Introduction

Peptides are small chain of amino acids. They are used in medicinal (Karahan 2021), cosmetic and dietary products due to the their unique properties (Park et al. 2021; Türkoǧlu and Erdem 2010).

Natural growth-modulating peptide (GHK) is a tripeptide discovered by Pickart et al. (Pickart et al. 1980) has many biological actions on the body. Some of the biological actions of GHK has including gastric ulcer and duodenal ulcer repair and influence the improvement of bone tissue, systemic wound healing initiation, dermal wound healing, skin remodeling, and hair follicle expansion. It also leads to an increase of anti-oxidant and anti-inflammatory activities, the root secretion of human keratinocytes, and decorin, angiogenesis, and nerve outgrowth (Choi et al. 2012; Kang et al. 2009; Pickart 2008, 2009; Pickart and Margolina 2012; Pickart et al. 2012). Age is an important factor in the achievement and capacity of these biological actions. The level of GHK in plasma is about 200 ng/mL (10−7 M) at age 20 but declines to 80 ng/mL by age 60. This decline in the GHK-level coincides with a noticeable decrease in the regenerative capacity of an organism (Pickart et al. 1980).

GHK has been responsible for gene expression changes for certain cancers and chronic obstructive pulmonary disease (Pickart et al. 2017). GHK effects on cancer are investigated in lots of research article. One of these studies was about the metastatic human colon cancer was investigated by Broad Institute's. In studies of metastatic human colon cancer, the Broad Institute's Connectivity Map showed that GHK among the 1309 bioactive molecules studied reversed the expression in 70% of 54 over-expressed genes (Pickart et al. 2017). In addition, GHK has been found to reactivate apoptosis different human cancer lines in cell culture.

Cancer is the most distressing disease that enforces a severe death toll worldwide. Chemotherapy is a common option for cancer treatment, but that is often limited because of its unwanted toxic effects on normal tissues. Because most anti-cancer agents are not distributed in the target tumor-bearing tissues, which results in their reduced therapeutic efficacy (Bilensoy et al. 2009; Guo et al. 2011). Over the past three decades, polymeric nanoparticles have a preferred nanoscale drug delivery vehicle to treat cancer because of their excellent endocytosis efficiency, passive tumor-targeting ability, high encapsulation efficiency, and ability to deliver a wide range of therapeutic agents (Brannon-Peppas and Blanchette 2004; Brigger et al. 2012; Budama-Kilinc et al. 2017; Çalman et al. 2019; Peer et al. 2007). PCL has been approved by the Food and Drug Administration for various biomedical applications as a non-toxic degradation product. Although there are numerous studies on the synthesis of PCL micro/nanoparticles for drug delivery applications (Paulo et al. 2011; Shuai et al. 2004), PCL is an extremely hydrophobic polymer that cannot form nanoparticles spontaneously in the absence of any emulsifier or stabilizer (Chawla and Amiji 2002). Therefore, coating/blending with large molecules such as polysaccharides or macromolecules is necessary to prevent aggregation and provide better stabilization to nanoparticles.

The docking method is a powerful tool and essential technique to describe protein–ligand and protein–protein interactions and the behavior of nanomaterials (Yuriev et al. 2011). This method can model atomic-level interactions between ligands and receptors (protein, macromolecule, DNA, RNA, etc.), which allows us to explain the basic biochemical processes by characterizing the behavior of small molecules in the binding site of target proteins (Taylor et al. 2002). The determination of the inhibition of the VEGF signaling pathway has become an important review topic for cancer treatment in recent years (Hicklin and Ellis 2005).

The aim of this study was to develop GHK-loaded PCL nanoparticles with appropriate surface modifications and well-defined particle sizes that can improve the cell uptake of GHK. This was the first time that GHK-loaded PCL nanoparticles were prepared by the double emission precipitation method. Meanwhile, the physicochemical characteristics and in vitro anti-cancer efficiency of nanoparticles were investigated. Furthermore, the anticancer activity of GHK and GHK-loaded PCL nanoparticles were studied by cellular metabolic activity assay (XTT). Intermolecular bonds that have a great influence on activity and selectivity were investigated with the help of molecular docking calculations to understand the most likely poses with which GHK binds to protein.

