Effect of regulated deficit irrigation on quality parameters, carotenoids and phenolics of diverse tomato varieties (Solanum lycopersicum L.)

Pag.1 Effect of regulated deficit irrigation and cluster position on quality parameters, carotenoids and phenolics 1 of diverse tomato varieties (Solanum lycopersicum L.) 2 Elena Coyago-Cruzabc, Mireia Corellad, Carla M. Stincob, Dolores Hernanze, Alfonso Morianaad, Antonio J. 3 Meléndez-Martínezb* 4 5 a. Dpto. Ciencias Agroforestales, Universidad de Sevilla, Escuela Técnica Superior de Ingeniería 6 Agronómica, Carretera de Utrera Km 1, 41013 Sevilla, Spain 7 b. Food Colour & Quality Lab., Department of Nutrition & Food Science, Universidad de Sevilla, Facultad 8 de Farmacia, 41012 Sevilla, Spain 9 c. Carrera de Ingeniería en Biotecnología de los Recursos Naturales, Universidad Politécnica Salesiana, Sede 10 Quito, Campus El Girón, Av. 12 de Octubre, Quito, Ecuador 11 d. Unidad Asociada al CSIC de Uso sostenible del suelo y el agua en la agricultura (US-IRNAS), Crta. de 12 Utrera Km 1, 41013 Sevilla, Spain 13 e. Dpto. Química Analítica, Universidad de Sevilla, Facultad de Farmacia, 41012 Sevilla, Spain 14 *Correspondence to: [email protected] 15 16 17 18 19 20 21 Código de campo cambiado Postprint of: Food Research International (96) 72-83 (2017) Pag.2 Abstract 22 This study aims to evaluate the effects of regulated deficit irrigation (RDI) and of cluster position (CI: first and 23 second cluster; CII: third and fourth cluster; CIII: fifth and sixth cluster) on fruit quality parameters, carotenoids 24 and phenolics in tomatoes. Three common (‘Tigerella’, ‘Palamós’ and ‘Byelsa’) and two cherry varieties 25 (‘Lazarino’ and ‘Summerbrix’) were studied. The results showed that the regulated deficit irrigation with 26 reduction of 40 and 50% in the leaf water potential in common and cherry tomatoes did not affect greatly the 27 organoleptic quality of common tomatoes and ʽSummerbrixʼ., while cherry varieties were significantly affected 28 with the cluster position. In most case, significant changes in the levels of carotenoids were observed depending 29 on the treatment and the cluster position in all varieties. Significant changes with the treatment and no change 30 with the cluster position were observed in phenolic compounds. Thus, in general, increased total carotenoid levels 31 and reduced the content of phenolic compounds were observed. Considering the significance of changes in the 32 levels of these groups of compounds it was concluded that ʽLazarinoʼ was more susceptible to water deficit, 33 whereas ʽSummerbrixʼ and ʽPalamósʼ were more resistant. On the other hand, the organoleptic and functional 34 quality changed with the variety. 35 Graphical abstract 36 37 RDI Control CIII CII CI CIII CII CI Fruit quality Functional compounds Carotenoids Phenolics Freeze-dried sample Pag.3 Highlights 38 Marked differences in some of the parameters studied were found across tomato varieties. 39 Regulated deficit irrigation (RDI) can enhance carotenoids and phenolics levels. 40 Overall RDI and cluster position had no major impact in the fruit commercial quality. 41 Key words: functional foods, hydrosustainable foods, lycopene, phytoene, phytofluene, leaf water potential, water 42 stress 43 1. INTRODUCTION 44 The tomato (Solanum lycopersicum L.) is one of the most important crops and widely consumed either fresh or 45 processed as food. Tomato is a source of fibre, potassium, phosphorous, vitamin C, B and E, carotenoids and 46 phenolic compounds (Ripoll J. , Urban, Brunel, & Bertin, 2016). 47 The functional quality of the tomato regarding carotenoids and phenolic compounds levels is affected by 48 agronomical factors, such as maturation in the plant, post–harvest treatment (Centeno, et al., 2011), collection 49 period, light intensity, temperature of cultivation (Dorais, 2007), CO2, vapour pressure deficit, air quality and 50 pollutants (Torres, Adrews, & Davies, 2006). Carotenoids of tomatoes are of great interest for different reasons, 51 thus, these are in general the primary source of lycopene in the human diet (Rao & Rao, 2007). These are also 52 one of the best dietary sources of the colourless carotenoids phytoene and phytofluene (Meléndez-Martínez, 53 Mapelli, Benítez, & Stinco, 2015) whose study is attracting increasing interest very recently (Meléndez-Martínez, 54 Paulino, Stinco, Mapelli-Brahm, & Xiang-Dong, 2014; Stinco, Heredia, Vicario, & Meléndez, 2016) as they may 55 provide health benefits (Meléndez-Martínez, Mapelli, Benítez, & Stinco, 2015; Meléndez-Martínez, et al., 2013). 56 On the other hand, phenolics have been shown to exhibit antioxidant properties in vitro, and are thought to 57 provide health benefits (Shadini & Ambigaipalan, 2015). 58 The tomato quality is a sum of several factors. The market quality change with water and environmental stress, 59 these can affect the fruit size and sugar content (Vinha, Alves, Barreira, Castro, & Costa, 2014), because they can 60 Pag.4 generate varied responses in the plant, such as the reduction of water potential, turgor, water content, growth and 61 crop productivity (Shao, Chu, Abdul, & Zhao, 2008). Deficit irrigation can be used to save water, This is 62 characterized by watering the root zone with less water and can be useful to improve the quality of some products 63 by enhancing the characteristic sensory attributes, which leads to and a greater level of satisfaction among 64 international consumers (Cano-Lamadrid, et al., 2015). 65 The nutritional quality of the tomato regarding carotenoids and phenolic compounds levels is affected by 66 environmental and other factors, such as maturation in the plant, post–harvest treatment (Centeno, et al., 2011), 67 collection period, light intensity, temperature of cultivation (Dorais, 2007), CO2, vapour pressure deficit, air 68 quality and pollutants (Torres, Adrews, & Davies, 2006). 