eIF2alpha (EIF2S1)

The alpha subunit of eukaryotic initiation factor 2 (eIF2) — a 315 aa subunit that forms the regulatory hub of the integrated-stress-response (ISR). Phosphorylation of a single residue, Ser51, by any of four stress-sensing kinases converts eIF2 from an activator of cap-dependent translation into a potent brake on global protein synthesis while simultaneously enabling selective translation of the stress-transcription factor atf4. Chronic Ser51 phosphorylation accumulates in aged brain tissue and is mechanistically linked to age-related proteostatic decline and cognitive impairment.

Identity

  • UniProt: P05198 (IF2A_HUMAN)
  • NCBI Gene: 1965
  • Ensembl: ENSG00000134001
  • Gene symbol: EIF2S1 (historical alias: EIF2A — distinct from the unrelated EIF2A gene; do not conflate)
  • Mouse ortholog: Eif2s1 (one-to-one ortholog; highly conserved across eukaryotes)
  • Length: 315 amino acids (canonical isoform)

eIF2 heterotrimer

EIF2S1 is the alpha (regulatory) subunit of the obligate heterotrimeric eIF2 complex:

SubunitGeneUniProtRole
alphaEIF2S1P05198regulatory; Ser51 phosphorylation site
betaEIF2S2P20042GTP-binding accessory
gammaEIF2S3P68104catalytic GTPase; binds Met-tRNAi

The gamma subunit holds Met-tRNAi (initiator methionyl-tRNA) in a GTP-loaded ternary complex (eIF2-GTP-Met-tRNAi). This ternary complex is essential for delivery of Met-tRNAi to the 40S ribosomal subunit to form the 43S pre-initiation complex (43S PIC) 1.

Function: ternary complex and translation initiation

Under basal conditions, eIF2-GTP-Met-tRNAi loads onto the 40S subunit and scans the 5’ UTR for an AUG start codon. GTP hydrolysis occurs at AUG recognition; the resulting eIF2-GDP is released and recycled by the guanine nucleotide exchange factor eif2b (eIF2B), which catalyzes GDP → GTP exchange to regenerate active ternary complex 1.

This recycling step is the rate-limiting bottleneck for global cap-dependent protein synthesis, and it is the step targeted by Ser51 phosphorylation.

The Ser51 phosphoswitch

Phosphorylation of Ser51 (located in the N-terminal regulatory domain of EIF2S1) converts eIF2 from a substrate to a competitive inhibitor of its own GEF, eIF2B:

  • Phospho-Ser51-eIF2 binds eIF2B with ~150-fold higher affinity than unphosphorylated eIF2 1.
  • Because eIF2B is present in limiting amounts relative to eIF2, even partial phosphorylation (~30%) is sufficient to stoichiometrically titrate eIF2B activity.
  • Result: global arrest of cap-dependent mRNA translation within minutes.

Paradoxical ATF4 induction

While global translation falls, a small subset of mRNAs bearing inhibitory upstream open reading frames (uORFs) in their 5’ UTRs are preferentially translated when ternary complex is scarce. The most consequential is atf4 mRNA, which encodes the master stress-response transcription factor ATF4 1. ATF4 drives transcription of:

  • CHOP/DDIT3 — pro-apoptotic transcription factor
  • GADD34 (PPP1R15A) — feedback phosphatase (see below)
  • Amino acid biosynthesis genes, redox genes, autophagy-related genes

This translation-level logic gate — global suppression plus selective ATF4 induction — is the central output of all four ISR kinases acting through the single Ser51 phosphosite.

The four eIF2alpha kinases

Four kinases, each with a distinct stress-sensing domain, converge on Ser51 1:

KinaseGeneStress sensedCanonical context
perkEIF2AK3Unfolded proteins in ER lumenER stress / UPR
gcn2EIF2AK4Uncharged tRNAs (amino acid deprivation)Nutrient stress, UV
hriEIF2AK1Heme depletion; mitochondrial stress; osmotic stressErythropoiesis; heat shock
pkrEIF2AK2Double-stranded RNAViral infection; innate immune activation

Each kinase shares a conserved kinase domain that phosphorylates Ser51 but carries a unique regulatory/sensor domain. This convergence on a single substrate residue is the mechanistic definition of the ISR as a pathway — four inputs, one molecular bottleneck.

