Intracellular Membrane Traffic

Eukaryotic cells are divided into membrane-enclosed compartments, each with a distinct protein and lipid composition. Yet these compartments are not isolated — they communicate continuously through an elaborate system of intracellular membrane traffic. Proteins synthesized in the endoplasmic reticulum must reach the Golgi apparatus for modification, then be sorted onward to lysosomes, the plasma membrane, or secretory vesicles. Material from the cell surface must be internalized and routed to the correct intracellular destination. All of this occurs with remarkable specificity: each transport carrier knows where it came from, what cargo it carries, and where it needs to go.

This lesson examines the molecular machinery that drives membrane traffic, the major trafficking routes within the cell, and bioinformatics approaches for analyzing vesicle transport pathways.

13.1 Molecular Mechanisms of Membrane Transport

Types of Coated Vesicles

Transport vesicles bud from donor compartments with the help of coat proteins that serve two critical functions: they curve the membrane into a bud and they select the correct cargo molecules. Three major types of coated vesicles operate in eukaryotic cells:

Each coat type is recruited to specific membranes and recognizes distinct cargo signals, ensuring that the right proteins are packaged into the right vesicles.

Clathrin Coat Assembly Drives Vesicle Formation

Clathrin is the best-understood coat protein. Each clathrin molecule is a three-legged structure called a triskelion, composed of three heavy chains and three light chains. Triskelions assemble into a polyhedral lattice — a basket-like cage — on the cytoplasmic face of the membrane, progressively curving the membrane into a coated pit.

Clathrin does not bind cargo or membrane directly. Instead, adaptor proteins bridge clathrin to the membrane and select cargo. The adaptor complex AP2 operates at the plasma membrane during endocytosis, while AP1 works at the trans-Golgi network. These adaptors recognize specific sorting signals in the cytoplasmic tails of transmembrane cargo receptors — often short peptide motifs such as the tyrosine-based YXXΦ motif (where Φ is a bulky hydrophobic residue) or the dileucine motif [DE]XXXL[LI].

Once the coated pit is fully invaginated, the GTPase dynamin assembles as a helical collar around the neck of the bud and uses GTP hydrolysis to pinch the vesicle free from the donor membrane. The clathrin coat is then rapidly disassembled by the ATPase Hsc70 and its cofactor auxilin, exposing the vesicle surface for targeting and fusion.

Not All Transport Vesicles Are Spherical

While textbooks often depict transport carriers as small, round vesicles, many carriers are tubular or irregularly shaped. Tubular carriers are particularly important for bulk membrane flow between compartments and for sorting in endosomes, where long tubules extend from the endosome body and concentrate recycling receptors by geometry alone — proteins in the tubule membrane are enriched relative to soluble contents because tubules have a high surface-to-volume ratio.

Phosphoinositides Mark Organelle Identity

How do coat proteins and other trafficking machinery know which membrane to bind? A key part of the answer lies in phosphoinositides (PIs) — phosphorylated derivatives of phosphatidylinositol. Different PI species are concentrated on different organelle membranes:

PI kinases and PI phosphatases generate and remove these lipid marks, creating a spatial code that recruits specific effector proteins to the correct membrane. For example, the FYVE domain found in many endosomal proteins binds specifically to PI(3)P, anchoring these proteins to early endosomes.

Membrane-Bending Proteins and BAR Domains

Coat proteins are not the only way to deform membranes. BAR domain proteins (named after Bin, Amphiphysin, and Rvs) are crescent-shaped dimers that bind to membranes and impose curvature matching their intrinsic shape. Different BAR domain subfamilies generate different degrees of curvature:

• N-BAR domains sense and stabilize moderate positive curvature
• F-BAR domains promote shallow curvature, involved in early stages of endocytic pit formation
• I-BAR domains induce negative curvature (membrane protrusions)

By inserting amphipathic helices into one leaflet of the bilayer, some BAR domain proteins can also actively generate curvature rather than merely stabilizing it. These proteins work in concert with coat proteins, the actin cytoskeleton, and lipid-modifying enzymes to drive the complex membrane remodeling required for vesicle budding and tubule formation.

