Fragaria viridis Fruit Metabolites: Variation of LC-MS Profile and Antioxidant Potential during Ripening and Storage

Fragaria viridis Weston or creamy strawberry is one of the less-known species of the Fragaria genus (Rosaceae family) with a wide distribution in Eurasia and is still in the shadow of more popular relatives F. ananassa (garden strawberry) or F. vesca (wild strawberry). Importantly, there is a lack of scientific knowledge on F. viridis compounds, their stability in the postharvest period, and bioactivity. In this study, metabolites of F. viridis fruits in three ripening stages were characterized with high-performance liquid chromatography with photodiode array and electrospray ionization triple quadrupole mass spectrometric detection (HPLC-PAD-ESI-tQ-MS). In total, 95 compounds of various groups including carbohydrates, organic acids, phenolics, and triterpenes, were identified for the first time. The quantitative content of the compounds varied differently during the ripening progress; some of them increased (anthocyanins, organic acids, and carbohydrates), while others demonstrated a decrease (ellagitannins, flavonols, etc.). The most abundant secondary metabolites of F. viridis fruits were ellagitannins (5.97–7.54 mg/g of fresh weight), with agrimoniin (1.41–2.63 mg/g) and lambertianin C (1.20–1.86 mg/g) as major components. Antioxidant properties estimated by in vitro assays (2,2-diphenyl-1-picrylhydrazyl radical (DPPH), 2,2′-azino-bis(3-ethylbenzothiazoline-6-sulfonic acid) cation radical (ABTS), ferric reducing antioxidant power (FRAP), and oxygen radical absorbance capacity (ORAC)) showed good antioxidant potential in all ripening stages of F. viridis fruits. The pilot human experiment on the effect of F. viridis fruit consumption on the serum total antioxidant capacity confirmed the effectiveness of this kind of strawberry. Postharvest storage of ripe fruits at 4 °C and 20 °C lead to declining content in the majority of compounds particularly ascorbic acid, ellagitannins, and flavonols, with the most significant loss at room temperature storage. These results suggest that F. viridis fruits are a prospective source of numerous metabolites that have potential health benefits.


Introduction
Genus Fragaria (strawberry) of the Rosaceous family is a well-known source of dietary fruits that are popular all over the world and is widely consumed due to its unique taste and fragrance. The global production of strawberries has reached 9 million tons per year with maximal production levels in China, USA, and Mexico [1]. Such high consumption makes it necessary to examine the various aspects of biology, chemistry, cultivation, and technology of strawberry manufacturing. The most common studies devoted to the metabolic diversity of the Fragaria genus include many groups of compounds such as carbohydrates, organic acids, vitamins, anthocyanins, ellagitannins, flavonoids, and minerals [2]. Special attention is also paid to understanding the nature of postharvest changes in Botanically, F. viridis is a green perennial herbaceous rhizomatous plant that can grow up to 25 cm tall with numerous adventitious roots. The stems are erect, and the length of the leaves are slightly longer, densely dressed with protruding trichomes. The stipules are narrow and brown, and the leaves have shaggy petioles from prominent trichomes that densely cover them. The inflorescences are small, corymbose, loose, and few-flowered and are dressed at the base with a solid or tripartite apical leaf. The pedicels are short and dressed with appressed or occasionally horizontally protruding trichomes. The flowers are relatively large, up to 2.5 cm in diameter, usually bisexual with triangular and lanceolate sepals. The petals are rounded, overlapping each other, short-clawed, and yellowish-white. The fragrant fruits are spherical, narrowed at the base, mostly yellowish-white, only reddish at the top, and rarely entirely pink or pale red, with achenes slightly immersed in the pulp, which are difficult to separate from the receptacle [11]. The fruits are separated from the stem together with sepals at inconsist density, and they are distinguished by good transportability, better that F. vesca and F. ananassa. The species is ecologically plastic. It grows in aspen-birch groves, on open grassy mountain slopes, on edges and glades of mountain forests, in meadows, and in meadow steppes in Europe, Russia, the Caucasus, and Western and Eastern Siberia. Sensory evaluation of F. viridis fruits demonstrated good taste and extraordinary fresh-fruity flavour [12].
The literature data about F. viridis are meagre and demonstrated that the essential oil of leaves consists of major components β-linalool, n-nonanal, tetradecanal, nerolidol, α-bisabolol, and phytol, which distinguishes it from fruit essential oil with the dominant m/p-xylene, isoledene, methyleugenol, α-cedrene, α-muurolene, and α-cedrol [13]. The known phenolics of F. viridis fruits are catechin, epicatechin, epigallocatechin gallate, cyanidin 3-O-glucoside, pelargonidin 3-O-glucoside, quercetin 3-O-glucoside, ellagic acid [14], quercetin 3-O-galactoside, quercetin 3-O-rutinoside, and chlorogenic acid [15], equalling the phenolic profile of F. viridis leaves. Both Botanically, F. viridis is a green perennial herbaceous rhizomatous plant that can grow up to 25 cm tall with numerous adventitious roots. The stems are erect, and the length of the leaves are slightly longer, densely dressed with protruding trichomes. The stipules are narrow and brown, and the leaves have shaggy petioles from prominent trichomes that densely cover them. The inflorescences are small, corymbose, loose, and few-flowered and are dressed at the base with a solid or tripartite apical leaf. The pedicels are short and dressed with appressed or occasionally horizontally protruding trichomes. The flowers are relatively large, up to 2.5 cm in diameter, usually bisexual with triangular and lanceolate sepals. The petals are rounded, overlapping each other, short-clawed, and yellowish-white. The fragrant fruits are spherical, narrowed at the base, mostly yellowish-white, only reddish at the top, and rarely entirely pink or pale red, with achenes slightly immersed in the pulp, which are difficult to separate from the receptacle [11]. The fruits are separated from the stem together with sepals at inconsist density, and they are distinguished by good transportability, better that F. vesca and F. ananassa. The species is ecologically plastic. It grows in aspen-birch groves, on open grassy mountain slopes, on edges and glades of mountain forests, in meadows, and in meadow steppes in Europe, Russia, the Caucasus, and Western and Eastern Siberia. Sensory evaluation of F. viridis fruits demonstrated good taste and extraordinary fresh-fruity flavour [12].
The literature data about F. viridis are meagre and demonstrated that the essential oil of leaves consists of major components β-linalool, n-nonanal, tetradecanal, nerolidol, α-bisabolol, and phytol, which distinguishes it from fruit essential oil with the dominant m/p-xylene, isoledene, methyleugenol, α-cedrene, α-muurolene, and α-cedrol [13]. The known phenolics of F. viridis fruits are catechin, epicatechin, epigallocatechin gallate, cyanidin 3-O-glucoside, pelargonidin 3-O-glucoside, quercetin 3-O-glucoside, ellagic acid [14], quercetin 3-O-galactoside, quercetin 3-O-rutinoside, and chlorogenic acid [15], equalling the phenolic profile of F. viridis leaves. Both fruits and leaf extracts showed good radical-scavenging and ferric-reducing ability [14]. Despite easy cultivation, high breeding potential, and good sensory parameters, F. viridis remains one of the underutilized strawberry species [12]. The growing interest in new strawberries as perspective food sources obliges us to do more in-depth research, particularly in the area of Fragaria metabolomics using high-performance Pharmaceuticals 2020, 13, 262 3 of 25 liquid chromatography-mass spectrometric techniques (nothing has been done previously with these techniques). It is, of course, of great interest in revealing specificities of metabolite changes in F. viridis fruits during the ripening progress as well as metabolite transformation during fruit storage.
In the present report, we realized the first detailed metabolomic profiling of F. viridis fruits in three stages of ripening (unripe, intermediate ripe, and fully ripe) using high-performance liquid chromatography with photodiode array and electrospray ionization triple quadrupole mass spectrometric detection (HPLC-PAD-ESI-tQ-MS), and antioxidant properties of F. viridis fruits were also studied in four in vitro models (scavenging capacity against 2,2-diphenyl-1-picrylhydrazyl radical and 2,2 -azino-bis(3-ethylbenzothiazoline-6-sulfonic acid) cation radical, ferric reducing antioxidant power, and oxygen radical absorbance capacity) during ripening progress and one pilot human experiment (serum total antioxidant capacity). Finally, the variation of selected compounds and antioxidant potential of ripe F. viridis fruits was investigated in response to cool and room temperature storage. To our knowledge, this is the first comprehensive study of F. viridis fruits.

