FOXO3 (Forkhead Box O3)

FOXO3 is a Forkhead family transcription factor and the most-replicated longevity-associated gene in human GWAS — its G allele at rs2802292 has been independently confirmed in Japanese-American, Ashkenazi Jewish, German, Italian, and Chinese cohorts. As the primary nuclear effector downstream of the insulin-igf1 / pi3k-akt-pathway axis, FOXO3 integrates nutrient and growth-factor signals to regulate apoptosis, stress resistance, autophagy, and cell-cycle arrest. When AKT is active (fed, growth-factor-replete state), FOXO3 is phosphorylated and sequestered in the cytoplasm; when AKT is suppressed (fasting, stress), FOXO3 enters the nucleus and drives a broad pro-longevity transcriptional program.

Identity

UniProt: O43524 (FOXO3_HUMAN) — reviewed Swiss-Prot entry
NCBI Gene: 2309
HGNC symbol: FOXO3 (formerly FOXO3A in older literature; FKHRL1 in earliest nomenclature)
Ensembl: ENSG00000118689
Mouse ortholog: Foxo3 (one-to-one ortholog; highly conserved forkhead domain)
C. elegans ortholog: daf-16 (the founding member of the FOXO longevity-regulatory axis)
Length: 673 amino acids (canonical isoform)
GenAge human entry: 123 (rs2802292 G allele associated with longevity, multiple populations)
Chromosomal location: 6q21; involved in t(6;11)(q21;q23) translocation fusing FOXO3 to KMT2A/MLL1 in secondary acute leukemia

Nomenclature note: The gene was formerly designated FOXO3A in the dbSNP/GWAS literature (e.g., rs2802292 is often described as a “FOXO3A SNP”). Current HGNC-approved symbol is FOXO3. Both names appear in cited literature; this page uses FOXO3 throughout with FOXO3A in aliases.

FOXO family in mammals

Four mammalian paralogs, all regulated by AKT:

Paralog	Old name	Primary tissue expression	Aging relevance
FOXO1	FKHR	Ubiquitous; high in adipose, liver	Metabolic regulation; no strong longevity GWAS signal
FOXO3	FKHRL1	Ubiquitous; highest in brain, heart	Strongest human longevity GWAS; canonical longevity TF
FOXO4	AFX	Ubiquitous	Senescent cell survival; FOXO4-p53 interaction targeted by senolytic peptides
FOXO6	—	Brain-restricted	Memory; minimal aging literature

FOXO1/3/4 share conserved AKT phosphorylation sites and DNA-binding specificity; FOXO6 lacks a nuclear export sequence and is constitutively nuclear.

Domain organization

N-terminus ←————————————————————————————→ C-terminus
  Thr32    [~aa 1–156]  [FH DBD 157–251]  [NLS 242–259]  [NES]  [TAD]  Ser315
  |                           |                                           |
  AKT site               FHRE binding                               AKT site
  (14-3-3)             (TTGTTTAC)                               (cytoplasmic export)
  
  Ser253 (central AKT site) falls within the forkhead domain C-terminal lobe

Key structural elements:

AKT phosphorylation sites (Thr32, Ser253, Ser315): Three conserved sites phosphorylated by AKT1/2/3; Thr32 and Ser253 create 14-3-3 docking surfaces; Ser315 promotes nuclear export
Forkhead DNA-binding domain (aa 157–251): Winged-helix structure; recognizes the canonical FHRE consensus 5’-[AG]TAAA[TC]A-3’ (UniProt lists 5’-[AG]TAAA[TC]A-3’)
Nuclear localization signal (NLS, aa 242–259): Overlaps with the DBD C-terminus; 14-3-3 binding masks the NLS
Nuclear export sequence (NES): CRM1-dependent export; phospho-Ser315 facilitates CRM1 association
C-terminal transactivation domain (TAD): Recruits transcriptional co-activators

