Quantcast

An Integrative Approach for Analyzing the Interplay of Genetic and Epigenetic Changes in Tumors

October 11, 2008

By Muradyan, Artur Boldt, Vivien; Steininger, Anne; Stabentheiner, Stephanie; Tebel, Katrin; Kreutzberger, Jurgen; Muller, Ines; Madle, Hannelore; Popper, Helmut H; Ullmann, Reinhard

The accumulation of chromosomal aberrations is a characteristic feature of tumor development. However, an understanding of tumorigenesis that assumes that changes in DNA copy number always cause equivalent changes in the corresponding RNA and protein levels is an oversimplification and completely ignores the individual genetic and epigenetic context in which an aberration has to be evaluated. We present a brief introduction to various techniques dedicated to the genome-wide analysis of genetic and epigenetic changes, and illustrate how complementary information derived from these various DNA array-based technologies can lead to a better understanding of the consequences of chromosomal aberrations. (Arch Pathol Lab Med. 2008;132:1557-1561)

DNA copy number changes have gained increasing attention from investigators as a means to identify genes associated with tumorigenesis as well as to provide markers assisting in differential diagnosis and in therapeutic decisions. For solid tumors, at least, this is mainly attributed to the introduction of techniques, such as comparative genomic hybridization (CGH),1 which for the first time, enabled the analysis of chromosomal gains and losses without the necessity of preparing metaphase spreads from the specimen to be analyzed. Thus CGH paved the way for the retrospective analysis of archival material and other samples that were difficult to analyze for DNA copy number changes by conventional karyotyping. In the course of a CGH experiment, the patient’s DNA and a reference DNA obtained from a healthy donor are labeled with different fluorochromes and cohybridized onto slides carrying metaphase spreads of another healthy donor. DNA copy number changes in the test DNA, relative to the reference DNA, result in different binding frequencies of the distinctively labeled DNAs to the respective region on the metaphase spreads and can be quantified by monitoring the fluorescence-signal intensity ratios of the 2 fluorochromes. Unfortunately, because metaphase chromosomes were used in the experimental design, the resolution is limited to about 3 to 10 Mb on average. These resolution limitations were abolished when the metaphase spreads were replaced by arrays of DNA spots, with each spot representing a specific chromosomal position. This variant of CGH was termed matrix2 or array CGH3 (Figure 1). Initially, clone-based array platforms dominated, primarily composed of bacterial artificial chromosome (BAC) clones. A BAC clone is a large insert clone that can carry, on average, 150-kb of insert DNA. Figure 2 gives the result of an array CGH analysis of a squamous cell carcinoma of the lung employing a submegabase resolution tilingpath BAC array, comprising more than 36 000 clones that cover the human genome in an overlapping fashion. Although BAC arrays are still used in many laboratories, they are increasingly being replaced by oligonucleotide arrays. These arrays are composed of small oligomers, ranging from 20 to 80 nucleotides, which are either presynthesized, like conventional primers for polymerase chain reaction, or more frequently, synthesized directly on the chip. Oligoarrays can offer great flexibility and enormous resolution; down to a few base pairs, if focused on certain regions. Figure 3 compares CGH results obtained with a BAC and an oligoarray, respectively.

The basic assumption underlying the use of chromosomal aberrations as diagnostic markers and predictors of biological behavior, however, is that a certain chromosome aberration always entails the same biological consequences. Yet, this scenario is rather unlikely, given the individual genetic background in each tumor, resulting in the abundance or depletion of different sets of transcription factors and other regulatory elements. The variable consequences of chromosomal aberrations can already be seen at the level of gene expression. Although some studies have used gene expression data to successfully predict the presence of DNA copy number changes, the integrative analysis of array CGH and gene expression data, derived from the same specimen, revealed that actually only 40% to 60% of genes directly reflect alterations of gene dosage at the DNA level (summarized in Stransky et al4). Moreover, there exist regions in which regulation of gene expression seems to be independent of DNA copy number,4 possibly reflecting epigenetic modifications, such as changes in DNA methylation or posttranslational histone modifications, which add another regulatory level of gene expression. Therefore, discussing DNA copy number changes in an isolated way does not give consideration to the complex interplay of genetic and epigenetic alterations.

Until recently, most attempts aimed at the genome-wide assessment of epigenetic changes in the individual patient failed because of the lack of appropriate analytical tools. Thanks to new technological advances, this shortcoming has been partly resolved, and the comparative analysis of genetic and epigenetic modifications at the genomic scale has become a realistic goal. One of these new powerful techniques is called chromatin immunoprecipitation on chip (ChIP on chip), a method that was designed initially to identify binding sites of transcription factors in yeast5 and cultured human cells.6 As for conventional chromatin immunoprecipitation, the method starts with cross-linking DNA and the proteins in the immediate vicinity (eg, by formaldehyde). Thereafter, the cells are lysed, and the chromatin is fragmented by sonication. Using an antibody linked to magnetic beads, the protein of interest is pulled down, coprecipitating the DNA attached to it. Enrichment of specific DNA sequences in the course of the separation is detected by comparison to the original input DNA following a hybridization procedure resembling the one already described for array CGH. Figure 4 shows the principle and the result of a ChIP on chip experiment analyzing the distribution of trimethylation of amino acid lysine at position 9 of histone 3 (H3K9me3), which is generally believed to be associated with a repressive state of chromatin.

