FGF signaling pathway

The Fibroblast Growth Factor (FGF) signaling pathway is a family of 22 secreted ligands (FGF1–FGF23, excluding FGF15 which is the mouse ortholog of human FGF19) that signal through four receptor tyrosine kinases (FGFR1–4). The pathway governs tissue development, repair, metabolic homeostasis, and — most relevant to aging — systemic endocrine crosstalk between organs. Three FGF subfamilies operate via distinct mechanisms: endocrine FGFs (FGF19/21/23) circulate through blood and require Klotho co-receptors; paracrine FGFs (FGF1–10 family) act locally and require heparan sulfate proteoglycans; and intracrine FGFs (FGF11–14) are non-secreted and modulate sodium channels. Aging biology is dominated by the endocrine subclass: fgf21 emerges as a systemic stress signal induced by mitochondrial dysfunction; fgf23 elevation is an early biomarker of kidney aging and cardiovascular risk; and klotho decline — the obligate co-receptor for FGF23 — phenocopies accelerated aging when genetically ablated.

FGF subfamily classification

Subfamily	Members	Range	Mechanism	Co-factor	Aging relevance
Endocrine	FGF19, FGF21, FGF23	Systemic (blood)	Hormone-like; low affinity for HSPG	αKlotho (FGF23) / βKlotho (FGF19, FGF21)	High — metabolic regulation, CKD/cardiovascular, longevity signaling
Paracrine	FGF1–10 (incl. FGF2/bFGF, FGF7/KGF)	Local (tissue)	Autocrine/paracrine; HSPG-stabilized	Heparan sulfate proteoglycans (HSPGs)	Moderate — vascular repair (FGF2), epithelial maintenance (FGF7)
Intracrine	FGF11–14	Intracellular	Non-secreted; no receptor signaling	None	Low — sodium channel modulation; not primary aging targets

KEGG hsa04010 covers FGF members as upstream activators of the broader MAPK cascade; the FGF-specific receptor signaling is organized under Reactome R-HSA-190236 with four FGFR sub-branches (R-HSA-5654736 through -5654743).

Receptors: FGFR1–4

FGFRs are receptor tyrosine kinases (RTK family) with a tripartite extracellular immunoglobulin-like domain structure (Ig-I, Ig-II, Ig-III), a single transmembrane helix, and a split intracellular tyrosine kinase domain. Ligand binding induces receptor dimerization and trans-autophosphorylation of the activation loop.

Critical splice variants: FGFR1–3 each exist as IIIb and IIIc isoforms (the Ig-III domain exon), generated by alternative splicing. The IIIb isoforms are expressed in epithelial cells and bind paracrine FGFs from the underlying mesenchyme; IIIc isoforms are expressed in mesenchyme and bind epithelial-derived FGFs. This splicing creates a reciprocal signaling logic during organogenesis. FGFR4 has no IIIb/IIIc variants.

Co-receptors for endocrine FGFs:

αKlotho (klotho): obligate co-receptor for FGF23 at FGFR1c. Without αKlotho, FGF23 cannot bind or activate its receptor at physiological concentrations. The klotho knockout mouse phenocopies aging (hyperphosphatemia, ectopic calcification, muscle wasting, premature death) ¹. Soluble αKlotho also acts as a direct signaling molecule independent of FGF23.
βKlotho: obligate co-receptor for FGF19 and FGF21 at FGFR1c/FGFR4. βKlotho expression is restricted to liver, adipose, and brain — accounting for the tissue-specific effects of hepatic FGF21 signaling. needs-canonical-id — βKlotho gene (KLB) protein page not yet seeded in this wiki.

Downstream signaling cascade

FGFR dimerization and trans-autophosphorylation initiates three main effector arms:

FGF + FGFR(+co-receptor) → FGFR dimerization → trans-autophosphorylation
         ↓
    FRS2α (docking protein) phosphorylation
         ↓
   ┌─────────────────┬─────────────────┐
   ↓                 ↓                 ↓
GRB2/SOS → RAS    GRB2/GAB1 →       PLCγ activation
   ↓               PI3K               ↓
RAS-MAPK          ↓                  DAG + IP3
(ERK1/2)       PI3K-AKT            PKC + Ca²⁺

RAS-MAPK arm (ras-mapk): FRS2α recruits GRB2/SOS → RAS-GTP → RAF → MEK → ERK1/2 → proliferation, differentiation, survival.
PI3K-AKT arm (pi3k-akt-pathway): GRB2/GAB1 scaffold recruits PI3K → PIP3 → AKT → mTORC1 (anabolic), BAD phosphorylation (survival), FOXO inhibition. This arm links FGF signaling to nutrient-sensing pathways.
PLCγ arm: Direct phosphorylation of PLCγ by FGFR → DAG (protein kinase C activation) + IP3 (calcium release). Relevant for FGF2-driven angiogenesis and FGF23-driven cardiac hypertrophy (FGF23 activates PLCγ in cardiomyocytes through FGFR4 independent of Klotho).

