GHK-CuvsMatrixyl
GHK-Cu has fifty years of research behind it, most of it in cell culture and wound healing rather than on wrinkles. Matrixyl has far less research, but what it has includes real randomized human skin trials, paid for by the company that sells it. And Matrixyl 3000, the version in most products, is built around the same GHK tripeptide with a fat tail instead of copper.
- Skin and collagen
- Topical
- One is a cosmetic ingredient, one is not
The most popular version of Matrixyl already contains GHK.
One has fifty years of research and a delivery problem. The other has two real human trials and a manufacturer behind them.
| Attribute | GHK-Cu | Matrixyl |
|---|---|---|
| What it is | A natural human tripeptide bound to copper | A designed cosmetic ingredient |
| Length | 3 amino acids | 5 amino acids plus a fat tail |
| Found in your body | Yes, first isolated from blood plasma in 1973 [3] | The peptide part is a collagen fragment |
| Largest human trial | 71 women, 12 weeks, cited inconsistently [4] | 93 women, 12 weeks, split face, placebo controlled [1] |
| Who funded it | Various | Procter and Gamble [1][2] |
| Regulatory | Cosmetic use allowed. Injectable use is not [7] | Cosmetic ingredient |
"Matrixyl" covers at least three different products. See the structure section below.
GHK-Cu
A three amino acid chain your own body makes, bound to a copper atom. Loren Pickart first isolated it from human blood plasma in 1973, and levels fall as you age. In plain terms: a natural signal that tells skin cells to rebuild, sold as a topical serum and, in some places, as an injection.
Matrixyl
A five amino acid piece of collagen with a fatty acid tail bolted onto it. The peptide part is a fragment your skin releases when collagen breaks down. In plain terms: it fakes the signal that collagen has been damaged, so fibroblasts start rebuilding. The fat tail is there to get it through the skin.
A good deal of this is GHK versus GHK.
"Matrixyl" is at least three products, all trade names from Sederma, now part of Croda. The original is palmitoyl pentapeptide-4, also written Pal-KTTKS. Matrixyl 3000 is two peptides together, and it is what sits in most products. Matrixyl synthe'6 is a different molecule again. Nearly all the human trial evidence people cite was run on the original.
Same core, different attachment. GHK-Cu carries copper. Palmitoyl tripeptide-1, the main active in Matrixyl 3000, carries a fatty acid. The copper and the fat tail do different jobs.
The trials and the products do not match. The human evidence is on Pal-KTTKS. Most shelves carry Matrixyl 3000.
Check the ingredient list, not the brand. The INCI name tells you which molecule you have. The marketing name does not.
One has more research. The other has more relevant research.
Isolated from blood plasma
Loren Pickart isolates the tripeptide from human blood plasma and reports that it prolongs liver cell survival. This is where the fifty year research trail starts. Decades of wound healing, collagen and gene expression work follow, almost all of it in cell culture, in animals, or in wounds rather than on intact aging skin.
Nature New Biology1973;243(124):85-7 · Peer reviewed
- peer reviewed
- not a skin trial
Abdulghani pilot
Creams applied to the thigh skin of 20 women for 12 weeks. Collagen production improved in 70 percent of those on GHK-Cu, against 50 percent on vitamin C and 40 percent on retinoic acid.
Disease Management and Clinical Outcomes1998;1:136-141 · Minor journal
- 20 subjects
- low tier journal
- not a face study
Leyden facial cream study
71 women with photoaging, 12 weeks, reporting better skin density and thickness and reduced wrinkle depth. It remains the most cited human evidence for GHK-Cu, and the citation for it is inconsistent across the literature.
Cited as an AAD presentation, as Dermatol Surg 2005, and as J Cosmet Dermatol 2002Citation unresolved · See the section below
- citation disputed
- source not verifiable
Robinson trial
A 12 week trial in 93 women aged 35 to 55. Double blind, placebo controlled, split face, randomized left to right. A moisturizer with 3 parts per million Pal-KTTKS against the same moisturizer without it. The treated side showed significant reduction in wrinkles and fine lines, measured both by image analysis and by expert graders.
