The composition and types of lipids largely determine the mechanical properties of the bilayer constituting a GUV. A bilayer composed of DMPC/Chol exhibits unique shape transitions when subjected to osmotic shock. Under a strong hypertonic osmotic shock, the aqueous fluid inside the sGUV is expelled, leading to the deformation of the GUV bilayer. The observed shape transitions are influenced primarily by the outward flux of water and the bilayer's bending rigidity. These deformations manifest in various forms, including starfish shapes and dumbbells, depending on the strength of the osmotic shock and the lipid composition of the bilayer.
Interestingly, the intrinsic properties of the DMPC-Cholesterol bilayer are favorable to induce a transition of the sGUV into a stomatocyte shape at room temperature. During this transition, the narrowing of the neck region results in vesicle-in-vesicle (VIV) formation. In the final VIV structure, the inner vesicles are completely separated from the outer vesicles. Previous studies have reported that for a bilayer composition of DMPC/Chol in a 70:30 molar ratio, complete separation of inner vesicles occurs within approximately 30 s when vesicles are synthesized in pure water and glucose is used for inducing an osmotic shock15. In contrast, we synthesized sGUVs with a composition of 67:37 DMPC: Chol molar ratio in a sucrose solution to create a density difference with the addition of glucose in the outer suspension, allowing them to settle at the bottom of the chamber for easier microscopic observation. Additionally, it is observed that the sGUV synthesis with a cholesterol content of lower than 30% molar ratio shows higher permeability and a decrease in the contrast of sucrose and glucose, which makes them difficult to settle at the bottom for microscopy observation. Under these conditions, we observed that the complete separation of the inner vesicle from the outer vesicle typically occurs over approximately 4-5 h. However, depending on the intrinsic membrane properties, it can take less or more time. Furthermore, when the obtained GUVs are left overnight after osmotic shock, nearly all VIV-shaped vesicles exhibit complete separation of the inner vesicle.
It should be mentioned that a significant percentage of vesicles transition into a VIV structure. At the same time, other shapes, such as dumbbells and distorted spheres, are also observed following the osmotic shock. Large sGUVs with a radius greater than 20 µm rarely transition into a stomatocyte shape. Instead, these sGUVs typically remain deformed in deflated non-spherical shapes, or nanotubules engulfed inside the GUV to release bending energy. The process of transformation of sGUV into cGUV proceeds as follows. When a freshly prepared sGUV sample (37 mol % cholesterol) is allowed to rest for an hour and then an osmotic shock is applied to this sGUV solution, an instantaneous transition stomatocytes is observed, and the vesicles showing VIV structure start to settle down at the bottom of the chamber; waiting for 2 h shows a significant population of VIV structure vesicles settled at the bottom of chamber. A mixed population was observed in these VIV transit vesicles. In most vesicles showing VIV structure, the outer and inner vesicles were completely separated, while in a few, the inner and outer vesicles remained attached through a neck.
In some instances, it is possible that freshly prepared GUVs cannot transit to the stomatocytic shape under osmotic shock. These cases are typically observed at 43 mol% cholesterol. In these cases, the sGUV samples were allowed to rest for more than 3-4 h at room temperature, followed by the application of osmotic shock, leading to a stomatocyte shape transition. Notably, the properties of the bilayer appear to undergo change over time, explaining the different rest times required for the two sGUVs of two different DMPC:Chol compositions to form stomatocytes and thereafter, cGUVs. The bilayers with lower cholesterol content (37 mol%) exhibit lower bending rigidity compared to those with higher cholesterol content (43 mol%). Bending rigidity tends to increase with cholesterol concentration, particularly when examined at temperatures above the melting transition temperature of DMPC (Tm = 24 °C)23. In the case of pure DMPC bilayer, a notable drop in bending rigidity occurs at the transition temperature (Tm), where the bilayer enters the ripple phase. In this phase, lipid molecules adopt a tilted packing configuration, reducing bending rigidity. Upon further increasing the temperature (Tm + 2 K), the bending rigidity rises to a specific value and stabilizes, showing only minor variations with additional temperature increases. This observation suggests that above the transition temperature, in the fluid state, the bending rigidity of the DMPC bilayer remains relatively constant24. A similar trend has been reported for bilayers composed of DMPC and lower cholesterol concentrations (<12.5%), and a tilted lipid packing is reported near the phase transition temperature of DMPC23,25. The existing literature states that the bending rigidity of bilayers containing more than 30 mol% cholesterol has higher bending rigidity at (T > Tm) but does not compare bending rigidity fluctuations around Tm at such higher cholesterol content23. Based on our experimental observations, GUVs composed of DMPC with 37-43 mol% cholesterol have suitable bending rigidity to deform GUVs into a stomatocyte shape at room temperature (~25-27 °C) under osmotic stress. GUVs incubated at T > 35 °C show tiny multiple vesicles engulfed inside the GUVs (data not shown).
