GHK-Cu is the copper(II) complex of the tripeptide glycyl-L-histidyl-L-lysine. The peptide is only three residues long, but it presents a histidine at position two, which turns the first two residues and the imidazole side chain into a chelating site that holds a single copper ion through the peptide backbone rather than loosely associating with it. The material supplied for research is the isolated complex, not the free peptide: the copper is part of the molecule, and its presence is what gives the lyophilized powder its blue color.

The tripeptide was first reported as a constituent of human serum in 1973, in work that identified it as a small factor active on cultured liver cells [1]. Almost everything published since has treated it as a copper carrier or copper-dependent effector, which makes the coordination chemistry inseparable from the biology.

Identity and characterisation

Echo Peptides publishes the following identifiers for this compound, and this guide uses them without alteration:

Two points about that block are worth stating plainly, because they are the commonest sources of confusion in secondary sources.

First, the formula contains copper, so the mass must too. A great deal of published and commercial material quotes the mass of the copper-free tripeptide against the name "GHK-Cu"; those two numbers differ by roughly the atomic mass of copper, and a molarity calculated from the wrong one is wrong by about eighteen percent. If a stock concentration is being computed from a weighed mass, the number that matters is the one for the complex actually in the vial.

Second, the absence of a PubChem CID here is deliberate rather than an omission. PubChem carries this complex as a charged species — as a cation and as an anion — and neither record corresponds to the neutral complex described by the formula above. Linking to a database entry whose formula contradicts the one on the page would look like more rigor while providing less, so no CID is quoted.

Independent characterisation of the complex itself is thin but not absent. A 2020 study that produced GHK recombinantly in E. coli rather than by synthesis used UV-visible spectroscopy to establish a 1:1 peptide-to-copper stoichiometry, and X-ray absorption near-edge spectroscopy to confirm that the bound metal is in the Cu(II) oxidation state; the authors report that the recombinant peptide's spectroscopic behaviour matched that of GHK isolated from human plasma [10].

Mechanism as studied: the copper site

The best-characterised property of GHK is its affinity for Cu(II), and here there are real numbers. Isothermal titration calorimetry, run with glycine as a deliberately weak competing ligand so that the binding constant falls inside the measurable range, gives a 1:1 complex with a conditional dissociation constant of 7.0 ± 1.0 × 10⁻¹⁴ M at pH 7.4 [2]. The same study measured the tetrapeptide DAHK — the N-terminal copper site of human serum albumin — under identical conditions and obtained 2.6 ± 0.4 × 10⁻¹⁴ M, that is, a slightly tighter constant than GHK's [2]. This is a useful corrective. GHK is often described as a high-affinity copper carrier, and it is; it is not described accurately as a molecule that outcompetes the dominant copper-binding site of serum albumin, at least by that measurement.

The two sites are also structurally distinct. Albumin's is the Xaa-Zaa-His (ATCUN) arrangement, with histidine third; GHK's histidine is second. A coordination position on the copper remains available in the GHK complex, and it can be filled by a third ligand: Bossak-Ahmad and colleagues showed that cis-urocanic acid, which co-occurs with GHK and copper in human tissue, also coordinates Cu(II) and forms a ternary [GHK][Cu(II)][cis-urocanic acid] species, which they argue may be a physiologically relevant form of the chelate [3]. The practical consequence for bench work is that "GHK-Cu in buffer" is a speciation problem, not a single defined molecule, whenever other copper ligands are present.

How such complexes form and exchange their metal has only recently been examined directly, and not on GHK. Stopped-flow and microsecond freeze-hyperquenching work on glycyl-glycyl-histidine, described by its authors as the simplest model of this class of N-terminal copper site, resolved partially coordinated intermediates including a two-nitrogen species that persists on a ~100 ms timescale at pH 6.0 and can sustain the Cu(II)/Cu(I) couple [4]. Whether GHK behaves the same way is an extrapolation, not a result.

Mechanism as studied: cellular readouts

The cell-culture literature is older and concentrated in a small number of laboratories. The foundational observation is that GHK-Cu increases collagen synthesis in fibroblast cultures. Maquart and colleagues reported stimulation beginning between 10⁻¹² and 10⁻¹¹ M, maximal around 10⁻⁹ M, and independent of any change in cell number — that is, more collagen per cell rather than more cells [5]. The same paper noted that a GHK triplet occurs in the α2(I) chain of type I collagen, and proposed that proteolysis at a wound site could release the tripeptide locally; that is a hypothesis about origin, and the paper presents it as one.

Later work from the same group extended the readouts beyond synthesis to turnover. In dermal fibroblast cultures, GHK-Cu raised matrix metalloproteinase-2 levels in conditioned media along with MMP-2 mRNA, and simultaneously increased secretion of the tissue inhibitors TIMP-1 and TIMP-2 [6]. The authors' framing is worth keeping: the complex is described as an activator of matrix remodeling — both arms of it — rather than simply of matrix production.

That study also contains the single most awkward control in this literature. The MMP-2 effect was reproduced by copper ions alone and not by the tripeptide GHK alone [6]. Read strictly, in that assay the peptide was a copper delivery vehicle rather than the active species.

