Antioxidant properties, phenolic composition and potentiometric sensor array evaluation of commercial and new blueberry (Vaccinium corymbosum) and bog blueberry (Vaccinium uliginosum)

Food Chemistry 188 (2015) 583–590

Contents lists available at ScienceDirect

Food Chemistry

journal homepage: www.elsevier .com/locate / foodchem

Antioxidant properties, phenolic composition and potentiometric sensorarray evaluation of commercial and new blueberry (Vacciniumcorymbosum) and bog blueberry (Vaccinium uliginosum) genotypes

http://dx.doi.org/10.1016/j.foodchem.2015.05.0310308-8146/� 2015 Elsevier Ltd. All rights reserved.

⇑ Corresponding author.E-mail address: [email protected] (P.R. Venskutonis).

Vilma Kraujalyte a, Petras Rimantas Venskutonis a,⇑, Audrius Pukalskas a, Laima Cesoniene b,Remigijus Daubaras b

a Department of Food Science and Technology, Kaunas University of Technology, Radvilenu pl. 19, Kaunas LT-50254, Lithuaniab Kaunas Botanical Garden of Vytautas Magnus University, Zilibero g. 6, Kaunas LT-46324, Lithuania

a r t i c l e i n f o a b s t r a c t

Article history:Received 8 January 2015Received in revised form 4 May 2015Accepted 7 May 2015Available online 8 May 2015

Keywords:Vaccinium corymbosumVaccinium uliginosumAntioxidant capacityPhenolic compoundsTotal anthocyanin contentElectronic tongueColour

Antioxidant properties of juices of newly bred and known blueberry (Vaccinium corymbosum) genotypesand wild bog blueberry (Vaccinium uliginosum) were evaluated by ABTS�+ scavenging capacity (RSC), ferricreducing antioxidant power (FRAP), oxygen radical absorbance capacity (ORAC), total phenolic content(TPC) and total anthocyanin content (TAC) assays. TPC varied in the range of 0.85–2.81 mg gallic acidequiv./mL, RSC, FRAP and ORAC values were 6.38–20.9, 3.07–17.8 and 4.21–45.68 lmol Trolox equiv./g,respectively. New blueberry genotypes and bog blueberry demonstrated stronger antioxidant propertiesand TAC than other studied genotypes. The content of quinic (203–3614 lg/mL), chlorogenic (20.0–346.8 lg/mL) acids and rutin (0.00–26.88 lg/mL) measured by UPLC/ESI-QTOF-MS varied dependingon the genotype. Juices were evaluated by electronic tongue; PCA score plot showed that the method dis-criminates different genotypes although some juice samples were located very closely and overlapping.Significant differences were observed between L⁄, a⁄, b⁄ colour parameters of some genotypes.

� 2015 Elsevier Ltd. All rights reserved.

1. Introduction

Berries of Vaccinium spp. have gained a remarkable worldwideinterest due to their excellent sensory properties and the presenceof healthy constituents (Scalzo, Stevenson, & Hedderley, 2013).Blueberries (Vaccinium corymbosum) and some closely related spe-cies are among the most popular berries. Continuous breeding pro-grams were focusing on higher commercial value genotypes withhigh productivity, large berry size, firmness, disease resistance,etc. Large fruits, waxy coating, light blue colour, firmness and a longshelf life are among the most important berry characteristics.However, due to an increasing demand of healthy foods, breedingprograms of new cultivars with higher concentrations of bioactivecompounds are continued in many horticultural units worldwide.For instance, antioxidants have become a desirable quality charac-teristic of berries due to their anti-inflammatory effects. Therefore,phenolics possessing radical scavenging capacity are among the pre-dictors for selecting new blueberry cultivars. It is well documentedthat blueberries are rich in phenolics, particularly hydroxycinnamic

and hydroxybenzoic acids and their derivatives such as chlorogenic,caffeic, gallic, p-coumaric, ferulic, ellagic, syringic, vanillic acids(Häkkinen & Törrönen, 2000; Mattila, Hellström, & Törrönen,2006; Rodriguez-Mateos, Cifuentes-Gomez, Tabatabaee, Lecras, &Spencer, 2012). Blueberries also accumulate flavonols (mainly quer-cetin derivatives), anthocyanins, flavan-3-ols, and proanthocyani-dins (Brownmiller, Howard, & Prior, 2009; Cho, Howard, Prior, &Clark, 2004; Häkkinen & Törrönen, 2000; Wang, Chen, Sciarappa,Wang, & Camp, 2008). Blueberry phenolics were reported as con-stituents, which are responsible for their high radical scavengingcapacity (Giovanelli & Buratti, 2009; Wang et al., 2008).

Sensory properties of berries, which traditionally are evaluatedby a panel of trained experts is also very important factor deter-mining consumer preferences. In addition, instrumental methodsand their combination with organoleptic assessment for the char-acterisation of flavour properties have been rapidly developing inthe past decades. For instance, identification of the key-odourcompounds currently is performed mainly by a combination ofgas chromatography and olfactometry. Although simultaneousapplication of these techniques provides valuable information theyare time-consuming and expensive. Therefore alternative rapidmethods in a reliable and cost-effective manner are of a great prac-tical interest, electronic tongues being good examples of such

584 V. Kraujalyte et al. / Food Chemistry 188 (2015) 583–590

techniques (Escuder-Gilaberta & Peris, 2010 and referencesherein). Electronic tongue is an analytical system applied for theanalysis of liquid multicomponent samples based on an array ofsensors coupled with chemometric processing tool and is consid-ered as an instrument artificially reproducing taste sensation. Itwas widely tested for liquid foodstuffs, including water (Siposet al., 2012), soft drinks (Diasa, Peresa, Barcelosa, Sá Moraisa, &Machado, 2011) and was proven as a promising, rapid andlow-cost method.

The formation of flavour and bioactive compounds in fruitsdepends on many factors such as genotype, environmental condi-tions, ripeness and storage (Castrejón, Eichholz, Rohn, Kroh, &Huyskens-Keil, 2008; Häkkinen & Törrönen, 2000; Scalzo, Politi,Pellegrini, Mezzetti, & Battino, 2005; Wang et al., 2008). On theother hand, recent studies indicated that the contents of antho-cyanins, chlorogenic acid and quercetin were highly dependentupon blueberry cultivar, while seasonal variations had a compara-tively minor effect (Brambilla, Lo Scalzo, Bertolo, & Torreggiani,2008; Scalzo et al., 2013).

Literature survey suggests that, considering many factorsinfluencing the composition of flavour and bioactive substancesas well as remarkable importance of blueberries in the diet, thestudies should be continued by focusing them on the comprehen-sive characterisation of different plant genotypes. The aim of thisstudy was to evaluate antioxidant properties and to measure theconcentrations of the main phytochemicals in several V. corymbo-sum genotypes and Vaccinium uliginosum, which was chosen forcomparison purposes as a wild species serving as a genetic mate-rial for breeding. Additionally, the juices were investigated byusing artificial taste sensor electronic tongue. The data obtainedin this study comprehensively characterize well-known and newgenotypes and may be useful for the selection of valuable cultivarsaccumulating higher contents of bioactive constituents with bene-ficial health effects both when used as fresh fruits and in processedfoods.

2. Materials and methods

2.1. Berries of V. uliginosum and V. corymbosum and juice preparation

Indigenous species of bog blueberry (V. uliginosum L.) and high-bush blueberry (V. corymbosum L.) genotypes (Table 1) were grown

Table 1Average berry mass and yield of blueberries.

