Determination of Nonanthocyanin Phenolic Compounds Using High-Resolution Mass Spectrometry (UHPLC-Orbitrap-MS/MS) and Impact of Storage Conditions in a Beverage Made from Strawberry by Fermentation

Depósito de investigación de la Universidad de Sevilla https://idus.us.es/ "This document is the Accepted Manuscript version of a Published Work that appeared in final form in JOURNAL OF AGRICULTURAL AND FOOD CHEMISTRY, copyright © American Chemical Society after peer review and technical editing by the publisher. To access the final edited and published work see https://doi.org. 10.1021/acs.jafc.5b05617 Determination of Nonanthocyanin Phenolic Compounds Using High-Resolution Mass Spectrometry (UHPLC-Orbitrap-MS/MS) and Impact of Storage Conditions in a Beverage made from Strawberry by Fermentation. M. Antonia Álvarez-Fernández, Ruth Hornedo-Ortega, A.B. Cerezo, Ana M. Troncoso and M. Carmen García-Parrilla*. Nutrición y Bromatología. Facultad de Farmacia. Universidad de Sevilla. C/ P. García González nº 2. Sevilla 41012, Spain. *Corresponding author fax: +34954233765; phone: +349545560; e-mail: mcparr[email protected]. ABSTRACT 1 Overproduction of strawberry leads to food waste, as it is very perishable. Therefore, 2 strategies to transform it into new products are appreciated. This research focuses on 3 characterization of the nonanthocyanin phenolic content of a beverage obtained from 4 strawberry by gluconic and acetic fermentation and subsequently monitored for 90 days 5 of storage, at two temperatures. 64 non-anthocyanin (poly) phenols were identified by 6 high-resolution mass spectrometry (UHPLC coupled with Linear Trap Quadropole and 7 OrbiTrap mass analyzer) and, for the first time, four compounds were reported in 8 beverages fermented from strawberry: aromadendrin hexoside, phloretin 2’-O-xylosyl 9 glucoside, dihydroferulic acid 4-O-glucuronide and kaempferol hexosil hexoside. 10 During the storage time the increased of protocatechuic acid content was thirteen times 11 and condensed tannins diminish, especially procyanidin trimer. 12 Statiscal analysis showed that the composition remains unchanged until day 15 of storage 13 at room temperature (27-30ºC) and day 30 under refrigerated conditions (4ºC). 14 Keywords: Stability, Orbitrap-mass spectrometer, Gluconic fermentation, Principal 15 component analysis, Sensory. 16 17 18 19 20 21 22 INTRODUCTION 23 Phenolic compounds are a group of ubiquitous compounds throughout the plant kingdom 24 and many occur in foods. In the late 20th century, interest in food rich in phenolic 25 compounds increased due to their antioxidant and anti-inflammatory properties, their 26 modulation of signal transduction and their anti-microbial and anti-proliferation 27 activities. 1 Consuming fruits with high polyphenol content has beneficial health 28 implications. 29 Strawberry (Fragaria x ananassa) is a source of many phenolic compounds that have 30 beneficial effects on health. This product is widely produced and consumed thoroughout 31 the world and Spain one of the leading producers, with a total production of 312,500 MT2; 32 this is not consumed or exported in its entirety. This situation generates overproduction 33 and, consequently, waste, which could be avoided by creating foodstuffs that maintain 34 unchanged the properties of the fruit used as raw material as much as possible as well as 35 its composition in bioactives. 36 Fermentation is an ancient process to transform and preserve foods. It involves raw 37 products undergoing chemical transformations as the result of the action of bacteria or 38 yeast enzymes 3 as in case of aroma and chemical composition being a method of food 39 processing that reduces sugar content. 4 Johnson et. al. reported that fermented berry 40 beverages have shown increased phenolic content and higher antioxidant activity than 41 their non-fermented counterparts. 5 In addition, in our previous work we found that after 42 gluconic fermentation the bioactive content remained practically unchanged. 6 Therefore 43 the study of fermented beverages deserves attention. 44 Reque et al reported that, under refrigeration conditions of storage, blueberries fruit and 45 juices present good stability with respect to antioxidant capacity, but with significant 46 anthocyanin loss, possibly due to oxidation and/or condensation reactions with other 47 phenolic compounds. 7 Glycosylated flavonols have also been reported to decrease during 48 storage, while the respective aglycones increase, indicating that enzymatic hydrolysis has 49 occurred. However, these reactions do not affect the total polyphenol content, which 50 remains constant. 