# Protein Quality Control and Organelle Turnover

> A crowded quality-control pathway can mean faster arrival or slower clearance. This lesson follows failed proteins from the endoplasmic reticulum to the proteasome, follows cytoplasmic cargo to the lysosome by autophagy, and uses one mass-balance rule to interpret LC3-II flux. Afterwards you can decide whether a mutant or drug changes folding, disposal or signalling.

Course: Cell Biology (Membranes and compartments). Level: core university (third year). About 22 minutes.
Lesson page (link this one when citing): https://stemhop.com/lessons/quality-control-and-autophagy/

Builds on: Protein Sorting into Compartments · Organelles as Distinct Environments

## What you'll learn

- Explain how BiP, the calnexin and calreticulin cycle and UGGT hold a misfolded protein in the ER, and trace ER-associated degradation from ubiquitination to the proteasome.
- Describe the three mammalian branches of the unfolded protein response (IRE1, PERK, ATF6), the single Ire1 branch of yeast, and why prolonged stress can kill a cell.
- Compare the proteasome and lysosome routes, including macroautophagy, ER-phagy and PINK1 and Parkin mitophagy.
- Calculate autophagic flux from LC3-II blots with and without a lysosomal inhibitor, and use the result to say whether a treatment changes induction or degradation.

**Starve a yeast cell of nitrogen and it starts to digest pieces of its own cytoplasm. A mutant that cannot do this loses viability faster than normal yeast during the starvation. Why?**

The first film followed cargo to two exits and ended on two questions: how fast does material arrive, and how fast does it leave? What it left out is how to measure those rates, which the worked problem below does with numbers, and how the ER and a damaged mitochondrion are handled, which come after.

In 1993, Miki Tsukada and Yoshinori Ohsumi found that the autophagy-defective yeast mutant *apg1* failed to induce protein degradation in the vacuole, the yeast counterpart of a lysosome, and lost viability faster during nitrogen starvation. The experiment supports a role in survival without establishing a sole cause.

> **The one idea to carry through this lesson.** For a defined cellular pool over a defined interval, its change equals arrival minus clearance. A high level measured at one time therefore cannot tell whether arrival increased or clearance slowed. Temporarily blocking a clearance route and measuring the additional accumulation can estimate flux through that route, if the block is complete and does not itself change arrival over that interval.

## Physical constraints on the two exits

The two exits have different physical jobs because they admit cargo differently. The **26S proteasome**, found in the cytosol and nucleus, has a hollow 20S core whose active sites face inward. Its regulatory caps recognise ubiquitin chains, remove ubiquitin for reuse, and use ATP to unfold and thread an engaged protein into the core. The tagging system is covered in [ubiquitin and proteolysis](https://stemhop.com/lessons/ubiquitin-and-proteolysis/), and the chemistry in [protein turnover](https://stemhop.com/lessons/protein-turnover-amino-acid-catabolism/).

In a **lysosome**, a V-ATPase—an ATP-powered proton pump—acidifies the lumen so digestive enzymes called hydrolases work well (see [organelles as environments](https://stemhop.com/lessons/organelles-as-environments/)). Material also arrives from outside by endocytosis. In macroautophagy, the main form of autophagy, a cup-shaped membrane grows around a portion of cytoplasm and closes into an autophagosome bounded by two membranes. The autophagosome fuses with a lysosome, and the inner membrane and its contents are digested; the building blocks return to the cytosol. Starvation can raise this through nutrient-sensing kinases such as mTOR (see [mTOR and AMPK](https://stemhop.com/lessons/mtor-and-ampk/)). Ohsumi’s yeast genes, later renamed *ATG* genes, encode much of the machinery, including a system that attaches LC3 (Atg8 in yeast) to phosphatidylethanolamine; this lipid-linked **[LC3-II](https://stemhop.com/methods/autophagic-flux-assay/)** coats autophagosome membranes, and inner-membrane copies are destroyed in the lysosome.

The first film used the simplest case, single proteins to one exit and bulk cargo to the other; cargo size alone does not decide fate, and Going deeper gives the overlap.

## Read a flow, not a still picture

Arrival minus clearance becomes useful when an experiment turns one of those rates down. Counting LC3 dots or measuring one [LC3-II](https://stemhop.com/methods/autophagic-flux-assay/) band still cannot distinguish faster carrier formation from slower destruction.

