JFS C: Food Chemistry and Toxicology

Carotenoids and Color of Fruit Juice

and Milk Beverage Mixtures
A. ZULUETA, M.J. ESTEVE, AND A. FRIGOLA

ABSTRACT: Seventeen commercially available, ready to drink fortified beverages consisting of mixtures of fruit
juices and milk were analyzed to evaluate their carotenoid profile (including their Z/E stereoisomers) and color

C: Food Chemistry & Toxicology

during their commercial shelf life. Lightness (L*) was found to be correlated with the content of milk in the mix-
tures (r = 0.649) whereas red-yellow colors were correlated with the contents of -carotene, -cryptoxanthin, and
-carotene supplied by the fruit. The beverages stored under refrigeration (4 2 C) showed higher luminosity (L*)
and higher saturation of color (C*) and yellowness (b*). Differences (P < 0.05) in the levels of carotenoids were found
among the beverages. Common fruits such as apple, lemon, pear, strawberry, kiwifruit, pineapple, and banana were
low in carotenoids. However, orange, apricot, mango, and peach contributed significantly (P < 0.05) to increasing
-cryptoxanthin and -carotene concentrations. Passion fruit supplied -carotene, and the presence of carrot in-
creased the levels of -carotene and -carotene. Conversion of all-E -carotene and -cryptoxanthin to their Z iso-
mers took place in 8 of the 17 samples containing -carotene and in 9 of the 13 samples containing -cryptoxanthin,
respectively, which resulted in some loss of provitamin A activity and nutritional value.
Keywords: color, carotenoids, fruit juicemilk beverages, provitamin A, Z/E isomers

Introduction of color, which is one of the most important attributes related to

C urrent developments in new ready-to-drink beverages are

based on drinks containing combinations of milk and fruit
juices, which have been receiving considerable attention as their
market potential is growing (Sharma 2005). Mixed fruit juice and
milk beverages fortified with vitamins, minerals, and fibers are
among the most widely consumed functional foods (Pszczola
2005). Consumption of these products has increased by 30% in
Europe in the last 10 y (MAPA 2003). Despite the growing con-
sumption of nutritional supplements (Kuzminski 1999) and forti-
fied foods such as ATBC drinks (supplemented with vitamins A, C,
and E as protective micronutrients) (Carle 1999; Marx and others
2000; Schieber and others 2002), little attention has so far been paid
to the carotenoid stereoisomers present in these types of food and
to the amounts that finally reach the consumer.
The importance of carotenoids is mainly based on their
provitamin A activity and their antioxidant capacity (Britton 1995),
although not all the carotenoids present in fruits and vegetables are
precursors of vitamin A. Certain carotenoids, including -carotene,
-carotene, and -cryptoxanthin, have provitamin A activity, be-
ing transformed into retinol by mammals (Simpson and Chichester
1981). Carotenoids predominantly occur in the thermodynamically
favored all-E isomers (Rodriguez-Amaya 2001). As they are highly
unsaturated, the application of some industrial treatments leads to
the formation of Z isomers, which do not possess the same provi-
tamin activity as E isomers. Heat, light, oxygen, and the presence of
electrophilic compounds are the factors that mostly cause degrada-
tion of carotenoids in foods (Dugave and Demange 2003).
The stability of carotenoids during their commercial shelf life is
very important if the final product is to be attractive and accept-
able. The isomerization and oxidation of carotenoids result in a loss
MS 20070168 Submitted 3/7/2007, Accepted 8/7/2007. Authors are with Area a shelf life of up to 12 mo. Six of them (1A, 2A, 3A, 4A, 9B, 10B)
de Nutricion y Bromatologa, Facultat de Farmacia, Univ. de Valencia, Ave. had been processed by pasteurization and were kept under refrig-
Vicent Andres Estelles s/n, 46100 Burjassot (Valencia), Spain. Direct in-
quiries to author Frgola (E-mail: ana.frigola@uv.es).
quality affecting choice of purchase (Calvo and others 2001; Baker
and Gunter 2004). Color is, alongside freshness, one of the main
criteria for food selection by the consumer. For this reason, color-
ings are added to many foods to make them appear healthier and
more appealing. The objective of the present study was to evalu-
ate the quality of commercial beverage mixtures of fruit juices and
milk in terms of their total carotenoid content, provitamin A, and
color.
Materials and Methods
Chemicals and reagents
-Carotene, lutein, and butylated hydroxytoluene (BHT) (special
grade) were purchased from Sigma (Steinheim, Germany). Zeaxan-
thin was purchased from Fluka (Steinheim, Germany). Ammonium
acetate (LC grade), petroleum ether, hexane (LC grade), potas-
sium hydroxide (85%), and tert-butyl methyl ether (TBME) (LC
grade) were purchased from Scharlau (Barcelona, Spain). Magne-
sium hydroxide carbonate (40% to 45%) was purchased from Pan-
reac (Barcelona, Spain); ethanol, acetonitrile (special grade), di-
ethyl ether, methanol, and sodium chloride (special grade) were
purchased from Baker (Deventer, The Netherlands); and chloro-
form was obtained from Merck (Darmstadt, Germany).
Samples
Two batches of 17 different juicemilk beverages made by 5 dif-
ferent manufacturers were purchased from a local supermarket
(Valencia, Spain) and were coded as A, B, C, D, and E (same letter
indicates same manufacturer). All of them were packaged in Tetra
Brik cartons. Eleven of them had been processed by UHT and were
kept in the supermarket at room temperature (20 2 C), having
eration (4 2 C), having a shelf life of up to 2 mo as declared on
the label by the producer. Table 1 gives the details (as indicated on


