Nanoliposome Use to Improve the Stability of Phenylethyl Resorcinol and Serve as a Skin Penetration Enhancer for Skin Whitening

Abstract

Phenylethyl resorcinol (PR) is a potent tyrosinase inhibitor and a cosmeceutical skin lightening agent. However, the application of PR is limited by photoinstability and poor solubility. In this study, we formulated and optimized phenylethyl resorcinol loaded nanoliposomes (PR-NLPs) to improve the stability and effective delivery of PR. PR-NLPs were prepared by the ethanol injection method and optimized by a single factor experimental and Box–Behnken design. In addition, Diethylamino Hydroxybenzoyl Hexyl Benzoate (DHHB) as the UBA absorber was added to PR-NLPs, which significantly improved the photostability of PR. The mean size, polydispersity index (PDI), and zeta potential of the optimized PR-NLPs were 130.1 ± 3.54 nm, 0.225 ± 0.02, and −43.9 ± 3.44 mV, respectively. The drug encapsulation efficiency (EE) and loading efficiency (LC) of PR-NLPs were 96.81 ± 3.46% and 8.82 ± 0.6%, respectively. These PR-NLPs showed good physicochemical stability for 3 months at 4 °C and 25 °C in the dark. They showed typical sustained and prolonged drug-release behavior in vitro. The in vitro cytotoxicity assay and cellular uptake demonstrated that the PR-NLPs had excellent biocompatibility and cell transport ability. It significantly inhibited tyrosinase activity and reduced melanin production in B16F10 cells at concentrations of 20 or 30 μg/mL. Moreover, the PR-NLPs enhanced the PR into the skin. These results indicate that PR-NLPs can be used as a nanocarrier to improve the transdermal delivery of PR.

1. Introduction

Whitening is a goal that most people have been pursuing, especially Asians. According to Global Industry Analysts, Inc (GIA), the global market for skin whitening agents will reach $23 billion by 2020 [1], which will make whitening cosmetics become the focus of attention of cosmetics companies. Various inhibitors of melanogenesis-related enzymes have been identified for whitening effects and the control of skin pigmentation in the medicine and cosmetic industries. Melanin is the primary pigment responsible for the color of human skin. Melanogenesis is a process in which melanin is produced within melanocytes and subsequently transferred to adjacent keratinocytes to protect keratinocytes from UV light [2]. However, overproduced melanin leads to various skin pathological problems, such as chloasma, age spots, freckles, and pigmentation caused by inflammation [3]. Therefore, excessive melanin production should be controlled to maintain the skin integrity.

In melanocytes, the production of melanin involves a variety of enzymatic and chemical reactions, which can be divided into two steps in mammals. First, phenylalanine in the cytosol is converted into L-tyrosine by phenylalanine hydroxylase [4]. Tyrosinase catalyzes the oxidation of L-tyrosine to produce L-DOPA [5]. L-DOPA continues to be oxidized by tyrosinase to dopaquinone (DQ). Subsequently, DQ is converted to either pheomelanin or eumelanin through a spontaneous reaction or tyrosinase-related protein-1 (TRP-1), or tyrosinase-related protein-2 (TRP-2) [6]. Tyrosinase is the rate-limiting enzyme of melanin synthesis, which is a glycoprotein carrying nearly 60–70 kDa. It is used as a target of melanogenesis inhibitors to reduce skin pigmentation [7].

Currently, skin whitening agents mainly include resveratrol, arbutin, tretinoin, kojic acid, vitamin E, and phenylethyl resorcinol (4-(1-phenylmethyl)1,3-benzenediol, PR). PR is a synthetic phenolic compound developed by Symrise (Holzminden, Germany) [8]. It is a new type of high-efficiency whitening active substance screened out by German researchers after chemically transforming the natural whitening ingredient silver pine. PR has attracted widespread interest as a skin whitening agent due to its excellent tyrosinase inhibition activity by obstructing the conversion of tyrosine to L-DOPA. According to relevant studies, PR is one of the highest tyrosinase inhibitors, and its tyrosinase inhibitory ability is 22 times stronger than that of kojic acid [9]. At a concentration of 0.5%, it can cause skin discomfort in Asians [10]. In addition, PR is a potential antifungal agent against dermatomycoses [11]. Unfortunately, some shortcomings, such as poor aqueous solubility (solubility of 4.05 ± 0.02 mg/g in water) [8], instability under light, and easy chelation by metal ions limit the application of PR as a formulation of cosmetic and pharmaceutical dermal products. When PR is exposed to light, it will be degraded, and the color of PR changes from white to pink. To improve the photostability of PR-NLPs against UV light, Diethylamino Hydroxybenzoyl Hexyl Benzoate (DHHB) was added to PR-NLPs to prevent the rapid degradation of PR [10]. DHHB is a UVA absorber without the drawback of photoinstability and with good solubility properties [12]. In addition, a 1% PR in the production can cause skin irritation, which results in low customer acceptance [13].

To overcome these limitations of PR, a drug delivery system is considered. Nanocarriers are widely applied in formulation and development due to their numerous advantages in drug delivery. Nanocarriers have emerged as a powerful approach to enhance bioavailability and solubility, prolong the duration of action, and improve drug stability [14]. Drug carriers are conceited to protect drugs, literally building layers of protection around them [15]. Nanoliposomes are also a potential drug delivery system due to their biodegradability, biocompatibility, low toxicity, and ability to encapsulate hydrophilic and lipophilic drugs simultaneously [16]. Additionally, the biggest obstacle to transdermal drug delivery lies in the barrier function of the epidermal stratum corneum of the skin, which makes the drug unable to penetrate smoothly. Nanoliposomes are composed of phospholipids and cholesterol, which are similar to skin stratum corneum lipids and have good compatibility. They can carry drugs to penetrate the skin and to be absorbed effectively [17]. Regarding applications to PR, previous studies reported that nanostructured carriers (NLCs) [18] and solid lipid nanoparticles (SLNs) [19] were used to increase the solubility and photostability of PR. Ethosomes, transferosomes, and liposomes have been used to increase the solubility and absorption of PR [13,20]. These vesicle carriers can reduce skin irritation of PR by avoiding direct contact of the encapsulated PR with the skin [21]. Additionally, to deliver PR to melanocytes, PR needs to penetrate the stratum corneum into the basal layer of the epidermis. Recently, Amnuaikit et al. [22] prepared a conventional liposome by the thin film hydration method. These conventional liposomes had a big vesicle size and low encapsulation efficiency. Panithi Raknam et al. [23] developed Liposomal Cream Formulation by the modified ethanol injection method. However, these liposomes ranged in size from 200–640 nm and were too large to penetrate the skin into the basal layer [23]. It has been reported that small liposomes have better transdermal delivery capability than larger liposomes [24]. For example, the amount of liposomes with a mean size of 120 nm that reached deeper skin areas (epidermis and dermis) was 4.6 times that of liposomes with a mean size of 191 nm and 33 times higher than that of liposomes with a mean size of 810 nm [24].

