Preliminary Observations on Viola calcarata as a Source of Bioactive Compounds: Antioxidant Activity and Phytochemical Proﬁle of Two Alpine Subspecies

: Viola L. is a botanical genus with approximately 525 to 620 species, spread worldwide. Several violets are traditionally used as edible ﬂowers and have been recently proved to be a source of bioactive compounds, including ﬂavonols, ﬂavanols, benzoic acids, and cinnamic acids. However, no information is available about the phytochemical proﬁle of the Viola calcarata complex, which is found in the Alpine environment. Thus, the present research aimed to assess the antioxidant activity and the presence of bioactive compounds (anthocyanins and phenolic compounds) in V. calcarata subspecies, to promote their biodiversity and use in the agrifood sector. Two V. calcarata subspecies were chosen, with different colors: V. calcarata subspecies calcarata L., with white (CW), yellow (CY), and violet ﬂowers (CV); and V. calcarata subspecies villarsiana (Roem & Schult.) Merxm., with bicolor (violet and yellow—VB) ﬂowers. CY showed a signiﬁcantly higher phenolic content (1116.43 mg GAE 100 g − 1 FW) than the other subspecies, while CV showed higher values in anthocyanins content (44.73 mg C3G 100 g − 1 FW). Regarding the antioxidant activity, CW (215.07 mmol Fe 2+ kg − 1 FW, 99.53 µ mol TE g − 1 FW, and 32.30 µ mol TE g − 1 FW for FRAP, DPPH, and ABTS, respectively) and VB (217.33 mmol Fe 2+ kg − 1 FW, 90.97 µ mol TE g − 1 FW, and 29.17 µ mol TE g − 1 FW for FRAP, DPPH, and ABTS, respectively) showed the highest values. Through HPLC, a total of eight phenolic compounds were quantitatively identiﬁed among the two subspecies, including ﬂavonols, cinnamic acids, benzoic acids, catechins, and vitamin C. Though different in their composition, the two subspecies are rich in phenolic compounds, highlighting the importance of preserving their biodiversity and their potential use in the agrifood sector.


Introduction
Wild flowers can be an effective source of phytochemicals with antioxidant activity important for human health [1][2][3][4]. Since the third millennium, several studies assessed their chemical composition [3], increasing the knowledge on their edible flowers properties, thereby globally increasing the demand for this kind of product [2,5,6].
Viola L. is a botanical genus with approximately 525 to 620 species, spread throughout the temperate regions and montane zones of the tropics worldwide [7]. Several studies have been conducted on some of these species [1,2,[8][9][10], investigating their potential as a source of bioactive compounds, including flavonols, flavanols, benzoic acids, and cinnamic acids [1,11].
Violet flowers have been used as food product since ancient times [5]. Viola tricolor L., having a refreshing and delicate taste, and many color combinations, is one of the most used often in sweets, salads, drinks, etc. [1]. Other species, such as Viola cornuta L., Viola decoctions, and syrups. Violet flowers are reported also to be useful for stomach ache, inflammation, pain, and cough, as they cause sweating, thus purifying the blood [20]. It could be important to conserve the diversity of V. calcarata subspecies for possible commercial exploitation, not only as ornamental flowers, but also as edible flowers or a source of interesting bioactive compounds.
The aim of this study was to explore the presence of bioactive compounds in V. calcarata species, in order to promote the local biodiversity and its use in the agrifood sector. We investigated the phenolic profile and the antioxidant activity (FRAP, DPPH, and ABTS) in two V. calcarata subspecies, with different colors: V. calcarata subspecies calcarata, with white, yellow, and violet flowers; and V. calcarata subspecies villarsiana, with bicolor (violet and yellow) flowers.

