Topic 5b — Protein Sorting

§Introduction

Compartmentalization (Topic 5a) creates a delivery problem: essentially all proteins are made in the cytosol, but each must reach a specific organelle — nucleus, mitochondrion, ER, Golgi, lysosome, plasma membrane, or the outside of the cell. The solution is an address system: proteins carry signal sequences that dedicated machinery reads and acts on. Get the signal wrong and the protein ends up in the wrong place — often with disease consequences.

Three facts organize the whole topic. First, where a protein is made forks the pathway: free ribosomes in the cytosol make proteins for the cytosol, nucleus, mitochondria, and peroxisomes; membrane-bound ribosomes on the rough ER make proteins destined for secretion, the plasma membrane, or the endomembrane system. Second, there are exactly three transport mechanisms — through nuclear pores, across membranes via protein translocators, and by vesicular budding/fusion. Third, the endomembrane route is a quality-controlled assembly line: the ER folds and checks proteins, the Golgi modifies and sorts them, and coated vesicles carry them onward — with the topological rule from Topic 5a (lumen ≡ cell exterior) holding throughout.

This guide follows the lecture closely — mitochondrial vs ER import, membrane-protein topology by start/stop signals, glycosylation-based folding control, and the secretory pathway. It sharpens one point the lecture left imprecise (which coat proteins carry ER→Golgi traffic — it's COPII, not clathrin), restores the SNARE/Rab targeting mechanism behind vesicle "docking," and reinforces a subtle point the lecture got right but common animations get wrong (the cleaved ER signal peptide stays in the translocon).

◎Learning Objectives

1The Sorting Problem & Three Delivery Routes

Bottom line: almost every protein is made in the cytosol but must reach a specific compartment. A protein's destination is encoded in a signal sequence — a short stretch of amino acids that acts as a molecular zip code — read by one of exactly three delivery systems.

Where a protein is made forks its fate

Ribosomes come from one shared cytosolic pool but operate in two locations, and that split predicts destinations:

• Free ribosomes (in the cytosol) make proteins that stay in the cytosol or are imported into the nucleus, mitochondria, or peroxisomes.
• Membrane-bound ribosomes (docked on the rough ER) make proteins destined for secretion, the plasma membrane, or the endomembrane system (ER/Golgi/lysosomes).

The same ribosome can end up either free or ER-bound; what recruits it to the ER is the signal sequence on the protein it happens to be making. After finishing, subunits return to the shared pool.

The three transport mechanisms

1. Through nuclear pores (gated transport). Proteins move between cytosol and nucleus through NPCs, generally folded, via NLS/NES signals (Topic 5a §7).
2. Across a membrane via a protein translocator (transmembrane transport). Proteins are threaded across the membrane of the ER, mitochondria, chloroplasts, or peroxisomes by dedicated translocator complexes — usually while unfolded.
3. By vesicular transport. Proteins already inside the endomembrane system move between compartments packaged inside membrane vesicles that bud from one compartment and fuse with the next — never crossing a membrane (the topological rule from Topic 5a).

2Signal Sequences by Destination

Bottom line: the address is written in the amino acids. Different destinations use chemically distinct signal sequences, and one of them (mitochondrial) is recognized by its 3-D shape, not its exact sequence.

The four you must know

• Nuclear import (NLS): a cluster of positively charged residues — lysine and arginine (roughly five basic residues).
• Nuclear export (NES): a different signal, based on hydrophobic residues (e.g. leucine).
• Mitochondrial matrix: positively charged residues spaced at intervals (every 2–3 amino acids), which fold into an amphipathic α-helix — positive residues cluster on one face, hydrophobic residues on the other.
• ER: a stretch of hydrophobic residues — N-terminal (for soluble/secreted proteins) or internal (for membrane proteins).

Why the mitochondrial signal is read by shape, not sequence

This is a favorite exam point. The mitochondrial matrix signal has its charged residues placed periodically (every few residues) precisely so that when the segment coils into an α-helix, all the positive charges land on one side and the nonpolar residues on the other — an amphipathic helix. The receptor recognizes this charged-on-one-face conformation, not a specific linear sequence. That's why many different mitochondrial signal sequences work: they share a shape, not a consensus string.