Experimental Details

Materials

PCL (Mn = 80,000) and polyvinyl alcohol (PVA) (Mw = 31.000–50.000, 87–89%) were purchased from Sigma–Aldrich (USA). GHK (Mw = 400.43) was purchased from Active Peptide Co. Ltd. (USA). Dichloromethane was also purchased from Merck Millipore (> 99.5%) (Darmstadt, Germany). All other chemicals used in this paper were analytical grade. Ultrapure water from Millipore Milli-Q Gradient System was used to prepare the solutions.

Synthesis of GHK Encapsulated PCL Nanoparticles

GHK-loaded PCL nanoparticles were synthesized by the double emulsion precipitation method (Lamprecht et al. 2000; Yang et al. 2001). Briefly, 1.5 mg of GHK was dissolved in 2 mL distilled water, 30 mg of PCL was dissolved in 2 mL dichloromethane (DCM), and the PCL solution was added to the GHK solution. They were mixed and sonicated for four minutes with an ultrasonicator (Bandelin, Sonopuls) at 55 W. After that, this solution was pulled with a syringe and then added dropwise into 4 mL of 5% PVA solution under continuous stirring. After that, the solution was sonicated again for four minutes. The mixture was left overnight under continuous stirring. Finally, the obtained GHK-loaded PCL nanoparticles were centrifuged at 10,000 rpm for 30 min (Hettich, Germany) then the supernatant was discarded; this step was repeated three times and re-suspended with 10 mL of distilled water solution for further characterization and applications. The optimum particle size and size distribution of nanoparticles was obtained by using different amounts and concentrations of PVA (1%; 2%; 3%; 4%; 5%), PCL (30; 60; 90; 120; 150 mg), and GHK (0.5; 1; 1.5; 2; 3 mg).

Average Particle Size, Polydispersity Index, Particle Size Distribution, and Zeta Potential Measurement

A Zeta-sizer Nano ZS (Malvern Instruments, UK) instrument equipped with a 4.0 mV He–Ne laser (633 nm) was used to examine the properties of the average size, polydispersity index (PdI), and size distribution of nanoparticles. Every electrophoretic light scattering measurement was performed at 25 °C and each sample was prepared with phosphate-buffered saline (PBS) before filtering with 0.2 μm regenerated cellulose membrane (Sartorius, Germany).

Transmission Electron Microscopy (TEM) Observation

The morphology of GHK-loaded PCL nanoparticles was observed via Transmission Electron Microscopy (TEM) (JEOL JEM 1400 Plus). Briefly, each sample was diluted with distilled water, and placed directly into an ultrasound bath at room temperature for two minutes, then two or three drops of the sample were deposited on a Formvar-coated carbon-supported copper grid, and TEM images were observed at 120 kV.

FT-IR and Raman Analysis

FT-IR and Raman spectral analysis were implemented to investigate the possible chemical interactions between GHK and PCL polymer. The IR analyses of the samples were carried out by using both IR transmittance and reflectance (ATR) techniques. The FT-IR spectra for GHK peptide and GHK-loaded PCL nanoparticles were obtained using a Jasco 6300 FT-IR spectrometer in the range 4000–400 cm−1 with 2 cm−1 resolution. The Jasco NRS-3100 micro-Raman spectrometer was used to record the Raman spectrum of the GHK peptide and GHK-loaded PCL nanoparticles. The Raman system, which was calibrated with a silicon semiconductor using the Raman peak at 520 cm−1, has three gratings (600, 1200, and 1800 lines/mm) and two laser sources (532 nm green laser and 785 nm red laser), three Olympus microscope lenses (20 × , 50 × , 100 ×), and a high-sensitivity cooled CCD detector. The GRAMS/AI 7.02 (Thermo Electron Corporation) software package was used to acquire the baseline adjustment, smoothing, second derivative and band fitting procedures. Using the Gaussian function, band fitting was performed until the reproducible and converged results were obtained from the squared correlations better than r2 ~ 0.9999. The 2nd derivative function that was obtained using the Savitzky–Golay function (two polynomial degrees, five points) was used to determine the band positions and band widths for band fitting.

