ANTICANCER RESEARCH 26: 851-856 (2006)

Hypermethylation in the Promoter Region in the Carcinogenesis of Endometrial in Japanese Patients

MEGUMI YANOKURA*, KOUJI BANNO*, NOBUYUKI SUSUMU, MAKIKO KAWAGUCHI, YOSHIKO KUWABARA, KASTUMI TSUKAZAKI and DAISUKE AOKI

Department of Obstetrics and Gynecology, Keio University School of Medicine, Tokyo, Japan

Abstract. Background: p16 has been reported to disappear by the inactivation of tumor suppressor by methylation hypermethylation in various . However, the relationship of the CpG island in the promoter region ( silencing) between the frequency of hypermethylation of p16 and the is well known. The methylation of the CpG island is mechanism of its inactivation has not been completely observed in most tumor suppressor genes, except , and elucidated in . Materials and Methods: gene silencing has been reported in most cancers (2). Hypermethylation in the promoter region of the p16 gene and Moreover, the DNA hypermethylation of some genes can the expression of the p16 in 51 specimens, including 8 be used as a molecular marker to predict carcinogenetic endometrial cancer cell lines, 7 normal endometrial tissues, 12 risk, since it is observed even in precancerous lesions. The atypical endometrial hyperplasia tissues and 32 endometrial application of DNA hypermethylation as an index in drug cancer tissues were analyzed. Results: Five out of 8 endometrial selection may be possible, since a correlation between cancer cell lines showed hypermethylation with high frequency, hypermethylation of checkpoint genes or although only 1 showed loss of . However, no -related genes and sensitivity to anticancer drugs endometrial cancer tissue of the 32 specimens showed has also been reported (3). hypermethylation. Furthermore, loss of expression was The hypermethylation of the p16 gene was reported to occur immunohistochemically observed in 3 out of the 20 specimens. in 13.5% and 33% (4, 5) of colon cancers, 25% (6, 7) of gastric Conclusion: These results suggest that hypermethylation of p16 cancers and 19% and 18% (8, 9) of breast cancers, which are rarely occurs and, thus, has no significant effect on the rather high values. However, there have been only a few carcinogenesis of endometrial cancer in Japanese patients. reports on the hypermethylation of the p16 gene in endometrial cancer; thus, the frequency is not clear. The The p16 gene, which is located on 9p21, has frequency of p16 gene hypermethylation in endometrial cancer been reported to be hypermethylated in tumors of various according to Wong F et al. was 10/49 (20%) (10), whereas organs. The p16 protein is a tumor suppressor that inhibits the Nakashima R et al. reported a frequency of 0/26 (0%) (11). activity of a CDK4-cyclinD complex and stops the cell cycle at In this study, the relationship between the frequency of the G1-phase by specifically binding to CDK4, therefore hypermethylation in the promoter domain of the p16 gene suggesting a relationship between abnormalities of the p16 and the inactivation of the p16 gene in endometrial cancer gene with carcinogenesis in various organs and prognosis (1). in Japanese patients is examined. Multistage carcinogenesis is assumed to occur by the accumulation of genetic and epigenetic abnormalities, and Materials and Methods

Cultured cells. In this study, 8 human endometrial cancer cell lines were used: HEC-108 (provided by Prof. Hiroyuki Kuramoto), *Both authors contributed equally to this work. SNG-II and SNG-M (provided by Prof. Shiro Nozawa), Ishikawa (provided by Dr. Masaki Nishida), HOOUA, HHUA, HEC-1B and Correspondence to: Dr. Kouji Banno, Department of Obstetrics and KLE. DMEM/F12 (1:1) (11330-132 GibcoBRL, Rockville, MD, Gynecology, Keio University School of Medicine, Shinanomachi USA) 10% fetal bovine serum (FBS) (CC 3008-504-10 SANKO 35, Shinjyuku-ku, Tokyo 160-8582, Japan. Tel: +81-3-3353-1211, JUNYAKU, Tokyo, Japan) was used for KLE, and F12 (N6658 Fax: +81-3-3226-1667, e-mail: [email protected] Sigma, St Louis, MO, USA) 10% FBS for the other cell lines. The origins of the 8 cell lines are shown in Table I. The cells were Key Words: p16, hypermethylation, endometrial cancer. cultured in a 5% CO2 incubator at 37ÆC in a 10-cm culture dish.