**Worked problem: autophagic flux.** Cultured cells are immunoblotted for LC3-II, with each band normalized to actin and control set to 1.0. Bafilomycin A1 inhibits the V-ATPase, suppressing lysosomal degradation under these assay conditions. It is added for the last few hours. Assume a saturating dose, high enough for a complete block, plus linear LC3-II accumulation and no change in carrier formation over that interval.

| Condition | Without bafilomycin | With bafilomycin |
| --- | --- | --- |
| Control | 1.0 | 2.0 |
| Starved | 1.5 | 4.5 |
| Drug D | 3.0 | 3.2 |

> flux over the interval = LC3-II with bafilomycin − LC3-II without bafilomycin

**Predict.** Drug D has the highest LC3-II level without bafilomycin. Does it also have the highest flux? Calculate all three differences before opening the answer.

> **Check your prediction.** No. Control lysosome-dependent turnover is 2.0 − 1.0 = 1.0. Starved turnover is 4.5 − 1.5 = 3.0. Drug D turnover is 3.2 − 3.0 = 0.2. Drug D builds the largest pool before inhibitor but clears only one fifth as much LC3-II as control over the interval. Its pattern points to a block at fusion or degradation, not induction; these lanes cannot say which step fails. Starvation gives three times the control increment, which fits more autophagosomes being made.

*Interactive on the lesson page:* LC3-II flux blot: the same band, two stories

LC3-II marks the carrier, not its cargo, so confirm the interpretation with cargo. p62, a receptor destroyed with its cargo, often falls during short starvation and should build up with Drug D. Its level also depends on production, and long starvation can raise p62 production, so read it alongside LC3-II. A tandem red-and-green LC3 reporter adds location: lysosomal acidity quenches green fluorescence, so normal delivery gives red-only dots, while a block before fusion leaves red-and-green dots.

For a real LC3-II turnover assay, use the inhibitor at a saturating dose for a short enough time that accumulation stays in its linear range and secondary effects are limited. The estimate remains conditional on the inhibitor not changing carrier formation; Going deeper gives one way that assumption can fail.

The same two questions apply upstream, where proteins are made rather than destroyed. The second film picks up here: what the ER does when unfinished proteins arrive faster than it can fold or remove them.

**Film transcript: When the ER Falls Behind**

### Film transcript: When the ER Falls Behind

Chapter 1 asked of any pile: how fast does it arrive, and how fast does it leave? Now ask that inside the ER. What happens when unfinished proteins arrive faster than the ER can fold or remove them? Many secreted and membrane protein chains enter the ER through a channel called the translocon. Chaperones then help them fold. BiP grips exposed hydrophobic stretches.

Lectin chaperones read sugar tags on many glycoproteins, supporting repeated folding attempts. Folded clients can leave for the Golgi. Some persistently misfolded clients enter ER-associated degradation, or ERAD. During return toward the cytosol, yeast Hrd1 adds a ubiquitin chain.

Cdc48 extracts the tagged client for destruction by the proteasome. Now count the flows. Arrival is ER-destined chains entering. Clearance is folding and export, plus removal by ERAD. When chains arrive faster than both exits can take them, unfolded proteins build up. The ER senses that stress and raises an alarm. This alarm is the unfolded protein response. Look at the equation. Which flows could the response change to bring the load down?

It can change both sides. In mammals, a sensor called PERK slows the making of most new proteins, so less arrives. The response also increases folding helpers and removal machinery. Less in and more out let the load fall. Mammals have three branches.

IRE1 splices the XBP1 message. PERK slows translation. ATF6 travels to the Golgi, where proteases release its transcription factor domain. Yeast relies on one ER-stress sensor, Ire1. The response buys time, but has limits. Persistent stress can also drive cell death. The ER can retry clients, remove them, slow arrival, or raise capacity. These moves rebalance arrival and clearance. When folding falls behind, reduce arrival and expand capacity.

From the film

- Many secreted and membrane protein chains enter the ER and fold with chaperones such as BiP; folded clients can leave for the Golgi.
- A persistently misfolded client can enter ERAD: it is moved toward the cytosol, ubiquitinated, extracted and destroyed by the proteasome.
- When unfolded proteins arrive faster than the ER can clear them, the unfolded protein response slows arrival and raises capacity.
- Mammals have three UPR branches, IRE1, PERK and ATF6; budding yeast relies on one ER-stress sensor, Ire1.
- If ER stress is not relieved, the response can also drive cell death.