C 2007 Institute of Food Technologists Vol. 72, Nr. 9, 2007JOURNAL OF FOOD SCIENCE C457
doi: 10.1111/j.1750-3841.2007.00550.x
Further reproduction without permission is prohibited
 Carotenoids and color of juicemilk mixtures . . .

the label) of each of the samples analyzed. All of them were ana-
lyzed at the middle of their shelf life period (Cortes and others 2005,
2006). Once the samples were in the laboratory, the packages were
kept under the same conditions as in the supermarket. The pH of
the beverages ranged from 2.96 to 4.11. Brix ranged from 4.00 to
14.43, in agreement with the declared addition of carbohydrates by
the manufacturer, and density was between 1.017 and 1.062 g/cm3 .
Analyses were performed in triplicate.
Analysis of carotenoids
The method applied was a modification of the method of Cortes
and others (2004).
The LC system used for the analysis consisted of a 1050 se-
C: Food Chemistry & Toxicology
(Agilent, 1100 series), an on-line degassing system, and a Chem-
Station (series A.06.03) data system (Hewlett-Packard, Waldbronn,
Germany). A 250 4.6 mm Vydac 201TP54 reverse phase C 18
column with a particle size of 5 m and a Vydac 201TP pre-
column (guard column) (4.6 mm i.d. cartridge with 5-m parti-
cles) (Hesperia, Calif., U.S.A.) were used. The mobile phase used
was methanol + 0.1 M ammonium acetate (A), tert-butyl methyl
ether (B), and water (C) in a concentration gradient at 0 min 95%
(A) and 5% (C), at 3 min 100% (A), at 5 min 95% (A) and 5% (B), at
10 min 86% (A) and 14% (B), at 15 min 75% (A) and 25% (B), at 22
min 95% (A) and 5% (B), and at 23 min 100% (A); and a temperature
gradient was applied (0 min, 20 C; 6 min, 30 C; 22 min, 20 C).
Carotenoids were extracted from 30 g of juicemilk beverage.

ries chromatograph with a quaternary pump system, a diode ar- An extraction process (35 mL ethanol/hexane, 4:3, v/v) was per-
ray detector (Hewlett-Packard, 1100 series), a column thermostat formed, followed by saponification with 10 mL diethyl ether/10 mL

Table 1 --- Composition of the various juicemilk beverages analyzed.

Sample Composition Juice type
a
1A Juice 25%, skim milk 20% Orange and pineapple
Water, sugar, calcium salts, stabilizer (pectin), acidulant (citric acid,
vitamin C, aromas, and vitamin A)
2Aa Juice 25%, skim milk 20% Orange and mango
Water, sugar, calcium salts, stabilizer (pectin), acidulant (citric acid,
vitamin C, aromas, and vitamin A)
3Aa Juice 25%, skim milk 20% Strawberry, banana, and orange
Water, sugar, calcium salts, stabilizer (pectin), acidulant (citric acid,
vitamin C, aromas, and vitamin A), colorant (allura red)
4Aa Juice 25%, skim milk 20% Peach and apricot
Water, sugar, calcium salts, stabilizer (pectin), acidulant (citric acid,
vitamin C, aromas, and vitamin A)
5A Juice, skim milk Orange, apple, pineapple, and lemon
Sugar, fiber, pectin, acidulant (citric acid), vitamins A, C, and E, and
aromas
6A Juice, skim milk Peach, apricot, apple, and lemon
Sugar, fiber, pectin, acidulant (citric acid), vitamins A, C, and E, and
aromas
7A Juice, skim milk Orange, mango, pineapple, and lemon
Sugar, fiber, pectin, acidulant (citric acid), vitamins A, C, and E, and
aromas
8A Juice, skim milk Orange, carrot, pineapple, passion fruit, mango, guava,
Sugar, fiber, pectin, acidulant (citric acid), vitamins A, C, and E, and apricot, and papaya
aromas
9Ba Juice 25.6%, skim milk 16.5% Peach, apricot
Water, sugar, calcium salts, pectin, acidifier (citric acid), vitamin C,
aroma, and colorant (-carotene)
10Ba Juice 25.6%, skim milk 16.5% Orange, pineapple
Water, sugar, calcium salts, pectin, acidifier (citric acid), vitamin C,
aroma, and colorant (-carotene)
11C Juice 20%, milk 3% to 5% Apple
12C Juice 20%, milk 3 to 5% Mango and peach
13D Juice 33.7%, skim milk 10% Orange, carrot, lemon, apricot, passion fruit, pineapple,
Saccharose, dextrose, fiber (pea, soy, orange, apple), pectin, natural and haw
aromas and aromas identical to natural ones, vitamins A, C, and E
14D Juice 42.5%, skim milk 10% Apple, kiwi, orange, and lime
Saccharose, dextrose, pectin, natural aromas and aromas identical to
natural ones, vitamins A, C, and E, colorants (extracts of chlorophyll
and curcuma)
15D Juice 7%, skim milk 10% Mango and pineapple
Saccharose, dextrose, pectin, natural aromas and aromas identical to
natural ones, vitamins A, C, and E, natural colorant (E 160a)
16D Juice 35%, skim milk 12% Apple (20%), orange (10%), banana (5%)
Saccharose, dextrose, pectin, natural aromas and aromas identical to
natural ones, vitamins A, C, and E, colorant (E 100)
17E Juice, skim milk Orange
Vegetable oil (rich in conjugated linoleic acid, 0.6%), citric acid,
stabilizers (E 440, E 471), sweeteners (E 950, E 951, E 959), and
natural aroma
a
Samples stored under refrigeration (4 2 C).
(AD) The same letters indicate the same manufacturer.