International Journal of Cosmetic Science2005;27(3):155-60 · PMID 18492182 · Peer reviewed
- placebo controlled
- randomized, split face
- manufacturer funded
Osborne split face study
An 8 week split face study in women aged 35 to 65, with about 60 people per product, also run at Procter and Gamble.
Journal of the American Academy of Dermatology2005 · Peer reviewed · Volume and pages not confirmed
- split face
- manufacturer funded
- citation incomplete
Watson randomized controlled trial
A double blind randomized controlled trial from Manchester, reporting improvements in fine wrinkles, pigmentation and roughness in photoaged skin. The best journal in the GHK-Cu file, and it tested a finished cosmetic product rather than the ingredient on its own.
British Journal of Dermatology2009;161(2):419-426 · PMID 19438432 · Peer reviewed
- randomized, double blind
- strong journal
- multi-ingredient product
Badenhorst head to head
The only direct comparison between these two, testing GHK-Cu in nano-lipid carriers against a control serum and against Matrixyl 3000. The GHK-Cu was delivered in a special carrier rather than an ordinary serum, so it is partly a test of the carrier.
Journal of Aging Science2016;4(3):166 · Low tier journal
- direct comparison
- low tier journal
- carrier confound
Two answers at the FDA
The FDA places injectable GHK-Cu in Category 2 of its 503A bulk substances list. Non-injectable GHK-Cu goes to Category 1, which permits compounding. Same molecule, two different answers depending on how it goes in. [7]
US Food and Drug Administration503A category updates, September 29 2023 · Regulatory document
Off Category 2, nothing resolved
Injectable GHK-Cu comes off Category 2 in a batch of twelve peptides, because the nominations were withdrawn rather than because anything was settled. It is scheduled for advisory committee review by February 2027. [8]
US Food and Drug AdministrationCategory 2 removals, April 2026 · Regulatory document
GHK-Cu has more research. Matrixyl has more of the research that answers the question people are actually asking. Fifty years of work on GHK-Cu tells you a great deal about what the molecule does to cells. Much less of it tells you what happens to a face over twelve weeks.
Everything that differs
| Attribute | GHK-Cu | Matrixyl (Pal-KTTKS) |
|---|---|---|
| Full name | Glycyl-L-histidyl-L-lysine copper complex | Palmitoyl pentapeptide-4 |
| Size | About 340 daltons for the peptide, about 404 with copper | About 802 daltons |
| Origin | Made naturally in the human body | Designed, based on a collagen fragment |
| Discovered | 1973, isolated from blood plasma [3] | Late 1990s, developed by Sederma |
| Proposed mechanism | Signals fibroblasts, regulates the enzymes that break down skin structure, carries copper into tissue | Mimics a collagen breakdown fragment, so fibroblasts rebuild |
| Gets through skin how | No delivery aid. It is charged and water loving, which skin resists | A fatty acid tail built in specifically to solve this |
| Human skin trials | Largest is 71 women, 12 weeks, cited inconsistently [4] | Two split face randomized trials, 2005, peer reviewed [1][2] |
| Who paid | Various, including independent groups | Procter and Gamble |
| Concentration in trials | Commonly 0.05 to 1 percent in topical studies | 3 parts per million, which is 0.0003 percent [1] |
| Other research | Fifty years across wound healing, gene expression, cell culture | Mostly cell culture plus the two trials |
| Regulatory | Allowed in cosmetics. Injectable use not permitted [7] | Cosmetic ingredient |
How each one works
GHK-Cu
Copper is needed for the enzymes that cross-link collagen and elastin, the proteins that keep skin firm. GHK binds copper tightly and appears to act as a carrier, delivering it where it is needed.
Beyond that, GHK-Cu is reported to signal fibroblasts to produce collagen, and to regulate the enzymes that break skin structure down, so remodeling stays balanced rather than tipping toward loss.
