3D Reconstruction of Inherently Flexible Macromolecular Assemblies.

University Of Vermont & State Agricultural College, Burlington VT

Investigators

Abstract

This research aims to develop new methods for carrying out the 3D reconstruction of flexible biological macromolecules, and determining their range of structural conformations. Biological macromolecules are highly regulated nanomachines that are dynamic and often show considerable flexibility while performing their functions. Knowing the full range of the conformations that molecules may take during a functional cycle is essential for understanding those functions. Methods for preparing biomolecules for imaging using electron microscopy can preserve this rich conformational landscape, allowing researchers to study the full range of structures. Improvements in electron microscope technology allow researchers to obtain resolutions as low as 0.5 nm when macromolecules have one or only a few conformations. However, the more flexible the macromolecule the more conformations it can take, and it has not been possible to obtain a similar level of resolution of molecules having global or continuous structure variations - this research aims to fill that gap. In this project we will develop multivariate statistical methods to facilitate the exploration of structural variations covering the full conformational space. We will implement highly adaptable non-linear dimensionality reduction techniques for applying to the results of established multivariate statistical image analysis techniques, or directly to the data. After sorting data according to similarity, a second step is necessary to determine the conformational spectrum. Sufficiently homogeneous subsets of projections or volumes are extracted and can be used for 3D structure determination, either through 3D reconstruction from projections or sub-tomogram averaging. For efficiently obtaining high resolution structures, these techniques will be combined with a new streamlined method for the correction of the microscope contrast transfer function (CTF) addressing specifically the continuously varying CTF in images of tilted specimens. The new methods will be applied to two biologically relevant systems: Complex I and EmaA. Complex I, the first and largest enzyme in the respiratory chain of mitochondria and bacteria, is essential for the survival of all aerobic living cells. However, its function is still little understood. EmaA is a very large glycosylated bacterial adhesin that binds to collagen during the first and essential step of the infectious process. The new tools developed will provide a timely, reliable and robust methodology to discern the different conformations of biological macromolecules and will be applicable to large number of biological complexes. The reconstruction of Complex I will contribute to the understanding of this enzyme, central to cellular energy metabolism. Knowledge of the structure of bacterial adhesins will help resolve between the existing paradigms explaining the role of glycosylation in bacterial adhesion and thus will help advance the understanding of key elements of the infectious process. These two biological systems are not unique, and the methods developed here will help advance the understanding of the structure function of an abundant number of macromolecules, whose function is still little understood. The results of the project will be presented at national and international meetings, and published in peer-reviewed journals. All software developed will be documented and made freely available to the scientific community under an open source license. Solved structures will be deposited in the EMDataBank.