C458 JOURNAL OF FOOD SCIENCEVol. 72, Nr. 9, 2007

Carotenoids and color of juicemilk mixtures . . .

methanolic KOH (0.1%, w/v, BHT) (1:1, v/v) for 0.5 h at room tem-
perature.
Identification and quantification of carotenoids. The
carotenoids (including geometrical isomers) in the beverages
analyzed were identified and quantified by UV-vis spectra and
retention times in HPLC using the HP ChemStation A.06.03 sys-
tem. The different Z isomers were identified using the Q ratio =
D IImax /D B , which is the quotient between band II (normally max )
and the cis peak band (Saleh and Tan 1991; Chen and Chen 1994;
Chen and others 1995).
We quantified the different carotenoids in accordance with their
similarity in terms of chemical-structural behavior and polarity,
with respect to the validated standards for which we had a cali-
Statistical analysis
Analyses were performed in triplicate. All statistical analyses
were performed using SPSS
14.0 for Windows (SPSS Inc., Chicago,
Ill., U.S.A.). Significant differences between the results were calcu-
lated by analysis of variance (ANOVA) using Tukeys test, and a mul-
tiple regression analysis was performed to study the influence of
different factors on a given parameter. Finally, Pearsons correla-
tion test was conducted to determine the linear correlations among
variables.
Results and Discussion

C: Food Chemistry & Toxicology

bration curve: lutein, zeaxanthin, and -carotene. Accordingly, for
the quantification we proceeded as follows: the chromatogram was
separated into 3 parts; all of the carotenoids up to and including
lutein were quantified as such, and the remaining xanthophylls
were quantified as zeaxanthin. The carotenoids were quantified
with the -carotene standard. Each carotenoid was quantified in
accordance with whether its maximum was close to 290, 350, 430,
or 450 nm (Cortes and others 2004).
Color measurement
The color of the different beverages was measured with a
Hunter Labscan II spectrophotometric colorimeter (Hunter As-
sociates Laboratory Inc., Reston, Va., U.S.A.) controlled by a
computer that calculated color coordinates from the reflectance
spectrum (Calvo and Duran 1997). The results were expressed in
accordance with the CIELAB system with reference to illuminant
D65 and with a visual angle of 10 . The samples were placed in an
optical glass tray, using the white plate of the colorimeter as the
background (standard white plate nr LS 13681 11/86, X = 78.50,
Y = 83.32, Z = 87.94). This background was used to standardize
the measurements. The results were expressed as tristimulus val-
ues corresponding to the CIELAB uniform color space (CIE 1978)
(L : lightness [0 = black, 100 = white], a [a = greenness, +a =
redness], and b [b = blueness, +b = yellowness]). Three consec-
utive measurements of each sample were taken.
These values were then used to calculate hue degree (h0 = arc
tangent [b /a ]) and chroma [C = (a2 + b2 )1/2 ], which is the in-
tensity or color saturation.
Color
Color values of the samples analyzed are shown in Table 2. One
of the peculiarities of these commercial samples was the addition
of artificial colorings which are allowed by the European Union
(EU) under Directive 94/36/EC (EPC 1994). Specifically, sample 3A
contained the artificial colorant allura red (E 129), which corre-
sponded to a high value of a (18.61 0.18). Sample 9B contained
the colorant -carotene, corresponding to the highest value of b
(62.76 0.09). Sample 14D contained kiwi and lime with chloro-
phyll and curcumin, corresponding to the most negative value of
a (13.96 0.16). Sample 15D contained natural colorant E 160a
(mixture of carotenoids), although no correlations with color pa-
rameters were found. Finally, sample 16D contained E 100 colorant
(curcumin), corresponding to yellowish colors.
The 1st attribute of color, the lightness value (L ), is associ-
ated with luminous intensity, which describes the light-reflecting
or -transmitting capacity of an object. This parameter correlated
with the percentage of milk in these beverages (r = 0.649, P < 0.01)
(n = 12, excluding samples which did not indicate the percentage
of milk on their labels).
CIE a has positive values for reddish colors and negative val-
ues for greenish ones. The presence of apple and kiwi + lime cor-
related negatively (P < 0.01) with CIE a and was associated with
green colors (r = 0.618, r = 0.591, respectively), whereas car-
rot, apricot, and passion fruit had statistically significant (P < 0.05)
correlations with CIE a and were accompanied by red colors
(r = 0.533, r = 0.566, r = 0.533, respectively). Finally, apricot and
peach had significant (P < 0.05) correlations with CIE b (r = 0.607,
r = 0.486, respectively) and were accompanied by yellow colors.

