MICROFLUIDIC PRODUCTION OF BIOFUNCTIONALIZED GIANT UNILAMELLAR VESICLES FOR TARGETED CARGO DELIVERY

Abstract

The present invention relates to a method for preparation of monodisperse cell-targeting giant unilamellar vesicles based on symmetrically division of a parent polymer shell-stabilized giant unilamellar vesicle into smaller polymer shell-stabilized giant unilamellar vesicles with a diameter between 1 μm and 10 μm using a microfluidic splitting device. The inventive method allows preparation of differently charged giant unilamellar vesicles as well as bioligand- and PEG-conjugated giant unilamellar vesicles, which are useful for targeted cellular delivery at high efficiency and specificity. A further advantage of the present invention is that the giant unilamellar vesicles can deliver huge cargos such as drug releasing porous microparticles, high amounts of in vivo imaging probes, viruses, or up-and-coming DNA origami robots.

Claims

1. A method for preparation of monodisperse cell-targeting giant unilamellar vesicles comprising: a) providing a polymer shell-stabilized giant unilamellar vesicle of diameter comprised between 1 μm and 100 μm, b) mechanically symmetrically dividing the polymer shell-stabilized giant unilamellar vesicle into two smaller polymer shell-stabilized giant unilamellar vesicles without harming the giant unilamellar vesicle by using a microfluidic device comprising a multi-Y-shaped division zone (7), c) repeating b) by mechanically symmetrically dividing the smaller polymer shell-stabilized giant unilamellar vesicles provided in b) until polymer shell-stabilized giant unilamellar vesicles reach a desired diameter between 1 and 10 μm, and d) optionally removing the polymer shell from the polymer shell-stabilized giant unilamellar vesicles obtained in c), wherein monodisperse means that the vesicles are of uniform size showing a coefficient of variation in size lower than 16%, wherein the diameter is measured by confocal microscopy, and wherein symmetrically dividing means that the change of molar percentage of the smaller vesicles is less than 5%, and that the change of the luminal content of the vesicles is lower than 20%, wherein said change of the luminal content is calculated as standard deviation/mean fluorescence*100.

2. The method according to claim 1, wherein the polymer shell-stabilized unilamellar vesicle provided in a) is obtained by: a′) merging a water phase comprising at least one lipid, and an oil phase comprising a surfactant of Formula (I): ##STR00034## wherein m is comprised between 5 and 150, and wherein n is comprised between 5 and 450, and wherein the oil phase consists of a solution of perfluorinated water-immiscible solvents, to form a polymer shell stabilized giant unilamellar vesicle, and a″) optionally integrating one or more proteins or fragments thereof into the polymer shell stabilized giant unilamellar vesicle provided in a′).

3. The method according to claim 1, wherein the polymer shell-stabilized unilamellar vesicle provided in a) is obtained by: a′) merging a water phase comprising at least one lipid and cations, and an oil phase comprising an amphiphilic copolymer to form a polymer shell stabilized giant unilamellar vesicle, wherein the oil phase consists of a solution of perfluorinated water-immiscible solvents; and a″) optionally integrating one or more proteins or fragments thereof into the polymer shell stabilized giant unilamellar vesicle provided in a′).

4. The method according to claim 1, wherein d) comprises removing the polymer shell from the polymer shell-stabilized giant unilamellar vesicles obtained in c) by adding a destabilizing agent, wherein the destabilizing agent is a demulsifier surfactant able to destabilize the structure of the polymer shell.

5. The method according to claim 1, further comprising e) after d): e) purifying the giant unilamellar vesicles by centrifugation.

6. The method according to claim 2, wherein the water phase of a′) comprises at least one lipid selected from the group comprising: a neutral lipid selected from the group comprising ceramide, sphingomyelin, cephalin, cholesterol, cerebrosides, diacylglycerols, phosphatidylcholines, lysophosphatidylcholines, phosphatidylethanolamines, lysophosphatidylethanolamine, lysoethanolamines, inverted headgroup lipids, sphingosins, sterol-modified phospholipids, ether ester lipids, diether lipids, vinyl ether (plasmalogen); an anionic lipid selected from the group comprising phosphatidic acids, lysophosphatidic acid derivatives, phosphatidylglycerols, lysophosphatidylglycerols, phosphatidylserines, lysophosphatidylserines, phosphatidylinositols, phosphatidylinositolphosphates, cardiolipins, bis(monoacylglycero)phosphate derivatives; a cationic lipid selected from the group comprising dioleyl-N,N-dimethylammonium chloride; N-(2,3-dioleyloxy)propyl)-N,N,N-trimethylammonium chloride; N,N-distearyl-N,N-dimethylammonium bromide; N-(2,3-dioleyloxy)propyl)-N,N,N-trimethylammonium chloride; 30-(N—(N′,N′-dimethylaminoethane)-carbamoyl)cholesterol; 1,2-dimyristyloxypropyl-3-dimethyl-hydroxy ethyl ammonium bromide; 2,3-dioleyloxy-N-[2(sperminecarboxamido) ethyl]-N,N-dimethyl-1-propanaminium trifluoroacetate; dioctadecylamidoglycyl carboxyspermine; N-(2,3-dioleyloxy)propyl)-N,N-dimethylammonium chloride and 1,2-dioleoyl-3-dimethylammonium-propane; a pH-sensitive lipid selected from the group comprising lipid N-(4-carboxybenzyl)-N,N-dimethyl-2,3-bis(oleoyloxy)propan-1-aminium, 1,2-distearoyl-3-dimethylammonium-propane, 1,2-dipalmitoyl-sn-glycero-3-succinate, 1,2-dioleoyl-sn-glycero-3-succinate, N-palmitoyl homocysteine; a photoswitchable lipid; acylglycine derivatives, prenol derivatives, prostaglandine derivatives, glyco-sylated diacyl glycerols, eicosanoid derivatives, (palmitoyloxy)octadecanoic acid derivatives, diacetylene derivatives, diphytanoyl derivatives, fluorinated lipids, brominated lipids, lipopolysaccharides; one of the aforementioned lipids coupled to a functional ligand selected from biotin, N-hydroxysuccinimide (NHS) ester, sulfo-NHS ester, nitrilotriacetic acid-nickel, amine, carboxylic acid, maleimides, dithiopyridinyl, pyridyl disulfide, pyridyldithiopropionate, N-benzylguanine, carboxyacyl, cyanur, folate, square, galloyl, glycan, thiol, arginylglycylaspartic acid, a fluorescent dye molecule, a magnetic resonance imaging reagent, a chelator; and one of the aforementioned lipids coupled to polyethyleneglycol with a molecular weight comprised between 350 and 50,000 g/mol.

7. The method according to claim 2, wherein the water phase of a′) comprises at least one anionic lipid, at least one neutral lipid, and optionally one neutral lipid functionalized with a fluorescent dye molecule.

8. The method according to claim 2, wherein the water phase of a′) comprises at least one cationic lipid, at least one neutral lipid, and optionally one neutral lipid functionalized with a fluorescent dye molecule.

9. The method according to claim 2, wherein the water phase of a′) comprises at least one lipid functionalized with a functional ligand selected from biotin, N-hydroxysuccinimide (NHS) ester, sulfo-NHS ester, nitrilotriacetic acid (NTA)-nickel, amine, carboxylic acid, maleimides, dithiopyridinyl, pyridyl disulfide, pyridyldithiopropionate, Nbenzylguanine, carboxyacyl, cyanur, square, galloyl, thiol; and wherein the method optionally comprises after d): d′) coupling the giant unilamellar vesicles with at least one macromolecule comprising at least one moiety reacting with one of said functional ligands, wherein the macromolecule is selected from the group comprising a carbohydrate, a nucleic acid, a protein or a fragment thereof, a polypeptide, a cell receptor, an imaging probe, a nanoparticle.

10. The method according to claim 2, wherein the water phase of a′) comprises at least one lipid coupled to polyethyleneglycol with a molecular weight comprised between 350 and 50,000 g/mol.

11. The method according to claim 2, wherein the water phase of a′) comprises at least one pH-sensitive lipid at a molar percentage comprised between 20%-80%, or wherein the water phase of a′) further comprises poly-ethylene-imine at a concentration comprised between 2-100 μg/ml.

12. The method according to claim 2, wherein the water phase of a′) further comprises at least one agent selected from the group comprising drug releasing porous particles, molecular imaging agents, diagnostic agents, therapeutic agents, proteins or fragments thereof, polypeptides, peptides, enzymes or fragments thereof, nucleic acids, oligonucleotides, polynucleotides, up-and-coming DNA origami robots, small molecule drugs, virus particles, virus-like particles, microbial antigens, steroids, proteoglycans, lipids, monosaccharides, oligosaccharides, polysaccharides, magnetic particles, nanorods, carbon nanotubes, dentritosomes, polymerosomes, metal nanoparticles and combinations or conjugates thereof.

13. The method according to claim 3, wherein the amphiphilic copolymer of a′) consists of (i) a triblock copolymer comprising two perfluorinated polymer end blocks and one polyether glycol block, or of (ii) a diblock copolymer comprising one perfluorinated polymer end block and a polyether glycol block, wherein the triblock or diblock copolymer is folded so that the perfluorinated polymer end blocks are arranged at the outer side and the polyether glycol block is arranged at the inner side of the polymer shell.

14. The method according to claim 1, wherein b) comprises mechanically dividing said polymer shell stabilized giant unilamellar vesicle into two smaller polymer shell stabilized giant unilamellar vesicles using a microfluidic device comprising a multi-Y-shaped division zone (7) comprising at least one Y-shaped junction, wherein said Y-shaped junction consists of one inlet channel and two outlet channels, and wherein c) comprises repeating the b) by using four or more sequential generations of Y-shaped junctions, wherein the inlet channel of each Y-shaped junction consists of the outlet channel of the previous Y-shaped junction.

15. A microfluidic device for preparing polymer shell stabilized giant unilamellar vesicles having a diameter between 1 and 10 m, wherein said diameter is measured by confocal microscopy, comprising: a multi-Y-shaped division zone (7) and a flow-rate control system, wherein the multi-Y-shaped division zone (7) comprises one or more sequential generations of Y-shaped junctions, wherein each Y-shaped junction consists of one inlet channel and two outlet channels, and wherein the inlet channel of each Y-shaped junction consists of the outlet channel of the previous junction; a stabilization plane (8) to stabilize the divided polymer shell stabilized giant unilamellar vesicles, one outlet channel (9) leading the divided polymer shell stabilized giant unilamellar vesicles to the outlet (10), and one outlet (10) where the divided polymer shell stabilized giant unilamellar vesicles exit out of the microfluidic device.

16. The microfluidic device of claim 15, further comprising a generation zone of a parent polymer shell stabilized giant unilamellar vesicle positioned upstream the division zone, said generation zone comprising: one oil phase inlet (1) introducing an oil phase into the microfluidic device, optionally one oil phase filter structure (2), one or more aqueous phase inlets (3) introducing the aqueous phase(s) into the microfluidic device, optionally one aqueous phase filter structure (4), when the aqueous phase inlets (3) are more than one, one junction (5) of the aqueous phase inlets (3); and a flow-focusing junction (6) consisting of a horizontal inlet channel and two vertical inlet channels, wherein said three inlet channels converge into an outlet channel through a narrow orifice, and wherein said outlet channel is connected to the division zone; wherein said parent polymer shell stabilized giant unilamellar vesicle has a diameter between 1 μm and 100 μm.

Description

DESCRIPTION OF THE FIGURES

[0802] FIG. 1 shows a phase-contrast microscopy image showing the microfluidic device for mechanical fivefold division of the parent polymer shell stabilized giant unilamellar vesicle, allowing formation of 32 daughter polymer shell stabilized giant unilamellar vesicle from a single parent vesicle having 60 μm diameter. Inset shows time course of mechanical division of a droplet at a Y-junction from a higher magnification recording in bright field. Scale bar is 300 μm, scale bar in inset is 60 μm.

