Ap

SD

a

ARR2A

KAASNH

1

it2ictd2

anaT(i

h0

Scientia Horticulturae 199 (2016) 170–177

Contents lists available at ScienceDirect

Scientia Horticulturae

journa l h om epa ge: www.elsev ier .com/ locate /sc ihor t i

ntioxidant power, anthocyanin content and organolepticerformance of edible flowers

tefano Benvenuti ∗, Elisa Bortolotti, Rita Magginiepartment of Agriculture, Food and Environment, Via Del Borghetto, 80, 56124 Pisa, Italy

r t i c l e i n f o

rticle history:eceived 21 September 2015eceived in revised form4 December 2015ccepted 26 December 2015

eywords:ntioxidantnthocyaninensory analysisew food

a b s t r a c t

The growing need for nutraceutical new foods has generated interest in edible flowers. This flower traitinspired us to conduct experiments aimed at evaluating both the antioxidant activity and anthocyanincontent in twelve species commonly used as ornamental plants. The antioxidant power of the edibleflowers was very high compared to common vegetables and/or fruits. Except for the low values of Boragoofficinalis (only 0.5 mmol FeSO4 100 g−1 fresh weight; FW), the antioxidant power in the edible flowersranged from 3.6 for Calendula officinalis to 70.4 for Tagetes erecta. Part of this high antioxidant activity isoften due to their high anthocyanin content at least in the case of the more pigmented flowers (red orblue). For example in the red varieties of Viola × wittrockiana, Dianthus × barbatus, Pelargonium peltatumthe high anthocyanin content (12.4, 13.3, 12.5 mg cyn-3-glu eq. 100 g−1 FW, respectively) was associatedto a high antioxidant activity. Indeed the best nutraceutical performances (antioxidant and/or antho-

ealth cyanin values) were shown by more pigmented flowers. A panel test was also carried out in order toevaluate the different degree of the flower’s palatability. This taste evaluation showed a high biodiversityof sensory profiles showing the greatest appreciation for Trapaeolum majus, Ageratum houstonianum andViola × wittrockiana. Finally, the overlap between nutraceuticals and organoleptic aspects highlightedpromising species for a potential market targeting new foods aimed at satisfying both taste and health.

© 2015 Elsevier B.V. All rights reserved.

. Introduction

The growing interest in nutraceuticals and functional foods hasncreased research into new foods that are beneficial to health. Thushe studies on fruits (Amagase et al., 2009), herbs (Wojdyło et al.,007) and seeds (Jayaprakasha et al., 2001) characterized by antiox-

dant, free radical scavenging and anti-aging activities assume arucial importance, since these properties are strongly linked tohe prevention and care of chronic illnesses such as cardiovasculariseases (Vivekananthan et al., 2003) and cancer (Greenlee et al.,012).

Although flowers were already used as food in ancient Greecend Rome (Melillo, 1994), they have only recently sparked offutraceutical research (Mlcek and Rop, 2011), focusing on newgronomic and economic horizons (Kelley and Biernbaum, 2000).heir rich pigmentation, which evolved to attract pollinators

Grotewold, 2006), suggests a high antioxidant activity that is ofnterest for human nutrition.

∗ Corresponding author. Fax: +39 0502216087.E-mail address: stefano.benvenuti@unipi.it (S. Benvenuti).

ttp://dx.doi.org/10.1016/j.scienta.2015.12.052304-4238/© 2015 Elsevier B.V. All rights reserved.

Anthocyanin pigments are primarily involved in this color-mediated attraction strategy and consequently their antioxidantactivity (Stintzing and Carle, 2004) makes the flowers an importantresource that could be agronomically and nutritionally enhanced.Indeed these pigmented flavonoids are considered a very impor-tant category of phytochemicals in plant foods due to their strongantioxidant activity and other beneficial physicochemical and bio-logical properties (De Pascual-Teresa and Sanchez-Ballesta, 2008).

Highly pigmented fruits, particularly small berries such as blue-berry, blackberry, cherry, raspberry and strawberry fruits, havebeen studied greatly due to their anthocyanin content and theirconsequent strong antioxidant activity. The interest in these phy-tochemicals has grown significantly in recent years due to theevidence that they play a crucial role in counteracting the oxidativestress related to chronic diseases (Li et al., 2012).

They are water-soluble compounds that impart color in plants(leaves, stems, roots, flowers and fruits) to appear red, purple orblue according to the pH and their structural features (Fossenand Andersen, 2003). Despite the fact that the main gastronomic

use of flowers stems from their attractive color (Kelley et al.,2001a, 2002), there is growing evidence of their role as anti-free radical functional-foods, as is well demonstrated in severalornamental species (Barros et al., 2010; Kaisoon et al., 2011;

orticu

Naa2(

inasn(

eknesflt

to

as

2

2

i(i8bwwotasag

1tttitw

2

cst

(ecw

S. Benvenuti et al. / Scientia H

avarro-González et al., 2014; Shi et al., 2009). As a source ofntioxidants (Chanwitheesuk et al., 2005), edible flowers havelso been shown to be effective as antitumor (Ukiya et al.,002), anti-inflammatory (Ukiya et al., 2006) and antimutagenicWongwattanasathien et al., 2010) biological agents.

Although the beneficial effects of flowers as a new promis-ng source of mineral elements in human nutrition should not beeglected (Rop et al., 2012), care needs to be taken regarding thenti-nutritional substances that are sometimes produced by somepecies (Sotelo et al., 2007). In any case, there is an increasingumber of ornamental (Mlcek and Rop, 2011) and wild speciesKucekova et al., 2013) grown as edible flowers.

