Physicochemical properties and bioavailability of naturally formulated fat‐soluble vitamins extracted from agricultural products for complementary use for natural vitamin supplements

Abstract The purpose of the current study was to evaluate the physicochemical properties, digestive stability, storage stability, and intestinal absorption of formulated natural vitamins (FNV) by mixing fat‐soluble vitamins extracted from agricultural products with their synthetic vitamin (SYNV) counterparts using a 6 to 4 ratio (w:w, dry weight). The FNV A, D, E, and K were evenly dispersed without crystal growth in the dispersion specifications for the functional tablet foods. The FNV A, D, E, and K had 89, 73, 65, and 36% of the digestive recovery, respectively, which was comparable to that of the SYNV. FNV D, E, and K were retained over 77%, but rapidly decreased to 15% after 6 months during accelerated storage at 25 30 and 35℃. The comparable radical scavenging capacity was found between the FNV and the SYNV. Results from the current study suggest that fat‐soluble vitamins extracted from agricultural products could be reasonable complementary use for natural vitamin supplements.

The metabolism of high lipophilic food microconstituents, such as fat-soluble vitamins, can start in the stomach where the foods are exposed to acidic conditions and gastric enzymes (Campbell, 2017).
In contrast to water-soluble vitamins, fat-soluble vitamins are not readily absorbed and are required to interact with bile acid and pancreatic lipase in digestive fluids, which is shown in Figure 1 ( Goncalves et al., 2015). In detail, vitamin A was frequently esterified and hydrolyzed to retinol and fatty acids. The free retinol was then integrated with other lipids into the formed micelles, and it was absorbed by the brush border membranes of the intestine cells (Kiokias, Proestos, & Varzakas, 2016). After digestion, the fat-soluble vitamins are typically absorbed at particular locations in the small intestine, and they are incorporated with the digestion of ingested dietary fats and oils (Goncalves et al., 2015;Wang et al., 2019). Once they are micellized, the physiological regulators in the intestinal membrane are responsible for their absorption (Schubert et al., 2018;Takada et al., 2015). For instance, vitamin D is absorbed by the passive and the active transport mechanisms (Reboul et al., 2011). The absorbed fat-soluble vitamins join into chylomicrons are activated in the liver and then they are released into the lymphatic circulation system (Porter & Charman, 1997). Fat-soluble vitamins generally have poor water solubility during digestion, and encapsulation within in the lipid-based delivery systems is suggested in order to improve their bio-accessibility. In contrast to vitamins A, D, and E, the intestinal absorption regarding the physiological regulator for vitamin K has not been well-documented.
Since vitamins are not synthesized or formed from inappropriate amounts from the human body, they have to be ingested as outsourced food and nutritional supplements in order to prevent various diseases related to their deficiencies (Challem, 1999). A previous study found that consumers prefer vitamin supplements that contain natural additives to synthetic additives in regard to efficacy and safety (Devcich, Pedersen, & Petrie, 2007;Topliss et al., 2002).
However, natural substances have drawbacks with the actual biological uses due to low stability during processing and storage (Pokorný, 2007). For instance, the biological activity of natural antioxidants is typically poorer than the biological activity of synthetic antioxidants, because natural antioxidants are not pure ingredients, and there is a dearth of information on their safety (Pokorný, 2007). In order to overcome these limitations, complementary use between the natural and the synthetic compounds as an approach for new vitamin supplements has been previously proposed. To be suitable as dietary fat-soluble vitamin supplements that originated from natural substances intended for human intake, these types of formulations must be enabled in order to be solubilized and absorbed from the digestive system and enterocyte beyond their stability. Furthermore, the stability of fat-soluble vitamins loaded in re-assembled casein micelles was mostly affected by the conditions of the storage (Loewen, Chan, & Li-Chan, 2018). It is plausible that their biological activities, such as radical scavenging activity, could be affected by the digestive and the storage stability once it is added into food or supplements. Thus, the purpose of the current study was to evaluate the physicochemical properties, which include the stability in the pH, dispersion, the storage stability, and the antioxidant activity as well as to estimate the digestive stability and the intestinal absorption of the natural fat-soluble vitamins formulated with vitamins extracted from the agricultural products.

