10.4: Diffusion Across a Membrane - Pores

Search Fundamentals of Biochemistry

Learning Goals (ChatGPT o3-mini)

(Learning goals written by Claude, Sonnet 4.6, Anthropic)

Pore Formation: Biophysical Principles and Protein-Independent Mechanisms

• Distinguish between ion channels and membrane pores with respect to pore diameter, ion selectivity, and the range of solutes that can pass, and explain how toroidal versus protein-lined pore architectures differ in the way lipid head groups are arranged at the pore wall.
• Describe the molecular events of electroporation — from initial membrane deformation to water wire formation to stable hydrophilic toroidal pore — and explain how this physical process is exploited to introduce DNA into prokaryotic and eukaryotic cells.

Pore-Forming Proteins: Structure, Mechanism, and Biological Roles

• Compare the structural transitions that convert a soluble monomeric pore-forming protein (using cytolysin A / ClyA as the example) into a membrane-embedded oligomeric pore, and explain how PFPs such as the complement membrane attack complex (MAC) exploit pore formation to kill target cells in the innate immune response.
• Describe the structural and functional properties of VDAC and aquaporin as examples of beta-barrel porins, explaining how VDAC's voltage-dependent open/closed transitions regulate metabolite and ATP flux across the outer mitochondrial membrane, and how aquaporin's conserved central asparagines block proton transfer while permitting water flow at rates of up to one billion molecules per second.
• Explain the gating mechanisms of small (MscS) and large (MscL) mechanosensitive channels, describing how changes in membrane tension caused by osmotic stress open the pore — allowing rapid efflux of cytoplasmic solutes to prevent cell lysis — and how pore diameter differences between MscS (~13 Å open) and MscL (~30 Å) correlate with their conductance values and selectivity.

Inter- and Intracellular Communication Through Pores

• Describe the structural organization of connexins, connexons, and gap junctions, explaining how six connexin subunits assemble into a hemichannel in one cell and dock with a partner hemichannel in an adjacent cell to form a continuous intercellular conduit for ions, metabolites, and small signaling molecules.
• Describe the architecture of the nuclear pore complex (NPC) — including the cytoplasmic ring, inner ring, nuclear ring, and FG-Nup meshwork — explain how the disordered FG-repeat domains act as a size-selective filter excluding molecules above ~40 kDa while allowing passive diffusion of smaller molecules, and explain why large cargoes such as proteins and RNA require karyopherin transport factors and GTP hydrolysis to transit the pore.

Pores and Pore-Forming Proteins (PFPs)

If you form a pore in a cell bilayer, molecules of all sizes could move either way based on their electrochemical potential. They will move from regions of a higher to lower electrochemical potential in a thermodynamically favorable process. Hence, movement through pores represents a special case of facilitated diffusion when part of the driving force is a concentration gradient and part an electrical potential. Several questions might come to mind.

• What proteins are involved in pore formation?
• How is the specificity of solute movement through the pore regulated?
• What is the mechanism of pore formation?

Pore formation can lead to cell death, which is the function of some pore-forming proteins (PFPs), including the toxin Hemolysin E (also known as HlyE, ClyA, SheA) secreted from E. Coli and S. Aureus. Human proteins also form a membrane attack complex (examples include the membrane attack complex-perforin/cholesterol-dependent cytolysin (MACPF/CDC) superfamily and the membrane attack complex (MAC). The MAC is an assembly of proteins involved in the complement system (part of the effector branch of the innate immune system) that leads to the death of Gram-negative bacteria, such as E. coli. Figure \(\PageIndex{1}\) illustrates the assembly process of the membrane attack complex and the complexity of interactions required to form a lethal pore in a cell.

Figure \(\PageIndex{1}\): Human membrane attack complex. Charles Bayly-Jones et al. Philosophical Transactions of the Royal Society B: Biological Sciences, Volume 372, Issue 1726. June 2017 https://doi.org/10.1098/rstb.2016.0221. The Royal Society under the terms of the Creative Commons Attribution License http://creativecommons.org/licenses/by/4.0/

Complement protein C9 can adopt a soluble or membrane form, which, when aggregated, forms a pore that leads to cell death. This is a common feature of PFPs.

