Stability of microemulsions containing red grape pomace extract obtained with a glycerol/sodium benzoate deep eutectic solvent ☆

– The valorization of red grape pomace is signi ﬁ cant as grape is one of the most cultivated fruits worldwide and generated by-product quantities are enormous. For this purpose, numerous encapsulation techniques have been developed. However, the studies on microemulsions composed with deep eutectic solvent extracts are very limited. In this study, red grape pomace extract (RGPE) was ﬁ rst prepared by deep eutectic solvent extraction and characterized by HPLC analysis. Rutin, quercetin, catechin and caftaric acid were identi ﬁ ed as the main non-pigment phenolic compounds. The RGPE was further encapsulated in microemulsions (MEs) following a low-energy approach using a mixture of low molecular weight surfactants, and the pseudo-ternary phase diagram was constructed. The physical and antioxidant stability of MEs containing 3 – 15 wt% RGPE was investigated for a period of 30 days. MEs were stable at an ambient temperature of 25 or 37 ° C. The radical scavenging activity of encapsulated RGPE was improved up to 13% compared to the free extract. Our results indicate that microemulsions provide protection of valuable phenolic constituents especially under elevated temperature conditions and can therefore be used as systems for applications in nutraceuticals or cosmetics.


Introduction
Wine (or grape) pomace is the residue generated during the fermentation process of pressed grapes. It is mainly composed from the grape skins, secondarily the seeds and small stems, and accounts between 20 to 30% of the initial grape mass (Jin et al., 2020). The global production of grapes intended for winemaking was estimated around 44 million t, according to 2018 data from the International Organization of Vine and Wine (OIV; Spigno et al., 2017).
In several regions, grape pomace has long been used for the production of traditional distilled spirits such as "Tsipouro" or "Raki" in Greece, Italian "Grappa", French "eau-de-vie de marc", etc. (Apostolopoulou et al., 2005). In its dried form, grape pomace is conventionally used as feed (Tayengwa et al., 2021) and after composting as soil conditioner (Ilyas et al., 2021).
However, the types of compounds identified and quantified each time depend on differences based on variety, regional climate and of course the wine-making procedure followed.
The beneficial effect of moderate red wine consumption is well documented and health benefits with regards to cardiovascular disease, diabetes or some types of cancers are attributed to the anti-inflammatory and antioxidant properties of the polyphenols it contains (Rosenzweig et al., 2017). Several studies have demonstrated the promising protective effect of polyphenols on neurodegenerative diseases such as Alzheimer's and dementia (El Gaamouch et al., 2021). Beneficial properties have also been established for resveratrol (trans-3,4 0 ,5-trihydroxystilbene) in in vitro studies focusing on epilepsy, amyotrophic lateral sclerosis (ALS), Parkinson's disease, Huntington's disease, or nerve injuries. The exact mechanism and direct targets involved remain unknown (Pasinetti et al., 2015). However, wine consumption is largely criticized, not recommended in cases of liver damage, or even prohibited for religious reasons. Therefore, red grape pomace may suggest an effective alternate source of wine polyphenols. According to a recent study, red grape pomace is approximately 20% richer in phenolic compounds than red wine, due to higher amounts of anthocyanins, flavonols, stilbenes and flavanals, and reduced the VLDL-cholesterol and triglyceride levels more effectively in in vivo studies with animal subjects (de Oliveira et al., 2017). Similarly, red grape pomace consumption improved the body weight, glucose tolerance, insulin sensitivity and hepatic steatosis of high-fat diet fed mice (Rosenzweig et al., 2017).
Microemulsion systems are an effective tool to improve their chemical stability, bioavailability and targeting properties of phenolic compounds. A wide range of studies offer information on the properties of pure compounds or plant extract microemulsion formulations (Chatzidaki et al., 2015;Fregapane et al., 2022;Cecchi et al., 2020), etc. With regards to grape derived constituents, anthocyanins have been incorporated in water-in oil microemulsions (Oh et al., 2006), resveratrol in both oil-in-water or water-in-oil formulations (Ju skaitė et al., 2015;Yutani et al., 2015) and more recently in deep eutectic solvent (DES)/water-in oil microemulsions (Sakuragi et al., 2020).
Deep eutectic solvents have gained interest as natural alternatives to more expensive or toxic solvents, as they combine tailor-made properties, biodegrability and enhanced extraction efficiency (Kaltsa et al., 2020). Their synthesis usually involves one constituent serving as a hydrogen bond donor (HBD) and another one as a hydrogen bond acceptor (HBA). Low-cost, food-grade substances such as organic acids, organic acid salts, polyols and amino acids are representative examples of materials used for DES preparation .
However, to our best knowledge, it is the first time to report the incorporation of grape pomace extract in DES solutioninoil microemulsions. Therefore, the purpose of this study was to prepare microemulsions containing red grape pomace extract (RGPE) produced by an eco-friendly/"green" extraction method (DES extraction) and examine major stability properties under different storage conditions.