Genotype Berry mass (g) Yield/bush (kg)

Bog blueberry 0.46 ± 0.01a 1.35 ± 0.10abcd

No. 16 0.55 ± 0.01ab 1.29 ± 0.36abcd

‘Danute’ (No. 17) 0.76 ± 0.02bc 1.39 ± 0.05abcd

‘Freda’ (No. 11) 0.93 ± 0.01c 1.32 ± 0.09abcd

‘Northland’ 1.72 ± 0.06e 1.37 ± 0.07abcd

‘Nui’ 3.33 ± 0.13k 0.52 ± 0.14a

‘Putte’ 1.26 ± 0.01d 1.91 ± 0.30bcd

‘Patriot’ 2.92 ± 0.27i 2.32 ± 0.52def

‘Weymouth’ 1.43 ± 0.03de 1.14 ± 0.35abcd

‘Northblue’ 1.73 ± 0.03e 1.20 ± 0.38abcd

‘Bluegold’ 2.67 ± 0.19hi 1.38 ± 0.56abcd

‘Aron’ 1.26 ± 0.03d 1.20 ± 0.24abcd

‘Bluehaven’ 2.34 ± 0.13fg 0.81 ± 0.18ab

‘Reka’ 2.49 ± 0.14gh 2.33 ± 0.35def

‘Puru’ 2.16 ± 0.06f 2.07 ± 0.21cd

‘Bluecrop’ 2.61 ± 0.07ghi 2.06 ± 1.04cd

‘Blueray’ 2.93 ± 0.12i 1.28 ± 0.56abcd

‘Bluejay’ 2.41 ± 0.04fgh 3.61 ± 0.42f

‘Toro’ 3.62 ± 0.13k 2.10 ± 0.22cd

‘Hardyblue’ 1.67 ± 0.03e 0.99 ± 0.12abc

Different superscript letter within the same column indicates significant differences(one way ANOVA and Duncan‘s test, p 6 0.05).

in the collection of Kaunas Botanical Garden of Vytautas MagnusUniversity (Lithuania). The genotypes ‘Danute’, ‘Freda’ and ‘No.16’ are newly developed in Kaunas Botanical Garden. It should benoted that half-highbush cultivars were distinguished for theirresistance to dangerous Botryosphaeria spp. fungal disease.Consequently, promising genotypes with good climatic adaptabil-ity and excellent agro-biological properties were selected. It is con-sidered that due to the similar growing conditions of all studiedaccessions, the observed differences should be dependent mainlyon the plant genotype.

Harvesting time was determined for each genotype, when berrycolour was fully blue and an average content of mature berries onthe bush was 30–50%. The berries were picked randomly fromdifferent parts of the plant, combined into one batch for each geno-type, packed in plastic bags, frozen and stored at �24 �C before use.The samples were defrosted at 24 �C during 2 h, crushed in ablender Braun Multimix (Munich, Germany) and centrifuged at10,000 rpm for 15 min. The supernatant of clear juice was kept at�24 �C in glass bottles until further analysis.

2.2. Chemicals

2,2-Diphenyl-1-picrylhydrazyl radical (DPPH�, 95%), gallic acid(GA), tetramethylchromane-2-carboxylic acid (Trolox 97%),anhydrous sodium carbonate, 2,20-azobis (2-amidinopropane)dihydrochloride (AAPH) were purchased from Sigma–AldrichChemie (Steinheim, Germany); 2,4,6-tripyridyl-s-triazine (TPTZ)and fluorescein (FL) from Fluka Chemicals (Steinheim, Germany);aluminium trichloride hydrate and sodium acetate from Reachim(Riga, Latvia); 2.0 M Folin–Ciocalteu’s phenol reagent, 2,20-azinobis(3-ethylbenzothiazoline-6-sulfonic acid) diammonium salt(ABTS), NaCl, KCl, Na2HPO4, and K2S2O8 were from Merck(Darmstadt, Germany); KH2PO4 from Jansen Chimica (Beerse,Belgium); methanol (99.8%) and acetic acid (98%) from Lachema(Brno, Czech Republic), quinic acid and rutin were fromChromadex (California, USA); chlorogenic acid (97%) from Roth(Karlsruhe, Germany).

2.3. Physicochemical characteristics of berries

The average weight of a berry was measured by weighing 50fruits on an analytical balance DJ-150E (Ishida, Japan). The contentof soluble substances was measured by refractometer PAL-3(Ataga, Japan) at 18 �C. Colour parameters L⁄, lightness (0 = black,100 = white), a⁄, (�a⁄ = greenness, +a⁄ = redness) and b⁄ �b⁄ =blueness, +b⁄ = yellowness) were evaluated by a MiniScan XE Plusspectrophotometer (Hunter Associates Laboratory, Reston,Virginia, USA). Additionally, hue degree [h� = arctangent (b⁄/a⁄)]and chroma (representing colour intensity) [C = (a⁄2 + b⁄2)1/2] val-ues were calculated.

2.4. Evaluation by electronic tongue (potentiometric sensor array)

Astree Electronic Tongue, consisting of 7 liquid cross selectivesensors and a reference electrode (Ag/AgCl), 48-positionauto-sampler, an electronic unit for data acquisition/processingand PC with a chemometric software (Alpha M.O.S. Co., Toulouse,France) was used. The instrument measures the potentiometricdifference between each of the 7 sensors and the reference elec-trode in the presence of dissolved juice compounds. The sensorsmeasure dissolved organic and inorganic compounds in liquidsincluding taste and aroma constituents. Any interaction at the sen-sor interface is detected and converted into electronic signals.Evaluation of juice samples was performed after achieving a steadypotential for all sensors. Between measurements the electronic

V. Kraujalyte et al. / Food Chemistry 188 (2015) 583–590 585

tongue was rinsed with distilled water. Ten replicate measure-ments of each sample were performed.

2.5. Determination of total phenolic content (TPC) and totalmonomeric anthocyanin content (TAC)

The TPC was measured with Folin–Ciocalteu reagent usingslightly modified procedure enabling to reduce the consumptionof reagents (Kraujalyte, Venskutonis, Pukalskas, Cesoniene, &Daubaras, 2013). Briefly, 10 lL of diluted juice (1:10 w/v) weremixed with 100 lL of 10-fold diluted Folin–Ciocalteu reagent,and after 4 min 100 lL of 7% Na2CO3 and 40 lL of distilled waterwere added. After mixing the microplate was placed in the reader,shaken for 30 s, incubated for 90 min and the absorbance read at725 nm. GA solutions (0–250 mg/L) were used for the calibrationcurve. The results a expressed in mg of GA equiv. per mL juice(mL GAE/mL).

The TAC in blueberry juice was determined by the pH differen-tial method (Lee, Rennaker, & Wrolstad, 2008). The absorbance wasmeasured at 520 and 700 nm, TAC were expressed ascyanidin-3-glucoside in mg/100 mL of juice.

2.6. ABTS�+ decolourisation assay

Radical scavenging capacity (RSC) of juices was measured bythe ABTS�+ scavenging assay (Re et al., 1999). Phosphate bufferedsaline (PBS) solution (8.18 g NaCl, 0.27 g KH2PO4, 1.42 g Na2HPO4

and 0.15 g KCl in 1 L of distilled water) pH was adjusted to 7.4 withNaOH. The ABTS�+ solution was prepared by mixing 50 mL ABTS(2 mM) with 200 lL K2S2O8 (70 mM); the mixture was kept inthe dark for 15–16 h before use. The working solution wasprepared by diluting with PBS to obtain the absorbance of0.700 ± 0.030 at 734 nm (Spectronic Genesys 8, ThermoSpectronic, Rochester, NY). Five mL of diluted (1:10 w/v) juice werepipetted into the microplate wells, 300 lL ABTS�+ working solutionadded; the plate was placed in a FLUOstar Omega reader (BMGLabtech, Offenburg, Germany), shaken for 30 s and the absorbanceread after 30 min at 734 nm. A series of Trolox solutions(0–1450 lM) were used for the calibration curve. The RSC values

Table 2Total phenolic content (TPC), total anthocyanin content (TAC), antioxidant capacity and so

Genotype TPC (mg GAE/mLjuice)

TAC (mg/100 mLjuice)

ABTS�+ (TEAC, lmoTrolox/g)

