A wide range of proteins have been reported to condensate into a dense liquid phase, forming a reversible droplet state. Failure in the control of the droplet state can lead to the formation of the more stable amyloid state, which is often disease-related. These observations prompt the question of how many proteins can undergo liquid–liquid phase separation. Here, in order to address this problem, we discuss the biophysical principles underlying the droplet state of proteins by analyzing current evidence for droplet-driver and droplet-client proteins. Based on the concept that the droplet state is stabilized by the large conformational entropy associated with nonspecific side-chain interactions, we develop the FuzDrop method to predict droplet-promoting regions and proteins, which can spontaneously phase separate. We use this approach to carry out a proteome-level study to rank proteins according to their propensity to form the droplet state, spontaneously or via partner interactions. Our results lead to the conclusion that the droplet state could be, at least transiently, accessible to most proteins under conditions found in the cellular environment.

Nine neurodegenerative diseases are caused by expanded polyglutamine (polyQ) tracts in different proteins, like huntingtin in Huntington’s disease (HD) and ataxin-3 in spinocerebellar ataxia type 3 (SCA3)1, 2. Age-at-onset decreases with increasing polyglutamine length in these proteins and the normal length is also polymorphic3. PolyQ expansions drive pathogenesis in these diseases, as isolated polyQ tracts are toxic, and an N-terminal huntingtin fragment comprising exon 1, which occurs in vivo due to alternative splicing4, causes toxicity. While such mutant proteins are aggregate-prone5, toxicity is also associated with soluble forms of the proteins6. The function of the polyQ tracts in many normal/wild-type cytoplasmic proteins is unclear. One such protein is the deubiquitinating enzyme ataxin 37, 8, which is widely expressed in the brain9, 10. Here we show that the polyQ domain in wild-type ataxin-3 enables its interaction with beclin 1, a key autophagy initiator11. This interaction allows the deubiquitinase activity of ataxin-3 to protect beclin 1 from proteasome-mediated degradation and thus enables autophagy. Starvation-induced autophagy, which is regulated by beclin 1, was particularly inhibited in ataxin-3-depleted human cell-lines, primary neurons and in-vivo. This activity of ataxin-3 and its interaction with beclin 1 mediated by its polyQ domain was competed by other soluble proteins with polyQ tracts in a length-dependent fashion. This resulted in impaired starvation-induced autophagy in cells expressing mutant huntingtin exon 1, which was also recapitulated in the brain of HD mouse model and in patient cells. A similar phenomenon was also seen with other polyQ disease proteins, including mutant ataxin-3 itself. Our data thus describe a specific function for a wild-type polyQ tract which is abrogated by a competing longer polyQ mutation in a disease protein. This also reveals a deleterious function of such mutations distinct from their aggregation propensity.

Misfolded α-synuclein is a major component of Lewy bodies, which are a hallmark of Parkinson’s disease. A large body of evidence shows that α-synuclein can aggregate into amyloid fibrils, but the relationship between α-synuclein self-assembly and Lewy body formation remains unclear. Here we show, both in vitro and in a Caenorhabditis elegans model of Parkinson’s disease, that α-synuclein undergoes liquid‒liquid phase separation by forming a liquid droplet state, which converts into an amyloid-rich hydrogel with Lewy-body-like properties. This maturation process towards the amyloid state is delayed in the presence of model synaptic vesicles in vitro. Taken together, these results suggest that the formation of Lewy bodies may be linked to the arrested maturation of α-synuclein condensates in the presence of lipids and other cellular components.

The WF-NTP extends significantly the power of existing automated platforms by combining enhanced optical imaging techniques with an advanced software platform. We anticipate that this approach will further extend the scope and utility of C. elegans as a model organism.