GADD34/PP1 feedback dephosphorylation

eIF2alpha-Ser51 phosphorylation is reversed by the GADD34–PP1 holophosphatase complex 2:

  • GADD34 (encoded by PPP1R15A) is itself an ATF4 transcriptional target, creating a delayed negative-feedback loop.
  • GADD34 acts as a regulatory subunit that recruits the catalytic PP1 (protein phosphatase 1) to phospho-Ser51-eIF2alpha, dephosphorylating it and restoring ternary complex assembly.
  • A constitutive paralog, CReP (PPP1R15B), performs basal eIF2alpha dephosphorylation independent of stress induction.

The GADD34/PP1 feedback loop gives the ISR transient rather than permanent translation arrest under mild stress, enabling an adaptive “pulse” followed by recovery.

Genetic evidence: the S51A knock-in

Scheuner et al. 2001 generated mice homozygous for an eIF2alpha S51A knock-in (Ser51 → Ala; abolishes all kinase phosphorylation) to dissect the physiological necessity of Ser51 regulation 3:

  • S51A/S51A homozygotes die perinatally with defective pancreatic beta-cell differentiation and impaired gluconeogenesis — establishing that Ser51 phosphorylation is essential for metabolic adaptation to the ER stress of insulin biosynthesis.
  • Heterozygotes (S51A/WT) survive but develop obesity and impaired glucose tolerance on high-fat diet.
  • The lethality of homozygous S51A confirms that the ISR is not a pathological response but a physiologically indispensable program.
DimensionStatusNotes
Pathway conserved in humans?yesEIF2S1 Ser51 and all four kinase families are fully conserved in humans; functional conservation well established
Phenotype conserved in humans?yesMutations in EIF2B subunits (Vanishing White Matter disease) recapitulate ISR dysregulation in humans
Replicated in humans?in-progressAging-specific ISR activation data in humans is limited; see Aging section below

needs-human-replication — The S51A survival data is from mouse; no human equivalent experiment exists.

ISRIB: chemical-genetic dissection of ISR output

Sidrauski et al. 2013 identified ISRIB (integrated stress response inhibitor) as a small molecule that bypasses Ser51-P to restore ternary complex assembly 4:

  • ISRIB acts downstream of Ser51 phosphorylation by stabilizing the decameric, active form of eIF2B, enhancing its nucleotide exchange activity even in the presence of phospho-eIF2alpha.
  • Because ISRIB does not block Ser51 phosphorylation itself, it separates eIF2B-dependent translation rescue from the upstream kinase signals — a clean chemical-genetic tool for dissecting ISR outputs.
  • In rodent memory assays, ISRIB administration enhanced long-term potentiation (LTP) and improved performance in spatial memory tasks, linking basal ISR tone to cognitive capacity 4.

ISRIB and structural analogs (including eIF2B activators like ABBV-CLS-7262 / DNL343, currently in clinical trials for neurodegenerative disease) have become the primary pharmacological tools for ISR intervention.

Aging relevance

Chronic ISR activation in aged brain

Basal ISR activity — evidenced by elevated phospho-Ser51-eIF2alpha and ATF4 protein — rises with age in rodent brain and in post-mortem human brain tissue from cognitively impaired donors 5. This chronic activation is hypothesized to progressively reduce ternary complex availability, suppressing translation of synaptic plasticity genes and contributing to age-associated cognitive decline 5.

DimensionStatusNotes
Pathway conserved in humans?yesISR machinery identical; age-related p-eIF2alpha elevation reported in human tissue
Phenotype conserved in humans?partialHuman post-mortem data correlative; causal direction not established
Replicated in humans?noNo interventional data in aged humans; correlation only

needs-human-replication — Causal link between ISR activation and human cognitive aging is not yet established; human trial data for ISR inhibitors in aging is absent.

