**1. Introduction**

**\***

> In recent years, edible flowers (EFs) have gained more attention due to their potential as a functional food with health benefits. This increased interest is also because customers are increasingly demanding in terms of nutrition. Flowers used in the human diet must be edible, which means harmless and non-toxic; thus, blossoms treated with chemical pesticides are unfit for consumption [1,2].

> The primary gastronomic use of blossoms stems from their alluring colour; the assortment of EFs includes several dozen species with a significant number of shapes, colours, and flavours [3,4]. According to Lu et al., the number of EFs varies depending on region, but roughly 97 families, 100 genera, and 180 species are found worldwide [5]. Popular edible ornamental flowers are begonia, borage, calendula, carnation, cornflower, daylily, chrysanthemum, hibiscus, nasturtium, pansy, rose, tulip, and others. In addition to the ornamental flowers described above, EFs include several vegetables (broccoli and cauliflower), herbs (common sage, chives, marjoram, mint, thyme, and summer savory), or the blossoms of some fruit trees (elderberry and apple).

> Nowadays, EFs are used in cuisine for flavour, garnish, and improved nutritional value of food, but they also represent a new opportunity for gastronomic innovations [6,7].

**Citation:** Mlcek, J.; Plaskova, A.; Jurikova, T.; Sochor, J.; Baron, M.; Ercisli, S. Chemical, Nutritional and Sensory Characteristics of Six Ornamental Edible Flowers Species. *Foods* **2021**, *10*, 2053. https://doi.org/ 10.3390/foods10092053

Academic Editor: Antonello Santini

Received: 10 August 2021 Accepted: 28 August 2021 Published: 31 August 2021

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

Flowers can be consumed in various forms, including fresh, dehydrated, lyophilised, cooked, and candied. Common uses for them are salads, sauces, jellies, soups, meat dishes, dyes, beverages, ice cubes, syrups, desserts, and cakes [8–10].

The main component of blossoms is water; its content ranges from 70 to 95% [11]. The content of nutrients like lipids, proteins, carbohydrates, and vitamins is similar to those in vegetables [12]. However, the mineral content of EFs shows the most significant variability regarding nutrient composition, as it is affected by the micronutrients in the soil [13]. The beneficial compounds for human health include antioxidants, vitamins, phenolics, carotenoids, flavonoids, minerals, and others [14]. Moreover, bioactive substances high content represents a beneficial component of the diet because of the possible development of valuable nutraceuticals [15]. However, the recommended limits of toxic agents must be regarded to utilise flowers as food [12].

Some edible flowers are traditionally used as medicinal herbs, and several previous studies revealed their biologically active molecules with potential health effects [16]. For example, these bioactive compounds can lower the risk of cardiovascular and cancer diseases, and they additionally have many beneficial properties like anti-inflammatory, antibacterial, antidiabetic, diuretic, antifungal and others [17–25]. The positive efficacy in lessening the risk of some diseases is due to these molecules' ability to reduce the damage induced by reactive oxygen species (ROS) [26]. These bioactive molecules probably may have a prolonged effect inside the body, as significant antioxidant activities (AA) have been found after the digestion in vitro of selected blossoms [27].

For example, the importance of AA in roses, begonias, and nasturtiums was mentioned by Friedman [28]. Because of the inclusion of polyphenols and ascorbic acid, nasturtium flowers demonstrated an exceptional tendency to exhibit unstable and highly reactive free radicals [29,30]. High values of AA recorded in extracts of rose flowers show a significant inhibitive effect on ROS [31–33]. In daylily extracts, intense scavenging ROS activity and lipid peroxidation were also observed [34,35]. *C. officinalis* extract had stronger AA when it came to scavenging free radicals than the synthetic antioxidant butylhydroxytoluene, commonly used as a food additive [18].

The utilisation of flowers for the human diet is associated with higher demands on their quality, sensory and nutritional properties [1] (Osimitz, Franzosa, Maciver, & Maibach, 2008). The larger size of blossoms with a simple collection is currently preferred if they can be economically effectively used [12]. The popular ornamental EFs of our gardens, specific in their size, colour, or aroma, could be convenient for the above properties and increase edible flowers' consumption. However, the sensory properties of selected cultivars of ornamental EFs are not sufficiently described in the literature. Additionally, only a small amount of EFs have been studied, and further research is needed to use them effectively. All flower samples in this research were non-toxic, allowing them to be included in human nutrition; nevertheless, it should be noted that the daily limit of intake for some of them is ye<sup>t</sup> unknown [14].

