What is AHK-Cu? The copper peptide studied in hair follicles

What is AHK-Cu? It is the copper(II) complex of a synthetic tripeptide ^[4], L-alanyl-L-histidyl-L-lysine, written Ala-His-Lys or AHK ^[1], ^[3]. Its sequence differs from the better-known GHK, glycyl-L-histidyl-L-lysine, in the first residue only: alanine stands where GHK has glycine ^[1], ^[9]. A 2026 review of copper-binding peptides describes AHK-Cu as synthetic, with histidine-anchored Cu(II) coordination, and as resting on a thin, poorly characterised evidence base ^[4].

A tripeptide is a chain of three amino acids; a copper complex is the peptide with a Cu(II) ion held in its structure. Everything below is a description of published research, and the research is sparse. The sources used here are one peer-reviewed study of AHK-Cu itself ^[1], one US patent holding the only animal experiments among them ^[2], two 2026 reviews that mention the compound briefly ^[3], ^[4], and one trial registration that lists it ^[5]. Two copper-binding papers study neighbouring peptides rather than AHK ^[6], ^[7], a cell-culture study uses the same three residues without copper ^[8], and the remaining sources concern GHK-Cu ^[9], ^[10], ^[11].

• The peer-reviewed study is Pyo and colleagues, 2007: human hair follicles kept alive outside the body, and cultured dermal papilla cells ^[1].
• The animal work is a patent from ProCyte Corporation, a legal document with worked examples rather than a journal paper ^[2].
• The copper-binding papers measured GHK, DAHK and AHH; none of them measured AHK-Cu ^[6], ^[7].
• One 2026 review classes the evidence as experimental ^[3]; another calls it thin and poorly characterised ^[4].
• The registered study is a single-group, early phase 1 plan that lists AHK-Cu among five peptides ^[5].

The timeline below is short because the record is short. It also shows that AHK-Cu is not new: the patent was filed in 1994 and the one peer-reviewed study is from 2007, so the thinness of the evidence is not the thinness of a recent compound.

1. 1973

   Pickart isolates GHK-Cu as an activity in human albumin; GHK is present in human plasma.[9]

2. 1994

   ProCyte Corporation files the application for US Patent 5,538,945.[2]

3. 1996

   US Patent 5,538,945, assigned to ProCyte Corporation, is granted: it discloses peptide-copper complexes, with hair-growth experiments in mice and infant rats in which AHK:Cu is among the complexes tested.[2]

4. 2007

   Pyo and colleagues report that AHK-Cu stimulated the elongation of human hair follicles ex vivo and the proliferation of dermal papilla cells in vitro.[1]

5. 2026

   One 2026 review describes AHK-Cu as supported by experimental, ex vivo evidence; another describes it as resting on a thin, poorly characterised evidence base.[3], [4]

6. October 2026

   A single-group early phase 1 study that lists AHK-Cu among five peptides in one topical serum is first posted on the trial registry.[5]

One further paper uses the same three residues without copper. A 2018 cell-culture study tested a vitamin C-linker-conjugated Ala-His-Lys tripeptide on mouse myoblast cells, which is a different molecule from AHK-Cu and is not evidence about it ^[8].

Whether a peptide holds copper depends on which atoms can reach the ion. A 2017 inorganic chemistry paper describes peptides and proteins with the N-terminal motif NH2-Xxx-His, where Xxx is any residue and histidine sits second, as forming well-established Cu(II) complexes. In that arrangement Cu(II) is bound by three nitrogens from the peptide and an external ligand in the equatorial plane, called the 3N form ^[6]. The paper names Gly-His-Lys as the canonical peptide of the Xxx-His motif and Asp-Ala-His-Lys, from human serum albumin, as the canonical peptide of the Xxx-Zzz-His motif, in which Cu(II) is bound by four nitrogens from the peptide (the 4N form) ^[6].

