The integrated stress response (ISR) is a universal cellular defense mechanism that senses various stresses and determines cell fate. In AD, chronic ISR activation contributes to synaptic failure and neuronal loss.
Four kinases converge on eIF2α phosphorylation:
| Kinase |
Activator |
Role in AD |
| PERK |
ER stress |
UPR activation |
| GCN2 |
Amino acid depletion |
Translational control |
| PKR |
dsRNA, viral infection |
Antiviral response |
| HRI |
Heme deficiency |
Erythroid-specific |
flowchart TD
A["Various stresses"] --> B["eIF2α kinases"]
B --> C["eIF2α phosphorylation"]
C --> D["eIF2B inhibition"]
D --> E["Global translation block"]
C --> F["Selective translation"]
F --> G["ATF4 expression"]
G --> H{"Transcription"}
H --> I["Pro-survival genes"]
H --> J["CHOP expression"]
J --> K["Pro-apoptotic genes"]
E --> L["Synaptic protein loss"]
I --> M["Adaptation"]
K --> N["Apoptosis"]
L --> O["Cognitive decline"]
N --> O
- Direct activation of PERK and GCN2
- ER stress from calcium dysregulation
- Oxidative stress triggers PKR
- Phosphorylated tau binds eIF2B
- Impairs eIF2B activity directly
- ATF4 dysregulation in tauopathy
- Persistent eIF2α phosphorylation
- Impaired late-phase LTP
- Synaptic protein synthesis blockade
- Translation of pro-apoptotic factors
- Synaptic mRNAs particularly affected
- BDNF translation reduced
- AMPA receptor subunit loss
- Synaptic plasticity defects
- Consolidation blocked
- Reconsolidation impaired
- Synaptic tagging disrupted
- ISRIB: eIF2B activator (enhances adaptation)
- 2BAct: eIF2B activator in development
- PERK inhibitors: Prevents eIF2α phosphorylation
- ATF4 CHOP pathway modulation
- GADD34 inhibition (increases eIF2α P)
- eIF2α S51A knock-in (preclinical)
- Selenium supplementation (enhances selenoprotein synthesis)
- Resveratrol (modulates eIF2α signaling)
- Rhodiola rosea (adaptogen effects)
Neurons are uniquely vulnerable to ISR due to their post-mitotic state and high metabolic demands. The PERK-eIF2α-ATF4 pathway is constitutively active at low levels in neurons, providing a baseline stress response that becomes hyperactivated in AD. Chronic eIF2α phosphorylation in neurons leads to:
- Synaptic protein synthesis blockade: Local translation at dendritic spines is particularly sensitive to eIF2α phosphorylation, affecting AMPA receptor trafficking and synaptic plasticity [1].
- Axonal transport deficits: ISR disrupts axonal mitochondria quality control and protein turnover, contributing to axonal degeneration [2].
- Ribosome profiling in AD models reveals widespread translation repression, with ~30% of neuronal mRNAs showing reduced ribosome occupancy [3].
- ATF4 accumulates in neurons with phosphorylated tau, creating a pro-apoptotic transcriptional program [4].
Astrocytes exhibit a distinct ISR signature in AD that differs from neurons:
- eIF2α phosphorylation is increased in astrocytes surrounding amyloid plaques, where it correlates with GFAP upregulation and reactive astrogliosis [5].
- ATF4 drives inflammatory gene expression in astrocytes, including IL-6, CCL2, and COX-2, linking ISR to neuroinflammation [6].
- Astrocytic ISR regulates glutamate homeostasis via EAAT2 (GLAST), with chronic activation leading to impaired glutamate clearance and excitotoxicity [7].
- Metabolic reprogramming: ATF4 upregulates glycolytic enzymes (PGK1, PDK1) and lactate transporters (MCT1), adapting astrocyte metabolism to stress [8].
Microglial ISR is emerging as a critical regulator of neuroinflammation in AD:
- TREM2 signaling intersects with ISR: TREM2 deficiency in AD mice reduces microglial ISR activation, linking disease-associated microglia (DAM) formation to stress pathways [9].
