⚠️ Auto-extracted by Claude on 2026-05-12 — canonical cGAS biology synthesized from UniProt Q8N884, NCBI Gene 115004, and training-era summaries of widely-cited primary sources (Sun 2013, Glück 2017, Liu 2018 — all archive-confirmed). NMR-cGAS-HR-suppression arm (Chen 2025 Science) cited from abstract only due to closed-access PDF (#gap/no-fulltext-access). Numerics may be approximate; verify before relying on quantitative claims.

cGAS (CGAS / MB21D1)

Cyclic GMP-AMP synthase is the upstream cytosolic DNA sensor of the cgas-sting innate immune pathway, and — crucially for aging biology — a dual-function protein: in the cytoplasm it is a pro-inflammatory sentinel that drives inflammaging, while on chromatin it acts as a suppressor of homologous recombination (HR) that impairs genome maintenance. Both roles are relevant to aging, and the recently characterized chromatin-suppressor arm provides a mechanistic explanation for why the naked mole-rat — which has four amino-acid divergences in cGAS that alter this arm — enjoys exceptional genome stability.

Identity

UniProt: Q8N884 (CGAS_HUMAN); Swiss-Prot reviewed entry
Gene symbol: CGAS; aliases MB21D1, C6orf150
NCBI Gene ID: 115004 (confirmed 2026-05-12)
HGNC: 21367 (confirmed 2026-05-12)
Ensembl: ENSG00000164430 (confirmed 2026-05-12)
Chromosomal location: 6q13
Mouse ortholog: Cgas (one-to-one; mouse cGAS shares high sequence identity but has functional divergences in the chromatin-binding domain — see NMR comparison below)
Length: 522 amino acids (canonical human isoform)

Key functional domains

Domain	Residues (approx.)	Function
N-terminal disordered region	1–160	Unspecific dsDNA-binding; liquid–liquid phase separation with DNA
DNA-binding region 1	173–215	Sequence-independent dsDNA recognition
Nucleotidyltransferase core	~160–490	Catalytic activity; GTP + ATP → 2’3’-cGAMP
KKH catalytic loop	427–429	Active-site motif; conserved across cGAS orthologs
Nuclear localization signal (NLS)	295–305 (KRKR-loop)	Required for chromatin association during S-phase
Nuclear export signal (NES)	169–174	Enables cytoplasmic translocation post-DNA-damage
DNA-binding region 2	384–407	Second contact patch; cooperativity with region 1

cGAS requires a minimum ~20 bp dsDNA duplex for full catalytic activation; longer DNA triggers cooperative oligomerization and proportionally greater 2’3’-cGAMP output ¹.

Function: Arm 1 — cytosolic DNA sensor (canonical)

Sensing mechanism

cGAS surveys the cytoplasm for double-stranded DNA. Under homeostasis, cytoplasmic dsDNA is essentially absent; its appearance signals danger (microbial invasion, organelle damage, genome instability). cGAS recognizes dsDNA in a sequence-independent manner — the relevant features are length (≥20 bp for minimal activation) and the double-stranded configuration ¹.

Upon dsDNA binding, cGAS undergoes a conformational change that repositions the active site, enabling ATP and GTP to be cyclized into 2’3’-cGAMP — cyclic GMP-AMP with a non-canonical 2’–5’/3’–5’ phosphodiester bond that gives it higher STING-binding affinity than bacterial cyclic dinucleotides ¹.

Output: STING → TBK1 → IRF3 / NF-κB

2’3’-cGAMP diffuses to the ER membrane, where it binds and activates sting (TMEM173). Downstream signaling proceeds through tbk1 → IRF3 (type I IFN transcription) and NF-κB (pro-inflammatory cytokines including IL-6, TNF-α, CXCL10, IL-1β). Full pathway architecture is documented on cgas-sting.

Aging-relevant cytosolic DNA sources

Source	Aging context
Mitochondrial DNA (mtDNA)	Leaked from damaged mitochondria during MOMP sub-threshold events; increases with mitochondrial-dysfunction
Cytosolic chromatin fragments (CCFs)	Released from nuclei of senescent cells via lamin B1 loss and nuclear envelope disruption ²
Micronuclei contents	Chromosome mis-segregation products that rupture spontaneously in aged cells
LINE-1 reverse-transcribed cDNA	Retrotransposon reactivation due to epigenetic derepression in aged tissues — cytosolic ssDNA/dsDNA hybrids activate cGAS

Glück et al. 2017 established that cGAS senses CCFs arising from lamin B1 degradation in senescent cells, activating cGAS-STING → SASP cytokines and IFN-I signaling across multiple senescence triggers (oxidative stress, irradiation, oncogene activation) in both human fibroblasts and mouse in vivo models ². This creates a feed-forward loop: DNA damage → senescence → CCF leakage → cGAS-STING → SASP → paracrine senescence in neighboring cells.