Table 2 --- Color characterization of samples analyzed.

Sample L a b h C
1A a
67.61 0.11 3.11 0.12 42.71 0.35 85.84 0.13 42.83 0.35
2Aa 66.25 0.01 7.79 0.17 56.20 0.63 82.11 0.08 56.75 0.64
3Aa 60.66 0.11 18.61 0.18 13.25 0.06 35.47 0.38 22.85 0.11
4Aa 64.83 0.03 8.90 0.01 58.39 0.12 81.34 0.01 59.07 0.12
5A 62.32 0.21 2.76 0.20 22.57 0.73 96.96 0.28 22.75 0.74
6A 60.15 0.05 7.81 0.04 56.41 0.23 82.12 0.01 56.95 0.23
7A 57.87 0.08 5.00 0.10 51.84 0.03 84.50 0.11 52.08 0.04
8A 53.05 0.01 20.76 0.04 48.57 0.64 66.86 0.23 52.83 0.60
9Ba 62.87 0.11 10.17 0.17 62.76 0.09 80.80 0.16 63.59 0.06
10Ba 69.79 0.12 1.57 0.09 42.22 0.07 87.87 0.12 42.25 0.07
11C 53.23 0.38 0.44 0.19 17.67 0.26 91.44 0.64 17.67 0.25
12C 67.60 0.03 2.62 0.01 38.27 0.03 86.08 0.08 38.36 0.03
13D 57.37 0.07 13.45 0.01 48.22 0.22 74.41 0.06 50.07 0.21
14D 54.89 0.03 13.96 0.16 29.18 0.03 115.57 0.26 32.35 0.07
15D 60.26 0.18 0.16 0.09 33.96 0.25 90.28 0.16 33.96 0.36
16D 65.62 0.06 6.20 0.05 31.00 0.25 101.31 0.06 31.62 0.43
17E 80.62 0.06 6.88 0.23 50.04 0.25 82.17 0.32 50.52 0.29
a
Samples stored under refrigeration (4 2 C).

Vol. 72, Nr. 9, 2007JOURNAL OF FOOD SCIENCE C459

 Carotenoids and color of juicemilk mixtures . . .

Hue value (h ) corresponds to whether the object is red, orange,
yellow, green, blue, or violet. The h value of the fruit juicemilk
beverages ranged from 35.47 to 115.57, owing to the high variabil-
ity in color among the samples. Typically yellow-red fruits and veg-
etables present in these beverages, such as peach, apricot, carrot,
guava, papaya, and passion fruit, were associated with h0 < 90 ,
whereas apple, lemon, lime, and kiwi fruits in the juices were as-
sociated with h0 > 90 . On the other hand, chroma (C ) describes
the degree of saturation, purity, or intensity of color. There was a
statistically significant correlation (P < 0.05) between the chroma
value and the presence of apricot (r = 0.646) or peach (r = 0.464),
indicating a high intensity of color.
With regard to storage conditions, samples stored under re-
C: Food Chemistry & Toxicology
mental carotenoids are more sensitive to isomerization than natu-
ral ones.
-Carotene was the predominant carotenoid in most of these
beverages, and mango and carrot were found to contribute signifi-
cantly to the content of -carotene (r = 0.490, r = 0.530, P < 0.05).
However, it was difficult to establish a correlation between
-carotene and the content of specific fruits because of the addi-
tion of supplemental -carotene by the manufacturers. Neverthe-
less, it is widely known that carotenoid levels in common fruits
such as apple, grape, lemon, pear, strawberry, kiwifruit, pineap-
ple, and banana are low (Yano and others 2005). Note that bev-
erage 11C, consisting exclusively of apple juice + milk, was poor
in carotenoids (Table 3). According to the Inst. of Medicine (IOM

