Comparative DNA methylation analysis in normal and tumour tissues and in cancer cell lines using differential methylation hybridisation

A Bayesian hierarchical model to estimate DNA methylation conservation in colorectal tumors

MOTIVATION

Conservation is broadly used to identify biologically important (epi)genomic regions. In the case of tumor growth, preferential conservation of DNA methylation can be used to identify areas of particular functional importance to the tumor. However, reliable assessment of methylation conservation based on multiple tissue samples per patient requires the decomposition of methylation variation at multiple levels.

RESULTS

We developed a Bayesian hierarchical model that allows for variance decomposition of methylation on three levels: between-patient normal tissue variation, between-patient tumor-effect variation and within-patient tumor variation. We then defined a model-based conservation score to identify loci of reduced within-tumor methylation variation relative to between-patient variation. We fit the model to multi-sample methylation array data from 21 colorectal cancer (CRC) patients using a Monte Carlo Markov Chain algorithm (Stan). Sets of genes implicated in CRC tumorigenesis exhibited preferential conservation, demonstrating the model's ability to identify functionally relevant genes based on methylation conservation. A pathway analysis of preferentially conserved genes implicated several CRC relevant pathways and pathways related to neoantigen presentation and immune evasion. Our findings suggest that preferential methylation conservation may be used to identify novel gene targets that are not consistently mutated in CRC. The flexible structure makes the model amenable to the analysis of more complex multi-sample data structures.

AVAILABILITY AND IMPLEMENTATION

The data underlying this article are available in the NCBI GEO Database, under accession code GSE166212. The R analysis code is available at https://github.com/kevin-murgas/DNAmethylation-hierarchicalmodel.

SUPPLEMENTARY INFORMATION

Supplementary data are available at Bioinformatics online.

Plasma cell free DNA methylation markers for hepatocellular carcinoma surveillance in patients with cirrhosis: a case control study

BACKGROUND

Hepatocellular carcinoma (HCC) is the leading cause of death in patients with cirrhosis, primarily due to failed early detection. HCC screening is recommended among individuals with cirrhosis using biannual abdominal ultrasound, for earlier tumor detection, administration of curative treatment, and improved survival. Surveillance by imaging with or without biomarkers such as alpha-fetoprotein (AFP) remains suboptimal for early stage HCC detection. Here we report on the development and assessment of methylation biomarkers from liquid biopsies for HCC surveillance in cirrhotic patients.

METHODS

DNA methylation markers including the HCCBloodTest (Epigenomics AG) and a DNA-methylation panel established by next generation sequencing (NGS) were assessed using a training/testing design. The NGS panel algorithm was established in a training study (41 HCC patients; 46 cirrhotic non-HCC controls). For testing, plasma samples were obtained from cirrhotic patients (Child class A or B) with (60) or without (103) early stage HCC (BCLC stage 0, A, B). The assays were then tested using blinded sample sets and analyzed by preset algorithms.

RESULTS

The HCCBloodTest and the NGS panel exhibited 76.7% and 57% sensitivities at 64.1% and 97% specificity, respectively. In a post-hoc analysis, a combination of the NGS panel with AFP (20 ng/mL) achieved 68% sensitivity at 97% specificity (AUC = 0.9).

CONCLUSIONS

Methylation biomarkers in cell free plasma DNA provide a new alternative for HCC surveillance. Multiomic panels comprising DNA methylation markers with other biological markers, such as AFP, provide an option to further increase the overall clinical performance of surveillance via minimally invasive blood samples.

TRIAL REGISTRATION

Test set study-ClinicalTrials.gov (NCT03804593) January 11, 2019, retrospectively registered.

Epigenetic IVD Tests for Personalized Precision Medicine in Cancer

Epigenetic alterations play a key role in the initiation and progression of cancer. Therefore, it is possible to use epigenetic marks as biomarkers for predictive and precision medicine in cancer. Precision medicine is poised to impact clinical practice, patients, and healthcare systems. The objective of this review is to provide an overview of the epigenetic testing landscape in cancer by examining commercially available epigenetic-based in vitro diagnostic tests for colon, breast, cervical, glioblastoma, lung cancers, and for cancers of unknown origin. We compile current commercial epigenetic tests based on epigenetic biomarkers (i.e., DNA methylation, miRNAs, and histones) that can actually be implemented into clinical practice.

