5 Cpg Island Methylation of the FHIT Gene Is Correlated with Loss Of

[CANCER RESEARCH 61, 3581–3585, May 1, 2001] Advances in Brief

5؅ CpG Island Methylation of the FHIT Gene Is Correlated with Loss of Gene Expression in Lung and Breast Cancer1

Sabine Zo¨chbauer-Mu¨ller, Kwun M. Fong, Anirban Maitra, Stephen Lam, Joseph Geradts, Raheela Ashfaq, Arvind K. Virmani, Sarah Milchgrub, Adi F. Gazdar, and John D. Minna2 Hamon Center for Therapeutic Oncology Research [S. Z-M., A. M., A. K. V., A. F. G., J. D. M.] and Departments of Pathology [A. M., R. A., S. M., A. F. G.], Internal Medicine [J. D. M.], and Pharmacology [J. D. M.], The University of Texas Southwestern Medical Center, Dallas, Texas 75390; Department of Thoracic Medicine, The Prince Charles Hospital, Brisbane 4032, Australia [K. M. F.]; British Columbia Cancer Agency, Vancouver V5Z 355, Canada [S. L.]; and Nuffield Department of Clinical Laboratory Sciences, University of Oxford, John Radcliffe Hospital, Oxford OX3 9DU, United Kingdom [J. G.]

