Enhanced protein encapsulation in polymer vesicles using hydrophobic membrane anchoring peptides
Hollow, spherical vesicles formed by a (mono- or bi-) layer of polymer molecules (polymersomes) have attracted attention for their versatile use in biotechnology. In combination with enzymes, polymersomes have been turned into functional vesicles that can e.g. respond to stimuli to act as sensors , specifically release encapsulated molecules  and are biocatalysts in multi-step reactions [3-5]. In the field of synthetic biology, these vesicles are a versatile chassis for the design of minimal cellular systems . Their robustness and low permeability (in comparison to e.g. liposomes) allow the assembly of compartmentalized systems, resembling the organizational principles of eukaryotic cells.
Polymersomes can be formed from diverse types of amphiphilic block copolymers. A triblock copolymer that has been extensively studied in biotechnology and synthetic biology is poly(2-methyloxazoline)-poly(dimethylsiloxane)-poly(2-methyloxazoline) (PMOXA-PDMS-PMOXA), as this material shows good biocompatibility and allows the functional integration of transmembrane proteins, such as proton pumps or porins, to mimic cellular functions. Whereas natural cells and organelles deal with molecular crowding, it is not trivial to encapsulate high concentrations of macromolecules in polymersomes. If PMOXA-PDMS-PMOXA vesicles are formed by the co-solvent method, i.e. injection of an organic solvent with dissolved polymer into an aqueous buffer, maximum protein concentrations of ~1.5 g L-1 are tolerated before the polymersome quality is impaired. From this it follows that the statistical encapsulation of proteins during the vesicle formation process is very inefficient resulting in the entrapment of, e.g., only 5 molecules of a 50 kDa protein in polymersomes with a diameter of ~100 nm.
In order to overcome this bottleneck, we have developed a new method for protein encapsulation, in which the proteins to be encapsulated are concentrated locally in an interaction with the polymer membrane. For this purpose, the proteins of interest are genetically fused to membrane anchoring domains. This allowed to increase luminal polymersome loadings by a factor of 20.
Since PMOXA-PDMS-PMOXA polymersomes have a symmetric membrane, membrane anchoring peptides, when incorporated during vesicle formation, are randomly distributed between the inner and outer surfaces of the membrane (Fig. 1). Therefore, a tobacco etch virus (TEV) protease recognition sequence or an intein domain were incorporated into the fusion proteins to enable the release of protein domains from the outer and/or inner polymer membrane, respectively. This is not only of interest for specific release but also to avoid crowding on the outer surface if other proteins are to be immobilized later. Further, immobilization can exert negative effects on protein functions, such as a reduction of the activity of enzymes. Therefore, the release of active domains in the vesicle lumen is enabled by intein splicing. Although polymersome technology has great potential for use in synthetic biology, the methods for the assembly of cell-like systems still need to be further developed in order to come close to the physicochemical conditions of cells and organelles. Our new method for directed protein encapsulation contributes to this and opens up new possibilities for the construction of functionalized polymersomes