Synthetic lipopeptides as potential topical therapeutics in wound and skin care: in vitro studies of permeation and skin cells behaviour

Abstract

Wound healing is an extraordinarily complicated process associating with the cell aging, slowing down of cell renewal mechanism and even loss of protective function to environmentally induced oxidative stress. A lack of effective treatment results in the appearance of skin lesions and chronic wounds. A satisfactory answer to skin care and therapy problems seems to be a molecule Gly-His-Lys of natural origin that promotes synthesis of collagen, inhibits the activity of matrix metalloproteinases (MMPs) and reduces the activity of collagenases. Several Gly-His-Lys analogues were obtained to investigate their antimicrobial properties. Three lipophilic analogues with the structure numbering 1b, 2b, 4b, exhibit significant effect against bacteria and were selected for in vitro evaluation of skin cells behaviour. In the present studies, an in vitro model of wound repair, proliferative cell staining, and tracking of living cells were used. Cell proliferation and cell migration, as important actions during the wound repair process, were studied. All the analogues provided an optimal environment for the growth, development and viability of skin cells. Microscopic changes in the morphology of fibroblasts and keratinocytes confirmed the positive effect of all the analogues on their growth and condition. Franz-type diffusion studies and the behaviour of analogues in the in vitro model of permeation were also performed. Permeability and diffusion coefficients were measured using UV-Vis spectrophotometry and RP-HPLC techniques. A substantial increase of permeability parameters for the lipophilic analogues was noticed. The outcomes of these studies suggest that lipophilic analogues possess a high potential as topical therapeutics. In relation with their proved antimicrobial activity, their safety profile and great tolerance for human skin show that they may be applied in both skin care and wound therapies.

---

1. Introduction

Despite the passage of years and vast knowledge in the field of cosmetology and medicine, the process of skin reconstruction and restoration constitutes still a subject of much scientific research. The complexity of this process is also associated with the syndrome of a dysfunction and slow down of cell renewal mechanism and even loss of protective functions to environmentally induced oxidative stress.^1–3 The lack of effective treatment and complications, especially in patients with diabetes or cardiovascular disease, results in the appearance of skin lesions and chronic wounds, i.e. diabetic foot ulcers, pressure ulcers or bedsores. In recent years, the incident rate of metabolic syndromes and cardiovascular disease has increased by around 5% and 7% per year, respectively. The process of healing is difficult not only due to many of severe symptoms of disease, but also progressive aging process. All the alterations in the skin over time are also visualized as cosmetic defects, such as hyperpigmentation, wrinkles, redness, stretch marks, scars and leading to the development of chronic skin failure, dysfunction of cell renewal mechanisms and risk of progression of disease entity.^4–7

Human growth factors (HGFs) such as epidermal growth factor (EGF), fibroblast growth factor (FGF2), transforming growth factor (TGF-β) and platelet-derived growth factor (PDFG) play a crucial role in both processes. The pleiotropic character of cytokines allows for suppressing the immune response and regulates the balance between the destruction process that is necessary to remove damaged tissues and the process of new tissue formation.^8–10 Their low activity contributes to impaired wound healing and involves long-term acute phase characterized for chronic wounds. Additionally, both syndromes inflammatory and microbe infections come from endogenic or exogenic factors, trigger the exudation. This exudation is associated to the HGFs and extracellular matrix (ECM) degradation. In fact, the migration of keratinocytes inside the wound is limited, complicating epithelialization of the wound surface and the scar formation.^11,12 Last year research proved that key HGFs, involved in the quality of wound healing, are essential features such as debridement, maintaining a moist wound environment and protection against the secondary infection.

The alternative to initiate the appropriate treatment is the use of the antiseptics such as chlorhexidine, iodine, hydrogen peroxide.^13–16 Clinically infected wounds usually require systemic antibiotic therapy. This way of treatment is associated with the development of multidrug resistance, side effects and low permeability of antibiotics to affected areas of the skin. Although topical antibiotics are important in soft tissue therapies, their current usage is usually limited to clinical applications. Applying modern drug for topical treatment of wounds allows to ensure the appropriate concentration of biologically active substance. Controlled delivery of therapeutic agents provides effective treatment.^17–21 Recently, HGFs and biomimetic substances (plant extracts, peptides), stimulating cell division, differentiation and growth, and also imitating the effect of MMP inhibitors, have become very important.^22–27

The primary object of our research is a natural origin tripeptide (Gly-His-Lys) that is involved in the regeneration process of damaged tissues. It promotes the synthesis of collagen and elastin, inhibits the activity of MMPs in the ECM and reduces the activity of collagenases. Furthermore, Gly-His-Lys is recognized as a copper ion carrier, improving the firmness of skin, smoothing wrinkles and inhibiting hyperpigmentation.^28–32 Nowadays, copper complex of Gly-His-Lys is commonly used in the anti-aging formulas as a factor to fight with the cosmetic defects. Numerous reports have proved clinical application of tripeptide in healing of wounds of various etiologies, especially by fibroblast proliferation and migration.^32–36 A wide range of biological activities, low toxicity, but on the other hand adversely pharmacological properties resulted from the rapid biodegradation, high hydrophilicity, low permeability and also its physical properties (e.g. stability, solubility), all are the reason for the modification of its structure. All the mentioned parameters are able to increase the possibility of its application in skin and tissue therapies. Interestingly, in our previous studies, we presented modification of Gly-His-Lys analogues with fatty acids that increased the affinity to phospholipids membrane in in vitro studies. The modifications had proved that peptides increased the bacterial and fungal membrane binding and insertion. Moreover, all the tested peptides exhibited strong to moderate activity and nine lipophilic analogues marked 1b–e, 2a, 2b, 2e, 4b, 5b were selected for studies based on the evaluation of safety profile to make the peptides very attractive candidates for topical therapeutics³⁷ (Fig. 1). Encouraged by the results obtained earlier, the goal of our research is to exploit the activity of synthetic lipophilic analogues of Gly-His-Lys as potential therapeutics in wound and skin care.

Fig. 1 The structure of Gly-His-Lys analogues and their conjugates with the fatty acids.

