
Biological membranes define distinct compartments such as cells and organelles essential for maintaining homeostasis and facilitating communication. Membrane proteins play pivotal roles in selectively regulating the exchange of molecules and information across these membranes by functioning as channels, receptors, enzymes, and transporters. To investigate these processes, researchers have developed model membrane systems consisting of chemically and physically well-defined lipid bilayers that can replicate the structure and function of biological membranes. These synthetic systems provide controlled environments that can facilitate a systematic study of membrane-associated processes.
Over time, model membranes have advanced from basic liposomes to more sophisticated systems such as black lipid membranes (BLMs) and nanodiscs tailored for specific experimental needs. These platforms have become indispensable tools for studying membrane protein structure and function (1-3). Their chemical compositions can be precisely adjusted by incorporating synthetic phospholipids and additives such as cholesterol, making the design of the membrane system critical for experimental success. Model membranes encompass both planar systems, including supported lipid bilayers (SLBs) and free-standing lipid bilayers (FLBs), as well as curved membranes such as small, large, and giant unilamellar vesicles (GUVs) distinguished by their sizes and shapes.
Single-molecule techniques provide us with the ability to directly observe movements, interactions, and conformational changes of individual biomolecules including membrane proteins. In contrast to bulk biochemical methods, which offer ensemble-averaged results, single-molecule approaches can uncover subtle molecular-level heterogeneities (4, 5). When integrated with model membranes, these techniques can yield unique insights into the structure and function of membrane proteins, revealing key aspects such as structural dynamics, stoichiometry, interaction kinetics, and rare molecular events. This deeper understanding of membrane-associated activities offers new perspectives on pathological dysregulation and drug-target interactions involving membrane receptor proteins (6, 7).
Given rapid advances in membrane biology and continuous development of new model systems, this review examined the evolution of various model membranes, including recent innovations such as large-area FLBs designed to integrate with single-molecule techniques. We highlighted key applications and explored how adaptations of these systems could enhance their utility for specific experimental purposes, advancing our understanding of membrane-related processes.
Many biological processes rely on voltage difference across membranes (known as membrane potential) and the flow of ions through membranes (referred to as ionic currents). These processes highlight capacitor-like properties of membranes and dynamic changes in membrane resistance during cellular activity. Traditional electrophysiology techniques have been essential for studying these phenomena (1), while single-molecule ion channel recordings provide deeper insights into individual channel behavior (8). By clamping transmembrane voltage, these methods allow precise control and measurement of electrical currents through ion channels. Model membrane systems such as BLMs, nanodiscs, and GUVs serve as complementary platforms for studying membrane behavior under controlled conditions, enabling investigations at both ensemble and single-molecule levels.
In a traditional BLM setup, a planar lipid bilayer spans an aperture, separating two aqueous compartments. Unlike closed vesicles that limit physical accessibility, BLMs provide unrestricted access to both sides of the membrane. This configuration enables precise control over ionic conditions and buffer composition in each compartment making it particularly suitable for studying gating dynamics, ion selectivity, and drug interactions with electrophysiology (9, 10).
Membrane proteins, such as ion channels, transporters, or pores, are reconstituted into the bilayer, allowing their function to be studied in isolation. To perform electrophysiological measurements, electrodes placed in each compartment measure ionic currents and control the transmembrane voltage, providing a stable platform to examine protein behavior (Fig. 1a). One example of using this technique at the single molecule level is the study of gramicidin, a protein that forms dimeric ion channels. Myers et al. were the first to observe ion transfer across BLM mediated by gramicidin using the setup described above (11). They subsequently investigated the effect of the membrane thickness and tension on channel lifetime by varying the lipid composition (12). At that time, the necessity of creating membranes individually limited the number of experiments. Today, however, technological advancements allow for high-throughput systems capable of recording the activity of single-molecule channel proteins in BLMs (13).
