Biotechnology and Bioprocess Engineering 23: 116-121 (2018)

DOI 10.1007/s12257-017-0442-3pISSN 1226-8372

eISSN 1976-3816

Fruit Juice Supplementation Alters Human Skin Antioxidant Levels

In Vivo: Case Study of Korean Adults by Resonance Raman

Spectroscopy

Moon-Hee Choi, Han-Gyo Jo, Min-Ju Kim, Min-Jung Kang, and Hyun-Jae Shin

Received: 1 November 2017 / Revised: 27 December 2017 / Accepted: 4 January 2018

© The Korean Society for Biotechnology and Bioengineering and Springer 2018

Abstract Carotenoids, found in many fruits and vegetables,

are antioxidants that protect human skin from UV radiation.

In humans, fruit and vegetable intake increases carotenoid

contents in skin, which are conventionally assessed by

invasive blood tests. In this study, 47 healthy Korean subjects

(volunteers) consumed fruit juice containing tomato, apple,

strawberry, or grape three times per week for 6 weeks. Skin

antioxidant levels were measured by non-invasive resonance

Raman spectroscopy. The correlation between skin carotenoid

(SC) score with demographic data (age, height, weight) and

juice supplementation and changes in SC scores among

groups were analyzed. Variations in skin antioxidant levels

increased with juice supplementation (p < 0.05). Fruit juice

intake was significantly correlated with SC score, indicating

increased skin antioxidant levels. Grape and tomato increased

skin antioxidant levels and showed higher antioxidant

activity than other fruits. Fruit juices containing high levels

of carotenoids and antioxidants may provide modest benefits

to human health.

Keywords: skin carotenoid, antioxidant activity, Raman

spectroscopy, fruit and vegetable juice, Korean case study

1. Introduction

Carotenoids are the dominant pigments in autumn leaf

coloration in approximately 15 ~ 30% of tree species, but

many plant colors, particularly reds and purples, are

generated by other classes of chemicals [1]. Carotenoids

are found in numerous types of fruits and vegetables such

as tomato, carrot, spinach, and grape [2]. The food sources

and well-known structures of carotenoids are summarized

in Table S1. The two predominant carotenoids in human

tissues, β-carotene and lycopene, represent approximately

70% of the carotenoids in the skin and are found in the

liver, adrenal glands, fat, pancreas, lungs, testes, and kidneys

[3]. Carotenoids containing unsubstituted β-ionone rings,

include β-carotene, α-carotene, β-cryptoxanthin, and γ-

carotene, exhibit vitamin A activity [4]. These compounds

are involved in protection against light and contribute to

the prevention of external damage in humans, such as those

by UV-irradiation, smoke exposure, and contact with the

environment [5]. Numerous studies have examined the role

of antioxidants in the skin, particularly studies related to

anti-aging effects [6]. Among the many carotenoids, β-

carotene is an endogenous photoprotector whose efficacy

in preventing UV-induced damage has been widely

demonstrated [7,8]. Epidermal and dermal accumulation of

β-carotene may protect human skin from light-induced

disease [9]. In addition, lycopene is the major carotenoid in

tomatoes and provides photo-protection and antioxidation

to human skin [10,11]. Apples contain a carotenoid

biosynthesis pathway [12]. Given the increasing importance

of carotenoids in health and disease, methods for detecting

carotenoids in living human tissue are needed. Resonance

Raman spectroscopy (RRS) is a novel approach for nonin-

Moon-Hee Choi, Hyun-Jae Shin*

Department of Biochemical and Polymer Engineering, Chosun University,Gwangju 61452, KoreaTel: +82-62-230-7518; Fax: +82-62-230-7226E-mail: shinhj@chosun.ac.kr

Han-Gyo Jo, Hyun-Jae ShinDepartment of Chemical Engineering, Graduate School of ChosunUniversity, Gwangju 61452, Korea

Min-Ju Kim, Min-Jung KangBio-Food Research Center, Hurom Co. Ltd., Gimhae 50969, Korea