A separate line of work points the other way. Lane and colleagues found that proteolysis of SPARC, a matrix-binding protein, releases KGHK, GHK and longer peptides containing those sequences, and that the KGHK sequence accounted for most of the angiogenic activity of the parent region in endothelial cord-formation assays. Substituting the histidine reduced the effect. Critically, although the peptides bound Cu(II) with high affinity, prior incubation with Cu(II) was not required for the stimulatory activity, and a second Cu(II)-binding peptide from SPARC had no angiogenic effect at all — leading the authors to conclude the activity was sequence-specific and independent of bound copper [8]. Two well-conducted studies, opposite conclusions about which half of the molecule is doing the work, in different assays and different cell types. Nobody has reconciled them.

Redox and iron-handling readouts

"Antioxidant" appears constantly in descriptions of this compound, and the primary data supporting it are narrower than the word implies. Miller and colleagues examined GHK-Cu(II) in lipid peroxidation systems and found it inhibited peroxidation only when the iron source was ferritin, and did not show significant superoxide-dismutase-like or ceruloplasmin-like activity; their proposed mechanism is physical blockade of ferritin's iron-release channels rather than radical scavenging [9]. That is a specific, mechanistically interesting, and quite limited finding, and it is not the same claim as general antioxidant capacity.

Model systems it has actually been examined in

  • Cultured fibroblasts. Human and rat dermal fibroblast monolayers supply nearly all the mechanistic data — collagen synthesis [5], MMP-2 and TIMP-1/-2 [6], and decorin expression [7].
  • A rat wound-chamber model. Repeated local delivery of GHK-Cu into implanted chambers increased chamber dry weight, total protein, hydroxyproline and uronic acid content, shifted glycosaminoglycan composition toward chondroitin and dermatan sulfate, raised decorin mRNA and lowered biglycan mRNA [7]. This is a tissue-deposition model measured by biochemistry, not an outcome study.
  • Zebrafish larvae. Used as a whole-organism copper-handling model, where recombinant GHK altered the cardiac response to waterborne Cu(II) exposure while leaving several other cardiovascular parameters unchanged [10].
  • Cell-free coordination chemistry. Calorimetry, spectroscopy and potentiometry on the complex itself [2, 3].
  • Conjugates and materials. Recent work attaches GHK to hyaluronan and studies the copper complex of the conjugate in vitro, reporting expression and release of trophic and osteogenic factors and nuclear translocation of the copper chaperones CCS and Atox1 [11]. Note that findings there describe the conjugate, which is a different molecule.

What the evidence does not establish

A large fraction of what circulates about GHK-Cu traces back not to primary studies but to review articles from a single group. The most widely cited of them attributes a very wide range of actions across many organ systems to the molecule and states that it is capable of up- and down-regulating at least 4,000 human genes [12]. Those reviews are indexed and readable, and they are also secondary literature written by authors with a commercial association with the compound. Claims sourced to them should be read as claims, and traced to the primary work before being relied on.

Beyond that, the following are open rather than settled:

  • Whether the peptide or the copper is the active moiety. The two most direct tests disagree [6, 8].
  • Whether the reported potency range generalises. The picomolar-to-nanomolar window [5] comes from a specific fibroblast culture system, with cell type, serum content and buffer composition all uncontrolled across the wider literature.
  • What the complex is in a given medium. Ternary species form with common physiological ligands [3], and albumin's own site binds copper at least as tightly [2], so the species distribution in a serum-containing culture is not the species that was weighed out.
  • Exchange kinetics. Only characterised for a model peptide, not for GHK [4].
  • Anything about outcomes in humans or animals. This guide describes assays and model systems. It makes no statement about effects in any organism outside those systems, and none should be inferred from it.

Handling and stability

The literature supports a small number of specific handling points, and it is worth separating those from general peptide advice.

The most concrete comes from work motivated by the problem itself: a 2025 study developing GHK-hyaluronan conjugates states as its premise that GHK and GHK-Cu are vulnerable to oxidative stress, hydrolytic attack and environmental conditions, and that conjugation was pursued to preserve molecular integrity and delay degradation [11]. The complex is not an unusually robust molecule.

The second follows from the coordination chemistry. The affinity measurement above required glycine, a weak amino-acid ligand, as a competitor precisely because it competes measurably for the copper [2]. Any buffer or medium component that chelates copper — and many do — will shift the equilibrium. Complexing buffers and chelator-containing media are therefore an experimental variable, not a neutral background, and a complexed-copper compound should be prepared in a defined, documented buffer if results are to be comparable between runs.

The third is speciation: because a coordination site remains open, third ligands present in the system can form ternary complexes [3]. What is in the well may not be the 1:1 complex.

The storefront specifies refrigerated storage at 2–8 °C for the lyophilized material. Beyond that, standard practice applies — desiccated and light-protected storage of the solid, aliquoting after reconstitution to avoid repeated freeze-thaw, and recording reconstitution date, solvent and buffer alongside the lot and its certificate of analysis. The published certificate for this material was issued against a 100 mg vial, which is worth noting when the certificate is used as the identity record for a different vial size.

Open questions

Three questions would change how this compound is interpreted if they were answered. Does GHK-Cu act by delivering copper to a cellular acceptor, or does the intact complex engage a specific target? The competing results in fibroblasts and endothelial assays are consistent with either, and no receptor has been identified [6, 8]. Second, what fraction of the effects reported in serum-containing culture survives correction for copper speciation, given that albumin's site binds at least as tightly and ternary complexes form readily [2, 3]? Third, how much of the modern gene-expression narrative rests on primary measurement rather than on review-level synthesis [12]?

None of these are obscure questions. They are the ones a careful reading of the primary literature raises immediately, and the fact that they remain open after fifty years is the most honest summary of where this molecule stands.

For research use only. Not for human or veterinary use.