8 Conversely, studies detail the individual nonanthocyanin phenolic 51 compounds required to gain deeper insight into chemical composition changes under 52 effects of different temperatures during storage. 53 The gluconic fermentation of strawberry presents advantages from a nutritional 54 perspective as Gluconobacter strains do not metabolize the fructose naturally present in 55 fruit, so it remains in the beverage as a sweetener, 6,9 while glucose is transformed into 56 gluconic acid. 57 As the gluconic fermentation is a very innovative process, few data have been reported in 58 the literature about the effects of storage on the nonanthocyanin composition of the 59 beverages produced by this mean, to the best of our knowledge. An accurate 60 characterization is required to discover how time and storage conditions [refrigeration (4 61 ºC) and room temperature (27-30 ºC)] could affect the beverages, in order to establish the 62 optimum conditions to maintain bioactive composition 63 The aims of this paper are, firstly, to establish the effects of storage for three months at 64 refrigeration temperature (4 ºC) and room temperature (27-30 ºC) on the non- 65 anthocyanin composition and antioxidant activity of an innovative gluconic-fermented 66 beverage as an alternative to take advantage of overproduction of strawberry and avoid 67 its waste and, secondly, to detail the chemical composition, to achieve complete 68 characterization of the products. 69 70 MATERIALS AND METHODS 71 Chemicals and reagents 72 The standard compounds used were acquired from Fluka Sigma-Aldrich (St. Louis, MO. 73 USA) [gallic acid, caffeic acid, p-coumaric acid, cinnamic acid, quinic acid, (-)- 74 epicatechin, (+)-catechin, chlorogenic acid, ellagic acid, (-)-epicatechin gallate, 75 kaempferol, kaempferol-3-glucoside, polydatin, apigenin, quinic acid, brevifolin, 76 protocatechuic acid, (+)-taxifolin, eriodictyol, rutin, quercetin, homovanillic acid and 77 naringenin], from Chromadex® Inc. (USA) [procyanidin B1] and from Extrasynthese 78 [luteolin, (+)-aromadendrin, phloridzin and flavonomarein]. DPPH (2,2-diphenyl-1- 79 picrylhydrazyl), AAPH (2,2´-diazo-bis-amidine-propane-dihydrochloride), Trolox (6- 80 hydroxy-2,5,7,8-tetramethylchroman-2-carboxylic acid) and dimethyl dicarbonate were 81 obtained from Sigma-Aldrich (St. Louis, MO. USA); acetonitrile and formic acid (LC 82 gradient) from Merck (Darmstadt, Germany), and methanol from Prolabo® (Obregón, 83 Mexico). 84 Samples 85 A beverage was made from strawberry by gluconic fermentation, [using strawberry pureé 86 as substrate into a biorreactor, and after few minutes (10-12) was added 125 ml of 87 inoculum of G japonicus strain E1, the end of fermentation process was established when 88 the glucose had been totally consumed] as previously reported 6 and mixed with a little 89 proportion of acetic fermented product. The strawberry vinegar was obtained by a semi- 90 continuous fermentation by Acetobacter malorum (CECT 7749), at 30 ºC, 70% O2, 13.88 91 g. 10 92 Subsequently, the product obtained was centrifuged 10 min at 1500 g and diluted with 93 soda water and dimethyl dicarbonate as a preservative to stop the fermentation process. 94 The beverage thus obtained was distributed into individual bottles and pasteurized at 90 95 ºC for 90 s then placed in the refrigerator (4 ºC) or at room temperature (about 30 ºC), 96 until samples were taken (0, 15, 30, 60 and 90 days). The pasteurization process was 97 performed to mimic the industrial process to make the beverages. The initial samples are 98 the same for both temperature conditions (R0/F0). As samples were taken, they were 99 immediately stored at -18 ºC until analysis. Table 1 details sample codes as follow: ‘F’ 100 for refrigerated samples and ‘R’ for room temperature samples, followed by a number 101 that indicates the storage days. 102 Extraction procedure 103 A sample of 30 g was extracted with 30 ml of methanol and 1% w/w of ascorbic acid and 104 then 30 min of sonication. It was centrifuged for 10 min at 1500 g. The supernatant was 105 collected and the pellet re-extracted with methanol with the same procedure. The fraction 106 was mixed, evaporated to dryness and then reconstituted in 50% (v/v) methanol. 11,12 107 High-resolution mass spectrometry (HRMS) analysis 108 The experiments were performed using a Thermo Fisher Scientific (Bremen, Germany) 109 liquid chromatography system hybrid Q-OT-qIT Mass Spectrometer (Hybrid 110 Quadropole-Orbitrap Elite Mass spectrometer). This bench-top UHPLC-MS/MS system 111 combines quadropole precursor ion selection with a high-resolution, accurate-mass 112 spectrometer. Identification was performed according to mass spectra, exact mass, 113 characteristic fragmentation and retention time. Xcalibur software (version 3.0.63) was 114 used for instrument control, data acquisition and data analysis. An Internet database of 115 accurate mass spectrometry data, mzcloud (www.mzcloud.org), Metlin, to identify 116 molecular characteristics (https://metlin.scripps.edu/index.php), Massbank 13 and 117 Respect for phytochemicals 14 were used as a reference library to identify compounds. 