## The ER: retry, remove, raise capacity

The same moving-pool logic applies upstream: the ER can retry a folding client, remove it, slow new arrival or raise capacity. Its lumen is the folding compartment for secreted and membrane proteins (see [protein sorting](https://stemhop.com/lessons/protein-sorting/)). A new chain enters through the translocon, a protein-conducting channel. BiP, an Hsp70-family chaperone, binds exposed hydrophobic stretches and helps it fold. What the second film left out is the sugar code its other chaperones read. Many entering proteins receive an **N-linked glycan**, a branched sugar tree attached to asparagine with three glucose residues at its tip. Glucosidases remove two. When one remains, the **lectin chaperones** calnexin, which is membrane-bound, and calreticulin, which is soluble, bind that sugar and hold the client for another folding attempt. Glucosidase II removes the last glucose and releases the protein.

Now comes the test. UGGT recognizes features of a still-non-native client and adds one glucose back. That glucose is not a permanent label for a bad protein; it is a ticket for another round with calnexin or calreticulin. A client that has reached its native state is not reglucosylated and can leave for the Golgi.

**Predict.** A client is released from calnexin but still exposes non-native features. What happens to its glycan, and where does the client go next?

> **Check your prediction.** UGGT adds one glucose back. The client can bind calnexin or calreticulin again and remains in the ER for another folding attempt. The glucose does not send it directly to the proteasome.

A client that cycles too long can be diverted to **ER-associated degradation** (**ERAD**). In the standard account, trimming mannose residues from its glycan helps mark it. ERAD must move the failed client from the ER lumen toward a proteasome in the cytosol. The return trip across the ER membrane is called **retrotranslocation**. Hrd1 is a yeast ER-membrane E3 ligase that adds a ubiquitin chain as the client moves toward the cytosol. The ATP-powered extraction machine Cdc48—p97 or VCP in mammals—pulls out the ubiquitinated client for delivery to the proteasome. The exact channel and the limits of the yeast reconstitution are in Going deeper.

*Figure (AI-generated illustration).* As a failed ER client returns toward the cytosol, it is ubiquitinated, pulled out by Cdc48 and destroyed; the dashed Hrd1 passage is only proposed. Image: an ER membrane with a misfolded protein moving from the lumen through a dashed, questioned Hrd1 block, pulled out by a Cdc48 ring and fed into a proteasome barrel, with a small inset of a lipid vesicle used in the reconstitution experiment.

**Predict.** In a cycloheximide chase, the drug cycloheximide stops protein synthesis so disappearance of an existing protein can be followed. An ERAD substrate falls by half every 20 minutes in wild-type yeast. What remains after 60 minutes, and what should happen if Hrd1 is required for its degradation?

> **Check your prediction.** Three half-lives leave one eighth, or 12.5%, in wild type. If Hrd1 is required for degradation, the band should fade much more slowly in the mutant. The stable band does not show that folding improved; a chase measures disappearance after synthesis stops.

When failures arrive faster than the ER can handle them, the **unfolded protein response** (**UPR**) raises folding and disposal machinery and, through PERK in mammals, slows the incoming load of new proteins. If stress is not relieved, the response can also drive cell death; how it does so is covered in [cell death](https://stemhop.com/lessons/cell-death/). In budding yeast, the single ER-stress sensor Ire1 cuts an intron from *HAC1* mRNA so the Hac1 transcription factor can be made. The three mammalian branches:

- **IRE1** removes a 26-nucleotide intron from *XBP1* mRNA, as shown for mouse XBP1, producing the active transcription factor XBP1s.
- **PERK** phosphorylates eIF2α on serine 51, slowing most translation while permitting preferential translation of ATF4 (see [stress signals](https://stemhop.com/lessons/stress-signals/)).
- **ATF6** travels to the Golgi, where Site-1 and Site-2 proteases cut it to release a transcription factor.

*Figure (AI-generated illustration).* Yeast relies on one ER stress sensor, Ire1, while mammals spread the alarm across three overlapping branches. Image: an ER membrane with three sensors; IRE1 removes an intron from XBP1 mRNA, PERK phosphorylates eIF2alpha, and ATF6 travels to the Golgi to be cut, with a small inset of the single yeast sensor Ire1 acting on HAC1 mRNA.

