Products are for research use only, not for human use

Small molecule

Dihexa

Dihexa is a synthetic small molecule derived from an angiotensin IV fragment. It is studied as a modifier of hepatocyte growth factor signalling at the receptor tyrosine kinase MET.

Updated · 38 sources read · How this library is built

  1. Hexanoyl1
  2. Tyr2
  3. Ile3
  4. NH4
  5. (CH2)55
  6. CONH26
Linear notation for a capped dipeptide amide rather than a peptide sequence: a hexanoyl group on the tyrosine alpha-amine, an L-Tyr-L-Ile amide bond, and the isoleucine carboxyl amidated to the amine of 6-aminohexanoic acid, whose own carboxyl is a primary amide. Only two of the three building blocks are alpha-amino acids, which is why no one-letter sequence is given.. Outlined positions carry a modification listed below.

Dihexa in brief

Dihexa is a synthetic capped dipeptide amide (C27H44N4O5, 504.7 g/mol) built from the Tyr-Ile core of an angiotensin IV analogue, with a hexanoyl cap at one end and a 6-aminohexanoic amide at the other; chemical databases index it as a small molecule. It was proposed to bind hepatocyte growth factor and promote activation of the receptor MET, but the paper reporting that binding was retracted in April 2025 and no other laboratory has measured it. Independent work found effects in zebrafish hair cells and substitution for the growth factor in two stem cell protocols, against one published negative result in a rat neurotoxin model. It holds no marketing authorisation and no International Nonproprietary Name, and the FDA records its nomination as a compounding bulk substance as withdrawn, stating it has identified no human exposure data for it. Its phosphate prodrug, fosgonimeton, met neither primary nor secondary endpoints in a phase 2/3 trial reported in 2025.[1,2,3,4,5,6,7,8,9]

Dihexa regulatory status

Two facts lead here. Three papers from the originating laboratory, including the one reporting binding to hepatocyte growth factor, were retracted in April 2025 after Notices of Concern issued in September 2021; the paper that names the molecule carries one of those notices and was not retracted. And the FDA lists dihexa acetate in the table of bulk drug substances nominated for compounding but withdrawn (page content current as of 22 April 2026), stating that it has not identified any human exposure data on drug products containing dihexa acetate and that it lacks important information regarding any safety issues raised by the substance, including whether it would cause harm if administered to humans. Beyond that: no FDA, EMA or Health Canada marketing authorisation was found for dihexa or for its phosphate prodrug fosgonimeton in Drugs@FDA (queried through openFDA), the EMA medicines database or the Health Canada Drug Product Database (searched November 2026), and no International Nonproprietary Name exists for dihexa, the name fosgonimeton belonging to the ester. The prodrug reached a completed phase 2/3 trial in mild-to-moderate Alzheimer's disease (NCT04488419) that met neither its primary nor its secondary endpoints. The 2026 WADA Prohibited List does not name dihexa; its class S0 prohibits at all times any pharmacological substance not addressed elsewhere on the List and with no current approval by a governmental regulatory health authority for human therapeutic use, and its class S2.3 prohibits hepatocyte growth factor by name together with other growth factors and growth factor modulators affecting the properties that class lists, so a compound presented as a modifier of that system falls to be considered under both.[10,3,8,11,12,13,9,7,14,15]

Dihexa at a glance

  • A capped dipeptide amide, C27H44N4O5, average mass 504.7, three defined stereocentres corresponding to L-tyrosine and L-isoleucine, with a hexanoyl group on one end and a 6-aminohexanoic amide on the other. No charge, no ring other than the tyrosine phenol, eighteen rotatable bonds, and no experimental or computed three-dimensional structure anywhere.[2,16]
  • It was designed rather than discovered: the active core of an angiotensin IV analogue was narrowed to the tripeptide Nle-Tyr-Ile and both ends were then modified with the stated aim of raising hydrophobicity and reducing hydrogen bonding.[1,17]
  • The mechanism it is named for, binding hepatocyte growth factor and promoting activation of the receptor tyrosine kinase MET, was established in papers from the originating laboratory that were flagged in 2021 and three of which were retracted in April 2025. The specific paper that reported the binding is one of the retracted three.[18,19,20,10,3]
  • Independent work exists and does not depend on those papers for its own measurements. Counting outcomes rather than papers: four independent studies reported an effect (concentration-dependent hair-cell survival endpoints in larval zebrafish, substitution for recombinant hepatocyte growth factor at the maturation step of two separate human stem cell differentiation protocols, and molecular and histological endpoints in APP/PS1 mice), and one reported none (a rat mitochondrial-toxin model, where no measured endpoint changed). None of the five measured binding to hepatocyte growth factor or activation of MET.[4,21,5,6,22]
  • Phosphorylating the tyrosine phenol gives fosgonimeton, the clinical candidate whose active metabolite the developer states is this molecule. A phase 2/3 trial of that prodrug reported in 2025 that it met neither its primary nor its secondary endpoints.[23,7,24]

What is Dihexa?

Dihexa is a laboratory-made molecule with a traceable design history. Work on angiotensin IV, a six-residue fragment of the renin-angiotensin cascade, had reported that certain analogues altered rodent memory-task performance while being degraded quickly and crossing biological barriers poorly. A 2011 study narrowed the activity of one analogue, Nle(1)-angiotensin IV, to its first three residues, Nle-Tyr-Ile. Dihexa is what resulted from replacing the ends of that tripeptide with a six-carbon acyl group and a six-carbon amino-acid amide, with the stated aim of making the molecule more hydrophobic and less able to donate hydrogen bonds. The 2013 paper reporting that series is where the name first appears; its authors were at Washington State University, and a company founded on the chemistry carried the phosphate ester of the molecule into clinical development, named Athira Pharma in the 2021 reporting on the image findings and appearing as LeonaBio on the trial registry record.[1,17,25,26,14]

Chemically the result is no longer a peptide in any useful sense. Two amino acids remain, both in natural L configuration, and the free amine and free acid that define a peptide's ends are both gone. What survives from the ancestry is a tyrosine phenol and an isoleucine side chain in a fixed relative arrangement; what is new is two flexible neutral chains. The chemical databases index it as a small molecule, which is the right reading.[1,2]

Dihexa itself is not a clinical substance, and the distinction from its ester is easy to lose. It has an FDA substance registry record and a UNII, 9WYX65A5C2, whose code list runs to a CAS number, a PubChem CID and an EPA CompTox identifier and carries no International Nonproprietary Name; the INN in this family, fosgonimeton, belongs to the phosphate ester. No regulator checked has authorised either molecule for anything, and the whole clinical programme, from the phase 1 study to the phase 2/3 trial that reported in 2025, ran on the ester rather than on this compound.[27,7,24,9]

