Impaginato 407 Adv. Hort. Sci., 2018 32(3): 407-420 DOI: 10.13128/ahs-22856 Quality and nutraceutical properties of mango fruit: influence of cultivar and biological age assessed by Time‐resolved Reflectance Spectroscopy M. Vanoli 1, M. Grassi 1, L. Spinelli 2, A. Torricelli 2, 3, A. Rizzolo 1 (*) 1 Consiglio per la ricerca in agricoltura e l’analisi dell’economia agraria, Centro di ricerca Ingegneria e Trasformazioni Agroalimentari (CREA- IT), sede di Milano, Via G. Venezian, 26, 20133 Milano, Italy. 2 Istituto di Fotonica e Nanotecnologie, Consiglio Nazionale delle Ricerche (IFN-CNR), Piazza L. da Vinci, 32, 20133 Milano, Italy. 3 Politecnico di Milano, Dipartimento di Fisica, Piazza L. da Vinci, 32, 20133 Milano, Italy. Key words: absorption coefficient, ascorbic acid, carotenoid composition, Mangifera indica L., total antioxidant capacity, total phenols. Abstract: The content and composition of the main antioxidants in the pulp of mangoes depend also on cultivar and maturity degree, the latter being non‐ destructively evaluated by the absorption coefficient measured by Time‐ resolved Reflectance Spectroscopy (TRS) at 540 nm (µa540). Aiming at evaluat‐ ing the levels of antioxidants [carotenoids (CAR), phenols (TPC), ascorbic acid (AA)] and antioxidant capacity (TAC) in relation to µa540 maturity class, select‐ ed ‘Haden’ and ‘Palmer’ mangoes were measured for µa540 by TRS, classified based on µa540 value as less (LeM), medium (MeM) and more (MoM) mature and analyzed for pulp firmness, pulp color (a*, h°, Yellowness Index), CAR (total and composition by HPLC‐DAD), TPC, AA and TAC. ‘Palmer’ fruit had higher TPC, AA and TAC than ‘Haden’ mangoes. On average MoM fruit showed higher TPC, total CAR, total all‐trans‐violaxanthin esters and all‐trans‐β‐carotene than MeM and LeM fruit. LeM fruit did not have compounds belonging to the 9‐cis‐violaxanthin group, while cis‐β‐cryptoxanthin was approx. 19% of total carotenoids. In MoM mangoes the main carotenoid was all‐trans‐β‐carotene (53%), followed by total all‐trans‐violaxanthin esters (30%), 9‐cis‐violaxanthin group (8%) and cis‐β‐cryptoxanthin (6%). The µa540 significantly correlated (r=0.78‐0.94) with total CAR, all‐trans‐β‐carotene, all‐trans‐violaxanthin no.3 (both cultivars), TPC, all‐trans‐violaxanthin no.1, no.2, no.6 (‘Haden’), and 9‐cis‐violaxanthin no.2, no.3 (‘Palmer’). Our results indicate that TRS is suitable to non‐destructively measure the pulp color of mangoes and to sort fruit with different ripening degree and nutraceutical properties. 1. Introduction Mango (Mangifera indica L.) is a climacteric fruit belonging to the fam- ily of Anacardiaceae grown particularly in tropical and subtropical coun- (*) Corresponding author: anna.rizzolo@crea.gov.it Citation: VANOLI M., GRASSI M., SPINELLI L., TORRICELLI A., RIZZOLO A., 2018 - Quality and nutraceutical properties of mango fruit: influence of cultivar and biological age assessed by Time-resolved Reflectance Spectroscopy. - Adv. Hort. Sci., 32(3): 407-420 Copyright: © 2018 Vanoli M., Grassi M., Spinelli L., Torricelli A., Rizzolo A. This is an open access, peer reviewed article published by Firenze University Press (http://www.fupress.net/index.php/ahs/) and distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Data Availability Statement: All relevant data are within the paper and its Supporting Information files. Competing Interests: The authors declare no competing interests. Received for publication 8 March 2018 Accepted for publication 13 September 2018 AHS Advances in Horticultural Science Adv. Hort. Sci., 2018 32(3): 407-420 408 tries, with an estimated world production of 42 mil- lion tons per year (FAO, 2015). Brazil is one of the first ten largest mango producers, and more than 25% of its production is exported to Europe (Mitra, 2016). Appreciated for its excellent eating quality due to attractive flesh color, juicy texture and sweet fla- vor, mango provides high contents of bioactive com- pounds including carotenoids, phenolic compounds, ascorbic acid and reducing sugars (Rocha Ribeiro et al., 2007; Manthey and Perkins-Veazie, 2009). Carotenoids are responsible for the yellow-orange color of mango pulp and their content and composi- tion depend mainly on cultivar and maturity degree, along with edaphic and climatic factors and posthar- vest handling, processing and storage conditions ( O r n e l a s - P a z e t a l . , 2 0 0 7 , 2 0 0 8 ; M a n t h e y a n d Perkins-Veazie, 2009; Hewavitharana et al., 2013; Liu et al., 2013; Vásquez-Caicedo et al., 2006; Vanoli et al., 2016). During mango fruit ripening, biosynthesis of carotenoids occurs, due to chloroplasts differenti- ation into chromoplasts by disintegration of the thy- lacoid membranes and by the development of new pigment-bearing structures (Vásquez-Caicedo et al., 2006). This process leads to carotenoids accumula- tion, which is usually accompanied by color changes of the pulp turning from white to yellow-orange (Ornelas-Paz et al., 2008; Vásquez-Caicedo et al., 2006). The accumulation of carotenoids in the meso- carp shows an exponential behavior during fruit ripening and cultivar-specific relationships between total or individual carotenoid (all-trans-β-carotene, all-trans-violaxanthin and 9-cis-violaxanthin) con- tents and mesocarp color (a*, h°) were established in different mango cultivars (Vásquez-Caceido et al., 2 0 0 6 ; O r n e l a s - P a z e t a l . , 2 0 0 8 ) . S i m i l a r l y t o carotenes, ascorbic acid content and total phenolic content vary according to the cultivars, maturity stage and cultural practices (Rocha Ribeiro et al., 2007; Valente et al., 2011; Liu et al., 2013; Oliveira et al., 2016; Septembre-Malaterre et al., 2016). In order to withstand shipping to distant markets and at the same time to have the optimum eating quality when ripe, mango fruit are harvested at the hard green stage, after having reached the physiolog- ical maturity, but before the onset of the climacteric rise. If mango fruit are immature at harvest, they do not reach an eating quality when ripe and, hence, the discrimination between mature and immature fruit at harvest is very important from the marketing point of view in order to minimize qualitative losses during the supply chain. Fruit shape (“shoulders” should be full), pulp color and firmness are the most used maturity indices for mangoes, but differences among cultivars and growing conditions have precluded uni- versal maturity indices. On the other hand, the cur- rent industry measurements of firmness and pulp color have the disadvantage of being destructive of fruit; hence, the development of a non-destructive technique could help the growers to pick fruit at the proper maturity degree for the different market des- tinations. Among the non-destructive techniques able to assess the maturity degree of fruit, Time-resolved Reflectance Spectroscopy (TRS) is gaining increasing interest (Nicolai et al., 2014). TRS is a non-destructive optical technique which can separate the effect of light absorption, due to chemical compounds such as pigments and water, and light scattering, due to microscopic changes in refractive index caused by membranes, vacuoles, starch granules, organelles, and air. By measuring photon time-of-flight distribu- tion with picoseconds temporal resolution, the absorption (µa) and reduced scattering (µs) coeffi- cients in the VIS-NIR wavelength range are quanti- fied, by probing pulp at a depth of 1-2 cm with no or limited influence from the skin (Cubeddu et al., 2001, Torricelli et al., 2008). TRS has been used to study the internal fruit attributes related to maturity in apples, peaches, nectarines and pears (Rizzolo