Matrixyl

Origin and structure

The story starts with a cell study. Collagen is made as a longer precursor, procollagen, whose end pieces are cut off when the fibres form. In 1993 Katayama and colleagues at the University of Tennessee took one of those cut-off pieces apart and found that five of its amino acids, lysine-threonine-threonine-lysine-serine (KTTKS), were the shortest sequence that still strongly increased collagen and fibronectin production in several types of connective-tissue cells.

On its own KTTKS is water-loving and electrically charged, so it barely enters skin. Cosmetic chemists therefore attached palmitic acid, a fatty acid with 16 carbon atoms, to the start of the chain. The result, pal-KTTKS, carries the cosmetic ingredient (INCI) name palmitoyl pentapeptide-4; that name also covers a variant with the sequence KTSKS. The French supplier Sederma sells it as Matrixyl, a ready-made solution that contains 100 ppm (0.01%) of the peptide, and researchers at Procter & Gamble ran the largest trial of the peptide on its own.

How it is thought to work

Palmitoyl pentapeptide-4 is classed as a matrikine: a small fragment of the skin's own structural proteins that acts as a signal. The idea, put forward in the 1993 paper, is a feedback loop: the end pieces cut off procollagen as collagen is made tell fibroblasts to build more matrix. The evidence for that signal comes from cultured cells. The human trials measured wrinkles on photographs and skin replicas, skin thickness or elasticity, not collagen itself.

Getting the peptide to the fibroblasts is a second hurdle, because they sit in the dermis, under the outer layers of the skin. In a laboratory test on hairless-mouse skin, plain KTTKS was found in none of the skin layers after 24 hours, while pal-KTTKS reached all of them: 4.2 µg/cm² in the stratum corneum, 2.8 µg/cm² in the epidermis and 0.3 µg/cm² in the dermis, with none passing all the way through. In skin extracts both forms were broken down quickly by enzymes, the palmitoylated form more slowly. These penetration data come from mouse skin, not from living human skin.

Is Matrixyl approved?

It does not need to be, because it is sold as a cosmetic ingredient, not as a medicine. The EMA has not authorised any medicine that contains palmitoyl pentapeptide-4.

In the EU, cosmetics fall under the Cosmetics Regulation (EC) No 1223/2009. Palmitoyl pentapeptide-4 is listed in the European Commission's CosIng ingredient database as a skin-conditioning ingredient and appears in none of the annexes that ban or restrict substances, so no specific limit applies. Ingredients like this need no approval in advance; instead, every finished cosmetic needs a safety assessment and must be notified in the EU's cosmetics portal (CPNP) before it goes on sale. In the Netherlands the NVWA supervises this. The regulation also states that a product intended to be swallowed, inhaled, injected or implanted is not a cosmetic.

In the United States cosmetics and their ingredients, apart from colour additives, need no FDA approval before they go on the market. A product becomes a drug in law when it is intended to treat disease or to affect the structure or function of the body. Anti-ageing creams therefore promise a smoother look rather than new collagen. The Cosmetic Ingredient Review (CIR), the US expert panel that assesses cosmetic ingredients, concluded in 2024 that palmitoyl pentapeptide-4 is safe in cosmetics at the concentrations reported to it.

In sport, palmitoyl pentapeptide-4 is not named on the WADA Prohibited List, and the list does not mention cosmetics. Class S0 prohibits at all times any pharmacological substance that no health authority has approved for human therapeutic use, and the 2027 list, in force from 1 January 2027, adds peptides to its examples. Pure pal-KTTKS sold as a research peptide is not an approved medicine. The Dopingautoriteit applies the same list in the Netherlands.

What the research shows

Beyond the cell work there is a handful of human studies, all of skin creams. They are small and short, and most were run or paid for by companies that make or use the ingredient:

• Cell studies (Katayama, 1993): KTTKS was the shortest piece of the procollagen fragment that still strongly stimulated collagen and fibronectin production in several types of connective-tissue cells.
• Robinson et al. 2005 (Procter & Gamble): 93 women aged 35 to 55, 12 weeks, double-blind, split face. The moisturiser with 3 ppm pal-KTTKS reduced wrinkles and fine lines significantly more than the same moisturiser without it, judged by image analysis and expert grading of photographs, and the women also rated their fine lines as improved. It was well tolerated. The abstract gives no effect size.
• Sederma studies (2002): a four-month comparison of a cream with 5 ppm against the plain cream (30 people per group) and a 16-person study with 3 ppm reported less roughness and shallower wrinkles on skin replicas, and in the smaller study thicker skin on ultrasound. Sederma researchers reported both in 2002 in short journal-supplement reports, not in full trial papers.
• Aruan et al. 2023 (Indonesia, funded by a cosmetics company): 21 women, 7 per group, applied a cream with palmitoyl pentapeptide-4, one with acetyl hexapeptide-3 (Argireline) or a placebo cream around the eyes for 8 weeks. None of the reported comparisons reached statistical significance, although the authors still describe the palmitoyl pentapeptide-4 cream as the best performer; they call the study too small and too short and ask for a larger, longer trial. The peptide concentration was not reported.
• Fu et al. 2010 (funded by Procter & Gamble): in 196 women with moderate to moderately severe wrinkles around the eyes, an 8-week regimen of three products with niacinamide, pal-KTTKS, a second palmitoyl peptide and carnosine, one of them also with retinyl propionate, improved wrinkle appearance more than 0.02% tretinoin and was better tolerated; after 24 weeks, in the roughly 25 women per group who continued, results were comparable. Because the products contained several actives, the effect cannot be put down to pal-KTTKS.

