Impact of thermal treatments on the bioaccessibility of phytoene and phytofluene in relation to changes in the microstructure and size of orange juice particles

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 1 Impact of thermal treatments on the bioaccessibility of phytoene and 1 phytofluene in relation to changes in the microstructure and size of 2 orange juice particles 3 4 Paula Mapelli-Brahma*, Carla M. Stincoa, María J. Rodrigob, Lorenzo Zacaríasb, 5 Antonio J. Meléndez-Martíneza 6 7 a Food Colour & Quality Lab., Area of Nutrition & Food Science, Universidad de Sevilla, 8 41012 Seville, Spain. 9 10 b Food Biotechnology Department, Institute of Agrochemistry and Food Technology (IATA-11 CSIC), 46980, Valencia, Spain. 12 13 * Author to whom correspondence should be addressed. 14 Paula Mapelli-Brahm 15 Área de Nutrición y Bromatología. Universidad de Sevilla. 16 Facultad de Farmacia. 41012. Sevilla, Spain. 17 TELEPHONE: ++34 95455 7017 18 FAX: ++ 34 95455 7017 19 e-mail: pmape[email protected] 20 21 Declarations of interest: 22 AJMM is a member of the advisory board of Israeli Biotechnology Research Ltd. 23 24 25 *Manuscript Click here to view linked References HIGHLIGHTS  Phytoene, phytofluene (PTF) and ζ-carotene behave similarly to thermal treatments.  Phytoene and PTF have higher bioaccessibility than some carotenes and xanthophylls.  TEM images show great disruptions of cell structures with the thermal treatments.  Pasteurized juices have higher carotenoid bioaccessibility than frozen/thawed juices. *Highlights 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 2 ABSTRACT 26 The interest in the colourless carotenoids phytoene and phytofluene is 27 expanding. In this study their bioaccessibility from thermally treated Pinalate orange 28 juices, which contains high concentrations of these carotenoids, was evaluated in 29 relation to microstructural and particle size changes. Other carotenoids were also 30 considered for comparison. Fresh, pasteurized, and ultrafrozen juices thawed at room 31 temperature (UF-RT), in microwave oven (UF-MW), and in the fridge (UF-FG) were 32 investigated. Colourless carotenoids suffered less and higher degradation as a result of 33 ultrafreezing and pasteurization, respectively, than xanthophylls. In the fresh juice the 34 carotenoid with the highest bioaccessibility was phytoene (10%), followed by 35 zeaxanthin (9%) and phytofluene (8%). Total carotenoid bioaccessibility followed the 36 order: Pasteurized > UF-MW > UF-RT > UF-FG > Fresh. Thermal treatments 37 decreased the particle sizes and ruptured the cell structures and hence increased the 38 bioaccessibility. The best source of bioaccessible colourless carotenoids was UF-MW 39 (0.93 mg/250 mL). 40 KEYWORDS 41 In vitro digestion; particle size distribution; pasteurization; transmission electron 42 microscopy; ζ-carotene. 43 ABBREVIATIONS 44 Conjugated double bonds (cdb) 45 Fresh juice (FRESH) 46 High performance liquid chromatography (HPLC) 47 Methanol (MeOH) 48 Methyl tert-butyl ether (MTBE) 49 One-way analysis of variance (ANOVA) 50 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 3 Particle size distribution (PSD) 51 Pasteurized juice (PAST) 52 Phytoene (PT) 53 Phytofluene (PTF) 54 Potassium hydroxide (KOH) 55 Strokes per minute (spm) 56 Transmission electron microscopy (TEM) 57 Ultrafrozen juice with thawing at room temperature (UF-RT) 58 Ultrafrozen juice with thawing in the microwave oven (UF-MW) 59 Ultrafrozen juice with thawing in the fridge (UF-FG) 60 61 1. Introduction 62 Carotenoids are a wide family of isoprenoid compounds with beneficial health 63 properties (Kulczyński, Gramza-Michałowska, Kobus-Cisowska, & Kmiecik, 2017) and 64 only a few of the more than 700 natural carotenoids have been studied in depth (Britton, 65 Liaaen-Jensen, & Pfander, 2009). Phytoene (PT) and phytofluene (PTF) have been 66 largely ignored for a long time as compared to others but are currently being a focus of 67 major interest for the scientific community. In fact, some recent reviews have been 68 published focusing in the importance of these two carotenoids in the diet as well as in 69 their potential interest in the context of health promotion and cosmetics (Engelmann, 70 Clinton, & Erdman, 2011; Meléndez-Martínez, Mapelli-Brahm, & Stinco, 2018; 71 Meléndez-Martínez, Mapelli-Brahm, Benítez-González, & Stinco, 2015). PT and PTF 72 are present in a wide variety of fruits and vegetables and their juices, and hence are 73 among the predominant carotenoids in the diet (Biehler et al., 2012; Mapelli-Brahm, 74 Corte-Real, Meléndez-Martínez, & Bohn, 2017). It is well known that tomato and 75 tomato-based food products have a high concentration of both compounds (Khachik et 76 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 4 al., 2002). However, another source rich in these colourless carotenoids has recently 77 been characterized, specifically the orange Pinalate (Lado et al., 2015; Rodrigo, Marcos, 78 Alférez, Mallent, & Zacarías, 2003). This orange is a spontaneous mutant derived from 79 the ordinary Navelate orange with a partial blockage at the ζ-carotene desaturation 80 which seems to be the cause of the accumulation of these lineal carotenes. The flavedo 81 of the Pinalate orange has one of the highest concentrations of PT in fruits described so 82 far (Lado et al., 2015) and also contains very high concentrations of PTF and ζ-carotene 83 (Rodrigo et al., 2003). However, it is important to consider that the beneficial actions of 84 carotenoids and other species do not only depend on how much we ingest, but also on 85 other factors, like for instance their bioavailability. Before their absorption, carotenoids 86 have to be incorporated into mixed micelles to become absorbable for the intestinal 87 enterocytes. It is widely accepted that the bioaccessibility of a given lipophilic 88 compound is the amount of such compound that is released from the food matrix and 89 incorporated into mixed micelles with respect to the initial amount present in the matrix. 