[0803] FIG. 2 shows single plane fluorescence confocal microscopy images of mechanically divided vesicles loaded with AlexaFluor 405 (links upper row), vesicles labelled with 1 mol % LissRhod PE (right upper row), green fluorescent protein (links, lower row) and 1 μm fluorescent polystyrene beads (right lower row). Upper right corner shows coefficient of variation (CV) of vesicle mean fluorescence intensity (n=665 single droplets). Scale bar is 40 μm.

[0804] FIG. 3 shows single plane fluorescence confocal microscopy images of giant unilamellar vesicles (20 mol % EggPG, 79 mol % EggPC and 1 mol % LissRhod PE) obtained by mechanical division and released in PBS. Scale bar is 25 μm.

[0805] FIG. 4 Optimization of intraluminal lipid concentration of small unilamellar vesicles for production by mechanical splitting of polymer shell stabilized giant unilamellar vesicles. Polymer shell stabilized vesicles containing lipid concentrations of 12 mM, 6 mM, 3 mM, 1.5 mM and 0.75 mM were mechanically splitted and the number of giant unilamellar vesicles before and after release of the polymer shell was counted to obtain the release efficiencies.

[0806] FIG. 5 Assessment of mechanical stability of giant unilamellar vesicles kept at 4° C. without mechanical agitation and giant unilamellar vesicles shaken at 800 rpm/37° C. Results are shown as average value with SD from three technical replicates.

[0807] FIG. 6 shows representative phase contrast image of mechanically splitted giant unilamellar vesicles incubated with rat embryonic fibroblasts REF52 cells. White arrows indicate single giant unilamellar vesicles. Scale bar is 10 μm.

[0808] FIG. 7 shows attraction quantification of differently charged giant unilamellar vesicles (Zeta-potential as indicated on x-axis) with cell lines of endothelial (MDCK), epithelial (A431D and A431) and adrenal (PC12) origin 24 hours after incubation (results are shown as average value normalized to −31 mV average and SD from three technical replicates).

[0809] FIG. 8 shows representative single plane fluorescence confocal microscopy images and schematic representation of charge-mediated interactions between giant unilamellar vesicles and A431 D cells, 24 hours after incubation and several washing steps. Nuclei (first column) were stained with Hoechst 33342, cell membranes (second column) were stained with WGA-AlexaFluor488 and giant unilamellar vesicles (third column) were visualized by the fluorescent signal of LissRhod-PE lipids incorporated into giant unilamellar vesicles. Respective giant unilamellar vesicle charge is indicated on the left side of the image. Scale bar is 20 μm.

[0810] FIG. 9 shows transmission electron microscopy (TEM) of internalized giant unilamellar vesicles. (a) Horizontal TEM overview of a REF52 cell layer incubated for 16 hours with negatively charged giant unilamellar vesicles. Arrows point to giant unilamellar vesicles within the cytoplasm. Scale bar is 2.5 μm. (b) Higher magnification horizontal TEM view of a giant unilamellar vesicle internalized into a cell. Black arrow point on the giant unilamellar vesicle membrane, cyan arrow points to the endosomal membrane. Scale bar is 200 nm.

[0811] FIG. 10 shows strategies for giant unilamellar vesicles functionalization. (a) Schematic representation and single plane fluorescence confocal microscopy images of triple-functionalized giant unilamellar vesicles produced by a splitting microfluidic device. The fluorescence corresponds to Biotin-binding Atto425-streptavidin, NTA.sup.2+-binding His-tagged GFP and ammine-binding NHS-Alexa647. (b and c) Schematic representations and microscopy images of the sequential giant unilamellar vesicles functionalization with 50 nm gold nanoparticles in (b) and with antibodies in (c). Right panels in (b) and (c) show phase contrast image of gold nanoparticle (indicated by white arrows) linked giant unilamellar vesicles and fluorescence confocal image of AlexaFluor488-linked Anti-CD3 IgG immobilized via His-tagged ProteinG to NTA.sup.2+ functionalized lipids, respectively. Scale bars are 1 μm and 2 μm for (b) and (c), respectively

[0812] FIG. 11 shows formation of an interaction area between anti-CD3 functionalized giant unilamellar vesicles and CD3.sup.+ Jurkat T-cells after incubation for 24 hours. Cell membranes were stained with WGA-AlexaFluor488 (upper left image), giant unilamellar vesicles were visualized by incorporation of LissRhod-PE fluorescent lipids (lower left image) and nuclei in composite were stained with Hoechst 33342 (right image). Arrow indicates site of lipid clustering at the giant unilamellar vesicle-cell interface. Right image shows bright field. Scale bar is 6 μm.

[0813] FIG. 12 shows attraction analysis of giant unilamellar vesicles functionalized with the RGD peptide at increasing concentrations (1-10%) and incubated with different cell lines. The presented results are normalized to the attraction of naive giant unilamellar vesicles without RGD, and shown as mean value±SD from three technical replicates. Attraction of RGD-functionalized giant unilamellar vesicles to suspension Jurkat cells was assessed by flow cytometry.

[0814] FIG. 13 Representative fluorescence confocal microscopy images of fluorescently labeled giant unilamellar vesicles (LissRhod-PE) decorated with 2 mol % RGD ligands and incubated with membrane stained (WGA-AlexaFluor488) A431 D cells. The insets show magnified images with accumulation of giant unilamellar vesicles at the cell periphery and in the perinuclear region observed after incubation for 24 hours. Scale bar is 60 μm.

[0815] FIG. 14 shows comparison of interactions between cells and giant unilamellar vesicle with or without NrCAM on their surface. Attraction values of naive and recombinant NrCAM functionalized non-charged giant unilamellar vesicles incubated for 24 hours with SH—SY5Y cells. Note that no significant increase in attraction can be achieved by functionalization as non-specific lipid-based interactions appear stronger than specific ligand-receptor based attraction.

[0816] FIG. 15 shows PEGylation of giant unilamellar vesicles for giant unilamellar vesicle-cell repulsion. (a) Grey scale representation of Zeta-potentials of small unilamellar vesicles and giant unilamellar vesicles containing different amounts of positively (DOTAP) or negatively (DOPG) charged lipids and decorated with varying amount of PEG of increasing molecular weight. (b) Attraction analysis of giant unilamellar vesicles from a) with six different cell lines.

[0817] FIG. 16 shows regulation of attractive and repulsive giant unilamellar vesicle-cell interactions. Heat map of the attraction analysis between A431 D cells and PEG-functionalized giant unilamellar vesicles containing positively charged lipids at 15 mol %.

[0818] FIG. 17 shows representative bright field (first column) and fluorescence (second column) confocal microscopy images of A431 cells interfaced with naive negatively charged giant unilamellar vesicles (upper row) or 50 mol % PEG1000-functionalized negatively charged giant unilamellar vesicles (lower row). Naive giant unilamellar vesicles are in direct contact with the cells while PEGylated giant unilamellar vesicles accumulate in the intercellular space (merged image, third column) and form contact inhibition zones with the cells (fourth column magnified from areas indicated in merged images). White dotted lines indicate periphery of cell groups deduced form the bright field image. Scale bars are 15 μm.

[0819] FIG. 18 shows attraction of biofunctionalized giant unilamellar vesicles. a) Attraction values of giant unilamellar vesicles decorated with 15 different peptides and proteins (coupled to the giant unilamellar vesicle surface via 1 mol % NHS lipids) for 6 cell lines. All values were normalized to the attraction of BSA coupled giant unilamellar vesicles. b) Attraction values of non-specific (i.e., BSA, poly-L-lysine (PLL), WGA and tat-peptide) and specific (i.e., anti-cadherin, recombinant cadherin and bradykinin) protein-biofunctionalized giant unilamellar vesicles to human fibroblasts BJ cells, neuroblastoma SH—SY5Y cells, endothelial MDCK cells, epithelial A431 cells, dexamethasone treated epithelial transformed A431 cells and Hela cells. All values were normalized to the attraction of BSA-coupled giant unilamellar vesicles.

[0820] FIG. 19 shows schematic representation (upper panel) of PEGylated giant unilamellar vesicles biofunctionalized with NrCAM in co-culture with astrocytes and neuronal cells, differing for the expression of axonin-1 receptor. Lower panel show representative single plane confocal microscopy images of the astrocyte-neuron co-culture following 24 hours of incubation with NrCAM-PEGylated giant unilamellar vesicles. Hs683 astrocytes and SH—SY5Y neurons were stained with CellTracker Green and Blue, respectively. Giant unilamellar vesicles were visualized by incorporation of LissRhod-PE fluorescent lipids. Scale bar is 50 μm.

[0821] FIG. 20 shows lysosomal degradation of the vesicles and the assessment of the mechanisms for lysosomal escape to allow intracellular giant unilamellar vesicle cargo release. (a) Representative fluorescence confocal microscopy images of REF52 cells loaded with endosomal entrapped negatively charged giant unilamellar vesicles. Nuclei (upper left image) were stained with Hoechst 33342, endosomes (upper center image) were labeled by staining with WGA-AlexaFluor488 for 24 hours, giant unilamellar vesicles (lower left image) were visualized by incorporation of LissRhod-PE fluorescent lipids and cytoplasm (lower center image) was stained by CellTracker Blue. Merged image (right image) shows that giant unilamellar vesicles reside within the cytoplasm and are entrapped in endosomal vesicles. Scale bar is 10 μm. (b) Representative fluorescence microscopy images of REF52 cells showing colocalization of giant unilamellar vesicles with lysosomal compartments 24 hours after incubation. Lysosomes (left panel) were stained with LysoTracker Green and giant unilamellar vesicles (center panel) were visualized by incorporation of LissRhod-PE fluorescent lipids. Right panel shows bright field (BF) images. Scale bar is 10 μm.

[0822] FIG. 21 shows representative fluorescence confocal microscopy images of the assessment of poly-ethylene-imine (PEI)-, GALA peptide- and DOBAQ-mediated lysosomal escape for HPTS-loaded giant unilamellar vesicles incubated with REF52 cells for 24 hours. Scale bar is 50 μm.

[0823] FIG. 22 shows dynamic light scattering measurements of giant unilamellar vesicle Zeta-potentials at different pH for giant unilamellar vesicles containing 60 mol % of the pH-sensitive lipid DOBAQ or giant unilamellar vesicles with pH-sensitive lipids. Note the transition of the DOBAQ containing giant unilamellar vesicles from negative to positive Zeta-potentials with the decrease of the pH. Results are shown as mean and SD values from three technical replicates.

[0824] FIG. 23 shows quantification of intracellular giant unilamellar vesicle degradation. A431 D cells were incubated with giant unilamellar vesicles and the giant unilamellar vesicles were quantified over time. While “empty” giant unilamellar vesicles show prompt degradation after uptake, giant unilamellar vesicles loaded with PEI show progressive intracellular accumulation indicating successful escape from lysosomal degradation.

[0825] FIG. 24 shows representative fluorescence confocal microscopy and bright field images of primary mouse hippocampal neurons (right panel) incubated with giant unilamellar vesicles (left panel, visualized by incorporation of 1 mol % LissRhod PE lipids) loaded with HPTS (center panel). Scale bar is 30 μm.

[0826] FIG. 25 shows delivery of heavy duty cargos by the giant unilamellar vesicles. (a) Representative bright field (left panel) and corresponding confocal microscopy (right panel) images of baculovirus-loaded giant unilamellar vesicles. Scale bar is 10 μm. (b) Maximal z-projection of fluorescence confocal microscopy images of REF52 cells (nuclei stained with Hoechst 33342, upper row left image) incubated with baculovirus (lower row left image, by oversaturation of Hoechst 33342 channel) containing DOBAQ carrier giant unilamellar vesicles (upper row right image) for 24 hours. Note expression of mitochondria targeted dsRed (lower row right image). Scale bar is 25 μm.