In this study, to optimize the mean size of PR nanoliposomes and transdermal efficiency, PR loaded liposomes (PR-NLPs) were prepared by the ethanol injection method and optimized by Box–Behnken design. The physicochemical properties, including the morphology, mean size, polydispersity index (PDI), zeta potential, entrapment efficiency, and FTIR spectrum, were characterized. The stability, cellular uptake, whitening efficacy and in vitro skin permeation were examined to evaluate the suitability of PR-NLPs for skin whitening applications. This study provides an experimental basis for overcoming the limitation of using PR for skin whitening products.

2. Materials and Methods

2.1. Materials

Phenylethyl resorcinol (PR) was obtained from Shanghai Acmec Biochemical Co., Ltd. (Shanghai, China). Lecithin and cholesterol were purchased from Macklin Biochemical Technology Co., Ltd. (Shanghai, China). A Cell Counting Kit-8 (CCK-8) was purchased from MedChemExpress (Princeton, NJ, USA). DMEM, 1640 medium, and fetal bovine serum (FBS), were purchased from Gibco (BRL, Gaithersburg, MD, USA). L-DOPA was purchased from Sigma-Aldrich (St. Louis, MO, USA). Triton X-100 and ultrafiltration centrifugal filter tubes were supplied by Beijing Solarbio Science & Technology Co., Ltd. (Beijing, China). Milli-Q water (NeoLab, Hangzhou, China) was used throughout the experiment. All chemicals were used as received without any modification.

2.2. Preparation of PR-NLPs

The preparation process of PR-NLPs is illustrated in Figure 1. In brief, The PR-NLPs were prepared by a modified ethanol injection method as follows: the lipid phase: lecithin (50 mg/mL) and cholesterol (8.75 mg/mL) were dissolved in ethanol under mixing at 800 rpm and 30 °C. Then, PR (5 mg/mL) and DHHB (1 mg/mL) were added to the lipid phase as the oil phase. Meanwhile, for the aqueous phase, Tween 80 (0.5% w/v) was dissolved in ultrapure water. A syringe pump was used to inject the oil phase into the aqueous phase under magnetic stirring at 30 °C. Due to the influence of injection velocity and the stirring rate on vesicle size (VS), the injection velocity and stirring rate were adjusted to 750 μL/min and 850 rpm, respectively. The aqueous phase immediately became milky as a result of spontaneous liposome formation. Second, the liposome was stirred for 30 min at 30 °C. The absolute ethanol was removed from the liposome suspension by a rotary evaporator at 40 ± 2 °C. The resulting liposome suspensions were stored at 4 °C. Fluorescein isothiocyanate-loaded PR-NLPs (FITC-PR-NLPs) were also prepared using an optimized molar ratio for cellular uptake studies and skin preparation studies.

2.3. Optimization of PR-NLPs Formulation

2.3.1. Investigating a Variety of Factors That Influenced EE and Mean Size of PR-NLPs

The effect of lecithin concentration (50–250 mg/mL), cholesterol content (0%–25%, w/w), phenylethyl resorcinol content (0%–25%, w/w), the volume ratios of nonsolvent and solvent (4, 5, 10, 15, 20), and Tween 80 content (0.1%–0.5%, v/w) were studied by single factor design as follow: one factor was changed while the other factor was kept constant. The major experimental factors were chosen to design Box–Behnken-RSM-based optimization.

2.3.2. Box–Behnken-RSM-Based (BBK-RSM) Optimization

The Box–Behnken-RSM process was utilized to optimize the prescription of PR-NLPs. The BBK was designed on the three factors of lecithin concentration (A), cholesterol content (B), and phenylethyl resorcinol content (C) that greatly influence the size (Y). A three-factor BBK at three levels was designed for screening the main effects of the three factors (

Table 1. Box–Behnken design-selected compositions and response variables of the prepared PR-NLP formulations.

No.	Lecithin Concentration (mg/mL)	Cholesterol Content (w/w %) a	Phenethyl Resorcinol Content (w/w %) a	Vesicle Size (nm) b
1	50	17.5	15	129.7 ± 0.8
2	60	10	15	151.8 ± 2.5
3	50	10	20	128.9 ± 0.2
4	50	17.5	15	133.3 ± 0.9
5	50	17.5	15	132.4 ± 1.1
6	50	17.5	15	124.8 ± 0.8
7	40	25	15	158.2 ± 0.4
8	50	17.5	20	195.2 ± 7.0
9	50	17.5	10	117.3 ± 2.1
10	50	10	10	88.61 ± 0.8
11	40	17.5	20	157.6 ± 1.2
12	50	25	15	171.6 ± 1.6
13	40	17.5	10	111.8 ± 0.3
14	40	10	15	108.4 ± 1.4
15	50	25	20	165.8 ± 3.2
16	50	25	10	106.2 ± 1.7
17	50	17.5	5	116.7 ± 0.8

^a The percentage of cholesterol to lecithin mass; ^b Mean values ± standard deviations of three measurements.

) on a significant response: size.

Design-Expert V10.0.1 software was used for the experimental design and processing of experimental data. The data were fitted and analyzed by the software to obtain model equations for significance evaluation. According to Design-Expert, the disturbance line graph and three-dimensional response surface graph between different factors were drawn to show the response relationship between different factors.