AHK has histidine second, so by sequence it fits the Xxx-His pattern, with alanine as Xxx ^[6]. That is as far as the sources go. They contain no spectroscopy, crystal structure or binding measurement of AHK-Cu, so how AHK-Cu actually sits around the copper ion has not been shown in them. The nearest measurement is of a different peptide, Ala-His-His (AHH), which the 2017 paper found could form both the 3N type and a four-nitrogen 4N type depending on pH: the 3N form predominated at about pH 4 to 6.5 and the 4N form at about pH 6.5 to 10 ^[6]. AHH is a neighbour of AHK, not AHK, and it shows only that small changes in a sequence can change the way a peptide holds copper.

GHK is the best-characterised relative. A 2011 study solved the crystal structure of the Cu(II) complexes of GHK and DAHK and examined them in solution. In solution, [Cu(II)(GHK)] is a monomer with three nitrogen ligands, from the amino group, an amidyl nitrogen and the imidazole of histidine, while in the solid state the complex is a dimer whose fourth site is held by an oxygen from a carboxylate ^[7]. DAHK, by contrast, binds through its amino group, two amidyl nitrogens and one imidazole nitrogen, with a water molecule apical to the copper ^[7].

The same paper shows why one cannot assume that near-identical peptides behave alike. Copper exchange between DAHK peptides was very slow, unlike the fast exchange seen with GHK, and [Cu(II)(GHK)] could be reduced to Cu(I) at about -0.62 V against a silver chloride reference with release of the copper, whereas [Cu(II)(DAHK)] could not ^[7]. Two plasma peptides that both bind copper through histidine therefore differed in how they exchange and release copper ^[7].

The ProCyte patent describes the family of complexes it covers in plain chemical terms: dipeptides or tripeptides chelated to copper at a molar ratio from about 1:1 to 3:1, with histidine, arginine or a derivative in the second position from the amino terminus ^[2]. It also records that the solution of one prepared complex was a dark blue-purple with a characteristic absorption maximum at 563 to 580 nm ^[2]. That colour is an observation about a solution, not a structure.

The only animal experiments among the sources are in US Patent 5,538,945, filed by ProCyte Corporation in 1994 and granted in 1996 ^[2]. A patent is a legal document that describes an invention and gives worked examples; it is not a journal article. The excerpts of it used here report group means, some with a ± spread, and no statistical comparisons ^[2]. The experiments are worth describing because they are the only place AHK:Cu was tried in a living animal, and for the same reason they should be read as the patent's own account.

In Example 12, AHK:Cu was injected into the skin of telogen-phase C3H mice, 60 days old, the resting phase of the hair cycle, and was followed by hair growth at the injection sites ^[2]. The investigators counted the mice growing hair at the injection site and measured the diameter of the hair-growth region at day 21 ^[2]. AHK:Cu was tested at a molar ratio of 2:1 in groups of four animals ^[2].

A further example repeated the intradermal experiment in C3H mice, 60 days old and in the telogen phase of hair growth, with versions of AHK:Cu built from D-amino acids, the mirror-image forms of the natural L-amino acids ^[2]. The patent states that substituting D-amino acids for the corresponding L-amino acids did not affect hair-growth activity ^[2]. Its table, which reports an area of hair growth in units the excerpt does not state, shows lower means for the version with D-alanine in the first position than for the other versions ^[2].

The topical experiment compared AHK:Cu with AHF:Cu, a complex in which lysine is replaced by phenylalanine, in telogen-phase female C3H mice aged 60 to 65 days ^[2]. The point of the experiment, as the patent reads it, is lysine against phenylalanine. The response, called Percent Treatment Area, was read at day 20, day 27 and day 34 ^[2]. At one of the two strengths tested, AHK:Cu reached a mean of 1.29% on day 20, 23.07% on day 27 and 90.14% on day 34, while AHF:Cu reached 0.00%, 0.00% and 12.91% ^[2].

The patent reads this as showing that complexes with hydrophilic residues, such as the lysine of AHK:Cu, are more active by topical application than similar peptides with hydrophobic residues such as the phenylalanine of AHF:Cu, and it notes that the reverse holds for injection, where the hydrophilic complexes were less active ^[2]. That is the patent's interpretation of a single comparison in mice, not an established rule.