118 The UHPLC separation was performed using a binary gradient consisted of (A) water 119 with 0.1% formic acid and (B) methanol with 0.1% formic acid: 0.0–1.0 min 5% B, 1.0– 120 12.0 min from 5 to 100% B, 12.0–13.0 min from 100% to 5% B and 13.0–15.0 min 5% 121 B. Twenty µL of sample were injected and flow rate was 400 µL min-1. The eluate was 122 analyzed using Full MS and Data Dependent Scanning in negative mode to MS/MS 123 analysis, and with Zorbax SB-C18 2.1x100 mm 1.8 µm column. The normalized collision 124 energy of the high-collision-induced-dissociation (CID) cell was set at 35 eV for data- 125 dependent scan. Helium was employed as sheath gas (53 arbitrary units), sweep gas flow 126 3 arbitrary units and auxiliary gas flow 14 arbitrary units. Automatic gain control was 127 established as follows: 128 - Ion trap Full, SIM and MSn AGC target: 10,000.00 129 - FTMS Full AGC target (1000.00) and SIM, MSn AGC target (50,000.00). 130 Data dependent scanning comprises a full MS scan (the range was from 100 to 1,500 m/z 131 and the resolution was 30,000 FWHM), followed by a data dependent scan (resolution of 132 15,000 FWHM). 133 To quantitate nonanthocyanin phenolic compounds, we used TraceFinderTM software 134 using the corresponding aglycone when the compounds were glycosides. In case of 135 hydrolizable tannins we used ellagic acid, and procyanidin B1 to quantitate condensed 136 tannins. Calibration curves were obtained by injecting standards diluted from seven 137 different concentrations (R2 0.9995-0.9979). A triplicate was performed at each point of 138 the calibration curve. 139 Sensory analysis 140 The sensory panel gathers ten trained assessors, seven females and three males, ranging 141 from 22 to 45 years old selected and trained according to ISO 1993.15 Most of them 142 participate regularly in sensory panels of beverages and accumulates more than 100 h 143 experience in sensory analysis. Additionally, a 20 hours specific training course was 144 accomplished by every participant and consisted in ordering scales of fructose, gluconic 145 acid, acetic acid, strawberry purées and fermented beverages from strawberry. 146 The trained panel selected the descriptors that better reflects the sensorial characteristics 147 of the beverages by open discussion with the panel leader and consensus. These attributes 148 were: strawberry taste, acidity, sweetness, strawberry aroma and overall impression. 149 The panel carried out 12 triangular tests 16 to differentiate samples stored at two different 150 temperatures: room (27-30ºC) and refrigeration temperature (4ºC), and six triangular tests 151 to differentiate between time of storage. 152 Eight samples were tasted in duplicate by the sensory panel using the selected attributes 153 on a ten centimeters unstructured scale ranging from 0 (‘non-exist’) to 10 (‘very strong’), 154 and for overall impression ranging from 0 (‘dislike extremely’) to 10 (‘like extremely’) 155 using standard wine tasting dark cup. To evaluate the color changes, transparent glass 156 cups have been used. 157 Antioxidant activity 158 Oxygen radical absorbance capacity (ORAC method). 159 The method used was as reported, with some modifications: 17 100 μL of fluorescein 160 solution (45 nM) and 50 μL of AAPH (15 mM) mixed with 50 μL of sample solution or 161 Trolox. The sample solutions were made in phosphate buffer (0.022:100 v/v). 162 Fluorescence was assessed with a fluorometer (multi-detector microplate Synergy HT, 163 Biotek®) with excitation, and emission wavelengths were 485 and 528 nm, respectively. 164 Data were recorded every 5 min after addition of AAPH until they reached less than 5% 165 of the initial value. Results were calculated using the areas under the fluorescein decay 166 curves between the blank (buffer + fluorescein + AAPH) and the sample, with Trolox as 167 a calibration standard (seven different concentration solutions ranged from 0.5 to 9.5 μM) 168 anthocyanins with other phenolic compounds such as procyanidins to form colored 308 polymer pigments. 