The NuRD complex cooperates with DNMTs to maintain silencing of key colorectal tumor suppressor genes

Many tumor suppressor genes (TSGs) are silenced through synergistic layers of epigenetic regulation including abnormal DNA hypermethylation of promoter CpG islands, repressive chromatin modifications and enhanced nucleosome deposition over transcription start sites. The protein complexes responsible for silencing of many of such TSGs remain to be identified. Our previous work demonstrated that multiple silenced TSGs in colorectal cancer cells can be partially reactivated by DNA demethylation in cells disrupted for the DNA methyltransferases 1 and 3B (DNMT1 and 3B) or by DNMT inhibitors (DNMTi). Herein, we used proteomic and functional genetic approaches to identify additional proteins that cooperate with DNMTs in silencing these key silenced TSGs in colon cancer cells. We discovered that DNMTs and the core components of the NuRD (Mi-2/nucleosome remodeling and deacetylase) nucleosome remodeling complex, chromo domain helicase DNA-binding protein 4 (CHD4) and histone deacetylase 1 (HDAC1) occupy the promoters of several of these hypermethylated TSGs and physically and functionally interact to maintain their silencing. Consistent with this, we find an inverse relationship between expression of HDAC1 and 2 and these TSGs in a large panel of primary colorectal tumors. We demonstrate that DNMTs and NuRD cooperate to maintain the silencing of several negative regulators of the WNT and other signaling pathways. We find that depletion of CHD4 is synergistic with DNMT inhibition in reducing the viability of colon cancer cells in correlation with reactivation of TSGs, suggesting that their combined inhibition may be beneficial for the treatment of colon cancer. Since CHD4 has ATPase activity, our data identify CHD4 as a potentially novel drug target in cancer.

Prognostic value of CpG island methylator phenotype among colorectal cancer patients: a systematic review and meta-analysis

BACKGROUND

Divergent findings regarding the prognostic value of CpG island methylator phenotype (CIMP) in colorectal cancer (CRC) patients exist in current literature. We aim to review data from published studies in order to examine the association between CIMP and CRC prognosis.

MATERIALS AND METHODS

A comprehensive search for studies reporting disease-free survival (DFS), overall survival (OS), or cancer-specific mortality of CRC patients stratified by CIMP is carried out. Study findings are summarized descriptively and quantitatively, using adjusted hazard ratios (HRs) as summary statistics.

RESULTS

Thirty-three studies reporting survival in 10 635 patients are included for review. Nineteen studies provide data suitable for meta-analysis. The definition of CIMP regarding gene panel, marker threshold, and laboratory method varies across studies. Pooled analysis shows that CIMP is significantly associated with shorter DFS (pooled HR estimate 1.45; 95% confidence interval (CI) 1.07-1.97, Q = 3.95, I(2) = 0%) and OS (pooled HR estimate 1.43; 95% CI 1.18-1.73, Q = 4.03, I(2) = 0%) among CRC patients irrespective of microsatellite instability (MSI) status. Subgroup analysis of microsatellite stable (MSS) CRC patients also shows significant association between shorter OS (pooled HR estimate 1.37; 95% CI 1.12-1.68, Q = 4.45, I(2) = 33%) and CIMP. Seven studies have explored CIMP's value as a predictive factor on stage II and III CRC patient's DFS after receiving adjuvant 5-fluorouracil (5-FU) therapy: of these, four studies showed that adjuvant chemotherapy conferred a DFS benefit among CIMP(+) patients, one concluded to the contrary, and two found no significant correlation. Insufficient data was present for statistical synthesis of CIMP's predictive value among CRC patients receiving adjuvant 5-FU therapy.

CONCLUSION

CIMP is independently associated with significantly worse prognosis in CRC patients. However, CIMP's value as a predictive factor in assessing whether adjuvant 5-FU therapy will confer additional survival benefit to CRC patients remained to be determined through future prospective randomized studies.

OT3-01-06: A Phase 2 Study Investigating the Safety, Efficacy and Surrogate Biomarkers of Response of 5-Azacitidine (5-AZA) and Entinostat (MS-275) in Patients with Advanced Breast Cancer

Background: Epigenetic alterations in the genome, including abnormal DNA methylation and histone hypoacetylation, initiate and promote cancerous changes via several mechanisms, including inactivation of tumor suppressor genes. Preclinical investigations in breast cancer suggest that use of epigenetic modifiers results in re-expression of aberrantly silenced genes and proteins that represent important therapeutic targets (e.g. estrogen receptor alpha, ER). Combination therapy with a DNA methyltransferase inhibitor (DNMTI) and a histone deacetylase inhibitor (HDACI) has yielded superior ER reexpression and greater restoration of tamoxifen responsiveness than with HDACI alone. We hypothesize that clinically tolerable doses of the DNMTI 5-azacitidine (5-AZA) and the HDACI entinostat may not only effect changes in DNA methylation and gene expression, but also yield objective disease responses in women with advanced breast cancer.

Trial design: This multicenter phase II study (NCT01349959) is enrolling patients with advanced human epidermal growth factor receptor 2 (HER2)-negative breast cancer with triple negative (ER/progesterone receptor [PR]/HER2−negative, Cohort A) or hormone-resistant (Cohort B) disease. Patients will receive 5-AZA 40 mg/m2 subcutaneously days 1–5 and 8–10 and entinostat 7 mg orally days 3 and 10 every 28 days. Because of the potential for re-expression of the ER with epigenetic agents, patients will be offered continuation of 5-AZA and entinostat at progression with the addition of hormonal therapy (investigator discretion). Mandatory tumor biopsies will be performed at baseline and after 8 weeks of therapy to evaluate correlative biomarkers.

Eligibility Criteria: Eligible patients must be ≥ 18 years, have measurable locally advanced/metastatic triple-negative (at least one prior chemotherapy received adjuvant/metastatic setting) or hormone-resistant (must have received two prior hormonal agents and one prior chemotherapy) disease, adequate organ function and ECOG PS ≤ 2.

Specific Aims:

1. Objective response rate (ORR) by RECIST 1.1 criteria.

2. Safety and tolerability

3. Progression-free survival, overall survival and clinical benefit rate.

4. Safety and toxicity data, feasibility and response rate where hormonal therapy is added to the combination under investigation at the time of progressive disease.

5. Pharmacokinetics, cytidine deaminase, changes from baseline of candidate gene methylation and expression in circulating deoxyribonucleic acid (DNA) and malignant tissue.

Statistical Methods:

Using a two-stage three-outcome design to assess the efficacy of the combination, a maximum of 30 patients (requiring 27 evaluable) will be accrued to each cohort unless undue toxicity is encountered for a maximum sample size of 60 patients. The study design tests the null hypothesis that the ORR is at most 5% against the alternative hypothesis that it is at least 20% with a type I error of 4% and power of 90%.

Present and Targeted Accrual: This study has just opened to patient enrollment. We anticipate a rapid accrual of 60 patients within 1 year.br](Funding from Stand Up to Cancer and CTEP).

Citation Information: Cancer Res 2011;71(24 Suppl):Abstract nr OT3-01-06.