and Vanoli, 2016). In nectarines, the µa measured at harvest at 670 nm (µa670), near the chlorophyll-a peak, can be considered an effective maturity index as it is linked to the biological age of the fruit (Tijskens et al., 2007) and has been successfully used to predict the soften- ing rate during shelf life in nectarines and to select fruit for different market destinations (Eccher Zerbini et al., 2009). Previous studies carried out on mango fruit have demonstrated the potential of TRS for the non- destructive determination of pulp color (Vanoli et al., 2013; Rizzolo et al., 2016; Vanoli et al., 2016), as well as the possibility of using TRS absorption in both the carotenoids (540 nm) and chlorophyll-a (670 nm) spectral regions to classify mango fruit according to maturity and to predict the ripening of individual fruit (Eccher Zerbini et al., 2015). The present work aimed at evaluating the levels of antioxidants and antioxidant capacity in the pulp of two cultivars of Brazilian mangoes in relation to fruit maturity class assigned according to the µa540 value, along with selected quality parameters. The relationships between µa540 maturity index and total carotenoid content, total phenolic compounds, ascorbic acid and the fourteen carotenoid com- Vanoli et al. - Quality and nutraceutical properties of mangoes according to TRS maturity class 409 pounds identified were also studied. 2. Materials and Methods Mango fruit On November 2011, ‘Haden’ and ‘Palmer’ man- g o e s w e r e p i c k e d i n e x p e r i m e n t a l o r c h a r d s o f EPAMIG (Empresa de Pesquisa Agropecuária de Minas Gerais) in Minas Gerais state (Brazil) and trans- ported to Milan (Italy) by plane soon after harvest. On arrival at CREA-IT lab (about 5-7 days from har- vest), 60 fruits of ‘Haden’ and 90 fruits of ‘Palmer’ without defects were selected and measured by TRS at 650 nm (‘Haden’) or at 690 nm (‘Palmer’) on two opposite sides in the fruit equatorial region and ranked by decreasing µa averaged over the two sides, that is from less (high µa) to more mature fruit (low µa). ‘Haden’ fruit were put at 20°C for 2 days, while ‘Palmer’ mangoes were randomized into 3 batches of 30 fruits, corresponding to 0, 4 and 11 days of shelf life at 20°C, in order to have the whole range of µa690 in each batch. Sub-samples of 10 fruits, cover- ing the whole range of µa, were selected for ‘Haden’ and for ‘Palmer’ batch held for 4 days at 20°C. Each selected fruit was measured by TRS at 540 nm on two opposite sides in the equatorial region; in the same positions of TRS measurements, skin was removed by a slicer, and, after measuring pulp color and firmness, the whole fruit were immediately deep frozen and kept at -30°C until carotenoid, ascorbic acid and total phenolic extractions. Time-resolved Reflectance Spectroscopy (TRS) A portable compact setup working at discrete wavelengths developed at Politecnico di Milano (Spinelli et al., 2012) was used. The light source is a supercontinuum fiber laser (SC450-6W, Fianium, UK) providing white-light picoseconds pulses, with dura- tion of few tens of picoseconds. A custom-made filter wheel loaded with 14 band-pass interference filters (NT-65 series; Edmund Optics) is used for spectral selection in the range 540-940 nm. Light is delivered to the sample by means of a multimode graded-index fiber and diffuse remitted light is collected by 1 mm fiber placed at 1.5 cm distance from the illumination point. A second filter wheel identical to the first one is used for cutting off the fluorescence signal origi- nated when illuminating the fruit in the visible spec- tral region. The light then is detected with a photo- multiplier (HPM-100-50, Becker & Hickl, Germany) and the photon time-of-flight distribution was mea- sured by a time-correlated single-photon counting board (SPC-130, Becker & Hickl, Germany). The instrumental response function has a full width at half maximum of about 260 ps and the typical acqui- sition time is 1 s per wavelength. A model for photon diffusion in turbid media was used to analyze TRS data to assess the bulk optical properties of the sam- ples (Martelli et al., 2009) to obtain the estimates of µa and µs at each wavelength. Firmness and pulp color Flesh firmness was measured using a Instron UTM model 4301 penetrometer (crosshead speed 200 mm min-1, 8 mm diameter plunger). Data were averaged per fruit. Pulp color was measured with a spectrophotome- ter (CM-2600d, Minolta Co., Japan), using the prima- ry illuminant D65 and 2° observer in the L*, a*, b* color space. Hue (h°) was computed from a* and b* values according to: h°= arctangent (b*/a*) x 360/(23 x 14) Yellowness index (I Y ) was computed as: I Y =[81.2746X−1.0574Z)/Y]×100 after converting L*a*b* parameters into the XYZ color space (Jha et al., 2006). Data were averaged per fruit. Carotenoids, ascorbic acid, total phenols and total antioxidant capacity analysis Carotenoids, ascorbic acid, total phenols and total antioxidant capacity analysis were carried out on frozen fruit after 30 min thawing at ambient temper- ature, by slicing pulp portions without peel near the positions of TRS measurements and pooling the slices coming from the two fruit sides. Carotenoids were extracted following the proce- dure described by Picchi et al. (2012) with slight mod- ifications. Briefly, on individual fruit, 2 g of pulp (two replicates) was extracted with 10 mL of a solution of hexane:acetone:ethyl acetate (2:1:1 v/v/v) contain- i n g 1 0 0 µ L o f 1 % b u t y l h y d r o x y t o l u o l ( B H T ) i n methanol, to prevent carotenoid oxidative degrada- tion, and centrifuged at 4°C at 15,000 rpm for 20 min. The extracts were stored at -80°C until spectrophoto- metric and high-performance liquid chromatographic (HPLC) analyses. Total carotenoid content (CAR) was determined measuring absorbance at 450 nm using a spectropho- tometer (UV-UVIDEC 320, Jasco, Japan). The hexa- ne:acetone:ethyl acetate solution was used as the blank. Total carotenoid content was estimated from a standard curve of all-trans-β-carotene and data w e r e e x p r e s s e d a s m i l l i g r a m s o f β - c a r o t e n e Adv. Hort. Sci., 2018 32(3): 407-420 410 equivalent (β-car) per kilogram of fresh weight (mg β- car kg FW-1). Carotenoid composition was determined on e x t r a c t s a c c o r d i n g t o A z e v e d o - M e l e i r o a n d Rodriguez-Amaya (2004) with some modifications. A Jasco (Tokio, Japan) HPLC system consisting of a PU- 1580 liquid chromatographic pump coupled to LG 1580-04 quaternary gradient unit, a model AS 2055- plus autosampler and an MD 2010-plus multi-wave- length detector was used. Separations were per- formed on an Inertsil ODS-3 column (4.6 mm i.d × 250 mm length, particle diameter 5 µm, GL Science) at the temperature of 40°C which was maintained using a Jasco Co-1560 Intelligent Column thermostat. The sample injection volume was 80 µL. The column was eluted with 20% methanol and 80% of a gradient mixture of acetonitrile (A) and ethyl acetate (B) at the flow rate of 0.6 mL min-1, with 10% B at 0-25 min, 10-20% B at 25-35 min, 20-50% B at 35-40 min, 50% B at 40-45 min, 50-10% B at 45-50 min. Spectra of all peaks were recorded in the 200-600 nm wavelength range, and peak areas were monitored at 450 nm. Carotenoids (Table 1) were identified by comparing their retention times and spectral characteristics with those of standards (all-trans-β-carotene and violax- anthin, obtained by pansy petals) and with those reported in the literature (Ornelas-Paz et al., 2007, 2008), considering the three maximum absorbance wavelengths (λmax) and the spectral fine structure (% III/II), which is the percentage of the peak height