Dihexa 10 mg

Compound profile

Dihexa: compound profile, literature landscape and handling notes

Where Dihexa came from

Dihexa came out of a long-running line of work at Washington State University on the angiotensin IV system, associated principally with the Harding laboratory and its collaborators. The starting point was not drug design at all. It was the observation, accumulated across the 1990s, that angiotensin IV, a hexapeptide fragment generated by further proteolysis of angiotensin II, had activities in brain tissue that could not be explained by the classical angiotensin AT1 and AT2 receptors. That implied a distinct binding site, which the field named AT4, and a substantial body of work went into characterizing what bound there and what happened when it did.

The problem with angiotensin IV as a research tool was practical rather than conceptual. It is a short unmodified peptide with free termini, which means aminopeptidases and carboxypeptidases dismantle it quickly, and it is polar enough that penetration into the central nervous system from a peripheral compartment is poor. Both of those properties are fatal for a compound intended to interrogate a brain system in an intact animal.

The medicinal chemistry that followed was therefore aimed at two specific liabilities, and this is the key to reading the structure. Successive rounds of work truncated the hexapeptide toward the residues that appeared to carry the activity, replaced the exposed termini with groups that peptidases do not recognize, and deliberately raised lipophilicity to improve the chance of crossing the blood-brain barrier. Intermediate compounds in that series, including the norleucine-substituted analog usually written as Nle1-angiotensin IV and the later compound named Norleual, appear throughout the literature and are frequently and wrongly treated as interchangeable with Dihexa.

The name itself encodes the chemistry rather than a target: it points at the two hexyl-length elements in the molecule. That naming convention is worth noting because it means the name tells you nothing about mechanism, which is unusual in this catalog and contributes to how often the compound is misclassified in vendor listings and in secondary summaries.

Reading the structure of Dihexa

The exact chemical name, formula and mass for this listing are printed in the specification table further down this page, and for this compound in particular they are worth actually reading rather than skimming, because the name is a chemical description and not a sequence. What follows is an interpretation of the design motifs.

Start with what was removed. Angiotensin IV is a six-residue peptide, running valine-tyrosine-isoleucine-histidine-proline-phenylalanine. Dihexa retains only a small fragment of that, essentially the tyrosine and isoleucine pair from positions two and three, extended by a non-proteinogenic amino acid spacer. Counted generously that is a tripeptide core, and only two of those units carry over from the parent. Describing Dihexa as a hexapeptide, which happens often, is simply wrong, and it leads people to expect peptide behavior that the molecule does not have.

Now the additions, which are where the molecule actually lives. The N-terminus is not a free amine; it carries an acyl cap, a six-carbon chain in the most commonly cited structure. The C-terminus is not a free carboxylate; it is capped as an amide. Each of those two changes does two things at once. First, they eliminate the recognition elements that exopeptidases require, since an aminopeptidase needs a free alpha-amino group and a carboxypeptidase needs a free carboxyl terminus, so a molecule capped at both ends is inert to the most common route by which a short peptide disappears. Second, and just as importantly, both caps are hydrophobic, so the calculated partition coefficient of the finished molecule sits in the range you associate with orally described small molecules rather than with peptides.

That second consequence is the one that drives everything practical about handling the material. A short peptide of comparable mass is usually a salt of a polar zwitterion, freely soluble in water, poorly retained on reversed phase, and prone to nothing worse than adsorption and deamidation. Dihexa is a capped, largely uncharged, aromatic, lipophilic molecule whose only ionizable groups of consequence are the tyrosine phenol and the amide backbone that does not ionize at all under normal conditions. It is best reasoned about, stored and analyzed as a small molecule that happens to contain amide bonds, and the failures researchers report with it are almost always failures of applying peptide intuition to a molecule that does not follow peptide rules.

The target and the pathway in more detail

There are two distinct target stories attached to Dihexa in the literature, they are not the same claim, and keeping them separate is the single most useful thing a researcher can do when reading about this compound.

The first is the inherited one. The parent compound angiotensin IV binds the site historically called AT4, and that site was subsequently identified as insulin-regulated aminopeptidase, an enzyme of the M1 aminopeptidase family also known by the abbreviation IRAP. Angiotensin IV behaves at that site as an inhibitor of its catalytic activity rather than as a classical agonist of a signaling receptor, which already complicates the language of receptor pharmacology. Dihexa inherited its lineage from this system, but inheritance is not mechanism, and there is no requirement that a heavily truncated and doubly capped derivative retain the binding mode of its parent.

