Changes in genomic DNA methylation patterns are one of the earliest and most consistent features of tumourigenesis. It has been demonstrated that aberrant DNA methylation profiles can be used as a valuable markers for clinical tumour characterisation. However, still relatively little is known about the nature of these changes and their role in cellular transformation.
...
We perform analyses toward the elucidation of the molecular regulation and functioning of epigenetic variations in cancer. The studies have a particular focus on the analysis of pancreatic cancer as well as breast and ovarial cancer.

Conversion of unmethylated cytosine to uracil by a treatment with bisulfite.

The detection of methylation variations is performed on oligonucleotide microarrays, using bisulfite treatment to uncover the methylation status. Sodium bisulfite induces methylation-dependent single nucleotide polymorphisms by converting unmethylated cytosine to uracil and, upon PCR amplification, to thymine. 5-methylcytosine is not affected by sodium bisulfite treatment and thus becomes amplified as cytosine.
...
The conversion can be identified by any means of sequence analysis. While microarray analysis was used until recently, we now perform high-throughput next-generation sequencing to this end. We utilise the technology toward genome-wide (if possible), high resolution (which is more important to us), gene-specific analyses of DNA methylation. This results in a detailed characterisation of genomic methylation patterns. The data are evaluated in combination with available clinical data and information from other analyses, such as transcriptional profiling. This may allow fundamental insights into the role of DNA methylation during tumourigenesis and provide the foundation for an epigenetic classification of tumours.

Correction of PCR-bais in quantitative DNA methylation studies by means of cubic polynomial regression

                         

DNA methylation profiling has become an important aspect of biomedical molecular analysis. PCR amplification of bisulfite-treated DNA is a processing step that is common to many currently used methods of quantitative methylation analysis. Preferential amplification of unmethylated alleles – known as PCR-bias – may significantly affect the accuracy of quantification. To date, reported processes for avoiding PCR-bias relied on an optimisation of PCR conditions. Although shown to be effective for particular genes, the implementation is time-consuming and labour-intensive, especially if multiple loci are analysed, thus the usefulness remains contradictory.
...
We developed an effective method of correcting biased methylation data. As opposed to refining experimental conditions, we approached the correction of amplification bias by accepting its occurrence during experimentation but adjusting the initial amplification result by a comparison to calibration data and the application of regression curves for deriving correction factors. As few as three calibration samples (0, 50, 100% methylation) were sufficient to define methylation degrees accurately. Two types of regression – hyperbolic and cubic polynomial – were applied. The correction process based on cubic polynomial regression was found to be superior overall.
...
The method is applicable irrespective of the locus that is interrogated or the number of sites analysed. Based on curve-fitting, the method is not influenced by the type of bias – preferential recovery of methylated or unmethylated alleles – and works equally well for both. Furthermore, any bias that was additionally introduced by the analysis procedure, for example a pyrosequencing readout, could be compensated by the very process. The method is also automatable for high-throughput analyses.
...

Moskalev et al. (2011) Nucleic Acids Res., in press.

Epigenetically de-regulated miR-375 is involved in a positive feedback loop with Estrogen Receptor alpha in breast cancer cells

                     

Breast cancer is the leading cause of cancer death in women worldwide. Although it is a heterogeneous disease, two-thirds of breast cancers share the common feature of being dependent on the presence and interaction of estrogen with the nuclear Estrogen Receptor α (ERα) protein. Approximately 70% of invasive breast cancers express ERα in actively proliferating cells.
...
It has become evident that ERα is up-regulated in luminal mammary epithelial cells during early stages of tumorigenesis and its overexpression is an important stimulatory factor for proliferation of mammary cells, leading to cell division and eventually to tumor development. The obvious role of ERα signalling in orchestrating the expression of genes involved in growth-related pathways, has established ERα as an important therapeutic target in breast cancer treatment. However, our understanding of the molecular mechanisms underlying deregulation of this signaling pathway is scarce.
...
We identified a high expression of microRNA 375 (miR-375) in ERα-positive cell lines. miR-375 overexpression is mainly caused by the loss of epigenetic marks, such as H3K9me2 and local DNA hypomethylation, which, in turn, triggers the dissociation of the transcriptional repressor CTCF from the promoter and enables interactions of ERα with regulatory regions of miR-375. Inhibition of miR-375 in ERα positive MCF-7 cells results in reduced ERα activation and cell proliferation.
...
A combination of expression profiling from tumour samples and miRNA target prediction identified RASD1 as a potential miR-375 target. Our findings show that miR-375 regulates RASD1 through targeting its 3' UTR. In addition, we demonstrated that RASD1 negatively regulates ERα expression. Our data indicate the existence of a positive regulation between ERα and miR-375 and suggest new strategies for the treatment of ER-positive invasive breast tumours.

de Soza Rocha Simonini et al. (2010) Cancer Res. 11, 162-167.