Unfortunately, despite their agronomic potential, the idea ofating flowers is still viewed with suspicion. Indeed it involves aind of neophobia (the reluctance to try novel foods) since often aew food generates an innate distrust (Pliner and Hobden, 1992)specially in children (Dovey et al., 2008). Consequently it is neces-ary, first of all, to develop nutrition education aimed at proposingowers as a common food. It is also important to verify consumerastes when selecting flowers for human nutrition.

Although there are some encouraging results on the nutraceu-icals of edible flowers, there is little information on theirrganoleptic appreciation by consumers.

The aim of this work is twofold: (i) to analyze the content ofntioxidants and anthocyanins in some well-known ornamentalpecies, and (ii) test their organoleptic appreciation by free-tasters.

. Material and methods

.1. Plant material

Twelve species of cultivated edible flowers (Table 1) were stud-ed. The fresh flowers (Fig. 1) were collected during autumn 2011October) and spring 2012 (April) from a greenhouse cultivationn Torre del Lago (LU) in north-west Tuscany, near the sea (43◦

5′N, 10◦ 27′E), in an area called Versilia, where flowers haveeen grown for ornamental purposes for many decades. The plantsere kindly provided by a floriculture company (Carmazzi Farm),hich for several years has specialized in the cultivation and sale

f edible flowers grown with organic agricultural systems. In brief,his cultivation was carried out in unheated greenhouses duringutumn and spring (mean temperature about 15–25 ◦C) in sandyoil using organic fertilizers and without pesticides. In addition, thegronomic management was conducted without pesticides and/orrowth regulating substances.

The fully open flowers were collected between 08.00 and0.00 AM and placed in special plastic containers (the same ashose used for the packets on sale). Absorbent paper was placed athe bottom of these containers to prevent any lymph leakage dueo guttation, thus ensuring optimal conservation. The packs weremmediately placed in refrigerator bags and stored at −80 ◦C onhe same day (within 12.00 A.M.). The material was then analyzedithin 3–4 weeks from collection.

.2. Laboratory analysis

Both the antioxidant activity and the total content of antho-yanins were expressed on a fresh weight (FW) basis. Floweramples (1 g) were extracted with 10 mL methanol 80% (v/v), con-aining 1% of HCl, for 12 h at 4 ◦C.

Antioxidant activity was determined on the extracts by the FRAP

ferric ion reducing antioxidant power) assay following Pellegrinit al. (2003). A calibration curve was prepared with increasing con-entrations of FeSO4 (reagent grade, Sigma–Aldrich), and resultsere expressed as mmol FeSO4 100 g−1 FW.

lturae 199 (2016) 170–177 171

Total content of anthocyanins was determined spectrophoto-metrically (UV-1204 Shimadzu, Tokyo, Japan). by measuring theabsorbance of the extracts at 535 nm (Hrazdina et al., 1982). Datawere expressed as mg cyn-3-glu eq. 100 g−1 FW.

For some of the species under examination (Viola × wittrockiana,Petunia × hybrida, Antirrhinum majus and Dianthus × barbatus),three or four cultivars were available which differed only by thecolor of the flower. Therefore, the relationship between the colorand the antioxidant activity or the concentration of anthocyaninscould also be investigated.

2.3. Sensory analysis

The sensory panel was carried out in April 2012. Eighty-sevenfree-tasters (37 males and 50 females, mean age 38 years) wererecruited by adverts among the university community (students,teachers, other staff, etc.) from the Department of Agriculture, Foodand Environment of Pisa University.

In order to evaluate only the real sensory profile of the variousflowers, it was decided to get the tasters to examine the flowerswithout any condiments, bread, crackers, etc. After a careful eval-uation of the perceived flavors, the tasters were asked to fill out aquestionnaire aimed at determining the performances of the edibleflowers. These experiments were based on previous experiences oftaste evaluation performed on vegetables (Zhao et al., 2007) and/orfruits (Tobin et al., 2013).

Five different organoleptic characteristics (spiciness, sweetness,softness, scent, bitterness) were included in the evaluation schemeand were expressed in a scale of 1–100. The data enabled the sen-sory profile to be highlighted with spider plots (Johansson et al.,1999). A synthetic evaluation (scale 1–10) for each flower was alsorequired in order to establish the effective degree of appreciationof each species. Finally, tasters were also asked to determine whichknown food each of the flowers resembled.

2.4. Statistical analyses

The experiments were replicated three times in each experi-mental period (autumn and spring).

Analysis of variance (ANOVA) in a completely randomizeddesign and the Student–Newman–Keuls test were used to com-pare any significant differences between samples. The confidencelimits used were based on 95% (P < 0.05). The lack of significancebetween the data of the laboratory analyses in the autumn andspring enabled them to be grouped into a single media. For thesynthetic taste evaluation (scale 1–10), values were expressed asmeans ± standard deviations. For each statistical analysis, commer-cial software (CoHort software, Minneapolis, MN) was used.

3. Results

3.1. Nutraceutical analysis

Table 2 shows the antioxidant activity and the anthocyanin con-tent of the various edible flowers. The antioxidant power of theedible flowers varied within a very wide range, encompassing twoorders of magnitude. The strongest antioxidant activity was dis-played by Tagetes erecta, which reached 70.4 mol FeSO4 100 g−1 FW.Similarly, Fuchsia hybrida showed a very high value (although sig-nificantly lower), which reached almost 50 mmol FeSO4 100 g−1

FW. Other rather high values were also shown by the red-floweredcultivars of D. barbatus, V. wittrockiana and Pelargonium peltatum,

with antioxidant powers of 38.6, 36.5 and 34.7 mmol FeSO4 100 g−1

FW, respectively.Values of the antioxidant power in the range 20–30 mmol FeSO4

100 g−1 FW were found in the pink cultivar of D. barbatus (29.1),

172 S. Benvenuti et al. / Scientia Horticulturae 199 (2016) 170–177

Table 1Botanical and agronomic information on the tested edible flowers.