Human age estimation from blood using mRNA, DNA methylation, DNA rearrangement, and telomere length

Establishing the age of unknown persons, or persons with unknown age, can provide important leads in police investigations, disaster victim identification, fraud cases, and in other legal affairs. Previous methods mostly relied on morphological features available from teeth or skeletal parts. The development of molecular methods for age estimation allowing to use human specimens that possess no morphological age information, such as bloodstains, is extremely valuable as this type of samples is commonly found at crime scenes. Recently, we introduced a DNA-based approach for human age estimation from blood based on the quantification of T-cell specific DNA rearrangements (sjTRECs), which achieves accurate assignment of blood DNA samples to one of four 20-year-interval age categories. Aiming at improving the accuracy of molecular age estimation from blood, we investigated different types of biomarkers. We started out by systematic genome-wide surveys for new age-informative mRNA and DNA methylation markers in blood from the same young and old individuals using microarray technologies. The obtained candidate markers were validated in independent samples covering a wide age range using alternative technologies together with previously proposed DNA methylation, sjTREC, and telomere length markers. Cross-validated multiple regression analysis was applied for estimating and validating the age predictive power of various sets of biomarkers within and across different marker types. We found that DNA methylation markers outperformed mRNA, sjTREC, and telomere length in age predictive power. The best performing model included 8 DNA methylation markers derived from 3 CpG islands reaching a high level of accuracy (cross-validated R(2)=0.88, SE±6.97 years, mean absolute deviation 5.07 years). However, our data also suggest that mRNA markers can provide independent age information: a model using a combined set of 5 DNA methylation markers and one mRNA marker could provide similarly high accuracy (cross-validated R(2)=0.86, SE±7.62 years, mean absolute deviation 4.60 years). Overall, our study provides new and confirms previously suggested molecular biomarkers for age estimation from blood. Moreover, our comparative study design revealed that DNA methylation markers are superior for this purpose over other types of molecular biomarkers tested. While the new and some previous findings are highly promising, before molecular age estimation can eventually meet forensic practice, the proposed biomarkers should be tested further in larger sets of blood samples from both healthy and unhealthy individuals, and markers and genotyping methods shall be validated to meet forensic standards.

De novo identification of differentially methylated regions in the human genome

Background

The identification and characterisation of differentially methylated regions (DMRs) between phenotypes in the human genome is of prime interest in epigenetics. We present a novel method, DMRcate, that fits replicated methylation measurements from the Illumina HM450K BeadChip (or 450K array) spatially across the genome using a Gaussian kernel. DMRcate identifies and ranks the most differentially methylated regions across the genome based on tunable kernel smoothing of the differential methylation (DM) signal. The method is agnostic to both genomic annotation and local change in the direction of the DM signal, removes the bias incurred from irregularly spaced methylation sites, and assigns significance to each DMR called via comparison to a null model.

Results

We show that, for both simulated and real data, the predictive performance of DMRcate is superior to those of Bumphunter and Probe Lasso, and commensurate with that of comb-p. For the real data, we validate all array-derived DMRs from the candidate methods on a suite of DMRs derived from whole-genome bisulfite sequencing called from the same DNA samples, using two separate phenotype comparisons.

Conclusions

The agglomeration of genomically localised individual methylation sites into discrete DMRs is currently best served by a combination of DM-signal smoothing and subsequent threshold specification. The findings also suggest the design of the 450K array shows preference for CpG sites that are more likely to be differentially methylated, but its overall coverage does not adequately reflect the depth and complexity of methylation signatures afforded by sequencing.

For the convenience of the research community we have created a user-friendly R software package called DMRcate, downloadable from Bioconductor and compatible with existing preprocessing packages, which allows others to apply the same DMR-finding method on 450K array data.

Electronic supplementary material

The online version of this article (doi:10.1186/1756-8935-8-6) contains supplementary material, which is available to authorized users.

BIMMER: a novel algorithm for detecting differential DNA methylation regions from MBDCap-seq data

DNA methylation is a common epigenetic marker that regulates gene expression. A robust and cost-effective way for measuring whole genome methylation is Methyl-CpG binding domain-based capture followed by sequencing (MBDCap-seq). In this study, we proposed BIMMER, a Hidden Markov Model (HMM) for differential Methylation Regions (DMRs) identification, where HMMs were proposed to model the methylation status in normal and cancer samples in the first layer and another HMM was introduced to model the relationship between differential methylation and methylation statuses in normal and cancer samples. To carry out the prediction for BIMMER, an Expectation-Maximization algorithm was derived. BIMMER was validated on the simulated data and applied to real MBDCap-seq data of normal and cancer samples. BIMMER revealed that 8.83% of the breast cancer genome are differentially methylated and the majority are hypo-methylated in breast cancer.