Abstract Aberrant methylation (referred to as methylation) of normally unmethylated CpG islands, located in the 5Ј promoter region of genes, Allele loss and loss of expression of fragile histidine triad (FHIT), a has been associated with transcriptional inactivation of several genes putative tumor suppressor gene located in chromosome region 3p14.2, are in human cancer and can serve as an alternative to mutational inac- frequent in several types of cancers. Tumor-acquired methylation of ␤ promoter region CpG islands is one method for silencing tumor suppres- tivation (12, 13). Several genes, including p16, RAR , TIMP-3, -sor genes. We investigated 5؅ CpG island methylation of the FHIT gene in DAPK, H-cadherin, and RASSF1A frequently undergo such methyl 107 primary non-small cell lung cancer (NSCLC) samples and corre- ation in lung and breast cancers (12, 14–23). In esophageal squamous sponding nonmalignant lung tissues, 39 primary breast carcinomas, as cell carcinomas, methylation of the FHIT gene was found in three of well as in 49 lung and 22 breast cancer cell lines by a methylation-specific four structurally unaltered but transcriptionally repressed tumor cell PCR assay. In addition, we analyzed brushes from the bronchial epithe- lines as well as in 5 of 35 primary tumors (24). lium of 35 heavy smokers without cancer. FHIT methylation was detected Because of the potential role of FHIT in the pathogenesis of lung in 37% of primary NSCLCs, 31% of primary breast cancers, and 65% of and breast cancers, we investigated the methylation status of FHIT in lung and 86% of breast cancer cell lines. The frequency of methylation in these tumors and correlated our findings with gene expression at the small cell and NSCLC cell lines were identical. Methylation was found in RNA and protein levels and with clinical features. 9% of the corresponding nonmalignant lung tissues and in 17% of bronchial brushes from heavy cigarette smokers. FHIT methylation was sig- nificantly correlated with loss of FHIT mRNA expression by Northern Materials and Methods blot analysis in lung cancer cell lines and with loss of Fhit expression in NSCLC and breast tumors by immunostaining. We conclude that meth- Tumor and Tissue Samples and Tumor Cell Lines. Tissues were col- ylation of FHIT is a frequent event in NSCLC and breast cancers and is lected after obtaining appropriate Institutional Review Board permission and ϭ an important mechanism for loss of expression of this gene. Methylation informed consent from the subjects. Primary NSCLC tumors (n 107) and ϭ of FHIT commences during lung cancer pathogenesis and may represent corresponding nonmalignant lung tissues (n 104) were obtained from a marker for risk assessment. patients who had received surgical resections with curative intent at The Prince Charles Hospital. Areas of viable tumor tissue were macrodissected and stored Introduction at Ϫ70°C. This cohort of patients had been investigated previously for various genetic abnormalities (2, 8, 18, 23, 25, 26). There were 76 males and 31 The FHIT3 gene, located in chromosome region 3p14.2, undergoes females, ages 28–81 years (mean, 61 years) at diagnosis. Sixty-one patients frequent allele loss (LOH) and occasional homozygous deletions in had stage I disease, 21 patients had stage II disease, 24 patients had stage IIIA various cancer types, including lung and breast cancer (1–5). FHIT disease, and 1 patient had stage IIIB disease. Histological subtypes included 45 transcripts of abnormal size, including deletions of exons and inser- adenocarcinomas, 43 squamous cell carcinomas, 11 adenosquamous carcino- tions, are found in a high percentage of lung and breast cancers, mas, 4 large cell carcinomas, 3 atypical carcinoids, and 1 typical carcinoid. whereas point mutations are rare (1, 2, 4, 5). Loss of Fhit protein Ninety-eight patients were ever smokers (consisting of current and former smokers) with mean exposure of 31 pack-years, and 9 were never smokers. expression also is frequent in these cancer types (6–10). Although Survival data of 5 or more years were available for most patients. Primary these findings suggest that FHIT is a TSG, which plays an important breast tumors (n ϭ 39) were obtained from women undergoing breast cancer role in the pathogenesis of lung and breast cancer, there remains resections at The University of Texas Southwestern Medical Center. The age considerable controversy as to the role of the gene in tumorigenesis of the patients ranged between 31 and 84 years (mean, 57 years) at diagnosis. (reviewed in Ref. 11). Three patients had stage I, 16 stage IIA, 3 stage IIB, 8 stage IIIA, 3 stage IIIB, and 6 stage IV disease. Histological subtypes included 33 ductal carcinomas ϭ Received 2/19/01; accepted 3/13/01. and 6 lobular carcinomas. Smokers (n 35) with at least 30 pack-years of The costs of publication of this article were defrayed in part by the payment of page smoking history and with morphometric evidence of sputum dysplasia were charges. This article must therefore be hereby marked advertisement in accordance with examined under fluorescence bronchoscopy at the British Cancer Agency (27). 18 U.S.C. Section 1734 solely to indicate this fact. Bronchial brushes were obtained from a predetermined site (apical segment of 1 Supported by Grants J1658-MED and J1860-MED from the Austrian Science Foun- dation, Lung Cancer Specialized Program of Research Excellence Grant P50 CA70907, right upper lobe). Bronchial biopsies were taken from multiple areas that and grants from The G. Harold and Leila Y. Mathers Charitable Foundation, The Susan demonstrated fluorescence abnormalities. Pathological examination revealed G. Komen Foundation and the Cancer Research Foundation of North Texas. dysplasia in at least 1 biopsy in 33 of 35 cases (mild dysplasia in 26 cases and 2 To whom requests for reprints should be addressed, at Hamon Center for Therapeutic moderate dysplasia in 7 cases). Twenty-five NSCLCs, 22 SCLCs, and 2 Oncology Research, The University of Texas Southwestern Medical Center, 6000 Harry Hines Boulevard, Dallas, TX 75390-8593. Phone: (214) 648-4900; Fax: (214) 648-4940; mesothelioma cell lines (referred to as lung cancer cell lines) and 22 breast E-mail: [email protected]. cancer cell lines generated by us have been described previously (28–30). 3 The abbreviations used are: FHIT, fragile histidine triad; TSG, tumor suppressor Corresponding B lymphoblastoid cell lines were available for 52 lines. gene; NSCLC, non-small lung cancer; SCLC, small cell lung cancer; MSP, methylation- MSP. DNA was prepared from tissue samples and cell lines by standard specific PCR; LOH, loss of heterozygosity; RT-PCR, reverse transcription-PCR; p16, p16INK4a; RAR, retinoic acid receptor; TIMP, tissue inhibitor of metalloproteinase; methods, and bisulfite modification of genomic DNA was performed as DAPK, death-associated protein kinase. reported by Herman et al. (31). Briefly, 1 ␮g of genomic DNA was denatured 3581