Plant Material
Fresh V. calcarata flowers were collected in 2017 around the Lake of Moncenisio, in the locality Plan des Fontainettes (Val-Cenis, France), at an altitude of 2090-2100 m.a.s.l. (Long. 338655.6; Lat. 5012069). The weather data of the sampling period (May), namely rainfall (83.4 mm) and maximum (14.5 • C) and minimum (7.2 • C) average temperatures, were not substantially dissimilar across the 10-year (2010-2020) average data of May in the same area: 93.4 ± 41.5 mm (mean value ± standard deviation), 13.5 ± 1.6 • C, and 5.9 ± 1.4 • C [21]. Each subspecies was associated with its phytosociological optimum: Elyno seslerieteae variae class for V. calcarata subsp. calcarata and Juncetea trifidi class for V. calcarata subsp. villarsiana [18]. V. calcarata subsp. calcarata was collected in three different colors: white, yellow, and violet (abbr. CW, CY, and CV, respectively) ( Figure 1A-C); V. calcarata subsp. villarsiana was bicolor (violet and yellow-abbr. VB) ( Figure 1D). decoctions, and syrups. Violet flowers are reported also to be useful for stomach ache, inflammation, pain, and cough, as they cause sweating, thus purifying the blood [20]. It could be important to conserve the diversity of V. calcarata subspecies for possible commercial exploitation, not only as ornamental flowers, but also as edible flowers or a source of interesting bioactive compounds.
The aim of this study was to explore the presence of bioactive compounds in V. calcarata species, in order to promote the local biodiversity and its use in the agrifood sector. We investigated the phenolic profile and the antioxidant activity (FRAP, DPPH, and ABTS) in two V. calcarata subspecies, with different colors: V. calcarata subspecies calcarata, with white, yellow, and violet flowers; and V. calcarata subspecies villarsiana, with bicolor (violet and yellow) flowers.
Approximately 100 g of flowers were collected per subspecies in summer 2017 at the optimal phenological stage (i.e., in full flowering), placed in sealed polyethylene bags, immediately stored at 4 °C in a portable refrigerator and brought to the laboratory for analysis.
Part of the material was weighed fresh and then oven-dried at 50 °C until a constant weight was reached, to determine the dry matter.  Approximately 100 g of flowers were collected per subspecies in summer 2017 at the optimal phenological stage (i.e., in full flowering), placed in sealed polyethylene bags, immediately stored at 4 • C in a portable refrigerator and brought to the laboratory for analysis.
Part of the material was weighed fresh and then oven-dried at 50 • C until a constant weight was reached, to determine the dry matter.

Extract Preparation
Samples of fresh flowers were grinded with a mortar and pestle into a fine powder using liquid nitrogen, and then stored at −80 • C until ultrasound extraction, performed to obtain the extracts. One gram of flower powder was extracted with 50 mL of a water:methanol solution (1:1), at room temperature with an ultrasound extractor (Sarl Reus, Drap, France) at 23 kHz for 15 min [11]. The obtained solution was filtered with one layer of filter paper (Whatman No. 1, Maidstone, UK), then with a 0.45 mm PVDF syringe filter (CPS Analitica, Milano, Italy). The extracts were stored at −20 • C until analyses, which were performed in three replicates for each subspecies.

Total Phenolic Compounds
The total phenolic content of the V. calcarata extracts was determined following the Folin-Ciocalteu method [11,22]. The analysis was performed as follows: 1000 µL of diluted 1:10 Folin-Ciocalteu reagent were mixed with 200 µL of flower extract in each plastic tube. The samples were left in the dark at room temperature for 10 min, and then 800 µL of Na 2 CO 3 (7.5%) were added to each plastic tube. After 30 min in the dark at room temperature, absorbance was read at 765 nm by means of a spectrophotometer (Cary 60 UV-Vis, Agilent Technologies, Santa Clara, CA, USA), and the results were expressed as mg of gallic acid equivalents (GAE) per 100 g of fresh weight (mg GAE 100 g −1 FW).

Total Anthocyanins
The total anthocyanin content was estimated by the pH differential method using two buffer systems, as described in the literature [2,23,24]. The analysis was performed as follows: 1 mL of phytoextract was diluted in a 10 mL volumetric flask, and then made up to volume with an aqueous buffer solution at pH 1 (KCl and HCl-25 mM). The same was made in a second volumetric flask with an aqueous buffer solution at pH 4.5 (C 2 H 3 NaO 2 and C 2 H 4 O 2 -0.4 M). Samples were left in the dark at room temperature for 20 min. Absorbance of both flasks was measured at 515 nm and 700 nm by means of a spectrophotometer (Cary 60 UV-Vis, Agilent Technologies, Santa Clara, CA, USA), and the results were expressed in milligrams of cyanidin3Oglucoside per 100 g of fresh weight (mg C3G 100 g −1 FW).

