Molecular mechanisms and therapeutic strategies: DNA methylation in pancreatic cancer

Background There is growing evidence that DNA methylation alterations contribute to carcinogenesis. While cancer tissue exhibits widespread DNA methylation changes, the proportion of tissue-specific versus tissue-independent DNA methylation alterations in cancer is unclear. In addition, it is unknown which factors determine the patterns of aberrant DNA methylation in cancer.ResultsUsing HumanMethylation450 BeadChips (450k), we here analyze genome-wide DNA methylation patterns of ten types of fetal tissue, in addition to matched normal-cancer data for corresponding tissue types, encompassing over 3000 samples. We demonstrate that the level of aberrant cancer DNA methylation in gene promoters and gene bodies is highly correlated between cancer types. We estimate that up to 60 % of the DNA methylation variation in a cancer genome of a given tissue type is explained by the corresponding variation in a cancer genome of another type, implying that much of the cancer DNA methylation landscape is tissue independent. We further show that histone marks in normal cells are better predictors of aberrant cancer DNA methylation than the corresponding signals in human embryonic stem cells. We build predictors of cancer DNA methylation patterns and show that although inclusion of three histone marks (H3K4me3, H3K27me3 and H3K36me3) improves model accuracy, the bivalent marks are the most predictive. Finally, we show that chromatin accessibility of gene promoters in normal tissue dictates the promoter’s propensity to acquire aberrant DNA methylation in cancer in so far as it determines its level of DNA methylation in normal tissue.ConclusionsOur data show that a considerable fraction of the aberrant cancer DNA methylation landscape results from a mechanism that is largely tissue specific. Histone marks as specified in the normal cell of origin provide highly predictive models of aberrant cancer DNA methylation and outperform those derived from the same marks in hESCs.Electronic supplementary materialThe online version of this article (doi:10.1186/s13072-016-0058-4) contains supplementary material, which is available to authorized users.

Chapter 2 - DNA Methylation and Hydroxymethylation in Cancer

Despite the clinical advancement, pancreatic cancer remains one of the most deadly cancers. Pancreatic cancers are often difficult to diagnose and there is no reliable screening biomarker for early detection of this cancer. Epigenetic alterations have emerged as a common hallmark of many types of human cancers as well as pancreatic cancers. Studies have shown that epigenetic dysregulation is associated with multiple steps during carcinogenesis. Therefore, detection of aberrant DNA methylation from blood samples may be a promising diagnostic tool for pancreatic cancers. First, we analyzed KRAS mutation status of 100 samples obtained by Endoscopic ultrasound-guided fine-needle aspiration (EUS-FNA), which is a safe and accurate technique to perform the pancreatic tissue biopsy, and found that more than 80% of the cells in the samples harbors KRAS mutation indicating high amount of cancer cells in the specimens. Next, we performed genome-wide DNA methylation analysis of 38 EUS-FNA specimens using Illumina Infinium HumanMethylation450 BeadChip. We further analyzed the deposited DNA methylation profiles of other tumor types in the public databases, such as TCGA, and identified 8 marker genes with high frequency and specificity of DNA methylation in pancreatic cancers. We validated the DNA methylation status of these genes by bisulfite pyrosequencing and quantitative methylation specific PCR. These markers were frequently methylated in EUS-FNA samples (95% of samples are methylation positive in at least one marker gene). Now we are analyzing the DNA methylation status in pancreatic cancers and corresponding blood samples in order to identify the most effective set of markers for diagnose pancreatic cancers. Our results suggest that analysis of DNA methylation in the circulating free DNA might be a safe and potent tool for evaluating and staging pancreatic cancers. Citation Format: Keiko Shinjo, Fumiharu Ohka, Keisuke Katsushima, Akira Hatanaka, Norihisa Ichimura, Zhao Juan, Kenji Yamao, Yutaka Kondo. DNA methylation in circulating free DNA as a new biomarker for pancreatic cancer. [abstract]. In: Proceedings of the 106th Annual Meeting of the American Association for Cancer Research; 2015 Apr 18-22; Philadelphia, PA. Philadelphia (PA): AACR; Cancer Res 2015;75(15 Suppl):Abstract nr 1069. doi:10.1158/1538-7445.AM2015-1069