Matrixyl versus Matrixyl 3000

The names are easy to mix up. Matrixyl is the solution of palmitoyl pentapeptide-4. Matrixyl 3000, despite the name, is built on two different peptides, palmitoyl tripeptide-1 (pal-GHK) and palmitoyl tetrapeptide-7 (pal-GQPR), rather than on palmitoyl pentapeptide-4. The trials above tested pal-KTTKS, so they say nothing direct about Matrixyl 3000, and studies of Matrixyl 3000 say nothing direct about pal-KTTKS.

Ingredient lists use the INCI names, not trade names: palmitoyl pentapeptide-4 is the Matrixyl peptide, while palmitoyl tripeptide-1 together with palmitoyl tetrapeptide-7 is the Matrixyl 3000 pair. A percentage of 'Matrixyl' in marketing refers to the supplier solution, not to pure peptide; because that solution contains 100 ppm, a cream with 3% Matrixyl holds about 3 ppm of palmitoyl pentapeptide-4.

Side effects and safety

On the skin, palmitoyl pentapeptide-4 has a clean record so far. In the 93-woman trial it was well tolerated, and in the small Indonesian trial nobody in the peptide group reported a complaint. The data the CIR panel reviewed show that it was not acutely toxic to rats or guinea pigs, not genotoxic in two bacterial tests and one test on mammalian cells, not irritating on human skin, including a 48-hour patch test, predicted not to cause allergy in two laboratory tests, and not sensitising in two human repeat-insult patch tests. The panel concluded in 2024 that it is safe in cosmetics as used; the highest concentrations reported were 12 ppm in face, neck and eye products and 35 ppm in hair conditioners.

The gaps are just as clear. The panel noted that there are no studies on reproduction and development and no cancer studies, and accepted that because use concentrations are low, little passes through the skin and the genotoxicity tests were negative. All of these data concern a few parts per million on intact skin. There are no human data for injection, for oral use or for amounts of pure peptide, and under EU law a product intended for injection is not a cosmetic.

Frequently asked questions

Does Matrixyl really work against wrinkles?

There is some evidence, but it is modest and mostly comes from the industry. In the largest trial, run by Procter & Gamble, 93 women had fewer visible fine lines after 12 weeks on the half of the face treated with 3 ppm pal-KTTKS than on the placebo half; the published abstract gives no effect size. Smaller supplier studies point the same way, while a 21-woman trial in Indonesia found no statistically significant differences. The trial that compared a peptide regimen with prescription tretinoin used products that also contained several other actives, so it says little about pal-KTTKS on its own.

What is the difference between Matrixyl and Matrixyl 3000?

They contain different peptides. Matrixyl is palmitoyl pentapeptide-4 (pal-KTTKS). Matrixyl 3000 is built on palmitoyl tripeptide-1 (pal-GHK) and palmitoyl tetrapeptide-7 (pal-GQPR) instead of pal-KTTKS. Research on one is not evidence for the other, and on an ingredient list you can tell them apart by these INCI names.

Is Matrixyl legal in the Netherlands?

Yes, as a cosmetic ingredient. Palmitoyl pentapeptide-4 is permitted without restrictions under the EU Cosmetics Regulation; each finished cream needs a safety assessment and a notification in the EU cosmetics portal, and the NVWA supervises this in the Netherlands. It is not a medicine. Pure peptide sold as research material has not been through the safety assessment a finished cosmetic needs and is not meant for use on or in people.

Can Matrixyl be injected?

No study has tested injection, and there are no safety data for it. The ingredient was developed and assessed for creams containing a few parts per million. Under EU law a product intended to be injected is not a cosmetic, and the EMA has authorised no medicine that contains palmitoyl pentapeptide-4.

Is Matrixyl banned in sport?

It is not named on the WADA Prohibited List, and the list does not mention cosmetics. Class S0 prohibits any pharmacological substance without approval for human therapeutic use, and from 1 January 2027 the list names peptides as an example; pure pal-KTTKS is not an approved medicine. The Dopingautoriteit uses the same list in the Netherlands.

How much peptide is in a cream with Matrixyl?

Very little. The supplier solution Matrixyl contains 100 ppm (0.01%) palmitoyl pentapeptide-4, so a cream with 3% Matrixyl holds about 3 ppm, the level used in the main trial. In an industry survey for the CIR panel, the highest concentrations in use were 12 ppm in face, neck and eye products and 35 ppm in hair conditioners.

Storage and handling

Store according to product-label instructions and batch handling notes. Lyophilized research peptides are commonly kept dry, protected from light, and handled with clean laboratory technique.

References

Checked against these sources on 28 October 2026.