2. Materials and methods

2.1. Cell culture and materials

2.1.1. Human adult calcium high temperature (HaCaT) keratinocyte cells. Human immortalized non-tumorigenic cell line was supplied by DKFZ Heidelberg. Cells were grown in Dulbecco's modified Eagle's medium (DMEM) with 10% fetal calf serum (FCS) and cultured in humidified atmosphere (5% CO₂) at 37 °C. Cells were grown with routine passage every 3 days until adequate numbers of cells were obtained.^38,39

2.1.2. Human primary fibroblast cells. All procedures for fibroblast cell isolation have been performed according to the approved guidelines of the Ethics Committee of Medical University of Gdansk (NKEBN/483/2011 Issued: 2011-12-19), and the samples of the skin were only taken when written informed consent from the patient was received. Biological material was obtained from healthy patients between 45 and 55 years of age. Skin biopsies were transferred into phosphate buffered saline (PBS). The dermis was subsequently separated from epidermis after cutting it into small pieces and digestion in dispase BD (6 U mL⁻¹) for 19 h at 4 °C. This was followed by incubation for 2 h at 37 °C in medium containing collagenase. Released cells and tissue fragments were then suspended in DMEM medium containing 10% FCS and centrifuged. The resulting suspension was reached at a density of 10⁴/cm⁻².

2.1.3. Cell proliferation assay. Cell proliferation assay was carried out according to the procedure previously described.³⁹ Cells in DMEM medium after the period of incubation provide a 100% relative numerous cells (K positive control). Cells in DMEM medium containing 10% FCS were also incubated under the same conditions (K⁺).

2.1.4. Cell migration assay. Cell migration was assessed by a scratch test. Both human skin cells were seeded in 6-well plate in an amount of 1 × 10⁵ cells in DMEM with 10% FCS and the addition of antibiotics per well and grown to confluence of 90–100% for 72–96 h (37 °C, 5% CO₂). Thereafter, using a sterile tip of each well was performed two perpendicular strokes to give a surface not covered with cells. The scratch cells were removed from each well by rinsing DMEM. The fresh DMEM and cells were treated with selected peptides (1b, 2b, 4b, 6i) with the final concentrations of 0.001 μg mL⁻¹, 0.010 μg mL⁻¹ and incubated for 24 h (37 °C, 5% CO₂). Both cultures in DMEM (K⁺) and in DMEM treated with 1 μM of dexamethasone (DXM) (K⁻) were incubated under the same conditions as described above. After overnight incubation, the media was removed, cells were rinsed with sterile PBS and then 1% paraformaldehyde (PFA) was added. After 20 minutes, the cells were washed with PBS and stained with 0.05% crystal violet solution. The stained cells were washed with PBS and then distilled water was added to each well. The wounded areas were photographed by microscope (Zeiss). The proximal distance extending between the explant biopsy and the distal edge of cellular migration as well as the surface area covered by cells outgrowth were measured.

2.1.5. Observation of changes in skin cells morphology. For the visualization of changes in skin cells morphology, skin cells were seeded and exposed at the same conditions as described in 2.1.4. Then the cells were also visualized with 0.05% crystal violet solution. Microphotographs were taken under an inverted microscope (Zeiss).

2.1.6. Evaluation of in vitro cytotoxicity assessment. Skin cells were plated in a 96-well plate with 4 × 10³ and 5 × 10³ cells of concentration administered in DMEM with 10% FCS. The cells were left for 24 h before exposed to peptide solutions administered in DMEM at the final concentration of 0.01 μg mL⁻¹, 0.10 μg mL⁻¹, 1.00 μg mL⁻¹, 10.00 μg mL⁻¹ and again incubated for 48 h. Subsequently, 5 mg mL⁻¹ 3-(4,5-dimethyl-2-thiazolyl)-2,5-diphenyltetrazolium bromide (MTT) in sterile PBS was added directly to the wells and the plates were incubated for 4 h (37 °C, 5% CO₂). Then media was discarded and 0.06 M HCl/iPrOH was added and shaken to dissolve MTT-formazan crystals. The optical density of solution was measured at wavelength of 570 nm with a microplate reader.

2.1.7. Statistical data analysis. All data obtained from measurements (3 repetitions in 3 independent experiments) were analyzed using Statistica ver.8.0 and Microsoft Excel ver.2010. The studies compared several variable parametric ANOVA test was used in combination with post hoc test NIR (p < 0.05 were considered statistically significant).

2.2. Franz diffusion cell system

In vitro study of permeability of peptides (1a–e, 1i, 2b, 2e, 3b, 4b, 4i, 6i, 6c) was performed using Franz-type diffusion cells consist of two vertical cells composed of both donor and acceptor cells and a liquid crystalline membrane. The diffusion area was placed in water bath (37.0 ± 1.0 °C). Surface diffusion cell was 1.77 ± 0.50 cm² and the volume of medium in the acceptor cell was 19.0 ± 0.1 mL. Semipermeable, porous films: polyester film (diameter: 24.0 mm, pore size 0.4 μm, thickness 12.0 μm) and two films of cellulose: cellulose film I (diameter: 24.0 mm, pore size: 0.9 ÷ 1.1 μm, thickness: 11.0 μm; water permeability: 0.8–1.9 dm³ m⁻² min⁻¹); cellulose film II (diameter: 24.0 mm, pore size: 0.7 ÷ 1.0 μm, thickness: 22.5 μm, water permeability: 1.9 ÷ 4.8 dm³ m⁻² min⁻¹) and the cerosome lipid mixture were used. Thickness of model lipid membranes before the stabilization was 0.078 ± 0.007 mm.

2.2.1. Compatibility of polymeric films with peptides. Polymeric films with a surface area of 2 cm² were shaken with 6 mL of peptide solutions (12 μg mL⁻¹ and 200 μg mL⁻¹) for 6 h at 37 °C. The samples were analyzed using reverse-phase high-performance liquid chromatography (RP-HPLC).

2.2.2. Preparation of model membrane. Lipid membrane was prepared in accordance with the methods described previously.^40–42 The lipid suspension in the amount of 0.125 cm³ was placed between two films. Then the model membrane was contained in the diffusion cell filled with a phosphate buffer solution at pH 7.4 and left on the 24 h in order to balance with the medium of the acceptor cell.