What dihexa is studied for is a protein interaction rather than a receptor in the usual sense. Hepatocyte growth factor is a secreted protein that activates MET, a receptor tyrosine kinase, and is generally understood to do so as a dimer or multimer, so a molecule that helps two copies associate could in principle raise signalling without being a receptor ligand at all. Atomic force microscopy from the same laboratory reported activated MET as multimeric and concentrated in dendrites while inactive MET was monomeric on the cell body, which is the physical picture the proposal was built on. Later papers from other groups describe the compound in those same terms, usually citing that work.[18,20,28,29,30,31]

That provenance is the most important thing to know here. Four papers from the originating laboratory, published between 2011 and 2014, received Notices of Concern from the Journal of Pharmacology and Experimental Therapeutics in September 2021 over possible image manipulation, with the journal awaiting a university inquiry; a company founded on the chemistry separately reported that an internal committee had found altered images in a doctoral dissertation and in at least four co-authored papers. In April 2025 three of the four were retracted, among them the 2014 paper reporting the binding to hepatocyte growth factor. The 2013 paper that names the molecule was not retracted and remains under its 2021 notice.[18,19,20,10,3,25,26]

Not to be confused with

Angiotensin IV (Val-Tyr-Ile-His-Pro-Phe)
The endogenous hexapeptide this chemical line descends from, and not the same molecule. Dihexa retains only two of its residues, Tyr and Ile, has lost the free N-terminal amine that angiotensin IV presents to an aminopeptidase, and carries two synthetic non-amino-acid ends. Calling dihexa an angiotensin IV analogue is accurate as lineage and misleading as chemistry.[1,17,2]
Nle(1)-angiotensin IV
The immediate predecessor, in which the first residue of angiotensin IV is replaced by norleucine. Its active core was narrowed to the N-terminal tripeptide Nle-Tyr-Ile, and dihexa is what resulted from modifying the ends of that tripeptide. The two are separate compounds with separate assay histories.[1,17]
Norleual
A close relative from the same programme with the opposite reported pharmacology. Norleual is a longer, reduced-bond peptide reported as an inhibitor of hepatocyte growth factor signalling, and it is still used in other laboratories as a pathway blocker. A near-identical scaffold in this family can be described as blocking rather than potentiating the same system, so names in this series must not be treated as interchangeable.[37,43,44]
6-AH (D-Nle-Tyr-Ile-NH-(CH2)5-CONH2)
The antagonist used as the specificity control in work on dihexa. It shares the Tyr-Ile pair and the aminohexanoic amide tail; the difference is a D-norleucine residue at position one where dihexa has a hexanoyl cap. That one substitution is the whole published distinction between the two opposite activities.[19,4]
Fosgonimeton (ATH-1017, CID 156596375)
Not a different molecule so much as this one with a phosphate group added to the tyrosine phenol, developed as a prodrug because the unphosphorylated compound was reported to be poorly stable to digestive enzymes. The developer states that its active metabolite is the compound described here, which PubChem indexes under this molecule's record as ATH-1001. Clinical results reported for fosgonimeton are results for the phosphate ester in people, not measurements of this compound.[23,27,7,2,24]
Hepatocyte growth factor itself
A large secreted protein, the natural ligand of MET. Dihexa is a 504 dalton synthetic amide and is not a fragment of it, does not share sequence with its receptor-binding domains, and is proposed to act on it rather than to replace it. Descriptions of dihexa as an HGF mimetic are shorthand for a proposed effect on the protein, not a statement of structural similarity.[18,4,31]
Lipids and dyes whose names contain the same letters
A literature search on this name returns several unrelated papers, because 1,2-dihexadecanoyl phospholipids, dihexa-oxacarbocyanine iodide and dimerised melanocyte-stimulating hormone constructs abbreviated diHexa all break across the same string. The three cited here are examples of each kind: none concerns this compound, and a search that is not restricted to the full chemical name or the supplementary concept term will mix them in.[45,46,47]

Dihexa structure and sequence

Read the molecule from one end. A straight six-carbon acyl chain forms an amide to the alpha-amino group of L-tyrosine; the tyrosine carboxyl forms an ordinary peptide bond to L-isoleucine; the isoleucine carboxyl forms an amide to the amino group of 6-aminohexanoic acid, whose own carboxyl is a primary carboxamide. Four amide bonds, two amino acids, no free amine, no free acid, no charge. The groups available for a specific interaction are the tyrosine phenol, the isoleucine branch and the backbone amides.[1,2]

Two features of the shape matter beyond the formula. The first is that the stereochemistry is fully specified and natural, (2S) at the tyrosine centre and (2S,3S) at the isoleucine centres, with no D-residue anywhere. That is the formal difference from 6-AH, the antagonist from the same programme used as the control in independent experiments, which carries a D-norleucine where this molecule carries the hexanoyl cap. Two compounds sharing the Tyr-Ile pair and the same tail have been reported with opposite activities on one protein system, so small changes here are consequential rather than cosmetic.[19,4,2]

The second is flexibility. Eighteen rotatable bonds and a single aromatic ring leave the molecule with no preferred shape a picture could stand for, which is part of why there is no three-dimensional view on this page: the Protein Data Bank holds no entry containing it, by descriptor search or by name, and PubChem holds no computed conformer for its record either. The protein side is better served, with crystal structures of MET bound to the NK1 splice variant of its ligand and to a bacterial agonist, and structures of insulin-regulated aminopeptidase, the competing candidate target for this chemical family, with a macrocyclic inhibitor in its catalytic cleft. Neither set says anything about where this compound goes.[2,16,32,31]

Modifications

  • Tyrosine alpha-amino group (N-terminus) A hexanoyl cap, that is a six-carbon straight-chain acyl group forming an amide. It removes the free alpha-amine, which is both the group an aminopeptidase would act on and the group the predecessor peptides present to the insulin-regulated aminopeptidase binding site. The stated design intent was to raise hydrophobicity and reduce hydrogen bonding; the acyl chain is also what separates this molecule from the antagonist 6-AH, which keeps an amino acid in that position.[1,19,2,40]
  • Isoleucine carboxyl (C-terminus) Extended as an amide to 6-aminohexanoic acid, itself terminating in a primary carboxamide. The result is a flexible five-carbon spacer ending in a neutral amide rather than a free acid, so the molecule carries no formal charge at either end. The same tail appears on the antagonist family, so the tail alone does not determine the reported direction of activity.[1,19,2]
  • Tyrosine phenol (4-hydroxyphenyl) Left free in this molecule, and the single point of attachment that produces the clinical prodrug: phosphorylating this hydroxyl gives fosgonimeton. The phenol is therefore the most chemically consequential handle on the structure.[23,2,24]

Three stereocentres, all defined in the reference record: the tyrosine alpha carbon (S), and the isoleucine alpha and beta carbons (2S,3S). Those assignments correspond to L-tyrosine and L-isoleucine, so the two amino acid units are of natural configuration and there is no D-residue anywhere in the molecule. That matters for comparison with the antagonist family, which is built on a D-norleucine. The hexanoyl cap and the aminohexanoic tail contain no stereocentres at all and are conformationally free: with eighteen rotatable bonds and a topological polar surface area of 151 square angstroms, this is a floppy molecule with no ring system outside the tyrosine phenol.[2,19]

What studies on Dihexa show

Each card names the experimental system and what was measured. A result in one system is not evidence about another.

