GHK-Cu and AHK-Cu: A Side-by-Side Evidence Ledger

Verified against PubChem PUG-REST property and synonym records for CIDs 73587, 7408502, 9831891, 139035031 and 168431292, retrieved 17 August 2026. Fields marked “not verified” are ones we could not confirm against a primary chemical database — we leave them blank rather than guess.

One methyl group, two literatures

Both molecules are copper(II) complexes of a three-residue peptide built on a shared His-Lys core. GHK is glycyl-L-histidyl-L-lysine; AHK is L-alanyl-L-histidyl-L-lysine. Glycine carries a hydrogen side chain and alanine a methyl group, so AHK is heavier by exactly one CH2 unit. PubChem records confirm the arithmetic without rounding error: free GHK is C14H24N6O4 at 340.38 g/mol (CID 73587) and free AHK is C15H26N6O4 at 354.41 g/mol (CID 7408502), a difference of 14.03 g/mol.

That is the entire structural distance between the two compounds. It is not the distance between their evidence bases. A PubMed search on the string "alanyl-histidyl-lysine" returns fifteen records. Fourteen of them concern DAHK — the aspartyl-alanyl-histidyl-lysine tetrapeptide at the N-terminus of human serum albumin — or unrelated tryptic fragments of alpha-lactalbumin. Exactly one is a primary study of the Ala-His-Lys copper complex.

GHK returns something else entirely: controlled rodent and rabbit wound models from the 1999–2006 period, fibroblast culture work, a physicochemical preformulation study, a 2012 gene-expression paper from a four-institution consortium, and an interventional Phase 2 trial recruiting as of August 2026. Anyone treating the two peptides as interchangeable because their structures are nearly identical is importing one compound's evidence into the other's file.

The copper site they share, and the measurement only one of them has

Both peptides coordinate a single Cu(II) ion through a histidine-anchored site, frequently described as analogous to the high-affinity copper transport site at the N-terminus of human albumin. In AHK the alanine methyl group sits away from the coordinating imidazole, so on structural grounds the binding pocket itself is conserved between the two peptides. That is a chemical inference, not a measurement.

For GHK the measurement exists. Trapaidze et al. (2011) used isothermal titration calorimetry with glycine as a weaker competing ligand and reported a conditional Cu(II) dissociation constant of 7.0 ± 1.0 × 10⁻¹⁴ M for GHK at pH 7.4, alongside 2.6 ± 0.4 × 10⁻¹⁴ M for DAHK, and found that the entropic and enthalpic contributions differ between the two peptides. No equivalent number was located for AHK-Cu.

Handling chemistry is similarly one-sided. Badenhorst et al. (2016) reported GHK-Cu distribution coefficients of log D between −2.38 and −2.49 across pH 4.5 to 7.4, first-order degradation under basic and oxidative stress with lesser acidic susceptibility, stability in water and pH 4.5–7.4 buffers for at least two weeks at 60 °C, and three degradation products identified by HPLC–MS, one of which was free histidine. No published preformulation study of AHK-Cu was found.

What was measured with GHK-Cu

The animal wound work is the densest part of the file. Canapp et al. (2003) created 6 mm full-thickness wounds inside ischemic bipedicle skin flaps on 24 adult male Sprague-Dawley rats, randomized to topical tripeptide-copper complex, vehicle, or no treatment, and traced wound margins daily for 13 days. At day 13 the recorded decrease from initial wound area was 64.5% in the complex group, 45.6% in the vehicle group and 28.2% in the untreated group, with lower TNF-alpha, MMP-2 and MMP-9 concentrations in biopsied treated wounds at days 6, 10 and 13.

Cangul et al. (2006) ran a randomized comparison in 18 New Zealand rabbits divided into three groups, with one full-thickness wound on each side of the dorsal midline and 21 days of daily topical application. Mean unhealed wound area was smaller in the copper-complex group than the zinc oxide group at day 7 and smaller than untreated control at days 7, 14 and 21.

Mechanistic work sits underneath. Siméon et al. (1999) implanted wound chambers under the skin of Sprague-Dawley rats with serial 2 mg injections and reported that GHK-Cu increased pro-MMP-2 and activated MMP-2 at days 18 and 22, with MMP-9 expression persisting in treated chambers to day 22, while interstitial collagenase activity was unchanged. The follow-on culture study (2000) found that copper ions reproduced the MMP-2 effect and the peptide alone did not.