Nanodiscs and bicelles, called membrane nanostructures, are small lipid bilayers (∼10-15 nm and ∼20-50 nm in diameter, respectively). Nanodiscs are stabilized by amphipathic membrane scaffold proteins that wrap around edges of the bilayer, maintaining its structural integrity. On the other hand, bicelles employ lipid molecules with different chain lengths to form stable lipid bilayers. While nanodiscs are discoidal, bicelles exhibit a dynamic morphology that varies with temperature or lipid composition (14). These systems provide a native-like lipid environment while keeping both sides of the membrane accessible, making them ideal for high-resolution single-molecule studies (15, 16). Unlike proteoliposomes, which may contain multiple proteins and introduce heterogeneity, membrane nanostructures can facilitate the incorporation of one or a few proteins, improving the precision of biophysical measurements.
Solid-state Nuclear Magnetic Resonance (ssNMR) is widely employed for determining protein structures. For membrane proteins, biologically relevant structures are obtained when the proteins are embedded within a lipid bilayer, as this mimics their native environment. To facilitate such studies, membrane nanostructures are frequently utilized due to their small size and ability to align either parallel or perpendicular to the magnetic field (Fig. 1b) (17, 18). For instance, Park et al. demonstrated the use of bicelles to investigate the structure of the transmembrane domain of the major coat protein Pf1 bacteriophage, revealing that these domains are tilted at specific angles within the membrane (19).
GUVs are large, spherical lipid bilayers (10-100 microns in diameter) that mimic properties of cellular membranes while offering precise control over experimental conditions. Their large surface areas facilitate the formation of stable seals with glass pipettes during patch-clamp experiments, making GUVs ideal for high-resolution single-channel recordings (20). These recordings enable studies of ion channels, transporters, and mechanosensitive proteins, providing insights into their gating dynamics, ion selectivity, and responses to chemical or mechanical stimuli (21).
Recent advancements such as automated planar patch-clamp systems developed by Barthmes et al. (22) have improved the efficiency of GUV-based studies (Fig. 1c). These systems use perforated substrates with suction to position GUVs over small pores, forming high-resistance seals without manual pipette handling. This configuration supports voltage-clamp measurements while providing precise control over membrane tension and curvature known to be critical parameters for investigating mechanosensitive ion channels. Although large sizes of GUVs can introduce variability in patch formation, their flexibility in controlling lipid composition and mechanical properties makes them valuable tools for studying protein functions under synthetic conditions.
Biomolecular interactions typically occur at the piconewton (pN) scale, necessitating precise techniques to measure forces and understand mechanics underlying molecular behavior. Force spectroscopy can detect nanometer-scale displacements caused by these small forces, providing insights into key biological processes such as ligand-receptor binding, protein and membrane mechanics, and mechanosensitive events (23). However, studying membrane-associated processes presents additional challenges due to thermal fluctuations inherent to biological membranes such as plasma membranes and lipid bilayers known to introduce noise and complicate measurements (24, 25). Stable experimental conditions, including precise control of temperature, ionic strength, and mechanical constraints, are essential for reliable data acquisition.
The following three principal techniques dominate the field of force spectroscopy: atomic force microscopy (AFM), optical tweezers, and magnetic tweezers. Each method has distinct strengths. AFM is best suited for high-force measurements, while optical and magnetic tweezers excel in capturing weaker interactions. Together, these tools enable both equilibrium and dynamic studies, facilitating exploration of binding affinities, unfolding events, and force-induced molecular changes in biomolecular systems (26).
AFM was initially developed for high-resolution surface imaging, achieving sub-nanometer resolution by scanning a sharp probe across surfaces (27). It has since evolved into a powerful tool for visualizing nanoscale topographies, including membrane structures (28). AFM has also found applications in single-molecule studies by functionalizing probes with specific ligands or biomolecules, allowing precise interactions with target samples. Through incremental force application, AFM can measure rupture forces and force-extension relationships, revealing key mechanical properties such as stiffness, unfolding events, and conformational changes.