Grape pomace extract preparation/grape marc extract preparation
Red grape pomace of Vitis vinifera cv. Muscat of Hamburg grapes was a kind donation from a local winery (Karditsa, Greece). All handling and processing procedures of the plant material and optimum extraction protocol followed are described in detail in our previous work . In brief, glycerol (hydrogen bond donor [HBD]) and sodium benzoate (hydrogen bond acceptor [HBA]) were mixed in a Duran TM bottle of appropriate volume and heated at 70°C, under stirring at 400 rpm, for approximately 60 min, to form perfectly transparent liquids. The selected mixing ratio of glycerol/sodium benzoate system was 9:1, as it was shown that this combination formed a transparent liquid which remained stable in terms of crystal formation over several weeks. The DES mixture was finally diluted with water at a level of 70% w/w as dictated by optimum extractability experiments. Ultrasonic (US) assisted deep eutectic solvent extraction parameters included: US pretreatment (15 min, room temperature, 25°C, 104.33 W) in Elma S 100 (H) heated ultrasonic bath (Elma Schmidbauer GmbH, Singen, Germany) followed by stirred-tank extraction (liquid-to-solid ratio: 10 mL of DES solution/g of dried pomace, 80°C, 240 min, 900 rpm) using a magnetic stirrer (Witeg, Wertheim, Germany). The final mixture was centrifuged (10,000 rpm) and the supernatant was collected for further use.

Phase diagram
The area of microemulsion formation and corresponding concentration ranges of each component (wine waste extract: S mix : IPM) was found by constructing a pseudoternary phase diagram following the titration method. Tween 80 and Span 20 were mixed at a weight ratio of 4:1 to obtain the surfactant mixture (S mix ). Then, the oil phase (IPM) was mixed with the surfactant mixture at different weight ratios (1:9, 2:8, 3:7, 4:6, 5:5, 6:4, 7:3, 8:2, and 9:1). The final mixtures were diluted with the wine waste extract in a dropwise manner (25 mL every 24 h), by stirring (600 rpm/2 min/30°C) and the point where samples became turbid was marked (Mitsou et al., 2019). The Microsoft Excel software was employed for the construction of the pseudoternary phase diagram.

Microemulsion selection and storage
The microemulsions prepared contained a constant amount of S mix at 25 wt%, according to the optimum area obtained from the pseudoternary diagram and was mixed with appropriate amounts of the oil phase (IPM). The concentration of the wine waste extract added dropwise was 3, 6, 9, 12 and 15 wt% and the total microemulsion obtained was 3 g. All components were mixed under stirring (600 rpm/30°C) and finally allowed to equilibrate prior to analysis. Finally, microemulsions were stored at room temperature (25°C) or elevated temperature (37°C) (CO 2 incubator model MCO-20 AIC, Sanyo, Japan) and their chemical and physical stability was evaluated over a period of 30 days. Measurements were performed at least in triplicate.

Extraction of phenolic components from microemulsions
A volume of 0.5 mL microemulsion sample was diluted 1:1 with methanol in a 1.5 mL Eppendorf tube and vortexed for a few seconds. Then, the tubes containing the mixture were centrifuged at 10,000 rpm for 10 min (20°C) (Ortoalresa Digicen 20-R, Los Frailes, Spain) and supernatants were collected for DPPH and HPLC analysis.

Radical scavenging activity determination
The radical scavenging activity of selected microemulsions was determined by the using the DPPH radical method following our previous methodology (Kaltsa et al., 2021). Briefly, 975 mL of DPPH solution (100 mM in methanol) was mixed with 25 mL of each supernatant and kept in the dark to react. The absorbance of the mixture was measured after 30 min at 515 nm (Shimadzu UV-1700, Shimadzu Europa GmbH, Duisburg, Germany). The lower the absorbance of the reaction mixture, the higher the scavenging activity. Simple aqueous solutions containing 3, 6, 9, 12 and 15 wt% of wine waste extract served as control samples. Results were expressed as percentage inhibition (A AR %) against blank according to Equation (1): where, A s and A 0 are the absorbance at 515 nm of the sample and the blank respectively.