[0827] FIG. 26 Microfluidic device comprising the generation zone of the parent polymer shell stabilized vesicle, and the splitting units to generate smaller giant unilamellar vesicles of diameter smaller than 10 μm. (1): Oil phase inlet, introducing the oil phase to the device; (2): optional oil filter structure, for retention and filtering of large contaminants like dust; (3): aqueous phase inlet, introducing the aqueous phase(s) into the devices. Optionally, more than one single inlet can be used (here two separate inlets); (4): optional aqueous phase filter structure for large contaminants, for retention and filtering of large contaminants like dust; (5): for multiple aqueous inlets: aqueous inlets junction. Merges the separate aqueous inlets, if only one aqueous inlet is used, not such structure is needed. (6): Flow focusing junction (example of a “Generation Zone”, area of formation of the parent polymer shell stabilized giant unilamellar vesicle). (7): splitting architecture. Sequential mechanical splitting of the parent droplet (“Splitting Unit”). (8): Stabilization plane. Wide area where splitted droplets can stabilize and mix, here the droplets are surrounded by excess oil and enough surfactant is provided to stabilized the small droplets (as splitting leads to a surfaces increase wherefore more surfactant is needed; (9): Outlet channel, leeds the droplets to the outlet; (10): Outlet, where splitted and stabilized polymer shell stabilized giant unilamellar vesicles exit the device.

[0828] FIG. 27 SUV dilution series for total lipid quantification of GUV solutions. An exemplary curve obtained with SUVs containing 1 mol % LissRhod PE lipids at 8 different concentrations is shown. Dotted line is the exponential fit with equation shown in graph. For example, GUVs produced with these SUVs had a fluorescence signal intensity of 5888 corresponding to a concentration of 72 μM.

[0829] FIG. 28 formation of polymer-shell stabilized GUVs by using a negative charged surfactant (A) or a positive charged surfactant (B).

[0830] FIG. 29 fluorescence microscopy pictures of polymer-shell stabilized GUVs produced by using the positive charged surfactant of formula (I) (A), and of GUVs produced by using said positive charged surfactant after release from the polymer shell (B). Scale bar in (A) is 100 μm, and in B is 10 μm.

[0831] The following examples are included to demonstrate preferred embodiments of the invention. It should be appreciated by those of skill in the art that the techniques disclosed in the following examples represent techniques discovered by the inventor to function well in the practice of the invention, and thus can be considered to constitute preferred modes for its practice. However, those of skill in the art should, in light of the present disclosure, appreciate that many changes can be made in the specific embodiments which are disclosed and still obtain a like or similar result without departing from the scope of the invention.

[0832] Further modifications and alternative embodiments of various aspects of the invention will be apparent to those skilled in the art in view of this description. Accordingly, this description is to be construed as illustrative only and is for the purpose of teaching those skilled in the art the general manner of carrying out the invention. It is to be understood that the forms of the invention shown and described herein are to be taken as examples of embodiments. Elements and materials may be substituted for those illustrated and described herein, parts and processes may be reversed, and certain features of the invention may be utilized independently, all as would be apparent to one skilled in the art after having the benefit of this description of the invention. Changes may be made in the elements described herein without departing from the scope of the invention as described in the following claims.

EXAMPLES

Materials

[0833] EggPG L-α-phosphatidylglycerol (Egg, Chicken), EggPC L-α-phosphatidylcholine (Egg, Chicken), 18:1 DOPG 1,2-dioleoyl-sn-glycero-3-phospho-(1′-rac-glycerol), 18:1 DOPC 1,2-dioleoyl-sn-glycero-3-phosphocholesteroline, 18:1 DOPE 1,2-dioleoyl-sn-glycero-3-phosphoethanolamine, LissRhod PE 1,2-dioleoyl-sn-glycero-3-phosphoethanolamine-N-(lissamine rhodamine B sulfonyl), 18:1 DGS-NTA(Ni) 1,2-dioleoyl-sn-glycero-3-[(N-(5-amino-1-carboxypentyl)iminodiacetic acid)succinyl] (nickel salt), 18:1 DOTAP 1,2-dioleoyl-3-trimethylammonium-propane, 18:1-12:0 Biotin PE 1-oleoyl-2-(12-biotinyl-(aminododecanoyl))-sn-glycero-3-phosphoethanolamine, DSPE-RGD 1,2-distearoyl-sn-glycero-3-phosphoethanolamine-N-[4-(p-(cysarginylglycylaspartate-maleimidomethyl)-cyclohexane-carboxamide], 18:1 PEG350 PE 1,2-dipalmitoyl-sn-glycero-3-phosphoethanolamine-N-[methoxy(polyethylene glycol)-350], 18:1 PEG750 PE 1,2-dioleoyl-sn-glycero-3-phosphoethanolamine-N-[methoxy(polyethylene glycol)-750], 18:1 PEG1000 PE 1,2-dioleoyl-sn-glycero-3-phosphoethanolamine-N-[methoxy(polyethylene glycol)-1000] and extrude set with 50 nm pore size polycarbonate filter membranes were purchased from Avanti Polar Lipids, USA. All lipids were stored in chloroform at −20° C. and used without further purification. DyLight 405 NHS Ester, AlexaFluor647-NHS, anti CD3 (16-0038-81), anti CD3-Alexa488 (53-0037-42), Hydroxypyrene-1,3,6-trisulfonic acid trisodium salt (HPTS), Hoechst 33342, CellTracker Blue CMAC dye, CellTracker Green CMFDA dye, LysoTracker Green DND-26 dye, wheat germ agglutinin (WGA)-AlexaFluor conjugates, Dulbecco's Modified Eagle Medium (DMEM) high Glucose, 1:1 DMEM:F12, RPIM-1640, FluoroBrite DMEM (high glucose), heat inactivated fetal bovine serum, penicillin-streptomycin (10,000 U/mL), GlutaMax Supplement, L-Glutamine (200 mM), trypsin-EDTA (0.05%) with phenol red, phosphate buffered saline, Basic fibroblast growth factor (aminoacids 10-155), Epithermal growth factor and AlexaFluor405 dye were purchased from Thermo Fischer Scientific, Germany. NHS Palmitic acid N-hydroxysuccinimide ester, 97% L-cysteine, 50 nm Au nanoparticles, heat-inactivated horse serum, Lectin (Wheat Germ Agglutinin), 1H,1H,2H,2H-perfluoro-1-octanol (PFO) de-emulsifier, Bradykinin, polyethylenimine (branched, Mw-25,000), Atto425-Biotin, human Interleukin 2, recombinant Insulin, fibronectin from bovine plasma, Poly-L-lysine and DOBAQ N-(4-carboxybenzyl)-N,N-dimethyl-2,3-bis(oleoyloxy)propan-1-aminium were purchased from Sigma Aldrich, Germany. Polydimethylsiloxan (PDMS) Sylgard 184 was purchased from Dow Corning, USA. Protein G His-tag was purchased from BioVision, USA. Bovine Albumin fraction V (BSA) was purchased from Carl Roth, Germany. His-tagged NrCAM 8425-NR-050 and human recombinant cadherin was purchased from was purchased from R&D Systems, USA. Fluoresbrite YG Microspheres 1.00 μm were purchased from Polysciences Europe, Germany. Perfluoropolyether-polyethylene glycol (PFPE-PEG) block-copolymer fluorosurfactant was purchased from Ran Biotechnologies, USA. Anti-VE-Cadherin and anti-alpha4-integrin (CD49d) antibodies were purchased from Santa Cruz (Sc-28644) and Millipore (MAB1383). Recombinant human CD95L was purchased from BioLegend, USA. A431, Hela, Hs683, SH—SY5Y and Jurkat cell lines were obtained from ATCC, USA. REF52 cell lines were a generous gift from Prof. Benjamin Geiger (Weizmann Institute Rechovot). PC12 cells were a generous gift from Amin Rustom (Institute for Neurobiology, Heidelberg). Primary mouse hippocampal neurons were obtained from the Institute of Neurobiology, Interdisciplinary Center for Neurosciences in Heidelberg, Germany. Purified Baculovirus were obtained from Martin Pelosse (Commissariat à l'énergie atomique et aux énergies alternatives, France), and were produced as described in Mansouri et al., 2016, Highly efficient baculovirus-mediated multigene delivery in primary cells, Nature Communications. Tat-HIV-GFP peptides were a generous gift from Rüdiger Arnold (Life Science Lab, German Cancer Research Center).

Microfluidic-Based Production of Giant Unilamellar Vesicles

[0834] For the production of polymer shell stabilized giant unilamellar vesicles (also referred to as “water-in-oil droplets” or only “droplets”) a solution of small unilamellar vesicles with lipid compositions as given in table 4 was prepared. Briefly, lipids dissolved in chloroform were mixed at receptive ratios in glass vials and dried under a gentle nitrogen stream. The dried lipid films were rehydrated to a final lipid concentration of 3 mM in production buffers given in table 4 for 30 min. Subsequently, the solution was shaken for 5 min at more than 600 rpm. The resulting liposome solution was extruded at least 9 times through 50 nm pore size polycarbonate filter. Small unilamellar vesicle solutions were stored at 4° C. for up to 3 days or used for polymer shell stabilized giant unilamellar vesicle production immediately.

[0835] Droplet-based microfluidic mechanical splitting devices were fabricated using poly(dimethylsiloxane) (PDMS). The complete devices were produced as previously described using photo- and soft-lithography methods (Soft Lithography, Xia Y and Whitesides, G M, 1998). Flow rates were controlled by a Elveflow OB1 MK3-microfluidic flow control system. If not stated otherwise, for the formation of giant unilamellar vesicles within microfluidic droplets, the small unilamellar vesicle solutions were diluted to a final lipid concentration of 1.5 mM. For droplet formation, the small unilamellar vesicle solutions were introduced into the aqueous channel of the microfluidic devices. Negatively charged giant unilamellar vesicles were formed using 1.25 mM PFPE(7000 g/mole)-PEG(1500 g/mole)-PFPE(7000 g/mole) triblock surfactant dissolved in FC-40. Positively charged giant unilamellar vesicles were formed using 0.5% PEG-based fluorosurfactant diluted in FC-40 (cat #: 008-FluoroSurfactant-1 G, RAN Biotechnologies). For formation of giant unilamellar vesicles containing DOBAQ lipids for lysosomal escape, 1.25 mM PFPE(2500 g/mole)-PEG(600 g/mole)-PFPE(2500 g/mole) triblock surfactant diluted in FC-40 was used. A ratio of the aqueous to oil phase of approximately 1:4 was used. Droplets were formed at the flow-focusing junction of the splitting device and collected from the outlet of the microfluidic chip into a microtube. Following the collection, polymer shell stabilized giant unilamellar vesicles were allowed to equilibrate for minimum 2 hours at 4° C. before the release.

Release of Giant Unilamellar Vesicles from the Polymer Shell

[0836] For the release of the giant unilamellar vesicles into an aqueous buffer, following the formation of polymer shell-stabilized giant unilamellar vesicles, excess oil phase was removed from the microtube and the layer of polymer shell stabilized giant unilamellar vesicles was mixed with the destabilizing agent PFO, added at a ratio 1:1 in the respect of the aqueous production “intraluminal” buffer (PBS, water, or DMEM). Thereafter, a volume of release buffer, also at a ratio 1:1 in the respect of the aqueous production “intraluminal” buffer, was added as a single drop or layer to the collected polymer shell stabilized giant unilamellar vesicles. The respective separated layers were mixed by gentle agitation of rotation of the collection tube.

[0837] Following 30 min of equilibration, the aqueous phase containing giant unilamellar vesicles was transferred into a 2 ml microtube.

[0838] Finally, a volume of release buffer was added at a ratio 1:1 in the respect of the aqueous production buffer and the released giant unilamellar vesicles were centrifuged at >10.000 g for 15 min. The supernatant was discarded and pellet was suspended to the desired concentration.

[0839] The release buffer is preferably the same buffer used as production buffer (also named “intraluminal” buffer), which is encapsulated by the polymer shell-stabilized giant unilamellar vesicles.

[0840] Table 6 reports the lipid compositions, functionalizations and buffers used in the Examples of the present invention.