3Mitochondrial Import (Post-Translational)

Bottom line: mitochondrial proteins are made completely in the cytosol first, kept unfolded by chaperones, and then threaded across the mitochondrial membranes by translocator complexes. This "finish, then import, unfolded" logic is the opposite of ER import (§4) — and that contrast is heavily tested.

The key players

• Chaperones (hsp70). They bind the newly made protein in the cytosol and keep it unfolded, because a folded protein is too bulky to thread through the translocator. Holding it unfolded costs energy — ATP hydrolysis drives chaperone binding/release.
• TOM complex (Translocase of the Outer Membrane). Its receptor recognizes the signal sequence and forms the channel across the outer membrane.
• TIM complexes (Translocase of the Inner Membrane — TIM23 and TIM22). They move the protein across the inner membrane.
• Signal peptidase. Once inside, it cleaves the matrix-targeting signal, activating the mature protein.

How many membranes it crosses depends on the destination

Two membranes → two possible depths:

• Matrix proteins cross both membranes — through TOM then TIM23 — into the matrix.
• Intermembrane-space proteins cross only the outer membrane (TOM only) and stop between the two membranes.

4ER Import (Co-Translational)

Bottom line: ER proteins are threaded into the ER as they are being translated — the ribosome docks onto a translocator and feeds the growing chain straight through. The signal is recognized twice, and (a favorite trap) the cleaved signal peptide does not float away.

The co-translational sequence

1. Translation begins on a free ribosome; an N-terminal signal sequence emerges.
2. The signal-recognition particle (SRP) binds the signal and pauses translation.
3. SRP docks onto the SRP receptor on the ER surface, delivering the ribosome to the translocon (the Sec61 protein channel).
4. SRP and its receptor are released and recycled; translation resumes, feeding the chain through the translocon into the ER lumen.
5. Signal peptidase cleaves the signal peptide; the finished protein folds in the lumen.

Recognized twice

The ER signal sequence is read two times: first by SRP in the cytosol (to target the ribosome to the ER), and again by the translocon itself, where it acts as a start-transfer signal that opens the channel. Both recognition events are required.

5Membrane Protein Topology: Start & Stop Signals

Bottom line: whether a protein ends up soluble in the lumen or embedded in the membrane — and how many times it crosses, and which way it faces — is set entirely by the arrangement of start-transfer and stop-transfer signals. Master this and you can predict topology from sequence.

Two signal types

• Start-transfer signal. Initiates translocation through the translocon (the ER signal sequence is one). For a soluble protein it's an N-terminal signal that gets cleaved; for a membrane protein it can be internal and become a membrane anchor.
• Stop-transfer signal. A hydrophobic stretch (~20–30 residues) that halts translocation, slips laterally out of the translocon, and stays in the bilayer as a transmembrane α-helix.

How arrangement builds topology

• Soluble/secreted: one cleaved N-terminal start-transfer signal → the whole chain enters the lumen; nothing stays in the membrane.
• Single-pass: a start-transfer signal begins translocation and a stop-transfer signal (or the start signal itself, if internal) anchors one segment in the membrane. Orientation (N-in vs N-out) depends on which signal is where.
• Multipass: alternating start-transfer and stop-transfer signals thread the chain in and out repeatedly — each pair adds membrane crossings, producing 2-, 6-, or 7-pass proteins (e.g. a 7-transmembrane receptor).

Detecting membrane segments

A membrane-spanning segment is ~20–30 hydrophobic residues in an α-helix. Hydropathy plots scan a sequence for such stretches, predicting both whether a protein spans the membrane and how many times. (A short 5-residue hydrophobic patch is too short to span — length matters.)

6ER Quality Control: Glycosylation & Folding

Bottom line: the ER is a folding factory with a built-in inspector. A sugar tag added to new proteins doubles as a folding-status flag, and proteins that fail to fold are pulled back out and destroyed. Nothing leaves the ER until it passes.

N-linked glycosylation

As a protein enters the ER lumen, an enzyme (oligosaccharyltransferase) transfers a preassembled oligosaccharide — carrying three terminal glucose residues — en bloc onto an asparagine (Asn/N) side chain. This is N-linked glycosylation, and about half of ER proteins receive it. Beyond structural roles, the sugar becomes a label for the folding machinery.