Plasma methylated septin 9: a colorectal cancer screening marker

Colorectal cancer (CRC) is a slow-developing cancer (10-15 years) with one of the highest frequencies in the world's population. Many countries have implemented various CRC screening programs, but have not achieved the desired compliance. Colonoscopy - considered the gold standard for CRC screening - has its limitations as well as the other techniques used, such as irrigoscopy, sigmoidoscopy, fecal blood and hemoglobin tests. The biomarker septin 9 has been found to be hypermethylated in nearly 100% of tissue neoplasia specimens and detected in circulating DNA fractions of CRC patients. A commercially available assay for septin 9 has been developed with moderate sensitivity (∼70%) and specificity (∼90%) and a second generation assay, Epi proColon 2.0 (Epigenomics AG), shows increased sensitivity (∼92%). The performance of the assay proved to be independent of tumor site and reaches a high sensitivity of 77%, even in early cancer stages (I and II). Furthermore, septin 9 was recently used in follow-up studies for detection of early recurrence of CRC. This article evaluates the opportunities, known limitations and future perspectives of the recently introduced Epi proColon(®) 2.0 test, which is based on the detection of aberrantly methylated DNA of the v2 region of the septin 9 gene in plasma.

Lung cancer epigenetics: emerging biomarkers

Lung cancer is the leading cause of cancer-related deaths worldwide, and the 5-year survival rate is still very poor due to the scarcity of effective tools for early detection. The discovery of highly sensitive and specific biomarkers highlighting pathological changes early enough to allow clinical intervention is therefore of great importance. In the last decade, epigenetics and particularly research on DNA methylation have provided important information towards a better understanding of lung cancer pathogenesis. Novel and promising molecular biomarkers for diagnosis and prognosis of lung cancer are continuously emerging in this area, requiring further evaluation. This process includes extensive validation in prospective clinical trials before they can be routinely used in a clinical setting. This review summarizes the evidence on epigenetic biomarkers for lung cancer, focusing on DNA methylation.

The presence of methylation quantitative trait loci indicates a direct genetic influence on the level of DNA methylation in adipose tissue

Genetic variants that associate with DNA methylation at CpG sites (methylation quantitative trait loci, meQTLs) offer a potential biological mechanism of action for disease associated SNPs. We investigated whether meQTLs exist in abdominal subcutaneous adipose tissue (SAT) and if CpG methylation associates with metabolic syndrome (MetSyn) phenotypes. We profiled 27,718 genomic regions in abdominal SAT samples of 38 unrelated individuals using differential methylation hybridization (DMH) together with genotypes at 5,227,243 SNPs and expression of 17,209 mRNA transcripts. Validation and replication of significant meQTLs was pursued in an independent cohort of 181 female twins. We find that, at 5% false discovery rate, methylation levels of 149 DMH regions associate with at least one SNP in a ±500 kilobase cis-region in our primary study. We sought to validate 19 of these in the replication study and find that five of these significantly associate with the corresponding meQTL SNPs from the primary study. We find that none of the 149 meQTL top SNPs is a significant expression quantitative trait locus in our expression data, but we observed association between expression levels of two mRNA transcripts and cis-methylation status. Our results indicate that DNA CpG methylation in abdominal SAT is partly under genetic control. This study provides a starting point for future investigations of DNA methylation in adipose tissue.

Genome-wide screen for differential DNA methylation associated with neural cell differentiation in mouse

Cellular differentiation involves widespread epigenetic reprogramming, including modulation of DNA methylation patterns. Using Differential Methylation Hybridization (DMH) in combination with a custom DMH array containing 51,243 features covering more than 16,000 murine genes, we carried out a genome-wide screen for cell- and tissue-specific differentially methylated regions (tDMRs) in undifferentiated embryonic stem cells (ESCs), in in-vitro induced neural stem cells (NSCs) and 8 differentiated embryonic and adult tissues. Unsupervised clustering of the generated data showed distinct cell- and tissue-specific DNA methylation profiles, revealing 202 significant tDMRs (p<0.005) between ESCs and NSCs and a further 380 tDMRs (p<0.05) between NSCs/ESCs and embryonic brain tissue. We validated these tDMRs using direct bisulfite sequencing (DBS) and methylated DNA immunoprecipitation on chip (MeDIP-chip). Gene ontology (GO) analysis of the genes associated with these tDMRs showed significant (absolute Z score>1.96) enrichment for genes involved in neural differentiation, including, for example, Jag1 and Tcf4. Our results provide robust evidence for the relevance of DNA methylation in early neural development and identify novel marker candidates for neural cell differentiation.

