Nucleation of Biomolecular Condensates from Finite-Sized Simulations

Area of Science:

• Biophysics
• Computational Biology
• Protein Science

Background:

• Protein condensate nucleation is concentration-dependent and influenced by finite-size effects in simulations.
• Understanding condensate thermodynamics and kinetics is crucial for biological processes.

Purpose of the Study:

• To develop a general and efficient method for determining macroscopic properties of protein condensates from finite-sized nucleation simulations.
• To estimate thermodynamic and kinetic parameters like equilibrium density, surface tension, and nucleation barriers.

Main Methods:

• Theoretical description of droplet nucleation in the canonical ensemble.
• Application to coarse-grained simulations of NDDX4 and FUS-LC proteins.
• Analysis of finite-size effects to extrapolate to the macroscopic limit.

Main Results:

• Developed a method to obtain ensemble properties of protein condensates in the macroscopic limit.
• NDDX4 condensates exhibit lower surface tension, higher solubility, and faster dynamics than FUS-LC condensates.
• NDDX4 condensates form with negligible nucleation barriers, whereas FUS-LC condensates form via an activated process.

Conclusions:

• The presented method accurately predicts macroscopic condensate properties from finite simulations.
• Physicochemical differences between NDDX4 and FUS-LC dictate distinct nucleation pathways and barriers.
• This work provides insights into the phase separation behavior of disordered proteins.