RESEARCH PAPER

Fruit Juice Supplementation Alters Human Skin Antioxidant Levels In Vivo: Case Study of Korean Adults by … 117

vasive, laser optical carotenoid detection. The RRS technique

is precise, specific, sensitive, and well suited to clinical and

field studies [13-15]. Most studies on the nutritional effect

of fruits and vegetables have been performed using Western

people. To analyze the effects of food intake in different

ethnic groups, information regarding the effect of fruit or

vegetable juice and/or extract intake in Korean adults is

necessary. In this study, RRS has been conducted to

determine the changes in antioxidant levels in the skin of

Korean adults following supplementation and consumption

of fruit juices containing high carotenoid levels.

2. Materials and Methods

2.1. Selection of test subjects and survey of health status

Changes in antioxidant levels following supplementation

and consumption of fruit juices were investigated. Forty-

seven healthy subjects (from among 60 people) from

Chosun University in Gwangju, aged 22 ~ 54 years, were

recruited for this study. Subjects completed a questionnaire

that included demographical, dietary, and lifestyle variables.

The questionnaires were reviewed to determine whether

the responses were inadequate and identify individuals

taking vitamin supplements (Table S2). This study was

approved by the IRB Committee of Chosun University

(IRB approval number: 2-1041055-AB-N-01-2015-0028).

2.2. Intake of fruit juices

Subjects were prohibited from taking other supplements for

1 month before and during fruit juice intake. The subjects

consumed fruit juices 3 times per week for 6 weeks (Fig. 1)

in consideration of the work schedules of the subjects. Fruit

juices were consumed a total 18 times. Fruits used in this

study included fresh tomato, apple, strawberry, or grape,

and juices were prepared with a blender (HE-DBF04,

Hurom, Gyeongsangnam-do, Korea). Participants were

divided into five groups: tomato, tomato-apple, strawberry-

apple, grape, and control. The control group did not

consume juice. The amount of juice provided was 300 mL

and the weight of fruit in the 300 mL of juice were

calculated by juice yield (Table 1).

Fig. 1. Experimental design for fruit-juice intake.

Table 1. Juice yield of fruits and amount of fruits per serving used in this study

FruitsJuice yield(%, v/w)

Amount of fruit per serving(g/300 mL)

Carotenoids content References

Tomato 44.6 ± 0.8 673 g 44552 6,620 μg/100 g dry wt. [18]

Apple 36.4 ± 2.2 824 g (412 g in mixed juice) 19.4 μg β-carotene/100 g [19]

Strawberry 32.8 ± 0.3 915 g (457 g in mixed juice) 500 ~ 2,000 μg/100 g dry wt. [20]

Grape 41.3 ± 0.8 728 g 567 μg/100 g dry wt [21]

Data were Mean ± S.D.

118 Biotechnology and Bioprocess Engineering 23: 116-121 (2018)

2.3. Detection of skin carotenoid score (SC score) by

resonance raman spectroscopy (RRS)

Skin carotenoid score as a measure of human antioxidant

level was detected using a resonance Raman spectro-

photometer (Bio Photonic scanner S3 Scanner, Pharmanex,

Provo, UT, USA). The S3 scanner uses blue light at a

wavelength of 478 nm; the wavelength shifts to 518 nm

upon detection of carotenoids because of photon reflection.

Photon reflection is quantified as the SC score (Fig. 2). An

iPad (Apple, Cupertino, CA, USA) was connected to an S3

scanner via Bluetooth and an iPad S3 scanner application

was used to operate the S3 scanner (Fig. 2A). Zero-

adjustment was conducted by covering the light exit

channel. Carotenoid levels in the skin of subjects were

measured on the palm of the hand. The SC score appeared

on the screen of the scanner when the measurement was

complete (Figs. 2B and 2C). The range of SC scores was

10,000 ~ 89,000, which was displayed as colors from red

to gray (Fig. 2D). The SC score of subjects was measured

once per week (Fig. 1B).