HPLC analysis
A Shimadzu CBM-20A liquid chromatograph (Shimadzu Europa GmbH, Duisburg, Germany) was used for HPLC analysis, coupled with an SIL-20AC autosampler and a CTO-20AC column oven. The detector was a Shimadzu SPD-M20A, which was connected to the computer via Shimadzu LC solution software. In a previous work , the characteristics of the column and the elution program were described in detail. Based on the calibration curves (1-50 mg mL À1 ) constructed with catechin (R 2 = 0.9999), caftaric acid (R 2 = 0.9999), rutin (R 2 = 0.9990), and quercetin (R 2 = 0.9999), the flavanols, hydroxycinnamates, and flavonols were quantified at 280, 320 and 360, respectively. Samples were analyzed in triplicate.

Statistical analysis
Statistical analysis of the results was performed with Statgraphics Centurion XV (Statgraphics, Rockville, MD, USA) and the LSD test was applied in order to compare the mean values of microemulsion and extract solution properties (A AR % and content of phenolic constituents by HPLC) at 99% level of confidence.

Phase diagram
A mixture of non-ionic surfactants, Tween 80 and Span 20 in 4:1 ratio was selected because it has been shown that this analogy was able to entrap high amounts of resveratrol diluted in DES composed by similar ingredients (∼50 wt%; Sakuragi et al., 2020). Isopropyl-myristate used as the oil phase finds many applications in cosmetics as well as in food grade delivery systems (Li et al., 2017;Chen et al., 2019;Mitsou et al., 2019).
The pseudo-ternary phase diagram of the wine waste DES extract: S mix : IPM system is depicted in Figure 1. As can be seen, a moderate/narrow microemulsion formation area was detected based on the optical translucency of the samples. The main microemulsion area was located within the region of S mix concentration between 20 and 30 wt%. According to our results, the above S mix ratio at the given concentration (∼22 wt%) was able to solubilize the RGPE at concentrations as high as ∼20 wt%. However, this sample presented phase separation after a few days of storage and hence it was omitted from further investigation. Therefore, formulations marked with green dots, representing samples with a fixed 25 wt% total surfactant concentration and varying wine waste extract concentration (3, 6, 9, 12 and 15 wt%), were selected to assess their physical and chemical stability.

Centrifugal and physical stability during storage
The appearance of freshly prepared microemulsion samples was red hued translucent, relatively low viscosity liquids that were free to flow when reversed. Upon the effect of centrifugal forces neither of the samples was stratified (Fig. 2a). All microemulsion formulations remained stable as well during storage under different temperature conditions (25 or 37°C) with no visible turbidity, phase separation, aggregation, or sedimentation of the components at the bottom during 30 days of storage in the dark (Fig. 2b and c). Therefore, the long-term gravitational stability of the formulations is ensured even when elevated storage temperatures may occur.

Effect of storage on radical scavenging activity
DPPH assay is a simple, versatile, and low-cost method for assessing the total radical scavenging activity of phenolic extracts.
In Fig. 3a, b , the effect of concentration and storage temperature on the radical scavenging activity of encapsulated and non-encapsulated/crude/free RGPE is presented. Extract concentration increase from 3 to 15 wt% caused a positive linear dose-dependent response to initial DPPH inhibition values, which varied between 12.59 to 55.82% and 12.7 to 59% for microemulsions and non-encapsulated extracts accordingly. However, similar values of scavenging activity were observed in most cases between microemulsions and plain extract homologues (P < 0.01).
Temperature had also a profound effect on the antioxidant activity of both types of samples during storage. For presentation reasons, data on the effect of different storage temperatures on microemulsions and extract solutions containing 15 wt% RGPE that presented the highest initial radical scavenging activity are displayed. DPPH values of microemulsions or extract samples (15% RGPE) stored under room temperature conditions (25°C) remained practically unaffected even after 30 days of storage ( Fig. 3b; P > 0.01). On the other hand, elevated storage temperatures (37°C), gradually decreased the scavenging activity of non-encapsulated extract. Fortunately, emulsified RGPE (RGPE MEs) was more resilient against storage and similar values were observed between freshly prepared and stored samples (day 30) (P > 0.01). It should be mentioned that the protective effect of emulsification against temperature and time degradation has an onset around 15-18 days, before which no statistical differences are detected between microemulsions and extract solutions. In total, a drastic reduction in scavenging values of almost 23% was estimated for non-emulsified 15 wt% extract solutions at 37°C, whereas the scavenging activity of microemulsions was reduced only by ∼2%.
Temperature is a key factor when it comes to the stability of microemulsions which serve as delivery systems of antioxidant compounds. However, the extent of its influence cannot be easily predicted, as formulations and extract antioxidant properties vary considerably. In general, temperature elevation is accompanied by a loss of the antioxidant potential of phenolic substances. The loss of antiradical activity of phenolic compounds caused by temperature is associated to their chemical degradation (Sólyom et al., 2014;Slavu et al., 2020;Zapata et al., 2021). For instance, the main derivative of gallic acid decarboxylation occurring at high temperatures (105-150°C) is pyrogallol (Boles et al., 1988). Quercetin glycosides under roasting conditions (180°C) form the quercetin aglycone (Rohn et al., 2007), whereas when heated at milder temperatures, protocatechuic acid is identified as the major product of cleavage (Buchner et al., 2006). The overall degradation phenomenon of phenolic antioxidants is enhanced over time even at temperatures under room temperature, whereas refrigerated storage conditions suggest an effective strategy for preserving both TPC content and radical scavenging activity (Gollücke et al., 2009;Tsantili et al., 2011).
Our findings are in line with other studies on antioxidant compounds embedded in microemulsions. The encapsulation of curcumin in O/W microemulsions significantly improved its antiradical properties (Li et al., 2017). Enhanced antiradical properties of Cymbopogon citratus essential oils encapsulated in microemulsions by 47% and 184% as assessed by ORAC (oxygen radical absorbance capacity assay) or PSC (peroxyl radical scavenging capacity assay) compared to the direct use of the oil have also been reported.
As shown by Mostafa et al. (2014) the retention of polyphenols in microemulsions stored at room temperature was improved by 8% compared to those stored at 40°C after 3 months of storage. In another study, no significant differences on the antioxidant activity of curcumin loaded O/W microemulsions when stored at lower temperatures (4 or 25°C) were observed (Amuti et al., 2021).
Hence, room or lower temperatures are suggested as an efficient means for prolonging their antioxidant stability of encapsulated RGPE.