You will also see the claim that GHK affects about 31 percent of human genes, resetting age-shifted patterns toward younger ones. That figure comes from a reanalysis of a public gene expression database, published in reviews authored by Loren Pickart, who discovered GHK and has commercial interests in it. [10] The underlying database is real. The framing comes from an interested party, and it is a long way from a measured change in anyone's skin.
Matrixyl
When collagen breaks down, it releases fragments. Your skin reads those fragments as damage and tells fibroblasts to build more collagen. Matrixyl is one of those fragments, made synthetically.
So it does not add anything to the skin. It sends a false alarm, and the repair response is the point.
The fatty acid tail is doing real work here. Skin is built to keep water loving molecules out. Attaching a fat tail lets the peptide slip through the outer layer and reach the fibroblasts in the dermis, which is where it needs to be.
And the version most people buy, Matrixyl 3000, is built on palmitoyl tripeptide-1, which is the GHK tripeptide with that same tail.
What separates them
Four gaps, and one of them is the whole comparison
One head to head trial exists, and it is not strong. Badenhorst and colleagues compared GHK-Cu delivered in nano-lipid carriers against a control serum and against Matrixyl 3000, reporting reductions in wrinkle volume and depth. [9] Two things about it. It tested GHK-Cu in a special delivery system rather than an ordinary serum, so it is partly a test of the carrier. And it was published in a low tier journal, so it has not had the scrutiny the Robinson trial received. It is the only direct comparison that exists, and it is thin.
The version most people buy has no independent trial. Matrixyl 3000 has not been through a published placebo controlled human study by anyone other than its manufacturer.
The delivery question for GHK-Cu is unsettled. You will read that GHK-Cu is too large to penetrate skin, citing a rule that molecules over 500 daltons cannot get through. That reasoning is wrong on its own terms. GHK is about 340 daltons and the copper complex about 404, both comfortably under 500. The real obstacle is that the molecule is charged and water loving, and skin is built to repel exactly that. Whether enough of it reaches the dermis from a serum, and at what concentration, has not been settled.
Matrixyl's trials tested one product, not an ingredient. Both 2005 trials tested a specific moisturizer at 3 parts per million. A different product, at a different concentration, in a different base, is not the thing that was tested. And most products today contain Matrixyl 3000, which was not the version in those trials at all.
Both are well tolerated. The buying problem is the label.
Topical use of either is low risk. Injecting GHK-Cu is a different matter, and it is not permitted for compounding in the US.
Reported effects
Both are considered well tolerated on skin.
- Pal-KTTKS reported as well tolerated across the 2005 trials [1]matrixyl
- GHK-Cu can cause irritation or redness in some peopleghk-cu
- Copper peptides can leave a temporary blue or green tint on skincosmetic
- Neither has a documented pattern of serious topical reactionsboth
Nobody has studied this
Nobody has run these studies. That is different from a clean result.
- The two compared head to head in an ordinary serum, rather than a special carrierno trial
- Either one used continuously for yearsno data
- Matrixyl 3000, the version most people buy, in a placebo controlled human trialno trial
- Injectable GHK-Cu for cosmetic purposes, in a controlled human trialno data
What is on the label
This is where money gets wasted.
- The brand name does not tell you the molecule. Check the INCI namebuying
- Matrixyl, Matrixyl 3000 and Matrixyl synthe'6 are three different thingsnaming
- Very few products state the concentration, and the trials used 3 parts per milliondose
- Injectable GHK-Cu sold as a research chemical carries the usual identity problems [11][12]supply
*Where this page says nothing is established, it means nobody has studied it. It does not mean a compound is safe.
The practical advice here is short. Read the ingredient list rather than the front of the bottle. If a product says Matrixyl, find out which one. If it says palmitoyl tripeptide-1, you are buying GHK with a fat tail, whatever else is on the label. Our guide on reading a certificate of analysis covers the injectable side of that problem.
