Development of a Green Extraction Process from Residues of Assyrtiko Wine Production for Cosmetic Applications

Abstract

Vitis vinifera L. cultivar, “Assyrtiko”, is a famous grape variety native to Santorini island. Its wine production residues are rich in bioactive polyphenols, making them valuable for extraction and use in cosmetics. The aim of this work was the development and optimization of an extraction process from “Assyrtiko” Wine Production Residue (AWPR), using a Natural Deep Eutectic Solvent (NaDES) as the extraction medium. Four NaDESs were synthesized and screened for the extraction, and the extracts were evaluated for Total Phenolic Content (TPC) and Total Flavonoid Content (TFC). The NaDES comprising betaine and 1,3-propanediol was chosen for further analysis because of its effectiveness as an extraction solvent. The extraction process was optimized using a Box–Behnken experimental design. The NaDES %w/w content in the NaDES/water system was found to play the most statistically significant role in the quality of the extracts, assessed via TPC and TFC values. The quality of the extract obtained from the optimal conditions was practically stable with respect to TPC and TFC after long storage, suggesting that NaDESs have a potential “protective” effect for the extracted phytochemicals and give energy-efficient character to the process. This extract was also directly incorporated into a moisturizing cosmetic formulation, which remained homogeneous and stable after testing, demonstrating the extract’s potential for cosmetic applications.

1. Introduction

Vitis vinifera L. cv. “Assyrtiko” is a rare ancient grape variety, indigenous to the island of Santorini, Greece, and inextricably linked to its characteristic volcanic environment [1]. “Assyrtiko” is well-known for the production of Protected Designation of Origin (PDO) Santorini white wines, which are characterized by their high acidity [1]. A biodegradable solid by-product of the wine production process, in general, can be collected after mechanical pressing or fermentation, and it consists of grape peels (skin), seeds and some parts of the stem [2]. It typically contains water (55–75%), polysaccharides (30%), proteins (6–15%), lipids, sugars and unsaturated fatty acids, but is also rich in polyphenols, including phenolic acids, anthocyanidin, proanthocyanidin and other flavonoids, especially catechin and epicatechin [2]. The disposal of wine production waste into soil emerges as a significant environmental issue, because of the phytotoxic and antimicrobial effect attributed to the tannins and polyphenols contained in the residue, as well as its high acidity [2]. Therefore, the need for management of this kind of waste from the winemaking process for the reduction in its ecological impact is crucial [2]. At the same time, the valorization of these residues is critical for their conversion into useful resources of the essential bioactive compounds occurring in the waste [2].

Natural Deep Eutectic Solvents (NaDESs) are characterized as “green” solvents with desirable properties [3]. They are synthesized from naturally occurring biocompatible metabolites such as choline, carbohydrates, amino acids, etc. [4,5,6]. NaDESs are eutectic mixtures of two or more components, and their formation results from an extended and complicated hydrogen bond network created among their ingredients [4]. The eutectic point temperature of the mixture is lower than that of an ideal liquid mixture, declaring important negative deviations from ideality [7]. NaDESs display biodegradability, chemical and thermal stability and dissolving capacity of a plethora of compounds [3]. Their physicochemical properties are adjustable for specific purposes, as different components can be combined at different molar ratios [3,7]. They are promising solvents for application in extraction processes because of the high extraction yields they provide, along with the stabilization of the resulting extracts and protection of the extracted compounds [3]. The final NaDES-extracts can be possibly incorporated “as obtained” (without any further solvent removal and purification steps) in final products [3]. This gives the whole process a more energy-efficient character and reduces its cost [8]. Furthermore, the production of ready-to-use NaDES-extracts is superior to traditional extraction methods in terms of environmental sustainability because of the reduced waste generation. Additionally, it has been proven that plant extracts based on NaDESs display enhanced biological activity compared to extracts obtained by conventional solvents, due to the specific and unique profile of the bioactive compounds present in the NaDES-extracts and the bioactivity of NaDESs themselves [9]. The potential of NaDESs as extraction media has been significantly broadened as they have been successfully used for the extraction of a variety of natural products, such as bioactive components, for the enhancement of drug delivery and also for the pretreatment and valorization of food wastes and by-products [3,7]. They can be used either in simple heating and stirring extraction processes or even combined with high-energy techniques such as ultrasound- or microwave-assisted extractions [9]. NaDESs often outperform conventional solvents in extraction processes due to their tunable physicochemical properties, achieving higher yields for both small and large bioactives [10]. Overall, NaDESs are potential substitutes for hazardous organic solvents, with a reduced environmental footprint, less impact on human health, higher extraction efficiency and selectivity [7]. The challenges and limitations regarding the use of NaDESs are related to their industrial implementation. The scale-up of NaDESs is feasible and promising, and they have been adopted at a larger scale in extraction processes (especially in cosmetics), with their viscosity being the main issue, requiring more energy-intensive procedures. In order for NaDESs to be successfully applied in scaled-up extraction processes, viscosity reduction, appropriate equipment design and economic evaluation need to be considered [9].

Skincare products have been drawing attention for thousands of years, but recently, with environmental consciousness being a significant aspect, the interest in plant-based cosmetics has been growing. In this frame, plant extracts containing bioactive phytochemicals have been employed for this purpose, and also because of their reduced adverse effects, their wide spectrum of action, their high effectiveness and their inexpensiveness. Besides the active ingredients occurring in plants, other constituents, like natural moisturizers, flavorings and pigments existing in them, make those extracts favorable for cosmetic applications. Furthermore, plant extracts generally fulfill the regulatory requirements due to their safety. Among the bioactive compounds present in plant extracts, phenolics seem to attract a lot of interest because of their anti-inflammatory, antimicrobial and antioxidant capacity. Therefore, they are used in cosmetology and dermatology as preventive or therapeutic agents for various skin conditions. More specifically, the strong antioxidant activity of phenolic compounds leads to their ability to reduce collagen degradation and protect from UV radiation, potentially making them suitable for anti-aging applications. Finally, natural extracts rich in antioxidant phenolics are ideal substitutes for synthetic antioxidants in skincare products [11,12]. In this frame, the incorporation of liquid and ready-to-use NaDES-extracts in cosmetic formulations is actually feasible, without drastic changes in rheological properties or sensory characteristics [9]. For example, for the preparation of a high added value cosmetic cream, the cream base is cooled down and then the NaDES-extract is added, in order to avoid thermal degradation during the incorporation.