69 . On the other hand, the increase of the population has provoked greater agricultural demand, with decrease of the 70 available water. Water stress can cause changes in the fruit quality and increase soluble solids content in late 71 growing seasons. It is a usual agronomical techniques for improving quality in the fruit ripening phase crops 72 (Johnstone, Hartz, LeStrange, Nunez, & Miyao, 2005). However, high levels of water stress deficit irrigation in 73 sensitive phenological phases (such as flowering) may reduce the sugar, acid and carotenoid content (Ripoll J. , 74 Urban, Brunel, & Bertin, 2016). Plant measurements have been considered very efficient tools for irrigation 75 scheduling (Turner, 1990). Deficit irrigation based on plant water status (especially leaf water potential) are 76 increasingly used for scheduling deficit irrigation (Corell, y otros, 2016). However, there are few studies on 77 deficit irrigation techniques based on plant water status and cluster position, and its effect on the fruit quality and 78 nutritional content commercial and functional tomato quality. . 79 In this regard, the main aim of the study was to determine the effect of a regulated deficit irrigation treatment 80 (based on crop water status, with physiological measurements, and compared with well-irrigated plants) and the 81 cluster position (CI: first and second cluster; CII: third and fourth cluster; CIII: fifth and sixth cluster) on fruit 82 quality parameters (diameter, weight, soluble solids and colour) as well as in carotenoids and phenolics levels. For 83 this purpose 5 tomato genotypes (three common ʽDaniella’ type and two cherry varieties) were considered. The 84 controlled deficit irrigation, consisted in reducing the water amount applied in the more resistant phases of the 85 Código de campo cambiado Con formato: Inglés (Estados Unidos) Código de campo cambiado Con formato: Inglés (Estados Unidos) Código de campo cambiado Con formato: Fuente: (Predeterminado) Times New Roman, 12 pto, Inglés (Estados Unidos), No revisar la ortografía ni la gramática Pag.5 crop, so as not to affect the production and yield. In this particular case, this technique is based on measures of 86 water status to optimize production, increasing the efficiency of water use. Measures of water status, with 87 thresholds of stress that in theory do not affect the production of the crop were used. This deficit irrigation 88 schedule is devised so as not cause a water stress in the plant that affects negatively the production. 89 2. MATERIALS AND METHODS 90 2.1 Reagents and standards 91 Methanol, trichloromethane and hydrochloric acid were of analytical grade and purchased from Labscan (Dublin, 92 Ireland). HPLC-grade methanol, HPLC-grade acetonitrile (PumChemCID: 6342), HPLC-grade ethyl acetate 93 (PumChemCID: 8857) and formic acid (PumChemCID: 284) were obtained from Panreac (Barcelona, Spain). 94 Water was purified in a NANOpureDlamondTM system (Barnsted Inc., Dubuque, IO). β-carotene (PumChem CID: 95 5280489) was purchased from Sigma-Aldrich (Taufkirchen, Germany) and lutein, lycopene, phytoene and 96 phytofluene were obtained from appropriate sources as described elsewhere (Meléndez-Martínez, Vicario, & 97 Heredia, 2007; Meléndez-Martinez, Stinco, Liu, & Wang, 2013). Quercetin (PumChem CID: 5280804), ferulic 98 acid, caffeic acid, p-coumaric acid (PumChem CID: 637542), were purchased from Sigma-Aldrich (Madrid, 99 Spain). 100 2.2 Plant materials 101 Five tomato (S. lycopersicum) varieties with indeterminate growth were studied, specifically three ‘Daniella’ type 102 varieties (Palamós, Tigerella and Byelsa) and two cherry varieties (Summerbrix and Lazarino) (Figure1). These 103 were provided by Fitó from Spain. All varieties have a red colour except ʽTigerellaʼ that is a striped round 104 medium to small variety. ʽPalamósʼ is a round medium to big variety, ʽByelsaʼ is a pear type of tomato, 105 ʽSummerbrixʼ is a pear small variety and ʽLazarinoʼ is a round variety. They were tested in a greenhouse 106 production at Escuela Técnica Superior de Ingeniería Agronómica (E.T.S.I.A.) in the Universidad de Sevilla 107 (Sevilla, South Spain, 37º21'09.71'' Lat. N, 5º56'19.13'' Long. W, 33 m a.s.l.). They were tested during spring 108 2015 (23rdFebruary to 15thJune). The varieties studied were germinated in seedlings and, when they developed 109 three or four true leaves, they were transplanted and secondary plant stems were pruned. Pest control was 110 Pag.6 performed with insecticides and acaricides specific for tomatoes. The irrigation of the plants was done by 111 dripping, with two daily cycles of irrigation that depended to crop evapotranspiration (ETc) of the plant. The 112 irrigation treatments were: Regulated deficit irrigation (RDI), with a water stress threshold of -1MPa of leaf water 113 potential (55.1 mm of applied water) from flowering , and a control treatment with irrigation requirements 114 determined according to ETc calculated with the FAO Penman-Monteith method Allen et al. (2006) (Allen, 115 Pereira, Raes, & Smith, 2006) (532.2 mm of applied water). The RDI was applied fifteen days after 116 transplantation. Each parcel was made up of 3 rows; tomatoes samples were collected at central rows plants. and 117 the main aspects of tomatoes were estimated on the plants of the central row. All the clusters were harvest at the 118 same time. Fruit harvest for the analysis was performed at full red ripening on June 21th. 119 Figure 1. Photograph of five tomato … 120 121 Samples included fruits representative of the three different clusters. The first and second clusters corresponded to 122 position I (CI), the third and fourth clusters corresponded to position II (CII), and the fifth and sixth clusters 123 corresponded to position III (CIII). Three fruits of seven plants for cherry varieties and one fruit of seven plants 124 for common varieties were collected at each position. These were selected from different parts of the cluster. In 125 most cases, fruit in the CI position was selected from the distal part, CII from medium part and CIII proximal part 126 of the cluster. Samples included a mix of fruits of each level previously characterised. This mix was divided into 127 two samples for the quantification of functional quality. The seeds and inside locular tissues were removed, cut 128 and quickly frozen at -80 °C, before being freeze-dried with a Cryodos system (Telstar, Japan). The water content 129 was determined by weighing the sample before and after freeze-drying. 