Cytoplasmic GTPases Control Coat Assembly

The recruitment of coat proteins to membranes is tightly regulated by small GTPases of the Arf/Sar1 family. These GTPases act as molecular switches:

1. In the GDP-bound (inactive) state, the GTPase is soluble in the cytoplasm
2. A guanine nucleotide exchange factor (GEF) on the donor membrane catalyzes the exchange of GDP for GTP
3. GTP binding exposes an amphipathic helix (in Sar1) or a myristoyl group (in Arf1) that inserts into the membrane, anchoring the GTPase
4. The membrane-bound GTPase recruits coat protein subunits
5. A GTPase-activating protein (GAP) stimulates GTP hydrolysis, triggering coat disassembly after budding

This GTPase cycle ensures that coat assembly is both spatially restricted (only occurring where the appropriate GEF is located) and temporally controlled (coats are removed after they have done their job).

Not All Transport Carriers Are Vesicles

Some transport between compartments does not involve vesicles at all. Membrane tubules and contact sites provide alternative routes:

• Membrane contact sites (MCS) are regions where two organelles are held in close apposition (10–30 nm apart) by tethering proteins. Lipids can be transferred directly between membranes at these sites by lipid transfer proteins, without the need for vesicular carriers.
• Tunnel-like connections between cisternae may allow direct transfer within the Golgi stack.

These non-vesicular pathways are especially important for lipid transport, which would be inefficient by vesicular means because each vesicle carries relatively few lipid molecules in its bilayer relative to the proteins in its lumen.

Rab Proteins Guide Vesicles to Their Target

Once a transport vesicle has budded and shed its coat, it must find and fuse with the correct target membrane. Rab GTPases — the largest family of small GTPases in the cell, with over 60 members in humans — are the master regulators of vesicle targeting.

Each Rab protein is associated with one or a few specific organelle membranes. In the GTP-bound (active) state, Rabs recruit effector proteins that mediate vesicle motoring along cytoskeletal tracks, tethering to the target membrane, and eventual fusion. For example, Rab5 marks early endosomes and recruits tethering factors such as EEA1 (early endosome antigen 1), while Rab7 marks late endosomes and lysosomes.

Rab Cascades Change Organelle Identity

A remarkable feature of the endomembrane system is that compartments can mature and change identity over time. This is driven by Rab cascades: one Rab protein recruits the GEF for the next Rab in the pathway while simultaneously recruiting a GAP that inactivates itself. For example, Rab5 on early endosomes recruits the Rab7 GEF (the Mon1-Ccz1 complex), which activates Rab7. Rab7, in turn, recruits a Rab5 GAP, causing Rab5 to be inactivated and removed. The net result is the conversion of a Rab5-positive early endosome into a Rab7-positive late endosome — a process called Rab conversion.

SNAREs Mediate Membrane Fusion

The final step in vesicle transport is membrane fusion, catalyzed by SNARE proteins (soluble NSF attachment protein receptors). There are about 38 SNAREs in a typical mammalian cell, and each fusion event uses a specific combination.

SNAREs on the vesicle (v-SNAREs, also called R-SNAREs) interact with complementary SNAREs on the target membrane (t-SNAREs, also called Q-SNAREs). The v-SNARE and t-SNARE coil around each other to form a tight, four-helix bundle called the trans-SNARE complex (or SNAREpin). The energy released by this coiling pulls the two membranes together, overcoming the electrostatic repulsion between them, and drives fusion of the lipid bilayers.

The specificity of SNARE pairing contributes to the accuracy of membrane trafficking: only the correct v-SNARE/t-SNARE combination forms a stable complex, ensuring that vesicles fuse with the right target.