The mtDNA-to-cGAS link is supported by Vizioli et al. 2020, demonstrating that mitochondria-to-nucleus retrograde signaling drives cytoplasmic chromatin formation and cGAS-mediated inflammation during senescence ³.

Function: Arm 2 — chromatin-bound HR suppressor (recently established)

Nuclear localization and chromatin tethering

Although classically described as a cytoplasmic protein, cGAS is constitutively present in the nucleus of proliferating cells, where it associates with chromatin — particularly centromeric satellite repeats and LINE elements — via its NLS (Lys295–Lys298 region). In this nuclear compartment, cGAS is normally held inactive by tight chromatin binding, which physically occludes DNA from accessing the catalytic site ⁴.

Suppression of homologous recombination

Liu et al. 2018 (Nature, 611 citations) established that nuclear cGAS suppresses DSB repair by homologous recombination and that this activity is separable from its cytoplasmic sensing function ⁴. Chromatin-bound cGAS restricts the access of HR effectors to double-strand break (DSB) sites, thereby reducing repair fidelity. In cell and mouse tumor models, nuclear cGAS deletion enhanced HR and reduced tumorigenesis. This paradoxically makes cGAS — a protein studied primarily as an antiviral innate immune sensor — also a brake on genome maintenance.

Chen 2025: TRIM41–P97 eviction and FANCI–RAD50 interference

The molecular mechanism of chromatin-bound cGAS’s HR-suppressive function was substantially clarified by Chen et al. 2025 (Science), which used the naked mole-rat as a comparative model ⁵:

Eviction mechanism in WT mammals. Upon DNA damage, cGAS is evicted from chromatin via a two-step process: TRIM41 (a RING E3 ubiquitin ligase) ubiquitinates cGAS at chromatin damage sites, targeting it for extraction by the P97/VCP segregase (AAA-ATPase). This ordered eviction clears cGAS from DSB repair foci.
Why eviction matters. When cGAS is cleared, the FANCI–RAD50 interaction at DSB sites is restored, enabling productive HR. Chromatin-bound (pre-eviction) cGAS physically limits the FANCI–RAD50 proximity required for HR initiation.
Naked mole-rat (NMR) divergence. NMR cGAS carries four amino-acid substitutions that weaken both TRIM41-mediated ubiquitination and P97 interaction, resulting in prolonged chromatin retention of NMR cGAS after DNA damage. This extended residency paradoxically potentiates HR by a mechanism not yet fully resolved (the Chen 2025 abstract implies that prolonged, correctly-positioned NMR cGAS at damage foci facilitates rather than suppresses FANCI–RAD50 interaction — possibly through a different surface interaction geometry than in human/mouse cGAS). needs-replication — full mechanistic model requires closed-access PDF verification.

Important caveat: Chen 2025 is closed-access; the above mechanistic description is derived from the published abstract and is not verified against the full-text Methods and Results. Quantitative claims (n, p-values, lifespan extension data) are NOT stated here. no-fulltext-access

In aging: the dual-burden model

cGAS imposes a dual aging burden in humans and mice that is absent (or mitigated) in the naked mole-rat:

Role	Direction in aging	Mechanism	Hallmark
Arm 1: cytosolic sensor	Pro-inflammatory (amplifies with age)	More cytosolic dsDNA with age → more cGAS-STING activation → SASP amplification, IFN-I elevation	chronic-inflammation, cellular-senescence
Arm 2: chromatin-bound HR suppressor	Pro-genomic instability (chronic brake on repair)	Nuclear cGAS occupancy limits FANCI–RAD50 interaction → reduced HR fidelity at DSBs	genomic-instability

These arms are not fully independent: accumulated genomic instability (Arm 2 consequence) generates more cytosolic DNA (Arm 1 input), and the resulting inflammation (Arm 1 output) accelerates mitochondrial dysfunction, which generates more mtDNA (Arm 1 input again). The circuit is self-reinforcing.

The NMR has evolutionarily addressed the Arm 2 problem via cGAS sequence divergence. Human/mouse cGAS has not. This cross-species contrast suggests the Arm 2 burden is not a fixed constraint of innate immune architecture but a tunable parameter — with potential therapeutic implications ⁵.

