What is Dihexa? origin, mechanism & research.

What Dihexa is, exactly

Dihexa is a small synthetic molecule whose formal chemical name is N-hexanoic-Tyr-Ile-(6) aminohexanoic amide. It is sometimes described in popular writing as a hexapeptide, but structurally it is closer to a dipeptide, the tyrosine-isoleucine core, flanked by two aliphatic chains: an N-terminal hexanoyl group and a C-terminal six-carbon amino acid amide. Those flanking chains are not decorative. They are the modifications that turn an unstable native peptide into a drug-like molecule.

The native molecule that Dihexa was derived from is angiotensin IV (Ang IV), a short hexapeptide fragment of the broader renin-angiotensin system. Angiotensin IV is a known ligand at the AT4 receptor, which has been characterized as insulin-regulated aminopeptidase (IRAP), and the AT4 / IRAP system is associated with memory and learning in rodent models. Native angiotensin IV, however, has two serious drawbacks as a drug candidate: it is rapidly degraded by peptidases in vivo, and it does not cross the blood-brain barrier in any therapeutically useful way.

Dihexa was designed to keep the procognitive activity of angiotensin IV at the AT4 site while solving both problems. The aliphatic modifications resist proteolysis, give the molecule oral bioavailability in rodents, and allow it to cross the blood-brain barrier. In short: peptide-like pharmacology, drug-like distribution. That design rationale is the foundation of the entire research program.

It is worth being clear about what Dihexa is not. It is not an FDA-approved drug. It is not a generic of any approved medication. It has not been the subject of a completed sponsor-led IND program. The published preclinical work is real and was peer-reviewed; the path from that work to a human therapeutic has not been walked.

The Wright and Harding lab origin story

The compound was developed at Washington State Universityby the laboratory groups of Dr. Joseph W. Harding and Dr. John W. Wright. Their work on the angiotensin system, and on AT4 / IRAP pharmacology specifically, predates Dihexa by roughly two decades. The lab’s long-standing hypothesis is that the AT4 receptor system is closely tied to memory and synaptic function, and that pharmacologically engaging this system in a stable, brain-penetrant form could yield a candidate molecule for Alzheimer’s disease and other neurodegenerative conditions.

The published trail of the Dihexa project itself spans roughly fifteen years. The most-cited preclinical paper is McCoy et al., 2013, “Evaluation of Metabolically Stabilized Angiotensin IV Analogs as Procognitive / Antidementia Agents,” published in the Journal of Pharmacology and Experimental Therapeutics (volume 344, pages 141 to 154). That paper laid out the preclinical case for Dihexa as a procognitive candidate, including in vivo behavioral findings in rodent cognition models. A subsequent review by Wright, Kawas, and Harding (2015), “The development of small molecule angiotensin IV analogs to treat Alzheimer’s and Parkinson’s diseases,” published in Progress in Neurobiology (volume 125, pages 26 to 46), synthesized the preclinical program and articulated the path the lab hoped to walk toward clinical translation.

A note on the integrity of the underlying record. In September 2021, the editors of Journal of Pharmacology and Experimental Therapeutics issued expressions of concern on four Harding-lab papers, including the McCoy 2013 paper above, after questions about possible image manipulation surfaced through the post-publication review process. A subsequent Washington State University investigation concluded that figures in several of those papers had been falsified or fabricated, with the central responsibility assigned to lead co-author Leen H. Kawas. In April 2025, the closely related follow-up paper Benoist et al., 2014, “The Procognitive and Synaptogenic Effects of Angiotensin IV-Derived Peptides Are Dependent on Activation of the Hepatocyte Growth Factor / c-Met System,” published in the same journal (volume 351, pages 390 to 402), was formally retracted on the basis of those findings. The McCoy 2013 paper itself remains published with an expression of concern attached. Any honest summary of the Dihexa preclinical record has to take the data-integrity history into account, particularly because the retracted Benoist 2014 paper is the canonical source of the HGF / c-Met dependence experiment that anchors the lab’s mechanistic story.

The Harding and Wright groups produced the foundational preclinical literature, and the broader independent replication base outside the originating laboratory is limited. This is a recurring pattern in academic peptide pharmacology and is not unique to Dihexa. The science of getting a procognitive small molecule from a university bench through a multi-year, multi-site Alzheimer’s clinical program is enormously expensive and requires industry partners, regulatory investment, and patient-recruitment infrastructure that academic labs do not possess.

The result is the situation Dihexa has been in since roughly 2014: the preclinical case is on the record, the lab’s aspirations are on the record, and the clinical program never started.

