🧬 Aging Wiki

growth-hormone

Properties1
aliasesGH, somatotropin, somatotrophin, GH1, growth hormone 1, pituitary growth hormone

16 min read

Growth Hormone (GH / GH1)

The master regulator of the somatotropic axis — a 191-amino-acid pituitary peptide secreted in pulsatile bursts by anterior pituitary somatotrophs that drives postnatal growth, anabolism, and body composition via hepatic IGF-1 induction. In the biology of aging, GH occupies a central position as the upstream driver of the most consistently pro-aging pathway identified in multicellular organisms: reduction of the GH → IGF-1 → IIS → AKT → FOXO axis extends lifespan in organisms from worms to mice, and loss-of-function mutations in the GH receptor protect humans from cancer and type 2 diabetes (Laron syndrome).

Identity

  • UniProt: P01241 (SOMA_HUMAN) — Swiss-Prot reviewed
  • NCBI Gene: 2688
  • HGNC symbol: GH1
  • Mouse ortholog: Gh1
  • Precursor length: 217 amino acids (includes 26-aa signal peptide)
  • Mature form: 191 amino acids (22 kDa) after signal peptide cleavage
  • Gene locus: 17q23.3; part of a 5-gene cluster (GH1, CSHL1, GH2, CSH1, CSH2) on chromosome 17

Molecular structure

GH is a member of the somatotropin/prolactin family of cytokine-like hormones, sharing deep sequence homology with human placental lactogen (~80% identical residues) and prolactin — a relationship established by comparative sequence analysis 1. Its 3D architecture is a compact four-helix bundle (helices I–IV) [#gap/unsourced — four-helix bundle description is from 3D crystal structures, e.g., de Vos et al. 1992 Science 255:306; primary crystallography source not yet on this page] with two cross-bracing disulfide bonds:

BondResidues (mature form numbering)Function
Disulfide 1Cys53–Cys165Stabilises helix I–loop topology
Disulfide 2Cys182–Cys189Maintains C-terminal mini-loop

(UniProt P01241 positions these as Cys79–Cys191 and Cys208–Cys215 in precursor numbering — shift by 26 for mature peptide.) The cDNA sequence and evidence for alternative splicing of the pre-mRNA were characterised by DeNoto et al. 1981 2.

Isoforms

The predominant circulating form is the 22-kDa monomer produced from the major transcript. Alternative splicing generates a 20-kDa isoform (lacking residues 32–46; ~10% of pituitary GH); its biological potency is similar but with reduced diabetogenic activity. Higher-molecular-weight oligomeric forms (dimer, polymer) circulate at low levels and have reduced specific activity.

Post-translational modifications

  • Phosphorylation at Ser132 and Ser176 (UniProt; functional significance not fully characterised)
  • Deamidation at Gln163 and Asn178 (storage artefact; reduces receptor affinity; also seen in recombinant preparations)
  • Signal peptide cleavage at Ala26|Phe27 (co-translational, in ER lumen)

The hypothalamic-pituitary-somatotropic axis

GH secretion is regulated by a dual hypothalamic control system operating on the anterior pituitary:

Hypothalamus
  ↓ GHRH (growth-hormone-releasing hormone)     → stimulates somatotrophs → GH pulse
  ↑ SST (somatostatin / SRIF)                   → inhibits somatotrophs → suppresses GH
Anterior pituitary (somatotroph cells, ~40-50% of anterior pituitary)
  ↓ GH (pulsatile bursts, 6-12 per day; largest pulse 60-90 min after sleep onset)
Liver (primary) + extrahepatic tissues
  ↓ IGF-1 synthesis and secretion
Long-loop negative feedback:
  IGF-1 ← inhibits → GHRH neurons + somatotrophs (short loop also: GH inhibits own GHRH)

The pulsatile secretion pattern is biologically critical: continuous GH infusion desensitises the GHR and has different transcriptional outputs than pulsatile delivery. Males exhibit higher-amplitude, more discrete GH pulses; females have more frequent, overlapping pulses. This sex difference in GH pulse pattern partly underlies sex differences in body composition and liver gene expression [^gap/unsourced — sex-specific GH pulse pattern reviewed but primary quantitative reference needed].