The last example used infant rats. Sprague-Dawley pups, 8 days old and kept in 5 litters of 10 to 11, were given one intradermal injection of AHK:Cu in sterile saline, or saline alone ^[2]. From the next day designated animals received 7 consecutive daily intraperitoneal injections of a chemotherapy drug, cytosine arabinoside (Ara-C), and on day 10 hair loss at the injection site was scored from 0 for normal to 4 for total loss of hair ^[2]. The scale runs through 1 for slight thinning, 2 for moderate thinning and 3 for sparse hair cover ^[2].

Ara-C caused significant hair loss by days 5 to 6 in most animals. Saline-only animals scored 0.00, animals given Ara-C alone scored 4.00, and every AHK:Cu group scored lower than that, between 3.25 and 1.11 ^[2]. The patent says the injection generally kept hair within a 0.25 cm radius of the site, and that sections showed normal-looking anagen follicles, the growing phase, at the site while follicles away from it were dystrophic and non-functional ^[2]. This was a localised effect in infant rats, reported in a patent, and is not evidence about people given chemotherapy.

The one peer-reviewed study of AHK-Cu itself was published in 2007 by Pyo and colleagues in Archives of Pharmacal Research ^[1]. Its title speaks of human hair growth in vitro, and its design has two parts. The first is an ex vivo model: human hair follicles are removed from skin and kept alive in culture, so that their lengthening can be measured outside the body. The second is a cell culture of dermal papilla cells, which the paper describes as specialised fibroblasts that are important in the morphogenesis and growth of hair follicles ^[1]. Neither part involves a person receiving anything; this is tissue and cells in a dish.

The abstract opens by recalling that a tripeptide-copper complex has been described as a growth factor for various differentiated cells, stimulating proliferation of dermal fibroblasts and raising production of vascular endothelial growth factor while lowering secretion of transforming growth factor-beta1 ^[1]. The authors then evaluated L-alanyl-L-histidyl-L-lysine-Cu2+ on human hair growth ex vivo and on cultured dermal papilla cells ^[1].

AHK-Cu at 10⁻¹² to 10⁻⁹ M, a range from picomolar to nanomolar concentrations, stimulated the elongation of human hair follicles ex vivo and the proliferation of dermal papilla cells in vitro ^[1]. The abstract does not give the lengths, the number of follicles or the donors, so the size of the effect cannot be stated from it ^[1]. A 2026 review tabulates the same ex vivo work as covering 10⁻¹³ M to 10⁻⁷ M; that differs from the abstract's range, and the abstract's is the one used here ^[3].

The authors also asked whether the cells were dying less often. Annexin V-fluorescein isothiocyanate and propidium iodide labelling with flow cytometry, a method that counts cells by stage of cell death, showed that 10⁻⁹ M AHK-Cu reduced the number of apoptotic dermal papilla cells, but the decrease was not statistically significant ^[1]. Alongside that count, they report that the ratio of Bcl-2 to Bax was elevated and the levels of the cleaved forms of caspase-3 and PARP were reduced by treatment with 10⁻⁹ M AHK-Cu, and they propose from this the preclusion of the apoptosis of dermal papilla cells ^[1].

The authors proposed that AHK-Cu promotes the growth of human hair follicles, and that the stimulatory effect may occur through stimulating the proliferation of dermal papilla cells and precluding their apoptosis ^[1]. The wording is a proposal. The one direct count of dying cells did not reach significance, and the supporting evidence is protein markers at a single concentration ^[1].

A 2026 review of short peptides in hair research summarises the study as having shown that AHK-Cu stimulated elongation of human hair follicles in ex vivo research, and that AHK-Cu promoted growth of dermal papilla cells in vitro ^[3]. The same review adds that AHK-Cu's specific signalling mechanism is less well characterised than that of GHK-Cu ^[3].