Metabolites of F. viridis Fruits: LS-MS Profile
Chromatographic profiling of F. viridis fruit metabolites was completed by high-performance liquid chromatography with photodiode array and electrospray ionization mass spectrometric detection (HPLC-PAD-ESI-tQ-MS). The identification of components found in F. viridis was done after a precise interpretation of chromatographic (retention times) and spectral data (ultraviolet-visible spectra and mass spectral patterns) in comparison with reference standards and literature data. The extraction procedures of fresh fruits were preliminarily tested with various solvents (methanol, ethanol, isopropanol, water, and acetone), solvent-material ratios, temperatures (20-90 • C), and methods of extraction (ultrasound-, microwave-, and water-bath-assisted). The resultant protocol used was 100% methanol with a solvent-material ratio of 1:1 followed by 5 min homogenization and sonification (30 min, 45 • C). After comparison of F. viridis fruit extracts in three ripening stages (unripe, intermediate ripe, and fully ripe), the HPLC-ESI-tQ-MS chromatogram of fully ripe fruit extract showed the presence of the maximal amount of compounds (95) with interpretable data (Figure 2), details of which are provided in Table 1.
Pharmaceuticals 2020, 13, x FOR PEER REVIEW 3 of 25 fruits and leaf extracts showed good radical-scavenging and ferric-reducing ability [14]. Despite easy cultivation, high breeding potential, and good sensory parameters, F. viridis remains one of the underutilized strawberry species [12]. The growing interest in new strawberries as perspective food sources obliges us to do more in-depth research, particularly in the area of Fragaria metabolomics using high-performance liquid chromatography-mass spectrometric techniques (nothing has been done previously with these techniques). It is, of course, of great interest in revealing specificities of metabolite changes in F. viridis fruits during the ripening progress as well as metabolite transformation during fruit storage.
In the present report, we realized the first detailed metabolomic profiling of F. viridis fruits in three stages of ripening (unripe, intermediate ripe, and fully ripe) using high-performance liquid chromatography with photodiode array and electrospray ionization triple quadrupole mass spectrometric detection (HPLC-PAD-ESI-tQ-MS), and antioxidant properties of F. viridis fruits were also studied in four in vitro models (scavenging capacity against 2,2-diphenyl-1-picrylhydrazyl radical and 2,2′-azino-bis(3-ethylbenzothiazoline-6-sulfonic acid) cation radical, ferric reducing antioxidant power, and oxygen radical absorbance capacity) during ripening progress and one pilot human experiment (serum total antioxidant capacity). Finally, the variation of selected compounds and antioxidant potential of ripe F. viridis fruits was investigated in response to cool and room temperature storage. To our knowledge, this is the first comprehensive study of F. viridis fruits.

Metabolites of F. viridis Fruits: LS-MS Profile
Chromatographic profiling of F. viridis fruit metabolites was completed by high-performance liquid chromatography with photodiode array and electrospray ionization mass spectrometric detection (HPLC-PAD-ESI-tQ-MS). The identification of components found in F. viridis was done after a precise interpretation of chromatographic (retention times) and spectral data (ultraviolet-visible spectra and mass spectral patterns) in comparison with reference standards and literature data. The extraction procedures of fresh fruits were preliminarily tested with various solvents (methanol, ethanol, isopropanol, water, and acetone), solvent-material ratios, temperatures (20-90 °C), and methods of extraction (ultrasound-, microwave-, and water-bath-assisted). The resultant protocol used was 100% methanol with a solvent-material ratio of 1:1 followed by 5 min homogenization and sonification (30 min, 45 °C). After comparison of F. viridis fruit extracts in three ripening stages (unripe, intermediate ripe, and fully ripe), the HPLC-ESI-tQ-MS chromatogram of fully ripe fruit extract showed the presence of the maximal amount of compounds (95) with interpretable data (Figure 2), details of which are provided in Table 1.  Table 1. IS-internal standard (genkwanin).  Table 1. IS-internal standard (genkwanin).