The proposed mechanism: HGF / c-Met

The mechanistic hypothesis at the heart of the Dihexa program is hepatocyte growth factor (HGF) and c-Met receptor agonism. HGF is a multifunctional growth factor, and c-Met is its receptor tyrosine kinase. Together, the HGF / c-Met system is implicated in cell growth, motility, tissue patterning, and, in the central nervous system, synaptic plasticity and neuronal survival.

The argument the Harding lab has advanced is that Dihexa potentiates HGF activity at c-Met in central nervous system tissue, and that this drives the downstream effects observed in their preclinical assays: increased dendritic spine formation, changes in synaptic connectivity markers, and, in living rodents, improvements in memory performance in standard cognitive paradigms.

Synaptogenesis

Synaptogenesis is the formation of new synaptic connections between neurons. In a brain undergoing neurodegeneration, the loss of these connections is one of the structural correlates of cognitive decline. A compound that drives new synapse formation, or that increases dendritic spine density (a structural proxy for excitatory synaptic input), is mechanistically interesting as a candidate for cognitive disorders. The Harding lab’s preclinical case is built on observations of exactly this kind in rodent hippocampal neuron cultures and in living rodents.

The HGF-dependence test, and why it now stands on contested ground

The cleanest piece of mechanistic evidence the lab offered for the HGF / c-Met hypothesis was the HGF-dependence experiment: in cell-culture systems where HGF / c-Met signaling was blocked (HGF antagonist or shRNA against c-Met), Dihexa exposure no longer produced the spine-density effect. That experiment is the centerpiece of the 2014 Benoist et al. paper in Journal of Pharmacology and Experimental Therapeutics, not of the McCoy 2013 paper.

The Benoist 2014 paper was retracted in April 2025 after the Washington State University investigation concluded that several figures had been manipulated. That retraction does not, by itself, prove the HGF / c-Met hypothesis is wrong. The hypothesis remains an active area of research interpretation in the angiotensin IV literature. It does mean the single piece of evidence most commonly cited in support of the mechanism has been formally withdrawn from the peer-reviewed record, and any clean mechanistic claim attributed to that paper is no longer one a careful reader can rely on.

Why this mechanism is also a safety question

The same HGF / c-Met pathway that makes Dihexa pharmacologically interesting in a cognitive context also has well-characterized roles in oncology. c-Met activation is a documented driver of cell proliferation, migration, and angiogenesis in multiple solid tumors. This is not speculation: HGF / c-Met is the target of approved oncology drugs that work by inhibiting the pathway, not potentiating it. A compound that potentiates HGF activity in a CNS-penetrant form inherits a theoretical concern about pro-tumor signaling, particularly in patients with undiagnosed malignancy. This concern has not been ruled out clinically because no clinical safety program has been conducted.

The state of the evidence: preclinical only

This is the single most important paragraph in this article. After roughly fifteen years of academic work, there are no completed, peer-reviewed human clinical trials of Dihexa. No published phase 1 safety study. No phase 2 efficacy data in Alzheimer’s disease, traumatic brain injury, mild cognitive impairment, or any other cognitive indication. No human pharmacokinetic profile published in a peer-reviewed venue. The compound’s reputation rests entirely on preclinical promise plus several years of online nootropic-community use, not on clinical evidence.

The preclinical evidence base, viewed honestly, contains the following:

• Cell-culture spine density work. Dose-dependent increases in dendritic spine density in dissociated rat hippocampal neuron cultures exposed to Dihexa at picomolar concentrations, with reported potency several orders of magnitude greater than the parent angiotensin IV in this specific in-vitro assay. (Several of the relevant papers carry expressions of concern, and the closely related Benoist 2014 paper was retracted in April 2025; see the integrity note above.)
• Rodent cognition assays. Improved performance in standard preclinical screens for cognitive-enhancing compounds: the Morris water maze for spatial learning, scopolamine-induced amnesia models for muscarinic-antagonist-induced cognitive deficit, and aging rodent paradigms.
• A coherent mechanistic story. The HGF-dependence finding, the AT4 / IRAP receptor pharmacology, and the blood-brain-barrier-penetrance characterization all line up into a single pharmacological hypothesis.
• A theoretical clinical case. Reviews from the originating laboratory have argued for Dihexa as a candidate in Alzheimer’s disease, traumatic brain injury, and conditions characterized by lost synaptic connectivity. These remain research hypotheses consistent with the preclinical mechanism, not established indications.