Role of nuclear architecture in epigenetic alterations in cancer

It is widely accepted that cancer results from an array of epigenetic and genetic alterations, particularly aberrant epigenetic patterns that are a hallmark of every cancer type studied. Another well-known feature of cancer cells is the array of abnormalities in their nuclear structure. Although it is known that nuclear structure has an important role in the regulation of gene expression, we know little about the direct relationship between nuclear structural alterations and aberrant epigenetic patterns in cancer. Here, we discuss some of the recent studies from our lab and others to understand the relationship between alterations of nuclear architecture and aberrant epigenetic patterns in cancer cells. Although the precise relationship remains elusive, we suggest that changes in nuclear structure and composition could alter long-range genomic interactions and cause global epigenetic changes during tumorigenesis. We emphasize the need for further studies to elucidate the direct relationship between nuclear structure alterations and aberrant epigenetic patterns in cancers.

Transcriptional regulation of Wnt inhibitory factor-1 by Miz-1/c-Myc

The Wnt signaling pathway is capable of self-regulation through positive and negative feedback mechanisms. For example, the oncoprotein c-Myc, which is upregulated by Wnt signaling activity, participates in a positive feedback loop of canonical Wnt signaling through repression of Wnt antagonists DKK1 and SFRP1. In this study, we investigated the mechanism of Wnt inhibitory factor-1 (WIF-1) silencing. Mapping of CpG island methylation of the WIF-1 promoter reveals regional methylation (-295 to -95 bp from the transcription start site) that correlates with transcriptional silencing. We identified Miz-1 as a direct activator of WIF-1 transcriptional activity, which is found at WIF-1 promoter. In addition, we show that c-Myc contributes to WIF-1 transcriptional repression in a Miz-1-dependent manner. Although the transient repression mediated by Miz-1/c-Myc is independent of de novo methylation, the stable repression by this complex is associated with CpG island methylation of the critical -295 to -95-bp region of the WIF-1 promoter. Importantly, Miz-1 and c-Myc are found at WIF-1 promoter in WIF-1 non-expressing cell lines DLD-1 and 209myc. Transient knockdown or somatic knockout of c-Myc in DLD-1 failed to restore WIF-1 expression suggesting that c-Myc is involved in initiating rather than maintaining WIF-1 epigenetic silencing. In a genome-wide screen, DNAJA4, TGFβ-induced and TRIM59 were repressed by c-Myc overexpression and DNA promoter hypermethylation. Our data reveal novel insights into c-Myc-mediated DNA methylation-dependent transcriptional silencing, a mechanism that might contribute to the dysregulation of Wnt signaling in cancer.

A potential tumor suppressor role for Hic1 in breast cancer through transcriptional repression of ephrin-A1

The tumor suppressor gene hypermethylated in cancer 1 (HIC1), which encodes a transcriptional repressor, is epigenetically inactivated in various human cancers. In this study, we show that HIC1 is a direct transcriptional repressor of the gene encoding ephrin-A1, a cell surface ligand implicated in the pathogenesis of epithelial cancers. We also show that mouse embryos lacking both Hic1 alleles manifest developmental defects spatially associated with the misexpression of ephrin-A1, and that overexpression of ephrin-A1 is a feature of tumors arising in Hic1 heterozygous mice in which the remaining wild-type allele is epigenetically silenced. In breast cancer, we find that ephrin-A1 expression is common in vivo, but that in cell culture, expression of the EphA receptors is predominant. Restoration of HIC1 function in breast cancer cells leads to a reduction in tumor growth in vivo, an effect that can be partially rescued by co-overexpression of ephrin-A1. Interestingly, overexpression of ephrin-A1 in vitro triggers downregulation of EphA2 and EphA4 levels, resulting in an expression pattern similar to that seen in vivo. We conclude that Hic1 spatially restricts ephrin-A1 expression in development, and that upregulated expression of ephrin-A1 resulting from epigenetic silencing of HIC1 in cancer cells may be an important mechanism in epithelial malignancy.

Interim analysis of a phase II trial of 5-azacitidine (5AC) and entinostat (SNDX-275) in relapsed advanced lung cancer (NSCLC)

8055

Background: Epigenetic gene silencing mediated through aberrant DNA methylation and histone deacetylation is a key contributor to lung carcinogenesis. Pre-clinical studies suggest that combining inhibitors of DNA methyltransferase (DNMT) with inhibitors of histone deacetylase (HDAC) synergistically induce re-expression of epigenetically-silenced tumor suppressor genes. Clinical studies at our institution combining the DNMT inhibitor, 5AC, with the HDAC inhibitor, entinostat, in hematologic malignancies have shown remarkable clinical activity. We hypothesized a similar effect would be seen in lung cancer. This study aims to assess the response rate and time to progression of 5AC and entinostat in NSCLC. Methods: Patients (pts) include adults with histologically confirmed recurrent NSCLC and progressive disease after ≥1 prior chemotherapy regimen. 40 mg/m2 of 5AC is administered SQ days 1–6 and 8–10 with 7 mg of entinostat PO days 3 and 10 of a 28 day cycle. A standard Simon two stage design is being used. The sample size is calculated with a power of 90% and a two-sided type 1 error allowance of 5%. Stage 1 included 18 pts, with subsequent expansion to a total of 32. Results: 25 pts have enrolled to date. Demographic characteristics include: mean age (range) - 63 (46- 80); M:F - 1:2; 80% former smokers; 80% adenocarcinomas; mean # of previous therapies - 3. One pt. has had a complete response. She remained on therapy for 14 m and came off therapy due to a new nodule which was resected. Molecular analysis suggests a second primary stage I NSCLC. She remains disease free at 20 m. Another man has stabilization of disease (SD) for ≥16 m with marked symptomatic improvement. Another pt had SD for 4 months. The remaining pts have progressed after 2 cycles of therapy. Main toxicities included injection site reactions, nausea/vomiting, constipation, fatigue, and hematologic toxicities. Conclusions: The combination of 5AC and entinostat is safe and well tolerated in advanced NSCLC pts. 2 pts have had durable benefit from treatment, including a complete response. Pharmacodynamic and pharmacokinetic analyses are being conducted to identify characteristics of the subset of pts responding to this novel therapy.

[Table: see text]

Methylation-specific PCR

Methylation-specific PCR is a rapid method used to determine the methylation status of DNA. Not only does methylation affect the expression of genes in normal cells, but it is now known that disease processes such as cancer can result in abnormal DNA methylation patterns. Methylation-specific PCR can be used to investigate imprinted genes, to assess human tumors for clonality by studying genes inactivated on the X chromosome, and to examine abnormally methylated CpG islands in neoplasia.

Frequent epigenetic inactivation of Wnt antagonist genes in breast cancer

Although mutation of APC or CTNNB1 (β-catenin) is rare in breast cancer, activation of Wnt signalling is nonetheless thought to play an important role in breast tumorigenesis, and epigenetic silencing of Wnt antagonist genes, including the secreted frizzled-related protein (SFRP) and Dickkopf (DKK) families, has been observed in various tumours. In breast cancer, frequent methylation and silencing of SFRP1 was recently documented; however, altered expression of other Wnt antagonist genes is largely unknown. In the present study, we found frequent methylation of SFRP family genes in breast cancer cell lines (SFRP1, 7 out of 11, 64%; SFRP2, 11 out of 11, 100%; SFRP5, 10 out of 11, 91%) and primary breast tumours (SFRP1, 31 out of 78, 40%; SFRP2, 60 out of 78, 77%; SFRP5, 55 out of 78, 71%). We also observed methylation of DKK1, although less frequently, in cell lines (3 out of 11, 27%) and primary tumours (15 out of 78, 19%). Breast cancer cell lines express various Wnt ligands, and overexpression of SFRPs inhibited cancer cell growth. In addition, overexpression of a β-catenin mutant and depletion of SFRP1 using small interfering RNA synergistically upregulated transcriptional activity of T-cell factor/lymphocyte enhancer factor. Our results confirm the frequent methylation and silencing of Wnt antagonist genes in breast cancer, and suggest that their loss of function contributes to activation of Wnt signalling in breast carcinogenesis.