Regulation — post-translational modifications

AKT phosphorylation (inhibitory)

The canonical inactivation mechanism ¹:

Growth factors / insulin activate pi3k-akt-pathway → AKT phosphorylates FOXO3 at Thr32, Ser253, and Ser315
Phospho-Thr32 and phospho-Ser253 recruit 14-3-3 proteins → sequester FOXO3 in cytoplasm
Phospho-Ser315 drives CRM1-dependent nuclear export
Net result: FOXO3 transcriptionally inactive; pro-survival gene expression dominates

Dimension	Status
Pathway conserved in humans?	yes
Phenotype (FOXO3 cytoplasmic in fed state) conserved?	yes
Replicated in humans?	yes (pharmacological + genetic evidence)

SIRT1 deacetylation (activating)

Under caloric restriction and nutrient stress, sirt1 deacetylates FOXO3 at multiple lysine residues. SIRT1-mediated deacetylation potentiates FOXO3 transcriptional activity, particularly toward stress-resistance (MnSOD, catalase) and cell-cycle arrest targets, while attenuating pro-apoptotic outputs. This provides a mechanistic link between caloric-restriction, SIRT1 activation, and FOXO3-driven longevity signaling. needs-replication (precise lysine sites and functional outcome hierarchy not fully resolved in primary literature)

MST1 phosphorylation (activating, stress-response)

STK4/MST1 (a Hippo pathway kinase) phosphorylates FOXO3 at Ser209 in response to oxidative stress, promoting nuclear translocation independently of AKT dephosphorylation. This provides an AKT-bypass route to FOXO3 activation under oxidative conditions. CDK1/2 also phosphorylate FOXO3 inhibitorily (during S and G2/M phases). unsourced (MST1-FOXO3 Ser209 phosphorylation — citation needed; Lehtinen et al. 2006 Nature likely source; not in Kops 2002)

AMPK phosphorylation (activating)

ampk phosphorylates FOXO3 at Ser30 (and additional sites in FOXO1), directly activating FOXO3 transcriptional activity under energy-deplete conditions. This positions FOXO3 as an output node where both the energy-sensing arm (ampk) and the growth-factor arm (pi3k-akt-pathway) converge. unsourced (the specific downstream target gene hierarchy activated by AMPK-pFOXO3 vs AKT-released FOXO3 is not well characterized)

IKKβ phosphorylation (inhibitory)

IKBKB/IKKβ (NF-κB pathway kinase) phosphorylates FOXO3 at Ser644, promoting cytoplasmic sequestration. This creates a feedback loop: inflammatory signals (which activate IKKβ) suppress FOXO3, consistent with the pro-inflammatory / pro-aging role of NF-κB signaling. needs-replication

Transcriptional targets

FOXO3 drives a broad pro-longevity gene expression program when active. Key targets:

Apoptosis

Gene	Product	Outcome
bim (BCL2L11)	BIM pro-apoptotic BH3-only protein	Apoptosis induction ²
PUMA (BBC3)	PUMA pro-apoptotic BH3-only protein	Apoptosis induction
FasL (FASLG)	Fas ligand	Death receptor pathway activation
TRAIL (TNFSF10)	TRAIL death ligand	Extrinsic apoptosis

FOXO3-driven apoptosis is context-dependent — it is most important in hematopoietic cells under cytokine withdrawal and in oncogenically transformed cells. In normal quiescent cells, FOXO3 predominantly activates stress-resistance rather than apoptosis programs ³.