Key negative regulator: Sprouty proteins (SPRY1–4) are transcriptionally induced by FGF-ERK signaling and act as feedback inhibitors upstream of RAS. CBL family E3 ubiquitin ligases promote FGFR ubiquitination and endocytosis, attenuating receptor-level output.

FGF21 — starvation hormone and longevity signal

fgf21 is a liver-derived (primarily) endocrine factor induced by fasting, ketogenic diet, mitochondrial stress, and exercise. It signals to adipose tissue (lipolysis), brain (FGF21/βKlotho in hypothalamus: appetite suppression, thermogenesis), and heart.

Lifespan extension

FGF21 transgenic mice (C57Bl/6J background, FGF21 expressed from hepatocyte-selective apoE promoter) live longer than wild-type littermates, with the magnitude sex-dependent: ~30% longer in males (median 27.9 → 36.2 months) and ~39% longer in females (median 28.8 → 40.1 months); combined-sex median extended from 28.1 to 38.2 months (~36%) ². Hazard ratio (combined): 0.22 [95% CI 0.15–0.34], p=2.7×10⁻¹². Mice also showed reduced body fat, improved insulin sensitivity, and suppressed GH/IGF-1 axis activity. The longevity effect is associated with — and possibly mediated by — growth axis suppression (FGF21 inhibits GH signaling at the hepatic level via JAK2/STAT5 inhibition → reduced IGF-1), not merely metabolic improvement. Notably, AMP kinase, mTOR, and sirtuin pathway markers were not elevated in FGF21-Tg tissues, ruling out those canonical longevity pathways as the mechanism.

Dimension	Status
Pathway conserved in humans?	yes
Longevity phenotype conserved in humans?	unknown — no human population data on FGF21 overexpression
Replicated in humans?	no — lifespan extension is mouse data only; metabolic effects (insulin sensitivity, weight) have human phase 2 analogue data

FGF21 as a mitochondrial stress marker (mitokine)

FGF21 is induced by the integrated-stress-response (ISR): mitochondrial stress causes cytosolic accumulation of mitochondrial RNAs, activating the eIF2α kinase PKR → eIF2α phosphorylation → selective translation of ATF4 → Fgf21 gene expression. CoQ deficiency specifically in brown adipose tissue (BAT) activates the PKR-eIF2α-ATF4-FGF21 axis; BAT-secreted FGF21 then drives systemic mitohormesis — increased whole-body metabolic rate and thermogenesis independently of UCP1 ³. FGF21 induction thus serves as a cell-nonautonomous (mitokine) signal that mitochondrial function is compromised in a tissue; elevated circulating FGF21 in humans correlates with mitochondrial myopathies and metabolic disease.

In aging, chronically elevated FGF21 may indicate sustained mitochondrial stress rather than adaptive longevity signaling. The acute vs. chronic distinction is biologically important and clinically underexplored. no-mechanism

Thymic aging (2025 data)

Two independent 2025 Nature Aging studies (different research groups, different model systems) established paracrine FGF21 roles in thymic maintenance:

Wedemeyer 2025 (closed-access, unverifiable): paracrine FGF21 modulates mTOR signaling in TECs to regulate thymus function across the lifespan; mTEC-derived FGF21 maintains cortical TEC function and T-cell output ⁴. no-fulltext-access — specific outcomes (rapamycin reversal claim, quantitative TEC numbers) cannot be verified against primary PDF.
Youm 2025 (Dixit lab, Yale): elevation of FGF21 specifically in TECs (FoxN1Cre-iFGF21Tg) protected cortical and medullary architecture in 24-month-old mice, increased naive CD8 T cells, and improved grip strength (healthspan marker); elevation in adipocytes (Adn-iFGF21Tg) independently protected against HFD-induced and age-related thymic involution. Conditional ablation of βKlotho in Foxn1⁺ TECs (FoxN1Cre:Klb^fl/fl) accelerated age-related loss of thymic cellularity and naive T cells, confirming FGF21/βKlotho signaling is partially required for TEC maintenance ⁵. n=6–13/group; hepatocyte-derived endocrine FGF21 at 5× elevation did not impact thymic biology, establishing the paracrine rather than endocrine source as essential.

needs-human-replication — both thymus studies are mouse models; whether FGF21 analogs can reverse human thymic aging is untested.