2.4. Antioxidant activity assay of fruits

In vitro antioxidant levels of fruits were measured by

determining the DPPH and ABTS radical-scavenging

activities. Measurements of scavenging activity of fruit

extracts against DPPH radicals were conducted as described

by Blois [16] with some modifications. Stock solutions of

samples (1.0 mg/mL) were diluted to final concentrations of

500 ~ 25 µg/mL in double-distilled water. Next, 0.3 mL of

a 0.1 mM DPPH methanol solution was added to 0.3 mL

of sample solutions of different concentrations and reacted

at 25°C. After 30 min, absorbance values were measured at

517 nm and converted into the percentage antioxidant

activity (AA) using the following formula:

DPPH scavenging activity of the fruit extract (IC50) is

the concentration required to reduce AA% by 50%.

Experiments were repeated three times. The ABTS assay

was conducted as described with some modifications [17].

The AA% formula was the same as that used in the DPPH

assay.

2.5. Statistical analysis

Statistical analysis was conducted using SPSS version 23

software (SPSS Inc., Chicago, IL, USA). T-test was performed

to analyze the antioxidant activities of fruits. Age, height,

and weight of subjects in each group were analyzed by

one-way analysis of variance (ANOVA). For the post-hoc

test, Duncan’s multiple range test was performed to identify

correlated variables. The correlation of age, height, or weight

and SC score was analyzed by two-way ANOVA. To

observe changes in the SC score, all scores were divided by

the SC score before fruit juice intake (normalization). The

normalized scores of each group were analyzed by one-

way ANOVA. The effects of fruits and time course in each

group were analyzed with the 95% confidence level.

AA % 100Abssample Absblank–( )

Abscontrol----------------------------------------------- 100×–=

NormalizationSC score after fruit juice intake

SC score before fruit juice intake-------------------------------------------------------------------------------=

Fig. 2. Resonance Raman apparatus used in this study. (A) Operation method, (B) function keys and monitors of S3 scanner, (C)instructions of the scanner, and (D) skin carotenoid score (SC score) range and its color variation.

Fruit Juice Supplementation Alters Human Skin Antioxidant Levels In Vivo: Case Study of Korean Adults by … 119

3. Results and Discussion

The carotenoid content in the fruit juices were measured

(Table 1). The antioxidant activity was highest in the grape

and tomato groups. The IC50 values for the DPPH

antioxidant activities of grape and tomato were 164 and

172 μg/mL, respectively (Fig. 3). The IC50 values of the

ABTS antioxidant activities of grape and tomato were 123

and 209 μg/mL, respectively. No significant differences in

demographic data were found, including age (F(47) = 0.136,

p > 0.05), height (F(47) = 1.385, p > 0.05), and weight

(F(47) = 0.955, p > 0.05), among the groups (Table 2).

There must be other flavonoids and phenolic compounds in

the juices. Thus, the antioxidant levels are thought to be not

only due to carotenoids. The correlation of age, height, and

weight with SC score were evaluated. The SC scores of

subjects are shown in Table 3. Age and height had no

correlation with SC score. However, significant SC score

changes in the tomato-apple group and control group were

found to be affected by weight [tomato-apple group F(47)

= 6.859, p < 0.01; control group F(47) = 3.708, p < 0.05].

After fruit juice intake, significant changes in the SC score

were observed starting after 3 weeks (3 weeks F(47) =

3.793, p < 0.05; 4 weeks F(47) = 2.748, p < 0.05; 5 weeks

F(47) = 2.756, p < 0.05; 6 weeks F(47) = 4.312, p < 0.01)

(Tables 2 and 3). Duncan’s multiple range test (p < 0.05)

Fig. 3. Antioxidant activity of fruits used in this study by (A) DPPH assay and (B) ABTS assay. Symbols: gallic acid (black circles),grape (black triangles), tomato (white inverted triangles), tomato and apple (black rectangles), and strawberry and apple (whitediamonds). Error bars indicate the variability of replicates within a single representative experiment (mean ± SD).

