The most widely taught plasma membrane model is the fluid mosaic model, proposed by S.J. Singer and Garth Nicolson in 1972, which describes the membrane as a two-dimensional fluid of phospholipids studded with proteins that drift and rotate within the plane. That picture remains the foundation of how biologists think about cell membranes, but five decades of research have layered so many refinements on top of it that the original sketch now looks almost quaint. The membrane turns out to be far more organized, asymmetric, and actively managed than the 1972 version suggested.
What the Fluid Mosaic Model Actually Says
Singer and Nicolson proposed that integral membrane proteins are globular molecules arranged so that their water-loving portions stick out into the surrounding fluid while their water-repelling portions sit buried in the fatty interior of the membrane. These proteins float in a matrix of phospholipids organized as a fluid bilayer, making the whole structure “formally analogous to a two-dimensional oriented solution” of proteins dissolved in a viscous lipid solvent.1PubMed. The fluid mosaic model of the structure of cell membranes The word “mosaic” captures the patchwork of different proteins; “fluid” captures their ability to move laterally through the lipid sheet.
This was a genuine conceptual leap. Earlier ideas had placed proteins in static layers coating the outside of the lipid bilayer, or alternating with lipids in a rigid lattice. The fluid mosaic model showed how thermodynamic principles predicted that proteins would naturally embed themselves in the bilayer rather than just coat it. And the model made a testable prediction: if the membrane is fluid, then proteins and lipids should diffuse laterally. That prediction was confirmed when researchers used photobleaching techniques to directly measure how fast molecules move within the plane of the membrane.2PubMed Central. Light and life in Baltimore–and beyond
Why the Original Model Is Not Enough
If the membrane were truly just a uniform fluid with randomly drifting proteins, cells would have a hard time doing most of the things they actually do: sorting signals, building specialized surface regions, or keeping different parts of the same cell biochemically distinct. Researchers noticed early on that proteins do not always diffuse as freely as expected. Some are practically immobile; others are confined to small corrals. Lipids themselves are not uniformly distributed. The original model treated the bilayer as a mostly passive solvent for proteins, but we now know the lipids are active players that organize into distinct domains and communicate with proteins in specific ways.
Even Singer and Nicolson acknowledged that some lipid might interact specifically with membrane proteins, but they treated that as a minor footnote. In the decades since, that footnote has grown into an entire subfield. The membrane is better understood today as a patchwork of microenvironments with different compositions, different physical properties, and different biological roles, all shifting on timescales from microseconds to minutes.
Lipid Rafts and Membrane Domains
One of the most debated refinements to the fluid mosaic model is the lipid raft hypothesis. Rafts are small, dynamic clusters enriched in cholesterol and sphingolipids that are thought to exist within the broader phospholipid sea. The idea is that these clusters form a more tightly packed, ordered phase that floats within the looser, more disordered surrounding membrane, a bit like ice patches floating in a pond (though the analogy is imperfect, since rafts are liquid, not solid).3PubMed Central. Plasma membrane organization and function: moving past lipid rafts
These cholesterol-and-sphingolipid-rich platforms are proposed to serve as organizing centers for cell signaling. Certain receptors preferentially partition into rafts, and the close quarters may help them interact with the molecules they need to relay a signal.4PubMed Central. Cholesterol and Sphingolipid Enriched Lipid Rafts as Therapeutic Targets in Cancer Beyond the classic cholesterol-sphingomyelin raft, researchers have identified a third type of microdomain called ceramide-rich platforms, which have a gel-like structure and play roles in processes ranging from immune signaling to cell death.5PubMed Central. Sphingolipids and lipid rafts: Novel concepts and methods of analysis
The raft concept is not without controversy. Rafts are tiny and transient in living cells, making them extremely difficult to observe directly. Much of the evidence comes from model membranes in the lab, where researchers can trigger large-scale phase separation by cross-linking specific lipid components.6PubMed Central. Crosslinking a lipid raft component triggers liquid ordered-liquid disordered phase separation in model plasma membranes In those artificial systems, the transition from nanometer-scale domains to micron-scale ones happens abruptly at a line tension of roughly 0.3 piconewtons.7PubMed Central. Line Tension Controls Liquid-Disordered + Liquid-Ordered Domain Size Transition in Lipid Bilayers Whether and how that translates to living cells is still being worked out. A commemorative review marking the fluid mosaic model’s 50th anniversary noted that researchers remain split between those who view the membrane as a simple fluid and those who see it as highly structured, with the truth almost certainly sitting between those extremes.8PubMed Central. Cholesterol- and actin-centered view of the plasma membrane: updating the Singer-Nicolson fluid mosaic model to commemorate its 50th anniversary