PFPs could create a pore by altering membrane lipid packing to form a toroid-like hole (Figure \(\PageIndex{2}\)) and/or by inserting into a membrane and forming a pore within the protein complex itself. In either mechanism, lipid packing is altered.

Figure \(\PageIndex{2}\): By GYassineMrabetTalk - Own work, CC BY-SA 3.0, https://commons.wikimedia.org/w/inde...?curid=3134923

Biophysical evidence supports the formation of "toroidal pores."

Lipid rearrangements in the membrane could lead to a hydrophilic or hydrophobic pore lining, as shown in Figure \(\PageIndex{3}\).

Figure \(\PageIndex{3}\): Diagram of possible lipid arrangements at the edge of a pore through a lipid bilayer. (above) In the absence of any rearrangement, the pore walls will be hydrophobic since the alkane tails are exposed. (below) Some researchers believe that the lipids at the edge reorient to line the pore wall with hydrophilic head groups. Lipid bilayer. (2023, January 15). In Wikipedia. https://en.Wikipedia.org/wiki/Lipid_bilayer

We started our study of lipid bilayers with pure lipid systems and then added membrane proteins. Let's do the same with pore formation. Electroporation is a common technique for forming pores in pure lipid bilayers and in cells. This technique is used to move a DNA with a target gene into either a prokaryotic cell (transformation) or eukaryotic cell (transfection) for exogenous gene expression. Without PFPs, this requires altering surface tension by applying an electrical potential. This forms depressions in the membrane, altering the packing of the nonpolar acyl chains. Small wire-like water columns appear in the process, which, like hydrophobic pores, ultimately rearrange into hydrophilic toroidal pores. Figure \(\PageIndex{4}\) shows snapshots of molecular dynamics simulations as a function of the time of pore formation in electroporation.

Figure \(\PageIndex{4}\): Zhou, C., Liu, K. Molecular dynamics simulation of reversible electroporation with Martini force field. BioMed Eng OnLine 18, 123 (2019). https://doi.org/10.1186/s12938-019-0743-1. Creative Commons Attribution 4.0 International License, http://creativecommons.org/licenses/by/4.0/

How does DNA pass through the pores in the bilayer? In pure lipid vesicles, it appears to pass through by electrophoresis. Most students are familiar with the movement of DNA fragments through pores during electrophoresis in agarose gels. In living cells, small nucleic acids such as small interfering RNA (siRNA) and antisense DNA molecules appear to pass through the bilayer by electrophoresis. Large DNAs, such as plasmids carrying an expression gene, bind to the cell and form cell-surface aggregates that appear to be endocytosed. Electroosmosis, the movement of liquids under the influence of an electric field, also plays a role.

Pores - Outer Membrane Factor (OMF) and Voltage-Dependent Anion Channel (VDAC)

Now, let's consider pores made of PFPs. We have already discussed two types of beta-barrel transmembrane proteins: the outer membrane factor (OMF) of Gram-negative bacteria and the voltage-dependent anion channel (VDAC). Both are examples of proteins called porins with typical beta-barrel topology.

VDAC: At low membrane potentials, VDAC (also known as mitochondrial porin), the most abundant protein in the mitochondrial outer membrane, moves metabolites and Ca²⁺ ions across the outer membrane of mitochondria. VDAC exist in an open state at zero or very low transmembrane potentials that allow for the transfer of key metabolic anions (pyruvate, oxaloacetate, malate, succinate, ATP, ADP, and P_i, which are involved in metabolism) and Ca²⁺, and in a closed state (above or below + 30 mV), which is not completely closed as it allows for the transfer of ions with a preference for cations. The closed state prevents ATP transfer. The transition to the closed state is promoted by tubulin and actin (cytoskeletal proteins), negatively charged lipids such as phosphatidyl ethanolamine and cardiolipin, and covalent phosphorylation by protein kinases. Bcl2, proteins that regulate programmed cell death (apoptosis), also interact with VDAC. In contrast to ligand-gated channels, which require ligand binding to open the channel, pore complexes are unusually open. Millions of ATPs/second move across the membrane through the open pore but none through the closed pore.

Figure \(\PageIndex{5}\) shows an interactive iCn3D model of mouse VDAC1 (4c69) with bound ATP loosely held in the site. An alpha helix partly occludes the central pore of this β-barrel protein.