2.2.3. Peptide solutions. All the peptides including Gly-His-Lys (standard peptide, 6i) were obtained in our laboratory in a solid phase method using 9-fluorenylmethyloxycarbonyl procedure. Peptides were obtained with a high purity (>98%) and completely characterized.^37,43 The structure of Gly-His-Lys analogues was presented in Fig. 1. The suspensions of peptides received by the 24 h shaking an excess of a suitable peptide in phosphate buffer or ethanol–phosphate buffer (30 : 70, v/v) at 37 °C. The solubility of peptides was determined after 1 h, 6 h and 24 h using ultraviolet-visible (UV-Vis) spectroscopy.

2.2.4. Evaluation of permeability of the peptides. The experiments were performed simultaneously using two Franz-cells diffusion cells. The solution of peptide in the concentration range of 0.02 ÷ 0.05% (w/v) was applied on the surface of membrane in the donor cell. Samples were received from the acceptor cells at time intervals. In case of the second cell, the experiment has continued by taking samples at 24 h, 48 h, 72 h. Each time after sampling the acceptor cells, it was filled with the same volume of fresh medium. All the samples were analyzed using matrix assisted laser desorption ionization time of flight mass spectrometry and RP-HPLC. Quantification of peptides in samples was performed by UV-Vis spectrophotometry and RP-HPLC methods. All the parameters such as flux (J), lag time (t_L), diffusion coefficient (D), permeability coefficient (K_p) were calculated.

2.2.5. Determination of peptides in the model membrane. The procedure for isolation of the peptides from the model membrane after 10 h and 72 h proceeded in two ways:
(a) Direct method. The model membrane was divided into: an outer film forming part of the donor (BI), the lipid layer (LII) and an inner film forming part of the acceptor compartment (BIII). The membrane was removed from the cell and prepared by separation into fractions. Both films BI and BIII were firstly separated from lipid fraction – LII. Each of these films BI, BIII and combined fractions of LII were extracted using phosphate buffer or ethanol–phosphate buffer (30 : 70, v/v) as well for 6 h at 37.0 ± 1.0 °C, in accordance with the procedure described previously.³⁷
(b) Indirect method. Method relied on the washing off absorbed compound from the membrane surface by pouring away of the content of the donor solution and rinsed quantities of phosphate buffer or ethanol–phosphate buffer (30 : 70, v/v) as well. The content of the peptides in a model membrane was calculated from the formula. The amount of peptides in the donor and acceptor cells was estimated by RP-HPLC.

Q_M = Q₀ − (Q_A + Q_D)

where Q₀ – the initial amount of compound in the donor cell [μg], Q_A – the amount of compound in the acceptor cell after the time of experiment [μg], Q_D – the amount of compound in the donor cell after the time of experiment [μg].

2.2.6. HPLC analysis. The samples were analyzed by RP-HPLC (linear gradient: 0–100% of solvent B in A in 30 min, 0–80% of solvent C in A in 30 min), eluents: (A) 0.1% trifluoroacetic acid (TFA) in water, (B) 0.1% TFA in 80% acetonitrile (ACN) solution, (C) 0.1% TFA in ACN, flow rate: 0.5 ÷ 1.0 mL min⁻¹, UV detection: 226 nm, 205 nm, columns: Jupiter 300 C18 4.6 × 250 mm (5.0 μm), ZORBAX SB-C8 Rapid Resolution HT 3.0 × 50.0 mm (1.8 μm). A calibration curve was calculated on the basis of peak area measurements of peptide solutions ranging 0.0125 ÷ 0.2000 mg mL⁻¹ with R² ≥ 0.9992. The limit of detection for tested peptides was found to be around 0.0100 μg mL⁻¹; the limit of reasonable quantification was set at 0.0200 μg mL⁻¹.

2.2.7. UV-Vis spectroscopic analysis. The quantification was performed using Spectrophotometer UV-VIS. The content of the samples was determined at 205 ÷ 214 nm. A correlation curve was established for peptide solutions ranging 0.00625 ÷ 0.20000 mg mL⁻¹ with R² ≥ 0.9989. Analytical wavelength was determined on the basis of the plot of absorbance vs. concentration at minimal and maximal wavelength. All the samples containing a higher peptide content were diluted prior to the analysis until the value within the linear range of the calibration curve was obtained.

2.2.8. Statistical data analysis. All data obtained from measurements (3 repetitions in 3 independent experiments) were analyzed using a computer program StatSoft Statistica ver.12.0 and Microsoft Excel ver.2010. Compliance distribution of quantitative variables analyzed the normal distribution using the Kolmogorov–Smirnov test. The discrepancies in the distribution of the results of individual samples contributed to the use of nonparametric ANOVA test, Kruskal–Wallis rank. In all calculations the values of p < 0.05 were accepted as statistically significant.

3. Results and discussion

A new series of synthetic Gly-His-Lys analogues were obtained to investigate their antimicrobial properties and then the safety profile for human skin to make the peptides good candidates for topical therapeutics. Interestingly, all the tested peptides exhibited strong to moderate activity and a few lipophilic analogues (1b–e, 2a, 2b, 2e, 4b, 5b) were selected for estimating the safety profile due to human skin cells. However, all the peptide conjugates with higher fatty acids, possessed high antibacterial activity, based on literature and our first experimental studies, were characterized by cytotoxic effect at lower concentrations.⁴⁴ Therefore, a series of Gly-His-Lys conjugates with lauric acid (1b, 2b, 4b) was selected to preliminary in vitro evaluation of the skin cell behaviour. Moreover, the analogues (1b, 2b, 4b) exhibited noticeable effect against both Gram-positive and Gram-negative bacteria and increased character of lipophilicity associated with high affinity to biological membranes.^37,44

3.1. In vitro evaluation of skin cell behaviour

In the present study, in vitro model of wound repair, proliferative cell staining and tracking of living cells was used. Both proliferation and migration activities are important during the wound repair process. The study in further analysis allowed for the determination of the positive effect of Gly-His-Lys analogues on human skin cells by measurement of proliferation and migration features. Gly-His-Lys (6i, standard peptide) as a natural tripeptide is well tolerated by skin. Its mechanism of molecular action, observed even at low concentration (0.1 nM), involves the stimulation of fibroblast cells, increasing the fibre biosynthesis and participation in both chondroitin, and dermatan sulfate production.⁴⁵ This effect is also a synergistic action of Gly-His-Lys with copper ions. Additionally, Gly-His-Lys is capable to stimulate proliferation of keratinocytes through the regulation of integrin α6, β1. The result is dependent on the tripeptide concentration and its feature seems to play a key role as a keratinocyte-growth factor. The increase of stem cells potential explained by the regulation of protein p63 and the ability to differentiation of keratinocyte cells contribute to continuous restoration, and regeneration of damaged skin.^46,47 Therefore, the goal of our study was to test the behaviour of skin cells in the presence of lipophilic analogues of Gly-His-Lys.