Aberrant DNA methylation patterns are a characteristic feature of cancer including myeloid malignancies such as acute myeloid leukemia (AML). The mechanisms behind aberrant DNA methylation have long remained obscure. New genome-wide studies have elucidated the genome and epigenome of solid tumors and AML. Molecular subtypes of AML were found to exhibit highly distinct DNA methylation profiles. Clonal evolution patterns of AML were recently dissected and might shape epigenetic dysregulation. Also, recurrent mutations in epigenetic modifying enzymes were identified in AML and linked to distinct DNA methylation signatures. The genetic background, thus, takes center stage as a driver of epigenetic dysregulation in AML. First mechanistic insights into the dysregulation of DNA methylation by recurrent mutations have already been gained. Other studies suggest that epigenomic plasticity and aging-associated changes in DNA methylation also contribute extensively to aberrant DNA methylation in cancer. Epigenetic dysregulation, therefore, seems to also occur independently of the genetic background. Furthermore, global changes in chromatin conformation and nuclear organization have also been proposed as potential contributors to aberrant DNA methylation. This review will summarize and discuss current concepts regarding the mechanisms behind aberrant DNA methylation in cancer and specifically AML.

Desmoplasia of Pancreatic Ductal Adenocarcinoma

DNA methylation and histone post-translational modifications (PTMs) represent two key epigenetic regulators of DNA information in chromatin. The faithful inheritance of DNA methylation is essential for normal mammalian development and long-term transcriptional silencing, and aberrant DNA methylation patterns are a hallmark of many cancers. Histone PTMs can alter the physical structure of chromatin and have been proposed to function in the context of a “histone code” to coordinate the recruitment of effector proteins that elicit selective effects on chromatin-templated cellular processes like gene transcription and DNA repair. Indeed, defects in histone recognition by effector proteins are also emerging as underlying causes of cancer initiation and progression. Effector proteins often contain multiple chromatin binding domains. While significant progress has been made characterizing individual effector domains, the role of paired domains and how they function in a combinatorial fashion within chromatin is poorly defined. The E3 ubiquitin ligase UHRF1 is a unique chromatin effector protein, in that it integrates the recognition of both histone PTMs and DNA methylation. We recently showed that H3K9 methyl binding by the tandem Tudor domain (TTD) of UHRF1 is required for DNA methylation maintenance in cancer cells — yet the function of its adjacent plant homeodomain (PHD) in this process was unknown. Here we show that the linked TTD and PHD of UHRF1 operate as a single functional unit in cells, providing a defined combinatorial readout of the modification state of the histone H3 tail that is essential for UHRF1-directed DNA methylation maintenance by DNMT1. These findings provide critical support for the ‘histone code' hypothesis, demonstrating that multivalent histone engagement plays a fundamental role in driving a downstream biological event in chromatin and suggest UHRF1 may be a favorable therapeutic target for regulating aberrant DNA methylation in cancers. This abstract is also presented as Poster B26. Citation Format: Scott B. Rothbart, Bradley M. Dickson, Michelle S. Ong, Krzysztof Krajewski, Scott Houliston, Dmitri B. Kireev, Cheryl H. Arrowsmith, Brian D. Strahl. Multivalent histone engagement by the linked tandem Tudor and PHD domains of UHRF1 is required for DNA methylation maintenance. [abstract]. In: Proceedings of the AACR Special Conference on Chromatin and Epigenetics in Cancer; Jun 19-22, 2013; Atlanta, GA. Philadelphia (PA): AACR; Cancer Res 2013;73(13 Suppl):Abstract nr PR03.

