Food Research International 38 (2005) 931–936 www.elsevier.com/locate/foodres
Color and carotenoid profile of Spanish Valencia late ultrafrozen orange juices A.J. Mele´ndez-Martı´nez a, G. Britton b, I.M. Vicario a, F.J. Heredia a
a,*
Department of Nutrition and Food Science, Faculty of Pharmacy, University of Seville, C/P. Garcı´a Gonza´lez 2, 41012 Seville, Spain b School of Biological Sciences, University of Liverpool, Crown Street, L69 7ZB Liverpool, United Kingdom Received 1 June 2004; accepted 29 January 2005
Abstract Twelve ultafrozen orange juices obtained from the variety Valencia late were analysed to characterize the color and the carotenoid profile of this recently developed product. Reflectance color measurements were carried out by spectroradiometry, using a white and a black background. In order to study the carotenoid pigments occurring in the orange juice, HPLC with gradient elution with methanol (MeOH), methyl tert-butyl ether (MTBE) and water as mobile phase was used. Separations were performed on a C30 column. Identification of carotenoids was made by comparison of their spectral and chromatographic characteristics with those corresponding to standards when possible. To help identify the peaks in the chromatograms, aliquots of orange juice extracts were analysed before and after treatment with hydrochloric acid (HCl) and sodium borohydride (NaBH4), to check the presence of 5,6-epoxy-carotenoids on one hand, and, on the other hand, ketocarotenoids or carotenals. 2005 Elsevier Ltd. All rights reserved. Keywords: C30; Carotenoids; CIELAB; Color; Orange juice; HPLC; TLC; Tristimulus colorimetry
1. Introduction Color is one of the most important attributes of orange juice and is mainly due to carotenoid pigments. Citrus fruits in general are a complex source of carotenoids, with the largest variety of these pigments found in any family of fruit (Gross, 1987). Several studies have shown the importance of color as a quality parameter in citrus products in general. Thus, in the United States, for instance, this attribute is evaluated for the commercial classification of the product in relation to its quality (Huggart, Fellers, De Jager, & Brady, 1979; Huggart, Petrus, & Buslig, 1977; Tepper, 1993). Regarding carotenoids, their role in juice color along with the recent growing interest in these pig*
Corresponding author. Tel.: +34 95455 6761; fax: +34 95455 7017. E-mail addresses:
[email protected] (G. Britton),
[email protected] (F.J. Heredia). 0963-9969/$ - see front matter 2005 Elsevier Ltd. All rights reserved. doi:10.1016/j.foodres.2005.01.014
ments owing to their potential health benefits, have prompted the development of many analytical methods for the characterization of these compounds in a wide variety of foodstuffs. Several characterization studies of carotenoids in different types of orange juices have been reported in recent years (Mouly, Gaydou, Lapierre, & Corsetti, 1999; Lee, 2001; Lee & Castle, 2001; Lee, Castle, & Coates, 2001). Ultrafrozen orange juice is a recently developed type of commercial juice that has not been subjected to high temperatures nor concentration processes. As a result of this novel procedure, the product has an unusual deep orange color, even compared to other juices obtained from Valencia oranges, which have long been appreciated worldwide due to their deep orange color (Francis & Clydesdale, 1975; Robards & Antolovich, 1995). This paper reports a study of the color of this novel product and a characterization of its carotenoid profile.