Encapsulation and Loading Efficiency of the Nanoparticles

First, the standard curve of GHK was obtained using a UV–Vis spectrometer (Shimadzu, Japan) at 220 nm. The encapsulation and loading efficiency of GHK-loaded PCL nanoparticles were measured by separating the nanoparticles from the aqueous nanoparticle suspension via centrifugation (Govender et al. 1999). The concentration of free GHK in the supernatant was determined by a UV–Visible Spectrometer via the GHK standard calibration curve.

The encapsulation and loading efficiency of GHK were respectively calculated by the following formulae:

$${\text{Encapsulation Efficiency}} = \frac{{{\text{Total GHK amount}} - {\text{Free GHK amount}}}}{{\text{Total GHK amount}}}{\text{x}}100$$
(1)
$${\text{Loading Efficiency}} = \frac{{\text{Encapsulated GHK amount}}}{{\text{Total nanoparticles weight}}}{\text{x}}10$$
(2)

In vitro Release Study

The in vitro GHK release profiles were determined from PCL nanoparticles using GHK-loaded PCL nanoparticles that were obtained from one set of preparations and re-dispersed in 10 mL of distilled water at a final concentration of 150 μg/mL. A time-dependent release study was carried out at time intervals of 0; 0.5; 1; 2; 3; 6; 7; 9; 10; 24; 72; 96; and 240 h. Samples were incubated at 37 °C under gentle agitation. At proper time intervals, the intake amounts of GHK in loaded nanoparticles were first extracted in PBS and then quantified spectrophotometrically. The concentration of GHK was determined via the GHK standard calibration curve. The release of GHK was determined as per the following equation:

$${\text{Release }}\left( {\text{\% }} \right) = \frac{{\text{Released GHK}}}{{\text{Total GHK}}}{\text{x}}100$$
(3)

In vitro Cytotoxicity of GHK, PCL Nanoparticles, and GHK-Loaded PCL Nanoparticles

A mouse fibroblast cell line (L929) was used to determine the cytotoxicity of GHK, PCL nanoparticles, and GHK-loaded PCL nanoparticles. L929 cells were cultured in a DMEM-F12 medium (including 10% Fetal Bovine Serum and Penicillium-Streptomycin) at 37 °C (5% CO2) and proliferated cells were trypsinized. Cells were seeded in sterile 96-well flat bottom microplates with 104 cells in each well. All samples were added to cultured cells at different concentrations (0–100 µg GHK in 1 mL) and incubated for 24 h at 37 °C (5% CO2). After that, the medium was replaced with 100 μL of 2,3-Bis-(2-Methoxy-4-Nitro-5-Sulfophenyl)-2H-Tetrazolium-5-Carboxanilide (XTT) solution in DMEM (0.5 mg/mL concentration with 7.5 μg/mL Phenazine methosulfate). Plates were incubated for four hours at 37 °C and the optical density was measured at 450 nm with a multiplate reader (Thermo LabsystemsMultiskan Ascent 354 Microplate Photometer).

In vitro Anticancer Efficiency of GHK

Human U87 glioblastoma cells were used to determine the anticancer efficacy of GHK tripeptide. L929 cells were used for comparison; U87 cells were cultured in DMEM-F12 medium (including 10% Fetal Bovine Serum and Penicillium-Streptomycin) at 37 °C (5% CO2) and proliferated cells were trypsinized. GHK was added at different concentrations (0–40 mg/mL) on both the U87 cells and L929 cells and incubated for 24 h (Kokcu et al. 2020). XTT was applied as mentioned above and the optical density was measured.