0250-7005/2006 $2.00+.40 851 ANTICANCER RESEARCH 26: 851-856 (2006)

Table I. Human endometrial cancer cell lines used in the experiment.

Cell line Tissue type Region of original tumor

SNG-II Well-differentiated adenocarcinoma Primary region Ishikawa Well-differentiated adenocarcinoma Primary region HHUA Well-differentiated adenocarcinoma Primary region HEC-1B Moderately-differentiated adenocarcinoma Primary region SNG-M Moderately-differentiated adenocarcinoma Lymph node metastasis HEC-108 Poorly-differentiated adenocarcinoma Primary region HOOUA Poorly-differentiated adenocarcinoma Primary region KLE Poorly-differentiated adenocarcinoma Primary region

Table II. Primer sequences used for MSP and RT-PCR.

Name Sense Antisense Size Anneal. temp. p16 methylated TTATTAGAGGGTGGGGCGGATCGC GACCCCGAACCGCGACCGTAA 150 bp 67ÆC p16 unmethylated TTATTAGAGGGTGGGGTGGATTGT CAACCCCAAACCACAACCATAA 151 bp 66ÆC p16 RT-PCR CCCGCTTTCGTAGTTTTCAT TTATTTGAGCTTTGGTTCTG 355 bp 50ÆC ‚-actin RT-PCR TTATTTGAGCTTTGGTTCTG CTCCTTAATGTCACGCACGATTTC 303 bp 50ÆC

Tissue specimens. There were 51 specimens, including the following: Table III. The results of endometrial cancer tissue specimen p16 7 normal endometrial tissue specimens, 12 atypical endometrial immunostaining. hyperplasia tissue specimens and 32 endometrial cancer tissue specimens, collected for the purposes of treatment and examination No. Tissue type Stage Differentiation Staining at Keio University Hospital, Japan, from 2003 to 2004. Informed consent was obtained and DNA was extracted from the specimens. 1EMIbG3+ Of the 7 normal endometrial tissue specimens, 6 were in the 2 AS IIIc G3 + secretory phase, while 1 was in the proliferative phase. Of the 32 3 EM IIIc G2 + endometrial cancer tissue specimens, 29 were endometrioid 4 EM IIb G1 + adenocarcinoma (G1:17, G2:6, G3:6) and 3 were adenosquamous 5 AS IIIa G2 + carcinoma. 6 EM IIa G3 + 7 EM IIIc G3 + 8 EM IIb G1 + DNA extraction and methylation-specific PCR (MSP). The DNA 9 EM IIIc G3 + extraction from cultured cell lines and clinical specimens was 10 EM IIa G1 + carried out with the Get Pure DNA Kit (GK03 DOJINDO, 11 EM IIa G2 – Kumamoto, Japan). The hypermethylation in the promoter domain 12 EM IIIb G1 + of the p16 gene was analyzed with MSP after sodium bisulfite 13 EM Ic G2 – treatment. The bisulfite treatment and MSP primer sequence were 14 EM Ia G1 + carried out in accordance with the method of Herman et al. (12). 15 EM IIIc G2 – 16 EM Ia G1 + RNA extraction and RT-PCR. Total RNA was extracted from the 8 17 EM Ia G1 + endometrial cancer cell lines using an RNeasy Mini Kit (74104 18 EM Ib G1 + Qiagen, Tokyo, Japan). The cDNA synthesis was carried out with 19 AS Ib G1 + 1 mg of total RNA using Super Script II Reverse Transcriptase 20 EM Ib G2 + (18064-022 Invitrogen, Carlsbad, CA, USA). The expression of the p16 gene was confirmed after first strand cDNA synthesis. The EM: endometrioid adenocarcinoma conditions and primer sequences for PCR are shown in Table II. AS:

Demethylation. One ÌM 5-aza-dC (A3656 Sigma) was added to the human endometrial cancer cell line medium, in which hypermethylation in the promoter region of the p16 gene was p16 protein expression analysis. The human endometrial cancer cell observed. On Day 2 after addition, the same concentration was lines, cultured on the chamber slide, were fixed with 95% cold added to the medium in the same manner. DNA and RNA were ethanol for 5 min. After washing with PBS, the cells were blocked extracted 4 days later. by incubation with 1% BSA in PBS for 30 min at room temperature.

