17.02.04 · mol-cell-bio / membranes

Vesicle trafficking: SNARE proteins, clathrin-coated vesicles, and the secretory pathway

stub3 tiersLean: nonepending prereqs

Anchor (Master): Rothman, J. E. — The Protein Life Cycle, 1st ed. (2021)

Intuition Beginner

Cells manufacture thousands of different proteins. Many of these proteins must travel from where they are made (the endoplasmic reticulum, or ER) to distant locations — the cell surface, the outside of the cell, or other organelles. The cell solves this delivery problem by packaging proteins into tiny membrane-bound sacs called vesicles.

Vesicles bud off from one membrane, carry their cargo through the cytoplasm, and fuse with a target membrane. This process — budding, transport, and fusion — is called vesicle trafficking. The main route is the secretory pathway: proteins move from the ER to the Golgi apparatus, through successive Golgi compartments, and finally to the plasma membrane or to lysosomes.

Delivering cargo to the correct destination requires specificity. A vesicle leaving the ER must fuse with the Golgi, not with the mitochondrion or the plasma membrane. This specificity is achieved by SNARE proteins, which act like molecular velcro. Every vesicle carries a v-SNARE on its surface, and every target membrane displays a matching t-SNARE. When the correct pair meets, they wind around each other and pull the two membranes together, driving fusion. Only matching SNARE pairs zip up, so vesicles dock only at their intended targets.

Visual Beginner

The diagram shows the secretory pathway from a bird's-eye view:

A protein (blue dot) is synthesised into the ER lumen. A COPII-coated vesicle buds from the ER, capturing the protein as cargo, and travels to the Golgi.
The vesicle fuses with the cis-Golgi (the Golgi face nearest the ER). The cargo protein passes through the medial- and trans-Golgi stacks, where it is modified (glycosylation, sorting).
At the trans-Golgi network (TGN), the protein is sorted into one of several destinations. A clathrin-coated vesicle buds from the TGN carrying cargo destined for the plasma membrane.
The clathrin coat is shed, and the vesicle fuses with the plasma membrane via SNARE pairing, releasing the protein to the cell exterior (secretion) or inserting it into the membrane.

Worked example Beginner

Insulin is a protein hormone that must be secreted from pancreatic beta cells into the bloodstream. Let us trace insulin through the secretory pathway.

Step 1. Synthesis and ER entry. The insulin gene is translated by ribosomes attached to the ER membrane. The nascent polypeptide chain (preproinsulin) is threaded through the Sec61 translocon into the ER lumen, where the signal sequence is cleaved to yield proinsulin. Disulfide bonds form, and proinsulin folds into its native conformation.

Step 2. ER-to-Golgi transport. Proinsulin is packaged into a COPII-coated vesicle that buds from a specialised region of the ER called an ER exit site (ERES). The vesicle sheds its COPII coat and fuses with the cis-Golgi.

Step 3. Golgi processing. As proinsulin passes through the Golgi stacks, it encounters convertase enzymes that cleave the connecting peptide (C-peptide), converting proinsulin into active insulin. This processing occurs primarily in the trans-Golgi network.

Step 4. Storage and secretion. Insulin is concentrated into dense-core secretory granules that bud from the TGN. These granules mature and wait near the plasma membrane. When blood glucose rises, calcium influx triggers SNARE-mediated fusion of the granule with the plasma membrane, releasing insulin into the extracellular space.

Check your understanding Beginner

Exercise (medium, short answer).

Explain the difference between anterograde and retrograde transport in the secretory pathway, and name the coat proteins responsible for each direction between the ER and Golgi.

Hint

Anterograde means forward (ER to Golgi to plasma membrane). Retrograde means backward. Each direction uses a different coat complex.

Answer

Anterograde transport moves cargo forward along the secretory pathway: from ER to Golgi, from Golgi to plasma membrane, or from Golgi to endosomes/lysosomes. COPII-coated vesicles carry anterograde cargo from the ER to the Golgi. Retrograde transport returns proteins backward: from the Golgi to the ER, or from later compartments to earlier ones. COPI-coated vesicles carry retrograde cargo from the Golgi back to the ER, recycling ER-resident proteins (such as BiP and protein disulfide isomerase) that escape by bulk flow. The balance of anterograde and retrograde traffic maintains the distinct protein compositions of each compartment.

Exercise (medium, multiple choice).

A protein carries a KKXX motif at its C-terminus. Where is this protein normally localised, and what is the function of the motif?

A. Plasma membrane; targets the protein for endocytosis B. ER lumen; signals retention by binding to the translocon C. ER membrane; mediates retrieval from the Golgi via COPI vesicles D. Lysosome; directs trafficking via mannose-6-phosphate tagging

Hint

The KKXX motif (two lysines followed by two any residues) is a cytoplasmic retrieval signal. Consider which coat complex recognises it.

Answer

Option C. The KKXX motif is a di-lysine retrieval signal found on the cytoplasmic tail of type I ER membrane proteins. When these proteins escape to the Golgi, the KKXX motif is recognised by the COPI coat (via the coatomer subcomplex), packaging them into retrograde vesicles that return them to the ER. This retrieval mechanism maintains the high concentration of ER-resident membrane proteins despite continuous vesicle traffic. Soluble ER-resident proteins use a different signal: the KDEL (Lys-Asp-Glu-Leu) sequence at their C-terminus, which is recognised by the KDEL receptor in the Golgi.