Cell-cycle arrest

Gene	Product	Outcome
CDKN1B (p27/Kip1)	CDK inhibitor	G1 arrest
CDKN1A (p21)	CDK inhibitor	G1/S arrest (FOXO3 also a p21 transcription target cooperatively with p53)
Cyclin G2	Inhibitory cyclin	G1/S checkpoint
RBL2 (p130)	RB-related pocket protein	E2F repression

Oxidative stress resistance

Gene	Product	Outcome
[[sod2	SOD2 (MnSOD)]]	Mitochondrial superoxide dismutase
CAT (catalase)	H2O2 detoxification enzyme	ROS detoxification
SESN3	Sestrin-3	mTOR inhibition + antioxidant

Kops et al. 2002 showed that FOXO3-deficient cells accumulate elevated ROS and are hypersensitive to oxidative stress, establishing FOXO3 as a direct transcriptional activator of MnSOD/SOD2 (demonstrated by ChIP, luciferase reporter, and northern blot in that paper) ³. Catalase as a direct FOXO3 target is supported by other literature but was not directly demonstrated in Kops 2002. needs-replication (FOXO3-driven catalase transcription needs independent citation)

Autophagy

Gene	Product	Outcome
BECN1 (beclin-1)	Beclin-1; autophagy initiation	Autophagosome nucleation
BNIP3 (bnip3)	Mitophagy receptor	Mitochondrial clearance
ATG12	ATG12; autophagy conjugation	Autophagosome elongation
MAP1LC3B (LC3B)	LC3-II; autophagosome membrane	Autophagosome formation
GABARAPL1	GABARAP-L1	Autophagosome maturation

Mammucari et al. 2007 demonstrated in skeletal muscle in vivo that FoxO3 is sufficient and necessary to activate autophagy gene expression; constitutively active FoxO3 induced atrophy and autophagic vacuole formation, while dominant-negative FoxO3 blocked denervation-induced autophagy ⁴.

Dimension	Status
Pathway conserved in humans?	yes
Phenotype (FOXO3→autophagy induction) conserved?	yes
Replicated in humans?	in-progress — autophagy upregulation in fasted humans is well-established but direct FOXO3-ChIP evidence in human tissue is limited

DNA damage response

Gene	Product	Outcome
GADD45A	Growth arrest / DNA damage gene	NER co-factor; G2/M checkpoint
GADD45B	GADD45 beta isoform	Similar to GADD45A

Muscle proteolysis (atrogenes)

Under prolonged fasting or denervation, FOXO3 transactivates atrogin-1 (FBXO32/MAFbx) and MuRF1 (TRIM63), muscle-specific E3 ubiquitin ligases that target sarcomeric proteins for proteasomal degradation. This is the molecular basis of FOXO3’s dual role: beneficial (nutrient sensing → autophagy → proteostasis) at physiological activation levels; pathological (atrogene induction → sarcopenia) at sustained high activation ⁴. contradictory-evidence — the dose-response relationship between FOXO3 activation level/duration and the balance of beneficial vs. muscle-wasting outputs is incompletely characterized. See sarcopenia.

Human longevity genetics

rs2802292 — the canonical longevity SNP

The strongest human genetic evidence for FOXO3 comes from GWAS and candidate-gene studies of exceptional longevity:

Willcox 2008 (Japanese-American Honolulu Heart Program cohort): Homozygous GG genotype at rs2802292 associated with exceptional longevity (OR = 2.75, P = 0.0007) vs TT genotype; heterozygous GT intermediate ⁵. G allele also associated with reduced all-cause mortality in the combined analysis. needs-replication only in the sense that effect sizes vary across cohorts; the direction of effect (G allele protective) is replicated.

Multi-cohort replication: The rs2802292 G allele association has been independently replicated in:

Ashkenazi Jewish centenarians (Albert Einstein College of Medicine cohort)
German long-lived individuals (Kiel cohort)
Italian nonagenarians/centenarians
Chinese long-lived individuals (multiple cohorts)

Mechanistic interpretation: rs2802292 is an intronic SNP; the G allele is associated with modestly higher FOXO3 expression in some tissues, consistent with a partial gain-of-function. The exact molecular mechanism (altered splicing, transcription factor binding, etc.) has not been fully resolved. no-mechanism (exact functional effect of rs2802292 G allele on FOXO3 expression/activity)

