Klotho Protein Longevity: The Aging Hormone Going Human in 2026

In 2023, a single low-dose injection of a kidney-made protein improved memory in aged rhesus macaques within four hours — and the effect was still measurable 23 days later. Higher doses did nothing. The protein is called klotho, and the work, led by Dena Dubal's group at UCSF and published in Nature Aging, produced one of the most unusual dose-response curves in longevity pharmacology: the low dose worked, the high doses failed (PMID: 37400721).

Soluble klotho in human blood starts dropping after age 40. By ages 50-64, circulating levels are roughly 20% lower than in 20-34 year-olds (PMID: 20599764). In a NHANES cohort of 13,746 US adults, anyone falling below 603.5 pg/mL carried a 34% higher risk of all-cause mortality (PMID: 38689889). "Klotho protein longevity" is no longer a mouse story. The first human gene-therapy trials began recruiting in 2025-2026 (ClinicalTrials.gov NCT07216781). Here is what klotho actually does, what raises it, and what the current hype is getting wrong.

What is klotho, and why did a Japanese lab name it after a Greek Fate?

Klotho was discovered by accident. In 1997, Makoto Kuro-o and colleagues at the National Institute of Neuroscience in Tokyo were studying hypertension when they generated a mouse line with a disrupted gene on chromosome 13. The mice looked fine at birth. Then, at around three to four weeks old, they started falling apart. They developed arteriosclerosis, osteoporosis, skin atrophy, pulmonary emphysema, and infertility, and most were dead by 8-9 weeks. It was every major aging phenotype at once, compressed into two months (PMID: 9363890).

Kuro-o's group named the gene klotho after Clotho, the youngest of the three Greek Fates, who spins the thread of human life. The naming was not modest. They were explicitly signaling that this gene controls lifespan (PMID: 9363890).

Klotho is produced mainly in the kidney's distal convoluted tubules and in the brain's choroid plexus. It exists in two forms: a membrane-bound receptor and a secreted hormone — α-klotho (the secreted, hormone-like form of the klotho protein that circulates in blood) — which ends up in blood and cerebrospinal fluid (PMID: 9363890; PMID: 17086194).

Eight years after the knockout paper, Kurosu and colleagues closed the loop. They built transgenic (genetically modified to carry an extra copy of a gene) mice that overexpressed klotho at roughly 2x normal levels. The males lived up to 31% longer than wild-type (936 days vs 715 days). Females lived 19% longer. This was the first proof that more klotho extends life, not just that losing it kills (PMID: 16123266). The gene earned its mythological name.

For other molecules in the same conversation, see other longevity compounds.

How does the klotho protein actually work in the body?

Klotho has at least four distinct jobs, and they don't all run through the same tissue.

Job 1 — Phosphate control. Klotho is the obligate co-receptor for FGF23 (fibroblast growth factor 23 — a phosphate-regulating hormone made by bone). Urakawa and colleagues showed in 2006 that klotho converts the ordinary FGFR1c receptor into an FGF23-specific receptor, so the kidney can excrete phosphate properly (PMID: 17086194). Without klotho, phosphate climbs, soft tissues calcify, and the body ages in fast-forward. Phosphate toxicity is the mechanism behind the entire klotho-knockout phenotype (PMID: 9363890; PMID: 17086194).

Job 2 — Repressing insulin/IGF-1 signaling. Circulating s-klotho binds a still-unidentified surface receptor and dampens the IGF-1 (insulin-like growth factor 1 — a growth signal linked to both body size and aging speed) axis (PMID: 16123266). This is the same evolutionarily conserved longevity pathway that extends life in C. elegans daf-2 mutants, Ames dwarf mice, and centenarians with IGF1R variants. Klotho sits in the same biology as the rest of the insulin/IGF-1 axis story.

Job 3 — Brain signaling through a blood messenger. This is the most surprising part. Klotho does not cross the blood-brain barrier (BBB) (the tight cell layer that keeps most blood molecules out of brain tissue). A subcutaneous injection in the leg has no business affecting hippocampal memory. Yet it does. Leon and colleagues showed in 2017 that peripheral klotho fragments improved cognition in aged mice even though the protein never entered the brain (PMID: 28793260). Park, Dubal and colleagues resolved the paradox in 2023: peripheral klotho activates platelets and triggers release of platelet factor 4 (PF4) (a small protein released from activated platelets that can cross into the brain). PF4 enters the brain, enhances NMDA receptor GluN2B (a subunit of a brain receptor involved in learning and memory) signaling, and boosts synaptic plasticity and long-term potentiation (the long-lasting strengthening of connections between neurons that underlies memory). Block platelet activation, and klotho's cognitive benefit disappears (PMID: 37587231; PMID: 24813892).

Job 4 — Mitochondrial support in muscle. Sahu and colleagues silenced klotho in young muscle progenitor cells and produced mtDNA (mitochondrial DNA — the separate genome inside your cellular power plants) damage and bioenergetic failure. Adding klotho back to aged progenitors restored mtDNA integrity to youthful levels and rescued muscle regeneration in vivo (PMID: 30451844). This ties klotho directly to mitochondrial dysfunction in muscle.