frigeration had greater lightness (L ) than those stored at room
temperature (65.33 3.28 and 59.24 4.71, respectively), greater
yellowness (45.92 17.25 and 38.88 12.64, respectively),
and higher color saturation (47.88 14.36 and 39.92 13.00,
respectively).
Carotenoid content
The concentrations of carotenoids with provitamin A activity are
shown in Table 3. When we quantified total carotenoids (including
geometrical isomers), large differences in the levels of carotenoids
among beverages were found. In Figure 1, some peaks stand out
from the others because in some cases exogenous carotenoids were
added by the manufacturers after treatment rather than before
treatment, in order to be present with fewer modifications in the
final product. Fourteen of the manufacturers declared the addi-
tion of vitamin A (Table 1), but none of them specified the form
in which vitamin A was added with the exception of manufacturer
B, who specified that the addition was in the form of -carotene.
Legal guidelines for vitamin food fortification indicate that vitamin
A can be added as retinol, retinyl acetate, retinol palmitate, or -
carotene (Annex II, BOE 2003). The quantities added should pro-
vide at least 15% and no more than 100% of the recommended daily
intake (RDI). Although samples 11C, 12C, and 17E did not indicate
the addition of vitamin A on their labels, the LC analysis clearly re-
vealed the addition of -carotene in 17E (Figure 2), because orange
juice is not as rich in -carotene as the chromatogram shows. More-
over, as can be seen from Figure 2, the -carotene is in the form
of its 13-Z isomer, as the spectrum exhibits the typical Z band be-
tween 300 and 350 nm. This could be due to the fact that supple-
2001), 1 retinol activity equivalent (mg RAE) = 1 mg of all-E retinol,
= 2 mg of supplemental all-E -carotene, = 12 mg of dietary all-E
-carotene, = 24 mg of other dietary provitamin A carotenoids. As it
was not possible to distinguish which amounts of -carotene corre-
sponded to the supplemental form and which belonged to the fruit
composition, the exact quantity of vitamin A is unknown.
On the other hand, -carotene was found in 14 of the samples
analyzed. The amounts of this carotenoid were statistically corre-
lated (P < 0.01) with the presence of apricot (r = 0.591) and carrot
(r = 0.926).
The xanthophyll -cryptoxanthin was found in 13 of the samples
analyzed. The amounts of this carotenoid were statistically corre-
lated (P < 0.05) with the presence of apricot and peach in the sam-
ples (r = 0.902, r = 0.520, respectively).
-Carotene was found in beverages 8A (14.24 2.19 g/100 g)
and 13D (25.31 2.32 g/100 g) because of the presence of passion
fruit in their composition. -Carotene is the principal carotenoid
in passion fruit, as reported by Mercadante and others (1998b) and
Yano and others (2005).
In some cases, the predominant carotenoid was lost. For exam-
ple, violaxanthin, which is the principal carotenoid in fresh mango,
was not detected in the processed beverages that contained this
fruit (2A, 7A, 8A, 12C, and 15D). However, auroxanthin, which
is produced by the conversion of the 5,6-epoxide groups of vio-
laxanthin to the 5,8-furanoxide groups of auroxanthin, appeared
at an appreciable level (27.90 6.96 g/100 mL) in sample 12C
(mango and peach fruit) and was the principal carotenoid (results
not shown). These results are in agreement with those obtained
by Mercadante and Rodriguez-Amaya (1998a), who found that

Table 3 --- Carotenoids with provitamin A activity and their stereoisomers in the beverages analyzed.
-crypto- Z - 9-Z 13-Z Total
xanthin cryptoxanthin -carotene -carotene -carotene -carotene carotenoids
Sample (g/100 g) (g/100 g) (g/100 g) (g/100 g) (g/100 g) (g/100 g) (g/100 g)
1Aa 12.30 2.66 14.33 4.95 3.84 0.27 25.30 3.17 54.86 0.59 --- 175.43 15.73
2Aa --- 28.18 1.60 9.18 0.20 83.60 1.58 --- 154.00 0.75 638.86 2.45
3Aa --- 14.45 2.48 --- 2.61 0.48 --- --- 36.88 5.18
4Aa 146.07 4.22 --- 6.24 0.40 75.55 5.48 --- 129.32 13.01 642.46 42.12
5A 12.91 2.12 --- 0.31 0.13 1.64 0.30 --- --- 64.49 1.16
6A 71.48 5.35 --- 6.34 1.05 53.39 10.23 --- 91.50 6.75 359.15 17.20
7A --- 29.68 3.86 4.62 0.28 29.03 4.47 --- 56.43 8.05 268.54 4.27
8A --- --- 74.28 8.58 159.31 30.42 --- --- 615.73 38.37
9Ba 45.95 5.61 9.09 0.06 140.78 3.45 --- 158.21 8.58 683.88 1.12
10Ba 20.92 8.20 --- 2.46 0.35 42.42 3.23 81.68 5.24 --- 181.31 14.14
11C --- --- --- 0.50 0.03 --- --- 8.83 0.36
12C --- 17.68 2.47 1.28 0.45 20.7 6.47 --- --- 170.24 46.35
13D --- 13.08 2.57 40.99 14.31 113.53 37.87 --- --- 312.86 41.08
14D --- --- --- 2.28 0.06 --- --- 20.17 2.74
15D --- --- 1.91 0.83 74.47 1.82 --- 168.56 21.80 257.92 18.27
16D --- 8.39 1.20 0.19 0.15 1.21 0.66 --- --- 27.67 5.29
17E --- 37.08 0.96 7.75 0.54 --- --- 247.81 8.48 331.45 7.09
a
Samples stored under refrigeration (4 2 C).