Formal definition Intermediate+

Vesicle trafficking is the process by which membrane-bound transport vesicles carry proteins and lipids between intracellular compartments and between the cell interior and the extracellular space. The process can be decomposed into five mechanistic steps: cargo selection, vesicle budding, vesicle transport, vesicle tethering and docking, and membrane fusion.

The secretory pathway describes the default route for proteins synthesised on ER-bound ribosomes:

Co-translational translocation. The signal recognition particle (SRP) binds the N-terminal signal sequence of the nascent polypeptide, targets the ribosome-nascent chain complex to the Sec61 translocon in the ER membrane, and the growing chain is threaded into the ER lumen or integrated into the ER membrane.
ER quality control. Proteins must fold correctly and, for oligomeric proteins, assemble before exiting the ER. Misfolded proteins are retained by chaperones (BiP, calnexin, calreticulin) and may be targeted for ER-associated degradation (ERAD) via retro-translocation to the cytoplasm and ubiquitin-proteasome degradation.
ER exit via COPII. Properly folded cargo is concentrated at ER exit sites (ERES) by the COPII coat. COPII assembly begins when Sec12 (a guanine nucleotide exchange factor, GEF) activates the small GTPase Sar1 by exchanging GDP for GTP. Sar1-GTP inserts its amphipathic N-terminal helix into the cytoplasmic leaflet of the ER membrane, recruiting the Sec23/Sec24 heterodimer (the inner coat).

Sec24 serves as the primary cargo adaptor, binding directly to di-acidic export signals (e.g., DxE) on cytoplasmic domains of transmembrane cargo or to cargo receptors (such as the lectin ERGIC-53 for glycoproteins). The Sec13/Sec31 heterotetramer (the outer coat) polymerises into a cage lattice that deforms the membrane into a bud.
Golgi processing. COPII vesicles fuse to form the ER-Golgi intermediate compartment (ERGIC) and then deliver cargo to the cis-Golgi. Cargo transits through the cis-, medial-, and trans-Golgi cisternae, undergoing sequential glycosylation modifications. The trans-Golgi network (TGN) serves as the major sorting station.
Sorting at the TGN. Three major sorting pathways diverge from the TGN: (a) constitutive secretion to the plasma membrane, (b) regulated secretion via secretory granules (hormones, neurotransmitters), and (c) lysosomal targeting via mannose-6-phosphate (M6P) tagging and M6P receptor-mediated sorting into clathrin-coated vesicles.

Vesicle coat proteins. Three major coat complexes operate in the secretory and endocytic pathways:

COPII (ER export). Inner coat: Sec23/Sec24. Outer coat: Sec13/Sec31. GTPase: Sar1. Cargo signals: di-acidic (DxE), di-hydrophobic (FF, YY), and cargo receptors. Vesicle diameter: $\sim 60$ – $80$ nm. The COPII cage is a cuboctahedral lattice with approximate octahedral symmetry, assembled from Sec13/Sec31 heterotetramers.
COPI (retrograde Golgi-to-ER transport and intra-Golgi retrograde). The coatomer is a seven-subunit complex ( $α$ , $β$ , $β^{'}$ , $γ$ , $δ$ , $ϵ$ , $ζ$ -COP) that forms both the inner adaptor and outer cage. GTPase: Arf1 (ADP-ribosylation factor 1). Cargo signals: KKXX (di-lysine) on ER membrane proteins, KDEL receptor-ligand complexes. Vesicle diameter: $\sim 60$ – $70$ nm.
Clathrin (TGN-to-endosome and plasma membrane endocytosis). Clathrin triskelions (three heavy chains + three light chains) polymerise into polyhedral cages. Adaptor proteins (AP-1 at the TGN, AP-2 at the plasma membrane, GGAs for lysosomal enzymes) link clathrin to cargo. GTPase: Arf1 (at TGN) or none (AP-2 is constitutively active at the plasma membrane). Cargo signals: tyrosine-based (YXX $Φ$ ) and dileucine-based ([DE]XXXL[LI]) motifs. Vesicle diameter: $\sim 80$ – $120$ nm.

SNARE proteins and membrane fusion. SNAREs (soluble NSF attachment protein receptors) are a family of membrane-anchored proteins that catalyse membrane fusion. They are classified by their location and by a conserved glutamine (Q) or arginine (R) residue in the SNARE motif:

v-SNAREs (on vesicles; also called R-SNAREs, carrying the arginine). Example: VAMP (vesicle-associated membrane protein) / synaptobrevin on synaptic vesicles.
t-SNAREs (on target membranes; also called Q-SNAREs, carrying the glutamine). Example: syntaxin and SNAP-25 on the presynaptic plasma membrane.

A functional SNARE complex consists of four SNARE motifs assembled into a parallel four-helix bundle: one R-SNARE and three Q-SNARE domains (classified Qa, Qb, Qc). The neuronal SNARE complex, for instance, contains synaptobrevin (R-SNARE, 1 helix), syntaxin-1 (Qa-SNARE, 1 helix), and SNAP-25 (contributing both Qb and Qc helices from a single polypeptide). Zippering of the four-helix bundle from the N-terminal toward the C-terminal (membrane-proximal) end pulls the two membranes together, overcoming the hydration repulsion barrier.

After fusion, the SNARE complexes (now all in the same membrane) are disassembled by the ATPase NSF (N-ethylmaleimide sensitive factor) with its co-factor alpha-SNAP, which binds the SNARE four-helix bundle and uses ATP hydrolysis to unwind it, recycling the SNAREs for another round of fusion.