Bog blueberry 2.81 ± 0.162l 14.19 ± 0.20k 20.90 ± 0.67k

No. 16 1.76 ± 0.036k 2.48 ± 0.03f 14.64 ± 1.18j

‘Danute’ (No. 17) 1.50 ± 0.079ij 2.56 ± 0.05f 12.00 ± 0.05i

‘Freda’ (No. 11) 1.50 ± 0.042ij 1.72 ± 0.04c 11.55 ± 0.93hi

‘Northland’ 1.54 ± 0.075j 1.90 ± 0.04d 11.08 ± 1.41ghi

‘Nui’ 1.42 ± 0.028ij 2.11 ± 0.04e 10.63 ± 0.48fgh

‘Putte’ 1.37 ± 0.027hi 2.10 ± 0.05e 10.41 ± 1.32efgh

‘Patriot’ 1.18 ± 0.033fg 3.20 ± 0.05j 10.30 ± 0.81defgh

‘Weymouth’ 1.17 ± 0.072efg 2.07 ± 0.06e 9.96 ± 0.59defg

‘Northblue’ 1.25 ± 0.052gh 2.52 ± 0.05f 9.58 ± 0.69def

‘Bluegold’ 1.47 ± 0.094ij 1.86 ± 0.02d 9.22 ± 0.63de

‘Aron’ 1.13 ± 0.039efg 2.14 ± 0.04e 9.13 ± 0.87cd

‘Bluehaven’ 0.97 ± 0.020abcd 1.69 ± 0.03c 7.90 ± 0.94bc

‘Reka’ 1.04 ± 0.037bcde 0.99 ± 0.02h 7.72 ± 0.13b

‘Puru’ 1.08 ± 0.042def 0.17 ± 0.00a 7.20 ± 0.61ab

‘Bluecrop’ 0.94 ± 0.034abc 0.68 ± 0.02b 7.15 ± 0.54ab

‘Blueray ‘ 0.93 ± 0.033ab 1.54 ± 0.032i 7.02 ± 0.56ab

‘Bluejay’ 0.85 ± 0.027a 1.77 ± 0.03c 6.79 ± 0.28ab

‘Toro’ 0.99 ± 0.038bcd 0.49 ± 0.01g 6.48 ± 0.42a

‘Hardyblue’ 1.07 ± 0.033cdef 0.62 ± 0.01b 6.38 ± 0.34a

The means of at least three measurements; different superscript letter within the same co

are expressed as Trolox equivalent antioxidant capacity (TEAC) inlmol Trolox/g juice.

2.7. Ferric reducing antioxidant power assay (FRAP)

FRAP was measured as described elsewhere (Kraujalyte et al.,2013). Briefly, the working reagent was prepared by mixing acetatebuffer (300 mM, pH 3.6), a solution of 10 mM TPTZ in 40 mM HCl,and 20 mM FeCl3 � 6H2O at 10:1:1 (v/v/v). Ten lL of diluted (1:10w/v) juices and working reagent (300 lL, heated to 37 �C) werepippeted into the microplate wells, 30 lL H2O added, the platewas placed in a FLUOstar Omega reader and shaken for a 30 s.After 30 min incubation the absorbance was read at 593 nm. A ser-ies of Trolox and Fe2SO4 solutions, 0–1500 lM and 0–1000 lM,respectively, were used for the calibration. The reducing power isexpressed as TEAC in lmol TE/g juice and Fe2+ antioxidant capacity(Fe2+ lmol/g juice).

2.8. Oxygen radical absorbance capacity (ORAC)

ORAC was measured as described by Prior et al. (2003). Briefly,25 lL of diluted (1:200 w/v) juice and fluorescein (120 lL; 14 lM)solution were placed in 96 microplate wells, the mixture waspreincubated 15 min at 37 �C followed by the rapid addition ofAAPH solution as a peroxyl radical generator (26 lL; 240 mM).The microplate was immediately placed in a FLUOstar Omegareader, automatically shaken before each reading and the fluores-cence was measured after each cycle (1 min � 1.1, 80 cycles). Thefilters of 485-P excitation and 520-P emission were used. Theraw data were exported from the Mars software to an Excel 2003(Microsoft, Roselle, IL) sheet for further calculations. Antioxidantcurves were first normalized and from the normalized curves, thearea under the fluorescence decay curve (AUC) was calculated as

AUC ¼ 1þPi¼80

i¼1f if 0; where f0 is the initial fluorescence reading at

0 min and fi is the fluorescence reading at time i. The final ORACvalues were calculated by using a regression equation betweenthe Trolox concentration and the net AUC. Trolox solutions (0–250 lM) were used for the calibration. The ORAC values areexpressed as TEAC in lmol TE/g juice.

luble solids of blueberry juices.

l FRAP ORAC TEAC(lmol Trolox/g)

Soluble solids(%)

TEAC (lmolTrolox/g)

Fe2+

(lmol/g)

17.80 ± 0.72j 31.99 ± 1.31j 45.68 ± 4.44l 12.2 ± 0.00c

8.56 ± 0.16i 15.07 ± 0.30i 25.98 ± 1.82k 12.6 ± 0.00d

6.89 ± 0.18h 11.99 ± 0.33h 20.76 ± 1.42j 13.5 ± 0.00f

5.38 ± 0.11ef 9.23 ± 0.21ef 12.56 ± 0.99fgh 11.8 ± 0.00b

6.85 ± 0.09h 11.93 ± 0.3h 20.17 ± 0.67j 13.0 ± 0.05e

5.84 ± 0.16g 10.08 ± 0.29g 18.69 ± 3.84ij 12.6 ± 0.1d

5.45 ± 0.24efg 9.36 ± 0.43efg 12.91 ± 0.49fgh 12.8 ± 0.05i

4.78 ± 0.11d 8.13 ± 0.2d 15.76 ± 2.20hi 13.9 ± 0.00j

5.30 ± 0.48e 9.08 ± 0.88e 11.69 ± 1.72efg 11.8 ± 0.00b

5.77 ± 0.09fg 9.94 ± 0.17fg 14.26 ± 1.13gh 13.7 ± 0.00g

5.21 ± 0.11e 8.92 ± 0.20e 14.83 ± 1.08gh 13.6 ± 0.00fg

4.50 ± 0.04d 7.62 ± 0.08d 11.89 ± 1.37efg 11.8 ± 0.00b

3.90 ± 0.09c 6.52 ± 0.17c 4.59 ± 1.11ab 11.9 ± 0.05b

4.66 ± 0.07d 7.91 ± 0.13d 10.43 ± 0.61def 11.9 ± 0.05b

3.69 ± 0.22bc 6.13 ± 0.41bc 8.53 ± 0.43cde 12.5 ± 0.00d

3.72 ± 0.16bc 6.19 ± 0.29bc 7.30 ± 2.23bcd 11.6 ± 0.05h

3.64 ± 0.16bc 6.05 ± 0.29bc 7.30 ± 2.65bcd 10.8 ± 0.00a

3.48 ± 0.12b 5.74 ± 0.22b 5.35 ± 1.46abc 12.9 ± 0.00e

3.07 ± 0.05a 4.99 ± 0.10a 4.21 ± 0.46a 12.1 ± 0.05c

3.68 ± 0.18bc 6.11 ± 0.32bc 5.36 ± 0.87abc 15.1 ± 0.00k

lumn indicates significant differences (one way ANOVA and Duncans’ test, p < 0.05).

Table 3Values of L⁄, a⁄, b⁄, h� (hue degree) and C (chroma) of blueberry juices.