Figure \(\PageIndex{5}\): Mouse voltage-dependent anion channel - VDAC1 (4c69) with bound ATP loosely held in the site. (Copyright; author via source). Click the image for a popup or use this external link: https://structure.ncbi.nlm.nih.gov/icn3d/share.html?Ueb4wg26f5oEVL4Y9

One ATP is bound in the barrel and interacts with Lys 12 and 20 at each end of the cavity-bound helix. The alpha-helix narrows the pore opening and presumably changes orientation in a voltage-sensitive fashion to gate the pore open and closed, hence regulating the conductance of ions through the pore. That the orientation of the charged arginine side chains would depend on the transmembrane potential should be somewhat obvious.

Aquaporins

Aquaporins can move a billion water molecules per second across membranes and exclude ions, including protons. Waters proceed in a single file through the pore. Instead of moving water through, it might simply move "H+ ions" by aligning successive H-bond donors and acceptors in a "wire" of water molecules. This is prevented by two conserved asparagine residues in the center of the channel, which disrupt the water-to-water hydrogen bond network in the channel waters and could facilitate proton transfer. Instead, the central waters form hydrogen bonds to the central asparagines. This, along with local membrane potentials, leads to opposite water orientations on the two leaflet sides of the membrane, precluding H⁺ transfer.

Here is a movie of a molecular dynamics simulation of water moving through the porin GlpF, an aquaporin in E. coli.

Science magazine (Tajkhorshid et al., Science Apr 2002, 296:525). Used with permission from the Theoretical and Computational Biophysics Group, the National Institutes of Health (NIH) Resource for Macromolecular Modeling and Bioinformatics, at the Beckman Institute, University of Illinois at Urbana-Champaign.

OMF: The outer membrane factor (see previous section) is one member of a class of bacterial porins, the most abundant proteins in the outer membrane of Gram-negative bacteria. They are classified as non-specific or specific (with respect to the solute that passes through) or monomeric, dimeric, or trimeric based on their structure. In Gram-negative bacteria with two lipid bilayers, solute movement from inside to outside includes at least three sets of proteins. Active transport (discussed in the next section) needs an energy source. It is used by inner membrane transport proteins, including ATP-binding cassette (ABC)-type, resistance-nodulation-division (RND)-type, and major facilitator superfamily (MFS)-type transporters. These are connected to membrane fusion proteins (MFP) that span the periplasm, which then interact with at least 21 types of porins. Molecules with molecular weights greater than 600 generally cannot cross the nuclear envelope of Gram-negative bacteria, limiting the size of potential antibiotics that must enter by passive diffusion.

Mechanosensitive ion channels - Mscs (which are pores!)

As the name applies, these ion channels (with openings large enough to be called pores) are gated open/closed by mechanical (physical) changes in the properties of the membrane, not extracellular/intracellular ligands or voltage changes. Certain bilayer lipids also activate the Mscs. There are two types of mechanosensitive ion channels: small (MscS) and large (MscL). Changes in local (boundary-layer) and nonlocal lipids are involved in channel gating (see the next section for lipid-gated ion channels). They are found in prokaryotes, archaea, and eukaryotes. They are also called stretch-gated ion channels.

Mcss transduces a physical force (stretching and change in turgor pressure) into an electric signal - a flow of ions across the membrane. Turgor pressure is the internal pressure that "presses" the cytoplasm and cell membrane towards the cell wall in bacteria and plant cells. It arises mostly from the osmotic flow of water into the cell. If bacterial cells are placed in a high-salt concentration solution (hypertonic), water flows out of the cell, and the cell membrane shrinks toward the inside, leaving the volume confined by the cell wall. When placed in a hypotonic solution, water flows into the cell, and the cell membrane swells to the cell wall. The response of these channels is fast, in the millisecond range, which is about as quick as a cellular response can be. Some variants of these channels are called piezochannels, based on the piezoelectric effect that describes how a voltage is produced when some materials are deformed by mechanical stress, which causes a redistribution of charges.

Bacteria normally have high concentrations of both K⁺ and negatively charged anions, especially glutamate, leading to high turgor pressure from the inward osmotic flow of water. At low external osmolarity, turgor pressure in the cell can reach 4 atm. When placed in high-osmolarity external solutions, bacterial cells respond by increasing the influx of solutes into the cell.