Rab GTPases. Rab proteins are small GTPases that serve as identity markers for intracellular membranes. Over 60 Rab proteins exist in mammalian cells, each associated with a specific compartment or transport step. Rab-GTP (active) recruits effector proteins including tethering factors (long-range connectors that capture incoming vesicles), motor proteins (for vesicle transport along microtubules and actin), and SNARE regulators. The Rab cycle alternates between GTP-bound (membrane-associated, active) and GDP-bound (cytosolic, inactive) states, regulated by GEFs (activating) and GAPs (inactivating). Example: Rab1 marks ERGIC and early Golgi membranes; Rab5 marks early endosomes; Rab7 marks late endosomes; Rab11 marks recycling endosomes.

Key mechanism Intermediate+

Mechanism: SNARE-mediated membrane fusion and the zippering free energy.

Membrane fusion is energetically unfavourable. Two lipid bilayers in aqueous solution repel each other through hydration forces (the ordered water shells surrounding the polar head groups resist displacement), steric repulsion of the head-group region, and the cost of exposing hydrophobic acyl chains to water during the hemifusion intermediate. The total energy barrier for spontaneous fusion of two bilayers is estimated at $\sim 40$ – $50 k_{B} T$ , which is prohibitive on biological timescales without catalysis.

SNARE proteins lower this barrier by converting the free energy of protein folding into mechanical work on the membrane. The mechanism proceeds through four stages:

Tethering. Rab-GTP on the vesicle recruits a tethering factor (e.g., the HOPS complex for vacuole-vacuole fusion, the exocyst for exocytosis) that captures the vesicle at a distance of $\sim 20$ – $50$ nm from the target membrane. Tethering is reversible and increases the local vesicle concentration near the target without committing to fusion.
SNARE pairing (trans-SNARE complex formation). The v-SNARE on the vesicle begins to associate with the t-SNARE on the target membrane, forming a trans-SNARE complex. This association is nucleated at the N-terminal of the SNARE motifs and proceeds toward the C-terminal (membrane-proximal) end.
Zippering. The progressive assembly of the four-helix bundle from N-to-C terminal is called zippering. Each layer of the four-helix bundle (numbered from $- 7$ at the N-terminal to $+ 8$ at the membrane-proximal end) contributes favourable free energy. Measurements of the zippering free energy using single-molecule force spectroscopy (Gao et al. 2012 Science 337, 1340-1343) give a total energy of approximately $\sim 35 k_{B} T$ ( $\sim 20$ kcal/mol) for complete assembly of the neuronal SNARE complex. This energy is deposited primarily in the membrane-proximal half of the complex (the "layer +5 to +8" region), which exerts the force needed to bring the two membranes within $\sim 1$ nm of each other.
Fusion pore opening. When the bilayers are brought into close apposition by SNARE zippering, the hydration barrier is breached and a fusion pore opens. The pore initially has a conductance of $\sim 100$ – $500$ pS, corresponding to a diameter of $\sim 1$ nm, and can either expand (full fusion, releasing all cargo) or close (kiss-and-run, releasing only small molecules through the transient pore).

The energetic budget is as follows. The barrier to spontaneous fusion is $\sim 40$ – $50 k_{B} T$ . The SNARE complex provides $\sim 35 k_{B} T$ of folding energy. The remaining $\sim 10$ – $15 k_{B} T$ is contributed by the tethering complex (which holds the membranes in proximity) and by the bending energy of the highly curved fusion pore neck. The match between the SNARE folding energy and the fusion barrier means that SNARE proteins are just sufficient to drive fusion — they are catalytic machines that convert the chemical energy of protein folding into the mechanical work of membrane merger.

NSF-mediated disassembly consumes one ATP per SNARE helix (three ATP molecules per NSF hexamer per SNARE complex). The NSF reaction is not required for fusion itself — it is required to recycle SNAREs after fusion, freeing them for subsequent rounds. In neurons, NSF inactivation (e.g., by N-ethylmaleimide treatment) does not block the first round of synaptic vesicle fusion but prevents subsequent rounds, depleting the readily releasable pool of vesicles.

Exercises Intermediate+

Exercise 2 (easy, multiple choice).

What is the function of the KDEL retrieval signal?

A. Directing proteins to the plasma membrane B. Returning escaped ER-resident soluble proteins from the Golgi to the ER C. Targeting proteins to lysosomes D. Initiating COPII vesicle budding

Hint

KDEL (Lys-Asp-Glu-Leu) is a tetrapeptide signal found at the C-terminus of soluble ER proteins such as BiP and PDI. Where do these proteins need to return if they escape?

Answer

Option B. The KDEL sequence is a C-terminal retrieval signal on soluble ER-resident proteins. It is recognised by the KDEL receptor in the cis-Golgi and ERGIC. The receptor-ligand complex is packaged into COPI-coated retrograde vesicles and returned to the ER, where the lower pH causes ligand release. This retrieval mechanism maintains the high concentration of chaperones and folding enzymes in the ER lumen despite continuous bulk-flow vesicle traffic.

Exercise 3 (medium, short answer).

Describe the sequence of events in COPII vesicle formation at an ER exit site, naming the key proteins and their functions.

Hint

The process begins with Sar1 activation and proceeds through inner coat recruitment, cargo capture, outer coat assembly, and finally scission.

