Influence of high pressure homogenization and pasteurization on the in vitro bioaccessibility of carotenoids and flavonoids in orange juice
Abstract
Production of high-quality healthy foods through sustainable methodologies is an urgent necessity. High pressure homogenization (HPH) is an interesting alternative to obtain premium citrus juices, but its effects on bioactive compounds are unclear. There was studied the influence of HPH (150 MPa) and pasteurization (92 °C for 30 s and 85 °C for 15 s) processing on physicochemical properties and in vitro bioaccessibility of carotenoids and flavonoids in orange juices. Regarding fresh juice, physicochemical properties of samples remained unchanged although cloudiness was improved by homogenization. Pasteurization did not affect total carotenoids content and retinol activity equivalents (RAE) of juices whereas homogenization yielded a significant reduction (1.37 and 1.35-fold, respectively). Interestingly, particle size reduction from homogenization drastically enhanced (about 5-fold) bioaccessibility of carotenoids including hardly bioaccessible epoxycarotenoids, finding unaltered rates in pasteurized samples. Bioaccessibility of flavonoids was constant in all cases. Results can promote HPH as an efficient option to obtain health-enhanced foods.
Full text
1 Influence of high pressure homogenization and pasteurization on the 1 in vitro bioaccessibility of carotenoids and flavonoids in orange juice 2 3 Carla M. Stincoa, Enrique Sentandreub*, Paula Mapelli-Brahma, José L. Navarrob, 4 Isabel M. Vicarioa, Antonio J. Meléndez-Martíneza 5 6 aFood Colour & Quality Lab., Dept. Nutrition & Food Science, Universidad de 7 Sevilla, 41012 Sevilla, Spain. 8 bInstituto de Agroquímica y Tecnología de Alimentos (IATA-CSIC). Ave. Agustín 9 Escardino 7, 46980 Paterna, Valencia, Spain. 10 11 12 13 *To whom correspondence should be addressed: [email protected] (Enrique 14 Sentandreu). 15 16
2 Abstract 17 Production of high-quality healthy foods through sustainable methodologies is an 18 urgent necessity. High pressure homogenization (HPH) is an interesting alternative to 19 obtain premium citrus juices, but its effects on bioactive compounds are unclear. There 20 was studied the influence of HPH (150 MPa) and pasteurization (92ºC for 30 s and 21 85ºC for 15 s) processing on physicochemical properties and in vitro bioaccessibility 22 of carotenoids and flavonoids in orange juices. Regarding fresh juice, physicochemical 23 properties of samples remained unchanged although cloudiness was improved by 24 homogenization. Pasteurization did not affect total carotenoids content and retinol 25 activity equivalents (RAE) of juices whereas homogenization yielded a significant 26 reduction (1.37 and 1.35-fold, respectively). Interestingly, particle size reduction from 27 homogenization drastically enhanced (about 5-fold) bioaccessibility of carotenoids 28 including hardly bioaccessible epoxycarotenoids, finding unaltered rates in pasteurized 29 samples. Bioaccessibility of flavonoids was constant in all cases. 30 Results can promote HPH as an efficient option to obtain health-enhanced foods. 31 32 33 Keywords: Citrus juices; carotenoids; health-promoting foods; high pressure 34 homogenization (HPH); sustainable processing; phytonutrients bioaccessibility. 35 36 Chemical compounds 37 -Z-Violaxanthin (PubChem CID: 6442428) 38 -Luteolin (PubChem CID: 5280445) 39 -Zeaxanthin (PubChem CID: 5280899) 40 -Zeinoxanthin (PubChem CID:5281234) 41
3 --Cryptoxanthin (PubChem CID: 5281235) 42 -α-Carotene (PubChem CID:4369188) 43 --Carotene (PubChem CID:5280489) 44 -Phytoene (PubChem CID:5280784) 45 -Phytofluene (PubChem CID: 6436722) 46 -Narirutin (PubChem CID: 442431) 47
4 1. Introduction 48 Orange juice is the most consumed fruit juice worldwide greatly appreciated by 49 consumers as a delicious natural source of vitamin C, fibre and nutritionally relevant 50 phytochemicals (mainly carotenoids and flavonoids). Epidemiological studies reported 51 that a diet rich in fruit phytochemicals prevents apparition of degenerative diseases, 52 finding that in the particular case of citrus carotenoids and flavonoids such benefits 53 include the protection against oxidizing agents, inflammation and modulation of gene 54 expression (Meléndez‐Martínez, 2019; Tripoli, Guardia, Giammanco, Majo, & 55 Giammanco, 2007). Health effects of such phytochemicals depend on the amount 56 consumed and their bioavailability. Intrinsic and extrinsic factors such as the 57 metabolite profile of citrus varieties, human metabolism and food/human genetics, 58 interactions within and between food matrices and industrial processing greatly affect 59 bioaccessibility of these bioactive compounds (Rodríguez-Concepcion et al., 2018). In 60 this line, bioaccessibility is considered a measure of potential bioavailability and can 61 be defined as the amount of a compound released from the food matrix during 62 digestion and potentially available for absorption (Rodríguez-Concepcion et al., 2018). 63 Interest of in vitro digestion methods to assess the bioaccessibility of food 64 phytochemicals is undeniable since they can facilitate the understanding of complex 65 interactions naturally occurring under in vivo conditions, facilitating the design of 66 commercial products with improved nutritional properties (Etcheverry, Grusak, & 67 Fleige, 2012). 68 Nowadays, consumers demand minimally altered and more natural products hence the 69 interest of the industry to implement innovative non-thermal stabilization technologies 70 such as dynamic high pressure homogenization (HPH) (Martínez-Monteagudo, Yan, 71 & Balasubramaniam, 2017). Similarly, there is an increasing interest to implement 72