of the longest wavelength absorption band (λmax III) to that of the middle absorption band (λmax II), taking the maximum of the valley between peak II and peak III as the baseline (Sajilata et al., 2008). Carotenoids were quantified referring to the total carotenoid con- tent estimated spectrophotometrically on the same extract in conjunction with the chromatogram per- c e n t c o m p o s i t i o n a n d d a t a w e r e e x p r e s s e d a s milligrams of β-carotene equivalent (β-car) per kilo- gram of fresh weight (mg β-car kg FW-1). All the mea- surements were carried out in triplicate. The vitamin A value, expressed as retinol equivalent (RE) was esti- mated from all-trans-β-carotene and cis-β-cryptoxan- thin amounts using as conversion figures 6 µg for Car and 12 µg for Crypt (Capra, 2006). Ascorbic acid was extracted following the proce- dure described by Robles-Sánchez et al. (2009 a) with slight modifications. Briefly, on individual fruit, 2 g of pulp (two replicates) was homogenized with 10 mL of 6% (w/v) aqueous solution of metaphosphoric acid (MPA), vortexed for 30 s, and centrifuged at 4°C at 15,000 rpm for 20 min and the extracts were kept at -20°C till HPLC analysis. Ascorbic acid was determined on just thawed extracts according to the conditions reported by Rizzolo et al. (2002), using a Jasco (Tokio, Japan) HPLC system consisting of a PU-980 liquid c h r o m a t o g r a p h i c p u m p , a m o d e l A S 1 0 5 5 - 1 0 autosampler and an UV-Vis 15770 detector set at 254 nm, coupled to an Inertsil ODS-3 column (4.6 mm i.d. × 250 mm length, particle diameter 5 µm, GL Science) at the temperature of 30°C, which was eluted with 0.02 M orthophosphoric acid at a flow rate of 0.7 mL min-1. Ascorbic acid was estimated from a standard curve of L-ascorbic acid in 6% MPA and data were Table 1 - Retention time (Rt, min), spectra characteristics [λmax (nm) in the mobile phase, obtained by DAD, spectral fine structure (% III/II)] and name abbreviation of tentatively identified compounds according to Ornelas-Paz et al. (2007, 2008) Peak no. Rt λmax I λmax II λmax III % III/II Tentative identification Abbreviation 1 5.24-5.89 419 439 471 82 unknown UNK 2 5.92-5.97 415 439 471 82 all-trans-violaxanthin various esters Viol no. 1 3 6.03-6.09 415 439 471 100 all-trans-violaxanthin various esters Viol no. 2 4 6.11-6.19 415 443 471 90 cis-β-cryptoxanthin Crypt 5 6.23-7.29 415 439 471 93 all-trans-violaxanthin various esters Viol no. 3 6 7.40-7.83 411 435 463 75 9-cis-violaxanthin 9-viol no. 1 7 8.03-8.77 415 435 467 83 9-cis-violaxanthin 9-viol no. 2 8 9.40-10.31 415 439 467 84 all-trans-violaxanthin various esters Viol no. 4 9 10.32-11.15 411 435 463 80 9-cis-violaxanthin 9-viol no. 3 10 33.48-36.19 419 439 471 n.c. all-trans-violaxanthin various esters Viol no. 5 11 37.97-40.61 451 479 23 all-trans-β-carotene Car 12 40.90-43.60 419 439 471 100 all-trans-violaxanthin various esters Viol no. 6 13 43.71-43.99 415 435 467 100 9-cis-violaxanthin 9-viol no. 4 14 44.00-44.60 415 439 467 100 all-trans-violaxanthin various esters Viol no. 7 Vanoli et al. - Quality and nutraceutical properties of mangoes according to TRS maturity class 411 3. Results TRS optical properties In ‘Palmer’ fruit, µa690 at harvest ranged from 0.074 cm-1 for the less mature fruit to 0.021 cm-1 for the more mature ones and decreased to 0.061 cm-1 and 0.019 cm-1, respectively, after 4 days of shelf life at 20°C; concomitantly, after shelf life, µa540 ranged from 0.117 cm-1 for the least mature fruit to about 0.33 cm-1 for the most mature ones. In ‘Haden’ man- goes, µa650 ranged at harvest from 0.231 cm-1 to 0.030 cm-1, with the majority of the fruit in the 0.030- 0.065 cm-1 range; after 2 days of shelf life at 20°C, µa650 decreased only in less mature fruit, whereas in all the other mango fruit it increased to values rang- ing from 0.036 cm-1 to 0.053 cm-1, while the µa540 values after shelf life ranged from 0.157 cm-1 for the least mature fruit to 0.835 cm-1 for the most mature ones. The µa540 maturity index, related to carotenoids content, was then used to classify the selected fruit within each cultivar in three TRS maturity classes: less mature (LeM) with low µa540, more mature (MoM) with high µa540 and medium mature (MeM) with intermediate values of µa540. Cultivar and TRS maturity class influenced the value of µa540 maturity index (Table 2); on average µa540 was higher in cv. H a d e n ( ‘ H a d e n ’ : 0 . 4 0 0 ± 0 . 0 2 5 c m - 1; ‘ P a l m e r ’ : 0.248±0.025 cm-1) and in the MoM class in both culti- vars, with MoM ‘Haden’ fruit being characterized by the highest µa540 value. Quality parameters TRS maturity class and cultivar greatly affected a* and h° pulp color parameters and had only a slight influence on firmness, probably due to the high stan- dard error values, whereas IY depended only on matu- rity class (Table 2). On average, ‘Palmer’ fruit had lower firmness and a* value, and higher h° than ‘Haden’ fruit. In ‘Palmer’ mangoes firmness did not vary with maturity class, while in ‘Haden’ firmness showed the highest values in LeM fruit and the lowest in MeM and in MoM ones. MoM ‘Palmer’ fruit had higher a* and IY and lower h° than LeM and MeM maturity classes, whereas LeM ‘Haden’ fruit had lower IY than MoM fruit, and a* increased and h° decreased from LeM to MeM and MoM maturity classes. Ascorbic acid, total phenolic content and total antiox- idant capacity AA content and TAC were significantly influenced only by cultivar, while TPC depended by both cultivar expressed as milligram per kilogram of fresh weight (mg kg FW-1). All the measurements were carried out in triplicate. Total phenol content (TPC) and total antioxidant capacity (TAC) were determined on the same extract (two replicates/fruit) obtained by homogenizing 2 g of pulp with 10 mL of acidic ethanol (ethanol:0.04 M HCl, 1:1 v/v), vortexed for 30 s and centrifuged at 4°C at 15,000 rpm for 20 min. Extracts were kept at -20°C till total phenol content and antioxidant capacity deter- minations. TPC was determined using the Folin- Ciocalteau method (Singleton et al., 1999) based on the reduction of a phosphowolframate phospho- molibdate complex by phenolics to blue reaction products, and measuring absorbance at 730 nm using a spectrophotometer (UV-UVIDEC 320, Jasco, Japan). The TPC was estimated from a standard curve of gallic acid and data were expressed as milligrams of gallic acid equivalents (GAE) per kilogram of fresh weight (mg GAE kg FW-1). All the measurements were per- formed in triplicate. TAC was evaluated using the free radical 1,1,-dyphenyl-2-picrylhydrazil (DPPH•) accord- ing to Brand-Williams et al. (1995) with modifications. Fifty microlitres of sample extract or Trolox standard solution (0.02-0.8 mM) were added to 2 mL of ethanol and 550 μL of DPPH• solution (0.05 mM in ethanol) and, during 5 min of incubation, the absorbance at 517 nm was measured with a Jasco 7800 UV/VIS spec- trophotometer (Jasco Europe S.r.l., Cremella, LC, Italy). The DPPH scavenging capacity of the samples was calculated using a standard curve of Trolox, and expressed as micromoles of Trolox equivalents (TE) per kilogram of fresh weight (µmol TE kg FW-1). All the measurements were performed in triplicate. Statistical analysis The Statgraphics v. 5.2 (Manugistic Inc., Rockville, MD, USA) software package was used. Data were s u b m i t t e d t o m u l t i f a c t o r a n a l y s i s o f v a r i a n c e (ANOVA) considering cultivar, TRS maturation class and their interaction as source of variation. In addi- tion, one-way ANOVA was used to study the main factors (cultivar, TRS maturity class), and the TRS maturity class within each cultivar. Percentage data of carotenoids were analyzed after arcsine transfor- m a t i o n . M e a n s w e r e c o m p a r e d b y 9 5 p e r c e n t Bonferroni’s test. Relationships between µa540 and pulp color parameters and between µa540, a*, h°, IY and ascorbic acid, TPC, TAC and carotenoids were studied using regression analysis. For each parame- ter, the model with the higher performance was con- sidered. 