Species Botanic family Biological cycle Origin Colors Cultivar

Ageratum houstonianum Asteraceae Annual Central America Blue Tycoon Blue

Antirrhinummajus

Scrophulariaceae Perennial Europe, CentralAmerica, North Africa

Red Montego RedRose Montego roseWhite Montego white

Begonia semperflorens Begoniaceae Perennial South and Central America, Africa, South Asia Red Eureka

Borago officinalis Boraginaceae Annual Mediterranean environment Blue Wild germplasm

Calendula officinalis Asteraceae Perennial South Europe Orange Alice Orange

Dianthus ×barbatus

Caryophyllaceae Perennial SouthEurope,Asia

Red Diabunda RedRose Diabunda RoseWhite Dulce White

Fuchsia hybrida Onagraceae Perennial South America Red Coral

Pelargonium peltatum Geraniaceae Perennial Southern Africa Red Tornado

Petunia ×hybrida

Solanaceae Annual SouthAmerica

Red Duvet RedRose Duvet PinkWhite Duvet White

Tagetes erecta Asteraceae Perennial Central America Orange Moonstruck

Tropaeolum majus Tropaeolaceae Perennial South America Orange African QueenViola × wittrockiana Violaceae Annual Europe,

WesternAsia

Red Delta Pure RedBlue Karma True BlueYellow Mammoth Prima YellorinaWhite Mariposa White

Fig. 1. Morphology of the tested edible flowers: (1) Ageratum houstonianum, (2) Antirrhinum majus, (3) Begonia semperflorens, (4) Borago officinalis, (5) Calendula officinalis,(6) Dianthus × barbatus, (7) Fuchsia hybrid, (8) Pelargonium peltatum, (9) Petunia × hybrid, (10) Tagetes erecta, (11) Tropaeolum majus, (12) Viola × wittrockiana.

S. Benvenuti et al. / Scientia Horticulturae 199 (2016) 170–177 173

Table 2Antioxidant activity (mmol FeSO4 100 g−1 FW) and anthocyanin content (mg cyn-3-glu eq. 100 g−1 FW) of the edible flowers. The ±standard deviation of the meansare shown. Means followed by different letters in the same row are not statisticallydifferent for p < 0.05.

Species Flower color Antioxidantactivity(mmol FeSO4

100 g−1 FW)

Anthocyanincontent (mg cyn-3-glu eq. 100 g−1

FW)

Ageratum houstonianum Blue 27.85 ± 3.3 d 2.99 ± 0.2 d

Antirrhinum majus Red 21.18 ± 2.6 d 7.37 ± 0.5 bRose 9.85 ± 1.1 e 9.73 ± 0.5 bWhite 4.71 ± 0.6 f 0.70 ± 0.1 f

Begonia semperflorens Red 13.24 ± 1.7 e 5.09 ± 0.4 c

Borago officinalis Blue 0.55 ± 0.1 g 1.43 ± 0.1 e

Calendula officinalis Orange 3.68 ± 0.3 f 0.47 ± 0.1 f

Dianthus × barbatus Red 38.67 ± 3.0 c 13.35 ± 1.2 aRose 29.12 ± 2.3 d 10.61 ± 1.0 aWhite 4.36 ± 0.8 f 0.73 ± 0.1 f

Fuchsia hybrida Red 47.52 ± 3.2 b 7.58 ± 0.6 b

Pelargonium peltatum Red 34.78 ± 2.9 c 12.52 ± 1.1 a

Petunia × hybrida Red 10.22 ± 0.8 e 14.44 ± 1.2 aRose 9.45 ± 0.6 e 12.85 ± 1.1 aWhite 5.40 ± 1.3 f 2.5 ± 1.1 d

Tagetes erecta Orange 70.42 ± 4.1 a 0.75 ± 0.1 f

Tropaeolum majus Orange 10.05 ± 0.8 e 8.27 ± 0.7 b

Viola × wittrockiana Red 36.55 ± 3.0 c 12.4 ± 1.1 aBlue 29.12 ± 2.1 d 13.6 ± 1.2 a

iarnThwadaa

s(s3D1c1t83halbi

3

f

Table 3Synthetic overall evaluation (scale 1–10) and prevailing reference flavor of a knownfood. The ±standard deviation of the means are shown.