130 The dried samples were ground in a basic IKA A 11 mill, then stored in a dark glass bottle and hermetically 131 sealed under nitrogen atmosphere. The samples were stored in a freezer at -21 °C until their analysis. 132 2.3 Measurements performed on the fruits 133 The measurements performed on the fresh tomato quantified equatorial and longitudinal diameter (cm), weight 134 (g), soluble solids (°Brix) and colour. The soluble solids were measured using a Hand-refractometer RHC- 135 200ATC (Huake, China), and the fruit colour was analysed using a CM-700d colorimeter (Minolta, Japan). For 136 Pag.7 this purpose the whole visible spectrum (380 – 770nm) was recorded with a bandwidth of 1nm. The colour 137 parameters corresponding to the uniform colour space CIELAB were obtained directly from the apparatus. 138 Illuminant D65 and 10° observer were considered as reference. 139 2.4 Carotenoid analysis 140 Sample preparation 141 Individual carotenoids were determined as described by Borghesi, et al. (2011) (Borghesi, et al., 2011) with slight 142 modifications. Approximately 20 mg of homogenized freeze-dried powder were used for the extractions. The 143 powder was mixed with 250 μL of methanol, 500 μL of trichloromethane and 250 μL of MiliQ-water and then 144 vortexed, sonicated for 2 min and centrifuged to remove the aqueous phase at 14 000x g for 3 min. After 145 recovering the coloured fraction, 500 μL of trichloromethane were added, and then the mixture was vortexed, 146 sonicated and centrifuged again. These operations were repeated until the solids did not have colour. The organic 147 coloured fractions were evaporated to dryness at a temperature below 30 °C in a vacuum concentrator and stored 148 under N2 at -20 °C until the analysis. 149 RRLC chromatography 150 The dry residue was re-dissolved in 40 μL of ethyl acetate prior to their injection in the RRLC system. The RRLC 151 analysis was carried out using the method reported by Stinco et al. on an Agilent 1260 system equipped with a 152 diode-array detector, C18 Poroshell 120 column (2.7 μm, 5 cm x 4.6 mm) (Agilent, Palo Alto, CA), 2.5 μL of the 153 injection volume at 30°C and 1 mL/min of the mobile phase (Stinco, Benítez, Hernanz, Vicario, & Meléndez, 154 2014). The mobile phase consisted of acetonitrile (solvent A), methanol (solvent B) and ethyl acetate (solvent C) 155 with the linear gradient elution: 85% A +15% B, 0 min; 60%A +20%B, + 20%C, 5 min; 60%A+20%B+20%C, 7 156 min; 85% A+ 15% B , 9 min; 85% A + 15% B, 12 min. The chromatograms were monitored at 285, 350 and 450 157 nm with the open lab ChemStation software. Carotenoids were identified by comparison of their chromatographic 158 and UV-vis spectroscopic characteristic with those of standard. The quantification of the carotenoids was 159 performed by external calibration from the areas of the chromatographic peaks obtained by Uv detector at the 160 following wavelengths: 285 nm for phytoene, 350 nm for phytofluene and 450 nm for lutein, lycopene and β- 161 Pag.8 carotene. All dry residues re-dissolved were injected twice and the concentration expressed in μg/g dry weight 162 (DW). Total carotenoids contents were calculated adding up all individual carotenoids. 163 2.5 Analysis of phenolic compounds 164 Sample preparation 165 Approximately 0.5 g of the freeze-dried was extracted with 15 mL of acidified methanol 0.1%. The mixture was 166 vortexed and sonicated for 15 min, and centrifuged at 4190 g for 7 min at 4 °C; the supernatant was collected and 167 the residue subjected to the same process twice, using only 5 mL of acidified methanol 0.1%. Finally, all the 168 supernatants were combined. The extract was stored at -20 °C until analysis. All the samples were extracted in 169 triplicate. 170 UHPLC analysis 171 One mL of the extract obtained was dissolved in 4 mL of 0.01% formic acid in water for injection in the UHPLC 172 system. Chromatography analyses were carried out in an Agilent 1290 chromatograph equipped with a diode- 173 array detector (Agilent Technologies, Palo Alto, CA. USA), an Eclipse Plus C18 column (1.8 um , 2.1 x 5 mm), 5 174 μL of injection volume, 1 mL/min of the mobile phase at 30°C and peaks were monitored at 220-500 nm. The 175 mobile phase consisted of 0.01% of formic acid in water (solvent A) and acetonitrile (solvent B) with the linear 176 gradient elution: 100% A, 0 min; 95%A +5%B, + 20%C, 5 min; 50%A+50%B, 20 min; washing and re-balancing 177 of the column, 22 min. The chromatograms were monitored at 280 and, 320 and 370 nm with the open lab 178 ChemStation software. Phenolic compounds were identified by their retention time and UV-vis spectra by 179 comparison with standards. The quantification of phenolic acids was achieved by comparison of their retention 180 times and spectra with those of appropriate standards, their levels being determined by external calibration 181 considering the following wavelengths: 280 nm for p-hydroxibenzoic acid, p-coumaric acid, caffeic acid, 182 chlorogenic acid, ferúlico acid, naringin and crisin; and 320 nm for quercetrin and quercetin . All the extracts 183 were injected twice and the concentration expressed in mg/100 g (DW). Total phenolic content was calculated 184 adding up all individual phenolics. 