Antioxidant Activity FRAP Assay
The first procedure adopted to evaluate the antioxidant activity in V. calcarata extracts was the ferric ion reducing antioxidant power (FRAP) method [11,25,26]. To obtain the FRAP solution, a buffer solution at pH 3.6 (C 2 H 3 NaO 2 ·3H 2 O + C 2 H 4 O 2 in water), 2,4,6tripyridyltriazine (TPTZ, 10 mM in HCl 40 mM), and FeCl 3 ·6H 2 O (20 mM) were mixed. Afterwards, 30 µL of flower extract were mixed with 90 µL of deionized water and 900 µL of FRAP reagent. The samples were then placed at 37 • C for 30 min and absorbance was measured at 595 nm by means of a spectrophotometer (Cary 60 UV-Vis, Agilent Technologies, Santa Clara, CA, USA). Results were expressed as millimoles of ferrous iron equivalents per kilogram of fresh weight (mmol Fe 2+ kg −1 FW).
The working solution of DPPH radical cations (DPPH, 100 µM) was obtained by dissolving 2 mg of DPPH in 50 mL of MeOH. The solution must have an absorbance of 1000 (±0.05) at 515 nm. The samples were prepared by mixing 40 µL of flower extract with 3 mL of the DPPH radical solution. The samples were then left in the dark at room temperature for 30 min. Absorbance was measured at 515 nm by means of a spectrophotometer (Cary 60 UV-Vis, Agilent Technologies, Santa Clara, CA, USA). The DPPH radical-scavenging activity was calculated as where Abs0 is the absorbance of the control (extraction solution without sample) and Abs1 is the absorbance of the sample. The antioxidant capacity was plotted against a Trolox calibration curve and results were expressed as µmol of Trolox equivalents per gram of fresh weight (µmol TE g −1 FW).
The working solution of ABTS radical cation (ABTS) was obtained by the reaction of 7.0 mM ABTS stock solution with a 2.45 mM potassium persulfate (K 2 S 2 O 8 ) solution. After an incubation time of 12-16 h in the dark at room temperature, distilled water was used to dilute the working solution, thus obtaining an absorbance of 0.7 (±0.02) at 734 nm. The antioxidant activity of the V. calcarata samples was assessed by mixing 30 µL of flower extract with 2 mL of ABTS radical solution. The samples were then left in the dark at room temperature for 10 min, and absorbance was measured at 734 nm by means of a spectrophotometer (Cary 60 UVVis, Agilent Technologies, Santa Clara, CA, USA). Th ABTS radical-scavenging activity was calculated as where Abs0 is the absorbance of the control (extraction solution without sample) and Abs1 is the absorbance of the sample. The antioxidant capacity was plotted against a Trolox calibration curve and results were expressed as µmol of Trolox equivalents per gram of fresh weight (µmol TE g −1 FW).

Phenolic Profile and Vitamin C
The bioactive compounds present in the extracts of V. calcarata flowers were determined using High-Performance Liquid Chromatography (HPLC) with Diode Array Detection (DAD) (Agilent 1200, Agilent Technologies, Santa Clara, CA, USA) [29]. The separation of compounds was obtained with a Kinetex C18 column (4.6 × 150 mm, 5 mm, Phenomenex, Torrance, CA, USA) and different mobile phases, according to a previously validated methodology (Table 1) [30,31]. The identification of the compounds was made by comparison with retention times and UV spectra of the analytical standards (purity ≥ 95%; Sigma Aldrich, St. Louis, MO, USA) and the quantification was achieved using calibration curves at the same chromatographic conditions. The following bioactive compounds were determined: phenolic acids (cinnamic acids: caffeic, chlorogenic, coumaric, and ferulic acid; benzoic acids: ellagic and gallic acid); flavonols (hyperoside, isoquercitrin, quercetin, quercitrin, and rutin); flavanols (catechin and epicatechin); and vitamin C. The results are expressed as mg 100 g −1 of fresh flower.

Statistical Analysis
All data were subjected to statistical analysis of the normality and homoscedasticity through a Shapiro-Wilk test and Levene test, respectively. Mean comparisons were computed using the one-way ANOVA test by means of SPSS 25 software (version 25.0; SPSS Inc., Chicago, IL, USA). A nonparametric Kruskal-Wallis test with stepwise comparison was performed on data groups where the variances were not homogeneous, as per the Levene test. Correlations among the bioactive compounds of the V. calcarata subspecies were calculated by Pearson's correlation coefficient test by means of PAST 4.03 software. A principal coordinate analysis (PCA) biplot was performed for the single phenolic compounds and categories of compounds (total phenolic and total anthocyanin) using the same software. Eigenvalues were calculated using a covariance matrix among the 240 traits used as input, and the two-dimensional PCA biplot was constructed.