| Sample preparations
The standard stock solution of the vitamins was prepared and stored in a dark at room temperature (+1~ −13 ℃) until further analysis. The lyophilized powders of the carrots, tomatoes, mushrooms, and spinach were supplied by Aribio. The extraction yield of the freeze-dried powder of the carrots, tomatoes, mushrooms, and spinach was 9.96%, 6.03%, 10.74%, and 10.35%, respectively. The extraction of the fatsoluble vitamins from each powder was carried out by the method described in the previous studies (Ostermeyer & Schmidt, 2006;Turner & Mathiasson, 2000). In detail, the freeze-dried powder of the carrots and the tomatoes was weighed in a round bottom flask. 260 ml of ethanol, 10 ml of pyrogallol ethanol solution, and 30 ml of 90% potassium hydroxide solution were added into it. The mixture went through saponification for 2 hr in a water bath at 65℃, and then it was quickly cooled, which was followed adding 300 ml of water. After that, it was transferred to a 2,000 ml separator funnel. 300 ml of petroleum ether was added into the separator funnel, and it was then gently shaken for 30 min. The petroleum ether layer was collected into a separate flask, and the compound was then re-extracted with 300 ml of petroleum ether, which was followed by collection once again as above. Both of the petroleum ether layers were concentrated under vacuum pressure from 40℃ to 50℃. The final solution was diluted one time with distilled water, and it was then filtered through 0.45-μm membrane filters.
The aliquot amount of mushrooms (30 g) was taken in a round bottom flask and mixed with 260 ml of ethanol, 10 ml of pyrogallol ethanol solution, and 30 ml of 90% potassium hydroxide solution.
The same method was applied as described above. According to a previous study (Jäpelt & Jakobsen, 2016), approximately 60 g of spinach weighed into a separator funnel was mixed with 240 ml of methanol and 400 ml of dichloromethane (V: V, 1:1.6), and it was then shaken 5 times. After 1 hr, the supernatant from the liquid layer was transferred to a new distillation flask, and a mixture of methanol and dichloromethane (V: V, 1:1.6) was added for shaking. After 1 hr, both of the liquid extracts were pooled together and then passed through Whatman filter paper No. 42 (MF-Millipore). The supernatant liquid layer collected after the filtering was concentrated under vacuum pressure from 40℃ to 50℃. The final solution was diluted with distilled water (V: V, 1:1) and filtered through 0.45-μm membrane filters.

| Formulation of tablet for natural vitamins
Formulated natural vitamins (FNV), which include vitamins A, D, E, and K, were prepared by Aribio Company. In detail, natural vitamins A, D, and E and extracted powder from agricultural products were mixed with synthetic vitamins (SYNV, 100% of chemicals) purchased from DSM Nutritional Products (LTD, Kaiseraugst, Switzerland) at the ratio of 60% to 40% in a dry weight base where 1% of magnesium stearate and 1% of silicon dioxide were mixed in order to make a tablet. In the case of vitamin K, it was formulated with a 25% mixture of synthetic vitamins, the vitamins extracted powder from agricultural products (4:6, w: w, dry weight), 1% of magnesium stearate, 1% of silicon dioxide, and 73% of crystalline cellulose. The aliquot amount (2.4 g) of each formulated natural vitamin (FNV) was used for further experiments.

| Identification and quantification of a vitamin
High-performance liquid chromatography (HPLC) with a photodiodearray detector (PDA) (Waters) that was equipped with an isocratic pump, a degasser, and automatic injection system (Model Ultimate 3000, Thermo Science) was used. The column used in this study was a YMC-Pack ODS-A C18 (150 mm x 4.6 mm, 5 μm). For the analysis of the β-carotene, the analytical column was maintained at a temperature of 22 ± 1 ℃. The mobile phase consisted of a mixture of methyl tertiary-butyl ether (MTBE) with methanol (60:40, v/v). The flow rate was set to 2 ml/min with a UV absorbance of 450 nm. To analyze the retinal acetate, ergocalciferol (D 2 ), cholecalciferol (D 3 ), DL-alphatocopherol acetate, and phylloquinone (K 1 ), and the temperature of column was maintained at 30 ± 1℃. Acetonitrile (ACN) was used as a mobile phase. The flow rate was maintained at 1 ml/min with a UV absorbance of 280 nm (Fanali, D'Orazio, Fanali, & Gentili, 2017). The chromatographic peaks for each vitamin were identified by comparing the retention times of the samples with those of the standard compounds. Quantification was conducted using a standard calibration equation obtained from each external standard.