Molecular Dynamic (MD) Simulation

Molecular dynamic simulation of Gly-His-Lys tripeptide with PCL and PVA polymers was performed with GROMACS 2019.1 program (Van Der Spoel et al. 2005) using a AMBER-99SB- ILDN force field (Hornak et al. 2006) for 10 ns and TIP4P water models (Abascal and Vega 2005). For the molecular dynamic study, Gly-His-Lys was prepared with the help of Gaussian software (Frisch et al. 2009) by using the results of optimization geometry that were calculated. PCL and PVA polymers were prepared for molecular dynamic study using Gaussian software program (Frisch et al. 2016), AmberTools18 (Case et al. 2018) and ACPYPE tool (Da Silva and Vranken 2012). 1 GHK molecule, 10 PCL and 4 PVA polymers were placed in a box (10x10x10).

The steepest descent integrator was initially performed for energy-minimization. The whole system was heated until the temperature and pressure were fixed using the NVT (for 100 ps) and NPT (for 100 ps) ensemble. Temperature coupling was implemented with a V-rescale (velocity rescaling) thermostat (Bussi et al. 2007) at 310 K while pressure coupling was carried out with an isotropic Parinello–Rahman barostat (Parrinello and Rahman 1981) at 1 bar with a coupling time constant of 0.1 ps and 2.0 ps with 4.5 × 10−5 compressibility, respectively. The system configurations (atom coordinates, velocities, and energies) were saved at every step. XMGRACE (Turner 2005) and VMD (Humphrey et al. 1996) softwares were used to view the energy graphs and system box.

Molecular Docking and ADME Properties

The docking calculations were achieved using the Glide SP (standard precision) module of the Maestro version 11.4 in the docking calculations were achieved using the Glide SP (standard precision) module of the Maestro version 11.4 in the Schrodinger Software program (Schrödinger Release 2017–4: Maestro, Schrödinger, LLC, New York, NY, 2017) (Friesner et al. 2004; Halgren et al. 2004). First, the most appropriate conformation of Gly-His-Lys (GHK) tripeptide obtained from the result of the molecular dynamic calculation was used to prepare for docking calculations by the LigPrep tool in the Maestro 11.4 version of the Schrödinger Software program using the OPLS force field (Harder et al. 2016). Thirty-two possible stereoisomers were produced for the ligand after selecting the ionization states at pH 7.0 ± 2.0. For protein preparation, the structure of Vascular Endothelial Growth Factor Receptor 2 (VEGFR2) kinase domain in a complex with a benzimidazole inhibitor (PDB code: 2QU5) (Potashman et al. 2007) was downloaded from the protein data bank. The SWISS-MODEL server was used to achieve a better protein homology model (Bienert et al. 2017). The receptor used as the target for molecular docking calculations was prepared using the Protein Preparation Wizard tool (Sastry et al. 2013). In the process of preparing the protein for docking, all water was removed, polar hydrogen was added, bond orders were assigned, charges were defined using PROPKA (Søndergaard et al. 2011) at pH 7.0, and receptor was optimized. Energy minimization was implemented by preferring 0.3 Å RMSD and the OPLS3 force field to converge heavy atoms. A receptor grid generation tool was utilized to generate a grid in which ligand protein binding poses were possible throughout the docking process in the active site. By defining routable groups of an active site, a cubic box was formed that was centered on the centroid of the ligand with specific dimensions and a ligand docking tool was employed to achieve ligand-receptor docking. Absorption, distribution, metabolism, and excretion (ADME) properties were also calculated using the Qik-Prop module of the Schrodinger software to specify the physicochemical and biological functional properties such as the molecular weight (MW), percent human oral absorption, predicted octanol/water partition coefficient (QPlogPo/w), polar surface area (PSA), and number of violations of Lipinski’s rule of five (Lipinski et al. 1997), which is important for generating an effective drug in new drug development.

