Inactivation of gene expression of some important tumor

GENERAL THORACIC Reduced Acetylated Histone H4 is Associated With Promoter Methylation of the Fragile Histidine Triad Gene in Resected Esophageal Squamous Cell Carcinoma Ching Tzao, MD, PhD, Guang-Huan Sun, MD, PhD, Ho-Jui Tung, PhD, Han-Shui Hsu, MD, Wen-Hu Hsu, MD, Yi-Ching Wang, PhD, Yeung-Leung Cheng, MD, PhD, and Shih-Chun Lee, MD Division of Thoracic Surgery and Department of Surgery, Cancer Epigenetics Laboratory, Tri-Service General Hospital, National Defense Medical Center, and Department of Humanity and Social Studies, National Defense Medical Center; Division of Thoracic Surgery, Veterans General Hospital; and Department of Life Sciences, National Taiwan Normal University, Taipei, Taiwan, China Background. Promoter methylation inactivates expression of some important tumor suppressor genes and may be associated with histone modification. The fragile histidine triad (FHIT) gene is considered a tumor suppressor gene in different human epithelial cancers. We investigated whether FHIT methylation is associated with aberrant expression of Fhit protein and acetylated histone, and whether aberrant expression of Fhit protein and acetylated histone are related to prognosis after resection for esophageal squamous cell cancer. Methods. We analyzed FHIT methylation using methylation-specific polymerase chain reaction and Fhit protein and acetylated histone H4 using immunohistochemistry in 60 resected tumor specimens. Concordance analysis was performed between FHIT methylation and expression of Fhit as well as H4. Results. The FHIT methylation was observed in 33(55%) specimens, and the aberrant expression of Fhit and acetylated H4 was found in 42 (70%) and 40 (67%) specimens, respectively. Expression of aberrant Fhit correlated positively with tumor staging (p < 0.017) and nodal involvement (p 0.004). Aberrant expression of acetylated H4 correlated positively with tumor staging (p < 0.001), nodal involvement (p < 0.001), and metastasis (p 0.004). Concordance rates of 75% and 81.7% were present between promoter methylation of FHIT and expression of Fhit (p 0.035) and acetylated H4 (p 0.02). Conclusions. Aberrant expression of Fhit and acetylated histone H4 are frequently associated with the presence of esophageal squamous cell carcinoma, and they are potential prognostic predictors for patients after resection of the tumor. (Ann Thorac Surg 2006;82:396 401) 2006 by The Society of Thoracic Surgeons Accepted for publication March 20, 2006. Presented at the Forty-second Annual Meeting of The Society of Thoracic Surgeons, Chicago, IL, Jan 30 Feb 1, 2006. Address correspondence to Dr Tzao, Division of Thoracic Surgery, Tri- Service General Hospital, National Defense Medical Center 325, Section 2, Cheng Gong Rd, Nei Hu, Taipei, Taiwan 114; e-mail: tzao@yahoo.com. Inactivation of gene expression of some important tumor suppressor genes such as p16 [1], E-cadherin [2], and the von Hippel Lindau (VHL) [3] gene has been implicated in human tumorigenesis. Among the various epigenetic alterations that lead to modified gene expression, the most important are believed to be DNA methylation [4] and chromatin remodeling by histone modification [5]. Promoter methylation is believed to be involved in the carcinogenic process in many cancers, including gastric, colorectal, breast, lung, ovarian, bladder, and oral cancers [6, 7]. Previous studies suggest a role of the FHIT (fragile histidine triad) gene, at chromosome 3p14.2, as a tumor suppressor gene in different epithelial cancers such as breast [8], gastric, colorectal, prostate, and lung [9]. Further investigations have shown that promoter methylation of FHIT plays an important role in tumorigenesis of breast [10], colorectal [11], and lung cancer [10, 12]. Aberrant FHIT expression is implicated in the development of esophageal squamous cell carcinomas, but the detailed mechanism responsible for its gene inactivation remains unclear [13, 14]. Reduction in acetylated histone H4 has been observed in gastric adenoma, gastric adenocarcinomas [15], and esophageal squamous cell cancer [16], suggesting that modification of histone is involved in their tumorigenesis. Of particular interest, recent studies show that there is a close association between histone acetylation and DNA methylation of some tumor suppressor genes in gastrointestinal carcinogenesis [17, 18]. A study by Fahrner and colleagues [19] suggested that promoter methylation of a number of tumor suppressor genes is related to modification of histone and leads to the development of cancer. Little is known about whether promoter methylation and aberrant expression of the FHIT gene are associated with histone modification, and there are no correlative 2006 by The Society of Thoracic Surgeons 0003-4975/06/$32.00 Published by Elsevier Inc doi:10.1016/j.athoracsur.2006.03.066