Dimension	Status
Gene conserved in humans?	yes — it IS a human gene
Longevity association replicated in humans?	yes — 5+ independent cohorts
Causal mechanism established?	no — GWAS association; functional mechanism of SNP unclear

Knockout phenotype in mice

Foxo3-/- (single KO) — Castrillon 2003

Foxo3-null female mice develop normally but show premature ovarian failure due to constitutive activation of primordial ovarian follicles, leading to follicle depletion ⁶. Male Foxo3-null mice are fertile. The ovarian phenotype was independently confirmed by Hosaka et al. 2004 ⁷ (all oocytes degenerated by 12 weeks in Foxo3a-/- females). Adult Foxo3-null mice are also reported to display lymphadenopathy and splenomegaly (myeloid infiltration) ⁶ — this claim is unverifiable against the full text (closed-access paper). no-fulltext-access (Castrillon 2003 — lymphadenopathy/splenomegaly claim cannot be confirmed; Hosaka 2004 found no consistent non-ovarian histological abnormalities in Foxo3a-null mice across 30+ tissues)

This phenotype — not a classic aging/lifespan phenotype — highlights that FOXO3’s most dosage-sensitive function in mice is follicle quiescence, not longevity per se. The longevity-relevant biology emerges more clearly in compound knockouts and in the context of nutrient-sensing manipulation.

Dimension	Status
Pathway conserved in humans?	yes (follicle activation via FOXO3 conserved)
Phenotype conserved in humans?	partial (premature ovarian insufficiency in humans associated with FOXO3 variants, but rarer and less dramatic)
Replicated in humans?	no direct equivalent KO; limited observational data

needs-human-replication — the aging/longevity phenotype of Foxo3 deficiency is not directly testable in humans.

Foxo1/3/4 single KOs — Hosaka 2004

Hosaka et al. 2004 generated individual germline knockouts of Foxo1, Foxo3a, and Foxo4 ⁷:

Foxo1-/-: embryonic lethal at E10.5 due to vascular defects (disorganized dorsal aorta, intersomitic vessels, absent yolk sac vasculature)
Foxo3a-/- females: premature ovarian failure (see above) — same paper confirms the Castrillon 2003 ovarian finding; all oocytes degenerated by 12 weeks
Foxo4-/-: grossly normal appearance, no consistent histological abnormalities in any tissue examined

The distinct phenotypes illustrate that FOXO paralogs have largely non-redundant in vivo roles. The paper does NOT describe a triple KO.

Attribution note: The “Foxo1/3/4 triple KO → hemangiomas and thymic lymphomas” phenotype frequently cited in the FOXO literature derives from Paik et al. 2007 (Cell 128:309–323) and/or Tothova et al. 2007 (Cell 128:325–339), not from Hosaka 2004. These papers used conditional (Cre-lox) triple deletion strategies. unsourced — these triple-KO claims require a separate citation to Paik 2007 or Tothova 2007 if added to this page.

Connection to C. elegans longevity

In caenorhabditis-elegans, the FOXO3 ortholog daf-16 is the primary transcriptional effector of the insulin/IGF-1 signaling (IIS) pathway. Mutations that reduce the worm IIS receptor daf-2 (the insulin/IGF-1 receptor ortholog) extend lifespan by up to 100% in worms — a lifespan extension that is entirely dependent on DAF-16/FOXO3. This is the most genetically clean demonstration that FOXO ortholog activation drives longevity in a metazoan.

Dimension	Status
Pathway conserved in humans?	yes — IIS→FOXO axis highly conserved
Lifespan extension via FOXO activation replicated in humans?	no — not testable directly; longevity GWAS provides indirect evidence
Effect magnitude in mammals?	much smaller — no analogous gain-of-function mammalian lifespan extension published

needs-human-replication — the remarkable lifespan extension seen in daf-16-dependent IIS reduction in worms has no equivalent magnitude in any mammalian model. The mechanism is conserved but effect sizes diverge dramatically.