Medicinal chemistry series with rodent behavioural endpoints; under a 2021 Expression of Concern and not retracted · Hippocampal synaptogenesis assays; Morris water maze in scopolamine-treated rats and in aged rats

The paper reports that several terminal modifications improved stability while keeping the water-maze and synaptogenesis effects, and that further modification aimed at raising hydrophobicity and reducing hydrogen bonding produced the capped amide described on this page, which it names and describes as metabolically stabilised and barrier-permeant. This is where the name and the design rationale first appear, and it is the paper every later description of the series traces back to.[1,33]

Material
The tripeptide Nle-Tyr-Ile and a series of N- and C-terminally modified analogues of it, among them the capped amide the paper names dihexa
Measured
Metabolic stability of each analogue, water-maze performance, and hippocampal synaptogenesis
Compared with
Nle(1)-angiotensin IV and the unmodified tripeptide; scopolamine-treated animals given nothing
Limitation
Carries an Expression of Concern issued by the journal in September 2021 over possible image manipulation; it was not retracted, but three companion papers from the same laboratory were retracted in April 2025. Read as an abstract, so no stability figure, no concentration and no group size is in hand. Rodent behavioural and histological endpoints only: nothing in the work measures a molecular target, and the stability and permeability claims are stated in the abstract without figures.

McCoy AT et al., 2013, The Journal of Pharmacology and Experimental Therapeutics; Journal of Pharmacology and Experimental Therapeutics editors, 2021, The Journal of Pharmacology and Experimental Therapeutics

Whole-organism pharmacology in a transparent model, with pathway inhibitors and an analogue control · Larval zebrafish lateral line neuromasts at 5 to 6 days post-fertilisation; hair-cell survival by DASPEI vital dye and by direct counts in a Brn3c:mGFP transgenic line; fluorophore-tagged gentamicin to measure entry into hair cells

Hair-cell survival varied with concentration, with the clearest effect at a 1 micromolar bath concentration (assay condition). Uptake of the tagged aminoglycoside into hair cells was unchanged, so whatever the compound does happens after the toxin is inside the cell rather than by blocking its entry. Co-application of 6-AH removed the effect entirely; Akt, MEK and TOR inhibitors each removed part of it. The analogue differing at the amino end produced no effect at all, and neither chronic gentamicin nor cisplatin exposure was affected.[4]

Material
Dihexa synthesised to order at 95.1 percent purity, plus an analogue differing at the amino end as the specificity control
Measured
Surviving hair cells after acute neomycin or acute gentamicin exposure; uptake of tagged gentamicin into hair cells; the same survival endpoint under each inhibitor
Compared with
Vehicle; the hepatocyte growth factor antagonist 6-AH; Akt, MEK and TOR inhibitors each at 1 to 10 micromolar assay concentrations; cisplatin as a mechanistically distinct toxin
Limitation
The hepatocyte growth factor dependence rests on 6-AH, a tool compound whose only characterisation is in a paper since retracted, and on pharmacological inhibitors rather than on any measurement of binding or of MET phosphorylation in this system. MET antibody labelling was widespread in skin epithelium as well as in neuromasts, so the cell type acted on is not resolved. The paper's abstract and its results text disagree about whether the control analogue had an amino group added or missing, and the paper mis-describes the 2013 origin study as a Parkinson's model when that study used scopolamine and aged rats.

Uribe PM et al., 2015, Frontiers in Cellular Neuroscience

Directed differentiation of human pluripotent stem cells with a defined chemical medium · Human embryonic and induced pluripotent stem cells taken through definitive endoderm to hepatocyte-like cells; transcript and protein markers; serum protein production, glycogen storage and cytochrome P450 activity as functional readouts

In a growth-factor-free protocol this molecule occupied the place of recombinant hepatocyte growth factor at the maturation step, and the authors state that both it and dexamethasone were required. The resulting cells showed hepatocyte morphology, marker expression and function comparable to growth-factor-derived cells.[6]

Material
Dihexa at 100 nanomolar (assay condition) in the final maturation medium, with dexamethasone
Measured
Whether hepatocyte maturation proceeds without recombinant growth factors, judged by morphology, marker expression and the three functional assays
Compared with
The same protocol driven by recombinant growth factors including hepatocyte growth factor
Limitation
A cell-fate outcome, not a receptor measurement: nothing in the study measures MET, phospho-MET or binding, and the description of the compound as a hepatocyte growth factor agonist is carried over from the originating laboratory rather than tested. The concentration optimisation is reported as data not shown, and the first base medium used turned out to contain EGF, which confounded the earliest growth-factor-free claim until the medium was changed.

Siller R et al., 2015, Stem Cell Reports

Independent replication of the growth-factor substitution in a different protocol · Human pluripotent stem cells taken to hepatoblasts and then to hepatocyte-like cells; AFP, HNF4alpha and albumin expression; immunostaining efficiency; urea and cytochrome P450 function; repopulation of injured mouse liver

The three-small-molecule combination reached marker co-expression efficiency of 95.5 percent against 96 percent for the growth-factor control, so in this protocol the compound stood in for hepatocyte growth factor alongside substitutes for the two other proteins. This is the same substitution result in a different laboratory with a different design.[22]

Material
Dihexa at 0.1 micromolar (assay condition) in the specification, expansion and maturation cocktails, with vitamin C and forskolin
Measured
Induction efficiency of hepatic markers against the growth-factor control
Compared with
BMP2 plus FGF4 as the growth-factor control
Limitation
The compound was never used alone: the paper reports that it or forskolin with vitamin C gave lower induction than growth factors, and only the triple combination matched them, so no single-agent contribution is measurable. Again no MET or phospho-MET measurement, and again the agonist attribution is cited rather than tested.

Pan T et al., 2022, Stem Cell Research & Therapy

Transgenic mouse study with a pathway inhibitor · Six-month-old male APP/PS1 mice against wild-type controls; water maze; Nissl-stained cortical neuron counts; ELISA for angiotensin IV and for TNF-alpha, IL-1beta and IL-10; western blots for PI3K, AKT and phospho-AKT

In the transgenic line, treatment was associated with higher PI3K and phospho-AKT, higher Nissl-positive neuron counts and synaptophysin, lower TNF-alpha and IL-1beta and higher IL-10, and every one of those changes was reversed by wortmannin. The experiment did not establish what receptor the compound engages.[21]

Material
Commercially sourced dihexa
Measured
Water-maze escape latency and platform crossings; cortical Nissl-positive neuron number and synaptophysin; three cytokines; PI3K and phospho-AKT levels
Compared with
Untreated transgenic mice, wild-type mice, and the PI3K inhibitor wortmannin as a mechanistic control
Limitation
No measurement of hepatocyte growth factor or MET appears anywhere in the paper, and PI3K and AKT sit downstream of many receptors, so the wortmannin reversal localises a required pathway and not a target. One result is unexplained by the paper itself: brain angiotensin IV was reported to rise after treatment, although this compound is not angiotensin IV and no metabolic route to it is reported, and that reading rests on a single ELISA. PI3K was normalised to beta-actin rather than to a pathway control, and all endpoints are organism-level rodent measures.