Initially, the study was focused on the evaluation of the influence of standard peptide on skin cells. The strongest biological effect was noticed at the lowest concentration of 0.01 μg mL⁻¹ and was comparable to the effect induced by the mitogenic factor, FCS (p < 0.05). Increasing 10-fold concentration of peptide 6i showed a decrease in relative number of human skin cells by about 20%. Only the effect of standard peptide at 10.0 μg mL⁻¹ on both cell lines resulted in 10% increase in cell number in comparison with its effect at 1.0 μg mL⁻¹ (Fig. 2A, p < 0.05). All the tested lipopeptides (1b, 2b, 4b) as well as standard peptide 6i exhibited proliferative capacity of skin cells in the whole range of concentrations in comparison with skin cells under physiological conditions. However, lipopeptide 1b in the concentrations of 0.01 μg mL⁻¹ and 0.10 μg mL⁻¹ induced 10% decrease of number of cells in keratinocyte cultures in comparison with peptide 6i. In two highest concentrations of 1.0 μg mL⁻¹ and 10.0 μg mL⁻¹, this correlation was opposite, increasing the growth of relative cell number in culture by 30% and 10%, respectively. The strongest proliferative effect on both cell lines was observed in the concentrations of 0.01 μg mL⁻¹ and 1.0 μg mL⁻¹. These effects on skin cells were comparable to the effect induced by FCS (p < 0.05).

Fig. 2 The effect of standard peptide (6i) and lipopeptides, respectively 1b, 2b, 4b in the concentrations of 0.01 μg mL⁻¹, 0.10 μg mL⁻¹, 1.00 μg mL⁻¹ and 10.00 μg mL⁻¹ on the proliferation of human skin cells: HaCaT keratinocytes (A), primary fibroblasts (B). K, control in DMEM, K⁺, positive control, in DMEM medium supplemented with FCS. The growth of cells was estimated by MTT test as described in Section 2. Results are presented as mean ± SD of triple wells; *p < 0.05 for cells growth with results without peptide. The obtained data shows that the majority of the examined peptides do not show a significant decrease in viability of human skin cells at whole concentration range at 48 h. Most of peptides presents an effective and safe potential for the treatment of cells. All the lipopeptides (1b, 2b, 4b) are able to increase the density of cells, especially in the lowest concentration of 0.01 μg mL⁻¹ as well as a standard peptide.

Strong proliferative capacity of skin cells was also observed for lipopeptide 1b at 0.10 μg mL⁻¹. Both lipopeptides 2b and 4b presented rather similar effect on skin cells and it was not comparable to FCS. Interestingly, taking into account the results obtained for two compounds 2b and 4b, we can surmise that the additional amino acids in basic sequence of Gly-His-Lys could contribute to less proliferative effect. Therefore, the lipophilic tripeptide 1b was more efficient to increase the number of skin cells than both lipophilic tetrapeptides (2b, 4b) (Fig. 2B). The higher concentration, the more number of cells in culture was observed in comparison with the number of cells in physiological conditions. For lipopeptide 1b at 10.0 μg mL⁻¹, only a slight decrease of number of fibroblast cells was observed (p < 0.05). It presented approximately 10% decrease in cell viability in comparison with the standard peptide. Peptide 6i caused 20% decrease of cell viability in comparison with control, but even at 10-fold lower concentration. Lipopeptide 2b at 0.10 μg mL⁻¹ initiated 35% decrease in cells population. Nevertheless, this negative effect was not observed for HaCaT cells treated with all the lipopeptides (1b, 2b, 4b). The minimal decrease of viability of cells isolated from the patients could be a result of individual characteristics (Fig. 2B). Selected peptides induced a protective effect on human cells. The most important effect was observed for lipopeptide 4b. This action could be explained by the presence of Met residue in the peptide sequence. Noteworthy, the Met residue takes part in reduction–oxidation reactions and plays a crucial role in control of homeostasis of biological processes and regulation of transduction of cell signal. Redox function of Met residue is also associated with disorders of aging but also with some physiological processes. Studies on the transformation of Met to oxidized form of Met are an important aspect in understanding cell signal transduction.⁴³ This process will be discussed in further studies focused on the evaluation of the effect of methionine-containing peptides on skin cells behaviour.

In addition to cell proliferation assay, cell migration assay was performed. The results indicated beneficial effect of peptides (6i, 1b, 2b, 4b) action. Peptide 6i in the lowest concentration of 0.001 μg mL⁻¹ inhibited the migration of keratinocytes as well as DXM, increasing the surface area not covered by cells in comparison with the effect observed for cells cultured under physiological condition (p < 0.05). In the presence of peptide 6i at 10-fold higher concentration, the surface area decreased by 10% and the increase of the migration capacity of cells was observed (Fig. 3A). The opposite and concentration-dependent effect was noticed in case of treatment of fibroblast cells. Peptide 6i in the lowest concentration of 0.001 μg mL⁻¹ significantly facilitated cell migration and reduced the surface area by 35% in comparison with the K⁺. In contrast, peptide 6i at 0.01 μg mL⁻¹ significantly reduced migration potential and showed similar effectiveness as DXM (Fig. 3B, p < 0.05). The images show the increase of the surface area in comparison with the K⁺ (p < 0.05), i.e. from 340 142.5 μm² to 508 382.6 μm² (data not shown).

Fig. 3 Analysis of cell migration capacity of human skin cells: HaCaT keratinocytes (A) and primary fibroblasts (B). K⁺ positive control in DMEM medium, K⁻ negative control in DMEM treated with DXM (1 μM), standard peptide (6i) and lipopeptides (1b, 2b, 4b) at 0.001 μg mL⁻¹ and 0.010 μg mL⁻¹. The cell migration capacity was measured by scratch assay as described in Section 2. Results are presented as the median and min–max of triple wells; * statistical difference in relation to control p < 0.05.

