGoogle Scholar
- Hanafusa T, Shinji T, Shiraha H, Nouso K, Iwasaki Y, Yumoto E, Ono T, Koide N: Functional promoter upstream p53 regulatory sequence of IGFBP3 that is silenced by tumor specific methylation. BMC Cancer. 2005, 5: 9- 10.1186/1471-2407-5-9PubMed CentralView ArticlePubMedGoogle Scholar
- Bell DA: Origins and molecular pathology of ovarian cancer. Mol Pathol. 2005, 18 (Suppl 2): S19-32. 10.1038/modpathol.3800306.View ArticleGoogle Scholar
- Eltabbakh GH, Belinson JL, Kennedy AW, Biscotti CV, Casey G, Tubbs RR, Blumenson LE: p53 overexpression is not an independent prognostic factor for patients with primary ovarian epithelial cancer. Cancer. 1997, 80: 892-898. 10.1002/(SICI)1097-0142(19970901)80:5<892::AID-CNCR10>3.0.CO;2-RView ArticlePubMedGoogle Scholar
- Høgdall EV, Christensen L, Høgdall CK, Frederiksen K, Gayther S, Blaakaer J, Jacobs IJ, Kjaer SK: Distribution of p53 expression in tissue from 774 Danish ovarian tumour patients and its prognostic significance in ovarian carcinomas. APMIS. 2008, 116: 400-409. 10.1111/j.1600-0463.2008.00917.xView ArticlePubMedGoogle Scholar
- Kurman RJ, Shih I-M: Pathogenesis of ovarian cancer: lesions from morphology and molecular biology and their clinical implications. Int J Gynecol Pathol. 2008, 27: 151-160.PubMed CentralPubMedGoogle Scholar
- Darcy KM, Brady WE, McBroom JW, Bell JG, Young RC, McGuire WP, Linnoila RI, Hendricks D, Bonome T, Farley JH, : Associations between p53 overexpression and multiple measures of clinical outcome in high-risk, early stage or suboptimally-resected, advanced stage epithelial ovarian cancers A Gynecologic Oncology Group study. Gynecol Oncol. 2008, 111: 487-495. 10.1016/j.ygyno.2008.08.020PubMed CentralView ArticlePubMedGoogle Scholar
- Bourdon JC, Deguin-Chambon V, Lelong JC, Dessen P, May P, Debuire B, May E: Further characterisation of the p53 responsive element--identification of new candidate genes for trans-activation by p53. Oncogene. 1997, 14: 85-94. 10.1038/sj.onc.1200804View ArticlePubMedGoogle Scholar
- Costello JF, Frühwald MC, Smiraglia DJ, Rush LJ, Robertson GP, Gao X, Wight FA, Feramisco JD, Peltomaki P, Lang JC, Schuller DE, Yu L, Bloomfield CD, Caligiuri MA, Yates A, Nishikawa R, Su Huang H, Petrelli NJ, Zang X, O'Dorisio MS, Held WA, Cavenee WK, Plass C: Aberrant CpG-island methylation has non-random and tumour-type-specific patterns. Nat Genet. 2000, 24: 132-138. 10.1038/72785View ArticlePubMedGoogle Scholar
- Butt AJ, Firth SM, King MA, Baxter RC: Insulin-like growth factor-binding protein-3 modulates expression of Bax and Bcl-2 and potentiates p53-independent radiation-induced apoptosis in human breast cancer cells. J Biol Chem. 2000, 275: 39174-39181. 10.1074/jbc.M908888199View ArticlePubMedGoogle Scholar
- Fowler CA, Perks CM, Newcomb PV, Savage PB, Farndon JR, Holly JM: Insulin-like growth factor binding protein-3 (IGFBP-3) potentiates paclitaxel-induced apoptosis in human breast cancer cells. Int J Cancer. 2000, 88: 448-453. 10.1002/1097-0215(20001101)88:3<448::AID-IJC18>3.0.CO;2-VView ArticlePubMedGoogle Scholar
- Llobet D, Pallarés J, Yeramian A, Santacana M, Eritja N, Velasco A, Dolcet X, Matias-Guiu X: Molecular Pathology of Endometrial Carcinoma: Practical aspects from the diagnostic and therapeutical view points. J Clin Pathol. 2009, 62: 777-785. 10.1136/jcp.2008.056101View ArticlePubMedGoogle Scholar
- Grimberg A: P53 and IGFBP-3: apoptosis and cancer protection. Mol Genet Metab. 2000, 70: 85-98. 10.1006/mgme.2000.3008View ArticlePubMedGoogle Scholar
- Kawasaki T, Nosho K, Ohnishi M, Suemoto Y, Kirkner GJ, Fuchs CS, Ogino S: IGFBP3 promoter methylation in colorectal cancer: relationship with microsatellite instability, CpG island methylator phenotype, and p53. Neoplasia. 2007, 9: 1091-1098. 10.1593/neo.07760PubMed CentralView ArticlePubMedGoogle Scholar
- Torng PL, Mao TL, Chan WY, Huang SC, Lin CT: Prognostic significance of stromal metalloproteinase-2 in ovarian adenocarcinoma and its relation to carcinoma progression. Gynecol Oncol. 2004, 92: 559-567. 10.1016/j.ygyno.2003.11.011View ArticlePubMedGoogle Scholar
- Chan MW, Wei SH, Wen P, Wang Z, Matei DE, Liu JC, Liyanarachchi S, Brown R, Nephew KP, Yan PS, Huang TH: Hypermethylation of 18S and 28S Ribosomal DNA predicts progression-free survival in patients with ovarian cancer. Clin Cancer Res. 2005, 11: 7376-7383. 10.1158/1078-0432.CCR-05-1100View ArticlePubMedGoogle Scholar
- Chang YS, Wang L, Suh YA, Mao L, Karpen SJ, Khuri FR, Hong WK, Lee HY: Mechanisms underlying lack of insulin-like growth factor-binding protein-3 expression in non-small-cell lung cancer. Oncogene. 2004, 23: 6569-6580. 10.1038/sj.onc.1207882View ArticlePubMedGoogle Scholar
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.