Answer

Sar1 activation. Sec12 (ER-resident GEF) exchanges GDP for GTP on Sar1. Sar1-GTP undergoes a conformational change that exposes its amphipathic N-terminal alpha helix, which inserts into the cytoplasmic leaflet of the ER membrane.
Inner coat recruitment. Sar1-GTP recruits the Sec23/Sec24 heterodimer. Sec24 is the cargo adaptor: it binds di-acidic (DxE), di-hydrophobic (FF), and other export signals on the cytoplasmic tails of transmembrane cargo proteins, or binds cargo receptors (such as ERGIC-53 for glycoproteins) that capture soluble cargo. Sec23 acts as a GAP for Sar1 (a timing mechanism for coat disassembly).
Outer coat polymerisation. The Sec13/Sec31 heterotetramer is recruited and polymerises into a polyhedral cage that deforms the membrane into a bud. Sec31 also stimulates the GAP activity of Sec23, coupling cage assembly to the GTP hydrolysis timer.
Scission. Continued cage polymerisation and Sar1-GTP hydrolysis drive membrane constriction at the bud neck, eventually severing the vesicle from the ER. The resulting COPII-coated vesicle is approximately 60–80 nm in diameter.

Exercise 4 (medium, short answer).

Explain why NSF and alpha-SNAP are required for SNARE-mediated fusion in living cells, even though purified SNARE proteins reconstituted into liposomes can fuse without them.

Hint

Consider what happens to SNARE proteins after fusion. Can they be reused without disassembly?

Answer

Purified SNAREs in liposomes can catalyse fusion because they assemble into trans-SNARE complexes (one SNARE in each membrane) that zipper and drive membrane merger. However, after fusion the SNAREs reside in the same membrane as cis-SNARE complexes (all four helices in one bilayer). These cis complexes are thermodynamically very stable (the four-helix bundle has a melting temperature above $90^{\circ}$ C) and cannot spontaneously disassemble at physiological temperature. NSF (an AAA+ ATPase) binds alpha-SNAP, which in turn binds the SNARE four-helix bundle; NSF uses ATP hydrolysis to mechanically unwind the complex, releasing individual SNAREs for reuse. In vivo, without NSF/alpha-SNAP, SNAREs become trapped as cis complexes after the first fusion event, depleting the pool available for subsequent rounds. This is why N-ethylmaleimide (which inhibits NSF) blocks synaptic transmission after the first few action potentials.

Exercise 5 (medium, multiple choice).

A patient has I-cell disease (inclusion cell disease, mucolipidosis II), caused by a deficiency in the phosphotransferase that adds the mannose-6-phosphate tag to lysosomal enzymes. What is the consequence?

A. Lysosomal enzymes are degraded in the ER B. Lysosomal enzymes are secreted extracellularly instead of being delivered to lysosomes C. Lysosomal enzymes are retained in the Golgi D. Lysosomal enzymes are correctly targeted but are catalytically inactive

Hint

Without the M6P tag, the M6P receptor in the TGN cannot capture lysosomal enzymes. Where do the enzymes go by default?

Answer

Option B. In the TGN, lysosomal enzymes are normally tagged with mannose-6-phosphate (M6P) on their N-linked oligosaccharides. The M6P receptor binds M6P-tagged enzymes and packages them into clathrin-coated vesicles for delivery to late endosomes and lysosomes. Without the phosphotransferase, lysosomal enzymes lack the M6P tag and are not captured by the receptor. They follow the default constitutive secretory pathway and are secreted from the cell. The result is lysosomes that lack hydrolases and accumulate undegraded substrates as inclusion bodies, giving the disease its name. M6P receptor is present and functional — the defect is in the tag, not the receptor.

Exercise 6 (hard, short answer).

The SNARE complex is described as a four-helix bundle with one R-SNARE helix and three Q-SNARE helices (Qa, Qb, Qc). Explain the structural basis for this classification and why the ionic layer (also called the "0 layer") is central to SNARE function.

Hint

The names Q-SNARE and R-SNARE come from a specific residue in the middle of the SNARE motif. What role does this residue play in the four-helix bundle?

Answer

At the centre of the four-helix bundle, at layer 0, three glutamine (Q) residues from the Qa, Qb, and Qc helices and one arginine (R) residue from the R-SNARE helix form an ionic interaction network. The three Q side-chain amide groups hydrogen-bond with the R guanidinium group, creating a highly conserved buried polar cluster in the otherwise hydrophobic core of the bundle. This Q-Q-Q-R "0 layer" is the defining feature of all SNARE complexes and is conserved from yeast to humans. Mutating any of these residues severely impairs fusion. The ionic layer serves as a nucleation point for zippering: assembly proceeds outward from layer 0 toward both termini, and the ionic layer must form correctly before the membrane-proximal layers can zipper and exert force on the membranes. The 1:3 stoichiometry (one R, three Q) ensures correct register of the four helices and prevents unproductive homotypic interactions.

Exercise 7 (hard, short answer).

Compare and contrast the roles of Rab GTPases and SNARE proteins in ensuring vesicle targeting specificity. Which provides the primary specificity determinant, and how do the two systems cooperate?

Hint

Rabs act early (tethering) and SNAREs act late (docking/fusion). Consider whether a vesicle can fuse without the correct Rab, or without the correct SNARE.

Answer

Rab GTPases and SNARE proteins operate at different stages of the targeting pathway and provide complementary layers of specificity. Rab GTPases act first: a Rab on the vesicle membrane (in its GTP-bound state) recruits tethering effectors on the target membrane that capture the vesicle at a distance of 20–50 nm. This provides the initial specificity filter — a vesicle displaying Rab1 will only be tethered by membranes displaying Rab1 effectors (ERGIC/early Golgi), not by membranes displaying Rab5 effectors (early endosomes). SNAREs act after tethering: the tethered vesicle is brought close enough for v-SNARE/t-SNARE pairing, and only cognate pairs zipper efficiently. Non-cognate SNARE pairs have much lower affinity and do not form productive trans complexes.

