FOXO4 (Forkhead Box Protein O4)

FOXO4 is the third mammalian FOXO paralog — sharing the canonical AKT-regulated cytoplasmic sequestration mechanism with foxo1 and foxo3, but distinguished by its central role in senescent cell survival via a FOXO4-p53 protein-protein interaction. Unlike FOXO1 (hepatic metabolic axis) or FOXO3 (multi-tissue longevity genetics), FOXO4’s most aging-relevant function is in maintaining senescent cells alive: in senescent cells, FOXO4 accumulates and interacts with p53 to prevent p53-driven apoptosis, thereby sustaining the senescent state and its associated SASP. The FOXO4-DRI senolytic peptide (Baar 2017) exploits this mechanism by disrupting the FOXO4-p53 interaction, releasing p53 to the cytoplasm and triggering selective apoptosis of senescent cells — a major proof-of-concept for protein-protein interaction senolytics.

Identity

UniProt: P98177 (FOXO4_HUMAN; reviewed Swiss-Prot entry)
NCBI Gene: 4303
HGNC symbol: FOXO4 (previously AFX, AFX1; chromosomal translocation designation MLLT7)
Mouse ortholog: Foxo4 (one-to-one ortholog)
Length: 505 amino acids (canonical isoform)
Chromosomal location: Xq13.1; involved in t(X;11)(q13;q23) translocation in acute leukemias
GenAge: No human entry confirmed needs-canonical-id; no models entry confirmed — FOXO4 aging evidence currently not curated in GenAge despite Baar 2017

Nomenclature note: The gene was originally identified as AFX (Acute lymphocytic leukemia [or AFX forkhead-domain gene]) and designated MLLT7 (Myeloid/Lymphoid or Mixed-Lineage Leukemia Translocation To 7) in leukemia contexts. Current HGNC-approved symbol is FOXO4. All three names appear in cited literature; this page uses FOXO4 throughout.

FOXO family context

Four mammalian FOXO paralogs, all regulated by AKT at three conserved phosphorylation sites:

Paralog	Old name	Primary tissue context	Key aging role
foxo1	FKHR	Liver, adipose, beta-cells	Hepatic gluconeogenesis; no strong longevity GWAS signal
foxo3	FKHRL1	Ubiquitous; brain, heart	Strongest human longevity GWAS (rs2802292); canonical longevity TF
FOXO4	AFX / MLLT7	Ubiquitous	Senescent cell survival via FOXO4-p53 interaction; FOXO4-DRI senolytic peptide
FOXO6	—	Brain-restricted	Memory consolidation; minimal aging literature

FOXO4 is the most senescence-associated FOXO paralog — its expression is selectively elevated in senescent cells relative to proliferating and quiescent cells ¹.

Domain organization

Region	Residues (approx.)	Function
N-terminal regulatory domain	1–99	Contains AKT phospho-site Thr32; 14-3-3 docking
Forkhead (winged-helix) DNA-binding domain	100–188	Sequence-specific DNA binding to FHRE consensus; contains Ser197
Nuclear export sequence (NES)	~200–220 (approx.)	CRM1-dependent cytoplasmic export after phosphorylation
C-terminal transactivation domain	~220–505	Contains Ser262 AKT site; transcriptional activation; acetylation by CREBBP/CBP; deacetylation by SIRT1

Residue boundaries from UniProt P98177 feature table (forkhead domain 100–188 confirmed); other boundaries approximate. unsourced — precise NES boundaries should be cross-checked against UniProt feature table and primary structural literature.

Note on phospho-site numbering: UniProt P98177 lists AKT phosphorylation sites as Thr32, Ser197, Ser262 — confirmed against the live UniProt P98177 feature table (evidence: ECO:0000269|PubMed:16272144 for Thr32; PubMed:10217147, 10518537, 16272144 for Ser197 and Ser262). These are analogous to FOXO3 Thr32/Ser253/Ser315 and FOXO1 Thr24/Ser256/Ser319 by sequence homology. An alternate numbering (Thr28/Ser193/Ser258) appears in older literature and may reflect a different isoform annotation; the current canonical UniProt canonical-isoform numbering is Thr32/Ser197/Ser262.