Sun X et al., 2021, Brain Sciences

Negative result in a rodent neurotoxin model · Forty male Wistar rats in three groups, chronic 3-nitropropionic acid exposure over five weeks, with body-weight, motor and cognitive assessment followed by histopathology

The toxin produced the expected deficits and the compound did not alter them on any measured endpoint. The authors report this plainly and suggest other models rather than reinterpreting the result.[5]

Material
PNB-0408, named in the paper as dihexa
Measured
Weight gain, motor function, spatial learning and memory consolidation, and histopathology of the affected regions
Compared with
Vehicle group and a toxin-only group
Limitation
Abstract-level reading only. A single model of mitochondrial poisoning in one laboratory, with no molecular endpoint measured, so it tests neither the proposed mechanism nor the compound's engagement with anything. Its value is that it is a published negative in a literature with very few.

Wells RG et al., 2024, Journal of Huntington's Disease

Primary neuron pharmacology with pathway inhibitors, company-authored · Primary rat cortical neurons challenged with amyloid-beta 1-42, and a rat model; mitochondrial oxidative stress and cytochrome c release; ERK, AKT and GSK3beta activity; ULK1 and Beclin-1; a glutamate challenge with AKT and MEK/ERK inhibitors

In challenged primary neurons, treatment was associated with greater survival and preserved neurite networks, less tau phosphorylation and lower GSK3beta activity, with greater ERK and AKT activation. Protection against the glutamate challenge was abolished by an AKT inhibitor or a MEK/ERK inhibitor, which is the same inhibitor pattern the zebrafish work reported in a different organism.[34]

Material
Fosgo-AM, which is this molecule under its programme name, and the prodrug fosgonimeton
Measured
Neuronal survival and neurite network integrity, tau phosphorylation, markers of mitochondrial stress, and the survival endpoint under each inhibitor
Compared with
Vehicle-treated challenged neurons; AKT and MEK/ERK inhibitors
Limitation
Read as abstract only, so no figure or concentration is in hand, and the work is by employees of the company developing the prodrug. As in every other study here, the inhibitors show which intracellular pathways the effect requires and not which molecule the compound binds; no binding or structural measurement is reported.

Reda SM et al., 2024, Neurotherapeutics

Retracted mechanistic study, cited for provenance only · Cell scattering and MET phosphorylation assays with subthreshold hepatocyte growth factor; hippocampal neuron spine and synapse counts; a rat water-maze experiment with a hepatocyte growth factor antagonist

This is the origin of the claim that the molecule binds hepatocyte growth factor with high affinity and potentiates MET activation, and it is the source nearly every later description cites. It was retracted in April 2025 after a 2021 Notice of Concern, so it supports no finding and is listed here because the mechanism cannot be discussed honestly without saying where it came from.[18,3,10]

Material
Dihexa and Nle(1)-angiotensin IV
Measured
Reported binding to hepatocyte growth factor, MET phosphorylation, cell scattering, spine and synapse number, and water-maze performance
Compared with
Hepatocyte growth factor itself; an antagonist; short hairpin RNA against MET
Limitation
Retracted. Nothing in it can be relied on, and no other laboratory has published a direct binding measurement for this molecule, so the central mechanistic claim currently rests on withdrawn work plus downstream pathway inhibitor experiments that are consistent with it without testing it.

Benoist CC et al., 2014, The Journal of Pharmacology and Experimental Therapeutics; Benoist CC et al., 2025, The Journal of Pharmacology and Experimental Therapeutics; Journal of Pharmacology and Experimental Therapeutics editors, 2021, The Journal of Pharmacology and Experimental Therapeutics

Randomised, placebo-controlled, quadruple-masked phase 2/3 trial of the PRODRUG, not of this molecule · LIFT-AD (NCT04488419) in participants with mild-to-moderate Alzheimer's disease; primary analysis population of 287 participants not taking an acetylcholinesterase inhibitor, randomised one to one against placebo; safety population 549; the registry records the study as completed, running from September 2020 to July 2024 with 554 participants enrolled

The trial achieved neither its primary nor its secondary endpoints. The between-group difference in least-squares mean change from baseline to week 26 was minus 0.08 (standard error 0.10, p = 0.70) on the Global Statistical Test, minus 0.70 (0.77, p = 0.35) on ADAS-Cog11 and plus 0.67 (0.92, p = 0.61) on ADCS-ADL23, all small differences favouring the ester. Among the plasma biomarkers only phosphorylated tau 217 changed nominally significantly. Serious adverse events were balanced between groups, 4.2 per cent against 6.9 per cent on placebo, while 14.2 per cent of the treated group withdrew because of adverse events against 4.6 per cent on placebo, most of them local reactions at the site of administration.[7]

Material
Fosgonimeton, the phosphate ester of this molecule, whose active metabolite the developer states is this molecule
Measured
A Global Statistical Test combining ADAS-Cog11 and ADCS-ADL23 as the primary endpoint at week 26; ADAS-Cog11, ADCS-ADL23 and neurofilament light as secondary endpoints; plasma Alzheimer biomarkers as exploratory endpoints
Compared with
Placebo
Limitation
A result about the phosphate ester in people and not a measurement of this molecule. No MET or hepatocyte growth factor engagement endpoint was reported, so the trial does not test the proposed mechanism directly, and the authors' reading of the consistent non-significant differences as possible biological activity is an interpretation the endpoints do not establish. The safety population is not the efficacy population: it includes a higher dose arm and participants taking an acetylcholinesterase inhibitor.

