ER-stress signaling and Alzheimer's proteins adjust the quality of human protein synthesis

⚕️

This is an AI-generated explanation of a preprint that has not been peer-reviewed. It is not medical advice. Do not make health decisions based on this content. Read full disclaimer

Imagine your body is a massive, bustling factory. Inside this factory, there are millions of tiny machines called ribosomes. Their job is to read blueprints (DNA) and build proteins, which are the bricks, tools, and workers that keep your body running.

Usually, these machines are pretty good, but they aren't perfect. Sometimes, they make a typo. They might put a "Z" where an "S" should be, or build a brick that's the wrong shape. In the world of biology, these mistakes create misfolded proteins. If too many of these "defective bricks" pile up, they can clump together, clog the factory, and cause the whole system to break down. This is what happens in diseases like Alzheimer's.

For a long time, scientists thought that as we get older, these machines just get "worn out" and start making more mistakes. But this new study from Ulm University tells a very different, and surprisingly hopeful, story about how human factories work compared to mouse factories.

Here is the breakdown of their discovery, using some everyday analogies:

1. The Human "Slow and Steady" Strategy

When the researchers looked at human cells from young people versus old people, they found something surprising.

The Mouse Factory: As mice age, their machines get sloppy. They make more mistakes, and the factory gets messy. This leads to early aging and signs of Alzheimer's.
The Human Factory: As humans age, our machines actually get more careful. They slow down their production line to ensure that every single brick is perfect. They make fewer mistakes, even though they are building fewer total bricks.

The Analogy: Imagine a young human worker who is a speed demon, building 100 widgets an hour but making 10 of them defective. As they age, they decide to slow down to 50 widgets an hour, but now they only make 1 defective one. They are trading quantity for quality.

2. The "Stress Alarm" (ER Stress)

Why do human cells do this? The study found that as we age, our cells sense a little bit of "stress" inside a specific part of the factory called the Endoplasmic Reticulum (ER). Think of the ER as the quality control department.

When the quality control department gets overwhelmed or senses that things aren't right, it sounds an alarm (called ER-stress signaling). This alarm tells the ribosomes: "Stop! Slow down! Don't rush! We need to make sure everything is perfect before we ship it out."

This alarm forces the cells to:

Produce fewer proteins overall.
Make those proteins with much higher accuracy.

3. The Alzheimer's Connection (The "APP" and "PSEN1" Duo)

Here is where it gets really interesting. The study found that two proteins famous for being linked to Alzheimer's disease—APP and PSEN1—are actually the managers of this quality control system in humans.

In Young Humans: These managers are active. They help keep the production line running smoothly but with a bit of "youthful" speed, meaning a few mistakes happen.
In Aging Humans: The system adjusts. The managers (APP and PSEN1) interact with the quality control alarm to tighten the screws. They help the cell switch to "High-Fidelity Mode."
The Twist: If you mess with these managers (by removing them or changing them, like in Alzheimer's mutations), the factory loses its ability to switch to "High-Fidelity Mode." It might start making too many mistakes again, or the quality control system breaks down, leading to the toxic clumps seen in Alzheimer's.

The Analogy: Think of APP and PSEN1 as the foremen on the construction site. In a healthy aging human, the foremen notice the site is getting a bit chaotic and tell the workers, "Hey, let's double-check every measurement." In Alzheimer's, the foremen are either missing or broken, so the workers keep rushing and making mistakes, causing the building to collapse.

4. Why Mice Are Different

The researchers tried this same experiment on mice, and it didn't work. Mouse cells didn't get "smarter" or more careful as they aged; they just got worse. This suggests that humans have evolved a special survival trick that mice don't have. Because humans live much longer, we needed a way to keep our protein factories running cleanly for decades. We evolved to prioritize accuracy over speed as we age.

The Big Takeaway

This paper changes how we look at aging and Alzheimer's.

Old View: Aging is just a decline where everything breaks and gets messy.
New View: Aging in humans is a strategic adjustment. Our bodies are actively trying to protect us by slowing down production to prevent errors.