90 In vitro methods have been widely used to determine the bioaccessibility of carotenoids 91 and to estimate their bioavailability (Estévez-Santiago, Olmedilla-Alonso, & 92 Fernández-Jalao, 2016; Ornelas-Paz, Failla, Yahia, & Gardea-Bejar, 2008; Rodríguez-93 Roque, Rojas-Graü, Elez-Martínez, & Martín-Belloso, 2014). The bioaccessibility of 94 PT and PTF has not been as widely studied as that of other dietary carotenoids but there 95 are several articles that demonstrate their high bioaccessibility compared to that of other 96 carotenes and even to some xanthophylls (Jeffery, Turner, & King, 2012; Mapelli-97 Brahm et al., 2017; Rodrigo, Cilla, Barberá, & Zacarías, 2015; Rodrigues, 98 Chitchumroonchokchai, Mariutti, Mercadante, & Failla, 2017). Thermal treatments are 99 widely used in the food industry for preservation (Ibarz, Pagán, & Garza, 1999) but 100 these processes can affect the bioaccessibility of carotenoids and lead to the degradation 101 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 5 of some of them (Aschoff et al., 2015). Thus, thermal treatments can produce changes in 102 food colour and be detrimental for the nutritional value (Palmero, Lemmens, Hendrickx, 103 & Van Loey, 2014). Within this context, the main goals of this study were two. On the 104 one hand, to understand how different thermal treatments of Pinalate orange juice affect 105 the stability and the bioaccessibility of the colourless carotenoids phytoene and 106 phytofluene and compare the results with those of other carotenoids present in the 107 orange juice. On the other hand, as the thermal treatments can disrupt the food matrix 108 and hence, enhance the release of the carotenoids and increase their bioaccessibility 109 (Hof van het, West, Weststrate, & Hautvast, 2000), the effects of the changes in particle 110 size distribution (PSD) and plastids integrity (assessed by transmission electron 111 microscopy (TEM)) were also evaluated. 112 113 2. Materials and methods 114 2.1. Standards and reagents 115 HPLC solvents, i.e. methanol (MeOH) and methyl tert-butyl ether (MTBE), were 116 of HPLC grade and were acquired from Panreac (Barcelona, Spain). The necessary 117 reagents to obtain the images with the electronic transmission microscope were 118 purchased from Ted Pella, Inc. (Redding, USA). Porcine bile extract, pepsin (porcine, 119 367 units/mg solid, measured as TCA- soluble products using hemoglobin as substrate) 120 and pancreatin (porcine, 8x USP specifications of amylase, lipase and protease) were 121 acquired from Sigma-Aldrich (Bornem, Belgium). All reagents were of analytical grade 122 or higher. 123 2.2. Samples 124 The mutant Pinalate orange was chosen for this study due to its large accumulation 125 in colourless carotenoids. It has been found that the flavedo and pulp of this orange are 126 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 6 rich in PT, PTF and ζ-carotene (Lado et al., 2015; Rodrigo et al., 2003). However, to the 127 best of our knowledge, the juice has not been characterized. Mature Pinalate orange 128 fruits (Citrus sinensis L. Osbeck), were harvested at full mature stage (11.9 ºBrix) in 129 January 2016, from adult trees grown at The Citrus Germplasm Bank at the Instituto 130 Valenciano de Investigaciones Agrarias (IVIA, Moncada, Valencia, Spain). All 131 analyses were made with the same primary juice, which was obtained by manually 132 squeezing representative orange replicates. Five aliquots (fresh samples) were stored in 133 the fridge under nitrogen until analysis (one day). Pasteurization was carried out by 134 immersion of 50 mL plastic tubes containing 20 mL of juice in a water bath at 120 ºC. 135 After 30 seconds at 90 ºC, the tubes were immediately cooled in an ice bath until the 136 samples reached 1.8 ºC. Pasteurized juices were stored under nitrogen atmosphere in the 137 fridge until their analysis a few hours later. Ultrafreezing was carried out by direct 138 immersion of 50 mL plastic tubes containing 20 mL of juice in liquid nitrogen for two 139 minutes. The ultrafrozen samples were stored at - 80 ºC until their analysis. These were 140 performed no more than three days later, after their thawing. The ultrafreezing and 141 thawing treatments evaluated were the same considered in a previous study (Stinco, 142 Fernández-Vázquez, Heredia, Meléndez-Martínez, & Vicario, 2013). On the day of the 143 analysis, five replicates of the ultrafrozen juices were thawed at room temperature 144 (avoiding light), another five were thawed in a fridge (4 ºC) and the last five were 145 thawed in a microwave oven at 800 watts during 20 s, these latter conditions not leading 146 to increases of temperature. Thus, five samples were analysed, fresh juice (FRESH), 147 pasteurized juice (PAST) and ultrafrozen juices with thawing at room temperature (UF-148 RT), in the microwave oven (UF-MW) and in the fridge (UF-FG). FRESH was taken as 149 reference sample in order to study the effect of the different thermal treatments on the 150 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 7 colour and particle size distribution of the juice and on the concentration and 151 bioaccessibility of the carotenoids. 152 2.3. Transmission electron microscopy 153 Fresh juice, processed juices and pulp were analysed by TEM. Regarding the juices, 154 1.5 mL of juice was centrifuged at 18000 g for 5 min at 4 ºC in 2 mL vials (microfuge 155 22R, Beckman Coulter, Krefeld, Germany) and the aqueous phase was discarded. In the 156 case of the pulp, few juice vesicles were manually cut by using a blade and were 157 introduce in a 2 mL vial. Immediately, 1 mL of modified Karnovsky fixative (0.5% 158 glutaraldehyde, 2.5% formaldehyde) was added to each vial. The cells were fixing 159 during few hours in darkness and then centrifuged at 18000 g for 15 min at 4 ºC. The 160 upper phase was discarded and the pellet was rinsed three times with 0.1 M sodium 161 cacodylate buffer (pH 7.4). Then, the post-fixation was performed with 1% osmium 162 tetroxide in the buffer for 1 h at 25 ºC. The sample was washed for 20 min at 4 ºC with 163 distilled water and then stained with 2% aqueous solution of uranyl acetate for 2 h at 25 164 ºC. Dehydration was made in an acetone series (50, 70, 90, 100%). Subsequently, the 165 sample was embedded in Spurr resin, following a gradual procedure with different 166 ratios of acetone/spurr. Polymerization was carried out overnight at 70 °C. Ultrathin 167 sections of 70-100 nm were obtained by cutting semithin sections with an 168 ultramicrotom (Leica UC7, Wetzlar, Germany). These ultrathin sections were examined 169 with a Zeiss Libra 120 transmission electron microscope (Oberkochen, Germany) 170 equipped with a SSCCD digital camera. 171 2.4. Epifluorescence microscopy 172 The pulp and juice of the Pinalate orange were observed under a BX61 motorized 173 epifluorescence microscope (Olympus, Tokio, Japan) in order to detect phytofluene. 