Unlocking the Skin Health-Promoting Ingredients of Honeysuckle (Lonicera japonica Thunberg) Flower-Loaded Polyglycerol Fatty Acid Ester-Based Low-Energy Nanoemulsions

Abstract

This study aims to provide a comprehensive evaluation of the bioactive compounds present in honeysuckle flower (Lonicera japonica Thunb.) extract (HSF) and their remarkable antioxidant activity. A docking simulation was performed to clarify the binding affinities of the identified phytochemicals to enzymes associated with anti-aging and anti-inflammatory activities. In addition, the low-energy nanoemulsions based on optimally formulated polyglycerol fatty acid esters (PGFEs), developed through D-optimality, were designed for the incorporation of HSF extract. The result revealed that HSF is a rich source of diverse phenolic and flavonoid compounds that contribute to its remarkable antioxidant capacity. Molecular docking analysis indicates that its compounds exhibit anti-aging and anti-inflammatory activities, particularly through collagenase, hyaluronidase, and TNF-α inhibition. Furthermore, D-optimality revealed that HSF-loaded nanoemulsions can be fabricated by a surfactant to oil ratio (SOR) of 2:1 with a ratio of low hydrophilic-lipophilic balance (HLB) surfactant to high HLB surfactant (LHR) of 1:2. Polyglyceryl-6 laurate as a high HLB surfactant produced the optimal nanoemulsion with small particle size and possessed an encapsulation efficiency (EE) of 74.32 ± 0.19%. This is the first report to combine D-optimal design-based nanoemulsion development with a multi-level analysis of HSF, including phytochemical profiling, antioxidant evaluation, and in silico molecular docking. These findings highlight that HSF-loaded polyglycerol fatty acid ester-based nanoemulsions could be a skin health-promoting ingredient and effective alternative for a variety of skincare applications.

1. Introduction

The burgeoning interest in natural ingredients for cosmetic formulations has led to the exploration of honeysuckle flowers (Lonicera japonica Thunb.), a species belonging to the family Caprifoliaceae [1], as a promising cosmeceutical ingredient. Honeysuckle is a captivating and versatile flowering plant genus that has long been recognized for its remarkable therapeutic properties. In traditional Chinese medicine, Lonicera japonica, known as Jinyinhua, has been widely used to treat inflammatory conditions, infections, and febrile illnesses [2]. Recent studies also highlight the bioactive potential of its flowers, stems, and leaves, which exhibit anti-inflammatory, antioxidant, and antimicrobial properties [1], supporting its relevance for skincare applications. As the cosmeceutical industry continues to evolve, the potential of honeysuckle flower as a valuable ingredient in skincare and personal care products has garnered increasing attention. The chemical composition of honeysuckle flowers is particularly noteworthy, as these blooms are known to be rich in a diverse array of bioactive phytochemicals, including an impressive repertoire of phenolic acids, flavonoids, iridoids, and triterpenoids [1,2,3]. The phytochemicals present in honeysuckle flowers have been associated with a wide range of beneficial biological activities, such as antioxidant, antimicrobial, and anti-inflammatory properties [1,2], which are highly desirable for incorporation into cosmeceutical formulations as they can help protect the skin, mitigate inflammation, and promote overall skin health and rejuvenation [4,5]. Interestingly, the chlorogenic acid and caffeic acid found within honeysuckle have been shown to exhibit enhanced efficacy, which help in mitigating UV-induced skin damage and enhancing skin repair mechanisms [4,5,6]. Honeysuckle flowers are rich in flavonoids such as luteolin, quercetin, and rutin. These compounds exhibit significant antioxidant activity, which is crucial for protecting skin cells from oxidative stress and premature aging [1,2]. The flavonoid content also contributes to anti-inflammatory and skin-soothing effects, further underscoring the remarkable potential of this floral source as a natural and effective alternative for a variety of skincare applications [7,8,9].

The use of lipid-based encapsulation systems has gained significant traction in the field of cosmeceutical ingredient delivery, with nanoemulsions in particular demonstrating immense potential as a versatile and effective approach [10,11,12]. These nano-scale dispersions, comprising one liquid phase dispersed within another immiscible liquid, exhibit unique physicochemical properties that make them well-suited for the encapsulation and targeted delivery of a wide range of bioactive compounds [11]. The notable increase in the attention towards low-energy self-emulsification techniques in nanoemulsion formation has been driven by several key factors, including the transition towards more environmentally friendly “green” technologies that prioritize reduced energy consumption, as well as the growing demand for delivery platforms capable of effectively encapsulating and transporting a diverse array of cosmeceutically relevant active ingredients, while maintaining their stability and bioavailability [12,13,14]. The use of nanoemulsions as delivery systems has garnered significant attention in the cosmeceutical industry in recent years, as they offer a means to improve the efficacy and bioavailability of active ingredients [11,15]. Nanoemulsions are defined as kinetically stable, thermodynamically unstable dispersions of oil and water phases with droplet sizes typically ranging from 20 to 200 nanometers. Compared to traditional emulsions, these nano-scale systems demonstrate enhanced permeability, increased solubilization of lipophilic compounds, and improved stability [11,12], rendering them a promising approach for the development of advanced cosmeceutical formulations.

In recent years, the personal care industry has undergone notable modifications, as cosmeceuticals have emerged as a major and innovative category that effectively combines the functionality of cosmetic products with the sophisticated features commonly associated with pharmaceutical formulations [16,17]. Polyglycerol fatty acid esters (PGFEs), a class of non-ionic surfactants, have emerged as an attractive option for the formulation of nanoemulsions due to their ability to facilitate the spontaneous formation of these nano-scale dispersions through a low-energy emulsification process [18,19,20]. These non-ionic natural surfactants are synthesized by esterifying natural glycerol, which is the hydrophilic component of a surfactant molecule, with fatty acids derived from various natural plant oils as its lipophilic moieties. The incorporation of polyglycerol fatty acid esters into nanoemulsion systems can not only enhance the solubilization and delivery of lipophilic cosmeceutical ingredients, but also offer additional functionalities such as improved stability, biocompatibility, and reduced skin irritation [18,19,20]. Over the last few decades, the esterification of polyglycerol with lauric acid has been extensively studied for their potential use in a variety of applications, including cosmetics and personal care products [21,22]. Polyglycerol fatty acid esters (PGFEs), derived from renewable sources such as plant-based oils and fats, have emerged as a promising class of cosmeceutical ingredients that can address this industry-wide shift towards a more sustainable and more eco-friendly profile.

