Enzyme-Assisted Extraction of Bioactive Compounds from Rosa canina L. Pseudofruit in Natural Deep Eutectic Solvents: Protease Stability and Biological Activities

TL;DR

This paper explores using enzyme-assisted extraction in natural solvents to efficiently extract bioactive compounds from Rosa canina pseudofruit, improving enzyme stability and extract effectiveness.

Contribution

The study introduces a novel green extraction method using NaDES that enhances enzyme stability and bioactive compound yield from Rosa canina.

Findings

EAE in NaDES increased enzyme half-life by up to 13-fold compared to buffer solutions.

Optimized EAE yielded high levels of phenolics (135.75 mg GAE/g DW) and flavonoids (65.05 mg CAE/g DW).

Extracts showed improved antioxidant, antidiabetic, anti-aging, and antibacterial activities over enzyme-free extracts.

Abstract

Enzyme-Assisted Extraction (EAE) in Natural Deep Eutectic Solvents (NaDES) was investigated as a green approach to extract bioactive compounds from the pseudofruit of Rosa canina L. Initially, the thermal stability of protease (Neutrase®) was evaluated at different temperatures (30–80 °C) in the NaDES Choline Chloride: Glycerol (1:2 molar ratio) (ChCl: Gly) with 20% (w/w) water as a cosolvent and in a buffer solution of the same pH. Kinetic and thermodynamic analyses revealed that ChCl:Gly markedly enhanced enzyme stability, extending half-life by up to 13-fold at 30–50 °C by increasing the enthalpic barrier to deactivation. EAE in NADES parameters, including enzyme loadings and extraction time, were optimized based on total phenolic (TPC) and flavonoid content (TFC), yielding maximum values of 135.75 ± 0.33 mg GAE/g DW and 65.05 ± 0.58 mg CAE/g DW, respectively. Extracts obtained under optimal conditions exhibited enhanced antioxidant, antidiabetic (α-amylase and α-glucosidase inhibition), anti-aging (tyrosinase inhibition), and antibacterial (inhibition of Escherichia coli growth) activities, outperforming enzyme-free extracts in all cases. The optimum extract also significantly reduced A431 cell viability (27–40%, p < 0.05). Overall, EAE in NaDES improved both enzyme stability and extraction efficiency, offering a sustainable and effective alternative for producing bioactive plant extracts.

Figures

Full text

1. Introduction

Plant extracts are complex mixtures whose biological effects depend on the identity, integrity, and relative abundance of their bioactive compounds (e.g., phenolic acids and flavonoids), which act as multifunctional agents by modulating critical cellular pathways related to inflammation, oxidation, and metabolism, offering multi-target therapeutic effects on chronic diseases, metabolic disorders, and wound healing [1,2]. Although the correlation of bioactive compounds with several biological activities has been consistently reported in the literature [3], the biological efficacy of plant extracts is not solely dependent on the total amount of bioactive compounds present, but also on their qualitative profile and structural stability [4]. Because these compounds are often sensitive to heat, oxygen, and harsh solvents, the applied extraction method can strongly influence both extract composition and measured bioactivity. Thus, extraction methods that increase extraction yield, while preserving bioactive compounds’ structural integrity, are of high importance for producing reproducible, functional extracts and for advancing greener processing within the One Health framework, where the development of optimized extraction methods is considered a critical step toward delivering health-promoting bioactive compounds while minimizing ecological burden and supporting sustainable plant resource utilization [5]. Within this context, Rosa canina L., commonly known as rose hip, represents a valuable natural source of bioactive compounds. In particular, the pseudofruit of Rosa canina L. is widely known for its high content of flavonoids, phenolic acids, vitamins, and carotenoids, which significantly contribute to its broad spectrum of biological activities. Rosa canina L. extracts have been reported to possess antidiabetic, antibacterial, antiaging, antioxidant, antiproliferative, anticancer, and neuroprotective activities [6,7,8].

The quality of the extract is directly influenced by the applied extraction method. Enzyme-assisted extraction (EAE) combined with natural deep eutectic solvents (NADES), EAE in NADES, has emerged as a promising green strategy capable of addressing these challenges. EAE is a green extraction method based on the inherent ability of enzymes to catalyze reactions with selectivity. The enzymes are capable of hydrolyzing the plant cell wall, thereby enhancing the release of entrapped bioactive compounds. By operating under mild conditions, EAE improves extraction efficiency while preserving the structural integrity and biological activity of sensitive phytochemicals [9,10]. NADES are eutectic mixtures formed by two or more naturally occurring components (such as sugars, amino acids, polyols, etc.), which act as hydrogen bond donors (HBD) and hydrogen bond acceptors (HBA). The new eutectic mixture possesses a melting point significantly lower than that of the individual components. Unlike conventional aqueous or organic solvent systems, NADES form a microenvironment, through hydrogen-bonding networks, that protects sensitive functional groups; thus, offering enzyme stability, increased catalytic activity, improved substrate solubility, and selectivity [11,12,13]. The synergistic action of enhanced enzymatic cell wall disruption and the tunable properties of NADES enables improved release and stabilization of phenolic and flavonoid compounds, while maintaining their quality.

The aim of the present study is the investigation of EAE in NADES as an efficient and green approach for the recovery of bioactive compounds from the pseudofruit of Rosa canina L. using a commercial protease (Neutrase^®^). Neutrase^®^ is a broad-spectrum neutral endoprotease (from Bacillus amyloliquefaciens) that cleaves internal peptide bonds in a wide range of proteins, generating smaller peptides. The selection of Neutrase^®^ was based on the compositional analysis of Rosa canina L. pseudofruit matrix, where protein was among the macromolecular constituents of the raw material (7.5% w/w). Although the specific identity of these proteins has not been characterized, this relatively high protein content supports the evaluation of a proteolytic enzyme, since proteolysis can contribute to matrix disintegration and facilitate the release of bound or entrapped bioactive compounds, complementing the solvent action of NADES. The kinetic and thermodynamic stability of the protease in a choline chloride–based NADES system will be studied, along with the influence of enzyme loading and extraction time on the total phenolic content (TPC; in gallic acid equivalents GAE) and total flavonoid content (TFC; in catechin equivalents CAE) of the resulting extracts. In parallel, several biological activities of selected extracts will be evaluated in vitro, namely antidiabetic (α-amylase and α-glucosidase inhibition), antiaging (tyrosinase inhibition), antioxidant, and antimicrobial, along with cytotoxicity assessment against cancer cells. To the best of our knowledge, this is the first study to explore the application of EAE in NADES of bioactive compounds from Rosa canina L. pseudofruit with a protease, providing new insights into the combined effects of enzymatic pretreatment and green solvent systems on both extraction efficiency and functional bioactivity.

2. Results and Discussion

2.1. Kinetic Stability of Protease in NADES

Enzyme deactivation involves reversible unfolding and subsequent irreversible inactivation, as described by the Lumry–Eyring model, and is governed by both thermodynamic and kinetic stability [14]. Beyond conventional aqueous buffers, NADES represent an emerging reaction medium with the potential to modulate enzyme stability [11].