Early AFM studies have used purple membranes from Halobacterium salinarum to investigate bacteriorhodopsin, a membrane protein with a dense, stable, and crystalline within the membrane. These features make purple membranes particularly well-suited for single-molecule force spectroscopy, enabling detailed studies of bacteriorhodopsin’s mechanical unfolding (29, 30). As this field progresses, supported lipid bilayers (SLBs) formed on solid substrates using synthetic lipids through the Langmuir–Blodgett technique or liposome rupture become widely adopted for membrane protein studies. Although SLBs provide stable platforms for high-resolution imaging of membrane topographies (31, 32), their interactions with a rigid substrate can alter membrane dynamics and protein behavior, compromising physiological relevance (33, 34).
To overcome these limitations, researchers have developed perforated substrates with nanopores, enabling FLBs to form over pores while being supported by adjacent SLBs (Fig. 2a). A two-chamber AFM setup using porous silicon nanowells allows FLBs to separate two aqueous compartments, mimicking native-like conditions (35). Petrosyan et al. (36) have further advanced this technique by creating lipid bilayers on polymethyl methacrylate (PMMA) substrates with submicron pores, successfully capturing the folding structure of bacteriorhodopsin. This configuration minimizes substrate interference, preserves native membrane properties, and ensures stability required for high-resolution measurements, bridging the gap between physiological relevance and experimental precision.
Optical tweezers use focused laser beams to trap and manipulate microscopic particles by exerting forces through optical radiation pressure and electromagnetic field gradients. First developed by Arthur Ashkin, who was awarded the 2018 Nobel Prize in Physics, this technique of using optical tweezers applies forces typically ranging from 0.1 to 100 pN, offering a less invasive alternative to AFM (37). Trapping requires objects with a refractive index higher than their surroundings, with polystyrene microspheres commonly used due to their optical and biocompatible properties.
Optical tweezers have been applied to study processes such as ligand binding on cell surfaces and membrane tether extraction (38, 39). While experiments on BLMs have been conducted (40), maintaining trapping stability remains a challenge. Spherical membranes provide more reliable trapping when filled with a higher refractive index material. For example, silica beads coated with lipid membranes have been used to create spherical supported bilayers, enabling stepwise observations of protein-membrane interactions (41, 42) (Fig. 2b). Similarly, Wang et al. have enhanced the trapping of GUVs by filling them with 60% iodixanol (refractive index: 1.429), enabling them to study DNA hairpin folding dynamics (43). Although their study was focused on DNA, it illustrated the potential of optical tweezers for membrane protein research by eliminating surface interference and providing a stable, non-toxic environment for precise measurements.
Magnetic tweezers offer a non-contact manipulation technique using external magnetic fields to exert forces on magnetic particles. They provide precise force control, typically ranging from 0.01 to 100 pN, with high temporal resolution exceeding 1 kHz (44). While magnetic tweezers historically had lower spatial and force resolution than optical tweezers, recent technological advances have significantly improved their precision (45, 46). Furthermore, enhancements in the mechanical stability of their molecular tethering systems have enabled the prolonged observation of numerous molecular transitions (47), making them competitive for many applications. A key advantage of magnetic tweezers is their ability to apply both tensile forces and torque, enabling more versatile experimental designs than other force spectroscopy techniques (48, 49).
In a typical setup, magnetic beads are attached to the sample and manipulated by external magnetic fields, while high-resolution tracking monitors the movement of the beads. In studies of various target proteins, several researchers have investigated proteins lacking their transmembrane domains and membrane environments, such as research on the zippering and unzipping of SNARE proteins—key components of the membrane fusion machinery (50-53). However, in other cases, such as studies on the folding pathways of integral transmembrane proteins, a membrane environment is essential to provide greater physiological relevance.