This study focuses on the development and optimization of PGFE-based nanoemulsions loaded with honeysuckle flower extract (HSF) using a D-optimal experimental design. The formulation process was assessed based on key physical parameters including droplet size, morphology, and encapsulation efficiency. The total phenolic and flavonoid content of the extract were quantified, and antioxidant activity was evaluated. In addition, molecular docking simulations were performed to predict interactions between HSF phytochemicals and key skin-aging and inflammation-related enzymes such as collagenase, hyaluronidase, and TNF-α. To the best of our knowledge, this is the first study to integrate a D-optimal experimental design with low-energy PGFE-based nanoemulsions loaded with honeysuckle flower extract, while concurrently investigating their phytochemical profiles, antioxidant potential, and in silico interactions with skin-aging and inflammatory enzymes. This comprehensive approach not only advances formulation strategies for cosmeceutical applications but also provides mechanistic insights into the bioactivity of HSF-derived phytochemicals, offering a novel route for multifunctional skincare innovations.

2. Materials and Methods

2.1. Chemical Materials

The study utilized the following chemicals and reagents: 2,2-Diphenyl-1-picrylhydrazyl (DPPH) and 2,4,6-tri[2-pyridyl]-s-tria-zine (TPTZ) were purchased from Sigma-Aldrich (Schnelldorf, Germany). FeSO₄·7H₂O and C₂H₉NaO₅ were purchased from Loba Chemie (Mumbai, India). Ethanol, FeCl₃·6H₂O, and methanol were purchased from Merck (Darmstadt, Germany). Deionized (DI) water and NaOH were purchased from RCI Labscan (Bangkok, Thailand). Polyglyceryl-3 polyricinoleate, polyglyceryl-3 laurate, polyglyceryl-4 laurate, polyglyceryl-6 laurate, propylene glycol, and spectrastat BHL were purchased from Chanjao Longevity Co., Ltd. (Bangkok, Thailand).

2.2. Sample Preparation

The dried HSF was purchased from Thaphrachan Herb Co. Ltd., a GMP-certified herbal supplier based in Bangkok, Thailand, during the period commencing in February 2024. The dried HSF was pulverized using a grinder from Panasonic Co., Ltd., Osaka, Japan, to achieve a fine powder. The sample was stored in a tightly sealed container.

2.3. Supercritical Fluid Extraction

A supercritical fluid extraction (SFE) device (SFC-CO₂-4000 analytical system, JASCO Inc., Tokyo, Japan) was employed to extract the HSF. An SFE vessel was used to hold 30 g of HSF powder for each experiment. The extraction was conducted at a pressure of 30 mPa, with the flow rates of CO₂ and ethanol (a co-solvent) both set to 1.0 mL/min. Triplicate extractions were implemented. The HSF extract was collected and stored at 4 °C in a container that was shielded from light until the next experiment [12].

2.4. Qualification of Polyphenolic Compounds

2.4.1. Determination of Total Phenolic Content (TPC)

Using the Folin–Ciocalteu reaction, the TPC in the HSF extract was evaluated, with a slight modification to the method of Yaowiwat et al. [12]. The standard curve in this investigation was established using gallic acid. The TPC was expressed as mg GAE per gram of sample, and the absorbance was measured at 765 nm. All experiments were conducted in triplicate.

2.4.2. Determination of Total Flavonoid Content (TFC)

A method that was slightly modified from the method of Yaowiwat et al. [12] was employed to determine the TFC in the HSF extract. The standard curve was prepared using quercetin. The TFC was expressed as mg QE per gramme of sample, and the absorbance was measured at 510 nm. All experiments were conducted in triplicate.

2.5. UHPLC-ESI-QTOF-MS/MS Analysis

Analysis of the HSF extract was conducted utilizing a UHPLC Agilent 1290 Infinity II System connected to an Agilent 6545 LC-QTOF/MS instrument (Agilent Technologies, Santa Clara, CA, USA) [12]. The separation process was conducted using a Waters XBridge C18 column (Waters Corporation, Milford, MA, USA) with dimensions of 100 mm × 2.1 mm and a thickness of 2.5 μm. The elution was accomplished by employing a binary gradient design, where eluent A consisted of 0.1% formic acid in deionized water and eluent B consisted of 0.1% formic acid in acetonitrile. The gradient sequence consisted of 5−17% B at 0–13 min, 17–100% B at 13–20 min, 100% B at 20–25 min, and 100–5% B at 25–27 min, with a flow rate of 0.3 mL/min. Finally, a post-run was scheduled to stabilize the column for 6 min between each analysis.

The LC-MS system utilized dual Agilent Jet System electrospray ionization (ESI) (Agilent Technologies, Santa Clara, CA, USA)as an interface. The system features specific parameters such as a sheath gas temperature of 250 °C, a sheath gas flow rate of 12 L/min, a gas flow rate of 11 L/min, a gas temperature of 300 °C, and a nebulizer pressure of 45 psig. The LC-MS full-scan mode was configured to function with both positive and negative ionizations. The scan range spanned from 50 to 1050 m/z, with a scan rate of 1 spectrum per second. The Auto-MS2 system was operated with predetermined collision energies of 10, 20, and 40 electron volts. The MS/MS scan range was configured to span from 50 to 1100 m/z, with a scan rate of 3 spectra per second. The isolation width of the MS/MS split was established at ±4 m/z. The reference solutions were included to establish internal reference masses for the calibration of the mass in both positive and negative operational modes.

2.6. Antioxidant Activity Assays

2.6.1. DPPH Radical Scavenging Assay

Radical scavenging activity was assessed using the DPPH assay, which was slightly modified from Yaowiwat et al. [12], where diluted HSF extract concentrations were prepared in ethanol. The standards utilized were gallic acid, ascorbic acid, and quercetin. In brief, the HSF solution was incubated with 167 μM DPPH• in ethanol at ambient temperature for 30 min in the dark. The absorbance was determined at 517 nm using a spectrophotometer microplate reader (SPECTROstar Nano, Ortenberg, Germany). All experiments were conducted in triplicate. The inhibiting effect of the HSF extract was calculated using the following Equation (1):

%inhibition = [(Ac − As)/Ac] × 100

(1)

where Ac is the absorbance of the blank and As is the absorbance of the test sample. The IC₅₀ was then calculated using a calibration curve of the HSF extract by plotting the sample concentration and the % inhibition.

2.6.2. The Ferric Reducing Antioxidant Ability (FRAP) Assay

The FRAP values of the HSF extract were determined compared with the results of a standard ferrous sulfate solution utilizing a technique that was slightly modified from that of Yaowiwat et al. [12]. The HSF extract samples were prepared in ethanol and combined with the FRAP reagent. After that, the mixtures were incubated for five minutes. Absorbance measurement was conducted at 593 nm. The regression equation, which was derived from the standard curve, was employed to determine the FRAP values of each sample. The regression coefficient (R²) of the calibration curve was 0.9999, and it was linear. All experiments were conducted in triplicate.