185 Con formato: Fuente: 12 pto, Cursiva Con formato: Fuente: 12 pto, Cursiva Pag.9 2.6 Statistical analysis 186 Statistical differences were determined by analysis of variance (simple and factorial ANOVA). The mean 187 separation was made via a Tukey’s test with 0.01 significant differences and correlations by Pearson with 95% 188 confidence level were employed to estimate the possible significance of the effect. The STATGRAPHICS 189 Centurion XVII software was used for statistical analyses. 190 3. RESULTS AND DISCUSSION 191 Regulated deficit irrigation affected plant water status significantly during the experiment. In the vegetative 192 development of the plant, the leaf water potentials in RDI ranging from -0.8 to -1.2 MPa, while in the control 193 ranging between -0.2 to -0.5 MPa, were observed. At harvest (June 21th) the leaf water potential was -1 MPa in 194 the RDI treatment and -0.6 MPa in the control sample for ʽPalamósʼ and ʽByelsaʼ; -1,2 MPa in the RDI treatment 195 and -0.7 MPa in the control sample for ʽTigerellaʼ and -1 MPa in the RDI treatment and -0.5 MPa in the control 196 sample for ʽLazarinoʼ and ʽSummerbrixʼ. These data showed that during experimentation, the RDI in the plants of 197 common and cherry tomatoes exhibited a reduction of 40 and 50%, respectively, of leaf water potential. 198 On the other hand, at harvest the plant height for ʽLazarinoʼ, ʽSummerbrixʼ and ʽTigerellaʼ was of 2.3 m, while 199 ʽPalamosʼ showed a height of 2 m and ʽByelsaʼ of 1.5 m. Thus, cherry varieties showed an average of 38 200 tomatoes per cluster, while ʽTigerellaʼ 16 fruits per cluster, ʽPalamósʼ 7 fruits per cluster and ʽByelsaʼ 5 fruits per 201 cluster. However, in this study it was observed that the load was not affected with the application of RDI. 202 The tomato quality data (diameter, weight, soluble solids and colour) are summarized in Table 1. 203 3.1 Estimation of fruit quality 204 3.1.1 Size 205 Fruit size keeps similarity with the data reported in other studies for cocktail tomato with normal water regime. 206 Thus, length values between 30 and 42 mm (Ripoll J. , Urban, Brunel, & Bertin, 2016; Choi S. , et al., 2014) have 207 been reported elsewhere. In most cases, the diameter of tomatoes did not change significantly with the treatment, 208 Pag.16 indicating that reduced irrigation and water stress can heighten the levels of lycopene (Wang, Kang, Du, Li, & 358 Qiu, 2011; Pernice R. , et al., 2010). However other authors reported that the behaviour of lycopene with respect 359 to the treatment was dependent on the variety in cherry tomatoes. Thus, in some cases the treatment led to 360 increases and in some others to decreases (Sánchez-Rodríguez R. , Leyva, Constán-Aguilar, Romero, & Ruiz, 361 2012). In the case of common varieties, increased levels of lycopene with the treatment have been reported by 362 other authors (Gahler, Otto, & Bphm, 2003). The cluster position led to statically significant differences in all 363 varieties in both RDI and control samples. In the case of ʽLazarinoʼ (RDI samples) and ʽSummerbrixʼ (control 364 samples), the levels of lycopene decreased with cluster position. Such levels were higher in the CI cluster in RDI 365 samples in cherry and common varieties. Hence, the lycopene values changed with the treatment in all varieties, 366 except in ʽPalamósʼ, while the cluster position changed in all varieties. Overall, the treatment, cluster position and 367 variety influenced the lycopene values in this study. 368 The levels of β-carotene in cherry and common varieties in normal water regime samples (ranging from 5.9 to 369 16.9 mg/100 g DW) were similar to those found in some studies (Meléndez-Martínez, Fraser, & Bramley, 2010; 370 Cortés-Olmos, Leiva-Brondo, Roselló, Raigón, & Cebolla-Cornejo, 2014) and higher compared to those reported 371 in others, which ranged from 10.0 to 25.1 µg/g DW (Verheul, Slimestad, & Holta, 2015; Choi S. , et al., 2014). 372 The treatment led to significant differences in some clusters in cherry and common varieties, with decreases in 373 ʽTigerellaʼ in all clusters. The results obtained in RDI samples agree well with those reported by other authors in 374 cherry varieties and common varieties, indicating that in the cherry varieties the amount of β-carotene increase or 375 decrease depending on the variety (Sánchez-Rodríguez R. , Leyva, Constán-Aguilar, Romero, & Ruiz, 2012; 376 Pernice R. , et al., 2010), while common varieties only decreased (Favati, et al., 2009). With respect to the cluster 377 position, cherry varieties showed statically significant differences in RDI and control samples and ʽTigerellaʼ only 378 in RDI. The β-carotene values increased with cluster position in ʽLazarinoʼ in RDI. In addition, β-carotene 379 exhibited changes with the treatment in common varieties, although the cluster position did not change in 380 ʽLazarinoʼ and ʽTigerellaʼ. Overall, β-carotene was largely influenced by the treatment and variety. 381 Table 2. Average values of carotenoids (mg /100 g DW) in cherry and common tomatoes studieda 382 Pag.17 Significant changes in the levels of all the studied carotenoids were observed in the five tomato varieties 383 depending on the treatment and the cluster position. Cherry varieties increased phytofluene and lycopene 384 concentrations with treatment and ʽTigerellaʼ decreased lutein, lycopene and β-carotene levels. Concerning the 385 cluster position, individual carotenoids in common and cherry varieties exhibited significant difference for RDI 386 and control treatments. This is another example that the carotenoid content is not only affected by the lack of 387 water, but also by varietal, agronomic and environmental factors, as noted in other studies (Pernice R. , et al., 388 2010). The results of this study showed that the treatment, the variety and the cluster position significantly affect 389 the individual carotenoid content, causing an increase in most tomatoes under study. 