Downloaded from cancerres.aacrjournals.org on October 2, 2021. © 2001 American Association for Cancer Research. FHIT METHYLATION with NaOH (final concentration, 0.2 M) and 10 mM hydroquinone (Sigma Frequency of FHIT Methylation. Using the primers we designed, Chemical Co., St. Louis, MO), and 3 M sodium bisulfite (Sigma Chemical Co.) we determined the frequency of FHIT methylation in primary NSCLC were added and incubated at 50°C for 16 h. Afterward, modified DNA was samples and corresponding nonmalignant lung tissues, primary breast purified using Wizard DNA purification resin (Promega Corp., Madison, WI), carcinomas, and lung and breast cancer cell lines by MSP (Fig. 1), and followed by ethanol precipitation. Treatment of genomic DNA with sodium the results are summarized in Table 1. FHIT methylation was also bisulfite converts unmethylated but not methylated cytosines to uracil, which found in 6 of 35 bronchial brushes from heavy smokers without lung are then converted to thymidine during the subsequent PCR step, producing sequence differences between methylated and unmethylated DNA. Primer cancer. Three of the positive cases were from subjects with mild sequences for the methylated FHIT reaction were 5Ј-TTG GGG CGC GGG dysplasia, and three were from subjects with moderate dysplasia. The TTT GGG TTT TTA CGC-3Ј (forward) and 5Ј-CGT AAA CGA CGC CGA unmethylated form of FHIT was found in 100% of the primary tumors CCC CAC TA-3Ј (reverse), and primer sequences for the unmethylated FHIT that had been grossly dissected and thus had at least some contami- reaction were 5Ј-TTG GGG TGT GGG TTT GGG TTT TTA TG-3Ј (forward) nation with normal cells and also in all nonmalignant specimens. In and 5Ј-CAT AAA CAA CAC CAA CCC CAC TA-3Ј (reverse). Primer contrast to the tumor samples, the cancer cell lines represent pure sequences were determined on the basis of the sequence data of the 5Ј CpG populations of tumor cells, and we determined their FHIT allele island of the gene as described in “Results.” The PCR mixture contained 10ϫ status. We found that the vast majority of these tumor lines contained PCR buffer (Qiagen, Inc., Valencia, CA), deoxynucleotide triphosphates (1.25 either the methylated or the unmethylated form, and only occasional mM), primers (final concentration, 0.6 ␮M each per reaction), 1 unit of cell lines contained both forms (Table 2). HotStarTaq (Qiagen, Inc.) and bisulfite-modified DNA (ϳ100 ng). A touch- down PCR with an annealing temperature between 71°C and 64°C was FHIT Methylation and FHIT Expression by Northern Blot performed. Amplification was carried out in a 9700 Perkin-Elmer Thermal Analysis. Expression of FHIT mRNA by Northern blot analysis in 19 Cycler. DNA from peripheral blood lymphocytes and buccal smears of healthy lung cancer cell lines had been reported previously by us (2). Unde- individuals were used as positive controls for the unmethylated form; DNA tectable or very low levels of FHIT mRNA were found in 16 of these from peripheral blood lymphocytes treated with SssI methyltransferase (New cell lines. We compared the FHIT methylation results of these cell England BioLabs, Inc., Beverly, MA) was used as a positive control for lines with the FHIT mRNA expression data and found a significant methylated alleles. Negative control samples without DNA were included for correlation between FHIT methylation and loss of FHIT mRNA each set of PCR. PCR products were analyzed on 3% agarose gels and expression (P ϭ 0.009). Fifteen of 16 cell lines that expressed no or visualized under UV illumination. The PCR reactions for all samples demon- very low levels of FHIT mRNA were methylated for FHIT, whereas strating methylation were repeated to confirm these results. 2 of 3 cell lines that expressed FHIT mRNA were not methylated. One 5-Aza-2؅-deoxycytidine Treatment and RT-PCR. The NCI-H1299 cell line did not express FHIT mRNA and was not methylated, NSCLC cell line was incubated in culture medium with and without 1 ␮M 5-aza-2Ј-deoxycytidine (Sigma Chemical Co.) for 6 days (16). RNA was presumably inactivating FHIT expression by another mechanism. prepared, and RT-PCR was performed to detect FHIT expression using FHIT Only one methylated cell line expressed FHIT mRNA. primers 5RT-F and 3D2, as described previously (24). Primers for glyceral- FHIT Methylation and Loss of Fhit Expression by Immuno- dehyde-3-phosphate dehydrogenase were used to confirm RNA integrity (16). staining. Fhit immunostaining was performed in 98 of the 107 PCR products were analyzed on 2% agarose gels. NSCLC samples and was absent in 53% as described previously by Other Molecular Markers. Data on immunostaining of Fhit have been Geradts et al. (8). In addition, immunostaining for Fhit expression was reported (in the case of NSCLCs) or were performed (in the case of breast also performed in 36 primary breast cancers. Loss of Fhit expression cancer samples) as described previously (8). LOH analysis at 3p14.2 using was observed in 26 of 36 (72%) breast cancer samples. polymorphic markers (D3S1300, D3S4103, and D3S1234) has been reported previously for lung cancer (2, 3) and was performed for the breast cancer samples as described (3). Other available molecular markers from previous studies included K-ras codon 12, p53 exons 5–8 mutations, and the methylation status of the genes RAR␤, RASSF1A, TIMP-3, p16, O6-methylguanine- DNA-methyltransferase (MGMT), death-associated protein kinase (DAPK), E-cadherin (ECAD), p14ARF (p14), and glutathione S-transferase P1 (GSTP1) in the 107 primary NSCLC samples (18, 23, 25). Statistics. Statistical analysis was performed using ␹2 test for differences between groups and t tests between means. Overall survival was calculated using Kaplan-Meier log rank testing.