Recent applications of magnetic tweezers have combined them with bicelle- or liposome-based model membranes to investigate folding dynamics of helical membrane proteins in native-like lipid environments (Fig. 2c). For example, bicelles have been used to study the unfolding and refolding of GlpG, a rhomboid protease (54), revealing cooperative unfolding and significant refolding hysteresis and demonstrating the importance of bicelles in preserving native-like folding dynamics more effectively than detergent-based systems. A follow-up study (55) has confirmed these findings and identified a conserved N-to-C-terminal folding pathway shared by GlpG and β2-adrenergic receptor. Notably, the study highlighted the impact of membrane curvature on protein folding dynamics: smaller vesicles with higher curvature improved folding efficiency by lowering the energy barrier for transmembrane helix insertion, a dynamic not observed with flatter membranes. Similarly, bicelles were employed to study the folding pathway of human glucose transporter 3 (GLUT3), a member of the solute carrier protein family, allowing the observation of folding states influenced by lipid composition. Additionally, the study demonstrated the effects of the endoplasmic reticulum membrane protein complex (EMC) on GLUT3 folding by incorporating EMC into the bicelles (56).
Fluorescence imaging is a fundamental tool in biological research, enabling visualization of specific molecules or structures through fluorescent markers and providing real-time insights into molecular processes. Among these methods, single-molecule fluorescence imaging has emerged as a powerful approach for directly observing individual molecules, offering unique perspectives on molecular dynamics (57). However, this approach presents several technical challenges. The fluorescence intensity of individual molecules is significantly lower than that of bulk fluorophores, making detection more difficult and increasing vulnerability to photobleaching. Additionally, single-molecule detection is more susceptible to noise compared to ensemble measurements, necessitating a high signal-to-noise ratio for reliable results.
Single-molecule fluorescence resonance energy transfer (smFRET) and single-particle tracking (SPT) are the most widely used single-molecule fluorescence techniques. These techniques enable precise measurements of intra- and intermolecular distance changes and allow researchers to track individual molecules over time. Together, these methods have revolutionized the study of biomolecular dynamics, providing unprecedented resolution and clarity. The following sections will explore these advanced techniques in detail, focusing first on smFRET, which measures molecular interactions and structural changes, followed by SPT, which monitors real-time movement of individual particles within model membrane environments.
smFRET is a highly sensitive technique used to measure distance changes within or between molecules at the single-molecule level (58). It involves attaching two fluorescent molecules — a donor and an acceptor — at specific sites on the target molecule and measuring energy transfer efficiency from the donor to the acceptor. This efficiency is highly distance-dependent, enabling precise measurements within the 1-10 nanometer range. Researchers have used smFRET to study dynamic processes such as protein folding, enzymatic structural changes, and nucleic acid dynamics (59).
A key advantage of smFRET lies in its compatibility with model membranes such as liposomes and nanodiscs, which can provide controlled environments for studying membrane-associated processes. For example, a study employing smFRET and vesicles reported the development of an efficient single-vesicle content-mixing assay for SNARE-mediated membrane fusion at the single-molecule level (Fig. 3a) (60). The researchers detected content mixing by observing changes in the FRET signal resulting from the fusion of a vesicle containing a fluorescently labeled DNA hairpin with another vesicle containing a complementary DNA strand. To further dissect these processes, nanodiscs can serve as native-like platforms for reconstituting single copies of membrane proteins. In one study, smFRET has been used to reveal dynamic conformational states of SNARE proteins during membrane fusion (Fig. 3b) (61). A follow-up study has demonstrated that the accessory protein complexin can inhibit SNARE assembly at the C-terminal region (62). Nanodiscs provide an optimal environment for such studies by isolating individual membrane proteins in a controlled, native-like lipid setting, ensuring precise measurements of membrane protein dynamics and advancing our understanding of key cellular processes.