34 309 Three hydrolized tannins were quantified: an ellagitannin (bis-HHDP-glucose), a 310 gallotannin (galloylquinic acid) and brevifolin carboxylic acid; the last two increased 311 during storage time and increased more at room temperature (74–28% and 34–27%, 312 respectively). Ellagitannin started to change at 15 days, and achieved its maximum 313 decrease of 65% at room temperature (26% at 4 ºC); this decrease was associated with 314 the increase of free ellagic acid (22% at room temperature). 315 Interestingly, protocatechuic acid significantly increased after 60 days of storage (thirteen 316 times its initial value) at 30 ºC, with its concentration reaching eleven times its initial 317 value after 90 days at 4 ºC. This increase may be due to the degradation process of 318 anthocyanin compounds (pelargonidin is a principal anthocyanin of strawberry), which 319 produced a cleavage of pelargonidin B-ring and, consequently, a molecule of 320 protocatechuic acid, as was reported in the thermal processing of food. 35 Additionally, 321 as anthocyanin are susceptible to thermal degradation, another mechanism proposed 322 describes the opening of the pyrylium ring and chalcone glycoside formation. 36 323 Kaempferol, apigenin and their derivatives increased (24–373%) during storage, 324 increasing more at 30 ºC than 4 ºC. Conversely, the quercetin 3-O-glucoside and rutin 325 content decreased during storage: the higher the temperature is, the greater the decrease 326 of the content of glycosides, producing an increase of the content of quercetin (Table S1). 327 The concentrations of (+)-taxifolin 7-O-glucoside underwent a significant reduction (75% 328 at 30 ºC–51% at 4 ºC). This result is associated with the increase of taxifolin (106%– 329 41%) because of the breakdown of the glucosidic bond releasing the corresponding 330 aglycone. 8 331 The major loss of non-anthocyanin phenolic compound content was observed at room 332 temperature. This observation was in accordance with what was reported before by 333 Oliveira et al. 37 They conclude that pasteurization treatment favors the rupture of cellular 334 structures, increasing the exposure of compounds to oxidation processes during storage. 335 This phenomenon is one of the main reasons for the loss of phenolic compounds. Apart 336 from the above mentioned changes, it can be highlighted that certain (poly)phenolic 337 compounds maintain their content practically unchanged at both tested temperatures as: 338 caffeic acid, caffeic acid hexoside and naringenin. Additionally, the concentration of two 339 compounds remained constant at refrigeration conditions (phloridzin and quercetin 340 glucuronide). All these data are shown Table S1. 341 PCA was carried out to explore the effects of storage time and temperature on the non- 342 anthocyanin phenolic compound content profile (Figure 1A and 1B). The matrix had 18 343 samples x 24 quantified non-anthocyanin phenolic compounds. The two principal 344 components (Factor 1 and Factor 2) accounted for 86.54% of the variability of the original 345 data. PCA showed clustering of the samples into four main groups, illustrated in Figure 346 1A. It was observed that samples belonging to the same storage period are very close to 347 each other, which generate four well-separated clusters, thus highlighting the effect of 348 time. Interestingly, R90 and R60 were very separately located from the other samples 349 (Figure 1A). The cluster presented in the middle of the Cartesian plane, includes samples 350 R30, F60 and F90, which indicates that storage under refrigeration has a preservative effect 351 on non-anthocyanin phenolic composition. The last cluster consists of initial samples 352 (represented by F0), F15, F30 and R15 with high weights on Factor 1, as determined by their 353 high content of procyanidin B1 (17); procyanidin trimer (18) and (+)-catechin (1). In 354 addition, the Figure 1B shows the variables separated into two groups, on the right: all 355 compounds, which experiments a decrease. On the left, those which increased theirs 356 concentrations along storage time at two temperature conditions. In all cases, the effect 357 was greater at room temperature than refrigerated conditions; specifically the composition 358 remains virtually unchanged until day 15 of storage at room temperature and day 30 under 359 refrigerated conditions. 360 Influence of storage conditions in antioxidant activity. 361 Figure 2 left and right represent antioxidant activity by ORAC and DPPH, respectively. 362 No significant differences were observed for ORAC results, but DPPH results were 363 statistically different (p<0.05). At room temperature, it has been observed a progressive 364 increase until 60 days of storage and then, it decreased. At refrigeration conditions, the 365 trend was similar. 366 This result was in accordance with reported before in mulberries alcoholic fermented. 