Results Genomic Sequencing of the FHIT 5؅ CpG Island after Bisulfite Treatment. Genomic sequencing of the FHIT 5Ј CpG island (Gen- Bank accession numbers U76262 and U76263) after bisulfite treatment of DNA from seven primary NSCLCs that lacked Fhit immunostaining was performed as described (8, 24). The methylation pattern of the 5Ј CpG island of the FHIT gene in these seven samples was very similar to the pattern described by Tanaka et al. (24) and indicated that the region between nucleotides 195 and 283 (from GenBank sequence number U76263) was highly methylated. Two sets of PCR primers that distinguish between methylated and unmethyl- Fig. 1. Methylation analysis of FHIT in various specimens. TU, tumor; NL, nonma- ated DNA sequences in the 5Ј region of the gene were designed based lignant lung tissue; Lane U, amplified product with primers recognizing unmethylated on these results. Using the original primer set (32), the correlation sequence (74-bp PCR product); Lane M, amplified product with primers recognizing methylated sequence (74-bp PCR product). Positive controls include peripheral blood between positive methylation and loss of immunostaining was poor lymphocytes from healthy individuals for the unmethylated form and in vitro methylated (data not shown). A second set of primers was designed, as described DNA for the methylated form of FHIT. Water blanks were used as negative controls. In the case of primary breast carcinomas, lung and breast cancer cell lines, and bronchial in “Materials and Methods,” and used for all experiments described brushes, the amplified product represents the methylated form of FHIT. Lanes that do not herein. show a band represent samples that are not methylated. 3582

Table 1 Frequency of FHIT methylation in primary tumors, tumor cell lines, and sample was scored as not informative. The concordance between LOH nonmalignant specimens and methylation was 56% for breast cancer cell lines and 48% for No. tested FHIT methylated primary breast carcinomas, respectively. Primary NSCLCs 107 40 (37%) Clinicopathological Characteristics. The methylation results Corresponding nonmalignant lung 104 9 (9%) from the primary NSCLCs and breast cancers were compared with Lung cancer cell lines 49 32 (65%) NSCLCs 25 16 (64%) clinicopathological characteristics from these patients including sex, SCLCs 22 14 (64%) age, histology, and Tumor-Node-Metastasis classification of the tu- Mesotheliomas 2 2 (100%) Bronchial brushesa 35 6 (17%) mors. For the NSCLC patients, data about smoking history and overall Primary breast carcinomas 39 12 (31%) survival also were available. No significant correlation between FHIT Breast cancer cell lines 22 19 (86%) methylation and any of these parameters was observed for the NSCLC a From heavy smokers without evidence of cancer. or breast cancer patients. However, there appeared to be a tendency for FHIT methylation to be associated with an increase in pack-years Table 2 Presence of methylated and unmethylated FHIT alleles in lung and breast smoked (mean pack-years, 38 versus 49; P ϭ 0.08) in the NSCLC cancer cell lines patients. We also observed that none of the lobular breast carcinomas As demonstrated here, 67 of 71 (94%) tumor cell lines had only methylated or ϭ unmethylated alleles. In only 4 tumor cell lines (6%) were both methylated and unmethy- examined were methylated. Although the sample size (n 6) is too lated forms present. We found no tumor cell line that lacked both methylated and small to reach statistical significance, this possible association needs unmethylated forms. to be investigated in a larger series. Unmethylated FHIT Methylation and Other Markers. No correlation between Methylated allele allele NSCLC SCLC Mesothelioma Breast Total FHIT methylation and K-ras codon 12 mutations and p53 exons 5–8 ϩϪ14 13 2 18 47 mutations were observed in the primary NSCLC samples. FHIT Ϫϩ98 0 320 ϩϩ21 0 14methylation was also not correlated with the methylation status of ϪϪ00 0 00RAR␤, RASSF1A, TIMP-3, p16, MGMT, DAPK, p14, and GSTP1 but Total 25 22 2 22 71 was correlated with the methylation status of ECAD (P ϭ 0.01) in the primary NSCLC samples.

