ER stress occurs when the endoplasmic reticulum, the cell’s main protein-folding factory, gets overwhelmed by misfolded or unfolded proteins and can no longer keep up with demand. Cells respond by activating a coordinated alarm-and-repair system called the unfolded protein response (UPR), which operates through three main signaling branches. Whether this response saves the cell or kills it depends on the severity and duration of the stress, a distinction that shapes diseases from diabetes to neurodegeneration to cancer.
What the Endoplasmic Reticulum Actually Does
The ER is a sprawling network of membranes inside nearly every cell in your body, and roughly a third of all the proteins a cell produces pass through it. Proteins enter the ER in an unfolded state and need the help of resident chaperones and folding catalysts to reach their correct three-dimensional shape.1PubMed Central. The endoplasmic reticulum and the unfolded protein response That folding process involves several modifications: a signal peptide gets clipped off, sugar chains are attached (a process called N-linked glycosylation), and disulfide bonds form to stabilize the protein’s structure.2PubMed Central. Protein folding in the endoplasmic reticulum The ER also stores calcium, manufactures lipids, and communicates directly with other organelles. So when the folding machinery gets backed up, the ripple effects extend well beyond one pile of bad proteins.
Anything that increases the protein load or disrupts folding conditions can trigger ER stress. Nutrient shortages, low oxygen, viral infections, genetic mutations, and even normal developmental demands like a cell gearing up to secrete large amounts of a single protein can all push the ER past its capacity. The cell’s first move is always the same: activate the UPR and try to restore balance.
How Cells Sense the Problem
The UPR relies on three sensor proteins embedded in the ER membrane. Each one detects trouble in a similar way but triggers a different downstream response. All three are held in an inactive state by a chaperone called BiP, which normally binds to their ER-facing domains and keeps them quiet. When misfolded proteins accumulate, BiP releases from the sensors and binds to the misfolded proteins instead, freeing the sensors to activate.3PubMed Central. UPR proteins IRE1 and PERK switch BiP from chaperone to ER stress sensor Think of BiP as a guard dog that abandons its post at the door because a fire broke out in the kitchen.
PERK and the Translation Brake
Once BiP lets go, PERK pairs up with another copy of itself and switches on its enzyme activity, which targets a protein called eIF2α. When eIF2α gets modified by PERK, the cell’s overall protein production slows down sharply. This is a logical first response: if the ER can’t fold what it already has, stop sending more.4PubMed Central. The eIF2 kinase PERK and the integrated stress response facilitate activation of ATF6 during endoplasmic reticulum stress At the same time, a small number of stress-related genes actually get translated more efficiently under these conditions. One of them, ATF4, switches on genes involved in amino acid metabolism, antioxidant defense, and autophagy, helping the cell cope.5PubMed Central. The Role of the PERK/eIF2α/ATF4/CHOP Signaling Pathway in Tumor Progression During Endoplasmic Reticulum Stress
IRE1 and Its Two Jobs
IRE1 is the oldest and most evolutionarily conserved UPR sensor. It has a dual function: one enzymatic activity that cuts RNA in a very precise way, and another that degrades RNA more broadly. The precise cut is reserved for one specific target, a messenger RNA called XBP1. IRE1 splices out a small segment of the XBP1 message, which changes the protein it encodes into a powerful transcription factor that ramps up ER folding capacity and quality-control machinery. The broader RNA-cutting activity, called RIDD, chews up other messenger RNAs heading for the ER, reducing the incoming protein load from a different angle. Research has shown that these two activities rely on different structural arrangements of IRE1 molecules: XBP1 splicing requires multiple IRE1 copies working cooperatively in clusters, while RIDD can be performed by a single IRE1 molecule acting alone.6PubMed Central. Ire1 has distinct catalytic mechanisms for XBP1/HAC1 splicing and RIDD
ATF6 and the Golgi Shortcut
ATF6 takes a completely different route. When stress hits, ATF6 physically travels from the ER to the Golgi apparatus, where two enzymes called S1P and S2P cut it in sequence.7PubMed. Dependence of site-2 protease cleavage of ATF6 on prior site-1 protease digestion is determined by the size of the luminal domain of ATF6 This releases ATF6’s business end, a fragment that travels to the nucleus and turns on genes for ER chaperones and other folding helpers.8PubMed. ER stress signaling by regulated proteolysis of ATF6 In effect, ATF6 boosts the ER’s workforce rather than reducing its workload. Between PERK slowing down new arrivals, IRE1 clearing the backlog, and ATF6 hiring more staff, the three branches cover the problem from all sides.
