Dihexa

Plain-English version: dihexa is a small, chemically modified molecule that scientists built by trimming and stabilizing a natural brain peptide called angiotensin IV. The goal was a compound that survives digestion, crosses into the brain, and switches on a growth-signaling system that helps neurons wire together. Its full chemical name is N-hexanoic-Tyr-Ile-(6)aminohexanoic amide.

Dihexa came out of structure-activity work on angiotensin IV analogs, specifically the Norleucine1-angiotensin IV (Nle1-AngIV) series, in the laboratories of Joseph Harding and John Wright at Washington State University.^[2] The brain's angiotensin IV / AT4 system had already been tied to spatial learning and memory in rats, which is the historical reason anyone thought to develop AngIV analogs as pro-cognitive agents in the first place.^[5] Dihexa was engineered to be metabolically stabilized (it resists the enzymes that normally chew up peptides), orally active, and able to cross the blood-brain barrier — three properties that most peptides lack.^[1]

About that famous number: the claim that dihexa is "roughly 10 million times more potent than BDNF" refers to one specific laboratory measurement — its ability to trigger new dendritic-spine formation in cultured hippocampal neurons, where it is active at picomolar (10^-12 M) concentrations while BDNF needs far higher concentrations for a comparable effect.^[1] It does not mean dihexa makes you 10 million times smarter, or that it is 10 million times "better" than anything. It is a potency figure for one cellular endpoint, and it gets badly mangled in marketing copy.

The credibility caveat, stated up front: dihexa has never been submitted for or granted FDA approval. It is not a prescription drug, a dietary supplement, or an over-the-counter product. It exists only as a preclinical academic research compound that vendors sell labeled "research chemical / not for human consumption." No investigational new drug (IND) program and no registered human clinical trial has ever been published. Everything below is animal and cell-culture science extrapolated by hopeful humans — treat it that way. For the legal picture, see Are Peptides Legal?

Takeaway first: dihexa's real mechanism is not what most sites say. It does not simply switch on the old "AT4 receptor." Instead, the leading hypothesis is that dihexa binds hepatocyte growth factor (HGF) and boosts signaling through its receptor, c-Met — a growth-and-repair pathway. Be aware up front, though: the primary paper behind that mechanism has since been retracted (see below), so this is a proposed mechanism, not settled science. Here is what the research claimed, and where it stands now.

1. HGF binding and c-Met activation (the proposed mechanism — now retracted)

Hepatocyte growth factor is a signaling protein; c-Met is the receptor it docks onto, a "receptor tyrosine kinase" that turns on cascades controlling cell growth, survival, movement, and — in the brain — synapse formation. The idea that dihexa works by binding HGF and activating c-Met came from a 2014 study by Benoist and colleagues, which reported high-affinity HGF binding and claimed that dihexa's effects disappeared when HGF pairing was blocked (with a dimerization antagonist called "Hinge") or when c-Met was silenced with shRNA.^[3] Critical caveat: that paper was retracted in 2025 over data-integrity problems (image manipulation), after a 2021 Notice of Concern. Its specific figures — including the frequently-quoted HGF binding constant and the c-Met-dependence knockdown experiments — can no longer be treated as reliable evidence. The HGF/c-Met story remains the most-repeated explanation for dihexa online, but with its key primary paper retracted, the mechanism should be regarded as unproven rather than established. Evidence level: a proposed mechanism from a now-retracted study — treat with caution.

2. Synaptogenesis — building new synapses

"Synaptogenesis" just means forming new synapses, the connection points between neurons that are the physical substrate of learning and memory. In cultured hippocampal neurons, dihexa induced robust new dendritic-spine growth at picomolar concentrations.^[1] This spinogenesis assay is the exact source of the "10 million times more potent than BDNF" line — it is a real in-vitro finding about one endpoint, not a claim about overall cognitive power. The underlying design work showing that the Nle1-AngIV core structure is what confers this synaptogenic, pro-memory activity came from the earlier Benoist study.^[2] Evidence level: in vitro and rodent.