The calnexin folding cycle (the checkpoint)

1. The three glucoses are trimmed to one. The chaperone calnexin (and calreticulin) bind the monoglucosylated protein, retaining the still-unfolded protein in the ER and assisting folding.
2. A glucosidase removes the last glucose → the protein is released from calnexin.
3. A sensor enzyme (UGGT) checks folding: if the protein is still misfolded, it re-adds a glucose, sending it back to calnexin for another round. Correctly folded proteins are not re-glucosylated and are exported to the Golgi.

This "re-glucosylate if still wrong" loop is a genuine quality-control cycle: the protein is held and re-checked until it folds or fails.

ERAD — disposal of failures

A protein that cannot fold is recognized as terminally misfolded, retrotranslocated back across the ER membrane into the cytosol, ubiquitinated, and degraded by the proteasome (Topic 6). This is ER-associated degradation (ERAD).

7Vesicular Transport & Coat Proteins

Bottom line: proteins move through the endomembrane system inside coated vesicles. The coat both bends the membrane to bud the vesicle and selects the cargo — and critically, different transport steps use different coats. Knowing which coat goes with which step is prime exam material.

How a coated vesicle forms

Coat proteins assemble on the cytosolic face of the donor membrane, forcing it to curve. Membrane-bending and fission proteins (e.g. dynamin) are recruited to the neck to pinch the vesicle free. Once released, the coat disassembles (it's no longer needed), exposing the membrane for targeting and fusion. Vesicle formation is energy-dependent — it works only in living, metabolically active cells.

8The Golgi & Vesicle Targeting

Bottom line: the Golgi is the cell's processing-and-sorting post office — proteins enter at the cis face, get modified as they progress to the trans face, and are dispatched from the trans-Golgi network. And vesicles don't fuse randomly: Rab proteins and SNAREs guarantee each vesicle fuses only with the correct target.

Golgi organization and cargo flow

• Polarized stack. From the ER side outward: cis → medial → trans cisternae → trans-Golgi network (TGN). Each cisterna holds different enzymes that modify passing proteins (trimming/adding sugars).
• Cisternal maturation model. Rather than proteins hopping between fixed compartments, the cisternae themselves progress: a new cis cisterna forms from ER vesicles, then matures into medial and then trans, carrying its cargo forward — while COPI vesicles recycle the resident enzymes backward. Sorting to final destinations happens at the TGN.

9Sorting to Lysosomes & Secretion

Bottom line: at the trans-Golgi network, proteins split toward their final fates. Lysosomal enzymes carry a specific sugar tag (mannose-6-phosphate); secreted proteins carry no signal at all and leave by default. And secretion itself comes in two modes: always-on or triggered.

Lysosome delivery: the mannose-6-phosphate route

1. Tag. Lysosomal hydrolases receive a mannose-6-phosphate (M6P) mark, added to their N-linked oligosaccharides in the cis-Golgi.
2. Sort. In the TGN, the M6P receptor binds M6P-tagged hydrolases and packages them into clathrin-coated vesicles bound for the endosome.
3. Release. The endosome's low pH makes the hydrolase dissociate from the M6P receptor.
4. Recycle. The receptor returns to the TGN for another round. A phosphate is removed from the M6P so the hydrolase can't rebind the receptor and be dragged back — committing it to the maturing lysosome.

Secretion: constitutive vs regulated

Proteins to be secreted carry no sorting signal directing them out of the Golgi — they follow the default (constitutive) pathway. The two secretory modes diverge at the TGN:

• Constitutive secretion. Operates in all cells, continuously: vesicles from the TGN fuse with the plasma membrane immediately, releasing cargo and adding new membrane lipids/proteins. No trigger, no control.
• Regulated secretion. In specialized cells, selected cargo is concentrated in secretory vesicles that wait near the surface until an extracellular signal triggers fusion. Examples: insulin from pancreatic β-cells (triggered by glucose), neurotransmitters from neurons (triggered by an action potential), and histamine from mast cells (triggered by allergen).

The logic of regulated secretion is demand-matching: insulin is only useful (and only safe) when glucose is high, so it's stored and released on cue — not dumped constitutively.

✦Core Concepts — Rapid Review

Everything most likely to be tested, compressed. If any line isn't instantly familiar, reread that section.

✎End-of-Topic Problems

Every Alberts end-of-chapter problem for this topic, one multiple-choice item each. Click any option to check it instantly — the correct choice is marked Answer as lectured. No submit step; feedback is immediate.