Species Tastingevaluation (1–10scale)

Flavor similarityto known food

Ageratum houstonianum 7.8 ± 0.4 CarrotAntirrhinum majus 6.2 ± 0.3 ChicoryBegonia semperflorens 7.1 ± 0.3 LemonBorago officinalis 7.0 ± 0.3 CucumberCalendula officinalis 5.5 ± 0.4 SaffronDianthus × barbatus 6.4 ± 0.2 ClovesFuchsia hybrida 5.2 ± 0.3 UnknownPelargonium peltatum 6.1 ± 0.3 GrapefruitPetunia × hybrida 6.3 ± 0.3 UnknownTagetes erecta 6.0 ± 0.2 PomegranateTropaeolum majus 8.2 ± 0.4 Radish

Yellow 3.31 ± 0.3 g 2.93 ± 0.2 dWhite 0.82 ± 0.1 g 0.35 ± 0.1 f

n blue flowered V. wittrockiana (29.1) and Ageratum houstoni-num (27.8), and in the red variety of A. majus (21.2). Fairly loweresults, in descending order, were obtained for red-flowered Bego-ia semperflorens (13.2) and P. hybrida (10.2), for orange-floweredrapaeolum majus (10.0), for pink-flowered A. majus (9.8) and P.ybrida (9.4). A weaker antioxidant power was displayed by threehite cultivars (P. hybrida, 5.4; A. majus, 4.7; D. barbatus, 4.4),

long with Calendula officinalis (orange, 3.7). Finally, the antioxi-ant activity values of yellow (3.3) and white (0.8) V.wittrockiana,nd especially Borago officinalis (blue, 0.5) were the lowest onesmong the samples under examination.

Similarly to the antioxidant power, the anthocyanin amounthowed great diversity depending on the species consideredTable 2). Nevertheless, one third of the flowers had statisticallyimilar high concentrations of anthocyanins, exceeding 10 mg cyn--glu eq. 100 g−1 FW. These included the red and pink cultivars of. barbatus (13.3 and 10.6, respectively) and P. hybrida (14.4 and2.8, respectively), along with P. peltatum (red, 12.5) and the blueultivar of V. wittrockiana (13.6). Concentrations ranging from 5 to0 mg cyn-3-glu eq. 100 g−1 FW were found in the red and pink cul-ivars of A. majus (7.4 and 9.7, respectively) and in T. majus (orange,.3), followed by B. semperflorens (red, 5.1). The range 1–5 mg cyn--glu eq. 100 g−1 FW included the similar concentrations of blue A.oustonianum, yellow V. wittrockiana and white P. hybrida (3.0), 2.9nd 2.5, respectively and that of blue B. officinalis (1.4). Extremelyow values, below 1 mg cyn-3-glu eq. 100 g−1 FW, were displayedy three white varieties (D. barbatus, 0,7; A. majus, 0,7; V. vittrock-

ana, 0.3) and two orange cultivars (T. erecta, 0.7; C. officinalis, 0.5).

.2. Organoleptic performances

Fig. 2 shows the sensory profiles of the various edible flowersollowing the panel test. The same graphic processing of a similar

Viola × wittrockiana 7.3 ± 0.3 Unknown

investigation was adopted (Johansson et al., 1999). First of all itshould be noted that the color of flowers did not determine anydifferent perception (data not shown) and consequently the datawere reported only in relation to the studied species.

In terms of spiciness, T. majus showed the highest values fol-lowed by A. houstorianum, C. officinalis, B. semperflorens and P.peltatum. In contrast, low values were reached by B. officinalis, P.hybrida and V. vittrockiana. As regards the flowers’ sweetness, thespecies with the best performances were T. majus and B. offici-nalis and, to a lesser extent, also V. vittrockiana. All other species,except C. officinalis, were not perceived as sweet in the taste test.Flower softness was judged as excellent for T. majus, V. vittrock-iana and P. hybrida with almost maximum range values. Satisfactoryorganoleptic evaluations of softness were also detected in A. majusand P. peltatum. On the other hand, the flowers of A. houstorianum,F. hybrida, D. barbatus, T. erecta and C. officinalis were described astough. In terms of the flowers’ scent, V. vittrockiana, T. majus and P.hybrida showed an excellent performance since they reached theapproximate maximum values. Suboptimal, but however satisfac-tory results, were found in terms of the scent of A. houstorianum,D. barbatus and B. semperflorens. In contrast A. majus, T. erecta, P.peltatum B. officinalis, F. hybrida and C. officinalis were reported asnot being very fragrant.

The majority of the species such as B. semperflorens (above all),A. houstonianum, D. barbatus, P. peltatum, F. hybrida, C. officinalisand T. erecta were reported as being notably bitter. Intermediatevalues of bitterness were reported for P. hybrida, A. majus and V.vittrockiana. B. officinalis and T. majus flowers were considered tobe unpleasantly bitter.

A comprehensive assessment (0–10 scale) was included in thequestionnaire regarding the sensory attractiveness of the differentflowers.

Table 3 shows these assessments followed by the prevailing ref-erence flavor perceived as similar to a known food. T. majus wasregarded the most attractive, since overall it scored 8.3, with a tastethat was judged to be similar to radish. This was the highest valueamong the twelve species tested,and was followed by those of A.haustorianum and V. wittrockiana (7.3 and 7.8, respectively). Theflowers of A. haustorianum were associated with the taste of car-rot, but V. wittrockiana did not remind tasters of any known food.The flowers of B. semperflorens and B. officinalis were also judgedto have a good taste since they achieved ratings of 7.1 and 7.2,respectively. For B. semperflorens, the flower’s flavor was associ-ated with lemon, and B. officinalis with cucumber. D. barbatus, P.hybrida, A. majus, P. peltatum and T. erecta were scored 6.4, 6.3, 62,

6.1 and 6.0, respectively. For P. peltatum, the flavor was defined asunknown. The others (in order of citation respectively) were found

174 S. Benvenuti et al. / Scientia Horticulturae 199 (2016) 170–177

F the tww

tg

s

t

4

4

pmFba

ig. 2. Sensory profile (spicy, sweet, soft, scent, bitter expressed as percentage) ofere selected.

o be similar, or at least comparable, to the following: cloves, grass,rapefruit, and pomegranate.

Despite C. officinalis was associated with the delicate taste ofaffron, it only scored 5.5.