Disposing of the Wreckage
Slowing protein production and boosting chaperones is only half the battle. The cell also needs to get rid of the misfolded proteins that have already piled up. Two major disposal systems handle this.
ERAD Pulls Defective Proteins Out for Recycling
ER-associated degradation, or ERAD, recognizes misfolded proteins inside the ER and threads them back through the membrane into the cytoplasm, a process called retrotranslocation.9PubMed Central. The cryo-EM structure of the human ERAD retrotranslocation complex Once on the cytoplasmic side, the proteins are tagged with a small marker called ubiquitin and fed into the proteasome, a molecular shredder that breaks them into amino acids for reuse.10PubMed Central. The recognition and retrotranslocation of misfolded proteins from the endoplasmic reticulum ERAD is particularly good at handling individual faulty proteins one at a time. But if misfolded proteins form large, proteasome-resistant clumps, the cell needs a bigger solution.
ER-phagy Removes Entire ER Sections
That bigger solution is ER-phagy, a selective form of autophagy in which the cell wraps whole sections of the ER in membranes and delivers them to lysosomes for digestion. Special receptor proteins on the ER surface mark the doomed segments, linking them to the autophagy machinery via binding sites on their cytoplasmic side.11PubMed Central. ER-phagy: mechanisms, regulation, and diseases connected to the lysosomal clearance of the endoplasmic reticulum ER-phagy handles situations ERAD cannot: clearing away ER regions containing aggregated proteins, damaged lipids, or simply excess membrane that expanded during the stress response and needs to shrink back to normal size afterward.12PubMed Central. Endoplasmic reticulum turnover: ER-phagy and other flavors in selective and non-selective ER clearance These two cleanup systems, ERAD and ER-phagy, work alongside the UPR and complement each other to maintain the ER’s overall size and function.13PubMed. ER-phagy responses in yeast, plants, and mammalian cells and their crosstalk with UPR and ERAD
When Repair Fails and the Cell Self-Destructs
All of the mechanisms described so far are pro-survival. But the UPR has a darker side. If stress persists and the ER cannot be restored, the same signaling pathways that tried to save the cell begin pushing it toward programmed death. The PERK branch is the clearest example: ATF4, which initially turns on helpful genes, also activates a transcription factor called CHOP when stress drags on. CHOP promotes cell death by suppressing survival proteins and increasing the production of reactive oxygen species.5PubMed Central. The Role of the PERK/eIF2α/ATF4/CHOP Signaling Pathway in Tumor Progression During Endoplasmic Reticulum Stress
CHOP is not the only death signal. Sustained IRE1 activity can activate a stress kinase called JNK, and rising calcium levels inside the cell can trigger a protein called caspase-12, which initiates the cell’s dismantling. Research using copper-sulfate-treated mice showed that all three pathways, CHOP, JNK, and caspase-12, can be activated simultaneously during severe ER stress, contributing to widespread cell death in the liver.14PubMed. Copper sulfate-induced endoplasmic reticulum stress promotes hepatic apoptosis by activating CHOP, JNK and caspase-12 signaling pathways The decision between survival and death is not a clean toggle but more of a tipping point: the longer the stress lasts and the more damage accumulates, the more the balance shifts from repair to apoptosis.
The ER-Mitochondria Calcium Pipeline
The ER does not operate in isolation. It forms physical contact sites with mitochondria, connected by protein tethers at structures called mitochondria-associated ER membranes (MAMs). These junctions are critical for transferring calcium between the two organelles, and that calcium flow influences metabolism, energy production, and cell death decisions.15PubMed Central. The correlation between mitochondria-associated endoplasmic reticulum membranes (MAMs) and Ca2+ transport in the pathogenesis of diseases
During ER stress, these contact sites can become a liability. Research on fatty liver disease found that ER stress increased the physical contact between the ER and mitochondria, driving excessive calcium transfer into the mitochondria. That calcium overload triggered the production of reactive oxygen species and activated the NLRP3 inflammasome, a potent inflammatory signaling complex. Blocking the ER stress with a chemical chaperone called TUDCA reduced mitochondrial calcium accumulation and dampened the inflammatory cascade.16PubMed Central. ER stress promotes mitochondrial calcium overload and activates the ROS NLRP3 axis to mediate fatty liver ischemic injury This ER-to-mitochondria calcium pipeline helps explain why ER stress so often leads to inflammation and tissue damage beyond the initial protein-folding problem.