390 3.3. Phenolic analysis 391 3.3.1. Total phenolic content 392 Quantitattive data about total phenolic content (TPC) are presented in Figure 3 and those for the individual 393 compounds in Table 3. 394 In some cases, the TPC in cherry varieties (ranging from 209 to 350 mg/100 g DW) were higher and in others 395 similar to those reported in common tomatoes (ranging from 210 to 286 mg/100 g DW). These data agree well 396 with those of other authors, for example Vinha et al. (2014) (Vinha, Alves, Barreira, Castro, & Costa, 2014), who 397 studied four tomato cultivars and reported TPC values between 54 to 67 mg gallic acid equivalent (GAE)/100 g 398 FW and Cortés-Olmos et al. (2014) (Cortés-Olmos, Leiva-Brondo, Roselló, Raigón, & Cebolla-Cornejo, 2014) , 399 who surveyed a collection of 126 populations of 16 traditional varieties and reported values between 30 to 79 mg 400 caffeic acid equivalent (CAE)/100 g FW. 401 On the other hand, TPC in common varieties in this study were higher compared with those found in others, 402 which reported variations of concentration from 18 to 31 mg GAE/100 g FW (Gahler, Otto, & Bphm, 2003) or 403 from 26 to 42 mg/100 g FW (Minutolo, Amalfitano, Evidente, Frusciante, & Errico, 2013). This may be due, at 404 least in part, to the fact that in this study acidified methanol was used and this favors the extraction of phenolics 405 according to the findings of some authors (Ju & Howard, 2003; Arranz, 2010). On the other hand, the freeze-dried 406 Pag.18 increase between 2.6 to 5.9% the amounts of TPC accordin to other authors which study the effects of different 407 drying procesees (Ching-Hi, Hsing-Yu, Chi-Yue, & Yung-Chuan, 2006). The data corresponding to cherry and 408 common varieties agree well with those found by other authors, who reported phenolic concentration ranges of 65 409 to 440 mg/100 g DW of 12 cherry varieties and 51 to 54 mg/100 g FW for three common tomatoes (Choi S. , et 410 al., 2014; Raiola, et al., 2015) . 411 Some significative diferences in the TPC values were observed as a result of the treatment in all varieties and 412 some clusters. In cherry varieties, the treatment caused an overall reduction in the total phenolic content, although 413 in ʽByelsaʼ it increased. Phenolic compounds are thought to have a strong link to the cultivar and water content of 414 the fruit (Minutolo, Amalfitano, Evidente, Frusciante, & Errico, 2013; Sánchez-Rodríguez, Ruiz, Ferreres, & 415 Moreno, 2012), which could explain in part these results. On the other hand, these data agree well with those of 416 other authors. For example Pernice et al. (2010) studied two tomato cherry varieties in normal, reduced and lack 417 of irrigation and concluded that the varieties had higher contents of phenolics in reduced irrigation, while a non 418 consistent behaviour was observed for the non irrigated samples (Pernice R. , et al., 2010); this behavior was also 419 noticed in other studies (Sánchez-Rodríguez R. , Leyva, Constán-Aguilar, Romero, & Ruiz, 2012; Sánchez- 420 Rodríguez, Ruiz, Ferreres, & Moreno, 2012). Again, the data demonstrate that the genotype has an important 421 effect on the levels of the compounds (Minutolo, Amalfitano, Evidente, Frusciante, & Errico, 2013). 422 As far as the cluster position was concerned, significant differences were observed in RDI and control samples in 423 all varieties. The concentration of total phenolic compounds was highest in CIII for cherry varieties and 424 ʽTigerellaʼ. However, the highest values were found in samples of the Byelsa variety in CI. The TPC values in 425 cherry varieties and ʽTigerellaʼ increased from CI to CIII. This could be due to some extent to the fact that 426 phenolics are components of plant tissues, and their distribution is influenced by light and environmental 427 conditions. As an example, some studies suggest that leaves growing under shade have a low concentration of 428 phenolics (Minutolo, Amalfitano, Evidente, Frusciante, & Errico, 2013; Rugna, Ricco, Gurni, & Wagner, 2008), 429 which may help explain, at least in part, the high TPC in the CIII cluster. 430 Pag.19 Hence, the treatment showed only change in ʽLazarinoʼ and the cluster position changed in all varieties. Overall, 431 the variety and cluster positon influenced the TPC values. 432 Figure 3. Total phenolic content (mg/100 g DW).. 433 3.3.2. Phenolics 434 In the five varieties studied, hydroxybenzoic acid derivatives (p-hydroxybenzoic acid, tR 0.5 min), 435 hydroxycinnamic acid derivatives (p-coumaric, tR 6.2 min,; caffeic, tR 4.5 min,; chlorogenic, tR 5.38 min,; ferulic, 436 tR 7.4 min), flavanones (naringin, tR 9.48 min), flavones (crisin, tR 16.5 min), flavonols (quercetrin, tR 8.9 min and 437 quercetin, tR 10.5 min) were quantified (Table 3 and Figure 4). The levels in control samples (ranging from 209 to 438 350 mg/100 g DW) were similar to those reported by other authors. For example Choi et al. (2014) studied 12 439 samples of cherry tomato and reported sums of individual phenolic compounds of 65 to 440 mg/100 g DW (Choi 440 S. , et al., 2014). 441 The results of the present study showed that p-hydroxybenzoic acid, p-coumaric acid, chlorogenic acid, and 442 quercetin were the predominant acids in common and cherry varieties. As it can be readily observed in Table 3, 443 there were statistically significant differences with treatment in some clusters. 444 p-Hydroxybenzoic acid levels changed significantly with the treatment in all varieties and all clusters with 445 decreases in ʽSummerbrixʼ and increases in ʽTigerellaʼ. Regarding the cluster position, all varieties exhibited 446 statically significant differences in RDI and control samples, with decreases from CI to CIII in ʽLazarinoʼ, 447 ʽSummerbrixʼ and ʽPalamósʼ varieties. In addition, p-hydroxybenzoic did not change with the treatment in some 448 varieties, while the cluster position exhibited changes in all the varieties. 