18 367 With the fermentation process hydroxycinamic acids increased while the progressive 368 growing in the content of protocatechuic acid indicated that it existed a degradation of 369 anthocyanins generating another compounds with antioxidant activity. 370 A similar behavior was reported before in the analysis of antioxidant activity in solutions 371 of phenolic compounds. 38 When the storage time was prolonged, the antioxidant activity 372 decreased and so maximum value was observed (Figure 2 right, maximum at 60 days and 373 then decreased). This increase in the overall antioxidant activity value could be probably 374 explained by the formation of oligomers from free polyphenols as (+)-catechin as its 375 content decreased as showed in Table 3. 376 Sensory analysis 377 The panel carried out 12 triangle tests to differentiate samples with different storage time 378 and 6 triangle tests to differentiate samples stored at different temperatures. In general, 379 sensory differences were perceived from 30 days of storage on, at room temperature 380 whilst the panel could not establish differences between refrigerated samples whatever 381 the storage time was. 382 On the other hand, the panel could not differentiate samples stored during 30 days at 383 different temperatures. However, differences between samples stored at 4ºC or 27-30ºC 384 for 60 could be perceived and even larger differences were perceived for samples stored 385 for 90 days. 386 The sensory profile of the beverage was built using the marks given for each attribute by 387 the panel. Figure 3 displays the spider charts for the samples stored at room temperature 388 (Figure 3A) and samples storage in refrigeration conditions (Figure 3 B). The attributes 389 (overall impression, strawberry aroma, and strawberry taste) have highest marks in 390 refrigeration conditions and the lowest marks were in samples stored at room temperature. 391 Additionally, the higher the time of storage is, the lower overall impression scores are. 392 These results should be taken into account to establish the shelf life of the beverage. 393 In conclusion, four new compounds were identified with HRMS technique in samples of 394 fermented strawberry beverage: aromadendrin hexoside, phloretin 2’-O-xylosyl 395 glucoside, dihydroferulic acid 4-O-glucuronide and kaempferol hexosil hexoside. 396 Nonanthocyanin phenolic composition underwent fewer changes under refrigeration than 397 at room temperature although 10 compounds declined when they were stored mainly at 398 room temperature. However our results show that 14 compounds: ellagic acid, ellagic 399 acid hexose, p-coumaric acid, p-coumaroylquinic acid, apigenin, apigenin pentose, 400 eriodictyol glucose, kaempferol, kaempferol glucuronide, protocatechuic acid, brevifolin 401 carboxylic acid, (+)-taxifolin, galloylquinic acid and apigenin-7-O-glucose increased 402 during storage. 403 Strawberry fermented beverage storage period should not exceed 30 days at room 404 temperature (27-30ºC) or 60 days at refrigeration conditions (4ºC). Gluconic fermentation 405 is an alternative process to prevent fruit waste while elaborating a glucose free product 406 that contains bioactive compounds. 407 Acknowledgments 408 The authors are very grateful to the Spanish Government for its financial assistance 409 (Project MICINN AGL2010-22152-01 and AGL2010-2215-03). The authors would also 410 like to thank the VPPI-US for the current ABC contract. We are also grateful to HUDISA 411 Desarrollo Industrial S.A., Lepe, Spain, for providing the strawberry purée samples. We 412 would like to thank Biology and Mass Spectrometry services (CITIUS) of the University 413 of Seville, for the multi-detector micro-plate reader and QTRAP instrument (Dr. M. 414 Carballo-Álvarez, Dr. M.E. Soria-Díaz and R. Valderrama-Fernández). 415 ASSOCIATED CONTENT 416 Supporting information 417 Table S1. Concentrations of non-anthocyanin phenolic compounds (µg/100g of 418 beverage). Table S2. Variable contribution, based on correlations. PCA analysis. Figure 419 S1. Total ion current chromatograms in ES (-) in full scan mode (FSM) of initial sample 420 (A), and R90 (B). The peaks are marked with numbers that refer to the tentative 421 identification in Table 1. Figure S2. Full scan and MS2 spectra with characteristic 422 fragmentation of Nº. 6 compound, with tentative identification of galloyl quinic acid. 423 Figure S3. MS2 spectra with a proposal fragmentation pattern characteristic of 424 propelargonidin dimer. (PDF) 425 426 References 427 (1) Oroian, M.; Escriche, I. Antioxidants: Characterization, natural sources, 428 extraction and analysis. Food Res. Int. 2015, 74, 10–36. 