Computational analysis of the S. cerevisiae proteome reveals the function and cellular localization of the least and most amyloidogenic proteins

Abstract

Protein sequences have evolved to optimize biological function that usually requires a well-defined three-dimensional structure and a monomeric (or oligomeric) state. These two requirements may be in conflict as the propensity for beta-sheet structure, which is one of the two most common regular conformations of the polypeptide chain in folded proteins, favors also the formation of ordered aggregates of multiple copies of the same protein (fibril, i.e., polymeric state). Such beta-aggregation is typical of amyloid diseases that include Alzheimer's, Parkinson's, and type II diabetes as well as the spongiform encephalopathies. Here, an analytical model previously developed for evaluating the amyloidogenic potential of polypeptides is applied to the proteome of the budding yeast (Saccharomyces cerevisiae). The model is based on the physicochemical properties that are relevant for beta-aggregation and requires only the protein sequence as input. It is shown that beta-aggregation prone proteins in yeast are accrued in molecular transport, protein biosynthesis, and cell wall organization processes while they are underrepresented in ribosome biogenesis, RNA metabolism, and vitamin metabolism. Furthermore, beta-aggregation prone proteins are much more abundant in the cell wall, endoplasmic reticulum, and plasma membrane than in the nucleolus, ribosome, and nucleus. Thus, this study indicates that evolution has not only prevented the selection of amyloidogenic sequences in cellular compartments characterized by a high concentration of unfolded proteins but also tried to exploit the beta-aggregated state for certain functions (e.g. molecular transport) and in well-confined cellular environments or organelles to protect the rest of the cell from toxic (pre-)fibrillar species.