OAS2

OAS2 is an interferon-stimulated 2'-5'-oligoadenylate synthetase that functions as a dsRNA sensor in innate antiviral immunity, coupling pathogen detection to RNase L activation and downstream restriction of viral replication. It exists as a zinc-mediated auto-inhibited dimer in which the catalytically deficient N-terminal OAS domain serves as a molecular ruler, requiring dsRNA of at least 35 bp for enzymatic activation — a stringency that distinguishes it from OAS1 and prevents autoreactivity to short endogenous RNAs (PMID:40412389, PMID:30965010). Myristoylation and dimerization target OAS2 to Golgi membranes, and this endomembrane localization is essential for its activation and restriction of viruses such as coronaviruses that exploit the endomembrane system; a patient loss-of-function mutation in OAS2 is associated with autoimmune disease (PMID:40412389). Alternative splicing generates two isoforms with distinct antiviral specificities: the shorter p69 isoform restricts coronaviruses via an RNase L-independent mechanism, while the longer p71 isoform restricts picornaviruses via RNase L, with specificity determined by the variable C-terminal tail (PMID:40412389, PMID:32276512).

1. 2009 Low

   Early pathway-level studies showed that constitutive OAS protein levels suffice for 2-5A/RNase L pathway activation during certain viral infections, raising the question of whether OAS2 specifically contributes to antiviral defense or is functionally redundant with OAS1/OAS3.

   Evidence Viral infection of FRhK-4 cells with CVB1/HAV combined with RNase L activity assay and OAS Western blotting

   PMID:19383565

   Open questions at the time

   • OAS2 role inferred by absence rather than direct manipulation
   • No OAS2-specific knockdown or knockout performed
   • Single cell line system
2. 2019 High

   Biochemical reconstitution established that OAS2 requires ≥35 bp dsRNA for activation — far longer than OAS1's 19 bp minimum — and that both OAS domains are necessary for catalytic activity, resolving the question of why OAS2 fails to respond to highly structured viral RNAs that activate OAS1.

   Evidence In vitro enzymatic assays with purified recombinant OAS2, enzyme kinetics, and domain mutagenesis

   PMID:30965010

   Open questions at the time

   • Structural basis of dsRNA length discrimination was unknown at this point
   • In vivo relevance of the 35 bp threshold not tested
   • Role of 3'-overhangs in physiological activation unclear
3. 2019 Medium

   Identification of miR-340-5p as a direct negative regulator of OAS2 mRNA revealed a host regulatory circuit whereby influenza A infection downregulates this miRNA to permit OAS2 upregulation as an antiviral response.

   Evidence Luciferase reporter assay confirming direct miR-340-5p targeting of OAS2 3'UTR, miRNA mimic/inhibitor transfection, viral replication assay

   PMID:30753994

   Open questions at the time

   • Physiological miR-340-5p stoichiometry in relevant tissues not assessed
   • Contribution of OAS2 versus co-targeted RIG-I not separated
4. 2020 Medium

   Functional gain- and loss-of-function experiments demonstrated that OAS2 restricts Zika virus replication through enhancement of IFNβ expression and JAK/STAT signaling, placing OAS2 not only downstream but also upstream of the interferon amplification loop via RIG-I-dependent induction.

   Evidence OAS2 overexpression/siRNA knockdown in A549 cells with RT-qPCR, ISRE reporter assay, and RNA-Seq

   PMID:32276512

   Open questions at the time

   • Mechanism of OAS2-mediated IFNβ enhancement beyond RNase L not defined
   • Single cell line
   • Unclear if effect is direct or via 2-5A/RNase L cleavage products
5. 2022 Medium

   An activating Oas2 mutation in mice demonstrated that constitutive OAS2 pathway activity can prevent pregnancy-driven mammary cancer metastasis and enhance anti-PD-L1 immunotherapy efficacy, establishing a role for OAS2 in tumor immune surveillance.

   Evidence ENU mutagenesis mouse model crossed with MMTV-PyMT; survival analysis, flow cytometry, immunohistochemistry

   PMID:35505346

   Open questions at the time

   • Specific downstream effectors of the constitutive IFN response in this context not identified
   • Relevance to human cancer not established
   • Metastasis-specific versus general immune activation not resolved
6. 2023 Medium

   OAS2 was placed downstream of AT1R signaling in vascular smooth muscle cells, where it mediates autoantibody-driven phenotypic switching from contractile to synthetic states, expanding OAS2 function beyond classical antiviral immunity.

   Evidence AT1R knockout rat model, RNA-Seq, OAS2 siRNA knockdown reversing VSMC phenotypic markers

   PMID:38092283

   Open questions at the time

   • Whether OAS2 enzymatic activity (2-5A synthesis) is required for VSMC phenotype switching unknown
   • Mechanism connecting AT1R activation to OAS2 transcription not defined
   • Single lab observation
7. 2025 High

   Cryo-EM and crystal structures revealed that OAS2 is auto-inhibited as a zinc-mediated dimer whose N-terminal catalytically deficient domain acts as a molecular ruler for dsRNA length discrimination; myristoylation-dependent Golgi localization was shown to be essential for activation and restriction of endomembrane-exploiting viruses, and a patient loss-of-function mutation was linked to autoimmune disease.

   Evidence Cryo-EM/crystal structure, in vitro enzymatic assays, mutagenesis, subcellular fractionation, live imaging, patient mutation analysis

   PMID:40412389

   Open questions at the time

   • Full-length activated OAS2–dsRNA complex structure not yet captured
   • Mechanism linking OAS2 loss-of-function to autoimmunity versus viral susceptibility not dissected
   • Whether zinc-mediated dimerization is dynamically regulated in cells unknown
8. 2025 Medium

   Alternative splicing was shown to generate two OAS2 isoforms (p69 and p71) with entirely distinct antiviral mechanisms and virus specificities, determined by the C-terminal tail — resolving why OAS2 can act through both RNase L-dependent and -independent pathways.

   Evidence Isoform-specific overexpression/knockdown, RNase L-deficient cell lines, domain swap experiments (preprint)

   PMID:bio_10.1101_2025.02.24.639105

   Open questions at the time

   • Preprint; not yet peer-reviewed
   • RNase L-independent effector mechanism of p69 isoform uncharacterized
   • Relative expression levels and tissue-specific splicing of p69 versus p71 unknown

• Key open questions include the identity of the RNase L-independent effector pathway used by the p69 isoform, the structural basis of the activated OAS2–dsRNA complex, the mechanism connecting OAS2 loss-of-function to autoimmunity, and whether OAS2 enzymatic activity is required for its non-canonical roles in VSMC phenotype switching and EMT suppression.
• No activated OAS2–dsRNA complex structure
• RNase L-independent mechanism uncharacterized
• Non-canonical (non-antiviral) functions lack mechanistic depth