Genotype L⁄ a⁄ b⁄ h� C

Bog blueberry 22.65 ± 0.02b 0.06 ± 0.01a 0.14 ± 0.02a 1.168 ± 0.1c 0.17 ± 0.04a

No. 16 21.77 ± 0.36a 1.68 ± 0.12b 0.30 ± 0.02ab 0.187 ± 0.02a 1.70 ± 0.20b

‘Freda’ (No. 11) 23.06 ± 0.05ef 2.86 ± 0.07d 0.86 ± 0.11ef 0.292 ± 0.02ab 2.98 ± 0.09ef

‘Danute’ (No. 17) 22.96 ± 0.03def 2.01 ± 0.01c 0.53 ± 0.04cd 0.256 ± 0.04ab 2.08 ± 0.02c

‘Northland’ 22.61 ± 0.45b 3.26 ± 0.18e 0.98 ± 0.09fg 0.291 ± 0.04ab 3.40 ± 0.16g

‘Nui’ 23.07 ± 0.04ef 2.19 ± 0.06c 0.68 ± 0.06de 0.301 ± 0.03ab 2.29 ± 0.09cd

‘Putte’ 23.20 ± 0.09fgh 2.87 ± 0.10d 0.86 ± 0.09ef 0.289 ± 0.02ab 2.99 ± 0.12f

‘Patriot’ 22.74 ± 0.02bcd 0.69 ± 0.01g 0.17 ± 0.02ab 0.236 ± 0.04ab 0.71 ± 0.01h

‘Weymouth’ 23.08 ± 0.05ef 2.26 ± 0.12c 0.58 ± 0.1cd 0.250 ± 0.02ab 2.34 ± 0.13d

‘Northblue’ 23.08 ± 0.07efg 3.16 ± 0.20e 0.87 ± 0.08ef 0.270 ± 0.03ab 3.28 ± 0.18g

‘Bluegold’ 23.07 ± 0.05ef 2.64 ± 0.18d 0.71 ± 0.12de 0.263 ± 0.03ab 2.73 ± 0.13e‘Aron’ 22.90 ± 0.09cde 1.59 ± 0.18b 0.48 ± 0.05cd 0.302 ± 0.02ab 1.66 ± 0.21b

‘Bluehaven’ 23.35 ± 0.05h 4.85 ± 0.21f 1.59 ± 0.12i 0.318 ± 0.03b 5.10 ± 0.18m

‘Reka’ 23.77 ± 0.06i 6.22 ± 0.22l 1.98 ± 0.2k 0.309 ± 0.02ab 6.53 ± 0.18m

‘Puru’ 23.39 ± 0.10h 4.25 ± 0.14j 1.20 ± 0.11gh 0.276 ± 0.03ab 4.42 ± 0.13k

‘Bluecrop’ 23.37 ± 0.05h 3.65 ± 0.29i 0.99 ± 0.08fg 0.267 ± 0.04ab 3.78 ± 0.28j

‘Blueray’ 22.82 ± 0.03bcd 1.59 ± 0.18b 0.49 ± 0.05cd 0.299 ± 0.04ab 1.66 ± 0.15b

‘Bluejay’ 23.33 ± 0.07gh 4.62 ± 0.17f 1.30 ± 0.1h 0.273 ± 0.04ab 4.79 ± 0.21l

‘Toro 22.71 ± 0.27bc 5.37 ± 0.09k 1.82 ± 0.17ijk 0.326 ± 0.04b 5.66 ± 0.01n

‘Hardyblue’ 23.16 ± 0.02fgh 1.20 ± 0.04h 0.39 ± 0.07bc 0.314 ± 0.05b 1.26 ± 0.03i

The means of at least three measurements; different superscript letter within the same column indicates significant differences (one way ANOVA and Duncans’ test, p < 0.05).

586 V. Kraujalyte et al. / Food Chemistry 188 (2015) 583–590

2.9. UPLC/ESI-QTOF-MS analysis

An Acquity UPLC system (Waters, Milford, USA) combined witha Brüker maXis UHR-TOF mass spectrometer (Brüker Daltonics,Bremen, Germany) was used. The UPLC was equipped with a bin-ary solvent delivery system, an autosampler with a 10 lL sampleloop, a photodiode array (PDA) detector, a column manager andan Acquity BEH C18 column (1.7 lm, 50 � 2.1 mm, i.d.) maintainedat 25 �C. The mobile phase was initially composed of 100% eluent A(1% v/v formic acid in ultra-pure water) followed by an increasefrom 0% to 35% of eluent B (methanol) over 3 min. During the fol-lowing 2 min, the eluent B was linearly increased from 35% to100%, where it was maintained for 1 min. Finally, the initial condi-tions were re-introduced over 1 min, and the equilibration timewas 1 min. The flow rate was 0.4 mL/min and the effluent wasmonitored at 254 nm. The effluent from the PDA detector wasintroduced directly into the UHR-TOF with an ESI source.Instrument control and data acquisition were achieved using theCompass 1.3 (HyStar 3.2 SR2) software. MS experiments were per-formed in negative ionization mode. The capillary voltage wasmaintained at +4000 V with the end plate offset at �500 V.Nitrogen was used as the drying and nebulizing gases at10.0 L/min and 2.0 bar pressure. Nitrogen was introduced intothe collision cell as the collision gas. The peaks were identifiedby comparing their retention times with standards and by thecharacteristic MS fragmentation patterns and accurate masses.

2.10. Statistical data handling

Mean values and standard deviations of antioxidant activitywere calculated using MS Excel 2003 from at least 3 replicates.Statistical analysis was performed by using ANOVA, the differencesbetween samples were evaluated by the Duncans’ test at p < 0.05(STATISTICA 8.0 software, 2007).

3. Results and discussion

3.1. General characteristics and total monomeric anthocyanin content(TAC)

The highbush blueberry genotypes ‘Reka’, ‘Patriot’, ‘Bluejay’,‘Toro’, ‘Bluehaven’, ‘Bluegold’, ‘Bluecrop’, ‘Blueray’, and ‘Nui’ pro-duced the largest berries with an average mass from 2.34 to

3.62 g (Table 1). An average berry mass of the highbush ‘Puru’ andthe half-highbush ‘Northblue’ was 2.16 and 1.73 g, respectively,whereas other genotypes, such as the highbush ‘Weymouth’ andwild bog blueberry had berry mass less than 1.5 g. It should be notedthat large berries are easier to harvest and cultivars with largerberries are more suitable for fresh consumption and are preferredby the consumers. The content of soluble solids was in the rangeof 10.8–15.1% (Table 2); their percentage in newly bred genotypes‘Freda’, ‘No. 16’, and ‘Danute’ were 13.5%, 11.8%, and 12.6%,respectively.

Colour of berries is associated with the presence of various bioac-tive compounds; e.g. orange, red, purple and blue colours are relatedto the presence of such pigments as anthocyanins, carotenoids andbetalains. The composition of blueberries depends on the plantgenotype and some other factors (Cho et al., 2004; Moze et al.,2011; Wu & Prior, 2005); for instance, the study of newly bredgenotypes revealed that malvidin constitutes 30–47% of the totalanthocyanins (Scalzo et al., 2013). The content of the anthocyaninsin blueberries was reported in the range of 0.24–0.42 mg/g(Anisimoviene et al., 2009) and 0.18–2.94 mg/g (Koca & Karadeniz,2009). The TAC values for V. corymbosum genotypes in our study(Table 2) were in the range from 0.17 (‘Puru’) to 3.2 mg/100 mL(‘Patriot’), whereas bog blueberry had remarkably higher TAC,14.19 mg/100 mL. It may be noted that newly bred genotypes,‘Danute’ (2.56 mg/100 mL) and No. 16 (2.48 mg/100 mL) wereamong the richest in TAC. Considerable variations in berry juiceTAC values may be observed in scientific articles, which may alsobe influenced by the applied procedure. We used a method of Leeet al. (2008) who reported the following TAC values for various fruitjuice, in mg/100 mL: blueberry 3.56, cranberry 1.31, grape 1.64,concord grape 3.12, tart cherry 2.09, black cherry juice 4.39,pomegranate 5.92. In general, it is in agreement with our study.