Finally, the flowers of F. hybrida were considered to be ratherough and with no known flavor, and thus only scored 5.0.

. Discussion

.1. Nutraceutical analysis

The antioxidant power of the edible flowers was very high com-ared with literature data on vegetables and fruits obtained by

eans of the FRAP assay. For example, a range of 0.02–2.44 mmol

eSO4 100 g−1 FW was reported for a screening of common vegeta-les (Llorach et al., 2008; Wold et al., 2006). Fruits have often higherntioxidant capacities compared with vegetables. For example, the

elve flower species. For this organoleptic test, the flower colors shown in Table 1

tissues (peel or pulp) of five different apple cultivars showed val-ues of the antioxidant power in the range from 1.6 to 21.0 mmolFeSO4 100 g−1 FW (Henrìquez et al., 2010). Guo et al. (2003) testedthe antioxidant power of 28 fruits commonly consumed in China,distinct for peel, pulp and seed tissues. With the exception of thepeel tissue of white pomegranate, which showed an antioxidantpower of 82 mmol FeSO4 100 g−1 FW, all the other fruit tissuesdisplayed values of this parameter similar or lower than thosethat we observed in our samples, ranging from 0.16 (watermelonpulp) to 55.5 (red rose grape seeds). Goji, a fruit originating fromAsia whose market has been expanding worldwide during the lastdecade, was recently found to have an average antioxidant power of1.9 mmol FeSO4 100 g−1 FW, comparable with those of traditional

fruits (Donno et al., 2015).

Similarly to goji, non-traditional food that is introduced on themarket is often recommended for its nutraceutical properties, suchas the antioxidant capacity. Hence, according to our results, edible

orticu

flssaofli

cfb

ttoDtcut

ctOcbliamanapmtWfleaptc2

co

oraift

osoustwcac

S. Benvenuti et al. / Scientia H

owers could be considered as a non-traditional food capable toatisfy this requirement, with an antioxidant capacity that is oftenimilar or higher than those of fruit tissues. For this reason, thegronomic perspective of edible flowers could be similar to thatf exotic fruits, like Pitahaya (Le Bellec et al., 2006). In addition,owers usually display very intense colors that could contribute to

ncrease their appeal toward the consumer (Kelley et al., 2001b).Table 2 shows that the flower’s color plays a role not only on the

ontent of antocyanins, but also on the antioxidant power. Exceptor B. officinalis, the intense colors of the flowers (especially red andlue) are good indicators for both parameters in the tested flowers.

Specifically, red color is generally associated to high values ofhe antioxidant power and white color to the lowest ones. Indeedhe higher values of the red-flowered cultivars, compared to thethers, were statistically significant in V. vittrockiana, A. majus and. barbatus. On the other hand, in P. hybrida both red and pink cul-

ivars displayed a much higher antioxidant power than the whiteultivars. The blue color of V. vittrockiana also showed higher val-es than the lighter colors (yellow and white). These results suggesthat the antioxidant activity is partly due to anthocyanins.

However, some species with a high antioxidant activity are notharacterized by a high anthocyanin content, and consequentlyhis antioxidant power is derived from other phytochemicals.ne example is T. erecta, which showed a very low anthocyaninontent, in spite of a strong antioxidant power. This imbalanceetween antioxidant power and antocyanin content indicates, at

east in this species, that the high antioxidant activity is primar-ly due to phytochemicals such as flavonoids, especially quercetinnd rutin (Kaisoon et al., 2011). Similarly, V. vittrockiana and T.ajus contain antioxidants having a different chemical nature from

nthocyanin, such as vitamin C (Proteggente et al., 2002), polyphe-ols (Paganga et al., 1999; Kaisoon et al., 2012), carotenoids (Raond Agarwal, 1999), chlorogenic acid (Shi et al., 2009). For exam-le the xanthophyll (lutein) content was found to be high in T.ajus flowers (Niizu and Rodriguez-Amaya, 2005), and capable

o confer a high antioxidant power to this species (Garzón androlstad, 2009). Moreover, the variety of V. vittrockiana with blue

owers had lower antioxidant power than the red-flowered vari-ty, although both were characterized by a similar concentration ofnthocyanins. This suggests the presence of additional antioxidanthytochemicals in the red cultivar, probably carotenoids. In fact inhis species, the cultivars with red flowers were found to be rich inarotenoids, as shown by spectroscopy studies (Gamsjaeger et al.,011).

However, in spite of these generally species-dependent phyto-hemicals, most of the nutraceutical features are also due to colorf the flower cultivars, which is strongly linked to pigment content.

This fully confirms experiments conducted in flowers of Althaeafficinalis, in which a higher antioxidant activity was found in theed flower varieties than in other cultivars with lighter colors. Thisctivity slightly decreases in the pink colored variety and then felln white flowers (Sadighara et al., 2012). The antocyanin contentollowed the same trend as a function of the color, decreasing inhe order red–pink–white.

The results reported in Table 2 also evidence the importancef flower color in terms of anthocyanin content within the samepecies. As expected, a higher antocyanin concentration in the redr pink cultivars than in the white ones was confirmed in the speciesnder examination: V. vittrockiana (where the blue variety had aimilar antocyanin content to the red one), P. hybrida, D. barba-us and A. majus. In any case, as with the antioxidant power, the

hite color was associated with the lowest values of anthocyanin

oncentration. In particular, in V. vittrockiana and D. barbatus thenthocyanin contents of the white flowers were well below 10%ompared with the related values found in the red ones.

lturae 199 (2016) 170–177 175

4.2. Organoleptic performances

The studied flowers showed a high taste biodiversity. Althoughthe prevailing gastronomic role of flowers is to make the chromaticdishes attractive, they are also able to confer a peculiar taste andimprove their palatability.