Of the 98 NSCLC samples studied, 38 were methylated; of these, Discussion 27 lacked immunostaining, 4 were uniformly positive, and 7 showed We studied the methylation status of the 5Ј CpG island of FHIT in a heterogeneous pattern of immunostaining, with focal areas of pos- lung and breast cancers and also their adjacent nonmalignant tissues itivity and negativity. Of the 36 breast cancer samples studied, 20 and the bronchial epithelium of smokers without lung cancer. We first were methylated, and of these, 16 lacked immunostaining, 1 was sequenced sodium bisulfite-treated DNA in the FHIT 5Ј CpG island uniformly positive for staining, and 3 showed a heterogeneous pattern. region as reported by Tanaka et al. (24). The methylation pattern we Because of the difficulty of interpreting the tumors showing hetero- obtained by sequencing several primary NSCLCs with loss of Fhit geneous staining patterns, they were excluded from the analysis of expression by immunostaining was very similar to the pattern that had concordance. Of the remaining 48 methylated lung and breast cancer been described previously (24). On the basis of the results, we samples, 43 lacked expression. Of the methylation-negative samples, designed two sets of primer pairs for use in MSP assays. The results 35 of 60 (58%) NSCLC samples and 6 of 16 (38%) breast cancer presented in this report exclusively used the redesigned primer set as samples were positive for immunostaining. Thus, of a total of 76 described in “Materials and Methods.” Using the redesigned primer methylation-negative lung and breast tumors, 41 (54%) lacked immu- set, we found a high percentage of primary NSCLC samples, primary nostaining, suggesting that methods of gene silencing other than breast carcinomas, as well as lung and breast cancer cell lines meth- methylation may exist. The correlation between FHIT methylation ylated for FHIT. The higher percentage of methylation found in the Ͻ and lack of Fhit expression was highly significant (P 0.0001). lung and breast cancer cell lines compared with primary tumors might -Reexpression of FHIT after Treatment with 5-Aza-2؅-deoxycy tidine. Loss of FHIT expression in the NSCLC cell line NCI-H1299 has been described previously by Northern blot analysis and RT-PCR (2) and confirmed in the present study. We found this cell line methylated for FHIT, making it a suitable candidate for treatment with the demethylating agent 5-aza-2Ј-deoxycytidine. Reexpression of FHIT was seen after treatment of NCI-H1299 cells with 5-aza-2Ј- deoxycytidine, confirming the role of the 5Ј region CpG methylation in regulating FHIT expression (Fig. 2). FHIT Methylation and LOH. We also compared the FHIT methylation results with data about LOH at the FHIT locus using polymorphic markers (D3S1300, D3S4103, and D3S1234) that had been reported previously in lung cancer cell lines and primary NSCLC samples (2, 3). The concordance between LOH and methylation was 63% for lung cancer cell lines and 43% for primary NSCLC samples, respectively. LOH analysis on the FHIT locus was also performed on 20 breast cancer cell lines. Allele loss was seen in 10 (50%) cell lines, whereas 8 (40%) cell lines were heterozygous and 2 (10%) cell lines were Fig. 2. Expression study of FHIT in the NCI-H1299 NSCLC cell line by RT-PCR. The classified as not informative. In addition, we analyzed 24 microdis- expression of FHIT is lost in the cell line NCI-H1299 but can be restored after treatment Ј Ϫ ϩ sected primary breast carcinomas and found FHIT locus LOH in 12 with 5-aza-2 -deoxycytidine (5-aza-dC). , NCI-H1299 without 5-aza-dC; , NCI- H1299 with 5-aza-dC; Lane H2O, water blank; Lane Pos. Control, positive control; (50%) samples. Eleven (46%) samples were heterozygous, and 1 (4%) GAPDH, glyceraldehyde-3-phosphate dehydrogenase. 3583