Objective To investigate the role of Stathmin in pancreatic cancer invasion and metastasis and its relationship with DNA methylation. Methods Immunohistochemical detection of MBDI and Stathmin protein expression in 40 cases of pancreatic cancer and 15 cases ot normal pancreatic tissue were performed,followed by analysis of their clinical and pathological relationship with pancreatic cancer; Human pancreatic cancer cell line BxPC-3 was treated with 5-Aza-2-dC (AZA).Both qRT-PCR and Western blot analysis of Stathmin expression were used before and after AZA treatment; Stathmin-siRNA transfected BxPC-3 cells were divided into the Stathmi-siRNA group and the empty vector control group.Transwell chamber invasion assay and animal experiment were performed to measure the changes in cell invasion and metastatic capability. Results lmmunohistochemistry showed positive MBDI and Stathmin expressions in 28 (70％) and 24 (60％) out of 40 cases of pancreatic cancer,respectively,which were significantly higher than that in the normal pancreatic tissue (P＜ 0.05); MBDI and Stathmin protein expressions were positively correlated (r =0.356,P =0.037),so were MBDI expression and lymph node metastasis (P=0.023).Stathmin expression was significantly correlated with clinical staging and lymph node metastasis (P =0.002,and P =0.001,respectively).After AZA treatment,both Stathmin mRNA and protein expression in BxPC-3 were significantly decreased.Transwell chamber invasion assay showed that compared with the control group,the cell invasion capability of the Stathmin-siRNA group was significantly decreased (P＜0.05).Animal experiment showed that the incidence of liver metastasis was significantly lower in the Stathmin-siRNA transfected group than the empty vector control group (P＜0.05).Conclusion Demethylation may contribute to the reduction of Stathmin expression in pancreatic cancer and further improve the prognosis of pancreatic cancer patients. Key words: Pancreatic neoplasms; DNA methylation; Neoplasm metastasis; Microtubule proteins

S-adenosylmethionine Levels Regulate the Schwann Cell DNA Methylome

DNA methylation acts in concert with other epigenetic mechanisms to regulate normal gene expression and facilitate chromatin organization within cells. Aberrant DNA methylation patterns are acquired during carcinogenic transformation; such events are often accompanied by alterations in chromatin structure at gene regulatory regions. The expression pattern of any given gene is achieved by interacting epigenetic mechanisms. First, the insertion of nucleosomes at transcriptional start sites prevents the binding of the transcriptional machinery and additional cofactors that initiate gene expression. Second, nucleosomes anchor all of the DNMT3A and DNMT3B methyltransferase proteins in the cell, which suggests a role for histone octamers in the establishment of DNA methylation patterns. During carcinogenesis, epigenetic switching and 5-methylcytosine reprogramming result in the aberrant hypermethylation of CpG islands, reducing epigenetic plasticity of critical developmental and tumor suppressor genes, rendering them unresponsive to normal stimuli. Here, we will discuss the importance of both established and novel molecular concepts that may underlie the role of DNA methylation in cancer.