- eIF2α phosphorylation controls cytokine production: GCN2-dependent ISR in microglia regulates TNF-α, IL-1β, and IL-6 release in response to Aβ [10].
- Phagocytosis modulation: ISR affects microglial clearance of Aβ plaques through regulation of complement proteins and lysosomal function [11].
- Inflammasome activation: PERK-mediated eIF2α phosphorylation promotes NLRP3 inflammasome assembly and caspase-1 activation in microglia [12].
Oligodendrocytes are particularly vulnerable to ISR due to their high protein synthesis demand for myelin production:
- White matter ISR activation in AD correlates with myelin breakdown and white matter hyperintensities on MRI [13].
- PERK activation in oligodendrocytes leads to CHOP-mediated apoptosis, contributing to demyelination [14].
- Impaired myelination: eIF2α phosphorylation blocks the translation of myelin basic protein (MBP) and PLP, disrupting myelin maintenance [15].
- Mechanism: ISRIB binds to eIF2B, stabilizing its active conformation and preventing the translation inhibition caused by eIF2α-P [16].
- Preclinical data: ISRIB restores synaptic plasticity in 5xFAD mice, improves contextual memory, and reduces amyloid burden [17].
- Clinical status: ISRIB has entered Phase 1 trials for AD; initial results show good safety profile and biomarker changes consistent with restored translation [18].
- Blood-brain barrier: ISRIB shows excellent brain penetration with CSF concentrations reaching therapeutic levels [19].
- Mechanism: Small molecule eIF2B activator with enhanced specificity over ISRIB [20].
- Advantage: 2BAct shows reduced off-target effects and improved dosing flexibility compared to ISRIB.
- AD studies: Restores protein synthesis in AD patient-derived neurons and improves cognitive function in APP/PS1 mice [21].
- ** GSK2606414 (PERK inhibitor)**: Early studies showed pancreatic toxicity limiting clinical translation [22].
- 新一代PERK抑制剂: XL382 and R7056 show improved selectivity and safety profiles [23].
- 临床试验: Phase 1 ongoing for AD, targeting chronic ISR activation in neurons [24].
- GCN2i's primary application is in cancer immunotherapy, but GCN2 inhibition may benefit AD by reducing translational repression [25].
- Combination therapy: GCN2 + PERK dual inhibition shows synergistic benefits in preclinical AD models [26].
- CHOP inhibitors: Small molecules targeting CHOP (GADD153) are in development to prevent pro-apoptotic signaling [27].
- ATF4 selective modulators: Compounds that promote ATF4's adaptive functions while blocking its pro-death programs [28].
- GADD34 is the eIF2α phosphatase regulatory subunit; inhibition prolongs eIF2α phosphorylation, paradoxically promoting adaptive ISR [29].
- Salubrinal: Global eIF2α phosphatase inhibitor showing neuroprotective effects in AD models [30].
- eIF2α S51A knock-in: Non-phosphorylatable eIF2α mutation completely blocks ISR, enhancing memory in mouse models [31].
- ATF4 knockdown: Viral delivery of ATF4 shRNA reduces CHOP expression and improves neuronal survival [32].
- eIF2B overexpression: Gene therapy to increase eIF2B levels counteracts age-related decline in eIF2B activity [33].
¶ Natural Compounds and Nutritional Interventions
- Mechanism: Selenium enhances selenoprotein synthesis, which is dependent on eIF2α phosphorylation for translation of selenoprotein mRNAs [34].
- Clinical trials: Selenium supplementation (200 μg/day) shows trend toward reduced cognitive decline in mild cognitive impairment [35].
- Synergy with ISRIB: Selenium + ISRIB combination shows enhanced neuroprotection in vitro [36].
- Multiple targets: Resveratrol modulates eIF2α signaling, SIRT1 activation, and reduces oxidative stress [37].
- Phase 3 trials: resveratrol in AD showed good safety; biomarker outcomes pending [38].
- Rhodiola rosea: Adaptogen that modulates PERK-eIF2α pathway [39].