Based on the results obtained in the single-factor experiment, The three factors were lecithin concentration (A), cholesterol content (B), and Phenylethyl resorcinol content (C). Three levels of independent variables were identified. For three factors, Box–Behnken-RSM can require fewer runs over the three-level full factorial design, a central composite design (CCD). In full factorial designs, the number of experiments increased exponentially with the number of factors. Seventeen experiments with five central points were carried out using the Box–Behnken design. The responses of the 17 experiments are shown in Table 1.

2.4. Characterization of PR-NLPs

2.4.1. Transmission Electron Microscopy (TEM) Analysis

The morphology and size of PR-NLPs were determined using a transmission electron microscope (JEOLTEM-1210, Shanghai, China) at 120 kV [19]. The PR-NLP suspension was diluted with ultrapure water at a ratio of 100:1 and dropped on a copper grid. After natural drying, the excess liquid was absorbed with filter paper, and the microscopic morphology was observed under TEM.

2.4.2. Atomic Force Microscopy (AFM) Analysis

The surface morphology of PR-NLPs was visualized by AFM (Bioscope Catalyst/Multimode, Santa Barbara, CA, USA). A muscovite mica coverslip was used in the analysis. It was cleaned with ultrapure water and treated to clean any further impurities. The PR-NLP suspension was diluted with ultrapure water at a ratio of 100:1 and placed on a mica coverslip. After natural drying, the excess liquid was absorbed with filter paper, and the microscopic morphology was observed under AFM.

2.4.3. Nanoparticle Size, Polydispersity Index (PDI), and Zeta Potential Analysis

An analysis of the nanoparticle size, PDI, and zeta potential of the samples were determined using a Zteasizer Nano ZS90 system (Malvern, Egham, U.K) at 25 °C. First, the nanoparticle size and PDI of the samples were diluted (1:5) with ultrapure water. The zeta potential of the samples was not diluted and determined at 25 °C with a scattering angle (θ) of 90 degrees.

2.4.4. High Performance Liquid Chromatography (HPLC) Measurement

The PR content was analyzed by high-performance liquid chromatography (Waters ACQUITY Arc bio, Milford, MA, USA). The analytical chromatography column was a C18 column (2.1 mm × 150 mm, 3.5 μm) (Waters, Milford, MA, USA). The injected volume was 2.5 μL. The mobile phase was a mixture of methanol and ultrapure water (50:50, v/v) at a flow rate of 0.35 mL/min, the wavelength was 281 nm, and the column temperature was 35 °C.

2.4.5. Drug Encapsulation Efficiency (EE) and Loading Capacity (LC)

The EE and LC of PR-NLCs were determined using an ultrafiltration centrifugation method. First, 0.5 mL of PR-NLPs suspension was diluted and demulsified with chromatographic methanol. The PR content in the diluted suspension was determined by HPLC, and this PR amount was designated as R_total. A sample of 0.5 mL PR-NLP suspension was added into ultrafiltration centrifugal filter tubes with a molecular weight cut-off of 30 KDa and centrifuged at 10,000 rpm for 1 h at 4 °C. After centrifugation, the ultrafiltration liquid was diluted with chromatographic methanol at the same speed, and the drug content in the obtained solution was analyzed under the same HPLC conditions. The amount of free PR was designated as R_free.

The EE and LC were calculated by the following equation:

$$\mathrm{EE} (\%)=\frac{{\mathrm{R}}_{\mathrm{total}}-{\mathrm{R}}_{\mathrm{free}}}{{\mathrm{R}}_{\mathrm{total}}}\times 100\%$$

$$\mathrm{LC} (\%)=\frac{{\mathrm{R}}_{\mathrm{total}}-{\mathrm{R}}_{\mathrm{free}}}{{\mathrm{R}}_{\mathrm{lipid}}+{\mathrm{R}}_{\mathrm{total}}-{\mathrm{R}}_{\mathrm{free}}}\times 100\%$$

where R_lipid is the weight of total amount of lipid, R_total is the weight of total amount of PR, and R_free is the weight of unloaded PR in PR-NLPs.

2.4.6. Fourier Transform Infrared Spectroscopy (FTIR) Analysis

FTIR spectra of the pure PR, the lyophilized blank NLPs, and the lyophilized PR-NLPs were obtained using a Fourier transform infrared spectrometer (VERTEX 70/70v FT-IR spectrometer, BRUKER OPTICS, Saarbrücken, Germany) to detect the interaction between PR and the lipid molecules. A total of 32 scans were used, and data were recorded over the range 4000–400 cm⁻¹. The sample was attached to the infrared transparent crystal surface of the ATR accessory, and the infrared spectrum of the sample was scanned. Software (OPUS 8.5, RUKER OPTICS, Saarbrücken, Germany) was used to capture and analyze spectra.

2.4.7. In Vitro Drug Release

To achieve controlled release and obtain prolonged-release information, the dialysis diffusion technique was used to investigate PR release from PR solution and PR-NLPs in phosphate-buffered saline (PBS) with 0.1% Tween 80. Basically, 7 mL of PR-NLPs suspensions were placed into a dialysis bag (molecular weight cut-off, 8000–14,000; Shanghai Acmec Biochemical Co., Ltd., (Shanghai, China). The release medium was PBS (pH = 6.8) containing 1% Tween 80. The dialysis bag was then placed into a 400 mL release medium, which was stirred at 100 rpm with the temperature kept at 37 ± 0.5 °C. The PR ethanol solution containing an equal amount of PR was evaluated by the same dialysis method as the control. At fixed time intervals, 1 mL of release medium was collected and replaced with 1 mL of fresh release medium to maintain sink conditions. Samples were determined by HPLC as described above.

2.5. The Photostability of PR-NLPs Study

To determine the photostability of PR-NLPs against UVA radiation, the method reported by Chirio et al. was adopted [25]. Briefly, samples were placed in a sealed vial, placed at a distance of 10 cm from the light source (commercial UVA sun lamp equipped with a 400 W lamp, Philips UV disinfection lamp). The output of UVA measured with a UV Power Pack Radiometer corresponded to 30 μW/cm². The samples were withdrawn at 4, 8, 12, 24, 36, and 48 h for analysis. All experiments were carried out in triplicate.