Introduction: Pancreatic cancer (PC) is a devastating disease, its etiology is not fully understood, and there is a paucity of non-invasive biomarkers for early diagnosis. While DNA methylation (DNAm) holds promise, associations between DNAm and PC risk are poorly understood, especially regarding temporal changes in associations. Therefore, we aimed to evaluate potential associations between DNAm in pre-diagnostic serum on 9 genes (BNIP3, CDKN1C, CLDN4, LCN2, p16, RASSF1, SFN, SPARC, and TFPI-2) and 2 repetitive elements (Alu and LINE-1) and PC risk in a US Department of Defense cohort. Methods : The study population included 82 PC cases, diagnosed 1991 - 2007, and 229 controls individually- matched on age (within 5 years), sex, race, and serum draw time period. Cases had up to 3 pre- diagnostic serum samples. Controls had 1 serum sample collected 0-10 years before their selection date (date of case diagnosis). DNA was extracted from serum and the percentage of cytosines at a given CpG site that were methylated was quantified via Pyrosequencing. For serum samples within a time window (<2, 2-<5, and 5-10 years pre-diagnosis), we computed odds ratios (ORs) and 95% confidence intervals (95% CIs) for the association of each CpG site with PC risk (66 sites in total across all genes and repetitive elements) with conditional logistic regression models. DNAm was evaluated as both a binary variable and a continuous variable. Results : In samples collected 5-10 years before diagnosis, those with no DNAm (0% DNAm) had increased odds of PC compared to those with DNAm (>0% DNAm) in the promoter region of BNIP3 (Chr10:131982293 OR, 95% CI: 5.63, 1.16 – 27.28) and those with lower DNAm (<median) had increased odds of PC compared to those with higher DNAm (≥median) in the promoter regions of CDKN1C (Chr11:2886398 OR, 95% CI: 4.07, 1.16 – 14.31; Chr11:2886388 OR, 95% CI: 3.51, 1.24 – 9.96) and RASSF1 (Chr3:50340841 OR, 95% CI: 3.98, 1.15 – 17.45). Similarly, in samples collected 2 - <5 years before diagnosis, lower DNAm (<median) was associated with increased odds of PC compared to those with higher DNAm (≥median) in the promoter region of TFPI-2 (Chr7:93890717 OR, 95% CI: 3.82, 1.20 – 12.15). In models with DNAm as a continuous measure and in samples collected <2 years before diagnosis, a 1% increase in DNAm in the promoter region of SFN was associated with increased odds of PC (Chr1:26863188 OR, 95% CI: 1.23, 1.01 – 1.51). There was no evidence of associations between DNAm on Alu, CLDN4, LCN2, LINE-1, p16, or SPARC and PC risk. Conclusions : Our findings of increased risk of PC associated with lower DNAm on BNIP3, CDKN1C and TFPI-2 and higher DNAm on SFN genes suggest there may be molecular patterns occurring prior to PC diagnosis that could be considered further for developing non-invasive biomarkers for early diagnosis of this aggressive carcinoma. The opinions and assertions expressed herein do not reflect the official policy or position of the Uniformed Services University, the Department of Defense, or the National Cancer Institute. Citation Format: Jordan McAdam, Ruth M Pfeiffer, Alexander P Keil, Ann Meyer, Anwar E Ahmed, Dechang Chen, Jennifer Rusiecki. DNA methylation and pancreatic cancer in select gene promoter regions: A repeated measures case-control study of US military service members [abstract]. In: Proceedings of the AACR Special Conference in Cancer Research: Advances in Pancreatic Cancer Research; 2024 Sep 15-18; Boston, MA. Philadelphia (PA): AACR; Cancer Res 2024;84(17 Suppl_2):Abstract nr B020.

Aim: We thoroughly discuss the interaction between the stemness index and DNA methylation in pancreatic cancer (PC). Materials & methods: First, the stemness indices of PC (denoted mRNAsi and mDNAsi) were calculated using a one-class logistic regression machine-learning algorithm. Second, we screened the central methylation sites associated with stemness and screened out the key genes. We investigated the DNA methylation regulators associated with the key genes. Finally, using CIBERSORT and TIMER, we assessed the influence of stemness indexes and key genes on PC microenvironment formation. Results: In this study we quantified the stemness indices for PC and screened 20 related central DNA methylation sites. Further analysis of the methylation site cg22687244, located in the 3'UTR, revealed that it promoted the expression of the key geneFAM81A. We show that FAM81A may be regulated by DNA methylation regulators. Furthermore, immune cells were found to be more abundant in PC microenvironments with high expression of FAM81A. Conclusion: We report for the first time that the 3'UTR methylation of FAM81A is closely related to PC stemness and contributes to tumor immune infiltration. Therefore FAM81A may serve as a potential marker to guide the treatment of PC.

DNA methylation acts in concert with other epigenetic mechanisms to regulate normal gene expression and facilitate chromatin organization within cells. Aberrant DNA methylation patterns are acquired during carcinogenic transformation; such events are often accompanied by alterations in chromatin structure at gene regulatory regions. The expression pattern of any given gene is achieved by interacting epigenetic mechanisms. First, the insertion of nucleosomes at transcriptional start sites prevents the binding of the transcriptional machinery and additional cofactors that initiate gene expression. Second, nucleosomes anchor all of the DNMT3A and DNMT3B methyltransferase proteins in the cell, which suggests a role for histone octamers in the establishment of DNA methylation patterns. During carcinogenesis, epigenetic switching and 5-methylcytosine reprogramming result in the aberrant hypermethylation of CpG islands, reducing epigenetic plasticity of critical developmental and tumor suppressor genes, rendering them unresponsive to normal stimuli. Here, we will discuss the importance of both established and novel molecular concepts that may underlie the role of DNA methylation in cancer.