Dynamic Light Scattering

[0841] Zeta potentials of giant unilamellar vesicles and original small unilamellar vesicles were measured with a Malvern Zetasizer Nano ZS system at a total lipid concentration of 15 μM in PBS. Equilibration time was set to 600 s at 25° C., followed by three repeat measurements for each sample at a scattering angle of 173° using the in-build automatic run-number selection. Material refractive index was set to 1.4231 and solvent properties to η=0.8882, n=1.33 and ε=79.0. For giant unilamellar vesicle zeta-potential measurements, equilibration time and number of individual measurements were set to 120 s and 2 repeats, respectively. Zeta-potential of DOBAQ containing giant unilamellar vesicles was assessed by diluting giant unilamellar vesicles to a final lipid concentration of 15 μM in PBS solution adjusted to desired pH with 4 N NaOH or 10% HCl. All zeta-potential measurements were performed at least in duplicates.

[0842] Also the distribution size of the original lipid formulations, such as small unilamellar vesicles or liposomes, was measured by dynamic light scattering. Analysis of the hydrodynamic radius of small unilamellar vesicles was performed with a Malvern Zetasizer Nano ZS system. Samples were diluted to a final lipid concentration of 15 μM in PBS filtered with through a 0.22 μm filter. The temperature equilibration time was set to 300 s at 25° C. Three individual measurements for each sample were performed at a scattering angle of 173° based on the built-in automatic run-number selection. The material refractive index was set to 1.4233 and solvent properties were set to η=0.8882, n=1.33 and ε=79.0.

Assessment of Droplet Homogeneity

[0843] To assess transmission heterogeneity of intraluminal droplet contents, water-in-oil droplets were produced at a flow focusing junction of a droplet splitting device. PBS aqueous phase containing 10 mM MgCl.sub.2, 1 mM AlexaFluor405, 1 μM His-tagged GFP, 1.08×10.sup.9 particles/ml Fluoresbrite YG Microspheres (diameter=1.00 μm) was used. Droplets were collected and mean droplet fluorescence intensity for all fluorophores was measured from single plane fluorescence confocal images using global thresholding segmentation and the Particle analyzer tool of ImageJ software.

Quantitative Mass Spectrometry (MS)

[0844] For quantitative mass spectrometry analysis of original small unilamellar vesicle and giant unilamellar vesicle lipid content, giant unilamellar vesicles were produced from the small unilamellar vesicle composed of 33 mol % DOTAP, 33 mol % DOPE, 33 mol % DOPC, 1 mol % LissRhod PE with 0.5 w/v % RAN Biotechnologies PEG-based fluorosurfactant diluted in FC-40. Relative quantitative mass spectrometry was performed using a Sciex QTRAP 4500 mass spectrometer hyphenated with a Shimadzu Nexera HPLC system. The instruments were controlled via Sciex Analyst 1.7 software. Samples were diluted 1 to 1000 in MeOH and subsequent, fractionation was performed using a Supelco Titan C18 column (0.21×10 cm, 1.9μ) operated at 45° C. The isocratic method featured a flow rate of 0.5 ml/min using a 10 mM NH4Ac solution in 98% MeOH. The MS experiments were performed in multiple reaction monitoring (MRM) mode using the following instrument settings: curtain gas 35 psi, ionization voltage 5500 V, nebulizer gas 30 psi, heater gas 60 psi, heater temperature 180° C. and a CAD gas set to 9. The following compound specific parameters were used:

TABLE-US-00003 Cell Pre- Frag- De- exit cursor ment Dwell clustering Collision po- mass mass time potential Energie tential ID [Da] [Da] [msec] [V] [V] [V] DOPE  744.498 603.500 110  91 33 16 DOTAP  662.528 603.500  50 166 41 20 DOPC  786.528 184.000  50 161 39 14 LissRhod 1301.605 682.000 110  40 67 24

[0845] Data analysis was performed using Sciex Analyst 1.7 and MultiQuant 3.0.2 software. Calculated concentrations were normalized using small unilamellar vesicle sample with the following initial lipid ratios DOPE 33/DOTAP 33/DOPC 33/Liss Rhod PE 1.

Quantification of Release Efficiency and Stability

[0846] Release efficiency and mechanical stability of the giant unilamellar vesicles after agitation was assessed by manually counting the polymer shell stabilized giant unilamellar vesicles and the released giant unilamellar vesicles with a Neubauer chamber mounted on a fluorescence microscope. Total lipid concentration of released and purified giant unilamellar vesicles was quantified by measuring giant unilamellar vesicle solution fluorescence (FIG. 27). Respective fluorescence was normalized to a standard small unilamellar vesicle dilution curve (fitted to one phase exponential decay) of known concentration and with equal ratio of fluorescently labeled lipids. For incubation of giant unilamellar vesicles with cell lines, total lipid concentrations between 1.5 μM-50 μM were used.

Quantification of Lysosomal Degradation

[0847] For quantification of lysosomal degradation of giant unilamellar vesicles, A431 D monolayers were incubated with fluorescently labeled giant unilamellar vesicles and monitored by live cell fluorescence time-lapse microscopy. Total giant unilamellar vesicle number in the field of view was counted by global threshold segmentation and the ImageJ build-in particle analyzer plug-in for each time frame.

Cell Culture

[0848] Fibroblast REF52 cells, endothelial MDCK cells, epithelial A431 and A431 D cells, cervical cancer Hela cells and astrocytes Hs683 cells were cultured in Dulbecco's Modified Eagle Medium supplemented with 4.5 g/L glucose, 1% L-glutamine, 1% penicillin/streptomycin and 10% fetal bovine serum. Neuroblastoma SH—SY5Y cells were cultured in a 1:1 mixture of F12:DMEM supplemented with 1% L-glutamine, 1% penicillin/streptomycin and 10% fetal bovine serum. Human T-lymphocyte Jurkat cells were cultured in RPMI-1640 medium supplemented with 1% penicillin/streptomycin and 10% fetal bovine serum. Adrenal PC12 cells were cultured in RPMI-1640 medium supplemented with 1% L-glutamine, 1% penicillin/streptomycin 5% fetal bovine serum and 10% heat-inactivated horse serum. Cell cultures were routinely cultured at 37° C. and 5% CO.sub.2 atmosphere and passaged at approx. 80% confluency using 0.05% trypsin/EDTA treatment. Jurkat cells were passaged by diluting 1 ml of Jurkat culture to 4 ml of fresh culture medium.

Cell Staining

[0849] Live cells were stained with Hoechst 33342 at a final concentration of 5 μg/ml to visualize cell nuclei. Cytoplasms were stained with CellTracker Blue CMAC dye and CellTracker Green CMFDA dye following the manufactures instructions. Lysosomes were stained with LysoTracker Green DND-26 dye following manufacturer's instructions. Cell membranes were stained with wheat germ agglutinin (WGA)-AlexaFluor conjugates. Membranes were stained by adding WGA conjugates to a final concentration of 50 μg/ml to cells grown in fully-supplemented growth medium for 10 min at room temperature. To reduce endocytotic dye uptake, stained cells were handled for imaging at room temperature. WGA is known to stain also endosomal and golgi-associated vesicles. To stain endosomal giant unilamellar vesicle uptake, cells incubated with giant unilamellar vesicles were incubated with a final concentration of 5 μg/ml WGA-AlexaFluor conjugates for 24 hours.

Confocal Microscopy and Live Cell Imaging

[0850] For fluorescence confocal microscopy observations, cell lines were cultured in 8-well Nunc LabTeK glass bottom culture slides filled with at least 400 μl of culture medium. Confocal microscopy was performed with a laser scanning microscope LSM 800 (Carl Zeiss AG). Images were acquired with a 20× (Objective Plan-Apochromat 20×/0.8 M27, Carl Zeiss AG) and a 63× immersion oil objective (Plan-Apochromat 63×/1.40 Oil DIC, Carl Zeiss AG). Images were analyzed with ImageJ (NIH), also to determine vesicle diameter. In brief, global intensity-based thresholding was performed on the confocal microscopy images, followed by watershed separation of overlying particles. Automated particle area measurement was performed on the binary images by the build-in particle analyzer. Adjustments of image brightness and contrast or background correction was performed always on the whole image and special care was taken not to obscure or eliminate any information from the original image. For images with speckled noise signals, 2 pixel median filters were applied. For cell fixation prior to confocal microscopy analysis, cell cultures were washed twice with PBS and subsequently fixed with 2-4% PFA for at least 20 min. For time-lapse live cell imaging a Leica DMi8 inverted fluorescent microscope equipped with a sCMOS camera and 10× HC PL Fluotar (NA 0.32, PH1) objective was used. Cells were cultured in 8-well Nunc LabTeK glass bottom culture slides in FluoroBrite DMEM (high glucose) medium supplemented with GlutaMAX, 10% FBS and 1% Pen/Strep.

Transmission Electron Microscopy

[0851] REF52 cells incubated for 16 hours with giant unilamellar vesicles were fixed in 2.5% glutaraldehyde dissolved in a 0.1 M Na.sub.3PO.sub.4 solution for 30 min at room temperature. Cells were further fixed in 0.4% uranyl acetate overnight. Fixed cells were subsequently dehydrated with a 50%, 60%, 70%, 80%, 90%, and 100% ethanol series and embedded in resin over night at 60° C. 85 nm ultrathin sections were prepared and contrasted with lead acetate or osmium tetroxide. A Zeiss EM 10 CR transmission electron microscope was used for imaging. If necessary, image contrast, brightness and sharpness were adjusted using the build-in ImageJ plug-ins.

Flow Cytometry

[0852] For flow-cytometry analysis of the attraction between RGD functionalized giant unilamellar vesicles and Jurkat cells, rhodamine B-giant unilamellar vesicles with varying RGD density (0, 1, 2, 10 mol %, see table 6) were incubated with Jurkat cells for 24 hours. Cell were subsequently centrifuged for 5 min at 250 g and the supernatant containing non-bound or non-uptaken giant unilamellar vesicles was discarded. Cells were resuspended in fresh culture medium and for each condition, giant unilamellar vesicle fluorescence within the cells was quantified with a BD LSR Fortessa Cell Analyzer (BD Bioscience) using the blue laser line in the PE channel (λ.sub.em max=575). From the acquisition of the scattered parameters, a gate was set to discriminate between Jurkat cells, debris and possible clumps. Subsequentially, singlet cells were identified from the aforementioned Jurkat population. Lastly, based on this gating strategy, fluorescence intensity associated with the cells periphery was recorded and quantified.

Attraction Assay

[0853] Cells were seeded in triplicates in 100 μl of their corresponding growth media to form a confluent monolayer after 24 h of incubation in 96 flat-bottom well-plates. Giant unilamellar vesicle (labeled with LissRhod PE lipids) solutions were added to a final lipid concentration of 1.5 μM and incubated for 24 h. Fluorescence of each well was measured using an Infinite M200 TECAN plate reader controlled by TECAN iControl software with in-build gain optimization and excitation/emission setting adjusted to 550/585 nm, respectively. Subsequently, wells were washed 3 times with 100 μl PBS using a multichannel pipette and residual fluorescence was measured again. Fluorescence intensity after washing was normalized to intensity before washing in order to account for any variation in sample preparation. All samples were measured in triplicates at 4 individual position/well in order to account for variation in cell monolayer density. For comparison of specific attraction of biofunctionalized giant unilamellar vesicles, all attraction values were normalized to the attraction of BSA functionalized giant unilamellar vesicles and the respective cell line in order to reference all attraction values for each cell type to a common moderately non-reactive protein.

[0854] A total lipid concentration of 3 μM of anti-CD3 functionalized giant unilamellar vesicles and Jurkats cells were incubated for 24 hours in order to perform fluorescence confocal microscopy observation of the attraction and formation of contact sides. Prior to imaging, cells were stained with Hoechst 33342 and WGA-AlexaFluor647 as described above.