Results and Discussion
The dry matter, total phenolic content, total anthocyanin content, and antioxidant activity (FRAP, DPPH, ABTS) of the four examined V. calcarata samples are reported in Table 2. Data are expressed on a fresh-weight basis. The statistical relevance is provided (*** = p < 0.001; ** p < 0.005; * p < 0.05; ns = not significant). Different letters inside a column indicate significant differences between the subspecies according to Tukey's post-hoc test (p < 0.05); nd = not detected.
No significant differences were found in dry matter percentage, ranging between 17.33% and 18.13% in VB and CV, respectively. The total phenolic content varied from 719.30 to 1116.43 mg GAE 100 g −1 , being significantly higher in CY. This result is higher than the values found by studies on other species: da Silva and colleagues (2020) found in yellow V. × wittrockiana flowers a total phenolic content of 725.50 mg GAE 100 g −1 FW, similar to data found by Demasi et al. (2020) in V. cornuta fresh flowers (767.26 mg GAE 100 g −1 FW). The CV and VB values (719.30 mg GAE 100 g −1 FW and 845.63 mg GAE 100 g −1 FW, respectively) are in agreement with data regarding the blue V. × wittrockiana flowers (716.50 mg GAE 100 g −1 FW) [32] and the V. cornuta flowers [2]. Conversely, CW showed a higher phenolic content than V. × wittrockiana white flowers (73.00 mg GAE 100 g −1 FW) [32]. All the four V. calcarata subspecies showed a higher phenolic content than V. odorata (428.40 mg GAE 100 g −1 FW) [11].
Anthocyanins were significantly higher in CV, were similarly present in CY and VB, but not detected in CW. The values ranged from 7.47 to 44.73 mg C3G 100 g −1 FW, being Despite showing slight differences in antioxidant activity, depending on the assay used (Table 2), the results showed that CW had the highest antioxidant activity (215.07 mmol Fe 2+ kg −1 FW, 99.53 µmol TE g −1 FW, and 32.30 µmol TE g −1 FW for FRAP, DPPH, and ABTS, respectively), together with VB (217.33 mmol Fe 2+ kg −1 FW, 90.97 µmol TE g −1 FW, and 29.17 µmol TE g −1 FW, for FRAP, DPPH, and ABTS, respectively), whereas CY and CV had a lower antioxidant activity..The antioxidant activity is mainly exerted by flavonoids and phenolic acids [33]. It is noteworthy that although CY has the highest phenolic content, its antioxidant activity is not the major one. A positive correlation between antioxidant activity and total phenol content was also not found in other crops such as apricot [34] and strawberry [35]. This might be due to the interaction of the Folin-Ciocalteu reagent with other compounds besides phenols, e.g., some inorganic ions, thiols, or proteins [36], suggesting the importance of deepening the analysis with the identification of single compounds by means of chromatographic methods. No significant differences were found in the FRAP assay. These values are higher than those obtained from fresh V. odorata flowers [11], collected as wild species in their natural habitat too; however, regarding the FRAP assay, Demasi and colleagues [2] found a higher result in V. cornuta. González-Barrìo et al. [37] found similar levels of antioxidant activity in V. × wittrockiana, with FRAP results ranging from 96.87 to 206.37 mmol Fe 2+ 100 g −1 DW, compared to V. calcarata results expressed in dry weight (CW: 122.2 mmol Fe 2+ 100 g −1 DW; CY: 127.97 mmol Fe 2+ 100 g −1 DW; CV: 92.7 mmol Fe 2+ 100 g −1 DW; and VB: 125.41 mmol Fe 2+ 100 g −1 DW).
HPLC analysis was performed to determine the compounds mainly contributing to the phenolic profile of the four V. calcarata subspecies (Table 3). Six phenolic acids (four cinnamic and two benzoic acids), five flavonols, and two flavanols (catechins) together with vitamin C were evaluated. The results showed that each subspecies had a specific phenolic composition. Among the 14 compounds investigated, five were found in CW (quercitrin, ferulic acid, ellagic acid, epicatechin, and vitamin C) and six were found in the other three subspecies (CY, CV, and VB), although each one had a different combination of them. The statistical relevance is provided (*** = p < 0.001; ** p < 0.005; * p < 0.05; ns = not significant). Different letters inside a column indicate significant differences between the subspecies according to Tukey's post-hoc test (p < 0.05); nd: not detected. Despite six compounds not being detected in any of the four subspecies, all classes of phenolic compounds were found to be present (flavonols, cinnamic acids, benzoic acids, catechins), and vitamin C. Four out of five flavonols were detected, thus being the most abundant class in V. calcarata subspecies, followed by benzoic acids and catechins (one out of two compounds detected), and lastly cinnamic acids (only one out of four compounds detected).
Regarding the ellagic acid and vitamin C content, no significant differences were found among the four Viola subspecies.
Conversely, hyperoside was found in CY, CV, and VB, ranging from 12.10 to 93.70 mg 100 g −1 FW, being less abundant in CV and VB, and showing a significantly higher value in CY. Similarly, quercitrin was found in CW, CY, and VB, ranging from 15.50 to 136.85 mg 100 g −1 FW, being less abundant in CY and showing a significantly higher content in VB.
