Integral membrane proteins are proteins permanently embedded in the lipid bilayer of a cell’s membrane, and they participate in nearly every major cellular process, from transporting molecules and relaying signals to anchoring cells together and catalyzing chemical reactions within the membrane itself. They account for roughly 30% of the proteins encoded in sequenced genomes, yet for decades they resisted structural study because removing them from their native lipid environment tends to destabilize them.1PubMed Central. Computational analysis of membrane proteins: the largest class of drug targets That tension between biological importance and technical difficulty has shaped what we know about them and how quickly we have learned it.
What Keeps Them Anchored in the Membrane
The core reason integral proteins stay put is thermodynamic: their membrane-spanning segments are rich in water-repelling amino acids, and pulling those segments out of the oily interior of the bilayer would be energetically costly. Statistical analyses of known transmembrane helices show that large hydrophobic residues like leucine, valine, isoleucine, and phenylalanine, along with the small amino acid glycine, dominate the stretches that sit inside the membrane’s hydrophobic core.2PubMed. Structure-based statistical analysis of transmembrane helices Polar and charged amino acids, by contrast, cluster at the boundaries where the membrane meets water, or face outward into the aqueous environment on either side of the bilayer. This arrangement is not random: charged amino acids like glutamate and aspartate appear at both the inner and outer membrane interfaces, while positively charged residues tend to gather on the side facing the cell’s interior.2PubMed. Structure-based statistical analysis of transmembrane helices
That asymmetry in charge distribution helps determine which end of the protein faces inward and which faces outward, a property called topology. A well-established principle known as the “positive-inside rule” captures this: positively charged amino acids in the loops connecting transmembrane segments tend to remain on the cytoplasmic side of the membrane, while negatively charged residues promote translocation of loops to the exterior.3PubMed Central. Interplay between hydrophobicity and the positive-inside rule in determining membrane-protein topology Together, hydrophobicity drives segments into the membrane and charge distribution orients them correctly.4PubMed Central. Lipid-protein interactions drive membrane protein topogenesis in accordance with the positive inside rule
The Main Structural Types
Not all integral proteins look alike once they are embedded. Three broad structural categories cover most of them, and each sits in the membrane differently.
Alpha-Helical Bundles
The most common type threads one or more alpha-helical segments across the bilayer. A single-pass (or “bitopic”) protein crosses the membrane once, with one helix connecting an extracellular domain to an intracellular domain. Multi-pass (or “polytopic”) proteins wind back and forth through the membrane multiple times. A comprehensive structural survey found that the typical transmembrane helix is about 24 residues long, and in a full-length integral membrane protein the transmembrane helices make up a little over half of the protein’s total amino acid content.5PubMed. Comprehensive analysis of the numbers, lengths and amino acid compositions of transmembrane helices in prokaryotic, eukaryotic and viral integral membrane proteins of high-resolution structure In polytopic proteins, helices can shield slightly polar residues from the surrounding lipids by packing them into the interior of the helical bundle, something a single-pass protein cannot do. This is why single-pass transmembrane helices tend to be even more hydrophobic on average than those in multi-pass proteins.6PubMed. Statistical analysis of predicted transmembrane alpha-helices
Beta-Barrel Proteins
A second structural family uses beta-strands instead of alpha-helices. These strands form a barrel-shaped cylinder that spans the membrane. Beta-barrel integral proteins are found mainly in the outer membranes of bacteria and mitochondria, not in the plasma membranes of animal cells. Getting them folded and inserted requires dedicated molecular machines: bacteria use a complex called the beta-barrel assembly machine (BAM), while mitochondria rely on the sorting and assembly machine (SAM). Recent work has revealed that beta-barrels fold through striking intermediate states in which the new barrel temporarily merges with the barrel of the assembly machine itself, forming a “hybrid barrel” before the finished protein is released into the membrane.7PubMed. β-barrel membrane proteins fold via hybrid-barrel intermediate states
Monotopic Proteins
A third category, sometimes overlooked, consists of monotopic membrane proteins. These embed into only one leaflet of the bilayer rather than spanning the entire membrane. They dip into the lipid from one side, often through hydrophobic loops, an amphipathic helix, or a wedge-shaped domain, without emerging on the other side.8PubMed Central. Monotopic Membrane Proteins Join the Fold Simulations suggest these proteins fall into two camps: shallow inserters that mainly interact at the lipid-water interface, and deep inserters that significantly distort the local bilayer thickness around them. Deeper insertion tends to correlate with more positively charged residues on the membrane-facing foot of the protein.9PubMed. Interaction of monotopic membrane enzymes with a lipid bilayer: a coarse-grained MD simulation study
How Integral Proteins Get Inserted
Cells do not simply shove these proteins into the membrane. Dedicated molecular machinery guides the process. In bacteria, the SecYEG translocon is the central channel: a ribosome synthesizing a new membrane protein docks onto SecYEG, and the growing polypeptide is threaded into the channel, which can open a lateral gate to release hydrophobic transmembrane segments sideways into the lipid bilayer.10Journal of Cell Science. Membrane insertases at a glance In the endoplasmic reticulum of animal and plant cells, the equivalent channel is the Sec61 complex, which works on the same general principle.