AKT phosphorylation and cytoplasmic sequestration

AKT (PKB) phosphorylates FOXO4 at Thr32, Ser197, and Ser262 — the three conserved sites shared across the FOXO paralog family. By analogy with the directly demonstrated FOXO3 (FKHRL1) mechanism ²:

Insulin/IGF-1 → pi3k-akt-pathway → AKT activation
AKT phosphorylates FOXO4 at Thr32 and Ser197 → recruits 14-3-3 proteins → cytoplasmic sequestration
Phospho-Ser262 drives CRM1-dependent nuclear export
Net result: FOXO4 transcriptionally inactive; proliferative/survival gene expression dominates

Dephosphorylation (under fasting, PTEN activity, or PI3K inhibition) allows nuclear re-import. UniProt P98177 notes that monoubiquitination under oxidative stress also promotes nuclear retention.

Dimension	Status
Pathway conserved in humans?	yes
Cytoplasmic sequestration in fed state conserved?	yes (by analogy with FOXO1/3; direct human data for FOXO4 specifically is limited)
Replicated in humans?	partial — mechanism confirmed in cell lines; in-vivo human data limited

needs-replication — direct evidence for AKT→FOXO4 nuclear exclusion in primary human cells (vs. cell lines) is limited compared with FOXO1/3.

Transcriptional targets

FOXO4 shares the core FHRE binding motif with FOXO1 and FOXO3 and likely activates overlapping target genes. Confirmed or well-supported targets include:

Gene	Product	Context
CDKN1B (p27/Kip1)	CDK inhibitor	G1/S cell-cycle arrest
BCL2L11 (bim)	BIM pro-apoptotic BH3-only protein	Apoptosis induction (shared with FOXO3)
IGFBP1	IGF-binding protein 1	Insulin-response element binding (earliest characterized FOXO4 target)
HIF1A	HIF-1alpha	FOXO4 suppresses HIF1A expression; hypoxia-response modulation

unsourced — FOXO4-specific target gene ChIP-seq data distinguishing FOXO4 from FOXO1/3 binding sites in aging or senescent cells is limited; most target gene claims are inferred from paralog family studies. A FOXO4-specific ChIP-seq in senescent cells would be highly informative.

Discovery — AFX/MLL translocation in acute leukemia

FOXO4 was identified as the gene at the Xq13.1 breakpoint in the t(X;11)(q13;q23) translocation in acute leukemias ³. This translocation fuses the N-terminal portion of FOXO4 (then called AFX) to the C-terminal SET/histone methyltransferase domain of KMT2A/MLL1, creating a constitutively active “rogue activator” fusion protein. The AFX designation — and the leukemia context — preceded the protein’s identification as a FOXO family member and a downstream AKT substrate. A parallel rearrangement fuses FOXO3 (then AF6q21) to MLL1 in secondary acute leukemia at 6q21, establishing that multiple FOXO paralogs are recurrent leukemia oncoproteins when fused to MLL1.

Dimension	Status
Cancer relevance of FOXO4-MLL1 fusion?	established in rare pediatric/secondary acute leukemias
Normal FOXO4 function in hematopoiesis?	maintains HSC quiescence / stress resistance as part of FOXO1/3/4 redundant network ⁴

Knockout phenotype — Foxo4-/- mice

Foxo4-null mice (Foxo4-/-) are viable and grossly normal with no consistent histological abnormalities in any tissue examined ⁵. This is in stark contrast to:

Foxo1-/-: embryonic lethal at E10.5 from vascular defects
Foxo3-/- females: premature ovarian failure

The lack of overt Foxo4-/- phenotype suggests that FOXO4’s functions are largely compensated by FOXO1 and/or FOXO3 in normal development. The FOXO4-specific phenotype (senescent cell survival) emerges primarily in aged, post-mitotic, or chemotherapy-stressed contexts that were not assessed in the original germline knockout study.