Statistical Analysis

Averages of the data obtained from the experimental and control groups were compared using the SPSS (version 24.0, SPSS Science, Chicago, IL) program. First, it was tested whether the obtained data showed normal distribution. Analysis of variance of the mean of the group with normal distribution was compared with a parametric test ANOVA. Homogeneous variances were compared with the Tukey HSD test. P < 0.05 was accepted as significant in all statistical evaluations.

Results and Discussion

Optimization studies were performed for blank PCL nanoparticles using different combinations of amounts of PVA (1%; 2%; 3%; 4%; 5%) and PCL (30; 60; 90; 120; 150 mg). The details of component amounts were given in Table S1. The optimum particle properties such as the average particle size, polydispersity index, particle size distribution, and zeta potential measurement were determined using a Zeta Sizer. The results of these experiments determined four PCL–PVA combinations: 1%–150 mg, 3%–150 mg, 4%–30 mg, and 5%–30 mg; these were determined to possibly be optimum blank PCL nanoparticles. Finally, these four combinations were used to assess the GHK-loaded nanoparticles using different amounts of GHK (0.5, 1, 1.5, 2, and 3 mg). All Zeta-Sizer results of nanoparticle optimization experiments were given in Table S2. Consequently, the component ratios of the optimum blank PCL were determined as 150 mg PCL for 1% or 3% PVA and 30 mg PCL for 4% or 5% PVA. Experiments were continued to prepare the GHK-loaded PCL optimum nanoparticles using five different amounts of GHK (0.5, 1, 1.5, 2, and 3 mg). In the direction of data obtained from optimization experiments, further studies were continued by selecting the 30 mg PCL–5% PVA–1.5 mg GHK combination as the optimum nanoparticle.

Figure 1a-b give the results of the Zeta-Sizer analysis of blank PCL nanoparticles. According to the results, in the particle size distribution by volume graphic, as seen in Fig. 1a, PCL nanoparticles had a narrow size distribution with 0.035 polydispersity index and 224.4 ± 3.87 nm average particle size. In the zeta potential graph of PCL nanoparticles as seen in Fig. 1b, the zeta potential value of PCL nanoparticles was obtained as − 9.79 ± 0.56 mV. Shown in the Fig. 1c, the GHK-loaded PCL nanoparticles also have a narrow size distribution with 0.029 polydispersity index and 232.5 ± 0.72 nm average particle size. In the Zeta potential graph of GHK-loaded PCL nanoparticles as seen from (Fig. 1d), GHK-loaded PCL nanoparticles were obtained with − 10.8 ± 0.64 mV zeta potential value.

Fig. 1
figure 1

Size distribution (a) and zeta potential (b) graphics of the PCL nanoparticles. Size distribution (c) and zeta potential (d) graphics of the GHK loaded PCL nanoparticle

Full size image

When compared our results with the literature values, it was observed that the average particle size, PdI and zeta potential values we obtained were appropriate (Kumar 2019; Öztürk et al. 2017; Tinca et al. 2020). For example, in the PCL nanoparticle studies conducted by Tinca et al., it was reported that the average particle size values varied from 210.4 ± 2.3 nm to 353.7 ± 5.6 nm, and the zeta potential varied from -6.3 ± 0.2 mV to -33.6 ± 0.5 mV. In addition, it was reported that PdI values varied from 0.033 ± 0.023 to 0.134 ± 0.026 in the same study (Tinca et al. 2020). In noscapine loaded PCL study, size range was determined from 150 to 818 nm and zeta potential range was determined from -8.8 to -16.6 mV (Kumar 2019). When we compare the blank nanoparticles with a study conducted in 2018 (Abamor 2018), the size determined as 220 nm in the literature was determined at 224 nm, which is a value very close to the value in the literature in our study. When we compare the PdI values and the zeta potential value with same literature, it was determined that our PdI values are quite low, and the zeta potential is higher compared to the literature. In the study conducted by Kolluru et al., The sizes of the blank PCL nanoparticles were lower than the value we obtained, but the PdI and zeta potential values are high. When looking at the loaded PCL nanoparticles, it was seen that our nanoparticle size is larger and the PdI and zeta potential values are lower (Kolluru et al. 2020).