Ann Thorac Surg TZAO ET AL 2006;82:396 401 ACETYLATED HISTONE AND FHIT METHYLATION studies of aberrant expression of Fhit and acetylated histone and prognosis of esophageal squamous cell cancer. This study was designed to investigate whether reduced acetylated histone is related to promoter methylation of FHIT and its aberrant protein expression, and whether results from these molecular analyses correlate with patients cancer staging and survival. Material and Methods Study Population and Tumor Samples After esophagectomy for 60 patients with squamous cell carcinoma, archival paraffin-embedded tissue blocks of all primary tumor specimens and 20 of 60 matched normal mucosa controls were collected from January 1999 to December 2003 [20]. The study was reviewed and approved by the Institutional Review Board with waiver of patient consent, which allowed us to get access to patients medical records and to obtain tissue samples and pertinent follow-up data. From review of medical records, clinical information was collected and TNM status was recorded based upon the American Joint Committee on Cancer (AJCC) staging system [21] for esophageal cancer. Survival and follow-up data were obtained from the Institutional Cancer Registry. Analysis of Protein Expression: Immunohistochemistry Assay Paraffin blocks of tumors were cut into 5- m slices and then processed using standard deparaffinization and rehydration techniques. After antigen retrieval using microwave heating, all tissue sections on slides were immunostained following instructions from the antibody suppliers. A polyclonal anti-fhit antibody (Zymed Laboratories, South San Francisco, California) was used to detect Fhit protein, and a rabbit polyclonal antibody against the acetylated peptide corresponding to the amino acid 2-19 of tetrahymena histone H4 (Upstate Biotechnology, Lake Placid, New York) was used to detect acetylated histone H4. The binding of primary antibody was visualized using a detection system (DAKO LSAB Kit K675; DakoCytomation California, Carpinteria, California). The normal staining patterns for Fhit and acetylated H4 are cytoplasmic and nuclear, respectively. Tumor cells that exhibited an absence of staining in the presence of surrounding nonneoplastic cells with cytoplasmic or nuclear staining were considered to have an abnormal pattern. Staining results were examined by two observers masked to the status of the molecular analyses. Another reading by a third observer was needed to reach a consensus when there was a significant discrepancy between initial readings. DNA Extraction and Methylation-Specific Polymerase Chain Reaction Assay for FHIT Serial 5- m sections were cut from formalin-fixed, paraffin-embedded tissue blocks. All slides were stained with hemotoxylin and eosin, and one of the slides was used as a guide to localize tumor regions. The tumor cells 397 were microdissected from three slides of serial sections. After deparaffinization in xylene, genomic DNA of recovered tumor cells was prepared using proteinase K digestion and phenol/chloroform extraction followed by ethanol precipitation. The promoter methylation status of the FHIT gene of all tumor samples and their normal mucosa controls were determined by chemical treatment with sodium bisulfite and subsequent methylation-specific (MS) polymerase chain reaction (PCR) analysis (MSP) as described by Herman and coworkers [22]. The methylation status in the exon 1 region of the FHIT gene was determined by chemical treatment with sodium bisulfite and subsequent MSP analysis. For bisulfite treatment of DNA used for MSP, extracted DNA (1 g) in a volume of 50 L was denatured by NaOH (final concentration, 0.2M) for 10 minutes at 37 o C for samples with nanogram quantities of human DNA. Salmon sperm DNA (Sigma- Aldrich, St. Louis, Missouri), 1 g, was added as carrier before modification. Then 30 L of 10 mm hydroquinone (Sigma-Aldrich) and 520 L of 3 M sodium bisulfite (Sigma-Aldrich) at ph 5, both freshly prepared, were added and mixed, and samples were incubated under mineral oil at 50 o C for 16 hours. Modified DNA was purified using the Wizard DNA purification resin according to the manufacturer (Promega, Madison, Wisconsin) and eluted into 50 L water. Modification was completed by NaOH (final concentration, 0.3 M) treatment