The Two Sides Are Not the Same
One feature the original fluid mosaic model largely glossed over is that the inner and outer leaflets of the bilayer have different lipid compositions, and cells spend energy keeping them that way. Phosphatidylserine and phosphatidylethanolamine are concentrated on the inner, cytoplasm-facing leaflet, while phosphatidylcholine and sphingomyelin dominate the outer leaflet.9PubMed Central. Regulation of phospholipid distribution in the lipid bilayer by flippases and scramblases
This asymmetry is not just a curiosity. Cells use it as a signaling system. When a cell is damaged or dying, phosphatidylserine gets exposed on the outer surface, acting as an “eat me” flag that tells immune cells called phagocytes to engulf and dispose of it. Maintaining the normal arrangement requires dedicated protein machinery. Flippases use energy from ATP to pull specific lipids inward. Floppases push lipids outward. Scramblases, by contrast, randomize lipids between the two leaflets without requiring energy, and their activation is what allows phosphatidylserine to appear on the surface during programmed cell death or blood clotting.10Current Biology. Transbilayer lipid asymmetry – Section: Transbilayer transport of lipids: flippases, floppases and scramblases
Recent work has identified specific scramblases responsible for regulating the distribution of particular lipids. A protein called TMEM63B, for instance, acts as a scramblase that responds to changes in membrane curvature and thickness, controlling how phosphatidylcholine and sphingomyelin are distributed. A single amino acid change in TMEM63B can make it permanently active, disrupting the normal lipid asymmetry of the membrane.11PubMed Central. Membrane structure-responsive lipid scrambling by TMEM63B to control plasma membrane lipid distribution
How Thickness Shapes Protein Behavior
A membrane protein that spans the bilayer has a hydrophobic core that needs to match the hydrophobic thickness of the surrounding lipid. When there is a mismatch, either the lipid distorts around the protein, or the protein adjusts its shape, or both. These adjustments are not trivial: they can change how the protein works.12PubMed. Lipids do influence protein function-the hydrophobic matching hypothesis revisited Molecular simulations show that the transport function of membrane channels can be directly modulated by how well the bilayer thickness matches the channel’s hydrophobic region.13PubMed. Bilayer hydrophobic thickness and integral membrane protein function
This has a striking organizational consequence. When two membrane proteins each distort the bilayer around them, the deformations overlap if the proteins are close enough, and the result can be an energetically favorable attraction. In other words, the membrane itself mediates protein-protein interactions that drive clustering, even when the proteins have no direct affinity for each other.14PubMed. Bilayer-thickness-mediated interactions between integral membrane proteins This adds yet another layer of organization that the fluid mosaic model, with its image of proteins freely drifting through a uniform solvent, did not anticipate.
The Sugar Coat That Bends the Membrane
Models of the plasma membrane often stop at the outer leaflet, but living cells are covered by a dense sugar-rich layer called the glycocalyx. Made of complex carbohydrate chains attached to lipids and proteins, this coat sits on every human cell and does more than just protect the surface. Large, flexible sugar polymers in the glycocalyx can generate crowding pressure strong enough to physically bend the membrane underneath them.15PubMed Central. Glycocalyx Curving the Membrane: Forces Emerging from the Cell Exterior
Biophysical modeling confirms that the amount of curvature depends on how densely the glycocalyx polymers are packed and how long they are. When the glycocalyx interacts with the membrane’s own spontaneous curvature tendency, the bending effect gets amplified. Experiments with glycosylated transmembrane proteins show they can drive the assembly of thin, highly curved protrusions called filopodia on the cell surface. The glycocalyx can also sense existing curvature, preferentially accumulating on parts of the membrane that are already bent.16PubMed Central. Biophysical modeling of membrane curvature generation and curvature sensing by the glycocalyx This feedback loop between shape and composition is the kind of active, dynamic behavior that static diagrams of the membrane completely miss.