852 Yanokura et al: DNA Hypermethylation of p16 in Endometrial Cancer

Figure 1. Analysis of p16 gene methylation in human endometrial cancer cell lines. Hypermethylation was observed in the promoter domain of the p16 gene in 5 out of 8 analyzed cell lines. M, marker; PC, positive control; NC, negative control.

Figure 2. RT-PCR p16 gene expression analysis in human endometrial cancer cell lines. Loss of expression was observed only in SNG-M in 5 cell lines, wherein the p16 gene was hypermethylated. M, marker; NC, negative control,

Figure 3. Fluorescent staining with an anti-p16 antibody in human endometrial cancer cell lines. The red staining is positive for nuclear staining by TO-PRO3 and the green staining is positive for anti-p16 antibody staining. Stainability by an anti-p16 antibody disappeared and loss of protein expression was observed in SNG-M.

853 ANTICANCER RESEARCH 26: 851-856 (2006)

Figure 4. Changes in p16 gene expression by addition of 5-aza-dC. Demethylation of the p16 gene was observed, but recovery of expression was not found by 5-aza-dC. M, marker; PC, positive control; NC, negative control

Figure 5. p16 gene methylation analysis in endometrial cancer cell lines. Hypermethylation of the promoter region of p16 in endometrial cancer cell specimens. M, marker; PC, positive control; NC, negative control; NE, nomal endometrium; AE, atypical endometrial hyperplasia; EC, endometrial cancer.

After washing with PBS, the cells were incubated overnight at 4ÆC deparaffinization, the specimens were antibody-inactivated in with an anti-p16 antibody (sc-1661 Santa Cruz, Delaware, CA, USA) 10 mM citric acid buffer (pH 7.0) in an autoclave for 10 min at at a 1:200 dilution. After washing with PBS, the cells were incubated 121ÆC. They were gradually cooled down to room temperature and with Alexa Fluor 488-labelled anti-mouse IgG antibody (A-11001 endogenous peroxidase activity was inhibited by incubation with Molecular Probes, Eugene, OR, USA) at a 1:500 dilution and H2O2 in PBS for 5 min. Subsequent staining was conducted in TO-PRO3 (T-3605 Molecular Probes) at a 1:1000 dilution, in the accordance with the manufacturer’s instructions. The same anti- dark for 1 h at room temperature. After washing with PBS, they p16 antibody (sc-1661 Santa Cruz) that was used for confocal laser were mounted with Mounting Medium (S3023 Dako Cytomation, scanning microscopy was used for staining as the primary antibody. Carpinteria, CA, USA) to observe p16 protein expression with a confocal laser scanning microscope (Carl Zeiss). Results Immunostaining of endometrial cancer specimens. The formalin- fixed and paraffin-embedded sections were sliced onto silane- Cell lines. Hypermethylation in the promoter region of the coated slides and were stained with a VECTESTAUN ABC Kit p16 gene was recognized with high frequency in 5 out of the (PK-6102 Vector Laboratories, Burlingame, CA, USA). After 8 endometrial cancer cell lines by MSP analysis (Figure 1).

854 Yanokura et al: DNA Hypermethylation of p16 in Endometrial Cancer

Figure 6. Immunostaining of endometrial cancer tissue specimens with an anti-p16 antibody. Positive cases are shown on the left panel and negative cases on the right panel. The nuclei of endometrial cancer cells are stained in positive cases.