429 (2) FAOSTAT. FAOSTAT. 2012. 430 (3) van Boekel, M.; Fogliano, V.; Pellegrini, N.; Stanton, C.; Scholz, G.; Lalljie, S.; 431 Somoza, V.; Knorr, D.; Jasti, P. R.; Eisenbrand, G. A review on the beneficial 432 aspects of food processing. Mol. Nutr. Food Res. 2010, 54, 9, 1215–1247. 433 (4) Reis, B. A. dos; Kosińska-Cagnazzo, A.; Schmitt, R.; Andlauer, W. Fermentation 434 of Plant Material - Effect on Sugar Content and Stability of Bioactive 435 Compounds. Polish J. Food Nutr. Sci. 2014, 64, 4, 235–241. 436 (5) Johnson, M. H.; de Mejia, E. G.; Fan, J.; Lila, M. A.; Yousef, G. G. 437 Anthocyanins and proanthocyanidins from blueberry-blackberry fermented 438 beverages inhibit markers of inflammation in macrophages and carbohydrate- 439 utilizing enzymes in vitro. Mol. Nutr. Food Res. 2013, 57, 7, 1182–1197. 440 (6) Álvarez-Fernández, M. A.; Hornedo-Ortega, R.; Cerezo, A. B.; Troncoso, A. M.; 441 García-Parrilla, M. C. Non-anthocyanin phenolic compounds and antioxidant 442 activity of beverages obtained by gluconic fermentation of strawberry. Innov. 443 Food Sci. Emerg. Technol. 2014, 26, 469–481. 444 (7) Reque, P. M.; Steffens, R. S.; Jablonski, A.; Flôres, S. H.; Rios, A. de O.; de 445 Jong, E. V. Cold storage of blueberry (Vaccinium spp.) fruits and juice: 446 Anthocyanin stability and antioxidant activity. J. Food Compos. Anal. 2014, 33 447 1, 111–116. 448 (8) Bimpilas, A.; Tsimogiannis, D.; Balta-Brouma, K.; Lymperopoulou, T.; 449 Oreopoulou, V. Evolution of phenolic compounds and metal content of wine 450 during alcoholic fermentation and storage. Food Chem. 2015, 178, 164–171. 451 (9) Attwood, M. M.; van Dijken, J. P.; Pronk, J. T. Glucose metabolism and gluconic 452 acid production by Acetobacter diazotrophicus. J. Ferment. Bioeng. 1991, 72, 2, 453 101–105. 454 (10) Hidalgo, C.; Torija, M. J.; Mas, A.; Mateo, E. Effect of inoculation on strawberry 455 fermentation and acetification processes using native strains of yeast and acetic 456 acid bacteria. Food Microbiol. 2013, 34, 1, 88–94. 457 (11) Álvarez-Fernández, M. A.; Hornedo-Ortega, R.; Cerezo, A. B.; Troncoso, A. M.; 458 García-Parrilla, M. C. Effects of the strawberry (Fragaria ananassa) purée 459 elaboration process on non-anthocyanin phenolic composition and antioxidant 460 activity. Food Chem. 2014, 164, 104–112. 461 (12) Álvarez-Fernández, M. A.; Cerezo, A. B.; Cañete-Rodríguez, A. M.; Troncoso, 462 A. M.; García-Parrilla, M. C. Composition of nonanthocyanin polyphenols in 463 alcoholic-fermented strawberry products using LC-MS (QTRAP), high- 464 resolution MS (UHPLC-Orbitrap-MS), LC-DAD, and antioxidant activity. J. 465 Agric. Food Chem. 2015, 63, 7, 2041–2051. 466 (13) Horai, H.; Arita, M.; Kanaya, S.; Nihei, Y.; Ikeda, T.; Suwa, K.; Ojima, Y.; 467 Tanaka, K.; Tanaka, S.; Aoshima, K.; et al. MassBank: A public repository for 468 sharing mass spectral data for life sciences. J. Mass Spectrom. 2010, 45, 7, 703– 469 714. 470 (14) Sawada, Y.; Nakabayashi, R.; Yamada, Y.; Suzuki, M.; Sato, M.; Sakata, A.; 471 Akiyama, K.; Sakurai, T.; Matsuda, F.; Aoki, T.; et al. RIKEN tandem mass 472 spectral database (ReSpect) for phytochemicals: A plant-specific MS/MS-based 473 data resource and database. Phytochemistry 2012, 82, 38–45. 474 (15) ISO. Analysis General Guidance for the Selection Trainning and Monitoring of 475 Assessors. Part I. Selected Assessors, Standart No8586-1; International 476 Organization for Standarization: Switzerland, 1993. 477 (16) ISO. Sensory Analysis Methodology Ranking.; International Organization for 478 Standarizationd: Switzerland, 2006; p 21. 479 (17) Ou, B.; Hampsch-Woodill, M.; Prior, R. L. Development and validation of an 480 improved oxygen radical absorbance capacity assay using fluorescein as the 481 fluorescent probe. J. Agric. Food Chem. 2001, 49, 10, 4619–4626. 482 (18) Brand-Williams, W.; Cuvelier, M. E.; Berset, C. Use of a free radical method to 483 evaluate antioxidant activity. LWT - Food Sci. Technol. 1995, 28, 1, 25–30. 484 (19) Pérez-Gregorio, M. R.; Regueiro, J.; Alonso-González, E.; Pastrana-Castro, L. 485 M.; Simal-Gándara, J. Influence of alcoholic fermentation process on antioxidant 486 activity and phenolic levels from mulberries ( Morus nigra L.). LWT-Food Sci. 487 Technol. 2011, 44 (8), 1793–1801. 488 (20) StatSoft, I. Electronic statistics textbook. StatSoft, Tulsa, OK 2007. 489 (21) Del Bubba, M.; Checchini, L.; Chiuminatto, U.; Doumett, S.; Fibbi, D.; Giordani, 490 E. Liquid chromatographic/electrospray ionization tandem mass spectrometric 491 study of polyphenolic composition of four cultivars of Fragaria vesca L. berries 492 and their comparative evaluation. J. Mass Spectrom. 