Lipid vesicle shape analysis from populations using light video microscopy and computer vision

Colony-like Protocell Superstructures

We report the formation, growth, and dynamics of model protocell superstructures on solid surfaces, resembling single cell colonies. These structures, consisting of several layers of lipidic compartments enveloped in a dome-shaped outer lipid bilayer, emerged as a result of spontaneous shape transformation of lipid agglomerates deposited on thin film aluminum surfaces. Collective protocell structures were observed to be mechanically more stable compared to isolated spherical compartments. We show that the model colonies encapsulate DNA and accommodate nonenzymatic, strand displacement DNA reactions. The membrane envelope is able to disassemble and expose individual daughter protocells, which can migrate and attach via nanotethers to distant surface locations, while maintaining their encapsulated contents. Some colonies feature "exocompartments", which spontaneously extend out of the enveloping bilayer, internalize DNA, and merge again with the superstructure. A continuum elastohydrodynamic theory that we developed suggests that a plausible driving force behind subcompartment formation is attractive van der Waals (vdW) interactions between the membrane and surface. The balance between membrane bending and vdW interactions yields a critical length scale of 236 nm, above which the membrane invaginations can form subcompartments. The findings support our hypotheses that in extension of the "lipid world hypothesis", protocells may have existed in the form of colonies, potentially benefiting from the increased mechanical stability provided by a superstructure.

Effects of hydrocarbon chain on the vesicle size distribution, kinetics of average size, bending modulus, and elastic modulus of lipid membranes

The effects of the hydrocarbon chain of lipids on the size distribution of giant unilamellar vesicles (GUVs), kinetics of average size, bending modulus, and elastic modulus of membranes have been investigated. 1,2-dioleoyl-sn-glycero-3-phosphocholine (18:1 (Δ9-Cis) PC (DOPC)), 1,2-dipalmitoleoyl-sn-glycero-3-phosphocholine (16:1 (Δ9-Cis) PC), and 1,2-ditridecanoyl-sn-glycerol-3-phosphocholine (13:0 PC (DTPC)) lipids were considered. The number of hydrocarbons in a chain of the corresponding lipid was 18, 16, and 13. GUVs were prepared using the natural swelling method under incubation times of 20, 40, 60, 90, 120, and 180 min. The size distribution of vesicles was fitted using the lognormal distribution. The average sizes of DOPC, 16:1 (Δ9-Cis) PC, and DTPC-GUVs increased with the incubation time until 120 min, and then remained steady at 16.7 ± 0.2, 15.2 ± 0.4 and 12.0 ± 0.3 µm for the corresponding lipids. The average size at equilibrium state increased with the number of hydrocarbons. The incubation time-dependent average size was fitted with an exponential growth equation, and then the kinetic constants of 0.028 ± 0.004, 0.036 ± 0.007, and 0.083 ± 0.009 min^-1 for DOPC, 16:1 (Δ9-Cis) PC, and DTPC-GUVs, respectively, were obtained. The equilibrium size distribution was fitted by the theoretical equation, and the bending modulus for DOPC, 16:1 (Δ9-Cis) PC, and DTPC membranes were 19.5 ± 0.2, 18.5 ± 0.1 and 14.3 ± 0.1 k_BT, respectively. The bending modulus increased with the number of hydrocarbons. The elastic modulus of these membranes was 261 mN/m with a 4% fluctuation. The correlation between the average size and the square root of the bending modulus was supported by theoretical analysis.

Effects of cholesterol on the size distribution and bending modulus of lipid vesicles

The influence of cholesterol fraction in the membranes of giant unilamellar vesicles (GUVs) on their size distributions and bending moduli has been investigated. The membranes of GUVs were synthesized by a mixture of two elements: electrically neutral lipid 1, 2-dioleoyl-sn-glycero-3-phosphocholine (DOPC) and cholesterol and also a mixture of three elements: electrically charged lipid 1,2-dioleoyl-sn-glycero-3-phospho-(1′-rac-glycerol) (DOPG), DOPC and cholesterol. The size distributions of GUVs have been presented by a set of histograms. The classical lognormal distribution is well fitted to the histograms, from where the average size of vesicle is obtained. The increase of cholesterol content in the membranes of GUVs increases the average size of vesicles in the population. Using the framework of Helmholtz free energy of the system, the theory developed by us is extended to explain the experimental results. The theory determines the influence of cholesterol on the bending modulus of membranes from the fitting of the proper histograms. The increase of cholesterol in GUVs increases both the average size of vesicles in population and the bending modulus of membranes.

Intelligent fluorescence image analysis of giant unilamellar vesicles using convolutional neural network

Background

Fluorescence image analysis in biochemical science often involves the complex tasks of identifying samples for analysis and calculating the desired information from the intensity traces. Analyzing giant unilamellar vesicles (GUVs) is one of these tasks. Researchers need to identify many vesicles to statistically analyze the degree of molecular interaction or state of molecular organization on the membranes. This analysis is complicated, requiring a careful manual examination by researchers, so automating the analysis can significantly aid in improving its efficiency and reliability.

Results

We developed a convolutional neural network (CNN) assisted intelligent analysis routine based on the whole 3D z-stack images. The programs identify the vesicles with desired morphology and analyzes the data automatically. The programs can perform protein binding analysis on the membranes or state decision analysis of domain phase separation. We also show that the method can easily be applied to similar problems, such as intensity analysis of phase-separated protein droplets. CNN-based classification approach enables the identification of vesicles even from relatively complex samples. We demonstrate that the proposed artificial intelligence-assisted classification can further enhance the accuracy of the analysis close to the performance of manual examination in vesicle selection and vesicle state determination analysis.

Conclusions

We developed a MATLAB based software capable of efficiently analyzing confocal fluorescence image data of giant unilamellar vesicles. The program can automatically identify GUVs with desired morphology and perform intensity-based calculation and state decision for each vesicle. We expect our method of CNN implementation can be expanded and applied to many similar problems in image data analysis.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12859-022-04577-2.

Size Distributions and Yields of Giant Vesicles Assembled on Cellulose Papers and Cotton Fabric

Lamellar phospholipid stacks on cellulose paper vesiculate to form cell-like giant unilamellar vesicles (GUVs) in aqueous solutions. The sizes and yields of the GUVs that result and their relationship to the properties of the cellulose fibers are unknown. Here, we report the characteristics of GUVs produced on four different cellulose substrates, three disordered porous media consisting of randomly entangled cellulose fibers (high-purity cellulose filter papers of different effective porosities), and an ordered network of weaved cellulose fibers (cotton fabric). Large numbers of GUVs formed on all four substrates. This result demonstrates for the first time that GUVs form on cotton fabric. Despite differences in the effective porosities and the configuration of the cellulose fibers, all four substrates yielded populations of GUVs with similar distribution of diameters. The distribution of diameters of the GUVs had a single well-defined peak and a right tail. Ninety-eight percent of the GUVs had diameters less than the average diameter of the cellulose fibers (∼20 micrometers). Cotton fabric produced the highest yield of GUVs with the lowest sample-to-sample variation. Moreover, cotton fabric is reusable. Fabric used sequentially produced similar crops of GUVs at each cycle. At the end of the sequence, there was no apparent change in the cellulose fibers. Cellulose fibers thus promote the vesiculation of lamellar phospholipid stacks in aqueous solutions.