Bioactive tetrapeptide GEKG boosts extracellular matrix formation: in vitro and in vivo molecular and clinical proof


Univ. Prof. Dr. med. Jean Krutmann, Director, IUF-Leibniz Research Institute for Environmental Medicine, Duesseldorf, Germany, Auf ’m Hennekamp 50, D-40225 Duesseldorf, Germany, Tel.: +49 211 3389 225, Fax: +49 211 3389 226, e-mail:


Abstract:  The ‘matrikine’ concept claims that processing of the precursors for collagen results in the formation of peptides such as KTTKS which in turn augments extracellular matrix (ECM) production. In the present study, we show the development of an anti-ageing active from an in silico approach by molecular design resulting in the tetrapeptide GEKG derived from ECM proteins. The efficacy of the peptide to significantly induce collagen production of the protein level and mRNA level has been demonstrated in vitro in human dermal fibroblasts and in vivo in a double-blind, randomized, placebo-controlled study enroling 10 volunteers with an average age of 48.2 years. The effect of GEKG on facial wrinkles was studied in 30 volunteers using state of the art fringe projection, which allows determination of surface roughness in three-dimensions. Here, only GEKG but not the placebo was able to significantly decrease skin roughness as a measure for wrinkles.


Rarefication of collagen fibres in ageing human skin results from increased breakdown and decreased de novo synthesis (1–3). Collagens are synthesized as propeptides extracellularly processed by specific peptidases. Even a pentapeptide sequence comprising KTTKS (Lys-Thr-Thr-Lys-Ser) derived from the carboxy terminal propeptide of type I collagen substantially augmented extracellular matrix (ECM) production (4,5). Therefore, a regulation loop has been postulated in which the release of peptides called ‘matrikines’ from the surrounding ECM stimulates the activity of the residing fibroblasts (6). Besides KTTKS, GHK and CNYYSNS have been identified as matrikines. With regard to skin ageing, GHK and KTTKS are effective for cosmetic indications (7), although true proof-of-principle studies including determination of molecular markers in addition to measuring non-invasive parameters are lacking (8–10).

Questions addressed

An in silico approach was used to develop efficient peptides by identifying highly repetitive amino acid motifs in several ECM proteins (11,12). The resulting tetrapeptide GEKG (Gly-Glu-Lys-Gly) was assessed for its effects on the formation of ECM components in vitro and in vivo in the present proof-of-principle study.

Experimental design, results and discussion

Primary human foreskin fibroblasts were incubated with GEKG or the benchmark KTTKS available as palmitoyl derivative (pal-KTTKS) for 24 h, and supernatants were assessed for procollagen formation. While the pentapeptide induced a minor increase, GEKG resulted in a significant procollagen secretion (Figure S1a). With regard to the level of mRNA expression [for details see Data S1, Table S1 and (13)], the main skin collagen COL1A1 was increased 1.8-fold by pal-KTTKs and 2.8-fold by GEKG by 1 ppm (Figure S1b). Posttranslational cross-linking of newly synthesized collagen has been excluded in these assays by inhibition of lysyl oxidase with β-aminopropionitrile. The net content of collagen is a delicate balance between its synthesis and degradation, e.g., by matrix metalloproteinase-1 which could also be an effective target for anti-ageing actives (14). In addition, dermal function can also be impaired by increased subcutaneous adipose tissue (15).

The anionic, non-sulphated glycosaminoglycan hyaluronan undergoes a daily turnover especially because of UVB radiation (16). Expression of hyaluronic acid synthase-1 (HAS1), one of the hyaluronan synthesizing enzymes, was 12.6-fold increased by pal-KTTKS, whereas GEKG induced mRNA expression of HAS1 5.7-fold. Fibronectin mRNA was moderately induced by pal-KTTKS and strongly upregulated by GEKG. Taken together, GEKG increased the formation of procollagen, HAS-1 and fibronectin.

To assess the efficacy of GEKG in vivo, 10 healthy volunteers (>35 years) were treated in a double-blind, randomized, placebo-controlled study once a day for a period of 8 weeks with vehicle only (=placebo 1) or the same vehicle containing 50 ppm GEKG. Treatment areas were either on buttock skin for extraction of skin biopsies or on volar forearms for skin physiological measurements. Expression of COL1A1 was significantly increased under both treatments after 8 weeks as compared with the untreated skin (Figure S1c, for details see Tables S1 and S2). The difference between the two treatment arms was also significant, GEKG being superior (P = 0.02, Wilcoxon signed rank test). Histochemical analyses indicate that treatment with tetrapeptide GEKG, but not with placebo, increased the formation of procollagen, hyaluronic acid and fibronectin (Fig. 1).

Figure 1.