Four jobs, four tissues, one protein. That multi-system reach is why klotho behaves less like a drug and more like an aging signal.

The evidence — from 9-week-old mice to aged rhesus macaques

The klotho literature has a clean chronological arc. The receipts:

• 1997 (mice). Kuro-o et al., Nature. Klotho knockout (a mouse engineered to have a specific gene deleted) mice show a compressed premature-aging syndrome and die by ~8 weeks (PMID: 9363890).
• 2005 (mice). Kurosu et al., Science. 2x klotho overexpression extends male mouse lifespan by up to 31% and female lifespan by 19% (PMID: 16123266).
• 2014 (humans + mice). Dubal et al., Cell Reports. Across 718 dementia-free adults, heterozygous KL-VS carriers (people with one copy of a beneficial variant of the klotho gene) outperformed non-carriers on cognitive tests. Transgenic klotho-overexpressing mice showed enhanced long-term potentiation and enriched synaptic GluN2B. First human genetics link (PMID: 24813892).
• 2011 (humans). Semba et al., InCHIANTI cohort of 804 Italian adults aged 65 and older. Those in the lowest tertile of serum klotho had roughly 1.78x higher all-cause mortality at a median six-year follow-up (PMID: 21474560).
• 2017 (mice). Leon/Dubal et al., Cell Reports. A peripheral klotho fragment acutely improved cognition and motor function in aged and α-synuclein transgenic mice — without crossing the BBB. Mechanism: NMDAR GluN2B cleavage and enhanced synaptic plasticity (PMID: 28793260).
• 2022 (humans, meta-analysis). de Luca Corrêa et al., Scientific Reports. Twelve studies, 621 participants. Chronic exercise raised s-klotho with a Hedge's g of 1.3 (95% CI 0.69-1.90, p<0.0001) — a large effect size (PMID: 36266389).
• 2023 (primates). Castner, Dubal et al., Nature Aging. Aged rhesus macaques received a single 10 μg/kg subcutaneous injection. High-load spatial memory improved by ~7% at four hours and held through 14-23 days. The 20 and 30 μg/kg doses did not work. The low dose did (PMID: 37400721).
• 2023 (mechanism). Park, Dubal et al., Nature Aging. Platelet factor 4 is the blood-to-brain messenger. Blocking platelet activation abolishes klotho's cognitive effect (PMID: 37587231).
• 2024 (humans). Yu et al., NHANES analysis of 13,746 US adults. Optimal serum klotho cut-off 603.5 pg/mL; below that, adjusted all-cause mortality hazard ratio 1.34 (PMID: 38689889).
• 2025 (mice, gene therapy). Roig-Soriano et al., Bosch lab at Universitat Autònoma de Barcelona, Molecular Therapy. A single AAV (adeno-associated virus — a harmless viral vector used to deliver gene therapy) injection delivering secreted klotho to young mice extended lifespan by 15-20%. At 24 months (roughly 70 human years), treated mice had larger muscle fibers, less fibrosis, better trabecular bone, and improved cognition. Three patents were filed covering bone, muscle, and longevity applications — the direct driver of 2026 commercial interest (Roig-Soriano 2025, Mol Ther 33(4):1449; PMID: 39988871).

Ten studies. Five species. Two decades. The signal is consistent in a way longevity biology usually is not.

What the hype gets wrong — KL-VS, "klotho supplements," and the dose problem

Klotho has a believable science story, and that makes the failure modes worth flagging explicitly.

The KL-VS cognition finding has not fully replicated. Dubal 2014 is the paper everyone cites, and it is the reason consumer genetics sites treat KL-VS variant (a specific DNA change in the klotho gene that raises klotho blood levels, carried by roughly 20-25% of people) status as a "longevity SNP." But a 2021 population-based study in Scientific Reports (55-87 year olds) found no cognitive benefit from KL-VS heterozygosity in the general population. Newer work from Belloy 2020 (PMID: 32282020) and Neitzel 2021 (PMID: 34158479) links KL-VS to lower amyloid and tau burden — but specifically in APOE4 carriers (APOE4 — the highest-risk common gene variant for Alzheimer's). Treat KL-VS as a suggestive, context-dependent finding, not settled science for cognition-in-the-general-population.

"Klotho supplements" on Amazon are not klotho. Klotho is a ~130 kDa protein. Swallow it and it gets digested in the stomach like any other dietary protein. Any pill marketed as a "klotho supplement" or on a list of klotho supplements is almost always an herbal blend claimed to "support klotho levels" with no human RCT evidence for raising the actual protein.

Offshore plasmid "gene therapy" is not peer-reviewed evidence. The Minicircle service — which offers a plasmid (a small circular piece of DNA sometimes used to deliver genes) version of klotho delivery outside FDA jurisdiction — has produced no peer-reviewed Phase 1 data. Anecdotes from users who report "feeling great" after offshore klotho treatment are not trial results. Treat them accordingly.