Carbohydrates
Two types of carbohydrates were discovered in F. viridis fruits including hexosyl-hexose (m/z 341; 1) and hexose (m/z 179; 2). The HPLC-MS method used does not allow for identification of the nature of carbohydrates, so we used the HPLC-DAD procedure demonstrating the presence of three compounds identified as glucose, fructose, and saccharose ( Figure S1), usual mono-and disaccharides of strawberry fruits [2].

Ellagic Acid Derivatives and Ellagitannins
Ellagic acid (53) The close mass spectrometric pattern gave the known strawberry ellagitannin fragariin A found in F. ananassa and has the structure of galloylated derivative of agrimoniin [29]. Two isomeric ellagitannins, 51 and 54, were identified as biogenetic relatives to agrimoniin compounds, agrimonic acids A and B, respectively [28].
After all, only ellagic acid was mentioned previously as a component of F. viridis [14], indicating that this is the first report describing the profile of ellagic acid derivatives from F. viridis fruits. Traditionally used strawberries, such as F. ananassa and F. vesca, are a good source of ellagitannins and ellagic acid glycosides [2,18]. Lambertianin C, saguiins, and agrimoniin were also found in many varieties of cultivated strawberries, demonstrating their obligate position in the Fragaria metabolome [2,18,26,27].
Quercetin glycosides were the most diverse group of F. viridis phenolics with 19 members, some of which were identified using standard references.

Triterpenes
Six compounds were triterpenes, and two reference standard defined compounds were tormentic acid (81) and pomolic acid (91). Both compounds are the usual Rosaceous metabolites [39,40] but were not found in Fragaria fruits early studies. Glycosidic derivatives of 81 and 91 were described as two O-hexosides, 78 and 83, and two di-O-hexosides, 75 and 77, and are unusual components of strawberry fruits.

Quantitative Content and LS-MS Profile Variation of F. viridis Fruits during Ripening
We studied F. viridis fruits at three different stages of ripening, including unripe fruits, the stage of technological ripeness (intermediate stage), and the stage of full ripeness ( Table 2). The total simple carbohydrate (mono-and disaccharides) content in F. viridis fruits varied from 41.14 mg/g in the unripe stage to 45.17 mg/g in ripe fruits. The main components were monosaccharides, glucose, and fructose, with a concentration of 41.10-45.16 mg/g responsible for the sweet taste of F. viridis fruits. Monosaccharides are the dominant sugars of F. ananassa [33,41,42] and F. vesca [43], but in some strawberry varieties, it happens that saccharose shows the highest content [34,44]. The sugar content of F. ananassa fruits demonstrated the same trend during ripening, with the lowest content in unripe fruits (3.61-4.45 mg/g) rising to the ripe stage with 4.82-8.20 mg/g [33], indicating the close character of carbohydrates changing in strawberries.
The highest total content of organic acids was found in ripe fruits of F. viridis (7.88 mg/g) and the lowest was found in the unripe stage (4.25 mg/g) including the greatest share of citric acid in all stages of ripening (2.83-5.63 mg/g). The remaining organic acids were minor components with concentration values 0.42-0.59 mg/g for malic acid, 0.37-0.42 mg/g for tartaric acid, 0.01-0.07 mg/g for fumaric acid, and trace-0.05 mg/g for oxalic acid. This results in an increase in the acidity of F. viridis fruits during ripening, which had a positive impact on strawberry taste. The domination of citric acid was demonstrated previously in many F. ananassa cultivars grown in Slovenia (4.4-10.5 mg/g) [42], Pakistan (12.0-14.3 mg/g) [33], and Turkey (5-10 mg/g) [45] as well as in F. vesca (5.6 mg/g) [34]. Moreover, the fruit development resulted in an increase in organic acids in strawberries [33], and malic, tartaric, fumaric, and oxalic acids were the minor acids in other Fragaria fruits [33,34,45].
Particularly noteworthy was the presence of a high level of ascorbic acid, up to 1.12 mg/g in ripe F. viridis fruits, that was significantly more than found in F. vesca (0.4 mg/g) and F. ananassa (0.25-0.9 mg/g) [33,45].
Galic acid derivatives showed trace (gallic acid, 1,2,3,4,6-penta-O-galloylglucose) or low-level content (1-O-galloyl glucose) without significant variation during ripening (0.04-0.05 mg/g). In contrast to the gallic acid derivatives, hydroxycinnamates were important compounds of F. viridis fruits with medium content although the amount decreased during ripening from 0.97 mg/g to 0.60 mg/g. The highest level was found for the p-coumaric acid 4-O-glucoside (0.29-0.35 mg/g), its isomer 24 (0.08-0.14 mg/g), and 5-O-caffeoylquinic acid (0.14-0.28 mg/g). The early study of 5-O-caffeoylquinic acid content variation in dry strawberries showed it ranging from 1.8 to 2.9 mg/g for F. vesca, from 1.2 to 1.7 mg/g for F. viridis, and from 0.7 to 1.8 mg/g for F. moschata [15]. The level of coumaroyl glycosides in Norway F. ananassa cultivars varied from 0.02 mg/g to 0.14 mg/g with maximal content in fully ripe fruits [25], which is different from our findings.
The most significant groups of phenolics with the highest content in all stages of maturity of F. viridis fruits were ellagitannins and ellagic acid derivatives. The general rule of ellagitannin variation in F. viridis fruits (with few exceptions) was decreasing content during ripening. This was particularly manifested in dominant compounds of agrimoniin (2.63→1.41 mg/g), lambertianin C (1.86→1.20 mg/g), fragariin A (0.93→0.63 mg/g), and sanguiin H6 (0.36→0.22 mg/g), where values were maximal in unripe fruits. The decrease of ellagitannin level in strawberry fruits during development was previously shown in some Norway cultivars of F. ananassa; the variation of agrimoniin content in Blink, Polka, and Senga cultivars was 0.72→0.57, 0.66→0.58, and 0.68→0.55 mg/g, respectively [25]. The most drastic fall in ellagitannin content, from 1.14 to 0.30 mg/g, was found in Italian cultivars of F. ananassa [18]. The most likely reason for the ellagitannin changes is due to increasing activity of specific enzymes, such as tannases, reaching the highest values in ripening fruits [46]. This is further illustrated by the slight rising content of ellagic acid, some ellagic acid O-glycosides, and low molecular weight ellagitannins (as pedunculagin, strictinin, castalagin, and casuarictin) that can be considered the breakdown products of ellagitannin molecules, but of course, this issue needs to be discussed additionally.
Anthocyanins, which are important phenolics of strawberries, were at a low level in F. viridis characterized by slight red pigmentation of the outer layer of fruits and depigmented pulp. It is for this reason that the unripe and pre-ripe fruits had trace anthocyanin content. In the ripe stage, the domination of pelargonidin 3-O-glucoside (0.06 mg/g), cyanidin 3-O-glucoside (0.05 mg/g), and pelargonidin 3-O-rutinoside (0.03 mg/g) was observed. The remaining pigments were found in trace levels. Pelargonidin and cyanidin glycosides were also found as components of phenolic pigments of all studied strawberries including F. ananassa [47], F. vesca, and F. moschata [14]. It is to be expected that the ripening of strawberry fruits resulted in intense pigmentation caused by the accumulation of anthocyanins, as in the case of Norway cultivars from 0.2 mg/g in the pre-ripe stage to 0.8 mg/g in fully ripe fruits of F. ananassa [25].   The total concentration of flavonols in F. viridis demonstrated decreasing levels during fruit development from 1.24 mg/g in unripened fruits to 1.01 mg/g in ripened fruits. Quercetin derivatives (0.66-0.82 mg/g) prevailed over kaempferol derivatives (0.35-0.42 mg/g) in all stages of ripening; this was also found in F. ananassa (0.01-0.05 mg/g for quercetin derivatives vs. 0.01-0.02 mg/g for kaempferol derivatives) [9]. The main components of flavonol complex of F. viridis were quercetin 3-O-rutinoside (0.25-0.32 mg/g) and kaempferol 3-O-rutinoside (0.11-0.28 mg/g), followed by two acylated compounds: quercetin 3-O-(6"-O-p-coumaroyl)-glucoside (0.05-0.11 mg/g) and kaempferol 3-O-(6"-O-p-coumaroyl)-glucoside (0.04-0.08 mg/g). The gradual decline of flavonoid concentration was found for flavonol di-O-glycosides and acylated flavonol O-glycosides in contrast to aglycones and flavonol mono-O-glycosides accumulated in ripe fruits. Again, the variation of enzymatic activity of hydrolases may be relevant in the progress of F. viridis fruits ripening. This phenomenon has not been mentioned previously in any strawberries and needs additional experimental data to confirm this finding.
Triterpenes found in F. viridis fruits were trace compounds, but despite this, the quantifiable levels of tormentic acid, its O-hexoside, and pomolic acid were found in the stage of full ripening. The variation of triterpenoids in any Fragaria species was not discussed previously, but it is known that the accumulation of triterpenoids reaches the maximal level in the fully ripe stage of fruits; this was also declared for Chardonnay grape [48], olive fruits [49], and tomato [50].