The second story, and the one on which essentially all current interest rests, is hepatocyte growth factor and its receptor c-Met. c-Met is a receptor tyrosine kinase, a single-pass transmembrane protein that dimerizes on ligand binding, trans-autophosphorylates on its intracellular tyrosines, and recruits adaptors that feed the phosphoinositide 3-kinase and mitogen-activated protein kinase cascades. Its physiological ligand, hepatocyte growth factor, is a large multidomain protein, and productive signaling depends on HGF dimerizing on the receptor rather than simply occupying it.

The mechanism reported for Dihexa is that it acts as a potentiator of this interaction, not as a ligand that activates the receptor by itself. The published account is that it associates with hepatocyte growth factor and facilitates the dimerization step, so that a given amount of HGF produces more c-Met activation than it otherwise would.

That mechanistic shape has a direct experimental consequence, and it is the question worth asking before designing anything with this material. A potentiator has no readout in the absence of its ligand. If an experiment reports an effect of Dihexa, the first thing to establish is what the baseline HGF concentration was, whether it was added deliberately or contributed by serum or by the cells themselves, and whether a no-HGF arm was included. An apparent concentration-response curve for a potentiator is a curve conditioned on that baseline, and it moves when the baseline moves.

What the published literature on Dihexa actually measures

The Dihexa literature is small enough that a careful reader can work through most of the primary reports, which is not true of most compounds in this catalog, and it separates cleanly into three kinds of measurement.

The first is cell-level morphology. The most-cited assays are performed on dissociated hippocampal neuron cultures, where the endpoint is dendritic spine density, spine morphology by class, or the number and apposition of pre- and postsynaptic marker puncta scored by immunofluorescence. These are counting assays on images. They are informative about synaptic structure and they are also intrinsically noisy, sensitive to culture age and plating density, and dependent on how the scorer defines a spine, which is why blinding and clearly stated counting rules matter more in this subfield than in most.

The second is receptor-proximal biochemistry, and this is the part that supports the c-Met claim directly. The relevant readouts are phosphorylation of c-Met and of downstream kinases by immunoblot, binding and dimerization measurements involving hepatocyte growth factor, and functional c-Met bioassays such as the classic epithelial scatter assay, in which HGF causes tightly packed epithelial colonies to disperse. The important design feature of this group of experiments is the use of c-Met inhibition or knockdown as a dependency test: the argument is not merely that Dihexa produces an effect, it is that the effect disappears when c-Met signaling is blocked.

The third is rodent behavioral work, principally spatial-learning tasks in animals in which learning has been impaired pharmacologically or by aging. These are whole-animal experiments with modest group sizes, and they measure an animal-level endpoint that sits many steps downstream of any molecular claim. Anyone citing this part of the record should be clear about which impairment model was used, because a scopolamine-type pharmacological model and an aged-animal model are not the same experiment and do not support the same inference. The endpoints themselves are latency and path measures in a maze, which are sensitive to swimming ability, motivation and stress as well as to anything the compound does.

What is largely absent is worth naming as well. There is very little independent structural work on the molecule, little published pharmacokinetic characterization by modern standards, and no substantial peer-reviewed human clinical literature. Statements that circulate about central nervous system exposure trace back to a small number of rodent reports.

Where the Dihexa literature is thin or frequently misread

This is the section that matters most for Dihexa, because the gap between how confidently the compound is described in secondary sources and how much primary evidence exists behind it is wider than for anything else in this catalog.

The most basic issue is concentration of evidence. A large fraction of the primary literature originates from one research group and its immediate collaborators, using overlapping assay systems and overlapping reagents. That is not a criticism of the work; it is the normal early history of a compound. It does mean that independent replication in unrelated laboratories, which is the thing that actually converts a finding into a fact, is thin, and it means that a systematic artifact in a shared method would propagate through the whole record without contradiction.

The second is the comparative potency claim. Dihexa is routinely described in circulated summaries as being some very large multiple more potent than a named neurotrophic protein. That comparison originates in specific in vitro assays with specific endpoints, run under specific conditions, and it is a statement about the concentration required to move that particular readout, not a general statement about the molecule. Detached from its assay, the comparison is close to meaningless, and it is almost always quoted detached from its assay.

The third is the lineage discontinuity already noted. The angiotensin IV and IRAP literature and the c-Met literature are frequently blended into a single narrative in which Dihexa is described as an AT4 compound that works through c-Met. Those are two claims from two eras of the work, and the evidence for each has to be weighed separately.