However, in Alzheimer's disease, this protective system seems to fail. The "managers" (APP/PSEN1) stop doing their job, and the factory reverts to a chaotic state, producing the toxic clumps that damage the brain.

In simple terms: Our bodies are trying to be perfect as we get older, but in Alzheimer's, the system that keeps us perfect gets hijacked. Understanding this "quality control" switch could help scientists design new drugs to help the brain stay accurate and healthy for longer.

1. Problem Statement

Proteostasis (the balance of protein synthesis, maintenance, and degradation) is critical for cellular health but is disrupted in neurodegenerative disorders like Alzheimer's Disease (AD) and during human aging. While ribosomal translation is the most error-prone step in gene expression (error rates of $10^{-4}$ to $10^{-3}$ ), it remains unclear how translational fidelity is regulated during human aging and whether AD-related proteins play a role in this process. Previous studies in rodents suggested that increased ribosomal error rates drive premature aging and AD-like phenotypes, but data on human aging is sparse and contradictory. This study aims to determine:

How ribosomal error rates change during human aging (in vivo and in vitro).
Whether Endoplasmic Reticulum (ER) stress signaling regulates these error rates.
Whether Alzheimer's-associated proteins, specifically Amyloid-beta Precursor Protein (APP) and Presenilin 1 (PSEN1), modulate translational fidelity.

2. Methodology

The researchers utilized a multi-faceted approach combining human and mouse cell models, pharmacological interventions, and omics technologies:

Cell Models:
- Human: Primary dermal fibroblasts from young (<1 year) and old (74–84 years) donors, as well as in vitro aged fibroblasts (serial passage to senescence).
- Mouse: Fibroblasts from young (12 weeks) and old (>100 weeks) C57BL/6J and NMRI mice.
- Engineered Lines: HEK293T cells and immortalized fibroblasts used for CRISPR/Cas9 knockouts (KO) and shRNA-mediated knockdown (KD) of APP, PSEN1, BiP, and Notch.
Translational Fidelity Assay:
- Used a NanoLuciferase reporter system containing inactivating mutations (near-cognate mutations or stop codons). Ribosomal errors (misincorporation or readthrough) restore NanoLuciferase activity, allowing quantification of the error rate relative to a Firefly luciferase control.
Protein Synthesis & Stress Markers:
- Synthesis Rates: Measured via metabolic labeling with $^{35}$ S-methionine and flow cytometry using o-propargyl-puromycin (OPP).
- ER Stress: Assessed via Western blot and qPCR for markers including p-eIF2 $\alpha$ , CHOP, ATF4, PERK, ATF6, IRE1 $\alpha$ , and BiP/GRP78.
- Proteome Stability: Heat shock followed by centrifugation to quantify aggregation-prone proteins.
Pharmacological Interventions:
- ER Stress Induction: Tunicamycin and Thapsigargin.
- UPR Inhibition: Specific inhibitors for PERK (GSK2606414), ATF6 (Ceapin A7), and IRE1 $\alpha$ (4 $\mu$ 8C).
- $\gamma$ -Secretase Inhibition: DAPT, Compound E, MRK560 (to block APP cleavage by PSEN1).
- Translation Inhibition: Low-dose Cycloheximide.
Omics Analysis:
- Transcriptomics: Nanopore long-read sequencing (Nanopore) to analyze gene expression changes.
- Proteomics: Mass spectrometry (LC-MS/MS) for protein quantification.
- Data Integration: Comparative analysis of Gene Ontology (GO) terms between aged cells and ER-stressed young cells.

3. Key Results

A. Ribosomal Error Rates Decrease in Human Aging (Contrary to Mice)

Human Cells: Both in vivo aged (old donors) and in vitro aged (late passage) human fibroblasts exhibited a decreased ribosomal error rate (increased fidelity) compared to young cells.
Mouse Cells: In contrast, mouse fibroblasts showed no significant decrease in error rates with age; in some strains, there was a non-significant trend toward increased errors.
Protein Synthesis: Overall protein synthesis rates declined significantly in aged human cells, accompanied by increased phosphorylation of eIF2 $\alpha$ (p-eIF2 $\alpha$ ), a hallmark of ER stress and the Integrated Stress Response (ISR).