p21(WAF1/CIP1) induction by 5-azacytosine nucleosides requires DNA damage

Decitabine (DAC) and 5-azacitidine have recently been approved for the treatment of myelodysplastic syndrome. The pharmacodynamic effects of DAC and 5-azacitidine outside their known activity as inhibitors of DNA methyltransferases (DNMTs) require further investigation. The purpose of this study was to investigate the effect of DAC on the expression of p21(WAF1/CIP1), a gene with a putative CpG island surrounding its promoter region. Promoter methylation analysis of p21(WAF1/CIP1) in leukemia cells revealed the absence of CpG methylation. However, DAC upregulated p21(WAF1/CIP1) expression in a dose-dependent manner (ED(50)=103.34 nM) and induced G2/M cell cycle arrest in leukemia cells. Sequential application of DAC followed by different histone deacetylase inhibitors induced expression of p21(WAF1/CIP1) synergistically. Upregulation of p21(WAF1/CIP1) paralleled DAC-induced apoptosis (ED(50)=153 nM). Low doses of DAC induced gamma-H2AX expression (ED(50)=16.5 nM) and upregulated p21(WAF1/CIP1) in congenic HCT 116 colon cancer cells in a DNMT-independent and p53-dependent fashion. Inhibition of p53 transactivation by pifithrin-alpha or the kinase activity of ATM by either the specific ATM inhibitor KU-5593 or caffeine abrogated p21(WAF1/CIP1) upregulation, indicating that DAC upregulation of p21(WAF1/CIP1) was p53- and ATM-dependent in leukemia cells. In conclusion, DAC upregulates p21(WAF1/CIP1) in DNMT-independent manner via the DNA damage/ATM/p53 axis.

Frequent epigenetic silencing of the bone morphogenetic protein 2 gene through methylation in gastric carcinomas

Recently, it was reported that exogenous bone morphogenetic protein (BMP)-2 acted as an antiproliferative agent in a variety of cell lines, including normal and cancerous gastric cell lines, indicating that BMP-2 plays an important role during cell growth. However, despite the loss of BMP-2 expression in several cancers, the underlying mechanism remains unknown. Epigenetic silencing through DNA methylation is one of the key steps during carcinogenesis. In this study, we found, through analysis by the methylation-specific polymerase chain reaction technique, CpG island methylation of the BMP-2 promoter region in gastric and colon cancer cell lines. BMP-2 mRNA was found to be activated after 5-aza-2'-deoxycytidine treatment of the methylation-positive cells. Moreover, 24 of the 56 (42.9%) gastric cancer tissues exhibited promoter methylation. Immunohistochemical staining revealed that 18 of the 24 (75%) gastric cancer tissues without methylation signals exhibited BMP-2 expression, whereas among 20 cancer tissues with strong methylation signals only four (20%) expressed BMP-2 (P = 0.0003). These findings indicate that BMP-2 methylation is strongly associated with the loss of BMP-2 protein expression in the primary gastric carcinomas. BMP-2 methylation was more often observed in diffuse type (60.7%) than in intestinal type (25%) gastric carcinomas (P = 0.007). Thus, aberrant BMP-2 methylation and the resultant loss of BMP-2 expression may be related to gastric carcinogenesis, particularly in the diffuse type.

Aberrant gene silencing in tumor progression: implications for control of cancer

Although it is clear that genetic alterations are critical for the initiation and maintenance of human cancer, it is also becoming evident that epigenetic changes may be essential for the development of these diseases as well. The best studied of these latter processes is heritable transcriptional repression of genes associated with aberrant DNA hypermethylation of their promoters. Herein we review how very early occurrence of these gene silencing events may contribute to loss of key gene functions which result in disruption of cell regulatory pathways that may contribute to abnormal cell population expansion. These altered regulatory events may then provide a setting where mutations in the same disrupted pathways may be readily selected and serve to lock tumor progression into place. This hypothesis has potential impact on means to prevent and control cancer and for the use of epigenetic markers for cancer risk assessment and early diagnosis.

DNA methylation patterns in hereditary human cancers mimic sporadic tumorigenesis

Cancer cells have aberrant patterns of DNA methylation including hypermethylation of gene promoter CpG islands and global demethylation of the genome. Genes that cause familial cancer, as well as other genes, can be silenced by promoter hypermethylation in sporadic tumors, but the methylation of these genes in tumors from kindreds with inherited cancer syndromes has not been well characterized. Here, we examine CpG island methylation of 10 genes (hMLH1, BRCA1, APC, LKB1, CDH1, p16(INK4a), p14(ARF), MGMT, GSTP1 and RARbeta2) and 5-methylcytosine DNA content, in inherited (n = 342) and non-inherited (n = 215) breast and colorectal cancers. Our results show that singly retained alleles of germline mutated genes are never hypermethylated in inherited tumors. However, this epigenetic change is a frequent second "hit", associated with the wild-type copy of these genes in inherited tumors where both alleles are retained. Global hypomethylation was similar between sporadic and hereditary cases, but distinct differences existed in patterns of methylation at non-familial genes. This study demonstrates that hereditary cancers "mimic" the DNA methylation patterns present in the sporadic tumors.

Dnmt3a and Dnmt3b are transcriptional repressors that exhibit unique localization properties to heterochromatin

We demonstrate that the recently identified DNA methyltransferases, Dnmt3a and Dnmt3b, like DNMT1, repress transcription in a methylation-independent manner. Dnmt3a and Dnmt3b repress transcription primarily through a plant homeodomain-like motif that is shared with the ATRX protein but is not present in DNMT1. Unlike DNMT1, which localizes to replication foci during S-phase in murine embryonic fibroblasts, Dnmt3a co-localizes with heterochromatin protein 1 alpha (HP1 alpha) and methyl-CpG binding proteins throughout the cell cycle to late-replicating pericentromeric heterochromatin. In contrast to Dnmt3a, Dnmt3b remained diffuse in the nucleus of embryonic fibroblasts at all cell cycle stages. However, Dnmt3a and Dnmt3b co-localize to these pericentromeric heterochromatin regions in murine embryonic stem cells. This finding is important to the fact that mutations in DNMT3B are found in the developmental syndrome, ICF (immunodeficiency, centromeric heterochromatin instability, and facial anomalies), which involves extensive loss of methylation from pericentromeric regions. The localization of Dnmt3a and Dnmt3b was unaffected in Dnmt1 null embryonic stem cells, which lose the majority of methylation at pericentromeric major satellite repeats, suggesting that these enzymes are not dependent upon preexisting methylation for their targeting. DNMT1 is then positioned to reestablish transcriptionally repressive chromatin as cells replicate, while Dnmt3a and Dnmt3b may help to establish such chromatin in late S-phase and maintain this repressive heterochromatin throughout the cell cycle in a developmentally and/or cell type manner.