In our screening, C. officinalis and F. hybrida were the only specieswhose flowers were quite difficult to masticate. This seems themain reason for which the flowers were not greatly appreciated(at least as raw and undressed food, as in the sensory analysis).

In all the other species, the preference resulted generally fromthe overall combination of softness, taste and flavor, although someflowers, such as A. majus, P. hybrida, A. houstorianum and B. officinaliswere mainly characterized by only one parameter. On the otherhand, T. majus, which had the highest score in the sensory analysis(Table 3), showed a strong spiciness together with sweetness, soft-ness and scent. Its flavor was associated with radish, therefore thetaste of this vegetable could be surprisingly perceived in a soft andunusual version. Similarly, both V. vittrockiana and D. barbatus wereappreciated for the combination of two features (softness/scent orbitterness/scent, respectively).

The appreciation levels of B. semperflorens, P. peltatum and T.erecta, which were strongly dependent on personal taste, wereprobably due to the unusual taste defined as “acidic”, since theirflavors were associated with lemon, grapefruit and pomegranate,respectively. Thus, an additional variety of tastes and aromas pro-vided by edible flowers could increase the biodiversity of sensoryprofiles that are required to food (Martin et al., 2014; Lease et al.,2015).

The combination of flowers with common vegetables or otherfood could generate new and appreciated tastes (Chambers andKoppel, 2013). Within this frame, possible taste affinities or con-trasts with wine could also promote an interest on edible flowers assensory “elicitors”. For example, recent studies have been publishedconcerning the consumer perception on food-beverage pairings(Paulsen et al., 2015), or the taste interaction between vegetablesand wine (Koone et al., 2014). Anyway, the consumption of edibleflowers is strongly linked to food education, as taste preference isdeveloped early in childhood (Mennella, 2014).

Finally, it is important to underline that the flowers used inthis study had been obtained by means of organic cropping. Thisaspect could play a critical role in order to attract possible con-sumers toward edible flowers as new food (Kelley and Biernbaum,2000), since organic cropping can properly preserve and ensureboth nutraceutical properties and food safety.

5. Conclusions

It is difficult to consider both nutraceutical and organolepticquality of edible flowers in predicting which species could be themost promising ones in a hypothetical market dedicated to thisnew food.

Organoleptic aspects are subjective and therefore susceptibleto change depending on the tastes of the consumers (and on thegeographical areas of the world) and also on how the flowers arecooked and seasoned. Nevertheless, there is no doubt that A. hous-torianum, B. semperflorens, V. vittrockiana and T. majus are veryattractive, and satisfy at the same time health requirements due totheir high antioxidant levels. The flowers of T. majushave alreadybeen investigated for the postharvest physiology (Friedman et al.,2005) and for the thermal requirements for their storage (Kelley

et al., 2003).

The cultivars with the most intensely colored flowers (such asred and blue) appear to be the most suited to a gastronomic andnutraceutical evolution in terms of the discovery of new food and

1 orticu

nndatn

cof

A

p

R

A

B

C

C

D

D

D

F

F

G

G

G

G

G

G

H

H

J

J

K

K

K

K

76 S. Benvenuti et al. / Scientia H

ew dishes. Moreover they could have an interesting application asatural colorants, representing an alternative to the use of syntheticyes in foods (Giusti and Wrolstad, 2003). Their organic productionppears to be an additional agronomic opportunity that fully meetshe future needs of food aimed at improving the quality of humanutrition.

Further research will be required to improve other nutraceuti-al parameters (carotenoids, micronutrients, vitamins, etc.) and toffer them as an additional source of vegetable biodiversity for theuture scenario of growing health needs.

cknowledgment

We thank “Carmazzi Farm” for having kindly provided thelants, organically grown, that have been studied in this work.

eferences

magase, H., Sun, B., Borek, C., 2009. Lycium barbarum (goji) juice improves in vivoantioxidant biomarkers in serum of healthy adults. Nutr. Res. 29, 19–25.

arros, L., Carvalho, A.M., Ferreira, I.C., 2010. Leaves, flowers, immature fruits andleafy flowered stems of Malva sylvestris: a comparative study of thenutraceutical potential and composition. Food Chem. Toxicol. 48, 1466–1472.

hambers, E., Koppel, K., 2013. Associations of volatile compounds with sensoryaroma and flavor: the complex nature of flavor. Molecules 18, 4887–4905.

hanwitheesuk, A., Teerawutgulrag, A., Rakariyatham, N., 2005. Screening ofantioxidant activity and antioxidant compounds of some edible plants ofThailand. Food Chem. 92, 491–497.

e Pascual-Teresa, S., Sanchez-Ballesta, M.T., 2008. Anthocyanins: from plant tohealth. Phytochem. Rev. 7, 281–299.

onno, D., Beccaro, G.L., Mellano, M.G., Cerutti, A.K., Bounous, S., 2015. Goji berryfruit (Licium spp.): antioxidant compound fingerprint and bioactivityevaluation. J. Funct. Foods 18, 1070–1085.

ovey, T.M., Staples, P.A., Gibson, E.L., Halford, J.C., 2008. Food neophobia and‘picky/fussy’eating in children: a review. Appetite 50, 181–193.

ossen, T., Andersen, Ø., 2003. Anthocyanins from red onion, Allium cepa, withnovel aglycone. Phytochemistry 62, 1217–1220.