Blueberry and their juice colour measurements were performedpreviously (Barba, Esteve, & Frígola, 2013; Barba et al., 2012;Giovanelli & Buratti, 2009). For instance, the values of untreatedberry juices were L⁄ = 31.54–32.66, a⁄ = 0.25–0.36, b⁄ =�6.83 � (�6.80), hue = 87.90, the Chroma = 6.81 (Barba et al.,2012, 2013). Our study showed that he lightness L⁄was in the rangefrom 21.7 (‘No. 16’, the darkest) to 23.39 (‘Puru’); significant differ-ences were observed between some studied genotypes (Table 3).The positive value of a⁄ indicates redness, which was considered tobe related with the presence of anthocyanins. The a⁄ values werefrom 0.06 (bog blueberry) to 6.22 (‘Reka’). The positive value of b⁄

indicates yellowness, while negative value is related to blueness.

Fig. 1. PCA score plots of sensor array (electronic tongue) data of blueberry juices: above – overall; below enlarged part.

V. Kraujalyte et al. / Food Chemistry 188 (2015) 583–590 587

The b⁄ values of the tested berry juices were from 0.14 (bog blue-berry) to 1.98 (‘Reka’). The highest value of hue angle (h�) was deter-mined for bog blueberry juice (1.16), while among V. corymbosumsamples it was in the range of 0.187 (‘No. 16’)–0.326 (‘Toro’).Strong positive correlation was found between h� and anthocyanincontent, 0.8593. The lowest chroma (C) value was determined forbog blueberry (0.17), while among blueberries of V. corymbosumspecies these values ranged from 0.71 (‘Patriot’) to 6.53 (‘Reka’).

3.2. Differentiation of berry juices by potentiometric sensor array(‘electronic tongue’)

Sensory quality characteristics of blueberries such asblueberry-like flavour, sweetness, juiciness and appearance (colourand size) were reported as being very important in consumer study(Saftner, Polashock, Ehlenfeldt, & Vinyard, 2008). Over 100 volatileconstituents were identified in blueberries depending on their geno-types, however ‘‘green notes’’ possessing compounds (E)-2-hexenal,(E)-2-hexenol, (Z)-3-hexenol and terpenes such as linalool andgeraniol were found in most studied cultivars and are consideredto be typical aroma constituents of blueberries (Du & Qian, 2010;Hirvi & Honkan, 1983).

Electronic tongues with potentiometric sensors, particularlyion-selective electrodes are the most widely used type of sensorarrays for the assessment of liquids (Bratov, Abramova, & Ipatov,2010; Ciosek & Wróblewski, 2011). Discrimination of food samplesby electronic tongue were compared with sensory evaluation andin some cases with chemical analysis data (Diasa et al., 2011;Kantor, Hitka, Fekete, & Balla, 2008; Kovács et al., 2010; Siposet al., 2012); these studies demonstrated that electronic tonguewas more sensitive to the differences between the tested samplescomparing with sensory analysis.

To the best of our knowledge Vaccinium species were not previ-ously evaluated by electronic tongue. In general, the principal com-ponent analysis (PCA) score plot shows that the instrumenteffectively discriminates different juices, although the data forsome samples were located very closely and overlapping (Fig. 1).Based on PCA all analysed samples could be grouped into the clus-ters depending on the berry genotype. The main group consisted of‘Freda’ (‘No. 11’), ‘Bluejay’, ‘Bluecrop’, ‘Toro’, ‘Weymouth’, ‘Puru’,‘Northblue’, ‘Patriot’ ir ‘Reka’. In this group, ‘Northblue’, ‘Patriot’and ‘Reka’ samples were located closely indicating similarities ofsensor’s responses to the juice from these genotypes. The smallergroups encompassed ‘Aron’, ‘Danute’ (‘No. 17’), ‘Blueray’ and ‘No.

588 V. Kraujalyte et al. / Food Chemistry 188 (2015) 583–590

16’ as well as ‘Northland’, and ‘Putte’ samples. Bog blueberry and‘Nui’ juice samples were closely located to each other, howeverrather distantly discriminated from the main group. ‘Bluegold’and ‘Bluehaven’ samples were clearly discriminated from the alltested samples, while ‘Hardyblue’ sample was located in the mostremote site of the plot. These results demonstrate that electronictongue may be a promising technique for discrimination of blue-berry juices produced from different plant genotypes; however,standardized datasets based on a large number of measurementsshould be compiled for this purpose.

3.3. Antioxidant properties of berry juices measured in electron/hydrogen transfer based assays

3.3.1. Total phenolic content (TPC)The values measured with Folin–Ciocalteu reagent are generally

accepted as representing the TPC although it is not fully correct;the reagent also reacts with other reducing ability possessing com-pounds. Consequently, the term TPC may be used rather condition-ally. The highest TPC values were found for bog blueberry (2.81 mgGAE/mL), whereas the values for V. corymbosum genotypes werefrom 0.85 (‘Bluejay’) to 1.76 mg GAE/mL (‘No. 16’) (Table 2). It isinteresting noting that the TPC values of newly bred genotypes,namely ‘No. 16’ (1.76 mg GAE/mL), ‘Freda’ (1.50 mg GAE/mL), and‘Danute’ (1.50 mg GAE/mL) were higher than those of thewell-known genotypes such as ‘Northland’, ‘Bluegold’, ‘Nui’,‘Putte’, ‘Northblue’ and ‘Patriot’. TPC in ‘Northland’ was reportedpreviously (1.84 mg GAE/g fresh weight (fw)) and it is in agree-ment with our results (Koca & Karadeniz, 2009). The TPC valueswere previously also measured for blueberry fruit extracts; forinstance, for ‘Bluecrop’ cultivated organically and conventionallythey were reported 3.2 and 1.9 mg/g (fw), respectively (Wanget al., 2008).

3.3.2. ABTS�+ decolourisation assayIn this assay bog blueberry juices demonstrated the highest

RSC, 20.9 lmol TE/g, whereas the TEAC values of V. corymbosumgenotypes were from 6.38 (‘Hardyblue’) to 14.64 (‘No. 16’) lmolTE/g (Table 2). Other newly bred genotypes, ‘Danute’ (12.0 lmolTE/g) and ‘Freda’ (11.55 lmol TE/g) also demonstrated high TEAC

Table 4Quantification of quinic acid, chlorogenic acid and rutin in blueberry juices.

Genotype Quinic acid(lg/mL)

Chlorogenic acid(lg/mL)

Rutin(lg/mL)

Bog blueberry 2673 ± 318.46g 20.0 ± 0.3a 0No. 16 3614 ± 205.30i 346.8 ± 4.4k 10.72 ± 0.96c

‘Freda’ (No. 11) 3267 ± 381.82hi 126.7 ± 2.5h 12.68 ± 0.26f

‘Danute’ (No. 17) 3335 ± 143.46hi 165.0 ± 5.9i 17.22 ± 1.96g

‘Northland’ 289 ± 9.67a 233.2 ± 4.3j 5.45 ± 0.68b

‘Nui’ 1902 ± 215.43f 139.9 ± 1.5g 26.88 ± 1.70e

‘Putte’ 2465 ± 104.22g 119.5 ± 1.8f 6.01 ± 1.11b

‘Patriot’ 3380 ± 355.61hi 53.4 ± 2.1de 9.59 ± 0.38c

‘Weymouth’ 621 ± 46.58bcd 120.5 ± 2.5f 5.80 ± 0.82b

‘Northblue’ 925 ± 45.55de 136.1 ± 3.6g 0‘Bluegold’ 1795 ± 52.35f 120.2 ± 1.9f 6.36 ± 0.49b

‘Aron’ 745 ± 15.28cd 55.7 ± 1.5e 2.93 ± 0.16a

‘Bluehaven’ 1083 ± 66.92e 43.8 ± 0.3c 3.08 ± 0.40a

‘Reka’ 1063 ± 26.51e 136.1 ± 1.1g 6.28 ± 0.31b

‘Puru’ 1938 ± 157.40f 30.1 ± 0.4b 27.10 ± 0.53e

‘Bluecrop’ 619 ± 20.41bc 50.1 ± 1.2d 9.76 ± 0.71c

‘Blueray’ 214 ± 6.49a 30.3 ± 0.5b 5.49 ± 1.00b

‘Bluejay’ 431 ± 1.87ab 30.5 ± 0.7b 2.65 ± 0.30a

‘Toro’ 337 ± 23.64ab 39.7 ± 0.3c 0‘Hardyblue’ 203 ± 8.14a 20.3 ± 0.4a 2.16 ± 0.29a

The means of at least three measurements; different superscript letter within thesame column indicates significant differences (one way ANOVA and Duncans’ test,p < 0.05).

values. Strong correlation was found between TPC and ABTS�+ scav-enging (r = 0.93). For comparison, in the previous study TEAC val-ues in ABTS�+ assay of commercial brand of blueberry juice wasin the range of 13.3–17.1 lmol TE/mL, which, in general, is inagreement with our results (Seeram et al., 2008).