449 p-Coumaric acid levels varied significantly with treatment in all varieties and all clusters except in the case of CI 450 in ʽLazarinoʼ. In ʽSummerbrixʼ, ʽTigerellaʼ and ʽPalamósʼ the treatment increased the concentration, while the 451 contrary was observed in ʽByelsaʼ. On the other hand, all the varieties showed significant difference in RDI and 452 control samples with higher values in CIII of RDI samples in ʽLazarinoʼ, ʽSummerbrixʼ and ʽTigerellaʼ. In 453 Pag.20 ʽTigerellaʼ RDI samples, p-coumaric acid values increased from CI to CIII. Hence, p-coumaric acid with the 454 treatment and cluster position changed in all varieties. 455 Raiola et al. (2015) reported in normally irrigated common varieties ranges between 0.92 and 1.80 mg/100 g FW 456 for caffeic acid, 15.63 and 23.16 mg/100 g FW for chlorogenic acid, 0.25 and 1.17 mg/100 g FW for quercetin 457 and 0.22 and 0.51 mg/100 g FW for naringenin. These values were similar (chlorogenic acid) or low (caffeic acid, 458 quercetin, naringenin) to the ones found in the present study (Raiola, et al., 2015). Caffeic acid concentrations 459 changed significantly with the treatment in all varieties and clusters, except in CIII for ʽLazarinoʼ. The treatment 460 decreased the levels in ʽSummerbrixʼ, ʽTigerellaʼ and ʽByelsaʼ and increased them in ʽPalamósʼ. With respect to 461 the cluster position, all varieties except ʽSummerbrixʼ in RDI showed statically significant difference. 462 ʽSummerbrixʼ, ʽLazarinoʼ and ʽPalamósʼ decreased with cluster position in control. These data agree well with 463 those of other authors, indicating no clear behaviour of caffeic acid with respect cluster position because this 464 changed depending of tomato variety (Arranz, 2010). In addition, the caffeic acid changed with the treatment in 465 all varieties, while the cluster position did not change in ʽTigerellaʼ. 466 The levels of chlorogenic acid changed significantly with the treatment in all varieties and clusters, except in the 467 case of CII samples in the variety ʽPalamósʼ. In this study, the concentration of this compound decreased with the 468 treatment in ʽTigerellaʼ and increased in ʽPalamósʼand ʽByelsaʼ. These data agreed well with those reported by 469 Sánchez-Rodíguez (2012), which showed both increases and decreases of chlorogenic acid with water stress in the 470 two tomato varieties studied (Sánchez-Rodríguez, Ruiz, Ferreres, & Moreno, 2012). Regarding the cluster 471 position, all the varieties exhibited significant differences and increases from CI to CIII. These observations 472 contrasted with those of other authors. For example Minutolo et al. (2013) reported that chlorogenic acid levels 473 decreased at the top of the plant (Minutolo, Amalfitano, Evidente, Frusciante, & Errico, 2013). Again these 474 differences could also be related with changes in the environmental conditions and harvest. In addition, ʽLazarinoʼ 475 did not change the chlorogenic acid values with the treatment, while the cluster position changed in all varieties. 476 The levels of ferulic acid changed significantly with the treatment in some clusters without presenting a defined 477 trend. ʽSummerbrixʼ, ʽTigerellaʼ and ʽPalamósʼ control samples showed the highest value in the CII cluster. These 478 Pag.21 findings agree well with those of other authors, suggesting that there is not a consistent behaviour of ferulic acid 479 with respect to cluster position as this changed depending on the tomato variety (Minutolo, Amalfitano, Evidente, 480 Frusciante, & Errico, 2013). Overall, the ferulic acid showed change with the treatment and this did not change 481 with the cluster position only in ʽSummerbrixʼ. 482 The concentration of naringin varied significantly with treatment in all the varieties except ʽPalamósʼ in CII and 483 ʽLazarinoʼ in the CI and CIII clusters. The treatment increased its concentration in ʽTigerellaʼ. These data are in 484 agreement with those of other authors. For example Pernice, et al. (2010) reported that reduced irrigation 485 increased naringin levels but stress decreased them (Pernice R. , et al., 2010). These authors reported naringin 486 levels ranging between 3.76 and 6.30 mg/100 g FW in normal irrigated samples, these values being higher than 487 those found in this study (ranging from 8.6 to 35.3 mg/100 g DW). Naringin values in RDI varied significant 488 with the cluster position in all varieties. ʽSummerbrixʼ and ʽTigerellaʼ showed high levels of naringin in the CIII 489 cluster. In addition, only ʽSummerbrixʼ did not change in the naringin values with the treatment, while ʽLazarinoʼ 490 and ʽPalamósʼ with the cluster position. 491 The treatment led to significantly changes in the levels of crisin in CIII in the case of ʽSummerbrixʼ in CI, CII and 492 CIII in the case of ʽTigerellaʼ, CII in the case of ʽPalamósʼ and CI in the case of ʽByelsaʼ. The treatment led to 493 decreases in ʽTigerellaʼ and ʽByelsaʼ. Regarding the effect of the cluster position, ʽSummerbrixʼ and ʽPalamósʼ in 494 control showed statically significant difference and high crisin value in CII, whileʽTigerellaʼ in RDI presented 495 statically significant difference and high value in the CIII cluster. Overall, ʽPalamósʼ and ʽByelsaʼ exhibited 496 change in the crisin values with the treatment, while cherry varieties did not change with the cluster position. 497 The treatment led to statically significant differences in the levels of quercetrin in all the varieties and some 498 clusters. Decreases in ʽTigerellaʼ and increases in ʽPalamósʼ were observed. With respect to the cluster position, 499 quercetrin showed statically significant difference in RDI and control in all varieties and the quercetrin value 500 increased from CI to CIII. In addition, only ʽByelsaʼ did not change with the treatment, while all varieties changed 501 with the cluster position. 