| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
© Oxford University Press 2006.
NEWS |
Momentum Building for Human Epigenome Project
Epigenetics does not get the recognition it deserves, say its proponents. "Our struggle in this field has been to convince the community that epigenetics really has a major role in human disease," said Peter Jones, D.Sc., director of the Norris Comprehensive Cancer Center at the University of Southern California in Los Angeles. Jones and 39 other scientists are now trying to convince funding agencies that epigenetics warrants a "Human Epigenome Project." This long-term effort will require substantial but as yet unspecified funds in an era of budget austerity for the National Institutes of Health.
|
There is a strong case for such a project. Epigenetics is the study of heritable changes in gene expression that occur without changes in DNA sequence—a vast and largely unexplored biological domain. Rudolph Jaenisch, Ph.D., of MIT's Whitehead Institute, has defined epigenetics as "the mechanism to transmit gene expression states from one cell generation to the next." So it's prima facie important, since epigenetics, by definition, governs development and differentiation. Stem cells, for example, give rise to differentiated tissues without accompanying gene mutations. "The whole essence of stem cells is epigenetic," Jones pointed out. And several diseases, including cancer, have been linked to epigenetic abnormalities.
But proponents still must overcome doubts in some quarters about the overall importance of epigenetics in cancer. No one knows the relative contributions of genetic and epigenetic changes in causing the disease. A human epigenome project could provide the answers—if it can convince the greater research community that it's worth supporting.
Charting a Course
The U.S. Human Epigenome Project idea is relatively new, evolving in a series of meetings separately sponsored by the American Association for Cancer Research and the National Cancer Institute. A "blueprint" published in the December 15 issue of Cancer Research laid out possible directions, and participants in a November workshop concluded that the project was worth undertaking and that the technology was available. They also selected tissues to be studied—a crucial decision because, unlike the human genome sequence, there are effectively infinite distinct epigenomes in any single organism. Epigenetic states differ widely among tissues, and changes are far more varied and much more frequent than DNA mutations. "Each differentiated cell has a different epigenome," said Jones. "That's the whole point."
Project planners have chosen to first catalog yeast and Drosophila epigenetics. Next would come mouse embryonic stem cells, followed by a human embryonic stem cell line. Finally, the epigenomes of up to 10 normal human tissues would be analyzed down to the single base pair level, with other tissues analyzed at lower resolutions. "For example, having a reference epigenome of the brain would be enormously helpful," said Andrew Feinberg, M.D., of Johns Hopkins, one of the project organizers.
|
Feinberg stressed the importance of a comprehensive approach to the epigenome. So far the best-studied epigenetic changes are DNA methylation and histone modifications, and the project would look at both. DNA methylation is the enzymatic addition of a methyl group to the cytosine base in DNA, creating a "fifth base." Almost all DNA methylation occurs at cytosine–guanine (C–G) pairs. So-called CpG islands, roughly 1 kilobase–long stretches of C–G repeats, are usually unmethylated in normal cells. Their methylation in the promoter regions of many tumor suppressor genes has been found in several cancers. The workhorse technology for global methylation analysis is bisulphite sequencing, which uses a chemical bisulphite to distinguish methylated from unmethylated cytosines.
The project would also comprehensively catalog changes in chromatin, the protein–DNA complex that forms chromosomes. Among other modifications, methylation, acetylation, and phosphorylation of histones—the bundled proteins that form the scaffold for the coiled DNA in chromatin—help regulate gene expression. The project would also look at chromatin binding proteins and DNAse hypersensitive sites, areas of chromatin associated with gene expression. "We're thinking about dozens of modifications that should be looked at systematically," said Feinberg. One important method is chromatin immunoprecipitation coupled with microarrays ("ChIP on chip"). Together with complementary methods, ChIP on chip "has evolved at ... a rapid rate," said Jones. "In fact, some genomes have been almost done already, looking at least for histone modification patterns, at a very decent level of resolution."
Joining the Epigenomics Boom
The proposed U.S. effort would not be the first epigenome project. In 1999, seven institutions in Britain, Germany, and France formed the Human Epigenome Consortium (HEC), whose ultimate goal is "to determine the methylation profile of all genes in all tissues," said Stephan Beck, Ph.D., of the Sanger Centre in Cambridge, U.K. (Beck coined the term "epigenome" that year.) The HEC's pilot project, funded by the European Union, was a methylation profile of the human major histocompatibility (MHC) locus, and was finished in 2004. The next stage, a methylation profile of all genes, excluding noncoding regions, in chromosomes 6, 13, 20, and 22 in about 200 normal human tissue samples, is now more than half finished. Upon completion of the project, "you will know exactly what the methylation is in each cell type," said Beck. "From this you can deduce which genes are turned on or which are switched off." Meanwhile, an EU-funded epigenome "network of excellence" is analyzing chromatin status and histone modifications.
"The Europeans quite frankly are way ahead of us," said Jones, who views the U.S. effort as complementary. The HEC's Beck agreed. "Everybody's clear desire is to avoid any redundancy here and to work together," he said. "With the U.S. effort coming online, I think the entire project will eventually have higher profile, and also there will be more funding opportunities, so it can be done fairly quickly."