3.3.3. Ferric reducing antioxidant power (FRAP)FRAP was calculated in Fe2+ equivalents and Trolox equivalents

(Table 2), because both values are informative, particularly forcomparison with the results reported in other studies. Again, thehighest values were determined for bog blueberry juice(31.99 lmol Fe2+/g and 17.80 lmol TE/g), whereas for V. corymbo-sum genotypes they were from 4.99 Fe2+/g and 3.07 lmol TE/g(‘Toro’) to 15.07 lmol Fe2+/g and 8.56 lmol TE/g (‘No. 16’).Strong positive correlation (r = 0.93) was observed between TPCand FRAP.

Previously reported FRAP values of blueberry juices obtainedfrom different producers were in the range of 3.6–4.7 lmolFe2+/mL (Seeram et al., 2008). Consequently, the newly bred geno-types of blueberries, ‘No. 16’, ‘Danute’ and ‘Freda’ had approxi-mately 2–3 times higher FRAP values than previously analysedcommercial juice samples. FRAP of blueberry extracts isolated withdifferent solvents was higher than that of berry juice as it wasshown for acetone/water extracts of ‘Northblue’, ‘Bluecrop’,‘Bluejay’ genotypes (20.1–25.5 lmol of TE/g fw) and bog blueber-ries harvested from different shrubs (24.0–28.2 lmol of TE/g fw)(Taruscio, Barney, & Exon, 2004). Koca and Karadeniz (2009)reported 13.69 Fe2+ lmol mg/g for the extract of ‘Northland’ and7.41–12.36 Fe2+ lmol mg/g of ‘Ivanhoe’, ‘Jersey’, ‘Rekord’, the culti-vars, which were not analysed in our study. Other reported FRAPvalues were even higher, 30.0 lmol Fe2+/g in the extract of unde-clared blueberry cultivar (Borges, Degeneve, Mullen, & Crozier,2010); 24.3 and 26.4 TE lmol/g in the extracts of ‘Patriot’ and‘Bluecrop’, respectively (Giovanelli & Buratti, 2009).

3.3.4. Oxygen radical absorbance capacity (ORAC)The ORAC values for studied V. corymbosum genotypes were in

the range from 4.21 (‘Toro’) to 25.98 (‘No. 16’) lmol TE/g (Table 2);however, the highest value was determined for bog blueberry(45.68 lmol TE/g). ‘Danute’, ‘Northland’ and ‘Nui’ genotypes maybe attributed to the second group of the V. corymbosum blueberries,possessing high ORAC values (p < 0.05). For comparison, previouslyreported ORAC values of blueberry juices from different producerswere from 14.5 to 23.9 lmol TE/mL (Seeram et al., 2008). Strongcorrelation (r = 0.92) was found between TPC and ORAC in ourstudy, which is in agreement with the previously published datafor blueberries (Wang et al., 2008).

The ORAC values of extracts of eight Vaccinium cultivars ofnorthern highbush blueberry measured using phycoerythrin werefrom 17.0 lmol of TE/g fw for ‘Bluecrop’ to 17–37 lmol of TE/gfw in other, not analysed in our study, genotypes (Prior et al.,1998). Hydrophilic extract isolated with acetone/water/acetic acidfrom a wide variety of fruits including undeclared blueberrycultivars constituted up to 90% of the total antioxidant capacityof blueberries (Wu et al., 2004). Cranberries and lowbush blueber-ries yielded the strongest hydrophilic antioxidants, followed by thecultivated blueberry with the average value of 61.84 lmol of TE/gfw. Cho et al. (2004) reported ORAC using fluorescein for five geno-types of cultivated blueberries in the range of 52 (‘Bluecrop’)–139 lmol of TE/g fw. The mean ORAC values of organically andconventionally cultured ‘Bluecrop’ cultivar extracts were 46.14and 30.76 lmol/g fw, respectively (Wang et al., 2008).

3.3.5. Factors influencing the antioxidant capacity of blueberriesIt is known that genetics, environmental conditions, harvesting

time and other factors may significantly influence the content of

Table 5Correlation matrix for berry mass, different biochemical components and ORAC, FRAP, and ABTS�+ values in berries of different blueberry genotypes.

BM TPC SS QA CHA RU FRAP ABTS�+ ORAC

BM 1.000 �0.557⁄l 0.143n 0.412n �0.533⁄l 0.020n �0.617⁄⁄ �0.624⁄⁄l �0.467⁄lTPC �0.557⁄l 1.000 0.272n 0.665⁄⁄l 0.858⁄⁄l 0.324n 0.917⁄⁄l 0.914⁄⁄l 0.903⁄⁄lSS 0.143n 0.272n 1.000 0.188n 0.077n �0.044n 0.203n 0.114n 0.236nQA 0.412n 0.665⁄⁄l 0.188n 1.000 0.467⁄l 0.520⁄l 0.588⁄⁄l 0.741⁄⁄l 0.639⁄⁄lCHA �0.533⁄l 0.858⁄⁄l 0.077n 0.467⁄l 1.000 0.154n 0.951⁄⁄l 0.863⁄⁄l 0.878⁄⁄lRU 0.020n 0.324n �0.044n 0.520⁄l 0.154n 1.000 0.240n 0.293n 0.372nFRAP �0.617⁄⁄l 0.917⁄⁄l 0.203n 0.588⁄⁄l 0.951⁄⁄l 0.240n 1.000 0.947⁄⁄l 0.955⁄⁄lABTS�+ �0.624⁄⁄l 0.914⁄⁄l 0.114n 0.741⁄⁄l 0.863⁄⁄l 0.293n 0.947⁄⁄l 1.000 0.924⁄⁄lORAC �0.467⁄l 0.903⁄⁄l 0.236n 0.639⁄⁄l 0.878⁄⁄l 0.372n 0.955⁄⁄l 0.924⁄⁄l 1.000

BM – average berry mass, TPC – total phenolic content, SS – soluble solids, QA – quinic acid, CHA – chlorogenic acid, RU – rutin, FRAP – ferric reducing antioxidant power,ABTS�+ – ABTS�+ radical cation decolorisation, ORAC – oxygen radical absorbance capacity.*, ** Significant at p 6 0.05 and 0.01, respectively. l – correlation linear, n – correlation non-linear.

V. Kraujalyte et al. / Food Chemistry 188 (2015) 583–590 589

secondary metabolites in fruits. In this study, the climatic and cul-tivation conditions were similar for all blueberry accessions; there-fore the variations in TPC, TAC and antioxidant activity indicatorsamong the tested genotypes, most likely, are determined by theirgenetic differences. Sample preparation for the evaluation ofantioxidant capacity is also an important issue, so far as extractionprocedure and solvents as well as handling after extraction mayinfluence the results. As it was reported, the mixtures of varioussolvents extract higher quantities of phenolic compounds fromblueberries comparing with their content in juice (Giovanelli &Buratti, 2009; Wang et al., 2008).