2.7. Molecular Docking

Molecular docking was employed to investigate the anti-aging and anti-inflammatory properties of the promising compounds, which included Congmunoside XII, Cynaroside, Dipsacoside B, Hyperoside, Madecassoside, and Undulatoside A, as identified through UHPLC-ESI-QTOF-MS/MS analysis. The PubChem database was used to determine the chemical structures of all compounds, which were then subjected to MM2 minimization before docking. The PDB database was used to obtain targets associated with aging (collagenase, elastase, and hyaluronidase) and pro-inflammatory (IL-1β, IL-6, and TNF-α) proteins. The protein structures have been stripped of all co-crystallized ligands and water molecules, with the exception of metal ions. Furthermore, the compounds and targets were assigned hydrogen atoms and potential charges for the purpose of docking preparation. AutoDock Vina [23] was employed to conduct molecular docking. The grid box was defined based on the position of the co-crystallized ligand present in each selected PDB structure. This approach ensures that the docking simulation targets the biologically relevant active site and allows for the accurate prediction of ligand–receptor interactions. Additionally, to ensure the reliability of molecular docking protocol, a redocking simulation was performed as a validation step. In this procedure, the internal ligand was removed from each crystallographic protein structure and subsequently redocked into the same binding pocket under identical docking parameters. The accuracy of the docking process was evaluated by calculating the root mean square deviation (RMSD) between the redocked ligand pose and the original ligand conformation in the crystal structure. All docking setups produced RMSD values of less than 2.0 Å. In this study, the docked compounds with the lowest binding energies were further analyzed for binding modes using Discovery Studio 2021 (BIOVIA DS. Discovery Studio 2021 Client. Dassault Systèmes, San Diego, CA, USA; 2021 and visualized using PyMOL2 (Schrodinger, LLC. The PyMOL Molecular Graphics System, Version 1.8. 2015). Using the ComplexHeatmap package [24] in RStudio software (version 4.2.2) (Team P. RStudio: Integrated Development Environment for R. Posit Software, PBC; 2024.), the lowest binding scores of all bound compounds were clustered and displayed in a heatmap. To categorize the molecular docking results, hierarchical clustering using Euclidean distance as the clustering metric was performed. This approach grouped compounds based on the similarity of their binding affinity profiles across the selected target proteins. The resulting clustering provided a visual and quantitative means of identifying compounds with similar docking behavior and potential biological relevance.

2.8. Preparation of HSF-Loaded Nanoemulsions

The phase inversion composition (PIC) methods for the preparation of the HSF nanoemulsions were partially modified from the method described by Yaowiwat et al. [12]. The oil phase was selected to consist of 1% HSF and Caprylic/Capric Triglyceride, while the continuous phase was selected to have propylene glycol dissolved in deionized water. Polyglyceryl-3 polyricinoleate (PGPR), polyglyceryl-3 laurate (PG-3L), polyglyceryl-4 laurate (PG-4L), and polyglyceryl-6 laurate (PG-6L) were used as the surfactant system in this investigation. In brief, an aqueous phase that contained propylene glycol, spectrastat BHL, and deionized water was carefully incorporated into the oil and surfactant at room temperature using a magnetic stirrer. Stirring was maintained until the nanoemulsion was completely formed.

2.9. Characterization of Nanoemulsions

2.9.1. Droplet Size Analysis

Analysis of nanoemulsion droplet size was performed using a zetasizer (Malvern Instruments Ltd., Malvern, UK) and a slightly modified technique from Yaowiwat et al. [12]. Each formulation was dispersed with water in a 1:100 ratio at a temperature of 25.0 ± 0.1 °C. Each measurement was analyzed according to 3 replicate analyses. Characterization of droplet size (nm) and distribution of size or polydispersity index were provided.

2.9.2. Morphology of Nanoemulsions by Transmission Electron Microscopy (TEM)

The morphology of the HSF nanoemulsions was examined in accordance with the methodology described by Yaowiwat et al. [12]. The contrast of the image was adjusted using 2% phosphotungstic acid, and each sample was prepared in a 300-mesh copper grid. The JEOL JEM-1200 EXII electron microscope (Tokyo, Japan) was employed to analyze the derived sample at 40,000× magnification and 80 kV.

2.9.3. Antioxidant Activity Assay of HSF-Loaded Nanoemulsions

The antioxidant activity of both the HSF-loaded nanoemulsion and the free HSF extract was assessed using the DPPH radical scavenging assay. For comparison, both the nanoemulsion and a 1% w/v HSF solution were tested at equivalent concentrations, and the percentage inhibition of DPPH radicals was measured spectrophotometrically at 517 nm. All samples were analyzed in triplicate, and the results were expressed as mean ± standard deviation.

2.9.4. Entrapment Efficiency

The EE of the HSF-loaded nanoemulsion was examined using a centrifugal filtering device equipped with a 100 kDa molecular weight cutoff filter (Microcon Millipore, Billerica, MA, USA) that had been slightly modified from the method described by Yaowiwat et al. [12]. In brief, the nanoemulsion was injected into the sample reservoir and subsequently subjected to centrifugation at 1500× g at 4 °C for 30 min in order to separate the entrapped and untrapped components. The EE of the HSF-loaded nanoemulsion was determined using the Folin-Ciocalteu reaction to quantify its TPC. The analysis of all experiments was conducted in triplicate.

2.10. Optimal Experimental Design

The optimal (custom) design of RSM was employed in this investigation. The statistical experimental design method was implemented in the design and modeling, and the coefficients were provided using the Design-Expert software version 10.0 (State-Ease, Minneapolis, MN, USA). The investigation was examined using D-optimality along with the quadratic model.

This study investigated the impact of independent factors, specifically X₁ (SOR), X₂ (LHR), and X₃ (Type of High HLB surfactant) on Y₁ (droplet size) and Y₂ (PDI). The coded levels of the independent variables are given in

Table 1. The coded levels of the categoric and numeric factors.

Independent Variables	Type	Level	Coded Level
			−1	0	1
SOR	Numeric	3	1:2	1:1	2:1
LHR	Numeric	3	1:2	1:1	2:1
Type of High HLB surfactant	Categoric	3	PG-3L	PG-4L	PG-6L

. Categoric factor (Type of High HLB surfactant) was selected as PG-3L, PG-4L, and PG-6L. Numeric factors were chosen as SOR and LHR.

2.11. Statistical Analysis

All experiments were performed in triplicate, and the findings were shown as mean ± SD. Statistical analysis was conducted using SPSS Software, Version 17.0 for Windows (IBM Co. Ltd., Armonk, NY, USA). The differences in the data were examined utilizing One-Way ANOVA with Tukey’s multiple comparison method. The level of significance was defined at p < 0.05.