2.6. Stability Test

The stability test was conducted to evaluate the long-term stability of PR-NLPs. The PR-NLP stability was studied over three months by investigating different changes in mean size and PDI. The samples were stored at 4 °C in the dark, 25 °C in the dark, and 25 °C in daylight, and aliquots of the samples were taken out for analysis every month. The mean size and PDI were determined by the Zteasizer Nano ZS90 system, as described in Section 2.4.3.

2.7. Cell Culture

HaCaT cells were cultured in DMEM supplemented with 10% FBS. B16F10 cells were cultured in 1640 medium with 10% FBS. Cells were incubated in 5% CO₂ at 37 °C.

2.7.1. Cytotoxicity Assay in HaCaT Cells and B16F10 Cells

Briefly, the cytotoxicity of PR, Blank-NLPs, and PR-NLPs were evaluated on HaCaT cells and B16F10 cells utilizing CCK-8 assay. The cells (1 × 10⁴ cells/well) were seeded in a 96-well plate and incubated at 37 °C, 5% CO₂, for 24 h. The cells were then incubated with a complete medium containing Blank-NLPs, PR, and PR-NLPs with varying concentrations (0–80 μg/mL, 0 μg/mL used as the control) for 24 h. After incubation, cells were washed once with PBS (pH 7.4) and incubated with 100 μL of 10% CCK-8 solution per well for 1.5 h at 37 °C in the dark. After the incubation, the 96-well plate was placed on a microplate reader (Thermo Fisher, Multiskan FC) to measure the absorbance at 450 nm. Cell viability was calculated by the following equation.

$$\mathrm{Viability} \mathrm{Rate} (\%)=\frac{{\mathrm{A}-\mathrm{A}}_0}{{\mathrm{A}}_{\mathrm{p}}{-\mathrm{A}}_{\mathrm{n}}}\times 100\%$$

where A, A_p, and A_n are the absorbances of the tested sample, positive control, and negative control, respectively.

2.7.2. Cellular Melanin Content Assay in B16F10 Cells

The B16F10 cells (1 × 10⁶ cells/well) were seeded in a 12-well plate and incubated with 1640 medium at 37 °C, 5% CO₂, for 24 h. The cells were then incubated with DMEM medium containing PR and PR-NLPs with varying concentrations (10, 20, and 30 μg/mL; 0 μg/mL used as the control) for 72 h. After incubation, cells were washed twice with cold PBS, and 500 μL of trypsin was added to each well. After 3 min, 500 μL of DMEM was added to stop digestion, and the samples were collected in centrifuge tubes and centrifuged at 13,000 rpm for 5 min. The centrifuged cells were dissolved in 300 μL of 1 M NaOH containing 10% DMSO, and the absorbance was measured at 405 nm. The experimental results are expressed as relative cellular melanin content.

2.7.3. Cellular Tyrosinase Inhibition Assay

Cells were seeded in 96-well plates at a density of 5×10³ cells per well. Cells were then incubated with 1640 medium at 37 °C, 5% CO₂, for 24 h. The cells were then incubated with DMEM medium containing PR and PR-NLPs with varying concentrations (10, 20, and 40 μg/mL; 0 μg/mL used as the control) for 72 h. After 72 h, cells were washed cold PBS and lysed with PBS (50 μL) containing 0.1% TritonX-100 at −80 °C for 30 min. The cellular tyrosinase activity was then measured after adding 150 µL of the reaction mixture containing 0.1% L-DOPA into 96-well plates. After incubation at 37 °C for 1 h, the absorbance was measured at 475 nm. The experimental results were expressed as relative tyrosinase activity. The cellular tyrosinase activity was calculated using the following equation:

$$\mathrm{Cellular} \mathrm{tyrosinase} \mathrm{activity} (\%)=\frac{{\mathrm{OD}}_{\mathrm{treat}}}{{\mathrm{OD}}_{\mathrm{control}}}\times 100\%$$

where OD_treat is the absorbance of the treated sample and OD_control is the absorbance of the control sample.

2.7.4. Celluar Uptake Study

The uptake of PR by B16F10 cells and HaCat cells were analyzed using flow cytometry. By replacing PR with liposoluble FITC, the FITC-PR-NLPs were prepared by the same method of PR-NLPs by labeling PR-NLPs with FITC. B16F10 cells and HaCaT cells were seeded at a density of 3 × 10⁵ cells per well in 6-well plates for 24 h. The cells were then treated with a FITC solution or FITC-PR-NLPs (the equivalent concentration of 5 μg/mL) at 37 °C, 5% CO₂ for 1 and 3 h. Cells were harvested and centrifuged to remove the supernatant. Subsequently, cells were washed twice with cold PBS and resuspended in 0.5 mL of PBS. Finally, quantitative analysis was carried out by flow cytometry (BD Biosciences). Data were processed using the FlowJo V10 software.

Confocal laser scanning microscopy (CLSM) was also used to qualitatively evaluate the cellular uptake of FITC from FITC-PR-NLPs. B16F10 cells or HaCaT cells were seeded at a density of 1 × 10⁵ cells per well in 24-well plates for 24 h. The cells were then treated with a FITC solution or FITC-PR-NLPs (the equivalent concentration of 5 μg/mL) at 37 °C, 5% CO₂, for 1 and 3 h. Cells were washed three times with PBS and fixed with 4% paraformaldehyde in PBS for 10 min. Finally, the nuclei were stained using DAPI for 10 min at room temperature, and images were taken using CLSM.

2.8. In Vitro Skin Permeation Study

To explore the distribution of drugs contained in PR-NLPs in various layers of the skin, in vitro permeation experiments were performed using a Franz diffusion cell. Pigskin was used as a diffusion membrane because it is similar to human skin in terms of structure, thickness, hair follicle content, and lipid composition. Full-thickness skin was acquired from the dorsal side of a freshly excised pig at the local market. A pigskin cut to an appropriate size was sandwiched between the injection cell and the receiving cell, with the cuticle facing the injection cell. An amount of 4 mL of FITC solution or FITC-PR-FITC (instead of PR or PR-NLPs) was added to each donor compartment, with an effective diffusion diameter of 2 cm. Each 20 mL receptor compartment was filled with freshly prepared 10% absolute ethanol (v/v) in PBS to provide sink conditions, thermostated at 37 ± 0.5 °C, and stirred with a magnetic bar at 450 rpm. To study the skin permeation of the fluorescent dyes over time, the skin was removed from the Franz diffusion cell after 4, 8, 12, 16, and 24 h. The treated pigskin surface was wiped clean with a cotton swab, embedded in the Tissue-Tek OCT compound (Sakura Finetek, Torrance, CA, USA) and frozen at –20 °C. The frozen skin was then sectioned with a cryotome (CryoStar NX50, Thermo Fisher, Waltham, MA, USA) into 10 μm slices. Keratinocyte nuclei were stained with DAPI. Finally, images were taken using CLSM. The skin delivery of FITC and FITC-PR-NLPs was visualized by assessing the penetration of the fluorescent dyes.