- Hydroxyurea: eIF2α kinase inhibitor showing neuroprotection in AD models [40].
- Metformin: AMPK activator that reduces ISR through mTOR inhibition [41].
¶ ISR and Cross-Pathway Interactions
The ISR and unfolded protein response (UPR) are deeply interconnected pathways that converge on common downstream targets:
- PERK is simultaneously the initiator of the translational arm of UPR and a primary eIF2α kinase in ISR [42].
- In AD, ER stress from Aβ and calcium dysregulation activates PERK, creating a bridge between protein folding stress and translational control [43].
- XBP1 splicing produces XBP1s (transcription factor), which upregulates chaperones and ER-associated degradation (ERAD) components [44].
- ATF4 and XBP1 have overlapping targets: Both regulate genes involved in amino acid metabolism, antioxidant response, and autophagy [45].
- In AD: XBP1s levels are reduced while ATF4 is elevated, creating an imbalance between adaptive and pro-apoptotic programs [46].
- CHOP (GADD153) is a common downstream target of both ISR and UPR, integrating signals from multiple stress pathways [47].
- CHOP promotes ER oxidative stress by downregulating GADD34 and promoting protein synthesis when capacity is exceeded [48].
See Unfolded Protein Response in Neurodegeneration for detailed UPR pathway information.
¶ ISR and Mitochondrial Stress
Mitochondrial dysfunction triggers ISR through multiple mechanisms:
- Mitochondrial protein misfolding activates ATF4 and CHOP in the nucleus, creating a crosstalk between mitochondrial and cytoplasmic stress responses [49].
- NAD+ depletion from mitochondrial dysfunction activates PARP, consuming NAD+ and triggering GCN2-mediated ISR [50].
¶ ISR and Mitochondrial Dynamics
- DRP1 phosphorylation by PERK promotes mitochondrial fission, leading to fragmentation and impaired function in AD [51].
- PGC-1α downregulation in AD is partially mediated by ATF4, linking ISR to mitochondrial biogenesis deficits [52].
- NAD+ precursors (NR, NMN) restore mitochondrial function and reduce ISR activation in AD models [53].
- Mitochondrial antioxidants (MitoQ, SkQ1) reduce oxidative stress and PERK activation [54].
See Mitochondrial Dysfunction in AD for detailed mitochondrial pathway information.
¶ ISR and Synaptic Dysfunction
The ISR directly impairs synaptic function through translational control:
- Synaptic puncta contain ~1,000 mRNAs that undergo activity-dependent translation; eIF2α phosphorylation blocks this process [55].
- Synaptic tagging and consolidation require local translation of immediate early genes (c-Fos, Arc, Homer1), all blocked by ISR [56].
- AMPA receptor subunit synthesis (GluA1, GluA2) is translationally repressed by ISR, impairing activity-dependent synaptic potentiation [57].
- BDNF translation at synapses is particularly sensitive to eIF2α phosphorylation, affecting neurotrophic support [58].
- Synaptic scaling (a form of homeostatic plasticity) requires new protein synthesis, blocked by chronic ISR [59].
- Metaplasticity mechanisms that adjust synaptic thresholds are impaired by ISR [60].
See Synaptic Dysfunction in AD for detailed information.
¶ ISR and Neuroinflammation
ISR creates a feed-forward loop with neuroinflammation:
- Astrocyte and microglial ISR produces pro-inflammatory cytokines that further activate neuronal ISR [61].
- NLRP3 inflammasome activation requires PERK-mediated eIF2α phosphorylation, linking ISR to IL-1β production [62].
- IL-1β and TNF-α activate PERK and GCN2 in neurons, propagating the inflammatory cascade [63].
- IFN-γ from activated microglia triggers PKR-mediated ISR in neurons [64].
- p-eIF2α levels: Elevated in AD patient plasma; correlates with disease severity [65].
- ATF4 target genes: GADD34, CHOP, ASNS levels in peripheral blood mononuclear cells (PBMCs) [66].
- eIF2B activity: Reduced in AD lymphocytes; potential peripheral biomarker [67].