The study aimed to investigate the nutritional composition, total phenolics and flavonoids contents, antioxidant capacity and mineral elements (P, K, Ca, Mg, Na, Fe, Mn, Cu, Zn, and Mo) of selected EFs. This work was supplemented and extended by studying the relationship between bioactive compounds and antioxidant capacity to assess their potential benefits to human metabolism. Furthermore, the sensory properties (appearance, fragrance, consistency, acid, bitter, astringent, sweet, spicy, overall taste, juiciness, and overall evaluation) were evaluated and described.

#### **2. Materials and Methods**

#### *2.1. Plant Material*

During the period of 2018–2019, the examined plants were grown in an unheated greenhouse on the plots of experimental orchards belonging to the Mendel University in Brno. These grounds are situated in the south-eastern part of Czechia at an elevation of 170 m above the sea level in Lednice town. Besides, the average yearly temperature and precipitation are 9.2 ◦C and 516 mm, respectively. The soil type was classified as black soil; the value of pH/KCl is 6.8. The agrochemical attributes of the used soil are shown in Table 1 [36].

> **1**


**Table 1.** Agrochemical characteristics of the soil.

Note: 1 All values of mineral content are expressed in mg/kg.

The criteria for selecting suitable ornamental flowers for our experiment include well-known ornamental edible species with larger blossoms that can be easily collected. The six selected flowers varied in species and colour, namely a pink cultivar of *Begonia* × *tuberhybrida*, a red cultivar of *Tropaeolum majus*, an orange cultivar of *Calendula officinalis*, a light yellow to a cream colour cultivar of *Rosa*, *Hemerocallis* cultivar with two-coloured petals (yellow and red), and orange cultivar of *Tagetes patula*. Some of their characteristics as shown in Table 2.

**Table 2.** Species and cultivars of edible flowers used in the present experiment.


#### *2.2. Preparation of Samples*

Flowers were collected in full ripeness from five randomly chosen plants of each species (cultivar). The degree of full readiness of flowers was determined from the blossom size, opening and colouring [37]. Five flowers from each cultivar were mixed and used for analyses.

The flowers of the unique cultivar were processed promptly after harvest (within 24 h at the latest). The reaped flowers were ground in a disc type mill SJ 500 (MEZOS, Hradec Králové, Czechia). Finally, the average sample was obtained by dividing pureed samples into quarters. Each parameter was measured in five replications. The obtained data were expressed as the average of a two-year experiment (2018–2019).

#### *2.3. Extraction of Samples*

The extraction of samples was performed according to the method developed by Kim et al. [38] and modified by Barros et al. [39]. The fresh flowers (10 g) were homogenised for 10 s in methanol; the plant and solvent ratio was 1:10 ( *w*/*v*). The subsequent slurries were left in a water bath at a constant temperature of +25 ◦C for 24 h. The exact amount of methanol (100 mL) was used twice to extract residues. Both portions of methanolic extracts were combined, and the final solution was concentrated through evaporation at +40 ◦C to dryness (rotary evaporator R-215, Buchi Ltd., Oldham, UK). The samples were redissolved in methanol at a 100 g/mL concentration and kept at +4 ◦C for further utilisation.

#### *2.4. Total Phenolic Content Assay*

Total phenolic content (TPC) was measured by the method presented by Kim et al. with some modifications [38]. The sample (500 μL of extract) was brought quantitatively into a volumetric flask and diluted with distilled water to a volume of 50 mL. Then Folin-Ciocalteu's reagen<sup>t</sup> (2.5 mL) and 20% solution of sodium carbonate (7.5 mL) were added to the sample. The resulting absorbance was measured at 765 nm against a blank on UV/VIS spectrophotometer LIBRA S6 (Biochrom Ltd., Cambridge, UK). TPC in different methanolic extracts was calculated and reported as g of gallic acid equivalents (GAE) per kg of fresh mass (FM).