Porsteinsson AP et al., 2025, Journal of Alzheimer's Disease Reports

What is actually established about how dihexa acts, and what rests on retracted work

In one sentence each, the proposal and the evidence. The proposal is that dihexa helps two copies of hepatocyte growth factor associate, so that the growth factor activates its receptor MET more strongly without dihexa touching the receptor at all. What independent laboratories have actually measured is narrower: that the effects they observe require AKT and MEK or ERK signalling inside the cell, with no measurement anywhere outside the retracted work of binding to the growth factor or of MET activation. The structural literature on MET activation shows ligand-mediated complexes with a ligand dimer at the centre and one receptor bound at each side, so the proposal is at least geometrically coherent; the originating laboratory advanced it on the basis of sequence similarity between these analogues and the growth factor's own dimerisation region, and reported that dihexa binds the protein and raises MET phosphorylation when the protein sits below its own threshold for signalling.[18,20,4,34,31]

The mechanism class is chemically plausible, although the 1999 work usually cited for it establishes less than it is read as establishing. That study measured binding of a small polyanionic ligand, heparin or the disaccharide analogue sucrose octasulfate, to the N-terminal domain of hepatocyte growth factor and to its NK1 splice isoform, by nuclear magnetic resonance and by fluorescence, with an apparent dissociation constant in the micromolar range, and showed by nitrogen-15 relaxation that backbone flexibility around the site fell sharply on binding. Stabilisation of a dimer was not measured: it was proposed as a model for the NK1 isoform, read off that isoform's crystal structure. The equivalent measurement for this molecule is missing altogether. The affinity claim sits in a retracted paper, no other group has published a binding constant or structural evidence of a complex, and the Protein Data Bank holds no entry containing the compound.[18,16,35]

What independent work establishes is narrower and still useful. Across three organisms and two cell systems, the measured effects require AKT and MEK or ERK signalling, and in one case TOR: inhibiting any of them removes part or all of the effect, in zebrafish hair cells, in primary rat cortical neurons, and in APP/PS1 mice where a PI3K inhibitor reversed each of the molecular and histological endpoints it was tested against, the water-maze measure not being among them. Those pathways lie downstream of MET, which is consistent with the proposal, and downstream of many other receptors, which is why consistency is as far as it goes. The strongest evidence that the target is specifically the hepatocyte growth factor system is the substitution result: in two laboratories and two protocols, one of them since published as a reusable protocol unit, the molecule took the place of the recombinant protein at a step where that protein had been required. The weakest link runs through all of it: 6-AH, the antagonist used as the specificity control, was characterised only in a retracted paper.[19,4,21,6,22,36,34]

An older question sits underneath this and has never been closed. The angiotensin IV analogues were first understood to act at a site called the AT4 receptor, which one group identified as insulin-regulated aminopeptidase, an enzyme with crystal structures and measured inhibition by angiotensin IV and its norleucine analogue. The laboratory that made dihexa argued instead for the hepatocyte growth factor and MET system, partly because effects persisted in cells they reported as lacking the aminopeptidase. For this molecule the aminopeptidase route is harder to invoke than for its ancestors, since the free alpha-amine that the catalytic cleft engages has been replaced by an acyl cap. That is an argument from structure rather than a measurement: no assay of this compound against that enzyme was located, and a 2018 systematic review of the wider nonhuman literature on the family did not turn one up.[37,38,39,32,40]

An abstract illustration standing for the kind of process this entry describes. It depicts no molecule and carries no text.
An abstract illustration, chosen to suit the kind of process this page describes. It is decoration rather than evidence: nothing in it is specific to this compound, nothing in it was measured, and it depicts no molecule. Generated from this entry's topic list. Volta Peptides compound library. CC0 1.0

What is still unknown about Dihexa

The integrity problem is the first thing to resolve and it is not resolvable from here. Four papers from the originating laboratory, covering the analogue series and the mechanism attributed to it, were flagged in September 2021; three were retracted in April 2025, including the one reporting binding to hepatocyte growth factor. The notices carry no retrievable body text in PubMed, PMC or the publisher's pages, so the grounds for retraction were not read first-hand here: what is confirmed is which papers, which notices and the dates. A research-integrity publication reported the reason for the 2021 notices as possible image manipulation and quotes a company filing describing an internal finding of altered images; a second outlet corroborates that in general terms. Those are reports, not documents read.[10,3,25,26]

What follows is a gap rather than a contradiction. Nothing published independently contradicts the mechanism; the one measurement that would anchor it has been withdrawn and never replaced. A direct binding experiment, or a structure of a complex, would close a question open since 2014, and the absence is conspicuous given how much later work cites the claim as settled, including 2025 papers whose introductions cite the retracted study. The same gap leaves one consequence of the proposal unexamined: MET is an established oncogene, initiating and sustaining transformation when genetically altered and supporting cancer cell survival and dissemination when transcriptionally activated, and the originating laboratory separately published work in which a close relative of this chemistry acted as a MET antagonist with anticancer activity. A molecule proposed to raise signalling through that receptor therefore raises a question that nothing in its literature answers: no carcinogenicity study, no tumour-promotion experiment and no long-term animal study of this compound was located.[18,20,28,16,41]

The clinical arc is easy to misread in either direction. The phosphate ester of this molecule became fosgonimeton, received an international nonproprietary name, went through a phase 1 study with electrophysiological markers and then a phase 2/3 trial, which reported in 2025 that it met neither its primary nor its secondary endpoints. That is a result about the prodrug in people and says nothing directly about this compound's chemistry; what it does mean is that the hypothesis behind the molecule has had its most expensive test and did not pass it.[27,7,24]

Stability and distribution are claimed rather than measured, and the one number in circulation has no source behind it. The origin series was designed for metabolic stability and barrier permeability, and its abstract states that the terminal modifications improved both; that is the paper now under a 2021 Expression of Concern, it reports no figure in its abstract, and nothing located anywhere gives a half-life, a plasma or brain concentration, a permeability coefficient, a protein-binding figure or a metabolite identification for this molecule. The figure of about a week that circulates as its half-life is traceable: it appears in the FDA substance registry record for the compound, in a 'manufacturer product information' field that quotes a nootropics website, which is not a measurement and is not presented as one by the registry. Measured physical properties are absent too, as the identity gaps record. The clinical pharmacokinetic work that does exist was done on the phosphate ester and belongs to that molecule.[1,27,9]

Two smaller uncertainties are each checkable and none has been checked. The 2021 mouse study reports a rise in brain angiotensin IV after treatment with a compound that is not angiotensin IV and has no reported route to it, on a single immunoassay, and nobody has followed it up. The zebrafish study's abstract and results section disagree about whether its inactive control analogue had an amino group added or removed, which matters because that analogue is the only published structure-activity datum of its kind here. Several nonhuman studies also used the compound only in combination with cells or another agent, so even their organism-level readings cannot be attributed to it alone.[4,21,42]

Key papers on Dihexa

  1. Evaluation of metabolically stabilized angiotensin IV analogs as procognitive/antidementia agents.

    McCoy AT, Benoist CC, Wright JW, Kawas LH, Bule-Ghogare JM, Zhu M, Appleyard SM, Wayman GA, Harding JW (2013), The Journal of Pharmacology and Experimental Therapeutics · Medicinal chemistry and rodent behavioural series

    The origin paper, and the only place the design logic is laid out: why the tripeptide core was capped at both ends and what those modifications were meant to achieve. Read it knowing it carries a 2021 Notice of Concern and was not retracted.

  2. Hepatocyte growth factor mimetic protects lateral line hair cells from aminoglycoside exposure.

    Uribe PM, Kawas LH, Harding JW, Coffin AB (2015), Frontiers in Cellular Neuroscience · Zebrafish lateral line pharmacology

    The most informative independent experiment on the molecule and open access. It has a concentration series, a negative control analogue, an uptake measurement that rules out one whole class of explanation, and four pathway inhibitors, and it is candid about what each step does and does not show.