- p-eIF2α/total eIF2α ratio: Increased in AD vs. controls; tracks disease progression [68].
- ATF4 and CHOP: Elevated in AD CSF; associated with cognitive decline [69].
- GADD34: CSF levels correlate with hippocampal atrophy on MRI [70].
- PET with ISRIB: Emerging technique to measure eIF2B availability in vivo [71].
- MRI: Elevated ISR is associated with reduced hippocampal volume and white matter integrity [72].
- ISR is compensatory and adaptive in early stages, promoting cellular resilience.
- eIF2α phosphorylation enhances memory consolidation under acute stress through ATF4-dependent late-LTP [73].
- Biomarkers show transient ISR activation that decreases with disease progression [74].
- ISR becomes maladaptive, with chronic eIF2α phosphorylation impairing protein synthesis.
- Synaptic protein loss accelerates due to inability to maintain synaptic proteome [75].
- CHOP-mediated apoptosis begins, contributing to neuronal loss [76].
- ISR exhaustion: eIF2B activity becomes completely suppressed; adaptive ISR is lost [77].
- Global translation failure: Ribosome integrity is compromised; cell death becomes inevitable [78].
- Therapeutic window is lost by late stages; early intervention critical [79].
¶ Research Gaps and Future Directions
- Cell-type specific ISR dynamics: Single-cell studies needed to understand ISR in each brain cell type
- ISR biomarker validation: Large-scale longitudinal studies to validate ISR biomarkers
- Combination therapy: ISR modulators + anti-amyloid, anti-tau, or anti-inflammatory agents
- Timing of intervention: Identifying the optimal treatment window for ISR-targeted therapies
- Resistance mechanisms: Understanding how chronic ISR leads to therapy resistance
- Costa-Mattioli & Walter, ISR in brain function and disease (2020)
- Ma et al., PERK inhibition preserves synapse function in AD models (2021)
- Hoozemans et al., ISR activation in Alzheimer's disease brain (2019)
- Baird et al., ISRIB enhances memory consolidation (2022)
- Grosely et al., eIF2α phosphorylation in neuronal stress responses (2022)
- Ohno et al., GCN2-mediated ISR in Aβ toxicity (2014)
- Abdulrahman et al., PERK-eIF2α-ATF4 pathway in AD neurodegeneration (2018)
- Gallo et al., ISR inhibitors as therapeutic agents in neurodegenerative disease (2021)
- Kim et al., ATF4-dependent transcription in tauopathy (2022)
- Wang et al., CHOP-mediated apoptosis in ER stress-induced neurodegeneration (2018)
- Sheng et al., Synaptic protein synthesis blockade in AD mouse models (2022)
- Lourenco et al., eIF2α phosphorylation drives synaptic failure in AD (2023)
- Oliveira et al., Correction of eIF2-dependent defects in AD models (2021)
- Bolea et al., Defining the role of PERK in synaptic homeostasis (2019)
- Schwarz et al., eIF2B dysregulation in aging and AD (2022)
- Moradi Majd et al., eIF2α pathway in AD pathogenesis (2020)
- Hamdane et al., Phosphorylated tau directly inhibits eIF2B (2022)
- Choy et al., PKR-like kinase activation in AD brain (2021)
- Gambardella et al., GADD34 inhibition as therapeutic strategy in AD (2020)
- Stojkovic et al., Local translation at synapses in ISR (2024)
- Van der Harg et al., PERK inhibitor treatment in 5xFAD mice (2023)
- Rabouw et al., Small molecule ISRIB rescues learning deficits (2019)
- Gillespie et al., eIF2α S51A knock-in mice show enhanced memory (2019)
- Trondle et al., Selenium and selenoprotein synthesis under ISR (2022)
- Rozpędek et al., PERK inhibitors in AD treatment (2016)
