Colorado State University

Numerous diseases, including Parkinson's disease, Alzheimer's disease, late-onset diabetes, and the transmissible spongiform encephalopathies, are associated protein aggregation. We are using budding yeast to study the basic steps involved in the formation of infectious protein aggregates (“prions”). In yeast, a growing list of proteins have the demonstrated to form prions. Many of these prion proteins contain a glutamine/asparagine-rich prion domain. Such glutamine/asparagine-rich regions are common in both the yeast and human genome, but predicting whether a given glutamine/asparagine-rich protein will form amyloid aggregates has proven difficult. We therefore have worked to develop improved prediction methods to predict prion activity.

The prevelance of glutamine/asparagine-rich prion-like domains also raises questions about the normal functions of these domains. While protein aggregation is often pathological, a growing body of research suggests that the reversible protein assembly in response to cellular stress can actually improve cell survival. Many of the proteins that form these reversible assemblies contain prion-like domains, which are themselves thought to contribute to assembly. Furthermore, mutations within prion-like domains involved in functional assembly can lead to the formation of pathological aggregates. Therefore, our lab is interested in understanding how assembly of prion-like domains can be harnessed and regulated by cells to promote cellular adaptation to unfavorable conditions, and how mutations can disrupt this process.

Finally, certain yeast prion domains can be completely scrambled (while preserving the amino acid composition) and still retain the ability to form infectious aggregates. This suggests that unlike most protein folding events, yeast prion formation is driven primarily by amino acid composition, not primary sequence. can misfold into their infectious form in a primary sequence-independent manner. This has led us to examine the role of amino acid composition in other cellular and molecular processes. We have developed a novel bioinformatic approach to link amino acid composition to protein fates and functions on a proteome-wide scale. Using this approach, we are currently exploring relationships between amino acid composition and the fundamental aspects of a protein life cycle (synthesis, abundance, and degradation), as well as defining the compositional features associated with proteins involved in known composition-driven processes.