To our knowledge, the literature regarding the extraction of “Assyrtiko” wine production residues using NaDESs is limited. More specifically, the use of betaine-based NaDESs and the incorporation of the NaDES-extracts in cosmetic formulations have not been investigated until now. Therefore, the aim of the present work was the development and optimization of an extraction process with “Assyrtiko” Wine Production Residue as feedstock (hereafter referred to as AWPR for simplicity), using NaDES as the extraction medium, in order for the resulting extract to be incorporated “as obtained” in an innovative cosmetic cream. More specifically, four NaDESs were initially prepared and used as green extraction media for phenolic and flavonoid compounds from AWPR. For comparison purposes, a system of conventional solvents widely used in this extraction process was also employed. Among the four NaDESs, the optimal extraction solvent was chosen for further investigation and incorporation in the final formulation. The extraction process was then optimized via the Box–Behnken experimental design, allowing the creation of regression equations that can be used as prediction tools for the selected responses (Total Phenolic Content and Total Flavonoid Content of the extracts).

2. Materials and Methods

2.1. Materials and Reagents

For the preparation of NaDESs, anhydrous betaine from Glentham Life Sciences Ltd. (Corsham, UK), D,L-lactic acid (80–85% aq. soln.) from Alfa Aesar (Ward Hill, MA, USA) and anhydrous glycerol from Penta (Prague, Czech Republic) were used. Folin–Ciocalteu reagent was obtained from Merck Millipore (Darmstadt, Germany). For the evaluation of the antioxidant activity of the extracts, 2,2-diphenyl-1-picrylhydrazyl (DPPH), 2,2′-Azobis(2-amidinopropane) dihydrochloride (AAPH) and linoleic acid (technical, 58–74% (GC)) were purchased from Aldrich (St. Luis, MO, USA). For the HPLC analysis, methanol HPLC grade acquired from Thermo Fisher Scientific (Waltham, MA, USA) was used. For the cosmetic cream formulation, beeswax, almond oil, avocado oil, aloe jelly and vegetable-based emulsifier (INCI name: Potassium Palmitol Hydrolyzed Wheat protein, Glyceryl Stearate, Cetearyl Alcohol) were purchased from a local pharmacy. 1,3-Propanediol (registered trade name ZEMEA, supplied by Primient Covation LLC, Loudon, TN, USA) for the preparation of NaDESs and “Assyrtiko” Wine Production Residue (AWPR) as a dried material were provided by KORRES Natural Products S.A (Athens, Greece). The wine residues of “Assyrtiko” were dried under the sun, resulting in the plant material with a content of moisture less than 10% (w/w). For all the experiments, ultrapure water was used.

2.2. Sample Preparation

AWPR was first ground using a domestic coffee grinder and was then filtered through a kitchen sieve (<0.2 mm) for the reduction in its particle size. It was stored in the dark at room temperature until further use.

2.3. NaDES Preparation

The NaDESs were prepared as described in our previous work, using the heating and stirring method, slightly modified [13]. The appropriate amounts of the starting materials for each NaDES were added to a round-bottom flask equipped with a magnetic stirrer. The mixture was stirred for 1–3 h at temperatures between 50 and 70 °C under inert conditions (nitrogen atmosphere), until a homogenous transparent liquid was formed. Then, the NaDES was transferred to hermetically closed glass vessels and stored at room temperature in the absence of light, until further use. The components and their molar ratio used in all four NaDESs prepared are demonstrated in

Table 1. Chemical composition of NaDESs used as extraction solvents.

NaDES	Component 1	Component 2	Component 3
Bet/La/W* (1:2:2.5*)	Betaine	D, L-Lactic acid	Water
Bet/Gly (1:2)		Glycerol	-
Bet/Gly (1:3)
Bet/Prop-1,3 (1:5)		1,3-Propanediol

* The 2.5 eq of water (W*) refers to the quantity of water contained in the commercially available D,L-lactic acid.

.

2.3.1. Betaine/D,L-Lactic Acid (Bet/La/W*) (1:2:2.5*)

For the preparation of the NaDES derived from betaine (Bet) and D,L-lactic acid (La), anhydrous betaine was mixed with D,L-lactic acid (80% aq. soln.) at a molar ratio of 1:2. The 2.5 eq of water (W*) refers to the quantity of water contained in the commercially available D,L-lactic acid. The mixture was stirred for 1 h at 50 °C under inert atmosphere, and a colorless transparent liquid was obtained.

2.3.2. Betaine/Glycerol (Bet/Gly) (1:2)

Anhydrous betaine was mixed with glycerol at a molar ratio of 1:2. The mixture was stirred for 3 h at 70 °C under inert atmosphere, and a colorless transparent liquid was obtained.

2.3.3. Betaine/Glycerol (Bet/Gly) (1:3)

Anhydrous betaine was mixed with glycerol at a molar ratio of 1:3. The mixture was stirred for 2 h at 70 °C under inert atmosphere, and a colorless transparent liquid was obtained.

2.3.4. Betaine/1,3-Propanediol (Bet/Prop-1,3) (1:5)

For the preparation of the NaDES derived from betaine (Bet) and 1,3-propanediol (Prop-1,3), which is the NaDES that was further used for the optimization of the extraction process, anhydrous betaine was mixed with 1,3-propanediol at a molar ratio of 1:5. The mixture was stirred for 2 h at 60 °C under inert atmosphere, and a colorless transparent liquid was obtained.

2.4. Physicochemical Properties of NaDESs

2.4.1. Polarity Measurements

The polarity of the NADESs prepared was estimated using the solvatochromic dye Nile Red, as previously reported [14]. This indicator interacts differently with various solvents, depending on their polarity, which causes changes in its color and light absorption properties. The initial ethanolic stock solution of Nile Red had a concentration of 0.1 mM. The absorption maxima of the solutions (λ_max) were measured using a BioTek Epoch (Santa Clara, CA, USA) 2 microplate spectrophotometer in the 400–700 nm range. The polarity of each NaDES was expressed in the form of the molar transition energy (E_NR), according to Equation (1).

E_NR (kcal·mol⁻¹) = 28,591/λ_max

(1)

2.4.2. pH Measurements

pH of the selected extraction medium containing 80% w/w NaDES and 20% w/w water was measured by a digital pH meter (METRIA M21, Barcelona, Spain) at room temperature (25 ± 5 °C).

2.5. Extraction Process Using NaDES/Water System as Extraction Medium

In a glass vial, a ground sample of AWPR was first suspended in the appropriate amount of NaDES/water system, as water was used along with the NaDES as a co-solvent. The extraction was performed at 45 °C for the selected extraction time, and filtration under vacuum was conducted afterwards for the separation of the extract (liquid phase) from the pulp. All extracts were stored at 4 °C until further analysis. Total Phenolic Content (TPC) and Total Flavonoid Content (TFC) of all extracts were then determined by standard colorimetric procedures. The solid raw material-to-liquid extraction medium ratio (mg/g), the extraction time (min) and the %w/w NaDES content in the NaDES/water system were fixed accordingly each time, because it was indicated by a first series of extraction experiments that those three parameters play a significant role in the phytochemical profile of the extracts.