High definition profiles of drug-resistent breast and ovarian carcinoma

                       

Acquired drug resistance represents a frequent obstacle which hampers efficient chemotherapy of cancers. Hence, treatment via standardised chemotherapy fails in a large fraction of patients, making additional therapy regimens necessary. In order to minimise side-effects caused by ineffective chemotherapy, prediction of individual patients’ response to therapeutic agents before the commencement of treatment would be desirable. The contribution of aberrant DNA methylation to the development of drug resistant tumour cells has gained increasing attention over the past decades. Hence, the objective of this study was to characterise DNA methylation changes which arise from treatment of tumour cells with the chemotherapeutic drug doxorubicin.
...
DNA methylation levels from CpG islands (CGIs) linked to twenty-eight genes were analysed, whose expression levels had previously been shown to contribute to resistance against DNA double strand break inducing drugs or tumor progression in different cancer types. High-definition DNA methylation profiles which consisted of methylation levels from 800 CpG sites mapping to CGIs around the transcription start sites of the selected genes were determined. In order to investigate the influence of CGI methylation on the expression of associated genes, their mRNA levels were investigated via qRT-PCR. It was shown that the employed method is suitable for providing highly accurate methylation profiles, comparable to those obtained via clone sequencing, the gold standard for high-definition DNA methylation studies.
...
In breast carcinoma cells with acquired resistance against the double strand break inducing drug doxorubicin, changes in methylation of specific cytosines from CGIs linked to thirteen genes were detected. Moreover, similarities between methylation profiles obtained from breast and ovarian carcinoma cell lines with acquired doxorubicin resistance were found. The expression levels of a subset of analysed genes were shown to be linked to the methylation levels of the analyzed CGIs.
...
Our results provide detailed DNA methylation information from two separate model systems for acquired doxorubicin resistance and suggest the occurrence of similar methylation changes in both systems upon exposure to the drug. This collaborative effort is being continued with the aim of stratification of patients prior to treatment dicisions.

Böttcher et al. (2010) PloS ONE 5, e11002.    
Böttcher & Hoheisel (2010) Curr. Genom. 11, 162-167.

As part of the National German Research Network (NGFN) and coordinated by us, we applied microarray technology toward a genome-wide and at the same time high-resolution analysis of DNA methylation patterns (see also article on page 30 of Science Inside).
...

...
Infrastructure was provided for collaborations within the NGFN-network and beyond. Different systems were provided, which meet the requirements of the respective project. By comparisons between them, also standards were defined that allow comparability of data across platforms and across various areas of analysis. Also for this purpose, the close connection and integration of the consortium with international partners was crucial. The platform combined groups with complementary but nevertheless sufficiently overlapping expertise, competence and capability. Parallel to data production and a biological, functionally oriented interpretation, the resulting information was used for biomedical applications. In addition, technical developments were performed in order to improve quality and reliability.
...
The results from the SMP’s activities were: (1) definition and establishment of standards for epigenetic analyses on microarray systems; evaluation and comparison to other systems via external networking; (2) establishment of genome-wide microarrays for a global analysis of disease-relevant variations in methylation; (3) identification of disease-specific markers or patterns for diagnosis and prognosis; (4) identification of methylation patterns that are indicative for the action of drugs (clinical and commercial utilisation by the respective SMP- and NGFN-partners). In terms of tumour entities, there was a focus on chronic lymphocytic leukemia, melanoma, pancreatic and breast cancer.

Specific results were a process for the correction of PCR biases during the measurement of methylation (see above), methylation patterns closely associated with the occurance of pancreatic cancer as well as the elucidation of functional mechanisms in breast and pancreatic cancer.

de Soza Rocha Simonini et al. (2010) Cancer Res. 11, 162-167.
Moskalev et al. (2011) Nucleic Acids Res., in press.