for 5 minutes at room temperature, followed by ethanol precipitation. The DNA was resuspended in water and stored at -20 o C until used for PCR amplification. As previously described [12, 13], primers for the methylated FHIT reaction were the D-SQ primer (forward) and 5= -CGTAAACGACGCCGAC- CCCACTA-3= (reverse); and primers for the unmethylated FHIT reaction were the D-SQ primer (forward) and 5=-CATAAACAACACCAACCCCACTA-3= (reverse). Polymerase chain reaction was performed for 45 cycles with annealing temperatures of 63 C and 68 C for unmethylated and methylated reactions, respectively, using 100 ng bisulfite-modified DNA. The DNA from BES6 normal bronchial cell line and H1299 lung cancer cell line was included in each assay to serve as unmethylated and methylated controls, respectively. Negative control samples lacking DNA were also included for each PCR set. All PCRs were performed with positive controls for both unmethylated and methylated alleles, and no DNA control. A ratio greater than 0.5 was defined as aberrant methylation. Statistical Analysis The Pearson 2 test was used to compare expression of Fhit and H4 protein, and methylation status of FHIT with different clinicopathologic characteristics of the patients in this study, including age, stage, and TNM category. Comparison of age between patients with and without alterations was made by the two-sample t test. The Kaplan-Meier method was used to estimate the probability of survival as a function of time and median survival [23]. The log-rank test was used to assess the significance of difference between pairs of survival prob- GENERAL THORACIC

GENERAL THORACIC 398 TZAO ET AL Ann Thorac Surg ACETYLATED HISTONE AND FHIT METHYLATION 2006;82:396 401 Table 1. Comparison Between Promoter Methylation, Protein Expression of FHIT, Expression of Acetylated Histone (H4) Protein, and Clinicopathologic Features for Patients Undergoing Esophagectomy for Squamous Cell Carcinoma of the Esophagus Methylation (FHIT) a Protein (Fhit) a Protein (H4) b Characteristics Patients (%) (%) (%) (%) (%) (%) Overall 60 27 (45%) 33 (55%) 18 (30%) 42 (70%) 20 (33.3%) 40 (66.7%) Age (mean SD) 63 10 61 11 64 12 62 10 64 11 62 10 Staging I II 39 21 (35%) 18 (30%) c 18 (30%) 22 (36.7%) d 19 (31.7%) 20 (33.3%) e III IV 21 6 (10%) 15 (25%) 3 (5%) 20 (33.3%) 1 (1.7%) 20 (33.3%) TNM T1 2 20 13 (21.7%) 7 (11.6%) f 10 (16.7%) 10 (16.7%) 7 (11.6%) 13 (21.7%) T3 4 40 14 (23.3%) 26 (43.4%) 8 (13.3%) 32 (53.3%) 13 (21.7%) 27 (45%) N0 35 18 (30%) 17 (28.3%) g 17 (28.3%) 21 (35%) h 20 (33.3%) 15 (25%) i N1 25 9 (15%) 16 (26.7%) 1 (1.7%) 21 (35%) 0 (0%) 25 (41.7%) M0 56 25 (41.7%) 31 (51.7%) 17 (28.3%) 39 (65%) 20 (33.3%) 36 (60%) j M1 4 2 (3.3%) 2 (3.3%) 1 (1.7%) 3 (5%) 0 (0%) 4 (6.7%) a These groups represent patients with alteration in the FHIT gene/protein. b These groups represent patients with H4 protein. c,d,e Value of p 0.017, 0.0001, and 0.0001, respectively, between stage I II and III IV. f Value of p 0.018 between stage T1 T2 and T3 T4. g,h,i Value of p 0.025, 0.004, and 0.0001, respectively, between N0 and N1. j Value of p 0.004 between M0 and M1. T tumor invasion; N regional lymph nodes; M distant sites (metastases). abilities [24]. A p value less than 0.05 was considered statistically significant. Results Expression of Fhit Protein and Acetylated Histone H4 in Relation to Patients Clinicopathologic Variables and Survival The associations of altered expression of Fhit protein and acetylated histone H4 with patients clinicopathologic parameters are summarized in Table 1. Aberrant expression of Fhit protein and acetylated histone H4 were observed in 42 (70%) and 40 (66.7%) patients, respectively. Aberrant expression for Fhit protein correlated positively with stage (p 0.017) and N status (p 0.004), whereas reduced expression of acetylated histone H4 showed significant positive correlation with tumor stage (p 0.001), N status (p 0.001), and M status (p 0.004). Median survival durations in months were 29 2 and 20 6 (SE) for patients with normal and aberrant expres- Fig 1. Kaplan-Meier survival curve with log-rank test for 60 patients after resection for esophageal squamous cell carcinoma. Comparison of survival between patients with normal (dotted line) and aberrant (solid line)expression of fragile histidine triad (Fhit) protein and acetylated histone H4 is shown in (A) and (B), respectively. The 95% confidence intervals for median survival in groups of normal and aberrant Fhit expression were 25 to 33 and 9 to 31, respectively, and in normal and aberrant H4 groups, 28 to 30 and 9 to 31, respectively.