Mscs are particularly important when bacteria are subjected to sudden osmotic shock. If they are placed in pure water, for example, water would flow down a concentration gradient into the cell, causing it to swell and lyse, killing it. This is done in the lab to prepare almost pure hemoglobin from ruptured red blood cells. The Mscs open under these conditions, and small molecules from the cytoplasm flow out, helping keep the cell viable. Their openings must be regulated to prevent too much outward flow, which would kill the cell.

In other organisms, they are also involved in touch (stretch), hearing (vibration, sound waves), and responses to gravity. Stimuli that activate them include fluid shear stress (relevant to endothelial cells that line blood vessels), membrane stretch (relevant to skeletal and cardiac muscle cells), or even the indentation of a bilayer by a pipette. Changes in transmembrane turgor or other mechanical pressures cause membrane tension. Yet even in the absence of these changes, Mscs can be activated by anesthetics, phospholipids lacking one fatty acid (lysophospholipids), and certain polyunsaturated lipids. These stimuli also perturb the membrane bilayers.

Given the pore size, these proteins are less selective to ion flow than voltage-gated channels (see next section). Depending on the amino acids that line the pore, some Mscs would allow the preferential flow of anions, while some allow cation flow.

Some somatosensory channels (i.e. not pore) proteins also respond to pressure. When open, these can be selective for specific ions such as Na⁺ or K⁺. Examples include some variants of the Transient Receptor Potential (TRP) ion channel. Other membrane proteins can also be activated by physical force. But true Mscs have some key characteristics. If mutated or deleted, the mechanosensory response is removed. If added to a cell, a mechanosensory response results.

a. Small-conductance mechanosensitive channel

This protein is a homoheptamer, with three helices from each monomer contributing to the overall structure. Two helices (1 and 2) interact more with the lipid components of the bilayer. Interactions of specific lipids with the helices seem to promote closure, but changes under high pressure lead to pore opening. Half of each helix three forms the pore, while the other half is more parallel to the membrane and interacts with a large cytoplasmic domain.

Figure \(\PageIndex{6}\) shows the differences between the closed form of E. Coli MscS ( 2oau) and the open form (2vv5) viewing down the pore axis. The heptameric protein is shown in gray. Two key valines (105 and 109) on each chain are shown in spacefill and colored cyan. These hydrophobic amino acids act like gatekeepers, helping keep water out and forming a "vapor seal." In the closed state, the pore is sealed by closing the leucine "rings" as one half of helix 3 packs more closely. Many members of the MscS family vary significantly in size and can have between 3-11 transmembrane regions. The closed pore has a diameter of about 4.8 Å, while the open pore is 13 Å across.

Figure \(\PageIndex{6}\): Differences between the closed form of E. Coli MscS ( 2oau) and the open form (2vv5) viewing down the pore axis.

Most of you would have studied Ohm's law, given by

\begin{equation}
\mathrm{V}=\mathrm{IR} \text { or } \mathrm{I}=\frac{\mathrm{V}}{\mathrm{R}}
\end{equation}

where I (amps) is the current, R (ohms) is the resistance and V (volts) is the voltage. A more general variant of this law used in physics is

\begin{equation}
\mathrm{J}=\sigma \mathrm{E}
\end{equation}

where J is the current density, E is the electric field, and σ is the conductivity (inverse of resistance), which depends on the material. The unit of sigma σ is ohms^-1 or mhos (ohm written backward). That unit has been renamed the Siemens (S). Mscs channels have a small conductance of approximately 1 nS in 400 mM salt solution.

b. Large conductance mechanosensitive ion channel (MscL):

The channels have large conductances (3 nS) and concomitantly larger pore sizes, allowing the flow of water, ions, and even small proteins. Again, they are involved in diffusion down an electrochemical gradient through pores, so they are not involved in active transporter but rather gated diffusion.

Figure \(\PageIndex{7}\) shows an interactive iCn3D model of the pentameric MscL from Mycobacterium tuberculosis (2oar), a Gram-positive bacterium that causes tuberculosis. About 23% of the world's population is affected by this pathogen. It causes about 1.5 million deaths each year. Compare this to 7 million confirmed deaths during four years of the COVID pandemic (with central estimates of 27 million excess deaths). It has killed over 1 billion people throughout human history (but not as many as malaria). The pore diameter is about 30 Å across.