Neither GHK-Cu nor any Matrixyl peptide is named on the 2026 WADA Prohibited List. [13] For topical cosmetic use this is not a realistic concern. For injectable GHK-Cu the picture is less certain, since the S0 catch-all covers substances with no approval from any government health authority, and non-injectable GHK-Cu is permitted for compounding in the US, which complicates that reading. If you compete under a testing body, ask that body directly.
The most quoted GHK-Cu study cannot be pinned down
The 71 women, 12 weeks figure is the number the copper peptide market runs on. It is cited three incompatible ways across the literature, and two of those citations place different papers at identical journal, volume, issue and page numbers.
Some sources describe the study as an American Academy of Dermatology presentation from 2002. Others cite a 2005 paper in Dermatologic Surgery, at volume 31, issue 7, pages 809 to 816. A third citation puts it in the Journal of Cosmetic Dermatology, 2002, volume 1, issue 3, pages 197 to 204.
We are not saying the study is bad. We are saying we could not establish what it is. The most quoted number in the copper peptide space rests on a citation nobody seems to have checked.
Matrixyl has the opposite problem. It has two trials that measured faces, with placebo groups and blinded assessment, and they are easy to look up. The catch is who paid for them. Both were run by Procter and Gamble, which sold the moisturizer. That does not make them wrong. It does mean nobody independent has repeated them at that standard.
Reaching that Dermatologic Surgery issue and seeing which paper is actually on those pages would settle it. Until somebody does, treat the 71 women figure as unverified rather than as evidence.
These two are not even in the same system
GHK-Cu is in two places at once
- September 2023: injectable GHK-Cu placed in Category 2 of the 503A list [7]
- At the same time, non-injectable GHK-Cu went to Category 1, permitting compounding
- April 2026: injectable GHK-Cu came off Category 2 because nominations were withdrawn [8]
- Advisory committee review is scheduled by February 2027
Matrixyl is a cosmetic ingredient
- Cosmetic ingredients do not go through FDA approval before sale in the US
- The manufacturer is responsible for safety; the FDA acts after the fact
- Being sold widely does not mean anyone reviewed it for effectiveness
- There is no injectable version to discuss
What this means in practice
- A GHK-Cu serum is a cosmetic and is sold freely
- Injectable GHK-Cu cannot currently be compounded in the US
- Matrixyl is a cosmetic ingredient and always has been
- Withdrawn nominations are not a safety finding either way
Same molecule, two different answers depending on how it goes in. That is the whole of the GHK-Cu regulatory story, and it is why a serum and a vial are not the same conversation.
Questions people ask
Which one works better?
Nobody has tested them against each other in an ordinary serum. Matrixyl has two randomized human trials on wrinkle appearance. GHK-Cu has far more research overall, most of it about cells and wound healing rather than wrinkles. That is a difference in the kind of evidence, not a verdict.
Is Matrixyl 3000 better than regular Matrixyl?
It is different, not proven better. The human trials people cite were run on the original Pal-KTTKS. Matrixyl 3000 is two other peptides, and it has not been through the same testing.
Can I use them together?
Many products already combine them. Nobody has tested the combination in a controlled trial, so the reasoning is mechanistic rather than measured.
Does GHK-Cu actually get into skin?
Unsettled. It is small enough, at about 404 daltons with the copper. The problem is that it is charged and water loving, and skin resists that. Formulation matters more than concentration here.
Why do copper peptides turn things blue?
The copper. It is cosmetic rather than harmful, and it fades.
Is injectable GHK-Cu legal?
Not for compounding in the US. Injectable GHK-Cu was placed in the FDA's restricted category in 2023, and while it came off that list in April 2026 because the nomination was withdrawn, it has not been approved and is not permitted. Topical GHK-Cu is a different matter and is sold as a cosmetic.
Are they banned in sport?
Neither is named on the 2026 Prohibited List. Topical use is not a realistic doping concern. Injectable GHK-Cu is less clear, because the S0 catch-all covers substances with no approval from any government health authority.