Loading and Encapsulation Efficiency of GHK

As shown in Fig. 2a, the standard curve of the GHK was prepared by UV–Vis analysis at 220 nm to calculate the encapsulation efficiency of the GHK-loaded PCL nanoparticles. The encapsulation efficiency of GHK was calculated via Eq. 1 as 82.3%. This indicated that the GHK-loaded PCL nanoparticles were obtained successfully. In a study conducted with PCL nanoparticles, it was found that the encapsulation efficiencies ranged from 38 to 79%. It was determined that the percentage value we obtained in our study was higher than that obtained in the literature (Kumar 2019). In the another study, encapsulation effectiveness was determined as 64% (Abamor 2018). The amount of GHK in the total GHK-loaded PCL nanoparticle mass was calculated via Eq. 2. The loading efficiency of the GHK-loaded PCL nanoparticles was determined as 0.73%. When compared with the study conducted by Kolluru et al., it was observed that the encapsulation efficiency was compatible with the literature and the loading efficiency was slightly lower (Kolluru et al. 2020).

Fig. 2
figure 2

The standard curve of the GHK (a), the in vitro release graphic of the GHK (b)

Full size image

In vitro Release Profile of GHK

In vitro release profile of the GHK was determined by placing 1 mL of GHK-loaded PCL nanoparticles in dialysis capsules. The dialysis capsules were placed in 250 mL of pH 7.4 PBS and UV–Vis measurements were taken at 220 nm. The process was continued for 240 h, and it was calculated via Eq. 3. It was found that 78.86% of the GHK was delivered in of the process Fig. 2b. In the literature, PCL nanoparticles released about 58% quercetin at the end of 240 h (Abamor 2018). In this study, it was seen that the release percentage is higher in the same period. In the another literature, it was stated that the drug release occurs at a rate of approximately 28% at the end of 24 h. However, in this study, it was observed that the amount of drug released in 24 h was less than 10% (Lokhande et al. 2015).

TEM Results

The morphology of the GHK-loaded PCL nanoparticles was determined by TEM. According to the TEM images, the GHK-loaded PCL nanoparticles showed distinct spherical particles with a solid dense structure. As shown in Fig. 3, the TEM images indicated that the GHK-loaded PCL nanoparticles showed a uniform size distribution with mean size 100 ± 20 nm.

Fig. 3
figure 3

Transmission Electron Microscope image of the GHK loaded nanoparticles

Full size image

In another study, peptide loaded PLGA nanoparticles were investigated and the particle size was measured as 139.01 nm (Kokcu et al. 2020).

MD Results

The system was generated using ideal geometries of GHK and polymers into a cubic box (see Fig. 4) with water molecules of the TIP4P water model. After the 1820-step energy minimization with the steepest-descent algorithm, potential energy converged to a large negative number, -1.7269 × 106 kj/mol in Fig. S1. The temperature (310 K) and pressure (1 bar) were stabilized with the NVT and NPT ensembles, respectively. Finally, MD simulation was performed to equilibrate the system for a 10 ns total of 5,000,000 steps with 2 fs time step. Average potential, kinetic and total energies of system were obtained and shown in Fig. S2. The Root Mean Square Deviations (RMSD) value of GHK in the system was calculated after the MD simulation. RMSD shows how much the structure has changed over the course of the simulation. Changes < 2 Å are acceptable, which mean that the peptide does not tend to deviate from its first form. The backbone RMSD graph for GHK in the system demonstrated that the peptide remained very stable and do not exceed 0.11 nm (see Fig. S3). Kokcu et al. showed that GHK has stable form in water and methanol environments by looking at RMSD changes (Kokcu et al. 2020). Compared with literature,in this study, it was observed that GHK in the system including water and polymer molecules stay in stable form by looking at RMSD changes. MD study of GHK tripeptide with the polymers used in the nanoparticle study was first in the literature.