Ann Thorac Surg TZAO ET AL 2006;82:396 401 ACETYLATED HISTONE AND FHIT METHYLATION 399 GENERAL THORACIC Fig 2. Representative results of promoter methylation for fragile histidine triad (FHIT) gene. Primer sets used for amplification were designated as U for the unmethylated product and M for methylated genes; N represents normal mucosa sample and T represents tumor sample. Normal mucosa from patients (pt.) 1 and 2 showed only unmethylated FHIT, whereas tumor samples from these 2 patients also showed methylated FHIT (bp base pair.) sion of Fhit protein, and 29 1 and 20 5 for patients with normal and reduced expression of acetylated histone H4, respectively. No significant difference in survival existed between patients with normal and aberrant expression of Fhit protein (p 0.23) or acetylated histone H4 (p 0.21; Fig 1). Promoter Methylation of FHIT and Its Association With Clinicopathologic Variables Representative results of promoter methylation of FHIT are shown in Figure 2. All normal mucosa controls showed only unmethylated FHIT. Thirty-three tumor specimens (55%) showed methylated FHIT (Table 1). Promoter methylation of FHIT correlated positively with tumor stage (p 0.017), T status (p 0.018), and N status (p 0.025; Table 1). Median survival durations in months were 22 7 and 29 4 for patients with unmethylated and methylated FHIT gene, respectively (p 0.60). Correlation of Promoter Methylation With Aberrant Protein Expression of FHIT A concordance rate of 75% was observed between promoter methylation and aberrant protein expression of FHIT (p 0.035; Table 2), suggesting that promoter methylation may play a role in the inactivation of protein expression for FHIT. Correlation of Promoter Methylation of FHIT With Aberrant Expression of Acetylated Histone H4 A concordance rate of 81.7% was observed between promoter methylation and aberrant expression of acetylated histone H4 (p 0.02; Table 3), suggesting that reduced expression of acetylated histone H4 may be one of the mechanisms that is associated with promoter methylation of the FHIT gene. Comment Aberrant expression of some important tumor suppressor genes has been implicated in carcinogenesis of human cancers [1 3]. Methylation of DNA [4] and chromatin remodeling by histone modification [5] are believed to be major mechanisms that leads to inactivation of expression of genes that are involved in tumorgenesis. To date, there are limited reports on whether promoter methylation and aberrant expression of the FHIT gene are associated with histone modification, and there are no correlative studies of aberrant expression of Fhit and acetylated histone with the prognosis of esophageal squamous cell cancer. We designed this study to investigate whether reduced acetylated histone was related to promoter methylation of FHIT and aberrant protein expression of Fhit, and whether results from these molec- Table 2. Concordance Analysis Between Promoter Methylation and Protein Expression of the FHIT Gene in Resected Esophageal Squamous Cell Carcinoma Table 3. Concordance Analysis Between Promoter Methylation of the FHIT Gene and Protein Expression of Acetylated H4 in Resected Esophageal Squamous Cell Carcinoma Protein Expression Methylated Promoter Methylation Unmethylated Protein Expression Methylated Promoter Methylation Unmethylated Aberrant 30 (50%) a 12 (20%) b Normal 3 (5%) b 15 (25%) a Concordant versus discordant Concordant 45 (75%) a Discordant 15 (25%) b a Concordant: methylated/aberrant and unmethylated/normal. b Discordant: unmethylated/aberrant and methylated/normal. Value of p 0.035 for correlation between promoter methylation and protein expression of the FHIT gene. Data were presented in actual number and percentage of cases in a total of 60 patients. Aberrant 31 (51.7%) a 9 (15%) b Normal 2 (3.3%) b 18 (30%) a Concordant versus discordant Concordant 49 (81.7%) a Discordant 11 (18.3%) b a Concordant: methylated/aberrant and unmethylated/normal. b Discordant: unmethylated/aberrant and methylated/normal. Value of p 0.02 for correlation between promoter methylation of the FHIT gene and protein expression of acetylated H4. Data were presented in actual number and percentage of cases in a total of 60 patients.