riedman, H., Vinokur, Y., Rot, I., Rodov, V., Goldman, G., Resnick, N., et al., 2005.Tropaeolum majus L. as edible flowers: growth and postharvest handling. Adv.Hortic. Sci., 3–8.

amsjaeger, S., Baranska, M., Schulz, H., Heiselmayer, P., Musso, M., 2011.Discrimination of carotenoid and flavonoid content in petals of pansy cultivars(Viola × wittrockiana) by FT-Raman spectroscopy. J. Raman Spectrosc. 42,1240–1247.

arzón, G.A., Wrolstad, R.E., 2009. Major anthocyanins and antioxidant activity ofNasturtium flowers (Tropaeolum majus). Food Chem. 114, 44–49.

iusti, M.M., Wrolstad, R.E., 2003. Acylated anthocyanins from edible sources andtheir applications in food systems. Biochem. Eng. J. 14, 217–225.

reenlee, H., Kwan, M.L., Kushi, L.H., Song, J., Castillo, A., Weltzien, E., Quesenberry,C.P., Caan, B.J., 2012. Antioxidant supplement use after breast cancer diagnosisand mortality in the Life After Cancer Epidemiology (LACE) cohor. Cancer 118,2048–2058.

rotewold, E., 2006. The genetics and biochemistry of floral pigments. Ann. Rev.Plant Biol. 57, 761–780.

uo, C., Yang, J., Wei, J., Li, Y., Xu, J., Jiang, Y., 2003. Antioxidant activities of peel,pulp and seed fractions of common fruits as determined by FRAP assay. Nutr.Res. 23, 1719–1726.

razdina, G., Marx, G.A., Hoch, H.C., 1982. Distribution of secondary plantmetabolites and their biosynthetic enzymes in Pea (Pisum sativum L.)leaves—anthocyanins and flavonol glycosides. Plant Physiol. 70, 745–748.

enrìquez, C., Almonacid, S., Chiffelle, I., Valenzuela, T., Araya, M., Cabezas, L.,Simpson, R., Speisky, H., 2010. Determination of antioxidant capacity, totalphenolic content and mineral composition of different fruit tissue of five applecultivars grown in Chile. Chil. J. Agric. Res. 70, 523–536.

ayaprakasha, G.K., Singh, R.P., Sakariah, K.K., 2001. Antioxidant activity of grapeseed (Vitis vinifera) extracts on peroxidation models in vitro. Food Chem. 73,285–290.

ohansson, L., Haglund, Å., Berglund, L., Lea, P., Risvik, E., 1999. Preference fortomatoes, affected by sensory attributes and information about growthconditions. Food Qual. Preference 10, 289–298.

aisoon, O., Siriamornpun, S., Weerapreeyakul, N., Meeso, N., 2011. Phenoliccompounds and antioxidant activities of edible flowers from Thailand. J. Funct.Foods 3, 88–99.

aisoon, O., Konczak, I., Siriamornpun, S., 2012. Potential health enhancingproperties of edible flowers from Thailand. Food Res. Int. 46, 563–571.

elley, K.M., Biernbaum, J.A., 2000. Organic nutrient management of greenhouseproduction of edible flowers in containers. Hortscience 35, 453.

elley, K.M., Behe, B.K., Biernbaum, J.A., Poff, K.L., 2001a. Consumer andprofessional chef perceptions of three edible-flower species. HortScience 36,162–166.

lturae 199 (2016) 170–177

Kelley, K.M., Behe, B.K., Biernbaum, J.A., Poff, K.L., 2001b. Consumer preference foredible-flower color, container size, and price. HortScience 36, 801–880.

Kelley, K.M., Behe, B.K., Biernbaum, J.A., Poff, K.L., 2002. Combinations of colors andspecies of containerized edible flowers: effect on consumer preferences.HortScience 37, 218–221.

Kelley, K.M., Cameron, A.C., Biernbaum, J.A., Poff, K.L., 2003. Effect of storagetemperature on the quality of edible flowers. Postharvest Biol. Technol. 27,341–344.

Koone, R., Harrington, R.J., Gozzi, M., McCarthy, M., 2014. The role of acidity,sweetness, tannin and consumer knowledge on wine and food matchperceptions. J. Wine Res. 25, 158–174.

Kucekova, Z., Mlcek, J., Humpolicek, P., Rop, O., 2013. Edible flowers—antioxidantactivity and impact on cell viability. Cent. Eur. J. Biol. 8, 1023–1031.

Le Bellec, F., Vaillant, F., Imbert, E., 2006. Pitahaya (Hylocereus spp.): a new fruitcrop, a market with a future. Fruits 61, 237–250.

Lease, H., Hendrie, G.A., Poelman, A.A., Delahunty, C., Cox, D.N., 2015. Asensory-diet database: a tool to characterise the sensory qualities of diets.Food Qual. Preference 49, 20–32.

Li, H., Deng, Z., Zhu, H., Hu, C., Liu, R., Young, J.C., Tsao, R., 2012. Highly pigmentedvegetables: anthocyanin compositions and their role in antioxidant activities.Food Res. Int. 4, 250–259.

Llorach, R., Martínez-Sánchez, A., Tomás-Barberán, F.A., Gil, M.I., Ferreres, F., 2008.Characterisation of polyphenols and antioxidant properties of five lettucevarieties and escarole. Food Chem. 108, 1028–1038.

Martin, C., Visalli, M., Lange, C., Schlich, P., Issanchou, S., 2014. Creation of a foodtaste database using an in-home taste profile method. Food Qual. Preference36, 70–80.