B. ER-Stress Signaling Regulates Fidelity via PERK

Inducing ER stress in young human cells (via Tunicamycin) mimicked the aging phenotype: it reduced the ribosomal error rate and total protein synthesis.
Mechanism: The reduction in error rate was dependent on the PERK arm of the Unfolded Protein Response (UPR). Inhibiting PERK prevented the Tunicamycin-induced increase in fidelity.
Elongation Dynamics: Interestingly, while initiation was suppressed, translation elongation speed actually increased under ER stress, suggesting a reallocation of resources (e.g., correct tRNAs) to improve accuracy rather than speed.
Omics Convergence: Transcriptomic and proteomic analyses revealed that ER-stressed young cells clustered with aged cells, sharing 50% of enriched Gene Ontology terms (e.g., ribosome biogenesis, RNA splicing), confirming that ER stress induces an "aged" transcriptional/proteomic state.

C. APP and PSEN1 Modulate Translation Quality

APP Knockdown: Reducing APP levels in human fibroblasts and HEK cells decreased the ribosomal error rate (increased fidelity) and reduced proteome instability. This was associated with increased p-eIF2 $\alpha$ .
PSEN1 Knockdown: Reducing PSEN1 (the $\gamma$ -secretase subunit) increased the error rate and total protein synthesis, creating a "youthful" but unstable proteome.
Mutant APP: Overexpression of wild-type (WT) APP increased error rates, whereas the "Swedish" mutation (associated with early-onset AD) did not, suggesting the mutation disrupts APP's regulatory function.
Mechanism: APP interacts with the chaperone BiP/GRP78. PSEN1 cleavage of APP appears to reduce BiP stabilization. When PSEN1 is inhibited or knocked down, APP accumulates, stabilizing BiP, which suppresses PERK signaling, leading to high translation rates but low fidelity. Conversely, APP reduction triggers PERK activation, lowering translation but increasing accuracy.

D. Species Specificity and Pharmacological Potential

Human Specificity: The regulatory effects of APP and PSEN1 on ribosomal error rates were exclusive to human cells; manipulating these proteins in mouse cells had no effect on translational fidelity.
Pharmacological Modulation:
- PERK inhibition reversed the fidelity changes caused by APP knockdown.
- $\gamma$ -secretase inhibitors increased error rates in human cells (mimicking PSEN1 KD), but this effect was abolished if APP was also knocked down, confirming APP is the essential mediator.
- Cycloheximide: Nanomolar doses of the translation inhibitor cycloheximide significantly reduced ribosomal error rates without impairing overall protein synthesis, suggesting a potential therapeutic avenue to enhance proteostasis.

4. Significance and Conclusion

Redefining Aging: The study challenges the notion that aging is solely characterized by a loss of fidelity. Instead, it proposes that human aging involves a compensatory mechanism where cells sacrifice protein synthesis quantity for quality (fidelity) to prevent the accumulation of toxic aggregates, mediated by ER stress signaling.
Novel Role for Alzheimer's Proteins: APP and PSEN1 are identified not just as sources of amyloid toxicity, but as physiological regulators of translational fidelity in human cells. Their dysregulation (e.g., via mutation or cleavage imbalance) may lead to proteome instability, a precursor to neurodegeneration.
Species Divergence: The findings highlight a critical divergence between mice and humans in aging mechanisms. Mouse models may not fully recapitulate the translational fidelity adjustments seen in long-lived human species, cautioning against direct extrapolation of rodent data to human AD pathology.
Therapeutic Implications: The discovery that low-dose cycloheximide or modulation of the PERK pathway can enhance translational fidelity offers a new strategy for treating age-related proteostasis collapse and Alzheimer's disease.

In summary, this paper establishes that human aging involves a PERK-mediated adaptation to increase ribosomal accuracy at the cost of synthesis speed, a process tightly regulated by APP and PSEN1, providing a new mechanistic link between ER stress, Alzheimer's pathology, and protein quality control.