5 innovative and environmentally friendly technologies for the sustainable production of 73 foods alternatively to traditional stabilization processes mainly ruled by thermal 74 treatments (Pereira & Vicente, 2010). Compared to heat processing, a significant 75 reduction of energy (about 15%) was stated for HPH preservation of milk (Valsasina 76 et al., 2017). Moreover, homogenization was also proposed to valorise biomass from 77 different sources such as milk whey and pistachio shell, among others (Marciniak, 78 Suwal, Britten, Pouliot, & Doyen, 2018; Özbek, Fockink, Yanık, Göğüş, & Lukasik, 79 2018). 80 81 The effect of processing on bioaccessibility of carotenoids and flavonoids from 82 foodstuffs seems to be intimately associated to the severity of treatments applied and 83 the food matrix composition (Mapelli-Brahm, Stinco, Rodrigo, Zacarías, & Meléndez-84 Martínez, 2018; Stinco, Fernández-Vázquez, Escudero-Gilete, Heredia, Meléndez-85 Martínez, & Vicario, 2012). Regarding the influence exerted by minimally processing 86 HPH treatment on in vitro bioaccessibility of carotenoids from foodstuffs, studies were 87 mainly addressed to tomato and tomato-based products with dissenting results 88 (Knockaert et al., 2012; Svelander et al., 2011). Disparate results can be understood by 89 factors outside those induced by the matrix disruption caused by HPH such as the 90 carotenoid type (Panozzo et al., 2013), rheological properties modified by processing 91 (Leite, Augusto, & Cristianini, 2014), the chromoplast substructures and cell wall 92 barriers (Palmero, Panozzo, Colle, Chigwedere, Hendrickx, & Van Loey, 2016). This 93 issue was assessed in some beverages made from the combination of fruit juices with 94 water, milk and soymilk suggesting how in vitro bioaccessibility of carotenoids was 95 generally increased by HPH and High Intensity Pulsed Electric fields (HIPEF) 96 compared to traditional thermal treatment (Rodríguez-Roque et al., 2016). To the best 97
6 of our knowledge, no studies were carried out to evaluate the effect of HPH processing 98 on in vitro bioaccessibility of carotenoids from orange juice. 99 Considering phenolics from orange juice, HPH processing does not seem to affect 100 their in vitro bioaccessibility (He et al., 2016). Moreover, human studies suggested 101 that in vivo bioavailability of orange flavanones was mainly ruled by individual factors 102 rather than technological treatments (Tomás-Navarro et al., 2014). 103 Main objective of this study was to assess the effect of HPH and a traditional 104 pasteurization on the in vitro bioaccessibility of carotenoids and flavonoids of Lane 105 Late orange juices. Furthermore, influence of treatments on physicochemical 106 properties of juices was also studied. Results achieved can promote the 107 implementation of more sustainable methodologies by food industry to produce high-108 quality health-promoting products to cater for the new trends in the demands of 109 consumers. 110 111 2. Experimental 112 2.1. Materials 113 Analytical-grade dichloromethane, acetone and methanol were from Carlo Erba 114 (CARLO ERBA Reagents S.r.l. Milan, Italy). HPLC-grade acetonitrile (ACN) and 115 formic acid (FA) were from Scharlab (Scharlab S.L., Barcelona, Spain). Methanol 116 (MeOH) and methyl tert-butyl ether (MTBE) of HPLC-grade were from Merck 117 (Merck KGaA, Darmstadt, Germany). Ultra-pure water was produced by a 118 NANOpureDiamond system (Barnsted Inc.). Mineral salts (KCl, NaCl), sodium 119 bicarbonate, monopotassium-phosphate, magnesium chloride hexahydrate, chlorhydric 120 acid, ascorbic acid (L-AAH, 99% purity grade), ammonium formate, pepsin (porcine 121 gastric mucosa), pancreatin (porcine pancreas), bile salt, α-carotene, β-carotene, β-122
7 cryptoxanthin, lutein and zeaxanthin were from Sigma-Aldrich (Merck KGaA, 123 Darmstadt, Germany). Phytoene and Phytofluene standards were extracted from 124 tomato fruits and purified following the procedure of Mapelli-Brahm, Corte-Real, 125 Meléndez-Martínez and Bohn (2017). 126 127 2.2. Preparation of orange juice assayed 128 Treatments assayed were decided on the basis of results from previous studies 129 addressing the obtaining of high quality orange juices with extended shelf life through 130 HPH (Carbonell, Navarro, Izquierdo, & Sentandreu, 2013) and pasteurization 131 (Sentandreu, Carbonell, Carbonell, & Izquierdo, 2005) processing. 132 Lane Late (Citrus sinensis L. Osb.) orange fruits were harvested in Lliria (Valencia, 133 Spain), collected in May 2017, washed with tap water, drained, sized and finally 134 squeezed in an Exzel industrial extractor from Luzzysa (Luzzysa, El Puig, Valencia, 135 Spain). A paddle finisher (model EPF 06, Luzzysa, El Puig, Valencia, Spain) with 136 holes of 0.4 mm Ø sieved raw juice to obtain 600 L of freshly prepared juice (FJ) that 137 was divided into four different batches (Fig. S1). One batch was homogenized at 150 138 MPa in a continuous high pressure system (NS3015H model, GEA Niro Soavi S.p.A., 139 Parma, Italy) with a digital thermometer (probe PT-100, range -10/110 ºC) installed in 140 the outlet section to monitor the temperatures reached by the samples. Tap water was 141 used to warm up the HPH system for 30 min operating at 150 MPa at a flow rate of 142 100 L/h. Tempered juice substituted water for 5 min to reach the stationary conditions 143 of 15 s of residence time and 68 ºC at the outlet of the HPH device. Then, the juice 144 was quickly cooled at 7–9 ºC in a plate heat exchanger (Junior model, APV Ibérica 145 S.A., Madrid, Spain) to obtain the H150 sample. A second batch was centrifuged in a 146 Westfalia system (model SAOH 205, GEA Westfalia, Nuremberg, Germany) to 147