412 Adv. Hort. Sci., 2018 32(3): 407-420 and maturity class (Table 3). On average, ‘Palmer’ mangoes had higher AA, TPC and TAC than ‘Haden’ fruit, and LeM fruit had lower TPC than MoM man- goes (Fig. 1). The average AA values were approx. 190 mg kg FW-1 for ‘Haden’ and 390 mg kg FW-1 for ‘Palmer’ and AA content did not change with maturity class in ‘Palmer’ mangoes while in ‘Haden’ showed the high- est values in MeM fruit (Fig. 1). TPC was higher in ‘Palmer’ than in ‘Haden’ cv., being on average approx. 316 and 264 mg kg FW-1 in ‘Palmer’ and ‘Haden’ fruit, respectively, as well as it was higher in MoM fruit from both cultivars. ‘Palmer’ fruit were characterized by higher TAC showing on average 2.59 times greater than that from ‘Haden’. TAC had signifi- cant positive correlations with TPC (r= 0.66, p= 0.002) and AA (r= 0.89, p<0.0001). Carotenoids Total carotenoids (CAR) depended only by maturi- ty class (Table 3), with LeM fruit having less CAR than MoM ones (Fig. 1). The chromatographic carotenoid patterns of LeM and MoM maturity classes in both cultivars showed 9 common carotenoids (Fig. 2) out of 14 peaks tentatively identified by comparing spec- tral characteristics with those previously reported u s i n g a s i m i l a r m o b i l e p h a s e ( T a b l e 1 ) . T h e carotenoid pattern includes seven all-trans-violaxan- thin (Viol) and four 9-cis-violaxanthin (9-Viol) con- taining compounds, cis-β-cryptoxanthin and all-trans- β-carotene. The most abundant carotenoid in both cultivars was all-trans-β-carotene (Tables 4 and 5), represent- ing 49-56% of the total carotenoid content, followed Fig. 1 - (A) Ascorbic acid (AA), (B) total carotenoids (CAR), total phenol content (TPC) and total antioxidant capacity (TAC) in ‘Palmer’ and ‘Haden’ mangoes in function of µa540 maturity class (LeM, less mature; MeM, medium mature; MoM, more mature). Bars refer to SE. Within each cultivar ANOVA results are indicated as follows: *, **, ***: significant at P≤0.05, 0.01, 0.001, respectively; NS, not significant. Mean±SE. Within each cultivar, means followed by different letters are statistically different (Bonferroni's test, P≤0.05). P-value of F-ratio: NS=not significant; *P<0.05; **P<0.01; ***P<0.001. Table 2 - Absorption coefficient at 540 nm (µa540, cm-1), flesh firmness (N) and pulp colour parameters (a*; hue, h°; yellowness index, IY) of 'Palmer' and 'Haden' mangoes of less mature (LeM), medium mature (MeM) and more mature (MoM) TRS maturity clas- ses and results of multifactor ANOVA (F-ratio value and P-value) Table 3 - Multifactor analysis of variance (F-ratio and P-value) for ascorbic acid (AA), total carotenoids (CAR), total plenolic content (TPC) and total antioxidant capacity (TAC) *P<0.05; **P<0.01; ***P<0.001; NS=not significant. Cultivar Maturity class µa540 Firmness a* h° IY Palmer LeM 0.196±0.027 b 8.33 ±0.55 a 0.86±0.79 b 89.13±0.75 a 130.7±1.7 b MeM 0.241±0.006 ab 7.70±0.45 a 3.64±1.03 b 86.85±0.89 a 153.3±1.6 ab MoM 0.310±0.017 a 6.12±0.63 a 9.84±1.75 a 81.72±1.30 b 165.7±9.5 a Haden LeM 0.191±0.011 b 37.29±9.89 a 2.98±1.62 c 87.18±1.55 a 130.0±6.0 b MeM 0.336±0.033 b 10.47±3.62 a 10.60±1.44 b 80.32±1.15 b 159.7±6.6 a MoM 0.677±0.158 a 5.85±0.65 b 20.46±1.12 a 72.37±0.84 c 191.2±6.1 a ANOVA A: cultivar 18.88 *** 4.81* 28.85 *** 32.20 *** 3.94 NS B: maturity class 23.52 *** 4.85 * 36.79 *** 32.25 *** 27.18 *** A × B 9.21 ** 3.98 * 3.94 * 4.16 * 2.01 NS Factors AA CAR TPC TAC main factors Cultivar (A) 38.79 *** 0.48 NS 5.12 * 32.75 *** Maturity class (B) 0.75 NS 7.28 ** 4.20 * 1.85 NS interaction A × B 0.68 NS 0.98 NS 0.81 NS 0.53 NS 413 Vanoli et al. - Quality and nutraceutical properties of mangoes according to TRS maturity class by cis-β-cryptoxanthin (6-18%) and Viol no.3 (11- 16%). The content of all-trans-violaxanthins was high- er than that of 9-cis-violaxanthins in both cultivars. The content of Viol no.3, no.4 and no.6, 9-Viol no.4 and all-trans-β-carotene, as well as the sums of a l l - t r a n s - v i o l a x a n t h i n s ( ∑ V i o l ) a n d o f 9 - c i s - violaxanthins (∑9-Viol) depended only on maturity class, that of 9-Viol no.1 only on cultivar, whereas those of 9-Viol no.2 and no.3 on both cultivar and maturity class (Table 4). In fact ‘Haden’ mangoes had higher amounts of 9-Viol no.3 (Table 4) and had lower proportion of 9-Viol no.4 than ‘Palmer’ fruit (Table 5). Viol no. 1 and 9-Viol no.1 were present only in ‘Haden’ and 9-Viol no. 2 only in ‘Palmer’ fruit. Fig. 2 - Typical chromatographic patterns at 450 nm of carotenoid extracts of (A) LeM ‘Palmer’, (B) MoM ‘Palmer’, (C) LeM ‘Haden’ and (D) MoM ‘Haden’ mangoes. For peak assignment see Table 1. Table 4 - Carotenoid compounds of 'Palmer' and 'Haden' mangoes (mg β-CARE kg FW-1) and vitamin A value (RE 100 g FW-1) influence of cultivar and of TRS maturity class and results of multifactor ANOVA (F-ratio value and P-value). For identification data of each carotenoid see Table 1 Mean±SE; 0=not detected. Within cultivar and within TRS maturity class means followed by different letters are statistically different (Bonferroni's test, (*) P<0.10; * P<0.05; **P<0.01; NS =not significant). ∑Viol= total all-trans-violaxanthin esters; ∑9-Viol= total 9-cis-vio- laxanthin. Cultivar Maturity class ANOVA Palmer Haden Less mature Medium mature More mature Cultivar (A) Maturity class (B) A × B Viol no. 1 0±0 a 0.49±0.49 a 0 ±0 a 0±0 a 0.99±0.99 a 3.25 (*) 2.89 (*) 2.89 (*) Viol no. 2 1.14±0.46 a 0.87±0.68 a 0.68 ±0.40 a 0.49±0.38 a 2.15±1.33 a 0.08 NS 2.11 NS 2.48 NS Crypt 2.58±0.55 a 1.82±0.38 a 1.66 ±0.37 a 2.69±0.47 a 2.18 ±0.97 a 1.53 NS 0.89 NS 0.39 NS Viol no. 3 2.43±0.31 a 2.49±0.55 a 1.53±0.28 b 2.82±0.67 ab 3.26±0.30 a 0.07 NS 3.08 (*) 2.37 NS 9-viol no. 1 0±0 b 0.81±0.42 a 0±0 a 0.67±0.42 a 0.68±0.68 a 4. 91 * 1.36 NS 1.36 NS 9-viol no. 2 0.33±0.16 a 0±0 b 0±0 b 0.06±0.06 ab 0.52±0.26 a 10.91 ** 6.23 * 6.23 * Viol no. 4 0.84±0.16 a 0.91±0.40 a 0.27±0.13 b 0.88±0.40 ab 1.72±0.37 a 0.65 NS 4.90 * 0.89 NS 9-viol no. 3 0.09±0.06 b 0.35±0.15 a 0±0 b 0.21±0.11 ab 0.56±0.25 a 7.62 * 7.57 ** 2.35 NS Car 9.36±1.14 a 9.63±2.72 a 5.05±0.67 b 8.85±1.81 ab 16.65±3.47 a 0.81 NS 8.37 ** 1.03 NS Viol no. 6 0.64±0.14 a 0.56±0.23 a 0.08±0.08 b 0.74±0.16 ab 1.12±0.25 a 0.12 NS 10.19 ** 1.66 NS 9-viol no. 4 0.55±0.19 a 0.22±0.15 a 0±0 b 0.55± 0.21 a 0.67±0.28 a 1.58 NS 4.15 * 0.99 NS Viol no. 7 0.15±0.10 a 0.19±0.10 a 0.07±0.07 a 0.17±0.11 a 0.30±0.18 a 0.19 NS 0.72 NS 0.97 NS ∑Viol 5.20±0.76 a 5.51±1.95 a 2.64±0.74 b 5.11±1.22 ab 9.54±2.82 a 0.88 NS 5.69 * 2.26 NS ∑9-viol 0.98±0.35 a 1.38±0.67 a 0±0 b 1.49±0.48 ab 2.43±1.06 a 1.06 NS 4.39 ** 0.32 NS Vitamin A value 258.2±22.3 a 162.0±45.3 a 85.6±11.4 b 149.7±30.5 ab 279.3±57.2 a 0.40 NS 8.38 ** 1.01 NS 414 Adv. Hort. Sci., 2018 32(3): 407-420 LeM mangoes had no 9-cis-violaxanthins and were characterized by lower contents of Viol no.3, Viol no.4, Viol no.6 and all-trans-β-carotene than MoM fruit, but higher proportion of all-trans-β-carotene than MeM ones (Tables 4 and 5). These carotenoids increased with advancing maturity degree, showing the highest contents in MoM mangoes. On average all-trans-β-carotene corresponded to 53% of total carotenoids; the all-trans-β-carotene proportion was not significantly affected by cultivar, whereas on average was lower in MeM fruit than in LeM ones, while MoM mangoes were not statistically different from fruit of the other two maturity classes (Table 5). The vitamin A value did not differ between culti- var, but significantly increased with maturity class from 86 of LeM fruit to 279 RE 100 g-1 of MoM man- goes (Table 4). Regression analysis The results of regression analysis between µa540 and pulp color parameters and between µa540, a*, h°, IY and ascorbic acid, TPC, TAC and carotenoids dif- fered for the two cultivars and data are summarized in Tables 6 and 7, reporting the type of the model having the best performance. The μa540 was positively related to a* and IY and negatively to h° (Table 6) with r ranging from 0.83 to 0.87 for ‘Palmer’ fruit and approx. 