Table 2. Characteristics of subjects classified by dietary group in this study

Variables Age (years) Height (cm) Weight (kg)

Tomato (N = 11) 32.5 ± 8.3a 169.6 ± 6.2a 63.7 ± 13.3a

Tomato-apple (N = 9) 31.8 ± 9.1a 167.8 ± 8.8a 62.8 ± 12.6a

Strawberry-apple (N = 8) 31.8 ± 7.1a 164.4 ± 4.4a 59.1 ± 9.5a

Grape (N = 9) 30.9 ± 7.2a 164.2 ± 6.6a 62.8 ± 14.1a

Control (N = 10) 30.3 ± 3.3a 169.8 ± 8.3a 69.8 ± 9.5a

Mean ± S.D.avalues in the same column not sharing a common superscript were significantly different by Duncan’s multiple range test (p < 0.05).

Table 3. Normalized average of skin carotenoids score

Groups Before intake 1st week 2nd week 3rd week 4th week 5th week 6th week

Tomato (N = 11) 1.00 ± 0.00 1.02 ± 0.10a 0.99 ± 0.09ab 0.97 ± 0.11ab 0.97 ± 0.11ab 0.99 ± 0.11ab 0.97 ± 0.10b

Tomato-apple (N = 9) 1.00 ± 0.00 1.00 ± 0.08a 1.05 ± 0.13b 1.01 ± 0.02b 1.06 ± 0.17b 1.05 ± 0.18b 1.03 ± 0.03b

Strawberry-apple (N = 8) 1.00 ± 0.00 1.02 ± 0.07a 0.97 ± 0.04ab 0.95 ± 0.06ab 0.92 ± 0.10a 0.98 ± 0.12ab 0.98 ± 0.14b

Grape (N = 9) 1.00 ± 0.00 1.00 ± 0.06a 0.96 ± 0.07a 1.03 ± 0.12b 1.01 ± 0.10ab 1.04 ± 0.14b 1.06 ± 0.15b

Control (N = 10) 1.00 ± 0.00 0.98 ± 0.03a 0.95 ± 0.06a 0.90 ± 0.06a 0.91 ± 0.06a 0.88 ± 0.07a 0.87 ± 0.08a

Data were Mean ± S.D.a–b values in the same column not sharing a common superscript were significantly different by Duncan’s multiple range test (p < 0.05).

120 Biotechnology and Bioprocess Engineering 23: 116-121 (2018)

was performed to evaluate the results from 3 to 6 weeks. At

3 and 5 weeks, the tomato-apple and grape groups showed

significant differences from the control group. At 4 weeks,

only the tomato-apple group showed a statistically significant

difference from the control group. All groups were

statistically different from the control. Next, analysis of SC

score changes by time course during fruit juice intake was

conducted in each group (Table 3). In the control group,

there was a significant difference in SC score (F(70) = 7.516,

p < 0.001) after 3 weeks. In contrast, no fruit juice-intake

groups showed significant differences in SC score (tomato

group F(77) = 0.346, p > 0.05; tomato-apple group F(63) =

0.516, p > 0.05; strawberry-apple group F(56) = 1.054,

p > 0.05; grape group F(63) = 0.871, p > 0.05).

Carotenoids are fat-soluble pigments obtained by fruit

and vegetable ingestion and are transported into the blood

as lipoproteins to accumulate in the lipid layer of the

human epidermis. Carotenoid levels can be assessed by

measuring their concentration in the blood or skin. As

reported previously, 2.75 ~ 11 mg carotenoid-rich juice

(2.75 mg/30 mL) significantly increased skin carotenoid

status over an 8-week period among children aged 5 ~ 17

years [22]. Similarly, a study of children who received rice

supplemented with fruits (3.5 mg carotenoids) or β-carotene

powder (5.0 mg carotenoids) compared to a control diet for

30 days showed significantly higher plasma β-carotene

concentrations [23]. However, no significant differences

were detected in skin carotenoid status between participants

consuming a carotenoid-rich high-dose and carotenoid-rich

low-dose juice during the study [24]. These results are

similar to those of following feeding of a high (11 mg

carotenoids), low (4.4 mg carotenoids), and no-carotenoid

muffin to 10 children [25]. The reason for this large variation

in serum carotenoid is unclear, but has also been observed

in other high-carotenoid foods and supplements [26,27].