3. Angiotensin IV heritage and the AT4/HGF link

Dihexa's lineage runs through the brain angiotensin IV system, which independent work tied to spatial learning in rats.^[5] The bridge between that system and HGF/c-Met is not accidental: related AngIV analogs have been shown to modulate the HGF/c-Met system directly — for example, the analog "norleual" can act as an HGF/c-Met inhibitor, the mirror image of what dihexa does.^[6] That family relationship is what let researchers tune a molecule from "memory system peptide" into "HGF amplifier." Evidence level: in vitro and rodent.

What we do NOT know

There is no human pharmacokinetic data — how dihexa is absorbed, distributed, metabolized, and cleared in people is uncharacterized. There is no human dose-response, no established therapeutic window, and no confirmed human half-life. The mechanism above is well-supported in animals and cells, but a clean mechanism in a rat does not guarantee a benefit — or safety — in a human.

How to read this section: for each area we give the plain-English claim, the proposed mechanism, the population actually studied, the observed effect, and the limitation. Every result below is from rats or cultured cells. That does not make it worthless — it makes it preliminary. Anecdotal human reports exist but are not controlled evidence, and no human trial has ever tested any of these claims.

Reversing chemically-induced memory deficits

Claim: may restore learning that has been pharmacologically impaired. Mechanism: HGF/c-Met-driven synaptogenesis in the hippocampus. Population: 3-month-old male Sprague-Dawley rats given scopolamine, a drug that blocks acetylcholine and induces a memory deficit resembling aspects of early dementia. Effect: oral dihexa at 2.0 mg/kg completely reversed the Morris water maze learning deficit; injections into the brain ventricles (ICV, 1 nmol) and into the abdomen (IP, 0.5 mg/kg) restored performance to be statistically indistinguishable from healthy vehicle controls.^[1] Limitation: a single acute pharmacological model in young rats; effect sizes come from small rodent cohorts and do not translate to a human dose or a human disease.

Improving memory in aged animals

Claim: may help brains with real, age-related decline — not just drug-induced deficits. Mechanism: restoring synaptic connectivity lost with aging. Population: 24-month-old aged Sprague-Dawley rats (roughly the rodent equivalent of advanced age). Effect: oral dihexa at 2 mg/kg significantly improved Morris water maze performance on most test days versus untreated aged controls.^[1] Limitation: small aged-rodent cohort, short duration, no human dose-response. "Improved on most test days" is a genuine but modest finding, not a cure for age-related memory loss.

Synaptogenesis and neural connectivity

Claim: promotes the physical formation of new neural connections. Mechanism: high-affinity HGF binding and c-Met activation driving dendritic-spine growth. Population: cultured hippocampal neurons. Effect: robust dendritic-spine formation at 10^-12 M (picomolar) concentrations in cultured neurons.^[1] This picomolar spinogenesis is the finding behind the potency headline. A 2014 follow-up reported that the effect depended on the HGF/c-Met pathway, but that paper was retracted in 2025 for data-integrity problems, so the mechanistic link should be treated as unconfirmed.^[3] Limitation: a cell-culture endpoint. New spines in a dish are a proxy for, not proof of, better memory in a person.

Neuroprotection and the Alzheimer's hypothesis

Claim: the pathway dihexa targets may be relevant to neurodegenerative disease. Mechanism: synapse preservation and formation via HGF/c-Met. Population: none — this is a review-level hypothesis, not an experiment on dihexa in disease. Effect / argument: Wright and Harding (2015) reviewed the brain HGF/c-Met system as a candidate Alzheimer's target, arguing that synapse loss correlates with cognitive decline better than amyloid plaque or tau tangle burden — so a compound that builds synapses could, in theory, address a core feature of the disease.^[4] Limitation: this is a hypothesis paper, not clinical evidence. Dihexa is not a treatment for Alzheimer's or any other condition, and no trial has tested it for one.