Triple Foxo1/3/4 conditional KO: Conditional deletion of all three somatic FOXO paralogs in the hematopoietic compartment (Mx1-Cre; Mx1-Cre+;FoxO1/3/4L/L strategy, induced by poly-IC at 4 weeks) causes severe hematopoietic stem cell (HSC) defects: a 4.6-fold decrease in the LSK (Lin-Sca-1+c-Kit+) compartment, myeloid lineage expansion (leukocytosis with relative neutrophilia), lymphoid developmental abnormalities, and impaired long-term repopulation capacity ⁴. HSCs accumulate elevated reactive oxygen species (ROS) (~2.5-fold increase, p<0.0001), and this phenotype is rescued by N-acetyl-L-cysteine (NAC) treatment in vivo. Critically, single-KO Foxo4-/- mice in the same hematopoietic conditional system show no HSC phenotype — only the triple-KO fully manifests the cell-cycle and apoptosis defects, establishing functional redundancy among all three paralogs in this compartment. This establishes FOXO1/3/4 functional redundancy in HSC maintenance — the individual Foxo4-/- showing no phenotype is explained by FOXO1/3 compensation.

Dimension	Status
Single Foxo4-/- phenotype in mice?	grossly normal — no overt phenotype ⁵
Triple Foxo1/3/4 conditional KO in HSCs?	severe HSC defect + ROS accumulation ⁴
Human equivalent?	no equivalent genetic data; FOXO4 variants not associated with specific human aging phenotypes

FOXO4-DRI senolytic peptide (Baar 2017) — major aging-relevant finding

This is the most significant aging-specific finding for FOXO4 and distinguishes it from its paralogs.

Background: FOXO4 in senescent cell survival

In senescent cells — whether induced by oncogene activation, genotoxic stress (doxorubicin), or replicative exhaustion — FOXO4 protein levels are selectively elevated relative to proliferating cells ¹. FOXO4 interacts directly with p53 in senescent cells, sequestering p53 in a nuclear complex that prevents p53 from translocating to the cytoplasm and mitochondria to drive apoptosis. This FOXO4-p53 interaction thus functions as a pro-survival mechanism specific to senescent cells — by sustaining p53 in a transcriptionally active but pro-apoptotic-output-inhibited state. The net result: senescent cells are resistant to spontaneous apoptosis, accumulate with age, and secrete the SASP that drives tissue deterioration.

This mechanism is distinct from the BH3-mimetic senolytic strategy (e.g., navitoclax, which inhibits BCL-2/BCL-xL to restore apoptotic priming) and represents a separate pro-survival axis specific to the senescent state.

FOXO4-DRI peptide design

Baar et al. 2017 designed a cell-permeable interfering peptide to disrupt the FOXO4-p53 interaction ¹:

Structure: D-amino acid retroinverso (DRI) peptide spanning FOXO4 residues 86–206 — covering the Forkhead domain and adjacent p53-interaction region; the DRI format uses D-amino acid isomers in reverse sequence, conferring protease resistance and cell permeability. The exact amino acid sequence (D-isoform): H-ltlrkepaseiaqsileaysqngwanrrsggkrppprrqrrkkrg-OH (MW 5358.2) ¹
Mechanism: FOXO4-DRI competes with endogenous FOXO4 for p53 binding → disrupts FOXO4-p53 nuclear complex → releases p53 to the cytoplasm → p53 drives mitochondrial apoptosis selectively in senescent cells
Selectivity basis: Non-senescent cells have low nuclear FOXO4 and/or p53 is in a different activation state; FOXO4-DRI preferentially triggers apoptosis in cells where both FOXO4 and activated p53 are co-elevated (the senescent state)

In vivo results (Baar 2017)

In vivo mouse models used in Baar 2017 ¹:

Doxorubicin chemotoxicity model: Wild-type C57BL/6 mice dosed with doxorubicin (0.1 µM × 2, with 2-day interval), analyzed 7 days later
Fast-aging model: XpdTTD/TTD mice (trichothiodystrophy DNA-repair model on C57BL/6 background) at 26–60 weeks of age
Naturally aged model: Wild-type C57BL/6J p16::3MR mice at 115–130 weeks of age (not “~2 years”; 115–130 weeks ≈ 2.2–2.5 years, reported as 104-week mice in Figure 7)

Key outcomes:

Senescence marker reduction: FOXO4-DRI reduced p16ink4a-driven RLUC bioluminescence and FOXO4 foci in liver cells of doxorubicin-treated and XpdTTD/TTD mice; reduced SA-β-GAL activity in tubular kidney cells of XpdTTD/TTD mice
Hair density restoration: Improved fur score (0–4 scale, from very patchy toward wild-type density) in XpdTTD/TTD mice; reduced abdominal temperature (a proxy for fur density loss) ¹
Renal function: Reduced plasma urea (primary endpoint) in XpdTTD/TTD (n=7–8/group) and naturally aged p16::3MR mice; plasma creatinine also reduced as independent confirmation (Figure 7I). The paper does not report glomerular filtration rate or creatinine clearance directly — plasma urea is the primary marker used throughout.
Fitness: Improved responsiveness to physical stimuli (scored 0–3) and increased voluntary running wheel distance (km/day) in XpdTTD/TTD mice; improved exploratory behavior in naturally aged mice ¹
Selectivity: Minimal apoptosis of non-senescent fibroblasts at tested doses; WT-p53 tumor cells also showed sensitivity (an on-target risk for therapeutic use); platelet levels not substantially affected (unlike pan-BCL inhibitors)

needs-human-replication — All FOXO4-DRI data are from mouse models (C57BL/6 doxorubicin chemotoxicity; XpdTTD/TTD fast-aging; naturally aged p16::3MR on C57BL/6J background). No human clinical trial data as of 2026.

needs-replication — Baar 2017 is a single landmark study; independent replication of the specific FOXO4-p53 mechanism and DRI peptide efficacy by other groups is limited needs-replication.

Dimension	Status
FOXO4-p53 interaction in senescent cells conserved in humans?	partial — FOXO4 and p53 are both expressed in human senescent cells; direct interaction not yet formally demonstrated in primary human tissue
Senolytic phenotype (apoptosis induction in senescent cells) conserved?	in-progress — human cell-line data supportive; in vivo human data absent
Replicated in humans?	no — no completed human trial of FOXO4-DRI as of 2026

FOXO4-DRI vs. other senolytics

FOXO4-DRI represents a distinct mechanistic class from the first-generation senolytics:

Senolytic	Primary target	Mechanism	Selectivity basis
Dasatinib + Quercetin (dasatinib + quercetin)	Multiple kinases / flavonoid pleiotropism	Multiple pro-survival pathway disruption	Broad; empirically discovered
navitoclax (ABT-263)	BCL-2, BCL-xL, BCL-w	BH3-mimetic; restores apoptotic priming	Requires BCL-2 family over-reliance
fisetin	Multiple (senomorphic + senolytic)	Not fully resolved	Empirical
FOXO4-DRI	FOXO4-p53 protein-protein interaction	Releases p53 from nuclear FOXO4 complex → cytosolic apoptosis	Mechanistically targeted to senescent cell survival node

FOXO4-DRI is notable as the first senolytic designed from mechanistic understanding of senescent cell survival rather than discovered empirically. This positions it as a proof-of-concept for a new generation of PPI (protein-protein interaction) senolytics.

Clinical status and trial pipeline

No completed Phase 1 or Phase 2 clinical trial of FOXO4-DRI as of 2026. The peptide’s pharmacokinetics (peptide delivery, bioavailability, dosing interval) and safety profile (p53 activation in WT-p53 tumor cells is a theoretical concern) remain early-stage challenges. long-term-unknown — long-term safety of periodic senolysis via FOXO4-DRI (stem cell reserve depletion, immune function) has not been evaluated.