GENERAL THORACIC 400 TZAO ET AL Ann Thorac Surg ACETYLATED HISTONE AND FHIT METHYLATION 2006;82:396 401 ular studies correlated with patients clinicopathologic features. Aberrant expression of Fhit protein was observed in a considerable number of our patients in both early and late cancer stages. This result supports a previous study [14] suggesting that loss of Fhit protein is an early event in the development of esophageal squamous cell cancer. Consistent with a frequency reported previously (68.4%) in invasive esophageal squamous cell carcinoma [16], a high frequency (66.7%) of loss of acetylated histone H4 protein was also observed in our patients. Neither of the studies mentioned above explored the correlations among aberrant protein expression of Fhit or acetylated histone and patient s clinicopathologic parameters as was performed in this study. Aberrant expression of Fhit protein and acetylated histone H4 correlated well with patients overall stage and N status, with loss of acetylated histone H4 also correlating well with M status, suggesting that they may serve as potential predictors of patients clinical outcome. Intriguingly, survival analysis did not show significant correlation between patients with normal and aberrant expression of Fhit or acetylated histone H4, although they showed significant correlation with cancer staging. The absence of correlation between survivals and protein expression of Fhit and acetylated histone H4 may be attributed to patient selection and the small sample size studied. A concordance of 75% was observed between promoter methylation of FHIT and its protein expression, confirming previous results that suggest promoter methylation of FHIT is a major mechanism responsible for silencing of its gene expression and an active player in carcinogenesis for some human cancers [10 12]. Within the remaining 25% of discordance, 12 tumor samples (20%) presented with unmethylatred promoter of FHIT but aberrant expression of Fhit protein, suggesting that there may be mechanisms other than promoter methylation leading to inactivation of gene expression, such as loss of heterozygosity or splicing of mrna transcript as reported previously [12, 13]. Promoter methylation of FHIT was observed in a fairly high frequency among our patients and it correlated well with tumor staging, T status, and N status, suggesting that it may well be a potential diagnostic marker and an outcome predictor for esophageal squamous cell cancer. The exact mechanisms for DNA methylation in human cancers remains unclear. Several mechanisms have been proposed, including increasing enzymatic activity of DNA methyltransferase and imbalance between histone acetylation and deacetylation [25, 15]. Recent studies show that here is a close association between histone acetylation and DNA methylation of some tumor suppressor genes in gastrointestinal carcinogenesis [17, 18]. The association of acetylated histone with a methylated promoter region has been demonstrated in a number of tumor suppressor genes such as p21 (WAF1/CIP1), hmlh1, p16 INK4a, and p14 ARF, which leads to the development of cancer [19, 26]. Similarly, our results demonstrated a close association between methylation of the FHIT with histone aceylation in esophageal squamous cell carcinoma. Taken together, these results suggest that promoter methyaltion and histone deacetylation are firmly linked. The exact mechanism of how histone deacetylation affects promoter methylation needs further investigation, possibly by modifying acetylation status of histone by using the histone deacetylase inhibitor trichostatin A to see whether methyaltion of FHIT could be reversed with its protein reexpressed [27] in an esophageal squmous cell cancer cell line. In conclusion, our results demonstrate that aberrant expression of Fhit and acetylated histone H4 frequently occur in patients with esophageal squamous cell carcinoma and are potential prognostic predictors. Promoter methylation of FHIT is at least one mechanism for the inactivation of its protein expression. A close association between FHIT methylation and aberrant expression of H4 suggests that reduced