Antioxidant Potential of F. viridis Fruits: Comparision with Other Strawberries
The activity of fruit total extracts of F. viridis in three stages of ripening was studied in four antioxidant assays including the scavenging capacity against 2,2-diphenyl-1-picrylhydrazyl radical (DPPH), and 2,2 -azino-bis(3-ethylbenzothiazoline-6-sulfonic acid) cation radical (ABTS), ferric reducing antioxidant power (FRAP), and oxygen radical absorbance capacity (ORAC) ( Table 3). In comparison, the activity of two extracts from commercially available ripe fruits of F. vesca (wild strawberry, Regina cultivar) and F. ananassa (garden strawberry, Senga Sengana cultivar) was also studied. Both species are much more common strawberries, and their antioxidant potential has been analysed many times [2]. Table 3. Antioxidant activity of Fragaria extracts in four assays, µM trolox-eq./g of dry weight ± S.D. The extracts of F. viridis fruits in all stages of ripening were effective radical scavengers against both radicals DPPH and ABTS. Variations of the antioxidant potential values were 27.53-29.18 µM trolox-eq./g in the DPPH assay and 35.07-36.22 µM trolox-eq./g in the ABTS assay, while the more active scavenger in the DPPH assay was the extract of unripe fruits and the ABTS assay gave the extract of ripe fruits as more active. The extracts of F. vesca and F. ananassa were less effective in DPPH/ABTS scavenging assays with values of trolox-equivalent content 15.21/19.73 and 9.33/14.67 µM/g, respectively. An early research of F. ananassa extracts in the DPPH assay showed a wide range of fluctuation of antiradical activity from 9.75-12.83 µM BHT-eq./g for Brazil cultivars [4] to 3.00-13.15 µM trolox-eq./g for Polish cultivars [10]. Sikmilar characteristics were found for ABTS assay data varying from 1.50-2.27 µM trolox-eq./g for Japan varieties [51] to 7.06-29.73 µM trolox-eq./g for Polish cultivars [10]. Raudonis et al. [14] analysed the activity of F. vesca and F. moschata extracts using HPLC-assisted ABTS assay, which gave the values of protection 25.11 and 8.24 µM trolox-eq./g, respectively.