Aging context — why FOXO3 matters

FOXO3 occupies a uniquely central position in aging biology:

Downstream effector of the most-replicated pro-aging pathway: insulin-igf1 → pi3k-akt-pathway → AKT → pFOXO3 (inactive). Reducing IIS activity extends lifespan in worms, flies, and mice — FOXO3/daf-16 activation is either necessary or sufficient for most of this extension.
Multi-hallmark integrator: Active FOXO3 simultaneously addresses deregulated-nutrient-sensing, disabled-macroautophagy (autophagy gene induction), and cellular stress resistance (antioxidant genes), positioning it as a master regulator across multiple hallmarks of aging.
Only longevity gene with >5 independent human cohort replications: The rs2802292 G allele association with exceptional longevity is the most-replicated human longevity genetic finding, distinguishing FOXO3 from most other candidate longevity genes.
Context-dependency limits therapeutic targeting: The same FOXO3 activation that promotes stress resistance and autophagy at low/moderate levels also drives atrophy (atrogenes) at sustained high levels. A pharmacological FOXO3 activator would need tissue-selectivity or dose-control not currently achievable.

Pathway membership

insulin-igf1 — primary upstream suppressor via IGF-1R → IRS → PI3K → AKT → pFOXO3
pi3k-akt-pathway — AKT is the proximal FOXO3 kinase (Thr32/Ser253/Ser315)
autophagy — FOXO3 transactivates BECN1, BNIP3, ATG12, LC3B, GABARAPL1
ampk — AMPK phospho-activates FOXO3 (Ser30) under energy stress
sirtuin — SIRT1 deacetylates FOXO3; SIRT3 interacts with FOXO3 in mitochondria
apoptosis — FOXO3 transactivates BIM, PUMA, FasL, TRAIL
dna-damage-response — FOXO3 transactivates GADD45A/B

Key interactors

akt — primary inhibitory kinase (Thr32/Ser253/Ser315)
14-3-3 — cytoplasmic retention after AKT phosphorylation
sirt1 — deacetylates FOXO3; activating; connects FOXO3 to caloric-restriction signaling
mst1 — phosphorylates Ser209 under oxidative stress; activating (AKT-bypass)
bim — direct transcriptional target; connects FOXO3 to intrinsic apoptosis; verified ²
beclin-1 — direct transcriptional target; connects FOXO3 to autophagy initiation

Limitations and gaps

no-mechanism — exact functional consequence of rs2802292 G allele (splicing? expression? chromatin?) is unresolved
needs-human-replication — FOXO3 activation sufficient to extend lifespan demonstrated only in invertebrates; mammalian data is suggestive (KO phenotypes, GWAS) but not mechanistically complete
dose-response-unclear — the FOXO3 activation threshold that tips from beneficial (autophagy, stress resistance) to harmful (atrogene-driven muscle wasting) is not quantified
contradictory-evidence — FOXO4-p53 axis in senescent cells (Baar et al. 2017 FOXO4 interfering peptide as senolytic) highlights that different FOXO paralogs can have opposing roles in specific aging contexts; FOXO3/FOXO4 interaction in senescence is incompletely characterized
unsourced — the precise tissue hierarchy of FOXO3 expression changes with human aging (age-related reduction vs. increase vs. tissue-specific) needs citation from Aging Atlas / GTEx data
long-term-unknown — no completed randomized trial of an intervention specifically targeting FOXO3 activation as a longevity endpoint