Melillo, L., 1994. Diuretic plants in the paintings of Pompeii. Am. J. Nephrol. 14,423–425.

Mennella, J.A., 2014. Ontogeny of taste preferences: basic biology and implicationsfor health. Am. J. Clin. Nutr. 99, 704S–711S.

Mlcek, J., Rop, O., 2011. Fresh edible flowers of ornamental plants—a new source ofnutraceutical foods. Trends Food Sci. Technol. 22, 561–569.

Navarro-González, I., González-Barrio, R., García-Valverde, V., Bautista-Ortín, A.B.,Periago, M.J., 2014. Nutritional composition and antioxidant capacity in edibleflowers: characterisation of phenolic compounds by HPLC-DAD-ESI/MSn. Int. J.Mol. Sci. 16, 805–822.

Niizu, P.Y., Rodriguez-Amaya, D.B., 2005. Flowers and leaves of Tropaeolum majus L.as rich sources of lutein. J. Food Sci. 70, S605–S609.

Paganga, G., Miller, N., Rice-Evans, C.A., 1999. The polyphenolic content of fruit andvegetables and their antioxidant activities. What does a serving constitute?Free Radic. Res. 30, 153–162.

Paulsen, M.T., Rognså, G.H., Hersleth, M., 2015. Consumer perception offood–beverage pairings: the influence of unity in variety and balance. Int. J.Gastr. Food Sci. 2, 83–92.

Proteggente, A.R., Pannala, A.S., Paganga, G., Buren, L.V., Wagner, E., Wiseman, S.,Van de Put, F., Dacombe, C., Rice-Evans, C.A., 2002. The antioxidant activity ofregularly consumed fruit and vegetables reflects their phenolic and vitamin Ccomposition. Free Radic. Res. 36, 217–233.

Pellegrini, N., Serafini, M., Colombi, B., Del Rio, D., Salvatore, S., Bianchi, M.,Brighenti, F., 2003. Total antioxidant capacity of plant foods: beverages and oilsconsumed in Italy assessed by three different in vitro assays. J. Nutr. 133,2812–2819.

Pliner, P., Hobden, K., 1992. Development of a scale to measure the trait of foodneophobia in humans. Appetite 19, 105–120.

Rao, A.V., Agarwal, S., 1999. Role of lycopene as antioxidant carotenoid in theprevention of chronic diseases: a review. Nutr. Res. 19, 305–323.

Rop, O., Mlcek, J., Jurikova, T., Neugebauerova, J., Vabkova, J., 2012. Edibleflowers—a new promising source of mineral elements in human nutrition.Molecules 17, 6672–6683.

Sadighara, P., Gharibi, S., Jafari, A.M., Khaniki, G.J., Salari, S., 2012. The antioxidantand flavonoids contents of Althaea officinalis L. flowers based on their color.Avicenna J. Phytomed. 2, 113–117.

Shi, J., Gong, J., Liu, J.E., Wu, X., Zhang, Y., 2009. Antioxidant capacity of extract fromedible flowers of Prunus mume in China and its active components. LWT-FoodSci. Technol. 42, 477–482.

Sotelo, A., López-García, S., Basurto-Pena, F., 2007. Content of nutrient andantinutrient in edible flowers of wild plants in Mexico. Plant Foods Hum. Nutr.62, 133–138.

Stintzing, F.C., Carle, R., 2004. Functional properties of anthocyanins and betalainsin plants, food, and in human nutrition. Trends Food Sci. Technol. 15, 19–38.

Tobin, R., Moane, S., Larkin, T., 2013. Sensory evaluation of organic andconventional fruits and vegetables available to Irish consumers. Int. J. Food Sci.Technol. 48, 157–162.

Ukiya, M., Akihisa, T., Tokuda, H., Suzuki, H., Mukainaka, T., Ichiishi, E., Yasukawa,K., Kasahara, Y., Nishino, H., 2002. Constituents of compositae plants: III.Anti-tumor promoting effects and cytotoxic activity against human cancer celllines of triterpene diols and triols from edible chrysanthemum flowers. CancerLett. 177, 7–12.

Ukiya, M., Akihisa, T., Yasukawa, K., Tokuda, H., Suzuki, T., Kimura, Y., 2006.Anti-inflammatory, anti-tumor-promoting, and cytotoxic activities of

constituents of marigold (Calendula officinalis) flowers. J. Nat. Prod. 69,1692–1696.

Vivekananthan, D.P., Penn, M.S., Sapp, S.K., Hsu, A., Topol, E.J., 2003. Use ofantioxidant vitamins for the prevention of cardiovascular disease:meta-analysis of randomized trials. Lancet 361, 2017–2023.

orticu

W

W

S. Benvenuti et al. / Scientia H

ojdyło, A., Oszmianski, J., Czemerys, R., 2007. Antioxidant activity and phenoliccompounds in 32 selected herbs. Food Chem. 105, 940–949.

old, A.B., Wicklund, T., Haffner, K., 2006. Antioxidant activity in commonlygrown and consumed vegetables: a screening survey. J. Appl. Bot. Food Qual.80, 111–115.

lturae 199 (2016) 170–177 177

Wongwattanasathien, O., Kangsadalampai, K., Tongyonk, L., 2010.

Antimutagenicity of some flowers grown in Thailand. Food Chem. Toxicol. 48,1045–1051.

Zhao, X., Chambers, E., Matta, Z., Loughin, T.M., Carey, E.E., 2007. Consumer sensoryanalysis of organically and conventionally grown vegetables. J. Food Sci. 72,S87–S91.