#### *2.5. Antioxidant Capacity by the DPPH Test Assay*

Total antioxidant capacity (TAC) assay was carried out according to Brand-Williams et al. [40,41] with some modifications to analyse the antioxidant capacity of given samples [40,41]. The determination of free radical scavenging activity of methanolic flower extracts on 2,2-diphenyl-1-picrylhydrazyl (DPPH) free radical was estimated. The preparation of the stock methanol solution of DPPH and then the working solution was performed according to the procedure described by Rop et al. [14] using a spectrophotometer (LIBRA S6). In this method, 150 μL of flower extract was allowed to react with 2.85 μL of the DPPH methanolic solution. After 1 h in the dark, the absorbance was measured at 515 nm, then the values were converted using a calibration curve and expressed as g of ascorbic acid equivalents (AAE) per kg of FM [42].

#### *2.6. Ferric Reducing Antioxidant Power Assay*

The total antioxidant potential was estimated spectrophotometrically using the ferric reducing antioxidant power (FRAP) assay determined by Benzie and Strain with a slight modification [43]. The FRAP reagen<sup>t</sup> was prepared from sodium acetate buffer (300 mM/L, pH 3.6), 10 mM/L TPTZ solution in 40 mM/L HCl and 20 mM/L FeCl3 solution in proportions of 10:1:1. The sample (50 μL) was added into a testing tube with 4 mL of FRAP reagent, and its absorbance was measured at 593 nm after 10 min of incubation. The standard curve was prepared using different gallic acid concentrations; the results were expressed as g GAE/kg of FM.

#### *2.7. Total Flavonoid Content Assay*

Total flavonoid content (TFC) was determined with the aluminium chloride colourimetric assay described by Singleton et al. [44]. Into a microcentrifuge tube, 0.3 mL of the flower extract, 3.4 mL of 30% ethanol, 0.1 mL of sodium nitrite (c = 0.5 mol/L) and 0.15 mL of aluminium chloride hexahydrate (c = 0.3 mol/L) were put and mixed. After 5 min of incubation, 1 mL of sodium hydroxide (c = 1 mol/L) was added to the mixture. The absorbance of samples was measured against the blank at 506 nm using the LIBRA S6 spectrophotometer. TFCs were calculated from the calibration curve using rutin as a standard and were expressed as g of rutin equivalents (RE) per kg of FM.

#### *2.8. Dry Matter and Mineral Content Assay*

The dry matter (DM) and the mineral content were measured using modified methods described by Higson and Novotny [45,46]. The plant samples were dried in a laboratory oven Venticell 111 (BMT, Brno, Czech Republic), at 105 ± 2 ◦C to a constant weight. The dried samples' weights were measured and expressed as a percentage of weight concentration ( *w*/*w*).

The next step was homogenisation of dried flowers to a particle size of up to 1 mm using a SJ500 laboratory grinder (MEZOS, Hradec Kralove, Czech Republic). About 1 g of DM of the homogenised sample was mineralised with concentrated sulphuric acid and 30% hydrogen peroxide in digestion tubes using a Bloc-digest M 24 apparatus (JP Selecta, Abrera, Spain). The digested samples were quantitatively added into a volumetric flask and then diluted to a final volume of 250 mL with double-distilled water.

The flower mineralizate was measured using an atomic absorption spectrometer PHILIPS PU 9200× with flame atomisation (Philips, Eindhoven, The Netherlands). A spectrophotometer Libra S6 was used to analyse the amount of phosphorus in the mineralizate

quantitatively. The sample was prepared in a 100 mL volumetric flask, where 10 mL of the mineralizate and the same volume of ammonium-vanadomolybdate reagen<sup>t</sup> were mixed. Lastly, samples were diluted up to a total volume of 100 mL with redistilled water and measured at 410 nm wavelength. For preparing the stock standard solution, potassium dihydrogen phosphate was used. The average contents of mineral elements were expressed as mg/kg of FM.

The content of total nitrogen was established according to the Kjeldahl method (ISO 1871:2009) using the analyser unit Kjeltec™ 2300 (Foss, Hillerod, Denmark). This apparatus provides automatic distillation and approves colourimetric titration. The crude protein in g/kg of FM was estimated by multiplying the determined nitrogen content by the standard default conversion factor of 6.25 [47].