Rutin was found in CV and VB, ranging from 16.73 to 329.63 mg 100 g −1 FW, being more abundant in CV; ferulic acid was found in CW, CY, and CV, ranging from 144.50 to 288.63 mg 100 g −1 FW, being more abundant in CY.
Epicatechin was found in three samples (CW, CY, and VB), showing a significantly higher value in VB (176.83 mg 100 g −1 FW), and similar lower values in CW and CY (22.33 and 21.00 mg 100 g −1 FW, respectively). Lastly, isoquercitrin was only detected in CV (5.40 mg 100 g −1 FW).
The correlation analysis highlighted that the total phenolic content was positively correlated (p < 0.05) with the FRAP assay (Figure 2), and the three methods for the evaluation of the antioxidant activity positively correlated with each other, thus confirming the positive link between phenolic compounds and antioxidant activity [3,39], as already highlighted in several edible flowers by Demasi et al. [11], who obtained the same correlation. The anthocyanins content was negatively correlated (p < 0.05) with the DPPH assay; however, it had a positive correlation (p < 0.05) with isoquercitrin and rutin. These two latter compounds were both negatively correlated with antioxidant activity, as were ferulic acid and hyperoside ( Figure 2). The latter were positively correlated with each other. Isoquercitrin and rutin were positively correlated (p < 0.05) with each other, as were quercitrin and epicatechin.
The relationship between single compounds and categories of compounds were evaluated through a two-dimensional PCA scatterplot (based on the first two principal components (PCs)) ( Figure 3). The first two PCs explained 92.5% of total variation. The first PC accounted for 56.4%, while the second PC accounted for 36.1%. The scatterplot showed that the four samples of V. calcarata are clearly distinguished.
For the positive values of both PCs, CY samples are found to be positively correlated mainly with phenolic compounds and hyperoside. For the positive PC1 values and the negative PC2 values, CW was found to be mainly correlated with quercitrin and epicatechin, as well as VB. For the negative PC1 values and the positive PC2 values, CV was mainly correlated with rutin. The relationship between single compounds and categories of compounds were evaluated through a two-dimensional PCA scatterplot (based on the first two principal components (PCs)) ( Figure 3). The first two PCs explained 92.5% of total variation. The first PC accounted for 56.4%, while the second PC accounted for 36.1%. The scatterplot showed that the four samples of V. calcarata are clearly distinguished.
For the positive values of both PCs, CY samples are found to be positively correlated mainly with phenolic compounds and hyperoside. For the positive PC1 values and the negative PC2 values, CW was found to be mainly correlated with quercitrin and epicatechin, as well as VB. For the negative PC1 values and the positive PC2 values, CV was mainly correlated with rutin.   The relationship between single compounds and categories of compounds were evaluated through a two-dimensional PCA scatterplot (based on the first two principal components (PCs)) ( Figure 3). The first two PCs explained 92.5% of total variation. The first PC accounted for 56.4%, while the second PC accounted for 36.1%. The scatterplot showed that the four samples of V. calcarata are clearly distinguished.
For the positive values of both PCs, CY samples are found to be positively correlated mainly with phenolic compounds and hyperoside. For the positive PC1 values and the negative PC2 values, CW was found to be mainly correlated with quercitrin and epicatechin, as well as VB. For the negative PC1 values and the positive PC2 values, CV was mainly correlated with rutin.

Conclusions
Promoting the biodiversity of the Alp territory could help the development of new supply chains, based on local genetic resources, for the sustainable development of mountain territories. In this study, the flowers of V. calcarata subspecies calcarata and villarsiana, with different colors, were interesting not only for their ornamental value but also for their antioxidant activities and phytochemical profiles.
The two subspecies showed the same amount of vitamin C, but different phenolic compositions, mainly consisting of flavonols. This is the reason why it should be important to preserve the biodiversity of these two subspecies since they can be a valuable source of different bioactive compounds.
Their exploitation as edible flowers could be positive for the alpine economy and in particular for plant growers. However, currently there is no information on the proper cultivation techniques to grow V. calcarata subsp. calcarata and villarsiana, this then being an opportunity for future study. Future studies could also focus on the evaluation of seed viability, germinability in a controlled environment and in the field, adaptation to transplanting, crop viability, and establishing crop cycles and cultivation parameters.