932
A.J. Mele´ndez-Martı´nez et al. / Food Research International 38 (2005) 931–936
2. Materials and methods 2.1. Orange juice samples Twelve ultrafrozen orange juices were supplied by the industry as representative samples from the 2002 season. In the industrial processing, once the oranges are squeezed, the juice is cooled and immediately frozen in a freezing tunnel by means of liquid nitrogen, so the product is subjected neither to high temperatures nor concentration processes. The commercial samples are kept in freezing chambers between 18 and 21 C until distribution. The samples analysed in this study were kept in a freezer at 21 C until analysis in the laboratory. Samples of juice (1 L) were thawed by means of a Samsung microwave oven at 750 W for about 9 min. These conditions allowed the samples to be rapidly thawed and risk of isomerization was avoided since the temperature of the samples during thawing was always under 30 C. 2.2. Extraction of carotenoids Extraction and saponification procedures have been reported elsewhere (Mele´ndez-Martı´nez, Vicario, & Heredia, 2003). Thawed ultrafrozen orange juice (10 mL) was centrifuged for 5 min at 5000 rpm. After discarding the clarified juice, the pellet containing the carotenoids was successively stirred and centrifuged, as described above, with methanol (containing 0.1% of 2,6-di-tert-butyl-p-cresol (BHT)) until no more color remained. All the methanolic extracts were combined and 10 mL of methanolic potassium hydroxide (10%, w/v) was added and the mixture allowed to stand in the dark at room temperature for 1 h. The mixture was transferred to a separatory funnel and carotenoids were extracted with 30 mL of dichloromethane containing BHT (0.1%). The traces of alkali were removed by washing four times with water and the yellow extract was evaporated to dryness at temperature lower than 35 C. Finally, the extract of carotenoids was re-dissolved in 1.5 mL of ethanol (containing 0.1% BHT) and filtered through Millipore PVDF Millex filters (13 mm · 0.45 lm) (Bedford, MA, USA). 2.3. HPLC analysis HPLC analyses were carried out by means of a Hewlett–Packard 1100 system, equipped with a quaternary pump, a photodiode array detector and a column temperature control module (Hewlett–Packard, Palo Alto, CA, USA). A 20 lL loop and a C30 YMC column (5 lm, 250 · 4.6 mm) (Wilmington, NC, USA) with a Kromasil guard-column (10 · 4 mm) (Hichrom Ltd., Reading, UK) were used. The column was kept at 17 C and
the flow rate was 1 mL min1. The diode array detector was set at 430, 450 and 486 nm. The gradient elution was the same as described by Mouly et al. (1999): 0 min: 90% MeOH + 5% MTBE + 5% water; 12 min: 95% MeOH + 5% MTBE; 25 min: 89% MeOH + 11% MTBE; 40 min: 75% MeOH + 25% MTBE; 60 min: 50% MeOH + 50% MTBE; 62 min: 90% MeOH + 5% MTBE + 5% water. MeOH and MTBE contained a small proportion of BHT (0.1%) and triethylamine (0.05%) in order to protect the carotenoids during the chromatographic analysis (Hart & Scott, 1995). Equilibrating time between consecutive injections was 12 min. At the end of each day the column was washed with MTBE. 2.4. Isolation of carotenoid standards Carotenoid standards were isolated according to standard procedures (Britton, Liaaen-Jensen, & Pfander, 1995). Neoxanthin, violaxanthin and lutein were obtained from spinach (Spinacia oleracea L.) leaves. b-Cryptoxanthin and zeaxanthin were obtained from red peppers (Capsicum annuum L.). a- and b-carotene were isolated from palm oil. Luteoxanthin was obtained by treating an ethanolic solution of violaxanthin with a few drops of aqueous HCl (0.1 M) for a few seconds. The reaction was stopped by means of the addition of diethyl ether and aqueous NaCl (10%) in the separatory funnel. The ether phase was washed with the NaCl solution four more times. Since some auroxanthin was also formed, the extract was chromatographied on plates of silica gel 60 GF254 (20 · 20 cm) (Merck, Darmstadt, Germany), using light petroleum ether (bp 65–95 C)–acetone– diethylamine (10:4:1) as solvent system (Mı´nguez-Mosquera & Hornero-Me´ndez, 1993). The main band (Rf = 0.26), corresponding to luteoxanthin, was scraped from the