174 The pulp and juice of the parental Navelate were used as a reference. Fluorescence 175 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 8 microscopy allows for identifying specific fluorescent substances, named 176 fluorochromes, by observing their characteristic emission properties when illuminated 177 with radiation of appropriate wavelength. PTF, upon being excited with near-UV light 178 (300 - 400 nm), emits light at approximately 510 nm and therefore can be detected by 179 fluorescence microscopy (reviewed in Meléndez-Martínez et al., 2015). The 180 arrangement of optical components of this microscope allows the illumination from 181 above of the sample and therefore offers a high signal-to-noise ratio. The microscope 182 was equipped with a mercury vapour lamp as light source (X-Cite 120PC) and with 183 10×/0.4 and 20×/0.75 UPLANAPO objective lenses. The images were acquired with an 184 Olympus camera with a CellSens Dimension Software. An U-MNU2 filter (exciting 185 filter BP 360 - 370 nm, emission filter BA420, and dichromatic mirror DM 400 nm) 186 were applied. To obtain the images one drop of juice or one small piece of a cut vesicle 187 juice was spotted on a microscope slide glass (26 × 76 mm) and a microscope cover 188 glass (18 × 18 mm) was used to cover the sample. The use of a mounting medium was 189 not necessary. 190 2.5. Particle Size Distribution 191 Particles scatter light in all directions with an intensity pattern, which is dependent 192 on particle size. On the basis of this principle, the particle size distribution (PSD) of the 193 juices was determined by laser diffraction using a Mastersizer 3000E equipped with a 194 He-Ne laser (Malvern Instruments, Inc., Worcs, U.K.). The Mie model of light 195 scattering recommended by ISO13320-1 November 1999 was used. This model 196 assumes that particles are uniform spherical particles, which are illuminated by a plane 197 wave of infinite extent with a known wavelength (633 nm). The optical properties of the 198 sample and of the surrounding medium should be known in this model; the refractive 199 index and absorption rate of the cloud particles for orange juice is 1.73 and 0.1, 200 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 15 only contain a small amount of PTF (Stinco, Escudero-Gilete, Heredia, Vicario, & 348 Meléndez-Martínez, 2016) in comparison with that of Pinalate (Table 1). Fluorescence 349 was hardly observed in the pulp of Navelate orange (Figures 1-C and 1-D). From these 350 results, it can be inferred that epifluorescence microscopy could be an efficient, rapid 351 and reliable technique to detect PTF in orange juices. 352 3.3. Particle Size Distribution (PSD) 353 Regarding the PSD, all parameters analysed followed the decreasing order: 354 FRESH > UF-FG > UF-RT > UF-MW > PAST, being the differences between FRESH 355 and PAST in all cases statistically significant (Table 2). The volume- and areabased 356 mean diameters (D[4,3] and D[3,2], respectively) were 14.4% and 12.5% respectively 357 lower in PAST than in FRESH (Table 2). The cloud size distributions were bimodal in 358 all juices and, as can also be seen in Figure 3, FRESH was the juice with the highest 359 percentage of particles with a bigger size and with the lowest percentage of particles 360 with a lower size followed by UF-FG, UF-RT, UF-MW and PAST. That is, PAST was 361 the juice with the lowest particle size. PSD results of FRESH were similar to those 362 found by Stinco et al. (2012). Other researches have shown conflicting results on the 363 change in the particle size due to the pasteurization of fresh orange juices (Leizerson & 364 Shimoni, 2005; Stinco et al., 2012). To the best of our knowledge, the effect of freezing 365 and thawing on the particle size of orange juices is not well described in the literature. 366 However, it should be pointed out that, in many articles, juices which are regarded as 367 fresh juices were in fact frozen at some point. This is an important aspect to take into 368 account, as it has been proven that this process can have important effects in the PSD 369 (Table 2) and therefore in the colour and bioaccessibility of the carotenoids (sections 3.4 370 and 3.7). 371 3.4. Colour measurement 372 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 16 Colour is an attribute of paramount importance for the consumers and in Pinalate 373 juice it could be strongly influenced by its peculiar profile of carotenoids. The main 374 coloured carotenoid in Pinalate juice, i.e. ζ-carotene (Table 3), has a b* and C*ab values 375 much lower than those of the main carotenoids in common orange juices, i.e. 376 violaxanthin and β-cryptoxanthin (Meléndez-Martínez, Britton, Vicario, & Heredia, 377 2007; Stinco et al., 2016). This may be one of the main reasons why fresh Pinalate juice 378 has a significantly lower b* and C*ab values, i.e. 30 and 32 respectively, compared to 379 those average values found in common orange juices, i.e. approximately between 59 380 and 62 and between 50 and 61, respectively (Fernández-Vázquez, Stinco, Meléndez-381 Martínez, Heredia, & Vicario, 2011; Stinco et al., 2016). As a result, Pinalate juice has a 382 light-yellowish colour, which correlates well with that of ζ-carotene (Meléndez-383 Martínez et al., 2007), while common orange juices are more orange. On the other hand, 384 the high content of colourless carotenoids PT and PTF and the low content of 385 xanthophylls could contribute to the pale colour of the Pinalate juice. The colour 386 attributes a* and b* were positive in all samples. 387 The colour of FRESH was affected by the thermal treatments. Considering the 388 significant decrease in the lightness value of PAST in comparison to that of FRESH 389 (Table 3) it can be concluded that pasteurization caused a darkening of the juice. 390 Regarding C*ab and hab parameters, no statistically significant differences were found 391 with any of the treatments. These results are in agreement with those of a study with 392 ultrafrozen juice of Valencia late oranges thawed under conditions analogous to those of 393 this study (Stinco, Fernández-Vázquez, Heredia, Meléndez-Martínez, & Vicario, 2013). 394 However, contradictory results have been found in the literature regarding the changes 395 in the colour parameters of thermally-treated orange juices (Ortiz et al., 2017; Stinco et 396 al., 2016; Stinco et al., 2013; Wibowo et al., 2015). These differences may be due to 397 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 17 some extent to differences in the carotenoid levels of the oranges (Wibowo et al., 2015). 398 However, it should be noted that changes in the concentration of ascorbic acid can also 399 influence the colour of orange juices (Meléndez-Martínez, Vicario, & Heredia, 2009; 400 Roig, Bello, Rivera, & Kennedy, 1999) and that orange juices browning could be due to 401 the oxidation of phenols to quinones (Eissa, Fadelb, Ibrahim, Hassan, & Elrashid, 402 2006). ΔE* is a good parameter for understanding how observers perceive colour 403 differences and it is considered that, ΔE* over 2.8 CIELAB units can be perceived by 404 even inexperienced observers (Melgosa, Pérez, Yebra, Huertas, & Hita, 2001). 