However, loss of expression was found in only 1 out of the 5 development. However, it has been suggested that gene hypermethylated endometrial cancer cell lines by RT-PCR hypermethylation of the CpG island in the promoter region expression analysis (Figure 2). p16 gene expression was plays an important role in gene inactivation in cases with no observed in all 3 cell lines, wherein no hypermethylation was genetic abnormality (2). found. On the other hand, fluorescence staining of the p16 The methylation of genes such as hMLH1, APC and protein was conducted for the SNG-II, Ishikawa and SNG-M E-cadherin has been reported and is also considered to have endometrial cancer cell lines and a loss of p16 protein a close relationship with carcinogenesis in endometrial expression was recognized only in SNG-M, as was the case cancer (13-15). with RT-PCR (Figure 3). In this study, hypermethylation in the promoter region of DNA demethylation and recovery of p16 gene expression the p16 gene was recognized with high frequency in 5 out of were examined with 5-aza-dC in the SNG-M cell line, 8 endometrial cancer cell lines, whereas no hyper- wherein p16 protein expression was not observed either by methylation was observed in the endometrial cancer tissue RT-PCR or by fluorescence staining. Although DNA specimens. Bearing in mind these results, the methylation demethylation was observed with a decreased band of status in cell lines may not necessarily reflect the in vivo methyl in MSP after adding 5-aza-dC, recovery of gene status. Thus, it may be difficult to explain all carcinogenetic expression was not found by RT-PCR (Figure 4). functions in vivo with the results of cell line experiments. This is because other DNA methylation may occur during Tissue specimens. Hypermethylation was not observed in the the inmortalization of cells and subsequent passage. 7 normal endometrial tissue specimens, 12 atypical Many researchers have reported a relationship between endometrial hyperplasia tissue specimens or 32 endometrial the hypermethylation of the CpG island in the promoter cancer tissue specimens (Figure 5). However, immunostaining region of the p16 gene and loss of p16 protein expression. of the p16 protein was conducted in 20 of the 32 endometrial Similar results have been reported from experiments, not cancer tissue specimens used for MSP analysis, which were only with human but also rat cell lines. The relationship available for preparation of paraffin sections, and a loss of p16 between methylation and loss of p16 gene expression, not in protein expression was found in 3 out of the 20 specimens the promoter region but in the upper region near the TATA (15%) (Table III, Figure 6). box in rat cancer cell lines, was recently reported (16). In addition, in an investigation with human bladder cancer cell Discussion lines, the methylation of the p16 gene, not at the initiation point of but at a 50-300 bp upper point, was To date, the inactivation of genes by mutation and reported essential for p16 gene expression (17). It is not has been considered essential for carcinogenesis and cancer clear as to which domain must be methylated for