expression of acetylated histone may be involved in promoter methylation of the FHIT gene. The authors thank Dr Mark Ferguson at the University of Chicago Hospitals for his review of this manuscript. The authors are also indebted to Drs Cheng-Ping Yu and Yi-Jen Perng at the Department of Pathology, Tri-Service General Hospital, for their review of histopathology and immunohistochemisty for tissue sections. This work was supported by grants from the National Science Council, Taiwan (NSC 92-2314-B-016-043), and the Research Foundation, Tri-Service General Hospital (TSGH-C95-4-S04). References 1. Herman JG, Merlo A, Mao L, et al. Inactivation of the CDKN2/p16/MTS1 gene is frequently associated with aberrant DNA methylation in all common human cancers. Cancer Res 1995;55:4525 30. 2. Graff JR, Herman JG, Lapidus RG, et al. E-cadherin expression is silenced by DNA hypermethylation in human breast and prostate carcinomas. Cancer Res 1995;55:5195 9. 3. Herman JG, Latif F, Weng Y, et al. Silencing of the VHL tumor-suppressor gene by DNA methylation in renal carcinoma. Proc Natl Acad Sci USA 1994;91:9700 4. 4. Baylin, SB, Herman JG. DNA hypermethylation in tumorigenesis: epigenetics joins genetics. Trends Genet 2000;16: 168 74. 5. Kouzarides T. Histone acetylases and deacetylases in cell proliferation. Curr Opin Genet Dev 1999;9:40 8. 6. Das PM, Singal R. DNA methylation and cancer. J Clin Oncol 2004;15:22:4632-42. 7. Leung SY, Yuen ST, Chung LP, et al. hmlh1 promoter methylation and lack of hmlh1 expression in sporadic gastric carcinomas with high-frequency microsatellite instability. Cancer Res 1999;59:159 64. 8. Negrini M, Monaco C, Vorechovsky I, et al. The FHIT gene at 3p14.2 is abnormal in breast carcinomas. Cancer Res 1996;56:3173 9. 9. Croce CM, Sozzi G, Huebner K. Role of FHIT in human Cancer. J Clin Oncol 1999;17:1618 24. 10. Zoechbauer-Mueller S, Fong KM, Maitra A, et al. 5 CpG island methylation of the FHIT gene is correlated with loss of gene expression in lung and breast cancer. Cancer Res 2001;61:3581 5. 11. Ahuja N, Mohan AL, Li Q, et al. Association between CpG island methylation and microsatellite instability in colorectal cancer. Cancer Res 1997;57:3370 4. 12. Tzao C, Tsai HY, Chen JT, Chen CY, Wang YC. 5 CpG island hypermethylation and aberrant transcript splicing both con-

Ann Thorac Surg TZAO ET AL 2006;82:396 401 ACETYLATED HISTONE AND FHIT METHYLATION tribute to the inactivation of the FHIT gene in resected non-small cell lung cancer. Eur J Cancer 2004;40:2175 83. 13. Tanaka H, Shimada Y, Harada H, et al. Methylation of the 5 CpG island of the FHIT gene is closely associated with transcriptional inactivation in esophageal squamous cell carcinomas. Cancer Res 1998;58:3429 34. 14. Mori M, Mimori K, Shiraishi T, et al. Altered expression of Fhit in carcinoma and precarcinomatous lesions of the esophagus. Cancer Res 2000;60:1177 82. 15. Ono S, Oue N, Kuniyasu H, et al. Acetylated histone H4 is reduced in human gastric adenomas and carcinomas. J Exp Clin Cancer Res 2002;21:377 82. 16. Toh Y, Yamamoto M, Endo K, et al. Histone H4 acetylation and histone deacetylase 1 expression in esophageal squamous cell carcinoma. Oncol Rep 2003;10:333 8. 17. Yasui W, Oue N, Ono S, Mitani Y, Ito R, Nakayama H. Histone acetylation and gastrointestinal carcinogenesis. Ann NY Acad Sci 2003;983:220 31. 18. Mitani Y, Oue N, Hamai Y, et al. Histone H3 acetylation is associated with reduced p21(waf1/cip1) expression by gastric carcinoma. J Pathol 2005;205:65 73. 19. Fahrner JA, Eguchi S, Herman JG, Baylin SB. Dependence of histone modifications and gene expression on DNA hypermethylation in cancer. Cancer Res 2002;62:7213 8. 20. Tzao C, Hsu HS, Sun GH, et al. Promoter methylation of the hmlh1 gene and protein expression of human mutl homolog 1 401 and human muts homolog 2 in resected esophageal squamous cell carcinoma. J Thorac Cardiovasc Surg 2005;130:1371 7. 21. Fleming ID, Cooper JS, Henson DE, et al. Digestive system: esophagus. In: Fleming ID, ed. AJCC cancer staging manual. 5th ed. Philadelphia: Lippincott Williams & Wilkins, 1997:65 9. 