3. Results and Discussion

3.1. Extraction of HSF

The result of the HSF extract in terms of percentage yield was 6.41 ± 1.01%, which presented a viscous semisolid with a dark green color. Among the innovative extraction methods that have gained significant attention in recent years is the use of supercritical fluids, particularly carbon dioxide, to selectively isolate target compounds from Lonicera japonica. The unique characteristics of supercritical fluids, particularly their ability to dissolve and extract a wide range of compounds, have made them a subject of extensive research and industrial exploration across a variety of sectors, including the extraction of bioactive molecules from natural raw materials [25,26,27]. Compared to traditional extraction methods, such as solvent extraction or distillation, supercritical fluid extraction offers several advantages, including relatively short processing times, the ability to extract target compounds while minimizing degradation, and the production of extracts with little or no organic co-solvent [25,26].

Previously, Wu et al. investigated a method of SFE for the extraction of chlorogenic acid from L. japonica flower buds and evaluated the antioxidant activity of the derived extract [28]. This study demonstrates that SFE is a suitable method of quality chlorogenic acid extraction from L. japonica flower buds, making it an efficient alternative to conventional solvent extraction. Furthermore, the results suggest that the flower buds of L. japonica can be a good source of natural antioxidants. The extraction of a diverse array of flower species using supercritical fluid extraction, which includes supercritical carbon dioxide, is described in numerous articles from the literature. The review demonstrates that researchers have been improving their studies by utilizing this appropriate method to ensure that the primary compounds of the raw material can be recovered in a mild condition. Additionally, SFE has demonstrated its significance as an extraction technique for enhancing the selectivity, solubility, and yield of the target compounds in flowers [27].

3.2. Total Phenolic Content and Total Flavonoid Content

The TPC and TFC of the HSF extract were 155.00 ± 1.90 mg of gallic acid equivalents (GAE)/g of extract and 399.55 ± 2.36 mg of quercetin equivalents (QE)/g of extract, respectively. Within the range of phytochemicals found in L. japonica, phenolic compounds and flavonoids have attracted considerable interest due to their significant involvement in the antioxidant properties of the plant. Extensive research has been conducted to elucidate the total phenolic content and total flavonoid content of L. japonica, providing valuable insights into the substantial contributions of these bioactive compounds to the overall antioxidant potential of the plant [1,2]. The phytochemical composition of L. japonica is characterized by a diverse array of phenolic compounds, including phenolic acids, flavonoids, and a wide range of other polyphenolic derivatives. Interestingly, the total phenolic content of L. japonica has been observed to vary considerably, with factors such as the geographical origin of the plant, the extraction method employed, and the developmental stage of the plant during sample collection [1,2]. Phenolic compounds, a diverse group of secondary metabolites found in plants, have been recognized for their remarkable ability to scavenge free radicals, inhibit lipid peroxidation, and modulate cellular signaling pathways. Flavonoids, another class of bioactive compounds found in L. japonica, have also been linked to skin health benefits and possess impressive therapeutic potential [29].

3.3. Identification of Bioactive Composition of HSF Extract

The untargeted assessment of the bioactive compound profile of HSF extract was performed using UHPLC-ESI-QTOF-MS/MS.

Table 2. Bioactive compounds identified in HSF extract by UHPLC-ESI-QTOF-MS/MS.

Compound	RT (min)	Molecular Formular	m/z	Mass (Theoretical)	Mass (Experimental)	Score	Error (ppm)
Phenolic acids
Glucocaffeic acid	8.851	C15H18O9	341.0885	342.0951	342.096	92.24	2.76
Chlorogenic acid	8.961	C16H18O9	353.0891	354.0951	354.0963	95.61	3.37
3-O-Caffeoyl-4-O-methylquinic acid	12.379	C17H20O9	367.1034	368.1107	368.1107	98.49	−0.06
4,5-Di-O-caffeoylquinic acid	16.598	C25H24O12	515.1197	516.1268	516.1269	99.71	0.31
1-Feruloyl-5-caffeoylquinic acid	17.327	C26H26O12	529.1337	530.1424	530.1409	91.07	−2.86
Flavonoids
Quercetin 3-alpha-L-arabionopyranoside-7-glucoside	13.857	C26H28O16	595.1309	596.1377	596.1384	97.06	1.12
Hyperoside	16.068	C21H20O12	463.0895	464.0955	464.0968	96.07	2.95
Luteolin 7-neohesperidoside	16.249	C27H30O15	593.1506	594.1585	594.1579	94.90	−0.96
Luteolin 7-gentiobioside	16.328	C27H30O16	609.1471	610.1534	610.1542	93.65	1.27
Cynaroside	16.663	C21H20O11	447.0939	448.1006	448.1011	99.07	1.23
Undulatoside A	16.804	C16H18O9	353.0885	354.0951	354.0958	98.35	1.9
Tricin 5-glucoside	16.968	C23 H24 O12	491.1196	492.1268	492.1277	87.05	1.87
Myrsinone	19.841	C17H26O4	293.1772	294.1831	294.1844	93.82	4.27
Iridoids
Hydroxyloganin	13.605	C17H26O11	451.1451	406.1475	406.147	99.07	−1.27
6′-O-β-Apiofuranosyl sweroside	13.817	C21H30O13	535.1672	490.1686	490.1692	95.96	1.23
Saponins
Congmunoside XII	17.158	C65H106O32	744.3332	1398.6667	1398.6702	93.86	2.5
Dipsacoside B	17.446	C53H86O22	582.279	1074.5611	1074.5617	98.94	0.55
Congmunoside XV	17.449	C54H88O24	559.2756	1120.5666	1120.5655	96.92	−0.96
Madecassoside	17.591	C48H78O20	973.5012	974.5086	974.5081	96.77	−0.61
Saccharides
Sucrose	1.779	C12H22O11	387.1153	342.1162	342.1171	97.52	2.52

illustrates the identified bioactive compounds together with their retention times, molecular formulas, and molecular weights (m/z). Five phenolic acids in the HSF extract were detected from matching the m/z from the system, along with eight flavonoids, two iridoids, and four saponins (as shown in Table 2).

One of the key bioactive compounds found in this plant is chlorogenic acid, a phenolic acid that has been extensively investigated for its antioxidant and anti-aging capabilities. These properties make them promising candidates for skin health applications as they can help protect the skin from oxidative stress [30,31,32]. Flavonoids, another class of bioactive compounds found in L. japonica, have also been linked to skin health benefits. Cynaroside and hyperoside, flavonoid glycosides isolated from L. japonica, have been shown to possess potent antioxidant and anti-inflammatory properties, which can help mitigate the effects of UV-induced skin damage and promote skin regeneration [33,34,35]. Additionally, the saponin compounds found in this plant, madecassoside, have been reported to exhibit a range of beneficial effects on skin health, including anti-aging and anti-wrinkle properties. Madecassoside has been the subject of growing interest due to its potent antioxidant properties, which can help neutralize the damaging effects of free radicals and reactive oxygen species that are known to accelerate the aging process. These phytocompounds are considered cosmetically beneficial for their role in skin hydration, collagen synthesis, UV protection, and reducing scars [36]. Consequently, the phytochemical compounds reported in the HSF extract were reported to exhibit a variety of potential biological activities, which may be associated with the biological activities of the HSF extract.