The two complexes are close in structure and far apart in how much has been written about them. GHK-Cu is the glycine peptide; AHK-Cu replaces that glycine with alanine ^[1], ^[9]. Both place histidine second, which is the position the Xxx-His copper-binding motif depends on ^[4], ^[6]. The 2026 review that compares them describes AHK-Cu as sharing the motif but resting on a thin, poorly characterised evidence base ^[4].

GHK-Cu's literature has a different texture. GHK is present in human plasma, saliva and urine ^[9]. Gene-expression profiling with the Broad Institute's Connectivity Map found that GHK changed the expression of 31.2% of human genes by 50% or more ^[10]. In people, a multicentre, randomised, evaluator-blinded, placebo-controlled trial of a GHK-copper gel in diabetic neuropathic ulcers was published in 1994 ^[11]. None of that transfers to AHK-Cu, which has not been put through the same experiments.

The comparison that would answer the obvious question, AHK-Cu and GHK-Cu in the same assay, does not appear in the sources. Pyo and colleagues studied AHK-Cu alone ^[1], the patent compared AHK:Cu with AHF:Cu and not with GHK:Cu in the excerpts used here ^[2], and the registration lists AHK-Cu and GHK-Cu among five peptides in one peptide serum intervention ^[5].

Taken together, the evidence is one cell-and-tissue study, one patent with animal experiments, and two reviews that each describe the compound briefly. The Biomedicines review lists AHK-Cu under experimental evidence, in the human hair follicle organ, with hair follicle elongation as the outcome and a proposed action of stimulating dermal papilla cell proliferation and preventing apoptosis ^[3]. The Pharmaceutics review calls the evidence base thin and poorly characterised and lists the compound's rationale as cosmetic angiogenesis, hair and skin claims ^[4].

The registered study is NCT07888595. It is interventional, early phase 1, single group, with no masking and an estimated enrolment of 48, and it measures transcriptomic biological age in treated against untreated skin; blood and skin samples are analysed, with IL6 and IL8 measured in serum ^[5]. It has a single group whose one arm covers systemic agents and/or a topical formulation. The topical part is a peptide serum applied to defined skin sites, with untreated skin on the same participant as the control, and the registration lists five peptides in that one intervention: GHK-Cu, AHK-Cu, Epitalon, PGP and CLV ^[5].

The status is enrolling by invitation, the record was first posted on 22 October 2026, the estimated start is 21 October 2026 and the estimated primary completion is 14 October 2028 ^[5]. The record is a study plan registered by its sponsor, a third party's intended research, and this guide describes it without suggesting any use. A registration describes a plan. It reports no findings, and because AHK-Cu is one of several peptides in the intervention, a result from it would not be a result for AHK-Cu alone ^[5].

No source in this guide records a marketing authorisation for AHK-Cu from the MHRA, the EMA, the FDA or Colombia's INVIMA. The research described here was done on isolated cells, cultured follicles and animals, and a registered plan that has not reported.

Because so little has been done, the list of what has not been measured is longer than the list of what has. Each item below is an absence in the sources used here, not a claim that the work does not exist anywhere.

• No results from any human study of AHK-Cu are among the sources; the one registered study has not reported and lists several peptides together ^[5].
• No peer-reviewed animal study of AHK-Cu is among them; the animal experiments are the patent's own ^[2].
• No structure, spectrum or binding measurement of AHK-Cu itself is among them; the copper-binding work covers GHK, DAHK and AHH ^[6], ^[7].
• No experiment places AHK-Cu and GHK-Cu side by side in the same assay ^[1], ^[2].
• The abstract of the ex vivo study gives no follicle lengths, follicle numbers or donors ^[1].
• Whether the cell-culture results in dermal papilla cells carry over to a hair follicle in a person is not addressed by any source here ^[1].

For a reader weighing how much to read into the name, the plain summary is that AHK-Cu is a copper-binding relative of GHK-Cu with one 2007 follicle study, one patent and a place in a recent registration, while GHK-Cu has a literature across gene expression, wounds, skin and a handful of human trials ^[1], ^[2], ^[5], ^[9], ^[10], ^[11]. The comparison above shows where the two overlap, and where AHK-Cu is simply unstudied.