SNAREs Need to Be Pried Apart by NSF and SNAP

After fusion, the v-SNARE and t-SNARE are trapped in the same membrane as a very stable cis-SNARE complex. To recycle the SNAREs for another round of fusion, the AAA+ ATPase NSF (N-ethylmaleimide-sensitive factor) and its adaptor protein α-SNAP (soluble NSF attachment protein) bind the cis-SNARE complex and use ATP hydrolysis to mechanically disassemble it. The separated SNAREs are then returned to their respective compartments — the v-SNARE is recycled by retrograde vesicle transport, and the t-SNARE remains on the target membrane.

Bioinformatics of Vesicle Trafficking Pathways

The vesicle trafficking machinery involves large families of paralogous proteins — Rabs, SNAREs, coat subunits, and tethering factors — that have diversified through gene duplication and divergence. Bioinformatics approaches are essential for classifying these families, predicting their interactions, and reconstructing trafficking pathways.

Coat protein family classification. The COPI, COPII, and clathrin systems share an ancient evolutionary origin. Structural bioinformatics has revealed that the β-propeller and α-solenoid domain architecture is conserved across all three systems and even in the nuclear pore complex, suggesting they evolved from a common protocoatomer ancestor.

Rab GTPase family analysis. With over 60 Rabs in humans, phylogenetic analysis is critical for predicting function. Sequence alignment of Rab family members reveals conserved GTPase domains alongside hypervariable regions that determine effector specificity. Databases such as Rabifier use hidden Markov models to classify Rab sequences from any organism into functional subfamilies.

SNARE complex interaction databases. Resources like SNAREbase catalog known SNARE complexes and their trafficking routes. Coiled-coil prediction algorithms help identify new SNARE proteins from genomic data, and coevolution analysis can predict which SNAREs form functional pairs.

Trafficking pathway reconstruction from proteomics. Organelle proteomics — isolating organelles by density gradient fractionation and identifying their protein content by mass spectrometry — allows computational reconstruction of trafficking pathways. By tracking how protein localization changes over time (dynamic proteomics), researchers can determine flux rates through the secretory pathway.

Let’s explore the protein families involved in trafficking by examining their sequences:

let snare_motif = "EIIKLENSIRELHDMFMDMAMLVESQGEMIDRIEYNVEHAVDYVE"
let props = Struct.protein_props(snare_motif)
print("SNARE coiled-coil domain properties:")
print(props)

The SNARE coiled-coil domain has a characteristic pattern of hydrophobic and charged residues that enables the four-helix bundle to form during membrane fusion. Analyzing the amino acid composition reveals the amphipathic nature of these domains.

let er_retention = "KDEL"
let lysosomal = "NPGY"
let endocytic = "YXXF"
print("ER retention signal (KDEL):")
print(Struct.protein_props(er_retention))
print("Lysosomal sorting signal:")
print(Struct.protein_props(lysosomal))
print("Endocytic signal (YXXΦ):")
print(Struct.protein_props(endocytic))

Sorting signals have distinct biochemical properties that determine how they are recognized by the trafficking machinery. The charged KDEL motif is bound by a pH-sensitive receptor, while the tyrosine-based endocytic signals interact with adaptor protein complexes.

let coat_data = '[{"label": "COPII (ER→Golgi)", "value": 70}, {"label": "COPI (Golgi→ER)", "value": 60}, {"label": "Clathrin (PM→endosome)", "value": 100}]'
let chart = Viz.bar(coat_data, '{"title": "Coated Vesicle Diameter (nm)", "color": "#8B5CF6"}')
print(chart)

13.2 ER to Golgi Transport

COPII-Coated Vesicles

The first leg of the secretory pathway — from the ER to the Golgi — is mediated by COPII-coated vesicles. COPII coat assembly begins when the small GTPase Sar1 is activated by its GEF (Sec12), which is an ER-resident transmembrane protein. GTP-bound Sar1 exposes an amphipathic helix that inserts into the ER membrane, then recruits the Sec23/Sec24 inner coat complex. Sec24 is the cargo-binding subunit — it recognizes ER export signals on transmembrane cargo proteins. The Sec13/Sec31 outer coat then polymerizes over the inner coat, providing the mechanical force to curve the membrane into a bud.