2.9. Statistical Analysis

In this paper, all values are reported as the mean and standard deviation. Differences between the two groups were compared by Student’s t-test. The above statistics were performed using GraphPad Prism 8 for the cell viability analysis, cellular melanin production, and tyrosinase activity. The acceptable level of significance was considered at * p < 0.05.

3. Results and Discussion

3.1. Investigating a Variety of Factors by Single-Factor Experiment

Excellent transdermal absorption requires a smaller mean size (<200 nm) and better stability of liposomes [26]. The mean size and encapsulation efficiency of liposomes are influenced by lecithin, cholesterol, phenylethyl resorcinol, the volume ratios of nonsolvent and solvent volume ratios, and Tween 80. To obtain the best ratio of these formulation components, a series of experiments were carried out.

3.1.1. Effect of Lecithin Concentration on the Mean Size and EE

The formation mechanism of liposomes has been explained by the bilayer planar fragments (BPFs) theory [27]. According to the BPF theory, lecithin dissolved in organic solvent (ethanol) would deposit at the water/organic solvent phase boundary. Upon complete diffusion of organic solvent to the external aqueous phase, BPF formed vesicles through self-assembly [28]. Studies have shown that as the concentration of lecithin in the formulation increased, the number of small unilamellar vesicles (SUV) increased, and the lipid surface thickened, which led to the formation of larger liposomes [29].

The effect of lecithin concentration on the mean size and encapsulation efficiency of PR-NLPs was investigated by varying the lecithin concentration and maintaining the mass ratio of lecithin to PR. With the increasing of lecithin concentration from 50 to 250 mg/mL, the mean size of PR-NLPs showed an obvious upward trend from 125.6 to 729.3 nm (Figure 2a). When the lecithin concentration was 50 mg/mL, the mean size of PR-NLPs was less than 150 nm and had a high encapsulation efficiency (>80%), which met the requirements that mean sizes are less than 200 nm. Therefore, this concentration was selected for the following experiments.

3.1.2. Effect of Cholesterol Content on the Mean Size and EE

Cholesterol was added to the structure of the liposomes to increase the stability of vesicles, making liposomes more stable by adjusting the fluidity of the lipid bilayer, preventing the crystallization of phosphatidyl chains and sterically hindering their movement, and reducing the permeability of the liposomes membrane to make the liposome more stable [30].

Figure 2b shows the mean size and encapsulation efficiency of PR-NLPs prepared with the different cholesterol contents. As the cholesterol content increased from 0% to 25%, the mean size of liposomes increased from 93.4 to 277.4 nm, and the encapsulation efficiency increased from 85% to 94%. The mean size and encapsulation efficiency of PR-NLPs increased with increasing concentrations of cholesterol. However, the addition of cholesterol increased the rigidity of liposomes and reduced permeability [13,31]. Once cholesterol content exceeded a certain level, the lecithin bilayer structure would be destroyed [32]. Therefore, an appropriate ratio of cholesterol to lipids was essential to improve the stability of nanoliposomes. Fifteen percent cholesterol content was selected to accommodate both adverse effects and benefits.

3.1.3. Effect of Phenylethyl Resorcinol Content on the Mean Size and EE

Liposomes can carry hydrophobic drugs by intercalating them into hydrophobic domains. Hydrophobic drugs can be directly captured into hydrophobic domains during the formation of liposomes [33]. Phenethyl resorcinol was a hydrophobic drug. PR was encapsulated into the lipid bilayer during vesicle formation using the ethanol injection method (Figure 1). Increasing the concentration of PR in lipid bilayers leads to a higher presence of PR, which may increase the mean size.

To examine the effect of PR content on mean size and encapsulation efficiency, the PR content was changed from 0% to 25%, maintaining the mass ratio of lecithin to PR. The results suggested that the mean size of PR-NLPs (from 99.1 to 106.0 nm) did not change significantly by increasing of PR content. However, when the PR content increased, the encapsulation efficiency of PR-NLPs showed a significant decline from 83% to 49%. Considering the significant influence of PR on the encapsulation efficiency, the subsequent optimization using a lower PR content from 10% to 20% was used for subsequent optimization.

3.1.4. Effect of the Volume Ratios of Nonsolvent and Solvent (NS/S) on EE and Mean Size

We investigated the effect of the NS/S on the mean size and encapsulation efficiency of PR-NLPs to examine the influence of the volume ratios of nonsolvent and solvent on liposome formation. The results are shown in Figure 2d. The mean size of PR-NLPs increased as the volume ratio of NS/N increased from 4 to 20, while the encapsulation efficiency decreased. As the volume ratios of nonsolvent and solvent decreased, the mean size of PR-NLPs increased from 100 to 128 nm, and the encapsulation efficiency decreased from 88% to 80%.

A previous research study indicated that NS/S volume ratios ranging from 1 to 10 have no significant effect on the mean size [34]. A low NS/S volume ratio reduces the possibility of PR being captured, which lowers the encapsulation efficiency of hydrophobic drugs. Considering the final PR load of PR-NLPs, the ratio was fixed at 5 for the following experiments.