How the Membrane Fuses and Repairs Itself
A flat bilayer is a useful starting picture, but cells constantly merge membranes together and tear them apart. Every time a vesicle delivers cargo to the cell surface, or a cell secretes a hormone, two lipid bilayers must fuse into one. The molecular machines responsible are SNARE proteins, which form tight bundles that pull opposing membranes close enough for their lipids to mix.17PubMed Central. The Multifaceted Role of SNARE Proteins in Membrane Fusion The process passes through identifiable intermediate stages: the membranes dock, the proximal leaflets merge into a hemifusion state, and then a full fusion pore opens.18PubMed Central. Membrane fusion intermediates via directional and full assembly of the SNARE complex Force-clamp experiments using atomic force microscopy have measured the energy barriers at each step, revealing that the main hurdle is getting past the thin layer of water between the two bilayers and coaxing lipids to splay outward from their orderly arrangement.19PubMed Central. SNARE-mediated membrane fusion trajectories derived from force-clamp experiments
Cells also need to patch the membrane when it tears. Mechanical stress, bacterial toxins, and immune attacks can all punch holes in the bilayer. The repair response kicks in within seconds, driven by a rush of calcium into the cell. One pathway uses ESCRT proteins, the same machinery cells use to bud vesicles during other processes. A calcium-sensing protein called ALG-2 detects the calcium spike and recruits ESCRT components to the wound, which then pinch off the damaged patch as a small vesicle shed into the surroundings.20PubMed Central. Mechanism of Ca2+-triggered ESCRT assembly and regulation of cell membrane repair A separate pathway uses annexin proteins to form a temporary “scab” over the breach, followed by calpain enzymes that cleave the scab and release it as microvesicles, clearing out the damaged material.21Journal of Cell Biology. Cellular wound healing: A two-step mechanism of plasma membrane repair by annexins and calpains – Section: Healing by shedding: A two-phase model of membrane repair Yet another mechanism involves lysosomal enzymes that trigger endocytosis of the lesion, pulling damaged membrane inside the cell for degradation.22PubMed Central. Damage control: cellular mechanisms of plasma membrane repair The cell does not rely on just one strategy; it deploys whichever combination fits the size and nature of the wound.
How Ion Channels Fit the Picture
Ion channels are among the most functionally important proteins in the membrane, and their behavior illustrates why a good membrane model matters. Potassium channels, for instance, achieve exquisite selectivity: they allow potassium ions through while blocking the slightly smaller sodium ions. Structural studies of the KcsA potassium channel revealed that the selectivity filter works not by preferring potassium through tighter binding but by presenting a large energy barrier that blocks sodium from entering the filter when potassium is present.23PubMed Central. Mechanism of potassium-channel selectivity revealed by Na(+) and Li(+) binding sites within the KcsA pore Molecular simulations of the MthK potassium channel showed that the channel’s gate and its selectivity filter are allosterically coupled, meaning the geometry at one end of the pore controls ion flow at the other end.24Nature Communications. Molecular mechanism of a potassium channel gating through activation gate-selectivity filter coupling
Super-resolution microscopy is now revealing that channels and transporters are not randomly scattered across the membrane surface. Instead, they cluster into nanodomains, and their spatial arrangement and dynamics within those domains play a regulatory role.25PubMed Central. Super-resolution microscopy to study membrane nanodomains and transport mechanisms in the plasma membrane This finding ties channel function back to the broader question of membrane organization: understanding where proteins sit, and why they sit there, is inseparable from understanding how they work.
Lab Models That Stand In for Real Membranes
Much of what we know about membrane physics comes from simplified systems built in the lab. The most commonly used model membranes include supported lipid bilayers (flat sheets of lipid deposited on a solid surface), giant unilamellar vesicles (cell-sized spherical bubbles made of lipid), and giant plasma membrane vesicles (blebs pinched directly from living cells, capturing some of the real membrane’s complexity).26PubMed. Model membrane platforms to study protein-membrane interactions Each system makes different trade-offs between experimental control and biological realism. A supported bilayer is easy to image with a microscope but sits on glass, which can restrict how lipids and proteins move. A giant vesicle floats freely but lacks the cytoskeleton and glycocalyx that real cells have.