ER Stress Feeds Into Inflammation
The connection between ER stress and inflammation runs through multiple channels. ER stress can activate the NLRP3 inflammasome through oxidative stress, disrupted calcium balance, and activation of the inflammatory master switch NF-κB.17PubMed. Crosstalk between ER stress, NLRP3 inflammasome, and inflammation NF-κB is especially interesting because each UPR branch can activate it through different mechanisms. The IRE1 branch does it through JNK signaling, the PERK branch does it by slowing the production of NF-κB’s inhibitor protein, and the ATF6 branch contributes through less well-characterized routes.18PubMed Central. NF-κB and its crosstalk with endoplasmic reticulum stress in atherosclerosis
This crossover matters clinically because it means ER stress does not just kill cells directly; it also recruits the immune system in ways that can damage surrounding tissue. In atherosclerosis, for example, ER stress in the cells lining blood vessels activates NF-κB-driven inflammation, which worsens plaque buildup and instability. The inflammation is not a separate problem layered on top of the ER stress; it is a direct downstream consequence.
Diseases Where ER Stress Plays a Central Role
Because the ER is involved in so many cell types and functions, chronic or severe ER stress shows up in a surprisingly wide range of diseases.
In neurodegenerative conditions like Alzheimer’s, Parkinson’s, and ALS, the accumulation of misfolded proteins is a hallmark. The UPR initially protects neurons by ramping up chaperones and clearance mechanisms, but prolonged stress from toxic protein buildup eventually triggers specific death pathways that destroy the very cells the brain cannot replace.19PubMed Central. Misfolded proteins, endoplasmic reticulum stress and neurodegeneration Understanding exactly when and how the UPR switches from protective to destructive in neurons is one of the most active areas of research in the field.
In type 2 diabetes, pancreatic beta cells face a double burden. They must produce and secrete large amounts of insulin, which places heavy demands on their ER, and they are exposed to metabolic stresses like high blood sugar and excess fatty acids. ER stress and JNK activation contribute to beta-cell death, reducing the pool of insulin-producing cells and worsening the disease. Inflammatory signaling through the NLRP3 inflammasome, triggered by ER stress, lipid injury, and other metabolic insults, compounds the damage.20PubMed. Role of pancreatic β-cell death and inflammation in diabetes
Cancer cells, paradoxically, often turn ER stress to their advantage. Tumors frequently grow in harsh conditions: low oxygen, limited nutrients, and acidic surroundings. These stresses activate the UPR, but rather than dying, many cancer cells co-opt the survival side of the response while evading the death signals.21PubMed Central. Cancer Microenvironment and Endoplasmic Reticulum Stress Response The UPR helps tumors adapt to hostile environments, supports their dormancy when conditions are worst, and may even suppress immune surveillance.22PubMed Central. Unfolded Protein Response (UPR) in Survival, Dormancy, Immunosuppression, Metastasis, and Treatments of Cancer Cells Conditions like oncogene activation and hypoxia keep the UPR chronically engaged in tumor cells, making it both a survival tool for the cancer and a potential therapeutic target.23PubMed. The UPRising connection between endoplasmic reticulum stress and the tumor microenvironment
Therapeutic Approaches to Taming ER Stress
If ER stress is at the root of so many diseases, can it be targeted with drugs? Several strategies are being explored. One approach uses chemical chaperones, small molecules that help proteins fold correctly and reduce the burden on the ER. Two of the best-studied are TUDCA (tauroursodeoxycholic acid) and PBA (4-phenylbutyric acid), both already approved for other medical conditions. In cell culture, TUDCA was more effective than PBA at preventing protein aggregation and protecting liver cells from stress-induced death. Interestingly, TUDCA appeared to work partly by engaging the PERK pathway itself, activating eIF2α phosphorylation and ATF4 expression in a way that promoted cell survival rather than death.24PubMed Central. Chemical chaperone, TUDCA unlike PBA, mitigates protein aggregation efficiently and resists ER and non-ER stress induced HepG2 cell death The fact that TUDCA reduced mitochondrial calcium overload in the fatty liver model discussed earlier shows how a single ER-stress-targeting compound can have broad downstream effects.