Effect of storage temperature on the content of major phenolic constituents
The quantification of the non-pigment phenolic compounds conducted by HPLC analysis is demonstrated in Figure 4. The samples tested are the initial free RGPE used as 15 wt% aqueous solution, and stored samples of the extract and its emulsified homologue after 30 days at room temperature (25°C) or at 37°C. As shown in our previous work , catechin is one of the main abundant compounds identified (17.74 mg g À1 ), followed by quercetin (6.68), caftaric acid (6.14) and rutin (1.48). In general, the emulsification of the RGPE enhanced the chemical stability of the polyphenols identified and this was more evident in the case of samples stored at elevated temperature (37°C) (P < 0.01). More specifically, catechin was not detected in samples of the free extract after being stored for 30 days at room temperature, whereas in microemulsions it was depleted almost by 66% compared to the initial concentration (P < 0.01). In free extract samples stored at 37°C, both catechin and quercetin could not be detected. On the contrary, in the form of microemulsionswith the exception of catechin -, the degradation of polyphenols was moderate (13.1% or less). Hence, the loss observed on the antioxidant capacity of free extracts at elevated temperature is possibly associated to the complete degradation of these major phenolic components, catechin and quercetin. Slightly different findings are reported by Tolun et al. (2020) who described the effect of storage conditions on the stability of grape pomace extract in polymer microcapsules. Their study revealed that the content of caftaric acid, catechin and rutin in free extract is reduced after 75 days by ∼70-80%, 57-65% and 60-80% respectively depending on relative humidity levels. On the contrary, when encapsulated in maltodextrin-gum Arabic spray dried microspheres the overall reduction in the concentration of the specific antioxidants was restricted to 2 up to 15%. However, they do report that catechin was the most affected compound among encapsulated samples and rutin the most resilient one, that is in line with our findings. Similar degradation levels are also reported for other maltodextrin microspheres containing red wine phenolics (Galmarini et al., 2013). Of course, direct comparisons cannot be made with the present study, but the fact that the polymer beads seemingly act more protectively towards the degradation of phenolic compounds could possibly be ascribed to the low mobility of the core compounds within the solid-like polymer matrices. The complexation of antioxidants with polysaccharides (Tolun et al., 2020) may also provide an additional barrier against their decomposition.

Conclusions
In this study, a new type of microemulsion (DES-in-oil) has been introduced for the successful emulsification of red grape pomace extract, which allowed the entrapment of high amounts of polyphenolic antioxidants with a relatively low concentration of low molecular weight surfactants. The incorporation of RGPE in microemulsions was more efficient in maintaining the antiradical activity of the extract during storage at elevated temperature (37°C), as evidenced by practically unaffected A AR % values. Catechin followed by quercetin were the most sensitive phenolic compounds, since their concentration totally depleted after 30 days of storage at 37°C in the free extract. Emulsification and storage at room temperature ensured the chemical stability and long-term antiradical activity of the red grape pomace extract.

Author contributions
S.I.L., D.P.M. and A.C., experimental design, supervision, monitoring of research work, manuscript reviewing; O.K. laboratory work, analysis, statistical analysis, writing; A.A., V. A., and D.P., laboratory work, analysis, editing; E.B.: manuscript revision. All authors have read and agreed to the published version of the manuscript.