Anecdotal, non-research reports

Outside the published literature, self-experimenters describe subjective effects like sharper focus, faster verbal recall, reduced "brain fog," and more vivid dreams. These are consistent with a synaptogenic mechanism, but they are uncontrolled, unblinded, and highly vulnerable to expectation bias — the "I paid for it, so I feel it" effect. Treat them as hypotheses about your own experience, not as evidence dihexa works.

The honest bottom line on "benefits"

• No human trials of any kind exist. Human effective doses, safety, and long-term outcomes are unknown.
• The evidence base is compromised. The foundational rat study (McCoy 2013) carries a 2021 journal Expression of Concern, and the key HGF/c-Met mechanism paper (Benoist 2014) was retracted in 2025 for data-integrity problems — so even the preclinical record must be read with caution.
• The research is concentrated in one group of collaborators (Harding/Wright and colleagues); independent replication is limited.
• Publication bias means positive rodent results are more likely to reach print than null ones.
• Potency is not efficacy. Being active at picomolar concentrations in a dish tells you nothing certain about real-world cognitive benefit or its size in a human.

For nootropic peptides with actual clinical track records, compare semax and selank, and see the cognitive enhancement guide for the full landscape.

Read this first: there is no validated human dose for dihexa, because no human dosing study of any kind has ever been done. The only doses that actually exist in the literature are rat doses. The "5-20 mg oral" figures you will see repeated across vendor pages and forums are anecdotal and vendor-derived — they are not supported by any human study and were not measured against any safety endpoint. We describe them below only for transparency and harm reduction, not as guidance to use an unapproved compound.

Doses that were actually studied (rats only)

Route	Dose range studied	Model / result	Note
Oral gavage	1.25–2.0 mg/kg	2.0 mg/kg reversed scopolamine deficit; 2 mg/kg improved aged-rat memory	Demonstrates oral activity and BBB penetration in rodents
Intraperitoneal (IP)	0.05–0.5 mg/kg	0.5 mg/kg restored maze performance to control level	Injection into the abdominal cavity (a rat research route)
Intracerebroventricular (ICV)	0.1–1 nmol	1 nmol restored performance	Direct brain-ventricle injection; research-only, not a human route

These come from the McCoy 2013 rat study^[1] — which itself carries a 2021 Expression of Concern from the journal over data-integrity questions in this body of work, so even the rodent dosing figures should be read as provisional rather than settled. Note the units: the published work is dosed per kilogram of body weight in rats, and there is no established, evidence-based way to convert that into a safe human dose. Naive body-surface-area scaling of a 2 mg/kg rat oral dose would land in the low tens of milligrams for an adult — which is roughly where the online "5-20 mg" figures come from — but that conversion assumes a human safety and pharmacokinetic profile that has never been measured. It is extrapolation stacked on extrapolation.

Circulating (unvalidated) human protocols

For completeness, the ranges that circulate in self-experimenter communities are oral doses of roughly 5–20 mg once daily and, less commonly, subcutaneous doses of a few milligrams daily, run in short 2–4 week blocks. None of this is trial-derived. There is no human data establishing that these doses are effective, that they are safe, or that the therapeutic window is wide enough to use them without harm. Anyone treating these numbers as a protocol is running an uncontrolled experiment on themselves.

Reconstitution math, with a worked example

Dihexa sold for injection ships as a lyophilized (freeze-dried) powder that must be mixed with bacteriostatic water first. The core formula is the same as for any peptide:

Concentration (mg/mL) = vial amount (mg) ÷ water added (mL)

Worked example: take a 10 mg vial and add 2 mL of bacteriostatic water. That gives 10 ÷ 2 = 5 mg/mL. To draw a hypothetical 2 mg amount: 2 ÷ 5 = 0.40 mL, which is 40 units on a standard 100-unit insulin syringe.