The primary specificity determinant is debated, but the current consensus is that both layers are required and neither is sufficient alone. Knocking out the correct Rab blocks tethering and fusion; providing a heterologous tether that forces a vesicle into proximity with a non-cognate SNARE does not produce efficient fusion. The two systems cooperate through kinetic proofreading: tethering (Rab-dependent) holds the vesicle in the vicinity of the target long enough for the correct v-SNARE to find its cognate t-SNARE by diffusion, while incorrect SNARE pairs dissociate before they can zipper. This two-layer system achieves high specificity (error rate $< 1 0^{- 3}$ ) despite the limited number of distinct SNARE and Rab isoforms.

SNARE zippering energetics, tethering factors, and the thermodynamics of membrane fusion Master

The fusion of two lipid bilayers is a nontrivial physical process that requires crossing an energy barrier of approximately $40$ – $50 k_{B} T$ . This barrier arises from three contributions: the hydration repulsion between the polar head-group surfaces (dominant at separations of $2$ – $3$ nm, decaying exponentially with a decay length of $\sim 0.2$ nm), the steric repulsion of the head-group glycerol and phosphate moieties, and the hydrophobic exposure cost of the hemifusion stalk intermediate, in which the proximal leaflets merge while the distal leaflets remain separate. SNARE proteins bridge this barrier by converting protein-folding free energy into mechanical work on the membrane, but the quantitative details of this conversion have only recently been measured with single-molecule precision.

Free energy of SNARE zippering. The neuronal SNARE complex (synaptobrevin-2 / VAMP2 + syntaxin-1A + SNAP-25A) assembles as a parallel four-helix bundle with 16 layers of interacting residues (layers $- 7$ to $+ 8$ , numbered from the ionic layer at position 0). Single-molecule optical-trap measurements by Gao, Zorman, and colleagues (2012 Science 337, 1340-1343) measured the force-extension profile of SNARE assembly and disassembly. The N-terminal half (layers $- 7$ to $- 1$ ) assembles with a free energy of approximately $\sim 8 k_{B} T$ , while the C-terminal half (layers $+ 1$ to $+ 8$ ) contributes $\sim 27 k_{B} T$ . The total zippering energy is $\sim 35 k_{B} T$ , in good agreement with the bulk thermodynamic measurement of $\sim 20$ kcal/mol ( $\sim 33 k_{B} T$ ) from denaturation studies.

The energy is deposited unevenly along the complex. The membrane-proximal C-terminal layers provide the largest contribution per residue because they are in direct mechanical contact with the membrane: zippering of these layers physically pulls the two bilayers together. The N-terminal half assembles relatively easily and serves primarily as a docking/nucleation step that positions the SNARE motifs for the energetically costly C-terminal zippering.

Regulatory proteins exploit this two-stage energetics. Complexin binds the assembled SNARE complex at the boundary between the N-terminal and C-terminal halves, clamping further zippering and preventing premature fusion. Synaptotagmin, the calcium sensor for neurotransmitter release, displaces complexin and triggers C-terminal zippering upon calcium binding, providing the final energy input that completes fusion. The calcium-dependent step contributes an estimated $\sim 5$ – $10 k_{B} T$ from synaptotagmin's insertion of its C2 domain hydrophobic loops into the membrane, which destabilises the bilayer and lowers the hemifusion barrier.

Tethering complexes as specificity gates. The tethering step, mediated by Rab-GTP effectors, operates before SNARE engagement and provides the first layer of targeting specificity. Tethering factors are classified by their architecture into two groups:

Long coiled-coil tethers (e.g., GM130, giantin, p115 for ER-to-Golgi transport; EEA1 for early endosome tethering). These are extended dimeric coiled coils of $100$ – $200$ nm in length that bridge the vesicle and target membranes. EEA1, an effector of Rab5, is a parallel coiled-coil homodimer that binds Rab5-GTP on the early endosome via its C-terminal FYVE domain and captures incoming vesicles via its N-terminal Rab5-binding site. Cryo-EM studies (Murray et al. 2016 Nature 537, 567-571) show that EEA1 undergoes a conformational transition from an extended rigid rod to a more flexible collapsed state upon Rab5 binding, effectively "reeling in" the vesicle.
Multi-subunit tethering complexes (MTCs). These are large assemblies ( $\sim 200$ – $800$ kDa) composed of $6$ – $10$ subunits. Eight MTCs have been identified, each associated with a specific transport step. The HOPS complex (homotypic fusion and vacuole protein sorting) is the best characterised: it mediates homotypic fusion of yeast vacuoles and consists of six subunits (Vps11, Vps16, Vps18, Vps33, Vps39, Vps41). Vps39 and Vps41 bind Rab-GTP (Ypt7 in yeast, Rab7 in mammals) on the two membranes, while Vps33 is a Sec1/Munc18 (SM) family protein that directly binds SNAREs and promotes their assembly. The exocyst (eight subunits: Sec3, Sec5, Sec6, Sec8, Sec10, Sec15, Exo70, Exo84) tethers secretory vesicles to the plasma membrane; Sec15 binds Rab11-GTP on the vesicle, and Sec3/Exo70 bind landmarks on the plasma membrane including PIP $_{2}$ and Rho GTPases.