Previous studies revealed strong correlations (r2 = 0.81 and

r2 = 0.79, respectively; both p < 0.001) between the amounts

of carotenoids measured in the blood by high-performance

liquid chromatography and those measured in the skin by

resonance Raman spectroscopy [15,28].

This study chose five fruits as sources of fruit juice

according to the productivity and consumption of fruits in

Korea. For SC score data, fruit juice intake positively

influenced SC scores. A correlation was observed between

SC score and fruit juice supplementation. The tomato-apple

group showed a larger change in SC score than the tomato

group. Two factors may explain the SC score changes in the

two groups. First, apples contain high levels of flavonoids

and a strong carotenoids biosynthesis pathway [12,29].

Thus, skin carotenoid levels may be increased when

consumed with mixed juices containing tomato and apple.

Second, these results may be related to differences in the

absorption of carotenoids or differences in the ability to

cleave beta-carotene to vitamin A. Significant correlations

between skin carotenoids and fruit and vegetable intake,

particularly high-carotenoid vegetable intake, were previously

observed based on the questionnaires like Table S2 and

24 h recall [15,28,30]. Intake of orange-colored fruits

increased β-carotene concentration in the serum of Indonesian

children compared to intake of green-colored vegetables

[31].

A high body weight is significantly correlated with a

decreased SC score. In previous studies reported that

decreases in SC score occurred with weight loss in all

intervention groups [32]. This result differs, who found a

positive association between skin carotenoid score and fruit

and vegetable intake, whereas, not between skin carotenoid

score and age, height, weight, or body mass index [33].

Although results suggest variations in the effects of

high-carotenoid fruits and vegetables on serum carotenoid

levels, most previous studies examined children. In this

study, participants who consumed fruit juices generally had

higher SC scores than those who did not consume fruits

juices (control group). These findings were significant, and

thus skin carotenoid status may be used as a biomarker to

determine changes in carotenoid intake in human skin.

This is the first study to measure antioxidant levels in the

skin of Korean adults using non-invasive resonance Raman

spectroscopy.

4. Conclusion

The information provide in this paper is rather preliminary

but this study is of great significance to measure the effect

of fruit juice intake on Korean adults. In conclusion,

supplementation with fruit juice containing high levels of

carotenoids and antioxidants may provide benefits to

human health, particularly antioxidant-related effects. For

practical application, this simple method can be applied as

a juice-quality indicator to determine antioxidant levels.

Electronic Supplementary Material (ESM) The online

version of this article (doi: 10.1007/s12257-017-0442-3)

contains supplementary material, which is available to

authorized users.

References

1. Varakumar, S., Y. S. Kumar, and O. V. S. Reddy (2011)Carotenoid composition of mango (Mangifera indica L.) wineand its antioxidant activity. J. Food Biochem. 35: 1538-1547.

Fruit Juice Supplementation Alters Human Skin Antioxidant Levels In Vivo: Case Study of Korean Adults by … 121

2. Owusu, J., H. Ma, Z. Wang, N. A. Afoakwah, C. Zhou, and A.Amissah (2015) Effect of pH and temperature on antioxidantlevels of tomato wine. J. Food Chem. 39: 91-100.

3. Darvin, M. E., J. W. Fluhr, M. C. Meinke, L. Zastrow, W. Sterry,and J. Lademann (2011a) Topical beta-carotene protects againstinfra-red-light–induced free radicals. Exp. Dermatol. 20: 125-129.