| Measurement of the water solubility
The experiments on the water solubility of the functional tablet food made of the agricultural products were conducted by the method developed in the Korean Food Standards Codex (Korea Ministry of Food and Drug Safety, 2018a) and in 1907 in the Edition of Pharmacopoeia Helvetica (Al-Gousous & Langguth, 2015). In order to measure the water solubility of the FNV A, D, E, and K, an aliquot amount of each FNV (2.4 g) was dispersed in 24 ml of distilled water and then was shaken for 30 min at 100 rpm in a water bath that was maintained at 37℃.

| Measurements of digestive stability and intestinal uptake
The digestive stability and the intestinal cellular uptake of the FNV A, D, E, and K were estimated by using the in vitro bio-mimic system coupled with a caco-2 cell as described by Jeong et al. (2018) with slight modifications. The small intestinal phase of digestion was initiated adjusting to a pH 2.0 ± 0.1 by 0.1M NaHCO3 and 4 ml of pancreatin (2 mg/ml PB), lipase (1 mg/ml PB), and bile acid (12 mg/ ml PB) were added. After being treated in a Caco-2 cell, the obtained cells were cultured at 37℃ for 30 min and sonicated for 30 min. The sonicated cells were centrifuged at 13,000 rpm for 5 min at 4℃. The supernatant was filtered through a PTFE syringe filter (Advantec, pore size 0.45 μm) and stored at −20℃ until the analysis.

| Prediction of shelf life
Formulated natural vitamins (FNV) A, D, E, and K were stored in a temperature/humidity-controlled chamber (HANBAEK) at base temperatures (30 ± 1 ℃) and two temperatures (25 ± 1 ℃ and 35 ± 1 ℃) for comparison. The relative humidity was set at a constant value of 90%. The FNV A, D, E, and K were kept at three temperatures of 25 ± 1℃, 30 ± 1℃, and 35 ± 1℃ and were stored for 6 months. After 1.2, 2.4, 3.6, 4.9, and 6 months, the stored samples were taken out in order to measure the contents of the FNV A, D, E, and K. The safety faction (0.8) was applied to determine the final shelf life of the FNV A, D, E, and K estimated for a temperature of 30℃.

| Measurement of oxygen radical absorbance capacity (ORAC)
The antioxidant capability was measured using the oxygen radical absorbance capacity (ORAC) assay described by (Son, Choi, Kim, Park, & Shim, 2016). The ORAC values were converted as micromolar Trolox equivalents (TE) per 1 mg of sample (TE µmol/g). The curves of the relative fluorescence intensity of vitamins A, D, E, and K from the FNV and the SYNV were expressed as the net area under curve (Net AUC = AUC of Fluorescence sample − AUC of Fluorescence blank).

| Statistical analysis
All the values were performed in triplicate. The analysis of variance (ANOVA) with the Tukey's post hoc test at a significant level of p < .05 was conducted using GraphPad Prism 3.0 software (GraphPad Software).

