Mammalian chromosomes are organized as a series of supercoilable looped domains about 100 kilobases in length, each tethered at its end to a nuclear matrix. These loops or chromatin domains are the functional units of transcriptional competence in that each independent domain can be assembled either as active chromatin or as inactive chromatin. Our long- term objective is to understand the function of chromatin domains and the elements that comprise them so that we may acquire a complete understanding of the regulation of gene expression and how it may be manipulated for basic research, medical (gene and cancer therapy) and commercial (drug production and delivery) applications. Most studies have focused on regulatory elements in the immediate vicinity of genes, including promoter and enhancer elements, proteins binding to these, and DNA modification around the promoter. It is now abundantly clear that crucial control elements often lie as much as several hundred kilobases from a gene within a structurally-defined unit which we have referred to as the chromatin domain. Few such functional domains have been characterized in any detail, and when they have the approach has generally has generally been a search for one or a few types of regulatory elements. A thorough understanding requires that we map a number of domains in detail, using a battery of methods to identify all of the potentially important regulatory elements in those domains. Although a matrix attachment region, a replication origin and an insulator, capable of dampening the interaction between a promoter in one domain and an enhancer in another, could co-occur in some instances, these elements may still be functionally distinct and not necessarily overlapping. It is only by mapping out all such elements in a number of cases that we expect to make these important distinctions. The detailed mapping of domains will allow us to identify the crucial regulatory elements in those domains, in preparation for future studies in which we plan to test these elements in experimental systems, or to delete or alter them. The initial mapping we propose here will not only allow us to understand and manipulate these particular domains better, but will hopefully reveal general features that will accelerate our ability to map other domains as well.
Showing the most recent 10 out of 114 publications
