 |
Protein Interactions in Health and Disease
Summary: David Eisenberg focuses on protein interactions. In his experiments he studies the structural basis for conversion of normal proteins to the amyloid state and conversion of prions to the infectious state. In bioinformatic work, he derives information on protein interactions from genomic and proteomic data. His goal is to learn the networks of interacting proteins in cells.
The Structure of Amyloid and Prions
Amyloid and prion diseases are diseases of protein conformation in which a normal, functional protein converts to an abnormal, aggregated, fibrillar state. Amyloid and prion fibrils are often associated with fatal diseases and are always characterized by a signature x-ray diffraction pattern, the "cross-β" pattern. Our goal is to understand the general features of the conversion, including what accounts for the transmissibility of these diseases and why there is a barrier between species, which sometimes can be penetrated. In 2005, we determined the first atomic-level structure for the cross-β spine of amyloid. This structure shows that the spine consists of two parallel β sheets, packed across a tight, dry interface that we call a steric zipper. The structure of the spine explains the stability of the amyloid, gives hints about the conversion process and species barrier, and also suggests why some proteins may form amyloid while others do not. In 2006, using joint bioinformatics and biochemical experimentation, we have determined 30 protein segments from fibril-forming proteins that themselves form both fibrils and microcrystals that have permitted structural determination of 13 more cross-β spines, revealing variations in amyloid structure.
Protein Interactions
The lives of cells are controlled by the networks of their macromolecular interactions. These interactions are the basis of the many metabolic and signaling pathways of the cell, and the molecular machines that carry out the work of the cell. We have inferred many of these pathways and complexes from analysis of fully sequenced genomes, using methods such as phylogenetic profiles, Rosetta Stone, and gene neighbor. A Prolinks database has been created for making this information generally available, and the inferred functional linkages in 330 genomes are available on the Web at http://mysql5.mbi.ucla.edu/cgi-bin/functionator/pronav. We have also devised methods for visualization and interpretation of the functional linkages, such as the clustered genome-wide functional linkage map. In recent work, we have extended the method of phylogenetic profiles from the study of pairs of proteins to the study of triplets, offering deeper insights into metabolic pathways.
We have gathered experimental reports of protein interactions in the Database of Interacting Proteins http://dip.doe-mbi.ucla.edu/. This open resource offers information on some 50,000 reports of protein interactions.
This work has also been supported by grants from the Department of Energy, the National Institutes of Health, and the National Science Foundation.
Last updated: August 10, 2006
|
|
|
|