0.98 for ‘Haden’ cultivar. In ‘Palmer’ mangoes (Table 7) µa540, a*, h° and IY were related to total carotenoids, Viol no.3 and no.4, 9-Viol no.2 and no.3, all-trans-β-carotene, ∑9-Viol and vitamin A value with lower r values (0.62- 0.84) for µa540 respect to those found for pulp color parameters (0.74-0.96). Only IY was related to Viol no.6, 9-Viol no.4 and ∑Viol with r≥0.7, and only a* was related to TPC, but with r<0.6. In contrast, in ‘Haden’ fruit µ a 540, a*, h° and I Y were related to total Table 5 - Carotenoid composition (percent to total carotenoids) of 'Palmer' and 'Haden' mangoes: influence of cultivar and of TRS matu- rity class and results of ANOVA (F-ratio value and P-value). For identification data of each carotenoid see Table 1 Table 6 - Results of regression analysis between absorption coefficient at 540 nm (μa540) and pulp color parame- ters For each regression, the following data are given: r = correlation coefficient, P = significance of the model (***, P<0.001; **, P< 0.01) and MT= model type (DR= doble reciprocal, E= exponential, L= linear, Ln= logarithmic-X, M= multiplicative, RX= reciprocal-X, Sc = S-curve). Cultivar Maturity class ANOVA Palmer Haden Less mature Medium mature More mature Cultivar (A) Maturity class (B) A × B Viol no. 1 0±0 a 0.87±0.87 a 0±0 a 0±0 a 0.36±0.36 a 3.25 (*) 2.89 (*) 2.89 (*) Viol no. 2 6.74±2.84 a 3.13±1.49 a 2.51±0.64 a 1.04±0.43 a 2.89±0.58 a 0.21 NS 0.36 NS 1.65 NS Crypt 14.87±3.01 a 15.40±2.77 a 18.18±0.53 a 15.31±0.07 a 5.88±1.04 a 0.01 NS 1.80 NS 1.40 NS Viol no. 3 14.32±2.05 a 16.30±2.07 a 16.33±0.18 a 15.96±0.09 a 11.31±0.07 a 0.04 NS 1.51 NS 2.63 NS 9-viol no. 1 0±0 b 2.44±1.30 a 0±0 a 1.122±0.26 a 0.25±0.25 a 7.84 * 2.48 NS 2. 48 NS 9-viol no. 2 1.36±0.60 a 0±0 b 0±0 b 0.06±0.06 ab 1.08±0.20 a 14.55 ** 5.95 * 5.95 * Viol no. 4 4.23±0.58 a 3.17±0.95 a 1.27±0.17 a 3.07±0.12 a 5.51±0.01 a 0.54 NS 2.49 NS 0.40 NS 9-viol no. 3 0.33±0.22 b 1.24±0.46 a 0±0 b 0.46±0.11 ab 1.21±0.08 a 7.51 * 8.04 ** 2.49 NS Car 50.62±1.03 a 54.25±2.31 a 55.65±0.08 a 48.86±0.02 b 53.26±0.02 ab 3.22 (*) 3.28 (*) 3.38 (*) Viol no. 6 3.20±0.71 a 1.98±0.68 a 0.07±0.07 b 3.27±0.10 a 3.57±0.005 a 1.66 NS 11.66 ** 0.62 NS 9-viol no. 4 2.63±0.91 a 0.71±0.52 b 0±0 b 1.73±0.23 a 1.14±0.20 ab 4.52 * 6.20 ** 2.26 NS Viol no. 7 0.77±0.51 a 0.95±0.51 a 0.06±0.06 a 0.25±0.11 a 0.56±0.21 a 0.19 NS 0.46 NS 1.24 NS ∑Viol 29.25±3.30 a 26.41±3.04 a 25.030±0.42 a 28.13±0.09 a 28.76±0.07 a 0.22 NS 0.17 NS 4.27 * ∑9-viol 4.31±1.33 a 4.39±1.71 a 0±0 b 6.07±0.19 a 5.79±0.21 a 0.01 NS 13.50 ** 0.00 NS Mean±SE; 0=not detected. Within cultivar and within TRS maturity class means followed by different letters are statistically different (Bonferroni's test, (*) P<0.10; * P<0.05; **P<0.01; NS =not significant). ∑Viol= total all-trans-violaxanthin esters; ∑9-Viol= total 9-cis-vio- laxanthin. µa540 ‘Palmer’ ‘Haden’ r P MT r P MT a* 0.831 ** L 0.975 *** E h° 0.826 ** RX 0.976 *** Sc I Y 0.872 ** Ln 0.977 *** DR Vanoli et al. - Quality and nutraceutical properties of mangoes according to TRS maturity class 415 carotenoids, Viol no.2, no.3, no.4 and no.6, 9-Viol no.1 and no.3, all-trans-β-carotene, ∑Viol, ∑9-Viol, vitamin A value and TPC, with higher r values (0.72- 0.95) for µa540 than for pulp color parameters. In addition, only µa540 was related to 9-Viol no. 4. No significant relationships were found between µa540, a*, h° and Iy and ascorbic acid and TAC, what- ever the cultivar, suggesting that μa540 was able to reveal the carotenoids content in the pulp, as this wavelength corresponds to the tail of carotenoid absorption. independently from cultivar. Rizzolo et al. (2016) also s h o w e d t h a t µ a5 4 0 m a t u r i t y i n d e x , r e l a t e d t o carotenoids content, successfully classified ‘Tommy Atkins’ mangoes at harvest. As for quality parameters, ‘Palmer’ mangoes had firmness values typical of fully ripe fruit (Beaulieu and Lea, 2003), independently from maturity class, whereas in ‘Haden’ fruit LeM class showed firmness values typical of firm-ripe stage and MeM and MoM classes values typical of ready-to-eat or ripe fruit (Eccher Zerbini et al., 2015). Pulp color parameters Table 7 - Results of regression analysis between absorption coefficient at 540 nm (μa540), pulp color parameters and total carotenoid (CAR), total phenolic compounds (TPC), carotenoid compounds (for identification data see Table 1) and vitamin A value µa540 a* h° IY r P MT r P MT r P MT r P MT 'Palmer' CAR 0.78 * L 0.909 *** L 0.902 *** RX 0.897 *** L Viol no.3 0.814 ** L 0.789 * L 0.783 * RX 0.735 * L 9-Viol no.2 0.839 ** L 0.959 *** L 0.958 *** RX 0.823 ** L Viol no.4 0.625 (*) L 0.908 *** L -0.912 *** L 0.784 * L 9-Viol no.3 0.765 * L 0.871 ** L 0.871 ** RX 0.778 * L Car 0.793 ** L 0.913 *** Sy -0.909 *** Sy 0.888 ** Sy Viol no.6 − − − -0.745 * RX 9-Viol no.4 − − − -0.847 ** RX ∑Viol 0.576 (*) Log − − 0.693 * Log ∑9-Viol 0.812 ** L 0.873 ** L 0.86 ** RX 0.954 *** L Vitamin A value 0.792 ** L 0.915 *** Sy -0.912 *** Sy 0.887 ** L TPC − 0.575 (*) Sy − − 'Haden' CAR 0.912 *** L 0.855 ** E 0.854 ** Sc 0.876 *** E Viol no.2 0.853 ** L 0.618 (*) L 0.634 * RX 0.601 (*) L Viol no.3 -0.824 ** Sc 0.723 * E -0.742 * E -0.805 ** Sc 9-Viol no.1 0.756 * L 0.666 * L 0.661 * RX 0.718 * L Viol no.4 0.815 ** Sx 0.789 ** L 0.79 ** RX 0.823 ** L 9-Viol no.3 0.95 *** L 0.83 ** L 0.834 ** RX 0.848 ** L Car 0.924 *** L 0.823 ** Sy 0.817 ** Sc 0.837 ** Sy Viol no.6 0.944 *** L 0.862 ** L 0.862 ** RX 0.873 *** L 9-Viol no.4 0.700 * L − − − ∑Viol 0.937 *** L 0.885 *** Sy -0.886 *** E 0.905 *** Sy ∑9-Viol 0.848 ** L 0.723 * L 0.721 * RX 0.756 * L Vitamin A value 0.923 *** L 0.823 ** Sy 0.817 ** Sc 0.838 ** Sy TPC 0.867 ** L 0.758 * L 0.761 ** RX 0.766 ** L For each significant regression, the following data are given: r= correlation coefficient, P= significance of the model (***, P< 0.001; **, P <0.01; *, P< 0.05; (*), P<0.10) and MT= model type (DR= doble reciprocal, E= exponential, L= linear, Ln= logarithmic-X, Log= logistic, RX= reciprocal-X, Sc= S-curve, Sx= square-root-X, Sy= square-root-Y). 