Promoter hypermethylation of the DNA repair gene O(6)-methylguanine-DNA methyltransferase is associated with the presence of G:C to A:T transition mutations in p53 in human colorectal tumorigenesis

Defects in DNA repair may be responsible for the genesis of mutations in key genes in cancer cells. The tumor suppressor gene p53 is commonly mutated in human cancer by missense point mutations, most of them G:C to A:T transitions. A recognized cause for this type of change is spontaneous deamination of the methylcytosine. However, the persistence of a premutagenic O(6)-methylguanine can also be invoked. This last lesion is removed in the normal cell by the DNA repair enzyme O(6)-methylguanine-DNA methyltransferase (MGMT). In many tumor types, epigenetic silencing of MGMT by promoter hypermethylation has been demonstrated and linked to the appearance of G to A mutations in the K-ras oncogene in colorectal tumors. To study the relevance of defective MGMT function by aberrant methylation in relation to the presence of p53 mutations, we studied 314 colorectal tumors for MGMT promoter hypermethylation and p53 mutational spectrum. Inactivation of MGMT by aberrant methylation was associated with the appearance of G:C to A:T transition mutations at p53 (Fischer's exact test, two-tailed; P = 0.01). Overall, MGMT methylated tumors displayed p53 transition mutations in 43 of 126 (34%) cases, whereas MGMT unmethylated tumors only showed G:C to A:T changes in 37 of 188 (19%) tumors. A more striking association was found in G:C to A:T transitions in non-CpG dinucleotides; 71% (12 of 17) of the total non-CpG transition mutations in p53 were observed in MGMT aberrantly methylated tumors (Fischer's exact test, two-tailed; P = 0.008). Our data suggest that epigenetic silencing of MGMT by promoter hypermethylation may lead to G:C to A:T transition mutations in p53.

DNA methylation, chromatin inheritance, and cancer

Cancer is a process driven by the accumulation of abnormalities in gene function. While many of these changes are genetic, epigenetically mediated changes in gene expression are being increasingly appreciated. This latter process emphasizes the need to understand two key components of heritable, but reversible, modulation of gene promoter function that are closely tied to one another - formation of chromatin which modulates transcription and establishing patterns of DNA methylation. The link lies first in the recruitment to methylated cytosines of a family of methyl-CpG binding domain proteins (MBDs), which are direct transcriptional repressors and can complex with transcriptional corepressors including histone deacetylases (HDACs). Additionally, the proteins that catalyze DNA methylation, DNA methyltransferases (DNMTs), also directly repress transcription and associate with HDACs. Regulation of these above chromatin-DNA methylation interactions as a function of DNA replication timing is emerging as a key event in the inheritance of transcriptionally repressed domains of the genome. Importantly, synergy between HDAC activity and DNA methylation is operative for a key epigenetic abnormality in cancer cells, transcriptional silencing of tumor suppressor genes. This change has now been recognized for genes that are essential for normal regulation of virtually every major cell function including cell growth, differentiation, apoptosis, DNA repair, and cell-cell, cell-substratum interaction. Understanding the molecular determinants of both normal and abnormal patterns of chromatin formation and DNA methylation thus holds great promise for our understanding of cancer and for means to better diagnose, prevent, and treat this disease.

A gene hypermethylation profile of human cancer

We are in an era where the potential exists for deriving comprehensive profiles of DNA alterations characterizing each form of human cancer. Such profiles would provide invaluable insight into mechanisms underlying the evolution of each tumor type and will provide molecular markers, which could radically improve cancer detection. To date, no one type of DNA change has been defined which accomplishes this purpose. Herein, by using a candidate gene approach, we show that one category of DNA alteration, aberrant methylation of gene promoter regions, can enormously contribute to the above goals. We have now analyzed a series of promoter hypermethylation changes in 12 genes (p16(INK4a), p15(INK4b), p14(ARF), p73, APC,(5) BRCA1, hMLH1, GSTP1, MGMT, CDH1, TIMP3, and DAPK), each rigorously characterized for association with abnormal gene silencing in cancer, in DNA from over 600 primary tumor samples representing 15 major tumor types. The genes play known important roles in processes encompassing tumor suppression, cell cycle regulation, apoptosis, DNA repair, and metastastic potential. A unique profile of promoter hypermethylation exists for each human cancer in which some gene changes are shared and others are cancer-type specific. The hypermethylation of the genes occurs independently to the extent that a panel of three to four markers defines an abnormality in 70-90% of each cancer type. Our results provide an unusual view of the pervasiveness of DNA alterations, in this case an epigenetic change, in human cancer and a powerful set of markers to outline the disruption of critical pathways in tumorigenesis and for derivation of sensitive molecular detection strategies for virtually every human tumor type.

Aberrant patterns of DNA methylation, chromatin formation and gene expression in cancer

Gene function in cancer can be disrupted either through genetic alterations, which directly mutate or delete genes, or epigenetic alterations, which alter the heritable state of gene expression. The latter events are mediated by formation of transcriptionally repressive chromatin states around gene transcription start sites and an associated gain of methylation in normally unmethylated CpG islands in these regions. The genes affected include over half of the tumor suppressor genes that cause familial cancers when mutated in the germline; the selective advantage for genetic and epigenetic dysfunction in these genes is very similar. The aberrant methylation can begin very early in tumor progression and mediate most of the important pathway abnormalities in cancer including loss of cell cycle control, altered function of transcription factors, altered receptor function, disruption of normal cell-cell and cell-substratum interaction, inactivation of signal transduction pathways, loss of apoptotic signals and genetic instability. The active role of the aberrant methylation in transcriptional silencing of genes is becoming increasingly understood and involves a synergy between the methylation and histone deacetylase (HDAC) activity. This synergy can be mediated directly by HDAC interaction with DNA methylating enzymes and by recruitment through complexes involving methyl-cytosine binding proteins. In the translational arena, the promoter hypermethylation changes hold great promise as DNA tumor markers and their potentially reversible state creates a target for cancer therapeutic strategies involving gene reactivation.

p14ARF silencing by promoter hypermethylation mediates abnormal intracellular localization of MDM2

The INK4a/ARF locus encodes two distinct tumor suppressors, p16INK4a and p14ARF. Although the contribution of p16INK4a to human tumorigenesis through point mutation, deletion, and hypermethylation has been widely documented, little is known about specific p14ARF lesions and their consequences. Recent data indicate that p14ARF suffers inactivation by promoter hypermethylation in colorectal cancer cells. Because it is known that p14ARF prevents MDM2 nucleocytoplasmic shuttling and thus stabilizes p53 by attenuating MDM2-mediated degradation, we studied the relationship of p14ARF epigenetic silencing to the expression and localization of MDM2 and p53. Cancer cell lines with an unmethylated p14ARF promoter showed strong nuclear expression of MDM2, whereas in a colorectal cell line with p14ARF hypermethylation-associated inactivation, MDM2 protein was also seen in the cytosol. Treatment with the demethylating agent 5-aza-2'-deoxycytidine was able to reinternalize MDM2 to the nucleus, and p53 expression was restored. No apparent changes in retinoblastoma localization were observed. We also studied the profile of p14ARF promoter hypermethylation in an extensive collection of 559 human primary tumors of different cell types, observing that in colorectal, gastric, renal, esophageal, and endometrial neoplasms and gliomas, aberrant methylation of p14ARF was a relatively common epigenetic event. MDM2 expression patterns revealed that lack of p14ARF promoter hypermethylation was associated with tumors showing exclusive nuclear MDM2 staining, whereas MDM2 cytosolic staining was frequently observed in neoplasms with aberrant p14ARF methylation. Taken together, these data support that epigenetic silencing of p14ARF by promoter hypermethylation is a key mechanism in the disturbance of the MDM2 nuclear localization in human cancer.