Fig. 4
figure 4

The cubic box including GHK tripeptide, PCL and PVA polymers

Full size image

Molecular Docking Results

The energy of the docking score obtained for GHK in the active side of the protein was -6.372 kcal/mol, as shown in Table S3 and Fig. 5a and Fig. S4. The binding pocket of the VEGFR2 protein for GHK contains a hydrophobic region as indicated by the green line, a polar region represented by the blue line, and hydrogen bonds that are effective at stabilizing the binding poses indicated by the purple line were picturized in the 2D ligand interaction graph in Fig. 5b. The GHK docked into the active side was bound with strong hydrogen bonds by the residues ASN-109, LYS-106, CYS-105, THR-102, and LEU-26. The ASN-109 (1.58 Å, 2.21 Å) and LEU-26 (2.36 Å) residues were strongly hydrogen bonded to the backbone of GHK, while the LYS-106 (1.76 Å), CYS-105 (2.17 Å), and THR-102 (1.93 Å) residues respectively made strong hydrogen bonds with the histidine, carboxyl group, and lysine moieties to obtain a stable conformation. The binding interactions of GHK with ASN-109, LYS-106, CYS-105, THR-102, and LEU-26 residues were as shown in Fig. 5c, d, e.

Fig. 5
figure 5

The docked pose of GHK to VEGFR2 protein (a), the 2D ligand interaction of GHK with VEGFR2 protein (b). The binding interaction of GHK with amino acid residue ASN-109, LEU-26 and CYS-105 (c), LYS-106 (d), CYS-105 and THR-102 (e)

Full size image

The electrostatic potential map surfaces of the GHK and VEGFR2 protein were also constituted to define the regions that were electron-rich and electron-poor. The lowest potential (electron-rich) regions were expressed in red; while those with the highest potential (electron poor) were shown in blue. Figure 6 also depicts the electrostatic potential map surfaces of ligand and receptors. The oxygen atoms in the carbonyl group with the red region interacted with the blue region with protein (high potential), while the hydrogen atoms in the histidine part tended to incline towards the protein’s red region. The pharmacokinetic parameters of drugs such as their permeability towards the blood–brain barrier, skin permeability, percentage oral absorption, and binding to human serum albumin were also calculated and tabulated using the QikProp application of the Maestro software package in Table S4.

Fig. 6
figure 6

The electrostatic potential map surfaces of the VEGFR2 protein and GHK tripeptide

Full size image

In summary, it was determined that the most stable binding poses and binding affinity energies of GHK were obtained as a result of the strong hydrogen bonds in the ASN-109, LYS-106, CYS-105, THR-102, and LEU-26 residues in the active region of the VEGFR2 protein. The pharmacokinetics such as the ADME properties of drug-candidate molecules are also an important point to consider during the drug development process at various points in biological systems. The permeability in the blood–brain barrier, skin permeability, percentage of oral absorption, binding to human serum albumin, and many more useful pharmacokinetic parameters provide substantial information to generate an effective drug in new drug development. The value of QPlogBB for the brain/blood partition coefficient parameter is -2.73, which indicates the drug’s ability to pass through the blood–brain barrier. Polymeric nanoparticles are promising drug delivery carriers for the central nervous system due to their encapsulating potential for drugs without excretion and metabolism within the body, and they do not cause any damage when delivered across the blood–brain barrier (Neha et al. 2013). Several in vivo studies have shown that 250 nm and smaller polymeric spherical nanoparticles can cross the blood–brain barrier (Parikh et al. 2010).

FT-IR and Raman Analysis Results

FT-IR, ATR, and Raman spectral analysis were performed to identify probable interactions between the GHK tripeptide and PCL polymer. Figure 7a, b indicate the FT-IR and ATR spectra of blank NPs, free GHK, and GHK-loaded NPs. The different characteristic bands that belong to different functional groups were also assigned and tabulated in Table 1.