405 Differences above this threshold were found in UF-FG, UF-MW, and PAST (Table 3). 406 The differences found in the ultrafrozen juices are very similar to those found in other 407 studies (Cinquanta, Albanese, Cuccurullo, & Di Matteo, 2010; Stinco et al., 2013). 408 Among the thermal treatments, pasteurization generated the greatest colour difference 409 (ΔE*) with respect to FRESH while the smallest difference occurred in the UF-RT 410 (Table 3). This could be related to the fact that UF-RT was the juice with the lowest 411 degradation of carotenoids and PAST the one that suffered the greatest degradation 412 (Supplementary Table 1). 413 3.5. Carotenoid profile in the fresh orange juice 414 Violaxanthin, antheraxanthin (cyclic epoxycarotenoids), zeaxanthin (cyclic 415 dihydroxycarotenoid), β-carotene (cyclic carotene) and three linear carotenes, i.e. ζ-416 carotene, PT and PTF, were detected. That is, three xanthophylls and four carotenes. β-417 Carotene is one of most studied carotenoids due to its high provitamin A activity and 418 zeaxanthin has also been extensively studied for its role in the human macula (Johnson 419 et al., 2000). On the other hand, although PT, PTF and ζ-carotene are found in human 420 serum they are very poorly studied carotenoids. The most abundant carotenoid was ζ-421 carotene followed by PTF and PT (Table 1). Other researchers have found a higher 422 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 18 concentration of PT with respect to that of ζ-carotene in the pulp of Pinalate (Lado et 423 al., 2015). This difference could be due to the possible difference in the degree of 424 release between the carotenoids when juicing, and this could, in turn, be due to the 425 difference in the subcellular structures where each carotenoid accumulates (Rodrigo, 426 Cilla, Reyes, & Zacarías, 2015). Regarding the isomers, 4 of ζ-carotene and PTF, 2 of 427 PT, and only the 9-cis isomer of violaxanthin and antheraxanthin were detected. In 428 addition, four compounds were detected but could not been identified. As their retention 429 times were similar to those of the common carotenoid esters, different saponification 430 conditions were tested in order to be certain whether they were free carotenoids or not. 431 The areas of their respective peaks were not altered with the different saponification 432 times and KOH concentrations so it was concluded that they were free carotenoids. 433 Taking into account that they could not be identified and that they were not one of the 434 six carotenoids that represent more than 95% of the total carotenoids in human blood 435 (Maiani et al., 2009), they were not included in the present study. However, their 436 chromatographic and spectroscopic characteristics are summarized in Supplementary 437 Table 2. The carotenoid profile found in the Pinalate juice is very different from that of 438 other coloured common citrus fruits. In the latter, xanthophylls represent up to 80% of 439 the total carotenoid content and linear carotenes account for no more than 20% (Rodrigo 440 et al., 2003) while, in the Pinalate juice, ζ-carotene, PT and PTF together represented 441 nearly 98% of the total carotenoids and only 1.2% of the carotenoids were xanthophylls 442 (Table 1). Thus, while in juices of common varieties of orange the concentration of PT 443 and PTF is approximately 1 and 0.35 mg/L respectively (Stinco et al., 2016), in Pinalate 444 juice the concentration was 14 and 19 mg/L, respectively. Although Cara Cara is also 445 considered an orange rich in these carotenoids, with reported values of 12 and 3 mg/L of 446 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 19 PT and PTF respectively (Stinco et al., 2016), the concentration is lower than that of 447 Pinalate. 448 3.6. Carotenoid degradation during thermal treatments 449 The degradation of carotenoids is an important aspect for the food industry since 450 it can affect the colour and the nutritional value of foods. Important percentages of 451 degradation of carotenoids were observed across the different treatments 452 (Supplementary Table 1). PT and PTF were the carotenoids which suffered the highest 453 degradation during pasteurization, with approximately 67% of degradation. On the other 454 hand, in all ultrafrozen juices, PT and PTF were more stable than xanthophylls 455 (Supplementary Table 1). These results seem to indicate that the colourless carotenoids 456 are less prone to degradation due to the combination of oxygen, enzymes and acid, but 457 less heat stable than the xanthophylls. This difference in the behaviour of PT and PTF 458 compared to the xanthophylls may be due to some extent to structural and physico-459 chemical differences, for instance their absence of terminal rings, differences in 460 polarity, or the smaller number of conjugated double bonds, which is known to have a 461 remarkable impact in the electron affinity and ionization energy of carotenoids 462 (Martínez, Stinco, & Meléndez-Martínez, 2014). Likewise, they could be due, at least in 463 part, to differences in their accumulation within the cell, like for instance the 464 substructures they may deposit or in the aggregates that might be formed. However, it 465 must be taken into account that orange juice is a complex matrix and that, to be able to 466 obtain more specific conclusions about the degradation of PT and PTF as compared to 467 other orange juice carotenoids, it seems more reasonable to carry out oxidation studies 468 with carotenoid standards under identical conditions. Violaxanthin was the carotenoid 469 which suffered the highest losses with the three types of ultrafreezing and thawing, with 470 an average degradation of 76.5% among the different thawing conditions. This is 471 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 20 consistent with the fact that 5,6-epoxycarotenoids suffer a re-arrangement to their 472 respective 5,8-furanoxide in acidic media (Meléndez-Martínez, Vicario, & Heredia, 473 2007). 