GH secretion peaks at puberty (coinciding with peak IGF-1 levels of 400–700 ng/mL) and declines progressively after age 30. By age 60–70, mean 24-hour GH secretion is approximately 30–50% of the young-adult rate, driven by increased somatostatin tone and decreased GHRH amplitude. This decline parallels the fall in circulating IGF-1 and has been called “somatopause.” Whether somatopause is a cause or consequence of aging physiology — and whether correcting it is beneficial — is the central clinical controversy (see Pharmacology below).

GH receptor signalling: GHR → JAK2 → STAT5

Growth hormone acts through the growth hormone receptor (GHR), a single-pass transmembrane receptor of the cytokine receptor superfamily (class I):

  1. Ligand-induced dimerisation: One GH molecule binds sequentially to two GHR monomers at distinct surface patches (site 1 and site 2 — identified by mutagenesis and crystallography). The 1:2 GH:GHR stoichiometry creates a symmetric transmembrane dimer.
  2. JAK2 trans-phosphorylation: Each GHR subunit constitutively associates with JAK2 kinase at its membrane-proximal Box 1 motif. Dimerisation brings the two JAK2 molecules into proximity; each phosphorylates the other (trans-activation at Tyr1007/Tyr1008 in the JAK2 activation loop).
  3. STAT5 recruitment and phosphorylation: Activated JAK2 phosphorylates tyrosines on the GHR intracellular tail (e.g., Tyr487, Tyr534, Tyr596, Tyr628), creating docking sites for STAT5b (the dominant GH-responsive STAT isoform). JAK2 then phosphorylates STAT5b at Tyr694.
  4. Nuclear translocation: Phospho-STAT5b dimerises, translocates to the nucleus, and binds GAS (gamma-activated sequence) elements in the promoters of target genes including IGF1, IGFBP3, ALS, and liver-specific metabolic genes.
  5. IGF-1 production: ~75% of circulating IGF-1 is produced by hepatocytes under GHR–JAK2–STAT5b control [^gap/unsourced — 75% hepatic fraction is textbook consensus; primary liver-specific KO data is in Liu 1998 + Sjögren 1999; exact fraction needs literature citation].

GHR also activates (via JAK2 scaffolding): PI3K–AKT (direct insulin-like metabolic effects), RAS–MAPK (proliferative), SOCS (suppressors of cytokine signalling — feedback off-switch), and PRLR-shared pathways.

Laron syndrome: GHR loss-of-function

Autosomal recessive loss-of-function mutations in GHR cause Laron syndrome (OMIM 262500): GH insensitivity, near-zero circulating IGF-1, severe proportional dwarfism, and paradoxically elevated serum GH (negative feedback fails because IGF-1 is absent). The Ecuadorian Laron cohort studied by Guevara-Aguirre et al. 2011 provides critical human evidence — covered in depth on igf-1; key GH-specific points here: (1) GHR mutations cause GH excess rather than deficiency despite the IGF-1–deficient phenotype; (2) serum GH is elevated precisely because the long-loop IGF-1 → hypothalamus/pituitary feedback is broken. This is mechanistically important: it shows GH itself is not the pro-aging signal — IGF-1 (downstream of GHR) is 3.