| Determination of fat-soluble vitamins from selected agricultural products
The retention time (Rt) of each fat-soluble vitamin standard simultaneously identified was as follows: β-carotene = 1.216 min, alpha-tocopherol acetate = 17.610 min, ergosterol = 22.550 min, and phylloquinone = 25.068 min (data not shown). Each standard curve from the β-carotene, ergosterol, alpha-tocopherol acetate, and phylloquinone was used for the quantification of each vitamins A, D, E, and K, respectively. Fat-soluble vitamins including vitamin A, D, E, and K extracted from selected agricultural products such as carrots, mushrooms, tomatoes, and spinach, respectively, were also F I G U R E 2 Identification and quantification of fat-soluble vitamins (A, D, E, and K) by HPLC-UV. The amount of vitamin extraction from the agricultural products, such as carrots, mushrooms, tomatoes, and spinach. The data include the means ± the SEM of the triplicated samples identified and quantified by comparison of its retention time and standard curve, respectively. The amount of vitamin A from carrots, vitamin D from tomatoes, vitamin E from mushrooms, and vitamin K from spinach was determined to be 0.921 (3,066.242 IU/g), 0.606 (2018.541 IU/g), 0.0955, and 0.0035 mg/g of dry weight, respectively ( Figure 2). The content (0.921 mg/g of dry weight) of vitamin A obtained from the carrots, which was calculated by the β-carotene, was higher than the levels obtained by the previous studies (0.789 mg/g of dry weight and 0.038 ~ 0.079 mg/g fresh weight, respectively) (Rawson, Tiwari, Tuohy, O'Donnell, & Brunton, 2011;Zaccari, Cabrera, Ramos, & Saadoun, 2015). A previous study found that the quantity of vitamin D 2 in mushrooms calculated from ergosterol was 0.0268 ~ 0.458 mg/g of dry weight (Gąsecka, Magdziak, Siwulski, & Mleczek, 2018), which was comparable to the result of the current study (0.606 mg/g of dry weight). The amount of vitamin E based on the standard calibration of alpha-tocopherol acetate was 0.0955 mg/g dry weight in tomatoes, which was higher than that of a previous finding (0.06 mg/g dry weight) from Raiola et al., (2015).
Vitamin K 1 calculated by phylloquinone amounted to 0.0035 mg/g dry weight. It was about 806 times higher than vitamin K 1 in fresh spinach (4,340 ug/100 g of dry weight of spinach), which was reported by Jäpelt and Jakobsen (2016).
At the same time, the current study estimated the contribution of the consumption of each representative agricultural product for each of the fat-soluble vitamins to the dietary reference intake (DRI) for Koreans or the adequate intake (AI) (Ministry of Health & Welfare, 2015). The DRI for vitamin A is 690 µg retinol activity equivalents (RAE)/day, and the AI for vitamins D, E and K is 11, 11.8, and 71 µg/ day, respectively. In the current study, the determined amount of vitamin A from the agricultural products was equal to 1.32% of the DRI, and vitamins D, E, and K reached to 60%, 0.008%, and 0.05% of the AI, respectively (data not shown). The results from the current study imply that carrots, mushrooms, tomatoes, and spinach could provide good natural sources of vitamins A, D, E and K, respectively.
Based on these results, the current study further examined the physicochemical properties and the bioavailability of the fat-soluble vitamin extracts from the agricultural products to see the possibility of complementary use as natural vitamin supplements.

| Dispersion of fat-soluble vitamins for functional tablet foods
Dispersion is a physical process related to mechanically decomposing tablets or capsules into small particles, and a tablet orally administered should be dissolved in the gastrointestinal tract fluid before absorption (Krsteska, Kostovski, Brzilova, Sejfulah, & Ugarkovic, 2010). To ensure that the active ingredient in the FNV A, D, E, and K is released efficiently from a formulated tablet and is then decomposed, the dispersion test was conducted as described in the materials and the methods. In the case of the FNV (A and E) and the SYNV (A and E), it was evenly dispersed without any crystal growth, and there was no significant difference between the natural and synthetic one (p > .05) (Figure 3). In contrast, vitamins D and K for both the FNV and the SYNV were initially dispersed in water without forming crystals, but they were gradually precipitated within 30 min after solubilizing in water at 37℃ (Figure 3). The results from the current study indicated that all single-tablet formulation fully dispersed in water at 37℃ within 30 min and met the dispersion time requirement of the regulation. Thus, the FNV A, D, E, and K formulated with vitamins extracted from the agricultural products followed by mixing the synthetic vitamins could be within the dispersion specifications. The rate of dissolution influenced the absorption rate and the efficacy of the tablet products, which significantly influenced its bioavailability (Silva, Webster, Bou-Chacra, & Löbenberg, 2018). Hence, the current study subsequently evaluated the bio-accessibility and the intestinal absorption in the gastrointestinal tract fluid by using an in vitro biomimetic system.