In this study, we also proved that lipophilic analogues of Gly-His-Lys are able to stimulate cell migration. Lipopeptide 1b at 0.01 μg mL⁻¹ not only increased the proliferation but also facilitated the migration of both skin cells. In the lowest concentration, it also possessed the potential of HaCaT cells migration, but conversely the inhibition effect of fibroblast cells migration was noticed. Noteworthy, lipopeptide 1b was selected for further studies based on determining the specific mechanisms of the interaction with skin cells. The results will be described further.

The effect induced by lipopeptides 2b and 4b in the migration assay exerted similar effect on skin cells. Both lipopeptides in the same concentration of 0.001 μg mL⁻¹ stimulated the migration of cells (p < 0.05). Although these lipopeptides at 10-fold higher concentration caused significantly inhibition of cells migration. However, only the effect of peptide 2b on fibroblasts and peptide 4b on HaCaT cells was statistically significant (Fig. 3B, p < 0.05).

Microscopic evaluation of skin cells was also performed. Noteworthy, the microscopic observation in the presented conditions for human skin cells of standard peptide has not been shown yet. Images of the fibroblast cells in the physiological conditions showed characteristic shape, indicating static cells with a large area of cytoplasm, probably fibrocytes. Fibroblast cells in the presence of peptide 6i in the lowest concentration 0.001 μg mL⁻¹ were much smaller and more numerous. It generalized changes in cell shape, cell surface and cytoplasm ultrastructure. Skin cells appeared as fusiform fibroblast-like cells and were in close proximity to each other. It has been shown that they were a population of rapidly dividing cells. This effect was not so evident in the presence of peptide 6i at 10-fold higher concentration. From microscopic image analysis, the mononuclear cells were large, branching and separated from one to another. Some of the spindle-shape fibroblasts changed their morphology to stellate cells. In comparison, the cells cultured in the physiological conditions were much larger. Most of the visible large and mononuclear cells resembled degenerating fibroblasts (Fig. 4A).

Fig. 4 Microscopic evaluation of human skin cells: primary fibroblasts (A) and HaCaT keratinocytes (B) in the presence of peptides (zoom ×40): (i) control – cells untreated with peptides, (ii) tested cells – cells in the presence of standard peptide (6i) in the lowest concentration of 0.001 μg mL⁻¹, (iii) tested cells – cells in the presence of standard peptide (6i) in the highest concentration of 0.010 μg mL⁻¹. Results are presented as microphotographs for cell growth with versus without peptide that were taken under an inverted microscope. Microscopic images show the beneficial effect of peptide 6i on the growth and condition of human skin cells, in particular at the lowest tested concentration.

Keratinocyte cells untreated with the peptides occurred in a small cluster, adhering tightly to each other. The cells did not display cuboidal cell-shape, typical of single epithelial cells and possessed small sized nucleus. Stimulation of cells by peptide 6i caused the increase of the undifferentiated cells population. The cells were nourished and possessed circular or oval shape with the revealed nucleus. At the lowest concentration of 0.001 μg mL⁻¹, they were deployed in a close proximity ensuring the cell interactions (Fig. 4B).

Microscopic observations confirmed the beneficial effect of all the lipopeptides on the growth and condition of human skin cells, in particular at the lowest tested concentration. Consistent with this, fibroblast cells cultured in the presence of all lipopeptides at 0.001 μg mL⁻¹ increased the population cells through stimulating division and proliferation of small, spindle-shape cells. Microscopic image showed adherent cells treated with peptides at 0.010 μg mL⁻¹. While increasing the concentration of the lipopeptides made the cells visible as large, remote, mononuclear cells displaying typical stellate morphologies and reflecting degenerative processes (Fig. 5A). The strongest effect was observed for lipopeptide 1b at 0.001 μg mL⁻¹. The increased cell density in the microscopic image indicated for the stimulation of proliferation process and could be allowed for tissue regeneration.

Fig. 5 Microscopic evaluation of human skin cells: primary fibroblasts cells (A) and HaCaT keratinocytes (B) in the presence of peptides (zoom ×40): (i) control – cells untreated with lipopeptides, (ii) tested cells – cells in the presence of lipopeptide 1b in the lowest concentration of 0.001 μg mL⁻¹, (iii) tested cells – cells in the presence of lipopeptide 1b in the highest concentration of 0.010 μg mL⁻¹, (iv) tested cells – cells in the presence of lipopeptide 2b in the lowest concentration of 0.001 μg mL⁻¹, (v) tested cells – cells in the presence of lipopeptide 2b in the highest concentration of 0.010 μg mL⁻¹, (vi) tested cells – cells in the presence of lipopeptide 4b in the lowest concentration of 0.001 μg mL⁻¹, (vii) tested cells – cells in the presence of lipopeptide 4b in the highest concentration of 0.010 μg mL⁻¹. Results are presented as microphotographs for cell growth with versus without lipopeptides that were taken under an inverted microscope. Microscopic images show the positive effect of lipopeptides on the condition and density of human skin cells, in particular at the lowest tested concentration.

The biological effects observed for lipopeptides 2b and 4b were closely comparable with lipopeptide 1b. The differences in size of cells also appeared in morphological evaluation. The cells were slightly larger. With the increasing the concentration of lipopeptides, the cells density decreased and the distance between them were significantly greater, that could have influenced on the decrease of cell interactions. The shape of cells was also less regular but more elongated and more similar to the cells untreated with lipopeptides, undergoing decreased aging process.

Keratinocytes untreated with peptides occurred in a small cluster, adhering tightly to each other. The cells did not display cuboidal cell-shape, typical of single epithelial cells and possessed small sized nucleus. Stimulation of cells by all tested peptides caused the increase of the undifferentiated keratinocyte population as well as for standard peptide. The cells were nourished and possessed circular or oval shape with the revealed nucleus. The close cell interactions were noticed by treated cells with lipopeptides at 0.001 μg mL⁻¹ (Fig. 5B).