SPT is an advanced microscopy technique that reconstructs trajectories of individual fluorescently labeled molecules, enabling high-resolution tracking in both space and time. Initially developed to study molecular dynamics in diverse biological contexts, SPT has become increasingly valuable for investigating membrane systems (63). Membrane-associated processes such as lipid diffusion, protein clustering, and vesicle trafficking rely heavily on spatial and temporal dynamics of individual molecules. To replicate native membrane environments for these studies, platforms such as SLBs and FLBs have been adopted. SLBs are often used because the objective lens can be positioned close to the samples, facilitating high-resolution imaging (64-66). However, SLBs can limit lateral diffusion (33, 34), reducing their ability to fully replicate native membrane conditions. This limitation has motivated researchers to explore alternative platforms such as BLMs.
Although BLMs offer improved physiological relevance, their configurations present challenges to high-magnification imaging due to the limited working distance of objective lenses. Fluidic chambers with small reservoirs provide partial solutions (67). However, challenges such as optical reflections from membrane support and physical distance between the membrane and the objective lens persist (68). These factors can reduce photon collection efficiency and introduce background noise, complicating single-molecule imaging. Pérez-Mitta et al. have addressed these challenges by developing a custom chamber for SPT on FLBs (69). Their design features an open top, allowing the objective lens to approach the membrane closely and improve photon acquisition. A 39° laser tilt can further minimize background fluorescence by reducing reflections from the membrane support, significantly enhancing imaging quality (Fig. 3c). This setup enables precise visualization of membrane protein dynamics in a native-like environment, overcoming limitations of SLBs.
SPT on FLBs offers new opportunities to study complex membrane phenomena, including protein clustering, lipid raft formation, and effects of membrane tension and curvature on various membrane-associated processes. Realizing these insights will require highly efficient labeling and detection of single molecules and vesicles, along with precise, quantitative analyses of their dynamics and interactions (70, 71). These advancements will be critical for advancing our understanding of cellular processes and guiding the development of targeted therapeutic strategies that exploit membrane-dependent pathways.
Model membrane techniques have evolved alongside advancements in single-molecule methods, undergoing structural and technical refinements to integrate with these state-of-the-art approaches. As highlighted earlier, early applications of BLMs were primarily focused on electrophysiology. With advancement of pore protein engineering, BLM technology has extended to applications such as nanopore-based protein sequencing. Similarly, force spectroscopy has benefited from innovations, including the development of FLBs on porous substrates. Techniques for filling GUVs with alternative media have further expanded their utility. Adjustments of FLB positioning and the use of tilted laser incidence have enhanced single-molecule fluorescence imaging, exemplifying continuous efforts to optimize model membrane systems for single-molecule studies.
The future of model membrane techniques combined with single-molecule approaches holds tremendous potential. Further miniaturization and integration could lead to lab-on-a-chip platforms capable of high-throughput membrane protein analysis. Emerging imaging technologies such as super-resolution microscopy and light-sheet microscopy may be adapted for use with model membranes, providing unprecedented spatial and temporal resolution. Advances in hybrid platforms that combine force spectroscopy with fluorescence techniques are likely to enable simultaneous mechanical and structural studies of membrane proteins.
As these technologies evolve, they are expected to provide deeper insights into membrane-mediated cellular processes, advancing our understanding of drug-target interactions, signaling pathways, and fundamental principles governing cellular organization and functions. Integrating model membranes with high-throughput and high-precision techniques are likely to lead to discovery of new membrane-related phenomena, opening new avenues for therapeutic development. These innovations are promising in bridging the gap between model systems and real biological environments, moving us closer to a comprehensive understanding of membrane dynamics in health and diseases.
We thank Minkwon Cha for his valuable discussions. This work was supported by the National Research Foundation of Korea (NRF) grant funded by the Korea government (MSIT) (NRF-2022R1C1C1012176, RS-2023-00218927, and RS-2024-00344154).
The authors have no conflicting interests.
![]() |
![]() |