In this work, the thermal stability of a commercial protease (Neutrase^®^) was evaluated in the NADES ChCl:Gly containing 20% (w/w) water as a cosolvent across a temperature range of 30 to 80 °C. Deionized water was employed instead of conventional buffer solutions to prevent the introduction of ionic species that could disrupt the intrinsic hydrogen-bonding network of NADES or interfere with enzyme–solvent interactions. The presence of water as a cosolvent, up to 50% w/w, is known to enhance enzyme stability in NADES by modulating enzyme folding, conformational flexibility, and catalytic function through hydrogen-bond interactions. Molecular dynamics studies indicate that at low water contents, water molecules are integrated within the NADES matrix, forming transient hydrogen bonds that increase molecular mobility while preserving solvent structural integrity [15,16,17]. For comparison, protease stability was also evaluated in aqueous buffer with a pH value corresponding to that of the NADES (pH = 5.3). Protease exhibited a strong, temperature-dependent enhancement of thermal stability in ChCl:Gly compared to the aqueous buffer, as reflected by consistently lower deactivation rate constants (k_d_), which resulted from plotting and substantially extended half-lives (t_1_/2) and D-values across all temperatures tested (Table 1). Deactivation rate constants (k_d_) at various temperatures were obtained from the slopes of the straight lines by plotting ln[A_t_/A_0_] vs. time (Figure 1). ANOVA followed by Tukey’s HSD confirmed a statistically significant difference (p < 0.05) between the NADES and buffer medium, supporting the stabilizing effect of ChCl:Gly. The deactivation constants increased in all media with increasing temperature, thus the irreversible thermal deactivation became progressively more significant. In ChCl:Gly at 30 °C, the protease showed a k_d_ an order of magnitude lower than in buffer, resulting in a ~9-fold longer half-life. This stabilization became even more pronounced at 40 and 50 °C, where k_d_ remained an order of magnitude lower and half-lives increased by 13- and 11-fold, respectively, relative to the corresponding buffer. At 60 °C, the NADES still provided a modest benefit, reducing k_d_ by 23% and extending half-life by 29%, indicating that protective interactions persist but no longer dominate. Even at 70 and 80 °C, the enzyme retained ~15% longer half-life in ChCl:Gly, highlighting residual thermoprotection. Together with statistical analysis (p < 0.05), these trends point to a critical destabilization threshold at 60 °C, beyond which solvent-mediated stabilization becomes insufficient to counteract thermal unfolding and irreversible deactivation. Nevertheless, while NADES cannot fully counteract high-temperature-induced denaturation, the measurable stabilizing effect highlights an enhanced resistance to thermal deactivation and an extension of enzyme functionality relative to conventional aqueous media.

The decimal reduction time (D, h) represents the time required at a specific temperature to reduce the enzymatic or microbial activity by 90%, or one logarithmic cycle [18]. D-values decreased with increasing temperature, with a more profound decline observed in the buffer, following the same trend as the half-life and demonstrating the superior thermal stability of protease in ChCl:Gly across all temperatures (Table 1).

Overall, the results demonstrate that ChCl:Gly induces up to an order-of-magnitude improvement in Neutrase^®^ stability within a defined thermal window (30–50 °C). This behavior strongly supports the hypothesis that NADES stabilizes the native and near-native conformational ensemble rather than the unfolded state, thereby delaying, but not eliminating, thermal deactivation [19]. From a bioprocess perspective, this selective stabilization is highly advantageous for prolonged or low-temperature enzymatic processes, storage, and batch processing, while also defining clear upper temperature limits for practical NADES-based applications [11].

2.2. Thermodynamic Stability of Protease in NADES

Energy of deactivation (E_(a)d_) is the energy required to initiate structural changes, such as unfolding, aggregation, or denaturation in the enzyme molecule [20]. Although the E_(a)d_ is typically derived from the temperature dependence of the deactivation rate constant (k_d_) using the Arrhenius equation, enzymes often exhibit non-Arrhenius behavior in complex environments, due to temperature-dependent pathways, multi-step deactivation, or intermediate states [21,22,23,24]. Non-Arrhenius behavior is typically classified as super-Arrhenius (convex curvature) or sub-Arrhenius (concave curvature), depending on whether the rate increases more or less rapidly than predicted [25,26,27]. Super-Arrhenius behavior may result from cooperative transitions or transport limitations, whereas sub-Arrhenius behavior may reflect localized unfolding or reduced molecular mobility. Several modified models have been developed, with the quadratic-modified Arrhenius equation accounting for temperature-dependent activation energies, better describing the underlying enzyme kinetics [28].

The experimental data were fitted to the Modified Arrhenius with Quadratic Term (Equation (4)) (Figure 2), illustrating the non- Arrhenius behavior along with the complex temperature dependence of enzyme stability in NADES. All coefficients of determination (R^2^) were greater than 0.98, indicating an excellent fit of the data to the model equation. The stability of protease in both NADES and buffer exhibited a sub-Arrhenius behavior, as indicated by C > 0 in the quadratic-modified Arrhenius equation (Table 2).

Across the tested temperatures, protease exhibited consistently higher apparent energies of deactivation (E_(a)d_) and enthalpy changes ( $[eqn]$ ) in NADES than in buffer, indicating that irreversible deactivation requires a substantially larger energetic input in the NADES system (Table 3). At 30 °C, E_(a)d_ in NADES was approximately 2-fold higher than in buffer, with this difference increasing progressively with temperature, reaching about 38% higher E_(a)d_ at 80 °C. This trend confirmed that NADES imposed a stronger energetic barrier against the transition from the native to the inactive state, in agreement with the markedly reduced k_d_ values observed kinetically.

The activation entropy ( $[eqn]$ ) exhibited a clear temperature-dependent shift from negative to positive values in both media, consistent with a mechanistic transition from a more ordered activated state at low temperatures to disorder-driven deactivation at elevated temperatures. Notably, $[eqn]$ values were more negative in buffer than in NADES, suggesting a more constrained transition state in aqueous conditions. In contrast, the less negative and eventually positive $[eqn]$ values in NADES indicate increased conformational changes during the deactivation pathway. However, the less favorable $[eqn]$ in NADES does not contradict its stabilizing effect, because the overall barrier ( $[eqn]$ ) remained higher in NADES at low–moderate temperatures, indicating enthalpy-dominated stabilization with enthalpy–entropy compensation. The protease in ChCl:Gly consistently exhibited higher $[eqn]$ values at low and moderate temperatures. At 30 and 40 °C, $[eqn]$ values were higher in NADES than in buffer by approximately 5–8 kJ·mol^−1^. Despite the moderate magnitude of the increase, statistical analysis demonstrated a significant effect of the reaction medium, whereas the temperature effects were not significant. At 70 and 80 °C the $[eqn]$ values decreased markedly in both systems, indicating advanced thermal deactivation.

