Phase Separation in Mixtures of Prion-Like Low Complexity Domains is Driven by the Interplay of Homotypic and Heterotypic Interactions

Abstract

Prion-like low-complexity domains (PLCDs) are involved in the formation and regulation of distinct biomolecular condensates that form via coupled associative and segregative phase transitions. We previously deciphered how evolutionarily conserved sequence features drive phase separation of PLCDs through homotypic interactions. However, condensates typically encompass a diverse mixture of proteins with PLCDs. Here, we combine simulations and experiments to study mixtures of PLCDs from two RNA binding proteins namely, hnRNPA1 and FUS. We find that 1:1 mixtures of the A1-LCD and FUS-LCD undergo phase separation more readily than either of the PLCDs on their own. The enhanced driving forces for phase separation of mixtures of A1-LCD and FUS-LCD arise partly from complementary electrostatic interactions between the two proteins. This complex coacervation-like mechanism adds to complementary interactions among aromatic residues. Further, tie line analysis shows that stoichiometric ratios of different components and their sequence-encoded interactions jointly contribute to the driving forces for condensate formation. These results highlight how expression levels might be tuned to regulate the driving forces for condensate formation in vivo. Simulations also show that the organization of PLCDs within condensates deviates from expectations based on random mixture models. Instead, spatial organization within condensates will reflect the relative strengths of homotypic versus heterotypic interactions. We also uncover rules for how interaction strengths and sequence lengths modulate conformational preferences of molecules at interfaces of condensates formed by mixtures of proteins. Overall, our findings emphasize the network-like organization of molecules within multicomponent condensates, and the distinctive, composition-specific conformational features of condensate interfaces.

Sequence-specific interactions determine viscoelasticity and aging dynamics of protein condensates

“…We focus on condensates formed by intrinsically disordered PLCDs [11][12][13] . These domains are prominent drivers or modulators of several functionally relevant and pathogenically significant condensates [14][15][16] . The stickers-and-spacers framework 17 has been useful for modeling and explaining the connections between sequence-encoded information and the driving forces for condensate formation by PLCDs [11][12][13] .…”

“…In cellular measurements of PLCDs that form solids, there is a preference for high numbers of Phe stickers and Ser residues as spacers 47 . Finally, Trp is used sparingly as a sticker in PLCDs 12,14 . Taken together, these trends suggest an evolutionarily encodable preference whereby certain PLCDs form condensates that are metastable Maxwell fluids.…”

Heterotypic interactions in the dilute phase can drive co-condensation of prion-like low-complexity domains of FET proteins and mammalian SWI/SNF complex

“…S10). These observations suggest that heterotypic interactions are highly cooperative with homotypic interactions in driving phase separation of the PLD mixtures 5 . In the co-PLD phase diagram (Fig.…”

“…The perceived physiological functions of biomolecular condensates range from signaling hubs under normal conditions to storage depots in response to cellular stress 3 . Sequence analyses of the proteins enriched in intracellular condensates both in the nucleus and cytoplasm have previously revealed an abundance of proteins containing long stretches of intrinsically disordered prion-like domains 4,5 . Prion-like domains (PLDs) are typically characterized by their low complexity sequence features with an overrepresentation of aromatic (Y/F) and polar amino acids (G/S/Q/N) and depletion of charged residues 6,7 .…”

“…Many intracellular biomolecular condensates, such as stress granules and transcriptional hubs, are known to contain a multitude of proteins containing prion-like low complexity domains 4,5,32 . Despite being broadly classified as prion-like based on the frequencies of certain amino acids in a protein sequence 6,7 , individual PLD chains typically feature distinct sequence composition, amino acid patterning, and chain length 4,22,33 .…”