What this page is built on
- 01Robinson LR, Fitzgerald NC, Doughty DG, Dawes NC, Berge CA, Bissett DL. Topical palmitoyl pentapeptide provides improvement in photoaged human facial skin. Int J Cosmet Sci. 2005;27(3):155-60. DOI 10.1111/j.1467-2494.2005.00261.x. PMID 18492182. Human clinical trial. 93 women, 12 weeks, double blind, placebo controlled, split face. Funded by Procter and Gamble.
- 02Osborne R, Robinson L, Mullins L, Raleigh P. Use of a facial moisturizer containing palmitoyl pentapeptide improves the appearance of aging skin. J Am Acad Dermatol. 2005. Human study. 8 weeks, split face, about 60 per product. Funded by Procter and Gamble. Volume, issue and pages not confirmed.
- 03Pickart L, Thaler MM. Tripeptide in human serum which prolongs survival of normal liver cells and stimulates growth in neoplastic liver. Nature New Biology. 1973;243(124):85-7. Original identification of GHK.
- 04Leyden JJ, et al. GHK-Cu facial cream, 71 women, 12 weeks. Citation disputed. Sources variously give this as an American Academy of Dermatology presentation, February 2002; as Dermatol Surg. 2005;31(7):809-816; and as J Cosmet Dermatol. 2002;1(3):197-204. Human study, citation unresolved. A 2026 review cites two Leyden papers with different titles at identical journal, volume, issue and pages, which cannot both be correct.
- 05Watson REB, Ogden S, Cotterell LF, Bowden JJ, Bastrilles JY, Long SP, Griffiths CEM. A cosmetic 'anti-ageing' product improves photoaged skin: a double-blind, randomized controlled trial. Br J Dermatol. 2009;161(2):419-426. PMID 19438432. Human randomized controlled trial. Tested a finished multi-ingredient product, not GHK-Cu alone.
- 06Abdulghani AA, Sherr S, Shirin S, Solodkina G, Tapia EM, Gottlieb AB. Effects of topical creams containing vitamin C, a copper-binding peptide cream and melatonin compared with tretinoin on the ultrastructure of normal skin. Disease Manag Clin Outcomes. 1998;1:136-141. Human pilot, 20 subjects, minor journal.
- 07US Food and Drug Administration. 503A bulk drug substances category updates, September 29, 2023. GHK-Cu placed in Category 2 for injectable routes and Category 1 for non-injectable. Regulatory document.
- 08US Food and Drug Administration. Category 2 removals, April 2026, following withdrawal of nominations. Regulatory document.
- 09Badenhorst T, Svirskis D, Merrilees M, Bolke L, Wu Z. Effects of GHK-Cu on MMP and TIMP expression, collagen and elastin production, and facial wrinkle parameters. J Aging Sci. 2016;4(3):166. Human study of a nano-carrier formulation, including a direct comparison against Matrixyl 3000. Low tier journal.
- 10Pickart L, Vasquez-Soltero JM, Margolina A. GHK and DNA: resetting the human genome to health. Biomed Res Int. 2014;2014:151479. Review by the discoverer of GHK, who has commercial interests. Based on reanalysis of a public gene expression database.
- 11Janvier S, Cheyns K, Canfyn M, Goscinny S, De Spiegeleer B, Vanhee C, Deconinck E. Impurity profiling of the most frequently encountered falsified polypeptide drugs on the Belgian market. Talanta. 2018;188:795-807. DOI 10.1016/j.talanta.2018.06.023. Peer-reviewed analytical study.
- 12NBC Washington. Lab finds problems in 30% of peptide vials tested. December 2024. News report of commercial laboratory testing, not peer reviewed.
- 13World Anti-Doping Agency. The 2026 Prohibited List. Effective 1 January 2026. Regulatory document.
Peptide Decoding is published by Decoded Sciences LLC. We take no payment from vendors for coverage, inclusion or ranking, and our affiliate relationships are disclosed in full.