The enhanced thermal stability of the protease in NADES relative to the buffer can be primarily attributed to the strong hydrogen-bond network of NADES, which interacts with the enzyme surface, limits solvent penetration into the protein core, and helps preserve the native conformation, thereby reducing the possible thermal denaturation [13,29]. Zhou et al. [30] reported that the enhanced stability of carbonic anhydrase in a choline chloride/betaine-based NADES was due to hydrogen bonding networks that stabilize enzyme structure, by creating a unique microenvironment around enzymes that influences their structural dynamics. In addition, the high viscosity of NADES has been shown to limit molecular mobility and conformational flexibility of enzymes, thereby reducing the conformational flexibility and susceptibility to thermal unfolding [31]. Tan et al. [32] studied the activity and stability of Amycolatopsis mediterranei cutinase (AmCut) in choline chloride-based DES and their individual components. They reported that at low DES concentrations (10% v/v) of choline chloride:glycerol (1:1 molar ratio), AmCut activity increased by more than twofold, while at higher DES concentrations (50% v/v) no loss of activity was observed after 2 h of incubation at 50 °C. The enhanced stability was attributed to the presence of the choline chloride moiety, while the hydrogen-bond donors (glycerol, urea, glucose) were, in some cases, inhibitory. The authors suggested that the interaction of choline chloride with AmCut induces a conformational state associated with both increased catalytic activity and improved thermal stability.

In conclusion, kinetic analysis clearly shows that the NADES ChCl:Gly significantly delays thermal deactivation of the enzyme, especially within the temperature range of 30–50 °C. Complementary thermodynamic analysis indicates that this enhanced stability arises from an increase in the enthalpic barrier to deactivation induced by the NADES, rather than any fundamental alteration of the enzyme denaturation pathway. Molecular dynamics simulations comparing the buffer and the NADES system could elucidate the molecular basis of the increased enthalpic barrier

2.3. Bioactive Compounds of EAE in NADES

The combined effects of enzyme loading and extraction time on total phenolic content (TPC) and total flavonoid content (TFC) obtained from Rosa canina L. pseudofruit by EAE in a ChCl:Gly NADES were evaluated (Figure 3).

The EAE in ChCl:Gly NADES significantly enhanced the extraction of both TPC and TFC. Based on ANOVA both factors were statistically significant (p < 0.05) in the tested range. At a fixed extraction time, TPC and TFC exhibited a significant positive response to enzyme loading, particularly in the range of 0.1–0.75% (v/v). Specifically, TPC and TFC were about 20% higher when enzyme loading 0.75% v/v was applied, compared to 0.10% v/v. However, beyond enzyme loading 0.75% v/v, a saturation trend was observed. Above this threshold, no statistically significant changes were detected, which may be attributed to limitations associated with the enzyme type, the structural characteristics of the biomass, or the availability of extractable bioactive compounds [33]. Similarly, both TPC and TFC followed the same trend and increased significantly (p < 0.05) with prolonged extraction time, specifically from 1 to 3 h, while only marginal and statistically not significant improvements were observed between 3 and 4 h, suggesting that near-equilibrium conditions were reached. The determination of an optimal extraction time requires balancing the enhanced release of bioactive compounds due to increased cell wall degradation against the increased susceptibility of the released compounds to degradation, which is influenced by the treatment temperature, which affects both release yield and compound stability [33]. Based on these considerations, the optimum extraction conditions were: enzyme loading of 0.75% v/v and extraction time 3 h, where TPC and TFC reached 135.75± 0.33 mg GAE/g DW and 65.05 ± 0.58 mg CAE/g DW, respectively. In the control experiment, where no enzyme was added, the yields of TPC and TFC were significantly (p < 0.05) lower across all extraction times studied (Figure 3). At the suggested optimum extraction conditions, the combined process EAE in NADES achieved approximately 40% higher yields in both TPC and TFC compared to the extraction without enzyme.

Pearson correlation analysis revealed strong positive correlations between the parameters tested (enzyme loading and extraction time) and both TPC and TFC. Specifically, a moderate but statistically significant correlation was observed between enzyme loading and TPC (r = 0.46, p <0.05) and TFC (r = 0.55, p < 0.05). In contrast, extraction time exhibited a strong positive correlation with both TPC (r = 0.83, p < 0.001) and TFC (r = 0.76, p < 0.001), demonstrating that extraction time is the dominant factor governing the extraction process. The moderate correlations observed between enzyme loading and both responses can be attributed to saturation effects during enzymatic extraction, as once a critical enzyme loading is reached, further increases result in limited additional phenolic and flavonoid release, as extraction becomes governed by substrate availability and mass transfer rather than enzyme [34].

These findings provide evidence that the enhanced extraction yields from EAE in NADES are due to the combination of the two non-conventional extraction methods, where the NADES solubilizes compounds, but auxiliary methods (such as enzymes, ultrasound, or microwave) break down tough plant cell walls, releasing more material for the NADES to extract [35,36]. It has been established that NADES as extraction media exhibit a strong affinity for phenolic compounds, even in the absence of additional cell-wall disruption methods. For instance, Sallustio et al. [37] applied a lactic acid:sodium acetate NADES (3:1, 35% water as co-solvent) for Rosa canina L. rosehip extraction and reported significantly higher TPC (35.26 ± 2.41 mg GAE/g FW) and TFC (0.42 ± 0.18 mg QE/g FW) compared to hydroalcoholic extraction (29.60 ± 3.56 mg GAE/g FW and 0.32 ± 0.06 mg QE/g FW, respectively). They attributed the high extraction capacity of NADES to the strong hydrogen-bonding interactions between the hydroxyl and carboxyl groups of the NADES constituents and the hydroxyl groups naturally present in phenolic compounds. However, the combined process of EAE in NADES has been reported to achieve even higher yields. Makkliang et al. [38] studied the simultaneous extraction and transformation of Pueraria mirifica isoflavone by applying EAE with cellulolytic enzymes in NADES choline chloride: propylene glycol 1:2. Under optimal conditions, the yields of the final products were higher compared to conventional methods. Additionally, Ricarte et al. [39] studied the EAE in DES of carotenoids and phenolics from sunflower wastes (petals and florets) and found that the combination of Viscozyme^®^ and the DES D, L-menthol:D, L-lactic acid (2:1) was more efficient than other green or conventional solvents. Liu et al. [40] investigated the integration of DES (choline chloride:urea 1:2) with UAE, MAE, and EAE for the extraction and stability of TFC from the jujube plant. Although EAE in DES did not yield the highest TFC among the tested methods, it resulted in about 47% higher yield relative to water-based extraction. Nevertheless, the key factor to this combined process (EAE in NADES) has been suggested to be the enhanced enzyme stability in NADES. Wang et al. [41] reported a substantial enhancement of saccharification efficiency of lignocellulosic biomass, with glucose yields increasing from 78% to 98% for certain substrates, attributing those results to the ability of NADES to provide a protected microenvironment for enzymes, in which reduced molecular vibration and friction slow conformational degradation over time. In addition, NADES were shown to form a protective layer on the enzyme surface, thereby reducing the unproductive lignin adsorption.

Therefore, the results clearly demonstrate the superiority of EAE in NADES compared to NADES alone, thereby supporting the integration of biocatalysis with green solvent systems. The observed plateau at high enzyme loadings and extended extraction times indicates that the process is well-controlled and could be further optimized, reinforcing its potential for sustainable and economically viable applications.