The tethering step is not merely a passive tether. MTCs actively promote SNARE complex assembly: Vps33 (in HOPS) and other SM-family subunits bind syntaxin-family Qa-SNAREs and catalyse the transition from closed (autoinhibited) to open (fusion-competent) conformations, effectively chaperoning the initial SNARE pairing event. This dual function — tethering and SNARE regulation — explains why tethering complexes are essential for fusion even when vesicles are artificially brought into proximity by other means.

Homotypic fusion and NSF/alpha-SNAP disassembly. After fusion, the cis-SNARE complex must be disassembled for the SNAREs to be recycled. NSF is a member of the AAA+ (ATPases associated with diverse cellular activities) family and assembles as a hexameric ring. Alpha-SNAP (soluble NSF attachment protein) binds the SNARE four-helix bundle and recruits NSF. Cryo-EM structures of the NSF/alpha-SNAP/SNARE complex (Zhao et al. 2015 Nature 527, 62-67; White et al. 2018 Nature 558, 382-386) reveal that NSF threads the SNARE complex through its central pore, with the D1 ATPase domains of each NSF subunit gripping the SNARE helices in a staircase-like arrangement. ATP hydrolysis drives a power stroke that pulls the SNARE helices through the pore, unwinding the four-helix bundle processively. Each NSF hexamer hydrolyses approximately 3 ATP molecules per disassembly event, providing $\sim 30 k_{B} T$ of free energy — sufficient to unwind the $\sim 35 k_{B} T$ SNARE complex because the disassembly is coupled to thermal fluctuations that transiently weaken individual layers.

Cargo sorting signals. The specificity of vesicle trafficking depends not only on correct targeting (SNAREs and Rabs) but also on correct cargo packaging. Major sorting signals include:

Di-acidic (DxE). Found on cytoplasmic tails of transmembrane cargo exported from the ER. Recognised by Sec24 in the COPII inner coat. First identified in vesicular stomatitis virus glycoprotein (VSV-G) by Nishimura and Balch (1997 Science 277, 556-558).
Di-hydrophobic (FF, YY, LIL). Also ER export signals recognised by Sec24. Different Sec24 isoforms (Sec24A-D in mammals) have distinct signal-binding specificities, providing cargo selectivity.
KKXX (di-lysine). C-terminal cytoplasmic retrieval signal on type I ER membrane proteins. Recognised by COPI coatomer (specifically the $δ$ -COP and $β^{'}$ -COP subunits), packaging the protein into retrograde COPI vesicles.
KDEL (and variants HDEL, RDEL). C-terminal retrieval signal on soluble ER-resident proteins. Recognised by the KDEL receptor (Erd2) in the cis-Golgi and ERGIC. The receptor cycles between the Golgi (where it binds KDEL at slightly acidic pH) and the ER (where it releases KDEL at neutral pH), using the pH gradient as a directional bias.
Mannose-6-phosphate (M6P). Added to N-linked oligosaccharides on lysosomal hydrolases in the cis-Golgi by the phosphotransferase (the enzyme deficient in I-cell disease) and the uncovering enzyme. Recognised by M6P receptors (cation-dependent and cation-independent) in the TGN, which package enzymes into clathrin-coated vesicles for delivery to late endosomes.
Tyrosine-based (YXX $Φ$ ). Internalisation signal on plasma membrane proteins. Recognised by the $μ$ 2 subunit of AP-2 (clathrin adaptor). The tyrosine fits into a hydrophobic pocket on $μ$ 2, and the $Φ$ (large hydrophobic) residue provides additional affinity.
Dileucine-based ([DE]XXXL[LI]). Endocytosis and lysosomal targeting signal. Recognised by the $σ$ / $δ$ subunits of AP complexes.

ER-associated degradation (ERAD). Proteins that fail to fold correctly in the ER are not exported but are instead targeted for degradation by the ubiquitin-proteasome system. The ERAD pathway has three branches distinguished by the topology of the substrate:

ERAD-L (luminal substrates). Misfolded soluble proteins in the ER lumen are recognised by lectins (OS-9, XTP3-B) and delivered to the retro-translocation channel (a complex of Hrd1, Derlin-1, and SEL1L). The substrate is ubiquitinated on the cytoplasmic side by the E3 ligase Hrd1 and extracted from the membrane by the AAA+ ATPase p97/VCP (valosin-containing protein), then degraded by the proteasome.
ERAD-M (membrane substrates). Misfolded transmembrane proteins are recognised by their aberrant transmembrane domains and retro-translocated through the same Hrd1 channel.
ERAD-C (cytoplasmic substrates). Misfolded cytoplasmic domains of ER membrane proteins are recognised by the Doa10 E3 ligase complex.

ERAD is not a minor quality-control pathway: in specialised secretory cells (plasma cells secreting antibodies, pancreatic beta cells producing insulin), up to $30%$ of nascent proteins are degraded by ERAD, reflecting the high error rate of protein folding in the ER lumen.