## Selective autophagy: choosing a mitochondrion

The first film wrapped a damaged mitochondrion without asking how it was picked; selective autophagy is how that cargo is chosen. In **selective autophagy**, receptors link chosen cargo to the LC3 family. **Mitophagy** is selective autophagy of mitochondria.

One well-tested route begins when a mitochondrion loses its membrane potential, the voltage across its inner membrane. In cultured cells, the kinase PINK1 is broken down on healthy, polarized mitochondria but builds up on depolarized ones and recruits Parkin, a ubiquitin ligase. Parkin adds ubiquitin to outer-membrane proteins, and ubiquitin-binding receptors recruit the autophagy machinery that engulfs the mitochondrion.

*Figure (AI-generated illustration).* In cultured cells, a depolarized mitochondrion is chosen for autophagy once PINK1 builds up, Parkin arrives and ubiquitin coats its surface. Image: two mitochondria side by side; on the healthy one PINK1 is cut into fragments and Parkin floats free, while on the depolarized one PINK1 builds up, Parkin arrives and ubiquitin tags the surface before a membrane cup engulfs it, with a questioned side arrow for other routes.

**Predict.** In mouse fibroblasts lacking Atg5, a protein needed for conventional autophagosome building, Parkin still gathers on depolarized mitochondria, but the mitochondria remain. Which step needs Atg5: Parkin recruitment, mitochondrial clearance, or both?

> **Check your prediction.** Clearance needs Atg5, but recruitment does not. Parkin still arrives, so tagging is upstream of conventional autophagosome building. The mitochondria remain, so a tag alone does not remove the cargo.

PINK1 and Parkin are a well-proven route for damaged mitochondria in cultured cells, not the only route in an animal. Going deeper adds the amplification step and the experimental limits behind that conclusion.

### Design problem

The ER’s two systems from the second film, a disposal route and a stress alarm, can fail separately, and telling which one broke is a design problem.

A mutant yeast strain accumulates a misfolded ER membrane protein. Is ER-associated degradation (ERAD) blocked, or is the unfolded protein response (UPR) broken? Design experiments that tell the two apart, and write down what each model predicts before you run them.

**Predict.** The substrate is stable in a cycloheximide chase, but tunicamycin, which blocks N-glycosylation, still triggers normal *HAC1* mRNA splicing, a yeast UPR readout. Which model is favoured, and why?

> **Check your prediction.** An ERAD block is favoured. The stable substrate shows slow destruction, while normal *HAC1* splicing shows that the Ire1 step of the UPR can still signal. The pair of results is stronger than either alone.

> **One way to do it.** An ERAD defect predicts normal synthesis, slow destruction and an intact UPR that may already be active. A UPR defect cannot raise folding and disposal capacity, so buildup may emerge mainly under stress while the alarm stays silent.
>
> 1. **Cycloheximide chase.** Stop synthesis and blot for the substrate at 0, 30, 60 and 90 minutes. A stable band in unstressed cells points to an ERAD block. Near-normal decay that slows only under stress points to the UPR. The chase alone cannot settle it: yeast need an intact UPR for efficient ERAD, so a UPR mutant can also degrade the substrate slowly. Step 3 decides.
> 2. **Ubiquitination.** Immunoprecipitate the substrate and blot for ubiquitin. No ubiquitin suggests a defect at or before tagging, such as loss of the E3 (Hrd1); ubiquitinated substrate stuck in membranes suggests a later step, such as extraction by Cdc48.
> 3. **UPR reporter.** Treat with tunicamycin and measure *HAC1* mRNA splicing by RT-PCR, or use a reporter that produces a measurable signal when a UPR regulatory element turns on. No splicing means the UPR is broken; normal splicing means at least the Ire1 step works.
> 4. **Complementation and rescue.** Transform the mutant with plasmids carrying candidate genes such as *HRD1* or *IRE1*. The gene that restores normal degradation (or normal splicing) is the lead candidate. Rescue alone is not proof, because extra copies of one gene can suppress a defect in another: overexpressed *APG1* rescues an *apg13* mutant. Sequencing the mutant allele, or remaking the mutation in a clean strain, shows that this mutation, not a second hit elsewhere, causes the phenotype.
>
> If the chase, the ubiquitin blot and the rescue all point to *HRD1*, and splicing is normal, ERAD is blocked. If splicing fails and *IRE1* rescues, the UPR is broken.