Firstly, a screening of the prepared NaDESs as extraction solvents was performed under the following conditions: solid-to-liquid ratio of 50 mg/g, extraction time of 60 min and NaDES content of 80% w/w. The TPC and TFC of the extracts were measured, and the optimal NaDES for the extraction was selected. The boundaries of the system were determined via some preliminary extraction experiments, in which the NaDES indicated by the screening procedure was used. Finally, the extraction process was optimized through a series of 15 experiments, using the Response Surface Methodology (RSM).

2.6. Extraction with Conventional Solvents

For comparison reasons, a conventional extraction of ground AWPR using a hydroethanolic solution as the extraction solvent was also implemented. The ethanol–water (80–20% w/w) system was added to a round-bottom flask, and the appropriate amount of solid raw material was suspended in it so that the solid-to-liquid ratio was 50 mg/g, as in the extractions using the four NaDESs during the screening process. Respectively, the mixture was stirred at 45 °C for 1 h using a reflux condenser, and then filtration under vacuum was performed. The solvent was removed through rotary vacuum evaporation.

2.7. Qualitative High Performance Liquid Chromatography (HPLC) Analysis of the Extracts

The identification of the extracted compounds in the extract obtained using the selected NaDES and the hydroethanolic extract from AWPR was based on a developed in-house library of 21 reference standards. A Shimadzu Prominance-i LC-2030C 3D Plus HPLC system (Kyoto, Japan) was used for the analysis, and the chromatographic conditions are shown in

Table 2. Chromatographic conditions of HPLC analysis.

Parameters	Chromatographic Conditons
HPLC	Shimadzu Prominance-i LC-2030C 3D Plus
Detector	LC-2030/2040 PDA Detector
Column	Reverse Phase Spherisorb ODS-2 Column C18 5 µm particle size, L × I.D. = 250 μm × 4.6 mm
Column temperature	30 °C
Detection wavelength	280, 370 nm
Flow rate	1 mL/min
Injection volume	10 μL
Mobile phase	Solvent A: water with 0.2% v/v phosphoric acid Solvent Β: methanol
Gradient program	0 min (90%, 10% B), 9.50 min (75% A, 25% B), 18 min (40% A, 60% B), 25 min (30% A, 70% B), 35 min (90% A, 10% B)

. An 80% w/w aqueous solution of the NaDES-extract was prepared for the analysis, whereas for the extract acquired using the system of conventional solvents, the aqueous solution was 8% w/w. Both samples were filtered using a 0.22 μm Nylon syringe filter before analysis.

2.8. Extraction Process Optimization Using Response Surface Methodology (RSM)

The optimal conditions of the extraction of valuable compounds from AWPR using the selected NaDES were identified by applying Response Surface Methodology (RSM). More specifically, a three-level, 15-run Box–Behnken experimental design (BBD) that included three replicates at the center point was applied, and for the statistical analysis of the results, Design-Expert 12.0 software (Stat-Ease Inc., Minneapolis, MN, USA—Trial version) was employed using the analysis of variance (ANOVA). The factor levels for the BBD of the three-variable system studied here were coded as −1, 0 and +1. The three factors (independent variables) studied were (A) NaDES content in the NaDES/water system (%w/w), (B) solid raw material-to-liquid extraction medium ratio (mg/g) and (C) extraction time (min). Their impact on the chosen responses (dependent variables), which were the extracts’ Total Phenolic Content (TPC) and Total Flavonoid Content (TFC), was examined. The level of statistical significance was set at p < 0.05. The three levels of each variable, standing for the conditions of the conducted experiments, are shown in

Table 3. Raw and coded values of independent variables of extraction process.

No of Experiment	NaDES Content (%w/w)	A	Solid/ Liquid Ratio (mg/g)	B	Extraction Time (min)	C
1	75	0	25	−1	260	+1
2	75	0	50	0	160	0
3	60	−1	25	−1	160	0
4	60	−1	50	0	60	−1
5	75	0	25	−1	60	−1
6	90	+1	75	+1	160	0
7	90	+1	50	0	60	−1
8	90	+1	50	0	260	+1
9	75	0	75	+1	60	−1
10	90	+1	25	−1	160	0
11	75	0	75	+1	260	+1
12	60	−1	75	+1	160	0
13	75	0	50	0	160	0
14	60	−1	50	0	260	+1
15	75	0	50	0	160	0

. The obtained experimental data were then fitted to a second-order polynomial equation that correlated each response to the three factors.

2.9. Colorimetric Determination of Total Phenolic Content (TPC) of Extracts

TPC of the extracts was evaluated through the reducing capacity of the Folin–Ciocalteu reagent, as described in our previous work [13], with slight modifications. The NaDES-extracts were first diluted in water, with the extract-to-water v/v ratios varying between 1:10 and 1.5:10, which emerged after preliminary test measurements for each extract. The TPC of the extract derived from the conventional extraction was also measured, using a 2 mg/g hydroethanolic stock solution (ethanol 80% w/w). 10 μL of the aforementioned solutions was mixed with 0.5 mL of water and 50 μL of the Folin–Ciocalteu reagent. The mixture was then stirred using a Vortex mixer and left for a 5 min incubation in the dark. 150 μL of saturated aqueous Na₂CO₃ solution was then added for pH control, and the final volume of the solution was adjusted to 1 mL by adding 290 μL of water. The mixture was stirred again and left for a 1 h incubation period in the dark at room temperature. The absorbance was measured at 755 nm using a BioTek Epoch 2 microplate spectrophotometer. The above-mentioned procedure was repeated for a blank sample, which contained water in the case of the NaDES-extracts and hydroethanolic solution in the case of the extract obtained from the conventional extraction. The absorbance of the blank was subtracted from that of each sample. Each experiment was performed in triplicate. The TPC of each extract was calculated as gallic acid equivalents (GAE) from a calibration curve by using gallic acid as a standard. It was finally expressed as mg GAE per g of AWPR.

2.10. Colorimetric Determination of Total Flavonoid Content (TFC) of Extracts

TFC of the extracts was evaluated by the aluminum chloride complexation colorimetric method, slightly modified [15]. The NaDES-extracts were diluted in water, with the extract-to-water v/v ratios varying between 1:10 and 1.5:10, which again emerged after preliminary tests. The TFC of the extract derived from the conventional extraction was also measured using a 2 mg/g hydroethanolic stock solution (ethanol 80% w/w). 50 μL of the solutions were mixed with 30 μL of 5% w/v aqueous NaNO₂ solution. The mixture was stirred using a Vortex mixer and left to incubate for 6 min. Afterwards, 60 μL of 10% w/v aqueous AlCl₃ solution was added, the mixture was stirred again and incubated for 5 min. Then, 300 μL of aqueous NaOH solution (1 M) was added, and the final volume of the solution was adjusted to 1 mL by adding 560 μL of water. After stirring and incubating for 15 min in the dark at room temperature, the absorbance was measured at 500 nm using a BioTek Epoch 2 microplate spectrophotometer. The above-mentioned procedure was repeated for a blank sample, which contained only water and hydroethanolic solution for the NaDES-extract and the extract derived from conventional extraction, respectively. The experiments were performed in triplicate. The TFC of each extract was calculated as catechin equivalents (CE) from a calibration curve using catechin as the reference molecule. The results were expressed as mg CE per g of AWPR.