Vial	BAC water	Concentration	Volume for 2 mg	Volume for 5 mg
10 mg	1 mL	10 mg/mL	0.20 mL (20 units)	0.50 mL (50 units)
10 mg	2 mL	5 mg/mL	0.40 mL (40 units)	1.00 mL (100 units)
25 mg	2.5 mL	10 mg/mL	0.20 mL (20 units)	0.50 mL (50 units)

Preparation steps: wipe the vial stopper and the water vial with an alcohol swab; draw the bacteriostatic water; inject it slowly down the inside wall of the vial rather than blasting the powder cake directly; swirl gently until dissolved; label the vial with the date, concentration, and compound name. Use the peptide calculator to double-check volumes and the reconstitution guide for a step-by-step walkthrough.

Timing, cycling, and stacking (convention, not evidence)

• Timing: those who use it favor morning dosing, reporting that the stimulating, focus-shifting effect can disrupt sleep if taken later. This is anecdotal.
• Cycle length: short 2–4 week blocks with extended breaks are the community norm, chosen precisely because long-term safety is unknown — not because any study established an optimal duration.
• Stacking: dihexa is sometimes paired with semax (a different, BDNF-oriented mechanism) in the Cognitive Stack. No interaction studies exist. Do not combine dihexa with angiotensin-pathway drugs (ACE inhibitors, ARBs) or with c-Met-targeting cancer drugs without medical oversight.

None of the above is a recommendation to self-administer an unapproved compound. It is a description of how the research doses were structured and how the online community talks about it.

Straight talk: dihexa is one of the least-characterized research peptides in circulation. There are no Phase 1 safety trials, no human pharmacokinetics, no dose-finding data, and no long-term studies. Everything we can say about safety comes from short rodent experiments (weeks long) and from anecdotal user reports — plus one important, mechanism-based concern about cancer that no study was ever designed to answer. This is not "probably fine"; it is "genuinely unknown."

Reported side effects (anecdotal, not from trials)

Because there are no controlled human studies, accurate frequency and severity cannot be established. The following come from self-experimenter reports.

Effect	Reported frequency	Severity	Notes
Headache	Common	Mild–moderate	Often early; may signal too high a dose
Insomnia / sleep disruption	Common	Mild	More likely with afternoon or evening dosing
Anxiety / overstimulation	Occasional	Mild–moderate	Dose-dependent; consistent with heightened neural activity
Blood-pressure changes	Reported	Uncertain	Angiotensin-family origin makes BP monitoring sensible
Injection-site reactions (SC)	Common (SC route)	Mild	Redness, swelling, itching at the site
GI discomfort (oral)	Occasional	Mild	Usually transient
Vivid dreams	Occasional	Mild	Common with synaptogenic compounds; not necessarily adverse

Most reported effects are mild and dose-dependent, and the usual community response is to lower the dose or move it earlier in the day. But remember: no one has systematically collected adverse events, so the absence of severe reports is not evidence of safety.

The HGF/c-Met cancer question (the concern that matters most)

This is the single most important safety point about dihexa, and it is mechanistic rather than anecdotal. The HGF/c-Met pathway that dihexa amplifies is a well-established proto-oncogenic signaling axis — meaning it is one of the systems cancers hijack. c-MET dysregulation drives tumor-cell proliferation, survival, motility, invasion, and metastasis across multiple cancers, and pharmaceutical companies have developed c-Met inhibitors specifically as anti-cancer drugs.^[7] Dihexa turns up the very pathway those drugs are designed to turn down.

What the evidence says: no published dihexa study reported tumor formation, and the rodent experiments did not flag it. But those studies were short (weeks) and were never designed to assess carcinogenicity. No study has measured the effect of chronic HGF/c-Met amplification in humans over months or years. The theoretical worry is greatest for anyone with an undiagnosed or pre-malignant condition, where amplifying this pathway could, in principle, accelerate disease. This risk currently cannot be quantified — and that uncertainty is itself the point.