Dimension	Status	Notes
Arm 1 (sensing) conserved in humans?	yes	cGAS-STING fully functional in human cells; UniProt confirms
Arm 2 (HR suppression) conserved in humans?	yes	Liu 2018 demonstrates in human cell lines
NMR-cGAS Arm 2 divergence replicated?	partial	Chen 2025 primary; independent replication pending needs-replication
Aged-tissue cGAS-STING activation in humans?	partial	Indirect evidence via plasma IFN-I signatures; direct tissue protein-level data limited needs-human-replication

Post-translational regulation

cGAS activity is tightly regulated at multiple levels to prevent autoimmunity from self-DNA sensing:

Inhibitory phosphorylation (Ser305 by AKT; Thr68/Ser213 by DNA-PK/PRKDC) — reduces DNA-binding affinity; enriched in proliferating cells to avoid aberrant sensing of nuclear/replication DNA
Inhibitory acetylation (Lys384, 394, 414) — blocks DNA binding; established histone-deacetylase-reversible mechanism
Activating acetylation (Lys47, 56, 62, 83 by KAT5/Tip60) — promotes catalytic activity; coupled to DNA-damage signaling
Sumoylation (Lys347, 384, 394) — inhibitory; blocks DNA binding; part of the nuclear-inactivation mechanism
Palmitoylation (Cys404, Cys405 — activating; Cys474 — inhibitory) — controls membrane association and signaling competence
Chromatin binding — direct nucleosome interaction occludes the DNA-binding groove, inhibiting enzymatic activity in the nucleus (preventing spurious nuclear cGAMP production); this tethering is released at DSB sites by the TRIM41–P97 axis ⁵

Druggability

Druggability tier: 2 (high-quality probes exist; no FDA-approved drug for any indication; no aging-validated inhibitor).

Aging-context rationale: Several selective, cell-active cGAS small-molecule inhibitors are established tool compounds:

RU.521 — potent non-nucleotide competitive inhibitor of the catalytic site; widely used in murine autoimmune models
CU-76 — structurally distinct cGAS inhibitor; activity in mouse AGS-like models
G140 and related scaffolds — reported in late-preclinical-stage programs for lupus/interferon-opathy indications

As of 2026-05-12 no cGAS inhibitor has entered a registered clinical trial (ClinicalTrials.gov query returned 0 active/recruiting cGAS-inhibitor trials). Ongoing pharmaceutical programs for Aicardi-Goutières syndrome (AGS) and systemic lupus erythematosus are tracking toward IND-enabling studies but have not yet dosed humans ⁶. Therefore the aging-context tier is 2: clinical-grade probes exist, but no approved drug and no aging-indication trial.

A key therapeutic design challenge: pan-cGAS inhibition would disable both Arm 1 (desirable for anti-inflammaging) and antiviral defense (undesirable in elderly immunosenescent hosts). Selective approaches — enhancing Arm 2 eviction (TRIM41 activation, P97 pathway potentiation, or AAV-delivered NMR-cGAS 4-AA variant) — would specifically address the HR-suppressor burden without compromising Arm 1 antiviral capacity. These strategies remain hypothetical. needs-human-replication

Key interactors

sting — primary effector; cGAS-synthesized 2’3’-cGAMP binds ER-resident STING, activating downstream kinase cascade
tbk1 — third effector (after STING); phosphorylates IRF3 and NF-κB pathway components
trim41 — E3 ubiquitin ligase mediating chromatin eviction in response to DNA damage (Chen 2025) ⁵
vcp (P97) — AAA-ATPase segregase; pulls ubiquitinated cGAS off chromatin after TRIM41-mediated marking ⁵
FANCI / RAD50 — HR effectors whose interaction is limited by chromatin-bound cGAS; restored when cGAS is evicted ⁵
KAT5 (Tip60) — acetylates activating lysines (Lys47/56/62/83) in response to DNA damage, coupling genotoxic sensing to cGAS activation
DNA-PK (PRKDC) — phosphorylates Thr68/Ser213 to inhibit cGAS during normal S-phase replication, preventing cytoplasmic DNA sensing from replication intermediates

Pathway membership

cgas-sting — cGAS is the pathway’s initiating sensor; see pathway page for full architecture
dna-damage-response — both as a responder (activated by DSBs) and as a modulator (chromatin-bound cGAS limits HR)

Recency search summary

PubMed recency searches conducted 2026-05-12:

“cGAS aging senescence inflammaging” (2024–2026): 23 hits; triaged top results. Highest-relevance recent: Salminen et al. 2025 Biogerontology (PMID 41241888; review of cGAS-STING in senescent cells and immune remodeling) ⁷ — integrated as a 2025 review citation.
“cGAS DNA repair homologous recombination aging” (2024–2026): 1 hit (PMID 39515594; nanoplastic-induced senescence via cGAS-STING — adjacent but not directly integrated).
Chen 2025 Science (PMID 41066557; DOI 10.1126/science.adp5056) confirmed via archive — closed access, abstract-only sourcing applied throughout.