474 Regarding total carotenoid concentration, FRESH was the juice with the highest 475 content, i.e. 7.7 mg/100 g, while PAST was the juice with the lowest concentration, i.e. 476 2.6 mg/100g (Table 1), ca. 3-fold lower compared to FRESH. Significant differences 477 were observed in total carotenoid content between fresh and ultrafrozen juices, although 478 previous work did not find differences (Stinco et al., 2013). This differential behaviour 479 may be related to the differences in carotenoid composition between standard sweet 480 orange juices and Pinalate juice. The percentages of carotenoid degradation in the juices 481 followed the order: UF-RT (32%), UF-FG (41%), UF-MW (42%) and PAST (67%) 482 (Supplementary Table 1). Interesting, this increasing order agreed well with that of the 483 degree of microstructural changes (section 3.2). In this sense, it seems reasonable to 484 hypothesize that the greater the degradation of the cell structures, the more exposed the 485 carotenoids are to the acid environment, oxygen, and enzymes, and so, the greater 486 degradation they suffer. The greater degradation of carotenoids and plastids found in 487 PAST could be due to the combination of the aforementioned factors and the heat, as it 488 is well known that the oxidative degradation of carotenoids is stimulated by heat 489 (reviewed in Rodríguez-Amaya & Kimura, 2004; Rodríguez-Amaya, 1999). Among the 490 ultrafrozen juices, UF-FG suffered the highest loss of carotenoids, a result that is in 491 good agreement with the fact that a slow defrosting causes greater carotenoid losses 492 than a rapid thawing (reviewed in Rodríguez-Amaya & Kimura, 2004). 493 3.7. Carotenoid bioaccessibility 494 In all the juices, PT had a bioaccessibility higher than that of PTF and the 495 bioaccessibility of both was higher than that of ζ-carotene. This order among the linear 496 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 21 carotenes might be due to the difference in the number of conjugated double bonds 497 (cdb). In this regard, it could be argued that the higher number of cdb of ζ-carotene (7 498 cdb) in comparison with those of PTF (5 cdb) and PT (3 cdb) makes this carotene more 499 rigid and prone to aggregation of their molecules, which could lead to a decrease in the 500 bioaccessibility, similarly to what were found when comparing PT, PTF and lycopene 501 (Mapelli-Brahm et al., 2017; Meléndez-Martínez, Paulino, Stinco, Mapelli-Brahm, & 502 Wang, 2014; Rodrigo, Cilla, Barberá, et al., 2015). In all the samples, the 503 bioaccessibility of the colourless carotenoids was even higher than that of the 504 xanthophylls antheraxanthin and violaxanthin. These epoxycarotenoids are not found at 505 detectable levels in human fluids or tissues (Khachik, 2006), while PT and PTF are 506 common circulating carotenoids and have been found in several organs (reviewed in 507 Meléndez-Martínez et al., 2015). The only carotenoid that showed greater 508 bioaccessibility than the colourless carotenoids was zeaxanthin, although in all cases the 509 difference with PT was not statistically significant. The high relative bioaccessibility of 510 PT and PTF compared to other carotenoids is consistent with previous investigations. 511 The colourless carotenoids showed bioaccessibilities higher than those of β-carotene, α-512 carotene and/or lycopene in carrot, papaya, different types of salads, apricot juice, 513 grapefruit, melon, watermelon, tomato, and tomato juice (Jeffery et al., 2012; Mapelli-514 Brahm et al., 2017; Rodrigues et al., 2017), and it was found to be even higher than that 515 of other xanthophylls such as β-cryptoxanthin, lutein and violaxanthin in certain 516 matrices (Jeffery et al., 2012; Mapelli-Brahm et al., 2017). 517 Total carotenoid bioaccessibility followed the order: PAST (26%) > UF-MW 518 (18%) > UF-RT (14%) > UF-FG (12%) > FRESH (8%). Thus, the pasteurization 519 increased such parameter over 3-fold and the freezing followed by thawing with 520 microwave over 2-fold. The same order was followed by all the carotenoids with the 521 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 22 exception of violaxanthin and antheraxanthin (Table 4). Thus, PAST was the juice with 522 the highest bioaccessibility of PT and PTF, i.e. 32 and 27%, respectively, while the 523 lowest was FRESH with 10 and 8% (Table 4), that is over 3-fold less for both carotenes. 524 The differences in the bioaccessibility of each carotenoid and the total bioaccessibility 525 between PAST and FRESH were statistically significant (P < 0.05). These results 526 highlight the importance of the food matrix effect on the bioaccessibility. Since 527 carotenoids must release from the food matrix in order to be incorporated into the mixed 528 micelles, it seems reasonable to expect that the more the matrix is disrupted, the higher 529 the bioaccessibility will be. In order to test this hypothesis, the particle size and the cell 530 structures degradation of the samples were studied, and the results of both analyses 531 confirmed this assumption. Thus, TEM images (Figure 2) showed the same increasing 532 order in the degradation of the plastids among the samples (section 3.2) than the order 533 found in the bioaccessibility. PSD results also agree with those of TEM and 534 bioaccessibility, indicating that FRESH was the juice with the highest percentage of 535 particles with a big size and the lowest percentage of particles with a low size, followed 536 by UF-FG, UF-RT, UF-MW and PAST (section 3.3). Taken together, it could be 537 concluded that the higher bioaccessibility found in PAST is due to the fact that the 538 effect of heat during pasteurization causes greater matrix degradation than the 539 ultrafreezing followed by thawing. On the other hand, it is considered that the higher the 540 concentration of a given carotenoid in a matrix, the lower bioaccessibility, among other 541 reasons because high concentrations could increase the possibility of molecular 542 aggregation (reviewed in Borel, 2003; Mapelli-Brahm et al., 2017). Consistent with the 543 above, it was found that, probably due in part to the degradation, the carotenoid 544 concentration in PAST was significantly lower than that of FRESH (Table 1). Other 545 authors have found increases in carotenoids bioaccessibility in orange juices with the 546 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 23 pasteurization (Aschoff et al., 2015). On the other hand, Stinco et al. (2013), also found 547 higher carotenoid bioaccessibility when orange juices were subjected to ultrafreezing 548 following to thawing in the microwave oven but this increase was not found when the 549 juices were defrosting at room temperature or in the fridge. 