57 Nucleosome [mis]positioning, chromatin folding, and a computational karyotype

Abstract

Given a set of experimentally or theoretically determined nucleosome positions, it is possible to rapidly construct and interactively display 3D models of entire chromosomes. Our Interactive Chromatin Modeling Web (ICM-Web) server can fold and display megabase segments of chromatin in real time (Stolz & Bishop, 2010). These models are first order approximations that assume each nucleosome is a canonical octasome and that the linker DNA assumes a sequence-specific conformation similar to free DNA. Thermal fluctuations are included in the model that alters the nucleosome wrapping (i.e. entry/exit angle) and linker conformation. The models provide valuable insights, even if the chromatin folds are not necessarily accurate, e.g. visualization of DNA rotational phasing on individual nucleosomes and how DNA defects alter global structure. The DNA defects arise from known sequence-specific conformations and thermal fluctuations of DNA. Nucleosome positioning or mispositioning alters chromatin topology by exposing or hiding DNA defects. We utilize the CHA1, MFA2, HIS3, PHO5, and MMTV promoter complexes to illustrate these ideas. The six positioned nucleosomes in the MMTV promoter complex yield a much more extended structure is typically represented in literature. A compact chromatin structure for the MMTV requires more than six nucleosomes and thus mispositioning of some nucleosomes. Nucleosome [mis]positoning hides or exposes a DNA defect in the MMTV that may regulate chromatin looping associated with this promoter. Using whole genome nucleosome positioning data, we generate chromatin folds for each chromosome of Saccharomyces cerevisiae to produce a computational karyotype. Thermal motion associated with linker DNA plays a critical role in chromatin flexibility, reduces the spatial range of each fiber, and allows it to be compacted into the nucleus. ICM is available on the Chromatin Folding tab at http://www.latech.edu/~bishop.