The “10 million times more potent than BDNF” claim

The single most-quoted claim about Dihexa is that it is roughly 10 million times more potent than BDNF (brain-derived neurotrophic factor). This claim has a real source: it originates from a specific in-vitro spine-density assay in the Harding lab’s cell-culture work, in which Dihexa drove a measurable spine-density effect at picomolar concentrations while BDNF required micromolar concentrations to produce a comparable effect in the same system. That ratio, roughly seven orders of magnitude in molar concentration, is what the “10 million times” (or, in some popular renderings, “100,000 times”) figure refers to. The seven-orders-of-magnitude framing is the one closest to what the originating laboratory reported in print.

What it does not refer to is a clinical comparison. It is not a head-to-head trial. It is not a measure of cognitive efficacy in humans. It is a relative potency value from one in-vitro assay context, in cell culture, where the readout was a structural proxy (spine density) rather than a behavioral or clinical outcome. Treating it as a clinical claim is a category error. It is an interesting preclinical observation; it is not evidence that Dihexa produces seven orders of magnitude more cognitive benefit than BDNF in any human.

The careful read of this number is: in one in-vitro spine-density readout, this specific synthetic small molecule was a far more potent driver of the measured signal than the canonical neurotrophic protein BDNF was. That is genuinely interesting at the laboratory level. It is not a statement about clinical effect. The integrity caveat above also applies: the spine-density assays sit inside the same cluster of papers that drew expressions of concern and, in the case of the Benoist 2014 paper, a formal retraction.

Why Dihexa never reached human trials

Readers reasonably ask: if the preclinical case has been on the record since 2013, why has there never been a published human clinical trial? Several factors converge.

1. Funding structures.A complete Alzheimer’s disease clinical program (phase 1 safety, phase 2 efficacy, phase 3 confirmatory) routinely costs hundreds of millions of dollars. That funding comes from pharmaceutical industry partners, large biotech investment, or NIH-scale grants in the rare cases where academic groups secure them. Academic peptide programs without industry partners struggle to clear that capital bar. Dihexa is, on the public record, an academic program without a large industry sponsor.
2. The Alzheimer’s clinical graveyard.Even when Alzheimer’s candidates do reach clinical trials, the historical failure rate is among the worst in pharmaceutical development. Many promising preclinical compounds have failed in human efficacy trials. Sponsors are reasonably reluctant to invest hundreds of millions in a candidate without strong, often replicated, clinical signals earlier in the pipeline.
3. Regulatory pathway difficulty. Alzheimer’s disease is a heavily regulated indication with stringent efficacy and safety expectations. The FDA’s endpoint requirements for cognitive disorders are demanding. A novel mechanism (HGF / c-Met agonism in CNS) compounds the regulatory complexity, particularly given the oncology associations of the same pathway.
4. Patentability and commercial pull. Small synthetic molecules of academic origin, at this stage, often face intellectual-property complications that affect industry licensing willingness. The economics of taking a non-blockbuster molecule through a multi-year cognitive trial are not always favorable.

None of those factors invalidate the preclinical findings. They explain why the evidence base remains preclinical.

FDA Category 2 and the post-April 2026 regulatory ambiguity

Dihexa’s regulatory status in the United States cannot be summarized as “legal” or “illegal” in the binary sense. It sits in a regulatory ambiguity shaped by the FDA’s recent reshuffle of bulk drug substances used in compounding.

The FDA maintains a categorization framework for substances compounding pharmacies want to use. Category 1 substances may be compounded under appropriate conditions. Category 2 substances are determined to be unsuitable for compounding at this time. Category 3 substances are still under evaluation.

On April 15, 2026, the FDA published a list of peptide-related compounds (commonly referenced as the transitional or post-April 15 list) that were removed from the prior Category 2 status but not affirmatively placed onto Category 1, pending review by the Pharmacy Compounding Advisory Committee (PCAC). Most of the listed compounds are scheduled for PCAC review on July 23 and 24, 2026, with the remainder to be reviewed by the end of February 2027. Dihexa Acetate is on that list.

The practical implication is that Dihexa Acetate’s regulatory posture today is ambiguous: removal from Category 2 lifts the prior compounding prohibition, but Dihexa Acetate has not been affirmatively cleared for pharmaceutical or 503A compounding use either. PepScribe’s pharmacy standard is 503A only, and our position is that a compound which has not been affirmatively cleared for 503A compounding cannot be commercially offered through PepScribe’s program. PepScribe does not currently offer Dihexa.

Safety considerations: the absence-of-evidence problem

Safety is the single most consequential dimension of any cognitive peptide conversation, and it is the area where the Dihexa community discussion most often falls short.

No completed human safety trials

There are no completed phase 1 safety studies of Dihexa in humans published in the peer-reviewed literature. Self-reported adverse effects in nootropic-forum use have included headache, lightheadedness, and reported blood pressure changes, but anecdote-grade reports are not safety data. They are not controlled, they are not blinded, they capture no incidence rate, they cannot establish a dose-response curve, and they tell us nothing useful about long-term consequences. Absence of well-documented harm in forum reports is not evidence of safety.