Role in aging — summary

FOXO4’s contribution to aging biology operates through two intersecting mechanisms:

Shared FOXO family role (nutrient sensing / stress resistance): Like FOXO1 and FOXO3, FOXO4 is a downstream nuclear effector of the insulin-igf1 / pi3k-akt-pathway axis. In low-nutrient / low-growth-factor conditions, FOXO4 enters the nucleus and contributes to antioxidant and cell-cycle arrest gene expression. In aged animals with chronic insulin signaling dysregulation, FOXO4 likely participates in the broader FOXO-mediated tissue aging program alongside FOXO1/3 — though FOXO4-specific (non-redundant) contributions in normal tissue aging are poorly characterized.
Senescent cell survival (FOXO4-specific, aging-dominant): FOXO4 accumulates in senescent cells and sustains their survival by sequestering p53. This function may be the primary non-redundant aging-relevant role of FOXO4 — one not easily compensated by FOXO1 or FOXO3, and now therapeutically targetable via FOXO4-DRI.

Pathway membership

insulin-igf1 — downstream nuclear effector (INSR/IGF1R → IRS → PI3K → AKT → pFOXO4 cytoplasmic)
pi3k-akt-pathway — direct AKT substrate; Thr32/Ser197/Ser262 phosphorylation
cellular-senescence — FOXO4 sustains senescent cell survival via FOXO4-p53 interaction; FOXO4-DRI disrupts this
p53-pathway — direct protein-protein interaction with p53 in senescent cells; modulates p53 nuclear vs. cytoplasmic localization

Key interactors

akt — primary inhibitory kinase; Thr32/Ser197/Ser262 phosphorylation → cytoplasmic sequestration
14-3-3 — cytoplasmic anchor; binds phospho-Thr32 + phospho-Ser197 (by analogy with FOXO3 mechanism)
p53 — key aging-relevant interaction; FOXO4 sequesters p53 in the nucleus of senescent cells to prevent apoptosis; targeted by FOXO4-DRI ¹
sirt1 — deacetylates FOXO4 (by analogy with FOXO1/3); activating under CR/fasting; UniProt confirms CREBBP/CBP acetylation with SIRT1 deacetylation unsourced (FOXO4-specific SIRT1 deacetylation site residue numbers not confirmed from primary literature; inferred from paralog homology)
usp7 — deubiquitinase; deubiquitinates FOXO4; UniProt P98177 lists monoubiquitination (oxidative stress-responsive, promotes nuclear localization) with USP7-mediated reversal unsourced (USP7-FOXO4 interaction needs primary citation)

Limitations and gaps

#gap/needs-human-replication — All in-vivo FOXO4-DRI senolytic data from mouse models (doxorubicin chemotoxicity model; XpdTTD/TTD fast-aging model; naturally aged C57BL/6 p16::3MR mice); no completed human clinical trial as of 2026.
#gap/needs-replication — Baar 2017 (10.1016/j.cell.2017.02.031) is the single primary study establishing FOXO4-p53 senescent survival mechanism and DRI peptide efficacy; independent replication of the mechanism by other groups using complementary approaches is limited.
AKT phospho-site residue numbers resolved: UniProt P98177 canonical isoform uses Thr32/Ser197/Ser262, confirmed against live UniProt feature table (2026-05-04). Alternate numbering (Thr28/Ser193/Ser258) in some older literature likely reflects a different isoform annotation.
#gap/unsourced — FOXO4-specific transcriptional target genes in aging/senescent cells: no FOXO4-specific ChIP-seq data distinguishing FOXO4 from FOXO1/3 binding sites; most target claims inferred from paralog family studies.
#gap/unsourced — USP7-FOXO4 interaction and monoubiquitination functional consequences need primary citation beyond UniProt annotation.
#gap/needs-canonical-id — GenAge human or models entry for FOXO4 not confirmed; database does not appear to curate FOXO4 despite clear aging relevance via Baar 2017.
#gap/long-term-unknown — safety of chronic or repeated FOXO4-DRI dosing (potential depletion of p53-expressing stem cell pools; tumor suppression risk in WT-p53 cancers) has not been evaluated in long-term studies.
#gap/dose-response-unclear — optimal dosing regimen for FOXO4-DRI as a senolytic (dose, frequency, duration) not established; Baar 2017 used a single dosing protocol.
The distinction between FOXO4’s contributions vs. FOXO1/3 in normal (non-senescent) aging tissue is incompletely characterized. Foxo4-/- single KO mice show no phenotype, making functional dissection difficult without conditional or paralog-competitive strategies.