Working alongside these translocons are insertases, smaller proteins that assist with folding and membrane insertion. In bacteria, a key insertase called YidC can operate either in partnership with the Sec translocon or independently. On its own, YidC inserts smaller membrane proteins; when cooperating with SecYEG, it helps fold the transmembrane helices of larger, more complex proteins.11PubMed Central. YibN, a bona fide interactor of the bacterial YidC insertase with effects on membrane protein insertion and membrane lipid production Recent structural work has shown that during insertion of multi-pass proteins, the nascent chain can cross over the translocon toward its backside, where YidC is recruited in a way that depends on the growing chain itself. YidC then promotes folding of transmembrane helices before they actually insert into the lipid, consistent with the idea that helices fold first and insert second rather than inserting as unstructured strands.12bioRxiv. Substrate-induced assembly and functional mechanism of the bacterial membrane protein insertase SecYEG-YidC
Hydrophobic Mismatch and Membrane Adaptation
Once a protein is sitting in the membrane, the hydrophobic thickness of its transmembrane region does not always match the hydrophobic thickness of the surrounding lipid bilayer perfectly. When there is a mismatch, something has to give. Molecular dynamics simulations show that when a protein’s hydrophobic segment is longer than the bilayer is thick (positive mismatch), the protein tends to tilt and the nearby lipids become slightly more ordered. When the protein’s hydrophobic segment is shorter than the bilayer (negative mismatch), the membrane bends locally and certain charged residues on the protein stretch, or “snorkel,” toward the water layer.13Biophysical Journal. Molecular Dynamics Simulations of Hydrophobic Mismatch in Transmembrane Peptides and Lipid Bilayers
The picture is more nuanced than the standard textbook version, though. While some proteins clearly force the surrounding membrane to thicken or thin to match their hydrophobic surface, experiments with synthetic peptides designed to test this showed no detectable thickness change in the lipid bilayer, even when a mismatch was deliberately introduced.14Biophysical Journal. Hydrophobic Mismatch between Helices and Lipid Bilayers Glycophorin A dimers, for instance, remain stable in the membrane despite poor hydrophobic match, using a combination of dimer tilting and local membrane distortion to accommodate the difference.15PubMed Central. The role of hydrophobic matching on transmembrane helix packing in cells The upshot is that the cell has multiple strategies for coping with mismatch, and real membranes are more flexible in this regard than a simple model would predict.
Transport Across the Membrane
One of the most visible jobs of integral proteins is ferrying molecules across the membrane that would not otherwise cross it. Ion channels represent the passive end of this spectrum. These proteins form pores through the bilayer that allow specific ions to flow down their concentration gradient. Selectivity is remarkably precise: individual amino acids lining the channel’s pore determine which ion passes through and which is excluded.16PubMed. Structural basis of ion channel permeation and selectivity Gating, or the opening and closing of the channel, can occur either at the bundle crossing formed where helices converge or at the selectivity filter itself, depending on the channel family. In certain potassium channels, for example, it is the selectivity filter, not the wider inner gate, that controls ion flow.17PubMed Central. Structural basis of ion permeation gating in Slo2.1 K+ channels
On the active-transport side, carrier proteins and pumps use energy to move substrates against their concentration gradient. Many secondary active transporters work by an “alternating access” mechanism, in which the protein flips between conformations that expose a binding site to one side of the membrane and then the other.18PubMed. Common folds and transport mechanisms of secondary active transporters Primary active transporters such as ABC exporters add another layer of complexity: they use ATP hydrolysis to drive large-scale conformational changes between inward-facing and outward-facing states, with the nucleotide-binding domains and the transmembrane domains cooperating in a choreographed sequence of movements.19PubMed Central. Mechanistic picture for conformational transition of a membrane transporter at atomic resolution
Signaling and Receptor Functions
G protein-coupled receptors, or GPCRs, are the largest and most diverse family of membrane receptors in the human genome.20PubMed. Transmembrane signaling by G protein-coupled receptors All GPCRs share the same basic architecture: seven transmembrane helices connected by loops, with an extracellular face that detects a signal and an intracellular face that activates downstream pathways. When a signaling molecule binds the extracellular surface, the receptor undergoes a conformational shift that allows it to activate a G protein on the cytoplasmic side, which in turn triggers enzymes or ion channels inside the cell. The range of stimuli GPCRs detect is staggering, spanning light, hormones, neurotransmitters, and odor molecules, all through variations on that same seven-helix scaffold.