Assay
The FRAP assay is one of the most popular assays to estimate antioxidant activity for analysis of edible fruits, and strawberries are not an exception. The level of ferric reducing antioxidant power of F. viridis extracts was high and increased during fruit ripening from 42.63 µM trolox-eq./g in the unripe stage to 47.11 µM trolox-eq./g in ripe fruit extract. The parameters of F. vesca (27.14 µM trolox-eq./g) and F. ananassa (21.06 µM trolox-eq./g) extracts were lower but close to known data (24.84 µM trolox-eq./g for F. vesca [14]). The level of antioxidant activity in the ORAC assay for F. viridis fruit extracts was similar to ABTS data and slightly decreased in ripening progress from 33.62 µM trolox-eq./g (unripe fruits) to 32.98 µM trolox-eq./g (ripe fruits). The ORAC data of F. vesca extract 25.05 µM trolox-eq./g was higher than for F. ananassa extract (18.87 µM trolox-eq./g) but at a level lower than F. viridis. The known information about ORAC potential of strawberries demonstrated high effectiveness of anthocyanins fraction of F. ananassa (2.7-24.46 mM trolox-eq./g) [52], opposite the activity of the total fruit extract (8.90-16.63 µM trolox-eq./g) [53].
Applying the DPPH-radical scavenging-assisted HPLC-PDA-ESI-tQ-MS assay, we identified the compounds responsible for the antioxidant defence of F. viridis extracts. For that to happen, an aliquot of the total extract was separated with chromatography and portions of the eluates were collected. A part of the eluates was used for the DPPH decolouration assay, and the remainder were used for HPLC-PDA-ESI-tQ-MS assay for the qualitative confirmation of compounds ( Figure S2). The results showed that the majority of compounds found in F. viridis were involved in the process of radical scavenging (most likely because of their phenolic nature), but 12 sites of elution gave more pronounced decolouration of the DPPH solution. There were ascorbic acid, ellagic acid, five ellagitannins (pedunculagin, sanguiin H6, lambertianin C, agrimoniin, and fragariin A), two anthocyanins (pelargonidin 3-O-glucoside, and cyanidin 3-O-glucoside), and three flavonols (quercetin 3-O-glucoside, quercetin 3-O-glucuronide, and quercetin 3-O-rutinoside). Three compounds, ascorbic acid, lambertianin C, and agrimoniin, were the most active due to their high content so they were the principal antioxidants of F. viridis fruits.
In brief, the information obtained in four in vitro assays demonstrates the high effectiveness of F. vesca fruit extracts as antioxidant agents, exceeding the activity of two well-known strawberries: F. vesca and F. ananassa. In light of the obtained results about the activity of F. vesca fruits, checking its usefulness in antioxidant protection of the human organisms was considered. To this end, this was done by pilot experiment by analysing total antioxidant capacity (TAC) of blood serum of healthy male volunteers after a 1-week intake of F. viridis fresh ripe fruits at doses of 100, 250, and 400 g/day. The level of TAC before F. vesca fruit intake was 510-516 µM trolox-eq./L. The week-long consumption of F. viridis fresh fruits gave a statistically significant increase of serum TAC level in all dose groups up to 524, 544, and 557 µM trolox-eq./L, respectively, for groups with 100, 250, and 400 g/day consumption (Figure 3).
By comparison, the results of F. vesca and F. ananassa fruit groups (both 250 g/day) were also positive-the consumption of both kinds of strawberries resulted in increases in serum TAC levels, but to a lesser degree (535 µM trolox-eq./L for F. vesca, 525 µM trolox-eq./L for F. ananassa).
This illustrates the good antioxidant potential of F. viridis fruits in any dose applied. To date, it is known that the consumption of F. ananassa resulted in increases in serum TAC by 7-25% [54]. The possible reasons for that phenomenon may be an increase in human serum of strawberry-related metabolites such as pelargonidin-glucuronide, urolithin A-glucuronide [55] and p-hydroxybenzoic acid [56] possessing high antioxidant potential and also the rising of serum antioxidants (glutathione) and serum level of antioxidant enzyme activity (catalase, glutathione peroxidase, and glutathione reductase) [57]. The metabolite profile of F. viridis is qualitatively and quantitatively close to F. ananassa; therefore, it is logical to assume that the increase of serum TAC after F. viridis consumption is caused by enhancement of the serum level of antioxidant metabolites of phenolic nature, serum antioxidants, and antioxidant enzymes. usefulness in antioxidant protection of the human organisms was considered. To this end, this was done by pilot experiment by analysing total antioxidant capacity (TAC) of blood serum of healthy male volunteers after a 1-week intake of F. viridis fresh ripe fruits at doses of 100, 250, and 400 g/day. The level of TAC before F. vesca fruit intake was 510-516 μM trolox-eq./L. The week-long consumption of F. viridis fresh fruits gave a statistically significant increase of serum TAC level in all dose groups up to 524, 544, and 557 μM trolox-eq./L, respectively, for groups with 100, 250, and 400 g/day consumption (Figure 3).
This illustrates the good antioxidant potential of F. viridis fruits in any dose applied. To date, it is known that the consumption of F. ananassa resulted in increases in serum TAC by 7-25% [54]. The possible reasons for that phenomenon may be an increase in human serum of strawberry-related metabolites such as pelargonidin-glucuronide, urolithin A-glucuronide [55] and p-hydroxybenzoic acid [56] possessing high antioxidant potential and also the rising of serum antioxidants (glutathione) and serum level of antioxidant enzyme activity (catalase, glutathione peroxidase, and glutathione reductase) [57]. The metabolite profile of F. viridis is qualitatively and quantitatively close to F. ananassa; therefore, it is logical to assume that the increase of serum TAC after F. viridis consumption is caused by enhancement of the serum level of antioxidant metabolites of phenolic nature, serum antioxidants, and antioxidant enzymes.