Notch signaling induces cell cycle arrest in small cell lung cancer cells

Among the various forms of human lung cancer, small cell lung cancer (SCLC) exhibits a characteristic neuroendocrine (NE) phenotype. Neural and NE differentiation in SCLC depend, in part, on the action of the basic-helix-loop-helix (bHLH) transcription factor human achaete-scute homologue-1 (hASH1). In nervous system development, the Notch signaling pathway is a critical negative regulator of bHLH factors, including hASH1, controlling cell fate commitment and differentiation. To characterize Notch pathway function in SCLC, we explored the consequences of constitutively active Notch signaling in cultured SCLC cells. Recombinant adenoviruses were used to overexpress active forms of Notch1, Notch2, or the Notch effector protein human hairy enhancer of split-1 (HES1) in DMS53 and NCI-H209 SCLC cells. Notch proteins, but not HES1 or control adenoviruses, caused a profound growth arrest, associated with a G1 cell cycle block. We found up-regulation of p21(waf1/cip1) and p27kip1 in concert with the cell cycle changes. Active Notch proteins also led to dramatic reduction in hASH1 expression, as well as marked activation of phosphorylated extracellular signal-regulated kinase (ERK)1 and ERK2, findings that have been shown to be associated with cell cycle arrest in SCLC cells. These data suggest that the previously described function of Notch proteins as proto-oncogenes is highly context-dependent. Notch activation, in the setting of a highly proliferative hASH1-dependent NE neoplasm, can be associated with growth arrest and apparent reduction in neoplastic potential.

Sequence-specific DNA binding activity of RNA helicase A to the p16INK4a promoter

p16(INK4a) is frequently altered in human cancer, often through epigenetically mediated transcriptional silencing. However, little is known about the transcriptional regulation of this gene. To learn more about such control, we initiated studies of proteins that bind to the promoter in cancer cells that do, and do not, express the gene. We identify RNA helicase A (RHA) as a protein that binds much better to the p16(INK4a) promoter in the expressing cells. RHA has not previously been characterized to manifest sequence-specific DNA interaction but does so to the sequence 5' CGG ACC GCG TGC GC 3' in the p16(INK4a) promoter. The Drosophila homologue to RHA, maleless (Mle), functions in the fly for 2-fold activation of male X-chromosome genes. In our experimental setting, RHA induces a similar modest up-regulation of the p16(INK4a) promoter that is dependent upon its sequence-specific interaction. Mle colocalizes with hyperacetylated H4Ac16 on the X-chromosome and some autosomal loci. The decreased binding of RHA to p16(INK4a) in our cells, where the gene is transcriptionally inactive, is associated with decreased amounts of RHA that immunoprecipitate with acetylated lysine antibodies. Finally, we show RHA to be a cellular substrate for caspase-3, which decreases its sequence-specific binding to p16(INK4a) by cleavage of the N terminus. Thus, we have identified a new protein interaction with the p16(INK4a) promoter that involves an important protein for transcriptional modulation. This interaction is decreased in cancer cells, where this gene is aberrantly transcriptionally silent.

Inactivation of the DNA-repair gene MGMT and the clinical response of gliomas to alkylating agents

BACKGROUND

The DNA-repair enzyme O6-methylguanine-DNA methyltransferase (MGMT) inhibits the killing of tumor cells by alkylating agents. MGMT activity is controlled by a promoter; methylation of the promoter silences the gene in cancer, and the cells no longer produce MGMT. We examined gliomas to determine whether methylation of the MGMT promoter is related to the responsiveness of the tumor to alkylating agents.

METHODS

We analyzed the MGMT promoter in tumor DNA by a methylation-specific polymerase-chain-reaction assay. The gliomas were obtained from patients who had been treated with carmustine (1,3-bis(2-chloroethyl)-1-nitrosourea, or BCNU). The molecular data were correlated with the clinical outcome.

RESULTS

The MGMT promoter was methylated in gliomas from 19 of 47 patients (40 percent). This finding was associated with regression of the tumor and prolonged overall and disease-free survival. It was an independent and stronger prognostic factor than age, stage, tumor grade, or performance status.

CONCLUSIONS

Methylation of the MGMT promoter in gliomas is a useful predictor of the responsiveness of the tumors to alkylating agents.

E-cadherin expression is silenced by 5' CpG island methylation in acute leukemia

E-Cadherin is a transmembrane glycoprotein that mediates Ca2+-dependent intercellular adhesion in normal epithelium. In tumors of epithelial origin, E-cadherin expression frequently is reduced, an event that contributes to tumor invasion and metastasis. The role of E-cadherin in hematopoietic tissues is less clear. In normal bone marrow, E-cadherin is expressed on erythroid progenitors, CD34+ stem cells, and stromal cells, where it likely contributes to intercellular interactions during hematopoiesis. In this study, we used a nested-PCR approach to examine the methylation status of the E-cadherin 5' CpG island in blood and bone marrow samples from normal donors and in bone marrow from patients with acute leukemia. In normal peripheral blood mononuclear cells and bone marrow, E-cadherin was completely unmethylated. In peripheral blood mononuclear cells, expression was evident by reverse transcription-PCR. Immunoblotting confirmed E-cadherin protein expression in two lymphoblastoid cell lines derived from normal donors. In contrast, E-cadherin was aberrantly methylated in 4 of 4 (100%) leukemia cell lines, 14 of 44 (32%) acute myelogenous leukemias, and 18 of 33 (53%) acute lymphoblastic leukemias. Genomic bisulfite sequencing of primary leukemias confirmed dense methylation across the CpG island. Methylation was associated with loss of E-cadherin RNA and protein in leukemia cell lines and primary leukemias. Following treatment with 5-aza-2'-deoxycytidine, a methylated leukemia cell line expressed both E-cadherin transcript and protein. Our results show that methylation of E-cadherin occurs commonly in acute leukemia and suggests a hypothesis for E-cadherin down-regulation in leukemogenesis.

Predicting lung cancer by detecting aberrant promoter methylation in sputum

Despite the promise of using DNA markers for the early detection of cancer, none has proven universally applicable to the most common and lethal forms of human malignancy. Lung carcinoma, the leading cause of tumor-related death, is a key example of a cancer for which mortality could be greatly reduced through the development of sensitive molecular markers detectable at the earliest stages of disease. By increasing the sensitivity of a PCR approach to detect methylated DNA sequences, we now demonstrate that aberrant methylation of the p16 and/or O6-methyl-guanine-DNA methyltransferase promoters can be detected in DNA from sputum in 100% of patients with squamous cell lung carcinoma up to 3 years before clinical diagnosis. Moreover, the prevalence of these markers in sputum from cancer-free, high-risk subjects approximates lifetime risk for lung cancer. The use of aberrant gene methylation as a molecular marker system seems to offer a potentially powerful approach to population-based screening for the detection of lung cancer, and possibly the other common forms of human cancer.