Integrins operate differently but are equally important. These are adhesion receptors that physically connect a cell’s interior scaffolding to the external environment. What makes integrins unusual is bidirectional signaling: signals from inside the cell can switch the receptor to a high-affinity state for its external ligand (“inside-out” signaling), and binding an external ligand can send a signal back into the cell to trigger events like spreading or migration (“outside-in” signaling).21PubMed Central. Integrin activation The transmembrane domain is critical for this two-way communication. Disrupting the alpha-helical continuity of the integrin transmembrane segment, for instance by introducing a proline mutation that kinks the helix, blocks both directions of signal transmission.22Journal of Biological Chemistry. Structural determinants of the integrin transmembrane domain required for bidirectional signal transmission across the cell membrane
Enzymes Embedded in the Membrane
Some integral proteins are full-fledged enzymes whose active sites sit within the lipid bilayer itself, a seemingly hostile environment for the water-dependent chemistry that most enzymes perform. Rhomboid proteases illustrate how evolution solved this problem. The crystal structure of a bacterial rhomboid shows six transmembrane helices arranged so that one short helix dips only partway into the membrane, starting deep inside it right at the catalytic residue. This arrangement creates a water-accessible cavity within the otherwise hydrophobic membrane interior, providing the hydrophilic environment needed for peptide bond cleavage.23PubMed Central. Structural basis for intramembrane proteolysis by rhomboid serine proteases
The surrounding lipids are not just passive bystanders, either. Work on a presenilin homolog, part of the same enzyme family linked to Alzheimer’s disease, showed that the protease’s activity was impaired in detergent micelles but promoted in an actual lipid bilayer. The bilayer stabilized the geometry of the enzyme’s active site and enhanced its interaction with a transition-state analog inhibitor.24bioRxiv. Active site geometry stabilization of a presenilin homolog by the lipid bilayer promotes intramembrane proteolysis The implication is that for intramembrane enzymes, the lipid environment is not just structural scaffolding; it actively shapes catalytic performance.
How Lipids Interact with Integral Proteins
For years, textbooks described a shell of “annular lipids” tightly wrapped around each integral protein, acting almost like a dedicated coat. That picture has largely been revised. Spectroscopic and biochemical evidence from the 1980s onward showed that these boundary lipids are not immobilized long enough to significantly influence enzyme or transporter activity, nor are they ordered in a rigid ring by the protein.25PubMed Central. The Myth of The Annular Lipids Lipids in the first shell exchange rapidly with the bulk membrane, on a timescale of nanoseconds to microseconds.
That does not mean lipid-protein interactions are unimportant, just that they are more selective and specific than a static “annular” model implies. Some lipids bind in grooves between transmembrane helices, in what is sometimes called the “non-annular” belt. These binding sites can be functionally significant. Mass spectrometry studies of the ammonia channel AmtB, for instance, measured the binding strength of different lipids at sequential binding sites and found that cardiolipin bound with a shallower increase in dissociation constant across successive binding events compared to other lipids, suggesting a higher overall affinity for these crevice sites.26PubMed Central. A step towards probing the non-annular belt of membrane proteins This kind of specific lipid binding can modulate protein conformation and activity in ways that a generic lipid shell does not.
After Insertion: Modifications That Matter
Once an integral protein is in the membrane, it can be chemically modified in ways that influence where it goes and how well it works. One well-studied modification is palmitoylation, in which a fatty acid chain is attached to a cysteine residue on the protein. For CFTR, the chloride channel whose malfunction causes cystic fibrosis, blocking palmitoylation decreases the amount of protein at the cell surface, disrupts its processing through the secretory pathway, and reduces its ion channel function.27Biochemical Journal. S-palmitoylation regulates biogenesis of core glycosylated wild-type and F508del CFTR in a post-ER compartment The enzyme DHHC-7 was identified as a palmitoyl transferase that interacts with CFTR and increases its steady-state levels, pointing to palmitoylation as a checkpoint during the protein’s trafficking from the endoplasmic reticulum through the Golgi to the plasma membrane.