plate. Auroxanthin was obtained by treating an ethanolic solution of violaxanthin with a few drops of aqueous HCl (0.1 N) for a few minutes. After stopping the reaction as mentioned above, the auroxanthin extract was washed four more times with aqueous NaCl (10%). Neochrome was obtained in the same way from neoxanthin. Purity of all the standards obtained was checked by means of spectrophotometry and HPLC. No pure standards of antheraxanthin and lutein 5,6epoxide were isolated, although spectra in the mobile phase were recorded for samples prepared from zeaxanthin and lutein, respectively, by treatment with 3-chloroperoxybenzoic acid (Barua & Olson, 2001). To obtain cis isomers of the pure carotenoids isolated, ethanolic solutions were heated at 80 C for 30 min in a water bath and subsequently illuminated overnight by means of a powerful lamp (500 W). The vials
A.J. Mele´ndez-Martı´nez et al. / Food Research International 38 (2005) 931–936
containing the carotenoid solutions were gently blanketed with nitrogen to avoid oxidation reactions. 2.5. Acidification and reduction of the extract of orange juice carotenoids To check the occurrence of 5,6-epoxides, an ethanolic extract of orange juice carotenoids was injected in the HPLC system before and after treatment with aqueous HCl (0.1 M) for 1 min. The reaction was stopped and the extract washed as detailed above. To check for the presence of ketocarotenoids and/or carotenals, an ethanolic extract was analysed by HPLC before and after treatment with NaBH4 in a fridge for 3 h (Mı´nguez-Mosquera, 1997). The reaction was stopped by partitioning into dichloromethane and washing several times with distilled water. 2.6. Objective color measurement Color measurements were made with a CAS 140 B spectroradiometer (Instrument Systems, Munich, Germany), equipped with a Top 100 telescope optical probe (Instrument Systems, Munich, Germany) and a Tamron zoom Model SP 23A (Tamron USA, Inc., Commack, NY, USA). Blank was measured by using a reference BaSO4 pressed plate Model USRS-99-010 (Labsphere, Inc. North Sutton, NH, USA). Each sample was measured against a white background and a black background, since some studies have shown that sometimes reflectance measurements against a black background are better correlated with visual assessment of color and other parameters (Huang, Francis, & Clydesdale, 1970a; Huang, Francis, & Clydesdale, 1970b; Terrab, Dı´ez, & Heredia, 2003). The whole visible spectrum (380–770 nm) was recorded (Dk = 1 nm) and Illuminant
933
D65 and 10 Observer were considered as references. The color parameters of the uniform color space CIELAB ðL ; a ; b ; C ab and hab Þ were obtained directly from the apparatus.
3. Results and discussion 3.1. HPLC analysis of carotenoids A typical chromatogram of carotenoids of Valencia late ultrafrozen orange juices is shown in Fig. 1. Carotenoids were identified by comparison of their spectral and chromatographic behavior with that of standards when possible. To help the identification of the peaks in the chromatograms, changes as a result of the treatments with HCl and NaBH4 were taken into account to reveal the presence of epoxycarotenoids and carotenoids with carbonyl groups, respectively. Results are summarized in Table 1. C30 columns provide a better resolution between some carotenoids and allow geometrical isomers to be separated (Emenhiser, Simunovic, Sander, & Schwartz, 1996; Sander, Sharpless, & Pursch, 2000). However, due to the complex carotenoid profile of orange juices, their usage lead to the appearance of many peaks corresponding to mixtures of carotenoids. Thus, we strongly recommend to compare the spectra taken at different points of the peak instead of considering an average spectrum, which could lead to misleading spectral data. Spectra of peaks 1–5 indicated clearly that they were carotenoid breakdown products with seven or fewer conjugated double bonds. The reduction of the extract indicated that peaks 2, 3 and 4 were mixtures probably containing carotenoids with carbonyl-groups.
Fig. 1. Chromatogram of Valencia late orange juices carotenoids plotted at 450 nm. For peak identification see Table 1.