COPII vesicles bud from specialized regions of the ER called ER exit sites (ERES), which are marked by clusters of Sec16 protein and lack bound ribosomes.

Only Properly Folded Proteins Leave the ER

The ER quality control system ensures that only properly folded and assembled proteins are packaged into COPII vesicles. Misfolded proteins are retained in the ER by chaperones such as BiP (Hsp70 family) and the lectin chaperones calnexin and calreticulin. Proteins that persistently fail to fold are targeted for ER-associated degradation (ERAD) — retrotranslocation to the cytoplasm and destruction by the proteasome.

Some cargo proteins are actively concentrated into COPII vesicles by specific export signals. Others may leave the ER by bulk flow — passive incorporation at the prevailing concentration — without requiring a specific signal.

Vesicular Tubular Clusters and the ERGIC

COPII vesicles do not travel directly to the Golgi. Instead, they fuse with one another and with incoming retrograde carriers to form vesicular tubular clusters (VTCs), also known as the ER-Golgi intermediate compartment (ERGIC). The ERGIC is a dynamic sorting station where:

• Cargo destined for the Golgi is concentrated and moved forward
• ER-resident proteins that escaped are retrieved and sent back
• Further quality control occurs

The ERGIC matures and moves along microtubules toward the Golgi, effectively becoming the cis-Golgi cisterna.

COPI Retrograde Retrieval

COPI-coated vesicles mediate retrograde transport from the Golgi back to the ER. This retrieval pathway has two critical functions:

1. Recovering ER-resident proteins that have accidentally escaped. Soluble ER-resident proteins (such as BiP and protein disulfide isomerase) carry a KDEL retrieval signal at their C-terminus. The KDEL receptor in the Golgi binds these proteins at the slightly acidic pH of the Golgi, packages them into COPI vesicles, and releases them back in the ER at the higher pH of the ER lumen. ER-resident transmembrane proteins carry a KKXX motif (dilysine signal) on their cytoplasmic tail, which is directly recognized by COPI coat subunits.

2. Recycling SNARE proteins and other trafficking machinery used in the anterograde pathway.

The Golgi: An Ordered Series of Compartments

The Golgi apparatus consists of a series of flattened, membrane-enclosed cisternae organized into functionally distinct regions:

• Cis-Golgi network (CGN): receives material from the ERGIC
• Medial-Golgi: contains enzymes for intermediate processing steps
• Trans-Golgi network (TGN): the major sorting station, directing proteins to lysosomes, the plasma membrane, or secretory granules

How proteins traverse the Golgi stack remains debated. The cisternal maturation model proposes that cisternae themselves mature from cis to trans, with Golgi-resident enzymes being recycled backward by COPI vesicles. The alternative vesicular transport model proposes that cisternae are stable and cargo moves forward in vesicles. Current evidence supports a hybrid view, with cisternal maturation being the predominant mechanism for large cargo (such as procollagen aggregates too big to fit in vesicles) and vesicular transport possibly operating in parallel.

Oligosaccharide Processing in the Golgi

As glycoproteins traverse the Golgi stack, their N-linked oligosaccharides — initially added in the ER as a uniform 14-sugar precursor — are extensively processed. Glycosidases trim mannose residues, and glycosyltransferases add sugars such as N-acetylglucosamine, galactose, sialic acid, and fucose. Because these enzymes are arranged in a cis-to-trans gradient within the Golgi, the order of processing is determined by the order of cisternal passage.

The result is enormous structural diversity: a single protein can carry many different oligosaccharide structures (glycoforms), which influence folding, stability, recognition, and half-life. O-linked glycosylation also begins in the Golgi, with sugars added to serine and threonine residues.