4. Discussion and Conclusions The absorption coefficient measured at 540 nm (µa540) showed different value ranges between ‘Haden’ and ‘Palmer’ mangoes and it was able to dis- tinguish more mature fruit from less mature ones differed among maturity classes, confirming previous results obtained for ‘Tommy Atkins’ cultivar. In fact Rizzolo et al. (2011) and Vanoli et al. (2011) found that LeM ‘Tommy Atkins’ mangoes were character- ized by higher h° and lower a* and I Y than MoM fruit. Moreover, Vanoli et al. (2011) found that with fruit Adv. Hort. Sci., 2018 32(3): 407-420 416 ripening at 20°C h° decreased and IY increased, con- firming that the trend of pulp color observed in our work with the TRS maturity classes was actually due to a different ripening degree. The average AA values found for ‘Palmer’ and ‘Haden’ fruit are comparable with the data by Rocha- Ribeiro et al. (2007) for the same cultivar, and with AA content reported for other cultivars by Liu et al. (2013) and Elsheshetawy et al. (2016). However, within the same variety, AA content may vary accord- ing to climatic conditions, cultural practices, maturity stage and postharvest factors. For ‘Keitt’ cultivar Ibarra-Garza et al. (2015) found that AA content var- ied from about 1300 mg kg FW-1 in fruit soon after harvest, to about 2500 mg kg FW-1 till 8 days of ripen- i n g a t r o o m t e m p e r a t u r e , f o l l o w e d b y a 5 4 % decrease in fully-ripe fruit. Similarly, Robles-Sánchez et al. (2009 b) reported for ‘Ataulfo’ fruit stored for 15 days at 12°C a 50% decrease of AA content at the end of shelf-life. The TPC contents found in this work are in agree- ment with Rocha-Ribeiro et al. (2007), even if these Authors reported lower AA contents than those found in our work for the same cultivars. No data on TPC content in mangoes in relation to TRS maturity classes, having same harvest time and same post- harvest management, are available in the literature. However, the TPC increasing trend from less to more mature fruit class found in this work is in agreement with the results for the final period of shelf life/stor- age reported by Ibarra-Garza et al. (2015) and Robles-Sánchez et al. (2009 b) when fruit are becom- ing softer and with a yellower pulp color. Ibarra- Garza et al. (2015) found higher TPC in ‘Keitt’ fruit at the beginning of a 10 d shelf life period at room tem- perature, with a sharp decrease from 2 to 4 days of shelf life, followed by a slight but significant TPC increase till the end of shelf life; a similar trend was also found by Robles-Sánchez et al. (2009 b) for whole and fresh-cut ‘Ataulfo’ mangoes stored at 5°C for 15 days. In agreement with data obtained for TPC and AA, ‘Palmer’ fruit showed higher TAC than ‘Haden’ ones. The positive correlations of TAC with TPC and AA found in this work are in agreement with literature data. In fact Silva and Sirasa (2018) reported signifi- cant correlations between ascorbic acid and TPC and FRAP and DPPH scavenging activity measured for sev- eral fruit species, and Palafox-Carlos et al. (2012) b e t w e e n D P P H s c a v e n g i n g a c t i v i t y a n d T P C i n ‘Ataulfo’ mangoes; on the other hand, Liu et al. (2013) and Ibarra-Garza et al. (2015) found in man- goes correlation between phenolic concentration and antioxidant activity measured with other meth- ods (FRAP, ORAC), but not between antioxidant activ- ity and ascorbic acid content; in contrast Rocha- Ribeiro et al. (2007) reported that DPPH radical scav- enging activity of mango extracts was strongly corre- lated with ascorbic acid content, but not with pheno- lic content. Liu et al. (2013) suggested that the differ- ence in relationships between antioxidant activity and ascorbic acid and phenolic compound contents found among authors could be due to a masking effect of phenolics present in far higher concentra- tion than ascorbic acid. Our results suggest that in this work the antioxidant activity can be attributed more to ascorbic acid than to total phenols. No data on total carotenoids in relation to TRS maturity classes are available in the literature. Vanoli et al. (2016) for ‘Palmer’ and ‘Haden’ mangoes reported a wide fruit-to-fruit variability in CAR con- tent for both cultivars and that CAR content had an increasing trend with µa540 value measured on fruit belonging to the same harvest date, i.e. that CAR content increases with advancing maturity degree. Similarly, an exponential increase in the carotenoid content with fruit ripening has been reported for ‘Ataulfo’ mangoes (Ornelas-Paz et al., 2008) and nine Thai cultivars (Vásquez-Caicedo et al., 2005). As for the chromatographic carotenoid patterns, the tenta- tive identification of peaks was carried out by com- paring spectral characteristics with those previously reported using a similar mobile phase. Seven peaks were tentatively identified as all-trans-violaxanthin (439 nm maximum absorption wavelength) and four as 9-cis-violaxanthin (435 nm maximum absorption wavelength) containing compounds. The spectral maximum for peak 4 was similar to that reported for cis-β-cryptoxanthin. A standard mixture of all-trans- β-carotene was used for the identification of peak 11; the retention time and the spectroscopic charac- teristics of reference material were identical to those observed for peak 11 in all the samples. In general, the spectral fine structures (% III/II values in Table 1) found in this work are in agreement with the values reported in the literature (Ornelas-Paz et al., 2007, 2008). In both cultivars the proportions of all-trans- β - c a r o t e n e a n d c i s - β - c r y p t o x a n t h i n t o t o t a l carotenoid content, as well as the higher content of all-trans-violaxanthins than that of 9-cis-violaxan- thins are very similar to findings reported in ‘Ataulfo’, ‘ K e i t t ’ , ‘ T o m m y A t k i n s ’ a n d ‘ K e n t ’ m a n g o e s (Mercadante and Rodriguez-Amaya, 1998; Pott et al., 2003; Ornelas-Paz et al., 2008). The differences in Vanoli et al. - Quality and nutraceutical properties of mangoes according to TRS maturity class 417 carotenoid composition among the maturity classes found in this research are consistent with literature data. Previous researches on carotenoid composition of various mango cultivars carried out by Godoy and R o d r i g u e z - A m a y a ( 1 9 8 9 ) , M e r c a d a n t e a n d Rodriguez-Amaya (1998), Yahia et al. (2006), and Ornelas-Paz et al. (2007) have shown that generally the most important carotenoid in mango is all-trans- β-carotene and its proportion to total carotenoids depends on cultivar and fruit maturity stage. Ibarra- Garza et al. (2015) reported for ‘Keitt’ fruit that all- trans-β-carotene corresponded to 33% of total carotenoids in unripe fruit at harvest, ranged from 37 to 44% during the first 6 day-period of ripening at room temperature and reached 61% in fully ripe fruit. The contents of all-trans-β-carotene in ‘Haden’ and ‘Palmer’ mangoes were similar to those reported by Rocha-Ribeiro et al. (2007) for the same cultivars and by Mercadante and Rodriguez-Amaya (1998) for ‘Keitt’ and ‘Tommy Atkins’ fruit, but lower than those found for ‘Haden’ and other cultivars by Ornelas-Paz et al. (2007). Also the amounts of all-trans- and 9-cis- violaxanthins in both cultivars were lower than those found by Ornelas-Paz et al. (2007) for ‘Haden’ and other cultivars, but similar to those reported by Low et al. (2015) for ‘Kensington Pride’ mangoes. The amounts of cis-β-cryptoxanthin in both cultivars were far higher than the maximum of 0.1 µg g-1 reported by Mercadante and Rodriduez-Amaya (1998) for ‘Keitt’ and ‘Tommy Atkins’ mangoes from mature- green to ripe stages, indicating that in the fruit of this experiment, not only β-carotene, but also cis-β-cryp- toxanthin was a contributor to the vitamin A value for these fruit. The differences in the single carotenoid concen- trations respect to literature data for the same culti- var could be due to the different maturity degree of mango fruit. In fact, referring to pulp color parame- ters, the fruit used in our experiment had pulp a* val- ues similar to those reported by Rocha-Ribeiro et al. (2007) for the same cultivars and in this case the carotenoid amounts were similar, whereas pulp h° values for ‘Haden’ fruit were higher than those reported by Ornelas-Paz et al. (2007), indicating a less advanced ripening degree consistent with the lower carotenoid content of our results. The results of regression analysis showed that µa540 was positively related to a* and IY and nega- tively to h° pulp color parameters, confirming the results