Downloaded from cancerres.aacrjournals.org on October 2, 2021. © 2001 American Association for Cancer Research. FHIT METHYLATION be explained by additional changes acquired in culture, or that the detected in the bronchial epithelium from heavy smokers. Moreover, tumor cell lines were derived from more aggressive tumors and that we were able to show that FHIT methylation is associated with both these had acquired more changes. loss of FHIT mRNA expression and loss of Fhit expression, and that We compared tumor FHIT methylation and loss of expression of methylation of this gene is reversible with 5-aza-2Ј-deoxycytidine. FHIT mRNA by Northern blot analysis and Fhit expression by im- Finally, our findings of a frequent acquired tumor-related epigenetic munostaining. A significant correlation between FHIT methylation alteration favor the candidacy of FHIT as a TSG. and loss of expression by Northern blot was seen for lung cancer and by immunostaining for lung and breast cancers. Seven methylated References NSCLCs and 3 breast cancer samples exhibited heterogeneity of Fhit 1. Sozzi, G., Veronese, M. L., Negrini, M., Baffa, R., Cotticelli, M. G., Inoue, H., immunostaining. A possible explanation for this finding is that there Tornielli, S., Pilotti, S., De Gregorio, L., Pastorino, U., Pierotti, M. A., Ohta, M., was tumor heterogeneity for methylation. Of 78 NSCLC and breast Huebner, K., and Croce, C. M. The FHIT gene 3p14.2 is abnormal in lung cancer. Cell, 85: 17–26, 1996. cancer specimens that completely lacked Fhit immunoreactivity, only 2. Fong, K. M., Biesterveld, E. J., Virmani, A., Wistuba, I., Sekido, Y., Bader, S. A., 43 (55%) were methylated. In the methylated tumors, we presume that Ahmadian, M., Ong, S. T., Rassool, F. V., Zimmerman, P. V., Giaccone, G., Gazdar, lack of expression occurred via biallelic inactivation, by a combina- A. F., and Minna, J. D. FHIT and FRA3B 3p14.2 allele loss are common in lung cancer and preneoplastic bronchial lesions and are associated with cancer-related tion of methylation and allelic loss. However, because 35 (45%) of the FHIT cDNA splicing aberrations. Cancer Res., 57: 2256–2267, 1997. tumors lacking protein expression were not methylated, we presume 3. Wistuba, I. I., Behrens, C., Virmani, A. K., Mele, G., Milchgrub, S., Girard, L., that mechanisms of inactivating the second allele other than methyl- Fondon, J. W., III, Garner, H. R., McKay, B., Latif, F., Lerman, M. I., Lam, S., Gazdar, A. F., and Minna, J. D. High resolution chromosome 3p allelotyping of ation must exist. Another fact that supports the importance of FHIT human lung cancer and preneoplastic/preinvasive bronchial epithelium reveals mul- methylation in inactivating this gene is the reexpression of the FHIT tiple, discontinuous sites of 3p allele loss and three regions of frequent breakpoints. gene after treatment with the demethylating agent 5-aza-2Ј-deoxycy- Cancer Res., 60: 1949–1960, 2000. 4. Negrini, M., Monaco, C., Vorechovsky, I., Ohta, M., Druck, T., Baffa, R., Huebner, tidine. This finding is in agreement with the results found by Tanaka K., and Croce, C. M. The FHIT gene at 3p14.2 is abnormal in breast carcinomas. et al. (24) in esophageal squamous cell carcinomas. Taken together, Cancer Res., 56: 3173–3179, 1996. these findings demonstrate that methylation of FHIT is an important 5. Ahmadian, M., Wistuba, I. I., Fong, K. M., Behrens, C., Kodagoda, D. R., Saboorian, M. H., Shay, J., Tomlinson, G. E., Blum, J., Minna, J. D., and Gazdar, A. F. Analysis mechanism for silencing this gene in lung and breast cancers. of the FHIT gene and FRA3B region in sporadic breast cancer, preneoplastic lesions, We also compared the results on FHIT methylation with data on and familial breast cancer probands. Cancer Res., 57: 3664–3668, 1997. 6. Sozzi, G., Pastorino, U., Moiraghi, L., Tagliabue, E., Pezzella, F., Ghirelli, C., LOH in lung and breast cancers. Although both events were frequent Tornielli, S., Sard, L., Huebner, K., Pierotti, M. A., Croce, C. M., and Pilotti, S. Loss in the two tumor types, the correlation between them was relatively of FHIT function in lung cancer and preinvasive bronchial lesions. Cancer Res., 58: low. 5032–5037, 1998. 7. Tomizawa, Y., Nakajima, T., Kohno, T., Saito, R., Yamaguchi, N., and Yokota, We did not find any significant correlation between FHIT methy- J. Clinicopathological significance of Fhit protein expression in stage I non-small cell lation status and clinicopathological characteristics of the NSCLC and lung carcinoma. Cancer Res., 58: 5478–5483, 1998. breast cancer patients. However, we observed a tendency that the 8. Geradts, J., Fong, K. M., Zimmerman, P. V., and Minna, J. D. Loss of Fhit expression in non-small-cell lung cancer: correlation with molecular genetic abnormalities and number of pack-years smoked by the NSCLC patients may be asso- clinicopathological features. Br. J. Cancer, 82: 1191–1197, 2000. ciated with FHIT methylation. This finding is particularly interesting 9. Tseng, J. E., Kemp, B. L., Khuri, F. R., Kurie, J. M., Lee, J. S., Zhou, X., Liu, D., because several authors reported an association between loss of Fhit Hong, W. K., and Mao, L. Loss of Fhit is frequent in stage I non-small cell lung cancer and in the lungs of chronic smokers. Cancer Res., 59: 4798–4803, 1999. expression and the smoking history of lung cancer patients (6, 7, 9). 