Storage Stability of Antioxidants and Antioxidant Potential of F. viridis Ripe Fruits
Twenty compounds with the most pronounced antioxidant effects were quantified in F. viridis fruits in two series of storage experiments (Table 4). Primarily, we studied the change in concentration of ripe fruits stored at two temperatures, 4 °C (cool temperature) and 20 °C (room temperature), to define the stability of antioxidants in fresh fruits for a short period. Table 4. Content of selected antioxidants in ripe fruits of F. viridis (mg/g of fresh fruit weight ± S.D.) and its total antioxidant potential (coulometric titration assay; μmol trolox-eq./g of fresh fruit weight ± S.D.) after storage at 4 °C (1 week) and 20 °C (3 days). . Changes in serum total antioxidant capacity (TAC) before (B) and after (A) 1-week intake of Fragaria fresh ripe fruits (group 1-F. viridis, 100 g/day, n = 5; group 2-F. viridis, 250 g/day, n = 5; group 3-F. viridis, 400 g/day, n = 5; group 4-F. vesca, 250 g/day, n = 4; and group 5-F. ananassa, 250 g/day, n = 6) and 10 g/day fructose (group 6, control group; n = 3). * p < 0.05 vs. control group after intake; ** p < 0.05 vs. same group before intake.

Storage Stability of Antioxidants and Antioxidant Potential of F. viridis Ripe Fruits
Twenty compounds with the most pronounced antioxidant effects were quantified in F. viridis fruits in two series of storage experiments (Table 4). Primarily, we studied the change in concentration of ripe fruits stored at two temperatures, 4 • C (cool temperature) and 20 • C (room temperature), to define the stability of antioxidants in fresh fruits for a short period.
The seven-day storage of ripe F. viridis fruits at 4 • C and three-day storage at 20 • C were the maximal periods of storage without external damage (browning, rotting, untypical smell, and taste) [58].
The storage of fresh F. viridis fruits at room temperature (20 • C) resulted in more drastic changes within a shorter period. The level of ascorbic acid declined from 1.14 mg/g to 0.36 mg/g (68.4%) for three days; additionally, anthocyanins became a trace compound. The strong reduction of content was detected for the polymeric ellagitannins, lambertianin C (42.6%), agrimoniin (36.7%), and fragariin A (66.2%) in opposition to ellagic acid, pedunculagin, and sanguiin H6, which increased at 150.0, 48.5, and 50.0%, respectively. The flavonoid biocide quercetin 3-O-rutinoside showed a statistically significant decrease of content from 0.24 mg/g to 0.18 mg/g (25%) coupled with a rising level of quercetin 3-O-glucoside and quercetin 3-O-glucuronide. Table 4. Content of selected antioxidants in ripe fruits of F. viridis (mg/g of fresh fruit weight ± S.D.) and its total antioxidant potential (coulometric titration assay; µmol trolox-eq./g of fresh fruit weight ± S.D.) after storage at 4 • C (1 week) and 20 • C (3 days).

Compound
T, • C Day The parameter of total antioxidant potential decreased from 4.12 to 0.52 µmol trolox-eq./g or 87.4% less antioxidant potential. Postharvest storage of ripe fruits is inextricably linked to senescence causing changes in biochemical profiles, biomolecules and polymers degradation, cell dysfunction and disintegration, and the leaking of enzymes [58]. Not long after, the fruits begin rotting, which reduces its alimentary value. In our study, the ripe F. viridis fruits after storage showed negative changes in content of ascorbic acid and anthocyanins, which are environmentally unstable plant compounds diminished in light and high humidity [59,60], just like polymeric ellagitannins and rutin typically degrading after contact with oxygen and esterase-like enzymes [61,62]. The preservative value of cool temperature (4 • C) was higher than room temperature (20 • C), saving antioxidants and the antioxidant potential of fruits longer. The decrease in phenolic compounds and ascorbic acid content in strawberries was shown in the number of papers. Anthocyanin content decreased in F. ananassa fruits during refrigerated storage at 4 • C in cultivars Camarosa (385→46 mg/kg) [63,64] and Elsanta (40→20 mg/g) [65]. The ascorbic acid level was also unstable at 0-20 • C with a loss of about 40% (cultivars Dover, Campineiro, and Mazi) [3] or more (cultivar Camarosa) [63]. The content of ellagic acid and flavonol monoglucosides in cool storage (5 • C) of F. ananassa fruits tends to rise as in the Selva cultivar from 19.9 to 26.8 µg/g for ellagic acid, from 40.1 to 44.1 µg/g for quercetin derivatives, and from 13.7 to 15.8 µg/g for kaempferol derivatives [66]. Our findings revealed that various strawberries (F. viridis and F. ananassa) have the same response during storage at cool and room temperature conditions.  (Table S1). Ellagitannins sanguiins H2, H6, and H10; agrimonic acids A and B; and agrimoniin were isolated previously in our laboratory from Rosaceous species (purity 90-95%) [28,67,68]

Total Extract Preparation from Fragaria Fruits
For preparation of the total extract of Fragaria fruits, the fresh material was homogenized in a Grindomix GM 200 grinder (Retsch GmbH, Haan, Germany) and 100 g was extracted twice with stirring in a glass flask (0.5 L) with methanol (100 mL) using an ultrasonic bath Sapphire 2.8 (Sapphire Ltd., Moscow, Russia) for 30 min and at 50 • C (ultrasound power 100 W and frequency 35 kHz).
The extracts were filtered through cellulose, concentrated in vacuo until dryness, and stored at 4 • C before use for chemical analysis and biological activity study. The yields of total extracts of Fragaria fruits were 10.63 g (F. viridis unripe fruits), 11.02 g (F. viridis intermediate ripe fruits), 12.43 g (F. viridis fully ripe fruits), 15.63 g (F. vesca fully ripe fruits), and 17.33 g (F. ananassa fully ripe fruits). The UV-Vis spectra were obtained in the spectral range of 200-600 nm. MS detection was performed in negative ESI mode using the parameters as follows: temperature levels of ESI interface, desolvation line, and heat block were 300 • C, 250 • C, and 400 • C, respectively. The flow levels of nebulizing gas (N 2 ), heating gas (air), and collision-induced dissociation gas (Ar) were 3 L/min, 10 L/min, and 0.3 mL/min, respectively. The MS spectra were recorded in the negative mode (−3-−5 kV source voltage) by scanning in the range of m/z 50-2000 at the collision energy of 5-40 eV. The system was managed under LabSolution's workstation software with the inner LC-MS library. The identification of compounds was done by the analysis of their retention time, ultraviolet, and mass-spectrometric data, comparing the same parameters with the reference samples and/or literature data. Before analysis, the sample of F. viridis fruits dry extract (10 mg) was dissolved in 50% MeCN (25 mL), filtered (0.22-µm PTFE syringe filter), and injected (1 µL) into the HPLC-DAD-ESI-tQ-MS system for analysis.