Fruit size and pigment distribution are other factors, which mayinfluence antioxidant capacity results. Highbush blueberries accu-mulate pigments only in the fruit peel, while in bog blueberry thepigments are present both in the peel and in the flesh. Breedingprograms often are seeking larger size fruits; so far as anthocyaninsare concentrated near the skin, and smaller blueberries would havelarger skin surface area per unit weight, they may contain morephenolics and consequently exhibit higher antioxidant capacity(Cho et al., 2004).

3.4. Quantification of quinic acid, chlorogenic acids and rutin in berryjuices by UPLC–MS/MS

Quinic acid, chlorogenic acid and rutin were quantified by arapid UPLC/ESI-QTOF-MS method (Table 4). Quinic acid does notcontain aromatic ring and phenolic group, however in some studiesit is discussed as a phenolic compound due to metabolic consider-ations (Robards, Prenzler, Tucker, Swatsitang, & Glover, 1999). Theconcentration of quinic acid was in a very wide range, from 203(‘Hardyblue’) to 3614 (‘No. 16’) lg/mL. Other newly bred geno-types, ‘Danute’ and ‘Freda’ also contained more than 3 mg/mL, aswell as known cultivar ‘Patriot’. Previously, an average concentra-tion of quinic acid in solvent extract of blueberry freeze-dried pow-dered material was reported 25.5 mg/g (Phillips et al., 2010).However, to the best of our knowledge there is no published dataabout quinic acid concentration in V. corymbosum berry juices.Comparing with such fruits as lemon, grape, peach, orange, andmelon containing 440, 35, 34, 140, and 0.8 lg/g fw of quinic acid,respectively (Flores, Hellín, & Fenol, 2012), blueberries may be con-sidered as a rich source of this constituent.

Chlorogenic acid has been reported as a major phenolic acid inblueberries (Häkkinen & Törrönen, 2000; Moze et al., 2011;Rodriguez-Mateos et al., 2012). Its concentration in juices was inthe range of 210–850 lg/mL (Brambilla et al., 2008), while in thewhole blueberries – 700 lg/g fw (Moze et al., 2011). In the juicesof V. corymbosum analysed in our study the content of chlorogenicacid (Table 4) was from 20.3 (‘Hardyblue’) to 346.8 lg/mL (‘No.16’), while in other new genotypes, ‘Freda’ and ‘Danute’ it wasremarkably lower. The majority of cultivars contained less than

140 lg/mL of chlorogenic acid except for ‘Northland’, which wasfound to be quite rich in this compound.

Rutin is found in many plant origin materials, including fruitsand berries. For instance, in strawberry juice it was reported4.41 lg/mL, sour cherry 38.18 lg/mL, black grape 1.54 lg/mL(Díaz-García, Obón, Castellar, Collado, & Alacid, 2013), blueberry31 lg/g fw (Moze et al., 2011). The amount of rutin in berry geno-types analysed in our study was from undetectable (‘Toro’,‘Northblue’, bog blueberry) to 27.1 lg/mL (‘Puru’). ‘Nui’ alsocontained high amount of rutin, while among other genotypes itsconcentration was higher than 10 lg/mL only in the juice of newlybred cultivars ‘Danute’ (17.22 lg/mL), ‘Freda’ (12.68 lg/mL), and‘No. 16’ (10.72 lg/mL). It is interesting noting that bog blueberry,although containing moderate amount of the quantified com-pounds, demonstrated the highest antioxidant capacity. Mostlikely, other constituents are important for bog blueberry as itmay be observed from its rather different chromatographic profile.

3.5. Correlation between berry mass, different biochemicalcomponents and the values measured by different antioxidantevaluation methods in various berry genotypes

The quality of berries depends on numerous physical, chemical,nutritional and sensory characteristics. Some studies observed thecorrelations between various berry quality and compositionalcharacteristics. Therefore, correlation coefficients were calculatedfor the data obtained in our study as well (Table 5). For instance,an inverse relationship between average berry mass and TPC aswell as the concentration of chlorogenic acid may be observed.Moderate negative correlation was found between berry massand ABTS�+, FRAP and ORAC. It may be suggested that berries pos-sessing higher mass accumulate more macrocomponents, e.g. car-bohydrates, which do not increase antioxidant capacity of the totaldry substances. In this case correlation coefficients provide ratherpreliminary information, however it may be suggested that breed-ing of half-highbush berry plants may be a promising means of cre-ating cultivars as a good source of antioxidants. The correlationsbetween similar characteristics but measured by different methodswere quite high. Thus, the genotypes containing larger amounts ofTPC possessed high values of ABTS�+, FRAP, and ORAC; correlationcoefficients were 0.914, 0.917, and 0.903, respectively. ABTS�+,FRAP, and ORAC values also correlated with the concentration ofquinic and chlorogenic acids (p 6 0.01).

4. Conclusion

Systematic evaluation of various quality characteristics of somewell-known and commonly cultivated blueberry genotypes, aswell as newly bred half-highbush blueberry genotypes (‘No. 16’,‘Northland’, ‘Danute’, ‘Freda’), demonstrated that the latter pos-sessed the highest TPC and TAC values, as well as the concentration

590 V. Kraujalyte et al. / Food Chemistry 188 (2015) 583–590

of quinic acid. The concentration of chlorogenic acid was high in‘No. 16’, ‘Danute’ and well-known genotype ‘Northland’. Berriesof the indigenous V. uliginosum species are distinguished for theexceptional high values of TPC and TAC. The results of this studysuggest that germplasm of half-highbush blueberry and V. uligi-nosum could be used in breeding new cultivars with enhancedantioxidant capacity.

Acknowledgement

This study was supported by the Research Council of Lithuania,Grant No. SVE-04/2011.

References

Anisimoviene, N., Rudzinskiene, M., Viškelis, P., Stankeviciene, E., Stanys, V.,Šikšnianas, T., et al. (2009). Anthocyanins in currants, cherries, blueberriesand antioxidative activity of berry extracts. Zemdirbyste – Agriculture, 96,158–167.

Barba, F. J., Esteve, M. J., & Frígola, A. (2013). Physicochemical and nutritionalcharacteristics of blueberry juice after high pressure processing. Food ResearchInternational, 50, 545–549.

Barba, F. J., Jäger, H., Meneses, N., Esteve, M. J., Frígola, A., & Knorr, D. (2012).Evaluation of quality changes of blueberry juice during refrigerated storageafter high-pressure and pulsed electric fields processing. Innovative Food Scienceand Emerging Technologies, 14, 18–24.

Borges, G., Degeneve, A., Mullen, W., & Crozier, A. (2010). Identification of flavonoidand phenolic antioxidants in black currants, blueberries, raspberries, redcurrants, and cranberries. Journal of Agricultural and Food Chemistry, 58,3901–3909.

Brambilla, A., Lo Scalzo, R., Bertolo, G., & Torreggiani, D. (2008). Steam-bleachedhighbush blueberry (Vaccinium corymbosum L.) juice: Phenolic profile andantioxidant capacity in relation to cultivar selection. Journal of Agricultural andFood Chemistry, 56, 2643–2648.

Bratov, A., Abramova, N., & Ipatov, A. (2010). Recent trends in potentiometric sensorarrays—A review. Analytica Chimica Acta, 678, 149–159.

Brownmiller, C., Howard, L. R., & Prior, R. L. (2009). Processing and storage effects onprocyanidin composition and concentration of processed blueberry products.Journal of Agricultural and Food Chemistry, 57, 1896–1902.

Castrejón, A. D. R., Eichholz, I., Rohn, S., Kroh, L. W., & Huyskens-Keil, S. (2008).Phenolic profile and antioxidant activity of highbush blueberry (Vacciniumcorymbosum L.) during fruit maturation and ripening. Food Chemistry, 109,564–572.

Cho, M. J., Howard, L. R., Prior, R. L., & Clark, J. R. (2004). Flavonoid glycosides andantioxidant capacity of various blackberry, blueberry and red grape genotypesdetermined by high-performance liquid chromatography/mass spectrometry.Journal of the Science of Food and Agriculture, 84, 1771–1782.