Proteoglycan Assembly

The Golgi is also the site of proteoglycan assembly. Proteoglycans consist of a core protein decorated with long, unbranched glycosaminoglycan (GAG) chains such as heparan sulfate, chondroitin sulfate, and keratan sulfate. GAG chain synthesis begins with a tetrasaccharide linker added to serine residues on the core protein, followed by alternating addition of two different sugars to build the repeating disaccharide chain. Subsequent sulfation by specific sulfotransferases in the Golgi creates the enormous structural diversity of GAGs, which is critical for their roles in growth factor signaling, cell adhesion, and extracellular matrix organization.

13.3 From the Trans-Golgi Network to Lysosomes

Lysosomes Are the Principal Sites of Intracellular Digestion

Lysosomes are membrane-enclosed organelles containing approximately 60 different acid hydrolases — enzymes that degrade proteins, nucleic acids, lipids, and polysaccharides. These enzymes operate optimally at the acidic pH (~4.5–5.0) maintained inside the lysosome by a V-type H⁺ ATPase (vacuolar proton pump) in the lysosomal membrane.

The lysosomal membrane protects the rest of the cell from these destructive enzymes. It is enriched in heavily glycosylated membrane proteins (such as LAMP-1 and LAMP-2) whose sugar coat forms a protective glycocalyx on the luminal surface, shielding the membrane from digestion.

Lysosomes Are Heterogeneous

Lysosomes are not uniform organelles. They vary in size, shape, and content depending on what they are currently digesting. Primary lysosomes are newly formed and contain only hydrolases. Secondary lysosomes (phagolysosomes, autolysosomes) have fused with endosomes, phagosomes, or autophagosomes and contain material undergoing degradation. Residual bodies contain indigestible remnants. This heterogeneity reflects the lysosome’s role as a dynamic endpoint for multiple degradation pathways.

Modern cell biology now recognizes lysosomes as more than simple degradation compartments. They also function as signaling platforms: the mTORC1 kinase complex is activated on the lysosomal surface in response to amino acid availability, linking lysosomal function to cell growth and metabolism.

Plant and Fungal Vacuoles

In plants and fungi, the functional equivalent of the lysosome is the vacuole — a large, acidic compartment that can occupy up to 90% of the cell volume. Beyond degradation, plant vacuoles serve additional roles in turgor pressure maintenance, pigment storage (anthocyanins), sequestration of toxic compounds, and nutrient storage. Yeast vacuoles are a powerful model for studying membrane traffic because the genes controlling vacuolar sorting (VPS genes) are conserved from yeast to humans.

Multiple Pathways Deliver Material to Lysosomes

Lysosomes receive material from several sources:

• Endocytosis: extracellular material arrives via early endosomes → late endosomes → lysosomes
• Phagocytosis: large particles (bacteria, dead cells) are engulfed and delivered in phagosomes
• Autophagy: cytoplasmic components and damaged organelles are enclosed in autophagosomes
• Direct delivery: some cargo is transported directly from the TGN

Autophagy Degrades Unwanted Proteins and Organelles

Autophagy (“self-eating”) is a conserved pathway by which cells degrade their own cytoplasmic components. In macroautophagy (the most studied form):

1. A cup-shaped phagophore (isolation membrane) nucleates and expands around a portion of cytoplasm or a targeted organelle
2. The edges of the phagophore fuse to form a double-membrane autophagosome
3. The autophagosome fuses with a lysosome, forming an autolysosome, and the inner membrane and contents are degraded

Autophagy is regulated by ATG (autophagy-related) proteins, originally identified in yeast. The process is induced by starvation (nutrient deprivation), as the mTORC1 kinase — which normally suppresses autophagy — is inactivated when amino acids are scarce. The ubiquitin-like conjugation systems ATG12-ATG5-ATG16L1 and LC3 (ATG8) are essential for phagophore expansion and closure. LC3 is lipidated (conjugated to phosphatidylethanolamine) and inserted into the phagophore membrane, where it recruits cargo receptors such as p62/SQSTM1 that recognize ubiquitinated substrates.

Selective autophagy pathways degrade specific targets: mitophagy (damaged mitochondria, mediated by PINK1/Parkin), pexophagy (peroxisomes), ribophagy (ribosomes), and xenophagy (intracellular pathogens).