8 separate low pulp (serum) and pulpy (Pu) fractions. The pulpy fraction was 148 homogenized at 150 MPa under the same conditions described above and blended with 149 serum in the appropriate 85/15 proportion to reconstitute the juice that was finally 150 pasteurized at 85 ºC for 15 s to obtain sample CHPuRP8515. Relevance of the 151 CHPuRP8515 sample concerns the capacity of the production line of a juice plant in 152 where the flow rate (volume/time) of reconstituted HPH juice is much higher than the 153 achieved by homogenization of the whole juice (H150). 154 Finally, FJ was pasteurized under strong (92 ºC for 30 s) and mild (85 ºC for 15 s) 155 thermal conditions to yield P9230 and P8515 samples, respectively. 156 157 2.3. Physicochemical parameters 158 2.3.1. Total soluble solids and acidity analyses 159 Total soluble solids were determined as ºBrix with a digital refractometer (Pal-1, 160 Atago Co. Ltd, Tokyo, Japan) and total acidity was assessed by titration with 0.1 N 161 NaOH (using phenolphthalein 0.1% in 20% EtOH) and expressed as % citric acid 162 (w/v). The Brix to acid ratio and pH of fresh juice were 13.8 and 14.9, respectively 163 164 2.3.2. Suspended pulp and cloudiness assessment 165 Determinations followed the methodology proposed by Cheng (2002) and samples 166 were analysed in triplicate expressing the results as average (mean ± Std. dev.). 167 168 2.3.3. HPLC-DAD analysis of ascorbic acid 169 It was determined using a liquid chromatograph-diode array (HPLC-DAD) detector 170 Agilent 1100 (Agilent Technologies, Palo Alto, CA, USA) equipped with a binary 171 pump, vacuum degasser and temperature-controlled autosampler. Separation was 172
9 achieved on a 250 mm x 4.6 mm i.d., 5 µm Phenomenex Luna (2) C18 column from 173 Phenomenex Inc. (Torrance, CA, USA) with the following conditions: solvent A, 174 water/ACN (95:5) containing 0.1% FA and ammonium formate (25 mM); solvent B, 175 ACN; initially 0% B hold for 10 min, 90% B at 10.5 min and purging up to 20 min, 176 column equilibration with 0% B from 20.5 min up to 35 min; injection volume, 3 µL; 177 flow rate, 1 mL/min; autosampler and column temperatures were 10 ºC and 23 ºC, 178 respectively; DAD detection wavelength, λ = 260 nm. Quantification of L-AAH in 179 samples was through external calibration using a preliminary calibration curve of a 180 commercial standard solution with a correlation coefficient (R2) of 0.999. 181 Samples were analysed in triplicate and results expressed as average (mean ± Std. 182 dev.). 183 184 2.3.4. Pectin methyl esterase (PME) activity assay 185 It was determined following the methodology proposed by Carbonell, Navarro, 186 Izquierdo and Sentandreu (2013) and samples were analysed in triplicate expressing 187 results as average (mean ± Std. dev.). 188 189 2.3.5 Particle size distribution analysis 190 It was determined using a Malvern Mastersizer 2000 system (Malvern Instruments 191 Limited, Worcestershire, UK) with a short wavelength blue light source with forward 192 and backscatter detection for enhanced performance in the 0.02–2000 lm range. 193 Values at 1.73 and 1.33 were used as refractive indexes of the juice and the dispersant 194 (water), respectively, using 0.1 for absorption index of cloud particles (Corredig, Kerr, 195 & Wicker, 2001). The equivalent volume mean diameter D[4,3] was calculated as 196 follows: 197
16 Homogenization decreased the total carotenoid content and RAE significantly (1.37 346 and 1.35-fold, that is, 27% and 26%, respectively) compared to FJ. Similarly, Suárez-347 Jacobo, Rüfer, Gervilla, Guamis, Roig-Sagués and Saldo (2011) described a decrease 348 in the vitamin A content of 18%, 20% and 33% when increasing the homogenization 349 pressure at 100, 200 and 300 MPa. These results suggested that degradation of 350 carotenoids could be mainly due to the cellular disruption promoted by HPH 351 treatments that enhanced their exposure to gastric enzymes and acidic pH, decreasing 352 their stability and favouring their oxidation (Meléndez-Martínez, Vicario, & Heredia, 353 2007). As expected, such negative effects were more pronounced in CHPuRP8515 354 sample since both HPH and pasteurization treatments were applied and led to a 355 reduction of 2.63- and 2.12-fold of its total carotenoid content and RAE, respectively. 356 357 3.3. Bioaccessibility of carotenoids from juices 358 Fig. 1 and Table 2 summarize results achieved regarding bioaccessibility of juice 359 carotenoids in percentage and the theoretical amount of carotenoids that could be 360 incorporated into micelles after the ingestion of 1L of juice (referred to as carotenoid 361 bioaccessible content, CBC), respectively. 362 Comparable results were found for FJ and pasteurized juices with the exception of β-363 carotene, (Z)-antheraxanthin isomers and (Z)-luteoxanthin isomers that had a higher 364 bioaccessibility by thermal treatment (Fig. 1). These findings followed, in general, 365 previous results about bioaccessibility of carotenoids from orange juices and 366 discrepancies can be explained through the different experimental conditions 367 considered (Mapelli-Brahm et al., 2018; Stinco, Fernández-Vázquez, Escudero-Gilete, 368 Heredia, Meléndez-Martínez, & Vicario,2012). 369