  3. Small-molecule-driven hepatocyte differentiation of human pluripotent stem cells.

    Siller R, Greenhough S, Naumovska E, Sullivan GJ (2015), Stem Cell Reports · Human stem cell differentiation

    The cleanest demonstration that the compound does something hepatocyte-growth-factor-shaped, in a system with functional readouts and no behaviour in it: it occupies the place of the recombinant protein at the maturation step of a growth-factor-free differentiation protocol.

  4. Efficiently generate functional hepatic cells from human pluripotent stem cells by complete small-molecule strategy.

    Pan T, Wang N, Zhang J, Yang F, Chen Y, Zhuang Y, Xu Y, Fang J, You K, Lin X, Li Y, Li S, Liang K, Li YX, Gao Y (2022), Stem Cell Research & Therapy · Human stem cell differentiation, independent replication

    Read immediately after the previous one. A different laboratory, a different protocol and the same substitution result, which is what turns a single finding into a replicated one.

  5. Retraction notice to "The Procognitive and Synaptogenic Effects of Angiotensin IV-Derived Peptides Are Dependent on Activation of the Hepatocyte Growth Factor/c-Met System" [J Pharmacol Exp Ther 351 (2014) 390-402]

    Benoist CC, Kawas LH, Zhu M, Tyson KA, Stillmaker L, Appleyard SM, Wright JW, Wayman GA, Harding JW (2025), The Journal of Pharmacology and Experimental Therapeutics · Retraction notice

    The April 2025 retraction notice for the mechanism paper, with its two companions for the 2011 and 2012 papers. Anyone reading the older literature on this molecule needs to know which parts of it have been withdrawn before reading any of it.

  6. Effects of an Angiotensin IV Analog on 3-Nitropropionic Acid-Induced Huntington's Disease-Like Symptoms in Rats.

    Wells RG, Azzam AF, Hiller AL, Sardinia MF (2024), Journal of Huntington's Disease · Rat neurotoxin model

    A published negative result, reported without reinterpretation, in a literature that has very few. It is also the only peer-reviewed source tying the code PNB-0408 to this molecule.

  7. Fosgonimeton in mild-to-moderate Alzheimer's disease.

    Porsteinsson AP, Sabbagh M, Tariot PN, Church KJ, San Martin J, Ooi KC, Daggett S, Hale MD, Holub R, Moebius HJ (2025), Journal of Alzheimer's Disease Reports · Phase 2/3 randomised controlled trial

    The end of the clinical programme built on this chemistry, for the phosphate prodrug rather than the compound itself. Cited bibliographically, as context for how much weight the mechanism has carried and what happened when it was tested at scale.

  8. Structural basis of MET receptor dimerization by the bacterial invasion protein InlB and the HGF/SF splice variant NK1.

    Niemann HH (2013), Biochimica et Biophysica Acta · Structural review

    For the structural background the dimerisation claim has to be read against: what is and is not known about how MET is activated by its ligand, from the crystal structures that do exist.

  9. Certain Bulk Drug Substances for Use in Compounding that May Present Significant Safety Risks (Category 2 of the bulk substances nominated under sections 503A or 503B), content current as of 04/22/2026

    U.S. Food and Drug Administration (2026), U.S. Food and Drug Administration · regulatory-document

    The FDA table recording the compounding nomination as withdrawn, and the agency's own words on what it does and does not know about this substance in people. Two sentences, and the most consequential regulatory statement about it anywhere.

