Saturday, February 18, 2012
Exhibit Hall A-B1 (VCC West Building)
BACKGROUND: Aggregation of misfolded proteins is a hallmark of many neurodegenerative diseases. Although amyloids and prions - characterized by cross-beta sheet structures, hydrophobic stretches, and glutamine-rich regions - have been identified in these insoluble deposits, these proteins represent only a subset of the proteins that aggregate. Here, we used a systems-biology approach to identify misfolding-prone proteins and to determine characteristics that affect aggregation propensity. METHODS: Given the wealth of genome-scale data from the model yeast organism S. cerevisiae, we adopted this systems-wide approach to analyze proteins that become insoluble or are targeted for proteolysis after heat-shock induced misfolding in the cell. Using quantitative mass spectrometry, we identified 240 misfolding-prone proteins and a reference set of 772 unaffected proteins. We developed a bioinformatics platform to integrate previous genome-scale studies and used computational prediction methods to analyze differences in molecular features. RESULTS: Partitioning the proteome into two classes described protein’s propensity to misfold: highly structured versus intrinsically disordered proteins. For structured proteins, misfolding propensity was independent of the complexity of the structure; instead, thermodynamic stability was a major determinant. In contrast, disordered proteins lack well-defined structures and misfolding was associated with a significant bias towards polar and charged residues. Misfolding of disordered proteins correlated with protein binding regions, where non-specific binding may aggravate non-native interactions. Proteins susceptible to misfold were also tightly regulated; their abundance was maintained at levels below most other proteins and many are not essential for viability. CONCLUSIONS: Our study identifies molecular properties that account for a protein’s propensity to misfold and suggests pathways used to minimize cellular damage. Understanding these features may explain the pathogenesis of many neurodegenerative diseases.