Structural biology of ATP-dependent chromatin remodeling

The extreme degree of compaction of DNA into chromatin would seem to create an insurmountable barrier to all DNA transactions. However, a variety of mechanisms have evolved that allow cells to turn this apparent impediment into a highly dynamic and finely tuned regulatory device that controls the accessibility of specific DNA sequences. A multitude of players are involved in achieving this regulation. A number of protein factors are involved in covalently modifying chromatin—either DNA itself or the histone octamer around which DNA is wrapped. Another group, the ATP-dependent chromatin remodeling complexes, utilizes the energy from ATP hydrolysis to non-covalently modify nucleosomes. These remodelers are large (often above 1 MDa) multi-subunit complexes that are conserved from yeast to humans and their activity leads to a number of different fates for the remodeled nucleosomes: sliding along DNA, transfer to a different DNA segment, formation of a “loose” nucleosome structure and exchange of histone components. Despite a significant body of work describing these outcomes we do not yet understand mechanistically how ATP-dependent chromatin remodelers actually remodel chromatin.

Our group is approaching this question from a structural point of view. We use primarily three-dimensional electron microscopy (3D EM) to study the ATP-dependent chromatin remodeling complex RSC (Remodels the Structure of Chromatin) from S.cerevisiae. This complex has been extensively characterized biochemically and genetically, making it an ideal model system. Electron microscopy is a powerful structural technique to study large macromolecular assemblies that are not amenable to more traditional methods either due to their low abundance (when purified from natural sources, for example) or refractory to crystallization (due to their intrinsic flexibility). More importantly, electron microscopy of vitrified samples (cryo-EM) holds the promise of allowing us to image—at high resolution and under physiological conditions—the conformational diversity present in and potentially central to the function of many macromolecular assemblies.

The long-term goal of our research is to visualize the entire remodeling mechanism at high resolution in order to understand its chemical underpinnings. In the meantime, we are starting to dissect RSC at lower resolution. We have obtained an initial reconstruction of the complex and are in the process of obtaining one of RSC bound to its substrate, the nucleosome. At the same time, we will build a map of the positions of a number of the 15 different subunits that comprise RSC by tagging them and probing their location; this will help us better understand their biological role within the complex. The group is also interested in what remains one of the major challenges in 3D EM—obtaining initial models of novel samples. We recently developed a new reconstruction method (Orthogonal Tilt Reconstruction) that addresses some of the problems and limitations of existing approaches. We are interested in further developing this method and eventually automating it to make the generation of initial models a robust and routine aspect of 3D EM.
home // research // people // publications // contact // intranet // MCB // Harvard