Analysis of adenomatous polyposis coli promoter hypermethylation in human cancer

Germ-line mutations in the tumor suppressor gene APC are associated with hereditary familial adenomatous polyposis (FAP), and somatic mutations are common in sporadic colorectal tumors. We now report that methylation in the promoter region of this gene constitutes an alternative mechanism for gene inactivation in colon and other tumors of the gastrointestinal tract. The APC promoter is hypermethylated in 18% of primary sporadic colorectal carcinomas (n = 108) and adenoma (n = 48), and neoplasia with APC methylation fails to express the APC transcript. Methylation affects only wild-type APC in 95% of cases and is not observed in tumors from FAP patients who have germ-line APC mutations. As with APC mutation, aberrant APC methylation occurs early in colorectal carcinogenesis. When other tumor types are analyzed (n = 208), methylation of the APC promoter is not restricted to the colon but is present in tumors originating elsewhere in the gastrointestinal tract but rarely in other tumors. Our data suggest that hypermethylation of APC provides an important mechanism for impairing APC function and further underscores the importance of the APC pathway in gastrointestinal tumorigenesis.

Constitutive achaete-scute homologue-1 promotes airway dysplasia and lung neuroendocrine tumors in transgenic mice

The transcription factor achaete-scute homologue-1 (ASH1) is essential for neural differentiation during fetal development and is a cardinal feature of neuroendocrine (NE) tumors such as small cell lung cancer. To explore the potential of ASH1 to promote NE differentiation and tumorigenesis in the lung, we constitutively expressed the factor in nonendocrine airway epithelial cells using transgenic mice. Progressive airway hyperplasia and metaplasia developed beginning at 3 weeks of life. ASH1 potently enhanced the tumorigenic effect of SV40 large T antigen in airway epithelium. These doubly transgenic animals developed massive NE lung tumors, implying that ASH1 may cooperate with defects in p53, pRb, or related pathways in promoting NE lung carcinogenesis.

DNMT1 binds HDAC2 and a new co-repressor, DMAP1, to form a complex at replication foci

DNA methylation can contribute to transcriptional silencing through several transcriptionally repressive complexes, which include methyl-CpG binding domain proteins (MBDs) and histone deacetylases (HDACs). We show here that the chief enzyme that maintains mammalian DNA methylation, DNMT1, can also establish a repressive transcription complex. The non-catalytic amino terminus of DNMT1 binds to HDAC2 and a new protein, DMAP1 (for DNMT1 associated protein), and can mediate transcriptional repression. DMAP1 has intrinsic transcription repressive activity, and binds to the transcriptional co-repressor TSG101. DMAP1 is targeted to replication foci through interaction with the far N terminus of DNMT1 throughout S phase, whereas HDAC2 joins DNMT1 and DMAP1 only during late S phase, providing a platform for how histones may become deacetylated in heterochromatin following replication. Thus, DNMT1 not only maintains DNA methylation, but also may directly target, in a heritable manner, transcriptionally repressive chromatin to the genome during DNA replication.

Promoter-region hypermethylation and gene silencing in human cancer

In summary, it is apparent that alterations in DNA methylation are a fundamental molecular change associated with the neoplastic process and have important biologic implications for tumor initiation and progression. The promoter-region hypermethylation events covered in the present chapter are especially critical and can frequently serve as alternative mechanisms for coding-region mutations for loss of key gene function in neoplastic cells. The mechanisms underlying the precise role of this hypermethylation in gene silencing must be further defined, as must the determinants of the hypermethylation changes themselves. The therapeutic implications of promoter-region hypermethylation must be explored, and a potential use for establishing this change as a sensitive biomarker for use in multiple types of cancer-risk assessment and detection assays has already emerged. The next few years should see exciting advances in our understanding of an epigenetic process which, in conjunction with genetic alterations, appears to drive the process of neoplasia.

Cloning and chromosomal localization of the human BARX2 homeobox protein gene

The human BARX2 gene encodes a homeodomain-containing protein of 254 amino acids, which binds optimally to the DNA consensus sequence YYTAATGRTTTTY. BARX2 is highly expressed in adult salivary gland and is expressed at lower levels in other tissues, including mammary gland, kidney, and placenta. The BARX2 gene consists of four exons, and is located on human chromosome 11q25. This chromosomal location is within the minimal deletion region for Jacobsen syndrome, a syndrome including craniosynostosis and other developmental abnormalities. This chromosomal location, along with the reported expression of murine barx2 in craniofacial development, suggests that BARX2 may be causally involved in the craniofacial abnormalities in Jacobsen syndrome.

Inactivation of the DNA repair gene O6-methylguanine-DNA methyltransferase by promoter hypermethylation is associated with G to A mutations in K-ras in colorectal tumorigenesis

O6-methylguanine DNA methyltransferase (MGMT) is a DNA repair protein that removes mutagenic and cytotoxic adducts from the O6 position of guanine. O6-methylguanine mispairs with thymine during replication, and if the adduct is not removed, this results in conversion from a guanine-cytosine pair to an adenine-thymine pair. In vitro assays show that MGMT expression avoids G to A mutations and MGMT transgenic mice are protected against G to A transitions at ras genes. We have recently demonstrated that the MGMT gene is silenced by promoter methylation in many human tumors, including colorectal carcinomas. To study the relevance of defective MGMT function by aberrant methylation in relation to the presence of K-ras mutations, we studied 244 colorectal tumor samples for MGMT promoter hypermethylation and K-ras mutational status. Our results show a clear association between the inactivation of MGMT by promoter hypermethylation and the appearance of G to A mutations at K-ras: 71% (36 of 51) of the tumors displaying this particular type of mutation had abnormal MGMT methylation, whereas only 32% (12 of 37) of those with other K-ras mutations not involving G to A transitions and 35% (55 of 156) of the tumors without K-ras mutations demonstrated MGMT methylation (P = 0.002). In addition, MGMT loss associated with hypermethylation was observed in the small adenomas, including those that do not yet contain K-ras mutations. Hypermethylation of other genes such as p16INK4a and p14ARF was not associated with either MGMT hypermethylation or K-ras mutation. Our data suggest that epigenetic silencing of MGMT by promoter hypermethylation may lead to a particular genetic change in human cancer, specifically G to A transitions in the K-ras oncogene.

CpG methylation is maintained in human cancer cells lacking DNMT1

Hypermethylation is associated with the silencing of tumour susceptibility genes in several forms of cancer; however, the mechanisms responsible for this aberrant methylation are poorly understood. The prototypic DNA methyltransferase, DNMT1, has been widely assumed to be responsible for most of the methylation of the human genome, including the abnormal methylation found in cancers. To test this hypothesis, we disrupted the DNMT1 gene through homologous recombination in human colorectal carcinoma cells. Here we show that cells lacking DNMT1 exhibited markedly decreased cellular DNA methyltransferase activity, but there was only a 20% decrease in overall genomic methylation. Although juxtacentromeric satellites became significantly demethylated, most of the loci that we analysed, including the tumour suppressor gene p16INK4a, remained fully methylated and silenced. These results indicate that DNMT1 has an unsuspected degree of regional specificity in human cells and that methylating activities other than DNMT1 can maintain the methylation of most of the genome.

Promoter hypermethylation and BRCA1 inactivation in sporadic breast and ovarian tumors

BACKGROUND

Inherited mutations in the BRCA1 gene may be responsible for almost half of inherited breast carcinomas. However, somatic (acquired) mutations in BRCA1 have not been reported, despite frequent loss of heterozygosity (LOH or loss of one copy of the gene) at the BRCA1 locus and loss of BRCA1 protein in tumors. To address whether BRCA1 may be inactivated by pathways other than mutations in sporadic tumors, we analyzed the role of hypermethylation of the gene's promoter region.

METHODS

Methylation patterns in the BRCA1 promoter were assessed in breast cancer cell lines, xenografts, and 215 primary breast and ovarian carcinomas by methylation-specific polymerase chain reaction (PCR). BRCA1 RNA expression was determined in cell lines and seven xenografts by reverse transcription-PCR. P values are two-sided.

