EIF4A2

EIF4A2 is an ATP-dependent DEAD-box RNA helicase of the eIF4F cap-binding initiation complex that serves as a principal effector of miRNA-mediated translational repression (PMID:23559250, PMID:31791371). It executes repression downstream of miRNA-loaded effectors by physically associating with the Ccr4-Not deadenylase complex and binding purine-rich motifs enriched in the 5'UTR immediately upstream of the start codon, blocking initiation as a primary event that precedes mRNA degradation; mRNAs with unstructured 5'UTRs are refractory to this control (PMID:23559250, PMID:31791371). Although eIF4A1 and eIF4A2 exchange freely within eIF4F, they are functionally distinct: eIF4A2 specifically governs 18S rRNA processing and 40S ribosome subunit biogenesis during B-cell development (PMID:39516355), and biochemically the isoforms differ in ATP affinity and ATPase behavior, with residues L98 and A100 underlying these differences and opposite sensitivity to the BLF1 toxin (PMID:40423315). Beyond its repressive role, eIF4A2 promotes translation of specific transcripts—pluripotency factors, histone variant H3.3, and Ddx6—and cooperates with Ddx6 to restrain the two-cell totipotency program in embryonic stem cells (PMID:35353581). The protein is sumoylated at K226, a modification induced by arsenite and ionizing radiation that drives its recruitment to stress granules (PMID:27160682). EIF4A2 is required for efficient HIV-1 cDNA synthesis and for transmissible gastroenteritis coronavirus replication via direct binding of the viral M protein (PMID:29842983, PMID:30583231). Human genetics establishes disease relevance: EIF4A2 haploinsufficiency causes a dominant dystonia-tremor syndrome associated with elevated IMP1 and diminished eIF4A2–Ccr4-Not colocalization (PMID:37485550), and rare de novo and bi-allelic variants cause a neurodevelopmental syndrome through both loss- and gain-of-function mechanisms (PMID:36528028).

1. 2013 High

   Established that eIF4A2, rather than acting only in general initiation, is the specific eIF4F helicase through which miRNAs repress translation, and that repression—not degradation—is the primary event.

   Evidence Knockdown/overexpression, ribosome profiling, reporter assays, and 5'UTR structure analysis of miRNA targets

   PMID:23559250

   Open questions at the time

   • Did not define the RNA sequence determinants of eIF4A2 selectivity
   • Mechanistic link to deadenylation machinery not yet established
2. 2016 High

   Identified a single SUMO acceptor site (K226) and connected its modification to stress-induced relocalization, linking eIF4A2 to stress granule dynamics.

   Evidence Mass spectrometry, K226R mutagenesis, and stress granule immunofluorescence under arsenite/heat-shock/hippuristanol

   PMID:27160682

   Open questions at the time

   • SUMO ligase responsible for K226 modification not identified
   • Functional consequence of stress granule recruitment for translation not defined
3. 2018 Medium

   Showed eIF4A2 is a host factor co-opted by viruses, with step-specific requirements distinguishing entry from downstream replication.

   Evidence shRNA knockdown in T cells with viral cDNA/entry assays (HIV-1); yeast two-hybrid, GST pull-down, Co-IP, and siRNA knockdown (TGEV M protein)

   PMID:29842983 PMID:30583231

   Open questions at the time

   • Molecular role of eIF4A2 in viral cDNA synthesis unresolved
   • Whether helicase activity is required for viral co-option untested
4. 2019 High

   Defined the molecular basis of eIF4A2-mediated repression by demonstrating direct Ccr4-Not association and binding to purine-rich 5'UTR motifs upstream of the AUG.

   Evidence Reciprocal Co-IP, ribosome profiling, RNA-seq, motif enrichment, and comparison with hippuristanol-inhibited eIF4A1

   PMID:31791371

   Open questions at the time

   • Stoichiometry and architecture of the eIF4A2–Ccr4-Not interaction not resolved
   • How RISC hands off to eIF4A2 not defined
5. 2022 Medium

   Revealed an activating, transcript-selective role for eIF4A2 in stem cell translation, beyond its repressive function, including control of the totipotency program.

   Evidence RNAi screen, ribosome fractionation, and polysome profiling in ESCs with pluripotency/Zscan4 readouts

   PMID:35353581

   Open questions at the time

   • Determinants directing eIF4A2 to activation versus repression unknown
   • RPS26-dependent versus -independent ribosome usage mechanism unresolved
6. 2022 Medium

   Demonstrated that EIF4A2 variants cause a neurodevelopmental syndrome and that disease arises by both loss- and gain-of-function, using cross-species complementation.

   Evidence Drosophila loss-of-function rescue with human wild-type and variant cDNAs, plus molecular modeling

   PMID:36528028

   Open questions at the time

   • Molecular nature of the gain-of-function variants not biochemically defined
   • Affected neuronal mRNA targets not identified
7. 2023 Medium

   Linked EIF4A2 haploinsufficiency to a dystonia-tremor syndrome and connected the disease to the Ccr4-Not repression pathway through a patient-cell target readout.

   Evidence Western blot, immunocytochemistry, IMP1 quantification, and eIF4A2–Ccr4-Not colocalization in patient fibroblasts

   PMID:37485550

   Open questions at the time

   • Causality between IMP1 derepression and disease phenotype not established
   • Neuronal cell-type-specific consequences untested
8. 2024 Medium

   Distinguished eIF4A2 from eIF4A1 functionally, assigning eIF4A2 a specific role in 18S rRNA and 40S subunit biogenesis despite free exchange within eIF4F.

   Evidence Conditional knockout mouse genetics, ribosome profiling, rRNA biogenesis assays, and B-cell flow cytometry

   PMID:39516355

   Open questions at the time

   • Mechanism by which eIF4A2 acts in rRNA processing unknown
   • Whether this role extends beyond B cells untested
9. 2025 High

   Resolved isoform-specific biochemistry by quantifying differential ATP affinity/ATPase activity and mapping the responsible residues, plus opposite BLF1 toxin sensitivity.

   Evidence In vitro ATPase assays, N-terminal domain swapping, L98/A100 mutagenesis, and BLF1 treatment of purified proteins

   PMID:40423315

   Open questions at the time

   • How ATPase differences translate to distinct cellular functions not shown
   • Structural basis for BLF1 isoform selectivity not determined
10. 2025 Medium

    Placed eIF4A2 upstream of mTOR-regulated translation in tumor biology, where its loss triggers senescence that pharmacological translation reduction reverses.

    Evidence Conditional eIF4A2 knockout in KRAS lung cancer mouse model, rapamycin rescue, secretome proteomics, and p21 staining (preprint)

    PMID:bio_10.1101_2025.03.19.644135

    Open questions at the time

    • Direct mRNA targets driving senescence not identified
    • Generalizability beyond KRAS-driven lung adenocarcinoma untested

• How a single helicase reconciles its opposing roles—miRNA/Ccr4-Not-dependent repression versus transcript-selective translational activation and 40S biogenesis—remains unresolved.
• No structural model of eIF4A2 engaged with Ccr4-Not
• No unified rule predicting repression versus activation of a given mRNA
• Connection between SUMO/stress regulation and target selection undefined