Lipid bilayers have served as laboratory stand-ins for biological membranes ever since the bimolecular lipid sheet was first identified as the membrane’s core structure.27Europe PMC. Model answers to lipid membrane questions A key early finding from model membranes was that cholesterol smooths out abrupt phase transitions. Pure phospholipid vesicles undergo a sharp shift from a rigid, gel-like state to a fluid state at a specific temperature, but adding cholesterol in equal amounts eliminates that sharp transition entirely.28Biochimica et Biophysica Acta (BBA) – Biomembranes. Phase transitions in phospholipid vesicles Fluorescence polarization and permeability measurements concerning the effect of temperature and cholesterol That discovery helped explain why cholesterol is so abundant in animal cell membranes: it keeps the membrane in a workable, intermediate state across a range of temperatures rather than snapping between extremes.
Advances in imaging are steadily closing the gap between model systems and living cells. Spectrally resolved super-resolution microscopy can now reveal compositional differences in live-cell membranes with about 30-nanometer spatial resolution, recording the positions and spectral signatures of individual polarity-sensing molecules.29PubMed. Spectrally Resolved, Functional Super-Resolution Microscopy Reveals Nanoscale Compositional Heterogeneity in Live-Cell Membranes These tools are beginning to show us the real membrane in something close to its native state, rather than forcing us to extrapolate from stripped-down lab versions.
Membranes That Break the Mold
Everything described so far applies primarily to the membranes of bacteria and eukaryotes, which are built from phospholipids linked to glycerol by ester bonds and arranged as a bilayer of two separate leaflets. Archaea, a third domain of life that thrives in extreme environments, use a fundamentally different chemistry. Their membrane lipids are ether-linked rather than ester-linked, and some archaeal lipids are large enough to span the entire membrane, creating a monolayer instead of a bilayer. This monolayer is nearly impermeable to ions and protons, making it well suited to survive in boiling acid or concentrated salt.30PubMed. The essence of being extremophilic: the role of the unique archaeal membrane lipids
Recent neutron diffraction studies have added a twist: mixing bilayer-forming diether lipids into archaeal monolayer membranes actually improves their structural stability under temperature stress compared to pure monolayer membranes. The pure monolayer system, counterintuitively, showed more variability in its spacing under fluctuating conditions.31PubMed Central. Bilayer-Forming Lipids Enhance Archaeal Monolayer Membrane Stability This challenges the simple story that monolayers are always tougher. It also reminds us that any single plasma membrane model is an approximation shaped by which organisms we happen to study most.
From Understanding to Application
Detailed knowledge of membrane structure has practical payoffs. Liposomes, which are artificial vesicles made from lipid bilayers, are among the most successful drug-delivery vehicles in medicine. Most clinically approved lipid nanoparticles fall in the range of 50 to 300 nanometers across, and their design draws directly on research into lipid-drug and lipid-protein interactions.32PubMed Central. Emerging research and clinical development trends of liposome and lipid nanoparticle drug delivery systems The lipid nanoparticles that delivered mRNA in widely used COVID-19 vaccines are perhaps the most prominent recent example: their effectiveness depends on the lipid shell fusing with cell membranes and releasing its cargo inside, a process governed by exactly the fusion mechanics and lipid-phase behavior that membrane researchers have spent decades characterizing. The cancer research community has also taken interest in lipid rafts as potential therapeutic targets, since raft domains can influence how receptors are activated and how tumor cells respond to drugs.33PubMed Central. Cholesterol and Sphingolipid Enriched Lipid Rafts as Therapeutic Targets in Cancer
The plasma membrane model has traveled a long way from the neat two-dimensional mosaic of 1972. What started as a useful simplification has been replaced, piece by piece, by a picture of a structure that is asymmetric across its thickness, heterogeneous across its surface, mechanically active, and constantly remodeled by the cell. No single diagram captures all of that, which is why researchers keep building new models, both conceptual and physical, to test how the pieces fit together.