C460 JOURNAL OF FOOD SCIENCEVol. 72, Nr. 9, 2007

Carotenoids and color of juicemilk mixtures . . .

violaxanthin, the principal carotenoid in fresh mango, was not de-
tected in 3 commercial brands of mango juice and auroxanthin ap-
peared at an appreciable level. Lutein and zeaxanthin were only
found in samples 10B (2.53 0.35 and 4.31 0.95 g/100 mL, re-
spectively) and 4A (2.36 0.14 and 4.06 0.57 g/100 mL, respec-
tively). The importance of these 2 carotenoids is due to their pro-
tection against macular degeneration connected with age, through
their ability to capture free oxygen and blue light in the retina
(Landrum and Bone 2001). However, owing to the presence of oxy-
gen in their chemical structures, these xanthophylls are more sus-
ceptible to degradation during thermal treatment and storage.
With regard to the occurrence of provitamin A carotenoid Z iso-
mers, 9 of the 13 beverages containing -cryptoxanthin were in
the form of its Z isomer, and 9 of the samples presented isomers
of 9-Z and 13-Z -carotene, which could originate from thermally
and light-induced isomerization of all-E -carotene during pro-
cessing and storage. According to Mnguez-Mosquera and others
(2002), considerably lower relative provitamin A activities of 53%
and 38% are observed for 13-Z -carotene and 9-Z -carotene,
respectively. Rodriguez-Amaya (1999) found that the principal Z
isomers in processed red, yellow, and orange fruits and vegeta-
bles are 13-Z, 9-Z, and 15-Z isomers. The production of these

C: Food Chemistry & Toxicology

mAU 11.101
11.101 min
Norm . 13-Z--carotene 11.344 min
Norm
700 all-E--carotene
350 300

300
250
600
250
200
200
500 150 150

100
100
400 50

0 50

250 300 350 400 450 500 550 nm

11.344
300 0

250 300 350 400 450 500 550 nm

200

100

0 2.5 5 7.5 10 12.5 15 17.5 20 22.5 min

Figure 1 --- Chromatogram of sample 15D at = 450 nm.

mAU
11.118
Norm.
11.118 min
1200
2000 13-Z--carotene
1000

800

600

1500 400

200

0
1000 250 300 350 400 450 500 550 nm

500

0 2.5 5 7.5 10 12.5 15 17.5 20 22.5 min

Figure 2 --- Chromatogram of sample 17E with spectrum of 13-Z -carotene ( = 450 nm).