502 Pag.22 Regarding quercetin, the treatment caused significant differences in its concentration in all the varieties except in 503 the case of CII and CIII in the ʽSummerbrixʼ variety. Decreases in all the varieties were observed, except in the 504 case of ʽByelsaʼ variety which the levels increased. These data contrast with those found by other authors. Thus, 505 Pernice, et al. (2010) concluded that both reduced irrigation (1.1 for normal irrigation and 2.1 mg/100 g FW for 506 reduced irrigation in the biotype PC01) and stress water (1.7 normal irrigation and 2.2 mg/100 g FW for none 507 irrigation in the biotype PC05) increased quercetin concentrations (Pernice R. , et al., 2010). They reported levels 508 of quercetin ranging between 1.07 and 1.71 mg/100 g FW in samples subjected to normal irrigations. These 509 values keep relationship with those found in this study (ranging from 37.6 to 99.0 mg/100 g DW). This 510 disagreement in the results of both studies can be due to a great extent to the fact that different varieties were 511 surveyed. The cluster position had a significant effect in both RDI and control samples in all varieties. The levels 512 increased from CI to CIII in all cases but ʽTigerellaʼ control samples, where the highest concentration was found 513 in CII. In addition, the treatment and the cluster position led to changes in the quercetin values. 514 Noticeable differences as a function of the variety were also observed. Thus for sample, the major phenolics in  515 Lazarin were quercetin and p-hydroxybenzoic acids, in this order. Their highest levels were found in CI and CII 516 independent of the treatment (Table 3). Overall, p-hydroxybenzoic acid, p-coumaric acid, ferulic acid, naringin 517 and crisin were largely influenced by the treatment and variety, while caffeic acid, quercetin and quercetrin only 518 by the variey and finally clorogenic acid by the cluster position and variety. 519 Interestingly, the treatment led to noticeable decreases in the levels of quercetin in all clusters. These increases 520 were particularly noteworthy (> 3-fold) in CIII and contributed to that the highest levels of phenolics were 521 detected in the samples corresponding to cluster position and treatment (Figure 32). On the other hand, several 522 studies indicate that the soluble solids values may be influenced by phenolics, in special flavonols, that can inhibit 523 glycolysis and improve gluconeogenesis (Stakhova, Ladygin, & Stakhov, 2001). Considering the treatment, this 524 relationship was found in all the varieties studied, especially in ʽLazarinoʼ in CI and CIII between SS and 525 quercetrin, ʽSummerbrixʼ in all clusters between SS and quercetrin and ʽTigerellaʼ in all clusters between SS with 526 quercetrin and quercetin. In ʽTigerellaʼ it was observed that, while SS increased, quercetrin and quercetin 527 Pag.23 decreased. With respect to the cluster position, cherry varieties RDI samples showed a relationship between SS 528 and flavonols. Thus, in ʽLazarinoʼ SS decreased and flavonols increased while in ʽSummerbrixʼ both SS and 529 flavonols increased. 530 Table 3. Average values of phenolic compounds (mg /100 g DW) .. 531 In summary, the present study supports the idea that regulated deficit irrigation did not change the fruit 532 organoleptic quality and enhance the levels of some secondary metabolites like certain carotenoids and phenolics. 533 In general, it can be concluded that regulated deficit irrigation with reduction of 40 and 50% in the leaf water 534 potential in common and cherry tomatoes and the cluster position had no major impact in the fruit quality except 535 in ʽLazarinoʼ but this did have a noticeable impact in functional quality.. The treatment did not allow: 1) 536 significant changes on equatiorial diameter, SS, a* and C*ab values; 2) lycopene and hence total carotenoids; 3) 537 caffeic acid, chlorogenic acid, quercetrin, quercetin and hence total phenolics. Thus, strong varieties to treatment 538 were observed. For example in most cases, ʽSummerbrixʼ did not change the organoleptic parameters; ʽPalamósʼ 539 in total carotenoid while ʽSummerbrixʼ, ʽTigerellaʼ and ʽPalamósʼ in total phenolic compounds. It was also 540 observed that cluster position did not lead to changes in the weight, phytoene, β-carotene and all individual 541 phenolics except chlorogenic acid. The organoleptic and functional quality exhibited greater influence with the 542 variety. 543 Overall, it can be concluded that the treatment and cluster position affected the organoleptic quality parameters to 544 the greatest extent in ‘Lazarino’, the carotenoids levels in cherry varieties and ʽTigerellaʼ and phenolics in all 545 varieties. 546 547 ABBREVIATIONS USED 548 E.T.S.I.A., Escuela Técnica Superior de Ingeniería Agronómica; a.s.l., above sea level; RDI, regulated deficit irrigation; 549 ETc, crop evapotranspiration; FAO, Food and Agriculture Organization of the United Nations; CIELAB, the Commission 550 International of IEclairage (CIE), defined colour spaces that includes CIE L*a*b*; UV-vis, ultraviolet-visible; RRLC, rapid 551 Pag.24 resolution liquid chromatography; UHPLC, ultra performance liquid chromatography; SS, soluble solids; TC, total 552 carotenoids; TPC, total phenolic content; FW, fresh weight; DW dry weight; 553 AUTHOR INFORMATION 554 Corresponding Author 555 *Phone: +34-954557017 556 E-mail: ajmelende[email protected] 557 Acknowledgements 558 The authors want to thank the Secretaría Nacional de Educación Superior, Ciencia, Tecnología e Innovación (SENESCYT) - 559 Ecuador for its financial support and the Technical Staff of the Service of Biology (SGI, Universidad de Sevilla). AJMM 560 acknowledges funding from the Spanish State Secretariat of Research, Development and Innovation (Ministry of Economy 561 and Competitiveness, project ref. AGL2012-37610, co-funded by FEDER). ECC, DH, CMS and AJMM thank the Ibero- 562 American Programme for Science, Technology and Development (CYTED, http://www.cyted.org) for the funding of the 563 IBERCAROT network (http://carotenoides.us.es, ref. 112RT0445). AJMM acknowledges funding from Carotenoid 564 Network: from microbia and plants to food and health (CaRed), funded by the Spanish Ministry of Economy and 565 Competitiveness (BIO2015-71703-REDT). Quality technological assistance from Ms. Ana Benítez is acknowledged. 