RESULTS

The BRCA1 promoter was found to be unmethylated in all normal tissues and cancer cell lines tested. However, BRCA1 promoter hypermethylation was present in two breast cancer xenografts, both of which had loss of the BRCA1 transcript. BRCA1 promoter hypermethylation was present in 11 (13%) of 84 unselected primary breast carcinomas. BRCA1 methylation was strikingly associated with the medullary (67% methylated; P =.0002 versus ductal) and mucinous (55% methylated; P =.0033 versus ductal) subtypes, which are overrepresented in BRCA1 families. In a second series of 66 ductal breast tumors informative for LOH, nine (20%) of 45 tumors with LOH had BRCA1 hypermethylation, while one (5%) of 21 without LOH was methylated (P =.15). In ovarian neoplasms, BRCA1 methylation was found only in tumors with LOH, four (31%) of 13 versus none of 18 without LOH (P =.02). The BRCA1 promoter was unmethylated in other tumor types.

CONCLUSION

Silencing of the BRCA1 gene by promoter hypermethylation occurs in primary breast and ovarian carcinomas, especially in the presence of LOH and in specific histopathologic subgroups. These findings support a role for this tumor suppressor gene in sporadic breast and ovarian tumorigenesis.

DNA methyltransferase levels and altered CpG methylation in the total genome and in the GSTP1 gene in human glioma cells transfected with sense and antisense DNA methyltransferase cDNA

This study examines the efficacy of using plasmid expression vectors containing sense and antisense DNA MTase cDNA to both up- and downregulate intracellular DNA MTase levels in human glioma cells. The effects of the changes in MTase levels on global genomic DNA methylation and on the methylation status of CpG dinucleotides in the GSTP1 gene were determined in a glioma cell line that overexpresses the GSTP1 gene. In cells transfected with sense DNA MTase cDNA, MTase gene transcripts increased to a maximum of 2. 5-fold at 24 h, while MTase activity increased to a maximum of 3. 6-fold at 48 h. The effects of antisense MTase cDNA transfections were less pronounced, and levels of MTase gene transcripts and enzyme activity in transfectants were decreased to only, approximately, one-half the levels of controls. The alterations in DNA MTase expression were associated with corresponding changes in the level of global DNA methylation and in the methylation of the GSTP1 gene in the cells, however, with no detectable morphological or cytotoxic effects on the cells. No significant changes in GSTP1 gene expression were detected after the transfections, presumably because of the high levels of basal GSTP1 expression in the cells. Consequently, the p16 gene, known to be repressed transcriptionally by DNA methylation, was examined for the functional effects of the altered MTase levels. The results showed a 2-fold decrease in p16 gene transcripts with the sense MTase transfectants, while in the MTase antisense-transfected cells p16 transcript levels increased by 30%. Together, these results demonstrate the feasibility of using both sense and antisense DNA MTase expression vectors to regulate DNA MTase levels in glioma cells and that, over relatively short periods of time, the alterations in MTase activities are not deleterious to the cells. The system provides a model with which the role of DNA methylation in critical genes and DNA sequences can be investigated in glioma cells.

DNA hypermethylation in tumorigenesis: epigenetics joins genetics

Recently, the concept that epigenetic, as well as genetic, events might be central to the evolution of human cancer is re-emerging. Cancers often exhibit an aberrant methylation of gene promoter regions that is associated with loss of gene function. This DNA change constitutes a heritable state, not mediated by altered nucleotide sequence, that appears to be tightly linked to the formation of transcriptionally repressive chromatin. This epigenetic process acts as an alternative to mutations to disrupt tumor-suppressor gene function and can predispose to genetic alterations through inactivating DNA-repair genes. Dissecting the molecular processes that mediate these methylation changes will enhance our understanding of chromatin modeling and gene regulation and might present novel possibilities for cancer therapy. Methylation changes constitute potentially sensitive molecular markers to define risk states, monitor prevention strategies, achieve early diagnosis, and track the prognosis of cancer.

Mice deficient in the candidate tumor suppressor gene Hic1 exhibit developmental defects of structures affected in the Miller-Dieker syndrome

HIC1 is a candidate tumor suppressor gene which is frequently hypermethylated in human tumors, and its location within the Miller-Dieker syndrome's critical deletion region at chromosome 17p13.3 makes it a candidate gene for involvement in this gene deletion syndrome. To study the function of murine Hic1 in development, we have created Hic1 -deficient mice. These animals die perinatally and exhibit varying combinations of gross developmental defects throughout the second half of development, including acrania, exencephaly, cleft palate, limb abnormalities and omphalocele. These findings demonstrate a role for Hic1 in the development of structures affected in the Miller-Dieker syndrome, and provide functional evidence to strengthen its candidacy as a gene involved in this disorder.

A bird's eye view of global methylation

Restriction landmark genomic scanning applied to a broad variety of cancer types can disclose tumour-specific and tumour-type-specific global methylation profiles. This and other genome-scanning approaches allows the rapid analysis of methylation profiles of thousands of genes in parallel-and promises to identify new genes critical to carcinogenesis and other biological processes.

Methylation patterns of the E-cadherin 5' CpG island are unstable and reflect the dynamic, heterogeneous loss of E-cadherin expression during metastatic progression

Metastatic progression of most common epithelial tumors involves a heterogeneous, transient loss of expression of the homotypic cell adhesion protein, E-cadherin, rather than the uniform loss of a functional protein resulting from coding region mutation. Indeed, whereas E-cadherin loss may promote invasion, reexpression may facilitate cell survival within metastatic deposits. The mechanisms underlying such plasticity are unclear. We now show that the heterogeneous loss of E-cadherin expression in primary human breast cancers reflects a heterogeneous pattern of promoter region methylation, which begins early prior to invasion. In cultured human tumor cells, such heterogeneous methylation is dynamic, varying from allele to allele and shifting in relation to the tumor microenvironment. Following invasion in vitro, which favors diminished E-cadherin expression, the density of promoter methylation markedly increased. When these cells were cultured as spheroids, which requires homotypic cell adhesion, promoter methylation decreased dramatically, and E-cadherin was reexpressed. These data show that the methylation associated with E-cadherin loss in human breast cancer is heterogeneous and unstable and suggest that such epigenetic plasticity may contribute to the dynamic, phenotypic heterogeneity that drives metastatic progression.

Epigenetic inactivation of LKB1 in primary tumors associated with the Peutz-Jeghers syndrome

Germ-line mutations of the LKB1 gene cause Peutz-Jeghers syndrome (PJS) characterized by mucocutaneous pigmentation, predisposition to benign hamartomas of the gastrointestinal tract and also to several types of tumors. However, somatic mutations of this gene are very rare. To examine inactivation of LKB1 by epigenetic mechanisms, we investigated a series of primary tumors and cancer cell lines, for hypermethylation affecting the CpG island located in the 5' region of the LKB1 gene using Methylation-specific PCR (MSP). First, we screened 51 cancer cell lines. Only three colorectal and one cervical carcinoma cell lines were methylated at LKB1, and loss of the LKB1 transcript was demonstrated. Treatment with the demethylating agent 5-aza-2'-deoxycytidine restored LKB1 expression. To address the incidence of LKB1 epigenetic inactivation in primary tumors, we analysed colorectal, breast, gastric, pancreatic, thyroid, bladder and testicular carcinomas (n=195). Normal tissues from the mentioned organs were unmethylated in this region. Among the described tumors, only one colorectal carcinoma and three testicular tumors displayed LKB1 promoter hypermethylation. Further study of those histological types more commonly associated with PJS, demonstrated that LKB1 promoter hypermethylation was present in five of 11 (45%) papillary breast carcinomas. Finally, in three patients with a strong family story suggestive of PJS disease, abnormal LKB1 methylation was found in four of 22 (18%) hamartomatous polyps lesions. Our findings provide an alternative pathway for inactivation of the LKB1 tumor suppressor gene involving promoter hypermethylation.