Vol. 72, Nr. 9, 2007JOURNAL OF FOOD SCIENCE C461

 Carotenoids and color of juicemilk mixtures . . .
beverages involves various technological steps that may affect
their carotenoid contents and isomeric composition. The tem-
peratures reached during the concentration of the juice and the
impact of light on the product are 2 important factors that in-
fluence the formation of Z isomers. Lessin and others (1997)
found an increase of 10% to 39% in the percentage of total Z
isomers of provitamin A carotenoids in various fruits and veg-
etables as a consequence of thermal treatment. Marx and others
(2000) found large relative amounts of 9-Z and 13-Z -carotene
(31.8% and 44.5%, respectively) in juices supplemented with syn-
thetic -carotene, compared with 6.7% to 13.6% of these stereoiso-
mers in natural carrot juices. This is in agreement with our re-
sults, as samples 9B and 15D, which declare the addition of
C: Food Chemistry & Toxicology
colorings (consisting of carotenoids), showed high levels of Z
isomers.
Relationship between color and carotenoids
Figure 3 shows the location of the beverages studied in the a b
plane. The dotted line represents the hue angle (h0 ). What is partic-
ularly interesting is the fact that the samples situated on the upper
right side, which indicates red-yellow colors, are the samples that
had the highest content in carotenoids, whereas the beverages lo-
cated on the left side (negative a values and h0 > 90 ) were char-
acterized by their low carotenoid content. Total carotenoid content
was well correlated (P < 0.05) with a (r = 0.755) and b (r = 0.776)
values.
In order to analyze the prediction of total carotenoids from color
parameters we performed a multiple regression analysis, fitted to
the following linear model:
Total carotenoids (g/100 g) = 125.93 + 11.56(a ) + 8.238(b ) among marketed beverages, especially those with provitamin A
The R-squared statistic indicated that the model fitted explained
69% of the variability in total carotenoids (P < 0.01).
With regard to individual carotenoids, a high correlation be-
tween -carotene and a values (r = 0.736, P < 0.01) was found.
Samples containing from 1.28 to 9.18 g/100 g of -carotene had
similar h values (from 80.80 to 87.87), being located in the reddish-
yellow color region (Figure 3). Sample 8A had the highest a value
C462 JOURNAL OF FOOD SCIENCEVol. 72, Nr. 9, 2007
and the highest amount of -carotene (74.30 8.58 g/100 g). Al-
though -carotene was not present in sample 3A, it had a high a
value, which could be attributed to the addition of allura red ar-
tificial coloring and/or to a high content of phenolic compounds
(anthocyanins) resulting from the inclusion of strawberry, as we
found in a previous work (Zulueta and others 2007). On the other
hand, -cryptoxanthin correlated with b (r = 0.507, P < 0.01) and
-carotene correlated with a (r = 0.604, P < 0.05) and b (r = 0.643,
P < 0.01). Melendez-Martnez and others (2007) studied the rela-
tionship between color and the chemical structure of carotenoid
pigments, finding that -cryptoxanthin and -carotene (both with
11 carbon double bonds) are yellowish.
Despite the addition of E 160a coloring (mixture of carotenoids),
beverage 15D showed an h value of approximately 90 , which
might be due to its low content of fruit juice (7%). Alternatively,
it might be the result of isomerization of -carotene to its 13-
Z isomer, leading to a decrease in color, as reported by Bauern-
feind (1984), who observed a hyposochromic shift in max and
smaller extinction coefficients for Z isomers. Z isomerization of
a chromophores double bond causes a slight loss in color, a
small hyposochromic shift (usually 2 to 6 nm for mono-Z), and a
hypochromic effect, as reported by Rodriguez-Amaya (2001). These
observations are in agreement with our results, as we found that
13-Z and 9-Z -carotene were correlated with a decrease in color in
comparison with their all-E isomers.
Conclusions
T hese beverages are proposed as a good vehicle for providing
consumers with vitamins and bioactive compounds. However,
the present study reveals the high variability in carotenoid contents
activity.
Acknowledgments
This study was carried out with funds from the Spanish Min-
istry of Science and Technology and European Regional Devel-
opment Funds (ERDF) (project AGL-2003-05236-C02-02 and AGL-
2006-13320-C03-03/Ali) and the Generalitat Valencianas Aid for
Research Groups (3/147 and GV/2007/048). Ana Zulueta holds an
Figure 3 --- Location of the beverages
studied within the a b plane.
Carotenoids and color of juicemilk mixtures . . .
award from this project. We are grateful to Pau Arce Vila for his tech-
nical assistance.
References
Baker R, Gunter C. 2004. The role of carotenoids in consumer choice and the likely
benefits from their inclusion into products for human consumption. Trends Food
Sci Technol 15(10):4848.
Bauernfeind JC. 1984. Natural food colors. In: Bauernfeind JC, editor. Carotenoids as
colorants and vitamin A precursors. New York: Academic Press. p 137.
[BOE] Boletn Oficial del Estado. 2003. 14 Octubre. Real Decreto 1275/2003 de 10 de
octubre relativo a los complementos alimenticios. Madrid.