| Bio-accessibility and intestinal cellular uptake of FNV
The initial amount of vitamins A, D, E, and K from each FNV was calculated according to each calibration curve of retinyl acetate, cholecalciferol, alpha-tocopherol acetate, and phylloquinone, respectively ( Figure 4a). The FNV A, D, E, and K contained 30.129 mg/g, 0.910 mg/g, 0.214 mg/g, and 0.032 mg/g, respectively, which was approximately 1.5, 1.1, 1.1, and 1 times higher than that of the SYNV A, D, E, and K, respectively. To estimate their bioavailability, the bioaccessibility and the intestinal cellular uptake were determined using the in vitro digestion model system coupled with a Caco-2 cell. The bio-accessibility was calculated by the relative percent of the content of the vitamin detected in the aqueous fraction followed by the digestion compared to the initial vitamin amount before digestion.
The bio-accessibility of the FNV A, D, E, and K was 89%, 73%, 65%, and 36%, respectively, which was comparable to the bio-accessibility of the SYNV (Figure 4b). Excluding the vitamins D and K, the bio-accessibility of the FNV A and the E (89% and 65%) were significantly lower than the SYNV A (99%) and the E (74%) (p < .05). The bio-accessibility of the FNV D (73%) was significantly higher than the SYNV D (57%) (p > .05). However, there was no significant difference between the FNV K (36%) and the SYNV K (37%) (p > .05).
The Caco-2 cellular uptake of the FNV was normalized by the cellular protein (µg/g protein), and it occurred in the following order: 0.0001 µg/g, 0.0007 µg/g, and 0.0004 µg/g for the FNV A, D, and E, which is illustrated in Figure 4c. In the case of the SYNV, 0.0002 µg/g of A, 0.0009 µg/g of D, and 0.0007 µg/g of E were up-taken by the Caoc-2 cell. Vitamin K was not detected in the Caco-2 cell, which indicates that it was below the limit of detection (<0.05 mg/L). The absorption of the fat-soluble vitamins from the gastrointestinal tract consists of several consecutive stages, which include the physicochemical and the enzymatic reactions (Borel, 2003). In particular, the bile acid played an essential role to facilitate the absorption of the fat-soluble vitamins (Saeed, Dullaart, Schreuder, Blokzijl, & Faber, 2017;Setchell et al., 2013). A previous study reported that the higher bio-accessibility of the fat-soluble vitamin incorporation into the micelles provided greater absorption efficiency (Reboul, 2015).
The micelles were generated by the lipolysis of the dietary fat where there are a mixture of phospholipids, cholesterol, lipid digestion products, and bile salts (Reboul, 2015). The FVN tablet, which was formulated by vitamins extracted from the agricultural products with synthetic vitamin in a 6 to 4 ratio (w:w), did not affect the digestive stability compared to the SYNV. This implies that the matrix from the agricultural products could be protected from the enzymatic reactions and formed micelle during the digestion. In contrast to the bio-accessibility, the intestinal absorption of the FNV was lower than that of the SYNV, which could be affected by the dietary lipids and Further studies to elucidate this mechanism should be conducted. In the case of vitamin K, which cannot be refluxed through the Caco-2 cells, it is still unknown whether vitamin K is mediated by the other active transporters.

| Prediction of shelf life for FNV during accelerated storage
The changes in the amount of the FNV A, D, E, and K at three temperatures, which included 25℃, 30℃, and 35℃, during storage are shown in Table 1. Overall, the vitamin content showed a F I G U R E 3 Dispersion test-left: before and after dispersing SYNV; right: before and after dispersing FNV A, D, E, and K in the water at 37℃ for 30 min decreasing tendency during storage in the FNV A, D, E, and K samples after accelerated storage for 6 months. In detail, the FNV A initially contained 148.83 ± 0.47 mg/g, and it significantly dropped to 22.85 ± 0.04 mg/g at 25℃, 27.27 ± 0.06 mg/g at 30℃, and 36.13 ± 0.06 mg/g at 35℃ after 6 months. The content of the FNV D was 1.66 ± 0.01 on day 0, 1.51 ± 0.01 for 6 months at 25℃6 months 1.52 ± 0.02 for 6 months at 30℃, and 1.54 ± 0.02 for 6 months at 35℃. In the case of the FNV E, it was 476.62 ± 0.56 on day 0, 420.77 ± 3.32 at 25℃, 427.89 ± 1.21 at 30℃, and 440.34 ± 0.69 at 35℃ during the 6 months. The amount of the FNV K was maintained at 100%, 77%, and 89% at 25℃, 30℃, and 35℃, respectively. Significant degradation was found in vitamin A that was stored at both 25℃ and 30℃, which had 15% and 18% remaining compared to 24% remaining at 35℃, which implies that the storage temperature for vitamin A was critical point rather than the oxygen and the humidity. The results were in agreement with the previous findings that reported that the storage length was by far the most decisive factor, and a negative correlation between vitamin A retention and storage time existed (Hemery et al., 2018;Wang, 2019). The stability of the FNV D and the E F I G U R E 4 Determination of vitamins A, D, E, and K in FNV (a), bio-accessibility of FNV (b) and the intestinal cellular uptake of FNV (c). The data include the means ± 95% CI of the triplicated samples. An asterisk indicates a significant difference from the positive control (SYNV). (**p < .01; *p < .05) ranged from 91% to 93% and 88% to 92%, respectively, during the 6 months of storage. This finding was similar to the previous results, which vitamin D and vitamin E were not easily degraded by processing or during storage (Loewen et al., 2018;Raikos, 2017).
To describe the temperature dependence of the reaction rate constant, the k values were used (Table 2), and they were plotted as linearity to the corresponding 1/T temperature. The obtained activation energy (Ea) and the k0 are shown in Figure 5. By combining the zero or the first-order motion model, a storage time calculation model with the temperature (T) and the quality factor variables were also obtained as shown in Table 3. Considering the statistical analysis, which is based on the R 2 , it was found that the contents of the FNV A, E, and K followed the first-order reaction, while the content of the FNV D fit the zero-order reaction as than the first-order reaction. Thus, each linear regression for the FNV A, E, and K was calculated by the first-order equations, and the linear regression of the FNV D was calculated by the zero-order equations that corre- A reliable shelf-life assessment for functional foods was suggested to be critical to verify how much vitamins will last in order to provide the proper bioactivity, which depends on the preferences of the customers (Calligaris, Manzocco, Anese, & Nicoli, 2016). Based on these results, some complementary ways could be recommended in order to optimize the retention of the vitamins in the FNV A, D, E, and K by modulating the exposure time to air as short as possible in order to delay or reduce the oxidation and the degradation reactions. Hence, the current study measured the radical scavenging capacity of the FNV in order to examine its oxidation during storage. The ORAC value for the FNV A (1,004.92 µmol TE/g) and the D (983.42 µmol TE/g) was slightly lower than that of the SYV, but it was not significantly different from the SYV A and the D (p > .05) ( Figure 5b). However, the FNV E and the K showed 1,019.71 µmol TE/g and 1,472.85 µmol TE/g, which indicates that it was higher than the SYN E (785.91 µmol TE/g) and the K (546.26 µmol TE/g).