Fig. 7
figure 7

The ATR spectrum of GHK, PCL and GHK loaded NPs (a), the FT-IR spectrum of GHK, PCL and GHK loaded NPs (b)

Full size image
Table 1 The ATR, FT-IR and Raman spectrum of GHK tripeptide and Peptide-loaded PCL NPs particles
Full size table

The FT-IR spectrum of free GHK observed absorption bands at 1682 cm−1, 1669 cm−1; 1646 cm−1, 1582 cm−1; 1515 cm−1, 1325 cm−1, 1307 cm−1, 669 cm−1, 652 cm−1 and 572 cm−1 corresponding to C = O ester stretching, C = O amide I, C-N–H amid II, C–O–C bending, Ring breathing (His), COOH bending, Ring def. (His), and COOH torsion, respectively. Although, this bands for the encapsulated nanoparticle was observed to be shifted by some amount for GHK-loaded NPs at 1670 cm−1, 1653 cm−1; 1646 cm−1, 1578 cm−1; 1521 cm−1, 1320 cm−1, 1295 cm−1, 710 cm−1, 670 cm−1 and 583 cm−1, respectively. Due to the interaction between polymer and GHK tripeptide, C = O amid I stretch vibrations was reduced from 1669 cm−1 to 1653 cm−1 and C-N–H amid II bending vibrations were shifted from 1515 cm−1 to 1521 cm−1. In addition, the carboxyl group stretching and bending and the ring bending, breathing, and deformation of the histidine group were also influenced by this interaction. The same shifts that indicate enhanced hydrogen bonding could also be observed in the ATR and Raman spectra in Table 1.

In addition, the peaks belonging to GHK, which remained under the PCL bands due to encapsulation, were also individually separated using band component analysis. The 1089 cm−1, 1050 cm−1 and 1352 cm−1 GHK peaks were detected using band component analysis for GHK-loaded PCL in the range 1120 -1000 cm−1 and 1440 -1340 cm−1 regions for FT-IR-ATR spectra in Fig. S5a-b. For FT-IR, peaks were found at 1307, 1184, 1161, and 1143 cm−1 with the aid of the same method by exploring the 1295 cm−1 and 1171 cm−1 peaks of GHK-loaded-PCL in the range 1120–1130 cm−1 and 1284–1320 cm−1 in Fig. S6a-b.

In vitro Cytotoxicity of GHK, PCL Nanoparticles, and GHK-Loaded PCL Nanoparticles

Figure 8a shows the cytotoxicity results of GHK, PCL nanoparticles, and GHK-loaded PCL nanoparticles. As seen in the figure, no investigated sample showed toxicity on cells. The results were similar for different concentrations and the control.

Fig. 8
figure 8

In vitro cytotoxicity of GHK tripeptide, PCL nanoparticles, and GHK-loaded PCL nanoparticles (a), In vitro anticancer effect of GHK on U87 cells

Full size image

The in vitro Anticancer Effect of GHK

As seen in Fig. 8a, b, GHK has no anticancer or toxic effect on cells in lower concentrations (p > 0,05). However, GHK decreased the viability of glioblastoma cells at higher concentrations while the viability of L929 cells remained the same as the control. Particularly at the 40 mg/mL concentration, the cell viability decreased to nearly 65%.

Conclusion

GHK is a tripeptide with multiple biological activities such as antioxidant, wound healing, anti-inflammation, anti-pain and anti-anxiety effects; it has also been proposed as a therapeutic peptide for cancer. Some in vitro studies have been conducted on the anticancer effect of GHK on prostate, breast cancer and human nerve cancer. In this study, the effect of GHK tripeptide and its nano-sized formulation on glioblastoma cells was investigated for the first time.

In our study, optimized GHK-loaded PCL nanoparticles were obtained with 232.5 ± 0.72 average particle size and − 10.8 ± 0.64 mV zeta potential. The usage potential of GHK nanoformulation for cancer threapy was investigated via in silico and in vitro methods. It is believed that the GHK-loaded PCL nanoparticles may demonstrate the potential to cross the blood barrier, and can be used as a therapeutic agent, particularly in the area of nervous diseases.