High-Performance Liquid Chromatography with Diode Array Detection (HPLC-DAD): Carbohydrate Analysis
The composition of free carbohydrates was analyzed by high-performance liquid chromatography with diode array detection (HPLC-DAD) using the procedure described previously [69]. To prepare the sample, dry extracts of F. viridis fruits (5 mg) were dissolved in 20 mL of deionized water and passed sequentially through a series of two cartridges Dowex ® 50WX8 (H + -form; 10 mL) and Dowex ® 1 × 8 (Cl − -form; 10 mL) eluted with water (20 mL). The final eluates were reduced in vacuo (20 mL) and filtered using 0.22-µm PTFE syringe filter before injection into the HPLC-DAD system for analysis.

HPLC-ESI-tQ-MS: Metabolite Quantification
To quantify compounds 1-95 in F. viridis fruits, we used HPLC-MS data (MS peak area) obtained in early conditions (Section 3.3). The reference standards (48 compounds; Table S2) were accurately weighed (10 mg) and individually dissolved in DMSO-50% methanol mixture (1:10) in a volumetric flask (10 mL). The stock solutions were used to build external standard calibration curves generated using six data points, 100, 50, 25, 10, 5, and 1 µg/mL followed by plotting the MS peak area vs. the concentration levels. The validation criteria (correlation coefficients, r 2 ; standard deviation, S YX ; limits of detection, LOD; limits of quantification, LOQ; and linear ranges) were calculated using the previous recommendations [70] (Table S2). All analyses were carried out in triplicate, and the data were expressed as mean value ± standard deviation (S.D.). The sample solution was prepared from homogenized F. viridis fruits (50 mg) and 5 mL of methanol in an Eppendorf tube. The mixture was sonicated for 30 min at 50 • C (ultrasound power 100 W, frequency 35 kHz), centrifuged (6000× g), filtered (using 0.22-µm PTFE syringe filter), and transferred to the volumetric flask (10 mL), and the final volume was reduced to 10 mL by 50% MeOH before HPLC-ESI-tQ-MS analysis. Genkwanin was used as the internal standard (final concentration 25 µg/mL in acetonitrile).

Antioxidant Activity: In Vitro Assays
Radical scavenging activity of Fragaria extracts against the 2,2-diphenyl-1-picrylhydrazyl radical (DPPH) and the 2,2 -azino-bis(3-ethylbenzothiazoline-6-sulfonic acid) cation radical (ABTS) was studied using microplate spectrophotometric decoloration assays as described previously [71,72]. The value of the ferric reducing antioxidant power (FRAP) was measured by spectrophotometric assay based on the reduction of the Fe 3+ -2,4,6-tri(2-pyridyl)-1,3,5-triazine complex to the ferrous form at low pH [73]. To determine the level of oxygen radical absorbance capacity (ORAC), we used an assay based on peroxyl radical generation by thermal decomposition of 2,2 -azobis(2-amidino-propane) dihydrochloride followed by fluorimetric detection [74]. All assays used trolox as a reference standard (methanolic solution 0.5-100 µg/mL), and the calibration curve was created by plotting the trolox concentration (µg/mL) vs. the absorbance (or fluorescence). The values of antioxidant parameters were expressed as µmol trolox-equivalents/g of dry weight. All the analyses were carried out five times and the data were expressed as mean value ± standard deviation (SD).

DPPH Radical Scavenging Assisted HPLC-PDA-ESI-tQ-MS Assay
High-performance liquid chromatography with photodiode array detection and electrospray ionization triple quadrupole mass spectrometric detection (HPLC-PDA-ESI-tQ-MS) assisted with spectrophotometric DPPH radical scavenging assay was realized in the chromatographic conditions described in Section 3.3 with enlarged injection volume at 30 µL. The eluates (50 µL) were collected every 30 s using an automated fraction collector (Econova, Novosibirsk, Russia) in 96-well microplates, then dried under a N 2 -stream, and redissolved in 50 µL of 50% methanol. An aliquot (25 µL) of the methanolic solution was mixed with DPPH solution (50 µg/mL in methanol) and absorbance was measured at 520 nm fifteen minutes later by a Bio-Rad microplate reader Model 3550 UV (Bio-Rad Labs, Richmond, CA, USA). The most active antioxidants gave strong decoloration of the DPPH solution, and corresponding eluates were separated in known HPLC-PDA-ESI-tQ-MS conditions again in order to confirm the presence of separate compounds.