2.11. In Vitro Evaluation of Antioxidant Activity of Extracts

2.11.1. Colorimetric Determination of DPPH Radical Scavenging Ability of the Extracts

The DPPH radical scavenging capacity of the extracts was studied using the procedure described in our previous work [13] with slight modifications. The method is based on the reduction of the stable DPPH radical when it reacts with the antioxidant molecules of the extract. 5 mg of 2,2-diphenyl-1-picrylhydrazyl (DPPH) was accurately weighed and dissolved in 50 mL of pure ethanol. The DPPH solution was kept at 4 °C in the dark until further analysis, but its storage should never last for more than a couple of hours, in order to avoid the self-reduction of the radical.

In a 96-well plate, 100 μL of DPPH solution was added to 100 μL of ethanolic solution of the extract (initial concentration C). The same procedure was then followed for further diluted samples, with concentrations 0.8 C, 0.6 C, 0.4 C and 0.2 C. The initial concentration C of the stock solutions varied between 10 μL/mL and 80 μL/mL, and it was determined after preliminary tests for each extract studied. In the case of the extract obtained from the conventional extraction, a hydroethanolic solution (80% w/w ethanol–20% w/w water) was used for the preparation of the stock solution of the extract instead of pure ethanol. The initial concentration C of this stock solution was 1 mg/mL. The samples were incubated for 30 min in the dark at room temperature, and the absorbance was measured afterwards at 515 nm using a BioTek Epoch 2 microplate spectrophotometer. The above-mentioned procedure was repeated for a blank sample, containing 100 μL of ethanol instead of extract (or 80% w/w ethanol–20% w/w water for the extract obtained from conventional extraction). All the experiments were performed in triplicate. The DPPH radical scavenging activity of the extract was quantified by the IC₅₀ value, which is the concentration of extract that is required to reduce the initial DPPH absorbance by 50%. The calculation of IC₅₀ was performed graphically from the plot of the % inhibition of the DPPH radical versus the extract concentration, and the results were expressed as μL of extract per ml of the final reaction mixture.

2.11.2. Inhibition of Linoleic Acid Oxidation Induced by 2,2′-Azobis(2-amidinopropane) Dihydrochloride (AAPH)

With this antioxidant activity assay, the capacity of the NaDES-extracts for protecting against lipid oxidation was measured in vitro. More specifically, this method aims at the determination of the % inhibition of linoleic acid oxidation, which is initiated by the peroxy radicals generated by the thermal decomposition of 2,2′-Azobis(2-amidinopropane) dihydrochloride (AAPH). The procedure described by Tzani et al. [8] was performed with slight modifications. 14 μL of 16 mM sodium linoleate solution was added to a UV cuvette containing 1.302 mL of a 0.05 M phosphate buffer, pH 7.4. Then, the oxidation reaction was initiated at 37 °C under air by the addition of 70 μL of a 400 mM AAPH solution. Oxidation was carried out in the presence of 14 μL of aqueous solution of the respective extract studied, with a concentration of 11.0–11.5 mg/mL. The oxidation reaction was monitored at 234 nm using a Jasco V-770 UV-Vis/NIR spectrometer (Tokyo, Japan), and the results were expressed as % inhibition of linoleic acid peroxidation.

2.12. Cosmetic Cream Preparation

A water-in-oil (W/O) emulsion with the NaDES-extract obtained from the optimal conditions was prepared according to Tzani et al. [8], with slight modifications. The oil phase consisted of beeswax, vegetable-based emulsifier, almond oil and avocado oil, and was heated up to 80 °C until the beeswax and the emulsifier were melted. The mixture was then removed from the heat, and aloe jelly, which constituted the water phase, was added slowly with constant stirring. The cream formed was left to cool at room temperature. An ultrasonic probe (VCX400 ultrasonic Vibrant Cell (Newtown, CT, USA), probe diameter 6 mm) was used at 160 W for 4 min intermittently in order for the final formulation to be smooth and homogeneous. The NaDES-extract was added at 10% w/w to the cooled emulsion. A control base cream (only with oil and water phases) was also formulated as a blank. All the aforementioned conditions are presented in

Table 4. Composition of cream formulations.

Ingredient	Control Base Cream (Blank)	NaDES-Extract Cream
Almond oil	3 mL	3 mL
Avocado oil	3 mL	3 mL
Beeswax	0.75 g	0.75 g
Vegetable-based emulsifier	0.75 g	0.75 g
Aloe jelly	6 g	6 g
NaDES-extract	-	10% w/w

.

2.13. Cosmetic Cream Characterization

2.13.1. Organoleptic Characteristics

The cream formulations were assessed by various organoleptic characteristics, including physical appearance, texture, phase separation, homogeneity and color, by visual and tactile observation. More specifically, the creams were first placed in clean containers to be inspected for their color, odor, uniformity and potential presence of any foreign particles. A small amount of each cream was taken with a spatula and was rubbed between fingers for the examination of its texture. A small quantity was also applied on the back of the hand in order for its absorbency into the skin to be observed. All the creams were characterized in daylight and at room temperature.

2.13.2. pH Measurement

pH of the creams was measured by a digital pH meter (METRIA M21) at room temperature (25 ± 5 °C). 0.5 g of cream was dispersed in 50 mL of water for the measurement.

2.13.3. Physical Stability of Cream

The formulations were subjected to a centrifugation test (7000 rpm for 40 min) and three alternate freeze–thaw cycles (at room temperature, 37 °C and −6 °C for 24 h each) to observe any phase separations or changes in appearance under stress conditions.

2.14. Release Studies of the NaDES-Extract from the Cream

The release study of the NaDES-extract from the cream formulation was conducted using the dialysis membrane method [8]. 5 g of cream with 10% w/w NaDES-extract was loaded into a dialysis tubing with a MWCO of 12–14 kDa (SERVA Electrophoresis GmbH, Heidelberg, Germany), which was immersed in 20 mL of phosphate buffer solution, pH 5.5, at 32 °C under magnetic stirring. At the time intervals 20, 60, 120 and 180 min, 1 mL of the dissolution medium was withdrawn and replaced with an equal volume of fresh medium. The samples gained were evaluated in vitro for their radical scavenging activity via DPPH assay. Again, a 96-well plate and a BioTek Epoch 2 microplate reader were used. 100 μL of the samples with released extract for each time interval was mixed in the plate with 100 μL of DPPH solution 0.1 mg/mL. The samples were incubated for 15 min in the dark at room temperature, and the absorbance was then measured at 515 nm. The above-mentioned procedure was also performed for a control sample, containing 100 μL of phosphate buffer solution instead of the released extract. All the experiments were conducted in triplicate, and the results were expressed as % of DPPH radical scavenging activity.