Figure \(\PageIndex{7}\): Closed state of the pentameric MscL from Mycobacterium tuberculosis (2oar). (Copyright; author via source). Click the image for a popup or use this external link: https://structure.ncbi.nlm.nih.gov/i...LsRZZbmFy5NRF6

Pore-forming alpha-helical toxins

We started this chapter by discussing the innate immune system's major attack complex (MAC). Pathogens also employ pore formation to kill host cells. Many secrete soluble proteins that aggregate in the membrane to form alpha-helical or beta-barrel pores. The proteins are called pore-forming toxins (PFTs). Killing occurs when cytoplasmic components leak out or when bacterial toxins, such as diphtheria and anthrax toxins, enter the cell.

Figure \(\PageIndex{8}\) shows an interactive iCn3D model of the pore formed by cytolysin A (ClyA, also known as HlyE), an alpha-PFT used by some E. Coli and Salmonella enterica strains. The pore is a large dodecamer that forms from soluble monomeric ClyA (2wcd). It has a pore diameter of about 40 Å,

Figure \(\PageIndex{8}\): Pore formed by cytolysin A (ClyA, also known as HlyE). (2wcd) (Copyright; author via source). Click the image for a popup or use this external link: https://structure.ncbi.nlm.nih.gov/i...RXUsE24d9d8NT6

Figure \(\PageIndex{9}\) shows the soluble monomeric form of cytolysin A (ClyA, HlyE) (1QOY). Nonpolar side chains are shown in cyan. The transparent surface is mostly polar, making the monomer soluble.

Figure \(\PageIndex{9}\): Soluble monomeric form of cytolysin A (ClyA, HlyE) (1QOY)

Figure \(\PageIndex{10}\) shows the changes in conformation between the single-chain soluble form (shown in the figure above) and a single chain of the membrane oligomeric channel.

Figure \(\PageIndex{10}\): Changes in conformation between the single-chain soluble form and the same chain in the channel

We started this chapter section by exploring electroporation and the formation of a toroidal-like hole in the lipid bilayer. In the case of ClyA, the protein aggregate itself forms the pore, not the lipids themselves, although lipid rearrangements are necessary to form the protein complex.

Gap Junctions

Connexins are voltage-gated channels that allow ions, metabolites, nucleotides, and small peptides to flow between cells. A connexin has four transmembrane helices and two extracellular loops. Six protomers combine in a single cell to form a channel complex called a connexon or hemichannel. Beta structures in the connexin hemichannel of one cell dock with a similar channel on an adjoining cell to form a full channel passing through the membranes of both cells, forming a gap junction between the cells. This is shown in Figure \(\PageIndex{11}\). There are 20 different connexins encoded in the human genome.

The left figure below shows the six protomers, each in a different color, and a gray rectangle representing the bilayer of a single connexon or hemichannel. The right side of the figure shows a full gap junction channel between two cells, with the membranes represented by gray rectangles. The connexin 26 monomer was used to create the diagram. Mutations in this protein are associated with hearing loss.

Figure \(\PageIndex{11}\): Connexon channel and the gap junction

The left figure above shows the six protomers, each in a different color, and a gray rectangle representing the bilayer of a single connexon or hemichannel. The right side of the figure shows a full gap junction channel between two cells, with the membranes represented by gray rectangles. The connexin 26 monomer was used to create the diagram. Mutations in this protein are associated with hearing loss.

The channel's cytoplasmic entrance is positively charged. It forms a funnel of six amino-terminal helices that leads to a negatively charged transmembrane lining. The entrance diameter is 14 Å.

Figure \(\PageIndex{12}\) shows an interactive iCn3D model of a full gap junction channel connecting two membranes using the human connexin 26 monomer (2zw3). Only the bottom membrane bilayer is represented by red and blue dummy atoms.