2.4. Antioxidant Activity

Antioxidant activity is a key indicator of the quality of plant-derived extracts, as it enables the neutralization of free radicals and the mitigation of oxidative stress. As shown in Figure 4, the presence of the enzyme in the extraction process significantly increased the antioxidant activity of the extracts. Specifically, in the absence of the enzyme, the extract exhibited the lowest antioxidant activity (IC_50_ = 1.25 ± 0.05 μL extract/mL), demonstrating limited radical scavenging capacity. In contrast, the presence of the enzyme resulted in a significant reduction in IC_50_ values up to an optimal level. Particularly, the increase in enzyme loading from 0.1 to 0.75% (v/v) resulted in a significant decrease in IC_50_ values (from 0.58 to 0.36 μL extract/mL), reflecting enhanced antioxidant activity likely due to improved enzymatic disruption of the plant matrix and increased release of bioactive compounds. The lowest IC_50_ was observed at 0.75% (v/v), suggesting the optimal enzyme loading under the studied conditions, while the IC_50_ value in the highest enzyme loading indicated that further enzyme addition did not proportionally increase antioxidant activity. Antioxidant activity was strongly and negatively correlated with both TPC (r = −0.98, p < 0.01) and TFC (r = −0.89, p < 0.05), demonstrating that the antioxidant capacity of the extracts is primarily driven by their phenolic and flavonoid content.

Previous studies have consistently reported strong antioxidant activity in Rosa canina L. extracts, along with other medicinal plants, and have linked the activity with their phytochemical profile. For instance, Ergün et al. showed that the methanolic extracts of Rosa canina L. galls exhibited stronger antioxidant activity (14.00 ± 3.01 µg mL^−1^) compared to extracts from adjacent ripe fruits (56.32 ± 0.94 µg mL^−1^) [42]. This difference was attributed to variations in TPC, as the gall extracts contained approximately twice the TPC of the fruit extracts. Another study examined the antioxidant activity through DPPH and the FRAP (ferric-reducing) assay from the methanolic extracts from five wild plants [3]. Although DPPH inhibitions were up to approximately 88%. They found that there is a strong positive correlation between TPC and FRAP (r = 0.94), whereas a weak and non-significant correlation with DPPH inhibition assay (r = 0.25). Through the correlation study, they indicated that phenolics and vitamin C are the main contributors to the reducing power of these extracts, whereas DPPH activity possibly depends on additional unstudied compounds. Hodoșan et al. [43] studied four Romanian ethnobotanical species and reported the strongest activity for peppermint dry extract (IC_50_ = 0.13 mg mL^−1^). Principal component analysis revealed strong negative correlations between TPC and TFC and both DPPH-IC_50_ and FRAP-IC_50_, whereas specific phenolics (protocatechuic, vanillic, caffeic, and ferulic acids) showed strong negative correlations with DPPH-IC_50_, confirming their key role in radical-scavenging activity. Finally, hydro-methanolic leaf extracts of seven medicinal and food plants demonstrated a strong positive correlation between antioxidant activity and TPC (ABTS, DPPH, FRAP; r > 0.98, p < 0.001) [44]. They confirmed that phenolics (total phenols, ortho-diphenols, tannins) largely drove the radical-scavenging activity, while flavonoids showed little or no correlation with antioxidant activity. Pomegranate leaves displayed the greatest DPPH inhibition (2.27 mM Trolox g^−1^), whereas parsley had minimal activity (0.01 mM Trolox g^−1^), confirming substantial variability in antioxidant potential among different plants. Considering that Rosa canina L. is a well-known source of ascorbic acid, which may be co-extracted to some extent depending on the solvent system and extraction conditions, the observed antioxidant activity could result from the combined action of phenolics/flavonoids and other redox-active constituents, including ascorbic acid; this could be confirmed by targeted vitamin C quantification (e.g., LC–MS/MS or HPLC-based methods). Overall, the literature indicates that antioxidant activity in plant extracts, including Rosa canina L., largely relies on phenolic composition, the antioxidant assay employed, and the variety of plant species.

2.5. Antibacterial Activity

The antibacterial activity of extracts obtained using ChCl:Gly NADES, with and without enzymatic assistance, was evaluated in vitro against E. coli at the optimal extraction time of 3 h. The results are expressed as percentage inhibition of E. coli growth (Figure 5). EAE in ChCl:Gly resulted in a marked enhancement of the antibacterial activity of the obtained extracts compared to extraction with NADES alone. Specifically, the latter exhibited moderate inhibition of E. coli growth (47.1 ± 0.6%), indicating that ChCl:Gly alone is capable of extracting bioactive compounds with potent antibacterial properties. However, the EAE in ChCl increased antibacterial activity up to 40%, with inhibition values ranging from approximately 66% to 77%, depending on enzyme loading. Even at low enzyme loadings (0.1–0.5% v/v), EAE in NADES produced extracts with higher antibacterial activity, which increased progressively with enzyme loading up to 0.75% v/v, beyond which no further significant enhancement was observed, suggesting a saturation effect associated with enzymatic hydrolysis and matrix accessibility limitations, as also observed in the study of TPC and TFC. Antibacterial activity showed strong positive correlations with TPC (r = 0.94, p < 0.05) and TFC (r = 0.88, p < 0.01), supporting a phenolic-driven antibacterial mechanism [45].

The increased antibacterial activity observed in the present study is consistent with the literature, where the benefits of EAE and NADES systems have been reported. Januskevič et al. [46] demonstrated that EAE significantly improved the antibacterial activity of sea buckthorn leaf extracts compared to conventional solid–liquid extraction, due to the increased release of bioactive compounds following enzymatic cell-wall disruption, as confirmed by SEM analysis. Similarly, Jurić et al. [47] and Memdueva et al. [48] demonstrated that choline chloride-based acidic NADES systems produce extracts with strong antibacterial activity against both Gram-positive and Gram-negative bacteria, comparable in some cases to conventional antibiotics, such as gentamicin. Although the acidity and ionic character of NADES are considered to contribute to antibacterial effects, there is evidence suggesting that the antibacterial activity is primarily driven by the extracted bioactive compounds rather than the solvent itself, due to a sharp decrease observed in acidity after extraction [48]. In this context, the antibacterial activity observed even at moderate enzyme loadings can be attributed to the combined effects of EAE in the favorable environment provided by ChCl:Gly.

2.6. Antiaging Activity (Inhibition of Tyrosinase Activity)

Excessive melanin accumulation is closely associated with hyperpigmentation disorders and contributes to age spots and skin discolorations. Tyrosinase (EC 1.14.18.1) serves as the rate-limiting enzyme in melanogenesis by catalyzing the oxidation of L-tyrosine to L-DOPA and the subsequent formation of reactive o-quinone intermediates that polymerize to form melanin [49]. Consequently, inhibition of tyrosinase has emerged as a strategy in anti-aging research aimed at maintaining skin homeostasis. Plant-derived extracts have been used in traditional medicine to modulate skin hyperpigmentation due to their rich phytochemical profile. Phenolic compounds have been recognized as natural inhibitors of melanogenesis due to their ability to interact with the catalytic site of tyrosinase or chelate its copper ions [50,51].