855 ANTICANCER RESEARCH 26: 851-856 (2006) inactivation. However, this may be the reason why the gene 7 Oue N, Oshimo Y, Nakayama H, Ito R, Yoshida K, Matsusaki expression was not inhibited though hypermethylation in the K and Yasui W: DNA methylation of multiple genes in gastric promoter domain of the p16 gene, as was observed in the carcinoma: association with histological type and CpG island methylator phenotype. Cancer Sci 94: 901-905, 2003. endometrial cancer cell lines used in this study. 8 Silva J, Silva JM, Dominguez G, Garcia JM, Cantos B, Although recovery of expression was not seen even after Rodriguez R, Larrondo FJ, Provencio M, Espana P and Bonilla 5-aza-dC treatment in the cultured cells and F: Concomitant expression of p16INK4a and in hypermethylation in the promoter region of the p16 gene primary breast cancer and analysis of inactivation mechanisms. was not observed in the endometrial cancer specimens, there J Pathol 199: 289-297, 2003. was loss of p16 protein expression. Thus, genetic functions 9 Parrella P, Poeta ML, Gallo AP, Prencipe M, Scintu M, such as DNA hypermethylation of upper domains other than Apicella A, Rossiello R, Liguoro G, Seripa D, Gravina C, Rabitti C, Rinaldi M, Nicol T, Tommasi S, Paradiso A, the promoter region, mutation and loss of heterozygosity Schittulli F, Altomare V and Fazio VM: Nonrandom seem to play important roles in the inactivation of the p16 distribution of aberrant promoter methylation of cancer- gene in endometrial cancer. Our results showed no related gene in sporadic breast tumors. Clin Cancer Res 10: hypermethylation in the promoter region of the p16 gene in 5349-5354, 2004. endometrial cancer, which is consistent with the report by 10 Wong YF, Chung TK, Cheung TH, Nobori T, Yu AL, Batova Nakashima et al. (11). We suggest that hypermethylation in A, Lai KW and Chang AM: Methylation of p16INK4A in the promoter region of the p16 gene may not be a common primary gynecologic malignancy. Cancer Lett 136: 231-235, 1999. phenomenon and that there is no relationship between 11 Nakashima R, Fujita M, Enomoto T, Haba T, Yoshino K, hypermethylation in the promoter region of the p16 gene Wada H, Kurachi H, Sasaki M, Wakasa K, Inoue M, Buzard G and inactivation of the p16 protein in the carcinogenesis of and Murata Y: Alteration of p16 and p15 genes in human endometrial cancer in Japanese patients. uterine tumours. Br J Cancer 80: 458-467, 1999. 12 Herman JG, Graff JR, Myohanen S, Nelkin BD and Baylin Acknowledgements SB: Methylation-specific PCR: a novel PCR assay for methylation status of CpG islands. Proc Natl Acad Sci USA Part of this work was supported by grants for Scientific Research 93: 9821-9826, 1996. from the Ministry of Education, Culture, Sports, Science and 13 Esteller M, Catasus L, Matias-Guiu X, Mutter GL, Prat J, Technology of Japan; contract grant number 17791135. Baylin SB and Herman JG: hMLH1 promoter hyper- methylation is an early event in human endometrial tumorigenesis. Am J Pathol 155: 1767-1772,1999. References 14 Zysman M, Saka A, Millar A, Knight J, Chapman W and Bapat B: Methylation of adenomatous polyposis coli in endometrial 1 Semczuk A, Boltze C, Marzec B, Szczygielska A, Roessner A cancer occurs more frequently in tumors with microsatellite and Schneider-Stock R: p16INK4a alterations are accompanied instability phenotype. Cancer Res 62: 3663-3666, 2002. by aberrant protein immunostaining in endometrial carcinomas. 15 Saito T, Nishimura M, Yamasaki H and Kudo R: J Cancer Res Clin Oncol 129: 589-596, 2003. Hypermethylation in promoter region of E-cadherin gene is 2 Ushijima T, Mihara M and Okouchi E: Tumorigenesis and associated with tumor dedifferention and myometrial invasion aberrant DNA methylation. New horizon for medicine. Tokyo, in endometrial carcinoma. Cancer 97: 1002-1009, 2003. Gendaiiryousya 35: 93-100, 2003. 16 Abe M, Okouchi E, Kuramoto T, Kaneda A, Takato T, 3 Toyota M, Satoh A, Akino K and Imai K: New aspects of Sugimura T and Ushijima T: Cloning of the 5’upstream region methylation study for cancer diagnosis and therapy. of the rat p16 gene and its role in silencing. Jpn J Cancer Res Experimental medicine. Tokyo, Yodosya 21: 161-167, 2003. 93: 1100-1106, 2002. 4 Lee S, Hwang KS, Lee HJ, Kim JS and Kang GH: Aberant 17 Gonzalgo ML, Hayashida T, Bender CM, Pao MM, Tsai YC, CpG island hypermethylation of multiple genes in colorectal Gonzales FA, Nguyen HD, Nguyen TT and Jones PA: The role neoplasia. Lab Invest 84: 884-893, 2004. of DNA methylation in expression of the p19/p16 in human 5 Burri N, Shaw P, Bouzourene H, Sordat I, Sordat B, Gillet M, bladder cancer cell lines. Cancer Res 58: 1245-1252, 1998. Schorderet D, Bosman FT and Chaubert P: Methylation silencing and mutation of the p14ARF and p16INK4a genes in colon cancer. Lab Invest 81: 217-229, 2001. 6 Ding Y, Le XP, Zhang QX and Du P: Methylation and mutation analysis of p16 gene in gastric cancer. World J Received September 16, 2005 Gastroenterol 9: 423-426, 2003. Accepted December 16, 2005

856