Mannose 6-Phosphate Receptor Sorts Lysosomal Hydrolases

Lysosomal enzymes are synthesized in the ER and travel through the Golgi like other secretory proteins. In the cis-Golgi, the enzyme GlcNAc phosphotransferase recognizes a three-dimensional signal patch on the surface of lysosomal hydrolases (not a linear sequence) and adds N-acetylglucosamine phosphate to mannose residues on their N-linked oligosaccharides. A second enzyme removes the GlcNAc, exposing the mannose 6-phosphate (M6P) tag.

In the trans-Golgi network, M6P receptors bind the tagged hydrolases and sort them into clathrin-coated vesicles destined for late endosomes. At the acidic pH of the late endosome, the hydrolases dissociate from the M6P receptor, which is recycled back to the Golgi for another round.

Lysosomal Storage Diseases

Defects in lysosomal enzyme targeting cause lysosomal storage diseases — a group of ~50 inherited disorders in which undigested substrates accumulate in lysosomes. If the GlcNAc phosphotransferase is defective, lysosomal enzymes fail to receive the M6P tag, are not sorted to lysosomes, and are instead secreted from the cell. This results in inclusion cell disease (I-cell disease), characterized by dense inclusions in lysosomes and severe developmental abnormalities.

Other lysosomal storage diseases arise from deficiency of a specific hydrolase (e.g., Gaucher disease from glucocerebrosidase deficiency, Tay-Sachs disease from hexosaminidase A deficiency), leading to accumulation of the substrate that enzyme normally degrades.

Bioinformatics of Autophagy and Lysosomal Biology

Autophagy gene databases. The Autophagy Database and HADb (Human Autophagy Database) catalog all known ATG genes and their orthologs across species. Sequence analysis reveals that the core autophagy machinery (ATG1/ULK1 complex, ATG6/Beclin1 complex, the two ubiquitin-like conjugation systems) is conserved from yeast to humans, with additional regulatory complexity added in metazoans.

LC3 interactome analysis. The LC3-interacting region (LIR) motif — a short linear motif [W/F/Y]XX[L/I/V] — mediates binding of cargo receptors and regulatory proteins to LC3 family members. Bioinformatics tools like iLIR scan protein sequences for potential LIR motifs, predicting new autophagy substrates and regulators. Given the short and degenerate nature of the motif, computational predictions must be validated experimentally, but they provide valuable starting points.

Lysosomal storage disease variant analysis. Clinical genomics databases (ClinVar, HGMD) contain thousands of disease-causing variants in lysosomal enzyme genes. Computational tools predict the effect of missense variants on protein stability and function, guiding diagnosis and informing the development of enzyme replacement therapy and pharmacological chaperone treatments.

Autophagy flux from imaging. Computational image analysis pipelines quantify autophagy by counting LC3-positive puncta (autophagosomes) in fluorescence microscopy images. Machine learning algorithms can distinguish genuine autophagosomes from background noise and track their dynamics over time, providing quantitative measures of autophagy flux.

Exercise: Identify Sorting Signals

Protein sorting in the secretory pathway depends on short peptide signals. Analyze the properties of different sorting signals to understand why they direct proteins to specific compartments:

let er_signal = "KDEL"
let nuclear = "PKKKRKV"
let mito = "MLSLRQSIRFFKPATRTLCSSRYLL"
print("ER retention (KDEL):")
print(Struct.protein_props(er_signal))
print("Nuclear localization signal:")
print(Struct.protein_props(nuclear))
print("Mitochondrial targeting:")
print(Struct.protein_props(mito))
// Which signal uses charged residues for receptor recognition?
let answer = "KDEL"
print(answer)

13.4 Endocytosis

Cells internalize material from their surface through several endocytic pathways, each suited to different cargo types and cellular needs.