17 Very interestingly, homogenization in H150 sample had a very significant (p<0.05) 370 positive impact on bioaccessibility (about 80%) and CBC (from 3- to 5-fold) of 371 carotenoids lutein, zeaxanthin, β-cryptoxanthin, α-carotene, β-carotene, lycopene, 372 phytoene and phytofluene compared to FJ (Fig. 1 and Table 2). In this line, RAE 373 activity of H150 sample had a 4.4-fold increase and suggested the relevance of HPH 374 processing to improve the vitamin A status of populations with high consumption of 375 orange juice. Epoxycarotenoids were also found in the micellar fractions from digested 376 H150 sample having a 2- (violaxanthin isomers) to 7.4-fold ((9Z)-violaxanthin + Z-377 isomer of antheraxanthin) CBC enhancement with a 3.86-fold increasing of total 378 carotenoid content compared to FJ (Table 2). Relevance of this finding deserves to be 379 highlighted since bioavailability of epoxycarotenoids used to be considered negligible 380 in humans (Meléndez-Martínez et al., 2017). Similarly, bioaccessibility and CBC of 381 carotenoids in CHPuRP8515 sample were significantly (p<0.05) higher than in FJ and 382 pasteurized juices, but lower than that of H150 juice as consequence of the 383 complementary pasteurization treatment assayed (Fig. 1 and Table 2). These firstly 384 reported findings in orange juice demonstrated the relevance of disruptive processes to 385 enhance in vitro bioaccessibility of pulp-related phytochemicals such as carotenoids 386 initially hypothesized by Sánchez-Moreno, Plaza, De Ancos, and Cano (2003) in the 387 same food matrix. 388 Since homogenization had a negative effect on carotenoids content from samples 389 assayed (Table 1), it is plausible to confer betterment of bioaccessibility and CBC of 390 carotenoids to changes induced by processing on size and structure of pulp particles of 391 juices. Results suggested that particle size reduction featured by homogenization 392 (Table S1) greatly increased SSA and boosted extractability of carotenoids. To assess 393 correlations between particle size parameters (D[4,3], D[3, 2] and SSA) and the 394
18 bioaccessibility of carotenoids, simple linear regression analyses were carried out 395 (Table S2). Good correlations between the particle size parameters and 396 bioaccessibility of all carotenoids were found when all the samples were considered 397 altogether. For the correlations between D[4,3] and D[3,2] and bioaccessibility, the 398 correlation coefficients were negative and significant for all cases (r ranging from 0.62 399 to 0.98), while the correlations between SSA and bioaccessibility were positive and 400 significant in all cases (r ranging from 0.55 to 0.86) except for bioaccessibility of α-401 carotene. According to the signs of the correlation coefficients, the lower the particle 402 size, (ie. lower D[4,3] and D[3,2]), the higher the bioaccessibility of carotenoids. On the 403 contrary, the bioaccessibility was directly proportional to SSA. These results were to 404 be expected, since D[4,3] and D[3,2] indicates the diameter of the average volume and 405 the diameter of the average area of a particle, respectively , SSA is related to the 406 density of each particle. Altogether, it can be argued that the reduction in the size of 407 the particles and the subsequent increase in SSA facilitates the release of carotenoids 408 during digestion. Thus, treatments having greater impact on particle size 409 (CHPuRP8515 and H150) can confer better bioaccessibility to carotenoids than non-410 disruptive thermal treatments. As it can be observed in Fig. S2, processing of 411 CHPuRP8515 sample yielded particles of lower size than that shown in H150 sample 412 although, in general, the volume in percentage of particles in CHPuRP8515 juice was 413 lower, explaining to a large extent the differences in bioaccessibility observed between 414 HPH samples assayed. 415 Results suggested that particle size of the food matrix plays a determinant role in the 416 bioaccessibility in of carotenoids. Importantly, the net result of treatments on the 417 actual amount of carotenoids that can be absorbed (referred to as CBC in this work) 418 also depends on how such treatments affect carotenoid retention in juices. In other 419