3. Results and Discussion

3.1. Task-Specific Design of NaDESs

One of the most significant characteristics of NaDESs is their design flexibility, meaning they can display desirable properties in specific applications depending on the selection of their constituents [8]. In the present work, the hydrogen bond donors and the hydrogen bond acceptors for the preparation of the NaDESs were task-specifically selected, in terms of their biocompatibility, safety and moisturizing capacity, in order for them to be directly incorporated in the final cosmetic formulations. All the ingredients employed herein are widely used in food, pharmaceutical and cosmetic industries.

More specifically, betaine is a naturally occurring quaternary ammonium salt, often found in whole grain products, spinach, red beetroot and quinoa, which finds application in cosmetic and skincare products due to its moisturizing properties [8,16]. Lactic acid is an alpha-hydroxy acid, produced by microbial fermentation of sugars and used as a flavoring additive in food and beverages, but also in personal hygiene products for pH adjustment and moisturization [8,17]. Glycerol is present in all natural fats and oils as fatty esters and is the most broadly used humectant in moisturizers for the improvement of skin surface hydration [18,19]. Moreover, it can enhance skin-barrier regeneration and maintain the integrity and stability of stratum corneum [19]. Hydration improvement and moisturizing ability are also properties of 1,3-propanediol, an alternative to 1,2-propanediol, which has a petrochemical origin, whereas 1,3-propanediol can be produced from renewable feedstocks (glycerol, sugars and other carbon sources) using microorganisms [20,21,22]. 1,3-propanediol improves skin barrier function and makes skin smoother, prevents water loss from skin tissue and even enhances hair moisture [20,21]. All in all, the task-specific selection of the individual NaDES components was performed so that their desirable skin-protective and moisturizing properties can hopefully enhance the final ready-to-use extract and the cosmetic formulation synergistically. In addition, the employment of NaDESs based on the aforementioned components enhances the green character and sustainability of the extraction process. Non-toxic, broadly available and biodegradable raw materials are used for NaDES preparation, which is an easy and scalable procedure (assuming the viscosity of NaDESs is taken into consideration). The extraction process has potential in being scaled up as well. Furthermore, it requires low temperatures to be conducted and thus lower energy consumption, which also results from the absence of solvent removal steps.

3.2. NaDES Physicochemical Characterization and Screening for Use as Extraction Media from AWPR

All four task-specifically designed NaDESs were preliminarily tested for their effectiveness as extraction solvents from AWPR. The extraction conditions selected for the screening were as follows: solid-to-liquid ratio of 50 mg/g, extraction time of 60 min and NaDES content of 80% w/w at 45 °C. All the extracts were evaluated in terms of their TPC and TFC, and the results are shown in

Table 5. NaDES screening as extraction solvents from AWPR for direct use in cosmetic formulations.

NaDES	TPC (mg GAE/gAWPR) of Extract	TFC (mg CE/gAWPR) of Extract	λmax (nm)	ENR (kcal mol−1)
Bet/La/W* (1:2:2.5*)	15.9 ± 1.5	12.1 ± 1.4	572	49.98
Bet/Gly (1:2)	22.0 ± 2.0	6.4 ± 0.1	568	50.34
Bet/Gly (1:3)	22.2 ± 1.4	9.4 ± 0.3	570	50.16
Bet/Prop-1,3 (1:5)	52.4 ± 1.0	14.4 ± 0.9	564	50.69

* The 2.5 eq of water (W*) refers to the quantity of water contained in the commercially available D,L-lactic acid. λ_max (water) = 589 nm, E_NR (water) = 48.54 kcal mol⁻¹, λ_max (ethanol) = 547 nm, E_NR (ethanol) = 52.27 kcal mol⁻¹.

, along with the polarity values of the four NaDESs used. Polarity is a crucial physicochemical property of the NaDESs and is considered a key parameter for extraction applications of NaDESs. The solubility of target molecules in NaDESs and thus the extraction efficiency are strongly dependent on the polarity of the solvents [6].

As for the polarity of NaDESs, in the Nile Red polarity scale, bathochromic shifts in the λ_max to higher wavelengths declare higher polarity of the tested solvents. In other words, low E_NR values indicate higher polarity, and higher E_NR means lower polarity of the solvent. According to the results of this study, all the NaDESs prepared were less polar than water (shown from their E_NR > 48.54 kcal mol⁻¹) and more polar than ethanol (E_NR < 52.27 kcal mol⁻¹). The NaDES with the highest polarity among the four tested NaDESs was Bet/La/W* (1:2:2.5*), whereas Bet/Prop-1,3 (1:5) seems to be the least polar among the four tested NaDESs.

The results indicated that the highest TPC and TFC values were achieved by the least polar NaDES Bet/Prop-1,3 (1:5), declaring the efficiency of this NaDES as an extraction solvent of phenolics and flavonoids from AWPR. A significant factor that needs to be taken into consideration, since the optimal extract is destined to be incorporated “as obtained” (without removal of the NaDES) in a cosmetic cream, is pH. Highly alkaline or strongly acidic pH values are not preferable for skincare products in order for skin health to be maintained [23]. It is noted that the aqueous system containing 80% w/w NaDES Bet/Prop-1,3 (1:5), which was used in the screening procedure as the extraction solvent, had a pH value of 7.33 ± 0.09. Finally, this NaDES is characterized by relatively low viscosity, which provides easier and more convenient handling. The high viscosity of NaDESs is a major property that must be taken into consideration, especially if the process is about to be scaled up [7]. Thus, the NaDES Bet/Prop-1,3 (1:5) was selected to be further investigated as an extraction medium for the optimization of the proposed methodology.

3.3. Comparison with Results Obtained from the Use of Conventional Extraction Solvents

For the sake of comparison, an extraction with an ethanol 80% w/w–water 20% w/w system in the same conditions as those selected for the screening process was performed. The extract was evaluated in terms of its TPC and TFC, and they were found to be 22.3 ± 0.3 mg GAE/g_AWPR and 19.8 ± 1.9 mg CE/g_AWPR, respectively. In the case of the extract based on the NaDES selected for the study, Bet/Prop-1,3 (1:5), TPC was higher than in the case of the extract derived from conventional extraction. Glycerol-derived NaDESs seemed to have equivalent extraction efficiency with the hydroethanolic solution used in the conventional extraction. The latter appears to be more efficient in providing extracts with higher TFC.