Quantification of Preferential Giant Unilamellar Vesicle Uptake in Co-Culture

[0855] For quantification of preferential giant unilamellar vesicle uptake in SH—SY5Y/Hs683 co-culture experiments, SH—SY5Y and Hs683 cell were separately stained with CellTracker Blue CMAC (and CellTracker Green CMFDA, respectively. Cells were co-seeded in a 10:1 SH—SY5Y:Hs683 ratio in a 1:1 mixture of F12:DMEM supplemented with 1% L-glutamine, 1% penicillin/streptomycin and 10% fetal bovine serum together with giant unilamellar vesicles composed of 20 mol % PEG750 PE, 20 mol % EggPG, 58 mol % EggPC, 1 mol % LissRhod PE and 1 mol % palmitic acid NHS coupled with 1.5 μM His-tagged recombinant NrCAM for 24 hours. Cell cultures were subsequently washed 3 times with PBS and fixed for 20 min with 4% PFA followed by fluorescence confocal microscopy analysis with appropriate laser excitation. Total area of SH—SY5Y and Hs683 cells in a field of view was calculated from single plane confocal images. Total number of giant unilamellar vesicles in respective areas was determined by global threshold segmentation and subsequently normalized to the total cell area. For example, total area of SH—SY5Y and Hs683 cells shown in FIG. 19 is 20083.37 μm.sup.2 and 40741.67 μm.sup.2, respectively. They contain 1491 (=0.0742 giant unilamellar vesicles/μm.sup.2) and 578 (=0.0141 giant unilamellar vesicles/μm.sup.2) giant unilamellar vesicles, respectively, which corresponds to an increase of 523% for SH—SY5Y cells.

Encapsulation of Baculoviruses

[0856] For intracellular delivery of baculoviruses (BVs) encoding a mitochondrial targeted dsRed (Discosoma red) fluorescent protein, solutions containing BVs were mixed in a 1:100 ratio with the small unilamellar vesicle solution. Importantly, for polymer shell stabilized giant unilamellar vesicle production, PBS with 60 mM MgCl.sub.2 was used, as lower manganese concentrations inhibited polymer shell stabilized giant unilamellar vesicle formation and release rates. Released and purified giant unilamellar vesicles containing baculoviruses were then incubated with REF52 cells for 24 h and subsequently stained with 8 μg/ml Hoechst33342. Expression of mitochondrial targeted dsRed protein and baculovirus localization was assessed my fluorescence confocal microscopy. For visualization of intracellular baculoviruses staining of baculovirus-DNA was imaged by overexposing the Hoechst33342 channel.

Giant Unilamellar Vesicle's Functionalization and PEGylation

[0857] Functionalization of giant unilamellar vesicles was always performed on released giant unilamellar vesicles. If not stated otherwise, giant unilamellar vesicle's functionalization was carried in PBS and in the dark on a horizontal shaker at room temperature. After functionalization, giant unilamellar vesicles were centrifuged at >10.000 g for at least 15 min and resuspended in PBS. For NHS based coupling reactions, giant unilamellar vesicles were kept whenever possible at 4° C. and released from droplets not later than 1 hour after production. NHS coupling reactions were performed for at least 3 hours. For NHS and NTA based functionalizations, respective proteins and peptides were added in 2-5 fold excess to total functionalized lipids as calculated from the total lipid concentration. For example, fluorescence quantification showed total lipid concentration of 150 μM (corresponding to approx. release efficiency of 10%) and giant unilamellar vesicles were produced from small unilamellar vesicles with 1 mol % palmitic acid-NHS lipids, a minimum of 1.5 μM of the protein to be coupled was added (about 50% of the NHS coupled lipids would reside within the inner membrane leaflet and not be accessible for coupling). In case of the WGA coupling, the lectin was coupled to the giant unilamellar vesicle in 10 times less molar concentration compared to the presented 5 mol % NHS ligand. Functionalization of giant unilamellar vesicles with RGD peptides was performed by introducing a desired amount of DSPE-RGD into the lipid mixture for small unilamellar vesicle production.

[0858] For the sequential functionalization of giant unilamellar vesicles with gold nanoparticles (AuNP), giant unilamellar vesicles containing 1 mol % NHS were incubated with 3 μM L-cysteine for 6 hours. Subsequently, 50 nm Au nanoparticles were added to a final concentration of 10 μg/ml and shaken at 300 rpm overnight. Functionalization of giant unilamellar vesicles with IgG antibodies was performed by incubating 3 μM His-tag Protein G to giant unilamellar vesicles containing 1 mol % 18:1 DGS-NTA(Ni) lipids for 1 hour. Subsequently, a final concentration of 3 μM of respective IgG dissolved in 1% BSA was added to mixture and incubated for 1 hour. In order to avoid cross reactions, whenever multiple functionalizations based on different coupling reactions were performed, as in the case of triple functionalization, proteins to be coupled via NHS were incubated with the giant unilamellar vesicles first before performing the other couplings, e.g. biotin-streptavidin coupling.

[0859] PEGylation of giant unilamellar vesicles with poly-ethylene-glycol polymers was performed by introducing a desired amount of PEG350-PE, PEG750-PE or PEG1000-PE into the lipid mixture for production of the initial small unilamellar vesicles.

Lysosomal Escape Mechanisms

[0860] For all tested lysosomal escape approaches, polymer shell stabilized giant unilamellar vesicles were produced in PBS+10 mM MgCl.sub.2+50 mM 8-Hydroxypyrene-1,3,6-trisulfonic acid trisodium salt (HPTS) using 1.25 mM triblock PFPE-PEG-PFPE 2500-600-2500 surfactant dissolved in FC-40. Released giant unilamellar vesicles were purified by centrifugation and incubated with REF52 cells for 24 hours. Intracellular HPTS fluorescence distribution was subsequently assessed by fluorescence confocal microscopy (Excitation 460 nm, Emission 510 nm). Imaging parameters were kept constant when comparing between the different approaches. For analysis of lysosomal escape via poly-ethylene-imine (PEI) proton sponge mechanism, giant unilamellar vesicles composed of 20 mol % EggPG, 79 mol % EggPC and 1 mol % LissRhod PE were loaded during droplet production with 44 μg/ml polyethyleneimine. For analysis of lysosomal escape via DOBAQ mediated intralysosomal fusion, giant unilamellar vesicles composed of 1 mol % LissRhod PE, 60 mol % DOBAQ, 20 mol % EggPG, and 19 mol % EggPC were produced.

Synthesis of Surfactant of Formula (I)

[0861] The method described in this section refers to the synthesis of one surfactant with PEG600 and PFPE having a particular molecular weight, and can be applied for the synthesis of surfactants having other molecular weights, in accordance with Formula (I), wherein m is an integer comprised between 5 and 150, and n is an integer comprised between 5 and 450.

[0862] PFPE-PEG600-aminium derivative, the surfactant of Formula (I), consists of three parts: (a) a PFPE block, which is immiscible with water, but with fluorinated oils; (b) a PEG block, which is immiscible with fluorinated oils, but with water; and (c) an aminium ion end, which causes a positive charge on the water-oil-interface. These three parts are connected using click reaction, first described by Sharpless K. B. et al., Angew. Chem. Int. Ed, 40: 2004-2021. The mild conditions, which are needed for this reaction type, enables a high yield and reduces side reactions. For this, PFPE acid is turned into a propargyl derivative by activating the carboxylic acid followed by amide formation with propargyl amine (Formula (II)). Both hydroxyl groups of PEG600 have to be transformed to azide groups. This can be achieved by tosylation (Formula (III)), followed by substitution with sodium azide (Formula (IV)). For the aminium ion part, bromoacetyl bromide is used as starting material, first functionalized with an alkyne (Formula (V)) and in a second step, the aminium ion is formed (Formula (VI)). In the last two steps, all three parts are clicked together (Formula (VI) and (I)).

Synthesis of Propargyl PFPE Amide (II)

[0863] ##STR00027##

[0864] The synthesis is based on the synthesis published by Scanga R. et al., RSC Adv. 2018, 8: 12960-12974.

[0865] PFPE acid (Krytox FSH, 10 mmol, 70 g) was placed in a flame-dried three-necked flask and dissolved in 125 mL HFE 7100 under inter atmosphere. After the polymere was completely dissolved, oxalyl chloride (30 mmol, 2.6 mL) was added. The mixture was refluxed for 18 hours. With stirring and heat applied, the solvent was removed under vacuum. The crude PFPE acid chloride was redissolved in HFE 7100 and filtered under inert conditions. The filtrate was added to a flame-dried three necked flask equipped with a dropping funnel under inert atmosphere. Tetrahydrofuran (30 mL), propargyl amine (10.5 mmol, 666 μL) and triethyl amin (15 mmol, 2.1 mL) were added to the dropping funnel. The mixture was added dropwise to the PFPE acid chloride solution and stirred for 18 hours. Afterwards, the crude product was filtered through celite and solvent was removed under reduced pressure. The product was received as yellow oil (9.33 mmol, 65.7 g, 93.3%).

Synthesis of PEG600 Ditosylate (III)

[0866] ##STR00028##

[0867] The synthesis is based on the synthesis published by Scanga R. et al., RSC Adv. 2018, 8: 12960-12974.

[0868] Sodium hydroxid (0.4 mol, 16 g) was dissolved in water (125 mL) under ice cooling and afterwards cooled to 0° C. PEG600 (0.1 mol, 60 g) was dissolved in Tetrahydrofuran (240 mL) and added dropwise through a dropping funnel to the sodium hydroxid solution, paying attention that the temperature must not increase above 5° C. Afterwards, the reaction mixture was allowed to reach room temperature and stirred for one hour. The mixture was cooled to 0° C. again. p-Toluenesulfonyl chloride (0.23 mol, 43.8 g) was dissolved in tetrahydrofuran (185 mL) and added dropwise through a dropping funnel to the cooled reaction mixture. The temperature must not rise above 5° C. during this process. The reaction mixture was stirred for 18 hours without further cooling. The obtained emulsion was separated and the solvent was removed from the organic layer. The crude product was redissolved in 900 mL ethyl acetate and washed twice with water and once with saturated sodium chloride solution. Subsequently, the purified solution was stirred over magnesium sulfate for one hour, filtered and solvent was removed under reduced pressure. The product was received as white amorphic substance (78.4 mmol, 71.3 g, 78.4%).

Synthesis of PEG600 Diazide (IV)

[0869] ##STR00029##

[0870] The synthesis is based on the synthesis published by Scanga R. et al., RSC Adv. 2018, 8: 12960-12974.

[0871] PEG600 ditosylate (III, 10 mmol, 9.1 g) and sodium azide (22 mmol, 1.43 g) were dissolved in dimethylformamide (25 mL) and stirred for 90 min at room temperature. Then, the mixture was stirred for 18 hours at 50° C. Afterwards, the suspension was filtrated and the solvent was removed from the filtrate under reduced pressure. The received oil was mixed with ethyl acetate (175 mL), sonicated for 20 min and filtered again. The filtrate was washed twice with water and once with saturated sodium chloride solution. The purified solution was dried over magnesium sulfate for one hour, filtered and solvent was removed under reduced pressure. The product was received as white amorphic substance (8.65 mmol, 5.62 g, 86.5%).

Synthesis of 2-bromo-N-(prop-2-yn-1-yl)acetamide (V)

[0872] ##STR00030##

[0873] The synthesis is based on the synthesis patented by Aulakh, V. S. et al., U.S. Pat. Appl. Publ., 20150266867, 24 Sep. 2015.

[0874] Bromoacetyl bromide (10 mmol, 877 μL) and triethylamine (10 mmol, 1.45 mL) were dissolved in dichlormethane (18 mL) in a flame-dried flask and cooled to 0° C. A solution of propargylamine (10 mmol, 641 μL) in dichlormethane (9 mL) was added dropwise and the mixture was stirred for 2 hours at 0° C. The received suspension was filtered and solvent of the filtrate was removed under reduced pressure. The crude product was further purified by gradient flash column chromatography (from 100% hexane to 50% ethyl acetate in hexane). The product was received as white powder (6.92 mmol, 1.22 g, 69.2%).

Synthesis of N,N,N-trimethyl-2-oxo-2-(prop-2-yn-1-ylamino)ethan-1-aminium (VI)

[0875] ##STR00031##

[0876] The synthesis is based on the synthesis patented by Aulakh, V. S. et al., U.S. Pat. Appl. Publ., 20150266867, 24 Sep. 2015.