obtained by Spinelli et al. (2012, 2013) and Vanoli et al. (2016) for the same cultivars and by Vanoli et al. (2013) for ‘Tommy Atkins’ mangoes. Color changes in the pulp of mango fruit are usually accompanied by carotenoid accumulation. In this work significant correlations between µa540, pulp color parameters a*, h°, IY and carotenoids were found for both cultivars. Vanoli et al. (2016) found an increasing trend of total carotenoids content with µa540 in ‘Palmer’ and ‘Haden’ mangoes; they also found high positive correlations between total carotenoids and a* and IY and a higher negative cor- relation with h° following a logarithmic-law function with higher correlation in ‘Palmer’ than in ‘Haden’ cv., confirming the better relationships of pulp color in ‘Palmer’ than in ‘Haden’ fruits. High correlations between pulp color and all-trans-β-carotene, all- trans-violaxanthin and 9-cis-violaxanthin were also observed in ‘Ataulfo’ and in ‘Manila’ mangoes (Ornelas-Paz et al., 2008) with the highest correlation coefficients for a* and h° parameters; in ‘Manila’ mangoes the best results were associated with the concentrations of all-trans-violaxanthin and 9-cis-vio- laxhantin, while in ‘Ataulfo’ with all-trans-β-carotene, confirming that there is a cultivar specific relation- ship between pulp color and carotenoids content. Similar correlations were also found by Vasquez- Caceido et al. (2005) in 9 Thai mango cultivars (power law functions) and by Bicanic et al. (2010) in 21 mango homogenates (second order polynomial dependence). Differently from carotenoids, ascorbic acid content and total antioxidant capacity for both cultivars and TPC for ‘Palmer’ were not related to pulp color, mea- sured both by a*, IY and h° color parameters and µa540; this is not surprising as literature reported that AA and TPC contents with shelf life does not follow a well-defined increasing or decreasing trend (Robles- Sánchez et al., 2009 b; Ibarra-Garza et al., 2015). In conclusion our results confirmed that the absorption coefficient at 540 nm (µa540) can be used as a non-destructive maturity index for mangoes. In fact it was able to classify intact fruit of two mango cultivars according to pulp color, the destructive maturity index commonly used for mangoes, as well as according to the contents of total carotenoids and of individual carotenoid compounds and vitamin A value. The good correlations between µa540, pulp color parameters and carotenoids indicate that TRS is a suitable tool to sort fruit with different ripening degree, having specific carotenoid pattern. Adv. Hort. Sci., 2018 32(3): 407-420 418 Acknowledgements This work was funded by Lombardia Region (Italy) a n d M i n a s G e r a i s R e g i o n ( B r a z i l ) ( P r o g e t t o d i Cooperazione Scientifica e tecnologica “Approccio multidisciplinare per l’innovazione della filiera di frut- ti tropicali - TROPICO” ID 17077, Rif. n° AGRO-16). Thanks to R.M.A. Pimentel, EPAMIG (Minas Gerais, Brazil) for ‘Palmer’ and ‘Haden’ mangoes supply from experimental orchards. References AZEVEDO-MELEIRO C.H., RODRIGUEZ-AMAYA D.B., 2004 - Confirmation of the identity of the carotenoids of tropi- cal fruits by HPLC-DAD and HPLC-MS. - J. Food Comp. Anal., 17: 385-396. BEAULIEU J.C., LEA J.M., 2003 - Volatile and quality changes in fresh-cut mangos prepared from firm-ripe and soft-ripe fruit, stored in clamshell containers and passive MAP. - Postharvest Biol. Technol., 30: 15-28. BICANIC D., DIMITROVSKI D., LUTEROTTI S., TWISK C., BUI- JNSTERS J.P., DÓKA O., 2010 - Estimating rapidly and precisely the concentration of beta carotene in mango homogenates by measuring the amplitude of optother- mal signals, chromaticity indices and the intensities of Raman peaks. - Food Chem., 121: 832-838. BRAND-WILLIAMS W., CUVELIER M.E., BERSET C., 1995 - Use of free radical method to evaluate antioxidant activity. - LWT, 28: 25-30. CAPRA S., 2006 - Vitamin A, pp. 55-65. - In: BAGHURST K. (ed.) Nutrient reference values for Australia and New Zealand: Including recommended dietary intakes. NHMRC report, Commonwealth of Australia, Canberra, Australia, pp. 332. CUBEDDU R., D’ANDREA C., PIFFERI A., TARONI P., TORRI- CELLI A., VALENTINI G., DOVER C., JOHNSON D., RUIZ- ALTISENT M., VALERO C., 2001 - Nondestructive quan- tification of chemical and physical properties of fruits by time-resolved reflectance spectroscopy in the wave- length range 650-1000 nm. - Appl. Opt., 40: 538-543. ECCHER ZERBINI P., VANOLI M., RIZZOLO A., GRASSI M., PIMENTEL R.M.A., SPINELLI L., TORRICELLI A., 2015 - Optical properties, ethylene production and softening in mango fruit. - Postharvest Biol. Technol., 101: 58-65. ECCHER ZERBINI P., VANOLI M., RIZZOLO A., JACOB S., TORRICELLI A., SPINELLI L., SCHOUTEN R.E., 2009 - Time-resolved Reflectance Spectroscopy as a manage- ment tool in the fruit supply chain: an export trial with nectarines. - Biosyst. Eng., 102: 360-363. ELSHESHETAWY H.E., MOSSAD A., ELHELEW W.K., FARINA V., 2016 - Comparative study on the quality characteris- tics of some Egyptian mango cultivars used for food processing. - Annals Agr. Sci., 61: 49-56. FAO, 2015 - FAOSTAT. - Food and Agriculture Organization o f t h e U n i t e d N a t i o n s , R o m e , I t a l y , http://faostat.fao.org. GODOY H.T., RODRIGUEZ-AMAYA D.B., 1989 - Carotenoid composition of commercial mangoes from Brazil. - Lebensm.-Wiss. Technol., 22: 100-103. HEWAVITHARANA A.K., TAN Z.W., SHIMADA R., SHAW P.N., FLANAGAN B.M., 2013 - Between fruit variability o f t h e b i o a c t i v e c o m p o u n d s , β - c a r o t e n e a n d mangiferin, in mango (Mangifera indica L.). - Nutr. Diet., 70: 158-163. IBARRA-GARZA I.P., RAMOS-PARRA P.A., HERNÁNDEZ- BRENES C., JACOBO-VELÁZQUEZ D.A., 2015 - Effect of postharvest ripening on the nutraceutical and physico- chemical properties of mango (Mangifera indica L. cv. Keitt). - Postharvest Biol. Technol., 103: 45-54. JHA S.N., KINGSLY A.R.P., CHOPRA S., 2006 - Non-destruc- tive determination of firmness and yellowness of mango during growth and storage using visual spec- troscopy. - Biosyst. Eng., 94: 397-402. LIU F.X., FU S.F., BI X.F., CHEN F., LIAO X.J., HU X.S., WU J.H., 2013 - Physico-chemical and antioxidant proper- ties of four mango (Mangifera indica L.) cultivars in China. - Food Chem., 138: 396-405. LOW D.Y., D’ARCY B., GIDLEY M.J., 2015 - Mastication effects on carotenoid bioaccessibility from mango fruit tissue. - Food Res. Int., 67: 238-246. MANTHEY J.A., PERKINS-VEAZIE P., 2009 - Influences of harvest date and location on the levels of β-carotene, ascorbic acid, total phenols, the in vitro antioxidant capacity, and phenolic profiles of five commercial vari- eties of mango (Mangifera indica L.). - J. Agric. Food Chem., 57: 10825-10830. MARTELLI F., DEL BIANCO S., ISMAELLI A., ZACCANTI G., 2009 - Light propagation through biological tissue and other diffusive media: Theory, solution, and software. - SPIE Press, Washington, DC, USA, pp. 298. MERCADANTE A.Z., RODRIGUEZ-AMAYA D.B., 1998 - Effect of ripening, cultivar differences and processing on the carotenoid composition of mango. - J. Agric. Food Chem., 46, 128-130. MITRA S.K., 2016 - Mango production in the world - pre- sent situation and future prospect. - Acta Horticulturae, 1111: 287-296. NICOLAI B.M., DEFRAEYE T., DE KETELAERE B., HERRE- MANS E., HERTOG M.L.A.T.M., SAEYS W., TORRICELLI A . , V A N D E N D R I E S S C H E T . , V E R B O V E N P . , 2 0 1 4 - Nondestructive measurement of fruit and vegetable quality. - Ann. Rev. Food Sci. Technol., 5: 285-312. OLIVEIRA B.G., COSTA H.B., VENTURA J.A., KONDRATYUK T.P., BARROSO M.E.S., CORREIA R.M., PIMENTEL E.F., P I N T O F . E . , E N D R I G E R D C . , R O M Ã