Role in aging: the somatotropic axis

GHR-/- mice: the longest-lived laboratory mouse on record

Coschigano et al. 2003 established that complete deletion of the GH receptor in mice (GHR-/- Laron mice) produces the most extreme longevity phenotype documented in a laboratory mouse model 4:

  • Significantly extended lifespan in GHR-/- mice compared to WT littermates; the paper is closed-access (not_oa) and exact median values cannot be confirmed from the full text here — commonly cited as ~40–55% depending on sex and colony (see table below) needs-replication (exact numbers need primary-source verification)
  • Drastically reduced serum IGF-1 (near undetectable)
  • Dramatically reduced fasting insulin
  • Protected from spontaneous neoplasms
  • Small body size (as expected from absent GH-driven growth)
  • This lifespan extension exceeds that reported for caloric restriction alone and rivals rapamycin at the highest doses tested in ITP mice
ManipulationApprox. median lifespan extensionReference
GHR-/- (Laron mice)~40–55%Coschigano 2003 4
Ames dwarf (Prop1df)~50%Brown-Borg 1996 5
Snell dwarf (Pit1dw)~40%Flurkey et al. 2001
Caloric restriction (C57BL/6)~20–30%Various ITP/NIA data
Rapamycin (ITP; started at 20 mo)~10–25%Harrison 2009 / ITP replications

Note: Ames (Prop1df) and Snell (Pit1dw) dwarf mice lack GH, TSH, and prolactin (pleiotropic pituitary deficiency); they are impure GH experiments. GHR-/- mice are the cleanest somatotropic-axis single-gene model because only GH signalling is abolished.

Extrapolation table:

DimensionStatus
Pathway conserved in humans?yes — GHR/JAK2/STAT5/IGF-1 axis is orthologous across mammals
Phenotype conserved in humans?partial — Laron syndrome protects from cancer and T2D; overall lifespan not significantly extended (confounded by accidents/alcohol)
Replicated in humans?no (lifespan) / limited (disease protection, single cohort)

needs-human-replication — Direct lifespan extension from GH reduction not demonstrated in humans.

Ames dwarf mice: first longevity proof

The first demonstration that GH deficiency extends mammalian lifespan came from Brown-Borg et al. 1996, reporting that Ames dwarf mice (Prop1df — recessive mutation in Prop1 transcription factor → fail to develop anterior pituitary cell types secreting GH, TSH, and prolactin) lived substantially longer than WT littermates 5. The widely-cited figure is ~50% extension, though the exact value cannot be confirmed from the primary source here (paper is closed-access; full text not retrieved) needs-replication. This landmark result, published as a brief Nature letter (~1,060 citations), launched the field of somatotropic axis longevity biology.

GH excess shortens lifespan: acromegaly and bovine GH transgenic mice

The pro-aging direction of GH action is confirmed in both directions:

  • Acromegaly (GH-secreting pituitary adenoma in humans): chronic GH excess → elevated IGF-1 → reduced lifespan (mean age at diagnosis ~40s, reduced life expectancy vs population), increased incidence of colorectal cancer, cardiovascular disease, hypertension, and insulin resistance. needs-replication (survival data from registry studies; confounded by late diagnosis and treatment era)
  • Bovine GH transgenic mice: overexpression of bGH in mice → giant phenotype (2–3× normal body weight) → markedly shortened lifespan to ~12–15 months (vs ~24–30 months WT); early insulin resistance, increased cancer, cardiomegaly [^gap/unsourced — multiple bGH transgenic studies; primary reference needed for exact lifespan numbers; Bartke lab refs recommended].

These bidirectional results in mice — and acromegaly data in humans — make the GH axis one of very few aging pathways with human evidence on both sides (reduction = protection; excess = acceleration).

Mechanism: why does reduced GH/IGF-1 extend lifespan?

Multiple downstream mechanisms proposed (not mutually exclusive):

  1. Reduced mTORC1 activity — IGF-1 → PI3K → AKT → TSC1/TSC2–RHEB–mTORC1; lower IGF-1 reduces mTOR, increasing autophagy and reducing anabolic drive (see mtor)
  2. FOXO transcription factor disinhibition — AKT phosphorylates and cytoplasm-sequesters FOXOs; reduced AKT → nuclear FOXO3/FOXO1 → upregulates stress resistance, antioxidants, DNA repair (see foxo3)
  3. Reduced proliferative signalling — less GH-driven MAPK/proliferation → reduced replicative senescence burden
  4. Improved stress resistance — GHR-/- and Ames dwarf fibroblasts are more resistant to paraquat, UV, and hydrogen peroxide in vitro; mechanism likely involves FOXO target genes needs-replication