Figure \(\PageIndex{12}\): Gap junction channel connecting two membranes using the human connexin 26 monomer (2zw3). (Copyright; author via source). Click the image for a popup or use this external link: https://structure.ncbi.nlm.nih.gov/icn3d/share.html?TpuQ6Agk8XWuApsWA

The Nuclear Pore Complex (NPC)

Channels have pores that can be gated open and allow the selective flow of ions. Pore-forming proteins have larger entrances that allow small and large molecules to pass through the bilayer. The pore opening in even large mechanically sensitive channels (about pale in size compared to the nuclear pore complex, which has a combined molecular mass of around 125,000,000! Its outer diameter is ≈1,200 Å, and its inner one is ≈ 425-Å. Figure \(\PageIndex{13}\) shows the relative size of the nuclear pore compared to other molecular structures, including the eukaryotic ribosome, nucleosome, a soluble tetrameric protein (rubisco, 270K), and MscL (shown as a circle, which represents the pore diameter).

Figure \(\PageIndex{13}\): Relative size of nuclear pore compared to other molecular structures, including the eukaryotic ribosome, nucleosome, a soluble tetrameric protein (rubisco, 70K) , and MscL (shown as a circle which represents the pore diameter)

Its job is to shuttle small molecules by passive diffusion down a concentration gradient through the pore. In addition, it transports large molecules and macromolecular structures (proteins, RNA, and perhaps ribosomes) across the nuclear membrane, a process that requires energy. The proteins that comprise this complex are called nucleoporins (nups), which appear to number around 34 in humans. Each NPC complex contains around 1000 nucleoporins. The complex fuses the inner and outer nuclear membranes.

We have focused so much on single bilayer membranes that comprise the plasma membrane and the membranes of organelles like the Golgi complex and lysosomes, it might come as a surprise (perhaps not to biology students) that the nuclear membrane, or envelope, appears to consist of two bilayers. Most know that mitochondria have two membranes, an inner and an outer membrane, similar to those of Gram-negative bacteria. Mitochondria are believed to have arisen from bacteria, so the double bilayer there makes sense. Figure \(\PageIndex{14}\) shows the nuclear membrane or envelope of two bilayers (1) with an outer ring (2), spokes (3), a basket (4), and filaments (5). The NPC spans both bilayers.

Figure \(\PageIndex{14\): Nuclear membrane or envelope of two bilayers (1) with an outer ring (2), spokes (3), a basket (4) , and filaments (5) https://en.Wikipedia.org/wiki/File:N...e_crop.svg.png; Creative Commons Attribution-Share Alike 2.5 Generic

The outer bilayer of the nuclear envelope is continuous with the endoplasmic reticulum, as shown in Figure \(\PageIndex{15\) below. The dots on the ER membrane are ribosomes, making this the rough ER (as opposed to smooth ER, which has no attached ribosomes).

Figure \(\PageIndex{15\): The outer bilayer of the nuclear envelope is continuous with the endoplasmic reticulum https://commons.wikimedia.org/wiki/F...18_Nucleus.jpg; OpenStax, CC BY 4.0 <https://creativecommons.org/licenses/by/4.0>, via Wikimedia Commons

Figure \(\PageIndex{16\) below shows a model of the basket-like structure of the nuclear pore complex (NPC). It shows that instead of two separate bilayers, there is just one bilayer, with each leaflet bending around at the NPC and reversing directions! Think of the interesting lipids and protein components that enable the bend! Alternatively, you could say that two different membranes fuse at the NPC.

Figure \(\PageIndex{16\): A model of the basket-like structure of the nuclear pore complex Citation: Azimi M, Mofrad MRK (2013) Higher Nucleoporin-Importinβ Affinity at the Nuclear Basket Increases Nucleocytoplasmic Import. PLoS ONE 8(11): e81741. https://doi.org/10.1371/journal.pone.0081741. Creative Commons Attribution License,

The NPC consists of 32 copies of each specific nucleoporin (Nup) except two. One has 48 copies and the other 16 (even these sum to 2x32 Nups). Three rings form and surround the pore. A 16-membered ring of Nups faces the cytoplasm (cytoplasmic ring), and another 16-membered ring of Nups faces the nucleoplasm (nuclear ring). There is 8-fold rotational symmetry in each ring, suggesting a dimeric repeat of Nups in the rings. Eight Nups in the cytoplasmic ring have a disordered end that sticks out into the cytoplasm as filaments. In contrast, the disordered ends of eight Nups in the nuclear ring form filaments that bind together to form a ring at the bottom of the nuclear basket.