Another epigenetic mark associated with suppression of gene expression is DNA methylation at CpG dinucleotides. Frequently, these CpGs are clustered in islands around transcription start sites of housekeeping genes, where they are usually unmethylated. Tumors use DNA methylation at CpG islands to silence tumor suppressor genes. Consequently, several techniques have been developed to analyze DNA methylation at such sites. However, only a small percentage of all CpGs that can be methylated are organized as CpG islands, with the remaining portion distributed across the whole genome.7 There, DNA methylation is implicated in ensuring genome integrity, for example, by silencing transposable elements.8 Not surprisingly, in addition to local hypermethylation of CpG islands, global hypomethylation can frequently be observed in tumors. Methylated DNA immunoprecipitation9 is a technique that is dedicated to the analysis of wholegenomic patterns of DNA methylation. The principle is similar to that of ChIP on chip, except that no cross- linking of DNA and adjacent proteins takes place and the antibody is targeted against 5′-methylcytidine, instead of a target protein.

In many cases, the detection of a chromosomal aberration will suffice to predict response to therapy or other relevant characteristics. The presence of a high-copy amplification of HER2/ neu, as a marker for the susceptibility to trastuzumab (Herceptin) treatment, is a good example. Unfortunately, the situation is not as simple in other instances, and only by the complementary application of the various techniques described above and their integrative analysis can we gain insight into the mechanisms that influence the course of that individual malignancy. Figures 5 and 6 exemplify how additional layers of information, such as ChIP on chip analysis of specific histone modifications, can clarify some of the variation in gene expression.

Although the benefits of this integrative approach will not immediately translate into clinical application, the more comprehensive understanding of the combined effects of genetic and epigenetic changes will lead to a more specific and accurate use of chromosomal markers.

References

1. Kallioniemi A, Kallioniemi OP, Waldman FM, et al. Detection of retinoblastoma gene copy number in metaphase chromosomes and interphase nuclei by fluorescence in situ hybridization. Cytogenet Cell Genet. 1992;60:190-193.

2. Solinas-Toldo S, Lampel S, Stilgenbauer S, et al. Matrix- based comparative genomic hybridization: biochips to screen for genomic imbalances. Genes Chromosomes Cancer. 1997;20:399-407.

3. Pinkel D, Segraves R, Sudar D, et al. High resolution analysis of DNA copy number variation using comparative genomic hybridization to microarrays. Nat Genet. 1998;20:207-211.

4. Stransky N, Vallot C, Reyal F, et al. Regional copy number- independent deregulation of transcription in cancer. Nat Genet. 2006;38:1386-1396.

5. Ren B, Robert F, Wyrick JJ, et al. Genome-wide location and function of DNA binding proteins. Science. 2000;290:2306-2309.

6. Weinmann AS, Farnham PJ. Identification of unknown target genes of human transcription factors using chromatin immunoprecipitation. Methods. 2002; 26:37-47. 7. Rollins RA, Haghighi F, Edwards JR, et al. Large-scale structure of genomic methylation patterns. Genome Res. 2006;16:157-163.

8. Wilson AS, Power BE, Molloy PL. DNA hypomethylation and human diseases. Biochim Biophys Acta. 2007;1775:138-162.

9. Weber M, Davies JJ, Wittig D, et al. Chromosome-wide and promoter-specific analyses identify sites of differential DNA methylation in normal and transformed human cells. Nat Genet. 2005;37:853-862.

10. Chen W, Erdogan F, Ropers HH, Lenzner S, Ullmann R. CGHPRO-a comprehensive data analysis tool for array CGH. BMC Bioinformatics. 2005;6: 85.

11. Olshen AB, Venkatraman ES, Lucito R, Wigler M. Circular binary segmentation for the analysis of array-based DNA copy number data. Biostatistics. 2004; 5:557-572.

Artur Muradyan, MSc; Vivien Boldt, MSc; Anne Steininger, MSc; Stephanie Stabentheiner, MSc; Katrin Tebel, BSc; Jurgen Kreutzberger, PhD; Ines Muller; Hannelore Madle; Helmut H. Popper, MD; Reinhard Ullmann, PhD

Accepted for publication May 20, 2008.

From the Department of Human Molecular Genetics, Max Planck Institute for Molecular Genetics, Berlin, Germany (Mr Muradyan, Mss Boldt and Steininger, Mrs Stabentheiner, Ms Tebel, Dr Kreutzberger, Mmes Muller and Madle, and Dr Ullmann); and the Institute of Pathology, Medical University of Graz, Graz, Austria (Dr Popper).

The authors have no relevant financial interest in the products or companies described in this article.

Reprints: Reinhard Ullmann, PhD, Department of Human Molecular Genetics, Max Planck Institute for Molecular Genetics, Ihnestr. 73, 14195 Berlin, Germany (e-mail: ullmann@molgen.mpg.de).

Copyright College of American Pathologists Oct 2008

(c) 2008 Archives of Pathology & Laboratory Medicine. Provided by ProQuest LLC. All rights Reserved.




comments powered by Disqus