Pharmacological ISR inhibition with ISRIB reversed age-associated long-term memory deficits in old (20-month) mice after a 3-day systemic treatment course 5. The effect was durable (weeks post-treatment) and accompanied by restoration of hippocampal translation rates and synapse density, suggesting that chronic ISR suppression of local dendritic translation underlies at least part of age-related memory decline.

needs-human-replication — All ISRIB aging rescue data is from rodent models.

ISR in neurodegeneration

Elevated p-eIF2alpha-Ser51 and downstream CHOP induction are documented in post-mortem tissue from Alzheimer’s disease, Parkinson’s disease, and prion disease brains. Whether ISR activation is causative or a secondary response to proteotoxic stress remains debated. contradictory-evidence — Multiple groups have shown ISR activation in neurodegeneration, but ISR inhibition studies report mixed results depending on disease model and timing of intervention.

Pathway membership

Cross-references

  • perk — ER-stress sensing kinase; phosphorylates Ser51 in UPR context
  • gcn2 — amino-acid-depletion sensing kinase
  • hri — heme/mitochondrial-stress sensing kinase
  • pkr — dsRNA/innate-immune sensing kinase
  • atf4 — primary transcriptional output downstream of Ser51 phosphorylation
  • eif2b — GEF; inhibited by phospho-Ser51-eIF2 in a competitive manner
  • gadd34 — ATF4-induced feedback phosphatase; restores translation
  • loss-of-proteostasis — hallmark; chronic ISR activation contributes via suppression of proteome renewal
  • integrated-stress-response — pathway page for full ISR circuit diagram

Limitations and gaps

  • #gap/needs-canonical-id — HGNC ID not populated; pull from https://www.genenames.org/data/gene-symbol-report/#!/symbol/EIF2S1 on next lint pass.
  • #gap/needs-human-replication — Aging-rescue experiments (cognitive and proteostatic) are mouse-only; no human interventional data with ISRIB or analogs in aging contexts.
  • #gap/needs-replication — Age-dependent p-eIF2alpha elevation in human brain is based on post-mortem cohorts; needs prospective or in-vivo imaging confirmation.
  • #gap/no-mechanism — The molecular mechanism by which ISR tone increases with normal aging (vs. acute stress) is poorly characterized; candidate mechanisms include mitochondrial dysfunction activating HRI, and increased ER load activating PERK, but tissue-specific data is sparse.
  • #gap/long-term-unknown — Long-term safety of systemic eIF2B activation (ISRIB-like compounds) is unknown; constitutive translation de-repression could impair stress responses to infection or acute injury.
  • #gap/unsourced — The ISRIB aging phenotype rescue figure (3-day treatment, durable weeks-long effect) is attributed to 5 from secondary summary; verify exact treatment protocol and effect duration against primary source PDF.

Footnotes

Footnotes

  1. doi:10.15252/embr.201642195 · Pakos-Zebrucka et al. · EMBO Rep 2016 · n/a · review · model: mechanistic synthesis across yeast, mouse, human; 2,458 citations 2 3 4 5

  2. doi:10.1083/jcb.153.5.1011 · Novoa et al. · J Cell Biol 2001 · in-vitro + in-vivo · model: HEK293 cells + UPR in MEFs; 1,351 citations

  3. doi:10.1016/s1097-2765(01)00265-9 · Scheuner et al. · Molecular Cell 2001 · n=litters (perinatal lethal) · in-vivo · model: S51A knock-in mouse (C57BL/6 background); 1,338 citations

  4. doi:10.7554/eLife.00498 · Sidrauski et al. · eLife 2013 · in-vivo · model: rat spatial memory + LTP; local PDF available at ; 717 citations 2

  5. doi:10.1038/s43587-021-00112-9 · Bhatt et al. / Bhatt & Bhatt (verify authorship) · Nature Aging 2021 · in-vivo · model: aged mice (20-month); ISR modulation and cognitive aging; 85 citations 2 3 4