550 In order to study whether the thermal treatments affect to the same extent the 551 bioaccessibility of the different carotenoids, the increase in the bioaccessibility for each 552 carotenoid with the thermal treatments, taking as the reference the respective 553 bioaccessibility in the FRESH, was calculated (Supplementary Table 3). The 554 percentages of increase in the bioaccessibility among the carotenes were very similar 555 while in the xanthophylls these values were less homogeneous. This may be due in part 556 to the greater difference with respect to the number and type of functional groups, i.e. 557 epoxides and hydroxyls, and with this the greater difference in the polarity, found 558 among the xanthophylls as compared to carotenes. Also the variety in the degree of 559 esterification and in the fatty acids that may be involved in each ester could be one of 560 the causes of this difference (Borel, 2003). The average increase in the bioaccessibility 561 of the carotenes with the pasteurization was 238% while in the UF-FG the 562 bioaccessibility was increased by only 49%. Interestingly, for all treatments, the 563 bioaccessibility increase was higher in PTF than in PT. 564 The pH of the samples was measured as it is known that the pH can affect the 565 transfer of carotenoids to mixed micelles during the digestion (reviewed in Reboul & 566 Borel, 2011). However, no significant changes in the pH were found with the thermal 567 treatments, being 3.8 the pH average (data not shown). In any case, the effect of 568 possible differences in the acidity of the juices was expected to be counteracted by the 569 pH reached during the gastric and intestinal digestion phases, which was homogeneous 570 across samples. 571 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 24 3.8. Bioaccessible carotenoid content 572 Information on the bioaccessible carotenoid contents can be more meaningful as 573 they represent the potentially absorbable amount of carotenoids. These data are 574 summarized in Table 5. In all samples, ζ-carotene was the carotenoid with the highest 575 bioaccessible content, followed by PT and PTF which showed virtually the same 576 bioaccessible content. The intake of approximately one glass of FRESH, i.e. 250 mL, 577 provided 0.7 mg of ζ-carotene and 0.4 mg of PT and PTF. With the exception of ζ-578 carotene, the amount of potentially absorbable PT and PTF that is provided with the 579 intake of any of the juices is much higher than that of the rest of carotenoids. Thus, for 580 example, the bioaccessible amount of PT or PTF is more than 100 times greater than 581 that of violaxanthin. 582 Noteworthy is that neither the PAST, which presented the highest total carotenoid 583 bioaccessibility, nor the FRESH, which exhibited the highest initial carotenoid 584 concentration were the juices with highest content of bioaccessible carotenoids. Thus, 585 the source that provided the highest quantities of bioaccessible PT, PTF, and ζ-carotene 586 and the highest total carotenoid bioaccessible content was UF-MW, while the lowest 587 was UF-FG. This clearly demonstrates the importance of using the bioaccessible 588 content rather than the percentage of bioaccessibility and the concentration in the 589 matrix, when it comes to evaluate the goodness of a food to raise the carotenoid status 590 in humans. UF-FG was the only juice with an bioaccessible content of PT, PTF and 591 total carotenoid lower than that of FRESH. In all the cases, the increase in the 592 bioaccessible content with the thermal treatments was higher for PTF compared to PT, 593 and this could mainly due to the higher increase in the bioaccessibility of PTF with the 594 different thermal treatments. 595 596 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 31 Zeaxanthin 9.24 ± 1.22Bab 15.50 ± 1.70Bab 15.10 ± 0.81Ba 30.73 ± 4.67Aa 39.57 ± 5.32Aa 9-cis-Antheraxanthin 1.56 ± 0.18Dd 6.14 ± 0.83Bc 4.49 ± 0.88BCf 8.91 ± 1.68Ac 3.57 ± 0.52CDc β-Carotene 6.21 ± 0.41Cc 11.72 ± 0.60BCb 8.98 ± 0.81BCde 14.57 ± 1.28Bbc 21.01 ± 3.87Ab Total 7.57 ± 0.28C 14.44 ± 0.99B 11.54 ± 0.75B 18.22 ± 1.89B 25.93 ± 4.35A FRESH, Fresh orange juice; UF-RT, ultrafrozen juice with thawing at room temperature; UF-FG, 708 ultrafrozen juice with thawing in the fridge; UF-MW, ultrafrozen juice with thawing in the microwave 709 oven; PAST, pasteurized juice. Values are the mean ± SD of 5 independent measures. Values within a 710 column and within a row with different lowercase and uppercase letters respectively indicate statistically 711 significant differences (P < 0.05). 712 713 714 715 716 717 718 Table 5. Bioaccessible carotenoid content (mg/250 mL) of Pinalate fresh orange 719 juice and Pinalate orange juices subjected to different thermal treatments. 720 Juice samples UF-RT UF-FG UF-MW PAST Phytoene 28.22 ± 3.60 40.53 ± 4.33 43.32 ± 5.97 67.85 ± 7.25 Phytofluene 31.22 ± 4.00 40.23 ± 4.45 42.58 ± 5.57 67.57 ± 5.95 ζ-Carotene 33.27 ± 4.32 40.59 ± 4.31 41.38 ± 5.82 66.21 ± 5.74 9-cis-Violaxanthin 85.71 ± 4.35 75.50 ± 7.37 68.33 ± 3.82 62.51 ± 4.38 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 32 Zeaxanthin 37.80 ± 5.46 49.35 ± 5.11 57.29 ± 0.47 58.46 ± 2.97 9-cis-Antheraxanthin 54.95 ± 5.63 55.52 ± 2.39 59.12 ± 1.60 55.98 ± 5.37 β-Carotene 22.01 ± 4.48 28.70 ± 3.30 31.02 ± 7.49 63.70 ± 4.62 Total 32.18 ± 4.12 40.67 ± 4.22 42.18 ± 5.76 66.74 ± 6.03 FRESH, Fresh orange juice; UF-RT, ultrafrozen juice with thawing at room temperature; UF-FG, 721 ultrafrozen juice with thawing in the fridge; UF-MW, ultrafrozen juice with thawing in the microwave 722 oven; PAST, pasteurized juice. Values are the mean ± SD of 5 independent measures. Values within a 723 column and within a row with different lowercase and uppercase letters respectively indicate statistically 724 significant differences (P < 0.05). 725 726 727 728 729 730 731 REFERENCES 732 Aschoff, J. K., Kaufmann, S., Kalkan, O., Neidhart, S., Carle, R., & Schweiggert, R. M. 733 (2015). In vitro bioaccessibility of carotenoids, flavonoids, and vitamin C from 734 differently processed oranges and orange juices [ Citrus sinensis (L.) osbeck]. 735 Journal of Agricultural and Food Chemistry, 63(2), 578–587. 736 https://doi.org/10.1021/jf505297t 737 Biehler, E., Alkerwi, A., Hoffmann, L., Krause, E., Guillaume, M., Lair, M. L., & 738 Bohn, T. (2012). Contribution of violaxanthin, neoxanthin, phytoene and 739 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 33 phytofluene to total carotenoid intake: Assessment in Luxembourg. Journal of 740 Food Composition and Analysis, 25(1), 56–65. 741 https://doi.org/10.1016/j.jfca.2011.07.005 742 Borel, P. (2003). Factors affecting intestinal absorption of highly lipophilic food 743 microconstituents (fat-soluble vitamins, carotenoids and phytosterols). Clinical 744 Chemistry and Laboratory Medicine : CCLM / FESCC, 41(8), 979–994. 745 https://doi.org/10.1515/CCLM.2003.151 746 Britton, G., Liaaen-Jensen, S., & Pfander, H. (2009). Nutrition and Health. In G. 747 Britton, S. Liaaen-Jensen, & H. Pfander (Eds.), Carotenoids (Vol. 5). Basel, 748 Switzerland: Birkhäuser. 749 Cinquanta, L., Albanese, D., Cuccurullo, G., & Di Matteo, M. (2010). Effect on orange 750 juice of batch pasteurization in an improved pilot-scale microwave oven. Journal 751 of Food Science, 75, E46–E50. 