10. Campiglio, M., Pekarsky, Y., Menard, S., Tagliabue, E., Pilotti, S., and Croce, C. M. Only nine never smokers were included in our study. Of these, five FHIT loss of function in human primary breast cancer correlates with advanced stage of the disease. Cancer Res., 59: 3866–3869, 1999. tumors were not methylated, and the other four were methylated. 11. Zo¨chbauer-Mu¨ller, S., Wistuba, I. I., Minna, J. D., and Gazdar, A. F. Fragile histidine However, the number of nonsmokers in our study is relatively small; triad (FHIT) gene abnormalities in lung cancer. Clin. Lung Cancer, 2: 141–145, 2000. therefore, a study with a larger number of nonsmokers is necessary to 12. Merlo, A., Herman, J. G., Mao, L., Lee, D. J., Gabrielson, E., Burger, P. C., Baylin, S. B., and Sidransky, D. 5Ј CpG island methylation is associated with transcriptional answer the question of whether FHIT methylation is associated with silencing of the tumour suppressor p16/CDKN2/MTS1 in human cancers. Nat. Med., smoking exposure. Interestingly, FHIT methylation was only detected 1: 686–692, 1995. in ductal breast cancers but not in lobular breast cancers. This finding 13. Baylin, S. B., Herman, J. G., Graff, J. R., Vertino, P. M., and Issa, J. P. Alterations in DNA methylation: a fundamental aspect of neoplasia. Adv. Cancer Res., 72: needs to be confirmed by study of a larger number of these cancers. 141–196, 1998. Sozzi et al. (6) reported loss of Fhit immunostaining in 85% of 14. Kashiwabara, K., Oyama, T., Sano, T., Fukuda, T., and Nakajima, T. Correlation bronchial dysplastic lesions and in 100% of carcinoma in situ lesions between methylation status of the p16/CDKN2 gene and the expression of p16 and Rb proteins in primary non-small cell lung cancers. Int. J. Cancer, 79: 215–220, 1998. adjacent to the tumor or at the resection margin of the surgical 15. Esteller, M., Sanchez-Cespedes, M., Rosell, R., Sidransky, D., Baylin, S. B., and samples. Our data published previously indicated that allele loss at the Herman, J. G. Detection of aberrant promoter hypermethylation of tumor suppressor genes in serum DNA from non-small cell lung cancer patients. Cancer Res., 59: FHIT locus (3p14.2) occurred relatively late (at the dysplastic stage) 67–70, 1999. during lung cancer pathogenesis (3). Therefore, we investigated 16. Virmani, A. K., Rathi, A., Zo¨chbauer-Mu¨ller, S., Sacchi, N., Fukuyama, Y., Bryant, whether FHIT methylation can also be detected in the bronchial D., Maitra, A., Heda, S., Fong, K. M., Thunnissen, F., Minna, J. D., and Gazdar, A. F. Promoter methylation and silencing of the retinoic acid receptor-␤ gene in lung epithelium of heavy smokers. Although almost all of the subjects had carcinomas. J. Natl. Cancer Inst., 92: 1303–1307, 2000. dysplastic changes in at least one of several bronchial biopsies, the 17. Sato, M., Mori, Y., Sakurada, A., Fujimura, S., and Horii, A. The H-cadherin brushes were obtained from a predetermined site, and cytological (CDH13) gene is inactivated in human lung cancer. Hum. Genet., 103: 96–101, 1998. 18. Zo¨chbauer-Mu¨ller, S., Fong, K. M., Virmani, A. K., Geradts, J., Gazdar, A. F., and examination of the cells was not performed. We found a relatively Minna, J. D. Aberrant promoter methylation of multiple genes in non-small cell lung high percentage (17%) of FHIT methylation in these samples, sug- cancers. Cancer Res., 61: 249–255, 2001. gesting that FHIT methylation is an early event in the pathogenesis of 19. Bachman, K. E., Herman, J. G., Corn, P. G., Merlo, A., Costello, J. F., Cavenee, W. K., Baylin, S. B., and Graff, J. R. Methylation-associated silencing of the tissue lung cancer. FHIT methylation in the nonmalignant lung tissues inhibitor of metalloproteinase-3 gene suggests a suppressor role in kidney, brain, and adjacent to resected NSCLC tumors may be explained by the fact that other human cancers. Cancer Res., 59: 798–802, 1999. 20. Bovenzi, V., Le, N. L., Cote, S., Sinnett, D., Momparler, L. F., and Momparler, R. L. methylation of certain genes occurs during multistage pathogenesis. DNA methylation of retinoic acid receptor ␤ in breast cancer and possible therapeutic Whether FHIT methylation represents a risk factor for development of role of 5-aza-2Ј-deoxycytidine. Anticancer Drugs, 10: 471–476, 1999. central or peripheral lung cancers needs to be studied. 21. Tang, X., Khuri, F. R., Lee, J. J., Kemp, B. L., Liu, D., Hong, W. K., and Mao, L. Hypermethylation of the death-associated protein (DAP) kinase promoter and aggres- In conclusion, we found methylation of FHIT frequently in primary siveness in stage I non-small-cell lung cancer. J. Natl. Cancer Inst., 92: 1511–1516, NSCLC samples and primary breast carcinomas, and it was also 2000. 3584