Hypermethylation-associated inactivation of p14(ARF) is independent of p16(INK4a) methylation and p53 mutational status

The INK4a/ARF locus encodes two cell cycle-regulatory proteins, p16INK4a andp14ARF, which share an exon using different reading frames. p14ARF antagonizes MDM2-dependent p53 degradation. However, no point mutations in p14ARF not altering p16INK4a have been described in primary tumors. We report that p14ARF is epigenetically inactivated in several colorectal cell lines, and its expression is restored by treatment with demethylating agents. In primary colorectal carcinomas, p14ARF promoter hypermethylation was found in 31 of 110 (28%) of the tumors and observed in 13 of 41 (32%) colorectal adenomas but was not present in any normal tissues. p14ARF methylation appears in the context of an adjacent unmethylated p16INK4a promoter in 16 of 31 (52%) of the carcinomas methylated at p14ARF. Although p14ARF hypermethylation was slightly overrepresented in tumors with wild-type p53 compared to tumors harboring p53 mutations [19 of 55 (34%) versus 12 of 55 (22%)], this difference did not reach statistical significance. p14ARF aberrant methylation was not related to the presence of K-ras mutations. Our results demonstrate that p14ARF promoter hypermethylation is frequent in colorectal cancer and occurs independently of the p16INK4a methylation status and only marginally in relation to the p53 mutational status.

Aberrant methylation in gastric cancer associated with the CpG island methylator phenotype

Aberrant methylation of 5' CpG islands is thought to play an important role in the inactivation of tumor suppressor genes in cancer. In colorectal cancer, a group of tumors is characterized by a hypermethylator phenotype termed CpG island methylator phenotype (CIMP), which includes methylation of such genes as p16 and hMLH1. To study whether CIMP is present in gastric cancer, the methylation status of five newly cloned CpG islands was examined in 56 gastric cancers using bisulfite-PCR. Simultaneous methylation of three loci or more was observed in 23 (41%) of 56 cancers, which suggests that these tumors have the hypermethylator phenotype CIMP. There was a significant concordance between CIMP and the methylation of known genes including p16, and hMLH1; methylation of p16 was detected in 16 (70%) of 23 CIMP+ tumors, 1 (8%) of 12 CIMP intermediate tumors, and 1 (5%) of 21 CIMP- tumors (P<0.0001). Methylation of the hMLH1 gene was detected in three of five tumors that showed microsatellite instability, and all three of the cases were CIMP+. The CIMP phenotype is an early event in gastric cancer, being present in the normal tissue adjacent to cancer in 5 of 56 cases. These results suggest that CIMP may be one of the major pathways that contribute to tumorigenesis in gastric cancers.

hMLH1 promoter hypermethylation is an early event in human endometrial tumorigenesis

It has recently been suggested that silencing of the hMLH1 gene by promoter hypermethylation is the mechanism underlying the presence of the microsatellite instability (MSI) phenotype in sporadic colon and endometrial carcinomas. To determine whether hMLH1 promoter hypermethylation is a relatively early event in endometrial tumorigenesis we evaluated endometrial hyperplasia (EH) characterized as simple, complex, and atypical (the direct precursor of endometrial carcinoma) for hMLH1 aberrant methylation. In addition, we studied the hMLH1, hMSH2, hMSH3, and hMSH6 promoter methylation and MSI status of those endometrial carcinomas with synchronous hyperplasias and those without them. We found that 11 of 12 (91%) cases of endometrial carcinoma (EC) displaying MSI had hMLH1 promoter hypermethylation, whereas aberrant methylation of any of the other mismatch repair genes was not observed. All 15 cases of EC without MSI were unmethylated at hMLH1. Abnormal methylation of hMLH1 was also present in 8 of 116 (7%) cases of EH and was restricted primarily to the atypical endometrial hyperplasia (AEH) type with coexisting endometrial carcinoma. In this set, half of EH methylated at hMLH1 displayed MSI, whereas none of the unmethylated EH had MSI. Our data suggest that hypermethylation of hMLH1 can be an early event in the pathogenesis of EC, preceding the development of an apparent MSI phenotype in a subset of cases.

In situ detection of the hypermethylation-induced inactivation of the p16 gene as an early event in oncogenesis

We have developed a technique, methylation-specific PCR in situ hybridization (MSP-ISH), which allows for the methylation status of specific DNA sequences to be visualized in individual cells. We use MSP-ISH to monitor the timing and consequences of aberrant hypermethylation of the p16 tumor suppresser gene during the progression of cancers of the lung and cervix. Hypermethylation of p16 was localized only to the neoplastic cells in both in situ lesions and invasive cancers, and was associated with loss of p16 protein expression. MSP-ISH allowed us to dissect the surprising finding that p16 hypermethylation occurs in cervical carcinoma. This tumor is associated with infection of the oncogenic human papillomavirus, which expresses a protein, E7, that inactivates the retinoblastoma (Rb) protein. Thus, simultaneous Rb and p16 inactivation would not be needed to abrogate the critical cyclin D-Rb pathway. MSP-ISH reveals that p16 hypermethylation occurs heterogeneously within early cervical tumor cell populations that are separate from those expressing viral E7 transcripts. In advanced cervical cancers, the majority of cells have a hypermethylated p16, lack p16 protein, but no longer express E7. These data suggest that p16 inactivation is selected as the most effective mechanism of blocking the cyclin D-Rb pathway during the evolution of an invasive cancer from precursor lesions. These studies demonstrate that MSP-ISH is a powerful approach for studying the dynamics of aberrant methylation of critical tumor suppressor genes during tumor evolution.

p15(INK4B) CpG island methylation in primary acute leukemia is heterogeneous and suggests density as a critical factor for transcriptional silencing

The promoter region of the cyclin-dependent kinase inhibitor p15(INK4B) contains a CpG island that is hypermethylated in many hematologic malignancies. To explore the relationship between patterns of methylation and gene transcription, we used bisulfite genomic sequencing to obtain a detailed analysis of methylation in acute leukemia, leukemia cell lines, and normal lymphocytes. The entire CpG island region of p15 was largely devoid of methylation in normal lymphocytes, but methylation of varying density was found in primary acute leukemia. Methylation density was generally conserved between the alleles from each sample, but marked heterogeneity for the specific CpG sites methylated was observed. Patterns of methylation were compared and expression assessed with reverse-transcriptase polymerase chain reaction (RT-PCR). The density of methylation within the CpG island, and not any specific location, correlates best with transcriptional loss. Leukemias with methylation of approximately 40% of the CpG dinucleotides on each allele had complete gene silencing, with variable, but diminished expression with less dense CpG island methylation. Our results suggest that the transcriptional silencing of p15 in conjunction with aberrant hypermethylation is best understood as an evolutionary process that involves progressively increasing methylation of the entire p15 CpG island.