Pinocytosis and Receptor-Mediated Endocytosis

Pinocytosis (“cell drinking”) is the constitutive uptake of small volumes of extracellular fluid. All eukaryotic cells perform pinocytosis continuously, non-selectively internalizing whatever solutes happen to be dissolved in the fluid.

Receptor-mediated endocytosis, by contrast, is highly selective. Specific receptors on the cell surface bind their ligands with high affinity and concentrate in clathrin-coated pits. When the pit invaginates and pinches off (aided by dynamin), the resulting clathrin-coated vesicle carries both the receptor-ligand complex and a small volume of extracellular fluid into the cell.

The LDL Receptor: A Paradigm

The low-density lipoprotein (LDL) receptor is the classic example of receptor-mediated endocytosis, elucidated by Goldstein and Brown (Nobel Prize, 1985). The LDL receptor:

1. Binds LDL particles (cholesterol-carrying lipoproteins) at the cell surface
2. Concentrates in clathrin-coated pits via an NPXY internalization motif in its cytoplasmic tail
3. Is internalized in coated vesicles that deliver the receptor-LDL complex to early endosomes
4. At the acidic pH of the endosome (~6.0), LDL dissociates from the receptor
5. The receptor is recycled back to the cell surface; the LDL is delivered to lysosomes, where cholesterol is released

Mutations in the LDL receptor cause familial hypercholesterolemia, leading to elevated blood cholesterol and premature cardiovascular disease.

Proteins Are Retrieved from Early Endosomes

Early endosomes are the first sorting station after internalization. They have a mildly acidic pH (~6.0–6.5), which is sufficient to dissociate many ligands from their receptors. From the early endosome, molecules can follow several routes:

• Recycling to the plasma membrane: receptors (such as the LDL receptor and transferrin receptor) are returned to the cell surface, either directly or via recycling endosomes
• Degradation: ligands and down-regulated receptors proceed to late endosomes and lysosomes
• Transcytosis: in polarized epithelial cells, some receptors are transported from one cell surface to another

Multivesicular Bodies

As early endosomes mature into late endosomes, their internal pH drops further (~5.0–5.5), and they acquire intraluminal vesicles (ILVs) — small vesicles formed by inward budding of the endosomal membrane. This gives them the appearance of multivesicular bodies (MVBs).

The formation of ILVs is driven by the ESCRT (endosomal sorting complexes required for transport) machinery, a set of four protein complexes (ESCRT-0, -I, -II, -III) that recognize ubiquitinated membrane proteins, sort them into the invaginating membrane, and catalyze membrane scission. Incorporation into ILVs effectively removes receptors from the cytoplasm, terminating their signaling. When the MVB fuses with a lysosome, the ILVs and their contents are degraded.

Alternatively, MVBs can fuse with the plasma membrane, releasing their ILVs as exosomes — small extracellular vesicles that mediate intercellular communication.

Recycling Endosomes

Recycling endosomes are a distinct compartment, often located near the cell center (pericentriolar), that acts as a reservoir for receptors and membrane awaiting return to the cell surface. They are marked by Rab11 and play important roles in polarized membrane traffic, cell migration, and cytokinesis.

Phagocytosis

Phagocytosis (“cell eating”) is performed by specialized cells — macrophages, neutrophils, and dendritic cells — that engulf large particles such as bacteria, dead cells, and debris. The particle binds to surface receptors (e.g., Fc receptors for antibody-coated particles, complement receptors), triggering actin-driven extension of pseudopods that wrap around and engulf the target. The resulting phagosome fuses with lysosomes to form a phagolysosome, where the ingested material is degraded.

Phagocytosis is crucial for innate immunity and for tissue remodeling during development.

13.5 Exocytosis

Exocytosis is the process by which intracellular vesicles fuse with the plasma membrane, delivering membrane proteins and lipids to the cell surface and releasing soluble cargo into the extracellular space.

Constitutive Secretion

Constitutive exocytosis operates continuously in all cells. Vesicles bud from the TGN and fuse with the plasma membrane without requiring any external signal. This pathway delivers newly synthesized membrane proteins and lipids to the cell surface and secretes extracellular matrix components, growth factors, and other molecules.