What was measured with AHK-Cu

Pyo et al. (2007) is the whole primary file. Human hair follicles were cultured ex vivo and dermal papilla cells in vitro, and AHK-Cu at 10⁻¹² to 10⁻⁹ M was reported to stimulate follicle elongation and dermal papilla cell proliferation. At 10⁻⁹ M the Bcl-2/Bax ratio was elevated and cleaved caspase-3 and PARP were reduced by Western blot. The paper is explicit about a limitation that secondary summaries usually drop: flow cytometry with annexin V–FITC and propidium iodide showed a reduction in apoptotic dermal papilla cells, but that reduction was not statistically significant.

Nothing else in the indexed literature isolates this molecule. There is no published copper-binding constant, no partition coefficient, no forced-degradation profile, no animal model and no registered clinical study. A ClinicalTrials.gov search on AHK-Cu and alanyl-histidyl-lysine as interventions returns zero records.

This matters for study design in a specific way. The alanine-for-glycine substitution is only informative as a variable if the baseline it is compared against is characterized. GHK-Cu is characterized. AHK-Cu is characterized in one assay system, by one group, in one paper, nineteen years ago.

The identifier problem

Identity for these compounds is not settled at the database level, and the confusion is not the reader's fault. PubChem CID 73587 is the free GHK tripeptide — C14H24N6O4, 340.38 g/mol, CAS 49557-75-7 — and yet its own synonym list carries "GHK-Cu" and "Copper Peptide" alongside "Prezatide" and "Tripeptide-1". A record for a metal-free peptide is therefore indexed under the name of its copper complex.

The complex itself appears under at least three distinct entries with three different molecular weights: CID 139035031 as a 1:1 anionic complex at 400.90 g/mol, CID 9831891 as bis-prezatide copper (C28H46CuN12O8, 742.3 g/mol, CAS 130120-56-8), and commercial listings almost universally under CAS 89030-95-5, labelled Copper Tripeptide-1, which does not correspond to any of those PubChem CIDs under that registry number.

AHK-Cu is cleaner but thinner: CID 168431292, C15H24ClCuN6O4−, 451.39 g/mol, CAS 682809-81-0, listed as the monohydrochloride complex. Any quantitative comparison between the two compounds that does not first fix which salt form and stoichiometry each number refers to is comparing molar amounts that differ by up to a factor of two.

What a side-by-side comparison cannot settle

No study has compared GHK-Cu and AHK-Cu against each other. Not in the same animal model, not in the same cell line, not in the same assay, not by the same group. Every apparent contrast between them in circulation is assembled across papers separated by different species, different endpoints, different concentration ranges and, in the wound-model case, different decades.

The endpoints do not overlap either. The GHK-Cu file is built on cutaneous wound closure, matrix metalloproteinase expression and dermal-matrix components. The AHK-Cu file is built on hair-follicle elongation and dermal papilla cell proliferation. These are not two answers to one question; they are one answer each to two different questions.

The human record for GHK-Cu is also thinner than the volume of preclinical work suggests. NCT07437586 — a Phase 2, randomized, double-blind, vehicle-controlled split-wound study of a topical GHK-Cu gel in 60 healthy adults aged 18 to 55, sponsored by Hudson Biotech and running at a single site — began on 2 February 2026 with primary completion scheduled for February 2027. It is recruiting. It has posted no outcome data. It is the only registered interventional human study of either compound.

What is not known

Neither compound has a completed human efficacy study. The single registered interventional trial for either — NCT07437586, a 60-participant Phase 2 split-wound study of topical GHK-Cu gel — began in February 2026, is recruiting at one site, and has posted no outcome data; its scheduled primary completion is February 2027. No study has compared GHK-Cu and AHK-Cu head-to-head under any conditions, so the apparent division of labour between them (wound and dermal-matrix endpoints for one, hair-follicle endpoints for the other) reflects which questions each was asked, not a demonstrated difference in behaviour. For AHK-Cu specifically, the gap is close to total: no copper-binding constant, no partition coefficient, no stability profile, no animal model, no dose-ranging, no comparison against GHK-Cu, and no data outside a single 2007 report in human hair follicles ex vivo and cultured dermal papilla cells. On the GHK-Cu side, the wound-model work used topical or intra-chamber routes in rats and rabbits and does not speak to any other route; the fibroblast work used nanomolar in-medium exposures that map to nothing in an intact organism; and populations were uniformly young healthy animals, with no aged, diabetic, immunosuppressed or pediatric models represented. Long-term exposure has not been characterized for either compound in any species. Copper handling itself — whether repeated delivery of a chelated copper load perturbs systemic copper homeostasis — is not addressed by any study in this file.

Questions