19 words, technological treatments that lead to an increased release of carotenoids during 420 digestion can result in an actual increase or decrease CBC depending on how the 421 treatments affected the total content of the carotenoids in the matrix. 422 423 3.4. Flavonoid content of juices 424 Table 3 shows the mean levels of flavonoids found in juices assayed. Results from 425 H150 and CHPuMP8515 samples were significantly (p<0.05) lower (about 18%) than 426 those reached by fresh and pasteurized juices. Controversial results were stated by 427 Velázquez-Estrada et al. (2013) that observed an increase in the flavonoid content 428 (hesperidin and kaempferol) of orange juices treated at 200 and 300 MPa. The total 429 flavonoid content determined in P9230 and P8515 samples were not significantly 430 affected, in agreement to previous results from Sentandreu, Navarro and Sendra (2007) 431 reporting that thermal pasteurization at 90ºC-30 s had a negligible effect on the 432 phenolic content of citrus juices. 433 434 3.5. Bioaccessibility of flavonoids from juices 435 As shown by Fig. 2, bioaccessibility of flavonoids from the juices assayed was in the 436 range of 3.1% to 11.2%, lower than the bioaccessibility of carotenoids. Overall, 437 bioaccessibility of flavones (Fig. 2A) and flavanones (Fig. 2B) were not significantly 438 affected by the pasteurization treatments. In the same way, HPH treatment at 150 MPa 439 did not induce significant differences in the bioaccessibility of flavonoids.These 440 results were in agreement with those reported by He et al. (2016) who found that the 441 bioaccessibility of phenolic compounds was not affected in orange juices treated with 442 HPH relative to the untreated juice. It was noticeable how sample CHPuMP8515 443 showed an increased bioaccessibility of total flavones (Fig. 2A) and flavanones (Fig. 444
20 2B) of 2.5 and 3.3%, respectively, compared to FJ. Bioaccessibility of individual 445 flavonoids were significantly higher than in pasteurized and fresh juices. Findings 446 suggested that after pulp homogenization, these compounds were not protected by cell 447 wall components that could improve extractability of flavonoids, and therefore, their 448 bioaccessibility. Several studies have reported that the processing of plant foods can 449 influence the bioaccessibility of phenolic compounds, mainly through changes in the 450 cell wall structure (Dutra et al., 2017). Likewise, Martínez-Huélamo et al. (2015) 451 suggested that mechanical and thermal treatments of tomato sauce increased 452 bioaccessibility of phenolics. 453 As far as flavonoid bioaccessible contents (FBC) (that is, the amount of flavonoid 454 present in the digestates per liter of juice) the value corresponding to total flavonoids 455 was not significantly affected by the pasteurization (Table 4). It was observed a slight, 456 but not significant, increase in juices pasteurized at 92 ºC. Regarding H150 sample, 457 only vicenin-2 was significantly increased compared with untreated juices, as a result 458 of which the bioaccessibility of total flavones was significantly higher too. However, 459 CHPuRP8515 sample exhibited a significant increase in the bioaccessibility of 460 flavanones narigin-d and nariturin compared to FJ. Bioaccessibility of total flavonoids 461 in this juice, albeit not significantly different, was 1.32-fold higher compared to 462 reference (Table 4). 463 464 4. Conclusions 465 Although HPH processing (150 MPa) altered colour of juices, no negative effects were 466 found regarding their ascorbic acid content finding in addition, how the particle size 467 reduction induced by homogenization improved cloudiness of samples. Moreover, 468 diminution of the particle size drastically increased (from 4 to 5-fold) bioaccessibility 469
21 and RAE value of carotenoids from homogenized juices. Very interestingly, 470 epoxycarotenoids, which are not perceived in human plasma or tissues, were readily 471 detected in the micellar fraction of digested homogenized samples. Thus, there was 472 evidenced the relevance of HPH processing to improve incorporation into organism of 473 pulp-linked bioactive nutrients such as carotenoids from natural extracts. 474 In contrast to traditional pasteurization treatments, results of this study can help food 475 industry to adopt greener HPH stabilization technologies for trading high-quality 476 products with enhanced health properties that can satisfy current trends demanded by 477 consumers. Reduction of costs from the use of energy saving homogenization 478 alternatives in addition to their high output rate can favour implementation of HPH 479 processing in a wide range of industrial applications. 480 481 Acknowledgements 482 Authors thank funding from the Consejería de Economía, Innovación, Ciencia y 483 Empleo, Junta de Andalucía (projects CAROTINCO-P12-AGR-1287 and P11-AGR-484 7783). Project RTA2014-00034-C04 from the Spanish Instituto Nacional de 485 Investigación y Tecnología Agraria supported contract of E. Sentandreu. Quality 486 assistance from the technical staff of the Service of Biology (SGI, Universidad de 487 Sevilla) is also acknowledged. 488 489 Declaration of competing interest 490 Authors declare no conflict of interest. 491 492 References 493 Arena, E., Fallico, B., & Maccarone, E. (2000). Influence of Carotenoids and Pulps on 494