SHOX2 DNA methylation is a biomarker for the diagnosis of lung cancer based on bronchial aspirates

BACKGROUND

This study aimed to show that SHOX2 DNA methylation is a tumor marker in patients with suspected lung cancer by using bronchial fluid aspirated during bronchoscopy. Such a biomarker would be clinically valuable, especially when, following the first bronchoscopy, a final diagnosis cannot be established by histology or cytology. A test with a low false positive rate can reduce the need for further invasive and costly procedures and ensure early treatment.

METHODS

Marker discovery was carried out by differential methylation hybridization (DMH) and real-time PCR. The real-time PCR based HeavyMethyl technology was used for quantitative analysis of DNA methylation of SHOX2 using bronchial aspirates from two clinical centres in a case-control study. Fresh-frozen and Saccomanno-fixed samples were used to show the tumor marker performance in different sample types of clinical relevance.

RESULTS

Valid measurements were obtained from a total of 523 patient samples (242 controls, 281 cases). DNA methylation of SHOX2 allowed to distinguish between malignant and benign lung disease, i.e. abscesses, infections, obstructive lung diseases, sarcoidosis, scleroderma, stenoses, at high specificity (68% sensitivity [95% CI 62-73%], 95% specificity [95% CI 91-97%]).

CONCLUSIONS

Hypermethylation of SHOX2 in bronchial aspirates appears to be a clinically useful tumor marker for identifying subjects with lung carcinoma, especially if histological and cytological findings after bronchoscopy are ambiguous.

SOX11 expression correlates to promoter methylation and regulates tumor growth in hematopoietic malignancies

Background

The transcription factor SOX11 plays an important role in embryonic development of the central nervous system (CNS) and is expressed in the adult immature neuron but is normally not expressed in any other adult tissue. It has recently been reported to be implicated in various malignant neoplasms, including several lymphoproliferative diseases, by its specific expression and in some cases correlation to prognosis. SOX11 has been shown to prevent gliomagenesis in vivo but the causes and consequences of aberrant expression of SOX11 outside the CNS remain unexplained.

Results

We now show the first function of SOX11 in lymphoproliferative diseases, by demonstrating in vitro its direct involvement in growth regulation, as assessed by siRNA-mediated silencing and ectopic overexpression in hematopoietic malignancies. Gene Chip analysis identified cell cycle regulatory pathways, including Rb-E2F, to be associated with SOX11-induced growth reduction. Furthermore, promoter analysis revealed that SOX11 is silenced through DNA methylation in B cell lymphomas, suggesting that its regulation is epigenetically controlled.

Conclusions

The data show that SOX11 is not a bystander but an active and central regulator of cellular growth, as both siRNA-mediated knock-down and ectopic overexpression of SOX11 resulted in altered proliferation. Thus, these data demonstrate a tumor suppressor function for SOX11 in hematopoietic malignancies and revealed a potential epigenetic regulation of this developmentally involved gene.

CDO1 promoter methylation is a biomarker for outcome prediction of anthracycline treated, estrogen receptor-positive, lymph node-positive breast cancer patients

Background

Various biomarkers for prediction of distant metastasis in lymph-node negative breast cancer have been described; however, predictive biomarkers for patients with lymph-node positive (LNP) disease in the context of distinct systemic therapies are still very much needed. DNA methylation is aberrant in breast cancer and is likely to play a major role in disease progression. In this study, the DNA methylation status of 202 candidate loci was screened to identify those loci that may predict outcome in LNP/estrogen receptor-positive (ER+) breast cancer patients with adjuvant anthracycline-based chemotherapy.