3.1.5. Effect of Tween 80 Content on EE, PDI, and Mean Size

Tween 80 could improve the stability of nanoliposomes [35]. We investigated the effect of Tween 80 on mean size and EE, and the results are shown in Figure 2e. As the content of Tween 80 increased from 0 to 0.5%, the mean size of PR-NLPs did not change significantly. With the increase in Tween 80 content from 0 to 0.5%, the encapsulation efficiency of PR-NLPs was obviously improved. In addition, the Tween 80 could significantly increase the encapsulation efficiency of PR-NLPs at a concentration of 0.3%, while the encapsulation efficiency did not change significantly when the content of Tween 80 exceeded 0.3%. However, the PDI of PR-NLPs decreased significantly when the content of Tween 80 increased (Figure 1f). Considering the PDI of PR-NLPs, the smaller the PDI, more uniform is the mean size distribution of liposomes. Based on these results, a Tween 80 content of 0.5% was selected for the following experiment.

3.1.6. Photostability of PR-NLPs

To improve the photostability of PR-NLPs against UV light, DHHB was added to PR-NLPs to prevent the rapid degradation of PR [10]. DHHB is a UVA absorber without the disadvantage of photoinstability and with good solubility [12].

In Figure 3, UV degradation curves are shown in different systems. The PR solution degraded rapidly, as only 36.5% of PR remained after 48 h of irradiation. Compared to the PR-Solution, effective protection against photodegradation was observed when PR was loaded in PR-NLPs (without DHHB) and PR-NLPs. In addition, when DHHB was added to the formulation of PR-NLPs, its photostability was higher than that of PR-NLPs without DHHB. Because the PR was encapsulated in the phospholipid bilayer of the liposome, a membrane was formed and may have blocked ultraviolet rays. These data suggested that DHHB was incorporated into the phospholipid bilayer of PR-NLPs, which enhanced the photostability of PR-NLPs.

3.2. Optimization of PR-NLPs Using Box–Behnken Design

Particle size is a key criterion for the characterization of liposomes, especially transdermal liposomes. The mean size of PR-NLPs ranged from 88.6 to 195.2 nm (Table 1). In the ANOVA result for mean size (

Table 2. ANOVA of quadratic model for responses of size.

Source	Sum of Squares	df	Mean Square	F-Value	p-Value
Model	11,997.55	9	1333.06	31.06	<0.0001
A	1248.33	1	1248.33	29.09	0.001
B	1821.66	1	1821.66	42.45	0.0003
C	6436.59	1	6436.59	149.99	<0.0001
AB	178.22	1	178.22	4.15	0.081
AC	206.88	1	206.88	4.82	0.0641
BC	93.9	1	93.9	2.19	0.1826
A2	1967.65	1	1967.65	45.85	0.0003
B2	1.76	1	1.76	0.041	0.8454
C2	80.13	1	80.13	1.87	0.2141
Residual	300.39	1	42.91	-	-
Pure Error	185.64	7	46.41	-	-
Corr. Total	12,297.95	4	-	-	-
R2	0.9756	16	-	-	-
C.V.%	4.85	-	-		-
Adeq Precisor	21.679	-	-	-	-

A: Lecithin concentration; B: Cholesterol content; C: Phenylethyl resorcinol content.

), the p-values of the lecithin concentration, cholesterol content, and phenylethyl resorcinol content were less than 0.05. This indicates that these factors had a significant effect on the mean size. According to ANOVA, the quadratic model suggested by the 17 BBK-RSM-based experiment runs and the final equation in terms of coded factors:

Size = 127.39 + 12.49A + 15.09B + 28.37C − 6.67AB + 7.19AC + 4.84BC + 21.62A² − 0.65B² − 4.36C²

where A, B, and C are the lecithin concentration, cholesterol content, and phenylethyl resorcinol content, respectively. AB, AC, and BC showed the interactions between two factors.

A three-dimensional (3D) response surface and two-dimensional contour were used to show the effect of independent variables on the size in Figure 4. The results in Figure 4 were consistent with the results of previous single-factor studies. After imposing constraints on three factors, the Design-Expert software proposed optimal PR-NLPs: A = 50 mg/mL, B = 17.5%, and C = 10%. The corresponding optimal response varial was a size of 127.7 nm. The optimized mean size would satisfy the mean size requirement of nanoliposomes, which would facilitate the transdermal delivery of PR. The optimized formulation was subjected to characterization analysis and cell assays.

3.3. Characterization of PR-NLPs

3.3.1. Morphology and FTIR analysis

Figure 5a shows that the PR-NLPs suspension was milky white. To obtain information regarding the physical appearance of the PR-NLPs, TEM and AFM were used to take photographs of the PR-NLPs, as shown in Figure 5b. All PR-NLPs were monodisperse (PDI < 0.3), and the shapes of PR-NLPs were close to spherical. The vesicular structure was discernible and the PR-NLPs were unilamellar vesicles. The outermost surface of PR-NLPs was a shell composed of phospholipid biomolecules. The resultant nanosystem displayed a spherical structure with a hydrodynamic diameter of 130.1 nm ± 3.54 nm (Figure 5c).

The PR-NLPs were further viewed by AFM to predict the particle topology. Figure 5d shows that the size of the PR-NLPs was uniform, and the particle size in AFM was consistent with the DLS test results. Figure 5e shows that the surface of the PR-NLPs was smooth and uniform, with a gentle peak shape. The obtained 2-D images related to the mean size, and 3-D image showed that the particles were uniformly distributed.

To determine the possible interaction and complex formation between PR and lipids during the preparation of the PR-NLPs, FTIR spectroscopic studies were performed. Figure 5f shows that the FTIR spectrum of pure PR exhibited infrared characteristic absorption peaks at 1600, 1509, 971, and 697 cm⁻¹. These absorption peaks only appeared in pure PR, but not in Blank-NLPs and PR-NLPs. The FTIR spectra of Blank-NLPs and PR-NLPs were almost identical. In the FTIR spectrum of PR-NLPs, the peaks corresponding to pure PR disappeared or were hidden in the peaks of Blank-NLPs, indicating that PR was encapsulated by the lipid matrix.

3.3.2. Measurements of Mean Size, PDI, Zeta Potential, EE, and LC

As shown in

Table 3. Diameter, zeta potential, PDI, and EE of Blank-NLPs and PR-NLPs (n = 3, mean value ± SD).