Palmitoylation does not matter equally for every membrane protein, though. Experiments with transmembrane adaptor proteins found that mutating the palmitoylation sites of LAT (a key signaling adaptor in immune cells) trapped it in the Golgi and led to its degradation rather than delivery to the cell surface. But for related adaptors PAG and NTAL, the same kind of mutation had only a weak effect on plasma membrane localization.28Journal of Cell Science. The role of palmitoylation and transmembrane domain in sorting of transmembrane adaptor proteins This suggests that some proteins rely heavily on palmitoylation for correct sorting, while others use alternative signals encoded in their transmembrane domains or other features.
Why They Are the Primary Targets for Drugs
The majority of marketed therapeutics act on membrane proteins, and this is no accident. Sitting on the cell surface, integral proteins are the most accessible points at which a drug can intercept a signaling cascade, block a transport process, or modulate an enzymatic reaction.29PubMed Central. Drugging Membrane Protein Interactions GPCRs alone are the targets of an estimated one-third of all approved drugs. Ion channels, transporters, and receptor tyrosine kinases account for much of the rest.
The challenge has always been on the research side. Producing integral membrane proteins in functional forms for drug screening has historically been difficult because detergent extraction tends to strip away the lipid interactions that keep the protein properly folded.30PubMed. Membrane protein production and formulation for drug discovery That is one reason why the structural biology of membrane proteins lagged so far behind that of soluble proteins for decades. The gap has narrowed considerably in recent years, driven largely by advances in cryo-electron microscopy.
How Scientists Now Visualize Integral Proteins
For most of the history of structural biology, X-ray crystallography was the only route to atomic-resolution structures, and membrane proteins were notoriously hard to crystallize. The field was transformed by improvements in single-particle cryo-electron microscopy. High-resolution three-dimensional reconstructions of membrane proteins by cryo-EM have become routine, with 300-kilovolt electron microscopes now the standard workhorse.31PubMed. Cryo-EM studies of membrane proteins at 200 keV New structures are being determined at a pace that would have been unthinkable twenty years ago.32PubMed Central. Membrane protein structural biology in the era of single particle cryo-EM
One innovation that has paired well with cryo-EM is the use of lipid nanodiscs, tiny disc-shaped patches of lipid bilayer held together by a belt protein. A membrane protein can be reconstituted into a nanodisc, giving it a near-native lipid environment while keeping the overall particle small and monodisperse enough for high-resolution imaging.33PubMed. Lipid Nanodiscs as a Tool for High-Resolution Structure Determination of Membrane Proteins by Single-Particle Cryo-EM Nanodiscs have also been used with solution NMR to study protein dynamics and lipid interactions in conditions that more closely mimic the real membrane than a detergent micelle does.34PubMed. Nanodiscs for Structural Biology in a Membranous Environment Researchers have even pushed cryo-EM to image membrane proteins embedded in intact liposomes, bypassing any reconstitution into artificial environments altogether.35PubMed Central. Cryo-EM analysis of a membrane protein embedded in the liposome
Evolutionary Patterns in Membrane Protein Architecture
Integral proteins are ancient. Analyses tracing the distribution of protein structural folds across the three domains of life found that the folds shared universally by bacteria, archaea, and eukaryotes sit at the base of the evolutionary tree of protein architecture, indicating they arose very early in life’s history.36PubMed. Universal sharing patterns in proteomes and evolution of protein fold architecture and life Among these early folds are the transmembrane helix bundles used by ion channels and transporters, which makes sense: controlling what crosses a membrane boundary is one of the most fundamental requirements for any cell.
Single-pass transmembrane proteins, though structurally simpler, show their own evolutionary trends. Genome-wide analyses have examined how properties like transmembrane segment length, topology, and the propensity to form dimers vary with organism complexity and the specific membrane a protein resides in. These studies reveal that single-pass proteins are not just stripped-down versions of multi-pass proteins. They have their own distinct evolutionary trajectories, with their transmembrane segments adapted to the particular thickness and composition of the membrane they inhabit, whether that is the plasma membrane, the endoplasmic reticulum, or the Golgi.37Biochimica et Biophysica Acta (BBA) – Biomembranes. Evolution and adaptation of single-pass transmembrane proteins The length of the transmembrane helix itself correlates with a protein’s intracellular localization, reflecting the varying bilayer thickness of different cellular compartments.