Serum Total Antioxidant Capacity
Twenty-eight men, aged 20-25 years, were recruited. All were free from hypertension, cardiovascular disorders, and alcohol abuse; none smoked or took any other drug and oral medication. We had the guarantee that all subjects had a similar diet and lifestyle because they were recruited from the same community with a refectory service. All subjects gave their informed consent for inclusion before they participated in the study. The study was conducted in accordance with the Declaration of Helsinki, and the protocol was approved by the Ethics Committee of Institute of General and Experimental Biology (protocol No. LM-0324, 27 January 2012). The volunteers were divided on six experimental groups: group 1-F. viridis fruits, 100 g/day (n = 5); group 2-F. viridis fruits, 250 g/day (n = 5); group 3-F. viridis fruits, 400 g/day (n = 5); group 4-F. vesca fruits, 250 g/day (n = 4); group 5-F. ananassa fruits, 250 g/day (n = 6); and group 6-fructose, 10 g/day (n = 3). Then, they took Fragaria fruits or fructose for 1 week (3 times a day in equal portions). Before and after the test, blood was drawn from the antecubital vein into a heparinized syringe, and immediately after blood drawing, serum was prepared by centrifugation (6000× g) and the serum total antioxidant capacity was estimated. Phosphate buffer (10 mM, pH 7.2; 100 µL), myoglobin solution (5 µM; 50 µL), ABTS solution (3 mM; 20 µL), and serum sample (20 µL) were mixed in 96-well microplate and incubated 3 min at 25 • C. Then H 2 O 2 solution (250 µM; 20 µL) was added and immediately measured at 600 nm for 5 min at 25 • C. A lag time (in sec) was estimated as the suppression period of ABTS oxidation (or absorbance increasing). The reference compound (trolox; 1, 2.5, 5, and 10 µM) was analyzed using the same protocol, and the calibration curve was created by plotting the lag time (in s) vs. the absorbance at 600 nm. The value of the serum total antioxidant capacity was expressed as µmol trolox-equivalents/L. All the analyses were carried out in triplicate, and the data were expressed as mean value ± standard deviation (SD).

F. viridis Fruit Storage Experiment
Seven and three portions of the fresh F. viridis fruits (200 g) were placed into individual polystyrene bags (300 mL) and incubated at 4 • C (7 days) or 20 • C (3 days), respectively, in a ventilated MK 53 thermostat (BINDER GmbH, Tuttlingen, Germany). Five portions (20 g each) of fresh F. viridis fruits were taken out of storage for analysis every 24 h, extracted as described previously (Section 3.5), and analyzed using HPLC-ESI-tQ-MS quantitative procedure (Section 3.5) or used without pre-extraction for the total antioxidant potential determination by coulometric assay (Section 3.10).

Total Antioxidant Potential of Fresh F. viridis Fruits: Coulometric Assay
The total antioxidant potential of fresh F. viridis fruits was found using a sightly modified bromine radical scavenging assay based on the coulometric titration method with electrogenerated bromine radicals [17,75]. Potentiostat Expert-006 (Econics Expert Ltd., Moscow, Russia) with a four-electrode two-compartment electrochemical cell was used for measurements. The working electrode was a bare platinum foil (surface area 1 cm 2 ), and the auxiliary electrode was a platinum wire isolated from the anodic cell with a semipermeable diaphragm. To detect the titration end-point (∆E = 200 mV), a pair of polarized platinum electrodes was used and the electrochemical generation was carried out from the supporting electrolyte (0.25 M KBr in 0.1 M H 2 SO 4 ) at a current density 5 mA·cm −2 , providing 100% current yield. To start the measurement, the portion fruit of F. viridis (50 g) with various storage periods was homogenized and 10 mg of homogenate was inserted into the coulometric cell (50 mL) containing 20.0 mL of supporting electrolyte. The time of titration was used for the total antioxidant potential calculation expressed in units of the quantity of electricity (Coulombs (C)) spent for titration of the full probe of homogenized fruits. The trolox solutions were used as a reference compound (500, 250, 100, 50, and 10 µg/mL in methanol) titrated coulometrically, and a calibration curve was plotted in coordinates "concentration (µg/mL)-the quantity of electricity (C)". Finally, the value of the total antioxidant potential was calculated as mg trolox-equivalents per g of fresh fruits. Values are expressed as mean obtained from ten independent experiments.

Statistical Analysis
Statistical analyses were performed using a one-way analysis of variance (ANOVA), and the significance of the mean difference was determined by Duncan's multiple range test. Differences at p < 0.05 were considered statistically significant. The results are presented as mean values ± S.D. (standard deviations) of some (3)(4)(5)(6)(7)(8)(9)(10) replicates.

Conclusions
The current study reported the metabolic profile of fruits of Fragaria viridis in various stages of ripening using the HPLC-DAD-ESI-tQ-MS technique not applied previously to this strawberry species. About a hundred compounds were characterized, and this is many more than the previously reported amount of F. viridis metabolites [14,15]. The largest number of components were phenolics, particularly ellagitannins and flavonol glycosides, forming the basis of F. viridis metabolome in all stages of ripening. In addition, derivatives of gallic acid, ellagic acid, hydroxycinnamates, coumarins, procyanidins, catechins, and anthocyanins were also found. Non-phenolic compounds, such as carbohydrates and organic acids, were quantitatively predominant, opposite triterpenes, with trace levels found. The concentrations of all compounds were affected by the ripening process with increased (anthocyanins and non-phenolics) or decreased (the majority of phenolic compounds) values to a fully ripe stage. This indicates that the ripening of F. viridis fruits is a complex process impacting the quantitative profile of metabolites. The high content of ascorbic acid and selected phenolics in F. viridis fruits were the source of strong antioxidant properties of fruit extracts, in particular free radical scavenging capacity, ferric reducing antioxidant power, and oxygen radical absorbance capacity studied in in vitro models. The same is true for human experiments, which demonstrated that the serum total antioxidant capacity increased significantly after a week's consumption of F. viridis fruits. Changes in antioxidant content and total antioxidant potential of fresh F. viridis fruits was found during storage at 4 • C and 20 • C, with the safest condition at 4 • C storage used within a week. The information received in our study highlighted the potential of F. viridis fruits as a source of antioxidant metabolites that need more scientific attention and wider implementation in the human diet.
Supplementary Materials: The following are available online at http://www.mdpi.com/1424-8247/13/9/262/s1, Figure S1: High-performance liquid chromatography with diode array detection chromatogram of free sugars in F. viridis ripe fruits, Figure S2: High-performance liquid chromatography with electrospray ionization triple quadrupole mass spectrometric detection chromatogram of F. viridis ripe fruits extract coupled with spectrophotometric DPPH radical scavenging assay, Table S1: Reference standards used for the qualitative and quantitative analysis by HPLC-DAD-ESI-tQ-MS assays, Table S2: Regression equations, correlation coefficients, standard deviation, limits of detection, limits of quantification, and linear ranges for 48 reference standards.