Group	Size (nm)	Zeta Potential (mV)	PDI	EE (%)	LC (%)
Blank-NLPs	106.0 ± 1.9	−27.9 ± 1.16	0.276 ± 0.02	-	-
PR-NLPs	130 ± 3.5	−43.9 ± 3.44	0.225 ± 0.02	96.81 ± 3.46	8.82 ± 0.6

, the mean size and PDI of PR-NLPs were 130.1 ± 3.54 nm and 0.225 ± 0.02, respectively. The PR-NLPs had negatively charged surfaces at -43.9 ± 3.44 mV. Generally, the nanocarrier system was relatively stable when the zeta potential was greater than ±30 mV. This was mainly due to the mutual repulsion of the charges, which prevented the contact between the particles to stabilize the nanocarriers to remain dispersed [9]. The stability of the formulations depended on the ingredient charges, and PR-LPs with large negative charges repelled each other and did not come together and aggregate. Hence, the higher zeta potential of PR-NLPs resulted in more stable formulations. The PDI of PR-NLPs was lower than 0.3, which indicated a uniform size distribution of nanoliposomes [36]. The mean size was less than 200 nm. Smaller size facilitated passage through biological barriers and enhanced the cellular uptake of PR-NLPs [37].

Drug encapsulation efficiency and loading efficiency are important aspects of the nanoliposomes as ideal drug delivery carriers [38]. EE referred to the percentage of the encapsulated substance in the total amount of the drug in the liposome suspension. It reflected the degree to which the drug was encapsulated by the carrier. LC referred to the amount of drug-loaded per unit weight or unit volume of liposomes. The EE and LC of PR-NLPs were 96.81±3.46% and 8.82±0.6%, respectively. These results indicated that almost all PR was efficiently encapsulated in the PR-NLPs and loaded in the lipid bilayer.

3.3.3. In Vitro Drug Release

For transdermal drug delivery systems, the nanocarriers exhibited slow-release capacities, which are beneficial to ensure high concentrations of the obtained components in the target site and prolong the half-life [39].

Figure 6 shows the release profile of PR from liposomes compared to PR release from solution. A PR solution in ethanol exhibited a sudden release pattern. Compared with the PR solution, PR-NLPs slowed down the cumulative release rate of PR at the same time point. The release rate of PR in solution was 94.21% ± 2.31%, which was faster than that of PR-NLPs (29.90% ± 4.05%) over 12 h. The PR-NLPs presented a sustained release manner, and the release behavior of PR trended toward an equilibrium release period for 72 h.

Due to high EE and sustained released capabilities, PR-NLPs have the capabilities to provide high concentrations of PR at local sites and release the PR in a controlled behavior. When PR-NLPs reach the site of action, they release PR slowly and prolong the duration of action of PR accumulation at the target site through sustained release. The slow-release of PR from liposomes would prevent high concentrations of PR from acting on the skin, thereby reducing the irritation of PR to the skin.

3.4. Stability Test

Storage stability is a key factor in transdermal delivery systems, including physical stability and chemical stability. Physical instability is induced by drug leakage from the delivery vehicle and aggregation or mutual fusion of drug carriers, while chemical instability is mainly caused by the oxidation of unsaturated fatty acids and the hydrolysis of ester bonds in the lipid components of the encapsulated drugs [40]. In the present study, the physical stability and photostability of PR-NLPs were studied. The stability of PR-NLPs was evaluated by determining the mean size and PDI at 4 °C in the dark and 25 °C in the dark and light. The result of PR-NLPs stability is shown in Figure 7.

Figure 7a shows the appearance of PR-NLPs after 3 months of storage under different conditions. When samples were stored at 4 °C in the dark, and 25 °C in the dark and in the light, there was no precipitation or turbidity in the PR-NLP suspension. Figure 7b,c shows the effect of storage conditions on the mean size and PDI of PR-NLPs. The mean size and PDI of PR-NLPs changed little at 4 °C in the dark for the whole storage period. In addition, the mean size and PDI both increased when stored in the dark or light storage at 25 °C. However, the mean size of PR-NLPs remained within 200 nm after 3 months when samples were stored in the dark and light at 25 °C, which met the transdermal delivery capability of PR-NLPs. Additionally, compared to those stored at 25 °C in the dark, the PDI of PR-NLPs significantly increased at the end of three months at 25 °C in the light.

Hence, these data indicate that PR-NLPs exhibited better stability at 4 °C in the dark. This could be explained by the idea that the low temperature inhibited and slowed down the decomposition of PR-NLPs [41]. When PR-NLPs were stored at 25 °C in the light, the increase in size and PDI might be caused by the aggregation and fusion of liposomes. The increase in PDI could be due to the hydrolysis of the lipid bilayer when exposed to daylight [39]. Nevertheless, the particle size of PR-NLPs remained within 200 nm, which can also effectively promote the transdermal absorption of PR and decrease skin irritation. Therefore, the PR-NLPs had favorable stability in the dark at 4 °C and 25 °C.

3.5. Evaluation of Biocompatibility of PR-NLPs In Vitro

To evaluate the potential cellular cytotoxic activity of PR-NLPs, cell viability was determined in HaCaT cells and B16F10 cells. Generally, cell viability above 80% was considered “no toxicity” [42]. HaCaT and B16F10 cells were treated with Blank-NLPs, PR solution, or PR-NLPs for 24 h, respectively. The cell viability of samples with different PR concentrations was examined. PR solution induced a significant decrease in cell viability at a concentration of 5–40 μg/mL, which could be related to the previously reported dose-dependent toxicity of the PR [18]. However, the cell viability of PR-NLPs was more than 80% when the concentration of PR was 40 μg/mL (Figure 8). There was no significant difference between blank NLPs and PR-NLPs.

PR was encapsulated in liposomes and slowly released from PR-NLPs. Unlike the concentration during PR solution treatment, when PR was added to the lipid matrix, it diffused slowly and was released to the outer region, thereby minimizing the toxic effects of PR on cells [43]. The higher the concentration of PR, the stronger the inhibitory ability of melanin synthesis under conditions where PR was not cytotoxic. PR loaded in NLPs could improve the biocompatibility of PR.