3.4. Antioxidant Activities of HSF Extract

In this study, the DPPH radical scavenging assay was utilized to evaluate the antioxidant activity of the HSF extract and comparable standards, while the FRAP assay was employed to assess the reducing capacity. According to the results, the IC₅₀ value of the HSF extract was 1.645 ± 0.040 mg/mL. Among the antioxidants, gallic acid possessed the lowest IC₅₀ value at 0.037 ± 0.020 mg/mL, whereas ascorbic acid had a value of 0.038 ± 0.001 mg/mL and quercetin possessed a value of 0.052 ± 0.001 mg/mL. There was a significant correlation between the concentration of bioactive substances found in the HSF extract and its antioxidant activity. These bioactive compounds, including flavonoids and other phenolic acids, especially chlorogenic acid, have been shown to scavenge free radicals and extensively studied for their potential health benefits [37,38,39].

Another measurement of antioxidant power is the reducing capacity, which could be measured directly using the FRAP method for ferric ion to ferrous ion reduction. In comparison to ascorbic acid (1.550 ± 0.001 mM FeSO₄/g extract), the HSF extract exhibited an effective reducing power of 1.17 ± 0.06 mM (mM FeSO₄/g extract). Previous studies have demonstrated a positive correlation between FRAP values and the total phenolic, total flavonoid, chlorogenic acid, caffeic acid, and quercetin content in L. japonica. Additionally, heating the plant during extraction for 60 min significantly increased the FRAP values [39].

3.5. Binding Affinity of Bioactive Constituents Related to Anti-Aging and Anti-Inflammatory Properties

In order to investigate the potential of active compounds (demonstrated in Figure 1A–F) identified from UHPLC-ESI-QTOF-MS/MS for their anti-inflammatory and anti-aging properties, molecular docking was implemented. Lower docking scores signify enhanced binding affinities of the compounds to their respective targets. As illustrated in Figure 2A, the docking results indicate that all compounds are capable of binding to the selected target proteins with binding scores below −6.0 kcal/mol, indicating the presence of strong affinities for these targets. According to the mean docking scores, the majority of compounds demonstrate potential anti-aging properties through robust binding to hyaluronidase, collagenase, and elastase. The binding affinities of the promising compounds for pro-inflammatory cytokines, such as TNF-α, IL-6, and IL-1β, were found to be moderately strong.

Focusing on the targets that promote anti-aging, madecassoside contributes to the strong binding affinity (−9.3 kcal/mol) with collagenase by forming key hydrogen bonds and interacting with residues G179, N180, L181, E219, and P238 through its glycoside moiety (shown in Figure 2B). Furthermore, madecassoside demonstrates a significant interaction with elastase (−8.3 kcal/mol), establishing hydrogen bonds with residues N99 and R177 through its pentacyclic triterpenoid structure, as well as with H57, S195, S214, and V216 via its glycoside moiety (shown in Figure 2C). The functional relevance of these docking results is well supported by experimental evidence. As reviewed by Tan et al., madecassoside has been extensively studied for its role in wound healing, fibroblast activation, and collagen synthesis [40].

Regarding the binding affinity to pro-inflammatory cytokines, cynaroside exhibits significant binding affinity to IL-1β, measured at −7.9 kcal/mol. The flavone structure of cynaroside interacts with residues M95, K97, and V100 through hydrophobic interactions, while its glycoside forms hydrogen bonds with residues N102 and A115 (shown in Figure 2E). These results suggest that cynaroside may exert anti-inflammatory effects by directly interacting with IL-1β and potentially modulating its biological activity. The predicted interaction is supported by in vitro and ex vivo findings reported by Lee et al., who demonstrated that cynaroside effectively attenuates IL-1β-induced inflammatory responses in primary rat chondrocytes and cartilage explants [41].

Additionally, hyperoside demonstrates a significant binding affinity for IL-6, with a binding energy of −8.3 kcal/mol. Multiple hydrogen bonds with residues V64, P65, G67, D68, and Q93 serve to facilitate this robust interaction (shown in Figure 2F). These interactions suggest that hyperoside could interfere with IL-6 activity and downstream inflammatory signaling pathways relevant to skin aging and inflammation. This molecular prediction is substantiated by experimental findings from Ku et al., who demonstrated that hyperoside significantly suppresses IL-6 expression in lipopolysaccharide (LPS)-stimulated human endothelial cells [42].

As shown in Figure 2D, congmunoside XII has the most robust interaction with hyaluronidase (−10.7 kcal/mol), establishing hydrophobic contacts with residue W324 through its sapogenin moiety and building numerous hydrogen bonds with residues S76, Y85, D126, E131, A132, W141, D142, K144, and Y247 via its glycoside moiety. These interactions collectively enhance binding affinity with the hyaluronidase enzyme. Moreover, it is interesting that congmunoside XII also exhibits a strong affinity for TNF-α (−9.1 kcal/mol). The glycoside moiety forms hydrogen bonds with residues Q61, E116, and P117, while the sapogenin backbone interacts with residues L57 and Y59 through hydrophobic interactions (shown in Figure 2G). These results suggest that congmunoside XII may potentially exert dual anti-inflammatory and anti-aging effects by modulating both cytokine signaling and matrix-degrading enzymatic activity. Although no direct in vitro or in vivo studies on congmunoside XII have been reported to date, its structural classification as a triterpenoid saponin lends biological plausibility to these findings. Triterpenoid saponins are known to inhibit pro-inflammatory mediators and matrix metalloproteinases in various tissue models. The multi-target binding profile observed here highlights congmunoside XII as a promising and novel bioactive candidate that warrants further investigation for inclusion in cosmeceutical formulations targeting inflammation-related skin aging.

The study reveals that the bioactive constituents in HSF extract have a high binding potential for both anti-aging and anti-inflammatory activities, as evidenced by their strong binding to hyaluronidase and TNF-α, according to molecular docking analysis. Through multi-target interactions, the HSF extract that contains these compounds could result in anti-inflammatory and anti-aging effects. Our findings are particularly remarkable in that they provide valuable insights into the molecular mechanisms that define the anti-aging and anti-inflammatory properties. This information will be further developed into a natural and effective alternative for a variety of cosmetic applications, thereby promoting skin health.

3.6. Optimization of Nanoemulsions

3.6.1. Fitting the Model

This investigation implemented a three-factor, three-level D-optimal design, as illustrated in

Table 3. D-optimal design of the experimental study. D-optimal experimental design of the parameters for nanoemulsion preparation in terms of mean values of droplet size and PDI.