Pharmacology

Recombinant human GH (somatropin, rhGH)

FDA-approved indications for somatropin (multiple brands: Norditropin, Genotropin, Humatrope, Saizen, Serostim, etc.):

  • Paediatric GH deficiency (confirmed pituitary/hypothalamic disease)
  • Short stature disorders: Turner syndrome, Prader-Willi, small for gestational age (no catch-up), idiopathic short stature (ISS; >2.25 SD below mean)
  • Adult GH deficiency (confirmed by stimulation testing)
  • AIDS wasting / cachexia (Serostim)
  • Short bowel syndrome (Zorbtive)

Mechanism: Replaces endogenous GH → GHR activation → JAK2/STAT5 → IGF-1 production → growth and anabolic effects.

Anti-aging use — not supported:

Several trials have tested rhGH in elderly individuals to reverse somatopause. The landmark Rudman et al. 1990 NEJM study reported increases in lean mass and decreases in fat mass over 6 months in 21 men aged 61–81 [^gap/needs-replication; dose: 0.03 mg/kg three times weekly]. However, subsequent meta-analyses and longer trials found: (1) the body composition effects reverse on cessation; (2) side effects are substantial at anti-aging doses (fluid retention, carpal tunnel, arthralgias, gynecomastia, glucose intolerance, potential increased cancer risk); (3) no evidence of mortality benefit or improved functional outcomes. The FDA has not approved rhGH for anti-aging and the American Association of Clinical Endocrinologists has explicitly cautioned against this use.

long-term-unknown — Long-term safety of supraphysiological GH supplementation in aging humans not established; cancer risk signal warrants monitoring.

GHRH analogues: sermorelin, tesamorelin

  • Sermorelin — synthetic 29-aa fragment of GHRH; stimulates endogenous GH secretion; used for paediatric GHD and off-label for adult somatopause; less potent than rhGH but physiological pulsatile pattern preserved
  • Tesamorelin — stabilised GHRH analogue; FDA-approved (Egrifta) for HIV-associated lipodystrophy (visceral fat excess); reduces visceral fat via GH → IGF-1 → lipolysis; studied in HIV-uninfected adults for cognitive outcomes (ongoing NCT trials)

GH secretagogues (GHSs): ghrelin mimetics

  • Ghrelin — endogenous 28-aa acylated peptide from gastric X/A cells; binds GHSR-1a in hypothalamus and pituitary → GH release; also stimulates appetite (orexigenic). Age-related ghrelin decline partially contributes to somatopause.
  • MK-677 (ibutamoren) — orally bioavailable non-peptide GHSR-1a agonist; sustained elevation of GH and IGF-1 over 24 h; Phase 2 trials in elderly for frailty/sarcopenia showed lean mass gain without significant strength improvement; used off-label and as research tool; long-term-unknown

Pegvisomant: GHR antagonist for acromegaly

Pegvisomant (Somavert) is a PEGylated GH analogue with mutations at the site-2 interface that allow site-1 binding but prevent productive GHR dimerisation → GHR blockade → reduced IGF-1. FDA-approved for acromegaly refractory to somatostatin analogues. From a longevity-biology perspective, pegvisomant is pharmacologically equivalent to the GHR-/- genetic model — the only approved drug that phenocopies the Laron mouse at the receptor level. It is not studied for longevity indications. needs-human-replication

Somatostatin analogues: octreotide, lanreotide, pasireotide

Used for acromegaly (suppress GH secretion from adenomas); not directly relevant to normal-aging GH reduction because they suppress endogenous GH axis from the hypothalamic side rather than directly blocking growth-promoting downstream effects.