References

  1. 1.McCoy AT, Benoist CC, Wright JW, Kawas LH, Bule-Ghogare JM, Zhu M, Appleyard SM, Wayman GA, Harding JW. (2013). Evaluation of metabolically stabilized angiotensin IV analogs as procognitive/antidementia agents. The Journal of Pharmacology and Experimental Therapeutics. 344:141-154. doi:10.1124/jpet.112.199497 · abstract only · expression of concern: Notice of Concern, J Pharmacol Exp Ther 2021;378(3):313, PMID 34551989. Not retracted as of 2026-11-02. This is the paper that first names and describes the molecule.
  2. 2.National Center for Biotechnology Information. PubChem Compound Summary for CID 129010512, Dihexa. PubChem. source · database record
  3. 3.Benoist CC, Kawas LH, Zhu M, Tyson KA, Stillmaker L, Appleyard SM, Wright JW, Wayman GA, Harding JW. (2025). Retraction notice to "The Procognitive and Synaptogenic Effects of Angiotensin IV-Derived Peptides Are Dependent on Activation of the Hepatocyte Growth Factor/c-Met System" [J Pharmacol Exp Ther 351 (2014) 390-402]. The Journal of Pharmacology and Experimental Therapeutics. 392:103567. doi:10.1016/j.jpet.2025.103567 · abstract only
  4. 4.Uribe PM, Kawas LH, Harding JW, Coffin AB. (2015). Hepatocyte growth factor mimetic protects lateral line hair cells from aminoglycoside exposure. Frontiers in Cellular Neuroscience. 9:3. doi:10.3389/fncel.2015.00003 · full text read
  5. 5.Wells RG, Azzam AF, Hiller AL, Sardinia MF. (2024). Effects of an Angiotensin IV Analog on 3-Nitropropionic Acid-Induced Huntington's Disease-Like Symptoms in Rats. Journal of Huntington's Disease. 13:55-66. doi:10.3233/JHD-231507 · abstract only
  6. 6.Siller R, Greenhough S, Naumovska E, Sullivan GJ. (2015). Small-molecule-driven hepatocyte differentiation of human pluripotent stem cells. Stem Cell Reports. 4:939-952. doi:10.1016/j.stemcr.2015.04.001 · full text read
  7. 7.Porsteinsson AP, Sabbagh M, Tariot PN, Church KJ, San Martin J, Ooi KC, Daggett S, Hale MD, Holub R, Moebius HJ. (2025). Fosgonimeton in mild-to-moderate Alzheimer's disease. Journal of Alzheimer's Disease Reports. 9:25424823251405817. doi:10.1177/25424823251405817 · full text read
  8. 8.U.S. Food and Drug Administration. (2026). Certain Bulk Drug Substances for Use in Compounding that May Present Significant Safety Risks (Category 2 of the bulk substances nominated under sections 503A or 503B), content current as of 04/22/2026. U.S. Food and Drug Administration. source · full text read
  9. 9.U.S. Food and Drug Administration; National Center for Advancing Translational Sciences. (2026). Global Substance Registration System record UNII 9WYX65A5C2, DIHEXA. FDA Global Substance Registration System (GSRS). source · database record
  10. 10.Journal of Pharmacology and Experimental Therapeutics editors. (2021). Notice of Concern: Benoist CC, Kawas LH, Zhu M, Tyson KA, Stillmaker L, Appleyard SM, Wright JW, Wayman GA, and Harding JW (2014) The Procognitive and Synaptogenic Effects of Angiotensin IV-Derived Peptides Are Dependent on Activation of the Hepatocyte Growth Factor/c-Met System, J Pharmacol Exp Ther, 351: 390-402. The Journal of Pharmacology and Experimental Therapeutics. 378:311. doi:10.1124/jpet.114.218735concern · abstract only
  11. 11.U.S. Food and Drug Administration. (2026). Drugs@FDA data, openFDA drugsfda endpoint, active-ingredient search for DIHEXA and FOSGONIMETON (searched November 2026). openFDA. source · database record
  12. 12.European Medicines Agency. (2026). Medicines data: all centrally processed human and veterinary medicines (JSON report), searched for dihexa and fosgonimeton, retrieved November 2026. European Medicines Agency. source · database record
  13. 13.Health Canada. (2026). Drug Product Database, active ingredient search for dihexa and fosgonimeton (searched November 2026). Health Canada Drug Product Database. source · database record
  14. 14.ClinicalTrials.gov. (2026). NCT04488419 (LIFT-AD): A randomized, placebo-controlled, double-blind study of ATH-1017 treatment in subjects with mild to moderate Alzheimer's disease, Phase 2/Phase 3, completed. ClinicalTrials.gov. source · database record
  15. 15.World Anti-Doping Agency. (2025). World Anti-Doping Code International Standard: Prohibited List 2026. World Anti-Doping Agency. source · full text read
  16. 16.RCSB Protein Data Bank. RCSB search API: chemical descriptor query on InChIKey XEUVNVNAVKZSPT-JTJYXVOQSA-N and full-text query on the compound name. RCSB PDB. source · database record
  17. 17.Benoist CC, Wright JW, Zhu M, Appleyard SM, Wayman GA, Harding JW. (2011). Facilitation of hippocampal synaptogenesis and spatial memory by C-terminal truncated Nle1-angiotensin IV analogs. The Journal of Pharmacology and Experimental Therapeutics. 339:35-44. doi:10.1124/jpet.111.182220 · abstract only
  18. 18.Benoist CC, Kawas LH, Zhu M, Tyson KA, Stillmaker L, Appleyard SM, Wright JW, Wayman GA, Harding JW. (2014). The procognitive and synaptogenic effects of angiotensin IV-derived peptides are dependent on activation of the hepatocyte growth factor/c-met system. The Journal of Pharmacology and Experimental Therapeutics. 351:390-402. doi:10.1124/jpet.114.218735 · abstract only · retracted: Retracted April 2025: J Pharmacol Exp Ther 2025;392(4):103567, doi 10.1016/j.jpet.2025.103567, PMID 40312093. Preceded by a Notice of Concern in 2021, PMID 34551987. This was the source of the claim that the molecule binds hepatocyte growth factor with high affinity; it is cited here as the provenance of that claim and not as evidence for it.
  19. 19.Kawas LH, McCoy AT, Yamamoto BJ, Wright JW, Harding JW. (2012). Development of angiotensin IV analogs as hepatocyte growth factor/Met modifiers. The Journal of Pharmacology and Experimental Therapeutics. 340:539-548. doi:10.1124/jpet.111.188136 · abstract only · retracted: Retracted April 2025: retraction notice doi 10.1016/j.jpet.2025.103566. Preceded by a Notice of Concern in 2021, PMID 34551990. This is where the 6-AH antagonist family was characterised.
  20. 20.Kawas LH, Yamamoto BJ, Wright JW, Harding JW. (2011). Mimics of the dimerization domain of hepatocyte growth factor exhibit anti-Met and anticancer activity. The Journal of Pharmacology and Experimental Therapeutics. 339:509-518. doi:10.1124/jpet.111.185694 · abstract only · retracted: Retracted April 2025: retraction notice doi 10.1016/j.jpet.2025.103568. Preceded by a Notice of Concern in 2021, PMID 34551988, and by an erratum in 2014. Source of the hinge-mimic hypothesis.
  21. 21.Sun X, Deng Y, Fu X, Wang S, Duan R, Zhang Y. (2021). AngIV-Analog Dihexa Rescues Cognitive Impairment and Recovers Memory in the APP/PS1 Mouse via the PI3K/AKT Signaling Pathway. Brain Sciences. 11:1487. doi:10.3390/brainsci11111487 · full text read
  22. 22.Pan T, Wang N, Zhang J, Yang F, Chen Y, Zhuang Y, Xu Y, Fang J, You K, Lin X, Li Y, Li S, Liang K, Li YX, Gao Y. (2022). Efficiently generate functional hepatic cells from human pluripotent stem cells by complete small-molecule strategy. Stem Cell Research & Therapy. 13:159. doi:10.1186/s13287-022-02831-1 · full text read
  23. 23.Johnston JL, Reda SM, Setti SE, Taylor RW, Berthiaume AA, Walker WE, Wu W, Moebius HJ, Church KJ. (2023). Fosgonimeton, a Novel Positive Modulator of the HGF/MET System, Promotes Neurotrophic and Procognitive Effects in Models of Dementia. Neurotherapeutics. 20:431-451. doi:10.1007/s13311-022-01325-5 · full text read