22 the Color Modification of Blood Orange Juice. Journal of Food Science, 65(3), 495 458–460. Retrieved from http://dx.doi.org/10.1111/j.1365-2621.2000.tb16027.x 496 Aschoff, J. K., Kaufmann, S., Kalkan, O., Neidhart, S., Carle, R., & Schweiggert, R. 497 M. (2015). In vitro bioaccessibility of carotenoids, flavonoids, and vitamin C 498 from differently processed oranges and orange juices [ Citrus sinensis (L.) 499 osbeck]. Journal of Agricultural and Food Chemistry, 63(2), 578–587. 500 https://doi.org/10.1021/jf505297t 501 Carbonell, J. V., Navarro, J. L., Izquierdo, L., & Sentandreu, E. (2013). Influence of 502 high pressure homogenization and pulp reduction on residual 503 pectinmethylesterase activity, cloud stability and acceptability of Lane Late 504 orange juice: A study to obtain high quality orange juice with extended shelf life. 505 Journal of Food Engineering, 119(3), 696–700. 506 https://doi.org/10.1016/J.JFOODENG.2013.06.041 507 Cerdán-Calero, M., Izquierdo, L., & Sentandreu, E. (2013). Valencia Late orange juice 508 preserved by pulp reduction and high pressure homogenization: Sensory quality 509 and gas chromatography-mass spectrometry analysis of volatiles. LWT - Food 510 Science and Technology, 51(2), 476–483. 511 https://doi.org/10.1016/j.lwt.2012.11.016 512 Cheng, G. (2002). Procedures for analysis of citrus products. Florida: Citrus Systems, 513 fourth ed. FMC Technologies Inc. 514 Corredig, M., Kerr, W., & Wicker, L. (2001). Particle Size Distribution of Orange 515 Juice Cloud after Addition of Sensitized Pectin. Journal of Agricultural and Food 516 Chemistry, 49(5), 2523–2526. Retrieved from 517 http://dx.doi.org/10.1021/jf001087a 518 Dutra, R. L. T., Dantas, A. M., Marques, D. de A., Batista, J. D. F., Meireles, B. R. L. 519
23 de A., de Magalhães Cordeiro, Â. M. T., … Borges, G. da S. C. (2017). 520 Bioaccessibility and antioxidant activity of phenolic compounds in frozen pulps 521 of Brazilian exotic fruits exposed to simulated gastrointestinal conditions. Food 522 Research International, 100, 650–657. 523 https://doi.org/10.1016/J.FOODRES.2017.07.047 524 Etcheverry, P., Grusak, M. A., & Fleige, L. E. (2012). Application of in vitro 525 bioaccessibility and bioavailability methods for calcium, carotenoids, folate, iron, 526 magnesium, polyphenols, zinc, and vitamins B(6), B(12), D, and E. Frontiers in 527 Physiology, 3(August), 317. https://doi.org/10.3389/fphys.2012.00317 528 Fernández-Vázquez, R., Stinco, C. M., Hernanz, D., Heredia, F. J., & Vicario, I. M. 529 (2013). Colour training and colour differences thresholds in orange juice. Food 530 Quality and Preference, 30(2), 320–327. 531 https://doi.org/10.1016/j.foodqual.2013.05.018 532 He, Z., Tao, Y., Zeng, M., Zhang, S., Tao, G., Qin, F., & Chen, J. (2016). High 533 pressure homogenization processing, thermal treatment and milk matrix affect in 534 vitro bioaccessibility of phenolics in apple, grape and orange juice to different 535 extents. Food Chemistry, 200, 107–116. 536 https://doi.org/10.1016/j.foodchem.2016.01.045 537 Institute of Medicine. Dietary reference intakes for vitamin A, vitamin K, arsenic, 538 boron, chromium, copper, iodine, iron, manganese, molybdenum, nickel, silicon, 539 vanadium, and zinc. (2001). Washington, D.C. Academy Press. 540 Knockaert, G., Pulissery, S. K., Colle, I., Van Buggenhout, S., Hendrickx, M., & 541 Loey, A. V. (2012). Lycopene degradation, isomerization and in vitro 542 bioaccessibility in high pressure homogenized tomato puree containing oil: Effect 543 of additional thermal and high pressure processing. Food Chemistry, 135(3), 544
24 1290–1297. https://doi.org/10.1016/j.foodchem.2012.05.065 545 Leite, T. S., Augusto, P. E. D., & Cristianini, M. (2014). The use of high pressure 546 homogenization (HPH) to reduce consistency of concentrated orange juice (COJ). 547 Innovative Food Science & Emerging Technologies, 26, 124–133. 548 https://doi.org/10.1016/J.IFSET.2014.08.005 549 Mapelli-Brahm, P., Corte-Real, J., Meléndez-Martínez, A. J., & Bohn, T. (2017). 550 Bioaccessibility of phytoene and phytofluene is superior to other carotenoids 551 from selected fruit and vegetable juices. Food Chemistry, 229, 304–311. 552 https://doi.org/10.1016/J.FOODCHEM.2017.02.074 553 Mapelli-Brahm, P., Stinco, C. M., Rodrigo, M. J., Zacarías, L., & Meléndez-Martínez, 554 A. J. (2018). Impact of thermal treatments on the bioaccessibility of phytoene and 555 phytofluene in relation to changes in the microstructure and size of orange juice 556 particles. Journal of Functional Foods, 46(February), 38–47. 557 https://doi.org/10.1016/j.jff.2018.04.044 558 Marciniak, A., Suwal, S., Britten, M., Pouliot, Y., & Doyen, A. (2018). The use of 559 high hydrostatic pressure to modulate milk protein interactions for the production 560 of an alpha-lactalbumin enriched-fraction. Green Chemistry, 20(2), 515–524. 561 https://doi.org/10.1039/c7gc03428h 562 Martínez-Huélamo, M., Tulipani, S., Estruch, R., Escribano, E., Illán, M., Corella, D., 563 & Lamuela-Raventós, R. M. (2015). The tomato sauce making process affects the 564 bioaccessibility and bioavailability of tomato phenolics: A pharmacokinetic 565 study. Food Chemistry, 173, 864–872. 566 https://doi.org/10.1016/j.foodchem.2014.09.156 567 Martínez-Monteagudo, S. I., Yan, B., & Balasubramaniam, V. M. (2017). Engineering 568 Process Characterization of High-Pressure Homogenization—from Laboratory to 569