Methods

Quantitative bisulfite sequencing was used to analyze DNA methylation biomarker candidates in a retrospective cohort of 162 LNP/ER+ breast cancer patients, who received adjuvant anthracycline-based chemotherapy. First, twelve breast cancer specimens were analyzed for all 202 candidate loci to exclude genes that showed no differential methylation. To identify genes that predict distant metastasis, the remaining loci were analyzed in 84 selected cases, including the 12 initial ones. Significant loci were analyzed in the remaining 78 independent cases. Metastasis-free survival analysis was conducted by using Cox regression, time-dependent ROC analysis, and the Kaplan-Meier method. Pairwise multivariate regression analysis was performed by linear Cox Proportional Hazard models, testing the association between methylation scores and clinical parameters with respect to metastasis-free survival.

Results

Of the 202 loci analysed, 37 showed some indication of differential DNA methylation among the initial 12 patient samples tested. Of those, 6 loci were associated with outcome in the initial cohort (n = 84, log rank test, p < 0.05).

Promoter DNA methylation of cysteine dioxygenase 1 (CDO1) was confirmed in univariate and in pairwise multivariate analysis adjusting for age at surgery, pathological T stage, progesterone receptor status, grade, and endocrine therapy as a strong and independent biomarker for outcome prediction in the independent validation set (log rank test p-value = 0.0010).

Conclusions

CDO1 methylation was shown to be a strong predictor for distant metastasis in retrospective cohorts of LNP/ER+ breast cancer patients, who had received adjuvant anthracycline-based chemotherapy.

Aberrant DNA methylation occurs in colon neoplasms arising in the azoxymethane colon cancer model

Mouse models of intestinal tumors have advanced our understanding of the role of gene mutations in colorectal malignancy. However, the utility of these systems for studying the role of epigenetic alterations in intestinal neoplasms remains to be defined. Consequently, we assessed the role of aberrant DNA methylation in the azoxymethane (AOM) rodent model of colon cancer. AOM induced tumors display global DNA hypomethylation, which is similar to human colorectal cancer. We next assessed the methylation status of a panel of candidate genes previously shown to be aberrantly methylated in human cancer or in mouse models of malignant neoplasms. This analysis revealed different patterns of DNA methylation that were gene specific. Zik1 and Gja9 demonstrated cancer-specific aberrant DNA methylation, whereas, Cdkn2a/p16, Igfbp3, Mgmt, Id4, and Cxcr4 were methylated in both the AOM tumors and normal colon mucosa. No aberrant methylation of Dapk1 or Mlt1 was detected in the neoplasms, but normal colon mucosa samples displayed methylation of these genes. Finally, p19(Arf), Tslc1, Hltf, and Mlh1 were unmethylated in both the AOM tumors and normal colon mucosa. Thus, aberrant DNA methylation does occur in AOM tumors, although the frequency of aberrantly methylated genes appears to be less common than in human colorectal cancer. Additional studies are necessary to further characterize the patterns of aberrantly methylated genes in AOM tumors.

DNA methylation biomarkers for blood-based colorectal cancer screening

BACKGROUND

Sensitive, specific blood-based tests are difficult to develop unless steps are taken to maximize performance characteristics at every stage of marker discovery and development. We describe a sieving strategy for identifying high-performing marker assays that detect colorectal cancer (CRC)-specific methylated DNA in plasma.

METHODS

We first used restriction enzyme-based discovery methods to identify marker candidates with obviously different methylation patterns in CRC tissue and nonpathologic tissue. We then used a selection process incorporating microarrays and/or real-time PCR analysis of tissue samples to further test marker candidates for maximum methylation in CRC tissue and minimum amplification in tissues from both healthy individuals and patients with other diseases. Real-time assays of 3 selected markers were validated with plasma samples from 133 CRC patients and 179 healthy control individuals in the same age range.

RESULTS

Restriction enzyme-based testing identified 56 candidate markers. This group was reduced to 6 with microarray and real-time PCR testing. Three markers, TMEFF2, NGFR, and SEPT9, were tested with plasma samples. TMEFF2 methylation was detected in 65% [95% confidence interval, 56%-73%] of plasma samples from CRC patients and not detected in 69% (62%-76%) of the controls. The corresponding results for NGFR were 51% (42%-60%) and 84% (77%-89%); for SEPT9, the values were 69% (60%-77%) and 86% (80%-91%).

CONCLUSIONS

The stringent criteria applied at all steps of the selection and validation process enabled successful identification and ranking of blood-based marker candidates.