[0877] 2-Bromo-N-(prop-2-yn-1-yl)acetamide (5 mmol, 882 mg) was dissolved in acetonitrile (6 mL). A trimethylamine solution (4.2 M in ethanol, 25 mmol, 6 mL) was added and the mixture was stirred for 20 hours. Then, solvent was removed under reduced pressure and the product was received as white powder by precipitating with diethyl ether (4.78 mmol, 1.12 g, 95.6%). For the calculation of molecular weight, bromium ion as counter ion was included.

Synthesis of azide-PEG600-aminium derivative (VII)

[0878] ##STR00032##

[0879] PEG600 diazide (IV, 5 mmol, 3.25 g), N,N,N-trimethyl-2-oxo-2-(prop-2-yn-1-ylamino)ethan-1-aminium (VI, 5 mmol, 1.17 g), copper(II) sulfate pentahydrate (0.5 mmol, 125 mg), (+)-Sodium ascorbate (0.2 mmol, 200 mg) and neocuproine (0.16 mmol, 165 mg) were dissolved in water (30 mL). The solution was stirred first for 1 hour at room temperature, followed by 48 hours at 50° C. Afterwards, the solution was freeze-dried. The crude product was suspended in ethanol, filtered and solvent was removed under reduced pressure. The solid material was redissolved in water and the remaining starting material was removed by extracting three-times with dichlormethan. After freez-drying the aqueous phase, the received product consists of one-side functionalized PEG600 derivative (azide-PEG600-aminium derivative) and two-side functionalized PEG600 derivative (PEG600 diaminium derivative) in an estimated ratio of 2:1. The two-side functionalized PEG600 derivative can not interact in the following reaction and can be separated easier afterwards. Therefore, no further purification steps were necessary. The product mixture was received as yellowish, amorphic substance (3.38 mmol, 2.93 g, 67.6%; calculated for aimed product: 2.25 mmol, 1.81 g, 45%). For the calculation of molecular weight, bromium ion as counter ion was included.

Synthesis of PFPE-PEG600-aminium derivative (I)

[0880] ##STR00033##

[0881] The synthesis is based on the synthesis published by Scanga R. et al., RSC Adv. 2018, 8:12960-12974.

[0882] Propargyl PFPE amide (II, 1 mmol, 7.04 g) was dissolved in HFE 7100 (6 mL). Azide-PEG600-aminium derivative (VII, 1.5 mmol, 1.3 g), copper(II) sulfate pentahydrate (0.1 mmol, 25 mg) and neocuproine (0.16 mmol, 33 mg) were dissolved in methanol (3 mL). (+)-Sodium ascorbate (0.2 mmol, 40 mg) was dissolved in water (3 mL). All three solutions were combined and stirred first for 1 hour at room temperature, followed by 60 hours at 50° C. Afterwards, methanol (12 mL) were added during stirring. The emulsion was destabilized and separated slowly. If no separation can be observed, add methanol in 1 mL-steps until the phases start to separate. Both phases were isolated. To the fluorinated oil phase, HFE7100 (6 mL) was added and the resulting separated aqueous phase was trashed. The solution was dried of magnesium sulfate for one hour, filtered and solvent was removed under reduced pressure. The product was received as high-viscous, yellowish oil (0.88 mmol, 6.93 g, 88.3%). For the calculation of molecular weight, bromium ion as counter ion was included.

Example 1. Microfluidic Mechanical Splitting of Polymer Shell Stabilized Giant Unilamellar Vesicles

[0883] Polymer shell stabilized giant unilamellar vesicles (“water-in-oil-droplet”) with size below 5 μm were obtained using a self-developed droplet-based microfluidic device consisting of a flow focusing junction for water-in-oil-droplet production and a multi-Y-shaped microfluidic droplet splitting unit (FIG. 1). After production of a parent water-in-oil-droplet with a 60 μm diameter, the splitting unit design allowed for high-throughput mechanical droplet splitting in up to five consecutive division steps creating droplets with a final diameter of 2.90 μm±0.45 μm (n=202), as measured by confocal microscopy (FIG. 1).

[0884] To assess transmission heterogeneity of intraluminal contents from the mother to the daughter droplets, fluorescence confocal microscopy was used to analyze droplet-entrapped fluorescent content and respective signal intensity distribution among droplets before and after splitting (FIG. 2). The results revealed only little inter-droplet variation of signal intensity for the low molecular weight fluorophore AlexaFluor 405 before splitting (Coefficient of variation (CV) before=5.6%, n=29) and after splitting (CV after =10.7%, n=665), green fluorescent protein (CV before=18.5%, n=29 and CV after =14.8%, n=665) and 100 nm fluorescently labelled small unilamellar vesicles composed of 20 mol % EggPG, 79 mol % EggPC and 1 mol % LissRhod PE (CV before=68.0%, n=29 and CV after =15.4%, n=665). Based on this, it could be concluded that luminal composition of splitted droplets resembles the composition of the mother droplets, showing that the mechanical splitting approach is suitable for the controlled production of small water in oil-droplets with defined and tunable compositions at high production rates (e.g. 2.5×10.sup.5 at the focusing T-junction and 8×10.sup.6 droplets/min after five-fold mechanical splitting, respectively). Importantly, peripheral distribution of lipid fluorescence in the divided droplets was observed, suggesting successful mechanical division of the polymer shell stabilized giant unilamellar vesicles.

Releasing of the Giant Unilamellar Vesicles from the Polymer Shell

[0885] Following addition of destabilizing surfactants to the collected mechanically-splitted polymer shell-giant unilamellar vesicles (see Methods), the inventors were able to release large quantities of giant unilamellar vesicles (diameter=1.400 μm±0.202 μm, n=122) into an aqueous phase (FIG. 3). A concentration of 1.5 mM was found to be the optimal required concentration for the initial solution of small unilamellar vesicles, achieving over 50% release efficiency of giant unilamellar vesicles, corresponding to a successful giant unilamellar vesicles release from approximately every second droplet (FIG. 4). Therefore, using optimized conditions, this method allows for a production rate of giant unilamellar vesicles of approximately 4×10.sup.6 giant unilamellar vesicles/min.

Symmetric Droplet Splitting

[0886] A full control over the physicochemical and biological properties of giant unilamellar vesicles is a pivotal requirement for biomedical and synthetic biology applications. Therefore, mass spectrometry (see Method section) was used to assess quantitatively the lipid composition of the formed polymer shell stabilized giant unilamellar vesicles. The results revealed that the lipid ratio of splitted giant unilamellar vesicles resembles that of the initial small unilamellar vesicles used during production of the parent polymer shell stabilized giant unilamellar vesicle and hence no lipid ratio change occurs during microfluidic handling and mechanical droplet splitting (Table 3 below).

TABLE-US-00004 TABLE 3 Quantitative mass spectroscopy (MS) was performed on small unilamellar vesicles containing DOPE, DOTAP, DOPC and LissRhod PE lipids and splitted giant unilamellar vesicles produced from these small unilamellar vesicles. Results are shown as LissRhod PE- normalized concentration ratios. Ratio change corresponds to the “change of molar percentage” defined at page 16. No considerable loss of specific lipid subsets was found. MS experiments were performed in triplicate. The coefficients of variation (CV) for the small unilamellar vesicle sample were <7.5 %. CVs for the giant unilamellar vesicle samples was <10 %. LissRhod Lipid Concentration DOPE DOTAP DOPC PE Concentration before division 33.00 33.00   32.99 0.99 Concentration ratio versus 33.22 33.22   33.21 1.00 LissRhodPE Concentration after division 21.49 21.10   21.95 0.65 Concentration ratio versus 32.98 32.38   33.68 1.00 LissRhodPE Ratio change  0.73  2.53  −1.39 0.00

Mechanical Stability

[0887] Additionally, a basic assessment of the mechanical stability of formed giant unilamellar vesicles revealed that approximately 90% of the giant unilamellar vesicles tolerate incubation at 37° C. and mechanical agitation on a horizontal shaker at 800 rpm for 24 h. This suggests that they can sustain considerable mechanical stress and are therefore potentially robust enough for drug delivery applications (FIG. 5).

Release of the Giant Unilamellar Vesicles and Maintenance Under Physiological Conditions

[0888] To successfully interface giant unilamellar vesicles with living cells, it is of major importance that the giant unilamellar vesicles can be produced and maintained under physiological buffer conditions. Therefore, release efficiency of mechanically splitted giant unilamellar vesicles in serum supplemented culture medium was systematically assessed using vesicles filled with serum supplemented cell culture medium. Similarly to PBS and water, it was obtained up to 45% release efficiency in cell culture medium (Table 4 below). Following the release, giant unilamellar vesicles were incubated with rat embryonic fibroblast (REF52) in cell culture. Importantly, time-lapse microscopy analysis showed that giant unilamellar vesicles remained stable during 20 h of incubation (FIG. 6).

TABLE-US-00005 TABLE 4 Quantification of release efficiencies of giant unilamellar vesicles for different combinations of intraluminal (production buffer) and extraluminal release buffers. Production release buffer buffer Water (eff %) PBS (eff %) DMEM (eff %) Water 17% 13%  3% PBS  5%  5%  4% DMEM  4%  5% 45%

Example 2. Fine-Tuning of Charge-Mediated Giant Unilamellar Vesicle—Cell Interactions

[0889] Charge-mediated interaction might serve, if precisely controlled, as a potent instrument to guide the interactions between cells and giant unilamellar vesicles. Therefore, the interaction spectrum of differently charged giant unilamellar vesicleswith various cell lines in vitro was systematically investigated. To this end, splitted giant unilamellar vesicles were produced with varying amounts of positively (DOTAP) and negatively (DOPG) charged lipids and their respective Zeta-potential was measured by dynamic light scattering after releasing from the polymer shell (Table 5 below). The results showed that the charge of giant unilamellar vesicles can be fine-tuned between highly positive to highly negative by adjusting respective lipid formulations (table 5 below).

TABLE-US-00006 TABLE 5 Zeta-potential values of the released giant unilamellar vesicles obtained by microfluidic mechanical division of polymer shell stabilized giant unilamellar vesicles having different lipid compositions. LIPID Molar percentage DOTAB 0 0 0 20 50 DOPC 49 79 99 79 49 DOPG 50 20 0 0 0 LissRhodPE 1 1 1 1 1 Zeta potential (mV) −31 −19 +2 +2 +28

[0890] To quantify the interactions between cells and giant unilamellar vesicles, a plate reader-based attraction assay was implemented and cell lines of endothelial (MDCK), epithelial (A431D and A431) and adrenal (PC12) origin were interfaced with respective giant unilamellar vesicles. These cell lines were selected in order to cover a wide spectrum of possible target tissues with different surface expression patterns. For all tested cell lines, it was found a strong correlation between the charge of giant unilamellar vesicles and the intensity of cell attraction, where higher charged giant unilamellar vesicles showed an increased attraction compared to less and non-charged giant unilamellar vesicles (FIG. 7). For example, in the case of A431 D cells, which is a frequently used model cell line in carcinoma research, giant unilamellar vesicles with a Zeta-potential of −31 mV showed an almost 100 times higher attraction when compared to giant unilamellar vesicles with +2 mV Zeta-potential. At the same time, giant unilamellar vesicles with a Zeta-potential of +28 mV increased in attraction by about 50 times compared to giant unilamellar vesicles with +2 mV Zeta-potential. However, as this quantitative assay is not able to discriminate between different types of interactions (e.g. uptake, attachment, fusion or engulfment), the qualitative nature of the interaction between giant unilamellar vesicles and cells was further investigated by fluorescence confocal microscopy. As shown in FIG. 8, three distinct types of interactions between giant unilamellar vesicles and A431 D cells could be induced: Endocytosis and attachment were mainly observed for the negatively-charged and neutral giant unilamellar vesicles, respectively. In case of positively charged giant unilamellar vesicles, colocalization of lipid fluorescence with the cell membrane staining (in many cases accompanied by morbid cell morphologies) was observed, indicating that fusion between both membranes occurred. To assure that negatively charged giant unilamellar vesicles are indeed taken up by the cells, two additional analyses were performed: First, the cell cytoplasm was stained and performed z-resolved confocal fluorescence microscopy of internalized fluorescently labeled giant unilamellar vesicles (data not shown). Second, he cells incubated with respective giant unilamellar vesicles were fixed and analysed at transmission electron microscopy (FIG. 9). Both assessments proved that giant unilamellar vesicles are indeed taken up by the cells and reside within their cytoplasm. Taken together, these results revealed an important effect of giant unilamellar vesicle charge on the nature of giant unilamellar vesicle-cell interactions and highlight the ability of the developed method for giant unilamellar vesicle charge control.