A.J. Mele´ndez-Martı´nez et al. / Food Research International 38 (2005) 931–936
934
Table 1 Chromatographic and spectral characteristics and preliminary identification of orange juice carotenoids Peak
RTa
Maxima observedb
Carotenoid
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31
4.61 5.05 5.54 5.85 6.33 7.24 7.98 8.52 9.32 9.92 10.52 10.99 11.49 12.60 12.97 13.47 14.37 16.38 17.58 20.06 21.21 21.95 22.72 27.12 29.52 33.27 38.96 42.90 44.75 49.63 52.01
356, 376, 374, 358, 372, 416, 332, 330, 330, 324,
374, 396, 396, 370, 392, 440, 418, 432 402, 420
396 420 420 392 416 468 442, 468
Mixture of apocarotenoids Mixture of apocarotenoids Mixture of apocarotenoids Mixture of apocarotenoids Apocarotenoid Neoxanthin Mixture containing a neoxanthin isomer
424, 448
Mixture containing neochrome
330, 418, 334, 330, 400,
408, 440, 438, 438, 424,
430, 458 468 464 464 448
a b
Mixture containing a cis-violaxanthin isomer Mixture containing a cis-violaxanthin isomer Violaxanthin
332, 436, 460 Upside 440, 468, downside 328, 414, 436, 464 404, 418, 444 428, 452 420, 446, 472 430, 452 452, 478 332, 440, 468 422, 446, 474 452, 478 380, 400, 424 424, 446, 474 452, 478 342, 446, 470
Luteoxanthin Mixture containing cis-antheraxanthin cis-Antheraxanthin isomer Mixture containing antheraxanthin and cis-violaxanthin Luteoxanthin isomer Mutatoxanthin isomer Lutein Mutatoxanthin isomer Zeaxanthin cis-Antheraxanthin isomer a-Cryptoxanthin or zeinoxanthin b-Cryptoxanthin f-Carotene a-Carotene b-Carotene cis-b-Carotene
RT, retention time (min). In nm.
Peak 6 showed a spectrum almost identical to that of the 9 0 Z-isomer of neoxanthin isolated from spinach. This fact, as well as its early elution and disappearance after treatment with HCl led to its tentative identification as neoxanthin. As a result of the acidic treatment of the carotenoid extract, peak 9 increased considerably its height. Taking into account that the injection of the standard carotenoids isolated revealed that 5,8-epoxycarotenoids eluted a little later than the corresponding 5,6-epoxycarotenoids as well as its spectral characteristics, it was tentatively identified as a mixture containing neochrome. Peak 13 matched with the chromatographic and spectral characteristics of the violaxanthin standard isolated from spinach. Retention times of peaks 11 and 12 and spectra recorded throughout the peaks matched with those corresponding to violaxanthin cis isomers formed during heating and illumination. The pronounced cis peak at 330 nm may indicate that they were 13 or 15-cis isomers. Peak 16 matched with the characteristics of the luteoxanthin isomer obtained from violaxanthin. Peak 19 (the main peak) showed different spectra depending on whether they were taken upslope or downslope. Spectra taken upslope and retention time of the
peak matched with those corresponding to antheraxanthin obtained from zeaxanthin by treating with 3-chloroperoxybemzoic acid. On the other hand, the spectra taken downslope, as well as the retention time, matched with those corresponding to a cis isomer of violaxanthin (with a smooth cis peak at 328 nm) obtained by heating and illumination. Peaks 17 and 18 disappeared after acidification with hydrochloric acid. The fine structure of their spectra was not as clear as those corresponding to other epoxycarotenoids such as neoxanthin and violaxanthin. These facts, along with their chromatographic behaviour, as compared with the cis isomers of those latter carotenoids, and the presence of pronounced cis peaks led us to identify them tentatively as cis isomers of antheraxanthin (probably 13 and 15 cis isomers). Peak 20 was identified tentatively as a luteoxanthin isomer from its spectrum, retention time and behavior after acidification. Peaks 21 and 23 increased considerably after acid treatment, which indicated that they corresponded to 5,8-epoxycarotenoids. This behavior, their retention times and their spectra may suggest that they were mutatoxanthin isomers.