The mechanism-level oncology flag

As noted in the mechanism section above, the HGF / c-Met pathway is a well-characterized driver of malignancy in oncology. A CNS-penetrant pro-growth signaling agent acting on a pathway with known oncology implications is a higher-stakes intervention than a peripheral peptide. The clinical safety work that would address this concern, in any rigorous way, has not been done.

Quality of research-channel material

Dihexa sold through research-chemical suppliers is not pharmaceutical-grade. There is no potency assay, no impurity profile, no sterility testing, no chain-of-custody documentation. Users obtaining material from these channels are dosing a compound whose identity, purity, and stability cannot be independently verified, and they are doing so without the human safety baseline that any cognitive therapeutic would normally require.

Administration routes in the published research

In the published preclinical work, Dihexa has been administered via two primary routes in rodents:

• Oral administration. Because the molecule was specifically engineered for oral bioavailability, much of the rodent pharmacokinetic and behavioral characterization is structured around oral dosing. The N-terminal hexanoyl group and the C-terminal aliphatic chain are explicitly there to resist gut peptidases and enable oral absorption.
• Intracerebroventricular (ICV) injection. In a subset of mechanistic studies, Dihexa was delivered directly into the rodent cerebroventricular system to characterize central effects without having to disentangle peripheral pharmacokinetics. ICV injection is a research route in animals, not a viable human route.

There is no FDA-approved dosing protocol for Dihexa, no validated human pharmacokinetic profile, and no clinical consensus on dosing in any indication. Discussion of preclinical administration routes is research context. It is not dosing guidance.

Red flags in pro and anti Dihexa content

The information environment around Dihexa is unusually unbalanced. Here is a framework for reading the claims you are likely to encounter.

Red flags in pro-Dihexa content

• “10 million times more potent than BDNF” (or “100,000 times”) presented without the in-vitro spine-density assay context that produced it, and without the data-integrity history attached to that body of work
• Specific cognitive outcome promises (“reverse cognitive decline,” “cure dementia,” “regrow brain connections”)
• Claims of clinical efficacy in Alzheimer’s disease, traumatic brain injury, or other cognitive indications, as if those indications had been studied in humans
• Vendors selling research-chemical Dihexa with implicit purchase-for-self-use framing
• Forum testimonials presented as evidence of efficacy or safety
• Failure to disclose that no completed peer-reviewed human clinical trial exists

Red flags in anti-Dihexa content

• Dismissing the published preclinical literature as if it did not exist or were inherently fraudulent
• Confusing “not proven in humans” with “proven not to work in humans”
• Treating the academic origin of the work as inherent evidence of low quality (the Harding lab’s preclinical publications are on the peer-reviewed record, even though several of them carry expressions of concern and one closely related paper has been retracted; the appropriate response is calibrated skepticism, not blanket dismissal)

What balanced evaluation looks like

• Acknowledging the preclinical case is on the public record and mechanistically coherent, while noting that several of the underlying papers carry expressions of concern and that the foundational HGF / c-Met dependence paper (Benoist et al., 2014) was retracted in April 2025
• Being explicit that no completed human clinical trial of Dihexa has been published
• Recognizing that the “10 million times more potent than BDNF” figure (often paraphrased as “100,000x”) is an in-vitro potency value from one assay, not a clinical claim
• Treating the HGF / c-Met oncology association as a real, unresolved mechanism-level safety question
• Distinguishing between the academic literature and the gray-market vendor ecosystem that has formed downstream of it
• Maintaining intellectual humility about a CNS-active compound with no human safety baseline

Where this leaves you

Dihexa is a compound the published evidence supports treating as an interesting research candidate, not as an available cognitive therapeutic. The proposed mechanism is coherent, the preclinical findings are striking, and the clinical evidence does not yet exist. The peer-reviewed record itself is also under stress: expressions of concern attached to several Harding-lab papers and the April 2025 retraction of the Benoist 2014 mechanistic paper mean a careful reader should weight even the preclinical claims more cautiously than they would for a comparable, untouched literature. For a CNS-penetrant pro-growth signaling agent with a known oncology-associated pathway, the absence of human safety data is not a technicality; it is the central fact of the compound’s current status.

If you are evaluating peptides for cognitive support, the responsible path is medical evaluation by a licensed clinician, not gray-market self-administration of a research compound. PepScribe does not currently offer Dihexa, and we will not market it as a commercial product. Transitional research peptides such as Dihexa are evaluated in a clinician-led conversation, not sold from a landing page.