In the present study, the antiaging activity (in vitro inhibition of tyrosinase) increased markedly with enzyme loading, as reflected by a progressive reduction in IC_50_ values (Figure 6), while the corresponding concentration–response curves (% inhibition versus extract concentration) used for IC_50_ determination are presented in Figure 7.

Kojic acid was used as the positive control, which is a known tyrosinase inhibitor that completely inhibits tyrosinase at a concentration of 1 mg/mL. The extract obtained without enzymatic assistance exhibited the highest IC_50_ value (83.37 ± 0.25 mg/mL), indicating the weakest inhibitory activity. In contrast, the extracts obtained via EAE in NADES showed enhanced bioactivity, lowering IC_50_ values to 47.41 ± 0.09, 41.55 ± 0.10, 33.03 ± 0.07, and 16.59 ± 0.20 mg/mL for enzyme loadings of 0.1, 0.5, 0.75, and 1% (v/v), respectively. The tyrosinase inhibition was significantly (p < 0.05) affected by enzyme loadings, suggesting that enzymatic treatment promotes the release of bioactive compounds, such as flavonoids and phenolic acids, which have known tyrosinase-inhibitory properties.

Strong negative correlations were observed between inhibition of tyrosinase activity and both TPC (r = −0.95, p < 0.05) and TFC (r = −0.87, p < 0.05), indicating that higher TPC or TFC values are linked to higher inhibition of tyrosinase. Also, as antioxidant activity increases, inhibition of tyrosinase also improves, as indicated by a strong positive correlation (r = 0.95, p < 0.05), which is in agreement with previous reports. For instance, Sari et al. supported that there is a positive correlation between antioxidant activity and tyrosinase inhibitory potency of the Intsia bijuga heartwood extract, largely attributed to quercetin [52]. According to the literature, the tyrosinase inhibitory potency of plant extracts is vastly affected by the plant species, as their phytochemical profiles differ. For example, the Ginkgo biloba tea extract showed an IC_50_ of approximately 200 μg/mL [53], the ethanolic extract of Eugenia dysenterica leaves exhibited the IC_50_ value of 11.88 mg/mL, while the Pouteria torta aqueous extract leaves exhibited an IC_50_ value of 30.01 mg/mL [54]. However, comparison with our previous study suggests that the extraction method influences the anti-aging activity of extracts obtained from the same plant species. In particular, extracts produced via EAE in NADES in the present study exhibited up to 70% higher tyrosinase inhibition compared to the optimal extract obtained by conventional EAE, which displayed an IC_50_ value of 65.77 ± 1.72 mg/mL [43]. However, a comparison with our previous publication suggests that the extraction method affects the antiaging activity of the extracts of the same plant species as well [55]. Specifically, the extracts from EAE in NADES (present study) achieved up to 70% higher inhibition of tyrosinase, compared to the optimum extract derived from EAE (IC_50_ of 65.77 ± 1.72 mg/mL). EAE–NADES likely promoted the extraction of phenolic subclasses with higher intrinsic tyrosinase inhibitory potency [56]. Therefore, these results confirm that EAE in NADES significantly improves tyrosinase inhibitory activity of the extracts, supporting its potential application in cosmetic and dermatological formulations.

2.7. Antidiabetic Activity

α-Amylase and α-glucosidase are carbohydrate-hydrolyzing enzymes involved in postprandial glucose elevation. α-Amylase catalyzes the initial hydrolysis of starch into oligosaccharides, and α-glucosidase contributes to the terminal cleavage of carbohydrates into absorbable glucose. The inhibition of those enzymes is believed to delay carbohydrate digestion, thereby decreasing the postprandial increase in blood glucose level. Plant-derived extracts are increasingly explored as natural inhibitors of both enzymes, due to their content in phenolics and flavonoids, which affect the enzyme function through hydrogen bonding and hydrophobic interactions [57,58].

Both α-amylase and α-glucosidase in vitro inhibitory activities increased with increasing enzyme loadings during EAE in NADES (Figure 8), while the corresponding concentration–response curves (% inhibition versus extract concentration) used for IC_50_ determination are presented in Figure 9. Acarbose was used as a positive control, which showed α-amylase inhibitory activity 100% at 1 mg/mL. The extracts obtained through EAE in NADES demonstrated IC_50_ values in an enzyme-loading-dependent manner for both enzymes tested. In particular, the IC_50_ values for α-amylase decreased to 41.02 ± 0.1, 24.25 ± 0.12, 20.50 ± 0.1, and 18.37 ± 0.2 mg/mL at enzyme loadings of 0.1, 0.5, 0.75, and 1% (v/v), respectively, while the corresponding IC_50_ values for α-glucosidase were 3.23 ± 0.25, 2.95 ± 0.09, 2.85 ± 0.1, and 2.15 ± 0.19 mg/mL. On the contrary, the enzyme-free extracts demonstrated IC_50_ values of 45.50 ± 0.2 and 3.98 ± 0.17 mg/mL for α-amylase and α-glucosidase, respectively. The difference in the enzyme-free extract and the ones from EAE in NADES could be attributed to the combined effects of the enhanced enzymatic hydrolysis of the plant, along with the complementary solvation properties of the NADES. This is further supported by Xia et al. [59] who used NADES to extract astilbin from Rhizoma Smilacis Glabrae and suggested that NADES not only protects the functional compounds, but also enhances their biological property in a “cellular ingredients bionic” way.

Both α-amylase and α-glucosidase inhibitory activities were strongly correlated with TPC, TFC, and enzyme loading, while positively correlated with antioxidant activity. Specifically, α-amylase inhibition was negatively correlated with TPC (r = −0.98, p < 0.01), TFC (r = −0.88, p < 0.05), and enzyme loading (r = −0.97, p < 0.01), whereas positively correlated with antioxidant activity (r = 0.94, p < 0.05). a-Glucosidase was negatively correlated with TPC (r = −0.94, p < 0.05), TFC (r = −0.88, p < 0.05), enzyme loading (r = −0.93, p < 0.05), whereas positively with antioxidant activity (r = 0.91, p < 0.05).

Evidently, the studied extracts exerted stronger inhibitory activity against α-glucosidase than against α-amylase, a trend widely reported for phenolic-rich plant extracts in the literature [59]. This behavior is attributed to the structural differences between the two enzymes; α-glucosidase has a more accessible and flexible active site, thus favoring interactions with low-molecular-weight phenolic compounds, such as flavonoids, while α-amylase has a deep and narrow catalytic cleft, more suitable for polysaccharides, such as starch [60].