22. Herman JG, Graff JR, Myohanen S, Nelkin BD, Baylin SB. Methylation-specific PCR: a novel PCR assay for methylation status of CpG islands. Proc Natl Acad Sci USA 1996; 93:9821 6. 23. Kaplan EL, Meier P. Nonparametric estimation from incomplete observation. J Am Stat Assoc 1958;53:457 81. 24. The Lifetest procedure. SAS technical report: P-179, additional SAS/STAT procedures, release 6.03. Cary, NC: SAS Institute, 1988:49-90. 25. Esteller M. CpG island hypermethylation and tumor suppressor genes: a booming present, a brighter future. Oncogene 2002;21:5427 40. 26. Baylin SB, Esteller M, Rountree MR, Bachman KE, Schuebel K, Herman JG. Aberrant patterns of DNA methylation, chromatin formation and gene expression in cancer. Hum Mol Genet 2001;10:687 92. 27. Xiong Y, Dowdy SC, Podratz KC, et al. Histone deacetylase inhibitors decrease DNA methyltransferase-3b messenger RNA stability and down-regulate de novo DNA methyltransferase activity in human endometrial cells. Cancer Res 2005;65:2684 9. GENERAL THORACIC DISCUSSION DR KING F. KWONG (Baltimore, MD): I would like to thank you for a nice paper, obviously encompassing a very difficult issue, that of esophageal cancer. I just have a few brief questions for you. As you are aware, in Western countries, such as the United States, we face a high preponderance of adenocarcinoma. So my first question to you is, do you see these same epigenetic changes occurring within a series of adenocarcinoma specimens? DR TZAO: Thank you, Dr Kwong. This is a very interesting issue. Actually in our country, we rarely see patients with adenocarcinoma. It has been reported that more than 85% of our patients were squamous cell carcinoma. We don t have enough tissues for adenocarcinoma in our tissue bank to be analyzed for similar epigenetic changes that have been studied for squamous cell carcinoma. However, it s certainly interesting to look into the differences of epigenetic alterations between these two cell types. DR KWONG: Obviously, it would be very attractive to be able to utilize this knowledge in order to strengthen our therapeutic potentials for these patients. Has your group utilized real-time PCR in order to identify some of these epigenetic changes, therefore allowing small tissue biopsy samples to be brought into use? And as a follow-up question, in the United States the National Cancer Institute has recently instituted and championed the use of histone deacetylase inhibitors in some of these aerodigestive malignancies. Is that under way or have plans been made in your country for trying to treat these patients with histone deacetylase inhibitors in a clinical trial? Once again, thank you for a nice paper. DR TZAO: Thanks for the excellent questions. To answer your first question on the real-time PCR, the molecular diagnosis after operation, I believe that it is a powerful tool for preoperative diagnosis and staging or intraoperative confirmation of nodal involvement so on so forth, using very tiny samples or cells that are harvested. Currently we are not doing that. As you know, there are a number of molecular markers identified, and FHIT is one of them for squamous cell carcinoma, both in lung and esophageal cancer, which is quite specific for these two types of cancers. So there is a potential for us explore that area in the future. The question is, we should probably look into what are the best markers or to establish a reliable marker profiling for all cancers of interests. As to your second question, actually, this is a translational study. We don t really know exactly how the epigenetic mechanism is working in these tissues. What we are doing is, to use 11 esophageal cancer cell lines, including squamous cell cancer and adenocarcinoma as well, we are trying to manipulate these cell lines in vitro and in vivo with known epigenetic drugs, like histone deacetylase inhibitors, the HDACIs, or azacitidine, the demethylating agent, to see if these treatments reverse the epigenetic alterations existing in these cell lines and whether that will inhibit cell growth. Hopefully, results from this study will shed some light on their therapeutic implication in terms of preoperative induction chemotherapy or postoperative adjuvant therapy. Thank you.