## Going deeper

These refinements show where the picture is client-dependent, where routes overlap and where an assay can mislead.

### UPR discovery and branch overlap

Jeffery Cox, Caroline Shamu and Peter Walter reported budding yeast Ire1 in 1993, in parallel with Kazutoshi Mori and colleagues. Ire1 spans the ER membrane and carries a kinase and a ribonuclease (RNase) on its cytosolic side; its cut in *HAC1* mRNA occurs outside the spliceosome.

Mammalian branches overlap but can be separated experimentally. In mouse fibroblasts lacking IRE1α, ATF6 was still cleaved, and tunicamycin, which blocks N-glycosylation, still induced BiP mRNA. One broken branch does not silence the response.

### The glycan cycle is client-dependent

The calnexin and calreticulin cycle has been followed in detail for particular clients. In HEK293-derived cells it was followed for the IGF-1 receptor, and the two human UGGT enzymes preferred different substrates. The retry loop is therefore a useful core model, not a claim that every client or UGGT behaves identically.

### Route overlap and flux limits

The proteasome cannot admit an intact organelle or large aggregate through its narrow pore; an individual substrate must be engaged and unfolded. Lysosomal routes can accept bulk cytoplasm and organelles, but cargo is not assigned by size alone, and a single protein does not automatically imply proteasomal degradation.

A lysosomal inhibitor can itself change carrier formation, for example by altering lysosomal mTORC1 signalling. That would break the worked problem’s assumption that arrival remains unchanged during the block.

### The Hrd1 channel: reconstitution and structure

In 2016 Ryan Baldridge and Tom Rapoport put purified yeast Hrd1 and a membrane-anchored misfolded substrate into synthetic lipid vesicles. Hrd1 ubiquitinating itself triggered movement of the substrate across the membrane, and the results suggest that Hrd1 forms a ubiquitin-gated channel, with Cdc48 completing extraction. That is a reconstitution with yeast proteins. In 2020 a cryo-EM structure of the full yeast Hrd1 complex, from the same laboratory, refined the picture for luminal substrates: Hrd1 and its partner Der1 each form a half-channel, and the two face each other in a thinned patch of membrane. Whether every ER-associated degradation (ERAD) substrate in every cell crosses this way is not settled.

### ER-phagy receptors

ER-phagy, the selective digestion of parts of the ER, uses several receptors. FAM134B, reported in 2015, was the first found in mammals. Another receptor, CCPG1, binds mammalian ATG8-family proteins and the autophagy protein FIP200 through distinct motifs. Mice with reduced *Ccpg1* accumulate insoluble protein in the ER of pancreatic acinar cells.

### PINK1 and Parkin: amplification and limits

PINK1 phosphorylates ubiquitin, which activates Parkin so that more ubiquitin accumulates on a damaged mitochondrial surface. Receptors such as optineurin and NDP52 bind the tagged mitochondrion and recruit autophagy machinery.

Derek Narendra, Atsushi Tanaka, Der-Fen Suen and Richard Youle reported Parkin movement to depolarized mitochondria in cultured cells in 2008. In 2010 the same laboratory reported that PINK1 was necessary and sufficient for Parkin recruitment in the cells tested. These experiments used uncouplers such as CCCP, which depolarize every mitochondrion at once, and largely overexpressed Parkin. Yet in reporter mice lacking *Pink1*, basal mitophagy in the sampled tissues was mostly preserved, with pancreatic islets the exception.

### Counting mitochondria after Parkin

In HeLa cells exposed to CCCP for 48 hours, [electron micrographs](https://stemhop.com/methods/electron-microscopy/) showed 0.62 mitochondrial profiles per µm² of cytoplasm without YFP-Parkin and 0.074 with it.

> reduction = (0.62 − 0.074) / 0.62 = 0.546 / 0.62 = 0.881, so 88.1% lower

That compares mean two-dimensional densities in two groups of cells at one time. It is not a rate.

---

From STEMhop (https://stemhop.com), free lessons by Christopher Deppmann. Lesson text © the authors. Practice questions: https://stemhop.com/lessons/quality-control-and-autophagy/