The potential role of herbal plants in the inhibition of α-glucosidase activity has been widely studied and is correlated with the phytochemical profile, without the individual compounds responsible for the activity being identified, with terpenes and flavonoids believed to be the largest chemical class that exhibit the strongest inhibitory activities. Several isolated compounds from medicinal plants have been reported to inhibit α-glucosidase more than the corresponding positive control [61,62]. Ratananikom et al. [63] investigated the inhibitory potential of Thai culinary vegetable extracts against α-glucosidase and α-amylase and reported inhibition ranges of 13.42 ± 0.23% to 79.84 ± 0.47% for α-glucosidase, while considerably lower inhibition was observed against α-amylase (4.82 ± 3.32% to 27.49 ± 1.67%). Another study on NADES extracts from eight Algerian date fruit cultivars [64] reported notable cultivar-dependent variability in α-amylase inhibition (highest in the Ourous variety, ~45%), with inhibition positively correlated with rutin content (r = 0.73), suggesting rutin may contribute to the effect. Additionally, it has been reported that solely NADES may exhibit moderate inhibitory activity. In particular, Popović et al. [65] studied several NADES systems along with their individual components and showed that both (ChCl and HBD) contributed comparably to the overall inhibitory effect of NADES. Notably, no synergistic effects between NADES components were observed, as the inhibitory activity of the NADES did not exceed that of their individual components. In addition, Jovanović et al. [66] examined NADES-based extraction of turmeric using different solvent systems and demonstrated that extracts obtained with ChCl:citric acid:water (1:2:3) and ChCl:lactic acid:water (1:2:5) exhibited the strongest α-amylase inhibitory activity, achieving inhibition levels of approximately 90%, which were higher than those of the positive control acarbose. The inhibition was attributed to the organic acid components of the NADES, indicating that the solvent itself was responsible for modulating the bioactivity of turmeric extracts. In the present study, although a contribution of the NADES itself cannot be excluded, comparison between enzyme-free extracts and those obtained via EAE in NADES clearly indicates that enzyme loading is the primary factor affecting the enhanced inhibitory activity. In conclusion, the results indicate that the process EAE in NADES favors both α-amylase and α-glucosidase inhibition activity, highlighting the importance of the applied extraction method.

2.8. Cytotoxicity of the Optimum Extract

The evaluation of cytotoxicity of plant extracts in cancer cell lines is essential for comprehensively assessing their therapeutic potential. Testing in cancer cells allows the identification of anticancer or antiproliferative effects and provides insight into possible mechanism-based cytotoxicity associated with bioactive compounds, such as phenolics and flavonoids. Additionally, when novel extraction methods or solvents are applied, this kind of evaluation is critical since they may alter the composition and bioavailability of plant components [67,68]. To this end, the cytotoxic effect of the optimum extract obtained via EAE in NADES ChCl:Gly, as well as the extract derived solely from EAE, was assessed to evaluate potential selectivity toward malignant cells. A human epidermoid carcinoma cell line (A431) was employed as a representative model of cancerous cells.

The assessment of plant extract cytotoxicity using the A431 human epidermoid carcinoma cell line is important for evaluating both antiproliferative activity and potential skin-related biological effects. A431 cells are a well-established in vitro model in skin cancer research and are characterized by high expression levels of the epidermal growth factor receptor (EGFR). Viability reduction in this cell line may reflect mechanism-driven antiproliferative effects rather than nonspecific toxicity [69]. The extract obtained via EAE in the NADES ChCl:Gly exhibited a markedly enhanced cytotoxic effect against A431 human epidermoid carcinoma cells compared to the extract produced using EAE in aqueous buffer (Figure 10). Specifically, the latter resulted in cell viabilities consistently around 100%, indicating the absence of cytotoxicity against A431 cells under the tested concentrations. In contrast, the extract derived from EAE in ChCl:Gly significantly (p < 0.05) reduced A431 cell viability across all concentrations, with values ranging from 72.83 ± 3.46% to 63.10 ± 2.55%, corresponding to an approximate 27–40% decrease in cell viability. Among the studied concentrations, the highest one (1000 μg/mL) resulted in the lowest A431 cell viability (63.10 ± 2.55%), although no statistically significant differences were detected (p > 0.05). This suggests that the cytotoxic effect of the EAE-NADES extract on A431 cells is not concentration-dependent within the examined range.

The results demonstrate that the Rosa canina L. extract obtained via EAE in NADES exhibits an antiproliferative effect against cancer cells, displaying enhanced cytotoxic activity compared to the extract obtained through EAE with aqueous buffer. The observed differences between the two extraction methods are likely attributed to qualitative and/or quantitative variations in the phytochemical profile of the resulting extracts. The use of NADES in combination with enzymatic treatment may enhance the solubilization, stabilization, and bioavailability of both polar and less polar bioactive compounds that are poorly extracted using aqueous buffer alone. The enhanced cytotoxicity toward cancer cells observed for the EAE–NADES extract may arise either from the extraction of additional bioactive compounds or from higher concentrations of the same ones, leading to amplified biological activities, while also from the NADES used [70,71]. The observed activity is favorable for both pharmaceutical and cosmetic applications, particularly for formulations intended for skin contact, and is in accordance with the literature. Rosa canina L. extracts have been reported to significantly reduce the viability of WiDr colon cancer cells, exhibiting an IC_50_ value of 270 µg mL^−1^, while normal colon epithelial cells (CCD 841 CoN) retained approximately 80% viability even at a concentration of 540 µg mL^−1^, thereby demonstrating clear selectivity toward cancer cells over healthy cells [72]. The authors attributed this activity to the polyphenol-rich composition of the extract, which was shown to repress telomerase (hTERT) expression and induce mitochondria-dependent apoptotic pathways, along with S-phase cell-cycle arrest. Similarly, Kilinc et al. [73] investigated the antiproliferative potential of Rosa canina L. extract using A549 lung carcinoma and PC-3 prostate cancer. The extract exhibited a cytotoxic effect against both A549 and PC-3 cancer cell lines, inhibiting cancer cell proliferation in a concentration-dependent manner.

3. Materials and Methods

3.1. Chemicals and Reagents

High-glucose Dulbecco’s Modified Eagle Medium (DMEM) was purchased from Biosera (Cholet, France). Dulbecco’s phosphate-buffered saline (DPBS), without CaCl_2_ and MgCl_2_, pH 7.4, Trypsin-EDTA, and fetal bovine serum (FBS) were obtained from PAN Biotech (Aidenbach, Germany). Gibco provided antibiotic–antimycotic and gentamycin. Dimethyl sulfoxide (DMSO), and 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl-2H-tetrazolium bromide (MTT) were purchased from Sigma-Aldrich (St. Louis, MO, USA). A431 cancer cells were obtained from the American Type Culture Collection (ATCC, Manassas, VA, USA). 2,2-Diphenyl-1-picrylhydrazyl (DPPH·); absolute methanol (CH_3_OH); and Folin–Ciocalteu were purchased from Sigma-Aldrich (St. Louis, MO, USA). Tyrosinase (EC 1.14.1.8.1) derived from mushroom, α-amylase (EC 3.2.1.1) from porcine pancreas, and α-glucosidase (EC 3.2.1.20) from Saccharomyces cerevisiae were obtained from Sigma-Aldrich (St. Louis, MO, USA). All other reagents were of the highest purity commercially available and were obtained from Sigma Chemical Co. (St. Louis, MO, USA).