| Radical scavenging capacity of FNV
Dietary antioxidants, which are predominantly found in fruits and vegetables, were used to prevent the process of oxidation and the chemical reaction promoted by the reactive oxygen species, such as oxygen peroxide (Park et al., 2017;Shinagawa, Santana, Araujo, Purgatto, & Mancini-Filho, 2018). A previous study found that fresh carrots and spinach had 1.93 µmol TE/g and 28.14 µmol TE/g of ORAC (Shiwakoti, Zheljazkov, & Schlegel, 2018). The results from the current study found that the FNV A and the K had 520 and 52 times stronger antioxidant capacity, respectively. In the case of the ORAC for the FNV D (983.42 µmol TE/g), it exceeded 9 times the amount found in mushrooms (113.9 µmol TE/g), which are known as an exceptional source of this nutrient (Smolskaitė, Venskutonis, & Talou, 2015). The effect of the ORAC for the FNV E (1,019.7 µmol TE/g) was about 317 times higher than a previous finding on fresh tomatoes (3.22 µmol TE/g) (Köstekli et al., 2016). Taken in conjunction, the FNV mainly used by agricultural products could be the TA B L E 1 Determination of the initial and the final contents of FNV A, D, E, and K at three temperatures

| CON CLUS IONS
The fat-soluble vitamins A, D, E, and K were extracted from carrots, mushrooms, tomatoes, and spinach, in order to formulate natural vitamins (FNV), which was followed by formulating with synthetic vitamins with 6 to 4 ratios. The FNV was well dispersed in water without a crystal form, which indicates that it was within the dispersion specifications for a functional table for foods. The bio-accessibility of the FNV D was higher than the bio-accessibility of the SYNV D, but the FNV A, E, and K were not significantly different than that of the SYNV A, E, and K, which was followed by the in vitro digestion.
However, the intestinal absorption by the Caco-2 cell of the FNV,

TA B L E 3
Prediction of the shelf life of the samples FNV A, D, E, and K which included the FNV D, was lower than the SYNV. The FNV A degradation was high and occurred rapidly, but the FNV D, E, and K contents did not significantly change during the 6 months storage time under the same storage conditions. The changes of the radical scavenging capacity in the FNV A, D, E, and K were comparable to those of the SYNV. The results from the current study suggest that the FNV A, D, E, and K could be used as complementary ingredients for the natural vitamin supplements.

ACK N OWLED G EM ENTS
This work was supported by the Technology development Program S2449295 funded by the Ministry of SMEs and Startups (MSS, Korea).

CO N FLI C T O F I NTE R E S T
The authors declare that they have no conflicts of interest.