FGF21 analogs in clinical development

FGF21 has a short half-life (~1.5 h) and is subject to proteolytic degradation, motivating engineered analogs:

Efruxifermin (AKero Therapeutics) — phase 2/3 in MASH/NASH; reduces hepatic fat, fibrosis
Pegbelfermin (Bristol-Myers Squibb) — phase 2 in NASH; hepatic efficacy shown
LY3502970 (Eli Lilly) — bispecific GLP1R/FGF21R agonist; phase 2 in obesity/T2D

None are in aging-indication trials. The closest aging rationale is metabolic / NAFLD (which overlaps deregulated-nutrient-sensing), not explicit geroprotection. needs-human-replication

FGF23 / Klotho axis — CKD-aging intersection

fgf23 is secreted by osteocytes in bone in response to dietary phosphate load and 1,25-dihydroxyvitamin D. Its canonical role is phosphate homeostasis: FGF23 acts on kidney (via αKlotho co-receptor) to suppress phosphate reabsorption and inhibit CYP27B1 (vitamin D activation). The axis becomes pathological in chronic kidney disease (CKD): declining kidney function → reduced αKlotho expression → FGF23 rises to compensate → persistently elevated FGF23 drives:

Left ventricular hypertrophy (via FGFR4/PLCγ in cardiomyocytes, Klotho-independent)
Vascular calcification (phosphate/calcium dysregulation)
Endothelial dysfunction

Elevated FGF23 and reduced Klotho are early biomarkers in CKD and correlate with cardiovascular mortality in the general aging population ⁶. Klotho decline is not restricted to CKD — circulating soluble αKlotho declines ~10% per decade from mid-life in otherwise healthy humans, suggesting a normative aging trajectory.

Dimension	Status
Pathway conserved in humans?	yes
FGF23 elevation in aging?	yes — seen in healthy aging; amplified in CKD
Replicated in humans?	yes — multiple observational cohorts

The aav-klotho gene therapy intervention (preclinical) aims to restore Klotho expression and reset the FGF23/Klotho axis.

Paracrine FGFs in tissue repair

FGF2 / bFGF: potent mitogen for endothelial cells and smooth muscle; drives angiogenesis in wound healing. Declines in aged tissue. No heparan sulfate → no FGFR signaling, explaining why HS-deficient tumors are refractory to FGF2.
FGF7 / KGF (Keratinocyte Growth Factor): acts through FGFR2-IIIb exclusively on epithelial cells; drives mucosal repair (gut epithelium, skin, lung). A recombinant form (palifermin) is FDA-approved to reduce oral mucositis in chemotherapy patients — the only approved FGF-family agent.

Both are less central to systemic aging than endocrine FGFs, but relevant to stem cell exhaustion in epithelial compartments and wound healing capacity.

Pharmacology and druggability

Druggability tier: 2 (aging-context; Open Targets Platform as of 2026-05-07 shows FGF21 [ENSG00000105550] and FGFR1 [ENSG00000077782] both at “Approved Drug” for SM/AB/PR modalities — tier 1 by max-druggability — but no approved drug targets the FGF21 axis for an aging indication, hence aging-context tier 2). FGFR inhibitors are oncology-approved but not aging-validated:

Erdafitinib — pan-FGFR1-4 inhibitor; FDA-approved for FGFR-altered bladder cancer (2019)
Pemigatinib — FGFR1-3 inhibitor; approved for cholangiocarcinoma with FGFR2 fusion
Infigratinib — FGFR1-3; approved for cholangiocarcinoma

Oncology-tier inhibition (continuous, high-dose) is not appropriate for aging: FGFR inhibitors cause hyperphosphatemia (by blocking FGF23 signaling), ectopic calcification, and epithelial toxicity. A longevity application — if ever pursued — would require tissue-selective agonism (FGF21 axis) rather than inhibition.

No FGFR inhibitor has aging-indication data. The aging-relevant pharmacological vector is FGF21 agonism (analogs above), which is metabolic/NASH phase 2. Aging-context tier is 2 (clinical-stage analog probes exist for the FGF21 arm; the FGF23/Klotho arm has recombinant Klotho protein in early-stage development). Contrast with maximum druggability of oncology FGFR inhibitors (tier 1 in oncology context), which is why the aging-context tier is lower.

Connection to aging hallmarks

altered-intercellular-communication (primary): FGF21 and FGF23 are paradigmatic endocrine aging signals. FGF21 is a mitokine — a cell-nonautonomous stress signal. FGF23/Klotho axis is a tissue-crosstalk system between bone, kidney, and heart that degrades with age.
deregulated-nutrient-sensing: FGF21 represses insulin-igf1 / GH axis signaling; the FGF21-PI3K-AKT arm intersects mTOR regulation. FGF21 transgenic longevity likely involves reduced IIS/mTOR activity.
chronic-inflammation: FGF23 excess contributes to systemic pro-inflammatory signaling. FGF2 from senescent fibroblasts is a reported SASP component. needs-replication — FGF2 in SASP is based on limited in-vitro data.