Another "big science" project with an epigenome component is the proposed $1.5 billion Human Cancer Genome Project. (See News, Vol. 97, No. 18, p. 1322, "Human Cancer Genome Project Moving Forward Despite Some Doubts in Community.") In addition to DNA resequencing and other approaches, its project planners intend to compare epigenetic patterns in tumors and normal tissues from the same cancer patient tissue donors to identify epigenetic defects in tumors. Feinberg said the Human Epigenome Project would add important, nonredundant information by generating a reference epigenome taken from healthy people. In cancer patients, said Feinberg, nonmalignant tissue could harbor epigenetic changes that have set the stage for carcinogenesis. Just looking at noncancerous and cancerous lung tissue, for example, from one patient wouldn't identify the changes because they might be present in both samples. A normal reference epigenome for that tissue, compared with the nonmalignant tissue in the cancer patient, could highlight these early epigenetic changes that predispose to cancer.
How Important in Cancer?
But the role of epigenetics in cancer is controversial. To begin with, tumors show less methylation than normal tissues, or more methylation, depending on where one looks. It seems unlikely that the same mechanism causes both phenomena, and their relative importance is unknown. "Methylation patterns are clearly abnormal in cancer," said Tim Bestor, Ph.D., of Columbia University in New York. "The dominant effect that we see, via a whole-genome approach, is global demethylation. If you use a candidate gene approach, what you mostly see ... is de novo methylation of CpG islands, especially associated with tumor suppressors." Both phenomena are probably important in cancer, with demethylation acting to activate oncogenes or promote genomic instability and de novo promoter methylation functioning to turn off tumor suppressor genes. "There's no question that the epigenetic silencing of tumor suppressor genes and the epigenetic activation of tumor promoting genes play a causal role in cancer," said Feinberg.
|
But how much of a role? Jones believes epigenetic changes are "just as important" as genetic changes in cancer. "The single-minded focus on the genetic basis of cancer has been really important ... but has actually missed a lot of changes, because many genes become epigenetically silenced but not mutated," he said. Jean-Pierre Issa, M.D., of the University of Texas M. D. Anderson Cancer Center in Houston, agreed. "We really have to add the epigenetic information there," he said. Looking only at gene mutations "is really like having the song without the music, or an orchestra that is missing half of the instruments."
But many known tumor suppressor genes with CpG island methylation are the same ones silenced through mutation or copy loss, raising the possibility that genetic and genomic methods have already identified most of them. However, in a global genomic and methylation analysis of 26 brain tumors published in 2002 in Nature Genetics, the group led by Joe Costello, Ph.D., at the University of California–San Francisco, showed that in fact most genes silenced by CpG island methylation are separate from genes that are mutated and deleted. "While there is overlap, more often the targets of nonrandom methylation are outside the areas that past decades of genome scanning have found," said Costello. It follows that a human epigenome project comparing the methylation patterns of normal and cancerous samples should find many new tumor suppressor genes.
Bestor said the DNA methylation portion of the project should be "tremendously useful," but he questioned the wisdom of analyzing histone modifications in the project, citing a lack of evidence that they regulate gene expression and that they are heritable across generations of cells. "I have deep doubts about the role of histone modifications both in gene regulation and the transmission of states of gene activity," he said. "Transcription could remodel chromatin, rather than chromatin allowing transcription to proceed." However, Feinberg said that most chromatin researchers do not share these doubts.
One factor that has limited the acceptance of epigenetics as a major cause of cancer is the paucity of familial cancer syndromes that involve the epigenetic machinery. The genes mutated in hereditary cancers, with rare exceptions, have nothing to do with epigenetics, so a skeptical view is that most sporadic cancers aren't likely to spring from epigenetic defects either. But Feinberg points out that the roughly 30% of cancers considered familial remain largely unexplained by known defects, genetic or epigenetic. Cancer, in these cases, is a complex trait, with causes yet to be determined. The lack of epigenetic familial cancer syndromes to date "is not really a ding on epigenetics, because the same problem visits the genetics of cancer, too," Feinberg said. "Most of the genetic cause of predisposition to cancer is still not explained, either genetically or epigenetically."
Beyond Cancer
Beyond cancer, Costello sees the human epigenome project as a crucial step toward an understanding of how organisms are organized. "That's one of the most fundamental questions in biology," he said. "How is it that you can have the exact same blueprint, the same genome, within different tissues, and have such different functions? ... Those differences are really epigenetic."
The Human Epigenome Project, however, must still be funded. "We know there's no money out there," said Feinberg. "At the same time, the group feels it's really important to articulate a vision of what we think should get done over time." As of mid-December, Jones was putting together a task force, through AACR, that will generate an action plan, a cost estimate, and a timeline. With a consensus blueprint achieved, and international cooperation, the project has momentum. The Human Epigenome Project "will make a fundamental difference ... in understanding human disease [and] development, what stem cells are, and how we're different evolutionarily," said Feinberg. "It's not all epigenetics, but it's half anyway, and we feel that this is a great frontier that should be pursued in an aggressive way."
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  M. O. Hoque, M. S. Kim, K. L. Ostrow, J. Liu, G. B. A. Wisman, H. L. Park, M. L. Poeta, C. Jeronimo, R. Henrique, A. Lendvai, et al. Genome-Wide Promoter Analysis Uncovers Portions of the Cancer Methylome Cancer Res., April 15, 2008; 68(8): 2661 - 2670. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Ongenaert, L. Van Neste, T. De Meyer, G. Menschaert, S. Bekaert, and W. Van Criekinge PubMeth: a cancer methylation database combining text-mining and expert annotation Nucleic Acids Res., January 11, 2008; 36(suppl_1): D842 - D846. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. M. Brena and J. F. Costello Genome-epigenome interactions in cancer Hum. Mol. Genet., April 15, 2007; 16(R1): R96 - R105. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Esteller The necessity of a human epigenome project Carcinogenesis, June 1, 2006; 27(6): 1121 - 1125. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||