Overall, NaDESs can be considered capable and competitive alternates over conventionally used extraction solvents. It is suggested that the extensive hydrogen bond network created between the constituents of the NaDESs is responsible for their capacity as extraction media, since it facilitates the dissolution of a plethora of bioactive compounds [14]. Moreover, this network improves the stability of the extracted compounds, enhancing extraction efficiency [14]. One of the major advantages of using NaDESs as extraction solvents and incorporating the extracts “as obtained” in the final cosmetic products is the potential stabilization and protection of the sensitive bioactive phytochemicals. In other words, NaDESs are promising biocompatible storage agents for the valuable molecules recovered from AWPR and other natural resources [13,24]. Since the NaDES Bet/Prop-1,3 (1:5) was chosen for further use in the present work, a DPPH assay was performed for both the extract obtained from this NaDES during the screening process and the conventional extract. The NaDES-extract showed an IC₅₀ value of 2.6 ± 0.1 μL_extract/mL, and the hydroethanolic extract of 9.5 ± 0.1 μL_extract/mL, after 30 min of incubation. It has become clear that the DPPH radical scavenging activity of the NaDES-extract is stronger than that of the conventional one, according to the lower IC₅₀ value corresponding to the NaDES-extract.

The inhibition of linoleic acid peroxidation by the NaDES-extract was determined using the AAPH assay in order to obtain a deeper insight into its antioxidant capacity in terms of scavenging the peroxy radicals generated by AAPH. It was found that the NaDES-extract accomplished 41.9 ± 5.0% inhibition at a concentration of 115 μg/mL.

These results serve as initial quantitative measures of the antioxidant potential of the extracts studied, reflecting the promising capability of the NaDES Bet/Prop-1,3 (1:5) to extract antioxidant phytochemicals more efficiently than the traditionally used hydroethanolic solution. It is important to mention that the pure NaDES Bet/Prop-1,3 (1:5) did not show antioxidant activity via both assays, so the potential antioxidant capacity of the NaDES-extract is attributed to the extracted phytochemicals only, and not the NaDES.

3.4. Qualitative High Performance Liquid Chromatography (HPLC) Analysis of the Extracts

In order to achieve a deeper insight into the compounds extracted from AWPR, HPLC analysis of the extract obtained using the NaDES Bet/Prop-1,3 (1:5) at screening conditions (solid-to-liquid ratio of 50 mg/g, extraction time of 60 min and NaDES content of 80% w/w) was applied. The chromatogram of the NaDES-extract is depicted in Figure 1. An indicative identification of 15 phenolic compounds in the extract was accomplished, most of which were phenolic acids and their derivatives.

The same analysis was conducted for the phytochemical profile of the extract acquired, using the hydroethanolic solution as extraction solvent, at the following conditions: solid-to-liquid ratio of 50 mg/g, extraction time of 60 min, 80% w/w ethanol–20% w/w water. The chromatogram is presented in Figure 2. In this case, 16 phenolic compounds were identified.

Interestingly, both HPLC chromatograms recorded at 280 nm showed an elevated baseline at elution times between 15 min and 23 min, approximately, on which some sharp characteristic phenolic peaks are superimposed. This broad diffuse “hump” is possibly attributed to the occurrence of tannins in the extracts [25,26,27,28,29], which are probably not completely separated during the elution process.

3.5. Preliminary Extraction Experiments

A series of preliminary extraction experiments using the NaDES Bet/Prop-1,3 (1:5) was carried out in order to select the margins of the extraction parameters that should be used in the experimental design. The TFC of the extracts obtained was the criterion for the investigation. The results of those experiments are presented in

Table 6. Preliminary extraction experiments using the NaDES Bet/Prop-1,3 (1:5).

No. of Preliminary Experiment	NaDES Content (%w/w)	Solid/Liquid Ratio (mg/g)	Extraction Time (h)	TFC (mg CE/g)
1	80	50	1	14.4 ± 0.9
2	80	50	4	20.8 ± 1.1
3	80	5	1	14.3 ± 0.5
4	50	50	1	18.1 ± 0.4

.

The parameters under investigation were initially selected as important factors according to our previous experience in NaDES extractions, and they were as follows: NaDES content in the NaDES/water system used as extraction solvent, solid raw material-to-liquid extraction medium ratio and extraction time [8,13,14].

According to the presented data, extended extraction time seemed to enhance the flavonoid content of the extract, so this parameter was further studied within a slightly broader range in the experimental design (60 min–260 min). As for the NaDES content, its decrease from 80% w/w to 50% w/w led to an increase in the TFC of the extract. In other words, the addition of 50% w/w water in the extraction medium favored the extraction of flavonoids, which is in agreement with previous analogous works [14]. In fact, water is usually added as a co-solvent in NaDES extractions to improve the extraction performance because of the resulting reduced viscosity of the extraction medium [13]. Thus, this factor was examined in the experimental design, but in a range of higher values (NaDES 60–90% w/w), in order to study the effect of the NaDES being the main component of the extraction medium. Furthermore, it is known that water in high percentages in NaDESs might weaken or even disrupt the hydrogen bond network of NaDESs [9,30], which could affect the capacity of the NaDESs to act as storage agents for the extracted phytochemicals. Finally, the ratio of solid raw material-to-liquid extraction solvent did not seem to significantly affect the TFC of the extracts in the investigated range, but it deserves to be further studied as an extraction parameter with higher values (25–75 mg/g) to see if it leads to important changes in phenolic and flavonoid content of the extracts.

3.6. Experimental Design for the Extraction Process Using NaDES

After the performance of the preliminary experiments, Response Surface Methodology (RSM) was implemented for the optimization of the extraction process, and, more specifically, the Box–Behnken design was employed. Then, an ANOVA test was conducted for the statistical analysis of the experimental data. The parameters under investigation were as follows: (A) NaDES content in the NaDES/water system (%w/w), (B) solid raw material-to-liquid extraction medium ratio (mg/g) and (C) extraction time (min).

For each extraction, the experimental values of the responses TPC (R₁), expressed as mg GAE/g of AWPR, and TFC (R₂), expressed as mg CE/g of AWPR, are presented in

Table 7. Experimental data of Box–Behnken design.

No. of Experiment	Factors			Responses
	A: NaDES Content (%w/w)	B: Solid/ Liquid Ratio (mg/g)	C: Extraction Time (min)	R1: TPC (mg GAE/g)	R2: TFC (mg CE/g)
1	75	25	260	40.7 ± 3.3	25.2 ± 0.6
2	75	50	160	37.3 ± 0.3	24.7 ± 1.7
3	60	25	160	38.7 ± 1.3	26.4 ± 1.0
4	60	50	60	36.7 ± 2.8	20.2 ± 1.2
5	75	25	60	36.6 ± 0.8	22.1 ± 0.6
6	90	75	160	22.3 ± 0.8	14.5 ± 0.9
7	90	50	60	24.1 ± 0.4	14.5 ± 1.3
8	90	50	260	31.0 ± 0.1	22.8 ± 0.1
9	75	75	60	33.5 ± 2.7	18.6 ± 0.1
10	90	25	160	22.8 ± 1.4	16.0 ± 1.5
11	75	75	260	36.0 ± 0.4	25.7 ± 0.7
12	60	75	160	37.3 ± 1.2	20.7 ± 0.5
13	75	50	160	35.5 ± 0.6	20.7 ± 0.9
14	60	50	260	45.0 ± 1.6	30.3 ± 0.5
15	75	50	160	35.1 ± 0.5	25.4 ± 0.5

.