3.6. Cellular Uptake Study

The stratum corneum of the skin is the main barrier for drug penetration. The key to successful topical drug treatment is that the drug must pass through the stratum corneum to the lesion site and maintain it for a certain period. To deliver PR to the target site of skin by PR-NLPs, one of the possibilities was that PR-NLPs could be actively internalized by keratinocytes and melanocytes. Immortalized HaCat and B16F10 cells, similar to normal keratinocytes and melanocytes, were chosen to evaluate the cellular uptake of PR-NLPs.

To study whether PR-NLPs could effectively deliver PR into HaCat cells and B16F10 cells, PR-NLPs were labeled with fluorescein isothiocyanate (FITC). The cellular uptake of FITC-PR-NLPs was evaluated by CLSM and flow cytometry. Figure 9a,b shows the cellular uptake of HaCaT cells and B16F10 cells at different time points after FITC solution and FITC-PR-NLP treatment, respectively. Compared with the FITC-PR-NLPs, the FITC solution presented negligible fluorescence in HaCaT cells and B16F10 cells at 1 h. This indicated that little FITC was taken up by HaCaT cells and B16F10 cells. However, when FITC was loaded in PR-NLPs, PR-NLPs were recruited around the cell membrane at 1 h incubation and internalized after 3 h of incubation.

From the results of flow cytometry, the fluorescence intensity of HaCaT cells treated with FITC-solution was 5.86 times and 5.06 times lower than those treated with FITC-PR-NLPs after 1 h and 3 h of incubation, respectively (Figure 9c). The fluorescence intensity of B16F10 cells treated with FITC solution at 1 h and 3 h incubation was, respectively, 3.28 times and 5.46 times lower than that of FITC-PR-NLPs (Figure 9d). After 1 h incubation, FITC-PR-NLP was taken up by most of the cells, while only a small amount of FITC solution was taken up by the cells. The results suggested that the prepared PR-NLPs effectively enhanced the delivery efficiency of loaded PR into HaCaT cells and B16F10 cells. There were various pathways through which cells could uptake nanoparticles, including clathrin or caveolae-mediated endocytosis, and the size of the nanocarrier was critical for receptor-mediated cellular uptake and the formation of vesicles close to the size of nanoparticles [14,44]. The nanosize of PR-NLPs and unique surface chemistry could activate endocytosis pathways [45]. The outer lecithin of PR-NLPs increased the affinity with the cell membrane and improved the uptake efficiency of cells. Therefore, these data also indicated that PR-NLPs could be accepted by HaCaT cells and effectively taken up in B16F10 cells.

3.7. Effect of PR-NLPs on Cellular Tyrosinase Activity and Melanin Production

Tyrosinase is a rate-limiting enzyme that plays an important role in melanogenesis. According to the cytotoxicity assay, PR-NLPs were not cytotoxic to B16F10 cells at concentrations less than 40 μg/mL. Therefore, B16F10 cells were treated with the formulation at 0, 10, 20, and 30 μg/mL for incubation of 72 h to evaluate the whitening effect of the PR solution, PR-NLPs.

Figure 10a shows the effect of the PR-NLPs on tyrosinase activity in B16F10 cells. The results showed that PR-NLPs and PR could significantly inhibit tyrosinase activity with increasing concentrations. At concentrations of 20 μg/mL and 30 μg/mL, PR-NLPs significantly decreased tyrosinase activity in B16F10 cells compared to the PR solution. At a concentration of 10 μg/mL, 20 μg/mL, and 30 μg/mL, both PR and PR-NLPs effectively reduced the production of melanin (Figure 10b). Moreover, the melanin content of B16F10 cells treated with PR-NLPs was significantly lower than that of B16F10 cells treated with PR at a concentration of 10 μg/mL, 20 μg/mL, and 30 μg/mL.

These results revealed that the PR-NLPs inhibited tyrosinase activity and melanin production. The strategy of PR loaded in NLPs was more effective than the PR solution, indicating that PR-NLPs enhanced the uptake efficiency of PR. PR-NLPs are suitable for PR delivery systems and can be used as nanocarriers to effectively apply PR.

3.8. Skin Permeation Study In Vitro

Fluorescently labeled drugs are widely used to investigate the transdermal absorption of drugs [17]. To visualize the transdermal absorption capacity of PR-NLPs, PR-NLPs were labeled with FITC. The Franz diffusion cell and pig skin were used for the diffusion study.

Figure 11 shows the transdermal absorption image of the FITC solution and FITC-PR-NLP treatment. The penetration depth of the FITC solution and FITC-PR-NLPs increased with the extension of incubation time. Figure 11a shows the skin transdermal absorption of FITC solution in vitro. The penetration depth of the FITC solution did not increase significantly before 16 h. Almost all fluorescence was observed in the epidermis without further penetration. The transdermal depth of FITC-PR-NLPs increased with incubation time (Figure 11b). Thus, the in vitro skin permeation study indicated that PR-NLPs could enhance the transdermal permeation of PR compared with PR solution. These results may contribute to the characteristics of PR-NLPs. They had a small mean size (<200 nm) and narrower particle distribution (PDI < 0.3), which made it easier for them to pass through the stratum corneum. Additionally, PR-NLPs had a lipid membrane similar to cell membranes. The structure may have the same mechanism for enhancing cell permeability [46]. Hence, the skin permeation study suggested that PR-NLPs could be used as PR carriers. PR was accumulated on the skin surface, resulting in a high local concentration of PR, which increased the tyrosinase inhibitory activity of PR.

4. Conclusions

In this study, we successfully developed a nanoliposome to facilitate the delivery of PR. PR-NLPs were prepared by ethanol injection, and the formulation of PR-NLPs was optimized by single factor experiments and Box–Behnken design. The PR-NLPs were nanosized, regular spherical, and negatively charged, with a high loading efficiency and encapsulation efficiency. The in vitro release study results showed that PR-NLPs exhibited a sustained-release effect. Additionally, loading PR onto nanoliposomes exhibited excellent physical and chemical stability at 4 °C during three months of storage and photostability against UV light. Compared with free PR, PR-NLPs effectively reduced the cytotoxicity of PR, enhanced the efficiency of cell uptake, inhibited tyrosinase activity, and reduced melanin production. In vitro skin permeation showed that PR-NLPs significantly enhanced PR permeation through and into skin. Based on these scientific results, PR-NLPs are expected to be a suitable delivery system for the application of PR in skin whitening agents.