4. Darvin, M. E., W. Sterry, J. Lademann, and T. Vergou (2011b)The role of carotenoids in human skin. Molecules 16: 10491-10506.

5. Fluhr, J. W., S. Sassnin, O. Lademann, M. E. Darvin, S.Schanzer, A. Kramer, H. Richter, W. Sterry, and J. Lademann(2011) In vivo skin treatment with tissue-tolerable plasmainfluences skin physiology and antioxidant profile in humanstratum corneum. Exp. Dermatol. 21: 130-134.

6. Masaki, H. (2010) Role of antioxidants in the skin: Anti-agingeffect. J. Dermatol. Sci. 58: 85-90.

7. Paiva, S. A. and R. M. Russell (1999) Beta-carotene and othercarotenoids as antioxidants. J. Am. Coll. Nutr. 18: 426-433.

8. Kohen, R. (1999) Skin antioxidants: Their role in aging and inoxidative stress-new approachs for their evaluation. Biomed.Pharmacother. 53: 181-192.

9. Kollias, N. and G. N. Stamatas (2002) Optical non-invasiveapproaches to diagnosis of skin diseases. J. Investig. Dermatol.Symp. Proc. 7: 64-75.

10. Scarmo, S., B. Cartmel, H. Lin, D. J. Leffell, E. Welch, P.Bhosale, P. S. Bernstein, and S. T. Mayne (2010) Significantcorrelations of dermal total carotenoids and dermal lycopenewith their respective plasma levels in healthy adults. Arch.Biochem. Biophys. 504: 34-33.

11. Hata, T. R., T. A. Scholz, I. V. Ermakov, R. W. McClane, F.Khachik, W. Gellermann, and L. K. Pershing (2000) Non-invasive raman spectroscopic detection of carotenoids in humanskin. J. Invest. Dermatol. 115: 441-448.

12. Ampomah-Dwamena, C., S. Dejnoprat, D. Lewis, P. Sutherland,R. K. Volz, and A. C. Allan (2012) Metabolic and gene expressionanalysis of apple (Malus x domestica) carotenogenesis. J. Exp.Bot. 63: 4497-4511.

13. Scarmo, S., K. Henebery, H. Peracchio, B. Cartmel, H. Lin, I. V.Ermakov, W. Gellermann, P. S. Bernstein, V. B. Duffy, and S. T.Mayne (2012) Skin carotenoid status measured by resonanceraman spectroscopy as a biomarker of fruit and vegetable intakein preschool children. Eur. J. Clin. Nutr. 66: 555-560.

14. van Rensburg, A. J. and F. Wenhold (2016) Validity andreliability of field esonance Raman spectroscopy for assessingcarotenoid status. J. Nutr. Sci. Vitaminol. 62: 317-321.

15. Zidichouski, J. A., A. Mastaloudis, S. J. Poole, J. C. Reading, andC. R. Smidt (2009) Clinical validation of a noninvasive, ramanspectroscopic method to assess carotenoid nutritional status inhumans. J. Am. Coll. Nutr. 28: 687-693.

16. Blois, M. S. (1958) Antioxidant determinations by the use of astable free radical. Nature 181: 1199-1200.

17. Arnao, M. B., A. Cano, and M. Acosta (2001) The hydrophilicand lipophilic contribution to total antioxidant activity. FoodChem. 73: 239-244.

18. Kim, H. K., J. H. Chun, and S. J. Kim (2015) Methoddevelopment and analysis of carotenoid compositions in varioustomatoes. Kor. J. Environ. Agric. 34: 196-203.

19. Kim, Y. N., D. W. Giraud, and J. A. Driskell (2007) Tocopheroland carotenoid contents of selected Korean fruits and vegetables.

J. Food Comp. Anal. 20: 458-465.20. Delgado-Pelayo, R., L. Gallardo-Gierrero, and D. Hornero-

Mendez (2015) Carotenoid composition of strawberry tree(Arbutus unedo L.) fruits. Food Chem. 199: 165-175.