More is not better. The 2023 rhesus macaque study is the single cleanest refutation of a "megadose it" strategy. Only the low dose worked. The two higher doses produced no cognitive benefit. This is an inverted-U dose response (a pattern where a middle dose helps but both lower and higher doses do not), and it is typical for neurosignaling molecules — think cortisol, dopamine, norepinephrine. Klotho is in that category (PMID: 37400721).

The "20% longer life" number is an extrapolation. The UAB 2025 result is preclinical. It is a mouse result with a gene therapy vector. No human lifespan data for klotho exist. They cannot exist yet. Phase 1 trials are just starting.

What actually raises klotho in humans?

This is the section where most write-ups quietly bail on the evidence. The honest version is narrower than the hype.

Chronic exercise is the only lever with meta-analysis-grade human evidence. The de Luca Corrêa 2022 meta-analysis pooled 12 studies and 621 participants. Twelve weeks or more of structured training raised circulating s-klotho with a Hedge's g of 1.3 — a large effect by any standard. The effect held across aerobic, resistance, and HIIT protocols as single modalities. The authors call klotho an exerkine (a signaling molecule released into the blood in response to exercise), a molecule released into the bloodstream in response to chronic training (PMID: 36266389). This is consistent with the broader chronic exercise literature and with fitness as a longevity lever.

Caveat the "do both" instinct. The same meta-analysis found a null effect for combined aerobic + resistance protocols. The likely explanation is a volume or intensity trade-off when you split a week across two modalities rather than stacking one. This is not a reason to drop lifting. It is a reason to stop assuming "more variety" automatically means "more klotho."

Vitamin D is directional, not dramatic. The Nikkhah et al. 2020 RCT gave 90 vitamin D-deficient adults over 60 50,000 IU of cholecalciferol (the D3 form of vitamin D) per week for 12 weeks. The treatment group showed a non-significant rise in plasma klotho and prevention of the placebo group's decline. It is the only human RCT on vitamin D and klotho, and the effect was modest. Anyone telling you "vitamin D boosts klotho" is stretching this one result. Deficiency correction still matters for reasons unrelated to klotho — see vitamin D in the compound index.

Sleep, phosphate moderation, and inflammation control have mechanistic support and essentially zero human RCT evidence for raising s-klotho. Put them in the "reasonable to do for other reasons" bucket, not the "raises klotho" bucket.

The honest version of the "how to raise klotho" story is one sentence: chronic single-modality training, twelve weeks minimum, is the only human-grade lever. Everything else is either suggestive or hypothetical.

The gene therapy frontier — what's actually in human trials

Klotho is too large to take orally. That leaves four delivery routes that actually get klotho into circulation: recombinant protein (a protein made in the lab from a cloned gene) injection (the 2023 monkey study route), AAV gene therapy (UAB preclinical), plasmid gene therapy (offshore only), and experimental mRNA. Gene therapy is the route with commercial momentum.

The UAB AAV program is the preclinical engine. Roig-Soriano et al. 2025 — Assumpció Bosch's lab at Universitat Autònoma de Barcelona — showed that a single AAV-s-KL treatment in young mice produced lifelong benefit: 15-20% longer lifespan, better muscle, better bone, better cognition at 24 months (PMID: 39988871). Three patents covering bone, muscle, and longevity applications are the direct reason 2026 is turning into a "klotho year" in biotech.

Klothea Bio, Klotho Neurosciences, and Minicircle. Phase 1 trials began recruiting in 2025-2026. Cohorts are small — the Klothea protocol registered on ClinicalTrials.gov is ~21 participants, healthy adults aged 25-75 (NCT07216781). A Phase 1 trial (the first stage of human testing, focused on safety in small groups) is a safety readout, not an efficacy one. This is the earliest edge of the gene-therapy frontier, and klotho joins other longevity proteins in the "plausible but not proven in humans" column. See also the broader gene therapy frontier.

There is an age window. Clemens et al. 2021 in eLife reported that klotho-based muscle regeneration rescue worked in old mice but was lost in the oldest-old (PMID: 33876724). Whether this window exists in humans is unknown. It matters a lot for who enrolls in trials and who actually benefits.

The realistic timeline. Phase 1 is safety. Even if every signal is clean, a licensed human klotho therapy is years away. What you can act on today is the exerkine data.

FAQ

Bottom line

Klotho is the rare longevity signal with converging evidence: a Greek-named gene in mice, a specific human variant tied to better aging in APOE4 carriers, blood levels that predict mortality in a 13,746-person cohort, and — as of 2023 — a single injection that sharpened memory in aged primates for two weeks at a dose where "more" actively failed. The pill version does not exist. The gene therapy version is just starting Phase 1. Until that reads out, the only honest lever is the boring one: chronic exercise, twelve weeks and up, as a single-modality habit.