Runs	Independent Variables			Dependent Variables
	SOR (X1)	LHR (X2)	Type of High HLB Surfactant (X3)	Droplet Size (Y1)	PDI (Y2)
1	2:1	1:2	PG-3L	123.6	0.2
2	1:2	2:1	PG-3L	912	0.52
3	2:1	2:1	PG-6L	250.5	0.29
4	1:1	2:1	PG-4L	495.3	0.36
5	1:2	2:1	PG-6L	664.2	0.42
6	1:1	2:1	PG-4L	491.3	0.35
7	2:1	1:1	PG-3L	154.2	0.27
8	2:1	1:1	PG-4L	103.5	0.21
9	1:2	1:2	PG-6L	386.2	0.41
10	1:1	1:2	PG-4L	292.3	0.38
11	1:1	1:2	PG-4L	298.4	0.37
12	2:1	1:1	PG-4L	97.5	0.23
13	1:1	2:1	PG-3L	521.3	0.39
14	1:2	1:1	PG-4L	738.2	0.5
15	1:2	1:1	PG-4L	748.2	0.51
16	1:1	1:1	PG-6L	285.3	0.31
17	1:2	1:1	PG-3L	812.6	0.49
18	2:1	1:2	PG-6L	58.9	0.16
19	2:1	2:1	PG-3L	234.1	0.32
20	1:1	1:1	PG-3L	452.9	0.36
21	2:1	1:2	PG-6L	60.9	0.17
22	1:2	1:2	PG-3L	707.3	0.47

, employing 22 runs to establish a correlation between the observed responses and the formulation ingredients. The dependent variables (responses) Y₁ (droplet size) and Y₂ (PDI) were statistically determined to determine the individual and combined effects of three independent variables (factors) of the nanoemulsions fabrication: X₁ (surfactant to oil ratio: SOR), X₂ (the ratio of low hydrophilic–lipophilic balance (HLB) surfactant to high HLB surfactant: LHR), and X₃ (Type of High HLB surfactant).

The quadratic model in Equation (2) was generated by the Design-Expert software (version 7.1; Stat-Ease Inc., Minneapolis, MN, USA). Droplet size and PDI are the dependent variables, and SOR, LHR, and Type of High HLB surfactant are the independent variables. The functional relationship between these factors is shown by the following equation:

$$\begin{array}{c}{Y}_i={\beta }_0+{\beta }_1{X}_1+{\beta }_2{X}_2+{\beta }_{3\left(1\right)}{X}_{3\left(1\right)}+{\beta }_{3\left(2\right)}{X}_{3\left(2\right)}+{\beta }_{12}{X}_1{X}_2+{\beta }_{13\left(1\right)}{X}_1{X}_{3\left(1\right)}\hfill \\ +{\beta }_{13\left(2\right)}{X}_1{X}_{3\left(2\right)}+{\beta }_{23\left(1\right)}{X}_2{X}_{3\left(1\right)}+{\beta }_{23\left(2\right)}{X}_2{X}_{3\left(2\right)}+{\beta }_{11}{X}_1^2+{\beta }_{22}{X}_2^2\hfill \end{array}$$

(2)

where $Y_i$ represents the responses, ${\beta }_0$ indicates a constant, and ${\beta }_i$, ${\beta }_{ii}$, and ${\beta }_{ij}$ are linear, quadratic, and interactive coefficients, respectively.

Table 4. ANOVA of three independent variables for droplet size (nm).

Source	Sum of Squares	df	Mean Square	F-Value	p-Value	Significance
Model	1.499 × 106	11	1.363 × 105	2376.47	<0.0001	*
${\mathrm{X}}_1$-SOR	1.117 × 106	1	1.117 × 106	19,479.19	<0.0001	*
${\mathrm{X}}_2$-LHR	1.240 × 105	1	1.240 × 105	2163.26	<0.0001	*
${\mathrm{X}}_3$-Type of HS	65,579.99	2	32,789.99	571.84	<0.0001	*
${\mathrm{X}}_1{\mathrm{X}}_2$	4424.77	1	4424.77	77.17	<0.0001	*
${\mathrm{X}}_1{\mathrm{X}}_3$	58,123.20	2	29,061.60	506.82	<0.0001	*
${\mathrm{X}}_2{\mathrm{X}}_3$	3896.42	2	1948.21	33.98	<0.0001	*
${\mathrm{X}}_1^2$	6419.20	1	6419.20	111.95	<0.0001	*
${\mathrm{X}}_2^2$	465.35	1	465.35	8.12	0.0173
Residual	573.41	10	57.34
Lack-of-Fit	476.80	5	95.36	4.94	0.0522
Pure Error	96.60	5	19.32
Cor Total	1.500 × 106	21
R2	0.9996
Adjusted R2	0.9992
C.V. %	1.87
Adeq Precision	152.5813

* p < 0.05 = significant.

and

Table 5. ANOVA of three independent variables for PDI.

Source	Sum of Squares	df	Mean Square	F-Value	p-Value	Significance
Model	0.2588	11	0.0235	179.67	<0.0001	*
${\mathrm{X}}_1$-SOR	0.2009	1	0.2009	1534.58	<0.0001	*
${\mathrm{X}}_2$-LHR	0.0076	1	0.0076	58.14	<0.0001	*
${\mathrm{X}}_3$-Type of HS	0.0108	2	0.0054	41.26	<0.0001	*
${\mathrm{X}}_1{\mathrm{X}}_2$	0.0048	1	0.0048	36.34	0.0001	*
${\mathrm{X}}_1{\mathrm{X}}_3$	0.0049	2	0.0025	18.84	0.0004	*
${\mathrm{X}}_2{\mathrm{X}}_3$	0.0061	2	0.0031	23.37	0.0002	*
${\mathrm{X}}_1^2$	0.0004	1	0.0004	2.87	0.1213
${\mathrm{X}}_2^2$	0.0001	1	0.0001	0.6595	0.4356
Residual	0.0013	10	0.0001
Lack-of-Fit	0.0009	5	0.0002	2.27	0.1942
Pure Error	0.0004	5	0.0001
Cor Total	0.9950
R2	0.9894
Adjusted R2	3.27
C.V. %	40.2891
Adeq Precision	3

* p < 0.05 = significant.

display the results of the ANOVA analysis, model adequacy, F-value, and p-value of each term. The final polynomial models were able to accurately explain 99.96% and 98.94% of the dependent variables, respectively, as indicated by the R² values of Y₁ (droplet size) and Y₂ (PDI), which were 0.9996 and 0.9894, respectively. Furthermore, the p-values for the Y₁ and Y₂ models are less than 0.0001 and 0.0001, respectively, which suggests that these final models are statistically significant. The final models were additionally considered adequate, as they met the acceptable criteria for a decent statistical model: the difference between the adjusted R² and the predicted R² is less than 0.2 when R² is greater than 0.9. The lack-of-fit term was not significantly correlated with the pure error of the process, as indicated by the non-significant lack-of-fit.