25 Industrial Scale. Food Engineering Reviews, 9(3), 143–169. 570 https://doi.org/10.1007/s12393-016-9151-5 571 Meléndez-Martínez, A. J., Pérez-Gálvez, A., Roca, M., Estévez-Santiago, R., 572 Olmedilla-Alonso, B., Mercadante, A. Z., & Ornelas-Paz, J. de J. (2017). 573 Biodisponibilidad de carotenoides, factores que la determinan y métodos de 574 estimación. In A. J. Meléndez-Martínez (Ed.), Carotenoides En 575 Agroalimentación Y Salud. (pp. 574–608). Ciudad de México, México: Editorial 576 Terracota. 577 Meléndez-Martínez, A. J., Vicario, I. M., & Heredia, F. J. (2007). Review: Analysis of 578 carotenoids in orange juice. Journal of Food Composition and Analysis, 20(7), 579 638–649. https://doi.org/10.1016/j.jfca.2007.04.006 580 Meléndez‐Martínez, A. J. (2019). An Overview of Carotenoids, Apocarotenoids and 581 Vitamin A in Agro‐Food, Nutrition, Health and Disease. Molecular Nutrition & 582 Food Research, 1801045, 1801045. https://doi.org/10.1002/mnfr.201801045 583 Minekus, M., Alminger, M., Alvito, P., Ballance, S., Bohn, T., Bourlieu, C., … 584 Brodkorb, A. (2014). A standardised static in vitro digestion method suitable for 585 food - an international consensus. Food & Function, 5(6), 1113–24. 586 https://doi.org/10.1039/c3fo60702j 587 Özbek, H. N., Fockink, D. H., Yanık, D. K., Göğüş, F., & Lukasik, R. (2018). The 588 green biorefinery concept for the valorisation of pistachio shell by high-pressure 589 CO 2 /H 2 O system. Journal of Cleaner Production, 196, 842–851. 590 https://doi.org/10.1016/j.jclepro.2018.06.062 591 Palmero, P., Panozzo, A., Colle, I., Chigwedere, C., Hendrickx, M., & Van Loey, A. 592 (2016). Role of structural barriers for carotenoid bioaccessibility upon high 593 pressure homogenization. Food Chemistry, 199, 423–432. 594
Table2.CarotenoidBioaccesibleContent(CBC,theoreticalamountofcarotenoidsinmgthat couldbeincorporatedintomicellesaftertheingestionof1Lofjuice)ofjuicesassayedΨ. Carotenoids Carotenoid Bioaccesible Content FJ P9230 P8515 H150 CHPuMP8515 (Z)‐ANTisomers 0.33±0.03a 0.42±0.07a 0.40±0.03a 0.89±0.03b 0.43±0.07a all‐(E)‐VIO+(Z)‐VIOisomers 0.09±0.01a 0.09±0.02a 0.11±0.01a 0.18±0.02b 0.08±0.01a (Z)‐LUTEOisomer 0.15±0.02a 0.19±0.04a 0.16±0.01a 0.51±0.05b 0.22±0.02a (9Z)‐VIO+(Z)‐ANTisomer 0.24±0.06a 0.33±0.07a 0.23±0.06a 1.77±0.09b 0.78±0.10c (Z)‐LUTEOisomer 0.06±0.01a 0.07±0.02ab 0.06±0.01a 0.23±0.03c 0.11±0.01b MUT‐epimerA 0.10±0.01a 0.13±0.03a 0.11±0.01a 0.32±0.03b 0.15±0.01a LUT 0.11±0.01a 0.14±0.03ab 0.12±0.01a 0.37±0.02c 0.17±0.02b MUT‐epimerB 0.23±0.01a 0.25±0.04a 0.25±0.04a 0.69±0.05b 0.34±0.01c ZEA 0.16±0.01a 0.19±0.06a 0.16±0.03a 0.49±0.02b 0.24±0.02a (9Z)‐or(9´Z)‐ANT 0.15±0.02a 0.17±0.06a 0.15±0.01a 0.77±0.02b 0.36±0.05c ZEI 0.04±0.01a 0.05±0.01a 0.04±0.01a 0.18±0.01b 0.09±0.01c BCR 0.13±0.02a 0.17±0.03a 0.15±0.01a 0.56±0.01b 0.25±0.03c ACAR 0.01±0.01a 0.02±0.01a 0.02±0.01a 0.06±0.01b 0.02±0.01a BCAR 0.03±0.01a 0.07±0.03ab 0.05±0.01ab 0.140±0.003c 0.07±0.01b PF 0.02±0.01a 0.02±0.01ab 0.02±0.01ab 0.08±0.01c 0.03±0.01b PT 0.07±0.01a 0.09±0.02a 0.08±0.01a 0.29±0.01b 0.13±0.02c TotalCarotenoids 1.95±0.16a 2.41±0.52a 2.11±0.10a 7.52±0.32b 3.48±0.36c RAE* 8.48±1.29a 13.44±3.68ab 11.19±1.20a 37.33±0.22c 17.20±1.72b ΨFJ,freshjuice;P9230,pasteurizedjuiceat92ºC30s;P8515,pasteurizedjuiceat85ºC15s;H20,homogenized juiceat20MPa;H150,homogenizedjuiceat150MPa;CHPuRP8515,centrifugedjuice,pulpfractionhomogenized at150MPa,reconstitutedwiththeserumfractionandpasteurizedat85ºC15s. *Retinolactivityequivalent:calculatedasRAE(µg/L)=(μgof β‐carotene)/12+(μg β‐cryptoxanthin+μ αcarotene)/24. Differentlowercaselettersinthesamerowshowsignificantdifferences(p<0.05)amongsamples. Nomenclatureusedforcarotenoids:ANT=(antheraxanthin),VIO=(violaxanthin),LUTEO=(luteoxanthin),MUT= (mutatoxanthin),LUT=(lutein),ZEA=(zeaxanthin),ZEINO=(zeinoxanthin), BCR = (β‐cryptoxanthin), ACAR= (αcarotene),BCAR=(β‐carotene),PT=(phytoene),PF=(phytofluene)
Table3.Flavonoidlevels(mg/L)found in juices assayedΨ. Flavonoids Orange juice samples FJ P9230 P8515 H150 CHPuMP8515 Vicenin‐2 10.07±0.19a 10.27±0.13ab 10.256±0.12ab 10.57±0.12bc 10.91±0.26c Apigenin‐d 7.82±0.04a 7.64±0.13a 7.89±0.208a 7.17±0.20b 6.08±0.10c ∑Flavones 18.65±0.25a 18.60±0.27a 18.68±0.34a 18.33±0.21ab 17.54±0.40b Naringin‐d 13.81±0.39a 14.60±0.25a 14.40±0.37a 14.03±1.04a 13.11±0.53a Narirutin 16.61±0.05ab 16.68±0.26ab 17.33±0.53b 15.68±0.43ac 15.03±0.66c Hesperidin 251.28±9.30a 271.02±4.27a 262.64±5.24a 207.66±5.49b 205.03±19.40b Didymin 6.42±0.42abc 7.14±0.39b 7.01±0.63ab 5.87±0.106bc 5.55±0.45c ∑Flavanones 288.12±9.51a 309.44±4.83a 301.37±5.08a 243.24±6.32b 238.72±20.99b ∑Flavonoids 306.77±9.76a 328.04±5.10a 320.06±5.42a 261.57±6.32b 256.26±21.35b ΨFJ,freshjuice;P9230,pasteurizedjuiceat92ºC30s;P8515,pasteurizedjuiceat85ºC15s;H20,homogenizedjuiceat20 MPa;H150,homogenizedjuiceat150MPa;CHPuRP8515,centrifugedjuice,pulpfractionhomogenizedat150MPa, reconstitutedwiththeserumfractionandpasteurizedat85ºC15s. Differentlowercaselettersinthesamerowshowsignificantdifferences(p<0.05)amongsamples.
Table4.FlavonoidBioaccesibleContent(FBC,theoreticalamountofflavonoidsinmgpresent inthedigestatesaftertheingestionof1Lofjuice)ofjuicesassayedΨ. Flavonoids Flavonoid Bioaccessible Content FJ P9230 P8515 H150 CHPuMP8515 vicenin‐2 0.82±0.05a0.89±0.05ab 0.79±0.07a 1.01±0.06bc 1.15±0.09c apigenin‐d 0.25±0.02a0.26±0.02a 0.25±0.01a 0.29±0.03a 0.29±0.03a ∑Flavones 1.07±0.07ab 1.15±0.07ab 1.04±0.08a 1.30±0.09bc 1.44±0.12c naringin‐d 0.78±0.10a 0.91±0.20a 0.90±0.09a 0.91±0.11a 1.46±0.23b narirutin 1.01±0.06a 1.05±0.13a 1.01±0.05a 1.19±0.09ab 1.38±0.16b hesperidin 16.33±1.16a 18.04±2.13a 16.17±1.32a 17.09±2.16a 19.69±1.44a didymin 0.46±0.05a 0.56±0.06a 0.47±0.05a 0.56±0.06a 0.59±0.07a ∑Flavanones 18.58±1.36a 20.55±2.50a 18.55±1.51a 19.74±2.41a 23.11±1.81a ∑Flavonoids 19.65±1.44a 21.70±2.57a 19.595±1.59a 21.04±2.49a 24.55±1.92a ΨFJ,freshjuice;P9230,pasteurizedjuiceat92ºC30s;P8515,pasteurizedjuiceat85ºC15s;H20,homogenized juiceat20MPa;H150,homogenizedjuiceat150MPa;CHPuRP8515,centrifugedjuice,pulpfractionhomogenized at150MPa,reconstitutedwiththeserumfractionandpasteurizedat85ºC15s. Differentlowercaselettersinthesamerowshowsignificantdifferences(p<0.05)amongsamples.