Example 3. Biofunctionalization of Giant Unilamellar Vesicles Formed by the Splitting Microfluidic Device

[0891] Despite the fact that charge-mediated cellular uptake of giant unilamellar vesicles is an efficient process, it fails to provide a cell type specific delivery of therapeutic compounds. Therefore, the inventors aimed to establish a ligand directed uptake of giant unilamellar vesicles by developing a toolbox of strategies for bio-orthogonal functionalization of the giant unilamellar vesicle surface with ligands targeting specific moieties and cell types. To exemplarily demonstrate the diversity of giant unilamellar vesicle biofunctionalization possibilities, giant unilamellar vesicles were produced by microfluidic mechanically splitting using biotinylated lipids for coupling to streptavidin tagged proteins, nitrilotriacetic acid-nichel (NTA-Ni.sup.2+)-functionalized lipids for coupling to histidine tagged proteins and DOPE lipids containing primary ammine for linking to N-hydroxysuccimid (NHS) functionalized molecules. Triple orthogonal functionalization of the released giant unilamellar vesicles was achieved by adding streptavidin functionalized Atto425, histidine tagged green fluorescent protein and NHS-functionalized Alexa647 (FIG. 10a). Moreover, more complex sequential functionalization strategies were tested. Towards this end, cysteine functionalized gold nanoparticles were immobilized to the giant unilamellar vesicle lipids via NHS-chemistry (FIG. 10b). Additionally, a multistep approach was tested to couple immunoglobulins (e.g. anti-CD3) via NTA-immobilized histidine tagged protein G (FIG. 10c).

[0892] Notably, antibodies offer great selectivity for particular cell surface antigens, wherefore antibody based targeting has previously been shown to greatly enhance specific small unilamellar vesicle delivery to defined cell subsets. To assess the functionality of anti-CD3-coated giant unilamellar vesicles, they were incubated with CD3.sup.+ Jurkat cells (see Method Section). When analyzed by confocal microscopy, the formation of an attachment site between the giant unilamellar vesicles and the cells, reminiscent of a “minimal” immunological synapse, was observed (FIG. 11), indicating successful giant unilamellar vesicle-cell coupling. In contrast, giant unilamellar vesicles without anti-CD3 coating did not show this complex interaction architecture.

RGD-Mediated Endocitosis

[0893] In order to systematically assess the possibility to apply attractive, receptor-specific giant unilamellar vesicle-cell interactions for targeted giant unilamellar vesicle delivery, negatively charged and RGD biofunctionalized giant unilamellar vesicles were produced with varying ligand densities. RGD was used for giant unilamellar vesicles biofunctionalization (see Method Section), since integrin receptor-based endocytosis has previously been tested to enhance liposomal drug delivery due to the ability of integrin proteins to function as natural intracellular signal transducer for the initiation of endocytic events. Therefore, RGD-giant unilamellar vesicles were interfaced with adherent cell lines expressing RGD-binding integrin receptors and their attractions measured. As a control, the same cells were incubated with non-biofunctionalized, naive giant unilamellar vesicles. For all tested cell lines (FIG. 12), it was found that RGD ligand density on the periphery correlates with giant unilamellar vesicle attraction and that giant unilamellar vesicle-cell coupling could be increased by approximately 10% when applying 10 mol %-RGD ligand decoration. In case of non-adherent Jurkat T-cells, which express high levels of α.sub.4β.sub.1 integrin, 10 mol %-RGD coating increased giant unilamellar vesicle-cell coupling even by 15-fold (as measured by fluorescence flow cytometry). The nature of interaction of fluorescently labelled RGD functionalized giant unilamellar vesicles was further analyzed by confocal microscopy, revealing that when interfaced with A431 D cells, giant unilamellar vesicles mostly accumulate at the cell periphery, the region of highest integrin density, and in the perinuclear region, suggesting RGD-integrin mediated endocytotic giant unilamellar vesicle uptake by the cells (FIG. 13).

Example 4. PEG-Based Passivation Strategy to Regulate Attractive and Repulsive Giant Unilamellar Vesicle-Cell Interactions for Enhance Targeted Delivery

[0894] Although biofunctionalization of giant unilamellar vesicles with anti-CD3 and RGD ligands successfully increased giant unilamellar vesicle-cell interactions, charge-driven and other non-specific attractions at the giant unilamellar vesicle-cell interface might still be high enough to interfere with the ligand-based cell type specific uptake. For example, when incubating non-charged giant unilamellar vesicles functionalized with NrCAM protein with SH—SY5Y neuroblastoma cells (NrCAM positive), the measured attraction was comparable to that of non-functionalized and non-charged giant unilamellar vesicles (FIG. 14b). This indicates that non-specific lipid-cell interactions are considerably high. Giant unilamellar vesicles covered by poly-ethylenglycol (PEG) were synthetized to shield these electrostatic and non-specific interactions. Initially it was tested the shielding potential of PEG at different concentrations and molecular weight for giant unilamellar vesicles of different charges. To this aim, giant unilamellar vesicles were produced with negatively and positively charged lipids at lipid ratios between 15 and 50 mol %. Lipids linked to PEG350, PEG750 or PEG1000 were tested at ratios comprised between 5, 10, 20 and 50 mol % for the. Zeta-potential of respective small unilamellar vesicles before production of droplet-stabilized giant unilamellar vesicles and Z-potential of released giant unilamellar vesicles were measured (FIG. 15a). The results revealed that the Z-potential of both, negatively and positively charged vesicles decreased with increasing PEG chain length and rate of PEGylation. Additionally, it was tested whether PEGylation of giant unilamellar vesicles might be used to shield giant unilamellar vesicle surface charge, thus introducing a repulsive behavior between giant unilamellar vesicles and cells. Towards this end, the PEGylated giant unilamellar vesicles were incubated with six different cell lines established from different tissues and measured respective giant unilamellar vesicle attraction (FIG. 15b). The results showed that for all tested cell lines, giant unilamellar vesicle PEGylation basically decreased the charge-mediated attraction between the giant unilamellar vesicles and the cells. Moreover, this effect was more pronounced in cases of higher PEGylation rates and longer PEG length. For example, in the case of giant unilamellar vesicles interactions with A431 D carcinoma cells, giant unilamellar vesicles equipped with 5 mol % PEG350 showed almost 50% more attraction compared to giant unilamellar vesicles equipped with 50 mol % PEG350 (FIG. 16). Consistently, confocal microscopy analysis showed that naive negatively charged giant unilamellar vesicles were usually localized within or above cells and only a small fraction was found between single cells or cell groups (FIG. 17 upper panel). In contrast, PEGylated giant unilamellar vesicles were observed mostly accumulating in the intercellular space, forming contact inhibition zones between the giant unilamellar vesicle-accumulations and individual cells (FIG. 17 lower panel). Most probably, this behavior can be attributed to the repulsive giant unilamellar vesicle-cell interactions.

[0895] In conclusion, the inventors were able to develop complementary strategies for the regulation of specific attractive interactions between cells and giant unilamellar vesicles, in order to promote the ligand-based specific interactions, by using a PEG-based passivation strategy to suppress charge-mediated interactions.

Example 5 Combination of PEGylation and Ligand-Directed Cell Interactions

[0896] After establishing the approaches to control and fine-tune the attractive and repulsive interactions between cells and giant unilamellar vesicles, both approaches were combined followed by the screening of cell type selectivity for giant unilamellar vesicle targeting. Towards this end, giant unilamellar vesicles were produced comprising 20 mol % negatively charged DOPG lipids, 59 mol % neutral DOPC lipids, 20 mol % PEG750-linked lipids and 1 mol % NHS coupled lipids for ligand immobilization. In this lipid composition, the net strength of ligand-receptor interactions between the giant unilamellar vesicle and the target cell needs to be strong enough to overcome the PEG-mediated repulsion, eventually allowing cell type specific endocytosis induced by the negative giant unilamellar vesicle charge. To assess the specificity, 15 types of differently functionalized giant unilamellar vesicles were first screened, each of which was expected to be either specific for a given cell type (e.g. anti-cadherin antibodies or bradykinin) or non-specific (e.g. poly-L-lysine) by measuring respective attraction values for six different carcinogenic cell lines. Respective carcinogenic cells might resemble potentially interesting targets in a giant unilamellar vesicle-based tumor treatment. In order to reference all attraction values for each cell type to a common moderately non-reactive protein, all values were normalized to the attraction of BSA-coupled giant unilamellar vesicles (FIG. 18a). FIG. 18b shows the summary of the attraction between the six differently functionalized giant unilamellar vesicles and six cell lines of different origin. Giant unilamellar vesicles coated with peptides and proteins which do not bind to cells in a specific manner showed high attraction to basically all cell types, as the non-specific attractions are able to overcome the repulsive PEG-barrier in all cases; examples thereof are poly-L-lysine, which mostly interacts based on electrostatic interaction, wheat germ agglutinin WGA, which binds to the glycocalyx of cells, or the HIV derived tat-peptide, which is an arginine rich peptide which penetrates cell membranes mostly based on hydrophobic interactions. However, when targeting more specific receptors, like the bradykinin specific G-protein coupled receptors, which are abundantly expressed in endothelial cells, by functionalizing giant unilamellar vesicles with the vasodilator bradykinin, specific attraction to endothelial cells was achieved. This attraction was increased up to 40% compared to the other cell lines and similar results could be obtained when targeting cadherin proteins by coating with recombinant cadherin or anti-cadherin antibodies. This comparison shows that fine tuning of attractive interactions and PEG-based shielding of giant unilamellar vesicle surface charges and ligands is a well-suited strategy to gain control over the giant unilamellar vesicle-cells interplay.

[0897] Finally, to assess whether the preferential attraction is able to induce a cell-type specific uptake under concurring conditions in a more complex multi-cell type environment, the giant unilamellar vesicle targeting approach was tested in co-culture experiments. Towards this end, astrocyte (Hs683) and neuronal (SH—SY5Y) model cell lines were chose, as these cell types closely grow and interact in the mammalian brain. Moreover, achieving preferential giant unilamellar vesicle uptake by neurons is desirable when developing therapeutic strategies for neuroblastoma or neurodegenerative diseases. To test the preferential attraction, giant unilamellar vesicles were prepared using 20 mol % PEG750, 20 mol % EggPG, 58 mol % EggPC, 1 mol % LissRhod PE and 1 mol % 18:1 DGS-NTA(Ni) lipids and functionalized with a His-tagged neuronal adhesion molecule NrCAM (extracellular domain aa20-630) which binds to axonin-1 on neuronal membranes (FIG. 19). Following 24 hours of co-culturing giant unilamellar vesicles with the two cell types, confocal microscopy was performed to analyze and to quantify the attraction of giant unilamellar vesicles to each cell type from respective images (see Method Section). The results revealed that neuronal cells contained up to 520% more giant unilamellar vesicles compared to astrocytes, highlighting the importance of fine-tuning of attractive and repulsive interaction towards targeted delivery of giant unilamellar vesicles in a complex environment.