The inner ring (the FG Nups layer in the figure above), between the cytoplasmic and nucleoplasmic rings, also consists of Nups with ordered domains and disordered parts. The disordered parts of the inner ring Nups have repetitive sequences enriched in phenylalanine and glycine (hence the name FG Nups) and stick out into the central pore. These disordered regions act as a filter, allowing certain molecules to pass while excluding those with molecular weights greater than 40 kDa. Large molecules need transport proteins called karyopherin transport factors to move through the pore.

The core structure of the NPC, obtained through cryoelectron microscopy, is shown in Figure \(\PageIndex{17\) below.

Figure \(\PageIndex{17\): The core structure of the Nuclear Pore Complex https://www.ebi.ac.uk/pdbe/entry/emd...-3103/analysis

The double bilayers are very evident. The nuclear and cytoplasmic filaments, and the disordered FG-repetitive sequences that protrude into the inner ring pore, are not observed because they are very flexible and don't adopt a single conformation using standard structure-determination methods.

Figure \(\PageIndex{18}\) shows an interactive iCn3D model of the nuclear pore complex (NPC) core (5a9q), whose structure was obtained using cryoelectron microscopy. (load slowly)

Figure \(\PageIndex{18}\): Nuclear pore complex (NPC) core (5a9q). (Copyright; author via source). Click the image for a popup or use this external link: https://structure.ncbi.nlm.nih.gov/icn3d/share.html?NP7oQb235BCgPARA8

A cartoon figure showing the variety of Nups in the NPC is shown in Figure \(\PageIndex{19}\).

Figure \(\PageIndex{19}\): A cartoon version of the various NUPS in the nuclear pore complex. Donnaloja F., Jacchetti E., Soncini M., Raimondi M. T.Front. Physiol., 12 July 2019 | https://doi.org/10.3389/fphys.2019.00896Creative Commons Attribution License (CC BY).

In this figure, the cytoplasmic and nucleoplasmic rings are shown in green, each formed mainly by 16 copies of the Y-complex arranged in two eight-membered rings. The inner ring, predominantly formed by 32 copies of the Nup93 complex, is shown in red. Transmembrane nucleoporins are depicted in violet, and the cytoplasmic filaments and nuclear basket structure are in orange. Attached to the inner ring are Nup62 complexes (depicted in blue), which form a cohesive meshwork within the central channel through their FG-repeat domains. The position of Nup98 is not indicated; it is an FG-repeat-containing nucleoporin important for the transport and exclusion function of NPCs; its position in the NPC is less defined, but it might be part of the inner ring. Similarly, Aladin (also known as AAAS), Gle1, Rae1, and Npl1 (also known as hCG1 and NUPL2) have been omitted.

Large proteins and RNA that pass through the pore must first be bound to a cargo receptor, which can move the "cargo" across the pore with concomitant GTP hydrolysis. This process is closer to active transport, which we will discuss in Chapter 11.3.

The entire nuclear pore complex was solved in 2022 using cryoEM. Here are two videos showing the dilated complex (7R5J). Click on the images to download mp4 animations of the complex.

Summary

(Summary written by Claude, Sonnet 4.6, Anthropic)

This chapter examines the largest class of membrane transport structures: pores. Unlike the ion channels described in the preceding chapter — which are gated, highly selective, and allow passage of specific ions — pores have larger openings that permit the simultaneous passage of multiple chemical species, with selectivity governed primarily by pore diameter, charge lining, and, in some cases, associated filter proteins.

The chapter begins by establishing the biophysical distinction between channels and pores, then addresses pore formation in protein-free lipid bilayers through electroporation. When a sufficiently strong electric field is applied across a bilayer, membrane tension increases, depressions form in the surface, and transient water columns (hydrophobic pores) appear in the hydrocarbon core. These spontaneously rearrange into more stable, hydrophilic toroidal pores, in which lipid head groups line the pore walls. Molecular dynamics simulations reveal this progression on a nanosecond timescale. Electroporation is the basis of bacterial transformation and eukaryotic transfection; small nucleic acids such as siRNA pass through pores by electrophoresis, while large plasmids appear to be endocytosed after binding to the cell surface.