 In an in vivo approach, 10 healthy human volunteers (six female and four male, 40–65 years, average 48.2; with 45.0 for women and 54.0 for men) were enroled after written informed consent into a double-blind, placebo-controlled study (approval had been obtained from the Ethics Committee of the Heinrich-Heine University) conducted according to the ethical rules stated in the Declaration of Helsinki Principles, and the ICH GCP guideline was observed insofar as applicable. After treatment with 50 ppm GEKG or O/W vehicle (=placebo 1) for 60 days on buttock skin, biopsies were taken for immunohistochemical assessment of procollagen I, hyaluronic acid and fibronectin as described in Supplementary Data.

Skin elasticity was measured and calculated as resilient distension (R1), gross elasticity (R2), net elasticity (R5) and elasticity/complete curve (R7). After 8 weeks of treatment, GEKG significantly improved in vivo R1 to 34.1% (P = 0.002 paired t-test before versus after), while the placebo treatment resulted in an improvement of 11.4%. R2 improved to 5.6% compared with −3.9% for the placebo, R5 to 10.7% compared with −0.6% for the placebo and R7 to 12.3% as compared with −1.5% for the placebo treatment (Figure S1d). The effects with regard to R2, R5 and R7 seen for tetrapeptide GEKG compared with placebo treatment were impressive but not significant probably because of the low number of 10 volunteers.

Therefore, in a second in vivo study enroling 60 volunteers with no history of skin diseases, the skin physiological effects of the tetrapeptide GEKG were compared with the effects of the palmitoyl derivative of pentapeptide KTTKS with regard to elasticity R1, volume and roughness after 8 weeks of twice daily applications. Elasticity on the inner forearm increased after two daily treatments over 8 weeks to 41.3% for 10 ppm GEKG, while palmitoyl-KTTKS showed an improvement of 35.6% (Figure S2a). To assess the anti-ageing effect of the tetrapeptide, photographs were taken from volunteers’ skin before and after the 8-week application period using Visioscan. A software compared the distribution of grey scales and calculated the theoretical amount of liquid that would be necessary to fill the wrinkles and generate a plane surface (17). A reduction in the volume is interpreted as an improvement in skin structure because of reduction of wrinkles in number and depth. Accordingly, a reduction in volume was observed for 10 and 100 ppm GEKG, respectively, resulting in a decrease of 8% and 12.2% (Figure S2b). In addition, the roughness parameters were also assessed via Visioscan. They originate from the DIN-parameters Ra − Rz describing the depth of fine and coarse wrinkles and give the maximum and average amplitude of a surface structure as well as the mean height level. Smoothness depth Rp was significantly improved to 41.9% by 100 ppm GEKG (Figure S2c), while pal-KTTKS resulted in an improvement of 18%.

Finally, a placebo-controlled study was carried out to analyse the effect of GEKG on facial wrinkles. Skin roughness was determined by a non-invasive in vivo state of the art technique, i.e., fringe projection area topography using the Primos Pico system (18). The parameters Ra and Rz only provide profile information but not surface area information (19). Therefore, they have been transformed and upgraded to the third dimension resulting in terms called Sa and Sz (20). Comparison of the periorbital wrinkles before and after treatment with GEKG (Figure S2d, for details see Table S2) indicated a significant change of SaΔSa after 4 weeks by 6.1% and after 8 weeks by 9.5% and of ΔSz by 3.3% after 8 weeks, while the placebo had no significant effect. Representative digital images of periorbital areas are presented in Fig. 2.

Figure 2.

 In a facial wrinkle study (n = 30), the effects of 50 ppm GEKG on facial wrinkles as compared with a O/W vehicle were assessed. Representative examples of periorbital wrinkles are given before and after a twice daily, 8-week treatment.


These studies for the first time show that topical application of the matrikine GEKG to human skin stimulates ECM protein expression of collagen at the mRNA and protein level, and of hyaluronan and fibronectin, respectively, and that this is associated with a significant improvement of skin physiological and clinical parameters such as skin wrinkles and skin roughness reflecting skin ageing. We also demonstrate that a ‘molecular design’ approach succeeded in improving the efficacy of matrikines in human skin.


The excellent technical assistance of M. Weber and U. Schubhart is gratefully acknowledged. Cutaneous roughness was studied using image analysis (Primos Pico) by ISPE, Milan, Italy. This work was supported by the Deutsche Forschungsgemeinschaft SFB 728 TP C1 and sponsored by Evonik Goldschmidt GmbH, Essen, Germany.

Authors contributions

MF, SGB, PL, TF, JK designed research. AM, UM, JL, TK, IF, HB, TJ, SF performed experiments. SGB and JK prepared the manuscript. The final manuscript was approved by all authors.

Conflict of interest

The authors state no conflict of interest.