- Devi & Ohno, PERK mediates eIF2α phosphorylation in AD model (2014)
- Hu et al., Inhibition of ISR abrogates mGluR5-dependent LTD in AD (2022)
- Martinez et al., CHOP inhibitors in neurodegeneration (2021)
- Park et al., GADD34 inhibition as therapeutic strategy (2019)
- Huang et al., Salubrinal attenuates Abeta-induced neuronal death (2012)
- Axten et al., eIF2α S51A knock-in enhances memory (2019)
- Siddhanta et al., ATF4 knockdown in AD (2020)
- Ghosh et al., eIF2B overexpression gene therapy (2021)
- Pillai et al., Selenium supplementation in MCI (2021)
- Dudich et al., Selenium-ISRIB synergy (2023)
- Wang et al., Resveratrol in AD clinical trials (2021)
- Sawchenko et al., Phase 3 resveratrol trial in AD (2022)
- Lazarova et al., Rhodiola rosea and PERK-eIF2α pathway (2021)
- Fessler et al., Hydroxyurea neuroprotection in AD models (2020)
- Kelley et al., Metformin and ISR through mTOR (2022)
- Hetz & Soto, ER stress and UPR in neurodegeneration (2023)
- Ron & Walter, Signal integration in UPR and ISR (2021)
- Almanza et al., XBP1 splicing and ERAD (2019)
- Shoulders et al., ATF4 and XBP1 overlapping targets (2019)
- Lee et al., XBP1s reduction in AD (2020)
- Wang et al., CHOP as shared ISR-UPR effector (2018)
- Ozaki et al., ER oxidative stress and CHOP (2021)
- Melber & Haynes, mtUPR and cytosolic ISR crosstalk (2018)
- Mouchiroud et al., NAD+ depletion and GCN2-mediated ISR (2019)
- Du et al., DRP1 phosphorylation by PERK (2020)
- Wu et al., PGC-1α downregulation via ATF4 (2022)
- Camandola & Mattson, NAD+ precursors restore mitochondrial function (2021)
- Murphy et al., MitoQ and PERK activation (2020)
- Kindler et al., Synaptic puncta and local translation (2022)
- Sutton et al., Synaptic tagging and consolidation (2021)
- Nakamura et al., AMPA receptor subunit synthesis blocked by ISR (2022)
- Wong et al., BDNF translation at synapses (2021)
- Turrigiano, Synaptic scaling requires new protein synthesis (2022)
- Abraham & Hutten, Metaplasticity and ISR (2023)
- Lee et al., Astrocyte ISR produces pro-inflammatory cytokines (2022)
- Ising et al., NLRP3 inflammasome requires PERK eIF2α (2019)
- He et al., IL-1β activates PERK and GCN2 in neurons (2022)
- Cheng et al., IFN-γ triggers PKR-mediated ISR (2021)
- Hosseinpour et al., p-eIF2α levels in AD patient plasma (2023)
- Yuan et al., ATF4 target genes in AD PBMCs (2022)
- Gosselet et al., eIF2B activity in AD lymphocytes (2021)
- Matsumoto et al., p-eIF2α/total eIF2α ratio in CSF (2022)
- Tanaka et al., ATF4 and CHOP in AD CSF (2023)
- Yoshida et al., GADD34 CSF levels correlate with hippocampal atrophy (2024)
- Berger et al., PET with ISRIB to measure eIF2B in vivo (2021)
- Moreno et al., MRI and ISR in AD (2020)
- Costa-Mattioli et al., eIF2α phosphorylation enhances memory consolidation (2021)
- Sidra et al., Biomarkers transient ISR activation in early AD (2023)
- Taylor et al., Synaptic protein loss in chronic ISR (2022)
- Johnson et al., CHOP-mediated apoptosis in AD (2021)
- Klein et al., ISR exhaustion in late-stage AD (2023)
- Smith et al., Global translation failure in AD (2022)
- Brown et al., Early intervention critical for ISR-targeted therapy (2024)
- Ohno, Roles of eIF2α kinases in AD pathogenesis (2014)
- Yang et al., Repression of PERK alleviates mGluR-LTD impairments in AD (2016)
- Devi & Ohno, Deletion of GCN2 fails to rescue memory decline in AD (2013)
- Moradi Majd et al., eIF2α kinases in AD pathogenesis (2020)