TableS1.PhysicochemicalparametersΨoforangejuicesassayed. Parameters FJ P9230 P8515 H150 CHPuRP8515 Pulp (%) 8.83±0.59a 7.56±0.31a 8.01±0.01a 2.01±0.35c 2.33±0.23c Transmitance (%) 13.02±0.91a 9.65±0.49 b 9.50±1.34 b 9.97±0.39 b 3.32±0.08c Ascorbic acid (mg/L) 428.70±0.35a 413.07±0.73a 414.17±0.02a 425.91±0.78a 388.07±0.66a PME activity (nkat/mL) 1.501±0.023a 0.045±0.001 d 0.067±0.001 d 0.614±0.001c 0.061±0.001 d L* 44.69±0.01f 45.73±0.07e 46.71±0.01 d 50.27±0.01a 49.91±0.04f a* 5.33±0.01a 4.95±0.01 d 5.07±0.01 b 3.24±0.01f 3.52±0.02 d b* 54.06±0.01c 51.06±0.27 d 50.91±0.11 d 58.49±0.01a 58.50±0.01a C*ab 54.32±0.01c 51.30±0.027 d 51.17±0.11 d 58.58±0.01a 58.61±0.01a hab 84.36±0.01e 84.46±0.03 d 84.31±0.01e 86.83±0.01a 86.55±0.02 b ΔE*ab FJ ─ 3.20c 3.75 b 7.42 a 7.09 a ΔE*ab P9230 3.20c ─ 1.02 d 8.87 a 8.65 a ΔE*ab P8515 3.75c 1.02 d ─ 8.56 a8.38a ΔE*ab H150 7.42 b 8.87 a 8.56 a ─ 0.45 d ΔE*ab CHPuRP8515 7.09a 8.65a 8.38 a 0.45 d ─ ¥D[4,3] (µm) 540.89±2.97a 369.50±4.10 b 355.13±7.13 b 70.36±3.52c 52.05±3.14c ¥D[3,2] (µm) 108.84±1.53a 67.82±2.87 b 64.14±0.85 b 31.18±2.70c 14.54±0.96 d ¥SSA (m2/g) 0.06±0.01 d 0.09±0.01 d 0.09±0.01 d 0.19±0.02c 0.42±0.03a ¥d(0.1) (µm) 70.16±1.70a 35.11±1.86 b 28.86±0.83 b 18.96±3.29c 9.79±1.96 d ¥d(0.5) (µm) 480.13±2.06a 325.16±4.46 b 320.28±7.28 b 60.28±3.29c 39.81±1.92c ¥d(0.9) (µm) 1096.51±4.16a 768.54±6.37 b 732.53±12.74 b 137.55±6.47c 114.68±7.47c ΨFJ,freshjuice;P9230,pasteurizedjuiceat92ºC30s;P8515,pasteurizedjuiceat85ºC15s;H150,homogenizedjuiceat150MPa;CHPuRP8515,centrifugedjuice,pulpfractionhomogenized at150MPa,reconstitutedwiththeserumfractionandpasteurizedat85ºC15s.Differentletterswithinthesamerowindicatestatisticallysignificantdifferences(p<0.05). ¥D[4,3],volumemeandiameter;D[3,2],surfaceareameandiameter;As,specificsurfacearea,d(0.1),d(0.5),andd(0.9),standardpercentilereading.
TableS2.Summaryofthesimpleregressionanalysis*betweentheparticlesizeparameters andthecarotenoidsbioaccessibilityfoundinorangejuicesassayed. Bioaccessibility(%) D[4,3] D[3,2] As d(0.1) d(0.5) d(0.9) (Z)‐VIOisomers ‐0.975 ‐0.918 0.813 ‐0.817 ‐0.977 ‐0.978 (9Z)‐VIO+(Z)‐ANTisomer ‐0.942 ‐0.873 0.859 ‐0.734 ‐0.946 ‐0.947 ∑(Z)‐LUTEOisomer ‐0.919 ‐0.821 0.705 ‐0.694 ‐0.921 ‐0.926 ∑MUT‐epimer ‐0.922 ‐0.827 0.736 ‐0.694 ‐0.924 ‐0.929 ∑(Z)‐ANTisomer ‐0.943 ‐0.856 0.759 ‐0.734 ‐0.946 ‐0.949 LUT ‐0.878 ‐0.765 0.620 ‐0.643 ‐0.880 ‐0.886 ZEA ‐0.908 ‐0.805 0.666 ‐0.688 ‐0.908 ‐0.915 ZEINO ‐0.924 ‐0.829 0.743 ‐0.701 ‐0.925 ‐0.930 BCR ‐0.892 ‐0.783 0.628 ‐0.667 ‐0.892 ‐0.900 ACAR ‐0.739 ‐0.628 0.336 ‐0.565 ‐0.736 ‐0.748 BCAR ‐0.966 ‐0.915 0.796 ‐0.824 ‐0.968 ‐0.968 PF ‐0.861 ‐0.758 0.557 ‐0.668 ‐0.860 ‐0.868 PT ‐0.909 ‐0.852 0.849 ‐0.723 ‐0.913 ‐0.912 TC ‐0.940 ‐0.852 0.765 ‐0.724 ‐0.942 ‐0.946 RAE ‐0.913 ‐0.816 0.656 ‐0.708 ‐0.914 ‐0.920 *Numbersinitalicandboldindicatetheexistenceofsignificantcorrelations(p<0.05).