Apoptotic peptidase activating factor 1 (APAF-1) is essential regulator of apoptosis and inactivation by DNA methylation is common event in numerous cancer types. We investigated the regulation of APAF-1 through DNA methylation in pancreatic cancer. Datasets from 44 patients after pancreatoduodenectomy and the pancreatic adenocarcinoma (PDAC) cell lines Capan-2 and MIA PaCa-2 treated with decitabine were analyzed by RT-PCR, immunoblotting, methylation-specific PCR analysis, apoptosis and viability assays to identify effects of APAF-1 regulation. APAF-1 mRNA and protein levels were significantly down-regulated, and APAF-1 methylation status was associated with perineural invasion in PDAC. Decitabine inhibited cell viability and increased apoptosis rates, however failed to restore APAF-1 mRNA and protein levels in cells. APAF-1 gene hypermethylation may contribute to the progression of PDAC through perineural invasion. Decitabine could sensitize pancreatic cancer cells to apoptosis and growth retardation, however, not directly through the APAF-1 demethylation process.

<div>Abstract<p>Global DNA hypomethylation occurs in many cancer types, but there is no explanation for its differential occurrence or possible impact on cancer cell physiology. Here we address these issues with a computational study of genome-scale DNA methylation in 16 cancer types. Specifically, we identified (i) a possible determinant for global DNA methylation in cancer cells and (ii) a relationship between levels of DNA methylation, nucleotide synthesis, and intracellular oxidative stress in cells. We developed a system of kinetic equations to capture the metabolic relations among DNA methylation, nucleotide synthesis, and antioxidative stress response, including their competitions for methyl and sulfur groups, based on known information about one-carbon metabolism and trans-sulfuration pathways. We observed a kinetic-based regulatory mechanism that controls reaction rates of the three competing processes when their shared resources are limited, particularly when the nucleotide synthesis rates or oxidative states are high. The combination of this regulatory mechanism and the need for rapid nucleotide synthesis, as well as high production of glutathione dictated by cancer-driving forces, led to the nearly universal observations of reduced global DNA methylation in cancer. Our model provides a natural explanation for differential global DNA methylation levels across cancer types and supports the observation that more malignant cancers tend to exhibit reduced DNA methylation levels. Insights obtained from this work provide useful information about the complexities of cancer due to interplays among competing, dynamic biological processes. <i>Cancer Res; 77(15); 4185–95. ©2017 AACR</i>.</p></div>

Global DNA hypomethylation occurs in many cancer types, but there is no explanation for its differential occurrence or possible impact on cancer cell physiology. Here we address these issues with a computational study of genome-scale DNA methylation in 16 cancer types. Specifically, we identified (i) a possible determinant for global DNA methylation in cancer cells and (ii) a relationship between levels of DNA methylation, nucleotide synthesis, and intracellular oxidative stress in cells. We developed a system of kinetic equations to capture the metabolic relations among DNA methylation, nucleotide synthesis, and antioxidative stress response, including their competitions for methyl and sulfur groups, based on known information about one-carbon metabolism and trans-sulfuration pathways. We observed a kinetic-based regulatory mechanism that controls reaction rates of the three competing processes when their shared resources are limited, particularly when the nucleotide synthesis rates or oxidative states are high. The combination of this regulatory mechanism and the need for rapid nucleotide synthesis, as well as high production of glutathione dictated by cancer-driving forces, led to the nearly universal observations of reduced global DNA methylation in cancer. Our model provides a natural explanation for differential global DNA methylation levels across cancer types and supports the observation that more malignant cancers tend to exhibit reduced DNA methylation levels. Insights obtained from this work provide useful information about the complexities of cancer due to interplays among competing, dynamic biological processes. Cancer Res; 77(15); 4185-95. ©2017 AACR.