3.2. Safety profile of peptides

In our study, the analysis of cytotoxicity of peptides (1b, 2b, 4b, 6i) was also presented. Both keratinocyte and fibroblast cells were exposed for 48 h to peptides at concentrations ranging 0.01 ÷ 10.00 μg mL⁻¹, as reflected by the lower absorbance in the colorimetric MTT test. All the lipopeptides at 0.01 μg mL⁻¹ showed a significant increase in viability of both skin cells (Fig. 6). Our data, especially antimicrobial activity described in previous article, safety profile and tolerance for skin suggest that may be applied not only on the non-injured but also injured skin.³⁷

Fig. 6 Analysis of cells viability of human skin cells: HaCaT keratinocytes (A) and primary fibroblasts (B) in the presence of peptides in whole range of concentrations of 0.010 μg mL⁻¹, 0.10 μg mL⁻¹, 1.00 μg mL⁻¹ and 10.00 μg mL⁻¹. The cytotoxic activity of peptide was evaluated using calorimetric MTT test as described in Section 2. Results are presented as the main values of triple wells as a percent of control. The obtained data show that the majority of tested lipopeptides (1b, 2b, 4b) do not exert any toxic effect on both human skin cells: HaCaT keratinocytes and primary fibroblasts at concentrations ranging from 0.010 μg mL⁻¹ to 1.00 μg mL⁻¹ in comparison with the standard peptide (6i) at 1.00 μg mL⁻¹ on primary fibroblasts. All the peptides present safe profile for the treatment of human skin cells.

3.3. Permeation studies

In order to investigate the behaviour of peptides in in vitro model of permeation studies, the permeation of standard peptide was evaluated and compared with that previously developed.^42,48 The total amount of peptide 6i in the acceptor cell after the experiment was determined as 27.06 μg mL⁻¹ (20.82%). The mass spectra confirmed the presence of the [M + H]⁺, [M + Na]⁺, [M + K]⁺ mass ions at m/z 341.2, 363.1, 379.5, respectively. It should be pointed out that it was not identified any other peak of peptide degradation in acceptor and donor solutions in the whole period of the experiment. Polyester film used in this studies appeared least permeable for unmodified peptides – peptides with free both N- and C-termini (1i, 4i, 6i). Although differences in permeability towards lipid membrane, based on polyester film of highly hydrophobic character, were fairly small. Numerous studies reported that the penetration of the hydrophilic drugs in ionic form is not effective through this type of membranes.^42,48 In this case, in our further experiments, it was replaced by the cellulose films (I and II type with different physical parameters). The resulting permeability and diffusion coefficients such as J_ss, t_L, K_p, D, and also Q_MA are shown in Table 1. A significant increase in permeability parameters was observed for peptides modified with fatty acids in comparison to unmodified ones. The K_p value for peptide 1i was calculated as 0.65 × 10⁻⁶ cm s⁻¹ and the value was 2.6-fold higher than in case of polyester film. The differences were also associated not only with the character of the films, but also with the pore size. Moreover, the differences in the use of cellulose film types I and II were not statistically significant. The polymeric membranes exhibited high compatibility with all tested peptides. All the peptides, with strong to moderate antimicrobial activity described in previous studies, presented the K_p ranges between (0.39 ± 0.23) × 10⁻⁶ cm s⁻¹ and (1.49 ± 0.16) × 10⁻⁶ cm s⁻¹, and the J_ss ranges between 0.10 ± 0.12 μg cm⁻² h⁻¹ and 1.16 ± 0.24 μg cm⁻² h⁻¹. For lipopeptide 1b, the K_p value increased to (1.28 ± 0.68) × 10⁻⁶ cm s⁻¹ and its amount in acceptor cell after 72 h was close to 69.54 μg cm⁻² (53.50%) (Fig. 7). In comparison, only 42.90 μg cm⁻² (13.76%) of peptide 1i was observed after the same period of experiment. Moreover, its concentration was detectable in the acceptor cells after 10 h of experiment. The K_p of lipopeptide 1c occurred at approximately the same time (Table 1).

Fig. 7 Diagrammatic illustration of a Franz-like diffusion cell and the behavior of the peptides (1b, 1i, 2b, 2e, 4b, 4i) in in vitro model of permeation studies. The solution of peptide in the concentration range of 0.02 ÷ 0.05% (w/v) was applied on the surface of membrane in the donor cell. An average peptide flux from the solution, total amount [%] of peptide in acceptor cell and membrane after 72 hours as well as permeability and diffusion coefficients were measured. Quantification of peptides in samples was performed by UV-Vis spectrophotometry and RP-HPLC methods. Samples were received from the acceptor cell at time intervals, every 30 min for 2 h and then every 1 h between 2 h and 10 h of experiments. In case of the second Franz-like diffusion cell, the experiment has continued by taking samples at 24 h, 48 h, 72 h. Each time after sampling the acceptor cell, it was filled with the same volume of fresh medium. Determination of peptides in a model membrane was performed by the procedure for isolation of the peptides from the model membrane after 10 h and 72 h in two ways: by direct and indirect methods. Results are presented as the main ± SD of 4 replicates.

Table 1 Summary of average peptide flux from the solution and total amount [%] of peptide in acceptor cell and membrane after 72 hours, permeability and diffusion coefficients^a

