We report the effect of neutral macromolecular crowders poly(ethylene glycol) (PEG) (8 kDa) and Ficoll (70 kDa) on liquid-liquid phase separation in a polyuridylic acid (polyU)/spermine complex coacervate system. The addition of PEG decreased both the amount of spermine required for phase separation and the coacervation temperature (T). We interpret these effects on phase behavior as arising due to excluded volume and preferential interactions on both the secondary structure/condensation of spermine-associated polyU molecules and on the association of soluble polyU/spermine polyelectrolyte complexes to form coacervate droplets. Examination of coacervates formed in the presence of fluorescently-labeled PEG or Ficoll crowders indicated that Ficoll is accumulated while PEG is excluded from the coacervate phase, which provides further insight into the differences in phase behavior. Crowding agents impact distribution of a biomolecular solute: partitioning of a fluorescently-labeled U15 RNA oligomer into the polyU/spermine coacervates was increased approximately two-fold by 20 wt% Ficoll 70 kDa and by more than two orders of magnitude by 20 wt% PEG 8 kDa. The volume of the coacervate phase decreased in the presence of crowder relative to a dilute buffer solution. These findings indicate that potential impacts of macromolecular crowding on phase behavior and solute partitioning should be considered in model systems for intracellular membraneless organelles.

We report the formation of coacervate-supported phospholipid membranes by hydrating a dried lipid film in the presence of coacervate droplets. Coacervate-supported membranes were characterized by fluorescence imaging, polarization, fluorescence recovery after photobleaching of labeled lipids, lipid quenching experiments, and solute uptake experiments. Our findings are consistent with the presence of lipid membranes around the coacervates, with many droplets fully coated by what appear to be continuous lipid bilayers. In contrast to traditional giant lipid vesicles formed by gentle hydration in the absence of coacervates, the coacervate-templated membrane vesicles are more uniform in size, shape, and apparent lamellarity. Due to their fully coacervate model cytoplasm, these simple artificial cells are macromolecularly crowded and can be easily pre-loaded with high concentrations of proteins or nucleic acids. Within the same population, in addition to coacervate droplets having intact lipid membrane coatings, other coacervate droplets are coated with membranes having defects or pores that permit solute entry, and some are coated with multilayered membranes. Membranes surrounding protein-based coacervate droplets provided protection from a protease added to the external solution. The simplicity of producing artificial cells having a coacervate model cytoplasm surrounded by a model membrane is at the same time interesting as a potential mechanism for prebiotic protocell formation and appealing for biotechnology. We anticipate that such structures could serve as a new type of model system for understanding interactions between intracellular phases and cell or organelle membranes, which are implicated in a growing number of processes ranging from neurotransmission to signaling.

No abstract