Pore-spanning membranes as a tool to investigate lateral lipid membrane heterogeneity

Over the years, it has become more and more obvious that lipid membranes show a very complex behavior. This behavior arises in part from the large number of different kinds of lipids and proteins and how they dynamically interact with each other. In vitro studies using artificial membrane systems have shed light on the heterogeneity based on lipid-lipid interactions in multicomponent bilayer mixtures. Inspired by the raft hypothesis, the coexistence of liquid-disordered (ld) and liquid-ordered (lo) phases has drawn much attention. It was shown that ternary lipid mixtures containing low- and high-melting temperature lipids and cholesterol can phase separate into a lo phase enriched in the high-melting lipids and cholesterol and a ld phase enriched in the low-melting lipids. Depending on the model membrane system under investigation, different domain sizes, shapes, and mobilities have been found. Here, we describe how to generate phase-separated lo/ld phases in model membrane systems termed pore-spanning membranes (PSMs). These PSMs are prepared on porous silicon substrates with pore sizes in the micrometer regime. A proper functionalization of the top surface of the substrates is required to achieve the spreading of giant unilamellar vesicles (GUVs) to obtain PSMs. Starting with lo/ld phase-separated GUVs lead to membrane heterogeneities in the PSMs. Depending on the functionalization strategy of the top surface of the silicon substrate, different membrane heterogeneities are observed in the PSMs employing fluorescence microscopy. A quantitative analysis of the heterogeneity as well as the dynamics of the lipid domains is described.

PDMS as a Substrate for Lipid Bilayers

PDMS (polydimethylsiloxane) is a cheap, optically clear polymer that is elastic and can be easily and quickly fabricated into a wide array of microscale and nanoscale architectures, making it a versatile substrate for biophysical experiments on cell membranes. It is easy to imagine many new experiments will be devised that require a bilayer to be placed upon a substrate that is flexible or easily cast into a desired geometry, such as in lab-on-a-chip, organ-on-chip, and microfluidic applications, or for building accurate membrane models that replicate the surface structure and elasticity of the cytoskeleton. However, PDMS has its limitations, and the extent to which the behavior of membranes is affected on PDMS has not been fully explored. We use AFM and fluorescence optical microscopy to investigate the use of PDMS as a substrate for the formation and study of supported lipid bilayers (SLBs). Lipid bilayers form on plasma-treated PDMS and show free diffusion and normal phase transitions, confirming its suitability as a model bilayer substrate. However, lipid-phase separation on PDMS is severely restricted due to the pinning of domains to surface roughness, resulting in the cessation of lateral hydrodynamic flow. We show the high-resolution porous structure of PDMS and the extreme smoothing effect of oxygen plasma treatment used to hydrophilize the surface, but this is not flat enough to allow domain formation. We also observe bilayer degradation over hour timescales, which correlates with the known hydrophobic recovery of PDMS, and establish a critical water contact angle of 30°, above which bilayers degrade or not form at all. Care must be taken as incomplete surface oxidation and hydrophobic recovery result in optically invisible membrane disruption, which will also be transparent to fluorescence microscopy and lipid diffusion measurements in the early stages.

Domain Localization by Graphene Oxide in Supported Lipid Bilayers

The gel-phase domains in a binary supported lipid bilayer (SLB) comprising dioleoylphosphatidylcholine (DOPC) and dipalmitoylphosphatidylcholine (DPPC) were localized on graphene oxide (GO) deposited on a SiO₂/Si substrate. We investigated the distribution of the gel-phase domains and the liquid crystalline (L_α) phase regions in DOPC+DPPC-SLB on thermally oxidized SiO₂/Si substrates with GO flakes to understand the mechanism of the domain localization on GO. Fluorescence microscopy and atomic force microscopy revealed that the gel-phase domains preferably distributed on GO flakes, whereas the fraction of the L_α-phase increased on the bare SiO₂ surface which was not covered with the GO flakes. The gel-phase domain was condensed on GO more effectively at the lower cooling rate. We propose that nucleation of the gel-phase domain preferentially occurred on GO, whose surface has amphiphilic property, during the gel-phase domain formation. The domains of the liquid ordered (L_o) phase were also condensed on GO in a ternary bilayer containing cholesterol that was phase-separated to the L_o phase and the liquid disordered phase. Rigid domains segregates on GO during their formation process, leaving fluid components to the surrounding region of GO.

In Vitro Membrane Platform for the Visualization of Water Impermeability across the Liquid-Ordered Phase under Hypertonic Conditions

Passive water penetration across the cell membrane by osmotic diffusion is essential for the homeostasis of cell volume, in addition to the protein-assisted active transportation of water. Since membrane components can regulate water permeability, controlling compositional variation during the volume regulatory process is a prerequisite for investigating the underlying mechanisms of water permeation and related membrane dynamics. However, the lack of a viable in vitro membrane platform in hypertonic solutions impedes advanced knowledge of cell volume regulation processes, especially cholesterol-enriched lipid domains called lipid rafts. By reconstituting the liquid-ordered (L_o) domain as a likeness of lipid rafts, we verified suppressed water permeation across the L_o domains, which had yet to be confirmed with experimental demonstrations despite a simulation approach. With the help of direct transfer of the L_o domains from vesicles to supported lipid membranes, the biological roles of lipid composition in suppressed water translocation were experimentally confirmed. Additionally, the improvement in membrane stability under hypertonic conditions was demonstrated based on molecular dynamics simulations.