Footnotes

doi:10.1016/S0092-8674(00)80595-4 · n=N/A (biochemical/cell study) · in-vitro · model: CCL39 fibroblasts (Chinese hamster lung), 293T cells, cerebellar granule neurons, Jurkat T cells · Cell 1999; local PDF available · First demonstration that AKT directly phosphorylates FKHRL1/FOXO3 at Thr32/Ser253/Ser315 in cells; Thr32 and Ser253 create 14-3-3 docking surfaces (S315A mutant retains 14-3-3 binding; T32A/S253A double mutant abolishes it); cytoplasmic sequestration confirmed; primary death gene identified as Fas ligand (FasL) ↩
doi:10.1016/s0960-9822(00)00728-4 · n=N/A (cell study) · in-vitro · model: IL-3-dependent hematopoietic cells (Ba/F3) · Current Biology 2000; 965 citations; local PDF available · FKHR-L1 (FOXO3) transactivates BIM (BCL2L11) promoter; BIM induction is required for FKHRL1-driven apoptosis; also verified on bim page ↩ ↩²
doi:10.1038/nature01036 · n=N/A (cell study) · in-vitro · model: DL23 (inducible FOXO3a human colon carcinoma line), MEFs (Sod2-/- and wild-type), 3T3-L6 cells, 293T cells · Nature 2002; local PDF available · FOXO3a-deficient cells accumulate elevated ROS and are hypersensitive to glucose-deprivation-induced apoptosis; FOXO3 directly activates MnSOD/SOD2 transcription via an inverse FOXO-binding element (DBE1) in the SOD2 promoter, confirmed by ChIP; catalase not directly demonstrated in this paper ↩ ↩²
doi:10.1016/j.cmet.2007.11.001 · n=N/A (mouse in vivo) · in-vivo · model: mouse skeletal muscle tibialis anterior (electroporation-based gene delivery); C2C12 cells; MEFs · Cell Metabolism 2007; local PDF available · FoxO3 constitutively active form sufficient to induce autophagy vacuole formation and muscle atrophy in vivo; dominant-negative FoxO3 blocks fasting/denervation-induced autophagy; primary direct transcriptional targets confirmed by ChIP and reporter assay: LC3 (MAP1LC3B), Bnip3, Bnip3l, Gabarapl1; Beclin-1 and Atg12 shown induced at 3-day denervation (Supplemental Fig S1, less prominently than LC3/Bnip3); atrogin-1 and MuRF1 controlled by FoxO3 via the ubiquitin-proteasomal pathway (independently of autophagy) ↩ ↩²
doi:10.1073/pnas.0801030105 · n=615 total (213 cases survived to ≥95y, mean attained age 97.9y; 402 controls died before age 81, mean attained age 78.5y) · nested case-control · p=0.0007 (GG vs TT at rs2802292; OR=2.75, 95% CI 1.51–5.02); TG vs TT OR=1.91 (95% CI 1.34–2.72, P=0.0003) · model: human (Japanese-American men; Honolulu Heart Program / HAAS cohort, Oahu) · PNAS 2008; local PDF available · GG homozygotes associated with exceptional longevity; also verified on pi3k-akt-pathway page ↩
doi:10.1126/science.1086336 · n=~N/A (Foxo3-/- mouse colony) · in-vivo · model: Foxo3-null mouse (C57BL/6) · Science 2003; 932 citations; not_oa (closed access) · no-fulltext-access · Foxo3-null females show constitutive primordial follicle activation leading to premature ovarian failure; Foxo3-null mice also develop lymphadenopathy and splenomegaly ↩ ↩²
doi:10.1073/pnas.0400093101 · n=N/A (germline single KO mouse colonies) · in-vivo · model: Foxo1-/-, Foxo3a-/-, Foxo4-/- individual germline KO mice (C57BL/6 and 129/Sv backgrounds) · PNAS 2004; local PDF available · Foxo1-/- embryonic lethal (E10.5; vascular defects); Foxo3a-/- females show premature ovarian failure (all oocytes degenerated by 12 weeks); Foxo4-/- grossly normal; no triple KO in this paper — the hemangioma/lymphoma triple-KO phenotype is from Paik et al. 2007 (Cell) or Tothova et al. 2007 (Cell) ↩ ↩²

🧬 Aging Wiki

Explorer

foxo3