3.2. Plant Material

Rosa canina L. pseudofruit was processed as previously described in earlier studies by our group [74]. Briefly, dried rosehips supplied by Korres S.A.—Natural Products were harvested in early October, manually deseeded, milled (<500 μm), and stored at room temperature until analysis.

3.3. Enzyme Preparation

The commercial enzyme preparation Neutrase^®^ was used in the present study and was a generous gift from Novozymes A/S (Bagsværd, Denmark). Neutrase^®^ is a commercial microbial neutral endo-protease and consists primarily of the active protease enzyme. No excipients or additives were removed from the commercial enzyme preparation, and the enzyme was used as received without further purification. Given the relatively low concentration of these additives in the formulation and the focus of the study, their impact on the results is considered to be minimal. Enzyme solutions were freshly prepared before each experiment.

3.4. General Procedure for the NADES Preparation

The NADES Choline Chloride: Glycerol in molar ratio 1:2 was prepared using the heating and stirring method as previously reported, until a clear, colorless liquid was formed [75]. The components of NADES were anhydrous components, and no water was present in the starting mixtures. The freshly prepared NADES was placed in a desiccator over P_2_O_5_ before the controlled addition of deionized water corresponding to 20% (w/w) of the total NADES mass. No further purification steps were applied. All results refer to water-containing NADES systems unless stated otherwise.

3.5. Enzyme Activity

The protease activity assay was based on the azocasein hydrolysis method [76]. Briefly, 50 μL of the sample was added to 2 mL of azocasein solution (0.2% w/v, in Tris–HCl buffer, pH 8). The mixture was vortexed and incubated at 40 °C for 10 min. After incubation, 2 mL of 0.1 M TCA solution was added to terminate the reaction, followed by centrifugation for 5 min. Subsequently, 2 mL of the supernatant was collected and mixed with 2 mL of 0.5 M NaOH. The solution was vortexed and allowed to stand at room temperature for 5 min. Finally, the absorbance was measured at 440 nm. An increase of 0.1 absorbance units was defined as 1 unit of protease activity. All measurements were performed in triplicate, and the results were expressed as mean values ± standard deviation.

3.6. Thermal Stability

Protease’s stability was assessed at six different temperatures (30, 40, 50, 60, 70, and 80 °C). The enzyme preparation was incubated in a thermostatically controlled water bath at a specified temperature after being diluted in the medium. Samples were withdrawn at different intervals, rapidly cooled in an ice bath, and protease activity was measured. The ratio of the initial activity and the activity during the deactivation process at different times was used to calculate the residual activity in each sample. The study was conducted in the NADES Choline Chloride: Glycerol in molar ratio 1:2 plus 20% water as cosolvent (referred to as ChCl:Gly). The experiments were carried out in a 50 mM citric acid-potassium phosphate buffer in the NADES equivalent pH of 5.3.

3.7. Kinetic and Thermodynamic Stability

Enzyme deactivation can be described by a first-order kinetic model (Equation (1)):

$$[eqn]$$

where A_t_ (units/mg protein) is the enzyme activity at time t, A_0_ (units/mg protein) is the initial enzyme activity, t (min) is the incubation time, and k_d_ (min^−1^) is the first-order deactivation rate constant.

The slope of the plot of $[eqn]$ versus t, at every temperature tested, gives the value of the deactivation rate constant k_d_.

The half-time (t1/2) and decimal reduction time (D) of the enzyme are the time needed for the reduction of 50% and 90% of the initial activity at a given temperature, respectively (Equations (2) and (3)).

$$[eqn]$$ $$[eqn]$$

Based on the Arrhenius equation, reaction rates increase exponentially with temperature; however, deviations from this linearity have been observed in more complex systems [77]. These deviations may appear as convex (Sub-Arrhenius) or concave (Super-Arrhenius) curvature and can be quantitatively described by introducing an additional quadratic term in 1/RT into the Arrhenius equation, thereby redefining it as Modified Arrhenius with Quadratic Term (Equation (4)) [78,79].

$$[eqn]$$

where C is calculated by Equation (5)

$$[eqn]$$

and dictates whether the plot is “concave” or “convex”, either sub- or super-Arrhenius behavior, and is positive or negative, respectively.

Accordingly, the deactivation energy, $[eqn]$ is no longer constant, but depends linearly on 1/T and can be calculated by Equation (6).

$$[eqn]$$

The change in enthalpy ( $[eqn]$ ), free energy ( $[eqn]$ ) and entropy ( $[eqn]$ ) for the deactivation of the enzyme in NADES or in buffer solution were determined using Equations (7), (8) and (9), respectively.

$$[eqn]$$ $$[eqn]$$ $$[eqn]$$

where, $[eqn]$ is the enthalpy of deactivation (kJ·mol^−1^), $[eqn]$ is Gibbs free energy of deactivation (kJ·mol^−1^), $[eqn]$ is the entropy of deactivation (kJ·mol^−1^·K^−1^), E_d_ is the deactivation energy (kJ·mol^−1^), T is the absolute temperature (K), k_d_ is the deactivation constant (h^−1^), R is the universal gas constant (8.314 J·mol^−1^·K^−1^), h is the Planck’s constant (6.626 × 10^−34^ J·s), κ is the Boltzmann’s constant (1.38 × 10^−23^ J·K^−1^)

3.8. Enzyme-Assisted Extraction in NADES

The milled material was mixed with the NADES Choline Chloride: Glycerol in molar ratio 1:2 plus 20% w/w water as cosolvent at 7% w/v (70 mg/mL) solid-to-liquid ratio, and a specific amount of enzyme was added. Extraction was performed in a thermoshaker (Thermomixer^®^, Eppendorf, Hamburg, Germany) under constant stirring conditions (1300 rpm), operating at 40 °C. After extraction, the enzyme was deactivated by heating the samples at 90 °C for 5 min. The samples were centrifuged at 10,000 rpm for 10 min at 4 °C, and the supernatants were collected and stored at −18 °C until further analysis.

3.9. Determination of Total Phenolic Content

The total phenolic content was evaluated based on the Folin–Ciocalteu method, following the procedure described previously [74]. The extract was mixed with Folin–Ciocalteu reagent, aqueous sodium carbonate solution (22% w/v) and distilled water. The reaction mixture was incubated at ambient temperature for 1 h, and the absorbance was read at 755 nm. The results were expressed as milligrams of gallic acid equivalents per gram of dry weight (mg GAE/g DW). All measurements were performed in triplicate.

3.10. Determination of Total Flavonoid Content

The total flavonoid content was determined using the aluminum chloride colorimetric assay, as described in a previous study [74]. Briefly, the extract was mixed with sodium nitrite (NaNO_2_), aluminum chloride (AlCl_3_), sodium hydroxide (NaOH), and distilled water. After incubation at ambient temperature for 15 min, absorbance was read at 510 nm. The results were expressed as milligrams of catechin equivalents per gram of dry weight (mg CAE/g DW). All measurements were conducted in triplicate.