The statistical analysis of the 15 runs revealed that the TPC of the extracts is best described by the following actual and coded equations of the reduced quadratic model (Equations (2) and (3), respectively):

TPC = −32.8473 + 2.2514 NaDES content + 0.2068 Solid/liquid ratio − 0.0473 Extraction time − 0.0182 NaDES content² − 0.0026 Solid/liquid ratio² + 0.0002 Extraction time²

(2)

TPC = 35.97 − 7.19A − 1.21B + 2.73C − 4.10A² − 1.60B² + 2.33C²

(3)

It is noted that the actual equation is used for the predictions of the response values, while the coded equation is useful for understanding how each factor influences each response by comparing their respective coefficients [8].

The proposed model was significant and valid, with the F-value equal to 38.72 and the p-value less than 0.0001. Furthermore, the coefficient R² was 0.9667, which indicates accuracy and a good fit of the model to the experimental data. The adjusted determination coefficient R²_adjusted confirmed the adequacy of the model, with a value of 0.9417. The R²_predicted was 0.8749, which was in agreement with R²_adjusted (they have a difference of less than 0.2), declaring that the model is able to reasonably predict future data (

Table 8. ANOVA test results for response R₁ (TPC).

	Model	Lack-of-Fit	A	B	C	A2	B2	C2
p-value	<0.0001	0.3582	<0.0001	0.0618	0.0012	0.0011	0.0883	0.0221
F-value	38.72	2.09	165.42	4.71	23.78	24.79	3.76	8.02

Model: R² = 0.9667, R²_adjusted = 0.9417, R²_predicted = 0.8749 and Adeq precision = 20.5948.

). The Adeq precision value indicates an adequate signal in comparison with the noise (it should be higher than 4).

It is shown that factor A, referring to NaDES %w/w content in the NaDES/water system used as an extraction solvent, is the most statistically significant term of the model, proven by its low p-value (<0.0001). In other words, it is the factor that seems to contribute the most to the effective extraction of phenolic compounds from AWPR.

The correlation between the studied factors and the TPC of the extracts is depicted in the 3D surface response plots (Figure 3).

The 3D plots presented above reveal that TPC was notably enhanced by NaDES content varying approximately between 60 and 70% w/w (Figure 3a,b), as illustrated by the red areas of the response surfaces, and, when solid-to-liquid ratio is kept constant at 25.4 mg/g, extraction durations exceeding 210 min favor the extraction of phenolics, leading to higher TPC (Figure 3b). Solid-to-liquid ratio did not seem to have significant influence on TPC, but lower values of this factor (25–45 mg/g) are preferable (Figure 3a).

As for the second response studied, TFC, the statistical analysis of the results provided the following actual and coded equations of the reduced quadratic model (Equations (4) and (5), respectively) for its description:

TFC = −33.3142 + 1.5158 NaDES content + 0.2204 Solid/liquid ratio + 0.0358 Extraction time − 0.0118 NaDES content² − 0.0027 Solid/liquid ratio²

(4)

TFC = 24.17 − 3.72A − 1.28B + 3.58C − 2.65A² − 1.70B²

(5)

The proposed model was significant with the F-value equal to 12.58 and the p-value 0.0008. The coefficient R² was 0.8749, declaring a satisfactory fit of the model to the data. The R²_predicted was 0.6685, which was in agreement with R²_adjusted value 0.8053, proving that the model can be used for future predictions with safety (

Table 9. ANOVA test results for response R₂ (TFC).

	Model	Lack-of-Fit	A	B	C	A2	B2
p-value	0.0008	0.7724	0.0006	0.1102	0.0008	0.0336	0.1428
F-value	12.58	0.5429	26.79	3.14	24.67	6.28	2.58

Model: R² = 0.8749, R²_adjusted = 0.8053, R²_predicted = 0.6685 and Adeq precision = 11.3403.

). Adeq precision is indicative of an adequate signal once again.

NaDES %w/w content in the NaDES/water system (factor A) was once more the parameter contributing the most to the extraction of flavonoids from AWPR (p-value 0.0006), followed by extraction time (p-value 0.0008). The 3D surface response plots for TFC are demonstrated below (Figure 4a,b).

Regarding the 3D plots for TFC, they highlight the significant role of NaDES content, particularly for values between 60 and approximately 70% w/w, which lead to the enhancement of TFC (Figure 4a,b). Extended extraction times were associated with a linear increase in total flavonoids, especially for values over 230 min (Figure 4b), when the solid-to-liquid ratio was kept constant at 25.4 mg/g. Once again, lower values of the solid-to-liquid ratio favor higher TFC (Figure 4a), and, more specifically, for an extraction time of 260 min, 25–55 mg/g seems to accomplish a higher TFC.

The optimization criterion of this study was the simultaneous maximization of the TPC and TFC of the extracts. It was indicated that the optimal extraction conditions within the studied boundaries of the experimental design were as follows: 60.4% w/w of NaDES in the NaDES/water system, 25.4 mg of solid raw material/g solvent and 260 min. The TPC and TFC of the extract obtained from the application of these conditions were evaluated, and the experimental values are presented in

Table 10. Experimental TPC and TFC of extract obtained from optimal extraction conditions.

NaDES Content (%w/w)	Solid/Liquid Ratio (mg/g)	Extraction Time (min)	TPC (mg GAE/gAWPR) of Extract	TFC (mg CE/gAWPR) of Extract
60.4	25.4	260	39.9 ± 0.1	20.7 ± 0.4

.

The validity of the proposed models was verified with the conduction of an additional experiment at random conditions within the design space (63.8% w/w of NaDES, 26.9 mg of solid raw material/g solvent and 210 min), the results of which are presented in

Table 11. Confirmation of the estimated models.

NaDES Content (%w/w)	Solid/Liquid Ratio (mg/g)	Extraction Time (min)	Response	Value Predicted by Model	95% Prediction Interval (Low)	Experimental Mean Value	95% Prediction Interval (High)
63.8	26.9	210	TPC	40.8	36.6	40.6	44.9
			TFC	27.0	21.7	24.7	32.2

, along with the 95% low and high prediction intervals estimated. Since the experimental values of both TPC and TFC are contained in the prediction intervals, the validity of the models was confirmed.

3.7. Stability Studies of NaDES-Extract After Long Storage

The NaDES-extract obtained from the optimal extraction conditions and the conventional extract were assessed again after three months of storage in terms of their TPC and TFC in order to obtain an indicative overview of the possible stabilization capacity of the NaDES used. The results are demonstrated in

Table 12. Stability studies of NaDES-extract and conventional extract after three months of storage in terms of TPC and TFC.