The fourth is rarely mentioned and should be. c-Met is a proto-oncogene product. Amplification and aberrant activation of the HGF and c-Met axis is an established driver in several tumor types, and a substantial part of the pharmaceutical industry is engaged in inhibiting it. A compound whose reported mechanism is potentiation of that same axis carries an obvious question about proliferative effects that the enthusiastic secondary coverage of Dihexa almost never raises. For laboratory work this is a reason to include proliferation and viability readouts alongside the morphological ones, not to assume they are unchanged.

How Dihexa behaves in solution

Dihexa does not behave like the other lyophilized materials in this catalog in solution, and the difference is not marginal. The two lipophilic caps that give the molecule its metabolic stability also make it poorly soluble in water. This is the single most important handling fact about it, and it is the one most often discovered the hard way.

In practice this means an organic co-solvent is normally involved. Dimethyl sulfoxide is the usual choice for a concentrated stock, with ethanol and other water-miscible organics also reported. That introduces a set of problems a peptide worker does not usually have to think about. Dimethyl sulfoxide is strongly hygroscopic and picks up atmospheric water on every opening of the vial, which lowers its solvent power for lipophilic solutes over time; a stock that dissolved cleanly in month one can throw material out of solution in month six with no change to the solute at all. It also freezes near room temperature, so a stock stored cold has to be thawed completely and mixed before any aliquot is taken, or the concentration drawn will not be the concentration calculated.

The second problem is dilution. Taking a concentrated organic stock into an aqueous buffer is a solvent exchange, and a lipophilic solute can come out of solution during that step even when the final nominal concentration looks modest. The precipitate is often fine, sometimes transient, and frequently invisible by eye. A clear solution is not evidence of a dissolved solute here, which is exactly the assumption peptide handling trains people to make. Dilution into a large volume of vigorously mixed warm buffer behaves differently from dropwise addition into a static cold one, and the two can give different delivered concentrations from the same stock.

Third, lipophilic molecules adsorb. Loss to polypropylene tubes, pipette tips, and particularly to filter membranes during sterile filtration can be substantial, and unlike protein adsorption it is not prevented by adding a carrier protein, since serum albumin will happily bind a lipophilic small molecule and carry it out of the aqueous phase in a different way. Every unnecessary transfer step is an opportunity for loss.

Finally, the tyrosine phenol is the reactive point of the molecule. It is oxidizable, and phenols in general are light-sensitive, so amber vessels, minimized headspace and avoidance of repeated freeze-thaw cycling of aqueous working solutions all apply.

Analytical notes specific to Dihexa

Analytically, the useful instruction for Dihexa is to stop treating it as a peptide and start treating it as a small molecule. Almost every difficulty reported with it follows from a peptide method being applied to something that is not one.

Reversed-phase behavior is the clearest example. A short polar peptide elutes early and is comfortably resolved by a shallow acetonitrile gradient running from a low starting percentage. Dihexa, carrying two hydrophobic caps and an aromatic ring, is strongly retained on C18 and elutes very late by comparison with peptides of similar mass. A generic peptide gradient will either not elute it inside the run at all or will push it out as a distorted late peak that carries over into the following run. The method needs a higher organic proportion, and it is worth confirming that the column is fully re-equilibrated between runs, because a late-eluting lipophilic solute is exactly the kind of analyte that produces ghost peaks in subsequent blanks.

Sample diluent matters more than usual for the same reason. Loading a concentrated organic stock directly onto a column equilibrated at low organic content produces solvent-strength mismatch, which shows up as peak fronting, splitting or shoulders that have nothing to do with sample purity. Matching the sample solvent to the starting mobile phase as far as solubility permits is the fix, and that is a constraint rather than a free choice here, because the compound will precipitate if diluted too far into aqueous.

Detection is comparatively easy, which is a small mercy. The aromatic content gives a genuine ultraviolet chromophore near 275 to 280 nanometers, so the compound can be seen without relying on end-absorbance at around 214 nanometers where every solvent impurity also absorbs. Area percent at the higher wavelength is therefore selective for aromatic species, which is useful for identity work but means that non-aromatic process impurities will be under-represented; a second wavelength, or a universal detector such as charged aerosol or evaporative light scattering, gives a fuller picture.

Mass spectrometry is likewise a small-molecule exercise. The molecule is light enough that electrospray produces a dominant singly protonated ion rather than the multiply charged envelope a peptide gives, so no deconvolution is involved and the observed value should be compared against a monoisotopic calculated mass rather than an average one. Peptide-mapping and sequencing workflows are simply the wrong tool: the molecule is too short and too capped to fragment informatively in the way those methods assume.

Compounds researchers confuse with Dihexa

Questions specific to Dihexa