Based on the DMPC/Chol phase diagram, the sGUV bilayer composition, which contains 37-43 mol% cholesterol, remains in the liquid-ordered (Lo) phase below and above the phase transition temperature Tm26,27. The presence of this phase is consistent with our observed membrane properties, including reduced membrane permeability and enhanced mechanical stability28,29. In Figure 7, the oblate and prolate deformation of the cGUV indicates that the outer and inner regions exhibit higher conductivity compared to the annular region. This confirms that the conductivity differences within the cGUV persist over an extended period. For a more detailed investigation of cGUV electrodeformation, refer to our previous work21. The low permeability of the membrane assists in altering and maintaining the conductivities in the different regions of cGUVs (outer, annular, and inner) for a reasonable time. The conductivity of the inner region is set during the osmotic shock to sGUVs by modifying the salt concentration of the solution used for the osmotic shock; the inner vesicle engulfed the solution inside with the adjusted conductivity, thereby defining the inner region's conductivity. The conductivity of the annular region is determined by the conductivity of the hydrating solution used during the formation of the cGUV. The conductivity of the outer region can be modified after the cGUVs are formed. This is achieved by either adding a salt solution to the external medium or through dilution to reach the desired conductivity level in the outer region.
In prepared cGUVs, both deflated and well-formed quasi-spherical are observed. sGUV synthesis through electroformation typically exhibits high membrane tension, ranging from 10-5 to 10-7 mN/m, which promotes the formation of highly spherical vesicles. However, when these GUVs were subjected to an osmotic shock, the membrane tension decreased due to efflux of water from the inside of the vesicle, leading to deflation of the vesicles. Following the dissipation of the osmotic shock, a mixed population of cGUVs with both good sphericity and deflated vesicles was observed. Our previous work found that the membrane tension for cGUVs lies between 10-7 and 10-9 mN/m. The membrane tension in this range shows a reasonable agreement with theoretical models predicting the electrodeformation behavior of cGUVs, further supporting the validity of our experimental observations21. This also indicates that the cGUVs have lower membrane tension compared to sGUVs synthesized using the electroformation method. It was also observed that with time (after 6 h), a significant population of cGUVs attained sphericity, indicating recovery of membrane tension.
This method shows that a significant population of cGUVs can be synthesized, as shown in Figure 6, when the lipid composition is made of DMPC/Chol (63:37) with good reproducibility. In some cases, when temperature and osmotic shock conditions are not adequately controlled, the yield can vary. Future studies should focus on analyzing the role of cholesterol under different temperature conditions and its domain formation within the bilayer. Additionally, understanding how osmotic shock redistributes cholesterol and alters lipid packing, leading to increased bilayer fluidity and changes in bilayer mechanical properties, would provide valuable insights.
In the literature, fabricated microfluidic devices employing jet flows have been used to project liposomes onto planar lipid bilayers to achieve vesicle-in-vesicle geometry. However, this method is typically suitable for the entrapment of smaller-sized liposomes and an uncontrolled number of them within the vesicle30. The method proposed in this work offers a convenient and straightforward approach for synthesizing compound vesicles (cGUVs) that contain a single micron-sized inner vesicle for conducting concise biophysical studies. It is particularly useful for experiments where spatial electrical conductivity variation and encapsulation of molecular dyes are required. Additionally, to extend the preparation technique, the electroformation of GUVs using the existing setup can be replaced with a portable GUV preparation kit. This kit includes a feature for heating electrodes as needed to reach 40 °C above the phase transition temperature of DMPC lipid, and is easy to operate31,32.
However, some limitations must be addressed to develop an improved method for cGUV formation. The synthesis process is optimal at low salt concentrations or without salt. Since higher salt concentrations modify bilayer properties, preventing the GUV from transitioning into the VIV structure, difficulties are encountered in their synthesis. This method to synthesize cGUVs is time-consuming, as it takes hours to complete the process. Furthermore, this method is limited to specific lipid compositions; adding cholesterol introduces complexity, making it less suitable for studying GUV bilayers composed of a single lipid type. Therefore, a comprehensive investigation of other lipid compositions is essential to explore this method further for forming cGUV.