From LUVs to GUVs─How to Cover Micrometer-Sized Pores with Membranes

Pore-spanning membranes (PSMs) are a versatile tool to investigate membrane-confined processes in a bottom-up approach. Pore sizes in the micrometer range are most suited to visualize PSMs using fluorescence microscopy. However, the preparation of these PSMs relies on the spreading of giant unilamellar vesicles (GUVs). GUV production faces several limitations. Thus, alternative ways to generate PSMs starting from large or small unilamellar vesicles that are more reproducibly prepared are highly desirable. Here we describe a method to produce PSMs obtained from large unilamellar vesicles, making use of droplet-stabilized GUVs generated in a microfluidic device. We analyzed the lipid diffusion in the free-standing and supported parts of the PSMs using z-scan fluorescence correlation spectroscopy and fluorescence recovery after photobleaching experiments in combination with finite element simulations. Employing atomic force indentation experiments, we also investigated the mechanical properties of the PSMs. Both lipid diffusion constants and lateral membrane tension were compared to those obtained on PSMs derived from electroformed GUVs, which are known to be solvent- and detergent-free, under otherwise identical conditions. Our results demonstrate that the lipid diffusion, as well as the mechanical properties of the resulting PSMs, is almost unaffected by the GUV formation procedure but depends on the chosen substrate functionalization. With the new method in hand, we were able to reconstitute the syntaxin-1A transmembrane domain in microfluidic GUVs and PSMs, which was visualized by fluorescence microscopy.

Pore-Spanning Plasma Membranes Derived from Giant Plasma Membrane Vesicles

Giant plasma membrane vesicles (GPMVs) are a highly promising model system for the eukaryotic plasma membrane. The unresolved challenge, however, is a path to surface-based structures that allows accessibility to both sides of the plasma membrane through high-resolution techniques. Such an approach would pave the way to advanced chip-based technologies for the analysis of complex cell surfaces to study the roles of membrane proteins, host-pathogen interactions, and many other bioanalytical and sensing applications. This study reports the generation of planar supported plasma membranes and for the first-time pore-spanning plasma membranes (PSPMs) derived from pure GPMVs that are spread on activated solid and highly ordered porous silicon substrates. GPMVs were produced by two different vesiculation agents and were first investigated with respect to their growth behavior and phase separation. Second, these GPMVs were spread onto silicon substrates to form planar supported plasma membrane patches. PSPMs were obtained by spreading of pure GPMVs on oxygen-plasma activated porous substrates with pore diameters of 3.5 μm. Fluorescence micrographs unambiguously showed that the PSPMs partially phase separate in a mobile ordered phase surrounded by a disordered phase, which was supported by cholesterol extraction using methyl-β-cyclodextrin.

How Does Liquid-Liquid Phase Separation in Model Membranes Reflect Cell Membrane Heterogeneity?

Although liquid-liquid phase separation of cytoplasmic or nuclear components in cells has been a major focus in cell biology, it is only recently that the principle of phase separation has been a long-standing concept and extensively studied in biomembranes. Membrane phase separation has been reconstituted in simplified model systems, and its detailed physicochemical principles, including essential phase diagrams, have been extensively explored. These model membrane systems have proven very useful to study the heterogeneity in cellular membranes, however, concerns have been raised about how reliably they can represent native membranes. In this review, we will discuss how phase-separated membrane systems can mimic cellular membranes and where they fail to reflect the native cell membrane heterogeneity. We also include a few humble suggestions on which phase-separated systems should be used for certain applications, and which interpretations should be avoided to prevent unreliable conclusions.

Critical Phenomena in Plasma Membrane Organization and Function

Lateral organization in the plane of the plasma membrane is an important driver of biological processes. The past dozen years have seen increasing experimental support for the notion that lipid organization plays an important role in modulating this heterogeneity. Various biophysical mechanisms rooted in the concept of liquid-liquid phase separation have been proposed to explain diverse experimental observations of heterogeneity in model and cell membranes with distinct but overlapping applicability. In this review, we focus on the evidence for and the consequences of the hypothesis that the plasma membrane is poised near an equilibrium miscibility critical point. Critical phenomena explain certain features of the heterogeneity observed in cells and model systems but also go beyond heterogeneity to predict other interesting phenomena, including responses to perturbations in membrane composition.

Direct imaging of liquid domains in membranes by cryo-electron tomography

Images of micrometer-scale domains in lipid bilayers have provided the gold standard of model-free evidence to understand the domains' shapes, sizes, and distributions. Corresponding techniques to directly and quantitatively assess smaller (nanoscale and submicron) liquid domains have been limited. Researchers commonly seek to correlate activities of membrane proteins with attributes of the domains in which they reside; doing so hinges on identification and characterization of membrane domains. Although some features of membrane domains can be probed by indirect methods, these methods are often constrained by the limitation that data must be analyzed in the context of models that require multiple assumptions or parameters. Here, we address this challenge by developing and testing two methods of identifying submicron domains in biomimetic membranes. Both methods leverage cryo-electron tomograms of ternary membranes under vitrified, hydrated conditions. The first method is optimized for probe-free applications: Domains are directly distinguished from the surrounding membrane by their thickness. This technique quantitatively and accurately measures area fractions of domains, in excellent agreement with known phase diagrams. The second method is optimized for applications in which a single label is deployed for imaging membranes by both high-resolution cryo-electron tomography and diffraction-limited optical microscopy. For this method, we test a panel of probes, find that a trimeric mCherry label performs best, and specify criteria for developing future high-performance, dual-use probes. These developments have led to direct and quantitative imaging of submicron membrane domains in vitrified, hydrated vesicles.