O W . , 2 0 1 6 - Chemical profile of mango (Mangifera indica L.) using electrospray ionisation mass spectrometry (ESI-MS). - Food Chem., 204: 37-45. ORNELAS-PAZ J.J., YAHIA E.M., GARDEA A.A., 2007 - Vanoli et al. - Quality and nutraceutical properties of mangoes according to TRS maturity class 419 Identification and quantification of xanthophyll esters, c a r o t e n e s , a n d t o c o p h e r o l s i n t h e f r u i t o f s e v e n Mexican mango cultivars by liquid chromatography- atmospheric pressure chemical ionization-time-of-flight mass spectrometry [LC-(Apcl+)-MS]. - J. Agric. Food Chem., 55: 6628-6635. ORNELAS-PAZ J.J., YAHIA E.M., GARDEA A.A., 2008 - Changes in external and internal color during posthar- vest ripening of ‘Manila’ and ‘Ataulfo’ mango fruit and relationship with carotenoid content determined by liq- uid chromatography-Apcl+-time-of-flight mass spec- trometry. - Postharvest Biol. Technol., 50: 145-152. P A L A F O X - C A R L O S H . , Y A H I A E . , I S L A S - O S U N A M . A . , G U T I E R R E Z - M A R T I N E X P . , R O B L E S - S Á N C H E Z M . , GONZÁLEZ-AGUILAR G.A., 2012 - Effect of ripeness stage of mango fruit (Mangifera indica L., cv. Ataulfo) on physiological parameters and antioxidant activity. - Sci. Hort., 135: 7-13. PICCHI V., MIGLIORI C., LO SCALZO R., CAMPANELLI G., FERRARI V., DI CESARE L.F., 2012 - Phytochemical con- tent in organic and conventionally grown Italian cauli- flower. - Food Chem., 130: 501-509. POTT I., BREITHAUPT D.E., CARLE R., 2003 - Detection of unusual carotenoid esters in fresh mango (Mangifera indica L. cv. ‘Kent’). - Phytochem., 64: 825-829. RIZZOLO A., BRAMBILLA A., VALSECCHI S., ECCHER ZERBINI P., 2002 - Evaluation of sampling and extraction proce- dures for the analysis of ascorbic acid from pear fruit tissue. - Food Chem., 77: 257-262. RIZZOLO A., VANOLI M., 2016 - Time-resolved technique for measuring optical properties and quality of food, pp. 187-224. - In: LU R. (ed.) Light scattering technolo- gy for food property, quality and safety assessment. CRC Press, Taylor & Francis Group, Boca Raton, FL, USA, pp. 439. RIZZOLO A., VANOLI M., SPINELLI L., TORRICELLI A., 2011 - Relationship between continuous wave reflectance measurement of pulp colour and the optical properties measured by Time-resolved Reflectance Spectroscopy in various fruit species, pp. 328-335. - In: ROSSI M. (ed.) Colour and colorimetry multidisciplinary contributions Vol. VII B. SIOF, Maggioli Editore, Santarcangelo di Romagna, Italy, pp. 348. RIZZOLO A., VANOLI M., SPINELLI L., TORRICELLI A., 2016 - Non-destructive assessment of pulp colour in mangoes by time-resolved reflectance spectroscopy: problems and solutions. - Acta Horticulturae, 1119: 147-154. ROBLES-SÁNCHEZ R.M., ISLAS-OSUNA M.A., ASTIAZARÁN- GARCÍA H., VÁZQUEZ-ORTIZ F.A., MARTÍN-BELLOSO O., GORINSTEIN S., GONZÁLEZ-AGUILAR G.A., 2009 b - Quality index, consumer acceptability, bioactive com- pounds, and antioxidant activity of fresh-cut “Ataulfo” mangoes (Mangifera indica L.) as affected by low-tem- perature storage. - J. Food Sci., 74: S126-S134. ROBLES-SÁNCHEZ R.M., ROJAS-GRAÜ M.A., ODRIOZOLA- SERRANO I., GONZÁLEZ-AGUILAR G.A., MARTIN-BEL- LOSO O., 2009 a - Effect of minimal processing on bioactive compounds and antioxidant activity of fresh- cut ‘Kent’ mango (Mangifera indica L.). - Postharvest Biol. Technol., 51: 384-390. ROCHA-RIBEIRO S.M., QUEIROZ J.H., LOPES RIBEIRO DE QUEIROZ M.E., CAMPOS F.M., PINHEIRO SANT’ANA H.M., 2007 - Antioxidant in mango (Mangifera indica L.) pulp. - Plant Foods Hum. Nutr., 62: 13-17. SAJILATA M.G., SINGHAL R.S., KAMAT M.Y., 2008 - The carotenoid pigment Zaexanthin. A Review. - Compr. Rev. Food Sci. Food Saf., 7: 29-49. SEPTEMBRE-MALATERRE A., STANISLAS G., DOURAGUIA E., GONTHIER M.P., 2016 - Evaluation of nutritional and antioxidant properties of the tropical fruits banana, lichi, mango, papaya, passion fruit and pineapple culti- vated in Réunion French Island. - Food Chem., 212: 225-233. SILVA K.D.R.R., SIRASA M.S.F., 2018 - Antioxidant proper- ties of selected fruit cultivars grown in Sri Lanka. - Food Chem., 238: 203-208. SINGLETON V.L., ORTHOFER R., LAMUELA-RAVENTOS R.M., 1999 - Analysis of total phenols and other oxidation s u b s t r a t e s a n d a n t i o x i d a n t s b y m e a n s o f F o l i n - Ciocalteu reagent. - Methods Enzymol., 299: 152-178. SPINELLI L., RIZZOLO A., VANOLI M., GRASSI M., ECCHER ZERBINI P., PIMENTEL R.M.A., TORRICELLI A., 2012 - Optical properties of pulp and skin in Brazilian mangoes in the 540-900 nm spectral range: implication for non- destructive maturity assessment by time-resolved reflectance spectroscopy. - Proc. 3rd CIGR Int. Conf. Agric. Eng. CIGR-AnEng2012, Valencia, Spain, 8-12 July. SPINELLI L., RIZZOLO A., VANOLI M., GRASSI M., ECCHER ZERBINI P., PIMENTEL R.M.A., TORRICELLI A., 2013 - Nondestructive assessment of fruit biological age in Brazilian mangoes by time-resolved reflectance spec- troscopy in the 540-900 nm spectral range. - Inside Food Symposium, Leuven, Belgium, 9-12 April. TIJSKENS L.M.M., ECCHER ZERBINI P., SHOUTEN R.E., VANOLI M., JACOB S., GRASSI M., CUBEDDU R., SPINEL- LI L., TORRICELLI A., 2007 - Assessing harvest maturity in nectarines. - Postharvest Biol. Technol., 45: 204-213. TORRICELLI A., SPINELLI L., CONTINI D., VANOLI M., RIZZO- L O A . , E C C H E R Z E R B I N I P . , 2 0 0 8 - T i m e - r e s o l v e d reflectance spectroscopy for nondestructive assessment of food quality. - Sens. Instrum. Food Qual. Saf., 2: 82- 89. VALENTE A., ALBUQUERQUE T.G., SANCHES-SILVA A., COSTA H.S., 2011 - Ascorbic acid content in exotic fruits: A contribution to produce quality data for food composition databases. - Food Res. Int., 44: 2237-2242. V A N O L I M . , P E R E I R A T . , G R A S S I M . , S P I N E L L I L . , FILGUEIRAS H., TIJSKENS L.M.M., RIZZOLO A., TORRI- CELLI A., 2011 - Changes in pulp colour during posthar- vest ripening of ‘Tommy Atkins’ mangoes and relation- ship with optical properties measured by Time-resolved Reflectance Spectroscopy. - Proceedings of the 6th Adv. Hort. Sci., 2018 32(3): 407-420 420 CIGR Section VI International Symposium “Towards a Sustainable Food Chain” Food process, Bioprocessing and Food Quality Management, Nantes, France , 18-20 April, pp. 5. VANOLI M., RIZZOLO A., GRASSI M., SPINELLI L., ECCHER ZERBINI P., PIMENTEL R.M.A., TORRICELLI A., 2013 - Quality of Brazilian mango fruit in relation to optical properties non-destructively measured by time-resolved reflectance spectroscopy, pp. 177-181. - In: BELLON MAUREL V., P. WILLIAMS, and G. DOWNEY (eds.) NIR 2013 Proceedings Picking up good vibrations. IRSTEA. France Institut National de recherche en sciences et technologies pour l’environnement et l’agriculture, Montpellier, France, pp. 749. VANOLI M., RIZZOLO A., SPINELLI L., AZZOLLINI S., TORRI- CELLI A., 2016 - Carotenoid content and pulp colour non-destructively measured by time-resolved reflectan- ce spectroscopy in different cultivars of Brazilian man- goes. - Acta Horticulturae, 1119: 305-312. VÁSQUEZ-CAICEDO A.L., HELLER A., NEIDHART S., CARLE R., 2006 - Chromoplast morphology and β-carotene accumulation during postharvest ripening of Mango cv. ‘Tommy Atkins’. - J. Agric. Food Chem., 54: 5769-5776. VÁSQUEZ-CAICEDO A.L., SRUMSIRI P., CARLE R., NEIDHART S., 2005 - Accumulation of all-trans-β-carotene and its 9-cis and 13-cis stereoisomers during postharvest ripening of nine Thai mango cultivars. - J. Agric. Food Chem., 53: 4827-4835. Y A H I A E . M . , O R N E L A S - P A Z J . J . , G A R D E A A . , 2 0 0 6 - Extraction, separation and partial identification of ‘Ataulfo’ mango fruit carotenoids. - Acta Horticulturae, 712: 333-338.