The Chen 2025 finding represents a significant post-2023 advance that is directly counter to the prior simplified view of cGAS as purely a pro-inflammatory cytoplasmic sensor. The dual-arm framing is the schema-critical update for this page.

Limitations and gaps

GenAge entry absent: CGAS not found in GenAge human or model-organism database (2026-05-12 search). May not qualify under GenAge inclusion criteria (requires demonstrated lifespan effect in a controlled experiment). needs-canonical-id
GTEx aging correlation unquantified: CGAS expression trajectory across tissue-age bins not queried from GTEx v2 API. Populate per sops/finding-tissue-expression.md. unsourced
MR evidence absent: no Mendelian randomization study for CGAS identified; no published GWAS instruments described as of 2026. mr-causal-evidence: not-tested.
Chen 2025 full-text access blocked: the chromatin-retention mechanistic model (TRIM41 → ubiquitination → P97 eviction → FANCI–RAD50 restoration) is from the abstract only. Mechanistic detail claimed for NMR cGAS four-AA variant should be verified against full Methods before relying on specific residue-level claims. no-fulltext-access
Liu 2018 closed-access: nuclear-cGAS HR-suppression paper (Nature, not_oa in archive). Quantitative details of HR suppression magnitude should be verified against the full text. no-fulltext-access
De Cecco LINE-1 citation: a paper by De Cecco and Sedivy group linking LINE-1 retrotransposon reactivation to cGAS-STING in aging is well-cited in the secondary literature but the specific DOI could not be confirmed via Crossref/PubMed during this seeding pass. Cited descriptively on the cgas-sting pathway page without a footnote. unsourced — needs DOI confirmation before adding a footnote here.
Human in-vivo aged-tissue data limited: direct measurement of cGAS-STING activation in human aged tissues at the protein level is sparse; most data come from cell models or mouse aging experiments.
Therapeutic strategies (Arm 2 enhancement) are entirely theoretical: TRIM41-agonist, P97-pathway-potentiation, and AAV-NMR-cGAS approaches have no published preclinical data as of 2026-05-12. needs-human-replication

Footnotes

sun-2013-cgas-cytosolic-dna-sensor · n=NR · in-vitro+in-vivo · Science 2013 · doi:10.1126/science.1232458 · 4,437 citations · local PDF confirmed in archive ↩ ↩² ↩³
gluck-2017-cgas-senescence-ccf · n=NR · in-vitro (human WI-38 fibroblasts) + in-vivo (mouse) · Nature Cell Biology 2017 · doi:10.1038/ncb3586 · 1,053 citations · local PDF confirmed in archive ↩ ↩²
doi:10.1101/gad.331272.119 · Vizioli MG et al. · Genes & Development 2020 · in-vitro+in-vivo (mouse) · mitochondria-to-nucleus retrograde signaling drives cytoplasmic chromatin and cGAS-mediated inflammation in senescence · DOI lookup pending ↩
liu-2018-nuclear-cgas-hr-suppression · n=NR · in-vitro+in-vivo · Nature 2018 · doi:10.1038/s41586-018-0629-6 · 611 citations · not_oa in archive · no-fulltext-access ↩ ↩²
chen-2025-nmr-cgas-hr-repair · n=NR · in-vitro+in-vivo · Science 2025 · doi:10.1126/science.adp5056 · not_oa in archive · abstract-only sourcing · no-fulltext-access ↩ ↩² ↩³ ↩⁴ ↩⁵ ↩⁶
doi:10.1080/1744666X.2019.1707663 · Tonduti D et al. · Expert Review of Clinical Immunology 2020 · review · cGAS-STING inhibitors in Aicardi-Goutières syndrome pipeline; near-clinical development stage overview ↩
doi:10.1007/s10522-025-10353-5 · Salminen A, Kaarniranta K, Kauppinen A · Biogerontology 2025 · review · cGAS-STING in senescent cells and immune system remodeling in aging · DOI lookup pending ↩

🧬 Aging Wiki

Explorer

cgas