Research
From Genome@Yale
It is a good rule not to put too much confidence in the observational results that are put forward until they are confirmed by theory.
-attributed to Sir Arthur Eddington in Horace Freeland Judson's The Eighth Day of Creation ISBN 0-87969-478-5
Contents |
Introduction
The human genome is predominantly composed of non-protein coding sequences (>98%) whose function remains largely undefined. A significant portion of the non-coding DNA is believed to serve as transcriptional regulatory elements that control how and when the coding fraction of the genome is used by a cell. There are four broad classes of transcriptional regulatory elements: enhancers, silencers, promoters and insulators. These elements are composed of a very short DNA sequences that serve as binding sites for transcription factors and are scattered throughout the genome. For each gene in the genome, there are usually multiple elements that control its expression. Conversely, each element may control multiple genes. Precise expression of each gene during development is achieved by a coordinated action of multiple transcriptional regulatory elements. In order to reconstruct and understand genome expression, we must identify most of these elements and determine how they are connected and controlled. By analyzing these transcriptional regulatory events in the entire genome, we hope to discover, reconstruct and analyze novel regulatory mechanisms and networks encoded in the genome.
Strategy
We employ ChIP-on-chip (chromatin immunoprecipitation couple to DNA microarrays) to define and analyze these non-coding, functional elements in the human genome. We utilize human cell lines and tissues (embryonic stem cells, primary fibroblasts, cancer/immortalized cells and primary cancer tissues) to identify those elements that are important for expression of the entire human genome (promoters and insulators) and those elements (enhancers and silencers) that are critical for defining cellular responses to extrinsic or intrinsic signals. We couple these experimental strategies with computational methods to systematically determine patterns, modes and mechanisms of binding by transcription factors that recognize these elements. We also utilize traditional molecular and biochemical methods for detailed analysis of the regulatory elements found in selected loci relevant for human diseases.
Regulatory Elements
Previously, we have defined and analyzed all active promoters that are responsible for expression of RNA polymerase II transcribed genes in primary human fibroblasts. About a third of the genes in the genome are associated with these active promoters. In addition, we have discovered a significant number of novel promoters that might drive expression of novel genes or transcripts, suggesting that we can use this promoter “map” to discover additional genes and transcripts that are missing in current gene annotations of the human genome. A subsequent analysis of chromatin structures on the promoters has revealed that all active promoters exhibit known chromatin structures associated with gene activation. Interestingly, an additional third of genes in the genome that are silent in this cell also display the same active chromatin structure, suggesting that chromatin structure plays a general role in defining which subset of the genome can be expressed in a given cell.
More recently, we have determined the locations of a large number of insulators in the genome of this primary fibroblast cells. Insulators are an important class of transcriptional regulatory elements that affect gene expression by preventing the spread of heterochromatin and restricting how enhancers select their target promoters. These insulators segment the genome into about 6,000 domains, with each domain containing an average of 2.5 genes. Surprisingly, many large gene families, such as the olfactory receptors, ribosomal proteins, keratins, metallothioneins, taste receptors, complement H groups, beta- and alpha-globins, alcohol dehydrogenases and house keeping genes, are bounded by insulators. We also find that distribution of insulators is cell type specific, suggesting that DNA methylation might be a possible epigenetic mechanism for regulated organization of the genome into distinct domains during development and disease.
We continue to characterize the active promoters, in terms of their strength, modes of regulation, and their role in pre-mRNA processing. We are also using molecular and biochemical approaches to understand mechanisms underlying insulator function and its role in organizing the genome. In addition, we have generated several “enhancer maps” in normal and cancer genomes and are now analyzing these elements. Combined knowledge of promoters, enhancers and insulators in the entire genome is enabling us to interrogate how these elements along with silencers and epigenetic modification of chromatin and DNA bring about specific patterns of genome expression. To facilitate a comprehensive analysis of these mechanisms, we are developing additional tools and methods that would enable a more integrated investigation of genome organization and epigenetic mechanisms responsible for regulated expression during cellular growth, differentiation, senescence and tumorigenesis.
Tumorigenesis
The WNT signaling pathway is essential for many developmental and adult tissue processes, including cell fate specification, proliferation, and differentiation. In CRC and cancers of other tissue, aberrant and persistent activation of the WNT signaling pathway is thought to be the initiating event responsible for tumorigenesis by altering expression of a set of genes. However, our understanding of how the WNT pathway regulates cell function and behavior in normal and disease states remains limited, because we do not yet have a coherent picture of how the WNT signal is translated into complex changes in cellular behavior and function. To date, only a limited number of WNT pathway targets have been identified. Even with the currently available list of target genes, the significance and general role of the WNT pathway in CRC and other cancers are beginning to be revealed. Therefore, understanding the entire repertoire of WNT pathway target genes and their concerted action will allow a system-wide examination of how this pathway promotes tumorigenesis.
Using the strategy that we have developed for determination of transcriptional regulatory maps in fibroblast and embryonic stem cells, we have mapped the binding sites of TCF4 and β-catenin, the downstream effectors of the WNT signaling pathway, within the entire genome of human colorectal cancer cells and generated a list of WNT-responsive enhancers that are aberrantly utilized in these cells. To determine of molecular mechanisms of WNT regulated expression of the target genes, other transcriptional regulatory elements, such as enhancers, silencers, insulators, and chromatin features and deletions and amplifications that may have occurred in the CRC genomes will be systematically analyzed. Genomic map of regulatory elements and mutations in the CRC genomes will provide a comprehensive view of genomic regulatory alterations that may contribute to its cancer phenotype. Furthermore, this transcriptional regulatory map of the CRC genome, when combined with the “normal” regulatory maps of diploid fibroblast and embryonic stem cells, will provide a detailed global view how transcription is regulated in three cell types, normal somatic, cancer, and stem cells, and potentially provide cancer-specific genome expression mechanisms that can be targeted for development of drugs and therapies.