Cross-references

  • igf-1 — primary downstream effector; GH → GHR → JAK2 → STAT5 → IGF-1 transcription; centenarian paradox, Laron syndrome, and Holzenberger 2003 IGF1R+/- data covered there
  • insulin-igf1 — the full IIS pathway hub; cross-organism lifespan data; FOXO biology; comparative pharmacology
  • igf1r — IGF-1 receptor; receives hepatic IGF-1 signal; Suh 2008 centenarian variants
  • jak-stat-pathway — JAK2-STAT5 signalling downstream of GHR
  • mtor — convergent downstream target of GH→IGF-1→AKT axis; mTORC1 inhibition is the shared mechanism between GH reduction and rapamycin longevity
  • foxo3 — key longevity transcription factor disinhibited by reduced IIS; nuclear FOXO3 drive stress resistance and autophagy
  • foxo1 — hepatic FOXO1 regulates gluconeogenesis downstream of AKT; GH/IGF-1 axis intersects insulin signalling at this node
  • deregulated-nutrient-sensing — the Hallmark that GH/IGF-1 biology most directly instantiates
  • homo-sapiens — Laron syndrome human evidence; acromegaly human evidence
  • cellular-senescence — GH/IGF-1 drives proliferation; excess IGF-1 may accelerate replicative senescence
  • ames-dwarf-mouse — Prop1df model organism (GH/PRL/TSH deficient; ~50% lifespan extension); implicit stub
  • ghr — GH receptor protein; implicit stub; GHR-/- Laron mouse and pegvisomant mechanism belongs there

Limitations and gaps

  • needs-human-replication — No controlled human trial has demonstrated that pharmacological GH reduction extends lifespan. Laron syndrome data (Guevara-Aguirre 2011) is observational, single-cohort, confounded by accidents/alcohol, and does not show lifespan extension per se.
  • needs-replication — Bidirectional lifespan effect of bGH overexpression (shortened) vs GHR-/- (extended) is well-established in mice, but the magnitude of extension in GHR-/- varies by genetic background and sex; tabulated lifespan values need primary-source verification per study.
  • dose-response-unclear — Optimal degree of somatotropic axis suppression for human healthspan benefit is unknown; too little GH → frailty/sarcopenia; too much → cancer/T2D risk.
  • long-term-unknown — Long-term safety of supraphysiological GH supplementation (anti-aging dosing in elderly) is not established; cancer risk signal in epidemiological literature warrants monitoring.
  • no-mechanism — Exact mechanism by which reduced GH/IGF-1 extends lifespan across such different organisms (worms through mice) is not fully resolved; FOXO, mTOR, and stress resistance pathways are all implicated but relative contributions unclear.
  • unsourced — Sex-specific GH pulse pattern primary reference needed (Male > Female pulse amplitude); textbook consensus cited here without primary paper.
  • unsourced — 75% hepatic fraction of circulating IGF-1 derived from GH signalling (vs extrahepatic paracrine) needs primary citation (Liu 1998 / Sjögren 1999 liver-specific Igf1 KO likely).
  • unsourced — Bovine GH transgenic mouse lifespan shortening exact values; Bartke lab primary references needed.
  • needs-canonical-id — GenAge gene entry for GH1 not confirmed; may be absent from GenAge-human (gene name: GH1, not a “longevity gene” in the conventional sense since the organism doesn’t live longer — the GHR-/- phenotype is on the receptor, not GH itself).
  • unsourced — Four-helix bundle structural description (helices I–IV) is from X-ray crystallography literature (de Vos et al. 1992 Science 255:306; Cunningham et al.); no primary crystallography source is currently cited on this page. Niall 1971 does not establish this structure.
  • no-fulltext-access — Brown-Borg 1996 (10.1038/384033a0): exact lifespan data (mean lifespan in days, statistical test, sex breakdown) unverifiable; DOI lookup failed and paper is behind Nature paywall with no PMC mirror. The commonly-cited ~50% extension figure should be confirmed when access becomes available.