Contraindications and who should avoid it

• Active or prior cancer, or elevated cancer risk: the HGF/c-Met mechanism makes dihexa inappropriate for anyone with current, recent, or suspected malignancy.
• Pregnancy and breastfeeding: no reproductive-safety data exist — avoid.
• Cardiovascular disease or on RAS-acting drugs: given its angiotensin-family origin, people with hypertension or on ACE inhibitors / ARBs should not use it without medical oversight.
• Under age 25: the brain is still developing, and the effect of driving synaptogenesis in a developing brain is entirely unstudied.
• On c-Met-targeted or other oncology therapy: dihexa opposes the mechanism of c-Met inhibitors — combining them would be counterproductive and potentially dangerous.

The limitations you must keep in mind

• No human pharmacokinetics: absorption, half-life, distribution, and metabolism in people are unknown.
• No human therapeutic window: the gap between an effective dose and a harmful one has never been measured.
• No long-term data: effects of repeated cycles on brain structure, cancer risk, or cardiovascular function are uncharacterized.
• No drug-interaction studies, and product-quality variability adds a separate, real hazard.

For broader context, see Are Peptides Safe?

Why this section matters: dihexa is an unregulated research compound, so identity and purity vary enormously from vendor to vendor. For an unproven, potent molecule, contamination and mislabeling are practical risks that sit right alongside the pharmacology. Knowing how to read a Certificate of Analysis (COA) is the most useful skill here.

What a credible product should show

• Third-party COA: independent HPLC purity testing (look for ≥98%) plus mass-spectrometry identity confirming the expected molecule — not a certificate written by the seller.
• Batch-specific results: the COA should reference the exact lot you are buying.
• Endotoxin testing (LAL): important for anything intended to be injected.
• Proper form and packaging: lyophilized powder (or capsules) in sealed, light-protected containers.

Red flags

• No COA, or a COA from the seller rather than an independent lab
• Pre-mixed "ready to use" liquid (shorter shelf life, contamination risk)
• Prices far below the market
• Explicit human-use or medical claims, which signal a non-compliant, higher-risk vendor

Legal and regulatory status (2026)

• Not FDA-approved for any use. Dihexa has never been submitted for or granted FDA approval and is not a prescription drug, dietary supplement, or over-the-counter product.
• Research-chemical only: it is sold labeled "research chemical / not for human consumption." No IND program and no registered human clinical trial has been published.
• Not a controlled substance: dihexa is not DEA-scheduled in the US, so possession is not criminalized. Legal status can vary by country.
• Sports: compounds that amplify growth-factor signaling are the kind anti-doping authorities scrutinize; competitive athletes should assume it is bannable and check current WADA guidance.

Storage

Keep unopened lyophilized vials cold and dark (freezer for long-term, refrigerator for shorter). Once reconstituted, refrigerate at 2–8°C and use within a few weeks; never freeze a reconstituted solution, and discard anything cloudy or discolored. Capsules should be kept cool, dry, and out of light. For the full legal picture, read Are Peptides Legal?

Dihexa is usually weighed against a few better-established nootropic peptides. None of these comparisons rest on head-to-head human trials — they contrast proposed mechanisms and evidence levels. The recurring theme: dihexa has the most dramatic preclinical results and by far the least human data.

Compound	Proposed mechanism	Evidence level	Key point
Dihexa	HGF/c-Met → synaptogenesis (proposed; key mechanism paper retracted 2025)	Preclinical only (rats, cells)	Most potent synaptogenic compound reported; no human data; cancer-pathway concern
Semax	BDNF modulation, neurotransmitter effects	Clinical use in Russia	Decades of human use, far better-characterized safety
Selank	GABA/serotonin modulation, anxiolytic + nootropic	Clinical use in Russia	Dual anti-anxiety and cognitive angle

Key distinction: if you want the biggest theoretical upside and are comfortable being an experiment of one with a genuinely unresolved cancer-pathway question, dihexa is the aggressive end. If you want something with an actual human track record, semax and selank are the more defensible starting points. See the cognitive enhancement guide for the full comparison.