752 Corredig, M., Kerr, W., & Wicker, L. (2001). Particle size distribution of orange juice 753 cloud after addition of sensitized pectin. Journal of Agricultural and Food 754 Chemistry, 49(5), 2523–2526. https://doi.org/10.1021/jf001087a 755 Eissa, H., Fadelb, H., Ibrahim, G. E., Hassan, I., & Elrashid, A. A. (2006). Thiol 756 containing compounds as controlling agents of enzymatic browning in some apple 757 products. Food Research International, 39(8), 855–863. 758 Engelmann, N. J., Clinton, S. K., & Erdman, J. W. J. (2011). Nutritional aspects of 759 phytoene and phytofluene, carotenoid precursors to lycopene. Advances in 760 Nutrition, 2(1), 51–61. https://doi.org/10.3945/an.110.000075.1 761 Estévez-Santiago, R., Olmedilla-Alonso, B., & Fernández-Jalao, I. (2016). 762 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 34 Bioaccessibility of provitamin A carotenoids from fruits: application of a 763 standardised static in vitro digestion method. Food & Function, 7(3), 1354–1366. 764 https://doi.org/10.1039/C5FO01242B 765 Fernández-Vázquez, R., Stinco, C. M., Meléndez-Martínez, A. J., Heredia, F. J., & 766 Vicario, I. M. (2011). Visual and instrumental evaluation of orange juice color: A 767 consumers’ preference study. Journal of Sensory Studies, 26(6), 436–444. 768 https://doi.org/10.1111/j.1745-459X.2011.00360.x 769 Gupta, R., Kopec, R. E., Schwartz, S. J., & Balasubramaniam, V. M. (2011). Combined 770 pressure–temperature effects on carotenoid retention and bioaccessibility in tomato 771 juice. Journal of Agricultural and Food Chemistry, 59(14), 1–20. 772 https://doi.org/10.1021/jf200575t.Combined 773 Hof van het, K. H., West, C. E., Weststrate, J. A., & Hautvast, J. G. A. J. (2000). 774 Dietary factors that affect the bioavailability of carotenoids. The Journal of 775 Nutritition, (14), 503–506. 776 Ibarz, A., Pagán, J., & Garza, S. (1999). Kinetic models for colour changes in pear 777 puree during heating at relatively high temperatures. Journal of Food Engineering, 778 39(4), 415–422. 779 Jeffery, J. L., Turner, N. D., & King, S. R. (2012). Carotenoid bioaccessibility from 780 nine raw carotenoid-storing fruits and vegetables using an in vitro model. Journal 781 of the Science of Food and Agriculture, 92(13), 2603–2610. 782 https://doi.org/10.1002/jsfa.5768 783 Johnson, E. J., Hammond, B. R., Yeum, K. J., Qin, J., Wang, X. D., Castaneda, C., … 784 Russell, R. M. (2000). Relation among serum and tissue concentrations of lutein 785 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 35 and zeaxanthin and macular pigment density. American Journal of Clinical 786 Nutrition, 71(6), 1555–1562. 787 Khachik, F. (2006). Distribution and metabolism of dietary carotenoids in humans as a 788 criterion for development of nutritional supplements. Pure and Applied Chemistry, 789 78(8), 1551–1557. https://doi.org/10.1351/pac200678081551 790 Khachik, F., Carvalho, L., Bernstein, P. S., Muir, G. J., Zhao, D.-Y., & Katz, N. B. 791 (2002). Chemistry, distribution, and metabolism of tomato carotenoids and their 792 impact on human health. Experimental Biology and Medicine (Maywood, N.J.), 793 227(10), 845–851. 794 Kulczyński, B., Gramza-Michałowska, A., Kobus-Cisowska, J., & Kmiecik, D. (2017). 795 The role of carotenoids in the prevention and treatment of cardiovascular disease – 796 Current state of knowledge. Journal of Functional Foods, 38, 45–65. 797 https://doi.org/10.1016/j.jff.2017.09.001 798 Lado, J., Zacarías, L., Gurrea, A., Page, A., Stead, A., & Rodrigo, M. J. (2015). 799 Exploring the diversity in Citrus fruit colouration to decipher the relationship 800 between plastid ultrastructure and carotenoid composition. Planta, 242(3). 801 https://doi.org/10.1007/s00425-015-2370-9 802 Lado, J., Zacarías, L., Gurrea, A., Page, A., Stead, A., & Rodrigo, M. J. (2015). 803 Exploring the diversity in Citrus fruit colouration to decipher the relationship 804 between plastid ultrastructure and carotenoid composition. Planta, 242(3), 645–805 661. https://doi.org/10.1007/s00425-015-2370-9 806 Leizerson, S., & Shimoni, E. (2005). Stability and sensory shelf life of orange juice 807 pasteurized by continuous ohmic heating. Journal of Agricultural and Food 808 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 36 Chemistry, 53, 4012–4018. https://doi.org/10.1021/jf047857q 809 Leong, S. Y., & Oey, I. (2012). Effects of processing on anthocyanins, carotenoids and 810 vitamin C in summer fruits and vegetables. Food Chemistry, 133(4), 1577–1587. 811 https://doi.org/10.1016/j.foodchem.2012.02.052 812 Maiani, G., Castón, M. J., Catasta, G., Toti, E., Cambrodón, I. G., Bysted, A., … 813 Schlemmer, U. (2009). Carotenoids: Actual knowledge on food sources, intakes, 814 stability and bioavailability and their protective role in humans. Molecular 815 Nutrition & Food Research, 53, S194–S218. 816 Mapelli-Brahm, P., Corte-Real, J., Meléndez-Martínez, A. J., & Bohn, T. (2017). 817 Bioaccessibility of phytoene and phytofluene is superior to other carotenoids from 818 selected fruit and vegetable juices. Food Chemistry, 229, 304–311. 819 https://doi.org/10.1016/j.foodchem.2017.02.074 820 Martínez, A., Stinco, C. M., & Meléndez-Martínez, A. J. (2014). Free radical 821 scavenging properties of phytofluene and phytoene isomers as compared to 822 lycopene: a combined experimental and theoretical study. The Journal of Physical 823 Chemistry. B, 118(33), 9819–9825. https://doi.org/10.1021/jp503227j 824 Meléndez-Martínez, A. J., Britton, G., Vicario, I. M., & Heredia, F. J. (2007). 825 Relationship between the colour and the chemical structure of carotenoid pigments. 826 Food Chemistry, 101(3), 1145–1150. 827 https://doi.org/10.1016/j.foodchem.2006.03.015 828 Meléndez-Martínez, A. J., Mapelli-Brahm, P., Benítez-González, A., & Stinco, C. M. 829 (2015). A comprehensive review on the colorless carotenoids phytoene and 830 phtyofluene. Archives of Biochemistry and Biophysics, 572, 188–200. 831 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 37 Meléndez-Martínez, A. J., Mapelli-Brahm, P., & Stinco, C. M. (2018). The colourless 832 carotenoids phytoene and phytofluene: From dietary sources to their usefulness for 833 the functional foods and nutricosmetics industries. Journal of Food Composition 834 and Analysis, 67(January), 91–103. https://doi.org/10.1016/j.jfca.2018.01.002 835 Meléndez-Martínez, A. J., Paulino, M., Stinco, C. M., Mapelli-Brahm, P., & Wang, X. 836 D. (2014). Study of the time-course of cis/trans (Z/E) isomerization of lycopene, 837 phytoene, and phytofluene from tomato. Journal of Agricultural and Food 838 Chemistry, 62(51), 12399–12406. https://doi.org/10.1021/jf5041965 839 Meléndez-Martínez, A. J., Vicario, I. M., & Heredia, F. J. (2007). Review: Analysis of 840 carotenoids in orange juice. Journal of Food Composition and Analysis, 20(7), 841 638–649. https://doi.org/10.1016/j.jfca.2007.04.006 842 Meléndez-Martínez, A. J., Vicario, I. M., & Heredia, F. J. (2009). Effect of ascorbic 843 acid on deterioration of carotenoids and colour in ultrafrozen orange juice. Journal 844 of Food Composition and Analysis, 22(4), 295–302. 