Pore-forming proteins (PFPs) use a fundamentally different mechanism: they are synthesized as water-soluble monomers that undergo dramatic conformational changes upon membrane binding, oligomerizing into stable protein-lined pores. Two classes of secondary structure are used. Alpha-helical PFTs, exemplified by cytolysin A (ClyA / HlyE) from E. coli and Salmonella, form dodecameric pores of approximately 40 Å diameter; the structural transition from soluble monomer to membrane-embedded channel subunit involves a large rearrangement that exposes nonpolar surfaces for bilayer insertion. Beta-barrel PFPs, including the bacterial outer membrane factor (OMF) and the mitochondrial voltage-dependent anion channel (VDAC), use antiparallel beta-strands to form the pore walls. VDAC is the most abundant protein in the outer mitochondrial membrane and, at low transmembrane potentials, operates as an open pore conducting metabolic anions — pyruvate, malate, succinate, ATP, ADP, and phosphate — and Ca²⁺ at rates of millions of molecules per second. Above approximately ±30 mV, a voltage-dependent conformational change driven by the repositioning of charged residues — notably the partially occluding N-terminal alpha-helix — switches VDAC to a closed state that excludes ATP while retaining some cation conductance. Regulation by tubulin, cardiolipin, phosphorylation, and Bcl-2 family proteins links VDAC gating to cytoskeletal state, membrane lipid composition, and apoptotic signaling.

Aquaporins represent a specialized class of beta-barrel pores optimized for one of the most demanding selectivity problems in biology: passing water molecules at rates approaching one billion per second while completely excluding protons. Water molecules traverse the pore in a single file. The potential for proton transfer via a Grotthuss mechanism — in which protons hop along a chain of hydrogen-bonded water molecules — is prevented by two conserved asparagine residues in the channel center, which disrupt the water-to-water hydrogen-bond network and reorient adjacent water molecules in opposing directions, breaking the continuity required for proton transfer.

Mechanosensitive channels (Mscs) are gated not by ligands or voltage but by changes in membrane tension arising from osmotic stress. When bacteria are placed in a hypotonic solution, water osmotically enters, increasing turgor pressure and deforming the membrane. This mechanical signal gates the opening of Mscs within milliseconds, allowing cytoplasmic solutes to efflux and relieve pressure before cell lysis. The small conductance channel MscS is a homoheptamer; hydrophobic valine residues (Val 105 and 109) form a vapor seal in the closed state, and mechanical opening involves rotation of transmembrane helix 3. The large-conductance channel MscL, a homopentamer with a 30 Å open-pore diameter and a conductance of 3 nS, can pass water, ions, and even small proteins. Both channels also respond to lysophospholipids and polyunsaturated lipids that perturb bilayer packing, further linking their activity to the lipid bilayer's physical state.

Gap junctions allow direct cytoplasmic communication between adjacent animal cells. Six connexin subunits — each with four transmembrane helices — assemble in a single cell's plasma membrane into a hemichannel (connexon). Beta-structured extracellular loops of a connexon dock with those of a connexon in the neighboring cell, forming a continuous intercellular channel approximately 14 Å in diameter at the cytoplasmic entrance. This conduit permits passage of ions, metabolites, nucleotides, and small peptides (below roughly 1 kDa) between cells, enabling electrical and metabolic coupling. Mutations in connexin 26 are a leading genetic cause of hearing loss.

The chapter concludes with the nuclear pore complex (NPC), the largest known protein transport machine, with a combined mass of ~125 MDa and an outer diameter of ~1200 Å. The NPC spans both bilayers of the nuclear envelope — which is itself a continuous double membrane with the outer bilayer connected to the rough endoplasmic reticulum — at a point where the two leaflets of each bilayer curve back on each other to create a fused toroidal structure. Approximately 34 distinct nucleoporin (Nup) proteins, present in ~32 copies each (~1000 total per NPC), are organized into cytoplasmic, inner, and nuclear rings with eight-fold rotational symmetry. Nups of the inner ring contain disordered FG-repeat domains (rich in Phe-Gly) that project into the central channel and form a selective meshwork: molecules below ~40 kDa diffuse passively through the FG-Nup phase, while larger cargoes, including proteins and RNA require karyopherin transport factors and GTP hydrolysis to transit the pore — a process functionally equivalent to active transport. Throughout, the chapter reinforces the broader course theme: the structural architecture of pore-forming assemblies — from the size of the opening to the chemical nature of the pore lining to the conformational transitions that gate them — determines what can pass, in what direction, and at what rate.