Regulated Secretion

Regulated exocytosis is restricted to specialized secretory cells — neurons, endocrine cells, exocrine cells, and mast cells. In this pathway:

1. Secretory proteins are concentrated and stored in secretory granules (also called dense-core vesicles), which accumulate in the cytoplasm
2. The granules are docked at the plasma membrane but do not fuse until triggered
3. An extracellular signal (e.g., a hormone, neurotransmitter, or antigen) raises the intracellular Ca²⁺ concentration
4. Ca²⁺ binds to the vesicle-associated sensor synaptotagmin, which interacts with the SNARE complex and the plasma membrane, triggering rapid membrane fusion and release of the granule contents

In neurons, this process is extraordinarily fast: the time from Ca²⁺ influx to neurotransmitter release can be less than one millisecond. The specialized SNARE complex at the synapse (VAMP2/synaptobrevin on the vesicle, syntaxin-1 and SNAP-25 on the plasma membrane) is primed and ready to fire, needing only the Ca²⁺-synaptotagmin trigger.

Secretory granule contents are often processed during storage — for example, proinsulin is cleaved to insulin and C-peptide within the granules of pancreatic β-cells.

Exercise: SNARE Domain Comparison

SNARE proteins drive membrane fusion by forming a four-helix bundle. Compare the coiled-coil domains of v-SNAREs and t-SNAREs:

let v_snare = "RDQKLSELDDRADALQAGASQFESSAAKLKQTQNKVEEL"
let t_snare = "EIIKLENSIRELHDMFMDMAMLVESQGEMIDRIEYNVEH"
print("v-SNARE domain:")
print(Struct.protein_props(v_snare))
print("t-SNARE domain:")
print(Struct.protein_props(t_snare))
// What structural motif do both domains form?
let answer = "coiled-coil"
print(answer)

Exercise: Map the Secretory Pathway

Visualize the relative protein traffic volume through different steps of the secretory pathway:

let traffic = '[{"label": "ER to Golgi", "value": 100}, {"label": "Golgi to PM", "value": 40}, {"label": "Golgi to lysosome", "value": 15}, {"label": "PM to endosome", "value": 30}, {"label": "Retrograde (Golgi to ER)", "value": 25}]'
let chart = Viz.bar(traffic, '{"title": "Relative Vesicle Traffic Volume", "color": "#F59E0B"}')
print(chart)
// Which pathway handles the most traffic?
let answer = "ER to Golgi"
print(answer)

Knowledge Check

Summary

In this lesson you learned:

• Coat proteins (clathrin, COPII, COPI) select cargo, curve membranes into vesicles, and are regulated by Arf/Sar1 family GTPases
• Phosphoinositides mark organelle identity, and BAR domain proteins assist in membrane bending
• Rab GTPases (~60 in humans) guide vesicles to target membranes; Rab cascades drive compartment maturation
• SNAREs mediate membrane fusion via trans-SNARE complex formation; NSF and α-SNAP recycle SNAREs after fusion
• COPII vesicles carry properly folded cargo from the ER through the ERGIC to the Golgi; COPI vesicles retrieve escaped ER residents via KDEL/KKXX signals
• The Golgi processes oligosaccharides and assembles proteoglycans as cargo traverses from cis to trans cisternae
• Lysosomes contain acid hydrolases targeted by mannose 6-phosphate; defects in M6P tagging cause lysosomal storage diseases
• Autophagy degrades cytoplasmic components via autophagosome-lysosome fusion, regulated by ATG proteins and mTORC1
• Endocytosis (receptor-mediated, pinocytosis, phagocytosis) internalizes material; ESCRT machinery sorts cargo into multivesicular bodies
• Exocytosis delivers material to the surface constitutively or in a regulated, Ca²⁺-triggered manner
• Bioinformatics tools classify trafficking protein families, predict SNARE interactions, identify autophagy substrates via LIR motifs, and reconstruct trafficking pathways from proteomics data