Peptide	J ss [μg cm^−2 h^−1]	t L [h]	K p × 10^6 [cm s^−1]	D × 10^6 [cm^2 h^−1]	Q MA [%]
a MA – the total amount of the peptide in acceptor cell and membrane, * – buffer solution, ** – methanol solution.
1b*	0.92 ± 0.49	2.78 ± 0.68	1.28 ± 0.68	3.65 ± 2.22	62.49 ± 3.54
1b**	0.65 ± 0.16	2.92 ± 0.34	0.90 ± 0.23	3.47 ± 0.74	—
2b*	0.92 ± 0.12	3.27 ± 0.01	1.27 ± 0.16	3.09 ± 0.50	65.50 ± 9.65
3b*	0.69 ± 0.12	2.37 ± 0.12	0.96 ± 0.12	4.26 ± 0.12	67.50 ± 10.37
4b*	0.75 ± 0.13	2.69 ± 0.12	1.04 ± 0.19	3.76 ± 0.57	66.27 ± 4.54
4b**	0.85 ± 0.19	3.24 ± 0.27	1.19 ± 0.27	3.12 ± 0.81	—
1i*	1.16 ± 0.24	9.38 ± 0.10	0.65 ± 0.14	1.08 ± 2.1	20.51 ± 4.60
1a*	0.49 ± 0.12	9.01 ± 0.29	0.68 ± 0.16	1.12 ± 0.97	40.01 ± 2.82
4i*	0.10 ± 0.12	9.12 ± 0.29	0.14 ± 0.16	1.11 ± 0.97	13.08 ± 1.87
1c*	0.91 ± 0.36	2.69 ± 0.59	1.26 ± 0.50	3.77 ± 0.53	69.99 ± 5.30
1c**	0.89 ± 0.26	3.02 ± 0.59	1.23 ± 0.35	3.35 ± 0.37	—
6i	0.22 ± 0.11	9.70 ± 0.18	0.31 ± 0.16	1.05 ± 0.96	—
6c	0.81 ± 0.26	2.73 ± 0.26	1.13 ± 0.35	3.71 ± 0.35	64.80 ± 6.75
1d*	1.07 ± 0.12	2.26 ± 0.29	1.49 ± 0.16	4.49 ± 0.97	62.98 ± 5.27
5e*	0.89 ± 0.12	1.83 ± 0.29	1.24 ± 0.16	5.54 ± 0.97	71.49 ± 3.88
2e*	0.99 ± 0.18	2.34 ± 0.10	1.37 ± 0.24	4.33 ± 0.69	67.03 ± 6.88
1e*	0.29 ± 0.17	1.94 ± 6.14	0.39 ± 0.23	5.23 ± 0.04	61.00 ± 3.21
1e**	0.73 ± 0.15	2.06 ± 1.14	1.02 ± 0.21	4.92 ± 0.04	—

---

The important differences in permeation profile for both lipopeptides 1b and 1c were observed in the first 10 h of experiment. These appear to be due to the improved affinity to the lipid membrane and the amount of lipopeptide release. The cumulative amount of lipopeptide 1c release was 2-fold higher than lipopeptide 1b in this period (41.10 μg cm⁻², 31.62%). However, the J_ss was observed after 24 h of experiment. Lipopeptide 6c presented a delayed release from the model membrane. The cumulative amount of the lipopeptide 6c release increased significantly within 10 h. After this period the amount of lipopeptide 6c in the acceptor cell has remained constant.

To summarize, a significant increase in permeability parameters such as steady-state flux and permeability coefficient was observed for lipopeptides in comparison to unmodified peptides, including the standard peptide. This is a reason of modification with different length of fatty acid, such as lauric, palmitic and stearic acid. Most of the lipophilic analogues possessed higher affinity to biological membrane than unmodified analogues, which was correlated with their lipophilicity. Moreover, the lipophilicity of peptide is also correlated with its antimicrobial activity.^37,44 The bacterial and fungal membrane binding and insertion of both unmodified and modified peptides was preliminary tested.³⁷ Current analysis of the relationship between lipophilicity and penetration process through the biological membrane will be widely discussed. Additionally, such work is aided by high profile of studies on skin cells. All performed in vitro studies confirmed high quality and great biological properties of the peptides.

4. Conclusion

We have described in vitro model of wound repair using proliferative cell staining and tracking of living cells. All the tested lipopeptides provided an optimal environment for the growth, development and viability of skin cells. Microscopic changes in morphology of fibroblasts and keratinocytes confirmed their positive effect on growth and condition of cells. The evaluation of skin cell behaviour treated by novel lipophilic analogues of Gly-His-Lys was also helped to select lipopeptides (1b, 4b) for further studies based on determining the specific mechanisms of the interaction with skin cells. All the tested peptides seemed to possess a high potential as topical therapeutics. Their antimicrobial activity, safety profile and great tolerance for human skin suggest that may be applied not only in skin care but also in wound therapy.

Declaration of interest

The authors report no conflicts of interest. Author contribution: MK, MKK wrote the manuscript, MK, MKK, MP, AS analyzed data, MK, MP designed experiments, KD, PT conceived and supervised the study, MK, AS performed experiments. MK and MP contributed equally to this work.

Acknowledgements

This work was supported by the Gdansk University of Technology (Grant DS No. 030386 and Grant DS No. 030861). Authors would like to thank the company Lipoid GmbH for supplying Cerasome 9005®. Authors also would like to thank Paulina Langa (Ph.D. student at the Department of Clinical Immunology and Transplantology, Faculty of Medicine, Medical University of Gdansk) for supporting during performing biological experiments.