3.6.2. Effects of Independent Variables on Responses

This investigation implemented a three-factor, three-level D-optimal design, as illustrated in Table 3, employing 22 runs to establish a correlation between the observed responses and the formulation ingredients. The dependent variables (responses) Y₁ (droplet size) and Y₂ (PDI) were statistically determined to determine the individual and combined effects of three independent variables (factors) of the nanoemulsion fabrication: X₁ (SOR), X₂ (LHR), and X₃ (Type of High HLB surfactant).

(1)

Droplet Size.

The mathematical model that demonstrates the substantial correlation between the independent variables under investigation and Y₁ (droplet size) is illustrated below.

$$\begin{array}{c}{\mathrm{Y}}_1\left(\mathrm{d}\mathrm{r}\mathrm{o}\mathrm{p}\mathrm{l}\mathrm{e}\mathrm{t} \mathrm{s}\mathrm{i}\mathrm{z}\mathrm{e}\right)=372.08-275.44X_1+97.53X_2+71.87X_{3\left(1\right)}+11.00X_{3\left(2\right)}-22.71X_1{X}_2-44.56X_1{X}_{3\left(1\right)}\hfill \\ -45.91X_1{X}_{3\left(2\right)}-21.05X_2{X}_{3\left(1\right)}+1.45X_2{X}_{3\left(2\right)}+39.84X_1^2+10.18X_2^2\hfill \end{array}$$

(3)

The model indicates statistical significance for all linear terms (${\mathrm{X}}_1, {\mathrm{X}}_2$, and ${\mathrm{X}}_3$), the interaction terms (${\mathrm{X}}_1{\mathrm{X}}_2, {\mathrm{X}}_1{\mathrm{X}}_3,$ and ${\mathrm{X}}_2{\mathrm{X}}_3$), and quadratic terms (${\mathrm{X}}_1^2, {\mathrm{X}}_2^2$). This has been confirmed by the results of the ANOVA presented in Table 4. The droplet size was primarily influenced by all independent variables (factors) in this significant model (p < 0.0001), including SOR, LHR, and the Type of High HLB surfactant. The particle size decreased as the concentration of surfactant and the concentration of high HLB surfactant were increased. In comparison with PG-3L and PG-4L, the formulation produced a smaller droplet size when PG-6L was employed. The particle size in the nanoemulsions decreased immediately as a result of the increasing surfactant concentration and the high HLB surfactant concentration, as demonstrated in Figure 3A–C. Furthermore, the statistically significant interaction term between ${\mathrm{X}}_1$ and ${\mathrm{X}}_2$ could be interpreted as indicating that the preparation condition containing a high surfactant concentration and a high quantity of high HLB surfactant provided the smallest size of particles. Consequently, the oil/surfactant/water ratio, surfactant blend, and Type of High HLB surfactant are the primary factors influencing the physicochemical parameters.

(2)

Polydispersity Index

The correlation between the independent variables and Y₂ (PDI) is represented by the following significant equation:

$$\begin{array}{c}{\mathrm{Y}}_2\left(\mathrm{P}DI\right)=0.3422-0.1174X_1+0.0211X_2+0.0205X_{3\left(1\right)}+0.0146X_{3\left(2\right)}+0.0235X_1{X}_2+0.0024X_1{X}_{3\left(1\right) }\hfill \\ -0.0251X_1{X}_{3\left(2\right)}+0.0175X_2{X}_{3\left(1\right)}-0.0311X_2{X}_{3\left(2\right)}+0.0096X_1^2+0.0044X_2^2\hfill \end{array}$$

(4)

The PDI is a highly significant physical characteristic that must be considered when developing nanosystems, as it indicates the dispersion of the nanocarrier size distribution. Moreover, the stability, efficacy, appearance, and characteristics of the formulation can be influenced by the PDI attributes of the lipid-based particle. In practice, nanoparticle-based substances are generally accepted with PDI values of 0.2 or lower. The ANOVA results in Table 5 demonstrate statistical significance for all linear variables (${\mathrm{X}}_1, {\mathrm{X}}_2$, and ${\mathrm{X}}_3$), the interaction terms (${\mathrm{X}}_1{\mathrm{X}}_2, {\mathrm{X}}_1{\mathrm{X}}_3,$ and ${\mathrm{X}}_2{\mathrm{X}}_3$). Even though the quadratic factors (${\mathrm{X}}_1^2, {\mathrm{X}}_2^2$) were considered insignificant for PDI, they were included in the final model to provide a high coefficient of determination (R²) of which most of the data fit the regression model. This significant model (p < 0.0001) indicated that all independent variables produced an important effect on the PDI, as demonstrated by the significant linear and interaction terms. The response surface plot for the significant interaction effects confirmed that the PDI decreased with increased surfactant concentration and high HLB surfactant concentration. Furthermore, while employing PG-6L, the formulation produced a reduced PDI value in comparison with PG-3L and PG-4L. The statistically significant interaction between ${\mathrm{X}}_1$ and ${\mathrm{X}}_2$ suggests that the preparation condition with increased surfactant concentration and high HLB surfactant concentration resulted in a low PDI value, as demonstrated in Figure 3D–F.

3.6.3. Optimization of Responses for Nanoemulsions

The optimal nanoemulsions can be produced with a SOR of 2:1, a surfactant concentration of 10.0% w/w, an LHR of 1:2, and a polyglyceryl-6 laurate as a high HLB surfactant. These nanoemulsions contain a minimal droplet size with a PDI of not higher than 0.2. Design-Expert software (version 7.1; Stat-Ease Inc., Minneapolis, MN, USA) was employed to perform numerical optimization, with a maximum desirability of 1.

Table 6. Predicted and actual value of responses at optimized conditions.

Response	Predicted Value	Actual Value	% Prediction Error
Y1; droplet size (nm)	59.831	60.2	0.61
Y2; PDI	0.168	0.17	1.18

illustrates the predicted values for the droplet size, PDI, and % EE that were obtained from the numeric optimization.

As illustrated in Table 6, the mathematically predicted values and the actual values obtained from the experiment were statistically compared. The model was accepted as appropriate because the percentage of prediction error was less than 5% for all responses.

3.7. Application of Optimal Conditions for HSF-Loaded Nanoemulsions

In order to enhance the stability of HSF against oxidation as a natural ingredient that promotes skin health, the nanoemulsions that were produced under optimal conditions were utilized to develop the HSF-loaded nanoemulsions. The HSF-loaded nanoemulsions can be fabricated with a SOR of 2:1, a surfactant concentration of 10.0% w/w, an LHR of 1:2, and a polyglyceryl-6 laurate as a high HLB surfactant. In brief, the nanoemulsion was completely formed by dispersing HSF in the oil phase and subsequently incorporating it into an aqueous phase using a magnetic stirrer. The particulate size and PDI values were obtained at 89.8 ± 1.2 nm and 0.25 ± 0.2, respectively.