  24. 24.National Center for Biotechnology Information. PubChem Compound Summary for CID 156596375, Fosgonimeton. PubChem. source · database record
  25. 25.Retraction Watch. (2021). Four papers by Athira CEO earn expressions of concern. Retraction Watch. source · full text read
  26. 26.Goldhill O. (2021). Athira chief executive resigns as company confirms she altered research. STAT. source · abstract only
  27. 27.Hua X, Church K, Walker W, L'Hostis P, Viardot G, Danjou P, Hendrix S, Moebius HJ. (2022). Safety, Tolerability, Pharmacokinetics, and Pharmacodynamics of the Positive Modulator of HGF/MET, Fosgonimeton, in Healthy Volunteers and Subjects with Alzheimer's Disease: Randomized, Placebo-Controlled, Double-Blind, Phase I Clinical Trial. Journal of Alzheimer's Disease. 86:1399-1413. doi:10.3233/JAD-215511 · abstract only
  28. 28.Martino KA, Nakhre A, Demarest RM, Devilbiss DM. (2025). Hepatocyte Growth Factor/MET Activator Rescues Working Memory Deficits after Repeated Mild Traumatic Brain Injury. Neurotrauma Reports. 6:991-998. doi:10.1177/2689288X251392851 · full text read
  29. 29.Wright JW, Harding JW. (2015). The Brain Hepatocyte Growth Factor/c-Met Receptor System: A New Target for the Treatment of Alzheimer's Disease. Journal of Alzheimer's Disease. 45:985-1000. doi:10.3233/JAD-142814 · abstract only
  30. 30.Kawas LH, Benoist CC, Harding JW, Wayman GA, Abu-Lail NI. (2013). Nanoscale mapping of the Met receptor on hippocampal neurons by AFM and confocal microscopy. Nanomedicine. 9:428-438. doi:10.1016/j.nano.2012.08.008 · abstract only
  31. 31.Niemann HH. (2013). Structural basis of MET receptor dimerization by the bacterial invasion protein InlB and the HGF/SF splice variant NK1. Biochimica et Biophysica Acta. 1834:2195-2204. doi:10.1016/j.bbapap.2012.10.012 · abstract only
  32. 32.Mpakali A, Saridakis E, Giastas P, Maben Z, Stern LJ, Larhed M, Hallberg M, Stratikos E. (2020). Structural Basis of Inhibition of Insulin-Regulated Aminopeptidase by a Macrocyclic Peptidic Inhibitor. ACS Medicinal Chemistry Letters. 11:1429-1434. doi:10.1021/acsmedchemlett.0c00172 · abstract only
  33. 33.Journal of Pharmacology and Experimental Therapeutics editors. (2021). Notice of Concern: McCoy AT, Benoist CC, Wright JW, Kawas LH, Bule-Ghogare JM, Zhu M, Appleyard SM, Wayman GA, and Harding JW (2013) Evaluation of Metabolically Stabilized Angiotensin IV Analogs as Procognitive/Antidementia Agents, J Pharmacol Exp Ther, 344: 141-154. The Journal of Pharmacology and Experimental Therapeutics. 378:313. doi:10.1124/jpet.112.199497concern · abstract only
  34. 34.Reda SM, Setti SE, Berthiaume AA, Wu W, Taylor RW, Johnston JL, Stein LR, Moebius HJ, Church KJ. (2024). Fosgonimeton attenuates amyloid-beta toxicity in preclinical models of Alzheimer's disease. Neurotherapeutics. 21:e00350. doi:10.1016/j.neurot.2024.e00350 · abstract only
  35. 35.Zhou H, Casas-Finet JR, Heath Coats R, Kaufman JD, Stahl SJ, Wingfield PT, Rubin JS, Bottaro DP, Byrd RA. (1999). Identification and dynamics of a heparin-binding site in hepatocyte growth factor. Biochemistry. 38:14793-14802. doi:10.1021/bi9908641 · abstract only
  36. 36.Mathapati S, Siller R, Impellizzeri AA, Lycke M, Vegheim K, Almaas R, Sullivan GJ. (2016). Small-Molecule-Directed Hepatocyte-Like Cell Differentiation of Human Pluripotent Stem Cells. Current Protocols in Stem Cell Biology. 38:1G.6.1-1G.6.18. doi:10.1002/cpsc.13 · abstract only
  37. 37.Yamamoto BJ, Elias PD, Masino JA, Hudson BD, McCoy AT, Anderson ZJ, Varnum MD, Sardinia MF, Wright JW, Harding JW. (2010). The angiotensin IV analog Nle-Tyr-Leu-psi-(CH2-NH2)3-4-His-Pro-Phe (norleual) can act as a hepatocyte growth factor/c-Met inhibitor. The Journal of Pharmacology and Experimental Therapeutics. 333:161-173. doi:10.1124/jpet.109.161711 · abstract only · corrected: Erratum, J Pharmacol Exp Ther 2015;352(2):379; the PubMed abstract carries a [Corrected] marker. Not retracted and not under a Notice of Concern.
  38. 38.Ho JK, Nation DA. (2018). Cognitive benefits of angiotensin IV and angiotensin-(1-7): A systematic review of experimental studies. Neuroscience and Biobehavioral Reviews. 92:209-225. doi:10.1016/j.neubiorev.2018.05.005 · abstract only
  39. 39.Wright JW, Kawas LH, Harding JW. (2015). The development of small molecule angiotensin IV analogs to treat Alzheimer's and Parkinson's diseases. Progress in Neurobiology. 125:26-46. doi:10.1016/j.pneurobio.2014.11.004 · abstract only
  40. 40.Ye S, Chai SY, Lew RA, Ascher DB, Morton CJ, Parker MW, Albiston AL. (2008). Identification of modulating residues defining the catalytic cleft of insulin-regulated aminopeptidase. Biochemistry and Cell Biology. 86:251-261. doi:10.1139/o08-037 · abstract only
  41. 41.Comoglio PM, Trusolino L, Boccaccio C. (2018). Known and novel roles of the MET oncogene in cancer: a coherent approach to targeted therapy. Nature Reviews Cancer. 18:341-358. doi:10.1038/s41568-018-0002-y · abstract only
  42. 42.Weiss JB, Phillips CJ, Malin EW, Gorantla VS, Harding JW, Salgar SK. (2021). Stem cell, Granulocyte-Colony Stimulating Factor and/or Dihexa to promote limb function recovery in a rat sciatic nerve damage-repair model: Experimental animal studies. Annals of Medicine and Surgery. 71:102917. doi:10.1016/j.amsu.2021.102917 · full text read
  43. 43.El-Habta R, Sloniecka M, Kingham PJ, Backman LJ. (2018). The adipose tissue stromal vascular fraction secretome enhances the proliferation but inhibits the differentiation of myoblasts. Stem Cell Research & Therapy. 9:352. doi:10.1186/s13287-018-1096-6 · abstract only
  44. 44.Church KJ, Vanderwerff BR, Riggers RR, Mateo-Victoriano B, Fagnan M, Harris PH, LeValley JC, Harding JW. (2018). Norleual, a hepatocyte growth factor and macrophage stimulating protein dual antagonist, increases pancreatic cancer sensitivity to gemcitabine. Anti-Cancer Drugs. 29:295-306. doi:10.1097/CAD.0000000000000598 · abstract only
  45. 45.Hu SK, Chen YM, Chao L. (2014). Phase segregation of polymerizable lipids to construct filters for separating lipid-membrane-embedded species. Biomicrofluidics. 8:052005. doi:10.1063/1.4895570 · abstract only
  46. 46.Bapst JP, Froidevaux S, Calame M, Tanner H, Eberle AN. (2007). Dimeric DOTA-alpha-melanocyte-stimulating hormone analogs: synthesis and in vivo characteristics of radiopeptides with high in vitro activity. Journal of Receptor and Signal Transduction Research. 27:383-409. doi:10.1080/10799890701723528 · abstract only
  47. 47.Kessel D, Conley M, Vicente MG, Reiners JJ. (2005). Studies on the subcellular localization of the porphycene CPO. Photochemistry and Photobiology. 81:569-572. doi:10.1562/2004-12-16-RA-403 · abstract only
  48. 48.Royal Society of Chemistry. (2026). ChemSpider record 57582587, Dihexa. ChemSpider. source · database record

Scope. This entry describes chemistry, structure and published laboratory findings. It is not guidance on use of any kind.

Sources read: 38 distinct works. Researched 2026-11-03. Detailed review 2026-11-02. How this library is built.

Your Cart

Your cart is empty

Browse our catalog to add research compounds.