Footnotes

Footnotes

  1. doi:10.1073/pnas.68.4.866 · in-vitro (protein chemistry / comparative sequencing) · Niall HD, Hogan ML, Sauer R, Rosenblum IY, Greenwood FC · PNAS 1971 · “Sequences of pituitary and placental lactogenic and growth hormones: evolution from a primordial peptide by gene reduplication” · What this paper shows: comparative amino acid sequence analysis of human placental lactogen (HPL), human growth hormone (HGH), and ovine prolactin — finding ~80% sequence identity between HPL and HGH (revised sequence); proposes the somatotropin/prolactin family evolved by gene reduplication from a common ancestral peptide; also reports identical chain length of ~190 aa for GH and HPL and identically placed half-cystine residues. Does NOT characterise the four-helix bundle 3D structure (that is from later crystallography, e.g., de Vos et al. 1992) · archive status: downloaded (green OA PMC389061) ↩

  2. doi:10.1093/nar/9.15.3719 · molecular biology (genomic DNA sequencing / S1 mapping) · DeNoto FM, Moore DD, Goodman HM · Nucleic Acids Res 1981 · “Human growth hormone DNA sequence and mRNA structure: possible alternative splicing” · What this paper shows: complete genomic DNA sequence of the hGH gene (2.6 kb EcoRI fragment); gene has four introns (IVS A = 256 bp, B = 209 bp, C = 93 bp, D = 253 bp); S1 nuclease mapping confirms 5’ end of mature mRNA; identifies alternative 3’ splice site in IVS B generating the 20-kDa isoform (15 aa deletion, ~10% of pituitary GH); signal peptide start confirmed; full amino acid sequence of 191 aa mature form derived from genomic sequence (Fig. 2) · archive status: downloaded (Europe PMC PMC327387) ↩

  3. doi:10.1126/scitranslmed.3001845 · observational (cross-sectional cohort with longitudinal mortality follow-up) · n=99 GHRD subjects monitored since 1988 (90 living, 9 deceased during monitoring; 53 additional pre-1988 deaths via questionnaire = 62 total deaths) + 1,606 unaffected relatives as controls · model: homo-sapiens (Ecuadorian founder cohort, GHR loss-of-function mutations) · Cancer: 0 cancer deaths in GHRD; 1 non-lethal cancer diagnosed (ovarian); cancer = 20% of deaths in relatives and 17% of all diseases in relatives; P=0.003, exact hypergeometric distribution (StatXact 7) · T2D: 0/90 in GHRD (95% CI 0–4%) vs ~5% in Ecuador/relatives; P=0.02 (exact binomial test) · IGF-1: ≤20 ng/mL in all GHRD vs 29–310 ng/mL (mean 144) in relatives (P<0.0001) · Lifespan: not significantly extended; 70% of GHRD deaths (of 30 deaths over age 10) were non-age-related (accidents, alcohol, convulsive disorders) · verified against PMC3357623 full text · Sci Transl Med 2011 · archive status: download failed (green OA PMC3357623; URL filter blocked) · see also igf-1 where this study is covered in depth ↩

  4. doi:10.1210/en.2003-0374 · in-vivo · genetic model (GHR-/- knockout) · model: Mus musculus (C57BL/6 background) · Coschigano KT et al. · Endocrinology 2003 · 509 citations · “Deletion, But Not Antagonism, of the Mouse Growth Hormone Receptor Results in Severely Decreased Body Weights, Insulin, and IGF-I Levels and Increased Life Span” · archive status: not_oa (no local PDF) ↩ ↩2

  5. doi:10.1038/384033a0 · in-vivo · observational (spontaneous mutant lifespan) · model: Mus musculus (Ames dwarf, Prop1df) · Brown-Borg HM, Borg KE, Meliska CJ, Bartke A · Nature 1996 · 1,060 citations · first demonstration of ~50% lifespan extension in GH-deficient Ames dwarf mice (exact lifespan values unverified — full text not accessible) · archive status: failed (bronze OA; no PMC mirror; Nature paywall) no-fulltext-access ↩ ↩2

Graph View

Backlinks