22. Dammann, R., Li, C., Yoon, J. H., Chin, P. L., Bates, S., and Pfeifer, G. P. Epigenetic intraepithelial neoplastic lesions by fluorescence bronchoscopy. Chest, 113: 696– inactivation of a RAS association domain family protein from the lung tumour 702, 1998. suppressor locus 3p21.3. Nat. Genet., 25: 315–319, 2000. 28. Phelps, R. M., Johnson, B. E., Ihde, D. C., Gazdar, A. F., Carbone, D. P., McClintock, 23. Burbee, D. G., Forgacs, E., Zo¨chbauer-Mu¨ller, S., Shivakumar, L., Fong, K. M., Gao, P. R., Linnoila, R. I., Matthews, M. J., Bunn, P. A., Jr., Carney, D., Minna, J. D., and B., Randle, D., Kondo, M., Virmani, A., Bader, S., Sekido, Y., Latif, F., Milchgrub, Mulshine, J. L. NCI-Navy Medical Oncology Branch cell line data base. J. Cell. S., Toyooka, S., Gazdar, A. F., Lerman, M. I., Zabarovsky, E., White, M., and Minna, Biochem. Suppl., 24: 32–91, 1996. J. D. Epigenetic inactivation of RASSF1A in lung and breast cancers and malignant 29. Wistuba, I. I., Bryant, D., Behrens, C., Milchgrub, S., Virmani, A. K., Ashfaq, R., phenotype suppression. J. Natl. Cancer Inst., in press, 2001. Minna, J. D., and Gazdar, A. F. Comparison of features of human lung cancer cell 24. Tanaka, H., Shimada, Y., Harada, H., Shinoda, M., Hatooka, S., Imamura, M., and lines and their corresponding tumors. Clin. Cancer Res., 5: 991–1000, 1999. Ishizaki, K. Methylation of the 5Ј CpG island of the FHIT gene is closely associated 30. Gazdar, A. F., Kurvari, V., Virmani, A., Gollahon, L., Sakaguchi, M., Westerfield, with transcriptional inactivation in esophageal squamous cell carcinomas. Cancer M., Kodagoda, D., Stasny, V., Cunningham, H. T., Wistuba, I. I., Tomlinson, G., Res., 58: 3429–3434, 1998. Tonk, V., Ashfaq, R., Leitch, A. M., Minna, J. D., and Shay, J. W. Characterization 25. Fong, K. M., Zimmerman, P. V., and Smith, P. J. Microsatellite instability and other of paired tumor and nontumor cell lines established from patients with breast cancer. molecular abnormalities in non-small cell lung cancer. Cancer Res., 55: 28–30, 1995. Int. J. Cancer, 78: 766–774, 1998. 26. Geradts, J., Fong, K. M., Zimmerman, P. V., Maynard, R., and Minna, J. D. 31. Herman, J. G., Graff, J. R., Myo¨ha¨nen, S., Nelkin, B. D., and Baylin, S. B. Methy- Correlation of abnormal RB, p16ink4a, and p53 expression with 3p loss of heterozy- lation-specific PCR: a novel PCR assay for methylation status of CpG islands. Proc. gosity, other genetic abnormalities, and clinical features in 103 primary non-small cell Natl. Acad. Sci. USA, 93: 9821–9826, 1996. lung cancers. Clin. Cancer Res., 5: 791–800, 1999. 32. Virmani, A. K., Muller, C., Rathi, A., Zo¨chbauer-Mu¨ller, S., Mathis, M., and Gazdar, 27. Lam, S., Kennedy, T., Unger, M., Miller, Y. E., Gelmont, D., Rusch, V., Gipe, B., A. F. Aberrant methylation during cervical carcinogenesis. Clin. Cancer Res., 7: Howard, D., LeRiche, J. C., Coldman, A., and Gazdar, A. F. Localization of bronchial 584–589, 2001.

3585

Downloaded from cancerres.aacrjournals.org on October 2, 2021. © 2001 American Association for Cancer Research. 5′ CpG Island Methylation of the FHIT Gene Is Correlated with Loss of Gene Expression in Lung and Breast Cancer

Sabine Zöchbauer-Müller, Kwun M. Fong, Anirban Maitra, et al.

Cancer Res 2001;61:3581-3585.

Updated version Access the most recent version of this article at: http://cancerres.aacrjournals.org/content/61/9/3581

Cited articles This article cites 31 articles, 17 of which you can access for free at: http://cancerres.aacrjournals.org/content/61/9/3581.full#ref-list-1

Citing articles This article has been cited by 31 HighWire-hosted articles. Access the articles at: http://cancerres.aacrjournals.org/content/61/9/3581.full#related-urls

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://cancerres.aacrjournals.org/content/61/9/3581. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.