Diseases of vesicle trafficking. Defects in vesicle trafficking components cause a range of human diseases:

I-cell disease (mucolipidosis II). Deficiency in GlcNAc-1-phosphotransferase, which adds the first GlcNAc-phosphate to mannose residues on lysosomal enzymes. Without M6P tags, lysosomal enzymes are secreted instead of being delivered to lysosomes. Lysosomes accumulate undegraded substrates as inclusion bodies. Severe skeletal abnormalities, developmental delay, and death in childhood.
Hermansky-Pudlak syndrome (HPS). Mutations in components of the BLOC (biogenesis of lysosome-related organelles complex) complexes or the AP-3 adaptor. Defective trafficking to lysosome-related organelles (melanosomes, platelet dense granules). Clinical features: oculocutaneous albinism, bleeding diathesis (absent platelet dense granules), and pulmonary fibrosis. At least 10 genes (HPS-1 through HPS-10) identified.
Chediak-Higashi syndrome. Mutations in LYST, a regulator of lysosomal trafficking. Giant lysosomal granules form in all granule-bearing cells. Features: partial albinism, recurrent infections (defective NK cell and cytotoxic T-cell granule release), and an accelerated phase of lymphoproliferative infiltration. The LYST protein is thought to regulate fusion/fission of lysosome-related organelles, and its absence leads to uncontrolled fusion.
Familial hypercholesterolaemia (class 2 mutations). Some LDL receptor mutations cause ER retention and ERAD of the receptor, reducing the number of receptors that reach the plasma membrane and impairing LDL uptake.

Connections Master

Cell membranes: structure 17.02.01. The bilayer architecture and membrane asymmetry established in the membrane-structure unit provide the physical substrate for vesicle budding and fusion. The bending rigidity and spontaneous curvature of the bilayer determine the energetic cost of vesicle formation, and the lipid composition of each organelle membrane contributes to its identity in the Rab/SNARE targeting system.
Membrane proteins 17.02.03 pending. The classification of integral and peripheral membrane proteins, the positive-inside rule, and the topology of transmembrane domains all determine how cargo proteins are oriented during translocation and how their cytoplasmic sorting signals (di-acidic, KKXX, YXX $Φ$ ) are presented to coat adaptors. The hydrophobic-mismatch-based retention mechanism introduced in the membrane-proteins unit also explains why short transmembrane proteins are retained in the ER while longer ones traffic forward.
Cell signalling 17.07.01. SNARE-mediated exocytosis is the effector mechanism for many signalling events: neurotransmitter release at synapses, insulin secretion from beta cells, and antibody secretion from plasma cells all depend on regulated vesicle fusion triggered by calcium influx. The Rab GTPase cycle introduced here parallels the G-protein cycle in GPCR signalling, both using GTP-GDP switching as a molecular timer.
Cellular organisation: organelles 17.03.01. Each organelle in the secretory pathway (ER, Golgi, endosomes, lysosomes) has a distinct identity maintained by its resident proteins and lipid composition. Vesicle trafficking is the mechanism that establishes and maintains these identities: anterograde traffic delivers new components, retrograde traffic retrieves escaped residents, and the balance of the two maintains steady-state organelle composition.
Cytoskeleton [17.03.01-17.03.02]. Long-range vesicle transport depends on microtubule motors (kinesin for anterograde, dynein for retrograde) and actin-based myosin motors for short-range movement near the cortex. Rab GTPases on vesicle membranes recruit motor adaptor proteins, coupling vesicle identity to directional transport.

Historical notes Master

The secretory pathway was first visualised by George Palade in the 1950s and 1960s using electron microscopy of pancreatic acinar cells, which are specialised for massive protein secretion. Palade traced the path of secretory proteins from the ER through the Golgi to secretory vesicles and finally to the cell surface, establishing the morphological framework for the entire field. He received the 1974 Nobel Prize in physiology or medicine for this work.

The signal hypothesis, proposed by Blobel and Dobberstein in 1975 (J. Cell Biol. 67, 835-851), explained how proteins are targeted to the ER for entry into the secretory pathway. Blobel received the 1999 Nobel Prize. The co-translational translocation mechanism, in which the SRP binds the signal sequence of the nascent polypeptide and targets the ribosome to the Sec61 translocon, was elucidated through the combined work of Walter, Blobel, and Meyer (Walter and Blobel 1981 J. Cell Biol. 91, 557-561).

The vesicle coat proteins were discovered in the 1980s and 1990s. Randy Schekman's genetic screen in yeast (conducted from the late 1970s through the 1980s) identified the SEC genes required for ER-to-Golgi transport, including SEC12, SEC13, SEC16, SEC23, SEC24, and SEC31 — the components of the COPII coat. The COPI coatomer was biochemically purified by Rothman and colleagues (Waters, Serafini, and Rothman 1991 Nature 349, 248-251). Schekman shared the 2013 Nobel Prize with Rothman and Sudhof.

Rothman's discovery of SNARE proteins began with the identification of NSF (N-ethylmaleimide sensitive factor) and SNAPs as the cytoplasmic factors required for intra-Golgi transport in a cell-free assay (Block, Glick, and Rothman 1988 Proc. Natl. Acad. Sci. 85, 7852-7856). The key breakthrough was the identification of SNAREs (SNAP receptors) as the membrane proteins that bind alpha-SNAP and NSF, by Sollner, Whiteheart, and colleagues (1993 Nature 362, 318-324). This paper showed that the synaptic proteins synaptobrevin (VAMP), syntaxin, and SNAP-25 are SNAREs, connecting the vesicle-fusion machinery to the neuroscience of neurotransmitter release.

Thomas Sudhof's work on synaptotagmin as the calcium sensor for neurotransmitter release (Brose, Petrenko, and Sudhof 1992 J. Biol. Chem. 267, 13237-13240; Chapman and Jahn 1994 Annu. Rev. Physiol. 56, 349-362) and on the Munc18/syntaxin regulatory interaction completed the picture of regulated exocytosis. Sudhof shared the 2013 Nobel Prize with Schekman and Rothman.