Footnotes

doi:10.1016/j.cell.2017.02.031 · Baar MP et al. 2017 · in-vivo (C57BL/6; XpdTTD/TTD fast-aging mice 26–60 wks; naturally aged p16::3MR mice 104–130 wks; doxorubicin chemotoxicity model) + in-vitro (IMR90 and WI-38 human fetal lung fibroblasts; BJ foreskin fibroblasts) · Cell 169:132-147 · FOXO4-DRI (residues 86–206 DRI peptide, MW 5358.2) disrupts FOXO4-p53 interaction → p53 nuclear exclusion → caspase-3/7-dependent apoptosis selectively in senescent cells (SI50=11.73 vs. IR-senescent IMR90; SI50=12.91 vs. doxo-senescent IMR90); in vivo: reduced plasma urea (primary kidney endpoint), improved fur density score, improved physical responsiveness, increased running wheel activity in fast-aging and naturally aged mice; platelet count unaffected; 1,365 citations (100th percentile FWCI) · archive: local PDF available ↩ ↩² ↩³ ↩⁴ ↩⁵ ↩⁶ ↩⁷ ↩⁸
doi:10.1016/s0092-8674(00)80595-4 · Brunet A et al. 1999 · in-vitro (CCL39 fibroblasts, 293T, cerebellar granule neurons) · Cell 96:857-868 · AKT phosphorylates FKHRL1/FOXO3 at Thr32/Ser253/Ser315; Thr32+Ser253 create 14-3-3 docking surfaces; Ser315 drives nuclear export independently; seminal paper establishing the AKT-FOXO paradigm — FOXO4 phospho-sites (Thr32/Ser197/Ser262 per UniProt P98177) assigned by sequence homology to this mechanism · local PDF available ↩
doi:10.1038/sj.onc.1200814 · Borkhardt A et al. 1997 · in-vivo (human acute leukemia tumor genetics) · Oncogene 14:195-202 · Cloning and characterization of AFX (FOXO4), the gene at Xq13 fusing to MLL/KMT2A in t(X;11)(q13;q23) acute leukemias; first identification of the FOXO4 locus · 242 citations · archive: not_oa (no local PDF); no-fulltext-access ↩
doi:10.1016/j.cell.2007.01.003 · Tothova Z et al. 2007 · in-vivo (Mx1-Cre+;FoxO1/3/4L/L conditional triple KO in hematopoietic compartment; C57BL/6; n=6-8 Cre+, n=3-7 Cre-) · Cell 128:325-339 · FOXO1/3/4 triple KO in HSCs → 4.6-fold LSK decrease, myeloid expansion, lymphoid abnormalities, ~2.5-fold ROS increase (p<0.0001), impaired long-term repopulation; single Foxo4-/- shows no HSC phenotype; NAC rescues HSC compartment in vivo; FOXOs are critical mediators of HSC resistance to physiologic oxidative stress · 1,533 citations (100th percentile FWCI) · archive: local PDF available ↩ ↩² ↩³
doi:10.1073/pnas.0400093101 · Hosaka T et al. 2004 · in-vivo (germline Foxo1-/-, Foxo3a-/-, Foxo4-/- single KO colonies; C57BL/6 and 129/Sv backgrounds) · PNAS 101:2975-2980 · Foxo4-/- mice grossly normal; no consistent histological abnormalities in any tissue; Foxo1-/- lethal E10.5 (vascular); Foxo3a-/- females show premature ovarian failure; no triple KO in this paper · 674 citations (100th percentile) · archive: local PDF available at (local PDF) ↩ ↩²

🧬 Aging Wiki

Explorer

foxo4