Extraction Solvent	TPC (mg GAE/gAWPR) on 1st Day	TPC (mg GAE/gAWPR) After Three-Month Storage	TFC (mg CE/gAWPR) on 1st Day	TFC (mg CE/gAWPR) After Three-Month Storage
60.4% w/w NaDES–39.6% w/w water	39.9 ± 0.1	41.0 ± 0.4	20.7 ± 0.4	19.8 ± 0.5
80% w/w ethanol–20% w/w water	22.3 ± 0.3	14.4 ± 0.2	19.8 ± 1.9	8.5 ± 0.7

.

It was proven that the quality of the NaDES-extract was practically maintained in terms of its TPC and TFC after three months, which reinforces the idea that the extracted phytochemicals were protected by the NaDES. On the contrary, in the conventional extract, a decrease of 35.4% in TPC and of 57.1% in TFC was observed, which serves as further evidence of our suggestion.

In addition, the capability of the NaDES Bet/Prop-1,3 (1:5) to act as a storage medium for the extracted bioactive compounds from AWPR is proven by the constancy of the IC₅₀ value of the NaDES-extract obtained from the optimal conditions, according to the DPPH assay, after three months of storage in the dark. More specifically, the IC₅₀ value of the NaDES-extract on 1st day was 13.7 ± 0.2 μL_extract/mL, while on the 90th day it was measured 13.5 ± 0.1 μL_extract/mL.

It is worth noting that the inhibition of linoleic acid peroxidation by the NaDES-extract obtained from the optimal extraction conditions was determined using the AAPH assay on the 1st day in order to complement its antioxidant profile. It was found that, at a concentration of 110 μg/mL, it achieved 41.8 ± 3.5% inhibition.

3.8. Cosmetic Cream Preparation and Characterization

The cream with incorporated NaDES-extract (obtained from the optimal extraction conditions), as well as the blank cream, were examined for their organoleptic properties. Both formulations had a cosmetically appealing appearance, white color, smooth and homogenous texture, with no signs of phase separation (Figure 5). They were not greasy and were easily absorbed by the skin. They also had a pleasant scent.

The pH values of the aqueous dispersions of the creams right after their preparation were found to be slightly alkaline (

Table 13. pH values of the aqueous dispersions of the cream containing the NaDES-extract and the blank cream.

Cosmetic Formulation	pH of Aqueous Dispersion
Blank cream	7.68 ± 0.07
Cream with NaDES-extract	7.71 ± 0.09

). The values obtained, since they reflect the pH of aqueous dispersions, do not represent the exact skin-contact pH of the creams; they only give an approximation of the behavior of the products on the skin. The pH of a W/O cream cannot be reliably measured without disrupting the emulsion, and this is the reason why the creams were first dispersed in water for the pH measurement.

Regarding the physical stability of the developed formulations, they remained homogeneous after 7000 rpm centrifugation for 40 min. Moreover, after three cycles of freeze–thaw testing at room temperature, 37 °C and −6 °C, for 24 h at each temperature, no significant changes in texture, viscosity or appearance were observed. Therefore, the prepared creams were considered physically stable under stress conditions.

3.9. Release Studies of the NaDES-Extract from the Cream

A release experiment was performed in order to investigate the DPPH radical scavenging activity of the cosmetic cream with the incorporated NADES-extract at 10% w/w content. At specific time intervals, a sample of the release medium (32 °C, pH 5.5) was withdrawn and assessed using the DPPH assay. The results are displayed in

Table 14. DPPH radical scavenging activity of the released NaDES-extract from cosmetic cream.

NaDES-Extract Release Time (min)	DPPH Radical Scavenging Activity (%)
20	14.0 ± 1.0
60	15.8 ± 2.4
120	16.3 ± 0.8
180	17.9 ± 1.8

. It is noted that the control base cream (blank) was also tested with the same assay, and it did not show any radical scavenging activity.

The DPPH radical scavenging activity achieved by the released extract was moderate and showed a slight increase over release time, with no significant change. This proves that the formulation containing the NaDES-extract from AWPR offers gentle antioxidant protection in terms of DPPH radical scavenging, in addition to the moisturizing effect provided by its constituents. In other words, its soothing capacity for the skin is complemented by a mild radical scavenging activity, thus the cream seems promising in maintaining overall skin health.

4. Conclusions

The extraction of bioactive compounds from “Assyrtiko” Wine Production Waste (AWPR) using Natural Deep Eutectic Solvents (NaDESs) was investigated and optimized in the present work. Among the tested solvents, the NaDES Bet/Prop-1,3 (1:5) was identified as optimal in terms of the Total Phenolic Content (TPC) and Total Flavonoid Content (TFC) of the extracts. For comparison reasons, a conventional extraction using a hydroethanolic solution as the extraction solvent was also investigated. The NaDESextract and the extract derived by the conventional extraction were assessed regarding their TPC, TFC and DPPH radical scavenging activity. The NaDES was proven to be more effective in the extraction of phenolics from AWPR, and the NaDES-extract showed higher DPPH radical scavenging capacity than the one obtained from the conventional extraction. The extraction process using the selected NaDES was optimized, implementing a Box–Behnken experimental design. The extraction time, the AWPR-to-solvent ratio and the NaDES content in the NaDES/water system used as the extraction solvent were studied for their impact on the TPC and TFC of the NaDES-extracts. The reliability of the prediction model obtained for the extraction process was verified with the appropriate confirmation experiments, and the most statistically significant factor for the process was found to be NaDES content in the extraction medium. The study indicated that the optimal extraction conditions (within the studied boundaries), so that an extract with maximum TPC and TFC is obtained, are 60.4% w/w of NaDES in the NaDES/water system, 25.4 mg of solid raw material/g solvent and an extraction time of 260 min. The quality of the NaDES-extract obtained from the aforementioned optimal conditions was practically stable in respect of TPC, TFC and DPPH radical scavenging activity after three months of storage, which suggests that the NaDES had a potential “protective” effect on the extracted phytochemicals. The NaDES-extract from the AWPR obtained from the optimal conditions was incorporated, “as obtained”, into a cosmetic cream, and sensory analysis was conducted to evaluate its organoleptic characteristics such as color, aroma and texture. The stability of the formulation was confirmed by performing centrifugation and freeze–thaw cycles. The DPPH radical scavenging activity of the cream after the incorporation of the NaDES-extract was assessed over time, and the mild DPPH radical inhibition accomplished was attributed to the released extract from the cream. According to all the aforementioned results, it can be concluded that NaDES-extracts from AWPR can be directly exploited in the development of added-value end-products in the pharmaceutical and cosmetic industry, and there are indications that the NaDES used is a promising storage medium and a possible stabilizing agent for the desirable bioactive molecules. The spectrum of potential applications of the NaDES-extract from AWPR can be further expanded in biomedicine, provided that the necessary biocompatibility and toxicity tests are conducted.