Ciosek, P., & Wróblewski, W. (2011). Potentiometric electronic tongues for foodstuffand biosample recognition – An overview. Sensors, 11, 4688–4701.

Diasa, L. G., Peresa, A. M., Barcelosa, T. P., Sá Moraisa, J., & Machado, A. A. S. C. (2011).Semi-quantitative and quantitative analysis of soft drinks using an electronictongue. Sensors and Actuators B, 154, 111–118.

Díaz-García, M. C., Obón, J. M., Castellar, M. R., Collado, J., & Alacid, M. (2013).Quantification by UHPLC of total individual polyphenols in fruit juices. FoodChemistry, 138, 938–949.

Du, X., & Qian, M. (2010). Flavor chemistry of small fruits: Blackberry, raspberry,and blueberry. In M. C. Qian & A. M. Rimando (Eds.). Flavor and Health Benefits ofSmall Fruits: ACS Symposium Series 1035 (pp. 27–43). Washington, DC: AmericanChemical Society.

Escuder-Gilaberta, L., & Peris, M. (2010). Review: Highlights in recent applicationsof electronic tongues in food analysis. Analytica Chimica Acta, 665, 15–25.

Flores, P., Hellín, P., & Fenol, J. (2012). Determination of organic acids in fruits andvegetables by liquid chromatography with tandem-mass spectrometry. FoodChemistry, 132, 1049–1054.

Giovanelli, G., & Buratti, S. (2009). Comparison of polyphenolic composition andantioxidant activity of wild Italian blueberries and some cultivated varieties.Food Chemistry, 112, 903–908.

Häkkinen, S. H., & Törrönen, A. R. (2000). Content of flavonols and selected phenolicacids in strawberries and Vaccinium species: Influence of cultivar, cultivationsite and technique. Food Research International, 33, 517–524.

Hirvi, T., & Honkan, E. (1983). The aroma of some hybrids between high-bushblueberry (Vaccinium corymbosum, L.) and bog blueberry (Vaccinium uliginosum,L.). Zeitschrift für Lebensmittel-Untersuchung und – Forschung, 176, 346–349.

Kantor, D. B., Hitka, G., Fekete, A., & Balla, C. (2008). Electronic tongue for sensingtaste changes with apricots during storage. Sensors and Actuators B: Chemical,131, 43–47.

Koca, I., & Karadeniz, B. (2009). Antioxidant properties of blackberry and blueberryfruits grown in the Black Sea Region of Turkey. Scientia Horticulturae, 121,447–450.

Kovács, Z., Dalmadi, I., Lukács, L., Sipos, L., Szántai-K}ohegyi, K., Kókai, Z., et al.(2010). Geographical origin identification of pure Sri Lanka tea infusions withelectronic nose, electronic tongue and sensory profile analysis. Journal ofChemometrics, 24, 121–130.

Kraujalyte, V., Venskutonis, P. R., Pukalskas, A., Cesoniene, L., & Daubaras, R. (2013).Antioxidant properties and polyphenolic compositions of fruits from differentEuropean cranberrybush (Viburnum opulus L.) genotypes. Food Chemistry, 141,3695–3702.

Lee, J., Rennaker, C., & Wrolstad, R. E. (2008). Correlation of two anthocyaninquantification methods: HPLC and spectrophotometric methods. FoodChemistry, 110, 782–786.

Mattila, P., Hellström, J., & Törrönen, R. (2006). Phenolic acids in berries, fruits, andbeverages. Journal of Agricultural and Food Chemistry, 54, 7193–7199.

Moze, S., Polak, T., Gasperlin, L., Koron, D., Vanzo, A., Poklar Ulrih, N., et al. (2011).Phenolics in Slovenian bilberries (Vaccinium myrtillus) and blueberries(Vaccinium corymbosum). Journal of Agricultural and Food Chemistry, 59,6998–7004.

Phillips, M. M., Case, R. J., Rimmer, C. A., Sander, L. C., Sharpless, K. E., Wise, S. A.,et al. (2010). Determination of organic acids in Vaccinium berry standardreference materials. Analytical and Bioanalytical Chemistry, 398, 425–434.

Prior, R. L., Cao, G., Martin, A., Sofic, E., McEwen, J., O’Brien, C., et al. (1998).Antioxidant capacity as influenced by total phenolic and anthocyanin content,maturity, and variety of Vaccinium species. Journal of Agricultural and FoodChemistry, 46, 2686–2693.

Prior, R. L., Hoang, H., Gu, L., Wu, X., Bacchiocca, M., Howard, L., et al. (2003). Assaysfor hydrophilic and lipophilic antioxidant capacity (oxygen radical absorbancecapacity (ORAC(FL)) of plasma and other biological and food samples. Journal ofAgricultural and Food Chemistry, 51, 3273–3279.

Re, R., Pellegrini, N., Proteggente, A., Pannala, A., Yang, M., & Rice-Evans, C. (1999).Antioxidant activity applying an improved ABTS radical cation decolorizationassay. Free Radical Biology and Medicine, 26, 1231–1237.

Robards, K., Prenzler, P. D., Tucker, G., Swatsitang, P., & Glover, W. (1999). Phenoliccompounds and their role in oxidative processes in fruits. Food Chemistry, 66,401–436.

Rodriguez-Mateos, A., Cifuentes-Gomez, T., Tabatabaee, S., Lecras, C., & Spencer, J. P.E. (2012). Procyanidin, anthocyanin, and chlorogenic acid contents of highbushand lowbush blueberries. Journal of Agricultural and Food Chemistry, 60,5772–5778.

Saftner, R., Polashock, J., Ehlenfeldt, M., & Vinyard, B. (2008). Instrumental andsensory quality characteristics of blueberry fruit from twelve cultivars.Postharvest Biology and Technology, 49, 19–26.

Scalzo, J., Politi, A., Pellegrini, N., Mezzetti, B., & Battino, M. (2005). Plant genotypeaffects total antioxidant capacity and phenolic contents in fruit. Nutrition, 21,207–213.

Scalzo, J., Stevenson, D., & Hedderley, D. (2013). Blueberry estimated harvest fromseven new cultivars: Fruit and anthocyanins. Food Chemistry, 139, 44–50.

Seeram, N. P., Aviram, M., Zhang, Y., Henning, S. M., Feng, L., Dreher, M., et al. (2008).Comparison of antioxidant potency of commonly consumed polyphenol-richbeverages in the United States. Journal of Agricultural and Food Chemistry, 56,1415–1422.

Sipos, L., Kovács, Z., Sági-Kiss, V., Csiki, T., Kókai, Z., Fekete, A., et al. (2012).Discrimination of mineral waters by electronic tongue, sensory evaluation andchemical analysis. Food Chemistry, 135, 2947–2953.

Taruscio, T. G., Barney, D. L., & Exon, J. (2004). Content and profile of flavanoid andphenolic acid compounds in conjunction with the antioxidant capacity for avariety of northwest Vaccinium berries. Journal of Agricultural and FoodChemistry, 52, 3169–3176.

Wang, S. Y., Chen, C. T., Sciarappa, W., Wang, C. Y., & Camp, M. J. (2008). Fruitquality, antioxidant capacity, and flavonoid content of organically andconventionally grown blueberries. Journal of Agricultural and Food Chemistry,56, 5788–5794.

Wu, X., Beecher, G. R., Holden, J. M., Haytowitz, D. B., Gebhardt, S. E., & Prior, R. L.(2004). Lipophilic and hydrophilic antioxidant capacities of common foods inthe United States. Journal of Agricultural and Food Chemistry, 52, 4026–4037.

Wu, X., & Prior, R. L. (2005). Systematic identification and characterization ofanthocyanins by HPLC–ESI-MS/MS in common foods in the United States: Fruitsand berries. Journal of Agricultural and Food Chemistry, 53, 2589–2599.