Other therapeutic strategies aim directly at individual UPR branches. Inhibitors of IRE1’s splicing activity are being tested in cancer models, where shutting down the tumor’s stress-coping mechanism could make it more vulnerable. PERK inhibitors have shown promise in animal models of neurodegeneration but carry the risk of pancreatic toxicity, since beta cells depend heavily on PERK to manage their protein-folding load. The field is still working out how to modulate these pathways precisely enough to help without causing collateral damage.
ER Stress Is Not Always Pathological
One of the more counterintuitive aspects of this field is that the UPR is not strictly a distress signal. Certain healthy cell types deliberately activate parts of the UPR during normal development. When B cells of the immune system mature into antibody-secreting plasma cells, they massively expand their ER to handle the flood of immunoglobulin production. This expansion involves activating the UPR, but in a selective and controlled way.25PubMed Central. Plasma cell differentiation initiates a limited ER stress response by specifically suppressing the PERK-dependent branch of the unfolded protein response Other secretory cell types, like the cells lining the gut that produce mucus, similarly depend on UPR signaling to build and maintain the ER infrastructure their jobs require. The line between “physiological UPR” and “pathological ER stress” has more to do with duration, intensity, and context than with the signaling itself.
Can ER Stress Be Detected Early?
Diagnosing ER stress in a living person is difficult because the markers are mostly intracellular. Tissue biopsies can reveal elevated levels of CHOP, spliced XBP1, or phosphorylated eIF2α, but biopsies are invasive and not always practical. Researchers have begun searching for secreted or excreted markers that could signal ER stress from a blood or urine sample. One promising candidate is a protein called CRELD2, which is secreted from cells under ER stress. In mouse models of ER-stress-driven kidney disease and in pediatric patients undergoing cardiac surgery, elevated urinary CRELD2 appeared before severe kidney injury became clinically obvious, suggesting it could serve as an early warning signal.26PubMed Central. Elevated urinary CRELD2 is associated with endoplasmic reticulum stress-mediated kidney disease Noninvasive biomarkers like this are still in the research stage, but they represent a step toward catching ER-stress-related disease before irreversible damage occurs.
How Viruses Exploit the System
Viruses that replicate inside cells often hijack the ER for their own purposes, since viral envelope proteins need the ER’s folding and modification machinery. The sheer volume of viral protein flooding into the ER triggers the UPR, but instead of letting the stress response shut down their replication, many viruses have evolved ways to selectively manipulate individual UPR branches. They may keep the branches that help with protein folding turned on while blocking the branches that would trigger antiviral immune responses or cell death.27PubMed Central. Hijacking the unfolded protein response (UPR) pathway: Balancing viral infection and host cell survival Some viruses even remodel ER membranes to create protected compartments for their replication machinery.28PubMed Central. The expanding roles of endoplasmic reticulum stress in virus replication and pathogenesis This arms race between viral exploitation and host defense is one of the reasons UPR signaling is so complex: the system has had to evolve safeguards against being co-opted, while viruses have evolved countermeasures against those safeguards.
An Ancient System That Keeps Getting More Complicated
The UPR is not the same in every organism. Yeast, the simplest eukaryote with a well-studied UPR, has only one sensor: IRE1. Invertebrate animals added PERK and ATF6. Vertebrates duplicated some of these sensors further, ending up with five: IRE1α, IRE1β, PERK, ATF6α, and ATF6β.29PubMed Central. Evolutionary Aspects of the Unfolded Protein Response Plants took a different route entirely, retaining IRE1 but developing unrelated sensors like bZIP28 and the GTP-binding protein AGB1 to handle their own unique stresses.30PubMed Central. Conserved and plant-unique strategies for overcoming endoplasmic reticulum stress The fact that every branch of the eukaryotic tree has independently elaborated on this system underscores just how fundamental protein-folding quality control is to complex life. It also means that findings from model organisms do not always translate directly to humans; the wiring is similar in broad strokes but different enough in the details to trip up researchers who assume the parallels are exact.