Prion-like proteins and the formation of liquid-like compartments

Proteins containing intrinsically disordered domains of low sequence complexity (also called prion-like proteins when they mostly consist of polar amino acids) are frequently found in association with compartments. However, what are the molecular properties of these proteins and what is their function? We proposed that these proteins promote the formation of compartments by liquid-liquid phase separation (Malinovska et al., 2013). We further argued that disordered prion-like domains promote multivalent and transient interactions, which are required for the formation of liquid-like compartments. The presence of conformationally flexible regions, however, comes with a cost, for many prion-like proteins are aggregation-prone and cause diseases (Malinovska et al., 2013).

Indeed, in collaboration with Tony Hyman at the MPI-CBG we could recently demonstrate, that the ALS-associated prion-like protein FUS forms dynamic compartments by liquid-liquid demixing in vitro and in vivo (Patel et al., 2015). We further found that liquid droplets assembled from patient-derived FUS show biophysical abnormalities with time and transition into an aberrant aggregated state when aged in vitro (Figure 1). These findings provide important insight into the molecular principles of compartment formation by prion-like proteins and suggest a molecular explanation for why these proteins are frequently associated with age-related diseases.

<b>Figure 1.</b> The prion-like protein FUS forms liquid compartments that convert with time to an aberrant fibrous state. (A) The mutant variants of FUS indicated above were purified from insect cells and subjected to an in vitro “aging” experiment for the indicated times. The G156E and R244C mutations are derived from ALS patients. (B) The conversion of wild type FUS is concentration-dependent, providing a link to frequently found mutations in the nuclear localization of FUS. These mutations are believed to increase the cytoplasmic concentration of FUS, thus increasing the probability for conversion.

Future Plans:

To find out whether our findings extend to other proteins, we aim to investigate the phase behavior of additional disease-associated prion-like proteins. We will also study the functional role of regulators of compartment formation such as polyADP-ribose and RNA. In addition, we will investigate how post-translational modifications (phosphorylation, arginine methylation) regulate liquid compartments. Finally, we will investigate the link between liquid compartments and age-related protein misfolding diseases using more relevant systems such as neuronal cells derived from patient iPS cells. By doing do, we expect to gain important insight into pathways underlying neurodegenerative diseases.