566 567 Notes 568 The authors declare that there are no competing financial interests 569 570 REFERENCES 571 Allen, R., Pereira, L., Raes, D., & Smith, M. (2006). Evapotranspiración del cultivo. Guías para la determinación 572 de los requerimientos de agua de los cultivos. Rome: Food and Agriculture Organization of the United 573 Nations. 574 Código de campo cambiado Código de campo cambiado Pag.25 Arranz, S. (2010). Compuestos polifenólicos (Extraíbles y no Extraíbles) en alimentos de la dieta española: 575 Metodología para su determinación e identificación. Spain: Complutense University of Madrid. 576 Pharmacy faculty. Department of Bromatology II. 577 Beckles, D. (2012). Factors affecting the postharvest soluble solids and sugar content of tomato (Solanum 578 lycopersicum L. ) fruit. Postharvest biology and technology 63, 129-140. 579 Borghesi, E., González, M., Escudero, M., Malorgio, F., Heredia, F., & Meléndez, A. (2011). Effects of salinity 580 stress on carotenoids, anthocyanins, and color of diverse tomato genotypes. Journal of agricultural 581 and food chemistry(59), 11676-11682. 582 Camejo, D., & Torres, W. (2001). High temperature effect on tomato (Lycopersicon esculentum) Pigment 583 and Protein Content and Cellular Viability. Cultivos Tropicales, 22(3), 13-17. 584 Cano-Lamadrid, M., Girón, L., Pleite, R., Burló, F., Corell, M., & Moriano, A. (2015). Quality attributes of table 585 olives as affected by regulated deficit irrigation. LWT-Food Science and Technology, 62, 19-26. 586 Centeno, D., Osorio, S., Nuness, A., Bertolo, A., Cameiro, R., Araújo, W., . . . Femie, A. (2011). Malate plays a 587 crucial role in starch metabolism, ripening, and soluble solid content of tomato fruit and effects 588 postharvest softening. The plant cell, Vol. 23, 162-184. 589 Ching-Hi, C., Hsing-Yu, L., Chi-Yue, C., & Yung-Chuan, L. (2006). Comparisons on the antioxidant properties 590 of fresh, freeze-dried and hot-air-dried tomatoes. Journal of food engineering, 77, 478-485. 591 Choi, S., Kim, D., Kozukue, N., Kim, H., Nishitani, Y., Mizuno, M., . . . Friedman, M. (2014). Protein, free amino 592 acid, phenolic, B-carotene, and lycopene content, and antioxidative and cancer cell inhibitory effects 593 of 12 greenhouse-grown commercial cherry tomato varieties. Journal of food composition and 594 analysis 34, 115-127. 595 Corell, M., Pérez-López, D., Martín-Palomo, M., Centeno, A., Girón, I., Galindo, A., . . . Moriana, A. (2016). 596 Comparison of the water potential baseline in different locations. Usefulness for irrigation 597 scheduling of olive orchards. Agricultural Water Management, 177, 308-316. 598 Cortés-Olmos, C., Leiva-Brondo, M., Roselló, J., Raigón, M., & Cebolla-Cornejo, J. (2014). The role of 599 traditional varieties of tomato as sources of funtional compounds. J Sci Food Agric, 94, 2888-2904. 600 Dorais, M. (2007). Effect of cultural management on tomato fruit health qualities. FAV Health 82005). QC, 601 Canadá. Acta Hort. 744, 279-293. 602 Favati, F., Lovelli, S., Galgano, F., Miccolis, V., DiTommaso, T., & Candido, V. (2009). Processing tomato 603 quality as affected by irrigation scheduling. Scientia horticulturae 122, 562-571. 604 Figás, M., Prohens, J., Raigón, M., Fita, A., García-Martínez, M., Casanova, C., . . . Andùjar, I. S. (2015). 605 Characterization of composition traits related to organoleptic and functional quality for the 606 differentiation, selection and enhancement of local varieties of tomato from different cultivar 607 groups. Food Chemistry, 187, 517-524. 608 Pag.32 Figure1. Photographs of the tomato varieties studied. A) ʽLazarinoʼ, B) ʽSummerbrixʼ, C)ʽTigerellaʼ, D) ʽByelsaʼ and E) ʽPalamósʼ Figure 21. Total carotenoids (mg/100 g DW) in the common and cherry tomatoes studied. Mean values in the same cluster for RDI and control columns followed by the same letter do not differ significantly at confidence level 95% and significance of differences between cluster height is given: n.s., not significant; *,P<0.1; **, P<0.01; ***, P<0.001. The values are means + SE (n = 4) RDI - CI RDI - CII RDI - CIII C- CI C - CII C - CIII 0 20 40 60 80 100 120 140 160 ´Lazarino` ´Summerbrix` a a a bb b a a a bbb *** *** ** n.s. RDI - CI RDI - CII RDI - CIII C- CI C - CII C - CIII 0 50 100 150 200 ´Tigerella` ´Palamós` ´Byelsa` b b b a a a a b ba a b ** *** *** * Con formato: Fuente: (Predeterminado) Times New Roman, 10 pto Pag.33 Figure 2. Cromatogram of tomato carotenoids at 285 , 350 and 450 nm of the variety Tigerella. 1. Phytoene; 2. Phytofluene; 3. Lutein; 4. Lycopene; 5. β-carotene Figure 3. Total phenolic content (mg/100 g DW) in the common tomato and cherry tomatoes studied. Mean values in the same cluster of RDI and control column followed by the same letter do not differ significantly at confidence level 95% and significance of differences between cluster height is given: n.s., not significant; *,P<0.1; **, P<0.01; ***, P<0.001. The values are means + SE (n = 6) RDI - CI RDI - CII RDI - CIII C- CI C - CII C - CIII 0 100 200 300 400 ´Lazarino` 'Summerbrix´ b b a a b aa a a b a a *** *** *** *** RDI - CI RDI - CII RDI - CIII C- CI C - CII C - CIII 0 100 200 300 400 `Tigerella´ `Palamós´ `Byelsa´ aa aa *** *** b a ba a b *** *** a b Pag.34 Figure 4. Cromatogram of tomato phenolics at 280 and 320 nm of the variety Tigerella. 1. p-Hydroxybenzoic acid; 2.Caffeic acid; 3. Chlorogenic acid; 4. p-Coumaric acid; 5. Ferulic acid; 6. Naringin; 7. Crisin; 8. Quercetrin; 9. Quercetin