2012, 47, 9, 1207–1220. 493 (22) Dias, M. I.; Barros, L.; Oliveira, M. B. P. P.; Santos-Buelga, C.; Ferreira, I. C. F. 494 R. Phenolic profile and antioxidant properties of commercial and wild Fragaria 495 vesca L. roots: A comparison between hydromethanolic and aqueous extracts. 496 Ind. Crops Prod. 2015, 63, 125–132. 497 (23) Fischer, U. A.; Carle, R.; Kammerer, D. R. Identification and quantification of 498 phenolic compounds from pomegranate (Punica granatum L.) peel, mesocarp, 499 aril and differently produced juices by HPLC-DAD-ESI/MSn. Food Chem. 2011, 500 127, 2, 807–821. 501 (24) Fischer, U. A.; Carle, R.; Kammerer, D. R. Thermal stability of anthocyanins and 502 colourless phenolics in pomegranate (Punica granatum L.) juices and model 503 solutions. Food Chem. 2013, 138, 2-3, 1800–1809. 504 (25) Gu, L.; Kelm, M. A.; Hammerstone, J. F.; Beecher, G.; Holden, J.; Haytowitz, 505 D.; Prior, R. L. Screening of foods containing proanthocyanidins and their 506 structural characterization using LC-MS/MS and thiolytic degradation. J. Agric. 507 Food Chem. 2003, 51, 25, 7513–7521. 508 (26) Hanhineva, K.; Rogachev, I.; Kokko, H.; Mintz-Oron, S.; Venger, I.; 509 Kärenlampi, S.; Aharoni, A. Non-targeted analysis of spatial metabolite 510 composition in strawberry (Fragaria × ananassa) flowers. Phytochemistry 2008, 511 69, 13, 2463–2481. 512 (27) Natić, M. M.; Dabić, D. Č.; Papetti, A.; Fotirić Akšić, M. M.; Ognjanov, V.; 513 Ljubojević, M.; Tešić, Ž. L. Analysis and characterisation of phytochemicals in 514 mulberry (Morus alba L.) fruits grown in Vojvodina, North Serbia. Food Chem. 515 2015, 171, 128–136. 516 (28) Pavlović, A. V.; Dabić, D. C. .; Momirović, N. M.; Dojinović, B. P.; 517 Milojković-Opsenica, D. M.; Tešić, Z. L.; Natic, M. M. Chemical Composition 518 of Two Different Extracts of Berries Harvested in Serbia. J. Agric. Food Chem. 519 2013, 61, 4188–4194. 520 (29) Quifer-Rada, P.; Vallverdú-Queralt, A.; Martínez-Huélamo, M.; Chiva-Blanch, 521 G.; Jáuregui, O.; Estruch, R.; Lamuela-Raventós, R. A comprehensive 522 characterisation of beer polyphenols by high resolution mass spectrometry (LC- 523 ESI-LTQ-Orbitrap-MS). Food Chem. 2015, 169, 336–343. 524 (30) Sun, J.; Liu, X.; Yang, T.; Slovin, J.; Chen, P. Profiling polyphenols of two 525 diploid strawberry (Fragaria vesca) inbred lines using UHPLC-HRMSn. Food 526 Chem. 2014, 146, 289–298. 527 (31) Vrhovsek, U.; Guella, G.; Gasperotti, M.; Pojer, E.; Zancato, M.; Mattivi, F. 528 Clarifying the identity of the main ellagitannin in the fruit of the strawberry, 529 Fragaria vesca and Fragaria ananassa Duch. J. Agric. Food Chem. 2012, 60, 10, 530 2507–2516. 531 (32) Mikuli Petkovšek, M.; Štampar, F.; Veberi, R. Accumulation of phenolic 532 compounds in apple in response to infection by the scab pathogen, Venturia 533 inaequalis. Physiol. Mol. Plant Pathol. 2009, 74, 1, 60–67. 534 (33) Figueiredo-González, M.; Cancho-Grande, B.; Simal-Gándara, J.; Teixeira, N.; 535 Mateus, N.; De Freitas, V. The phenolic chemistry and spectrochemistry of red 536 sweet wine-making and oak-aging. Food Chem. 2014, 152, 522–530. 537 (34) Wojdyło, A.; Teleszko, M.; Oszmiański, J. Antioxidant property and storage 538 stability of quince juice phenolic compounds. Food Chem. 2014, 152, 261–270. 539 (35) Patras, A.; Brunton, N. P.; O’Donnell, C.; Tiwari, B. K. Effect of thermal 540 processing on anthocyanin stability in foods; mechanisms and kinetics of 541 degradation. Trends Food Sci. Technol. 2010, 21, 1, 3–11. 542 (36) Oliveira, A.; Coelho, M.; Alexandre, E. M. C.; Almeida, D. P. F.; Pintado, M. 543 Long-Term Frozen Storage and Pasteurization Effects on Strawberry 544 Polyphenols Content. Food Bioprocess Technol. 2015, 8, 9, 1838–1844. 545 (37) Oliveira, A.; Almeida, D. P. F.; Pintado, M. Changes in Phenolic Compounds 546 Table 3. Concentrations of non-anthocyanin phenolic compounds (µg/100g of beverage) R0/F0 R15 R30 R60 R90 (+)-catechin 92.95bcde ± 4.02 83.89ae ± 0.28 61.10abe ± 0.09 20.47abce ± 0.20 14.33abcd ± 0.09 ellagic acid hexose 185.73e ± 2.19 184.33e ± 14.22 206.65 ± 5.45 212.22 ± 4.41 226.05ab ± 0.0 9 ellagic acid 134.12cde ± 6.80 132.39cde ± 2.06 156.49ab ± 0.66 151.04ab ± 1.10 166.69ab ± 3.06 bis HHDP glucose 17.22bcde ± 0.43 13.47ae ± 0.11 13.35ae ± 1.13 11.80ae ± 0.55 6.04abcd ± 0.03 brevifolin carboxilic acid 1.72cde ± 0.07 1.78cde ± 0.07 2.14ab ± 0.06 2.36ab ± 0.03 2.30ab ± 0.07 galloylquinic acid 1.00de ± 0.03 1.03de ± 0.08 1.33e ± 0.17 1.65ab ± 0.13 1.72abc ± 0.03 dimer of caffeic acid-O-hexoside 7.01bcde ± 0.42 6.76a ± 0.04 6.33ab ± 0.32 5.25abc ± 0.04 5.09abc ± 0.01 p-coumaric acid 98.53cde ± 2.41 99.15cde ± 0.41 111.09abde ± 0.22 121.75abce ± 2.03 133.25abcd ± 0.07 p-coumaroyl