Example 6. Lysosomal Escape of Giant Unilamellar Vesicles for Efficient Cytoplasmic Cargo Delivery

[0898] To investigate the mechanism of giant unilamellar vesicles intracellular uptake, negatively charged giant unilamellar vesicles were incubated with REF52 cells stained for endosomal vesicles and cytoplasm. Confocal microscopy of respective cultures revealed that uptaken giant unilamellar vesicles are surrounded by endosomal membranes (FIG. 20a). This observation confirmed that giant unilamellar vesicles enter into the cells by endocytic pathways, e.g. micropinocytosis or phagocytosis, excluding other uptake mechanisms such us direct penetration or sole engulfment of the giant unilamellar vesicles. However, following uptake, a progressive loss of giant unilamellar vesicle fluorescence was observed over time and the giant unilamellar vesicles colocalized with lysosomes (identified by staining with LysoTracker Green DND-26), following 24 hours of incubation (FIG. 20b). This observation might be attributed to lysosomal degradation of the vesicles, a process which is frequently observed in case of small unilamellar vesicle-based delivery methods. Yet, many pharmacological compounds target cytoplasmic components, therefore efficient lysosomal escape mechanisms, which allow for the release of giant unilamellar vesicle cargo into the cell and avoid lysosomal degradation, are a pivotal requirement for such applications.

[0899] To circumvent this degradation, two independent lysosomal escape mechanisms were assessed (see Method Section), both based on the inhibition of the sudden decrease in pH occurring during endosome-lysosome fusion: 1) Lysosomal escape via a proton sponge mechanism by incorporation of high molecular weight poly-ethylene-imine (PEI) into the giant unilamellar vesicles; 2) Lysosomal escape via intra-lysosomal fusion by incorporation of the pH sensitive lipid DOBAQ into the giant unilamellar vesicle membrane. The retention, degradation and release of giant unilamellar vesicle cargo was exemplarily assessed by loading the giant unilamellar vesicles with the membrane impermeable dye HPTS and observing its intracellular fluorescence distribution 24 hours after incubation of respective giant unilamellar vesicles with A431D cells (FIG. 21). HPTS is a highly water-soluble pH indicator with a pKa of approx. 7.3 in aqueous buffer. In the cases of PEI-loaded giant unilamellar vesicles, HPTS fluorescence was exclusively detected in punctuated form, colocalizing with the giant unilamellar vesicle fluorescence inside the cells. This indicates HPTS retention inside the giant unilamellar vesicles and therefore endosomal or lysosomal entrapment, suggesting no successful cytoplasmic cargo release. However, for giant unilamellar vesicles containing 60 mol % DOBAQ, HPTS fluorescence could be found distributed within the whole cell body, proving successful release of HPTS from the giant unilamellar vesicles into the cytoplasm. Indeed, DOBAQ-containing giant unilamellar vesicles show polarity switching of their Z-potential at low pH, as assessed by dynamic light scattering (FIG. 22). Notably, when tracking the total intracellular giant unilamellar vesicle number over time, the inventors found that PEI loaded giant unilamellar vesicles accumulated within the cells and did not undergo degradation (FIG. 23). This suggest that, even though no cargo release is taking place, PEI loaded giant unilamellar vesicles are protected from degradation, potentially because PEI can act in the giant unilamellar vesicle lumen as a potent pH buffer preventing acidic degradation and mature lysosome formation. For approaches in which stable incorporation of giant unilamellar vesicles and their cargo into cells is required, this might represent a promising implementation mechanism.

[0900] The dynamics of endocytosis and lysosomal activity can significantly vary between transformed and non-transformed cells, therefore the inventors aimed to test the approach on primary cells as well. Towards this end, in vitro cultured primary hippocampal neurons, which represent a medically-relevant cell type, were used. Indeed, these cells represent an important target in many therapeutic procedures for neurodegeneration or neuronal tumors. When interfaced with HPTS loaded and DOBAQ containing negatively charged giant unilamellar vesicles, extensive uptake of the giant unilamellar vesicles into the neurons was observed 24 hours after incubation (FIG. 24). In order to enhance their attraction to the sialic acid containing glycocalyx, the giant unilamellar vesicles were functionalized with wheat germ agglutinin (WGA). Their accumulation in the perinuclear region suggests incorporation into the cell's intracellular trafficking machinery. Importantly, widespread distribution of HPTS fluorescence within the neuronal soma and in the dendrites was observed. This observation proves that DOBAQ based lysosomal escape of giant unilamellar vesicle cargo is also functional and compatible with primary cells.

Example 7. Targeted Delivery of Large Heavy Duty Cargoes

[0901] In order to ultimately test the cargo capacity of microfluidically-formed giant unilamellar vesicles for novel therapeutic approaches, microfluidic loading of purified baculoviruses (BV) was performed. BVs are considered as promising future candidates for transduction and handling of large amounts of genetic material in genome engineering approaches. Successful assembly and release of BV-loaded giant unilamellar vesicles was achieved (FIG. 25). Following the formation, the BV-loaded DOBAQ-containing giant unilamellar vesicles were incubated with REF52 cells for 24 hours. Confocal microscopy analysis revealed intracellular uptake of the giant unilamellar vesicles, as well as release of the BVs (stained by Hoechst 33342, see Method Section) into the cell cytoplasm. This was accompanied by the expression of mitochondrial targeted dsRed protein encoded by the BV, indicating successful giant unilamellar vesicle based transduction of mammalian cells with the baculovirus and its content. These results highlight the advantages of using giant unilamellar vesicle-based drug delivery for more efficient drug administration of advanced cargoes that would not be possible by conventional delivery methods.

Example 8. Integration of Proteins by Microfluidic Techniques

[0902] Integrin α.sub.IIbβ.sub.3 (Uniprot ID: P08514/P05106) was purified from outdated human blood platelets using TBS and Triton X-100 as described by WO 2018/228894 A1. Integrin α.sub.IIbβ.sub.3 was reconstituted into large unilamellar vesicles by the detergent removal method. Therefore, dried egg PC was dissolved in a buffer containing 0.1% of Triton X-100. Integrin α.sub.IIbβ.sub.3 was added to a 1:1000 integrin-lipid ratio. The solution was incubated at 37° C. for 2 hours in a shaker at 600 rpm. Triton X-00 was removed in two subsequent washing steps of 3.5 hours using 50 mg/ml SM-2 Biobeads. The size distribution of liposomes and integrin-liposomes was measured by dynamic light scattering in a Malvern Zetasizer Nano ZS setup (Malvern, UK) to be around 100 to 140 nm.

[0903] Simultaneously, polymer shell-stabilized giant unilamellar vesicles were formed as described in example 1.

[0904] Following these preparatory steps, the parent polymer shell stabilized giant unilamellar vesicles were fused with the Integrin-liposomes using a pico-injection device as described in WO 2018/228894 A1.

[0905] Thereafter, the formed polymer-shell stabilized giant unilamellar vesicles containing Integrin α.sub.IIbβ.sub.3 integrated in the lipid bilayer have been splitted in smaller polymer-shell stabilized giant unilamellar vesicles by using the microfluidic splitting unit as disclosed in Example 1.

[0906] A similar procedure has been used to integrate other transmembrane proteins into the lipid bilayer such as, for example, F.sub.0F.sub.1-ATP synthase (Uniprot ID: P0ABA0), Alpha(α)-hemolysin (Uniprot ID: P09616), and Gramicidin D (PubChem CID: 45267103).

Example 9. Production of Polymer-Shell Stabilized Giant Unilamellar Vesicles by Using a Positive Charged Surfactant

[0907] The parent polymer-shell stabilized giant unilamellar vesicles have also been produced following the inventive method and using a positive charged surfactant of Formula (I). For this purpose, the surfactant of Formula (I) has been dissolved in HFE 7100 at a concentration of 5 mM. The water phase has consisted of 1.5 mM lipids (30 mol % POPG, 69 mol % POPC, 1 mol % ATT0488 DOPE) in PBS buffer. The polymer-shell stabilized GUVs have been produced with a water to oil phase ratio of 1:5. The polymer-shell stabilized GUVs have been imaged after 1 hours (FIG. 29A). Afterwards, the polymer-shell stabilized GUVs have been released in PBS buffer using 1H,1H,2H,2H-Perfluoro-1-octanol and have been imaged without further purification (FIG. 29B).

[0908] When using this surfactant, a positive charge is produced directly at the periphery of the polymer shell, so that the polymer-shell stabilized giant unilamellar vesicles can be formed without the presence of magnesium, through direct interaction of the lipids with the surfactants (FIG. 28).

[0909] Fluorescence microscopy pictures of polymer-shell stabilized GUVs produced by using the positive charged surfactant of formula (I), and of GUVs produced by using said positive charged surfactant after release from the polymer shell are shown in FIGS. 29A and 29B.

TABLE-US-00007 TABLE 6 Lipid compositions and buffer used for production of the giant unilamellar vesicles in the Examples of the disclosure Functional Ligand/ Production b./release b./ Example modification Moiety/Macromolecule Composition functionalization mix Nr NHS-based: WGA(wheat germ agglutinin); + 5 mol % Palmitic acid NHS PBS + 10 mM MgCl.sub.2, palmitic acid lysosomal escape 1 mol % LissRhod PE, PBS, NHS lipid 20 mol % EggPG, 10 mg/ml Wheat germ agglutinin 14 mol % EggPC 60 mol % DOBAQ, AuNP 1 mol % palmitic acid NHS, PBS + 10 mM MgCl.sub.2, 1 mol % LissRhod PE, PBS, 20 mol % DOPG, 6 μM L-cysteine, 10 μg/ml 50 nm 78 mol % DOPC Au nanoparticles, 1% BSA BSA, poly-L-Lysine; bFGF, EGF, 5 mol % palmitic acid NHS, PBS + 10 mM MgCI.sub.2 WGA, anti-cadherin, anti-α4 1 mol % LissRhod PE, PBS, integrin, IL2, insulin, bradykinin 54 mol % EggPC, 0.5-6 μM proteins and peptides Tat-(HIV)-GFP, cystein, CD95L, 20 mol % EggPG, e-cadherin, fibronectin 20 mol % PEG-750 PE NTA-based: add of His-Tag ProtG to released 1 mol % 18:1 DGS-NTA(Ni), PBS + 10 mM MgCl.sub.2, DGS-NTA giant unilamellar vesicles , and 1 mol % LissRhod PE, PBS, (Ni) lipid then IgG antibody 20 mol % EggPG, 3 μM His-tagged Protein G, 78 mol % EggPC 3 μM anti-CD3 AF488, 1% BSA add of NrCAM 1 mol % 18:1 DGS-NTA(Ni), PBS + 10 mM MgCl.sub.2 1 mol % LissRhod PE, PBS, 20 mol % EggPG, 3 μM His-tagged NrCAM 58 mol % EggPC, 20 mol % PEG750 PE DSPE-RGD RGD 0/1/2/10 mol % DSPE-RGD, 1 mol % LissRhod PE, 20 mol % DOPG, 79/78/77/69 mol % DOPC NTA and 1% mol 18:1 DGS-NTA(Ni), PBS + 10 mM MgCl.sub.2, Biotin bound 1 mol % 18:1-12:0 Biotin PE, PBS, to the lipid 1 mol % LissRhod PE, 1.5 μM 6xHis- GFP, 1 mol % DOPE, 1.5 μM Atto425-SAV, 20 mol % DOPG, 1.5 μM Alexa Fluor 647-NHS 77 mol % DOPC Ester PEG: 1 mol % LissRhod PE, PBS + 10 mM MgCl.sub.2 or only PEG750 PE, 5/10/20/50 mol % PBS for DOTAP; PEG1000 PE PEG350/750/1000 PE, PBS in the lipid 15 mol % EggPG, or DOTAP mixture for 79/74/64/34 mol % EggPC SUV production PEG350, PE, 1 mol % LissRhod PE, PBS + 10 mM MgCl.sub.2 or only 5/10/20/50 mol % PBS for DOTAP; PEG350/750/1000 PE, PBS 50(49) mol % EggPG or DOTAP 44/39/29/0 mol % EggPC Lysosome Example Escape Composition Production buffer/release buffer Nr PEI 1 mol % LissRhod PE, PBS + 10 mM MgCl.sub.2 + 44 μg/ml 20 mol % EggPG, Polyethylenimine (PEI), 79 mol % EggPC, PBS DOBAQ 1 mol % LissRhod PE, PBS + 60 mM MgCl.sub.2 20 mol % EggPG, PBS 19 mol % EggPC, 60 mol % DOBAQ,