21. Crupi, P., A. Coletta, and D. Antonacci (2010) Analysis ofcarotenoids in grapes to predict norisoprenoid varietal aroma ofwines from Apulia. J. Agric. Food Chem. 58: 9647-9656.

22. Aguilar, S. S., H. J. Wengreen, and J. Dew (2015) Skincarotenoid response to a high-carotenoid juice in children: Arandomized clinical trial. J. Acad. Nutr. Diet. 115: 1771-1778.

23. Vulong, L. T., S. R. Dueker, and S. P. Murphy (2002) Plasmabeta-carotene and retinol concentrations of children increase aftera 30 day supplementation with the fruit Momordica cochinchinensis(gac). Am. J. Clin. Nutr. 75: 872-879.

24. Holden, J. M., A. L. Eldridge, G. R. Beecher, I. M. Buzzard, S.Bhagwat, C. S. Davis, L. W. Douglass, S. Gebhardt, D. Haytowitz,and S. Schakel (1999) Carotenoid content of U.S. foods: Anupdate of the database. J. Food Comp. Anal. 12: 169-196.

25. Tanumihardjo, S. A., M. A. Horvitz, M. P. Dosti, and P. W.Simon (2009) Serum alpha- and beta-carotene concentrationsqualitatively respond to sustained carrot feeding. Exp. Biol. Med.(Maywood). 234: 1280-1286.

26. Borel, P., P. Grolier, N. Mekki, Y. Boirie, Y. Rochette, B. Le Roy,M. C. Alexandre-Gouabau, D. Lairon, and V. Azais-Braesco(1998) Low and high responders to pharmacological doses ofbeta-carotene: Proportion in the population, mechanisms involvedand consequences of beta-carotene metabolism. J. Lipid Res.39: 2250-2260.

27. Crane, T. E., C. Kubota, J. L. West, M. A. Kroggel, B. C.Wertheinm, and C. A. Thomson (2011) Increasing the vegetableintake dose is associated with a rise in plasma carotenoids withoutmodifying oxidative stress or inflammation in overweight orobese postmenopausal women. J. Nutr. 141: 1827-1833.

28. Aguilar, S. S., H. J. Wengreen, M. Lefevre, G. J. Madden, and J.Gast (2014) Skin carotenoids: A biomarker of fruit and vegetableintake in children. J. Acad. Nutr. Diet. 114: 1174-1180.

29. Lee, H. S., Y. H. Cho, J. Park, H. R. Shin, and M. K. Sung (2013)Dietary intake of phytonutrients in relation to fruit and vegetableconsumption in Korea. J. Acad. Nutr. Diet. 113: 1194-1199.

30. Resnicow, K., E. Odom, T. Wang, W. N. Dudley, D. Mitchell, R.Vaughan, A. Jackson, and T. Baranowski (2000) Validation ofthree food frequency questionnaires and 24-hour recalls withserum carotenoid levels in a sample of African-American adults.Am. J. Epidemiol. 152: 1072-1080.

31. de Pee, S., C. E. West, D. Permaesih, S. Martuti, Muhilal, and J.G. Hautvast (1998) Orange fruit is more effective than are dark-green, leafy vegetables in increasing serum concentrations ofretinol and beta-carotene in schoolchildren in Indonesia. Am. J.Clin. Nutr. 68: 1058-1067.

32. Zukley, L., P. Legowski, V. Nguyen, T. Geise, J. Lowndes, and K.Melanson, T. Angelopoulos, and J. Rippe (2004) The effect ofweight loss on dietary carotenoid and skin carotenoid levels insubjects participating in a weight loss study. Obesity Res. Suppl.12: A57.

33. Rerksuppaphol, S. and L. Rerksuppaphol (2006) Effect of fruitand vegetable intake on skin carotenoid detected by non-invasiveRaman spectroscopy. J. Med. Assoc. Thai. 89: 1206-1212.