hexose 141.51bcde ± 2.44 137.32acde ± 0.10 135.28abde ± 0.98 124.39abce ± 0.81 120.73abcd ± 0.19 p-coumaroylquinic acid 49.27de ± 1.18 48.53de ± 0.50 52.05de ± 0.18 56.65abce ± 0.82 62.06abcd ± 0.56 apigenin 0.05bcde ± 0.00 0.06acde ± 0.00 0.09abde ± 0.00 0.11abc ± 0.00 0.11abc ± 0.00 apigenin pentose 0.72c ± 0.02 0.73c ± 0.02 0.89ab ± 0.02 0.79 ± 0.00 0.84 ± 0.05 apigenin-7-O-glucose 1.42cde ± 0.03 1.41cde ± 0.01 1.64abe ± 0.04 1.76abce ± 0.01 1.98abcd ± 0.02 eriodictyol glucose 20.76cde ± 0.96 22.32cde ± 0.05 25.95abde ± 0.29 27.74abc ± 0.71 28.14abc ± 0.37 aromadendrin 52.51de ± 0.55 51.72de ± 0.56 52.12de ± 0.47 52.12abc ± 0.16 48.03abc ± 1.08 kaempferol 6.61bcde ± 0.17 12.46acde ± 0.09 18.82abde ± 2.60 22.72abce ± 0.11 24.65abcd ± 0.45 kaempferol glucuronide 18.76cde ± 0.19 18.36cde ± 0.36 20.43ab ± 1.05 20.74ab ± 0.04 20.61ab ± 0.48 procyanidin B1 63.42cde ± 3.41 62.86cde ± 0.30 44.83abde ± 0.29 4.55abc ± 0.19 3.11abc ± 0.28 procyanidin trimer 8.59bc ± 0.99 5.79ac ± 0.13 2.08ab ± 0.01 nd ± nd ± protocatechuic acid 16.10de ± 3.99 13.28de ± 2.27 21.90de ± 0.05 213.90abce ± 0.75 80.90abcd ± 1.92 quercetin 3-O-glucoside 2.65de ± 0.04 2.56de ± 0.15 2.46de ± 0.03 2.20abce ± 0.02 1.93abcd ± 0.01 rutin 2.20bcde ± 0.04 1.88acde ± 0.01 1.38abde ± 0.00 0.90abce ± 0.03 0.59abcd ± 0.04 (+)-taxifolin 0.80cde ± 0.03 0.94cde ± 0.00 1.13abde ± 0.00 1.33abce ± 0.00 1.65abcd ± 0.11 (+)-taxifolin-7-O-glucoside 8.67bcde ± 0.14 4.23ade ± 0.09 4.18ade ± 0.18 2.81abce ± 0.24 2.12abcd ± 0.08 Table 3. Continued. R0/F0 F15 F30 F60 F90 (+)-catechin 92.95bde ± 4.02 105.95a ± 1.25 95.22bc ± 0.40 60.10abc ± 0.24 53.63abcd ± 0.04 ellagic acid hexose 185.73 ± 2.19 195.60 ± 10.05 186.99 ± 1.87 193.16 ± 14.1 7 193.44 ± 9.42 ellagic acid 134.12 ± 6.80 148.59 ± 0.71 148.95 ± 1.04 146.17 ± 1.69 152.05 ± 0.28 bis-HHDP-glucose 17.22 ± 0.43 19.58 ± 0.92 17.57 ± 0.39 13.85abc ± 0.01 12.81abc ± 0.27 brevifolin carboxilic acid 1.72de ± 0.07 1.92 ± 0.10 1.86 ± 0.07 2.20ac ± 0.00 2.18ac ± 0.10 galloylquinic acid 1.00 ± 0.03 0.99 ± 0.03 1.02 ± 0.05 1.19 ± 0.04 1.27 ± 0.03 dimer of caffeic acid-O-hexoside 7.01e ± 0.42 7.45e ± 0.17 7.23 ± 0.04 6.94 ± 0.14 6.65ab ± 0.05 p-coumaric acid 98.53 ± 2.41 99.15d ± 2.16 97.37de ± 0.46 104.35bc ± 1.17 103.33c ± 0.27 p-coumaroyl hexose 141.51cde ± 2.44 141.06cde ± 0.04 137.14abde ± 0.60 134.73abc ± 0.45 132.54abc ± 0.20 p-coumaroylquinic acid 49.27 ± 1.18 48.91 ± 1.33 47.70 ± 0.25 50.49 ± 1.72 50.03 ± 1.92 apigenin 0.05e ± 0.00 0.05e ± 0.00 0.05 ± 0.00 0.05 ± 0.00 0.06ab ± 0.00 apigenin pentose 0.72 ± 0.02 0.69d ± 0.01 0.71 ± 0.01 0.83b ± 0.00 0.74 ± 0.07 apigenin-7-O-glucose 1.42 ± 0.03 1.44 ± 0.03 1.42 ± 0.01 1.50 ± 0.10 1.49 ± 0.04 eriodictyol glucose 20.76bcde ± 0.96 24.05ade ± 0.25 23.84ade ± 0.25 25.59abc ± 0.23 26.26abc ± 0.13 aromadendrin 52.51 ± 0.55 53.38 ± 0.64 53.24 ± 0.94 52.43 ± 0.06 51.92 ± 0.56 kaempferol 6.61bcde ± 0.17 9.87acde ± 0.06 11.95abde ± 0.02 16.35abce ± 0.10 19.97abcd ± 0.30 kaempferol glucuronide 18.76 ± 0.19 19.22 ± 0.36 19.81 ± 0.00 20.00 ± 0.26 19.97 ± 0.10 procyanidin B1 63.42bcde ± 3.41 74.90ade ± 0.33 72.90ade ± 0.18 40.65abce ± 0.17 23.03abcd ± 0.10 procyanidin trimer 8.59bcde ± 0.99 10.53acde ± 0.09 8.23abde ± 0.08 3.04abce ± 0.08 1.84abcd ± 0.03 protocatechuic acid 16.10e ± 3.99 14.81e ± 1.17 19.09e ± 1.21 17.55e ± 0.86 186.35abc d ± 7.99 quercetin 3-O-glucoside 2.65 ± 0.04 2.60 ± 0.04 2.56 ± 0.04 2.47 ± 0.04 2.40 ± 0.01 rutin 2.20bcde ± 0.04 1.89acde ± 0.07 1.79abde ± 0.05 1.65abce ± 0.00 1.44abcd ± 0.02 (+)-taxifolin 0.80cde ± 0.03 0.91de ± 0.01 0.99a ± 0.01 1.09ab ± 0.02 1.13ab ± 0.02 (+)-taxifolin-7-O-glucoside 8.67bcde ± 0.14 4.27a ± 0.03 4.12a ± 0.13 4.08a ± 0.05 4.21a ± 0.06 Mean values and standard deviation. A superscript letter ‘a’ indicates a significant difference (p< 0.05) between the initial samples (R0/F0) and the other samples under the same temperature conditions. A superscript letter ‘b’ indicates a significant difference (p<0.05) between the samples at day 15 (R15/F15) and the other samples under the same temperature conditions. A superscript letter ‘c’ indicates a significant difference (p<0.05) between the samples at day 30 (R30/F30) and the other samples under the same temperature conditions. A superscript letter‘d’ indicates a significant difference (p<0.05) between the samples at day 60 (R60/ F60) and the other samples under the same temperature conditions, obtained through ANOVA and TUKEY’S HSD (honest significant difference) statistical tests. No superscript letter indicates no significant differences. Samples were determined in duplicate. nd: no data Figure 1. Álvarez-Fernández et al. A Figure 2. Álvarez-Fernández et al. Figure 3. Álvarez-Fernández et al. TOC graphic