A.J. Mele´ndez-Martı´nez et al. / Food Research International 38 (2005) 931–936
Peaks 22, 24, 27, 29 and 30 matched with the characteristics of the standards of lutein, zeaxanthin, b-cryptoxanthin, a-carotene and b-carotene, respectively. Peak 25 decreased substantially in height as a result of the treatment with HCl. Its spectrum, containing a spectral shoulder, was not typical of most of epoxycarotenoids since it did not have much fine structure. Taking into account all this, it has been tentatively identified as a cis isomer of antheraxanthin. Peaks 26 and 28 have been tentatively identified as a-cryptoxanthin or zeinoxanthin and f-carotene, respectively, considering their retention times and spectral characteristics, which agreed well with those found in the literature (Britton, Liaaen-Jensen, & Pfander, 2004; Rodriguez-Amaya, 2001). Peak 31 was identified as a cis-b-carotene isomer on the basis of its chromatographic and spectral characteristics, which were in accordance with those corresponding with one isomer of b-carotene obtained by heating and illumination. Auroxanthin and lutein 5,6-epoxide were not detected in the juice. 3.2. Chromatic characterization The color coordinates of the juice are summarized in Table 2. There was great homogeneity among the color parameters measured for 10 of the 12 orange juices analysed, but the remaining two showed considerably lower values of L ; a ; b and C ab (45.51 and 44.97, 8.34 and 8.93, 51.61 and 52.64, 52.28 and 53.39, respectively, for white background; 36.34 and 36.73, 4.67 and 5.02, 37.97 and 41.02, 38.25 and 41.32, respectively, for black background), which may indicate that they were obtained from slightly unripe oranges. Of all the chromatic parameters measured, the psychometric ones ðL ; hab and C ab Þ are usually those that indicate more intuitively the color of the juices analysed. L* values agreed with bright colors. The values obtained for hue (hab), the qualitative attribute of color, matched with orange colors, as distinct from the yellowish colors that characterize many other orange juices. Regarding chroma, the quantitative attribute of colorfulness, its values matched with vivid colors.
Table 2 Chromatic coordinates of the juice White background
L* a* b* C ab hab
Black background
Mean ± SD
Range
Mean ± SD
Range
68.22 ± 10.77 13.83 ± 2.65 67.46 ± 7.45 68.88 ± 7.80 78.52 ± 1.14
44.97–74.24 8.34–16.53 51.61–74.02 52.28–75.84 77.27–80.82
57.11 ± 9.67 7.96 ± 1.77 53.19 ± 6.95 53.79 ± 7.11 81.58 ± 0.97
36.34–63.24 4.67–10.26 37.97–61.15 38.25–61.97 83.02–80.27
935
Acknowledgements We gratefully acknowledge Zumos Vitafresh (Almonte, Spain) for supplying the samples analysed and for collaboration in this study.