Britton G. 1995. Structure and properties of carotenoids in relation to function. FASEB
J 9:15518.
Calvo C, Duran L. 1997. Propiedades fsicas. II. Opticas y color. In: Temas en tec-
nologa de alimentos.: Programa iberoamericano de ciencia y tecnologa para el
desarrollo (CYTED). Mexico, DF: Instituto Politecnico Nacional.
Calvo C, Salvador A, Fiszman SM. 2001. Influence of colour intensity on the percep-
tion of colour and sweetness in various fruit flavoured yoghurts. Eur Food Res Tech-
nol 213(2):99103.
Carle R. 1999. Physical and chemical stability of ATBC drinks. Fruit Processing 9:342
9.
Chen BH, Peng HY, Chen BH. 1995. Change of carotenoids, color, and vitamin A con-
tents during processing of carrot juice. J Agric Food Chem 43(7):19128.
Chen TM, Chen BH. 1994. Optimization of mobile phases for HPLC of cis-trans
carotene isomers. Chromatographia 39(6):34654.
[CIE] Commission Intl de lEclairage. 1978. Recommendations on uniform color
spaces, color-difference equations, psychometric color terms. CIE publication no.
15 (E-1.3.1), Supplement 2. Vienna: Bureau Central de la CIE.
Cortes C, Esteve MJ, Frgola A, Torregrosa F. 2004. Identification and quantification
of carotenoids including geometrical isomers in fruit and vegetable juices by liq-
uid chromatography with ultraviolet-diode array detection. J Agric Food Chem
52(8):220312.
Cortes C, Esteve MJ, Frgola A, Torregrosa F. 2005. Changes in carotenoids includ-
ing geometrical isomers and ascorbic content in orange-carrot juice during frozen
storage. Eur Food Res Technol 221(12):12531.
Cortes C, Torregrosa F, Esteve MJ, Frgola A. 2006. Carotenoid profile modification
during refrigerated storage in untreated and pasteurized orange juice and or-
ange juice treated with high-intensity pulsed electric fields. J Agric Food Chem
54(17):624754.
Dugave C, Demange L. 2003. Cis-trans isomerization of organic molecules and
biomolecules: implications and applications. Chem Rev 103(7):2475532.
[EPC] European Parliament and Council. 1994. Directive 94/36/EC of 30 June 1994
on colours for use in foodstuffs. OJ No. L237, European Parliament and Council. p
1329.
[IOM] Institute of Medicine. 2001. Dietary reference intakes for vitamin A, vitamin K,
arsenic, boron, chromium, copper, iodine, iron, manganese, molybdenum, nickel,
silicon, vanadium, and zinc. Washington, D.C.: Natl. Academy Press.
Kuzminski LN. 1999. Issues and pressures for food and beverage research
and development in the 21st century. Crit Rev Food Sci Nutr 39(1):1
11.
Landrum JT, Bone RA. 2001. Lutein, zeaxanthin, and the macular pigment. Arch
Biochem Biophys 385(1):2840.
Lessin WJ, Catignani GL, Schwartz SJ. 1997. Quantification of cis-trans isomers of
provitamin A carotenoids in fresh and processed fruits and vegetables. J Agric Food
Chem 45(10):372832.
Marx M, Schieber A, Reinhold C. 2000. Quantitative determination of carotene
stereoisomers in carrot juices and vitamin supplemented (ATBC) drinks. Food
Chem 70(3):4038.
Melendez-Martnez A, Britton G, Vicario I, Heredia F. 2007. Relationship be-
tween colour and the chemical structure of carotenoid pigments. Food Chem
101(1):114550.
Mercadante AZ, Rodriguez-Amaya DB. 1998a. Effects of ripening, cultivar differences,
and processing on the carotenoid composition of mango. J Agric Food Chem
46(1):12830.
Mercadante AZ, Britton G, Rodriguez-Amaya DB. 1998b. Carotenoids from
yellow passion fruit (Passiflora edulis). J Agric Food Chem 46(10):4102
C: Food Chemistry & Toxicology
6.
Mnguez-Mosquera MI, Hornero-Mendez D, Perez-Galvez A. 2002. Carotenoids and
provitamin A in functional foods. In: Hurst WJ, editor. Methods of analysis for func-
tional foods and nutraceuticals. Boca Raton, London, New York, Washington, D.C.:
CRC Press. p 10157.
Ministerio de Agricultura, Pesca y Alimentacion (MAPA) of Spain. 2003. Spain total
citric transformation 20022003 campaign.
Pszczola DE. 2005. Ingredients. Making fortification. Food Technol 59(4):44
61.
Rodriguez-Amaya DB. 1999. Changes in carotenoids during processing and storage of
foods. Arch Latinoam Nutr 49(3 Suppl 1):38S47S.
Rodrguez-Amaya DB. 2001. A guide to carotenoid analysis in foods. Washington,
D.C.: ILSI Press.
Saleh MH, Tan B. 1991. Separation and identification of cis/trans carotenoid isomers.
J Agric Food Chem 39(8):143843.
Schieber A, Marx M, Reinhold C. 2002. Simultaneous determination of carotenes and
tocopherols in ATBC drinks by high-performance liquid chromatography. Food
Chem 76(3):35762.
Sharma R. 2005. Market trends and opportunities for functional dairy beverages. Austr
J Dairy Technol 60:1958.
Simpson KL, Chichester CO. 1981. Metabolism and nutritional significance of
carotenoids. Annu Rev Nutr (1):35171.
Yano M, Kato M, Ikoma Y, Kawasaki A, Fukazawa Y, Sugiura M, Matsumoto H, Oohara
Y, Nagao A, Kazunori O. 2005. Quantitation of carotenoids in raw and processed
fruits in Japan. Food Sci Technol Res 11(1):138.
Zulueta A, Esteve MJ, Frasquet I, Frgola A. 2007. Vitamin C, vitamin A, phenolic com-
pounds and total antioxidant capacity of new fruit juice and skim milk mixture bev-
erages marketed in Spain. Food Chem 103(4):136574.
Vol. 72, Nr. 9, 2007JOURNAL OF FOOD SCIENCE C463