References

1. G. Kaya and J. H. Saurat, Dermatologia, 2007, 215, 284–294.
2. J. H. Saurat, Dermatologia, 2007, 215, 271–272.
3. M. Pawlaczyk, M. Jelonkiewicz and A. Witkowska, Geriatria, 2011, 5, 303–308.
4. P. E. Bermann, Nurs. Clin. North Am., 2007, 42, 485–500.
5. M. Farage, K. W. Miller, E. Berardesca and H. I. Maibach, Am. J. Clin. Dermatol., 2009, 10, 75–86.
6. F. Masaki, K. Riko, H. Seiji, Y. Shuhei and A. Yakabe, Wounds, 2007, 19, 13–19.
7. G. Kaya, Rev. Med. Suisse, 2008, 30, 1078–1082.
8. J. J. An, W. S. Eum, H. S. Kwon, J. S. Koh, S. Y. Lee, J. H. Baek, Y. J. Cho, D. W. Kim, K. H. Han, J. Park, S. H. Jang and S. Y. Choi, J. Cosmet. Dermatol., 2013, 12, 287–295.
9. L. B. Lopes, E. J. Furnish, P. Komalavilas, C. R. Flynn, P. Ashby, A. Hansen, D. P. Ly, G. P. Yang, M. T. Longaker, A. Panitch and C. M. Brophy, J. Invest. Dermatol., 2009, 129, 590–598.
10. M. Limbowski and A. Zieliński, Pol. Med. Paliatywna, 2003, 2, 283.
11. E. Gospodarek, Pol. Med. Paliatywna, 2003, 2, 283.
12. S. Dabrowiecki, Pol. Med. Paliatywna, 2003, 2, 290–291.
13. J. Stefańska, Mikrobiologia Medycyna, 2000, 22, 17–24.
14. G. W. Denton, Chlorhexidine, in Disinfection, Sterilization, and Preservation, ed. S. S. Block, Lippincott Williams & Wilkins, Philadelphia, Pa, 5th edn, 2001, pp. 321–336.
15. M. Bartoszewicz, A. Junka, D. Amutnicka, B. Mączyńska, A. Janczura, J. Nowicka and A. Secewicz, Leczenie Ran, 2011, 8, 21–27.
16. M. A. Zych, E. B. Górska, U. Jankiewicz and P. Kowalczyk, New Medicine, 2013, 1, 31–34.
17. J. Garau, E. Bouza, J. Chastre, F. Gudiol and S. Harbarth, Clin. Microbiol. Infect., 2009, 15, 125–136.
18. J. R. Lentini, M. Narita and V. L. Yu, Eur. J. Clin. Microbiol. Infect. Dis., 2008, 27, 3–15.
19. A. N. Neely, J. Gardner, P. Durkee, G. D. Warden, D. G. Greenhalgh, J. J. Gallagher, D. N. Herndon, R. G. Tompkins and R. J. Kagan, J. Burn Care Res., 2009, 30, 19–29.
20. I. Karpiuk and S. Tyski, Przegl. Epidemiol., 2012, 66, 567–573.
21. S. R. Norrby, C. E. Nord and R. Finch, Lancet Infect. Dis., 2005, 5, 115–119.
22. M. J. Thornton, Derm.–Endocrinol., 2013, 5, 264–270.
23. R. A. Baxter, Journal of Cosmetic Dermatology, 2008, 7, 2–7.
24. L. Rittie, G. J. Fischer and C. E. M. Griffiths, Anti-aging effects of retinoids and mechanism of action, in Retinoids and Carotenoids in Dermatology, ed. A. Vahlquist and M. Duvic, CRC Press, 2007, pp. 77–94.
25. M. H. Gold and M. P. Goldman, J. Drugs Dermatol., 2007, 6, 1018–1023.
26. K. P. Mukherjee, N. Maity, K. N. Nema and B. K. Sarkar, Phytomedicine, 2011, 19, 64–73.
27. M. P. Lupo and A. L. Cole, Dermatol. Ther., 2007, 20, 343–349.
28. F. X. Maquart, A. Simeon, S. Pasco and J. C. Monboisse, French J. Soc. Biol., 1999, 193, 423–428.
29. A. Simeon, Y. Wegrowski, Y. Bontemps and F. X. Maquart, J. Invest. Dermatol., 2000, 115(6), 962–968.
30. A. Simeon, H. Emonard, W. Hornebeck and F. X. Maquart, Life Sci., 2000, 67, 2257–2265.
31. L. Pickart, D. Downey, S. Lovejoy and B. Weinstein, Gly-L-His-L-Lys: copper(II) – a human plasma factor with superoxide dismutase-like and wound-healing properties, in Superoxide and Superoxide Dismutase, ed. G. Rotilio, Elsevier, 1986, pp. 555–558.
32. A. A. Abdulghami, A. Sherr, S. Shirin, G. Solodkina, E. M. Tapia, B. Wolf and A. B. Gottlieb, Dis. Manag. Clin. Outcome, 1998, 1, 136–141.
33. G. D. Mulder, L. M. Patt, L. Sanders, J. Rosenstock, M. I. Altman, M. E. Hanley and G. W. Duncan, Enhanced healing of ulcers in patients with diabetes by topical treatment with glycyl-L-histidyl-L-lysine copper, Wound. Repair. Regen., 1994, 2, 259–269.
34. F. X. Maquart, G. Belfon, B. Chaqour, J. Wegrowski, M. L. Patt, R. E. Trachy, J. C. Monboisse, F. Chastang, P. Birembant, P. Gillery and J. P. Borel, J. Clin. Invest., 1993, 92, 2368–2376.
35. J. B. Bishop, L. G. Phillips, T. A. Mustoe, A. J. Vander Zee, L. Wiersema, D. E. Roach, J. P. Heggers Jr, D. P. Hill, E. L. Taylor and M. C. Robson, J. Vasc. Surg., 1992, 16, 251–257.
36. L. Pickart and S. Lovejoy, Methods Enzymol., 1987, 147, 314–328.
37. M. Kukowska, M. Kukowska-Kaszuba and K. Dzierzbicka, Bioorg. Med. Chem. Lett., 2015, 25, 542–546.
38. P. Boukamp, R. T. Dzarlieva-Petrusevska, D. Breitkreuz, J. Hornung, A. Markham and N. E. Fusenig, J. Cell Biol., 1988, 106, 761–771.
39. W. Barańska-Rybak, M. Pikula, M. Dawgul, W. Kamysz, P. Trzonkowski and J. Roszkiewicz, Acta Pol. Pharm., 2013, 70, 795–801.
40. T. J. Franz, Curr. Probl. Dermatol., 1978, 7, 58–68.
41. K. Matsuzaki, T. Imaoka, M. Asano and K. Miyajima, Chem. Pharm. Bull., 1993, 4, 575–579.
42. L. Mazurowska and M. Mojski, Talanta, 2007, 72, 650–654.
43. M. Kukowska, M. Kukowska-Kaszuba and K. Dzierzbicka, Tetrahedron Lett., 2015, 56, 525–528.
44. M. Kukowska and K. Dzierzbicka, Curr. Med. Chem., 2014, 21, 1505–1521.
45. A. Simenon, Y. Wegrowski, Y. Bontemps and F. X. Maquart, J. Invest. Dermatol., 2000, 115, 962–968.
46. A. Simeon, H. Emonard, W. Hornebeck and F. X. Maquart, Life Sci., 2000, 67, 2257–2265.
47. Y.-A. Kang, H.-R. Choi, J.-I. Na, C.-H. Huh, M.-J. Kim, S.-W. Youn, K. H. Kim and K. C. Park, Arch. Dermatol. Res., 2009, 301, 301–306.
48. J. J. Hostynek, F. Dreher and H. I. Maibach, Inflammation Res., 2010, 59, 983–988.

---

Footnote

† These authors contributed equally to this work.

---