Limitations and knowledge gaps

Human longevity translation: the FGF21 transgenic mouse lifespan extension is one of the cleanest single-factor longevity results in aging biology, but there is no human population evidence that elevated FGF21 extends lifespan — paradoxically, high circulating FGF21 in humans correlates with metabolic disease (an indicator of stress, not health). The mouse-to-human direction of the FGF21 longevity effect is unresolved. needs-human-replication
FGF21 analog aging trials: all clinical FGF21 analog programs are targeting MASH/metabolic disease, not aging per se. No aging-indication RCT has been run. needs-human-replication
αKlotho supplementation: recombinant soluble Klotho and AAV-Klotho (aav-klotho) are preclinical. Human aging indication is distant. needs-human-replication
FGF19 in aging: minimal data; FGF19 regulates bile acid homeostasis and hepatic glucose metabolism but does not appear in aging-longevity literature. Likely a minor player. unsourced
Intracrine FGFs (FGF11-14) in aging: sodium channel modulation in neurons; relevance to age-related neurodegeneration is speculative and uncited. unsourced
FGF23 causality vs correlation: elevated FGF23 predicts cardiovascular mortality in CKD but Mendelian randomization evidence for a causal role of FGF23 in general-population cardiovascular aging is limited. needs-replication

Cross-references

Implicit stubs (pages not yet seeded): fgfr1, fgfr4, frs2, fgf2, fgf7, klb (βKlotho), ppar-alpha, atf4, phex, plcgamma

Footnotes

doi:10.1038/36285 · Kuro-o M et al. · Nature 1997 · in-vivo (mouse) · Klotho knockout mice show premature aging syndrome: ectopic calcification, skin atrophy, arteriosclerosis, osteoporosis; n=~50 homozygotes · cited 3,766× (archive: not_oa, closed-access) · no-fulltext-access — the widely-cited “~61-day median lifespan” figure for Klotho-KO mice is consistent with secondary literature but is NOT verifiable from the primary PDF (closed-access); should be re-anchored to a verified secondary source on next pass ↩
doi:10.7554/elife.00065 · Zhang Y et al. · eLife 2012 · in-vivo (mouse, C57Bl/6J) · FGF21-Tg mice (n=77 Tg, n=67 WT; combined male+female) live longer than WT: combined median 28.1→38.2 months; males ~30% longer (27.9→36.2 mo), females ~39% longer (28.8→40.1 mo); HR=0.22 [0.15–0.34], p=2.7×10⁻¹²; log-rank test; mechanism: JAK2/STAT5 suppression → reduced IGF-1; AMP kinase/mTOR/sirtuin markers not elevated · cited 408× (archive: gold OA, PDF verified locally) ↩
doi:10.1038/s44318-023-00008-x · Chang CF, Gunawan AL et al. · The EMBO Journal 2024 · in-vivo + in-vitro (mouse BAT; murine brown adipocytes; human beige adipocytes) · CoQ deficiency in BAT triggers cytosolic accumulation of mitochondrial RNAs → PKR activation → eIF2α phosphorylation → ISR → ATF4-FGF21 axis; BAT-secreted FGF21 drives whole-body increased metabolic rate (mitohormesis) independently of UCP1; ATF4 KO abolishes FGF21 induction; genetic model: BAT-specific PDSS2 KO (PDSS2^BKO) in C57 background · cited 8× (archive: diamond OA via PMC10897314, PDF verified locally) ↩
doi:10.1038/s43587-024-00801-1 · Wedemeyer SA et al. · Nature Aging 2025 · in-vivo (mouse) · paracrine FGF21 modulates mTOR signaling to regulate thymus function across the lifespan · cited 2× (archive: closed-access, not_oa — specific claims unverified against primary PDF) no-fulltext-access ↩
doi:10.1038/s43587-025-00813-5 · Youm YH, Gliniak C et al. (Dixit lab, Yale) · Nature Aging 2025 · in-vivo (mouse) · TEC-specific FGF21 overexpression (FoxN1Cre-iFGF21Tg) protected thymic architecture and increased naive CD8 T cells in 24-month-old mice; adipocyte-specific FGF21 overexpression (Adn-iFGF21Tg) independently protected against age/HFD-induced thymic involution; hepatocyte-derived FGF21 (5× elevation) did not affect thymus; βKlotho ablation in Foxn1⁺ TECs accelerated thymic aging; n=6–13/group · cited 6× (archive: PMC12003152, PDF verified locally) ↩
doi:10.1016/j.bone.2016.11.012 · Yamada S, Giachelli CM · Bone 2017 · review · FGF23/Klotho axis in CKD-mineral bone disorder; vascular calcification mechanisms · cited 308× (archive: green OA / PMC5429216; pending download) ↩

🧬 Aging Wiki

Explorer

fgf-signaling