845 https://doi.org/10.1016/j.jfca.2008.12.005 846 Melgosa, M., Pérez, M. M., Yebra, A., Huertas, R., & Hita, E. (2001). Algunas 847 reflexiones y recientes recomendaciones internacionales sobre evaluación de 848 diferencias de color. Óptica Pura Y Aplicada, 34, 1–10. 849 Ornelas-Paz, J. D. J., Failla, M. L., Yahia, E. M., & Gardea-Bejar, A. (2008). Impact of 850 the stage of ripening and dietary fat on in vitro bioaccessibility of B-carotene in 851 “Ataulfo” mango. Journal of Agricultural and Food Chemistry, 56(4), 1511–1516. 852 https://doi.org/10.1021/jf072751r 853 Ortiz, L., Dorta, E., Gloria Lobo, M., Antonio González-Mendoza, L., Díaz, C., & 854 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 38 González, M. (2017). Use of banana peel extract to stabilise antioxidant capacity 855 and sensory properties of orange juice during pasteurisation and refrigerated 856 storage. Food and Bioprocess Technology, 10(10), 1883–1891. 857 https://doi.org/10.1007/s11947-017-1964-6 858 Palmero, P., Lemmens, L., Hendrickx, M., & Van Loey, A. (2014). Role of carotenoid 859 type on the effect of thermal processing on bioaccessibility. Food Chemistry, 157, 860 275–282. https://doi.org/10.1016/j.foodchem.2014.02.055 861 Reboul, E., & Borel, P. (2011). Proteins involved in uptake, intracellular transport and 862 basolateral secretion of fat-soluble vitamins and carotenoids by mammalian 863 enterocytes. Progress in Lipid Research, 50(4), 388–402. 864 https://doi.org/10.1016/j.plipres.2011.07.001 865 Rodrigo, M. J., Cilla, A., Barberá, R., & Zacarías, L. (2015). Carotenoid bioaccessibility 866 in pulp and fresh juice from carotenoid-rich sweet oranges and mandarins. Food & 867 Function, 6(6), 1950–1959. https://doi.org/10.1039/C5FO00258C 868 Rodrigo, M. J., Cilla, A., Reyes, B., & Zacarías, L. (2015). Carotenoids bioaccessibility 869 in pulp and fresh juice from carotenoid-rich sweet oranges and mandarins. Food & 870 Function, 6(Cc), 1950–1959. https://doi.org/10.1039/C5FO00258C 871 Rodrigo, M. J., Marcos, J. F., Alférez, F., Mallent, M. D., & Zacarías, L. (2003). 872 Characterization of Pinalate, a novel Citrus sinensis mutant with a fruit-specific 873 alteration that results in yellow pigmentation and decreased ABA content. Journal 874 of Experimental Botany, 54(383), 727–738. https://doi.org/10.1093/jxb/erg083 875 Rodrigues, D. B., Chitchumroonchokchai, C., Mariutti, L. R. B., Mercadante, A. Z., & 876 Failla, M. L. (2017). Comparison of two static in vitro digestion methods for 877 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 39 screening bioaccessibility of carotenoids in fruits, vegetables and animal products. 878 Journal of Agricultural and Food Chemistry, 65, 11220−11228. 879 https://doi.org/10.1021/acs.jafc.7b04854 880 Rodríguez-Amaya, D. B. (1999). Changes in carotenoids during processing and storage 881 of foods. Archivos Latinoamericanos de Nutricion, 49(3(1-S)), 38S–47S. 882 Rodríguez-Amaya, D. B. (2001). A guide to carotenoid analysis in foods. Washington, 883 D.C., Estados Unidos.: ILSI Press. 884 Rodríguez-Amaya, D. B., & Kimura, M. (2004). HarvestPlus Handbook for Carotenoid 885 Analysis. Tropical Agriculture. Washington, DC and Cali: IFPRI and CIAT. 886 Rodríguez-Roque, M. J., Rojas-Graü, M. A., Elez-Martínez, P., & Martín-Belloso, O. 887 (2014). In vitro bioaccessibility of health-related compounds from a blended fruit 888 juice-soymilk beverage: Influence of the food matrix. Journal of Functional 889 Foods, 7(1), 161–169. https://doi.org/10.1016/j.jff.2014.01.023 890 Roig, M. G., Bello, J. F., Rivera, Z. S., & Kennedy, J. F. (1999). Studies on the 891 occurrence of non-enzymatic browning during storage of citrus juice. Food 892 Research International, 32, 609–619. 893 Sánchez-Moreno, C., Plaza, L., Elez-Martínez, P., De Ancos, B., Martín-Belloso, O., & 894 Cano, M. P. (2005). Impact of high pressure and pulsed electric fields on bioactive 895 compounds and antioxidant activity of orange juice in comparison with traditional 896 thermal processing. Journal of Agricultural and Food Chemistry, 53(11), 4403–897 4409. https://doi.org/10.1021/jf048839b 898 Stinco, C. M., Escudero-Gilete, M. L., Heredia, F. J., Vicario, I. M., & Meléndez-899 Martínez, A. J. (2016). Multivariate analyses of a wide selection of orange varieties 900 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 40 based on carotenoid contents, color and in vitro antioxidant capacity. Food 901 Research International, 90, 194–204. 902 https://doi.org/10.1016/j.foodres.2016.11.005 903 Stinco, C. M., Fernández-Vázquez, R., Escudero-Gilete, M. L., Heredia, F. J., 904 Meléndez-Martínez, A. J., & Vicario, I. M. (2012). Effect of orange juices 905 processing on the color, particle size, and bioaccessibility of carotenoids. Journal 906 of Agricultural and Food Chemistry, 60(6), 1447–1455. 907 https://doi.org/10.1021/jf2043949 908 Stinco, C. M., Fernández-Vázquez, R., Heredia, F. J., Meléndez-Martínez, A. J., & 909 Vicario, I. M. (2013). Bioaccessibility, antioxidant activity and colour of 910 carotenoids in ultrafrozen orange juices: Influence of thawing conditions. LWT - 911 Food Science and Technology, 53(2), 458–463. 912 https://doi.org/10.1016/j.lwt.2013.04.003 913 Wibowo, S., Vervoort, L., Tomic, J., Santiago, J. S., Lemmens, L., Panozzo, A., … Van 914 Loey, A. (2015). Colour and carotenoid changes of pasteurised orange juice during 915 storage. Food Chemistry, 171, 330–340. 916 https://doi.org/10.1016/j.foodchem.2014.09.007 917 918 4 Figure 3 PHYTOENE (PT) PHYTOFLUENE (PTF) ζ-Carotene β-Carotene Violaxanthin Antheraxanthin Zeaxanthin Carotenoid analysis (HPLC) Colour analysis PSD: FRESH > FG > RT > MW > PAST BIOACCESSIBILITY: FRESH < FG < RT < MW < PAST FRESH < FG < RT < MW < PAST PINALATE ORANGE JUICE SAMPLES 1. Fresh juice (FRESH) 2. Ultrafrozen juices thawed at room temperature (RT) 3. Ultrafrozen juices thawed in microwave oven (MW) 4. Ultrafrozen juices thawed in the fridge (FG) 5. Pasteurized juice (PAST) Particle Size Distribution (PSD) TEM images DEGRADATION OF SUBSTRUCTURES (TEM): BIOACCESSIBILITY: PT > PTF > ζ-Carotene PT and PTF > Xanthophylls PT and PTF < Xanthophylls CAROTENOID DEGRADATION: Ultrafrozen juices: Pasteurized juices: Graphical abstract Click here to download Figure: Graphical Abstract_Thermal Treatments PT PTF.pptx