3.7.1. Morphology of HSF-Loaded Nanoemulsions

The morphological features of nanoemulsion droplets, including their size, shape, and surface properties, are critical factors that influence the overall performance and stability of these systems. The droplets in the nanoemulsions exhibited a spherical morphology, including a hydrophobic oil core encased by a thin interfacial layer of surfactant. One of the key factors that determines the morphology of nanoemulsion droplets is the type of surfactant used in their formulation. The selection of surfactant is essential, as it can significantly impact the characteristics of the nanoemulsion. Polyglycerol fatty acid esters have emerged as a promising class of emulsifiers for the fabrication of these nano-scale delivery systems due to their ability to facilitate the spontaneous formation of these nano-scale dispersions through a low-energy emulsification process. This investigation revealed that the morphology of the HSF-loaded nanoemulsions, examined using TEM, predominantly exhibited a spherical form surrounded by adsorbed surfactants, as illustrated in Figure 4. The HSF-loaded nanoemulsions were successfully fabricated by polyglyceryl-6 laurate and polyglyceryl-3 polyricinoleate, a class of non-ionic surfactants that are used as surfactant systems. These ingredients have emerged as a promising class of cosmeceutical ingredients that are contributing to the development of more sustainable and environmentally responsible formulations.

3.7.2. Antioxidant Activity of HSF-Loaded Nanoemulsions

To evaluate the antioxidant performance of the developed nanoemulsion, the DPPH radical scavenging assay was employed and compared with that of the free HSF. The IC₅₀ value of the crude HSF extract was determined to be 1.645 ± 0.040 mg/mL, indicating a strong intrinsic antioxidant capacity. Based on this result, 1% w/v of HSF extract was incorporated into the nanoemulsion formulation for further comparison. The antioxidant activity of the HSF-loaded nanoemulsion was evaluated under the same DPPH assay conditions and expressed as percentage inhibition. The nanoemulsion demonstrated 94.7 ± 0.98% DPPH inhibition, whereas the aqueous solution of free HSF extract at the same concentration exhibited a slightly higher inhibition value of 97.1 ± 0.81%. This small difference in antioxidant activity suggests that the nanoemulsion retained the functional integrity of the phenolic compounds during formulation. The slightly reduced radical scavenging efficiency in the nanoemulsion may be attributed to the presence of an oil phase and surfactant barrier, which could moderately limit the immediate availability of phenolic constituents to interact with free radicals in the aqueous assay medium. Nonetheless, the high percentage inhibition observed for the nanoemulsion confirms that the encapsulated form still possesses robust antioxidant properties, making it suitable for cosmetic formulations targeting oxidative stress-related skin damage. Moreover, the added advantages of nanoemulsions, such as enhanced skin permeability, controlled release, and formulation stability, further support the use of HSF-loaded nanoemulsions as effective antioxidant systems in cosmeceutical applications.

3.7.3. Entrapment Efficiency of HSF-Loaded Nanoemulsions

EE is a critical parameter reflecting the capacity of a delivery system to encapsulate bioactive compounds while ensuring their physicochemical stability. The EE of HSF was determined using the TPC method based on the Folin–Ciocalteu assay. This method has been widely applied for the quantification of phenolic-rich plant extracts in nanoemulsion systems and is considered appropriate for encapsulation studies due to its simplicity, reproducibility, and adequate sensitivity in complex matrices. Previous research has demonstrated the effectiveness of this method in nanoformulation contexts. For example, Sari et al. applied the TPC method for curcumin-loaded nanoemulsions and confirmed its ability to accurately reflect the efficiency of encapsulation and release behavior [43]. Similarly, Yaowiwat et al. employed the TPC assay to quantify EE in Camellia sinensis flower extract-loaded nanoemulsions, successfully detecting phenolic compounds with acceptable precision and stability under formulation conditions [12]. While the Folin–Ciocalteu method does not provide compound-specific quantification, its use is widely accepted in evaluating the global encapsulation of phenolic constituents. Considering the high phenolic content of HSF and the primary goal of capturing total phenolic retention within the system, the TPC method offers a reliable, standardized approach for assessing EE. The obtained EE value, measured via TPC, is thus representative of the encapsulation performance and is consistent with findings in similar systems.

In the present study, the optimized HSF-loaded nanoemulsion exhibited an EE of 74.32 ± 0.19%, indicating that the phenolic-rich extract was effectively encapsulated within the oil phase of the nanoemulsion system. This level of EE is consistent with values reported in previous studies involving plant extract-loaded nanoemulsions. For example, Chaiittianan and Sripanidkulchai reported an EE of 67.99 ± 0.87% for a nanoemulsion containing Phyllanthus emblica L. branch extract, which was considered efficient for dermal application [44]. Similarly, Yaowiwat et al. formulated nanoemulsions incorporating Camellia sinensis flower extract and achieved an EE of 72.85 ± 0.21%, using a low-energy emulsification technique [12]. Moreover, Yanasan et al. developed nanoemulsions containing Passiflora quadrangularis L. fruit extract and obtained EE values of 51.38% and 60.30%, depending on the formulation variables [45]. These comparative data suggest that EE values above 70% are widely accepted as satisfactory for nanoemulsion-based delivery systems, particularly when formulated using low-energy processes and natural surfactants. The EE value achieved in this study demonstrates that the polyglycerol fatty acid ester-based nanoemulsion system provides efficient encapsulation of HSF. The result not only supports the physical stability and delivery capacity of the system but also affirms its suitability for use in cosmetic and cosmeceutical formulations.

4. Conclusions

This study underscores the potential of Lonicera japonica flower extract (HSF) as a bioactive cosmeceutical ingredient through the integration of phytochemical profiling, molecular modeling, and formulation engineering. The UHPLC-ESI-QTOF-MS/MS analysis revealed a diverse composition of phenolic acids, flavonoids, and saponins, several of which demonstrated strong theoretical interactions with inflammation- and aging-related molecular targets. These findings provide mechanistic insight into the extract’s therapeutic potential. The development of a low-energy nanoemulsion delivery system, optimized via D-optimal design and response surface methodology, represents a strategic advancement in enhancing the solubility, stability, and delivery efficiency of HSF’s bioactives. The resulting formulation exhibited favorable physicochemical characteristics, indicating its suitability for dermal applications.

Crucially, this work not only establishes the scientific rationale for HSF-based nanoformulations but also presents a replicable platform for translating traditional medicinal ingredients into modern skincare innovations. Moving forward, experimental validation through enzymatic inhibition assays, cellular anti-inflammatory testing, and skin permeation studies will be essential. These efforts will bridge the gap between computational predictions and clinical relevance, enabling the development of safe, sustainable, and efficacious plant-based cosmeceutical products.