3.11. Cell Culture Conditions

Human epidermoid cancer cells (A431 cells) were grown in 25 cm^2^ culture flasks (Corning) in high glucose Dulbecco’s Modified Eagle Medium (DMEM), supplemented with 10% heat inactivated fetal bovine serum (FBS), as well as 1% antibiotic–antimycotic, and 0.07% gentamicin. Cells were maintained in an incubator at 37 °C in a 5% CO_2_ atmosphere. Then, cells were detached with Trypsin-EDTA and re-seeded into fresh medium every 3 days.

3.12. Cell Viability Evaluation

Cell survival was estimated via the MTT [3-(4,5-dimethylthiazol2-yl)-2,5-diphenyl-2H-tetrazolium bromide] colorimetric assay. This method is based on the ability of mitochondrial dehydrogenase to form purple formazan crystals via the reduction of MTT [80]. A431 cells were seeded in 96-well plates at a density of 6000 cells/well and maintained in the incubator for 24 h. Then, cells were treated accordingly with the examined compounds for 24 h, kept at the humidified incubator, and MTT solution (final concentration 0.65 mg/mL) was added to each well upon the removal of cells’ medium. Cells were incubated for 3 h to allow MTT metabolism, the cells’ medium with MTT was removed, and the produced formazan crystals were solubilized upon the addition of 200 μL per well. An Epoch 2 microplate reader (Bio Tek Instruments, Winooski, VT, USA) was employed to record the absorbance at 570 nm. Cell viability was calculated as a percentage of (mean optical density of treated cells/mean optical density of control group) × 100. All the experiments were performed in triplicate for repeatability, and the results were expressed as means ± standard deviation.

3.13. Cytotoxicity Study

A431 cells were incubated for 24 h with the examined extracts in the dark for the following concentrations: 150, 250, 400, 650, and 1000 μg/mL in 0.5% deionized water. Cell survival was further assessed via MTT assay.

3.14. Antioxidant Activity

The antioxidant activity was evaluated by the DPPH radical scavenging assay, as previously reported [74]. Absorbance was read at 515 nm, and the results were expressed as IC_50_ values (μL extract/mL), calculated from the linear regression of percentage inhibition of DPPH radicals vs. extract concentrations.

3.15. Antibacterial Activity

The antibacterial activity of the extracts was tested against the Gram-negative bacteria Escherichia coli (E. coli) using the broth microdilution method, as previously reported [74]. Optical density was recorded at 600 nm after 24 h of incubation. Antibacterial activity was expressed as % inhibition of E. coli growth.

3.16. Anti-Aging Activity (Inhibition of Tyrosinase Activity)

Inhibition of tyrosinase activity was measured using a modified dopachrome assay with L-DOPA as the substrate, following the procedure described previously [74]. Samples were incubated with tyrosinase (200 U/mL), phosphate buffer (50 mM), and L-DOPA (10 mM). The reaction mixtures were incubated at 25 °C for 10 min, after which absorbance was read at 492 nm. Kojic acid served as the positive control representing complete enzymatic inhibition. All assays were performed in triplicate.

3.17. Antidiabetic Activity

3.17.1. Inhibition of a-Amylase Activity

α-Amylase inhibition was evaluated using a modified Caraway–Somogyi iodine/potassium iodide (IKI) assay previously described [74]. Briefly, the sample was mixed with α-amylase solution (0.5 U/mL) in a 96-well microplate and pre-incubated at 37 °C for 10 min. The enzymatic reaction was initiated by adding starch solution (0.025% w/v), followed by a second 10-min incubation at 37 °C. The reaction was terminated by adding 1 M HCl, after which iodine–potassium iodide reagent was added. Absorbance was read at 630 nm. Acarbose was used as a reference inhibitor corresponding to complete enzymatic inhibition (positive control). All measurements were performed in triplicate.

3.17.2. Inhibition of a-Glucosidase Activity

Inhibition of α-glucosidase activity was assessed following a previously established protocol [74], utilizing p-nitrophenyl-α-D-glucopyranoside (pNPG) as the enzymatic substrate. Samples were incubated with α-glucosidase at 37 °C, after which pNPG was added, and the reaction was allowed to incubate for 5 min at 37 °C. The absorbance was read at 405 nm against a blank containing buffer instead of the sample. Acarbose was used as the positive control. All measurements were performed in triplicate.

3.18. Statistical Analysis

The data were confirmed to be normally distributed and to exhibit homogeneity of variances according to Shapiro–Wilk Test and Levene’s Test, respectively. Analysis of variance (ANOVA) and Tukey’s HSD test were used to assess the possible differences among the means. Pearson’s correlation analysis was performed using the R programming. Non-linear regression was used to determine the parameters of the Modified Arrhenius with Quadratic Term from the experimental data, using SigmaPlot software (Version 12.5; Systat Software Inc., San Jose, CA, USA). The results were expressed as mean values with the standard deviation (SD) of three independent measurements (n = 3), and significance was assumed at p < 0.05.

4. Conclusions

This research clearly shows that EAE in NADES is an effective and sustainable approach for the recovery of bioactive compounds from Rosa canina L. pseudofruit. The key innovation of this work is that it mechanistically links the NADES solvent microenvironment to enzyme stability and, consequently, to extraction performance and bioactivity. In particular, the choline chloride:glycerol NADES significantly enhanced the thermal stability of the protease, as confirmed by both kinetic and thermodynamic analyses. The observed increase in the enthalpic barrier to enzyme deactivation indicates that the NADES environment delays irreversible unfolding and preserves enzymatic functionality, particularly advantageous for prolonged extraction processes. The combined application of EAE in NADES resulted in markedly higher TPC and TFC compared to enzyme-free extract, with clear dependence on enzyme loading and extraction time. The findings indicate that moderate enzyme loadings are sufficient to maximize bioactivity, while preserving an optimal balance between extraction efficiency and economic feasibility. Extracts obtained through EAE in NADES exhibited significantly improved antioxidant, antibacterial, antiaging, and antidiabetic activities, which were strongly correlated with the TPC and TFC of the extracts. Additionally, the extracts displayed antiproliferative activity against cancer cells, supporting their functional relevance. To the best of our knowledge, this is the first study integrating enzyme stability assessment, extraction optimization, and biological evaluation for the EAE of Rosa canina L. pseudofruit in a NADES system. The findings provide new mechanistic insight into the green extraction methods for the development of high-value extracts. EAE in NADES emerges as a promising strategy for the sustainable valorization of plant biomass, with strong prospects for application in cosmetic, nutraceutical, and pharmaceutical formulations. Compared with our previous studies, where only EAE was applied, the present work advances the field by integrating NADES as a functional extraction medium for enzymes. Overall, EAE in NADES emerges as a promising strategy for sustainable biomass valorization with prospects for cosmetic, nutraceutical, and pharmaceutical applications. Future work should investigate enzyme–NADES interactions via molecular dynamics, formulation compatibility and extract stability, and process scale-up to confirm industrial feasibility. Future work should focus on the investigation of molecular dynamics to gain deeper insight into enzyme-NADES interactions, formulation compatibility, and extract stability in products, along with process scale-up to confirm industrial feasibility.