References Barua, A. B., & Olson, J. A. (2001). Xanthophyll epoxides, unlike bcarotene monoepoxides, are not detectibly absorbed by humans. The Journal of Nutrition, 131, 3212–3215. Britton, G., Liaaen-Jensen, S., & Pfander, H. (1995). Carotenoids: Isolation and analysis (Vol. 1A). Basel, Switzerland: Birkha¨user. Britton, G., Liaaen-Jensen, S., & Pfander, H. (2004). Carotenoids. Handbook. Basel, Switzerland: Birkha¨user. Emenhiser, C., Simunovic, N., Sander, L. C., & Schwartz, S. J. (1996). Separation of geometrical carotenoid isomers in biological extracts using a polimeric C30 column in reversed-phase liquid chromatography. Journal of Agricultural and Food Chemistry, 44, 3887–3893. Francis, F. J., & Clydesdale, F. M. (1975). Food colorimetry: Theory and applications. Westport: AVI Publishing Company. Gross, J. (1987). Pigments in fruits. London: Academic Press. Hart, D. J., & Scott, K. J. (1995). Development and evaluation of an HPLC method for the analysis of carotenoids in foods, and the measurement of the carotenoid content of vegetables and fruits commonly consumed in the UK. Food Chemistry, 54, 101–111. Huang, I.-L., Francis, F. J., & Clydesdale, F. M. (1970a). Colorimetry of foods. 3. Carrot puree. Journal of Food Science, 35, 772–773. Huang, I.-L., Francis, F. J., & Clydesdale, F. M. (1970b). Colorymetry of foods. 2. Color measurement of squash puree using the Kubelka–Munk concept. Journal of Food Science, 35, 315–317. Huggart, R. L., Fellers, P. J., De Jager, G., & Brady, J. (1979). The influence of color on consumer preferences for Florida frozen concentrated grapefruit juices. Proceedings of the Florida State Horticulture Society, 92, 148–151. Huggart, R. L., Petrus, D. R., & Buslig, B. S. (1977). Color aspects of Florida commercial grapefruit juices, 1976–77. Proceedings of the Florida State Horticulture Society, 90, 173–175. Lee, H. S. (2001). Characterization of carotenoids in juice of red navel orange (Cara Cara). Journal of Agricultural and Food Chemistry, 49, 2563–2568. Lee, H. S., & Castle, W. S. (2001). Seasonal changes of carotenoid pigments and color in Hamlin, Earlygold, and Budd Blood orange juices. Journal of Agricultural and Food Chemistry, 49, 877–882. Lee, H. S., Castle, W. S., & Coates, G. A. (2001). High-performance liquid chromatography for the characterization of carotenoids in the new sweet orange (Earlygold) grown in Florida, USA. Journal of Chromatography A, 913, 371–377. Mele´ndez-Martı´nez, A. J., Vicario, I. M., & Heredia, F. J. (2003). A routine high-performance liquid chromatography method for carotenoid determination in ultrafrozen orange juices. Journal of Agricultural and Food Chemistry, 51(15), 4219–4224. Mı´nguez-Mosquera, M. I. (1997). Clorofilas y carotenoides en Tecnologı´a de los Alimentos. Sevilla: Secretariado de publicaciones de la Universidad de Sevilla. Mı´nguez-Mosquera, M. I., & Hornero-Me´ndez, D. (1993). Separation and quantification of the carotenoid pigments in red peppers (Capsicum annuum L.), paprika, and oleoresin by reversed-phase HPLC. Journal of Agricultural and Food Chemistry, 41, 1616–1620. Mouly, P. P., Gaydou, E. M., Lapierre, L., & Corsetti, J. (1999). Differentiation of several geographical origins in single-strength Valencia orange juices using quantitative comparison of carotenoid profiles. Journal of Agricultural and Food Chemistry, 47, 4038–4045.
936
A.J. Mele´ndez-Martı´nez et al. / Food Research International 38 (2005) 931–936
Robards, K., & Antolovich, M. (1995). Methods for assessing the authenticity of orange juice. Analyst, 120, 1–28. Rodriguez-Amaya, D. B. (2001). A guide to carotenoid analysis in foods. Washington, DC: ILSI Press. Sander, L. C., Sharpless, K. E., & Pursch, M. (2000). C30 stationary phases for the analysis of food by liquid chromatography. Journal of Chromatography A, 880, 189–202.
Tepper, B. J. (1993). Effects of a slight color variation on consumer acceptance of orange juice. Journal of Sensory Studies, 8, 145–154. Terrab, A., Dı´ez, M. J., & Heredia, F. J. (2003). Chromatic characterisation of Moroccan honeys by diffuse reflectance and Tristimulus Colorimetry – Non-uniform and uniform colour spaces. Food Science and Technology International, 8(4), 189–195.