The Rab GTPase family was discovered by Chavrier and colleagues (1990 Mol. Cell. Biol. 10, 6530-6539) and Zerial and McBride (2001 Nat. Rev. Mol. Cell Biol. 2, 107-117) mapped the functional specificity of individual Rabs to specific transport steps. The concept of Rabs as membrane identity markers, with their GTP/GTP cycle providing a timing mechanism for membrane association, has become the organising principle for understanding organelle specificity in vesicle trafficking.

The clathrin coat was the first vesicle coat to be characterised, initially by Pearse (1975 J. Mol. Biol. 97, 93-98) who purified coated vesicles and identified the 180 kDa clathrin heavy chain. The triskelion structure was visualised by negative-stain electron microscopy (Ungewickell and Branton 1981 Nature 289, 420-422). The adaptor protein complexes (AP-1, AP-2) were identified by Keen and colleagues (Robinson and Pearse 1986 J. Cell Biol. 102, 48-54) as the link between clathrin and cargo.

The single-particle cryo-EM revolution of the 2010s transformed the structural understanding of vesicle trafficking machinery. Structures of the COPII cage (Stagg et al. 2008 J. Mol. Biol. 383, 519-528), the clathrin coat at near-atomic resolution (Fotin et al. 2004 Nature 432, 573-579; Morris et al. 2019 Science 366, eaax2879), the HOPS complex (Baker et al. 2015 J. Cell Biol. 210, 277-292), and the NSF/alpha-SNAP/SNARE disassembly machine (White et al. 2018 Nature 558, 382-386) provided the structural framework for interpreting decades of genetic and biochemical data.

Bibliography Master

Alberts, B. et al. Molecular Biology of the Cell, 7th ed. Garland Science, 2022. Ch. 13 (Intracellular Membrane Traffic) and Ch. 14 (Intracellular Compartments).
Lodish, H. et al. Molecular Cell Biology, 9th ed. W. H. Freeman, 2021. Ch. 14 (Vesicular Traffic, Secretion, and Endocytosis).
Rothman, J. E. The Protein Life Cycle, 1st ed. W. H. Freeman, 2021. Ch. 1-4 (SNARE machinery and vesicle fusion).
Sollner, T. et al. "SNAP receptors implicated in vesicle targeting and fusion." Nature 362 (1993): 318-324. The foundational SNARE identification paper.
Block, M. R., Glick, B. S., and Rothman, J. E. "Purification of an N-ethylmaleimide-sensitive protein catalyzing vesicular transport." Proc. Natl. Acad. Sci. USA 85 (1988): 7852-7856.
Palade, G. "Intracellular aspects of the process of protein synthesis." Science 189 (1975): 347-358. Nobel Prize lecture describing the secretory pathway.
Novick, P., Field, C., and Schekman, R. "Identification of 23 complementation groups required for post-translational events in the yeast secretory pathway." Cell 21 (1980): 205-215. The SEC gene screen.
Barlowe, C. et al. "COPII: a membrane coat formed by Sec proteins that drive vesicle budding from the endoplasmic reticulum." Cell 77 (1994): 895-907.
Waters, M. G., Serafini, T., and Rothman, J. E. "'Coatomer': a cytosolic protein complex containing subunits of non-clathrin-coated Golgi transport vesicles." Nature 349 (1991): 248-251.
Pearse, B. M. F. "Coated vesicles from pig brain: purification and biochemical characterization." J. Mol. Biol. 97 (1975): 93-98.
Gao, Y. et al. "Single reconstituted neuronal SNARE complexes zipper in three distinct stages." Science 337 (2012): 1340-1343.
Zhao, Y. et al. "Mechanistic insights into the recycling machine of the SNARE complex." Nature 527 (2015): 62-67.
White, K. et al. "Structures of the NSF N-domain and D2-ring reveal the mechanism of SNARE complex disassembly." Nature 558 (2018): 382-386.
Zerial, M. and McBride, H. "Rab proteins as membrane organizers." Nat. Rev. Mol. Cell Biol. 2 (2001): 107-117.
Fotin, A. et al. "Molecular structure of a clathrin coat at 10 A resolution." Nature 432 (2004): 573-579.
Murray, D. H. et al. "An endosomal tether undergoes an entropic collapse to bring vesicles together." Nature 537 (2016): 567-571.

Prerequisites

17.02.01 pending
17.02.03 pending

Tier anchors

beginner: Alberts et al., Molecular Biology of the Cell, 7th ed. (2022), Ch. 13-14
intermediate: Lodish et al., Molecular Cell Biology, 9th ed. (2021), Ch. 14
master: Rothman, J. E. — The Protein Life Cycle, 1st ed. (2021)

References

Alberts et al. — Molecular Biology of the Cell, 7th ed. (Garland Science, 2022) · Ch. 13 Intracellular Membrane Traffic; Ch. 14 Intracellular Compartments
Lodish et al. — Molecular Cell Biology, 9th ed. (W. H. Freeman, 2021) · Ch. 14 Vesicular Traffic, Secretion, and Endocytosis
Rothman, J. E. — The Protein Life Cycle (W. H. Freeman, 2021) · Ch. 1-4 SNARE machinery and vesicle fusion
Sollner, T. et al. — SNAP receptors implicated in vesicle targeting and fusion · Nature 362 (1993) 318-324

Estimated time

beginner: 15m
intermediate: 30m
master: 45m